Impact of environmental factors on viability and stability and high pressure pretreatment on
stress tolerance of Lactobacillus rhamnosus GG (ATCC 53103) during spray drying
vorgelegt von
Diplom-Ingenieur
Edwin Ananta
Von der Fakultät III – Prozesswissenschaften
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Ingenieurwissenschaften
- Dr.-Ing. -
genehmigten Dissertation
Promotionsausschuss:
Vorsitzender: Prof. Dr. H. Kunzek
Berichter: Prof. Dipl.-Ing. Dr. U. Stahl
Berichter: Prof. Dr. Dipl.-Ing. D. Knorr
Tag der wissenschaftlichen Aussprache: 29 April 2005
Berlin 2005
D 83
KURZFASSUNG
Es ist aus der Literatur bisher bekannt, dass die gesundheitsfördernde Wirkung von
probiotischen Bakterien (u.a. Wiederherstellung und Stabilisierung einer ausgewogenen
Darmflora, Verdrängung von Krankheitserregern und unerwünschten Bakterien im Darm,
Beeinflussung des Immunsystems und Verbesserung der natürlichen Abwehrkräfte des
Körpers, Senkung der Konzentration gesundheitsschädigender Stoffwechselprodukte und
krebsfördernder Enzyme im Dickdarm, Verhinderung bzw. Verkürzung von
Durchfallerkrankungen, usw.) nur mit lebenden Bakterien zu erzielen ist. Vorgeschlagen wird
eine tägliche Dosis von mindestens 108 lebenden Zellen, um einen gesundheitsrelevanten
Effekt beim Verzehr probiotischen Produkts zu bewirken. Damit diese Lebendkeimzahl
eingehalten werden kann, ist die Fähigkeit der probiotischen Bakterien in Lebensmitteln
sowie bei technologischen Verarbeitungsprozessen bis hin zum Ende der
Mindesthaltbarkeitsfrist sicherzustellen.
In dieser Hinsicht ist die Trocknung von probiotischen Bakterien zur Herstellung vom
probiotischen Pulver (als Endprodukt oder als Halbfabrikat), die üblicherweise durch die
Gefriertrocknung erfolgt, als ein kritischer Verarbeitungsschritt anzusehen. Da diese als sehr
schonend geltende Trocknungsart allerdings zeitaufwendig und relativ kostenintensivr
Prozess ist, wird in der letzten Zeit vermehrt untersucht, ob dieses Trocknungsverfahren
durch Sprühtrocknung ersetzt werden kann.
Diese Arbeit befasst sich in erster Linie mit der Evaluierung der Anwendbarkeit von
Sprühtrocknung als eine alternative Methode zur Herstellung von Probiotika enthaltenden
Trockenpräparaten. Dazu wurden optimale Trocknungsbedingungen ermittelt und die
verwendeten Schutzmedien auf thermo-physikalische Beschaffenheit und Wechselwirkung
mit einem synthetischen Membransystem untersucht. Eine Trocknungstemperatur von 80 °C
wurde als ein annehmbarer Kompromiss ermittelt. Bei dieser Temperatur ließ sich eine
probiotische Pulverzubereitung auf Milchbasis mit einer Restfeuchte von 4 Prozent und
einem Keimgehalt von ca. 109 Zellen pro g herstellen. Der Auswahl des Trägermediums
wurde eine große Rolle zugeschrieben, da dies das Überlebensverhalten während der
Lagerzeit determiniert, was letztlich das Qualitätsmerkmal der probiotischen Produkte und
somit die gesundheitsfördernde Wirkung vom Probiotikaverzehr beeinflußt. Obwohl partieller
Austausch der Magermilch durch kommerzielle präbiotische Substanzen unter Beibehaltung
einer konstanten Trägerstoffkonzentrationen keinen negativen Einfluss auf die unmittelbare
Überlebensrate von Lactobacillus rhamnosus GG bei der Sprühtrocknung hatte, fand im
Vergleich zu Magermilch eine stärkere Keimzahlreduktion in den beiden präbiotischen
Kombinationspräparaten während der Lagerung bei 37°C statt. Desweiteren wurde
festgestellt, dass das Vorliegen des Glaszustandes in dem Trägermaterial nur eine
untergeordnete Rolle bei der Gewährleistung der Lagerstabilität der darin befindlichen Keime
spielt. Die unterschiedliche Schutzwirkung der verwendeten Trägermaterialien müssen
vielmehr in der räumlichen bzw. chemischen Struktur der Schutzmoleküle liegen, die
wiederum Einfluss nehmen auf die Art und Weise, wie sie mit Zellmembranen bzw. ihre
Bestandteile in Wechselwirkung treten. Diese Sicht wurde durch weitere Arbeit mit der
Trocknung von den als Modellsystem für Doppellipidmembran dienenden Liposomen im
Beisein von Zuckermolekülen bestätigt. Ferner wurde zur weiteren Klärung über den
herausragenden Schutzeffekt von Magermilch auf die Lagerstabilität getrockneter Bakterien
der Einfluss von Milchproteinen untersucht. Es zeigte sich, dass bei der Lagerung die
Keimzahlreduktion bei Bakterien in proteolytisch behandelter Magermilch wesentlich
schneller erfolgte als die in nativer Magermilch.
Unterstützend wurde an einer analytischen Methode zur Charakterisierung der
Zellschädigung während der Sprühtrocknung gearbeitet. Dazu wurde die
durchflußzytometrische Analysenmethode verwendet. Dies benötigt entsprechende
Färbungs- und Messungsstrategie, die im Vorfeld etabliert werden musste. Bei der
Etablierungsphase wurde die Methode eingesetzt zur Beurteilung der Art und des Ausmaßes
der zellulären Schädigung, die unter Anwendung verschiedener physikalischer Behandlung,
wie beispielsweise Hochdruck-, Hitze-, oder Ultraschallbehandlung an den Zellen von L.
rhamnosus GG herbeigeführt wurde. Dafür wurde die Farbstoffkombination
Carboxyfluorescein-diacetat und Propidiumiodid angewandt. Es zeigt sich, dass während bei
der Hitzebehandlung oberhalb 60°C der Verlust der Membranintegrität als primäre Ursache
des hitzebedingten Zelltodes ausgemacht werden konnte, führte offensichtlich die
irreversible Schädigung der membrangebundenen Enzymsysteme zum
hochdruckinduzierten Zelltod. Anders als die hitzegetöteten Zellen bestand die Population
der druckinaktivierten Zellen hauptsächlich aus Zellen, die noch intakte Membranen und
aktives intrazellüläres Enzym besitzen. Hinsichtlich der Inaktivierung von L. rhamnosus GG
während Sprühtrocknung führten den Ergebnissen zufolge höhere Temperaturen während
der Trocknung zu stärkeren Schäden in der Zellmembran, welche als die Hauptursache für
die beobachtete Inaktivierung identifiziert werden kann.
Letztlich wurde die Eignung der Hochdruckvorbehandlung zur Verbesserung der
Hitzestabilität von L. rhamnosus GG überprüft. Perspektivisch ließe sich diese Art von
Vorbehandlung einsetzen, um die probiotischen Keime mit Hilfe ihrer induzierbaren
Abwehrmechanismen gegenüber der Hitze- und Dehydrationsstress bei der Sprühtrocknung
resistenter zu machen. Es zeigte sich, dass durch Druckvorbehandlung die Keime kurzzeitig
gegenüber letaler Hitze resistenter waren. Es wurde ebenfalls festgestellt, dass offensichtlich
die als Reaktion auf die Druckvorbehandlung synthetisierten Stressproteine an der erhöhten
Hitzeresistenz beteiligt sind. Anhand durchflusszytometrischer Analyse war es ersichtlich,
dass diese Metaboliten die bakteriellen Membranen einen erhöhten Schutz vor
hitzebedingter Schädigung verliehen.
ABSTRACT
Current literature data suggest that the health-promoting effect of probiotic bacteria (re-
establishment and stabilization of a balanced gut flora, prevention of pathogens outgrowth in
the intestine, impact on the immune system, reduction of the concentration of toxic
metabolites and carcinogenic enzymes in the large intestine, prevention of gastro intestinal
diseases, etc.) is only achieveable upon consumption of living bacteria. A daily dose of at
least 108 living cells has been suggested to assure health-relevant effects following the
consumption of probiotic product. To maintain this proposed number of living bacteria, the
survivability of probiotic bacteria during processing as well as in food has to be guaranteed
up to the end of the shelf life.
In this respect the drying process for the production of probiotic powder (as final or as
semifinished product), which is generally performed by freeze drying can be considered as a
critical processing step. However, since freeze drying is a time and cost intensive process,
there is an growing interest on the application of spray drying.
This work deals primarily with the evaluation of the applicability of spray drying as an
alternative method for the production of probiotic powder. Optimal drying conditions were
determined and the protective media were characterized in terms of their thermophysical
properties and capability of direct interaction with a synthetic membrane system. Spray
drying at temperature of 80 °C allowed production of probiotic powder on skim milk basis with
a residual moisture of 4% and a bacterial load of 109 cells per g. The selection of the carrier
medium is regarded crucial, since this medium determines the survival behavior during the
storage, which in turn affects the quality criteria of the probiotic products as well as the
declared health-promoting effect. Partial exchange of the skim milk solids with commercial
prebiotic compounds did not have any detrimental effect upon spray drying. However, the
inactivation of spray dried Lactobacillus rhamnosus GG during storage at 37°C was more
pronounced when the bacteria were dried in prebiotic preparations compared to the ones
dried in skim milk. Furthermore it was found that the formation of a glassy state contributed
only little to the maintenance of storage stability. The different protective effect of the drying
media applied was thought to be governed by the spatial and/or chemical structure of the
protective compounds, which could facilitate a direct interaction with cell membranes and/or
their components. This view was confirmed by further work regarding the drying of model
phospholipid bilayers, i.e. liposomes in the presence of different types of sugar molecules.
The influence of milk proteins was examined to further clarify the outstanding protection of
bacteria dried in skim milk. It was shown that probiotic bacteria dried in proteolytically treated
skim milk were inactivated faster than the one dried in native skim milk during storage.
Furthermore, work on flow cytometric analysis was conducted in order to allow the
characterization of cell damage during spray drying. This required appropriate staining and
measurement strategy, which had to be established first. Initially, this method was applied to
characterize cellular damage as affected by different physical treatments, including high
pressure, heat, or ultrasound. It was shown that during heat treatment beyond 60°C the loss
of the membrane integrity could be identified as the primary cause of heat-induced cell
death. In contrast, the irreversible damage of the membrane-bound enzyme systems was
responsible for high pressure-induced cell death. Compared to the heat-killed cells the
population of the pressure-inactivated cells consisted mainly of cells, which still had intact
membranes. With help of flow cytometric analysis it was demonstrated that upon spray
drying, the higher level of inactivation of L. rhamnosus GG at higher temperatures was
closely related to increased damage in the cell membrane.
The application of sub-lethal high pressure pretreatment was examined in terms of
evaluating possible approaches for the improvement of heat stability of L rhamnosus GG.
The rationale of applying this pretreatment is to take advantage of the inducible defense
mechanisms of the organisms to make them more resistant against stresses related to spray
drying, i.e. stress due to thermal exposure and dehydration. It was shown that pressure
pretreated bacteria showed higher thermotolerance as compared to unadapted ones.
Furthermore, it was found that apparently the stress proteins synthesized as response to the
pressure pretreatment are involved in the increased heat tolerance. On the basis of flow
cytometric analysis it was demonstrated that these proteins may have a protective effect on
the bacteria membranes, which led to an increased protection against heat-induced damage.
Table of contents
TABLE OF CONTENTS
1 GENERAL INTRODUCTION ................................................................................................................ 1
1.1 Probiotic products ................................................................................................................................. 2
1.2 Impact of probiotic on consumer’s health – Mode of action and selection criteria................................. 4
1.3 Technological approaches for improvement of viability retention.......................................................... 6
1.3.1 Genetic engineering of microorganism ............................................................................................. 7
1.3.2 Cultivation......................................................................................................................................... 8
1.3.3 Downstream processing ................................................................................................................... 9
1.4 The EU-Project PROTECH - QLK1-CT-2000-30042........................................................................... 14
1.5 Structure of PhD thesis ....................................................................................................................... 15
1.6 References.......................................................................................................................................... 16
2 FLOW CYTOMETRIC ANALYSIS FOR INACTIVATION STUDIES.................................................... 26
2.1 Introduction ......................................................................................................................................... 27
2.1.1 Effect of physical inactivation treatments on microorganisms......................................................... 27
2.1.2 Flow cytometry................................................................................................................................ 31
2.1.3 Determination of viability status of microorganism with fluororescence probes .............................. 34
2.1.4 Objective......................................................................................................................................... 41
2.2 Material and methods.......................................................................................................................... 41
2.2.1 Test organism................................................................................................................................. 41
2.2.2 Inactivation treatments and microbiological analysis ...................................................................... 41
2.2.3 Staining procedure and measurement strategies ........................................................................... 43
2.2.4 Flow cytometric measurement........................................................................................................ 43
2.2.5 Analysis of flow cytometric data...................................................................................................... 44
2.2.6 Statistical analysis .......................................................................................................................... 45
2.3 Results and discussion ....................................................................................................................... 45
2.3.1 Basic pattern................................................................................................................................... 45
2.3.2 Inactivation mechanisms by heat treatment.................................................................................... 46
2.3.3 Inactivation mechanisms by high hydrostatic pressure................................................................... 51
2.3.4 Combined application of heat and pressure ................................................................................... 61
2.3.5 Inactivation mechanism by high-intensity ultrasound...................................................................... 63
2.4 Conclusion .......................................................................................................................................... 66
2.5 References.......................................................................................................................................... 70
3 SPRAY DRYING OF PROBIOTIC BACTERIA ................................................................................... 78
3.1 Introduction ......................................................................................................................................... 79
3.1.1 Drying of microorganism................................................................................................................. 79
3.1.2 Spray drying ................................................................................................................................... 92
3.1.3 Spray drying works on lactic acid bacteria......................................................................................94
3.2 Objective............................................................................................................................................. 95
3.3 Material and methods.......................................................................................................................... 97
3.3.1 Test organism and preparation of bacterial suspension ................................................................. 97
3.3.2 Preparation of carrier solution......................................................................................................... 97
3.3.3 Spray drying ................................................................................................................................... 98
3.3.4 Determination of moisture content in spray dried powders............................................................100
Table of contents
3.3.5 Enumeration of probiotics after spray drying .................................................................................100
3.3.6 Staining procedure and flow cytometric assessment.....................................................................101
3.3.7 Storage test ...................................................................................................................................101
3.3.8 Differential scanning calorimetry measurement.............................................................................102
3.3.9 Calculation of glass transition temperatures ..................................................................................102
3.3.10 Monitoring direct interaction of sugar-membranes using liposomes ..............................................103
3.4 Results and discussion ......................................................................................................................107
3.4.1 Identifying critical processing conditions........................................................................................107
3.4.2 Flow cytometric analysis of spray dried bacteria............................................................................112
3.4.3 Incorporation of prebiotics in the spray drying medium..................................................................116
3.4.4 Storage test at non refrigerated conditions ....................................................................................119
3.4.5 The role of glassy state on bacterial storage stability ....................................................................123
3.4.6 Monitoring direct interaction of sugar-membranes using liposomes ..............................................129
3.4.7 The role of milk constituents in the protection................................................................................137
3.4.8 Role of milk constituents in conferring stability against low pH and bile acids...............................140
3.5 Conclusion .........................................................................................................................................142
3.6 References.........................................................................................................................................145
4 PRESSURE INDUCED STRESS RESPONSE..................................................................................154
4.1 Introduction ........................................................................................................................................155
4.2 Objective............................................................................................................................................163
4.3 Material and methods.........................................................................................................................163
4.3.1 Test organism................................................................................................................................163
4.3.2 Preparation of bacterial suspension...............................................................................................163
4.3.3 High pressure treatment ................................................................................................................164
4.3.4 Assessment of growth behavior after pressure treatment..............................................................165
4.3.5 Lethal heat challenge at 60°C........................................................................................................165
4.3.6 Plate enumeration method.............................................................................................................166
4.3.7 Mathematical description of heat inactivation kinetics ...................................................................166
4.3.8 Staining procedure with LIVE/DEAD® BacLight™ Bacterial Viability Kit........................................166
4.3.9 Flow cytometric measurement and data analysis ..........................................................................167
4.3.10 Statistical analysis .........................................................................................................................167
4.4 Results and discussion ......................................................................................................................168
4.4.1 Heat inducible thermotolerance of L. rhamnosus GG....................................................................168
4.4.2 Identification of non-lethal pre-treatment condition ........................................................................169
4.4.3 Post treatment growth behaviour of L. rhamnosus GG..................................................................170
4.4.4 Heat treatment at 60°C..................................................................................................................172
4.4.5 Determination of the role of de novo protein synthesis in inducible heat tolerance .......................177
4.4.6 Flow cytometric assessment of damaged on cellular membrane as affected by heat ...................179
4.4.7 Pressure induced tolerance against nisin and bile acid .................................................................181
4.4.8 Spray drying of pressure pre-treated bacteria ...............................................................................183
4.4.9 Pressure induced thermotolerance on other L. rhamnosus strain..................................................185
4.5 Conclusion .........................................................................................................................................185
4.6 References.........................................................................................................................................190
5 SUMMARY AND OUTLOOK..............................................................................................................196
Table of contents
6 ANNEXES..........................................................................................................................................203
6.1 Annex 1 : Fermentation profile of L. rhamnosus GG (API 50 CHL, Bio Merieux, France)..................203
6.2 Annex 2 : Detector’s configurations for flow cytometric analysis........................................................204
6.3 Annex 3 : Estimated residence time of dried particle in spray dryer...................................................205
6.4 Annex 4 : Technical specifications of Raftilose®P95 (Orafti, Tienen, Belgium) ..................................206
6.5 Annex 5 : Technical specifications of Polydextrose (Danisco, Copenhagen, Denmark) ....................207
6.6 Annex 6 : Technical specification of COROLASE®PP (AB Enzymes, Darmstadt, Germany).............208
6.7 Annex 7 : Kinetic of lactose degradation using ß-galactosidase (G-3665, Sigma, St. Louis, MO) .....209
6.8 Annex 8 : Regression parameters for heat inactivation curves (4.4.4)...............................................210
7 LIST OF DISSEMINATION ACTIVITIES............................................................................................212
8 CURRICULUM VITAE........................................................................................................................217
1
1 GENERAL INTRODUCTION
Technological aspects of the production of probiotic containing food products
General introduction 2
1.1 Probiotic products
Probiotic products represent a strong growth area within the functional foods group and
intense research efforts are under way to develop dairy and non-dairy products into which
probiotic bacteria such as Lactobacillus and Bifidobacterium species are incorporated. These
bacteria represent a unique group of lactic acid bacteria which is claimed to benefit human
health upon consumption, by improving the endogenous micro-flora of the gut, provided that
a sufficiently high number of viable and highly functional cells are consumed regularly [1, 2].
Moreover, a general consensus regarding the importance of high levels of live
microorganisms for probiotic products has already been highlighted [3]. The use of the word
probiotic is therefore restricted to products which contain live microorganisms in an adequate
dose in order to exert the desirable effects [4].
Microorganism, which is regarded to exert probiotic effect, is not exclusively lactic acid
bacteria. For instance, the application of yeast species Saccharomyces boullardii as probiotic
microorganism was regarded to be promising [5]. Moreover, spore forming bacteria, primarily
of the genus Bacillus have also been studied and commercialized as probiotics for human
and animal use [6]. Several species from the genus Enterococcus are also used as
probiotics. However, in recent years safety concerns have been raised on their use as
probiotics, since they have been associated with some serious infective diseases and with
multi-resistance to antibiotics – especially concerning the transmission of these resistances
to other strains [7].
When the application of probiotic is characterized in terms of the type of food products in
which they are harboured in, yoghurt is certainly one of the mostly used food vehicle for
probiotic consumption. Probiotic yoghurt was reported to account for 13% of all yoghurt sold
in Germany [8], where in year 2003 as many as 1,5 million tons of yoghurt were produced
and a yoghurt consumption per head of 15 kg was observed [9]. In Europe, the probiotic
yoghurt market alone was estimated to be worth a value in the region of € 900 million [10]
and made up approximately 65% of the total European functional food market [8].
Apart from yoghurt as the classical food vehicle, during recent years probiotics have been
increasingly incorporated into non-dairy food system. Based on the water activity of the food
products, where probiotics are incorporated in, there are two major food ecosystems, i.e. the
ecosystem of dehydrated products with low aw-value and the one of products with
intermediate to high aw-values. Trials already conducted to incorporate probiotic in different
food systems are documented in Table 1.
General introduction 3
Table 1.
Overview of works dealing with incorporation of probiotics into dairy or non-dairy products
Low aw food product
Food product Organism Reference
Oat-based cereal bar Bifidobacterium lactis Bb12 [11]
Freeze-dried yoghurt L. acidophilus, B. bifidum, B. longum [12]
Spray dried skim milk powder
with or without prebiotics
L. paracasei, L. salivarius, L. rhamnosus [13, 14]
Dry sausage L. acidophilus, L. crispatus, L. amylovorus,
L. gasseri, L. johnsonii, L. gallinarum, L.
rhamnosus
[15-18]
Dried fruits L. casei [19]
Intermediate to high aw food products
Food product Organism Reference
Yoghurt L. acidophilus, B. bifidum, L. casei, B.
breve, B. longum
[12, 20-24]
Skim milk with prebiotics Bifidobacterium [25]
Soymilk B. infantis, B. longum [26, 27]
Low fat quark cheese L. acidophilus [28]
Infant formula B. bifidum, B. breve, B. infantis, B. longum [29, 30]
Tomato juice L. acidophilus, L. plantarum, L. casei [31]
Mayonnaise B. bifidum, B. infantis [32]
Cheese B. lactis, L. acidophilus [33-36]
Cheese-based dip L. acidophilus, L. paracasei, L. rhamnosus,
B. animalis
[37]
Ice cream B. longum, B. brevi, B. infantis, L.
acidophilus, B. bifidum
[38-43]
Frozen yoghurt L. acidophilus, B. bifidum [44-46]
Chocolate L. reuteri [47]
General introduction 4
1.2 Impact of probiotic on consumer’s health – Mode of action and selection criteria
The probiotic concept was worked upon by Metchnikoff at the beginning of the century, who
tried to link the health and longevity of Bulgarian peasant, who consumed large quantity of
fermented milk, with the composition of their internal flora [48]. As a consequence, he
suggested to manipulate the enteric flora in a beneficial way, i.e. by replacing the harmful
microbes by useful microbes, so as to achieve health benefits in the host.
A huge number of reviews have summarized the health benefits in the ingestion of probiotics
along with their potential and established modes of action in maintaining intestinal and
urogenital health [1, 49-53]:
1. Aid in lactose malabsorption due to bacterial lactase activity
2. Protection against gastro-intestinal infections such as traveller’s diarrhoea, infantile
diarrhea, antibiotic-induced diarrhoea, inflammatory bowel diseases, Helicobacter pylori
associated infections by
• Competition for nutrients and adhesion sites on intestinal mucosa, which facilitate
preferential colonization
• Secretion of antimicrobial substances such as organic acids, hydrogen peroxide,
bacteriocins, antibiotics and deconjugated bile acids, which prevent the outgrowth of
pathogens
• Formation of short-chain fatty-acids (e.g. butyric acid, propionic acid), which reduce
gut pH and simultaneously serve as a substrate for colonic mucosa through
• Attenuation of virulence
• Blocking of toxin receptor sites
• Stimulation of mucosal and systemic host immunity, either general (increased levels
of cytokines, IgA, γ-interferon, increased phagocytic activity) or specific response
(specific antibodies to certain pathogens)
• Suppression of toxin production
3. Suppression of cancer by
• Mutagen binding
• Deactivation of carcinogen and/or procarcinogen
• Inhibition of carcinogen-producing enzymes of colonic microbes
• Inhibition of tumour formation and proliferation
4. Reduction of the risk of coronary heart disease as a consequence of
• Interference with cholesterol absorption from the gut
• Direct assimilation of cholesterol
• Production of metabolites that affect the systemic levels of blood lipids
5. Prevention of urogenital infections by
• Adhesion to urinary and vaginal tract cells
General introduction 5
• Colonization resistance
• Production of inhibitors (H2O2, biosurfactant)
6. Alleviation of constipation
7. Improvement of the nutritional value of foods
As documented by the aforementioned overview, the types of health-associated effects upon
probiotic consumption are relatively broad. Furthermore, apart from the strain dependency of
probiotic effect, the efficacy of probiotic consumption for prophylactic or therapeutic use can
be different, if other characteristics of the consumer (e.g. age, ethnic group, diet behaviour) –
which determine the composition of gut microflora – are also taken into account.
Thus, in order to facilitate harmonization the scientific evidences on the safety, efficacy and
effectiveness of probiotic consumption as well as to further consolidate the acceptance of
probiotics, it is important to generate guidelines and recommend criteria and methodology for
the evaluation of probiotics, and to identify and define what data need to be available to
accurately substantiate health claims. A working group was convened by FAO/WHO to
identify and outline the minimum requirements needed for probiotic status, as schematically
shown in Figure 1 [4].
Strain identification by phenotypic and genotypic method
• Genus, species, strain
• Deposit strain in international culture collection
Functional characterization
•In vitro tests
• Animal studies
Safety assessment
•In vitro and/or animal
• Human study
Efficacy test with double blind, randomized,
placebo-controlled (DBPC) human study
Labeling
• Contents – genus, species, strain designation
• Minimum numbers of viable bacteria at end of shelf-life
• Proper storage conditions
• Corporate contact details for consumer information
Effectiveness trial to compare
probiotics with standard treatment of
a specific condition (human study)
Probiotic Food
Confirmation by
second independent
DBPC study
PHASE 1
SAFETY
PHASE 2
EFFICACY
PHASE 3
EFFECTIVENESS
PHASE 4
SURVEILLANCE
PHASE 1
SAFETY
PHASE 2
EFFICACY
PHASE 3
EFFECTIVENESS
PHASE 4
SURVEILLANCE
Figure 1
Proposed guidelines for the evaluation of probiotics for food use
General introduction 6
The efficacy of therapeutic or prophylactic application of probiotic is highly dependent on
some functional characteristics, which in turn serve as selection criteria. Some in vitro tests
such as resistance to gastric acidity, bile acid resistance, adherence to mucus and/or human
epithelial cells and cell lines, antimicrobial activity against potentially pathogenic bacteria,
reduction of pathogen adhesion to surfaces, bile salt hydrolase activity, resistance to
spermicides (applicable to probiotics for vaginal use) were broadly applied [4].
1.3 Technological approaches for improvement of viability retention
According to the proposed general concept on the efficacy of probiotic application probiotic
function is only obtained with living cultures [4, 54, 55]. The probiotic bacteria must therefore
be viable at the time of consumption and maintain their viability throughout the
gastrointestinal tract. Recommendations for the minimum suggested level for probiotics in
the food to attain this viability are quite variable [56]. In general, a level of 106 CFU g-1 at the
time of consumption is required [3, 57], although in some cases a minimal level of 105 cfu per
gram till the end of best before used period was considered as sufficient [21, 58, 59]. Official
standards requiring a minimum of 106-107 CFU g-1 have been introduced by several food
organizations worldwide. In Japan, a guideline has been developed by the Fermented Milks
and Lactic Acid Bacteria Beverages Association which requires a minimum of 107 viable
bifidobacteria cells mL-1 to be present in dairy products [60]. The Federation Internationale de
Laiterie/International Dairy Federation (FIL/IDF) requires 107 CFU of L. acidophilus in
products such as acidophilus milk and 106 CFU g-1 of bifidobacteria in fermented milks
containing bifidobacteria at the time of sale [61]. Likewise the Swiss Food Regulation as well
as the MERCOSOR regulations requires a minimum of 106 CFU of viable bifidobacteria in
similar products [62].
Rather than to achieve a specific health effect in humans, all these standards were primarily
adopted to provide bacterial concentrations that were technologically attainable and cost-
effective [63]. More critical than the concentration of the probiotic bacteria in the food,
however, is the minimal daily intake of the probiotic bacteria necessary to attain a therapeutic
effect. A daily dose of at least 108 cells was proposed to elicit the health promoting effect on
consumers health [51]. These high numbers – achievable by consuming 100 g or mL of food
products containing a minimal level of 106 CFU g-1 or mL-1 – have been suggested to
appropriately compensate for the possible loss in the numbers of probiotic organisms during
passage through the stomach and intestine.
The criteria regarding critical concentration of probiotics in food must be considered in the
production of probiotic containing food products throughout the whole processing line, in
order to achieve and maintain high level of viability under retention of the health-related
General introduction 7
functionality. Additional technological issue addressed to probiotic bacteria is that they are
not creating unpleasant flavours or textures upon when incorporated into food products [55].
Genetic modification
Microorganism
Genetic modification
Microorganism
Development of suitable media for maximal yield
and improved technological properties
Modification of fermentation conditions (pH, T, t,...)
Cultivation
Development of suitable media for maximal yield
and improved technological properties
Modification of fermentation conditions (pH, T, t,...)
Cultivation
Addition of protective compounds
Encapsulation/Microstructurization
Adaptive physical/chemical treatment
Optimization of drying/freezing conditions
Strategies in incorporating probiotics in food ecosystem
Storage condition – T, aw, atmosphere composition, packaging
Processing
Addition of protective compounds
Encapsulation/Microstructurization
Adaptive physical/chemical treatment
Optimization of drying/freezing conditions
Strategies in incorporating probiotics in food ecosystem
Storage condition – T, aw, atmosphere composition, packaging
Processing
Improvement of
viability and stability
Figure 2
Relevant target sites where improvements on the stability of probiotic bacteria can be achieved
Figure 2 outlines some feasible approaches addressed to the improvement of the viability
and stability of probiotic cells. Modifications to improve these crucial technological traits can
either be aimed on genetic level by using genetic engineering or on physiological level in
upstream (pre-harvesting phase) and downstream processing steps (post-harvesting phase).
1.3.1 Genetic engineering of microorganism
Since technological properties are genetically determined [64], it is of interest to make use of
already existing database on bacterial genes being activated under certain stressful
conditions or genes differentially expressed in resistant mutant in order to create food-grade
mutants [65-70]. Modifications can be performed by applying food-grade plasmids,
integration of foreign DNA in the chromosome of target bacteria and application of regulating
system [71, 72]. With help of this strategy it is expected that technologically sensitive but
highly effective probiotic bacteria can potentially be manipulated to become more robust for
survival under harsh conditions, such as food product development and gastrointestinal
transit.
For instance, recombinant sucrose-6-phosphate synthase (SpsA) was synthesized in
Escherichia coli by using the spsA gene of the cyanobacterium [73]. Transformants exhibited
General introduction 8
a 10,000-fold increase in survival compared to wild-type cells following either freeze-drying,
air drying, or desiccation over phosphorus pentoxide. Trials have also been made to produce
mutants which overexpress protective stress metabolites, i.e. heat shock proteins GroEL and
GroES, thereby improving tolerance against abusive heat treatments [74]. Increasing the
available GroES and GroEL concentration prior to the stresses associated with freezing,
lyophilization, or spray-drying may offer additional protection against protein denaturation
and produce a more viable and physiologically active product [66]. Attempts had been made
to convert Saccharomyces cerevisae into a yeast preferring growth under high hydrostatic
pressure (piezophile) by manipulating the genome and by introducing genes that control high
pressure growth in yeast so as to allow unique microbial biotransformation [75]. Another
possible approach in terms of reducing detrimental effect of freezing dealt with the
introduction of ice nucleation gene into yeast cells [76]. The mutants were able to synthesize
ice nucleating proteins, which were able to trigger ice formation at temperatures, where ice
nucleation has not occurred and thus – due to a higher ∆T – potentially increase freezing
rate.
Apart from the context of viability enhancement, genetically modified dairy starter culture had
already been developed for cheese production [77]. With help of the genetic engineering it
also seems possible to improve acidification, flavouring and texturizing properties of yogurt
and cheese starters, which can lead to the removal of chemical additives from the
formulation [71, 78]. For instance, works have been performed to transfer genes coding for
production of exocellular polysaccharide (EPS), which increases the viscosity of yogurt and
decreases susceptibility to syneresis [78]. Furthermore, research is under way to develop
strains of starter culture bacteria that are resistant to bacteriophage infection or to equip
them with beneficial traits, such as antimicrobial production [79]. Ultimately, genetic
manipulation could also be made to improve specific health-related effects of probiotic
bacteria [78].
However, within the current legislative situation, i.e. Novel Food Regulations, which sets a
high standard of safety precautions as well as the current rejective behaviour of consumer on
the application and presence of genetically modified microorganism in food [79], this
approach seems not likely to be applicable in the near future.
1.3.2 Cultivation
Another approach is dealing with the improvement of cultivation techniques. Since the growth
condition is also known to dictate the robustness of industrial microbial strains, efforts are
continuously done by modifying the composition of fermentation media or by altering
fermentation conditions. The addition of Tween 80 or calcium in fermentation media was
reported to positively affect the survival characteristics during freezing [80-82]. Similarly, the
General introduction 9
type of base used to control pH was found to account for good cryotolerance [82]. Moreover,
harvesting time, growth temperature and pH of the fermentation broth are considered as
crucial operating factors during fermentation, which need to be adjusted properly, since they
also determined survival properties during freezing and freeze-drying [83-89]. However,
Lactobacillus acidophilus grown in free-pH fermentation runs (final pH 4.5) tended to be
resistant to low acidity, high ethanol concentration, freezing/thawing cycle, H2O2, and
lyophilization, whereas cells from cultures under controlled pH (pH = 6.0) were very sensitive
[90]. This findings questions the validity of growing cultures near neutrality under controlled
pH to ensure maximum biomass and active cells, since growth under achievement of low
final pH assured a better survival during the industrial processes and gastro-intestinal transit.
1.3.3 Downstream processing
Inclusion of protective compounds
Addition of protective compounds to prevent cell death during drying and freezing is one of
the most feasible approach to confer protection [91-94]. The additives include a variety of
simple or more complex chemical compounds. Cryoprotective chemicals can be divided in
permeating cryoprotectants, e.g. dimethyl sulphoxide (DMSO), glycerol, which can pass
through cell membranes and non-permeating cryoprotectants, e.g. hydroxyethyl starch,
various sugars, which cannot enter cells [95]. Permeating cryoprotectants were found to
reduce harmful concentrations of solute/electrolytes in the cell, stabilize cell proteins,
stabilize plasma membrane by electrostatic interaction and prevent intracellular ice formation
by lowering the intracellular freezing point. High concentration of extra- and intracellular
cryoprotectants is reported to facilitate vitrification due to dramatic reduction of ice nucleation
and crystal growth [95]. Furthermore, in the presence of cryoprotectants, the effect of
extracellular solution changes can be minimized [96]. The effect of different protective agents
differs according to the microorganism used, but the selection made was rather empirical
than based on mechanistic background.
In the vast majority of probiotic products the protective agent is restricted to milk-based
carrier category. Although decent protection without including milk-based additives and/or
ingredients seems to be very difficult to achieve [97-99], technological improvements are
important when aiming at diversified application of probiotic in novel and non-traditional
products. Especially the incorporation of probiotics in non-dairy products stored at room
temperature, such as cereal products and chocolate can create an overwhelming challenge
for their stability [2, 55].
General introduction 10
Encapsulation
Another practicable approach is by means of encapsulating the cells into a protective matrix,
through which their survival in harmful environmental conditions can be enhanced [100]. A lot
of works have evidenced, that encapsulated cells had improved survival properties under in
vitro gastrointestinal condition as compared to control population [24, 32, 101-109]. Similarly,
they protective coatings or matrices can facilitate better survival in food products [32, 43,
110-112]. Trials have already been made to evaluate the contribution of encapsulation to
enhancement of survival during drying and storage [99, 107, 113-116]. In many cases it was
not clearly validated, whether the cells were fully immobilized in an external protective matrix
or they merely adhered to the protective compounds.
Adaptive treatment for stress induction
Furthermore, the properties of the cells themselves can be improved by induction of their
protective mechanisms. Particularly, this approach utilize their stress adaptation and cross-
protection responses, which could enhance the survival of probiotics in stressful conditions
and to improve their technological properties [69, 117, 118]. The importance of inducing
beneficial stress responses of probiotic microorganism have only recently gained increasing
research interest. Studies have been performed in assessing homologous or heterologous
stress responses on probiotic lactobacilli and bifidobacteria to increase tolerance against
acid [119, 120], bile [118, 120], sodium chloride [118, 121], freeze-thawing cycle [65, 68, 122,
123], heat [68, 118, 120, 124-128], spray drying [115, 129, 130], fluidized bed drying with
subsequent storage [131], desiccation [132] and high hydrostatic pressure [133]. The applied
stress inducers for the pre-treatment step are either of physical or chemical nature. This
adaptive response is characterized by physiological changes so that the bacteria become
more tolerant to adverse conditions following exposure to mild, non-lethal stress conditions.
Physiological changes reported in bacteria include the transient induction of general or
specific proteins. With respect to heat shock response a set of conserved heat shock
proteins are generally overexpressed [134]. Classical heat shock proteins are the molecular
chaperones (e.g. DnaK, GroEL, etc.) or ATP-dependent proteases (ClpP). These proteins
play roles in protein folding, assembly, and repair and prevention of aggregation under stress
and nonstress conditions [117, 135]. Moreover, the accumulation of compatible solute, such
as betaine and trehalose during osmotic stress, was accounted for enhancement of viability
upon drying [132, 136, 137]. Compatible soIutes can counterbalance the external osmotic
conditions without adversely affecting the structure of proteins and other macromolecules
within cells [138]. Bacteria could adapt to environmental stress conditions such as cold
temperature by modifying the fatty acid composition of the cellular membrane, in particular by
increasing the proportion of unsaturated fatty acid residues [81].
General introduction 11
Most of the studies on adaptive response were performed with cultures from exponential
growth phase, since bacteria that enter into stationary phase had already developed
resistance against various types of environmental stress. However, data on stress response
studies with culture from stationary growth phase revealed the potential of pretreatment of
cells in this particular growth stage to improve survival during subsequent treatment [90, 120,
123, 131].
Optimization of drying or freezing conditions
Optimization of drying or freezing conditions is a multi-disciplinary task, where innovative
engineering solutions can be expected since this step not only determines survival behaviour
considerably, but potentially also the whole processing line, energy consumption and the
resulted product characteristics.
Freezing is performed either to produce frozen bacterial culture as an end product or to
prepare intermediate product for subsequent freeze-drying process. Moreover, there are also
some end products harbouring frozen microorganism such as frozen dough containing
baker’s yeast with conserved desirable metabolical features [139].
However, freezing exerts different injury effects on lactic acid bacteria. In particular, freezing
rate seems to be crucial in achieving high level of survival [82]; however the effect of freezing
rate was different for the different microbes [89]. It was proposed, that freezing rate
principally influences the size of ice crystals and the site, in which ice nucleation and crystal
growth occur, which ultimately determine the type of cellular damage experienced by frozen
cells, or whether vitrification takes place or not [95, 140]. Rapid freezing (achieved by
freezing at –196°C) was reported to have a better effect on microbial survivability and
storage stability [141, 142] although other studies suggested that high cooling rate did not
improve viability retention [143] or might even have detrimental effects on cells [123].
A general explanation of the effect of freezing rate on biological cells was proposed by Mazur
et al, who introduced the two-factor hypothesis of freezing damage, according to which there
are two independent mechanisms of damage during freezing, one active at low freezing
rates, the other at high freezing rates [140]. Although the optimal cooling rate can vary by
orders of magnitude for different cell types, the qualitative behaviour appears to be universal.
In both cases, mechanical damage, linked to the interaction between cells and ice crystals,
highly impact the integrity of the cell structure [144].
At low freezing rates, cellular damages were related to the exposure of highly concentrated
intra- and extracellular solution [95]. The removal of water (as ice crystallizes) in the
surrounding medium results in increased extracellular solute concentration [96]. On the other
hand, osmotic-driven migration of water from the cell increases intracellular solute
concentration. Although this dehydration can reduce the probability of intracellular ice
General introduction 12
formation, the solute concentration might reach a detrimental level. In contrast, at high
freezing rates, cell injury is attributed to the mechanical forces as a consequence of the
formation of intracellular ice. Membrane rupture due to osmotic fluxes might also contribute
to cell damage [145]. However, there were some cases reported, in which the intracellular ice
per se did not cause cell death but even improve survival [146, 147]. In both cases,
mechanical damage, linked to the interaction between cells and ice crystals, may affect the
integrity of the cell structure [144].
By using very high cooling rates at low temperatures molecular motion is arrested prior to
crystal formation [96]. In particular, the increased viscosity of the solution reduces the
molecular diffusion and the rate of ice nucleation as well as crystal growth, so that ice
formation decreases and all phase changes can be inhibited [95]. The unfrozen solution
remains in a metastable state, with an amorphous, non crystalline structure.
Despite of the general notion of an optimal rate for freezing of biological cells [140], operating
at higher freezing rates seems to be more attractive when economical value of the freezing
process such as reduced processing time is taken into account. Moreover, the possibility to
reach the vitrified state, where the dangers of intracellular ice formation and injury by
concentrated solutes can be avoided, can only be realized by rapid freezing.
An increase of the freezing rate can also be achieved by increasing surface area per unit
weight; thus accelerating the removal of crystallization heat. This process improvement could
be realized by generating droplets of bacterial suspension along with protective agent, which
are then immediately immersed in liquid nitrogen at –196°C. Another possibility to have a
sufficiently high freezing rate is by producing a controlled spray of high surface-to-mass ratio
droplets (size of approximately 5 to 30 µm) in air blast freezer.
Drying is also widely used as a means of preservation of bacterial cells, although the process
itself and subsequent storage are known to be lethal to a large fraction of the dried
population. Compared to frozen concentrated bacterial preparation, the use of dried bacterial
population do not require cryogenic shipment and storage. In addition, owing to removal of
water the weight of the product can be markedly reduced. However, it was reported that the
time lag before acidification begins is longer for the dried than for the frozen cultures [91].
Viability loss during drying was related to damage to the cell wall and cytoplasmic
membrane, so that the dried cells became more sensitive to NaCl [13, 14, 148]. Damage on
cell membrane could be detected by increased permeability of ß-galactosidase substrate,
higher diffusion rate of DNase into cells and by leakage of UV-absorbing materials from the
cells [148-150]. Following drying changes were reported to take place in the
unsaturated:saturated fatty acid ratio, a loss of ∆pH and a decrease in the activity of
membrane bound H+-ATPase [151]. Damage of membrane bound H+-ATPase, which is
responsible for pH homeostasis in acidic environment by discharging H+ from the cell reduce
General introduction 13
the ability to tolerate acidic conditions [152]. Protective compounds, primarily saccharides
protect membrane and proteins from dehydration damage, most likely by hydrogen bonding
to polar residues in the dry macromolecules, as described by the water replacement
hypothesis [153, 154]. Protective effects of saccharides are also related with the ability of
sugar to form a high viscous glassy matrix during dehydration [155].
Bacteria can be dried either with freeze drying, vacuum drying, spray drying, or fluidized bed
drying [91, 129, 156-158]. Most of dried bacterial preparations are currently produced with
freeze-drying due to the possibility to operate at mild conditions, so that the degree of injury
could be minimized. However, some drawbacks of freeze drying process, such as long
processing time and high energy consumption, led to efforts in evaluating alternative drying
processes. Spray drying is one of the promising process for production of dry probiotic
preparations [13, 14, 115, 130, 159], since under optimized conditions it allows high
processing rates and low operating costs under maintaining a high degree of survivability.
Storage condition
Furthermore, the storage conditions (packaging material, temperature, humidity etc.) can
greatly affect the stability of the probiotic product. In general frozen and dried product should
preferably be stored at low temperature to ensure viability retention.
Storage temperatures lower than –80°C are found to be sufficient in maintaining high level of
viability [82, 95, 143] and shelf-life can be dramatically increased as the storage temperature
is reduced. At –196°C it was reported that there is insufficient thermal energy for
deteriorative chemical reaction [160].
When probiotic dried preparation is subjected to prolonged storage, low relative humidity (11
– 22%) was found to enhance bacterial stability [161]. Storage at zero humidity seems to be
disadvantageous due to the apparent increased rate of lipid oxidation [162], particularly on
bacterial membranes. Membrane deterioration as a result of exposure to oxygen was based
on the oxidation of unsaturated fatty acid in cellular membrane [163]. Oxidation process is
reported to be activated by an increase in the residual humidity [161] and could be effectively
counteracted by storage in the absence of oxygen [164, 165]. Enhancement of shelf life of
dried probiotic bacteria could be also achieved when storage temperature was lowered [166,
167]. Similar to chemical reactions in general, the effect of storage temperature on the rate of
viability loss could be described by Arrhenius equation.
The packaging material can be considered as a critical factor during storage. Compared to
storage polyester (PET) bottles, skim milk-encapsulated bifidobacteria stored in glass bottles
showed a relatively low viability reduction during prolonged storage at 4°C, regardless of the
presence of oxygen scavenger and desiccant [99]. This difference was attributed to the
relatively high oxygen permeability of PET bottles [60].
General introduction 14
Strategies to incorporate probiotics in cultured food ecosystem
Futhermore, when cultured dairy products are considered as food vehicle for probiotic
consumption, there are numerous challenges related to the instability of some strains of
probiotic bacteria in fermented milk products. Careful selection of yoghurt starter is
prerequisite in order to maintain high viability level of probiotic, since yoghurt culture is
capable of creating environments that inhibit not only undesirable microbial contaminants but
also the co-existing probiotic bacteria. This inhibitory activity is attributed to several factors,
including production of lactic and other organic acids, hydrogen peroxide, and bacteriocins,
as well as reduced availability of nutrients [168-170]. To enhance the growth and high
viability level of probiotics in yoghurt some practical solutions such as the use of higher
inocula of probiotics, the addition of growth promoting factors such as amino acids, peptides
and other micronutrients as well as the addition of ascorbic acid or cysteine, which decrease
redox potential, were proposed [171-173]. Apart from addition of growth-promoting
substrates the conditions in the manufacture and storage of yoghurt could be manipulated in
order to increase probiotic survival [174]. The methods applied encompassed termination of
fermentation at pH>5.0, reduction of storage temperature to less than 3-4°C, addition of pH
buffering agents such as whey protein concentrates, heat shock for prevention of excessive
acid production, reduction of incubation temperature in favour of bifidobacterial growth,
mechanical rupture of yoghurt bacteria and attenuation of yoghurt bacteria by high pressure
treatment.
1.4 The EU-Project PROTECH - QLK1-CT-2000-30042
The present PhD thesis is performed in TU Berlin, Department of Food Biotechnology and
Food Process Engineering, within the frame of an EU funded project PROTECH (Nutritional
enhancement of probiotics and prebiotics: Technology aspects on microbial viability, stability,
functionality and on prebiotic function, QLK1-CT-2000-30042), in which the impact of
processing technologies on probiotic and/or prebiotic based functional food is evaluated.
Figure 3 illustrated the objects of investigation and how the project partners are interlinked in
the multitude of tasks [175]. In brief, development of a fermentation medium as well as
optimization of harvesting time with respect to post-harvesting stability were pursued. In the
drying study freeze-drying and spray-drying were evaluated, in terms of identification of
suitable processing regimes, performance of different protectants in offering high survival
during drying and storage, factors governing storage stability during storage, etc. The use of
sub-lethal stress to improve technological behaviours (i.e. heat and oxygen tolerance) as well
as to enhance resistance against extreme conditions in gastro intestinal tract (bile tolerance)
was evaluated. Proteomic approach was applied in order to examine the specific response of
the cells. The survival of probiotics in yoghurt was assessed to determine critical factors
General introduction 15
governing survival in well-established food system and how modifications on this system
could be made to reduce cell death, especially by incorporating prebiotics. The possibility of
enzymatically modifying prebiotics into a more complex structure, thus making them less
fermentable, was investigated. Feeding trials of commercial prebiotics supplemented with
probiotics were performed in order to assess whether these properties could induce
beneficial changes on the composition of short chain fatty acid in different parts of rat colon.
DRYING AND
STORAGE
STRESS
INCORPORATION IN
FOOD PRODUCT
RAT FEEDING
TRIALS
VTT
pH, Bile, O2: Nestlé
Pressure : TU Berlin
Heat : UCC, TU Berlin, Teagasc
Freeze-drying : VTT, Valio, TU Berlin
DANISCO, ORAFTI
Wageningen
University,
TU Berlin
Lund University
Stability test in food :
Nestlé, ORAFTI, VTT, Valio
SELECTION OF
PREBIOTICS
PREBIOTIC
MODIFICATION
FERMENTATION
Spray-drying : TU Berlin, UCC, Teagasc, VTT
Figure 3
General and specific objects of investigation, which were accomplished by partners participating in
PROTECH project along with the network of interaction among partners.
1.5 Structure of PhD thesis
The work is focused on the exploration of spray drying as an alternative processing method
to produce dried probiotic preparation. In this context investigations were performed on
studying the mechanism leading to cell damage during drying and the role of physical state
of the drying matrix and the interaction of protective compounds with cellular membrane in
dehydration tolerance. Moreover, a pre-adaptation step under sub lethal high pressure level
was assessed on its potential in increasing heat tolerance. The linkage between the separate
works were shown in Figure 1.4.
In the first chapter flow cytometric analysis was applied to evaluate the impact of physical
and chemical treatments on viability state of probiotic microorganism. Particularly, the effect
of heat treatment and high hydrostatic pressure treatment as well as the combined effect of
both treatments was assessed on Lactobacillus rhamnosus GG (ATCC 53103). The impact
of low pH was studied in order to draw conclusion about survival mechanism of probiotic. A
multiple staining strategy using fluorescent dyes carboxyfluoresceindiacetate (cFDA) and
propidium iodide (PI) was applied.
General introduction 16
Spray drying
Pressure-induced
thermotolerance
Flow cytometry
Direct interaction
sugar- membrane
Glassy state
Figure 1.4
Structure of the works performed within this PhD thesis
In the chapter dealing with spray drying, the works comprised of the selection of optimal
processing condition, in which a compromise has to be taken in order to achieve powder with
an appreciable level of viable probiotic bacteria and sufficiently low residual moisture to
enable stable storage. The nature of drying induced cellular injury was also a subject of
investigation and therefore evaluated with flow cytometric analysis previously established.
Differential Scanning Calorimetry (DSC) was applied to investigate the role of physical state,
in particular the presence of drying medium in glassy state, on survival behaviour during
storage. The work with liposome as a simple model for bacterial membrane was done due to
the fact that storage stability was found to be governed not only by physical state of the
extracellular matrix, but also by direct interaction between saccharides, which constituted the
protective compound, with cellular membrane.
The work on the application of pressure pre-treatment to induce beneficial stress response
was done to evaluate whether and to which extent heat resistance of L. rhamnosus is
affected by mild pressure treatments prior to exposure to lethal temperature. Apart from
identifying the optimal combinations of pressure, temperature and treatment time for the pre-
treatment steps, it was attempted to elucidate the possible cellular mechanisms involved in
the acquisition of heat tolerance with help of flow cytometry technique.
1.6 References
1. Fooks, L.J., Fuller, R., and Gibson, G.R. 1999. Prebiotics, probiotics and human gut microbiology.
International Dairy Journal. 9: 53-61.
2. Mattila-Sandholm, T., Myllärinen, P., Crittenden, R., Mogensen, G., Fonden, R., and Saarela, M.
2002. Technological challenge for future probiotic foods. International Dairy Journal. 12: 173-182.
3. Kurman, J.A., Rasic, R.L. 1991. The health potential of products containing bifidobacteria, in
Therapeutic properties of fermented milk, Robinson, R.K., Editor. Elsevier Applied Food Science Series:
London. p. 117-158.
General introduction 17
4. FAO/WHO. 2002. Guidelines for the evaluation of probiotics in food, in Report of a Joint FAO/WHO
Working Group on Drafting Guidelines for the Evaluation of Probiotics in Food. Food and Agriculture
Organization of the United Nations and World Health Organization: Ontario, Canada.
5. Lourens-Hattingh, A., Viljoen, B.C. 2001. Growth and survival of a probiotic yeast in dairy products.
Food Research International. 34: 791-796.
6. Sanders, M.E., Morelli, L., Tompkins, T.A. 2003. Sporeformers as human probiotics: Bacillus,
Sporolactobacillus, and Brevibacillus. Comprehensive Reviews in Food Science and Food Safety. 2:
101-110.
7. Wessels, S., Axelsson, L., Hansen, E.B., De Vuyst, L., Laulund, S., Lähteenmäki, L., Lindgren, S.,
Mollet, B., Salminen, S., and von Wright, A. 2004. The lactic acid bacteria, the food chain, and their
regulation. Trends in Food Science & Technology. 15: 498-505.
8. Stanton, C., Gardiner, G., Meehan, H., Collins, K., Fitzgerald, G., Lynch, P.B., and Ross, R.P. 2001.
Market potential for probiotics. American Journal of Clinical Nutrition. 73(Suppl.): 476S-483S.
9. Milch-Markt. 2004. Zahlen und Daten der deutschen Milchindustrie. http://www.milch-
markt.de/de/milchaktuell/branchenzahlen_aktuell/milchaktuell_zahlen_daten.html.
10. Shortt, C. 1998. Living it up for dinner. Chemistry and Industry. 8: 300-303.
11. Ouwehand, A.C., Kurvinen, T., and Rissanen, P. 2004. Use of a probiotic Bifidobacterium in a dry food
matrix, an in vivo study. International Journal of Food Microbiology. 95: 103-106.
12. Rybka, S., Kailasapathy, K. 1995. The survival of culture bacteria in fresh and freeze-dried AB yoghurt.
The Australian Journal of Dairy Technology. 50: 51-57.
13. Gardiner, G.E., O'Sullivan, E., Kelly, J., Auty, M.A.E., Fitzgerald, G.F., Collins, J.K., Ross, R.P., and
Stanton, C. 2000. Comparative survival rates of human-derived probiotic Lactobacillus paracasei and L.
salivarius strains during heat treatment and spray drying. Applied and Environmental Microbiology. 66:
2605-2612.
14. Corcoran, B.M., Ross, R.P., Fitzgerald, G., Stanton, C. 2004. Comparative survival of probiotic
lactobacilli spray dried in the presence of prebiotic substances. Journal of Applied Microbiology. 96:
1024–1039.
15. Arihara, K., Ota, H., Itoh, M., Kondo, Y., Sameshima, T., Yamanaka, H., Akimoto, M., Kanai, S., and
Miki, T. 1998. Lactobacillus acidophilus group lactic acid bacteria applied to meat fermentation. Journal
of Food Science. 63: 544-547.
16. Erkkilä, S. and Petaja, E. 2000. Screening of commercial meat starter cultures at low pH and in the
presence of bile salts for potential probiotic use. Meat Science. 55: 297-300.
17. Erkkilä, S., Venalainen, M., Hielm, S., Petäjä, E., Puolanne, E., and Mattila-Sandholm, T. 2000.
Survival of Escherichia coli O157:H7 in dry sausage fermented by probiotic lactic acid bacteria. Journal
of the Science of Food and Agriculture. 80: 2101-2104.
18. Erkkilä, S., Petäjä, E., Eerola, S., Lilleberg, L., Mattila-Sandholm, T., and Suihko, M.-L. 2001.
Flavour properties of dry sausages fermented by selected novel meat starter culture. Meat Science. 58:
111-116.
19. Betoret, N., Puente, L., Díaz, M.J., Pagan, M.J., Garcia, M.J., Gras, M.L., Martínez-Monzó, J., and
Fito, P. 2003. Development of probiotic-enriched dried fruits by vacuum impregnation. Journal of Food
Engineering. 56: 273-277.
20. Hull, R.R., Roberts, A.V., and Mayes, J.J. 1984. Survival of Lactobacillus acidophilus in yoghurt.
Australian Journal of Dairy Technology. 39: 164-166.
21. Shah, N.P., Lankaputhra, W.E.V., Britz, M., and Kyle, W.S.A. 1995. Survival of L. acidophilus and
Bifidobacterium bifidum in commercial yoghurt during refrigerated storage. International Dairy Journal. 5:
515–521.
General introduction 18
22. Nighswonger, B.D., Brashears, M.M., and Gilliland, S.E. 1996. Viability of Lactobacillus acidophilus
and Lactobacillus casei in fermented milk products during refrigerated storage. J. Dairy Sci. 79: 212-219.
23. Dave, R.I. and Shah, N.P. 1997. Viability of yoghurt and probiotic bacteria in yoghurts made from
commercial starter cultures. International Dairy Journal. 7: 31-41.
24. Picot, A. and Lacroix, C. 2004. Encapsulation of bifidobacteria in whey protein-based microcapsules
and survival in simulated gastrointestinal conditions and in yoghurt. International Dairy Journal. 14: 505-
515.
25. Shin, H.S., Lee, J.H., Pestka, J.J., and Ustunol, Z. 2000. Growth and viability of commercial
Bifidobacterium ssp. in skim milk containing oligosaccharides and inulin. Journal of Food Science. 65:
884-887.
26. Hou, J.W., Yu, R.C., and Chou, C.C. 2000. Changes in some components of soymilk during
fermentation with bifidobacteria. Food Research International. 33: 393-397.
27. Chou, C.C. and Hou, J.W. 2000. Growth of bifidobacteria in soymilk and their survival in the fermented
soymilk drink during storage. International Journal of Food Microbiology. 56: 113-121.
28. Kunz, B., Schuth, S., Stefer, B., and Sträter, S. 2001. Produktentwicklung unter Nutzung
multifunktioneller Lebensmitteladditive aufgezeigt am Beispiel eines synbiotischen Magerquarks.
Ernährungs-Umschau. 48: 195-199.
29. Dubey, U.K. and Mistry, V.V. 1996. Growth characteristics of bifidobacteria in infant formulas. Journal
of Dairy Science. 79: 1146-1155.
30. Dubey, U.K. and Mistry, V.V. 1996. Effect of bifidogenic factors on growth characteristics of
bifidobacteria in infant formulas. Journal of Dairy Science. 79: 1156-1163.
31. Yoon, K.Y., Woodams, E.E., and Hang, Y.D. 2004. Probiotication of tomato juice by lactic acid bacteria.
The Journal of Microbiology. 42: 315-318.
32. Khalil, A.H. and Mansour, E.H. 1998. Alginate encapsulated bifidobacteria survival in mayonnaise.
Journal of Food Science. 63: 702-705.
33. Gardiner, G., Ross, R.P., Collins, J.K., Fitzgerald, G., Stanton, C. 1998. Development of a probiotic
cheddar cheese containing human-derived Lactobacillus paracasei strains. Applied and Environmental
Microbiology. 64: 2192-2199.
34. Gomes, A.M.P. and Malcata, F.X. 1998. Development of probiotic cheese manufactured from goat milk:
response surface analysis via technological manipulation. Journal of Dairy Science. 81: 1492-1507.
35. Gomes, A.M.P., Vieira, M.M., and Malcata, F.X. 1998. Survival of probiotic microbial strains in a cheese
matrix during ripening: simulation of rates of salt diffusion and microorganism survival. Journal of Food
Engineering. 36: 281-301.
36. Vinderola, C.G., Prosello, W., Ghiberto, D., and Reinheimer, J.A. 2000. Viability of probiotic
(Bifidobacterium, Lactobacillus acidophilusand Lactobacillus casei) and nonprobiotic microflora in
argentinian fresco cheese. Journal of Dairy Science. 83: 1905–1911.
37. Tharmaraj, N. and Shah, N.P. 2004. Survival of Lactobacillus acidophilus, Lactobacillus paracasei
subsp. paracasei, Lactobacillus rhamnosus, Bifidobacterium animalis and Propionibacterium in cheese-
based dips and the suitability of dips as effective carriers of probiotic bacteria. International Dairy
Journal. 14: 1055-1066.
38. Modler, H.W., McKellar, R.C., Goff, H.D., and Mackie, D.A. 1990. Using ice cream as a mechanism to
incorporate Bifidobacteria and fructooligosaccharides into the human diet. Cultured Dairy Products
Journal. 25: 4-9.
39. Hekmat, S. and McMahon, D.J. 1992. Survival of Lactobacillus acidophilus and Bifidobacterium bifidum
in ice cream for use as a probiotic food. Journal of Dairy Science. 75: 1415-1422.
General introduction 19
40. Christiansen, P.S., Edelsten, D., Kristiansen, J.R., and Nielsen, E.W. 1996. Some properties of ice
cream containing Bifidobacterium bifidum and Lactobacillus acidophilus. Milchwissenschaft. 51: 502-504.
41. Hagen, M. and Narvhus, J.A. 1999. Production of ice cream containing probiotic bacteria.
Milchwissenschaft. 54: 265-268.
42. Haynes, I.N. and Playne, M.J. 2002. Survival of probiotic cultures in low-fat ice-cream. Australian
Journal of Dairy Technology. 57: 10-14.
43. Godward, G. and Kailasapathy, K. 2003. Viability and survival of free, encapsulated and co-
encapsulated probiotic bacteria in ice cream. Milchwissenschaft. 58: 161-164.
44. Holcomb, J.E., Frank, J.F., and McGregor, J.U. 1991. Viability of Lactobacillus acidophilus and
Bifidobacterium bifidum in soft-serve frozen yogurt. Cultured Dairy Products Journal. 26: 4-5.
45. Laroia, S. and Martin, J.H. 1991. Effect of pH on survival of Bifidobacterium bifidum and Lactobacillus
acidophilus in frozen fermented dairy desserts. Cultured Dairy Products Journal. 26: 13-14, 16, 18, 20-
21.
46. Davidson, R.H., Duncan, S.E., Hackney, C.R., Eigel, W.N., and Boling, J.W. 2000. Probiotic culture
survival and implications in fermented frozen yogurt characteristics. Journal of Dairy Science. 83: 666-
673.
47. Informationsdienst-Wissenschaft. 2004. Probiotische Schokolade. http://idw-
online.de/pages/de/news92389.
48. Metchnikoff, E. 1907. Lactic acid as inhibiting intestinal putrefaction, in The prolongation of life:
Optimistic studies, Heinemann, W., Editor: London. p. 161-183.
49. Sanders, M.E. 1999. Probiotics. Food Technology. 53: 67-77.
50. Reid, G. 1999. The scientific basis for probiotic strains of Lactobacillus. Applied and Environmental
Microbiology. 65: 3763-3766.
51. Lourens-Hattingh, A. and Viljoen, B.C. 2001. Yogurt as probiotic carrier food. International Dairy
Journal. 11: 1-17.
52. McNaught, C.E. and MacFie, J. 2001. Probiotics in clinical practice: a critical review of the evidence.
Nutrition Research. 21: 343-353.
53. O'Sullivan, D.J. 2001. Screening of intestinal microflora for effective probiotic bacteria. Journal of
Agriculture and Food Chemistry. 49: 1751-1760.
54. Ouwehand, A.C. and Salminen, S.J. 1998. The health effects of cultured milk products with viable and
non-viable bacteria. International Dairy Journal. 8: 749-758.
55. Saarela, M., Mogensen, G., Fondén, R., Mättö, J., and Mattila-Sandholm, T. 2000. Probiotic bacteria :
safety, functional and technological properties. Journal of Biotechnology. 84: 197-215.
56. Charteris, W.P., Kelly, P.M., Morelli, L., and Collins, J.K. 1998. Ingredient selection criteria for
probiotic microorganisms in functional dairy foods. International Journal of Dairy Technology. 51: 123–
136.
57. Boylston, T.D., Vinderola, C.G., Ghoddusi, H.B., and Reinheimer, J.A. 2004. Incorporation of
bifidobacteria into cheeses: challenges and rewards. International Dairy Journal. 14: 375–387.
58. Shah, N.P. and Lankaputhra, W.E.V. 1997. Improving viability of Lactobacillus acidophilus and
Bifidobacterium spp. in yogurt. International Dairy Journal. 7: 349-356.
59. Schillinger, U. 1999. Isolation and identification of lactobacilli from novel-type probiotic and mild
yoghurts and their stability during refrigerated storage. International Journal of Food Microbiology. 47:
79-87.
60. Ishibashi, N. and Shimamura, S. 1993. Bifidobacteria: research and development in Japan. Food
Technology. 46: 126-135.
61. IDF. 1992. General standard of identity for fermented milks. International Dairy Federation. 163.
General introduction 20
62. Talwalkar, A. and Kailasapathy, K. 2004. A review of oxygen toxicity in probiotic yogurts: Influence on
the survival of probiotic bacteria and protective techniques. Comprehensive Reviews in Food Science
and Food Safety. 3: 117-124.
63. Roy, D. 2001. Media for the isolation and enumeration of bifidobacteria in dairy products. International
Journal of Food Microbiology. 69: 167-182.
64. Carvalho, A.S., Silva, J., Ho, P., Teixeira, P., Malcata, F.X., and Gibbs, P. 2004. Relevant factors for
the preparation of freeze-dried lactic acid bacteria. International Dairy Journal. 14: 835-847.
65. Kim, W.S. and Dunn, N.W. 1997. Identification of a cold shock gene in lactic acid bacteria and the effect
of cold shock on cryotolerance. Current Microbiology. 35: 59-63.
66. Walker, D.C., Girgis, H.S., and Klaenhammer, T.R. 1999. The groESL chaperone operon of
Lactobacillus johnsonii. Applied and Environmental Microbiology: 3033-3041.
67. Tanghe, A., Teunissen, A., Van Dijck, P., and Thevelein, J.M. 2000. Identification of genes
responsible for improved cryoresistance in fermenting yeast cells. International Journal of Food
Microbiology. 55: 259-262.
68. Schmidt, G. and Zink, R. 2000. Basic features of stress response in three species of bifidobacteria: B.
longum, B. adolescentis, and B. breve. International Journal of Food Microbiology. 55: 41-45.
69. van de Guchte, M., Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S.D., and Maguin, E. 2002.
Stress responses in lactic acid bacteria. Antonie van Leeuwenhoek. 82: 187-216.
70. De Angelis, M., Di Cagno, R., Huet, C., Crecchio, C., Fox, P.F., and Gobbett, M. 2004. Heat shock
response in Lactobacillus plantarum. Applied and Environmental Microbiology. 70: 1336-1346.
71. Barach, J.T. 1985. What's new in genetic engineering of dairy starter cultures and dairy enzymes? Food
Technology. 39: 73-84.
72. Heller, K.J. 1998. Innovative Milchprodukte durch gentechnisch veränderte Mikroorganismen. dmz -
Deutsche Molkerei Zeitung. 119: 1074-1080.
73. Billi, D., Wright, D.J., Helm, R.F., Prickett, T., Potts, M., and Crowe, J.H. 2000. Engineering
desiccation tolerance in Escherichia coli. Applied and Environmental Microbiology. 66: 1680-1684.
74. Desmond, C., Fitzgerald, G.F., Stanton, C., and Ross, R.P. 2004. Improved stress tolerance of
GroESL-overproducing Lactococcus lactis and probiotic Lactobacillus paracasei NFBC 338. Applied and
Environmental Microbiology. 10: 5929–5936.
75. Abe, F. and Horikoshi, K. 2000. Tryptophan permease gene TAT2 confers high-pressure growth in
Saccharomyces cerevisiae. Molecular and Cellular Biology. 20: 8093-8102.
76. Hwang, W.-Z., Coetzer, C., Tumer, N.E., and Lee, T.-C. 2001. Expression of a bacterial ice nucleation
gene in a yeast Saccharomyces cerevisae and its possible application in food freezing processes.
Journal of Agriculture and Food Chemistry. 49: 4662-4666.
77. Bockelmann, W. and Heller, K.J. Cheese ripening with genetically engineered lactic acid bacteria. in 3rd
Karlsruhe Nutrition Symposium. 1998. Karlsruhe.
78. Mollet, B. 1999. Genetically improved starter strains: opportunities for the dairy industry. International
dairy journal. 9: 11-15.
79. Rastall, R.A. and Maitin, V. 2002. Genetic engineering: threat or opportunity for the dairy industry.
International Journal of Dairy Technology. 55: 161-165.
80. Wright, C.T. and Klaenhammer, T.R. 1983. Survival of Lactobacillus bulgaricus during freezing and
freeze-drying after growth in the presence of calcium. Journal of Food Science. 48: 773-777.
81. Beal, C., Fonseca, F., and Corrieu, G. 2001. Resistance to freezing and storage of Streptococcus
thermophilus is related to membrane fatty acid composition. Journal of Dairy Science. 84: 2347-2356.
82. Fonseca, F., Beal, C., and Corrieu, G. 2001. Operating conditions that affect the resistance of lactic
acid bacteria to freezing and frozen storage. Cryobiology. 43: 189–198.
General introduction 21
83. Beal, C., Louvet, P., and Corrieu, G. 1989. Influence of controlled pH and temperature on the growth
and acidification of pure cultures of Streptococcus thermophilus 404 and Lactobacillus bulgaricus 398.
Applied Microbiology and Biotechnology. 32: 148-154.
84. Gilliland, S.E. and Rich, C.N. 1989. Stability during frozen and subsequent refrigerated storage of
Lactobacillus acidophilus grown at different pH. Journal of Dairy Science. 73: 1187-1192.
85. Brashears, M.M. and Gilliland, S.E. 1995. Survival during frozen and subsequent refrigerated storage
of Lactobacillus acidophilus cells as influenced by the growth phase. Journal of Dairy Science. 78: 2326-
2335.
86. Reilly, S.S. and Gilliland, S.E. 1999. Bifidobacterium longum survival during frozen and refrigerated
storage as related to pH during growth. Journal of Food Science. 64: 714-718.
87. Murga, M.L.F., Bernik, D., de Valdez, G.F., and Disalvo, A.E. 1999. Permeability and stability
properties of membranes formed by lipids extracted from Lactobacillus acidophilus grown at different
temperatures. Archives of Biochemistry and Biophysics. 364: 115-121.
88. Palmfeldt, J. and Hahn-Hägerdal, B. 2000. Influence of culture pH on survival of Lactobacillus reuteri
subjected to freeze-drying. International Journal of Food Microbiology. 55: 235-238.
89. Peter, G. and Reichart, O. 2001. The effect of growth phase, cryoprotectants, and freezing rates on the
survival of selected micro-organisms during freezing and thawing. Acta Alimentaria. 30: 89-97.
90. Lorca, G.L. and G.F., d.V. 2001. A low-pH-inducible, stationary-phase acid tolerance response in
Lactobacillus acidophilus CRL 639. Current Microbiology. 42: 21-25.
91. Champagne, C.P., Gardner, N., Brochu, E., and Beaulieu, Y. 1991. The freeze drying of lactic acid
bacteria. A review. Canadian Institute for Science and Technology Journal. 24: 118-125.
92. Hubalek, Z. 2003. Protectants used in the cryopreservation of microorganisms. Cryobiology. 46: 205-
229.
93. de Valdez, G.F., de Giori, G.S., de Ruiz Holgado, A.P., and Oliver, G. 1983. Comparative study of the
efficiency of some additives in protecting lactic acid bacteria against freeze-drying. Cryobiology. 20: 560-
566.
94. de Valdez, G.F., de Giori, G.S., de Ruiz Holgado, A.P., and Oliver, G. 1985. Effect of drying medium
on residual moisture content and viability of freeze-dried lactic acid bacteria. Applied and Environmental
Microbiology. 49: 413-415.
95. Karlsson, J.O.M. and Toner, M. 1996. Long-term storage of tissues by cryopreservation: critical issues.
Biomaterials. 17: 243-256.
96. Darvall, J.G.L. 2000. Preservation of microorganisms. Culture. 21: 1-5.
97. Champagne, C.P., Mondou, F., Raymond, Y., and Roy, D. 1996. Effect of polymers and storage
temperature on the stability of freeze-dried lactic acid bacteria. Food Research International. 29: 555-
562.
98. Lian, W.-C., Hsiao, H.-C., and Chou, C.-C. 2002. Survival of bifidobacteria after spray-drying.
International Journal of Food Microbiology. 74: 79-86.
99. Hsiao, H.C., Lian, W.C., and Chou, C.C. 2004. Effect of packaging condition and temperature on
viability of microencapsulated bifidobacteria during storage. Journal of the Science of Food and
Agriculture. 84: 134-139.
100. Krasaekoopt, W., Bhandari, B., and Deeth, H. 2003. Evaluation of encapsulation techniques of
probiotics for yoghurt. International Dairy Journal. 13: 3-13.
101. Rao, A.V., Shiwnarain, N., and Maharaj, I. 1989. Survival of microencapsulated Bifidobacterium
pseudolongumin simulated gastric and intestinal juices. Canadian Institute for Science and Technology
Journal. 22: 345-349.
General introduction 22
102. Sultana, K., Godward, G., Reynolds, N., Arumugaswamy, R., Peiris, P., and Kailasapathy, K. 2000.
Encapsulation of probiotic bacteria with alginate-starch and evaluation of survival in simulated
gastrointestinal conditions and in yoghurt. International Journal of Food Microbiology. 62: 47-55.
103. Lee, K.-Y. and Heo, T.-R. 2000. Survival of Bifidobacterium longum immobilized in calcium alginate
beads in simulated gastric juices and bile salt solution. Applied and Environmental Microbiology. 66: 869-
873.
104. Crittenden, R., Laitila, A., Forssell, P., Matto, J., Saarela, M., Mattila-Sandholm, T., and Myllarinen,
P. 2001. Adhesion of bifidobacteria to granular starch and its implications in probiotic technologies.
Applied and Environmental Microbiology. 67: 3469-3475.
105. Sun, W. and Griffiths, M.W. 2000. Survival of bifidobacteria in yogurt and simulated gastric juice
following immobilization in gellan-xanthan beads. International Journal of Food Microbiology. 61: 17-25.
106. Hansen, L.T., Allan-Wojtas, P.M., Jin, Y.-L., and Paulson, A.T. 2002. Survival of Ca-alginate
microencapsulated Bifidobacterium spp. in milk and simulated gastrointestinal conditions. Food
Microbiology. 19: 35-45.
107. Lian, W.-C., Hsiao, H.-C., and Chou, C.-C. 2003. Viability of microencapsulated bacteria in simulated
gastric juice and bile solution. International Journal of Food Microbiology. 86: 293-301.
108. Krasaekoopt, W., Bhandari, B., and Deeth, H. 2004. The influence of coating materials on some
properties of alginate beads and survivability of microencapsulated probiotic bacteria. International Dairy
Journal. 14: 737-743.
109. Chandramouli, V., Kailasapathy, K., Peiris, P., and Jones, M. 2004. An improved method of
microencapsulation and its evaluation to protect Lactobacillus spp. in simulated gastric condition. Journal
of Microbiological Methods. 56: 27-35.
110. Sheu, T.Y., Marshall, R.T., and Heymann, H. 1993. Improving survival of culture bacteria in frozen
desserts by microentrapment. Journal of Dairy Science. 76: 1902-1907.
111. Shah, N.P. and Ravula, R.R. 2000. Microencapsulation of probiotic bacteria and their survival in frozen
fermented dairy desserts. Australian Journal of Dairy Technology. 55: 139-144.
112. Adhikari, K., Mustapha, A., and Grün, I.U. 2003. Survival and metabolic activity of microencapsulated
Bifidobacterium longum in stirred yogurt. Journal of Food Science. 68: 275-280.
113. Champagne, C.P., Morin, N., Couture, R., Gagnon, C., Jelen, P., and Lacroix, C. 1992. The potential
of immobilized cell technology to produce freeze-dried, phage-protected cultures of Lactococcus lactis.
Food Research International. 25: 419-427.
114. O'Riordan, K., Andrews, D., Buckle, K., and Conway, P. 2001. Evaluation of microencapsulation of a
Bifidobacterium strain with starch as anapproach to prolonging viability during storage. Journal of Applied
Microbiology. 91: 1059-1066.
115. Desmond, C., Ross, R.P., O´Callaghan, E., Fitzgerald, G., and Stanton, C. 2002. Improved survival of
Lactobacillus paracasei NFCB 338 in spray-dried powders containing gum acacia. Journal of Applied
Microbiology. 93: 1003-1011.
116. Wang, Y.-C., Yu, R.-C., and Chou, C.-C. 2004. Viability of lactic acid bacteria and bifidobacteria in
fermented soymilk after drying, subsequent rehydration and storage. International Journal of Food
Microbiology. 93: 209-217.
117. Abee, T. and Wouters, J.A. 1999. Microbial stress response in minimal processing. International
Journal of Food Microbiology. 50: 65-91.
118. Kim, W.-S., Perl, L., Park, J.-H., Tandianus, J.E., and Dunn, N.W. 2001. Assessment of stress
response of the probiotic Lactobacillus acidophilus. Current Microbiology. 43: 346-350.
119. Lorca, G.L., Raya, R.R., Taranto, M.P., and de Valdez, G.F. 1998. Adaptive acid tolerance response in
Lactobacillus acidophilus. Biotechnology Letters. 20: 239-241.
General introduction 23
120. Saarela, M., Rantala, M., Hallamaa, K., Nohynek, L., Virkajärvi, I., and Mättö, J. 2004. Stationary-
phase acid and heat treatments for improvement of the viability of probiotic lactobacilli and bifidobacteria.
Journal of Applied Microbiology. 96: 1205-1214.
121. Arihara, K. and Itoh, M. 2000. UV-induced Lactobacillus gasseri mutants resisting sodium chloride and
sodium nitrite for meat fermentation. International Journal of Food Microbiology. 56: 227-230.
122. de Urraza, P. and de Antoni, G. 1997. Induced cryotolerance of Lactobacillus delbrueckii subsp.
bulgaricus LBB by preincubation at suboptimal temperatures with fermentable sugar. Cryobiology. 35:
159-164.
123. Bâati, L., Fabre Gea, C., Auriol, D., and Blanc, P.J. 2000. Study of the cryotolerance of Lactobacillus
acidophilus: effect of culture and freezing conditions on the viability and cellular protein levels.
International Journal of Food Microbiology. 59: 241-247.
124. Teixeira, P.M., Castro, H.P., and Kirby, R. 1994. Inducible thermotolerance in Lactobacillus bulgaricus.
Letters in Applied Microbiology. 18: 218-221.
125. Gouesbet, G., Jan, G., and Boyaval, P. 2001. Lactobacillus delbrueckii ssp. bulgaricus
thermotolerance. Lait. 81: 301-309.
126. Gouesbet, G., Jan, G., and Boyaval, P. 2002. Two-dimensional electrophoresis study of Lactobacillus
delbrueckii subsp. bulgaricus thermotolerance. Applied and Environmental Microbiology. 68: 1055-1063.
127. Ananta, E. and Knorr, D. 2003. Pressure-induced thermotolerance of Lactobacillus rhamnosus GG.
Food Research International. 36: 991-997.
128. Ananta, E. and Knorr, D. 2004. Evidence on the role of protein biosynthesis in the induction of heat
tolerance of Lactobacillus rhamnosus GG by pressure pre-treatment. International Journal of Food
Microbiology. 96: 307-313.
129. Teixeira, P.M., Castro, H.P., and Kirby, R. 1995. Spray drying as a method for preparing concentrated
cultures of Lactobacillus bulgaricus. Journal of Applied Bacteriology. 78: 456-462.
130. Desmond, C., Stanton, C., Fitzgerald, G.F., Collins, K., and Ross, R.P. 2001. Environmental
adaptation of probiotic lactobacilli towards improvement of performance during spray drying. International
Dairy Journal. 11: 801-808.
131. Prasad, J., McJarrow, P., and Gopal, P. 2003. Heat and osmotic stress responses of probiotic
Lactobacillus rhamnosus HN001 (DR20) in relation to viability after drying. Applied and Environmental
Microbiology. 69: 917-925.
132. Kets, E.P.W., Teunissen, P.J.M., and de Bont, J.A.M. 1996. Effect of compatible solutes on survival of
lactic acid bacteria subjected to drying. Applied and Environmental Microbiology. 62: 259-261.
133. Scheyhing, C.H., Hörmann, S., Ehrmann, M.A., and Vogel, R.F. 2004. Barotolerance is inducible by
preincubation under hydrostatic pressure, cold-, osmotic- and acid-stress conditions in Lactobacillus
sanfranciscensis DSM 20451. Letters in Applied Microbiology. 39: 284–289.
134. Yura, T. and Nakahigashi, K. 1999. Regulation of the heat-shock response. Current Opinions in
Microbiology. 2: 153-158.
135. Hendrick, J.P. and Hartl, F.-U. 1993. Molecular chaperone functions of heat-shock proteins. Annual
Reviews in Biochemistry. 62: 349-384.
136. Welsh, D.T. and Herbert, R.A. 1999. Osmotically induced intracellular trehalose, but not glycine betaine
accumulation promotes desiccation tolerance in Escherichia coli. FEMS Microbiology Letters. 174: 57-
63.
137. de Castro, A., Bredholt, H., Strøm, A.R., and Tunnacliffe, A. 2000. Anhydrobiotic engineering of
gram-negative bacteria. Applied and Environmental Microbiology. 66: 4142-4144.
138. Ko, R., Smith, L.T., and Smith, G.M. 1994. Glycine betaine confers enhanced osmotolerance and
cryotolerance in Listeria monocytogenes. Journal of Bacteriology. 176: 426-431.
General introduction 24
139. Diniz-Mendes, L., Bernardes, E., de Araujo, P.S., Panek, A.D., and Paschoalin, V.M.F. 1999.
Preservation of frozen yeast cells by trehalose. Biotechnology and Bioengineering. 65: 572-578.
140. Mazur, P., Leibo, S., and Chu, E.H.Y. 1972. A two factor hypothesis of freezing injury. Experimental
Cell Research. 71: 345-355.
141. Tsvetkov, T. and Shishkova, I. 1982. Studies on the effects of low temperatures on lactic acid bacteria.
Cryobiology. 19: 211-214.
142. Ryhänen, E.-L. 1991. Über den Einfluss der Gefriergeschwindigkeit auf Lebensfähigkeit und
Stoffwechselaktivität gefrorener und gefriergetrockneter Lactobacillus acidophilus Kulturen. Finnish
Journal of Dairy Science. 49: 14-36.
143. Foschino, R., Fiori, E., and Galli, A. 1996. Survival and residual activity of Lactobacillus acidophilus
frozen cultures under different conditions. Journal of Dairy Research. 63: 295-303.
144. McGann, L.E. 1978. Differing actions of penetrating and non-penetrating cryoprotective agents.
Cryobiology. 15: 382-290.
145. Muldrew, K. and McGann, L.E. 1990. Mechanism of intracellular ice formation. Biophysical Journal. 57:
525-532.
146. Mazur, P. 1977. The role of intracellular freezing in the death of cells cooled at supraoptimal rates.
Cryobiology. 14: 251-272.
147. Acker, J.P. and McGann, L.E. 2002. Innocuous intracellular ice improves survival of frozen cells. Cell
Transplantation. 11: 563-571.
148. Brennan, M., Wanismail, B., Johnson, M.C., and Ray, B. 1986. Cellular damage in dried Lactobacillus
acidophilus. Journal of Food Protection. 49: 47-53.
149. Johnson, J.A.C. and Etzel, M.R. 1995. Properties of Lactobacillus helveticus CNRZ-32 attenuated by
spray-drying, freeze-drying, or freezing. Journal of Dairy Science. 78: 761-768.
150. Lievense, L.C., Verbeek, M.A.M., Noomen, A., and van't Riet, K. 1994. Mechanism of dehydration
inactivation of Lactobacillus plantarum. Applied Microbiology and Biotechnology. 41: 90-94.
151. Castro, H.P., Teixeira, P.M., and Kirby, R. 1997. Evidence of membrane damage in Lactobacillus
bulgaricus following freeze drying. Journal of Applied Microbiology. 82: 87-94.
152. Matsumoto, M., Ohishi, H., and Benno, Y. 2004. H+-ATPase activity in Bifidobacterium with special
reference to acid tolerance. International Journal of Food Microbiology. 93: 109-113.
153. Crowe, J.H., Crowe, L.M., and Carpenter, J.F. 1993. Preserving dry biomaterials: The water
replacement hypothesis, Part 1. BioPharm. 6: 28-33.
154. Crowe, J.H., Crowe, L.M., and Carpenter, J.F. 1993. Preserving dry biomaterials: The water
replacement hypothesis, Part 2. BioPharm. 6: 40-43.
155. Crowe, J.H., Hoekstra, F.A. , Nguyen, K.H.N., Crowe, L.M. 1996. Is vitrification involved in depression
of the phase transition temperature in dry phospholipids? Biochimica et Biophysica Acta. 1280: 187-196.
156. King, V.A.-E. and Lin, H.-J. 1995. Studies on the effect of protectants on Lactobacillus acidophilus
strain dehydrated under controlled low-temperature vacuum dehydration and freeze-drying by using
response surface methodology. Journal of the Science of Food and Agriculture. 68: 191-196.
157. Bayrock, D. and Ingledew, W.M. 1997. Mechanism of viability loss during fluidized bed drying of baker's
yeast. Food Research International. 30: 417-425.
158. Bayrock, D. and Ingledew, W.M. 1997. Fluidized bed drying of baker's yeast : moisture level, drying
rates, and viability changes during drying. Food Research International. 30: 407-415.
159. Ananta, E., Volkert, M., Knorr, D. 2005. Cellular injuries and storage stability of spray dried
Lactobacillus rhamnosus GG. International Dairy Journal. 15: 399-409.
160. McGee, H.A. and Martin, W.J. 1962. Cryochemistry. Cryogenics. 2: 1-11.
General introduction 25
161. Castro, H.P., Teixeira, P.M., and Kirby, R. 1995. Storage of lyophilized cultures of Lactobacillus
bulgaricus under different relative humidities and atmospheres. Applied Microbiology and Biotechnology.
44: 172-176.
162. SLMB. 1991. Wasseraktivität. Schweizer Lebensmittelbuch. Kapitel 64.
163. Castro, H.P., Teixeira, P.M., and Kirby, R. 1996. Changes in the cell membrane of Lactobacillus
bulgaricus during storage following freeze drying. Biotechnology Letters. 18: 99-104.
164. To, B.C.S. and Etzel, M.R. 1997. Survial of Brevibacterium linens ATCC 9174 after spray drying, freeze
drying, or freezing. Journal of Food Science. 62: 167–170.
165. Bozoglu, T.F., Özilgen, M., and Bakir, U. 1987. Survival kinetics of lactic acid starter cultures during
and after freeze drying. Enzyme and Microbial Technology. 9: 531-537.
166. King, V.A.E., Lin, H.J., and Liu, C.F. 1998. Accelerated storage testing of freeze-dried and controlled
low-temperature vacuum dehydrated Lactobacillus acidophilus. Journal of General and Applied
Microbiology. 44: 161-165.
167. Achour, M., Mtimet, N., Cornelius, C., Zgouli, S., Mahjoub, A., Thonart, P., and Hamdi, M. 2001.
Application of the accelerated shelf life testing method (ASLT) to study the survival rates of freeze-dried
Lactococcus starter cultures. Journal of Chemical Technology & Biotechnology. 76: 624-628.
168. Gilliland, S.E. and Speck, M.L. 1977. Instability of Lactobacillus acidophilus in yoghurt. Journal of Dairy
Science. 60: 1394-1398.
169. van de Guchte, M., Ehrlich, S.D., and Maguin, E. 2001. Production of growth-inhibiting factors by
Lactobacillus delbrueckii. Journal of Applied Microbiology. 91: 147-153.
170. Shah, N.P. 2000. Probiotic bacteria: selective enumeration and survival in dairy foods. Journal of Dairy
Science. 83: 894-907.
171. Dave, R.I. and Shah, N.P. 1997a. Effectiveness of ascorbic acid as an oxygen scavenger in improving
viability of probiotic bacteria in yoghurts made with commercial starter cultures. International Dairy
Journal. 7: 435–443.
172. Dave, R.I. and Shah, N.P. 1997b. Effect of cysteine on the viability of yoghurt and probiotic bacteria in
yoghurts made with commercial starter cultures. International Dairy Journal. 7: 537-545.
173. Gomes, A.M.P. and Malcata, F.X. 1999. Bifidobacterium spp. and Lactobacillus acidophilus: Biological,
biochemical, technological and therapeutical properties relevant for use as probiotics. Trends in Food
Science and Technology. 10: 139-157.
174. Kailasapathy, K. and Chin, J. 2000. Survival and therapeutic potential of probiotic organisms with
reference to Lactobacillus acidophilus and Bifidobacterium spp. Immunology and Cell Biology. 78: 80-88.
175. Ananta, E., Birkeland, S.-E., Corcoran, B., Fitzgerald, G., Hinz, S., Klijn, A., Mättö, J., Mercernier,
A., Nilsson, U., Nyman, M., O'Sullivan, E., Parche, S., Rautonen, N., Ross, R.P., Saarela, M.,
Stanton, C., Stahl, U., Suomalainen, T., Vincken, J.-P., et al. 2004. Processing effects on the
nutritional advancement of probiotics and prebiotics. Microbial Ecology in Health and Disease. 16: 113-
124.
26
2 FLOW CYTOMETRIC ANALYSIS FOR INACTIVATION STUDIES
Application of flow cytometric analysis to evaluate the mechanism of inactivation by physical
treatment methods
Flow cytometric analysis for inactivation studies 27
2.1 Introduction
2.1.1 Effect of physical inactivation treatments on microorganisms
Heat
Reduction or inactivation of microbial populations by thermal processes is a common
process of food preservation in use nowadays. Cells contain several targets for the action of
heat. The basal heat resistance of microorganism is related to the intrinsic heat stability of
the essential macromolecules, including ribosomes, nucleic acids, enzymes and proteins
inside the cells and the membrane [1, 2]. During exposure to heat structural changes in these
critical cellular components may lead to cell death, for instance loss of specific secondary
and tertiary structure of ribosomal subunit, protein coagulation, degradation of RNA and
membrane damage. However, the exact prime cause for cell death upon heat exposure is
still not clearly understood [3]. Heat treatment at temperatures in the vicinity of 60°C was
reported to cause damage in the cytoplasmic membrane of Lactobacillus bulgaricus,
whereas for temperature of 65°C and immediately above, ribosomes and/or proteins
denaturation as well as cell wall damage may be responsible for thermal death [4]. Electron
microscopy study on heat treated Bacillus cereus showed that following exposure to 62°C for
2 min or 15 min or to 100°C for 5 min the appearance of membrane changed: it developed
holes and fractures [5]. Concentric rings appeared inside the cytoplasm and the ribosomes
disappeared. After heating to higher temperature, dense areas of precipitation owing to
protein coagulation appeared in the cytoplasm [6].
High pressure
To meet the requirement of producing high quality food under appropriate minimization of
microbial contamination, processing concepts based on the use of emerging technologies
have been developed, into which high hydrostatic pressure treatment could be classified.
The efficacy of this novel processing in inactivating different types of microorganisms is well
documented [7-10]. Moreover, the potential of high pressure technology as an alternative tool
on modifying macromolecules (proteins, polysaccharides, etc.), as well as to assist and/or to
substitute conventional freezing methods, has been reviewed [11-13].
Many studies had been initiated to provide improved knowledge on the mode of action of
high pressure on microorganisms.
Hydrostatic pressure was reported to affect the intracellular pH of microorganisms by
enhancing the dissociation of weak organic acid, increasing the permeability of the
cytoplasmic membrane and inactivation of enzymes required for pH homeostasis [14-16].
The major constituents of this crucial cellular function are located in cellular membrane. The
regulation of a fairly constant internal pH (pHin) was considered as crucial for maintaining
Flow cytometric analysis for inactivation studies 28
microbial viability, and consequently a substantial pressure-induced loss of this transport
functionality would reduce the ability of microorganisms to survive harsh environment during
and after pressure treatments [17].
Lactic acid bacteria, which do not possess an electron transport chain, maintain a pH
gradient by proton-translocating activity of F0F1 ATPase. This membrane-bound enzyme
appears to be a possible target for pressure induced inactivation of microorganisms. This
multimeric enzyme can either synthesize ATP using protons or conversely expulse protons
out of the cell with the energy provided by the ATP hydrolysis [18]. In lactic acid bacteria, the
latter activity increases at low pH and is crucial to maintain the ∆pH [19]. Damage of
membrane bound H+-ATPase, which is responsible for pH homeostasis in acidic environment
by discharging H+ from the cell, can reduce the ability to tolerate acidic conditions [20, 21].
Early study on the effect of relatively low pressure (50 MPa) on the F0F1 ATPase of the
isolated membranes of Streptococcus faecalis showed that under the investigated condition
the proton-translocation step, and not the ATP hydrolytic step was inhibited by pressure [22].
Upon pressurization of Lactobacillus plantarum at 250 MPa, the activity of F0F1 ATPase was
decreased [23]. Along with that, acid efflux was impaired and the regulation of pHin was
hampered. Upon observing the ATP pool and acid efflux it was also noted that the glycolysis
(ATP generating system) was less sensitive to pressure than F0F1 ATPase (ATP utilizing
system). Exposure to high pressure (at 200 or 300 MPa) on lactic acid bacteria suspended in
acidic environment resulted in a decrease of the pHin to the extracellular pH value and the
capacity to restore the pHin was totally lost [17].
Furthermore, not only pH homeostasis mechanisms were affected by pressure, but also
other critical membrane-bound transporters, particularly multi drug resistance (MDR)
transport system. It is a group of integral membrane proteins that transport hydrophobic
drugs and lipids across the cell membrane. MDR transporters can be divided into two
classes based on their source of energy: Secondary transporters, which use proton gradients
to facilitate an antiporter mechanism, and ATP binding cassette (ABC) transporters that
couple the efflux of substrate across the cell membrane with energy derived from ATP
hydrolysis. ABC transporters belong to one of the largest superfamilies of proteins and that
either import or export a broad range of substrates that include amino acids, ions, sugars,
lipids, and drugs [24].
Pressure as high as 200 MPa was reported to inactivate HorA of L. plantarum [15]. This
functional molecule is a membrane-bound, ATP-dependent MDR enzyme, which confer hop
resistance on beer spoilage bacteria and has high homology to other bacterial ATP-binding
cassette-type multidrug transporters. Following exposure to a sub-lethal level of pressure,
which is sufficient to inactivate HorA, cells of L. plantarum failed to survive during subsequent
storage in media containing hop extract [25].
Flow cytometric analysis for inactivation studies 29
Furthermore, high pressure studies on LmrP activity of L. lactis revealed that the loss of
viability after pressure treatment correlated with reduced LmrP activity and the loss of the
ability to restore a ∆pH after pressure treatment [26]. This enzyme is one MDR transport
enzyme in L. lactis, which is involved in drug/toxin extrusion in a proton motive-dependent
(∆pH dependent) manner. High pressure impaired the activity of the enzyme and not the
proton motive force.
The composition and the phase behaviour of cytoplasmic membranes were also implicated in
irreversible pressure denaturation of membrane-bound proteins, such as HorA [27]. It is
known that the function of membrane-integral proteins is not only influenced by its three
dimensional structure but also to the composition and the phase behaviour of the
cytoplasmic membrane.
Elevated pressure was reported to reduce the activity of transmembrane enzyme Na+/K+-
ATPase either by induced dissociation and/or unfolding of protein subunits of the enzyme or
by alterations of membrane fluidity which hinders conformational transition of the protein
required for the reaction [28]. Similarly, other research group came to the conclusion that the
inhibition of this enzyme, which couples the chemical energy from hydrolysis of ATP to
dynamic gradients of Na+ and K+ across the plasma membrane, was due to pressure induced
ordering of the acyl-chains of lipid matrix or due to subunit dissociation [29]. At least three
stages of damage was proposed to take place upon pressurizing this enzyme, which was
embedded in phospholipids bilayer. Pressure of 100 MPa or lower induces a decrease in the
fluidity of the liposome’s lipid bilayer and reversible conformational changes of the
transmembrane protein, resulting in the functional disorder of the enzyme Na+/K+-ATPase
[30]. Pressure of 100 to 220 MPa causes a reversible phase transition of the lipid bilayer and
the dissociation of protein subunits. These changes bring about the separation of protein and
the lipid bilayer, producing transmembrane tunnels. Pressure of 220 MPa or higher is
accompanied by irreversible protein unfolding and fragmentation of the lipid bilayer, thus
destroying the gross membrane structure. However, due to the presence of cytoskelett and
other constituents of cell envelope, which confer higher mechanical or structural stability to
microbial membranes, critical pressure level in inducing irreversible damage on microbial
membrane should be higher compared to the one required to irreversibly degrade artificial
membrane.
As already mentioned before, pressure upshift induced in one-component phospholipids
bilayer a decrease in the membrane fluidity, which is as a result of phase transition from the
liquid crystalline state to the gel phase. It was previously suggested that the fluidity of the
bacterial membrane had a direct role in the pressure resistance of bacteria [31]. FT-IR
spectroscopy of the bacterial membrane revealed that the addition of sucrose can reduce the
melting temperature (Tm) of the gel-liquid crystalline phase transition [26]. At higher
Flow cytometric analysis for inactivation studies 30
pressures the Tm of bacterial membranes were lower in the presence of sucrose than in its
absence so that the pressure required to induce a phase transition from the liquid crystalline
phase to gel phase is increased when sucrose is present. For a given pressure level addition
of sucrose can prevent bacterial membranes from experiencing phase transition and thus
maintained in them a more fluid state during pressure treatment and partly contribute to
enhanced resistance of bacteria to pressure [26].
Irreversible degradation of intracellular components was also deemed responsible for viability
loss. A correlation was observed between loss of cell viability and decrease in ribosome-
associated enthalpy in cells subjected to pressures of 50-250 MPa for 20 min [32]. Cell death
and ribosome damage were therefore closely related phenomena. Transmission electron
microscopy (TEM) micrographs showed that pressure induced denaturation of ribosomes
may be manifested by the presence of dense compacted interior regions of the cytoplasm
after pressure treatment at 500 MPa [33]. The authors of the latter study also observed a
reduction of the area of the ribosomal peak as a function of pressure, which emphasized the
implication of damage on the ribosomal subunit in cell death.
In addition, a study on the electrophoretic mobility of pressure treated intracellular enzymes
of L. monocytogenes strains was made to relate the cell death of these organisms to the
pressure induced conformational modifications of those enzymes [34]. Although no positive
correlation was found between the overall pressure resistance of the organisms and the
pressure resistance of the investigated enzymes (13 types), a wider range of metabolic
enzymes should be assayed in order to elucidate the contribution of intracellular enzyme to
overall pressure resistance.
The role of certain proteins on bacterial inactivation was emphasized by Ludwig et al (1996)
and Perier-Cornet et al (2005), who based their ideas on the similarity of elliptic form of the
p,T-isokinetic stability diagram of E. coli with the elliptic character of protein denaturation
diagram [35, 36].
Studies on changes on cell morphology following pressure treatment revealed that despite
the clear evidence on pressure induced damage on this membrane-bound enzyme no direct
morphological changes on the membrane of L. plantarum could be observed [23]. Scanning
electron micrograph of L. monocytogenes showed that no morphological changes were
observed on cells inactivated by pressure as high as 345 MPa [37]. Pressure induced
damage on cell morphology was more pronounced in exponential growth phase as
evidenced by observations made with transmission electron microscopy on L. lactis or E.coli,
which stressed that exposure to high pressure caused cell envelope damage [38, 39].
Flow cytometric analysis for inactivation studies 31
Ultrasound
The application of ultrasound processing in the food industry for preservation purposes has
received increasing attentions. Briefly, it is broadly accepted that ultrasound alone is not
effective enough to inactivate bacteria in food [40]. However, improvements could be made
by coupling of ultrasound with heat treatment. The concept of ultrasound assisted thermal
processing (thermosonication), which is based on the synergy between ultrasound and heat
for bacterial inactivation, has been proven to be of potential interest in food preservation,
especially to enhance the lethal effect of conventional thermal treatments. Several research
groups extensively studied the bactericidal effect of ultrasound and particularly its synergistic
potential when applied simultaneously with heat [41-43]. Recent investigations have shown
the influence of amplitude, external static pressure and temperature as well as pH and
composition of treatment medium as the key processing variables [44, 45]. Generally, the
mechanism underlying microbial inactivation during ultrasound treatment in liquid medium
was related to physical disruption (shear stress, localized heating) and chemical reactions
(production of free radicals) within the microorganisms’ cell [40]. These degradative events
occurred as a consequence of the microscopic shock waves, which were generated upon
implosion of gas bubbles [46]. The implosion itself resulted from the pulsation of the gas
bubbles, which underwent regions of alternating compression and expansion within the
propagated longitudinal waves [44].
2.1.2 Flow cytometry
The need to have a quasi real-time assessment microbiological method in order to describe
the viability state of bacteria in a more precise manner triggers the development of
microbiological rapid methods or improvement of already existing ones. Flow cytometry is
regarded as one versatile tool for research in microbiology, which exhibits three unique
technical properties of high potential to be used in various microbiological studies including
assessment of microbial viability or metabolic activity, monitoring of gene expression system,
as well as identification and enumeration of microorganisms [47, 48] :
(i) its tremendous velocity to obtain and process data; allowing analyses to be
performed at a flow rate of 10 – 100 µL min-1 and detection of up to 10000 events s-1,
when the microbial concentration in the processed sample is sufficiently high;
(ii) high-speed multiparametric data acquisition and multivariate data analysis, which
combine direct and rapid assays to determine numbers, cell size distribution and
additional biochemical and physiological characteristics of individual cells, thus
revealing the heterogeneity present in a population; and
Flow cytometric analysis for inactivation studies 32
(iii) the sorting capacity of some cytometers, which allows the physical separation or
transfer of specific populations or even single cells into tubes, onto slides or onto agar
plates, thus allowing further physical, chemical, biological or molecular analysis and
establishing a link between the reproductive viability and the staining pattern of
bacteria.
Figure 1
(a) Schematic diagram of the basic components of a Beckman Coulter flow cytometer
(BeckmanCoulter Inc., Miami, Florida, USA)
(b) More detailed view of the interaction between cells and light in the illumination zone as well as the
types of signals detected by the sensors.
Forward angle light scatter is measured with FS sensor at low angle with a photodiode. Right
angle light scatter (detected by SS sensor) and fluorescence (detected by FL1 to FL3 sensors) are
collected at 90°, split by a series of dichroic mirrors and filters in different colors, and measured by
photomultiplier tubes. Graph was taken from http://www.cytobuoy.com/index.html
(c) Hydrodynamic focusing of the sample stream through flow cell by regulating the pressure of the
sheath fluid against the cell suspension in order to align the particles in single file. Graph was
taken from http://www.facslab.toxikologie.uni-mainz.de/zytometrie.jsp
Figure 1 shows the basic components of a flow cytometer, which is constituted of a flow cell,
a light source, optics, detectors, electronics and computer. In addition, a flow cytometer can
be equipped wit a cell-sorting device. The working principle of a flow cytometer can be briefly
described as follows: Microbial cells in suspension flow in single-file through a laser-
illuminated zone where they scatter light and emit fluorescence that are collected, filtered,
amplified and converted to digital values that are stored in a list mode data files on a
Flow cytometric analysis for inactivation studies 33
computer, where each event (i.e. presence of a microbial cell) with the corresponding data
for each parameter is recorded sequentially. The magnitudes of the parameter measured
(SS, FS, or fluorescence) are sorted electronically into ‘bins’ or ‘channels’. As a result, a set
of rapid, multiparametric measurements could be performed on each single cell of interest.
The delivery of the cells in a single file modus, i.e. one after another through a focused laser
beam (usually 488 nm blue laser) is ensured by hydrodynamic focusing using a sheath fluid,
normally a saline solution (Fig. 1c).
The scatter parameters measured by flow cytometer are known as forward scatter (FS,
measured in forward angle direction) and side scatter (SS, measured in the right angle
direction), which provide information about physical characteristics of a cell such as its size
and granularity, respectively [49]. The measurement of FS/SS parameters can be used to
distinguish cells in a mixed sample according to their morphological properties, thus allowing
exclusion of background from the cell of interest. More interesting for microbiological analysis
is the possibility to stain cells with fluorescent dyes or fluorogenic substrates, which then
allow the analysis of structural properties or biological activities as well as taxonomic analysis
on single cell level. As can be seen in Figure 2 the fluorescent dyes usually used may be [50-
52]:
(i) stains which bind to (or react with) particular molecules such as DNA, RNA or protein,
(ii) fluorogenic substrates which reveal distributions in enzymatic activity,
(iii) indicators which change their property as a function of pHin or which are taken up in
response to the state of membrane polarization, or
(iv) antibodies or oligonucleotides tagged with a fluorescent probe.
For visualization purposes, data are displayed either as a frequency distribution where the
magnitude of the parameter measured (SS, FS, or fluorescence) is plotted against the
number of cells. Alternatively, the data can be represented by two- or three-parameter
density plots of light scattering versus fluorescence or – in case of dual staining procedure –
the fluorescence from a DNA stain versus the fluorescence owing to microbial enzyme
activity. Thus, an impression about the distribution of a variety of properties of interest
amongst the cells in the population as a whole can be gained.
Flow cytometric analysis for inactivation studies 34
Figure 2
Different cellular target sites for physiological and taxonomical probes used in combination with flow
cytometric analysis. Figure is taken from [52].
2.1.3 Determination of viability status of microorganism with fluororescence probes
The determination of the impact of industrial treatment on microorganisms – either during
cultivation, inactivation or preservation steps – basically relies on the use of classical plate
count methods, which is still regarded as the benchmark method for the determination of
viability. According to this method viability is defined as the ability to reproduce and form a
visible colony, which typically contains at least 106 cells [53]. Reproductive growth or
culturability is regarded as the highest level of physiological fitness [51], for which some
requirements need to be fulfilled: an intact cytoplasmic membrane which functions as a
barrier between the cytoplasma and the extracellular environment, DNA transcription, RNA
translation, generation of energy for maintenance of cell metabolism, biosynthesis of
essential molecules such as proteins, nucleic acids, polysaccharides, as well as growth and
multiplication [54].
However, classic culture technique has some drawbacks related to the fact that it could not
give insights about population heterogeneities or the physiology of individual organism [53].
Besides, long determination time which principally arises from the application of this method
leads to significant limitations in anticipating abnormalities in growth or surviving behaviour of
bacteria during industrial processes. Moreover, culturing technique may underestimate the
numbers of truly viable bacteria. It is known that under stress conditions or limitation of
nutrient availability or due to imposed sub-lethal injury, some cells can enter a non-culturable
state, yet they can still exhibit metabolic activity [55, 56]. In natural environments such as soil
or seawater or gastro-intestinal tract bacteria from this viable but not culturable (VBNC) state
can be encountered as well [57-60]. Reports on bacteriological quality of different types of
Flow cytometric analysis for inactivation studies 35
water revealed that the ratio of colony forming unit to total bacterial number were less than
2% [61] or even 2-4 log orders of magnitude lower [62]. Apparently, in this VBNC state
bacteria undergo metabolic change leading to the production of cells that no longer actively
form colonies on solid media, but retain other indicators of cell viability, such as active
membrane potential, maintenance of cellular integrity and the capacity for metabolic activity
[63]. It was thought that the risk of the residual metabolic activities of such bacteria not
detected by standard culturing technique (or non-growing) could lead to food spoilage or
accumulation of toxins due to retention of gene encoding virulence and resuscitation of
nonculturable cells, so that nonculturable pathogens may still pose a hazard to public health
[47, 63, 64].
The application of flow cytometry analysis permits simultaneous evaluation of multiple
cellular parameters, both structural and functional on single cell level. This approach would
then allow an extended description of bacterial viability state beyond the one based on
reproductive capacity as well as identification of heterogeneities within population with regard
to structure and function [48, 65].
In the following the application of some physiological probes as well as possible combined
application among them, which are relevant for use in combination with flow cytometry, are
briefly reviewed.
Esterase activity
Esterases are present in all living organisms [66]. The fluorogenic substrate cFDA
(carboxyfluorescein-diacetate) is used primarily for the evaluation of cellular enzymatic
activity. It is a lipophilic, non-fluorescent precursor that readily diffuses across the cell
membranes [67]. In the intracellular compartment it undergoes hydrolysis of diacetate groups
by unspecific esterases into a polar, membrane-impermeant fluorescent compound cF
(carboxyfluorescein). Carboxyfluorescein is a derivative of fluorescein which are more
negatively charged at physiological pH, and thus less likely to leak from the cells [54]. The
substrate cleaving reaction of intracellular esterase is typically not energy dependent [64]
and the enzyme will remain functional in cells as long as it is retained by the intact
membrane and protected from the environment [48]. Consequently, the cells only remain
fluorescent if their membranes are intact and cF are unable to diffuse out; thus for cells to be
associated as viable, this probe requires both active intracellular enzymes and intact
membranes [68].
Many works had been performed to enumerate viable bacteria with cFDA [58, 62, 68-72].
cFDA was proved to be more effective at labelling Gram positive bacteria than Gram
negative bacteria [58]. Several studies had been conducted in order to evaluate the
application of other esterase substrates such as Calcein acetoxy methyl ester (calcein AM)
Flow cytometric analysis for inactivation studies 36
and carboxyfluorescein diacetate-succinimidyl ester (cFDA-SE) on bacteria. Although calcein
AM has greater fluorescence intensity, reduced bleaching of fluorescence, reduced leaching
from cells and is insensitive to pH changes between pH 5.5 and 10, it was found to be less
reliable for viability assessment of yeasts and bacteria [73]. cFDA-SE is another fluorogenic
esterified substrate similar to cFDA but differing by the presence of a succinimidyl ester (SE)
group that can bind strongly to free amines [67]. cFDA-SE is also cell permeant and the DA
groups are hydrolysed intracellularly by nonspecific esterases, resulting in a highly
fluorescent amine reactive fluorophore (cF-SE). This molecule can react with amine
containing residues of intracellular proteins, forming highly stable dye–protein adducts.
However, it was reported that cFDA-SE is a poor marker of bacterial activity due to the
occurrence of non-specific labelling of all cells, irrespective of their metabolic state [68]. Apart
from identification of intracellular esterase activity, the pH-sensitivity of some esterase
substrates such as BCECF (Bis-carboxyethyl-carboxyfluorescein) and cFDA-SE lead to their
application in determine bacterial intracellular pH [17, 74].
The major drawback in using esterase substrate to identify the bacterial viability status is
particularly based on the fact that intracellular fluorescein accumulation owing to enzyme-
substrate reactions in cells exposed to cFDA is not energy-dependent. As a consequence, it
may be expected that this staining technique does not reflect the energetic status of a cell
very directly and will not therefore adequately distinguish degrees of cell viability reflecting a
generalised physiological or energetic capacity [60].
In line with the proposed drawback it was reported that even cells killed by H2O2, γ-
irradiation, and heat still showed esterase activity and cF accumulation [48, 75]. On the other
hand it is also possible that cells with damaged membranes contain active esterases; they
just can not retain the products. Irreversible membrane permeabilization in absence of
esterase inactivation was achieved by applying pulsed electric fields on LGG (data not
shown), where following the pulsed electric fields treatment at 35 kV/cm the cells were
stained by PI and simultaneously extracellular cF fluorescence was observed.
The enzymatic conversion proceeds approximately as rapidly in cells treated with inhibitors
of energy metabolism as in control cells (Shapiro, H., personal communication in Cytometry
Mailing List). As typically used, the conversion of cFDA into cF discriminates between cells
with intact membranes, which retain the dye, and cells with damaged membranes, from
which dye leaks out much more rapidly. Thus, at the end, the information obtained from cF
retention (dye retention assay) and from exclusion of nucleic-acid excluded dyes such as PI
(dye exclusion assay) is basically the same, i.e. that the membrane is intact [64]. In
conclusion, the occurrence of esterase activity and intracellular accumulation of cF does not
necessarily reflect crucial metabolic activities which are involved in the maintenance of
reproductive growth [48]. Metabolic activity should be better demonstrated using dyes
Flow cytometric analysis for inactivation studies 37
responsive to energy metabolism, for example, indicators of membrane potential. The
measurement of membrane potential allows a more accurate evaluation of functional cell
integrity, because it corresponds to the energetic state of the membrane and the cell’s
capacity to synthesize ATP [76].
Pump activity
Bacteria have a very efficient efflux pump, which result in rapid efflux of dye, thus hampering
interaction with the target molecule and is regarded as a major obstacle for multiparametric
measurement. On the other hand dye efflux can serve as an additional measure of cell
viability [77]. For instance, extrusion of carboxyfluorescein (cF) from intact cells of L. lactis
[78] and S. cerevisae [75] was found to take place in an energy dependent manner. The
efflux experiments showed an excellent correlation between the viability of S. cerevisiae cells
and the ability to translocate cF [75]. Labelling of L. lactis with cF combined with the ATP-
driven efflux of cF was proven to be suitable as an additional indicator of metabolic
performance, i.e. reproduction and acidification of the stressed cells [79]. With help of the cF-
efflux assay population’s heterogeneity following treatment could be resolved, i.e. cells that
are capable of performing glycolytic activity and getting energized upon sugar addition are
distinguished from the cells that are not [51].
Active dye extrusion from energized cells as described for rhodamine 123 and or fluorescein
may be linked to existing multidrug resistant pumps [66]. Proton gradient could be detected
by observing the efflux of ethidium bromide (EB). This nucleic acid dye can cross the intact
cytoplasmic membrane but is actively pumped out of the cells via a non-specific proton
antiport transport system [80]. Active exclusion of EB was reported to correlate with
metabolic activity [47].
Membrane integrity
The presence of an intact membrane is prerequisite for maintaining the capability of
metabolic activity. Cells can recover from a transient permeabilization, but if the membrane is
irreversibly compromised the cell is doomed to die. Without an intact membrane a cell can
not maintain electrochemical gradients so it will loose its membrane potential and pH
gradient. As the intracellular compartment is no longer separated from the environment,
components leak out of the cell and potentially toxic chemicals from the environment diffuse
freely into the cell. Under this circumstance breakdown of cell components and finally the
degradation of the whole cell occur [51]. However, whether or not the criterion of membrane
integrity is a reliable indicator of viability is a matter of controversy; basically due to the
presence of a significant proportion of bacteria after heat inactivation trials, which was poorly
stained by propidium iodide (PI) [79, 81]. PI is most commonly applied for the determination
Flow cytometric analysis for inactivation studies 38
of membrane integrity [54, 66]. It is a membrane-impermeant, nucleotide-binding probe
which is excluded from cells with intact membrane. Following loss of membrane integrity PI
diffuses into the cells and intercalates into the double stranded helical structure of nucleic
acids (DNA or RNA) forming a red-fluorescent complex. Dye exclusion assay using PI has
been considered as the most reliable stain as PI positive cells have not yet been shown to
grow upon sorting [47]. This supports the hypothesis that with the breakdown of cell wall
integrity, irreversible damage is achieved, thus it remains the best indicator for cell death.
Other impermeant dyes which can only enter cells with sufficient membrane damage and are
excluded by cells with intact membrane are the cyanic nucleic acid dyes TOTO-1, TO-PRO-
3, Sytox Green™ [67]. Compared to PI these new DNA probes are highly fluorescent [82]
and allow better discrimination of intact and membrane-compromised cells [83].
EB, which is a homolog of PI is not suitable for use as membrane integrity indicator since it is
actually taken up by bacteria and is rapidly removed by an efflux pump [84]. Compared to EB
PI carries an additional positive charge over ethidium and is therefore more likely to remain
excluded from membrane-intact cells [76]. Thus, it is more appropriate to use EB as
supravital or total cell stain in combination with a mixture of sodium azide and Tween-20,
which could overcome the dye extrusion of EB without compromising membrane integrity
[47].
Membrane potential
Membrane potential plays a critical role in bacterial physiology. As a component of the proton
motive force, it is intimately involved in the generation of ATP, in the bacterial autolysis, in
glucose transport, in chemotaxis as well as in survival at low pH [85]. The transmembrane
electrical potential gradient in metabolically active bacteria is typically 100 mV, with the
interior negative, originates from selective permeability and the active transport of ions
across the cytoplasmic membrane, which causes differences in the concentrations of ions on
opposite sides of the cell membrane [54, 64]. The magnitude of membrane potential is
reduced to zero in dead cells, particularly when the integrity of membrane is destroyed by
physical or chemical agents or by certain classes of antimicrobial drugs [85]. Alternatively,
collapse in the membrane potential can be due to treatment with proton ionophores by
eliminating the proton gradient across membrane. Any treatment that reduces the magnitude
of membrane potential is said to depolarize the cell.
To analyze the membrane potential, distributional probes are usually applied. These are
lyphopilic dyes that can readily pass the cell membrane and accumulate according to their
charge [64]. Positively charged lipophilic rhodamine 123 and cyanine dyes (DiOC6) can pass
cell membranes, but are only retained in cells with a polarized cytoplasmic membrane. The
fluorescence level is determined by the magnitude of membrane potential. DiOC6 was used
Flow cytometric analysis for inactivation studies 39
to detect decrease of membrane potential in Listeria cells in response to bacteriocin, which
reflects the mode of action of these antimicrobial peptides in inducing pore formation leading
to ionic leakage [86].
However, cationic membrane potential probe is not quite suitable for Gram negative bacteria,
since they frequently do not take up cationic stains unless the outer membrane of the cells is
permeabilized [81, 87]. Furthermore, the measurement of membrane potential with
rhodamine 123 is complicated due to the presence of active transport system, which pump
out the fluorescence stain [66, 88].
On the other hand, anionic oxonol dyes such as DiBAC4(3), which is a negatively charged
molecule, can enter depolarized cells or is excluded if a membrane potential is present. In
depolarized cells this dye binds to lipid-rich intracellular components, resulting in bright green
fluorescence. The aforementioned problems with cationic dyes are not an issue with these
anionic dyes because they only enter the cell once the active transport system have ceased
and the membrane potential is lost [87]. The use of this dye allow a clear discrimination
between viable and depolarized/dead cells of E. coli after various treatments [81, 89, 90] or
of B. lactis cells after exposure to bile salts [91]. Moreover, a strong relationship was found
between the percentages of depolarized cells stained with DiBAC4(3) and the degree of cell
membrane damage in dried yeast as measured by the more traditional method od leakage of
intracellular compounds [92].
It was noted that fluorescence signals for both cationic and anionic lipophilic dyes are
strongly dependent on cell size [53]. Thus, for precise and accurate estimation of membrane
potential using DiOC2(3) a ratiometric technique is developed. With help of this approach the
influence of cell size or size variations on fluorescence signal from this dye could be
decoupled [85]. The numerical value of membrane potential is described as the ratio of size
but potential independent green fluorescence to red fluorescence, which is both dependent
on size and potential [53].
Multiple physiological probes
Multiple physiological probes, which are used in multiple assay or multi-staining approach,
could facilitate acquisition of information about various cell properties and improved
dissection of several sub-populations based on their differential dyes uptake. Care has to be
taken in selecting appropriate combination of dyes with regard to excitation and emission
wavelength so as to allow distinction of each probe in the presence of other [60]. Software
compensation is sometimes necessary when spectral overlap between the emitted
fluorescence of stain mixtures occurs.
The multiple staining strategy using fluorescent dyes cFDA and PI was broadly applied. This
staining strategy was successfully used to characterize the effect of heat treatment on
Flow cytometric analysis for inactivation studies 40
Lactobacillus plantarum [66], the permeabilizing effect of bile salt and acid on Lactococcus
lactis [83], the effect of ethanol stress on Oenococcus oeni [93], and the response of
Bifidobacterium lactis towards bile stress [91]. Moreover, cFDA/PI staining had also been
applied for the determination of viability of Trichomonas vaginalis [94] and for the analysis of
bacterial activity in the aquatic environment [61, 95].
The applicability of the commercially available LIVE/DEADBacLight bacterial viability kit
has been evaluated on a wide spectrum of bacteria [62, 69, 96-100]. This kit was developed
to differentiate live and dead bacteria based on plasma membrane permeability. The staining
mechanism using LIVE/DEAD kit on bacterial cells is based on the attachment of the non-
fluorescent agents on nucleic acids [67]. Once the DNA-dye complex is built fluorescence
could be measured. This kit is constituted of two fluorochromes, which have distinct
fluorescent behaviour in terms of emission wavelengths and membrane permeability. The
first component is the membrane-permeant stain SYTO9®, which stains all cells and can be
used to distinguish particles from cells. Due to membrane damage, the second dye, the
membrane-impermeant dye propidium iodide (PI) penetrates into cells and quenches the
green SYTO9® fluorescence. When used in combination, intact cells are labeled green and
cells with damaged membranes are labeled red.
Simultaneous staining with membrane impermeant and supravital DNA stains such as PI and
EB combined with azide as decouplers for EB-efflux transporter was reported to be useful to
differentiate between dead and potentially viable bacteria [47].
A triple fluorochrome staining procedure involving EB, PI and DiBAC4(3) was developed to
differentiate starved cells of Salmonella typhimurium according to their dye uptake behavior
[56, 64]. Compared to the triple staining method using rhodamine 123 as membrane potential
dye the staining method with DiBAC4(3) has a wider range of application since rhodamine
123 staining does not work on pumping cells, i.e cells with an active efflux mechanism [64].
The sub-populations observed were categorized as metabolic active (actively excluding EB),
deenergized but with a polarized cell membrane (uptake of EB but exclusion of DiBAC4(3)),
depolarized (uptake of both dyes) and permeabilized (uptake of PI). In combination with
sorting and reproductive growth assay of separate sub-populations on agar it was found that
most polarized cells could be recovered as well as a significant fraction of the depolarized
cells. The latter result indicate that cells without membrane potential are not necessarily non
viable [54, 81]. Pump activity was found to be a more sensitive indicator of cell stress since
this activity already ceases prior to electrical depolarisation [64]. The triple staining technique
with EB, PI and DiBAC4(3) was also applied to examine cell physiology of S. cerevisae and
E. coli under various fermentation conditions [56, 87]. Considerable drop in cell viability,
which correlated with cytoplasmic membrane depolarisation and increase of permeability,
Flow cytometric analysis for inactivation studies 41
was observed in the latter stages of fed-batch fermentations, at which the cells were exposed
stress associated with glucose limitation [101].
These examples emphasize the potential of using multiple probes in giving single-cell based
information on the physiological condition of bacteria and the use for monitoring changes
following imposed stress or environmental changes as well as for investigating
heterogeneities.
2.1.4 Objective
The objective of this study was to characterize the physiological/metabolic behaviour of
Lactobacillus rhamnosus GG population following exposure to heat, high hydrostatic
pressure and high intensity ultrasound with help of flow cytometric analysis. The investigated
organism served as a model system prior to further investigations with typical food spoilage
bacteria. Flow cytometric analysis was performed under application of double staining
method with cFDA (carboxyfluorescein diacetate) and PI (propidium iodide). With this dye
combination the treatment effects on bacterial intracellular enzymatic activity and integrity of
cytoplasmic membrane could be determined. Furthermore, the ability of Lactobacillus
rhamnosus GG to extrude accumulated cF upon energization with glucose represents a
physiological characteristic, which could be ascertained as an additional vitality marker. In
conjunction to conventional cultivation assays the application of this rapid, fluorescence-
based method along with a suitable measurement strategies allow the mechanism of
bacterial inactivation by means of physical treatments to be better characterized and
considered during optimization of food decontamination processes.
2.2 Material and methods
2.2.1 Test organism
Lactobacillus rhamnosus GG (ATCC 53103) – thereafter abbreviated with LGG – was
obtained from Valio R&D (Helsinki, FI). This probiotic strain is of human origin. The beneficial
effects of LGG have been shown in many types of intestinal disturbances caused by
pathogenic bacteria and viruses, as well as in prophylactic use [102]. For long-term
maintenance LGG was stored as glass bead cultures (Roti-Store, Carl-Roth, Karlsruhe, D)
in a -80°C freezer (U101, New Brunswick Scientific, Nürtingen, D).
2.2.2 Inactivation treatments and microbiological analysis
One bead from deep-frozen culture was transferred into MRS broth (Oxoid, Basingstoke, UK)
and incubated overnight at 37°C. This broth was then used to inoculate the final broth under
adjustment of N0 ~ 103 CFU mL-1. The culture was incubated at 37°C up to stationary growth
Flow cytometric analysis for inactivation studies 42
phase for 24 h. Cells were harvested by centrifugation at 2700 X g for 10 min at 10°C,
washed twice with 10 mM sterile phosphate-buffered saline (PBS) at pH 7.0, and finally
resuspended in 50 mM PBS (pH 7.0) to an OD600nm of 10 (corresponding to a cell
concentration of 109 CFU mL-1).
For thermal treatment two hundred µL of cell suspension were transferred into glass vials
and immersed in water bath at 60, 68 or 75°C for different exposure time.
For high pressure treatment the suspension was filled into 1.8 mL - cryovials (Type 375299,
Nunc, Roskilde, DK). Pressure treatment was performed with a multi-vessel high pressure
unit (Type U111, Unipress, Warsaw, PL). This unit consists of five pressure chambers, which
are separated from each other via high pressure valves (Fig. 3). All chambers are immersed
in a water bath equipped with a thermostat, which allows a simultaneous treatment of five
different samples in one pressure build-up step at close to isothermal conditions. High
pressure treatments at different pressure levels were conducted at 37°C for 10 min.
Figure 3
Schematic hydraulic diagram of multivessel high pressure apparatus U111. The intensifier is
connected with the pressure vessels through high pressure valves (1-5). The multiplication factor
(~11) of the intensifier leads to a maximum pressure of 700 MPa. Valves 6-11 are used for loading
and unloading the pressure medium (silicon oil).
High-intensity ultrasound treatments were carried out using a Sonopuls HD 2070
homogenizer (Bandelin Electronic, Berlin, Germany). This unit was composed of a frequency
generator, a piezoelectric transducer UW70 and an amplifying horn SH70 with a sonotrode
KE76. The electrical energy, which produces oscillations of the piezoelements in the
Flow cytometric analysis for inactivation studies 43
transducer, is transformed by the horn and the sonotrode. The sonotrode surface vibrates at
a frequency of 20 kHz.
For all experiments a wave amplitude of 160 µm (equivalent to a power input of 17.6 W) was
applied. In detail, approximately 10 mL of microorganism suspension was filled into a glass
beaker. Then the sonotrode was immersed into the suspension. Samples for microbiological
analysis were taken periodically during the ultrasound application. The temperature changes
during ultrasound treatments were monitored with a K-type thermocouple.
After treatments the samples were rapidly cooled on ice, diluted in Ringer’s solution (No.
15525, Merck, Darmstadt, DE) and drop plated in duplicate on MRS agar (Oxoid,
Basingstoke, UK). Plates were placed in an anaerobic jar under anaerobic atmosphere,
generated by an anaerobic kit (AnaerocultA, Merck, Darmstadt, DE). The viable cell
numbers were determined after 48 h of incubation at 37°C.
All inactivation data were expressed as logarithm of the relative survivor fraction (log N/N0). N
refers to the bacterial count following exposure to a particular treatment, whereas N0
represents the initial count prior to the treatment. All experiments were performed at least in
triplicate.
2.2.3 Staining procedure and measurement strategies
Control or treated cells were initially incubated with 50 µM cFDA (Molecular Probes, Inc.
Leiden, The Netherland) at 37°C for 10 min to allow intracellular enzymatic conversion of
cFDA into cF. Immediately after this labelling, the cells were spun down and resuspended in
50 mM PBS-buffer (pH 7.0). This step is the followed by addition of 30 µM PI (Molecular
Probes, Inc. Leiden, The Netherland). The cell suspension was kept in ice bath for 10 min to
allow labelling of membrane-compromised cells prior to measurement in flow cytometer. To
measure the performance of treated cells in extruding intracellular accumulated cF activity,
cF-stained cells were incubated together with glucose 20 mM for a fixed holding time of 20
min at 37°C, as adapted from previous studies [75, 79]. The kinetics of cF-release from
glucose energized cells were monitored by incubating cF-labeled cells at 37°C in the
presence of glucose 20 mM and measuring the progress of cF-extrusion every 5 min.
2.2.4 Flow cytometric measurement
Flow cytometric analysis was performed on a Coulter®EPICS®XL-MCL flow cytometer
(BeckmanCoulter Inc., Miami, Florida, USA) equipped with a 15 mW 488 nm air-cooled
argon laser. All the parameters were collected as logarithmic signals. Green fluorescence of
cells stained with cF was collected in the FL1 channel (525 ± 20 nm), whereas red
fluorescence of cells labelled with PI was collected in the FL3 channel (620 ± 15 nm).
Flow cytometric analysis for inactivation studies 44
Spectral overlap between the emitted fluorescence of stain mixtures was eliminated by
appropriate software compensation. Acquisition of fluorescence data was performed by pre-
setting a gate in the forward-angle light scatter (FS) vs. sideward scatter (SS) plot, with help
of which bacterial cells of interest and artefacts could be discriminated. The flow rate was set
at typical values of 300-600 bacterial cells per s. The software package Expo32 ADC
(BeckmanCoulter Inc., Miami, Florida, USA) was used to analyse flow cytometry data. All
detectors were calibrated with FlowCheck Fluorospheres (BeckmanCoulter Inc., Miami-FL,
USA). Other specific settings of flow cytometer for this measurement are listed in Annex 2.
2.2.5 Analysis of flow cytometric data
Density plot analysis of green fluorescence (FL1) versus red fluorescence (FL3) was applied
to resolve the fluorescence properties of the population measured by flow cytometer (Fig. 1
and Fig. 2). With this graph the population was able to be graphically differentiated according
to their fluorescence behaviours. Table 1 describes the quadrant designation of stained cells.
Table 1
Quadrant designation of cells stained with cF and PI
Quadrant Labelling behaviour Cellular mechanism involved
#1 cF- and PI+ cF-accumulation as a result of esterase activity not
detectable
Membrane compromised
#2 cF+ and PI+ Active esterase
Membrane minimally damaged
#3 cF- and PI- Esterase activity not detected or cF extruded out of the cells
Intact membrane
#4 cF+ and PI- Active esterase
Intact membrane
Residual cF-accumulation activity following pressure treatments was calculated using
Equation 1, in which the post-pressure activity of the population framed in quadrants #2 and
#4 was set in relation to the activity of untreated cells, which were encountered in quadrant
#4.
Based on the shift of cF-stained population upon glucose addition from quadrant #4 into
quadrant #3 after a 20 min incubation period cF-extrusion in response to glucose
energization was able to be monitored (Fig. 2) . Equation 2 calculated the performance of
dye extruding mechanism of pressure treated cells. Similar to aforementioned staining
strategy, glucose energized cells were measured every 5 min to follow the kinetics of cF-
Flow cytometric analysis for inactivation studies 45
extrusion of pressure treated cells. In Equation 3 the relative number of population losing
intracellular accumulated cF could be followed over time.
100
4
4
[%] ⋅
=
Ctrl
p
Q
Q
EA Equation 1
EA : Residual enzymatic activity in response to a particular pressure treatment
Q4p : Percentage of population in quadrant A4 following pressure treatment
Q4Ctrl : Percentage of population in quadrant A4 prior to pressure treatment
100
4
4
1[%] ⋅
−= Q
Q
cFA Glu Equation 2
cFA : Measure of performance in extruding cF
Q4Glu : Percentage of population in quadrant #4 following glucose addition and 20 min
incubation
Q#4 : Percentage of population in quadrant #4 prior to glucose addition
100
4
4
[%]
0
_⋅
=
=t
Glut
Q
Q
RcF Equation 3
RcF : Relative number of cells still stained with cF in quadrant #4 following glucose addition
Q4t_Glu: Percentage of cells still stained with cF in quadrant #4 following glucose addition and
incubation for t min
Q4t=0: Percentage of cells still stained with cF in quadrant A4 prior to glucose addition
2.2.6 Statistical analysis
Statistical significance of the effect of pressure treatments on cell viability and pressure-
induced changes on physiological status of LGG was examined using one-way ANOVA test.
Differences were considered significant at the p<0.05 level of probability. All statistical
analysis were performed with Origin7 software package (OriginLab, Northhampton, MA,
USA)
2.3 Results and discussion
2.3.1 Basic pattern
To differentiate bacterial population based on their fluorescence properties, the dual-
parameter density plot of the green fluorescence (x-axis) and the red fluorescence (y-axis)
Flow cytometric analysis for inactivation studies 46
was used (Fig. 4). Each dot, which constitutes the cell cloud, represents one single cell,
which is plotted as a co-ordinate of its green and red fluorescence value. Principally, the
effect of physical treatments on LGG was evaluated as the ability to accumulate and retain
cF as an indicator of membrane integrity and enzyme activity and the uptake of PI to assess
membrane damage.
ab
Figure 4
Basic fluorescence density plots (cF, green fluorescence vs PI, red fluorescence) of intact and heat
treated cells of L. rhamnosus GG (a and b, respectively) following staining with cFDA
(carboxyfluorescein diacetate) and PI (propidium iodide). Heat treatment at 75°C for 30 sec was
performed to yield dead, membrane-compromised cells. The figures (in %) following the quadrant
number are associated with the percentage of the population in the corresponding quadrant.
The quadrants arrangement on the dot plot were set so that viable cells of LGG with intact
membranes were framed in quadrant #4 (Fig. 4a). Within this quadrant only the population,
which both actively accumulated cF and excluded PI, thus showing high green fluorescence
and low red fluorescence, was encountered. Prior to the treatment, all LGG cells were
encountered in quadrant #4 (Fig. 4a). Upon heat-induced rupture of cell membrane and loss
of cF-accumulation capacity the cells are not capable of excluding PI. This particular
population, which was solely labeled by PI, showed low green fluorescence and high red
fluorescence. Membrane damaged population was thus encountered in quadrant #1 (Fig.
4b).
2.3.2 Inactivation mechanisms by heat treatment
Figure 5 shows a sequence of fluorescence density plots; each of them showing the result of
cFDA/PI labeling on LGG after different holding time at 60°C. Prior to heat challenge, cells
were found in quadrant #4, which indicated that they were solely stained by cF (Fig. 5a).
Exposure to 60°C at increasing holding time resulted in a gradual increase of cells framed in
quadrant #3, while simultaneously the number of cells framed in quadrant #4 was decreasing
Flow cytometric analysis for inactivation studies 47
(Fig. 5b to 5f). It was also observed, that practically no cells are found in quadrant #1. After
300 s of thermal treatment the majority of the cells are encountered in quadrant #3 (Fig. 5f).
According to the quadrant designation described in Table 1, the occurrence of cells in
quadrant #3 indicates that these cells were labeled neither by cF nor by PI. The increasing
occurrence of a sub-population in this quadrant indicated that due to heat challenge the
energy-independent accumulation of cF – which is a physiological feature of untreated cells –
was considerably reduced. The absence of PI labeled population suggests that the
membrane integrity was still relatively high, thus not permitting PI penetration across the
membrane to intercalate with nucleic acids. Although the treatment time was prolonged to
300 s, only max. 15% of the population was showing positive PI fluorescence, suggesting
that the majority of the treated cells had still high membrane integrity (Fig. 5g). The absence
of PI labeling accompanied by increasing loss of cF-accumulating activity suggested that in
the cells exclusively heat inactivation of esterase occurred, whereas membrane remained
unaffected. Furthermore, when data from plate count method were taken into account (Tab.
2), it was evident, that even after only 120 s of exposure to 60°C 99.99999% of all bacteria
were inactivated (log N/N0 ~ -7). However, only as many as 4.5% of the whole population
were positively stained with PI (Fig. 5g).
Flow cytometric analysis for inactivation studies 48
bc
def
a
0 50 100 150 200 250 300
0
10
20
30
40
50
60
70
80
90
100
g
Percentage of cells (%)
Treatment time (sec)
g
Figure 5
Fluorescence density plots of L. rhamnosus GG in response to staining with cFDA and PI after heat
challenge at 60°C for different exposure time: 0 (a), 30 s (b), 60 s (c), 120 s (d), 240 s (e) and 300 s
(f). The figures (in %) following the quadrant number are associated with the percentage of the cells in
the corresponding quadrant.
(g) Kinetics of the increase in the percentage of L. rhamnosus GG cells in quadrants #1 and #3 after
heat treatment at 60°C, as derived from the sequence of density plots Fig. 2a to 2f. Cells encountered
in quadrant #3 (z) are neither labeled with cF nor with PI (cF- and PI-), whereas in quadrant #1 ()
only the sub-population solely stained by PI (cF- and PI+) are framed. The results are means based on
data from three or more independent experiments with error bars indicating standard deviations.
This findings suggests that thermal death could be achieved in absence of membrane
degradation, as emphasized by flow cytometric fluorescence pattern of cells killed at 60°C.
Lievense et al (1994) who attempted to distinguish between dehydration and thermally
induced damage on L. plantarum concluded, that drying at 5°C resulted in membrane
damage and cell death, as manifested by increased penetration of Dnase into cells, whereas
cell inactivation could be achieved upon exposure to 60°C in absence of membrane rupture
[103]. Studies conducted by Jepras et al (1995) revealed the occurrence of a significant
fraction of heat killed E. coli cells (65°C for 30 min) which was poorly stained by PI [81].
Similar findings has also been reported by Bunthof et al (1999), who demonstrated that from
L. lactis cells treated at 60°C for 90 s, as much as 69% were not PI labeled although these
cells could not be recovered on plates [79]. Alternatively, the cytoplasmic membrane of cells
killed by heat at 60°C might be ruptured, but the pore size induced might be too small to
allow proper diffusion of PI into cell interior. As a result, PI staining in such cells was not
observed within the incubation period after PI addition, i.e. 10 min in ice, as described in the
standard incubation protocol. A possible approach to confirm the delayed penetration of PI
would be a prolonged contact time with PI. Data on the delayed diffusion of PI into
lymphocytes that had been made necrotic by high pressure indicated that this is due to
Flow cytometric analysis for inactivation studies 49
gelatinised cytoplasmic proteins and steady-state level PI fluorescence was achievable only
after a contact time of 300 min [104]. A general concern which is often put forward upon
using PI, i.e. the low extinction coefficient and hence the relatively low fluorescence of this
probe [54], did not hold true, since a relatively high PI fluorescence could indeed be detected
on cells heat killed at 75°C (Fig. 4b).
A different fluorescence behavior was obtained upon heat kill of LGG at higher treatment
temperatures. When the fluorescence density plots of cells treated for a fixed holding time
(90 s) were compared, it is evident, that at temperatures of 68 and 75°C cells were
exclusively found in quadrant #1, suggesting that PI stained cells predominated (Fig. 6b and
6c), whereas only minor population was encountered in quadrant #1 after treatment at 60°C
(Fig. 6a).
In Fig. 5d the kinetics of change in the percentage of PI stained cells upon heat treatment at
various temperature is shown. Following treatment at 60°C only max. 10% of the population
was found in quadrant #1. In contrast, PI fluorescence occurred in more than 90% of the
cells even after only exposing them to 75°C for 30 s. Similarly, at 68°C after a holding time of
90 s more that 85% of the cells were exhibiting PI fluorescence. Taking a treatment time of
90 s as a base for comparison, it is clear that according to D-values obtained from heat
inactivation trials at these three temperatures (Tab. 2) 99.9999% of the initial population
could not resume growth on agar.
Table 2
Inactivation rates (k) and decimal reduction time (D-value) of L. rhamnosus GG treated in sterile 50
mM phosphate-buffered saline (pH 7.0) at different temperatures. Data were calculated from two or
more replicate heat inactivation experiments.
Treatment temperature (°C) k ± SD (s-1) R2 ± SD D- value = 1/k (s)
60 0.061 ± 0.0068 0.978 ± 0.016 16.3
68 0.205 ± 0.071 0.949 ± 0.072 4.9
75 0.325 ± 0.148 0.951 ± 0.064 3.1
R2 : correlation coefficient
SD : Standard deviation
The findings demonstrates that the response of the heat killed cells towards PI labelling
could be differentiated according to the temperature levels. Furthermore, the difference in the
PI staining capacity might reflect the difference in target sites of heat inactivation depending
on the temperature level used. The possibility to inactivate intracellular esterase without
seriously compromising cellular membrane might give evidence of an alternative inactivation
pathway preceding or apart from heat-induced membrane damage. Inactivation of esterase,
which is not involved in the maintenance of viability, is therefore indicative for a substantial
Flow cytometric analysis for inactivation studies 50
defect in an unknown, heat-sensitive cellular component, which ultimately led to cell death. In
addition, this result also emphasizes the general importance of exposing vegetative cells to
temperatures above 60°C, when the inactivation of intracellular esterase and membrane
damage were primarily aimed.
a
b
c
0 30 60 90 120
0
20
40
60
80
100
Percentage of cells in gate #1 (%)
Treatment time (sec)
d
Figure 6
Fluorescence density plots of cFDA/PI labeled L. rhamnosus GG cells after exposure to 60°C (a),
68°C (b) and 75°C (c) for 90 s. The figures (in %) following the quadrant number are associated with
the percentage of the cells in the corresponding quadrant.
(d) Kinetics of the increase in the number of PI labeled population (cells framed in quadrant #1) of L.
rhamnosus GG after heat treatment at 60°C (), 68°C (z), and 75°C (▲). The results are means
based on data from three or more independent experiments with error bars indicating standard
deviations.
Another cellular constituent which was reported to be implicated in thermal induced cell
inactivation is ribosome [2]. A strong relationship was observed between thermal death of
bacteria and the first major peak in DSC thermograms (temperature between 60 to 80°C)
which is attributed to ribosomal melting [4]. Viability loss of L. plantarum cells occurred when
the microorganisms were subjected to heat treatment in the range of 55 to 70°C for 60 s prior
to DSC scan [105]. The viability loss is related to the irreversible change of apparent
enthalpy and in the shapes or position of peak temperatures associated with ribosome
subunits, which was much more evident the higher the applied challenge temperature was.
The authors also found no apparent influence on the thermally induced transitions of other
cellular structures. This findings suggest that heat induced cell death can possibly occur in
absence of membrane deterioration due to ribosomal denaturation. In the latter work, the
Flow cytometric analysis for inactivation studies 51
temperature range, where ribosome denaturation was detected by DSC measurement, was
found qualitatively similar to the treatment temperature used in this study (60°C), in which
cells of LGG were inactivated without significant loss of membrane integrity.
In contrast, intracellular DNA denaturation was most likely not involved in the thermal
inactivation at temperatures evaluated in this study, i.e. between 60 and 75°C. According to
the results of previous thermal analysis studies irreversible denaturation of cellular DNA
required temperatures well above the temperature of cell inactivation [106]. Irreversible DNA
denaturation on L. plantarum was only observed after the cells were preheated to 100°C
[105]. At temperatures that cause ribosome denaturation, the DNA transition is reversible
[107].
2.3.3 Inactivation mechanisms by high hydrostatic pressure
Fluorescence properties of pressure treated LGG in response to cFDA/PI staining
It is shown in Figure 7b, 7c and 7d that cells exposed for 10 min to 200, 400, and 600 MPa
still possessed residual esterase activity, respectively. This is well documented by the
presence of the greater part of pressure-treated population in quadrant #2 and #4, in which –
similar to untreated sample (Fig. 7a) – cells with high cF fluorescence and thus high green
fluorescence value were encountered. The cells accumulating cF are further separated into
two sub-populations: cells solely labelled by cF in quadrant #4 and a sub-population in
quadrant #2, where in consequence of a certain intensity of the applied pressure treatment,
part of the population was double stained by cF and PI. This unique sub-population was
observed especially following pressure treatment at pressures beyond 400 MPa (Fig. 7c and
7d). Furthermore, another sub-population which exhibited only PI fluorescence (located in
quadrant #1) was found upon application of pressures in excess of 400 MPa.
Flow cytometric analysis for inactivation studies 52
a
e
bc
fg
d
h
Figure 7
Flow cytometric fluorescence dot-plots of L. rhamnosus GG treated with various pressure levels in
response to cFDA/PI staining (upper figures, a: control, b: 200 MPa, c: 400 MPa, d: 600 MPa) and
after glucose energization assay to determine the activity of the cells in extruding the intracellular
accumulated cF (lower figures, e: control, f: 200 MPa, g: 400 MPa, h: 600 MPa ).
As already mentioned before, at pressures higher than 400 MPa three sub-populations could
be identified based on their differential uptake of cF and PI: cF-stained (quadrant #4), cF and
PI double stained (quadrant #2), and PI stained population (quadrant #1). In the special case
of the double stained population, it was reported that the occurrence of a sub-population with
this unique staining characteristics was also observed in bile salt stressed bifidobacterial
cells and in ethanol stressed malolactic starter cultures [91, 93, 108]. Double stained
population could be regarded as an intermediate state of membrane damage. The cell
membranes of these cells seemed to be irreversibly damaged to a low extent, under which
relatively big molecule PI (molecular weight: 668 g·mol-1) could get into the cells, whereas
simultaneously cF was still intracellularly retained. The fact that cF (molecular weight: 376
g·mol-1) which is far smaller than PI did not passively diffuse out of the cells, is surprising.
Compared to the original non-fluorogenic substrate cFDA (molecular weight: 460 g·mol-1),
which is moderately permeant to cell membrane, the presence of additional negative charges
on cF at physiological pH may possibly inhibit its leakage out of the cells [54, 67], unless a
certain degree of membrane degradation was exceeded. Strong intracellular binding of cF
was documented by the retention of cF in high pressure killed cells incubated for 120 min at
37°C (Fig. 10).
The occurrence of double stained cell population indicated the presence of a threshold
pressure level for irreversible membrane rupture. In excess of this critical level, which
obviously settled at 200 MPa and which is marked by a drastic increase of the percentage of
Flow cytometric analysis for inactivation studies 53
PI labelled cells (Fig. 8d), PI could penetrate some cells, while for some cells cF could still be
retained but for some other not. The threshold level seemed to be non-homogeneously
distributed within the population, since major part of the population did not accumulate PI and
consequently was still encountered in quadrant #4 (Fig. 7). Up to 600 MPa only a small part
of the whole population was stained by PI (Fig. 8d); thus membrane rupture occurred only on
the most sensitive populations.
0 100 200 300 400 500 600
0
20
40
60
80
100
cF-extrusion activity (%)
Pressure (MPa)
0 100 200 300 400 500 600
-8
-6
-4
-2
0
log N/N0 (-)
0 100 200 300 400 500 600
60
70
80
90
100
110
cF accumulation activity (%)
Pressure (MPa)
0 100 200 300 400 500 600
0
5
10
15
20
25
30
Percentage of PI stained cells (%)
Pressure (MPa)
ab
cd
Figure 8
Impact of pressure treatment at 37°C for 10 min on several viability indicators of L. rhamnosus GG:
survival ratio as determined by plate count on MRS agar (¡), and metabolic activities as derived from
flow cytometric analysis including cF-accumulation activity (), cF-extrusion activity (T) and number
of cells with damaged membrane (z). The calculation of metabolic activities was based on the
analysis of fluorescence density plots in Fig. 6.
Regarding the viability status of this double stained population, sorting of the population of
stressed B. lactis, which were double stained with cFDA and PI, showed that a significant
part could resume growth on plates after resuscitation [91]. This physiological status was
considered as a transient phase in the progressive change towards cell dying, however,
death is not irreversible and double stained cells may still recover [51]
An alternative view to explain the cause of double stained population was suggested by
Shapiro (2001), who speculated that this particular population was resulted from clumps
Flow cytometric analysis for inactivation studies 54
containing bacteria with opposing physiological status [53]. Some events associated with
both high TO-PRO-3 fluorescence and high DiIC1(3) fluorescence, which are nucleic acid
stain and membrane potential stain, respectively might represent clumps containing viable
and dead cells of S. aureus. The viable cells have normal membrane potential and do not
take up TO-PRO-3. The dead, membrane-damaged cells have zero membrane potential and
do take up TO-PRO-3.
Furthermore, flow cytometric analysis combined with cFDA/PI staining strategy was also
applied to evaluate the effect of high pressure on Listeria innocua at low temperatures, in
which phase change between water and different ice modifications might occur [109].
Following pressurization at 300 MPa in liquid state (0°C), with which reduction by 3 log
cycles was achieved, three different sub-populations, i.e. cF+PI-, cF+PI+, and cF-PI+, as
encountered in the present study, were observed. In contrast, when phase change was
induced (ice I to ice II) by means of pressurization at 300 MPa and –45°C the majority of the
population was solely stained by PI. The difference in the cF/PI uptake properties might
indicate different mode of action of pressure on bacterial cells depending on the physical
state of water.
Comparison of viability status as determined by cellular response to cFDA/PI labelling
and standard culturability assay
Based on these fluorescence density plots in response to cF/PI uptake, the effect of different
levels of pressure on cF accumulation activity could be further quantified and compared with
standard viability assessment, i.e. culturing method on MRS agar (Fig. 8a and 8b).
The results of the latter method showed that pressures lower than 400 MPa were considered
to be non-lethal, whereas treatment at pressures higher than 500 MPa led to bacterial
reduction by more than 7 log cycles (Fig. 8a). In contrast, it was observed that even after
pressurization at pressures higher than 500 MPa, cells which were scored as dead by
cultivation method still showed a high level cF accumulation capacity similar to untreated
sample (Fig. 8b). These observations suggested that the cF accumulation capacity of the cell
did not correlate with culturability, i.e. ability to reproduce and form visible colonies on MRS
agar. Apparently the retention of intracellular esterase activity was not crucial in the
maintenance of reproducibility/culturabilty. This result is further supported by other studies
revealing considerable esterase functionality in cells killed by H2O2, γ-irradiation, and heat
[48, 75, 79].
As previously mentioned, upon exceeding a threshold pressure level of 200 MPa bacterial
membranes were compromised, thus leading to PI labelling of parts of the treated population.
Nevertheless, the disruption of membranes alone could not be made responsible for loss of
reproductive capacity, since upon pressure treatments at 300 and 400 MPa no significant
Flow cytometric analysis for inactivation studies 55
inactivation (Fig. 8a) was achieved although the magnitude of membrane ruptures at these
pressures, i.e. the total percentage of PI stained cells, was as high as the one at higher,
lethal pressure levels (Fig. 8d). Consequently, exclusion of PI from cell or the absence of PI
labelling could not give profound evidence about their reproductive capacity.
Likewise, it was reported that pressure treated cells of L. plantarum, which were not
recoverable by plate counts, were not stained by PI either [15]. Membrane damage was
observed with the PI assay only for treatments resulting in greater than 5-log reductions of
viable cells. The authors concluded that membrane damage owing to pressure treatment as
determined with PI staining was observed later than cell death. In line with the latter
observation a study which compared the type of cellular damage induced by high pressure at
different growth phases showed that the inactivation of stationary-phase cells by high
pressure may occur by a mechanism other than the permanent loss of membrane integrity
[39]. The authors observed high pressure induced morphological and physiological changes
in E. coli cells using lipophilic dye, which preferentially binds to cytoplasmic membrane. It
was found that while in exponential-phase cells the loss of viability is always accompanied by
a loss of the physical integrity of the membrane, the cell membrane in stationary-phase cells
have a cytoplasmic membrane that is robust enough to withstand pressurization up to very
intense treatments (600 MPa) and thus remain physically intact. It was further reported, that
after high pressure treatment at 400 MPa, which was effective to kill L. monocytogenes and
S. aureus by 6 to 7 log cycles, only ca. 75% of the population was positively labelled by PI
[110]. Similarly, the fact that pressure inactivation is not based on membrane rupture is
supported by the data on flow cytometric analysis of stationary phase cell of L.
monocytogenes, which were fully inactivated using pressure of 400 MPa [111]. The authors
demonstrated that following PI labelling there was a significant sub-population which did not
take up PI, suggesting that their membranes were not seriously damaged, whereas another
part of the population appeared to have been stained by PI, like heat killed cells (121°C, 15
min). Thus, the response of cells towards PI labelling, which reflects their status of
membrane integrity, could not be regarded as a reliable indicator of cell viability [54]
Fluorescence properties of pressure inactivated cells in comparison to that of heat
killed cells – Difference in the inactivation pathway
In correlation to the results of classical culturing method, the fact that pressure killed cells
could still accumulate cF may put forward new considerations regarding the difference of
mechanism of cell inactivation by both heat and pressure. According to Figure 8a, pressures
as high as 500 or 600 MPa at 37°C gave approximately similar plate count results, i.e. more
than 7 log-cycle reduction (Fig. 8a), as heat inactivation at 75°C in ambient atmosphere (Tab.
2). However, the inactivation pathway by pressure and heat differed pronouncedly, since
Flow cytometric analysis for inactivation studies 56
deviations were observed in the fluorescent profiles of the pressure (Fig. 7d) and heat killed
cells (Fig. 6b). Basically, the fluorescence density plot patterns demonstrated different
magnitude of membrane degradation as a result of exposure to these physical treatments.
For cells inactivated by heat it was shown that the membranes of all cells were ruptured,
leading to the occurrence of cells solely labelled by PI throughout the population; therefore
framed in quadrant #1 (Fig. 6b). In contrast, only as many as 22% of pressure inactivated
population was PI stained (Fig. 8d). This number is constituted of cells solely stained by PI
(quadrant #1) and cells double-stained with cF and PI (quadrant #2). This comparison study
revealed that that membrane integrity and esterase activity at the greater part of the pressure
inactivated cells were not completely diminished. Compared to degradative effect of heat on
cell membrane pressure induced cellular injury on bacteria, which ultimately lead to
inactivation, is therefore not governed by the loss of membrane permeability.
Activity of extruding intracellular accumulated cF in response to glucose addition
Apart from the ability to accumulate cF and to exclude PI, which characterize intact cells of
LGG (Fig. 4a), the ability to extrude accumulated cF upon energization using fermentable
sugar, could also be ascertained as an additional vitality marker in order to study the mode of
action of pressure on bacteria. This pump activity is most likely mediated by an ATP-driven
transport system, since ATP production and rapid extrusion of cF upon energizing was
observed despite dissipation of proton motive force by addition of ionophores valinomycin
and nigericin [77, 79].
At physiological pH, cF has predominantly a threefold negative charge and can thus be
considered practically membrane impermeable, or leaks very slowly [112]. An extensive
extrusion of intracellular cF upon energization with 20 mM glucose could be followed by
apparent shift of initially stained population from quadrant #4 (cF-stained) to quadrant #3
(unstained) owing to the loss of intracellular cF fluorescence (Fig. 7e to 7h). Based on the
extent of population shift after 20 min incubation in presence of glucose the degree of injury
on this cellular pump activity was able to be determined. Data on kinetic study of cF efflux on
L. lactis an incubation time of 20 min in the presence of lactose was regarded as sufficient to
effectively remove cF out of the cells [79].
This dye extrusion mechanism was found to be substrate specific, since upon addition of
lactose, the extrusion did not take place (Fig 11a). It has been evaluated before (Annex 1),
that LGG could not utilize lactose [102].
Flow cytometric analysis for inactivation studies 57
0 5 10 15 20
0
20
40
60
80
100
Percentage of cF-stained cells in quadrant #4 (%)
Incubation time at 37°C (min)
Figure 9
Kinetics of cF extrusion from L. rhamnosus GG cells, represented by the decrease in the percentage
of cF-stained population in the presence of 20 mM glucose. The performance of this metabolic activity
was measured on control population () and on cells treated at 200 MPa for 10 min at 37°C (▲).
It was observed that this performance of transporting cF out of the cells was more pressure
sensitive in contrast to cF accumulation activity. After pressurizing the cells at pressures up
to 200 MPa cells were able to survive and cF-efflux performance was as effective as control
sample at a fixed glucose incubation period of 20 min (Fig. 7e and 7f). Following Fig. 8, no
significant differences (p>0.05) could be observed between cells treated at 200 MPa and
control group regarding the reproductive capacity, esterase activity and cF-extrusion
performance. In order to profoundly elucidate cellular damage affected by incubation under
elevated pressure, cF-extrusion performance was not determined at steady-state level,
achievable after 20 min, but was continuously measured upon incubation at 37°C in
presence of glucose (Fig. 9). It was shown from the kinetics of the migration of cF-stained
cells from quadrant #4 to quadrant #3, that indeed the cF-extrusion of cells treated at 200
MPa was not as effective as untreated population, in which nearly all cells had already
extrude intracellular accumulated dye within the first 5 min, whereas in the case of the
pressure treated sample, the same level of dye extrusion was reached after 20 min.
According to culturing method, cells pressurized at 400 MPa were still able to recover on
agar with a slight loss of viability (log N/N0 < -0.5). In contrast, flow cytometric data showed
massive loss of efflux activity (Fig. 7g). Only 20% of initially cF-stained populations were
becoming unstained upon glucose addition and thus separated in quadrant #3 (Fig. 8c). The
results of these two viability assessments suggest, that this considerable disturbance of
extrusion activity might not be detrimental for the reproductive capacity of LGG. Alternatively,
the cells might be able to cope with or repair the defected and/or reduced functionality during
Flow cytometric analysis for inactivation studies 58
cultivation on MRS agar, thus not leaving it as a growth-limiting factor. When incubation
period at 37°C for glucose energization assay was extended up to 120 min, it was observed
that the percentage of population in quadrant #4 decreased, indicating that cells exposed to
400 MPa were still able to extrude intracellular accumulated cF (Fig. 10). The fact that
compared to untreated sample (Fig. 9) the cF efflux activity of cells treated with 400 MPa
was markedly reduced suggests that the occurrence of sub-lethal injury on this energy
dependent dye extruding system.
At higher pressures of 500 or 600 MPa, not more than 10% of cF stained population in
succeeded to transport intracellular accumulated cF across the cell membranes (Fig 8c).
Passive efflux of cF owing to leakage during incubation with glucose could be excluded,
since cF staining was still observed on cells previously inactivated by 600 MPa even after
incubating them in the presence of glucose for 120 min (Fig. 10a) or 24 h (data not shown).
According to plate count method cells treated with pressures higher than 500 MPa lost their
reproductive capacity/culturability (Fig. 8a). Consequently, cF-efflux activity and culturability
seemed to be strongly correlated at lethal pressure levels beyond 500 MPa. Cells which
completely lost the ability to extrude dye were not able to be cultured and thus scored as
dead, though still possessing enzymatic activity and intact membrane, which allow them to
accumulate cF. Thus, the lethal effect of pressure (at pressures higher than 400 MPa) could
was based on the irreversible perturbation of ATP-mediated dye extrusion system (Fig. 10b),
without of which cells could not resume growth even on nonselective media. An irreversible
rupture in this transport system did not allow the bacteria to reproduce and form colonies. In
line with this observation, it was also noted that in most cases pressure induced structural
changes in membrane proteins, in particular membrane-bound enzymes, generally occurred
at pressures > 400 MPa [113]. The fact that efflux of fluorescence compound can give
information about crucial transport mechanism located in membrane of cells of lactic acid
bacteria was shown upon using ethidium bromide as a reporter probe for HorA and LmrP
activity [15, 26].
Flow cytometric analysis for inactivation studies 59
0 20406080100120
0
10
20
30
40
50
60
70
80
90
400 MPa
600 MPa
Percentage of cF-stained cells in quadrant #4 (%)
Incubation time with glucose at 37°C (min)
a b
0 100 200 300 400 500 600
0
20
40
60
80
100
Glucose incubation 20 min
Glucose incubation 120 min
cF-extrusion activity (%)
Pressure (MPa)
Figure 10
(a) Kinetics of cF extrusion from high pressure treated L. rhamnosus GG cells, represented by the
decrease in the percentage of cF-stained population in the presence of 20 mM glucose during
incubation up to 120 min. Similar to previous pressure treatments cells were subjected to elevated
pressure conditions at 37°C for 10 min.
(b) Summary of pressure induced changes in cF extrusion activity of L. rhamnosus GG after glucose
incubation for 20 min or 120 min
Membrane-bound enzymes and cytoplasmic membrane are regarded as a major target for
pressure mediated sub-lethal injury or cell death [114, 115]. When bacterial cells are
subjected to pressure treatment, membrane-bound enzymes, such as HorA [15, 27], LmrP
[26] and F0F1-ATPase [23], are irreversibly inactivated prior to cell death. Defect in this
crucial transport mechanisms must not necessarily lead to cell death, if cells are transferred
to nonselective media, where they can initiate de novo protein synthesis or refolding of
integral membrane proteins [26]. However, sub-lethally injured cells were rapidly inactivated
in additional stresses exposed to them after pressure treatment due to the absence of
cellular homeostasis mechanism. Pressure treatment in range between 300 to 400 MPa,
which was identified as the appropriate pressure level in inducing sub-lethal injury to
pressure resistant mutants of enterohemorrhagic E. coli to resulted in an accelerated low-pH
inactivation (pH 3 to 4) during subsequent storage [116]. After treatment at a relatively
moderate pressure of 300 MPa, which sufficient to inactivate the hop resistance protein
HorA, sub-lethally injured L. plantarum cells could not survive in model beer containing 50
ppm hop extract [25]. With this regards, the authors of the latter work considered that
pressure inactivation is a two-stage process involving initial sub-lethal damage on
cytoplasmic membrane or membrane-associated transport systems in the first step, which is
then followed by cell death as a result of adverse environmental conditions.
Flow cytometric analysis for inactivation studies 60
Further characterization of a putative ATP-dependent transport mechanism on LGG in
response to glucose energization was observed upon exposing LGG to pH 3.0 at 37°C in the
presence or absence of glucose. Exposure of LGG to pH 3.0 was found to be non-lethal up
to an exposure time of 180 min (data not shown). According to the cF-fluorescence
histograms in Figure 11a, which basically shows the response of pH stressed cells towards
cFDA staining, cells which were incubated in the presence of glucose had higher cF
fluorescence values compared to the ones without glucose. The cF-fluorescence values of
acid stressed LGG were still low despite addition of lactose, which can not be utilized by this
strain. Since cF fluorescence intensity is reported to be highly pH dependent (Fig. 11b), this
findings may emphasize the role of an ATP-dependent proton extrusion system in
maintaining intracellular physiological pH, which in turn results in high intracellular
fluorescence intensity of cF. In absence of fermentable sugar proton gradient could not be
maintained so that intracellular pH value reached levels in the vicinity of extracellular pH
value, resulting in reduced cF fluorescence intensity of LGG. Exposure to pH 3.0 itself and
decrease of intracellular pH did not affect viability, but an irreversible impairment on this
proton translocating activity by means of pressure was reported to reduce the overall
physiological fitness of the cell, especially in their ability to restore ∆pH [17]. Apart from
inactivating ATP dependent dye extrusion system pressure has a deleterious effect on this
energy dependent proton extrusion machinery, which then restrict the pH range tolerated by
bacteria [117]. Previous work already showed the implication of pressure induced damage on
the proton translocating activity of F0F1 ATPase in Lactobacillus plantarum, leading to
impairment of acid efflux and maintenance of pH gradient. This event already took place
before cell death [23].
Flow cytometric analysis for inactivation studies 61
ab
0 200 400 600 800 1000
0
20
40
60
80
100
120
140
160
180
200
Number of events (-)
cF fluorescence intensity (-)
Figure 11
(a) Flow cytometry data on frequency distribution of the cF-fluorescence values of L. rhamnosus GG
before (T) and after exposure to pH 3 at 37°C for 1 h in the presence of glucose (), lactose (z)
or without any carbon source (S).
(b) pH-dependent fluorescence of carboxyfluorescein (cF) fluorophores. Fluorescence intensities
were measured for equal concentrations of the three dyes using excitation/emission at 490/520
nm. Source: http://www.probes.com/handbook/figures/0495.html
2.3.4 Combined application of heat and pressure
The heat stability of intracellular esterase under high pressures was assessed by
determining cell response toward cFDA/PI staining following pressure treatment at elevated
temperature, in which esterase already inactivated and membrane damage occurred. As
already discussed in Section 1.3.2. a temperature of 60°C was found to be effective in
selectively inactivating esterase without affecting membrane integrity. Combined application
of pressure and temperature processing was thought to have a synergistic effect on the
inactivation processes by specifically affecting different cellular target sites, i.e. esterase by
heat and membrane-bound enzymes by pressure.
In order to achieve constant temperature of 60°C throughout the whole pressurization phase
bath temperature was kept at 65°C. Assisted by this experimental setup adiabatic heating
generated during compression phase up to 600 MPa could elevate the temperature of the
cell suspension (initial sample temperature in the centre ~ 35°C) to 60°C (Fig. 12bb).
Pressure was released after a holding period of 300 s.
Flow cytometric analysis for inactivation studies 62
0 50 100 150 200 250 300
30
35
40
45
50
55
60
65
70
75
80
0
1000
2000
3000
4000
5000
6000
7000
Temperature (°C)
Time (s)
Pressure
Sample temperature
Pressure (MPa)
Bath temperature
ab
Figure 12
(a) Thermal behavior of water as a function of pressure and initial temperature which occurs during
compression
(b) Pressure and temperature profiles recorded in the sample (suspension of LGG in PBS buffer, pH
7.0) during the high pressure processing cycle
In contrast to the fluorescence pattern of LGG treated at 60°C in ambient pressure (Fig. 13a),
where due to esterase inactivation practically only cF-PI- population was identified, there is a
higher degree of heterogeneities in the response of LGG towards cFDA/PI labelling after
combined pressure-heat treatment (Fig 13b).
A striking evidence of the positive effect of heat treatment at elevated pressure is that the
major part of the treated population was PI stained. Membrane damage, which did not occur
on cells treated at 60°C and ambient pressure, could be induced in a significant proportion of
the population (single stained by PI or double stained population) by elevating treatment.
However, following pressurization at 600 MPa and 60°C a significant part of the LGG
population could still accumulate cF, either with or without PI uptake (Fig. 13b). Obviously, in
the cF accumulating fraction thermal damage of esterase and p-T induced degradation of
membrane took place in a reduced rate at higher pressure; resulting in the enzymatic
conversion of cFDA to cF and retention of the dye in the cell (single stained by cF or double
stained population).
Flow cytometric analysis for inactivation studies 63
ab
Figure 13
Fluorescence density plots of L. rhamnosus following cFDA/PI staining on (a) heat treated cells (60°C,
300 s) and (b) pressure treated cells (600 MPa, 60°C, 300 s).
In line with the latter result, it was notable that a significant number of proteins (DNA
polymerases, hydrogenases, etc.) shows improved heat stability under pressure [118]. It was
also found that upon elevating pressure the heat stability and in situ substrate conversion
rate of some glycolytic enzymes could be enhanced at temperatures, at which those
enzymes are degraded under ambient pressure [119]. Studying the properties of proteins
potentially adaptable to high temperature by means of pressure may further open the
possibility to identify processing or environmental conditions that allow heat sensitive
proteins to remain stable and microorganisms viable, so as to improve biotransformation
processes or to allow novel catalytic reactions to take place.
Furthermore, the results of high pressure inactivation at higher temperatures suggest that
under high pressure the heterogeneity of the cells in terms of the stability of the cellular
components are more pronounced and thus can be elucidated more effectively. Whereas at
ambient pressure the lethal effect of heat was experienced in a nearly uniform manner
throughout the population treated, the presence of more than one population differing in the
types of damage induced by pressure or by combined application of pressure and
temperature might open a discussion about the safety of the process, with which some of the
treated cells were still capable of accumulating certain – potentially problematic –
metabolites.
2.3.5 Inactivation mechanism by high-intensity ultrasound
The propagation of ultrasound waves in the liquid medium is always accompanied with heat
generation (Fig. 14). Since it is of importance to sufficiently exclude thermal effect in order to
exclusively assess ultrasound effect on bacteria, temperature of the medium was controlled
by placing the sample container in an ice bath. With help of this procedure heat dissipated
upon generation of ultrasound waves could be instantly removed; thus leaving only an initial
Flow cytometric analysis for inactivation studies 64
temperature rise of approximately 10°C. Temperature equilibrium was achieved after 2 min
and the temperature never exceeded 20°C (Fig. 14). Since the temperature never exceeded
20°C, thermal effect could be excluded, and the inactivation could exclusively be attributed to
ultrasound effect.
0 5 10 15 20
-2,6
-2,4
-2,2
-2,0
-1,8
-1,6
-1,4
-1,2
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
log N/N0 (-)
Treatment time (min)
Temperature (°C)
Figure 14
Ultrasound inactivation kinetics of Lactobacillus rhamnosus (∆) and Escherichia coli () in phosphate
buffer at pH 7.0 along with the temperature increase () in the treatment medium during propagation
of ultrasonic waves. Data are the means of three replicate ultrasound treatments; the error bars
represent the standard deviations of the mean.
It was evident that, the LGG was more resistant against lethal effect of ultrasound in
comparison to the E. coli (Fig. 14). When the decimal reduction time (D-value) of ultrasound
death kinetics was calculated, the D-value of LGG was more than two times higher than that
of E. coli (18.8 min and 8.3 min, respectively). Gram negative bacteria were known to be
more susceptible towards ultrasound compared to Gram positive ones [120, 121]. Gram
positive bacteria were less sensitive to ultrasound since they usually have a thicker and a
more tightly adherent layer of peptidoglycans than Gram negative bacteria [122].
Following exposure to ultrasound up to 20 min the major part of the LGG cells (more than
80%) were still encountered in quadrant #4 (Fig. 15) This fluorescence behaviour indicated
that the majority of the treated cells was still able to retain cF and was not stained by PI. In
this particular sub-population the integrity of cytoplasmic membrane was not seriously
affected by ultrasound. From Figure 16a it is also evident, that by applying ultrasound for 20
min only maximal 7% of the cell population showed membrane damage, which allowed them
to be labelled with PI. In contrast, according to the results of cultivation method only 8% of
Flow cytometric analysis for inactivation studies 65
the initial population could resume growth on agar following ultrasound propagation at the
same condition (Fig. 16a).
abcde
Figure 15
Fluorescence density plots of L. rhamnosus GG in response to staining with cFDA and PI after
ultrasound treatment at different exposure time. Duration of treatment was 0 min (a), 5 min (b), 10 min
(c), 15 min (d) or 20 min (e).
Bacterial membranes are considered as an ultimate requirement for the retention of viability.
Cells which irreversibly lost their membrane integrity could neither maintain any of the
electrochemical gradients necessary to remain functional nor have the potential to give rise
to metabolism or proliferation due to the absence of selective permeability [48]. Therefore,
such cells can be classified as dead cells, which lost their capacity to form colonies on agar
[64]. In agreement with this classification, the PI labelled cells, which are encountered in
quadrant #1, were considered as dead. However, the percentage of PI labelled cells (max.
7% of the whole population) are far below the magnitude of cells not able to grow on agar
(92% of the initial population lost their growth capacity on agar) following ultrasound
treatment. Thus, the metabolically active population in quadrant #4 (80% of the whole
population) was most likely constituted of cells able to resume growth on MRS agar (viable)
and the ones not able to resume growth (dead); with dead cells as the major constituent.
These findings led to believe, that due to ultrasound effect cell death could even occur
without any severe damage of membranes, since despite of the absence of growth on agar,
the major population of ultrasound killed cells was still emitting green fluorescence as a
consequence of the retention of their enzymatic activity and membrane integrity. Moreover,
the presence of at least two sub-populations with regards to their fluorescence behaviour
after ultrasound treatment indicated the presence of heterogeneity within the population in
their capacity of resisting the deteriorating impact of ultrasound.
Flow cytometric analysis for inactivation studies 66
0 5 10 15 20 25
0
5
10
15
75
80
85
90
95
100
Percentage of cells in quadrants #1 () and #4 (%) (%)
Treatment time (min)
0 5 10 15 20 25
0
20
40
60
80
100
120
Survival rate, N/N0 (%)
Treatment time (min)
Figure 16
a. Changes in the percentage of Lactobacillus rhamnosus cells encountered in quadrant #1 (∆) and
#4 (▼) in response to ultrasound propagation. PI labelled cells are framed in quadrant #1,
whereas the ones accumulating cF are to be found in quadrant #4. Data for this figure were
derived from the density plots shown Figure 15. Data were means of three replicates of ultrasound
treatments; error bars indicate the standard deviations of the mean.
b. Kinetic of viability loss upon ultrasound exposure ({) as determined by plating cells on MRS agar.
Data were means of at least two replicates of ultrasound treatments; error bars indicate the
standard deviations of the mean.
In conclusion, the major action site of ultrasound on inducing lethal effect was not necessarily
the cytoplasmic membrane, since it was observed, that the majority of the ultrasound treated
population was still able to accumulate cF and did not allow PI penetration. Thus, the
degradative effect of ultrasound on the cytoplasmic membrane was less pronounced. This
observation is in contrast to the suggestion found in the literature, which proposed that the
target of ultrasound damage might be the inner (cytoplasmic) membrane consisting of a
lipoprotein bilayer [122]. Although the power input of the ultrasound applied with regards to
bactericidal efficacy was sufficient, as assessed previously [43], the treatment temperature
(below 20°C) applied was possibly too low to synergistically induce membrane deterioration.
Cell death which was observed upon applying high-intensity ultrasound seemed to result
from non-membrane related degradation.
2.4 Conclusion
The present work deals with the application of flow cytometric analysis to evaluate the
mechanism of microbial inactivation with LGG as model organism by means of physical
Flow cytometric analysis for inactivation studies 67
treatments. A well-described multiple staining strategy, which is composed of physiological
dyes carboxyfluoresceindiacetate (cFDA) and propidium iodide (PI) was applied to examine
specific cellular metabolic activities and their relative changes following inactivation
treatments. It was expected that additional insights on process-induced changes in cellular
integrity or metabolic activities, which were not explicitly assessable by culture techniques,
could be achieved using this measurement technique. Furthermore, it is also noteworthy to
differentiate the mechanisms of microbial inactivation occurred during different treatments, in
order to allow problematic contaminants to be injured or inactivated more effectively as well
as to effectively combine different treatments, which have different cellular target sites.
Figure 17 shows the fluorescence density plots of LGG after being exposed to thermal
treatment (Fig. 17b and 17c), supercritical CO2 treatment (Fig. 17d), pulsed electric fields
treatment (Fig. 17e), high-intensity ultrasound treatment (Fig. 17f), high pressure treatment at
moderate temperature (Fig. 17g), and high pressure treatment at elevated temperature (Fig.
17h). From this overview it is obvious that different treatments led to different response of the
cell to cFDA/PI labelling, indicating that the applied treatments differed in the cellular sites
being primarily affected, although the survival rates according to the plate counts result were
in the same range.
Regarding the heat induced damage on LGG, when these cells were exposed to 60°C at
ambient pressure the cF-accumulation activity was considerably reduced without significant
loss of membrane intactness. This particular physiological state left cells stained neither with
cF nor with PI, even when the cells were heat treated up to 300 s. Thermal-induced death
could therefore be achieved in absence of membrane degradation. In contrast, when cells
were subjected to temperatures above 65°C, PI uptake already occurred in the first 90 s.
From this staining behaviour it was concluded that at higher temperatures the primary target
of lethal effect of heat is the bacterial cytoplasmic membrane. The findings demonstrates the
differences in the target sites of heat inactivation depending on the temperature level used.
Furthermore, this result also emphasizes the general importance of treating vegetative cells
at temperatures above 60°C, when the inactivation of intracellular esterase and membrane
damage were aimed.
Flow cytometric analysis for inactivation studies 68
d
abc
efgh
Figure 17
Flow cytometric fluorescence density plots of L. rhamnosus GG subjected to different treatment
methods: a) untreated sample (log N/N0 = 0); b) heat treatment at 60°C for 300 s (log N/N0 = -6.8); c)
heat treatment 75°C for 30 s (log N/N0 = -7.5), d) supercritical CO2 treatment at 40 MPa and 40°C for
10 min (log N/N0 = -7.3); e) pulsed electric fields treatment with field strength of 35 kV/cm and specific
energy input of 300 kJ/kg at 30°C (log N/N0 = -1.2); f) high-intensity ultrasound 17.6 W for 20 min (log
N/N0 = -1.2); high pressure treatment at 600 MPa and 37°C for 10 min (log N/N0 = -7.3) and high
pressure treatment at 600 MPa and 60°C for 5 min (log N/N0 ~ -7). Treatment media were phosphate
buffer saline (pH 7.0)
In contrast to heat killed cells which resulted in homogeneous population with regards to their
response to cFDA/PI staining, three populations differing in their behaviour upon cFDA/PI
uptake were observed when high pressure killed cells were analysed. The major population
of high pressure inactivated cells of LGG could accumulate fluorescent molecule
carboxyfluorescein (cF), which indicated that some of the dead cells were still enzymatically
active and not severely membrane compromised. The fact, that pressure inactivated bacteria
could perform enzymatic conversion of cFDA into cF needs further attention, since the
presence of such metabolically active, but dead bacteria in food might be critical in terms of
their potential activity on excreting toxic or food spoiling metabolites. It is also obvious that
the heterogeneity among the population of pressure killed cells towards cFDA/PI staining is
more pronounced compared to heat killed cells. This findings suggested that the distribution
in the tolerance towards deteriorative effect of pressure on membrane or intracellular enzyme
– which in turn might reflect the overall resistance of the organism and their capacity of
Flow cytometric analysis for inactivation studies 69
repairing the imposed injuries – might be more pronounced in comparison to the
corresponding distribution against lethal effect of heat.
Moreover, according to plate count method pressures up to 400 MPa was regarded as non-
lethal on LGG. However, below this threshold level differences observed in the physiological
activity of pressure-treated cells, especially in the reduced rate of dye extrusion upon glucose
energization, indicated that they were sub-lethally injured. Pressure induced sub-lethal injury,
which was observable up to 400 MPa, seemed to correlate more profoundly with perturbation
of cellular transport mechanism rather than with the occurrence of double stained population,
since the diffusion of PI in cF stained cells not only occurred on cells lethally injured by
pressure as high as 600 MPa but also on cells surviving pressure treatment (for instance
after treatment at 300 MPa).
Measurement of cF-extrusion activity in response to glucose energization indicated that the
lethal effect of pressure (at pressures higher than 400 MPa) was related to the irreversible
perturbation of dye extrusion machinery, which is most likely mediated by an ATP-driven
transport system. Accordingly, dead cells, which lost the capacity to reproduce themselves
and grow on agar, are the one which were not able to extrude cF, although the membrane
was still intact and esterase remained active. This findings underlines the results from
previous works on pressure induced damage on other ATP-dependent, membrane bound
enzymes, which are crucial in maintaining viability [15, 17, 23, 26, 27].
Cell death occurred upon exposure of high intensity ultrasound under continuous removal of
generated heat to LGG. In absence of thermal effect it could be shown that only a small
population was labelled by propidium iodide (PI) following exposure to ultrasound up to 20
min. Within the experimental conditions investigated ultrasound did not considerably affect
the cytoplasmic membrane, although according to plate count results viability loss occurred.
It could be concluded that, cell death which was observed upon applying high-intensity
ultrasound seemed to result from non-membrane related degradation.
Taken together the results of flow cytometric measurement on cFDA/PI labelled bacteria
following exposure to heat, pressure and ultrasound one could elucidate the mode of action
of these physical stressors on cellular activities or integrity. Ultimately, the use of this
technique might lead to an improved design of inactivation processes, i.e. one could then
choose a certain type of cellular damage preferred and then select the type of treatments
required to achieve this goal. Tailor-made non-viable bacteria might be an interesting
research object in the field of probiotic research, where the importance of the ingestion of
viable bacteria in eliciting health effect is sometimes questioned, since non-viable bacteria
were reported to be effective as well [123-125]. Non-viable probiotic cells are of interest due
to easy-handling and longer shelf life [126]. However, systematic studies on the relationship
between probiotic effect and the type of inactivation treatments used to produce non-viable
Flow cytometric analysis for inactivation studies 70
bacteria are still lacking. In this context, high pressure killed cells might be one of the
promising candidate to be investigated, since the fluorescence pattern of pressure
inactivated cells – which is indicative for a lower extent of damage on metabolic activity and
on membrane – is quite similar to the one of viable cells.
2.5 References
1. Somero, G.N. 1992. Adaptations to high hydrostatic pressure. Annual Reviews in Physiology. 54: 557-
577.
2. Abee, T. and Wouters, J.A. 1999. Microbial stress response in minimal processing. International
Journal of Food Microbiology. 50: 65-91.
3. Earnshaw, R.G., Appleyard, J., and Hurst, R.M. 1995. Understanding physical incativation processes:
combined preservation opportunities using heat, ultrasound and pressure. International Journal of Food
Microbiology. 28: 197-219.
4. Teixeira, P., Castro, H., Mohácsi-Farkas, C., and Kirby, R. 1997. Identification of sites of injury in
Lactobacillus bulgaricus during heat stress. Journal of Applied Microbiology. 83: 219-226.
5. Silva, M.T. and Sousa, J.C.F. 1972. Ultrastructural alterations induced by moist heat in Bacillus cereus.
Applied Microbiology. 24: 463-476.
6. Hurst, A. 1984. Reversible heat damage, in Repairable lesions in microorganisms, Hurst, A. and Nasim,
A., Editors. Academic Press, Ltd.: London.
7. Cheftel, J.C. 1995. Review: High pressure, microbial inactivation and food preservation. Food Science
and Technology International. 1: 75-90.
8. Palou, E., López-Malo, A., Barbosa-Cánovas, G.V., and Swanson, B.G. 1999. High-pressure
treatment in food preservation, in Handbook of food preservation, Rahman, M.S., Editor. Marcel Dekker,
Inc.: New York. p. 533-576.
9. Knorr, D. and Heinz, V. 2001. Development of nonthermal methods for microbial control, in Disinfection,
sterilization, and preservation, Block, S.S., Editor. Lippincott Williams&Wilkins: Philadelphia. p. 853-877.
10. Heinz, V. and Knorr, D. 2001. Effects of high pressure on spores, in Ultrahigh pressure treatment of
foods, Knorr, M.L.G.H.a.D., Editor. Aspen Publication: Gaithersburg. p. 77-113.
11. Cheftel, J.-C. 1992. Effects of high hydrostatic pressure on food constituents : an overview, in High
pressure and biotechnology, Balny, C., et al., Editors. Colloque INSERM/John Libbey: London. p. 195-
209.
12. Stute, R., Klingler, R.W., Boguslawski, S., Esthiaghi, M.N., and Knorr, D. 1996. Effects of high
pressures treatment on starches. Starch. 48: 399-408.
13. Knorr, D., Schlüter, O., and Heinz, V. 1998. Impact of high hydrostatic pressure on phase transitions of
foods. Food Technology. 52: 42-45.
14. Benito, A., Ventoura, G., Casadei, M., Robinson, T., and Mackey, B. 1999. Variation in resistance to
natural isolates of Escherichia coli O157 to high hydrostatic pressure. mild heat, and other stresses.
Applied and Environmental Microbiology. 65: 1564-1569.
15. Ulmer, H.M., Gänzle, M.G., and Vogel, R.F. 2000. Effects of high pressure on survival and metabolic
activity of Lactobacillus plantarum TMW1.460. Applied and Environmental Microbiology. 66: 3966-3973.
16. De Angelis, M. and Gobetti, M. 2004. Environmental stress responses in Lactobacillus: A review.
Proteomics. 4: 106–122.
Flow cytometric analysis for inactivation studies 71
17. Molina-Gutierrez, A., Stippl, V., Delgado, A., Gänzle, M.G., and Vogel, R.F. 2002. In situ
determination of the intracellular pH of Lactococcus lactis and Lactobacillus plantarum during pressure
treatment. Applied and Environmental Microbiology. 68: 4399-4406.
18. van de Guchte, M., Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S.D., and Maguin, E. 2002.
Stress responses in lactic acid bacteria. Antonie van Leeuwenhoek. 82: 187-216.
19. Nannen, N.L. and Hutkins, R.W. 1991. Proton-translocating adenosine triphosphatase activity in lactic
acid bacteria. Journal of Dairy Science. 74: 747–751.
20. Poolman, B. 1993. Energy transduction in lactic acid bacteria. FEMS Microbiology Reviews. 12: 125–
147.
21. Matsumoto, M., Ohishi, H., and Benno, Y. 2004. H+-ATPase activity in Bifidobacterium with special
reference to acid tolerance. International Journal of Food Microbiology. 93: 109-113.
22. Marquis, R.E. and Bender, G.R. 1987. Barophysiology of prokaryotes and proton translocating
ATPases, in Current perspectives in high pressure biology, Jannasch, H.W., Marquis, R.E., and
Zimmerman, A.M., Editors. Academic Press Ltd.: London, United Kingdom.
23. Wouters, P.C., Glaasker, E., and Smelt, J.P.P.M. 1998. Effects of high pressure on inactivation kinetics
and events related to proton efflux in Lactobacillus plantarum. Applied and Environmental Microbiology.
64: 509-514.
24. Chang, G. 2003. Multidrug resistance ABC transporters. FEBS Letters. 555: 102-105.
25. Gänzle, M.G., Ulmer, H.M., and Vogel, R.F. 2001. High pressure inactivation of Lactobacillus plantarum
in a model beer system. Journal of Food Science. 66: 1174-1181.
26. Molina-Höppner, A., Doster, W., Vogel, R.F., and Gänzle, M.G. 2004. Protective effect of sucrose and
sodium chloride for Lactococcus lactis during sublethal and lethal high-pressure treatments. Applied and
Environmental Microbiology. 70: 2013-2020.
27. Ulmer, H.M., Herberhold, H., Fahsel, S., Gänzle, M.G., Winter, R., and Vogel, R.F. 2002. Effects of
pressure-induced membrane phase transitions on inactivation of HorA, an ATP-dependent multidrug
resistance transporter, in Lactobacillus plantarum. Applied and Environmental Microbiology. 68: 1088-
1095.
28. Chong, P.L.G., Fortes, P.A.G., and Jameson, D.M. 1985. Mechanism of inhibition of (Na, K) - ATPase
by hydrostatic pressure studied with fluorescent probes. The Journal of Biological Chemistry. 260:
14484-14490.
29. Janosch, S., Kinne-Saffran, E., Kinne, R.K.H., and Winter, R. 2003. Inhibition of Na+,K+-ATPase by
hydrostatic pressure, in Advances in High Pressure Bioscience and Biotechnology II, Winter, R., Editor.
Springer Verlag: Berlin. p. 215-219.
30. Kato, M., Hayashi, R., Tsuda, T., and Taniguchi, K. 2002. High pressure-induced changes of biological
membrane - Study on the membrane-bound Na+/K+-ATPase as a model system. European Journal of
Biochemistry. 269: 110-118.
31. Casadei, M.A., Manas, P., Niven, G., Needs, E., and Mackey, B.M. 2002. Role of membrane fluidity in
pressure resistance of Escherichia coli NCTC 8164. Applied and Environmental Microbiology. 68: 5965–
5972.
32. Niven, G.W., Miles, C.A., and Mackey, B.M. 1999. The effects of hydrostatic pressure on ribosome
conformation in Escherichia coli: an in vivo study using differential scanning calorimetry. Microbiology.
145: 419–425.
33. Kaletunc, G., Lee, J., Alpas, H., and Bozoglu, F. 2004. Evaluation of structural changes induced by
high hydrostatic pressure in Leuconostoc mesenteroides. Applied and Environmental Microbiology. 70:
1116-1122.
Flow cytometric analysis for inactivation studies 72
34. Simpson, R.K. and Gilmour, A. 1997. The effect of high hydrostatic pressure on the activity of
intracellular enzymes of Listeria monocytogenes. Letters in Applied Microbiology. 25: 48-53.
35. Ludwig, H., Scigalla, W., and Sojka, B. 1996. Pressure and temperature inactivation of
microorganisms, in High Pressure Effects in Molecular Biophysics, Markley, J.L., Northrop, D.B., and
Royer, C.A., Editors. Oxford University Press: New York.
36. Perrier-Cornet, J.M., Tapin, S., Gaeta, S., and Gervais, P. 2005. High-pressure inactivation of
Saccharomyces cerevisiae and Lactobacillus plantarum at subzero temperatures. Journal of
Biotechnology. 115: 405–412.
37. Kalchayanand, N., Dunne, P., Sikes, A., and Ray, B. 2004. Viability loss and morphology change of
foodborne pathogens following exposure to hydrostatic pressures in the presence and absence of
bacteriocins. International Journal of Food Microbiology. 91: 91-98.
38. Malone, A.S., Shellhammer, T.H., and Courtney, P.D. 2002. Effects of high pressure on the viability,
morphology, lysis, and cell wall hydrolase activity of Lactococcus lactis subsp. cremoris. Applied and
Environmental Microbiology. 68: 4357-4363.
39. Manas, P. and Mackey, B.M. 2004. Morphological and physiological changes induced by high
hydrostatic pressure in exponential- and stationary-phase cells of Escherichia coli: Relationship with cell
death. Applied and Environmental Microbiology. 70: 1545-1554.
40. Piyasena, P., Mohareb, E., and McKellar, R.C. 2003. Inactivation of microbes using ultrasound: a
review. International Journal of Food Microbiology. 87: 207-216.
41. Ordoñez, J.A., Aguilera, M.A., Garcia, M.L., and Sanz, B. 1987. Effect of combined ultrasonic and
heat treatment (thermoultrasonication) on the survival of a strain of Staphylococcus aureus. Journal of
Dairy Research. 54: 61-67.
42. Garcia, M.L., Burgos, J., Sanz, B., and Ordonez, J.A. 1989. Effect of heat and ultrasonic waves on the
survival of two strains of Bacillus subtilis. Journal of Applied Bacteriology. 67: 619-628.
43. Zenker, M., Heinz, V., and Knorr, D. 2003. Application of ultrasound-assisted thermal processing for
preservation and quality retention of liquid foods. Journal of Food Protection. 66: 1642-1649.
44. Sala, F.J., Burgos, J., Condón, S., López, P., and Raso, J. 1995. Effect of heat and ultrasound on
microorganisms and enzymes, in New methods of food preservation, Gould, G.W., Editor. Blackie
Academic & Professional: London. p. 176-204.
45. Pagán, R., Mañas, P., Raso, J., and Condón, S. 1999. Bacterial resistance to ultrasonic waves under
pressure at nonlethal (manosonication) and lethal (manothermosonication) temperatures. Applied and
Environmental Microbiology. 65: 297-300.
46. Sams, A.R. and Feria, R. 1991. Microbial effects of ultrasonication of broiler drumstick skin. Journal of
Food Science. 56: 247-248.
47. Nebe von Caron, G., Stephens, P., and Badley, R.A. 1998. Assessment of bacterial viability status by
flow cytometry and single cell sorting. Journal of Applied Microbiology. 84.
48. Vives-Rego, J., Lebaron, P., and Nebe-von Caron, G. 2000. Current and future applications of flow
cytometry in aquatic microbiology. FEMS Microbiology Reviews. 24: 429-448.
49. Davey, H.M., Davey, C.L., and Kell, D.B. 1993. On the determination of the size of microbial cells using
flow cytometry, in Flow cytometry in microbiology, Lloyd, D., Editor. Springer-Verlag: London. p. 49-65.
50. Davey, H.M. and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogeneous microbial
populations: The impartance of single cell analyses. Microbiological Reviews. 60: 641-696.
51. Bunthof, C.J. 2002. Flow cytometry, fluorescent probes, and flashing bacteria, in Department of Agro-
Technology and Food Sciences. Wageningen University: Wageningen. p. 160.
52. Ben-Amor, K. 2004. Microbial eco-physiology of the human intestinal tract: a flow cytometric approach,
in Department of Agro-Technology and Food Sciences. Wageningen University: Wageningen. p. 166.
Flow cytometric analysis for inactivation studies 73
53. Shapiro, H.M. 2001. Multiparameter flow cytometry of bacteria: Implications for diagnostics and
therapeutics. Cytometry. 43: 223-226.
54. Breeuwer, P. and Abee, T. 2000. Assessment of viability of microorganisms employing fluorescence
techniques. International Journal of Food Microbiology. 55: 193-200.
55. Barer, M.R. and Harwood, C.R. 1999. Bacterial viability and culturability. Advances in Microbial
Physiology. 41: 93-137.
56. Hewitt, C.J. and Nebe-von-Caron, G. 2004. The application of multi-parameter flow cytometry to
monitor individual microbial cell physiological state. Advances in Biochemical Engineering/Biotechnology.
89: 197-223.
57. Roszak, D.B. and Colwell, R.R. 1987. Survival strategies of bacteria in the natural environment.
Archives in Microbiology. 141: 348-352.
58. Diaper, J.P. and Edwards, C. 1994. The use of fluorogenic esters to detect viable bacteria by flow
cytometry. Journal of Applied Bacteriology. 77: 221-228.
59. Porter, J., Edwards, C., and Pickup, R.W. 1995. Rapid assessment of physiological status in
Escherichia coli using fluorescent probes. Journal of Applied Bacteriology. 79: 399-408.
60. Davey, H.M., Jones, A., Shaw, A.D., and Kell, D.B. 1999. Variable selection and multivariate methods
for the identification of microorganisms by flow cytometry. Cytometry. 35: 162-168.
61. Yamaguchi, N. and Nasu, M. 1997. Flow cytometric analysis of bacterial respiratory and enzymatic
activity in the natural aquatic environment. Journal of Applied Microbiology. 83: 43-52.
62. Hoefel, D., Grooby, W.L., Monis, P.T., Andrews, S., and Saint, C.P. 2003b. Enumeration of water-
borne bacteria using viability assays and flow cytometry: a comparison to culture-based techniques.
Journal of Microbiological Methods. 55: 585-597.
63. Keer, J.T. and Birch, L. 2003. Molecular methods for the assessment of bacterial viability. Journal
Microbiological Methods. 53: 175-183.
64. Nebe von Caron, G., Stephens, P.J., Hewitt, C.J., Powell, J.R., and Badley, R.A. 2000. Analysis of
bacterial function by multi-colour fluorescence flow cytometry and single cell sorting. Journal of
Microbiological Methods. 42: 97-114.
65. Kell, D.B., Ryder, H.M., Kaprelyants, A.S., and Westerhoff, H.V. 1991. Quantifying heterogeneity:
Flow cytometry of bacterial cultures. Antonie van Leeuwenhoek. 60: 145-158.
66. Ueckert, J., Breeuwer, P., Abee, T., Stephens, P., von Caron, G.N., and ter Steeg, P.F. 1995. Flow
cytometry applications in physiological study and detection of foodborne microorganisms. International
Journal of Food Microbiology. 28: 317-326.
67. Haugland, R.P., Handbook of fluorescent probes and research products. 9th Edition ed. 2002, Eugene,
OR, USA: Molecular Probes, Inc.
68. Hoefel, D., Grooby, W.L., Monis, P.T., Andrews, S., and Saint, C.P. 2003a. A comparative study of
carboxyfluorescein diacetate and carboxyfluorescein diacetate succinimidyl ester as indicators of
bacterial activity. Journal of Microbiological Methods. 52: 379-388.
69. Jacobsen, C.N., Rasmussen, J., and Jakobsen, M. 1997. Viability staining and flow cytometric
detection of Listeria monocytogenes. Journal of Microbiological Methods. 28: 35-43.
70. Budde, B.B. and Rasch, M. 2001. A comparative study on the use of flow cytometry and colony forming
units for assessment of the antibacterial effect of bacteriocins. International Journal of Food
Microbiology. 63: 65-72.
71. Malacrino, P., Zapparoli, G., Torriani, S., and Dellaglio, F. 2001. Rapid detection of viable yeasts and
bacteria in wine by flow cytometry. Journal of Microbiological Methods. 45: 127-134.
Flow cytometric analysis for inactivation studies 74
72. Forster, S., Snape, J.R., Lappin-Scott, H.M., and Porter, J. 2002. Simultaneous fluorescent gram
staining and activity assessment of activated sludge bacteria. Applied and Environmental Microbiology.
68: 4772-4779.
73. Kaneshiro, E.S., Wyder, M.A., Wu, Y.P., and Cushion, M.T. 1993. Reliability of calcein acetoxy methyl
ester and ethidium homodimer or propidium iodide for viability assessment of microbes. Journal of
Microbiological Methods. 17: 1-16.
74. Molenaar, D., Abee, T., and Konings, W.N. 1991. Continuous measurement of the cytoplasmic pH in
Lactococcus lactis with a fluorescent pH indicator. Biochimica et Biophysica Acta. 1115: 75-83.
75. Breeuwer, P., Drocourt, J.L., Rombouts, F.M., and Abee, T. 1994. Energy-dependent, carrier-
mediated extrusion of carboxyfluorescein from Saccharomyces cerevisiae allows rapid assessment of
cell viability by flow cytometry. Applied and Environmental Microbiology. 60: 1467-1472.
76. Shapiro, H.M., Practical flow cytometry. 4th ed. 2003, Hoboken, New Jersey: John Wiley & Sons, Inc.
681.
77. Bunthof, C.J., Braak, S.v.d., Breeuwer, P., Rombouts, F.M., and Abee, T. 2000. Fluorescence
assessment of Lactococcus lactis viability. International Journal of Food Microbiology. 55: 291-294.
78. Molenaar, D., Bolhuis, H., Abee, T., Poolman, B., and Konings, W.N. 1992. The efflux of a fluorescent
probe is catalyzed by an ATP-Driven extrusion system in Lactococcus lactis. Journal of Bacteriology.
174: 3118-3124.
79. Bunthof, C.J., van den Braak, S., Breeuwer, P., Rombouts, F.M., and Abee, T. 1999. Rapid
fluorescence assessment of the viability of stressed Lactococcus lactis. Applied and Environmental
Microbiology. 65: 3681-3689.
80. Midgley, M. 1987. An efflux system for cationic dyes and related compounds in E coli. Microbiologal
Sciences. 4: 125–128.
81. Jepras, R.I., Carter, J., Pearson, S.C., Paul, F.E., and Wilkinson, M.J. 1995. Development of a robust
flow cytometric assay for determining numbers of viable bacteria. Applied and Environmental
Microbiology. 61: 2696–2701.
82. Roth, B.L., Poot, M., Yue, S.T., and Millard, P. 1997. Bacterial viability and antibiotic susceptibility
testing with SYTOX green nucleic acid stain. Applied and Environmental Microbiology. 63: 2421-2431.
83. Bunthof, C.J., Bloemen, K., Breeuwer, P., Rombouts, F.M., and Abee, T. 2001. Flow cytometric
assessment of viability of lactic acid bacteria. Applied and Environmental Microbiology. 67: 2326-2335.
84. Jernaes, M.W. and Steen, H.B. 1994. Staining of Escherichia coli for flow cytometry: influx and efflux of
ethidium bromide. Cytometry. 17: 302-309.
85. Novo, D., Perlmutter, N.G., Hunt, R.H., and Shapiro, H.M. 1999. Accurate flow cytometric membrane
potential measurement in bacteria using diethytoxacarbocyanine and a ratiometric technique. Cytometry.
35: 55-63.
86. Ratinaud, M.H. and Revidon, S. 1996. A flow cytometric method to assess functional state of the
Listeria membrane. Journal of Microbiological Methods. 25: 71-77.
87. Hewitt, C.J. and Nebe-Von-Caron, G. 2001. An industrial application of multiparameter flow cytometry:
assessment of cell physiological state and its application to the study of microbial fermentations.
Cytometry. 44: 179-187.
88. Nebe von Caron, G. and Badley, R.A. 1995. Viability assessment of bacteria in mixed populations
using flow cytometry. Journal of Microscopy Oxford. 179: 55-66.
89. Mason, D.J., Lopez-Amoros, R., Allman, R., Stark, J.M., and Lloyd, D. 1995. The ability of membrane
potential dyes and calcafluor white to distinguish between viable and non-viable bacteria. Journal of
Applied Bacteriology. 78: 309-315.
Flow cytometric analysis for inactivation studies 75
90. Lopez-Amoros, R., Comas, J., and Vives-Rego, J. 1995. Flow cytometric assessment of Escherichia
coli and Salmonella typhimurium starvation-survival in seawater using rhodamine 123, propidium iodide
and oxonol. Applied and Enviromental Microbiology. 61: 2521-2526.
91. Ben-Amor, K., Breeuwer, P., Verbaarschot, P., Rombouts, F.M., Akkermans, A.D.L., Vos, W.M.d.,
and Abee, T. 2002. Multiparametric flow cytometry and cell sorting for the assessment of viable, injured,
and dead Bifidobacterium cells during bile salt stress. Applied and Environmental Microbiology. 68:
5209-5216.
92. Attfield, P.V., Kletsas, S., Veal, D.A., van Rooijen, R., and Bell, P.J.L. 2000. Use of flow cytometry to
monitor cell damage and predict fermentation activity of dried yeast. Journal of Applied Microbiology. 89:
207-214.
93. da Silveira, M.G., V, S.R., Loureiro-Dias, M.C., Rombouts, F.M., and Abee, T. 2002. Flow cytometric
assessment of membrane integrity of ethanol-stressed Oenococcus oeni cells. Applied and
Environmental Microbiology. 68: 6087–6093.
94. Humphreys, M.J., Allman, R., and Lloyd, D. 1994. Determination of the viability of Trichomonas
vaginalis using flow cytometry. Cytometry. 15: 343-348.
95. Tanaka, Y., Yamaguchi, N., and Nasu, M. 2000. Viability of Escherichia coli O157:H7 in natural river
water determined by the use of flow cytometry. Journal of Applied Microbiology. 88: 228-236.
96. Terzieva, S., Donnelly, J., Ulevicius, V., Grinshpun, S.A., Willeke, K., Stelma, G.N., and Brenner,
K.P. 1996. Comparison of methods for detection and enumeration of airborne microorganisms collected
by liquid impingement. Applied and Environmental Microbiology. 62: 2264-2272.
97. Boulos, L., Prevost, M., Barbeau, B., Coallier, J., and Desjardins, R. 1999. LIVE/DEAD BacLight :
application of a new rapid staining method for direct enumeration of viable and total bacteria in drinking
water. Journal of Microbiological Methods. 37: 77-86.
98. Auty, M.A.E., Gardiner, G.E., McBrearty, S.J., O'Sullivan, E.O., Mulvihill, D.M., Collins, J.K.,
Fitzgerald, G.F., Stanton, C., and Ross, R.P. 2001. Direct in situ viability assessment of bacteria in
probiotic dairy products using viability staining in conjunction with confocal scanning laser microscopy.
Applied and Environmental Microbiology. 67: 420-425.
99. Bunthof, C.J., Schalkwijk, S.v., Meijer, W., Abee, T., and Hugenholtz, J. 2001. Fluorescent method
for monitoring cheese starter permeabilization and lysis. Applied and Environmental Microbiology. 67:
4264-4271.
100. Alonso, J.L., Mascellaro, S., Moreno, Y., Ferrús, M.A., and Hernández, J. 2002. Double-staining
method for differentiation of morphological changes and membrane integrity of Campylobacter coli cells.
Applied and Environmental Microbiology. 68: 5151-5154.
101. Hewitt, C.J., Boon, L.A., McFarlane, C.M., and Nienow, A.W. 1998. The use of flow cytometry to study
the impact of fluid mechanical stress on Escherichia coli W3110 during continuous cultivation in an
agitated bioreactor. Biotechnology and Bioengineering. 59: 612-620.
102. Saxelin, M. 1997. Lactobacillus GG - A human probiotic strain with thorough clinical documentation.
Food Reviews International. 13: 293-313.
103. Lievense, L.C., Verbeek, M.A.M., Noomen, A., and van't Riet, K. 1994. Mechanism of dehydration
inactivation of Lactobacillus plantarum. Applied Microbiology and Biotechnology. 41: 90-94.
104. Herrmann, M. Treatment of mammalian cells with high hydrostatic pressure results in the induction of
either apoptosis or necrosis. in Gemeinsames Statusseminar von BMBF-, DBU- und DFG-geförderten
Forschungsprojekten - Hochdrucklebensmitteltechnologie und -bioverfahrenstechnik“. 2004.
Freising.
105. Lee, J. and Kaletunc, G. 2002. Evaluation of the heat inactivation of Escherichia coli and Lactobacillus
plantarum by differential scanning calorimetry. Applied and Environmental Microbiology. 68: 5379–5386.
Flow cytometric analysis for inactivation studies 76
106. Mackey, B.M., Miles, C.A., Parsons, S.E., and Seymour, D.A. 1991. Thermal denaturation of whole
cells and cell components of Escherichia coli examined by differential scanning calorimetry. Journal of
General Microbiology. 137: 2361–2374.
107. Mohacsi-Farkas, C., Farkas, J., Meszaros, L., Reichart, O., and Andrassy, E. 1999. Thermal
denaturation of bacterial cells examined by differential scanning calorimetry. Journal of Thermal Analysis
and Calorimetry. 57: 409–414.
108. Ananta, E., Heinz, V., and Knorr, D. 2004. Assessment of high pressure induced damage on
Lactobacillus rhamnosus GG by flow cytometry. Food Microbiology. 21: 567–577.
109. Luscher, C., Balasa, A., Frohling, A., Ananta, E., and Knorr, D. 2004. Effect of high-pressure-induced
ice I-to-ice III phase transitions on inactivation of Listeria innocua in frozen suspension. Applied and
Environmental Microbiology. 70: 4021-4029.
110. Arroyo, G., Sanz, P.D., and Préstamo, G. 1999. Response to high-pressure, low-temperature treatment
in vegetables: determination of survival rates of microbial populations using flow cytometry and detection
of peroxidase activity using confocal microscopy. Journal of Applied Microbiology. 86: 544-556.
111. Ritz, M., Tholozan, J.L., Federighi, M., and Pilet, M.F. 2001. Morphological and physiological
characterization of Listeria monocytogenes subjected to high hydrostatic pressure. Applied and
Environmental Microbiology. 67: 2240-2247.
112. Martin, M.M. and Lindqvist, L. 1975. The pH dependence of fluorescein fluorescence. Journal of
Luminescence. 10: 381-390.
113. Vogel, R.F., Ehrmann, M.A., Gänzle, M.G., Kato, C., Korakli, M., Scheyhing, C.H., Molina-Guiterrez,
A., Ulmer, H.M., and Winter, R. 2003. High pressure response of lactic acid bacteria, in Advances in
High Pressure Bioscience and Biotechnology II, Winter, R., Editor. Springer Verlag: Berlin. p. 249-254.
114. Marquis, R.E. 1984. Reversible actions of hydrostatic pressure and compressed gases on
microorganisms, in Repairable lesions in microorganisms, Hurst, A. and Nasim, A., Editors. Academic
Press, Ltd.: London.
115. Macdonald, A.G. 1984. The effects of pressure on the molecular structure and physiological functions of
cell membranes. Philosophical Transactions of The Royal Society: Biological Sciences. 304: 47-68.
116. Garcia-Graells, C., Hauben, K.J.A., and Michiels, C.W. 1998. High-pressure inactivation and sublethal
injury of pressure-resistant Escherichia coli mutants in fruit juices. Applied and Environmental
Microbiology. 64: 1566-1568.
117. Ritz, M., Jugiau, F., Rama, F., Courcoux, P., Semenou, M., and Federighi, M. 2000. Inactivation of
Listeria monocytogenes by high hydrostatic pressure: effects and interactions of treatment variables
studied by analysis of variance. Food Microbiology. 17: 375–382.
118. Abe, F. and Horikoshi, K. 2001. The biotechnological potential of piezophiles. Trends in Biotechnology.
19: 102-108.
119. Buckow, R., Heinz, V., and Knorr, D. 2005. Two fractional model for evaluating the activity of
glucoamylase from Aspergillus niger under combined pressure and temperature conditions.
Biotechnology Progress. Submitted for publication.
120. Hülsen, U. 1999. Alternative heat treatment processes. European Dairy Magazine. 3: 20-24.
121. Villamiel, M. and de Jong, P. 2000. Inactivation of Pseudomonas fluorescens and Streptococcus
thermophilus in tripticase soy broth and total bacteria in milk by continuous-flow ultrasonic treatment and
conventional heating. Journal of Food Engineering. 45: 171-179.
122. Scherba, G., Weigel, R.M., and O'Brien, W.D.J. 1991. Quantitative assessment of the germicidal
efficacy of ultrasonic energy. Applied and Environmental Microbiology. 57: 2079-2084.
123. Kaila, M., Isolauri, E., Saxelin, M., Arvilommi, H., and Vesikari, T. 1995. Viable versus inactivated
lactobacillus strain GG in acute rotavirus diarrhoea. Archives of Disease in Childhood. 72: 51-53.
Flow cytometric analysis for inactivation studies 77
124. Ouwehand, A.C. and Salminen, S.J. 1998. The health effects of cultured milk products with viable and
non-viable bacteria. International Dairy Journal. 8: 749-758.
125. Simakachorn, N., Pichaipat, V., Rithipornpaisarn, P., Kongkaew, C., Tongpradit, P., and
Varavithya, W. 2000. Clinical evaluation of the addition of lyophilized, heat-killed Lactobacillus
acidophilus LB to oral rehydration therapy in the treatment of acute diarrhea in children. Journal of
Pediatric Gastroenterology and Nutrition. 30: 68-72.
126. Ouwehand, A.C., Kirjavainen, P.V., Shortt, C., and Salminen, S. 1999. Probiotics: mechanisms and
established effects. International Dairy Journal. 9: 43-52.
78
3 SPRAY DRYING OF PROBIOTIC BACTERIA
Effect of processing conditions and drying media on survival characteristics during drying
and storage at non refrigerated conditions
Spray drying of probiotic bacteria 79
3.1 Introduction
3.1.1 Drying of microorganism
The industrial manufacture of fermented dairy products relies on the use of bacterial starter
culture for the fermentation of yoghurt, cheese, sour cream, etc. Furthermore, growing
interest is clearly observed in the use of probiotic bacteria, either in dairy or non-dairy
products. These bacteria are commonly preserved and distributed either in frozen or dried
form before they are inoculated in the fermentation tank or incorporated in the food product.
Bacteria stored in frozen or dried form may be maintained for a long-term storage period.
Besides, the possibility to work with frozen or dried concentrate also meets the expanding
interest in the application of ready-to-use culture concentrates for direct inoculation of milk
vats, which is – compared to previous practice using sub-culturing method – less prone to
bacteriophage attack and contamination with other problematic microbes. Preservation of
bacteria in dried form is preferred, since frozen cultures occupy a large volume, heavy and
require storage at subzero temperatures, all of which results in high costs for storage,
shipping and energy [1].
However, it is generally anticipated, that the transfer of bacteria into dried form, which
involves physical removal of water, may have unexpected side effect; primarily loss of activity
and/or ultimately loss of viability. The loss of these crucial bacterial properties has to be
minimized or effectively controlled since it finally affects the productivity of the fermentation
process and also the characteristics of food products containing living bacteria cells.
Unexpected deviation in the functional behaviours or in the number of viable cells is usually
compensated by adding higher concentration of bacteria into the fermentation broth or into
the final product. Since the suggested correction procedure would result in higher production
cost, it is of utmost importance to know the nature of cellular damages as induced during
water removal and to find ways how to effectively protect the crucial sites in order to
minimize viability loss not only during drying but also during subsequent storage.
Cellular damage observed in bacteria during dehydration
Dehydration decreases water availability inside or in the vicinity of the dried cells such that
they reach a dormant state during which the metabolism is slowed down and even stopped
completely [2]. Viability and activity loss occurred on bacteria during drying was reported to
be closely related to damage to the cell wall, cytoplasmic membrane and the DNA [3, 4].
Cytoplasmic membrane damage could be detected by increased sensitivity to NaCl [4-7].
Moreover, either increased permeability of ß-galactosidase substrate [5] or increased
leakage of ß-galactosidase in the supernatant fluids [4] were indicative for the loss of
membrane integrity. Similarly, evidences of membrane disruption could be found in the
higher diffusion rate of DNase into cells and by leakage of UV-absorbing materials or
Spray drying of probiotic bacteria 80
potassium ions from the cells [1, 4, 5, 8]. Although antagonic activity of bacteriocin producing
bacteria was still retained after spray drying [6, 9], it was also reported that plasmic loss
occurred [10], presumably due to increased membrane permeability, leading to an absence
of a specific inhibitory activity mediated by the lost plasmid.
Since permeability control is mainly associated with cell membrane, it seemed probable to
relate damage in the dried cells with changes in the profile of cellular lipids, which primarily
constitute the cytoplasmic membrane. Previous work on drying of Lactobacillus bulgaricus
showed a decreased ratio of unsaturated/saturated fatty acids compared to normal cells,
indicating that spray drying induced lesions in the cellular lipid-containing structures [11].
During subsequent storage in air further decrease of the unsaturated/saturated fatty acids
could be observed [11, 12]. It is possible that two different mechanisms of phospholipids
degradation were involved: oxidation of unsaturated fatty acid and lipolysis [12]. Furthermore,
it was suggested that products from lipid peroxidation might be involved in DNA damage [11].
However, another study on the comparison of the effect of different drying methods
documented that the fatty acid composition of freeze dried cells did not vary greatly from the
normal cells, whereas in vacuum dried cells, the fatty acid spectra shifted towards shorter
chain length [5].
Apart from affecting the cellular lipids drying induced damage could also occur on cellular
proteins, either membrane-bound or cytosolic proteins [13]. It was reported that drying
resulted in damage of membrane-bound proton translocating ATPase of L. bulgaricus [12,
14]. Following either vacuum or freeze drying the attachment of a 46 kDa surface layer
protein of Lactobacillus acidophilus to the cell wall was destabilized, resulting in partial loss
of surface proteins or making the protein easily extractable with 0.1% sodium dodecylsulfate
[5]. Drying was reported to be deleterious on the functional integrity of membrane-bound
proteins indirectly by either disrupting the bilayer integrity, which resulted in the displacement
of the transporting proteins, or by directly denaturing these proteins [14-16].
In order to improve understanding of the drying induced damage on crucial cellular
components, such as cytoplasmic membranes and proteins, and relate the damage on these
macromolecules to the overall viability of the microorganism as well as to suggests
approaches to effectively protect this complex biological system during drying, a lot of
systematic studies have been dedicated to evaluate and differentiate the mode of action of
dehydration on model systems, encompassing model membranes and proteins.
Dehydration damage occurred on model membranes
The effect of drying on membranes has been thoroughly evaluated on liposomes, which are
artificial vesicles composed of concentric lipid bilayers separated by water compartments.
Spray drying of probiotic bacteria 81
The typical characteristic of bilayer-forming lipids is their amphiphilic nature: a polar
headgroup covalently attached to one or two hydrophobic hydrocarbon tails. When these
lipids are exposed to an aqueous environment, interactions between themselves (hydrophilic
interactions between polar headgroups and van der Waal’s interactions between
hydrocarbon chains) and with water (hydrophilic interactions, hydrophobic effect) lead to
spontaneous formation of closed bilayers.
Liposomes can differ in size: they can range from the smallest vesicle (SUV) obtainable on
theoretical grounds (diameter ~20 nm) to liposomes which are visible under the light
microscope, with a diameter of 1 µm or greater (LUV), equal to the dimensions of living cells
(Fig. 1). They can also differ in terms of lipid composition and structural organization,
corresponding to uni-, oligo- or multi-lamellar vesicles (MUV o MVV). Liposomes are built in
such a way that the solute can be encapsulated in the aqueous compartment (polar solutes)
or embedded in the lipid bilayers (lipophilic or amphiphilic solutes).
SUV
Ø 15 – 25 nm
Unilamellar
Multilamellar
LUV
LUV
Ø 1000 – 2500 nm
MLV
Ø 100 – 1000 nm
MVV
Ø 100 – 1000 nm
Figure 1
Schematic view of the liposomes of different sizes and lamellar structure. SUV: Small unilamellar
vesicle; LUV: Large unilamellar vesicle; MLV: Multilamellar vesicle; MVV: Multivesicular vesicle
The properties of liposomes and their subsequent applicability depend on the physical and
physico-chemical characteristics of the liposomal membrane. Usually, a zwitterionic or non-
ionic lipid is used as the basic lipid for the preparation of liposomes. The net surface charge
of liposomes can be modified by the incorporation of positively charged lipids, such as
stearylamine, or negatively charged lipids, such as dicetylphosphate, phosphatidyl glycerol or
phosphatidyl serine [17].
Spray drying of probiotic bacteria 82
The fluidity of the liposomal bilayer, made from a single lipid, depends on the lipid phase
transition temperature (Tm). As can be seen in Figure 2, when the liposome is brought to Tm,
the membrane passes from a solid gel phase, where the lipid hydrocarbon chains are in an
ordered state, to a fluid liquid-crystal phase, a disordered state, where molecules have more
freedom of movement [18]. Hence, depending on lipid Tm, different membranes composed of
distinct lipids can exhibit different fluidity levels at the same temperature. Membrane
permeability is highest at the phase transition temperature [13, 19], and is lower in the gel
phase than in the fluid phase [17]. A general sequence of hydrophilic solute permeability is:
water > small non-electrolytes > anions > cations ≅ large non-electrolytes > large
polyelectrolytes [18].
DEHYDRATION REHYDRATION AT
ROOM TEMPERATURE
hydrated
liquid crystalline
liquid crystalline
liquid crystalline
liquid crystalline
gel
leakage
retention
dehydration
dehydration
withs
ugar
Sugar molecule
withoutsugar
Figure 2
A diagrammatic representation of the effect of sugar on leakage of trapped solute when phospholipid
bilayers are dried and rehydrated at room temperature, as adapted from [20]. Note that this scheme
only valid for lipids having Tm lower than ambient temperature.
Drying experiments with liposomes showed that the removal of water, which profoundly
affects the stability of biological membranes, alters the physical properties of membrane
phospholipids and leads to destructive events including fusion between adjacent vesicles
and lipid phase transition from liquid crystalline to gel (Fig. 2) [21]. The destruction of dried
biological membranes is manifested upon rehydration, i.e. by increased permeability of
entrapped water soluble substance and lateral phase separation of membrane constituents
[16, 22].
Spray drying of probiotic bacteria 83
In detail, if a lipid bilayer is dried then as the water molecules are removed, the phospholipids
headgroups are forced closed together. These water molecules spatially separate the
phospholipids headgroups in the hydrated state. The close approach of lipid molecules leads
to increased van der Waal’s interactions between the fatty acyl chains. Consequently, the
lipid would be more likely to undergo transition from liquid crystalline into gel phase [22]. This
event appears as a shift in the gel-to-liquid crystalline phase transition toward higher
temperatures, i.e. the dry phospholipids exist in a gel phase at temperatures, at which they
would be in liquid crystalline phase if they were in the hydrated state [20, 22]. When these
gel-phase lipids are rehydrated at room temperature, they undergo another phase transition
to the liquid crystalline phase (Fig. 2). However, during membrane phase transitions, there
regions of gel phase and liquid crystalline phase might coexist or there are packing defects in
some regions or non-bilayer phases are formed. As a result, the membranes do not provide
adequate barrier properties leading to transient leakage, which is thought from mismatch
between molecules in the gel and those in the liquid crystalline state. Catastrophic
membrane leakage, when occurred in biological cells, may ultimately lead to cell death [23].
In addition to a transition from the liquid crystal to the gel phase, certain phospholipids can
undergo a transition from liquid crystal to hexagonal II phase as water is removed [24].
Hexagonal II phase is produced by an increased interaction between adjacent phospholipid
headgroups as the separating water is removed. This type of transition is especially common
in membranes high in phosphatidylethanolamine (DOPE; Tm ~ -5°C [25]), such as the inner
membrane of E. coli, and may play a role in the cell mortality.
Dehydration damage occurred on proteins
Most studies to elucidate the effect of drying on functionality of proteins were not conducted
on complex systems such as membrane-bound proteins, since the implication of protein-lipid
interaction may complicate the interpretation [16]. Instead, extensive studies have been
made on soluble proteins including ß-galactosidase [26-28], trypsinogen [29], trypsin [30, 31],
alkaline phosphatase [32], amylase [33], restriction enzymes [34-36], invertase [37],
dehydrogenase [38], lysozyme [39] and bovine serum albumin [40, 41].
The biological function of proteins depends on their three-dimensional structure, which is
determined largely by water. Loss of water, which form the hydration shell of the protein, can
lead to loss of native structure, resulting in loss of biological function upon rehydration [42].
For instance, enzyme inactivation as a result of drying may involve a conformational change
in the active site, which can be due to protein denaturation (unfolding and/or inter- or
intraprotein hydrophobic aggregation) or to blockage of specific active groups by the
formation of covalent links (i.e. condensation of amino groups of proteins with carbonyl
Spray drying of probiotic bacteria 84
compounds) that modify the active site of the enzyme or make it inaccessible to its substrate
[28].
Protection against drying induced injury on membranes
Water replacement hypothesis
Certain sugars may confer protection to liposomes and isolated biological membranes
against dehydration damage [15, 16, 20, 22]. The protective effect of sugar relies on a direct
physical interaction between the hydroxyl groups of the sugars and the polar residues of the
phospholipids head groups in dehydrated state, as described by the water replacement
hypothesis [21, 43, 44]. Accordingly, sugar molecules are thought to substitute for water
molecules between the lipid headgroups during dehydration and sugar molecules need to be
present on both sides of liposome membrane [20] (Fig. 2). Liposomes dried without internal
sugar did not remain intact and released their contents. The aforementioned hypothesis is
primarily evidenced by infrared spectroscopy measurements, which showed that in the
presence of sugar, the phosphate group in dry phospholipids behaves as in it is fully
hydrated and vice versa, OH stretching bands in sugar are strongly affected by the presence
of phospholipids, which suggests that the phosphate of the phospholipid and the OH group of
the sugar are interacting in the dry state, possibly by hydrogen bonding [43, 45]. Adequate
spacing between the lipid headgroups owing to the insertion of sugar is deemed responsible
for the substantial depression of liquid-crystalline-to-gel phase transition temperatures (Tm),
resulting in the preservation of membrane in a liquid-crystalline state, even when dry (Fig. 2).
Consequently, the membrane would not pass through a phase transition during rehydration
and leakage of entrapped aqueous solution could be prevented.
Formation of high viscous glassy matrix
The stabilizing properties of sugar can also be explained by the formation of glassy state by
the sugar upon dehydration [46-48]. A glass is a kinetically metastable, time-dependent
physical state presented in amorphous or partially crystalline materials, characterized by
near absence of molecular movement [49]. The most important parameter describing the
glassy state of amorphous materials is the glass transition temperature (Tg), below which the
materials exhibit extremely high viscosities which gives them "solid-like" properties. Above
the glass transition temperature, viscosity drops sharply in the "rubbery" state and the
mobility of the system increases accordingly. As documented in Figure 3, glass transition
temperature (Tg) is highly dependent on molecular weight. Furthermore, the glass transition
temperatures of low molecular weight carbohydrates is specific to each anhydrous material
but is extremely sensitive to water, which plasticizes the amorphous structure and reduce the
Tg [50]. The formation of a glassy solid results in a reduction of translational molecular motion
Spray drying of probiotic bacteria 85
and rates of chemical reactions and relaxation rates for various processes in glassy matrices
may be very low, especially at temperatures well below the glass transition temperature [51].
Figure 3
Glass transition (Tg) of common carbohydrates and sugars (anhydrous) as a function of molecular
weight. Graph is adapted from [52].
One special dehydration damage which could be minimized upon entrapment of liposome in
high viscous glassy matrix is fusion; an event which is characterized by the occurrence of
large, multilamellar vesicles and leakage of the contents of the original liposomes [53, 54].
The increased stability of dry liposomes in the glassy state may be associated with elevated
energy barrier for liposome fusion, which considerably reduced the mobility of molecules and
facilitated physical separation of dry liposomes [47]. Alternatively, the prevention of fusion
could be explained by the ability of sugar to work as a spacing matrix between liposomes
[55].
It was reported that polymers, such as hydroxyethyl starch and dextran, both good glass
formers which have high glass transition temperature (Tg) are capable of inhibiting fusion
between liposomes during drying due to glass formation [46, 53, 54]. However, these
polymers were not effective in preventing leakage from liposome during drying, since they
did not lower Tm of the lipid [45, 53]; a property which is prerequisite for prevention of leakage
owing to membrane phase transition [54]. Consequently, both glass formation and decrease
of Tm in the dry lipids is required to retain structural integrity. Although it may be required, the
thermodynamically unstable nature of the glassy state might explain why it itself is not a
sufficient requirement for stabilization of dehydrated liposomes [47, 56].
A special case was observed on trehalose glass, which was found to be more superior in
conferring dehydration tolerance compared to other sugar glasses. At the same water
content, the Tg of trehalose (100°C [57]) is much higher than sucrose (62°C [57]) or other
sugars and liposome stabilized in trehalose could retain aqueous content more effectively
Spray drying of probiotic bacteria 86
than the one entrapped in sucrose [58]. Apparently this remarkable effect of trehalose is
resulted from its ability to form dihydrate as it absorbed water, thereby sequestering water,
which might otherwise participate in lowering the Tg to below ambient. Thus, trehalose could
render dried biomaterials more stable when stored under high humidities and temperatures.
Protection capacity of sugars with respect to the types of sugar moieties, types of
glycosidic linkage and degree of polymerisation
Data from a comparative study of the efficiency of sugars from different degree of
polymerisation (DP) in conferring protection for liposomes against dehydration induced
rupture showed that while glucose, maltose or maltotriose completely prevented the
aggregation and/or fusion of liposomes during lyophilization, other malto-oligosaccharides
(DP = 4 to 7) induced them due to the increase in hydrophobicity with the number of glucose
residues [59].
Other polysaccharides with high Tg such as dextran or hydroxyethyl starch (MW = 450000)
were capable of inhibiting fusion of liposomes; however these polymers were completely
ineffective in stabilizing membrane during freeze drying [45, 53], since they are thought to be
sterically hindered from penetrating the bilayer in the dry state so that the gel to liquid
crystalline phase transition temperature was not depressed and leakage of entrapped
aqueous solution was not prevented [54]. Surprisingly, inulins of a DP ranging between 10
and 30 from chicory and dahlia could stabilize liposomes during freeze-drying and the
stabilization is mediated by a direct interaction of the polysaccharides with membrane lipids
despite of the proposed problem with steric hindrance [45] and even a high molecular mass
bacterial levan (DP > 25000) was able to directly interact with membranes [60].
Consequently, size-related effects of steric hindrance can not fully account for the observed
protective effects of oligo- or polymeric sugars. Furthermore, with increasing chain length,
fructo-oligosaccharides were more effective than gluco-oligosaccharides in stabilizing dried
liposomes against leakage of aqueous content after rehydration [61]. According to FTIR
spectroscopy data it was observed that the ability of gluco-oligosaccharides to hydrogen
bond to the head group of dry lipids decreased dramatically with increasing DP, whereas
chain length hardly affected the ability of fructo-oligosaccharides to interact.
Despite the ability of both monosaccharide and disaccharide to form hydrogen-bonding with
liposomes, the interaction between acyl chains is stronger in the monosaccharide system,
indicating that the interaction between monosaccharide and liposome is weaker than that of
disaccharide [62]. As a result, leakage of entrapped fluorescence marker was more
pronounced after drying in the presence of monosaccharide, although this was capable of
inhibiting fusion. In contrast, it was reported that glucose did not inhibit fusion during drying
and it did not prevent leakage [54, 63]. Other study also confirmed the higher efficacy of
Spray drying of probiotic bacteria 87
disaccharide to minimize drying induced damage on membrane [63, 64]. However, as
already mentioned above, the efficacy of different disaccharides in protecting dried
liposomes varies greatly [58]. When the ability of sucrose and trehalose to preserve a model
membrane system, sarcoplasmic reticulum, were compared, it was observed that although
both sugars could confer good protection, higher concentration of sucrose was required to
achieve equal extent of stabilization [65]. Despite the observed superiority of trehalose over
other sugars it was proposed that as freeze-drying technology improved or at ideal drying
conditions the differences between disaccharides tended to disappear, and the protective
effect of sugar encompasses disaccharides in general [65].
Protection against drying induced injury on proteins
Dehydrated proteins could also be stabilized in the presence of sugar molecules [27, 29, 39,
40]. In analogy to protective effect of sugar on dehydrated membrane stabilization of native
protein state during dehydration can be explained by water replacement hypothesis. Using
infrared spectroscopy method it was shown that in the presence of sugar the amide bands of
the dried proteins are similar to those of hydrated proteins and the OH vibrations of sugar are
altered by the protein so that they closely resemble the OH groups of the hydrated sugar [13,
39, 43]. Similarly, drying of proteins in the presence of sugar rendered the conformational
state or the secondary structure of dried proteins more similar to the one of hydrated
proteins, as measured with derivative infrared spectroscopy [27, 40, 66]. Sugar molecules
are thought to replace water and be capable of hydrogen bonding to the polar and charged
groups of the protein in the place of lost water, leading to the preservation of the native,
aqueous structure in the dried state [42, 66]. Consequently, the occurrence of hydrogen
bonding of sugar to the dry proteins is required for inhibition of dehydraton-induced protein
unfolding [39]. Upon rehydration protein that was dried in the presence of sugar induced the
refolding of protein structure, whereas the protein dried alone did not show any recovery of
its native structure [40]. However, the failure of glucose to prevent lysozyme unfolding during
freeze-drying showed that hydrogen bonding between carbohydrate and protein alone is not
sufficient to confer protection to a protein during lyophilization [66]. The difference in the
efficacy for protein preservation of different sugar despite the ability to hydrogen bond to
protein – at sites which normally binds water – could be explained by the extent of
interaction between protein and sugar [31]. The authors suggested that trehalose interacted
more strongly with trypsin and hence it is a better preservative than sucrose because
trehalose is unable to form strong hydrogen bond with other trehalose molecules in the
anhydrous amorphous state. Besides, the efficacy of different sugars for protein protection
could be explained by Maillard (browning) reaction between reducing sugars and proteins in
the dry state, which has often been invoked as a major source of damage [28, 32, 65]. When
Spray drying of probiotic bacteria 88
a freeze-dried model system was incubated with sucrose, trehalose, and glucose, the rate of
browning seen with sucrose approached that of glucose – as much as 2000 times faster than
that with trehalose [67]. In a comparative study on the protective effect of different types of
sugars (mono-, di-, tri- or polysaccharides) it was shown that only trehalose could offer
protection of dried preparation of restriction enzymes even at storage temperature up to 70°C
and this superiority was thought to be based on chemical stability and non-reducing nature of
trehalose [34].
The role of glassy state in the preservation of structure of native proteins have been
discussed and analyzed extensively. It was hypothesized, that the glassy state is the sole
important factor in long-term stabilization of stored restriction enzymes dried in a trehalose
matrix because translational diffusion-limited relaxation processes taking place during protein
denaturation are strongly inhibited [68]. Loss of activity of ß-galactosidase dried in lactose-
containing matrix, which seemed to correlate with accumulation of the product of non-
enzymatic browning reaction between lactose and enzyme, was found to be reduced by
addition of maltodextrin, a polymer with high Tg [28]. Addition of glass formers which increase
the Tg was effective in lowering sugar crystallization [28], which is known to promote water
redistribution, increase the rate of diffusion-controlled reactions and affect biomolecule
stability [69].
As already indicated above and also in the drying study on membrane, the entrapment of
protein molecules in an amorphous, glassy matrix was not fundamental requisite for
stabilization, since enzyme inactivation was observed in heated glassy matrices well below
their glass transition temperature [37]. The absence of break in the Arrhenius plot for thermal
inactivation of lactase in the vicinity of glass transition indicated that the formation of glassy
state did not influence the deteriorative effect of heat on enzyme [26]. In addition, the
restriction enzyme EcoRI was very stable during storage at 37/45°C in spite of the fact that
sugar matrices were completely rubbery [36]. Dextran, a high molecular weight
carbohydrates which remains glassy when dried, was not capable of inhibiting protein
unfolding during dehydration [34, 42]. It was demonstrated in a work on drying of alkaline
phosphatase that different inulin preparations with Tg > 100°C gave different degree of
protection during storage at 60°C and better retention of activity was achieved when inulins
of relatively high DP and low content of reducing groups [32]. In line with this observation,
again, non-enzymatic browning reaction, which is related to the presence of reducing sugar
was proposed to be implicated in the loss of functional integrity of amino compounds present
in proteins [70].
Furthermore, care has to be take in defining optimal sugar concentration added in protective
matrix. One work on spray drying of trypsinogen showed that preservation of complete
activity can be achieved by addition of sucrose at 1:1 mass ratio [29]. However, at high
Spray drying of probiotic bacteria 89
carbohydrate concentrations, preferential sugar-sugar interactions prevailed, resulting in a
phase separation within the formulation matrix. The preferential incorporation of the sucrose
molecules in a sugar-rich phase reduced the actual amount of the carbohydrate available to
interact with the protein and thereby decreased the number of effective protein-sucrose
contacts. As a consequence, the protein could not be effectively protected during spray
drying. Upon differentiating the individual stresses occurred during spray drying, i.e. heat,
atomization pressure, and dehydration, it was found that dehydration was the major stress
responsible for protein denaturation [29]. This observation could be confirmed by the data on
high pressure study presented in Chapter 2 of this work, in which it was shown that the
application of high pressure up to 400 MPa did not have any lethal effect on LGG suspended
in phosphate buffer.
Protection against drying induced injury on living cells
A study which attempted to elucidate the impact of added trehalose on dehydration tolerance
of Saccharomyces cerevisae and on phase transition temperature of the cell revealed that
the reduction of the dry Tm in the presence of sugar was deemed responsible for the high
survivability of these cells after drying and rehydration [71]. The same conclusion was drawn
as well upon drying of Escherichia coli and Bacillus thuringiensis, although it was also noted
that sugar not only prevented drying-induced membrane phase transition but also maintained
the dry proteins in their hydrated conformations [13]. The maintenance of dried proteins in a
conformation similar to that of the hydrated protein is proposed to be based on the binding of
sugars to the hydrophilic domains of the proteins and preventing inter- and intraprotein
hydrogen bonding during drying and rehydration [72].
In contrast, added sugars (maltose, trehalose, sorbitol) did not depress Tm in dry cells of L.
plantarum, irrespective of their beneficial effect of desiccation tolerance of liposomes [73].
Therefore, it was speculated that instead of direct interaction with the polar lipid headgroups
primarily the radical scavenging (antioxidative) activity of the sugar conferred dehydration
protection [73], as also supported by Andersen et al (1999) [74]. On the other hand, it was
also mentioned that the failure to provide desiccation tolerance was attributed to the fact that
the sugars might not get access to the cytoplasm, since the cells were grown on glucose and
fructose and thus not adapted to the uptake of the evaluated carbohydrates.
Sugar has to be present in both side of the membranes (extra- and intracellular) to allow
sufficient protection of cellular components critical for viability including membranes and
proteins [13, 20, 75, 76]; however the threshold level of internal sugar concentration sufficient
to confer protection varies greatly depending on the cell type, drying conditions and residual
moisture content. A minimum amount of sugar is needed intracellularly to hydrogen-bind to
all of the lipid molecules in the plasma membrane or to form intracellular glassy state [76]. In
Spray drying of probiotic bacteria 90
general, anhydrobiotic organisms have concentrations of internal sugar (especially trehalose)
that range from 20 to 50% of the dry weight of the organism [56]. To allow internal
accumulation of sugar, the impermeability of cell membrane to sugar has to be overcome.
Some treatments have been proposed to improve diffusion of sugars across the cell
membrane.
One approach made use of the membrane leakage upon temperature-induced membrane
phase transition. Penetration of trehalose and sucrose into cells of E. coli and B.
thuringiensis (0.43 µmol trehalose per mg dry weight) was reported to be achievable after
incubation in 100 mM sugar solution at hydrated Tm (at around 10°C), when membranes
have greater permeability and the sugar flew down its concentration gradient and into the
cells [13]. Similarly, trehalose could also be introduced into insulin-producing cells from
mammalian pancreas with help of inherent leakiness of the membranes during membrane
lipid phase transition at Tm [19].
Furthermore, improved diffusion was achievable by long soaking time. Wolkers et al (2001)
reported on an efficient uptake of trehalose (ca. 20 mM cytosolic concentration) into human
blood platelets at 37°C in a time span of only several hours, which improved the survival rate
during lyophilization [77]. Soaking of baker’s yeast in 1 M trehalose solution for several days
was proved sufficient to load the sugar into cells (up to 200-250 mg trehalose per g of dry
cells) and it was proposed that exogeneous trehalose was incorporated by the low-affinity
trehalose transport system (a facilitated diffusion process) [78].
It had been shown that osmotic induction of trehalose synthesis in E. coli could increase the
rate of survival of desiccation [79, 80]. Drying of osmotically shocked E. coli (resulting in 300
mM internal trehalose prior drying) in the presence of 1 M trehalose improved their viability
after drying and storage stability compared to untreated sample [80].
More sophisticated approaches on mammalian cells relied on the reversible permeabilization
of plasma membranes using a switchable recombinant pore-forming hemolytic proteins α-
hemolysin, which allowed introduction of trehalose into the cells and facilitated retention of
the integrity of plasma membrane during drying [81, 82]. Chen et al (2001) reported that with
this technique more than 0.1 M intracellular trehalose could be accumulated, leading to
increased membrane integrity of fibroblasts after drying [76]. On the other hand, expression
of recombinant genes that encode for trehalose-6-phosphate synthase and trehalose 6-
phosphate phosphatase in human foreskin fibroblasts, resulted in trehalose accumulation as
high as 0.3-0.4% of the dry weight of the cell and increased desiccation tolerance [83].
Furthermore, it was also shown that the electropermeabilization technique, which is widely
used for introduction of DNA and other foreign molecules into cells, could be applied for the
intracellular delivery of trehalose, enabling cryo- and lyopreservation of human T
lymphocytes, mouse myeloma cells, and fibroblasts [84].
Spray drying of probiotic bacteria 91
In the study of plant desiccation tolerance (pollen, plant seeds, resurrection plants, etc.) it
was observed that mechanism of protection was related not only to the considerable
accumulation of di- and oligosaccharides, as can be explained by water replacement
hypothesis; the presence of compatible solutes and specific proteins, such as the late
embryogenesis abundant proteins and heat shock proteins, and the accumulation of
antioxidative compounds may also play a role in guaranteeing survival in dehydrated state
[85].
The protective role of sugar in maintaining high survivability of dried organism may also rely
on the formation of glassy state; although it was always clearly indicated whether
cytoplasmic or extracellular vitrification or vitrification at both sides are mainly responsible for
the limitation of dehydration-induced damage. Generally, it was proposed that immobilization
by vitrification may minimize stress damage on the cellular structure and thus protect their
biological capabilities during dehydration and rehydration [48]. When vitrification state is lost,
free radical oxidation, phase separation and cytoplasmic crystallization would occur and
impose real threat to the survival of dry organism. However, significant losses of fermentation
activity of commercial dry yeast were still observed in vitrified yeast sample [86]. No break or
step change was observed in the Arrhenius plot for the rates of activity loss in the vicinity of
the Tg, indicating that the degradation took place with the same rate regardless of the
thermophysical state of the matrix. Furthermore, when exposed to 70°C for 24 h yeast
viability decreased with increasing molar mass of the maltodextrine, although maltodextrine
is regarded as a good glass-former polymers with Tg values proportional to molar mass [87].
The authors found that despite the inability of maltose to protect dried lactase this reducing
sugar was more effective in ensuring high viability of dried yeast compared to maltodextrin,
indicating the predominant contribution of hydrogen bonding to yeast protection. It was
shown further that vitrification was not sufficient to explain protective mechanism of sugar
since the viability after drying of S. cerevisae, which was internally loaded with 10-20%
trehalose, could be dramatically improved although this amount had only a minor effect on
the Tg of the samples [88]. Likewise, one study on mammalian cells showed that chemical
activity leading to degradation of fibroblast membrane still occurred during storage below Tg
[76]. It was suggested that the absence of protection in glassy state was caused by local
microheterogeneities within the dried biological sample, resulting in spatial distribution of
glassy and rubbery states throughout the sample volume [86].
Taking together, drying, as a preservation method of preference for living cells, induced
various damages on crucial cellular components, as evidenced by many works on
dehydration of membranes and proteins. This cellular injury can by chance lead to cell
inactivation and therefore has to be minimized in the production of stable bioactive
Spray drying of probiotic bacteria 92
preparations. Some approaches to render cells more tolerant against dehydration were
investigated on model systems as well as on prokaryotic or eukaryotic cells. Besides, many
studies take dehydration tolerant plants into consideration, in order to learn the mechanism
of maintaining viability in dehydrated state. The superior role of sugar in conferring protection
against dehydration induced cellular damage is undoubted. Although the real mechanism
could not be unequivocally understood yet, the stabilizing effect of sugar relies on its ability to
directly interact with membranes and proteins in dried state; thus taking over the role of water
in maintaining the native or hydrated structure of these macromolecules and preventing
membrane phase transition. Second, the formation of a highly viscous glassy state during
dehydration in the presence of sugar could also be deemed responsible – at least partly – for
the marked reduction of deleterious chemical reactions occurring in cells during water loss.
Thus, both criteria have to be fulfilled by a certain sugar or a certain mixture of different types
of sugar – each fulfilling one of the proposed protective mechanism mentioned above –
which constitute the protective matrix, so that good survival after drying and rehydration
could be ensured. Additional criteria upon selecting effective sugar as single compound or
used in combination might be the chemical inertness and the inherent possibility to act as
antioxidative agent. Furthermore, sufficient protection could only be achieved when direct
interaction as well as formation of glassy state occurred in both sides of the membranes;
thus the added sugar has to be able to penetrate cell membrane appropriately. Apart from
adding sugar, incorporation of other protection-relevant compounds (e.g. antioxidant, etc.) in
the drying matrix seems to be a practical approach in improving the protective capacity.
Finally, the defense mechanism of the cells itself in responding to the changes in their
environment during the first stages of desiccation (by accumulation of compatible solutes,
synthesis of stress proteins, modified metabolic pathways, etc.) need to be explicitly studied
so as to effectively utilize this cellular response in combination with an optimized protective
formulation towards a better survivability.
3.1.2 Spray drying
In the spray drying process the feed solution is transformed from a fluid state into a dried
form by spraying the feed into a hot drying medium. The process itself involves the
atomization of a liquid feedstock into a spray of droplets, which come in direct contact with
electrically heated air in a drying chamber. There are three modes of contact between hot air
and liquid feed: co-current, counter-current and mixed flow [89]. Figure 4 shows a schematic
drawing of a lab-scale spray dryer working with in a co-current flow of drying air and feed
solution.
Spray drying of probiotic bacteria 93
Feed pump
1
2
3
4
5
6
7
8
9
10
11
Product Fine particles
TT
TT
Figure 4
Schematic view of a spray dryer installation with its essential components. The role of each
components are described in text.
The feed (1) is pumped from the product feed tank to the atomizing device which is normally
located in the air disperser in the top of the drying chamber. The sprays are produced by a
rotary (wheel) or nozzle atomizer (2), resulting in fine droplets of the range 2 – 150 µm
diameter. When two-fluid nozzle is used to generate spray, pressurized air (3) is required.
The drying air (4) is drawn from the atmosphere via a filter (5) by an aspirator (6) and is
passed through the air heater (7) to the air disperser. In the drying chamber (8), The
atomized droplets meet the hot air and the evaporation takes place cooling the air at the
same time. Evaporation of moisture from the droplets and formation of dry particles proceed
under controlled temperature and airflow conditions. The construction of the drying chamber
is made under consideration of the adequate residence time and droplet trajectory distance
for achieving the heat and mass transfer. After the drying of the spray in the chamber, the
majority of the dried product falls to the bottom of the chamber. The dried product can either
be discharged continuously from the bottom of the drying chamber or passed into a solid-gas
separator (9) where the solids from the gas stream are recovered and the powder is
collected at its bottom (10). The fine particles are usually collected in an outlet filter (11).
Spray drying of probiotic bacteria 94
Operating conditions and dryer design are selected according to the drying characteristics of
the product and powder specification, most importantly the residual moisture content of the
powder and the particle size. The first variable is affected by the evaporation rate and the
dryer ∆T (inlet air temperature minus the outlet air temperature), which in turn dictates the
amount of drying air needed and ultimately the sizing and cost of almost all of the system
components. The particle size requirement affects the choice of atomization method and can
also affect the size of the dryer.
The dehydration of the atomised liquid particles proceeds from the particle surface to the
inner core, resulting in the formation of protective vapour film, which surrounded the droplet
and keeps the particle at the vapour saturation temperature. As long as the particle does not
become completely dry, evaporation still takes place and the temperature of the solids may
decrease (due to evaporative cooling) or does not approach the dryer outlet temperature [90,
91]. This is why many heat sensitive products (enzymes, microorganisms, volatile aroma
compounds, etc.) can be spray dried at relatively high temperatures to produce powders with
low moisture load without the danger that the product may be harmed.
3.1.3 Spray drying works on lactic acid bacteria
As shown in Table 1, a lot of studies have been undertaken on evaluating the applicability of
spray drying to produce lactic acid bacteria preparations for use as dairy starter (cheese or
yoghurt), bacteriocin-producing and probiotic cultures.
The driving force for these studies was mainly to demonstrate the capability of spray dried
cultures in replacing the usual liquid or frozen bulk starter or freeze-dried cultures in the
production of fermented products. In comparison to the latter techniques of culture
production, spray drying is claimed to be more cost effective and less time consuming [92]. It
was reported in more detail that the energy consumption upon using spray drying is much
lower (4000 – 6000 kJ/kg of evaporated water) in comparison to that of freeze drying, where
as much as 100.000 kJ was required to evaporate 1 kg of water [93, 94]. Besides, freeze
dried cultures often have extended lag phases since they were not only exposed to
attenuating effect of freezing but they also are subjected to attenuation by dehydration
effects and the destabilizing effect of freezing may increase the susceptibility of the cell to
subsequent drying step [1, 95, 96].
During spray drying bacteria are faced to multiple stresses , i.e. heat (both wet and dry),
oxidative, dehydration-related stresses (osmotic, accumulation of toxic compounds, etc.)
acting either simultaneously or successively on bacteria, which potentially lead to cell death.
Besides, as already mentioned above, the removal of water, which contributes to the stability
of biological molecules, may cause irreversible changes in the structural and functional
integrity of bacterial membranes and proteins. Preservation of these essential functions and
Spray drying of probiotic bacteria 95
structure is crucial for the survival of bacteria and the retention of their functionality. Until now
the commercial application of this preservation method for lactic acid bacteria has not won
broad recognition yet. This scepticism originates mainly from the survival aspects during
drying, which is suspected to be very low [4]. Besides, the storage stability and the
rehydration properties of the spray dried bacteria are considered as poor. However, there is
a body of evidence, which demonstrates the possibility to spray dry various strains of lactic
acid bacteria without a drastic loss of viability and activity, or at least show survival rates
during spray drying which are comparable to that on freeze-drying [4, 9, 10, 97].
3.2 Objective
The overall objective of the current study was to evaluate the feasibility of spray drying to
produce dry preparations containing probiotic bacteria Lactobacillus rhamnosus GG. The
outlet temperature, i.e. the temperature measured at between drying chamber and cyclone,
which was regarded as the drying temperature, was evaluated on its effect on both the
residual moisture content and the survival rate of L. rhamnosus GG in the resulted powder.
Flow cytometric analysis in combination with carboxyfluoresceindiacetate-propidium iodide
(cFDA-PI) dual staining strategy was applied to the identify the nature of spray drying-
induced cellular injury as induced upon application of different drying temperatures.
Moreover, the suitability of prebiotic substances as a part of the drying medium was
assessed so as to demonstrate the possibility of producing pro- and prebiotic containing
preparations. The storage stability of L. rhamnosus GG spray dried in different carrier media
was investigated with respect to their capability of forming glassy state, primarily in order to
find out, whether a correlation exists between the glass forming ability of a specific carrier
and the storage stability of probiotic cultures dried with it. Apart from evaluating the
contribution of glassy state to dehydration tolerance, direct interaction between protective
media with liposomes was assessed using flow cytometric measurement.
Spray drying of probiotic bacteria 96
Table 1
Overview of existing spray drying works performed on lactic acid bacteria
Microorganism Reference Inlet Temperature
(IT). Hot air flow Atomization
pressure
Outlet Temperature
(OT) Composition of feed solution Spray drying
unit Special remarks
L. acidophilus [98] 170°C n.d. n.d. 75-85°C 25 & 40% solid milk n.d. moisture content 3.5 - 6.1%
S. salivarius subsp. thermophilus, L. delbrueckii
subsp. Bulgaricus [97] 140 – 180°C (160°C) 17m3/h = 0.28 m3
/
min 0.98 bar = 98 kPa 60-90°C (60°C) yoghurt, 13,63% solid non fat n.d.
L. helveticus [99] 220°C 30m3/h = 600 g/min n.d. 65-130°C condensed skim milk (15-34% solid) Lab-Plant SD-04 OT<80°C caused fouling on the walls, Residence
time 0.5 s, Particle size ~12 µm
L. helveticus [1] 220°C 30m3/h = 600 g/min n.d. 82 or 120°C 19% maltodextrin Lab-Plant SD-04 OT adjustment using feed flow rate
L. bulgaricus [3] 200°C n.d. 5 bar 62-105°C skim milk Niro
L. bulgaricus [4] 200°C n.d. n.d. 80°C 40% maltodextrin in water or 11% milk Niro
L. bulgaricus [11] 200°C n.d. n.d. 70°C skim milk (LabM) Niro
L. lactis, L. casei, S. thermophilus [100] 220°C n.d. n.d. 70-90°C 25% (w/w) solid maltodextrin Lab-Plant SD-04
L. lactis [9] 160 – 200°C 28m3/h = 0.47 m3/min compressed air 600
l/h 68°C 20% RSM Büchi 191 Flow rate of feed solution = 10,13,17 mL/min
S. thermophilus, L. delbrueckii subsp. bulgaricus [101] 180°C n.d. n.d. 60-80°C yoghurt Luwa water evaporating capacity : 1645 kg/h
L. paracasei NFBC338, L. salivarius UCC118 [6] 170°C n.d. n.d. 60-120°C (80-85°C) 20% RSM Büchi B191 moisture content in appreciable range (4%)
L. paracasei NFBC338 [102] 170°C n.d. n.d. 85-105°C 20% RSM Büchi B191 Moisture:1.7 - 3.3 % g/g
OT adjustment using feed flow rate
Bifidobacteria [103] 100°C n.d.
5-8 bar (Compressed
air 700 l/h) 45°C 10% (encapsulation) Büchi B190
B. infantis, B. longum [104] 100°C n.d. n.d. 50-60°C
Gelatin, gum arabic, skim milk, soluble
starch 10% (scan 2-30%) Büchi B191 Moisture content 7-9%
L. salivarius, L. sakei [10] 200°C n.d. 5 bar 70°C 11% milk Niro
L. paracasei NFBC338 [105] 170°C n.d. n.d. 95-100°C or 100-
105°C
20% RSM or 10% RSM with 10% gum
accacia n.d. Moisture: 2.5 - 3.2% g/g,
OT adjustment using feed flow rate
L. acidophilus, B. lactis Bb12 [106] 130 – 190°C n.d. 5 kgf/cm2 75 – 120°C Cellulose acetate phthalate, Glycerol,
Maltodextrin, Raftilose, milk Lab-Plant SD-04 Avg. particle size 22 µm
L. rhamnosus GG, L. rhamnosus E800, L. salivarius
UCC500 [7] 170°C 100% n.d. 85-90°C RSM 20% w/v or 10% RSM with 10%
prebiotics Büchi B191 Moisture <4%,
OT adjustment using feed flow rate
n.d. not described, RSM: reconstituted skim milk
Spray drying of probiotic bacteria 97
3.3 Material and methods
3.3.1 Test organism and preparation of bacterial suspension
Lactobacillus rhamnosus GG (ATCC 53103), thereafter abbreviated as LGG, was obtained
from Valio R&D (Helsinki, Finland). For long-term maintenance this organism was stored as
glass bead cultures (Roti-Store, Carl-Roth, Karlsruhe, Germany) in freezer at -80°C (U101,
New Brunswick Scientific, Nürtingen, Germany).
One bead of a deep-frozen culture was transferred into MRS broth (Oxoid, Basingstoke, UK)
and incubated overnight at 37°C. Afterwards a final broth was inoculated with the overnight
growth culture prior to another incubation period at 37°C for 24 h. Cells were centrifuged at
2000 g for 10 min and washed twice with PBS-buffer (phosphate-buffer-saline) pH 7.0.
Following washing the pellet was resuspended in an equal volume of the final carrier
solution.
3.3.2 Preparation of carrier solution
Reconstituted skim milk powder, hereafter abbreviated with RSM (Oxoid, Basingstoke, UK)
at a concentration of 20% (w/v) was used as the reference medium. The evaluated prebiotics
were: Raftilose®P95, an oligofructose produced by partial enzymatic hydrolysis of chicory
inulin (ORAFTI, Tienen, Belgium) and Polydextrose (DANISCO, Copenhagen, Denmark).
Thereafter, P95 and PDX, are used as the abbreviations for Raftilose®P95 and Polydextrose,
respectively. Chemical structures of these preparations are shown in Figure 5. Prebiotic
media used for spray drying consisted of an equal ratio of reconstituted skim milk and each
of these prebiotics (20%, w/v, total solids). The media were decontaminated by heat
treatment in a water bath at 90°C for 30 min.
Spray drying of probiotic bacteria 98
HO
CH2OH
O
OH
HO
OCH
2
OR
OH
OH
O
CH2
O
OH O
O
OH
OH
O
CH2
CH2OH
O
OH
OH
O
O
HO
OH
OH
O
CH2OH
HO
HO
OH
OH
O
CH2OH
O
HO
OH
OH
O
CH2OH
CH2OH
O
OH
OH
HO
OCH
2
O
OH
HO
O
O
HO
OH
OH
O
CH2
O
ab
Figure 5
Basic chemical structure of prebiotic molecules incorporated in spray drying media (Annex 4 and 5)
(a) General structure of the two basic types of ß(2→1) fructans, which are the major constituents of
the commercial oligofructose preparation Raftilose®P95. The two fructan types (GFn: glucosyl type
and Fn: fructosyl type) exist normally as a mixture. The degree of polymerisation (DP) of
Raftilose®P95 ranges typically from 2 to 8.
(b) General structure of polydextrose, which is a randomly cross linked polymer of glucose with a
highly branched complex 3D structure. All bonds are present as 1 – 6 and 1 – 4 linkages. In the
polymer, R can be hydrogen, sorbitol-bridge or more polydextrose molecule. The commercial
name of this polydextrose preparation is Litesse®.
3.3.3 Spray drying
The spray-drying process of LGG in the various media was undertaken in a laboratory scale
spray dryer (Büchi B-191, Flawil, Switzerland), which is schematically shown in Figure 6. The
feed solution was pneumatically atomized into a vertical, co-current drying chamber using a
two-fluid nozzle at a constant flow rate (5 mL min-1). The outlet temperature was adjusted
from 70 to 100°C by varying the air inlet temperature. The dried powder was collected in a
product container connected in bottom part of the single cyclone separator. Once the outlet
temperature stabilized, the heated glass container was disconnected and replaced with
another container in order to minimize uncontrolled thermal stress on the dried bacteria.
Spray drying of probiotic bacteria 99
ab
Figure 6
(a) Pathway of drying air in the spray dryer Büchi B191
Cold air is aspirated through the air inlet tunnel (1) and then electrically heated (2) prior to
entrance in the spray cylinder, in which drying of the droplets into solid particles takes place (3).
Dried powder is separated from fine particles in the cyclone (4) and collected in the glass
container (8). Outlet filter is placed to remove fine particles and to prevent them from entering
aspirator (5) which generates the air flow. Temperatures are measured in the entrance of spray
cylinder (6, termed as air inlet temperature) and in the intermediate piece between spray cylinder
and cyclone (7, termed air outlet temperature).
(b) Pathway of feed solution and pressurized air in the spray dryer Büchi B191
Feed solution (A) conveyed by peristaltic feed pump (B) and atomizing air (D, inlet) are passed
separately to the nozzle head (C), where the atomization of the feed solution into fine droplets
takes place. The co-current two-fluid nozzle is located at the centre of the upper part of the spray
cylinder. Atomization is created by compressed air at a pressure of 0.5 to 2 bar. Nozzle diameter
is 0.7 mm. Powders produced with this adjustment have particle size ranged from 5 to 15 µm on
average [105].
Generally, the level of outlet temperature which is determined by the drying rate, could be
adjusted by two different settings of the spray dryer. In this study, the adjustment of outlet
temperature was performed by holding flow rate of the feed suspension at a constant value
(25% pump capacity ~ 5 mL min-1) for all outlet temperatures, whereas the inlet temperature
was varied, as shown in Table 2. A rise in the outlet temperatures due to the fouling of the
inner wall of the spray chamber can be compensated by slightly increasing the feed flow rate.
Spray drying of probiotic bacteria 100
Table 2
Applied parameters for spray drying of L. rhamnosus GG
Pump capacity : 25 ± 2 % ~ flow rate of feed suspension 5 mL min-1
Flow rate of drying air : 100% ~ 60 m3 h-1
Atomization pressure : 6 bar
Flow rate of pressurization air : 700 L h-1
Aimed outlet temperature (°C) Adjusted inlet temperature (°C)
70 115
80 130
90 145
100 155
On the other hand, it is also possible to vary the flow rate of the feed solution (thereby
varying the amount of water per time unit which needs to be evaporated) under holding the
inlet temperature at typically high level to obtain the aimed outlet temperature, as used in
previous studies (Tab. 1). This latter operational procedure is thought to be more flexible
when different drying rates are frequently applied since changing the feed flow rate is faster
and easier than changing inlet temperature.
3.3.4 Determination of moisture content in spray dried powders
The residual moisture content of spray dried powders was determined by oven-drying at
102°C [107]. This involved determination of the difference in weight before and after
overnight storage in he oven dryer. Moisture content was then expressed as a percentage of
the initial powder weight.
3.3.5 Enumeration of probiotics after spray drying
To determine the survival rate of the probiotic bacteria, spray dried powders were rehydrated
with sterile Ringer’s solution (No. 15525, Merck, Darmstadt, Germany) to obtain the same
solids concentration as the initial feed solution. Afterwards, they were serially diluted and
drop plated in duplicate on MRS agar (Oxoid, Basingstoke, UK). Plates were placed in an
anaerobic jar (AnaerocultA, Merck, Darmstadt, Germany) and incubated at 37°C for 48 h.
Survival rate were calculated as follows: %survival = N/N0 x 100, where N0 represented the
number of bacteria before drying and N was the number of bacteria after drying.
Spray drying of probiotic bacteria 101
3.3.6 Staining procedure and flow cytometric assessment
Double staining with carboxyfluorescein diacetate (cFDA) and propidium iodide (PI)
A 100 µL aliquot of rehydrated powder was mixed with 900 µL Ringer solution and
centrifuged for 10 min at 4000 g. The pellet was resuspended in 100 µL PBS buffer 0.05 M
pH 7.0 and mixed together with 100 µL of 100 µM cFDA stock solution (Molecular Probes,
Inc. Leiden, The Netherlands), so that the concentration of cFDA in the reconstituted pellet
suspension was 50 µM. The suspension was incubated at 37°C for 10 min to allow
intracellular enzymatic conversion of cFDA into cF (Carboxyfluorescein). After excessive
cFDA was removed by centrifugation, 30 µM PI (Molecular Probes, Inc. Leiden, NL) was
added. The cell suspension was kept in an ice bath for 10 min to allow labelling of the
membrane-compromised cells prior to flow cytometric measurement.
Flow cytometric measurement
All measurements were made with a Coulter®EPICS®XL-MCL flow cytometer
(BeckmanCoulter Inc., Miami, USA) with 488 nm excitation from an argon-ion laser at 15
mW. The green fluorescence from carboxyfluorescein was collected through a 525 nm band-
pass filter; and a 620 nm band-pass filter was used to collect red fluorescence from
propidium iodide. Data were analysed with the software package Expo32 ADC
(BeckmanCoulter Inc., Miami-FL, USA). Acquisition of fluorescence data was performed by
pre-setting a gate in the forward-angle light scatter (FS) versus sideward scatter (SS) plot,
which enabled bacterial cells of interest and artefacts to be discriminated. The flow rate was
set at typical values of 300-600 bacterial cells per s. Further settings are listed in Annex 2.
3.3.7 Storage test
The dried samples were stored at a relative humidity of 11%, which was regarded as optimal
in preserving dried bacteria [108]. The relative humidity was maintained constant by storing
the powder in hermetically closed jar above saturated lithium chloride solution [109]. The
samples were kept at storage temperatures of 25 or 37°C. Only powders dried at an outlet
temperature of 80°C were subjected to the storage test. The storage inactivation data were
expressed as logarithmic value of relative survival fraction (log N/N0). N refers to the bacterial
count at a particular storage period, whereas N0 represents the bacterial count at the
beginning of the storage. The viability loss during storage was assumed to follow first order
reaction kinetics, and the first order rate constants were calculated. Three replicate storage
trials were undertaken.
Spray drying of probiotic bacteria 102
3.3.8 Differential scanning calorimetry measurement
The glass transition temperatures (Tg) of spray dried preparations were determined by
differential scanning calorimeter (DSC). With this technique the difference in heat flow to or
from a sample, and to or from a reference material is monitored as a function of temperature,
while the sample is subjected to a controlled temperature program. Simultaneously,
thermogravimetric measurement (TG) was applied to monitor the change in the mass of the
sample as a function of temperature. All calorimetric measurements were performed using a
Netzsch STA-409C thermoanalyzer unit (Netzsch GmbH, Selb-Bayern, Germany).
All measurements were made at a linear heating rate of 20 K⋅min-1 using an opened platinum
pan and an empty pan as a reference material. Samples of 10 – 20 mg were initially heated
up linearly to 150°C to remove moisture from the sample. Prior to initiation of the second
scan, the heated, water-free samples were rapidly cooled down to -30°C with nitrogen flush.
3.3.9 Calculation of glass transition temperatures
The glass transition temperature (Tg) was determined during the second run and was defined
as the midpoint value of the change in specific heat observed as an endothermic shift in the
baseline of the DSC signal. Due to the aforementioned limitation, the Tg of the fresh,
moisture-loaded spray dried powder could not be directly determined. This value was
therefore estimated by Gordon Taylor equation (Equation 1). This empirical equation is
typically applied to predict the Tg values of a solid substrate at various water contents. The
reliability of this equation for calculating Tg of various food systems has already been
demonstrated in several studies [47, 58, 110-113].
For the prediction of Tg in sugar/water glasses, the Gordon Taylor equation is as follows :
21
2211
wkw
TwkTw
Tgg
g⋅+
⋅
⋅
+
⋅
=
Equation 1
where w1 and w2 are the weight fractions of the solute and water, respectively; Tg1 is the
glass transition temperature of the water-free solute; Tg2 is the glass transition temperature of
water (-135°C); and k is a constant [57, 114].
Spray drying of probiotic bacteria 103
0 20406080100120
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Galactose
Glucose
Fructose
Sucrose
Maltose
Lactose
k (-)
Glass transition temperature Tg (°C)
Trehalose
Monosaccharide Disaccharide
Figure 7
The k-values of some important mono- and disaccharides glasses (anhydrous) as a function of glass
transition temperatures (Tg). Data were taken from [57], in which the onset of the glass transition
according to DSC measurement was regarded as Tg. This empirical relationship allow Equation 2 to
be established (see Material and Methods).
For a given sugar molecule, the Tg values of several combinations of experimental weight
fractions must first be determined in order to estimate k. The values of k calculated for
relevant sugars (mono- and disaccharides) with Tg values at several water contents are
shown in Figure 7. It could be observed, that the plot of Tg for the anhydrous sugars against k
was linear. The regression equation obtained (Equation 2) was used in Equation 1 for the
prediction of the Tg value of spray drying medium used in this study at various water
contents, so as to enable the generation of state diagram, which shows the glass transition
temperatures over a wide moisture range.
61.30293.0 1
+
⋅
=g
Tk Equation 2
3.3.10 Monitoring direct interaction of sugar-membranes using liposomes to determine the
protective effect of sugar during drying
Preparation of liposome
Egg phosphatidylcholine (EPC) was purchased from Avanti Polar Lipids (Alabaster, AL) and
carboxyfluorescein (cF) was obtained from Molecular Probes (Eugene, OR)
100 µL of EPC was dried from chloroform under a stream of N2 and stored under vacuum
overnight to remove traces of solvent (Fig. 8). Afterwards 1 mL of 1000 µM cF 50 mM PBS
Spray drying of probiotic bacteria 104
buffer (pH 7.0) was added to hydrate the dried lipids. Liposomes were prepared from these
hydrated lipids using a LiposoFast-Basic hand-held extruder (Avestin Europe GmbH,
Mannheim, Germany) with two layers of polycarbonate membranes with 1000 nm pores. The
LiposoFast-Basic produces unilamellar liposomes by the manual extrusion of the lipid-cF
suspension through a polycarbonate membrane of defined pore size, using gas-tight, glass
syringes. The sample is passed through the membrane by pushing the sample back and
forth between two syringes.
0,5 ml EPC in NAP-5 column
+ 1 mL PBS-EDTA buffer
(0,9 mL PBS 0,05M + 0,1 mL EDTA 10 mM)
100 µL EPC in
Chloroform
EPC vesicle
Rehydration
Flow cytometry
measurement
FS – Particle size
FL1 – cF leakage
24 h in Desiccator under
Vacuum
+ 1 mL Carboxyfluorescein
(1000 µM)
Extrusion with LiposoFast 15x
(Pore size: 1000 nm)
Drying in desiccator
Figure 8
Brief overview of the procedure of preparation of liposomes made of EPC (Egg phosphatidylcholine)
using LiposoFast extrusion system as well as the assessment of drying induced damage by flow
cytometry.
Leakage experiments
For leakage experiments, 0.5 mL the vesicles obtained from extrusion was passed through a
NAP-5 column (Amersham Biosciences AB, Uppsala, Sweden) and equilibrated in 1 mL
PBS-EDTA buffer (50 mM PBS, 1 mM EDTA at pH 7.0), to remove the CF not entrapped by
the vesicles. The eluted samples a lipid concentration of approximately 10 mg·mL-1.
Spray drying of probiotic bacteria 105
Liposomes were mixed with an equal volume of concentrated solutions of sugars in PBS and
filled into 1.5 mL microcentrifuge vials (Eppendorf AG, Hamburg, Germany) to reach a final
lipid concentration of 5 mg·mL-1. Twenty µL was filled in each vial. The vials were then dried
in desiccators at room temperature over silica gel for 24 h. Damage to the liposomes was
determined in flow cytometer after rehydration with 20 µL of PBS-EDTA buffer either as
leakage of the soluble marker cF or as occurrence of aggregration.
Flow cytometric measurement and data analysis
Assessment of liposome was performed on Coulter®EPICS®XL-MCL flow cytometer
(BeckmanCoulter Inc., Miami, USA), as already described above. Exact configurations are
listed in Annex 2. A separate protocol for the flow cytometric measurement of liposome was
created. The application of flow cytometry in this field of research is relatively new. The most
frequently applied technique to assess leakage on liposome is the fluorometer technique,
which basically measures the cF-fluorescence in the suspension medium. The fluorescence
of carboxyfluorescein is strongly quenched at the high concentration inside the vesicles
(Concentration ≥ 100 mM) and is increased when cF is released into the medium [45]. The
cF concentration in the liposomes has to be quite high so that upon leakage of a certain
amount of cF into the surrounding medium measurable fluorescence signal could be
detected. The total cF content of the vesicles (0% retention value) was determined after lysis
of the membranes with 50 mL of 1% Triton X-100, whereas the 100% retention values were
determined with freshly prepared liposomes. In contrast, using flow cytometric technique the
fluorescence of cF entrapped in the liposome is measured. This method allows a
considerably lower concentration of cF (Concentration ~ 1000 µM) to be used. A decrease in
the fluorescence intensity as measured inside the liposome after drying and rehydration may
reflect the leakage of certain amount of cF into the surrounding medium. To check this,
liposomes loaded with different concentration of cF were made and measured with flow
cytometer. The green fluorescence from cF was collected through a 525 nm band-pass filter.
Fluorescence data was plot as frequency histogram. In general, frequency histograms can
be used to display relative fluorescence or scattered light signals plotted against the number
of events. With this plot the distribution in the fluorescence intensity within the detected
liposomes can be observed. The mean of the fluorescence intensity of the liposomes was
plotted against the corresponding cF concentration in liposomes (Fig. 9a). It is obvious, that
the fluorescence intensity increased as the concentration of cF increased, indicating that the
intensity of the fluorescence signals collected at 525 nm can give information about the
residual cF concentration in the liposomes. A shift in the distribution towards lower
fluorescence value is therefore related to the leakage of the enclosed dye in the external part
of liposomes.
Spray drying of probiotic bacteria 106
0 20406080100
0.0
0.2
0.4
0.6
0.8
1.0
cF-fluorescence intensity (-)
EDTA-concentration (mM)
µ
0 200 400 600 800 1000
0
5
10
15
20
25
30
Fluorescence intensity (-)
Carboxyfluorescein in Liposome (µM)
0123456
0
20
40
60
80
100
120
Mean or mode of population's FS (-)
Microsphere diameter (µm)
mean
mode
ab c
Figure 9
Correlation of signals measured by flow cytometer, i.e. signals measured in the 535 nm green
fluorescence channel and forward scatter channel, with fluorescence intensity (Fig. 9a) and particle
size of polystyrene beads (Fig. 9b), respectively. Figure 9c shows the influence of increasing
concentration of EDTA on cF fluorescence intensity.
The same also apply for the determination of the distribution of particle size, which correlate
with the forward scatter signal. Standard reference polystyrene beads (Polysciences Europe
GmbH, Eppelheim, Germany) were used to validate a relative correlation between the
intensity of forward scatter (FS) signal with particle size. It was observed that either the mean
or the mode of the frequency distribution of the beads measured positively correlated with
the bead size (Fig. 9b). A shift towards higher value of forward scatter is indicative for the
occurrence of liposomes with increased size. This change accompanies drying of liposomes
and can be related to either fusion or aggregation.
Besides, it was found that the presence of EDTA could delete cF-fluorescence, as measured
by fluorescence photometer (Perkin Elmer 650-10S, Perkin-Elmer Corp., Norwalk, USA).
Figure 9c demonstrates the effect of increased concentration of EDTA on the fluorescence of
a 1µM cF solution. As high as 1 mM concentration of EDTA was found to be sufficient to
cause maximal reduction of cF fluorescence, since at higher concentrations of EDTA no
further decrease of cF fluorescence could be achieved. Thus, EDTA (end concentration
1mM) is incorporated in the PBS buffer used to dissolve sugar for leakage experiments. The
addition of EDTA is thought to be necessary in order to prevent re-attachment of leaked cF
on the outer phospholipid layer. It is likely, that this unexpected re-attachment would still
render leaked liposomes fluorescent and thereby producing false-negative results, i.e. leaky
liposomes might still be detected as intact liposome due to possible attachment of leaking cF.
With EDTA outside of liposome the cF diffusing out the liposomes could be effectively
sequestered, thus eliminating a potential source of erroneous fluorescence.
Spray drying of probiotic bacteria 107
3.4 Results and discussion
3.4.1 Identifying critical processing conditions
Initial spray-drying experiments were designed to investigate the effect of process
parameters, primarily different adjustment of outlet air temperatures on bacterial survival rate
and residual moisture content of the powders. Taking values of critical water content for skim
milk powder from literature into account, it is expected to identify optimal drying temperature,
which is to be used for storage stability test.
The total solids content of the RSM as the drying medium was held constant at 20% (w/v).
This solids content was frequently used and has been regarded as optimal for assuring high
residual viability of different strains of lactic acid bacteria [6, 9, 99, 102]. Increasing the total
solids content of the feed solution decreased the percentage of surviving bacteria [98, 99,
115], although batch-kinetics studies demonstrated that the thermoresistance increased with
increasing solids content. Aside from increased osmotic stress which eventually might occur
at increasing solid contents, drying conditions was expected to vary with difference in solids
concentration of feed solution [98]. Increases in solids content result in increases in
suspension viscosity and more concentrated feed suspensions produce larger particles
[116]. These particles present relatively smaller ratios between the surface area and the
volume and greater core retention, thus requiring longer drying times to achieve a given level
of residual moisture; this can reduce cell viability due to the longer contact time of the
particles with the hot air .
Daemen et al (1982) attributed the decrease in cell viability to a decrease in the drying rate
for higher solids content feed solutions. They estimated that the increase in the drying time
was approximately proportional to the square of particle size. This condition led to a higher
moisture content at the centre of the dried particle and higher moisture contents decrease
thermoresistance [115]. It was already reported that dried organisms become more resistant
to heat damage [117-120]. Furthermore, spray drying conducted on spray drying of
trypsinogen showed that preservation of complete activity can be achieved by addition of
sucrose at 1:1 mass ratio [29]. However, at higher carbohydrate concentrations in the feed
solution, preferential sugar-sugar interactions prevailed, resulting in a phase separation
within the formulation matrix, which led to a reduction of the protection capacity of the sugar
during spray drying.
A range of outlet temperatures between 70°C and 100°C was used in the preliminary
experiments to spray dry LGG. The survival rate of LGG was inversely proportional to air
outlet temperatures (Fig. 10a) and the residual moisture content increased as the air outlet
temperature was reduced (Fig. 10b). It is obvious that drying should not take place at very
high temperatures (T ≥ 90°C) not only due to higher inactivation but also due to more
Spray drying of probiotic bacteria 108
pronounced browning reaction [97, 101]. Low viability level due to application of high
temperature was found to extend the lag phase required for complete recovery and
multiplication to prior to lactic acid production [100]. On the other hand, drying at very low
temperatures (T ≤ 60°C) showed increased tendency of powder lumping due to higher
moisture content [101].
0
1
2
3
4
5
70 80 90 100
0
15
30
45
60
75
90
Survival rate (%)
Outlet temperature (°C)
Moisture content (%)
Figure 10
Effect of outlet temperature during spray drying on the moisture content (upper figure) and survival of
L. rhamnosus GG (bottom figure) in probiotic powder prepared from 20% (w/v) reconstituted skim milk
(RSM). Initial cell count of the feed solution was ~ 109 cfu mL-1. Data are means of two, or more, spray
drying experiments, whereas the error bars represent the standard deviation of the means.
Therefore, a compromise in terms of selection of air outlet temperature is required for optimal
drying results. A higher viability level, as achievable by drying at lower temperatures, is
clearly preferable; however the knowledge of a critical water content should rather dictate the
selection the optimal drying temperature, since a lot of technological properties of the powder
are dependent on the water content. In particular, as can be seen in Figure 11a, the
flowability aptitude of skim milk powders is largely influenced by water activity, aw of the
powder [121]. With help of this Figure and Figure 11b, which shows the sorption isotherm of
skim milk powder, the relationship between moisture content, water activity and flowability
could be better examined. It is known from the literature, that a residual moisture of 4% (w/w)
was regarded as a good quality parameter of dried dairy products [122]. This moisture level
corresponds to an aw value of around 0.2, which in turn give a high flowability index (Fig.
11a). Residual moisture contents of around 4% (w/w) were achieved upon spray drying at an
outlet temperature of 80°C. These results fit well with results made by other groups working
with the same spray drying equipment as used in this study [6] or with spray dryer from
Spray drying of probiotic bacteria 109
another company [1]. The range of critical moisture content as well as aw for good powder
characteristics discussed here is also in line with data from literature describing desirable aw
value for the survival of bacteria dried in milk-based media [123].
A different concept was followed in determining critical water content for skim milk powder.
Based on the measurement of glass transition temperatures, a critical water content of 7%
(w/w) was proposed for the storage of skim milk powder at 25°C (Fig. 11b) [124]. The latter
critical moisture value was considerably higher than the one proposed by Masters (1985) and
would theoretically allow spray drying at lower temperatures; which are less harsh for
bacteria and energetically more advantageous than higher outlet temperatures.
Nevertheless, a moisture content lower than 7% (w/w) after drying is preferred so as to
minimize the risk of storage-related defects such as crystallization of lactose, because of the
better buffering effect towards fluctuation in storage temperature during shipping or
processing. Consequently, an air outlet temperature of 80°C during spray drying was used
for further assessments.
Water activity
Flowability
00.2 0.4 0.6 0.8
40
50
60
70
80
90 a
20151050
-40
-20
0
20
40
60
80
100
0
0.2
0.4
0.6
0.8
1.0
Water content (g/100g of Solids)
Temperature (°C)
Water activity (at 24°C)
b
Figure 11
(a) Influence of water activity (aw) on the flowability properties of skim milk powder [121]
(b) Glass transition temperature, Tg of skim milk powder (S) as influenced by its moisture content.
Another plot shows the sorption isotherm of skim milk powder (z). Graph is adapted from [124,
125].
The role of thermal effect in viability loss during spray drying
It was clearly shown that the number of cells not surviving spray drying conditions increased
as the air outlet temperature was elevated (Fig. 10). This observation is of qualitatively in
agreement with the findings of many other studies, which examined survivability at different
air outlet temperature. This tendency pointed out that thermal stress was apparently more
pronounced at higher temperatures. However, during spray drying bacteria are faced not
only to heat (both wet and dry), but also different other stresses, including oxidative,
dehydration-related stresses (osmotic, accumulation of toxic compounds, etc.). Each of them
Spray drying of probiotic bacteria 110
can potentially lead to death and it is of interest to differentiate the contribution of each of this
factor to bacterial inactivation. Knowledge about the type of stress predominantly affecting
bacteria or how these stresses are interrelated in inducing cellular damage during spray
drying may help in identifying the processing condition with reduced lethality and effective
selection of protective media. In the scope of this study the lethal effect of wet heat was
evaluated. This stress was simulated by challenging LGG suspended in RSM to heat
treatment at 60, 65 and 70°C. The role of heat stress was determined by analyzing kinetic
parameters derived from thermal inactivation curves and comparing these with survival rates
of LGG spray dried at relevant temperatures. The thermal death curves are shown in Figure
12a. From this figure it can be seen that these temperatures were found to be suitable for
assessing inactivation rates in an adequate time domain. The slopes of the curves (k in s-1)
allow calculation of D-values at the treatment temperatures applied. Figure 12b was
constructed upon extrapolation of the Arrhenius plot (ln k versus T
-1) towards higher
temperatures, in order to estimate the D-values, i.e. time required to kill 90% of the initial
population, within the temperature ranges similar to the air outlet temperature used for spray
drying (70 to 100°C). Accordingly, the D-values for 70, 80, 90, and 100°C are approximately
6.8, 0.4, 0.03, and 0.002 s, respectively.
60 70 80 90 100
1E-3
0.01
0.1
1
10
100
D-values (s)
Temperature (°C)
0 100 200 300 400 500 600
-8
-7
-6
-5
-4
-3
-2
-1
0
k= -0.14531
R2= 0.998
60°C
65°C
70°C
k= -0.04196
R2= 0.977
log N/N0 (-)
Treatment time (s)
k= -0.00756
R2= 0.997
ab
Figure 12
(a) Thermal inactivation curves of L. rhamnosus GG in reconstituted skim milk 20%. Regression lines
representing first order thermal death kinetics, which were used to fit the survival data, are drawn
along with the calculated inactivation rates at each temperature, k (in s-1). The D-values at each
treatment temperature are calculated as reciprocal value of k.
(b) D-values of L. rhamnosus GG in reconstituted skim milk 20% as a function of treatment
temperature. Data points represented as closed symbols indicate D-values calculated from
original data (Fig. 10a), whereas data points shown as open symbols were D-values estimated
using Arrhenius equation, which in turn was constructed using original inactivation data.
Accordingly, activation energy, Ea, of LGG inactivation in reconstituted skim milk 20% was 281.15
kJ mol-1.
Spray drying of probiotic bacteria 111
Further, when the exposure of bacteria to air outlet temperatures used during spray drying
was assumed to be around 0.5 to 2 s, as already proposed in previous study [98, 99], then
the percentage survival after exposure to this fixed treatment time at different treatment
temperatures could be calculated using the calculated kinetic parameters as well as
Arrhenius relationship. It was assumed that the total volume of the Büchi B-191 passed by
the drying air was about 7 L; thus the estimated residence time at maximal aspirator capacity
(60 m3/h) should be ca. 0.4 s (Annex 3). Upon comparing the inactivation results obtained
from spray drying and thermal kill in solution (for 0.4 s treatment time) it was obvious that –
except at 70°C, where the difference in death rates is within the error range of the
microbiological analysis – thermal inactivation in aqueous solution always gave higher
inactivation results than spray drying (Table 3). Consequently, heat stress only played a
minor role in contributing to bacterial death during spray drying, since there would be higher
inactivation results during spray drying if the contribution of heat were indeed much more
pronounced. Cell death during spray drying was therefore more related to non-thermal effect.
Table 3
Percentage survival after spray drying of L. rhamnosus GG at different outlet temperatures and after
exposure of the bacterial solution to the treatment temperatures for 0.4 s. In both cases bacteria were
suspended in reconstituted skim milk, 20% and subjected either to spray drying or to thermal
treatment.
Temperature
(°C)
N/N0, spray drying
(%)
N/N0, thermal inactivation
(%)
70 68.21298 87.4734
80 54.26882 9.54993*
90 26.83412 << 0.001*
100 6.25322 << 0.001*
* estimated values
Furthermore, these data also suggested that the real temperature experienced by bacteria
was much lower than the adjusted outlet temperature. It is indeed one feature of spray drying
process that dehydration of the atomised liquid particles proceeds from the particle surface to
the inner core, resulting in the formation of protective vapour film, which surrounds the
droplet, keeps the particle surface wet and maintain the temperature at the vapour saturation
temperature (wet-bulb temperature). At this drying stage, the drying rate is constant. As long
as the particle does not become completely dry, thermal inactivation will be limited since
evaporation still takes place. As a result, the temperature of the solids may decrease (due to
Spray drying of probiotic bacteria 112
evaporative cooling) or does not approach the dryer outlet temperature [90, 91]. This is why
many heat sensitive products can be spray dried at relatively high temperatures without the
danger that the product may be harmed. At the subsequent drying stage, the particle surface
becomes dry and the temperature may increase maximally to the dryer temperature since
evaporative cooling is no longer available. Consequently, effect of temperature would be
higher, but – as already mentioned above – due to the lower moisture content, the microbial
cells will show higher resistance [117-120]. It is also possible microbial cell are entrapped in
the solid matrix. With regard to the latter issue it was shown by confocal scanning laser
microscopy (CSLM) technique, which allow cross-sectional analysis of the dried powder, that
the spray dried cells were encapsulated in the milk powder particles, which may have
protected the culture during spray drying [6].
Previous works on spray drying of Salmonella and S. cerevisae already indicated that
thermoresistance of the bacteria (in aqueous state) and drying temperature are not the
predominant factors affecting lethality [117, 126]. Therefore, extrapolation of heating death in
solution is not suitable for the prediction of survival data during spray drying [3]. Likewise,
Lievense et al (1994) already demonstrated that during dehydration at 5°C, cell death
occurred due to damage in their cell membrane not related to thermal effects [8]. In contrast,
they found out that exposure of cells in suspension to 60°C led to cell death but there was no
indication about the occurrence of membrane damage.
3.4.2 Flow cytometric analysis of spray dried bacteria
Flow cytometric method has continuously been developed and evaluated in the area of dairy
industry, particularly for rapid detection, enumeration and differentiation of bacterial
contaminants in milk [127, 128] as well as to analyze subpopulations of bacteria in probiotic
products and dairy starters [129]. Besides, this technique was evaluated on its efficiency in
determining the viability of freeze-dried bacterial cells [130] and in monitoring the cell
damage and fermentation activity of dried yeast [131].
In most cases the presence of matrix in which the bacteria are embedded, such as milk
proteins and lipid particles, as well constituents of protective media, might hamper the
application of this technique, since fluorescence stains might bind non-specifically to proteins
and lipid globule [127]. For the case of milk as matrix, clearing of milk was necessary. This
can be achieved by using enzyme cocktails, which can effectively degrade milk proteins and
lipid [127], as well as a special milk-clearing solution, which contains non-ionic detergent and
a chelating agent as reactive ingredients [129].
In this study flow cytometric analysis was applied to evaluate cellular injury sites affected by
spray drying as well as to relate data on the physiological characteristics of spray dried
bacteria as obtained by flow cytometric analysis with survival data as obtained with plate
Spray drying of probiotic bacteria 113
count method. To achieve these goals, LGG was stained with both cFDA and PI. In contrast
to previous studies which implied the necessity of clearing prior to staining and
measurement, clearing step was not applied in this study.
Briefly, the staining mechanism using cFDA and PI is based on the capability of viable to
enzymatically convert non-fluorescent cFDA into a membrane-impermeant green fluorescent
product cF, which can be accumulated in their cytoplasm. Thus, retention of the dye by the
cells requires a high degree of membrane integrity and functional cytoplasmic enzymes.
Apart from retaining cF intracellularly cells with intact membranes are able to exclude the
membrane-impermeant, nucleic acid dye PI. This dye can only enter cells with compromised
membranes. Intracellularly it binds the RNA or DNA; the resulted PI-nucleic acid complex
emits red fluorescence upon excitation.
The dual-parameter dot plot of the green fluorescence (x-axis) and the red fluorescence (y-
axis) in Figures 13a to 13f was used to differentiate bacterial populations based on their
fluorescence properties in response to cFDA-PI staining. Each dot, which constitutes the cell
cloud, represents one single cell, which is plotted as a co-ordinate of their green and red
fluorescence value. The different intensities of the shaded area in the cell clouds signify the
population density. The quadrants on the dot plot were set so that viable cells with intact
membranes were in quadrant 4 (Fig 13a). This quadrant only included bacterial cells, which
actively accumulated cF and excluded PI, and which therefore showed high green
fluorescence and low red fluorescence. Prior to spray drying, all LGG cells were encountered
in quadrant 4 (Fig. 13a). Upon rupture of the cell membrane and the concomitant loss of the
CF-accumulation capacity (simulated by thermal treatment at 75°C for 90 s) the cells are not
capable of excluding PI. The particular bacterial population, which was solely labelled by PI,
showed low green fluorescence and high red fluorescence. Thus, the membrane damaged
population was found in quadrant 1 (Fig 13b).
Spray drying of probiotic bacteria 114
bc
ef
a
d
g
0 102030405060708090
0
10
20
30
40
50
60
70
80
90
Percentage of surviving cells (%)
Percentage of cells showing cF-accumulation (%)
Figure 13
Flow cytometric fluorescence density-plot analysis (cF, green fluorescence versus PI, red
fluorescence) of L. rhamnosus GG spray dried in reconstituted skim milk, RSM (20%, w/v) at different
air outlet temperatures following staining with cFDA (carboxyfluorescein diacetate) and PI (propidium
iodide). Conditions applied: viable cells prior to spray drying (a); dead, membrane compromised cells
following heat treatment at 75°C for 90 s (b); and bacteria following spray drying at air outlet
temperature of 70 (c), 80 (d), 90 (e), and 100°C (f). The figures (in %) following the quadrant number
(top left hand corner) represent the percentage of the cells in the corresponding quadrant.
(g) Correlation between survival rates of L. rhamnosus GG spray dried at different air outlet
temperatures in reconstituted skim milk (RSM), as determined by plating on MRS agar, and the
percentage of carboxyfluorescein (cF)-accumulating population in quadrant 4 of the flow
cytometry dot plots (Figure 6). The results are means based on data from three, or more, replicate
experiments; error bars show the standard deviations of the means.
The fluorescence profile of the spray dried bacteria indicated that the population in quadrant
4 decreased as the air outlet temperature increased (Fig. 13c to Fig. 13f). Simultaneously,
the population labelled with PI (in quadrant 1) increased. Both trends indicated that the
degree of damage of cell membranes increased as the air outlet temperature increased.
Spray drying of probiotic bacteria 115
When the survival rates of LGG spray dried at different air outlet temperatures, as
determined by plating on MRS agar, were plotted against the percentage of cF-stained cells
(in quadrant 4), a strong correlation was obtained (Fig 13g). This indicates, that cells, which
were still capable of retaining cF after spray drying, were capable of forming colonies on
MRS agar plates. Since the decrease in the population of cF stained bacterial cells was
indicative of membrane deterioration, it is concluded that the cells suffering damage to their
membranes during spray drying did not grow on MRS agar and were thus dead.
According to data obtained using confocal scanning laser microscopy it was shown that at
lower spray drying temperatures (70°C) membrane intact cells predominated, whereas drying
at 120°C resulted in cells with compromised membranes, which were stained by PI [6].
Similarly, Johnson et al (1995) also found that the loss of membrane integrity in L. helveticus
was greater in cells spray dried at higher outlet temperature (82°C compared to 120°C), as
determined by ß-galactosidase assay [1]. Evidences about increased permeability of cell
membrane following drying could also be documented by increased sensitivity of dried cells
to NaCl, increased leakage of potassium ions, UV-absorbing materials and ß-galactosidase
in the supernatant fluids [4-7]. In terms of the exact mode of action of drying induced
membrane damage it was demonstrated in a previous work on L. bulgaricus that spray
drying induced lesions in the cellular lipid-containing structures and resulted in a reduced
ratio of unsaturated/saturated fatty acids; thereby pointing out the possible implication of lipid
oxidation in membrane degradation upon excessive contact with air [11]. Similarly, freeze
drying is also accompanied by a decrease in the unsaturated/saturated fatty acids ratio [14].
Furthermore, it was found that membrane damage leading to increased permeability to
Dnase could be achieved following dehydration in absence of heat [8]. However, as
evidenced by the present study, thermal induced membrane rupture could not be fully
excluded since higher magnitude of membrane damage and viability loss were obtained at
higher outlet temperatures (Fig. 13). Taken together, it seems probable to conclude that
damage on cell membrane occurs during spray drying and a massive injury of this cellular
component (i.e. increase of permeability) results in cell death. Membrane rupture occurring
during spray drying is most likely caused by synergistic effect of thermal and non-thermal
(dehydration, oxidative) stresses.
In addition, on the latter stages of the work it was found that the current sampling method for
the determination of the initial count of the bacteria prior to spray drying has to be re-
assessed. This conclusion was triggered by the fact that in some cases the bacterial count
after spray drying was higher than before, especially when low outlet temperatures were
applied. This observation might be caused by the chain formation of lactic acid bacteria,
which is frequently observed. Upon plating on agar, both several viable bacteria in aggregate
or one single cell form one colony. It was proposed that following spray drying the chains are
Spray drying of probiotic bacteria 116
broken due to shear stress, resulting in higher amounts of viable single cells; each of them
are capable of forming a visible colony on agar. Eventually, in particular when the lethal
effect of drying is low, the amount of single, viable cells might be higher than the initial count
taken prior to drying owing to the aggregation of many cells in the latter case. A practical way
to solve this problematic was suggested: instead of taking the initial count prior to drying it
was proposed to consider the initial count from bacteria suspension sprayed through the two-
fluid-nozzle with help of pressurized air in absence of drying air. Using this approach the
same shear stress experienced by spray dried samples is also applied on the initial count. As
a result, the bacteria count after atomizing through the nozzle was 2.50 ± 0.16 times more
than initial count without atomization. Taking this result into account the survival rates (N/N0
in percent) determined in this and many other works should therefore be less than currently
detected. However, this approach was still not yet used in the present study but should be
considered in further investigations in order to obtain the real magnitude of viability loss
during spray drying.
3.4.3 Incorporation of prebiotics in the spray drying medium
As already concluded by some authors, survival of bacteria during drying is highly dependent
not only on the processing conditions but also on the type of drying media regardless of the
type of drying method used [7, 132-134]. With regards to the mass and heat transfer, which
extensively take place during the passage of the dried bacteria in the dryer, Lian et al (2002)
suggested that thermal properties of the drying medium, such as thermal conductivity and
diffusivity can affect survival of spray dried probiotics [104].
The interest in supplementing food with prebiotic substances is increasing. By definition,
prebiotic is a non-digestible food ingredient that beneficially affects the host by selectively
stimulating the growth and/or activity of one or a limited number of bacteria in the colon, and
thus improves the health of host [135]. The prebiotics developed so far are the non-digestible
oligosaccharides and non-digestible fructans [136]. Furthermore, when in a single product
both probiotic bacteria and prebiotic compounds exist, the product is called synbiotic. It is
defined as a mixture of probiotic and prebiotic that beneficially affects the host by improving
the survival and the implantation of live microbial dietary supplements in the gastrointestinal
tract, by selectively stimulating the growth and/or by activating the metabolism of one or a
limited number of health-promoting bacteria, including the ones in the synbiotic mixture [136].
Spray drying of probiotic bacteria 117
0
20
40
60
80
RSM PDX P95 RSM/PDX RSM/P95
0
10
20
30
40
50
60
70
Powder recovery (%)
b
Survival rate (%)
a
Figure 14
The effect of total or partial substitution of reference drying medium, i.e reconstituted skim milk (RSM)
by prebiotic substances Polydextrose or Raftilose®P95, on survival rate of L. rhamnosus GG (Fig 12a)
and on powder recovery (Fig. 12b). Figure 12a was adapted from [7].
The media were RSM, reconstituted skim milk powder (20%, w/v, total solids); Polydextrose (20%,
w/v, total solids); Raftilose®P95 (20%, w/v, total solids); RSM/Polydextrose, a 50:50 (v/v) blend of
RSM and reconstituted Polydextrose (both 20%, w/v, total solids) and RSM/Raftilose®P95, a 50:50
(v/v) blend of RSM and reconstituted Raftilose®P95 (both 20%, w/v, total solids). Data are the means
of three, or more, replicate spray drying trials; the error bars represent the standard deviations of the
mean.
As already mentioned above, fructan is considered as a good prebiotic compound and it is
known from the literature that this oligosaccharide is the one specifically accumulated by
various plants upon dehydration [45, 61]. Based on these positive characteristics of
prebiotics, it seems probable to evaluate the possibility to use commercial non-digestible
prebiotic preparation, especially the one containing fructan, as the drying medium for LGG so
as to allow a one-step production of dried synbiotic preparation. The prebiotic compounds
used in this study are Polydextrose (Danisco, Copenhagen, Denmark) and Raftilose®P95
(Orafti, Tienen, Belgium). Similar to the spray drying study using RSM as drying medium, the
total solids content of the spray drying medium was held constant at 20% (w/v).
Unfortunately, as can be seen in Figure 14, full substitution of RSM by any of the prebiotics
investigated resulted in poor survival characteristics (Fig. 14a, redrawn from [7]) and low
powder recovery (Fig. 14b). Such spray dried bacteria were reported to experience
considerable damage on cell membrane during spray-drying, as evidenced by the higher
Spray drying of probiotic bacteria 118
sensitivity to 5% NaCl [7]. Compared to spray drying using reconstituted skim milk as carrier,
the survival of LGG in polydextrose or Raftilose®P95 was more than 100 times lower (Fig.
14a). Lian et al (2002) also found that inclusion of prebiotic as the sole carriers during spray
drying did not afford protection to cells compared with RSM [104]. In contrast, a partial (50%)
substitution of milk solids by any of the prebiotics tested resulted in survival rates slightly
lower than the one obtained with RSM. Furthermore, the presence of skim milk solids in
spray drying media were proven to be essential in maintaining high powder recovery. Partial
substitution of milk solids by prebiotic brought about a powder recovery level of ca. 50%. This
yield resembles the one achieved when RSM was used as drying medium.
The poor powder yield during the production of prebiotic powder was caused by the higher
stickiness of these products on the walls of drying chamber and aerocyclone, which in turn is
highly influenced by the thermoplastic and hygroscopic nature of these materials. When the
feed solution is being atomized into the drying chamber at any axial position of along the wall
of drying chamber when the temperature becomes high enough a sticky material forms and
deposits there [137]. This problem can be reduced by building large diameter dryers to keep
hot parts of the wall outside the path of the sprayed particles. Alternatively, the application of
a wide-body drying chamber is suggested [90]. In this system, air goes in at the top of the
dryer and is discharged out the top after passing down the centre of the dryer and back up
the walls. This design is suggested to provide maximum flexibility in keeping walls of the
dryer cool and keeping the dry products off the walls, since the walls are relatively cold. In
addition, it has been proposed to modify Büchi spray dryer to a scraped surface chamber to
continuously remove deposits from the walls [137]. The authors also suggest the substituting
wall material (borosilicate glass – standard in Büchi spray dryer) with cast iron. The latter
material is capable of absorbing about twice and conducting about 50 times more energy
than borosilicate glass, which then render a more stable and uniform wall temperature. Apart
from these constructional options to increase the yield, the presence of skim milk in the
drying medium is critical in maintaining acceptable level of powder recovery. The pronounced
presence of sticking on drier wall when using prebiotics is thought to be influenced by their
glass transition temperatures, since stickiness and caking can be prevented when the
surface of the dried particle did not reach 10 to 20°C above glass transition temperatures
[91].
Own data on comparing the effect of drying media on survival rate and moisture content
during spray drying are shown in Figure 15. Accordingly, no substantial differences between
the evaluated media (RSM, RSM:Polydextrose 50:50, or RSM:Raftilose®P95 50:50) could be
observed regarding their protection capacity against dehydration at an outlet temperature of
80°C. These experimental data demonstrated that both prebiotic substances could be
Spray drying of probiotic bacteria 119
incorporated in the spray drying medium without any adverse impact on the survivability and
the residual moisture content.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
0
10
20
30
40
50
60
70
80
90
Survival rate (%)
RSM/Raftilose P95
RSM/Polydextrose
Moisture content (%, w/w)
RSM
Figure 15
Effect of media type on the survival of L. rhamnosus GG (grey bars) and moisture content of powder
(open bars) prepared by spray drying various media at an outlet temperature of 80°C.
The media were RSM, reconstituted skim milk powder (20%, w/v, total solids); RSM/Polydextrose, a
50:50 (v/v) blend of RSM and reconstituted Polydextrose (both 20%, w/v, total solids) and
RSM/Raftilose®P95, a 50:50 (v/v) blend of RSM and reconstituted Raftilose®P95 (both 20%, w/v, total
solids). Data are the means of three, or more, replicate spray drying trials; the error bars represent the
standard deviations of the mean.
The approach of only partially replacing milk solids against other carrier compound was also
followed in a previous work on using gum acacia as a constituent of spray drying medium
[105]. The authors did not give detailed explanation why they did not perform full
replacement of RSM, which was regarded as reference medium. However, it was shown that
probiotics dried in the presence of gum acacia showed improved storage stability and gave
better protection against low pH conditions compared to the one dried with RSM alone. It
was speculated that this beneficial effect may be due to encapsulation. Furthermore, apart
from their prebiotic properties, gum acacia seems was also found to give good protection
against H2O2-induced oxidative damage [105]. Other carrier material, starch, which was
thought to be effective for microencapsulation proved to be unsuitable for use as protective
matrix for Bifidobacterium strain during spray drying, storage and stress conditions [103].
3.4.4 Storage test at non refrigerated conditions
Data from literature showed that storage stability was inversely proportional to storage
temperature and storage at refrigerated temperature allowed higher shelf-life of spray dried
Spray drying of probiotic bacteria 120
probiotic bacteria [6, 105, 138]. However, refrigerated storage is expensive to both suppliers
and retailers of probiotic products and thus there is a need to produce probiotic products that
are stable at ambient temperature. Taking this consideration into account it was aimed in this
work to evaluate the survival characteristics of spray dried LGG upon storage at elevated
temperatures, i.e. 25 and 37°C. These temperatures are also suitable for accelerated storage
test, with help of which the storage effect could be determined in a feasible time frame (within
5-8 weeks).
In this study the storage test was performed without modification of atmosphere under a
constant relative humidity of 11%, which was found to be optimal for the maintenance of high
level of viability during the storage of dried bacteria [108]. This behaviour might be correlated
with the influence of aw or relative humidity on the rates of various chemical reactions which
might have deteriorative effects on cell constituents, as evidenced by Figure 16. It is obvious
that within an aw range between 0.1 to 0.2 the rates of different chemical reactions (non
enzymatic browning, non enzymatic hydrolysis, enzyme catalyzed reactions, fat oxidative
etc.) are very low. This range also corresponds with the relative humidity values of 11% to
23%, recommended by Castro et al (1995) for the stable storage of dried L. bulgaricus.
Storage at 0% humidity was found to be more detrimental for stability of dried bacteria [108],
presumably due to the increased rate of oxidation on polyunsaturated fatty acids of cell
membrane as hydrate shell surrounding fatty acids is lost [91, 139]. These considerations are
regarded as helpful in identifying and confirming the presence of an optimal storage condition
for dried bacteria with regards to relative humidity.
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00
Rate of deteriorative reactions (-)
aw(-)
Figure 16
Rate of deteriorative reactions, which may lead to food spoilage, as a function of water activity [139].
The potential sources of damage on food systems are: fat oxidation (1), non-enzymatic browning (2),
non-enzymatic hydrolysis (3), enzyme activity (4), growth of moulds (5), growth of yeast (6), growth of
bacteria (7)
Spray drying of probiotic bacteria 121
The different media, spray dried at an air outlet temperature of 80°C, were compared for the
effect on the viability of LGG during prolonged storage at 25 or 37°C in a relative humidity
value of 11%. Figures 17a and 17b show the loss of viability during storage at 25 and 37°C,
respectively. The decline of the bacterial load was represented by the logarithmic values of
the survival fractions after different storage periods. The loss of viability was accelerated at
higher storage temperature. This observation is quantitatively described in Table 5, in which
the inactivation rate constants (s in week-1) of the bacteria dried in different matrices are
presented. For the calculation of the inactivation rates, first order inactivation kinetics are
assumed.
01234567
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
log N/N0 (-)
Storage time at 25°C (week)
01234567
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
log N/N0 (-)
Storage at 37°C (week)
ab
Figure 17
Viability loss of spray dried L. rhamnosus GG during storage at 25°C (a) or 37°C (b) at a constant
relative humidity of 11%, which is expressed as the logarithmic values of relative survival fraction (log
N/N0), as described in Material and Methods. For storage stability experiments the bacteria were spray
dried at an air outlet temperature of 80°C in different media, as defined in Figure 15: RSM (),
RSM/Polydextrose (c), and RSM/Raftilose®P95 (V). Data are the means of three replicate spray
drying and storage trials; the lines through the data points were fitted using linear regression of the
data. The slopes of the regression lines (s) were considered as inactivation rates (in week-1) and
tabulated in Table 5 along with the corresponding correlation coefficient (R2).
At both storage temperatures protective medium supplemented with Raftilose®P95 showed
the poorest protection performance, whereas the one containing polydextrose was better at
37°C or even gave protection capacity equivalent to RSM at 25°C (Tab. 5). It can be
concluded that replacement of milk solids with any of the prebiotics tested facilitated the
degradation of crucial cell components, which ultimately led to higher viability loss during
storage compared to LGG dried with RSM alone.
Taking literature data into consideration, it was documented that membrane damage occurs
during storage despite the presence of skim milk solids as protectant and storage at optimal
relative humidity of 11% [12, 108]. The damage in membrane was highly related to lipid
Spray drying of probiotic bacteria 122
peroxidation, as expressed by a decrease in the ratio of the unsaturated to fatty acids [12].
This oxidative damage has at least two indirect consequences leading cell death: the product
of lipid peroxidation may lead to DNA damage [140] and alteration of membrane lipid
composition may cause dysfunction of membrane-associated enzymes, such as ATPase due
to a decrease in membrane fluidity or weakening of hydrophobic interactions [12]. The
implication of oxidative damage leading to membrane rupture and cell death during storage
was further demonstrated to be effectively suppressed by incorporating antioxidants [138].
Alternatively, modification in the gas composition of the storage atmosphere, i.e. by replacing
air with nitrogen or by applying vacuum proved to improve cell survival [108, 141]. In line with
these results it could be assumed that the partial substitution of milk solids with prebiotics
increased the susceptibility of fatty acids of cellular membrane to oxidative damage and that
components in skim milk solids should be more effective in conferring protection against lipid
oxidation. Moreover, it was found that inactivation of LGG in prebiotic containing drying
media was reduced in lower storage temperature (Fig. 17a). This approach seems feasible to
compensate the presumed increase of oxidation-induced viability loss upon incorporating
prebiotics at cost of skim milk solids. It was suggested before that increased survival of dried
bacterial culture at low temperatures might be due to a reduction of fatty acid oxidation [108].
Table 5
Inactivation rate constants (s in week-1) of L. rhamnosus GG in different spray drying media during
storage at 25 or 37°C.
Spray drying medium s25°C (week-1)a R
2b s
37°C (week-1) R2
RSM 0.087 0.892 0.115 0.972
RSM : Raftilose®P95 (1:1) 0.189 0.937 0.304 0.995
RSM : Polydextrose (1:1) 0.089 0.826 0.231 0.956
aThe slopes of the regression lines, as shown in Figure 3, were taken as the inactivation rates.
Bacteria were dried at an air outlet temperature of 80°C in reconstituted skim milk (RSM). Presented
data are the means of the inactivation rates obtained from three, or more, replicate storage
experiments.
bR2: correlation coefficient
Generally, for a shelf life period of one month only a slight loss of viability occurred in spray
dried skim milk (reduction of 0.25 log unit at 25°C). The high storage stability of probiotic
bacteria in spray dried skim milk at non-refrigerated temperatures showed that the bacteria
were sufficiently protected. This evidence justified the suitability of skim milk as a medium for
the large-scale production of shelf-stable spray dried probiotic bacteria, as recommended by
Spray drying of probiotic bacteria 123
Carvalho et al (2004), who suggested to use skim milk as drying medium unless a relevant
information of a specific culture of lactic acid bacteria is present [141]. However, synbiotic
powders showed reduced survival rate during storage at non-refrigerated temperature. The
negative effect resulted from partial substitution of milk solids with the evaluated prebiotics
could be most likely counteracted by storing this powder in refrigerated conditions.
3.4.5 The role of glassy state on bacterial storage stability
Entrapment of a living system in a glassy matrix upon dehydration was suggested as being
responsible for their long term stability [48, 142]. The glassy structure of the external matrix is
a highly effective environmental barrier with an extremely low molecular mobility. With
respect to the preservation of bacteria, this condition leads to a suppression of unexpected
deteriorating events on bacterial membranes, which constitute the interface to the
surroundings and are predominantly exposed to various environmental abuses. Lipid
oxidation of membrane fatty acid was deemed responsible for cell death during storage [11].
Since translational diffusion is drastically restricted in the glassy state [68], the diffusion of
oxygen, which preceded oxidative damage [74], and chemical reactions requiring diffusion
[48], could most likely be limited. Other degradative events such as fusion of membranes and
protein unfolding could also be prevented [56].
According to the results from storage test of spray dried probiotic (Fig. 17), it was found that
the viability retention of the spray dried LGG was markedly influenced by the composition of
the drying medium although a nearly identical level of viable count was determined
immediately after drying (Fig. 15). As mentioned above, the formation of glassy state is
thought to be essential in preventing deteriorative events during prolonged storage under a
controlled condition and to ensure good stability the dried materials should be stored under
their glass transition temperatures (Tg). Thus, it was therefore attempted to verify whether the
decrease in the protection performance of prebiotic supplemented RSM-based media was
related with a change in the glass forming capability as resulted from the substitution. In
particular, it was hypothesized that the partial substitution of the milk solids with prebiotic led
to a considerable reduction of the Tg of RSM to values lower than the storage temperatures
at 25 and 37°C. The Tg of RSM was found to be similar to the one of lactose (Tg = 101°C,
according to [124]), which makes up to more than 50% of the total solids contents the skim
milk. Under the hypothesized condition (Tg < Tstorage) the bacteria were not entrapped in a
glassy matrix but in the rubbery state, which are thought to be more susceptible to various
deleterious events, as already noted above.
Spray drying of probiotic bacteria 124
Figure 18 shows typical DSC- and TG-curves as obtained from thermal analysis of the
constituents of the drying media. All samples were scanned twice. Between the scans the
heated sample was rapidly cooled to –30°C. During the first scan, a broad endothermic peak,
which is characteristic for water evaporation, was observed in all media at temperatures
between 60 and 140°C. Within this particular temperature range, the thermogravimetric
signal, which was simultaneously measured, showed a significant mass loss. It was therefore
assumed, that moisture was removed from the sample during the first scan. This occurred in
all samples. No further mass loss was observed during the second scan, when the
dehydrated materials were heated up to 130°C. Moreover, it was also evident in all samples,
that an endothermic shift of the baseline of the DSC signal occurred during the second scan.
This latter phenomena is characteristic for glass transition. The key results of these
measurements, i.e. the glass transition temperatures (Tg) of the investigated media, are
summarized in Table 6.
Table 6
Glass transition temperatures (Tg) of the constituents of the drying media as determined by Differential
Scanning Calorimetry.
Spray drying medium na Mean value of Tg (°C) Range of Tg (°C)b
RSMc 2 109 5.8
Raftilose®P95 4 82 8.9
Polydextrose 4 102 11.2
RSMc: Raftilose®P95 (1:1) 1 102 -
RSMc : Polydextrose (1:1) 2 108 3.7
an: Number of measurements performed on each medium to determine Tg
bRange is defined as the difference between the largest and the smallest values
cReconstituted skim milk
Based on the experimental Tg data of the water-free spray dried formulations, a material-
specific state diagram was generated using the Gordon-Taylor equation (Equation 1). The
material specific constant, k, was derived from an empirical equation (Equation 2) proposed
by Roos et al (1993) [57]. The state diagram shows the effect of the residual water content of
the dried samples on Tg (Fig. 19a). It also serves as a stability map to ascertain specific
combinations of water content/storage temperature in the glassy state that are suitable for
stable storage. The prevalence of glassy state could only be assured as long as the samples
did not absorb moisture during storage (data not shown). Moisture uptake would decrease
Spray drying of probiotic bacteria 125
the Tg of the system, and consequently a second-order transition of the glass towards the
rubbery state (devitrification) could occur. As already noted, under this condition the
entrapped bacterial samples would be more susceptible to various deteriorating reactions. To
avoid moisture uptake, the bacterial samples were stored at a low relative humidity, i.e. 11%.
Glass transition
Onset : 106.1°C
Midpoint : 110.6°C
Offset : 115.0
Relative mass (%)
Temperature (°C)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 106.1°C
Midpoint : 110.6°C
Offset : 115.0
Relative mass (%)
Temperature (°C)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 94.0°C
Midpoint : 101.5°C
Offset : 109.1
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 94.0°C
Midpoint : 101.5°C
Offset : 109.1
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 94.0°C
Midpoint : 101.5°C
Offset : 109.1
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 87.8°C
Midpoint : 96.3°C
Offset : 104.8
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
TgGlass transition
Onset : 87.8°C
Midpoint : 96.3°C
Offset : 104.8
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
TgGlass transition
Onset : 87.8°C
Midpoint : 96.3°C
Offset : 104.8
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 68.2°C
Midpoint : 79.5°C
Offset : 90.9
Relative mass (%)
Temperature (°C)
Heat flow (mW mg-1)
TgGlass transition
Onset : 68.2°C
Midpoint : 79.5°C
Offset : 90.9
Relative mass (%)
Temperature (°C)
Heat flow (mW mg-1)
TgGlass transition
Onset : 68.2°C
Midpoint : 79.5°C
Offset : 90.9
Relative mass (%)
Temperature (°C)
Heat flow (mW mg-1)
Tg
Glass transition
Onset : 94.2°C
Midpoint : 101.5°C
Offset : 108.8
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
TgGlass transition
Onset : 94.2°C
Midpoint : 101.5°C
Offset : 108.8
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
TgGlass transition
Onset : 94.2°C
Midpoint : 101.5°C
Offset : 108.8
Temperature (°C)
Relative mass (%)
Heat flow (mW mg-1)
Tg
a
b
c
d
e
TG
TG
DSC
DSC
TG
TG
DSC
DSC
TG
TG
DSC
DSC
TG
TG
DSC
DSC
TG
TG
DSC
DSC
Figure 18
Typical thermograms showing DSC (Differential Scanning Calorimetry) and TG (Thermogravimetry)
signals of the constituents in different spray dried media: (a) reconstituted skim milk (RSM); (b)
Raftilose®P95; (c) Polydextrose; (d) RSM/Raftilose®P95; (f) RSM/Polydextrose.
The heat flow (DSC signals, left y-axis) and relative mass change (TG signals, right y-axis) of the
samples are plotted as a function of temperature. Thermograms obtained during the first and second
heating cycle are represented by the solid and dashed lines, respectively. See Material and Methods
for details of glass transition temperature (Tg) determination.
Figure 19b shows the comparison of the Tg data available for skim milk. It was found that the
data from Palzer & Zürcher (2004) [113] showed a marked deviation from the values
calculated with Gordon Taylor equation (∆T ~ 20°C) or from values obtained from other
Spray drying of probiotic bacteria 126
works [112, 124]. Commonly, the precision of Tg data for sugars, as calculated by DSC, was
approximated to be in a range between ±1 and ±5°C [50]. However, large deviation of Tg
values on a nearly identical material has already been observed on trehalose, where the Tg
values compiled from different works span between 73 and 115°C [143]. The apparent
discrepancies of Tg values could be attributed to impurities in the material as well as residual
water content [143]. Compared to Tg values of skim milk powder at different moisture level as
provided by Jouppila & Roos (1994) [124] and Vuataz (2002) [112], it can be observed that
the Tg values in this work, which were predicted with help of Gordon Taylor equation, were
qualitatively in an appropriate agreement. These considerations may justify the use of this
empirical equation to estimate the presence of glassy state in the spray drying media based
on their anhydrous Tg values, as measured by DSC (Tab. 6)
Commonly, spray drying at an air outlet temperature of 80°C results in a moisture content of
no higher than 4.5%. At this residual moisture content Tg values of 50.6, 49.5, and 44.5°C
were calculated using Gordon-Taylor equation for RSM, RSM/PDX and RSM/P95,
respectively. Since the calculated Tg values are higher than the applied storage temperatures
(25 and 37°C), it can be assumed that upon spray drying glassy state was achieved and thus
bacteria were stored in the glassy state. It has been proposed before that when a non-
crystallised milk concentrate is spray-dried, the lactose is rapidly solidified in an amorphous
solid structure [144], in which milk proteins are coated [112]. Nevertheless, the glass is in a
meta-stable state and will tend to convert to the crystal eventually, with a rate depending
upon temperature and moisture content [114]. Since relative humidity of the storage room
was held constant at 11%, there was no further uptake of moisture (moisture content at
equilibrium 3.5 % w/w) according to sorption isotherm of skim milk powder [125].
Consequently, an increase of moisture content, which would lead to a depression of Tg, could
be inhibited during the storage. Furthermore, no lactose crystallization could occur at the
applied storage conditions (Tstorage < Tg), since lactose can only crystallize at temperatures
above Tg [124]. It was reported in many studies that full crystallization of sugar led to phase
separation and thus to a loss of stabilization of the entrapped proteins [33, 35], thus making
the enzymes more susceptible to non enzymatic browning reactions [91].
However, when the bacterial inactivation rate constants during storage were compared,
differences in the protection performance of the drying media could be ascertained.
According to the data in Table 2, the protection capacity of RSM (s37°C 0.115 week-1) was the
highest, followed by RSM/PDX (s37°C 0.231 week-1). Probiotic bacteria dried in RSM/P95
showed the lowest storage stability (s37°C 0.309 week-1). A similar trend was observed at the
storage temperature of 25°C, suggesting that RSM was the most effective protective media
Spray drying of probiotic bacteria 127
(s25°C 0.087 week-1). In contrast, viability retention with RSM/P95 as drying media (s25°C 0.189
week-1) was poor .
0246810
-20
0
20
40
60
80
100
120
Temperature (°C)
Moisture content (%)
0246810
0
20
40
60
80
100
120
Glassy state
Temperature (°C)
Moisture content (%, w/w)
Rubbery state
ab
Figure 19
(a) Glass transition temperatures (Tg) of spray dried media at different moisture contents (0 to 10%,
w/w): RSM (-∆-), RSM/Polydextrose (-ο-), RSM/Raftilose®P95 (-□-); the composition of the media
is defined in Fig. 2. The Tg values of dry, moisture-free samples (total solids content, 100% w/w)
were experimentally determined and are tabulated in Table 2. Data for Tg for pure water (0% total
solids) were adopted from literature [57, 143]. The intermediate Tg values of the dual-phase
system were calculated with Gordon-Taylor equation, as defined in Material and Methods.
(b) Comparison of the Tg at different moisture contents as determined by different authors: this study
(solid line); Jouppila & Roos, 1994 () [124]; Vuataz, 2002 (S) [112] and Palzer & Zürcher, 2004
(z) [113].
While in both storage temperatures all media should theoretically be in the glassy state (Fig.
19a), differences in protection performance between the media was observed (Fig. 17).
Thus, it seems that the entrapment in a glassy matrix alone was not sufficient for maintaining
stability of the spray dried bacteria during storage. If the formation of glassy state alone were
indeed sufficient for preventing deleterious events from occurring, then no difference in the
protective capacity during storage would be seen. However, as obvious from the
incorporation of prebiotics at cost of skim milk solids the present data corroborate with
previous results which showed that the type of sugar constituting the glass apparently also
influenced the protection capacity conferred to the entrapped biological molecule [58].
Previous dehydration studies on microorganisms, enzymes and liposomes [26, 32, 36, 47,
86, 145] already demonstrated the occurrence of deteriorative events below Tg. The
inappropriateness of glassy state (Tg) to be considered as an absolute threshold of stability
may be explained by the possibility that the internal cytoplasm of the cell may not yet be a
glass even though the external solution has vitrified [76]. As the rate of water transport
across the plasma membrane may be less than the rate of water evaporation from the
Spray drying of probiotic bacteria 128
solution, there may be too much water inside of the cells to form a glass. After the outside
solution has vitrified, water transport across this vitrified layer may be extremely slow. The
integrity of the plasma membrane may thus not be adequately protected by the non-vitrified
internal cytoplasm and hence results in the long-term degradation of the membrane.
Similarly, the cytoplasmic content may also be subject to degradation due to the absence of
glassy state. Moreover, based on their investigation on instant active dry yeast Schebor et al
(2000) also suggested that biological materials could not regarded as homogeneous
materials; thus it is possible that local microheterogeneities exist [86]. This would lead to a
coexistence of glassy and rubbery states within the material [26] and thus an extrapolation of
the formation of glassy state to the whole volume of the biological material is not necessarily
appropriate. Furthermore, there are many examples in the literature indicating that molecular
motions do occur below Tg [146, 147], as opposed to the general view that translational
diffusion is considered to be virtually nonexistent in glassy state [68]. In the prebiotic
preparations applied in this work it is possible that the presence of impurities, i.e. low
molecular weights sugars might also contribute to the existence of microheterogeneities in
the dried samples.
Instead of or in conjunction to their capability of forming glass upon water removal, the use of
sugars as protectants of dehydrated biomaterials such as enzymes, proteins, liposomes, red
blood cells and bacteria can alternatively be explained by the water replacement hypothesis
[56], which envisages the function of sugars as water substitutes when the hydration shell of
proteins as well as water molecules around polar residues in membrane phospholipids are
removed. In terms of membrane stabilization the protective effect of sugar relies on a direct
physical interaction between the hydroxyl groups of the sugars and the polar residues of the
phospholipids head groups in dehydrated state so that phospholipids bilayers remain at their
hydrated spacing [21, 43, 44]. Adequate spacing between the lipid headgroups owing to the
insertion of sugar is deemed responsible for the substantial depression of liquid-crystalline-
to-gel phase transition temperatures (Tm), resulting in the preservation of membrane in a
liquid-crystalline state, even when dry. Consequently, the membrane would not pass through
a phase transition during rehydration and leakage of entrapped aqueous solution could be
prevented. Likewise, the capability of sugars to efficiently stabilize proteins during drying is
attributed to the formation of hydrogen bonds with the polar and charged groups of proteins
when water is removed [13, 39, 43]. This would then lead to the preservation of the native,
aqueous structure in the dried state [42, 66].
Data from microbiological analysis suggested, that although partial replacement of skim milk
in the carrier by prebiotic substance did not negatively impact on the spray drying survival of
Spray drying of probiotic bacteria 129
probiotic bacteria (Fig. 15), the protection performance during prolonged storage was lower
as the amount of skim milk solids in the carrier was reduced (Fig. 17a and 17b). In the light of
the water replacement theory [56], skim milk constituents, most likely lactose, seemed to be
more superior in directly interacting with the polar headgroups of membrane phospholipids,
and thereby minimizing the damage on the cellular membranes during spray drying and
prolonged storage. Principally, disaccharides were regarded as being effective in protecting
both bacterial membranes and proteins during drying [13]. In contrast, some polysaccharides
(such as dextran and hydroxyethyl starch) did not interact directly with the polar headgroups
of membrane phospholipids, and thus did not protect the membranes during drying [58]. The
absence of direct interaction has been attributed to the large size of the polymers, which
would sterically prevent them from interacting with membrane lipids [45, 46, 54] or with
proteins [148] although they commonly have high Tg, as also demonstrated in this study.
Thus, lactose might play a dominant role in bacterial protection by means of direct interaction
with sensitive biomolecules. Partial substitution of skim milk with oligosaccharides such as
Raftilose®P95 (degree of polymerization, DP = 2-8) or polydextrose (DP = 12 or more) was
thought to increase the amount of oligosaccharides incapable of stabilizing membrane at
cost of the net quantity of lactose, which may undergo direct interaction with either bacteria
membrane or proteins. As a consequence, bacterial stability during storage was adversely
impacted.
In order to substantiate this hypothesis, direct interaction of different types of sugar
molecules, especially the ones present in spray drying media, with phospholipid bilayers is
studied on liposomes.
3.4.6 Monitoring direct interaction of sugar-membranes using liposomes
Establishing the flow cytometry-based analytical procedure
Liposomes are regarded as a suitable model system for biological membranes, on which the
effect of drying can be investigated and how protective compounds, especially sugars, can
protect membranes upon dehydration [23]. The effect of drying can be assessed by
monitoring the leakage of aqueous fluorescent marker carboxyfluorescein (cF), which were
previously enclosed in the inner part of liposomes. This can be realized by measuring the
fluorescence of the suspending medium, in which cF is released using spectrofluorometry
[45, 54, 61, 63], or by determining the fluorescence retained in the liposome at single particle
level using flow cytometer, as applied in this study. Moreover, as drying may induce fusion or
aggregation of liposomes, determination of particle size might give indication of the
occurrence of this drying-related damage. This can be achieved by using particle size
analyzer which operates on the principle of dynamic light scattering, thereby allowing
measurement of the absolute liposome diameter [54, 59, 63]. The special case of membrane
Spray drying of probiotic bacteria 130
fusion is usually determined by fluorescence resonance energy transfer method [61, 62]. Due
to the possibility of simultaneous measurement of light scattering intensity using flow
cytometer, this technique is applied in this study to monitor relative changes in particle size,
as caused by dehydration. As already described in the section of Material and Methods, the
applicability of flow cytometric analysis for assessing drying effect on liposomes was
validated by using latex beads (validation of size) and liposomes with different concentration
of cF (validation of fluorescence intensity).
Before drying After drying
20 mg/mL sucrose 100 mg/mL sucrose
0 mg/mL sucrose
Figure 20
Flow cytometric fluorescence histograms showing the effect of drying of liposomes with different
concentrations of sucrose on distribution of cF fluorescence intensity (upper figures) and on particle
size distribution (lower figures), as examined using flow cytometric analysis. Liposomes (5 mg mL-1)
were dried in the presence of sucrose at various concentrations. The figures under the gate
designations indicate the percentage of liposomes encountered in those gates with respect to their
sizes or fluorescence intensities
Figure 20 visualizes the way of presenting and analyzing the data obtained from flow
cytometer. The distributions of fluorescence intensity (upper figure) and particle size (bottom
figure) were shown in the histograms. The manual fixation of gates (designated as “cF” and
“G” in fluorescence and particle size histograms, respectively) was performed on liposomes
prior to drying. The analysis software shows percentage of liposomes encountered in these
gated regions. This gate analysis allows the extraction of quantitative information about the
extent of damage experienced by liposomes during drying instead of qualitatively observing
Spray drying of probiotic bacteria 131
the appearance of histograms. A reduction of the percentage of liposomes in any of both
gates after drying and rehydration indicates that the original characteristics gated in those
regions, i.e. either size distribution or cF loading, could not be restored and would thus
illuminate the evidence of drying-induced damage.
As can be seen in Figure 20, marked changes were observed in both fluorescence and size
distributions following drying in the absence of sucrose; not only in the shrinking and
broadening of the fluorescence distribution but also a shift towards higher sizes in the
histogram corresponding to liposome size. These changes indicate the leakage of cF,
resulting in the marked decrease of liposomes with high fluorescence intensities and the
aggregation of several liposomes, resulting in increased number of liposomes with higher
size, which were previously not detectable. These indications of damage as manifested by
cF leakage and size increase, are in good agreement with results from other groups, who
mostly determined these changes with analytical methods other than flow cytometry [45, 61,
63]. Drying of liposomes in the presence of exogenous sugar markedly increased the
retention of cF in liposomes as well as reduced the fusion/aggregation events (Fig. 20). This
degree of improvement was found to be highly proportional to the amount of added sugar.
0 20406080100
0.5
0.6
0.7
0.8
0.9
1.0
cF fluorescence
Liposome size
cFd/cF0 or Sized/Size0 (-)
Sucrose concentration (mg/mL)
Figure 21
Effect of exogenously added sucrose on liposome’s integrity after drying in terms of the retention of
the original fluorescence distribution () and particle size distribution (T) in the corresponding gates
(see Figure 20). An index of 1 indicates that the size or fluorescence distribution before drying are
completely restored. Liposome concentration in the liposome-sugar mixture was as high as 5 mg mL-1.
Data were means of the results from three or more independent experiments.
Furthermore, based on the results obtained from this gate analysis, the percentage of
liposomes encountered in both gates “cF” or gate “G” (Fig. 20) was set in relation to the
Spray drying of probiotic bacteria 132
corresponding values prior to drying in the presence of various amounts of sugar, so as to
follow the magnitude of improvement in dependence on the exogenous sugar concentration
(Figure 21). This plot also enables the critical sugar concentration which allows good
prevention against cF leakage and/or aggregation to be determined.
It could be observed from Figure 21 that prevention of cF leakage and inhibition of liposome
fusion were more pronounced the higher amount of sucrose used as drying protectant. In
particular, upon addition of highest concentration of exogenous sucrose, i.e. at 100 mg mL-1,
which corresponds to a sucrose:EPC liposome mass ratio of 20:1, the original characteristics
of liposomes in terms of size and fluorescence intensity could nearly be restored. Other
studies reported the use of a sucrose:EPC liposome mass ratio of 11:1 [47, 53], or 17:1 [63]
or 20:1 [23, 54, 149] in order to precent cF leakage and membrane fusion. The air drying
procedure used in these studies reported that a residual water content of about 0.02 to 0.04
g H2O per g dry weight was achieved [23, 47, 54, 63].
The sucrose:EPC liposome ratio of 20:1 seems to be independent on the size or specific
surface area of liposome. In this study the size of liposome made was 1000 nm whereas in
other studies liposomes with smaller size, i.e. 100 nm [23, 47, 54, 63] were used. Tanaka et
al (1992) also used similar concentration of 100 mg mL-1 sucrose as used in this study to
stabilize 3 mg mL-1 sonicated EPC liposomes with diameter of 26 nm [62]. Taken together,
although the specific surface area (µm2 g-1) of liposomes of 100 nm is estimated to be 10
times higher than that of 1000 nm, the amount of sucrose required to efficiently stabilize
liposomes was in the same magnitude.
A general conclusion that can be drawn from this study is that sucrose is capable of
minimizing drying induced damage. Using FTIR spectroscopy it was observed that the sugar
OH groups interact directly through hydrogen bonding with the phosphate of the phospholipid
headgroups of liposome [63, 64]. This direct interaction result in the reduction of gel to liquid
lipid phase transition temperature Tm. In the absence of sugar Tm of dried liposome was
40°C, whereas sucrose lowered the dried Tm of liposome to 7°C, as observed by FTIR
spectroscopy of the CH2 symmetric stretch band of liposomes [149]. A substantial depression
of Tm, results in the preservation of membrane in a liquid-crystalline state, even when dry [21,
43]. Consequently, the membrane would not pass through a phase transition during
rehydration and leakage of entrapped aqueous solution could be prevented.
Furthermore it is obvious that fusion and leakage were closely related (Fig. 21). This close
relationship was also evidenced by other studies [54, 59].
It is noteworthy to suggest that the stabilization of liposomes during drying in the presence of
sucrose can be partially attributed to the ability of sucrose to form glassy state upon water
removal. The formation of glassy state could also explain the requirement of a high mass
ratio of sucrose to efficiently prevent drying induced leakage and fusion of liposome. The Tg
Spray drying of probiotic bacteria 133
of air-dried liposomal sample dried with sucrose was observed between 48 and 63°C [47].
Long-term preservation of liposomes in sugar matrix could be facilitated in glassy state.
Retention of cF in liposomes remained quite high after storage at temperatures below Tg, and
decreased remarkably above Tg. Likewise, fusion was inhibited below Tg but more
pronounced above Tg [47, 149].
020406080100
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
cFd/cF0 (-)
Sucrose concentration, outside (mg/mL)
Sized/Size0 (-)
Figure 22
The influence of exogenously added sucrose at various concentrations on liposome’s integrity in terms
of the recovery of the original fluorescence distribution (squares) and size distribution (circles) in the
corresponding gates (see Figure 20). Three different internal sucrose concentration were applied: 0
mg mL-1 (filled symbols), 40 mg mL-1 (crossed symbols), or 80 mg mL-1 (open symbols). An index of 1
indicates that the size or fluorescence distribution before drying are completely restored. Liposome
concentration in the liposome-sugar mixture was as high as 5 mg mL-1. Data were means of the
results from two independent experiments.
In addition, it was found that the presence of sucrose inside the liposomes did not have any
positive effect on either cF leakage or fusion of liposomes. With regards to prevention of
fusion, it was indeed expected that fusion would not be influenced by internal sugars [149]. In
terms of leakage prevention, it was observed that at high concentration of internal sugar (80
mg mL-1), the retention of cF is decreased; possibly due to harmful osmotic conditions.
According to this findings, it seems that only the outer side of phospholipid layer needs to be
stabilized by sugar in order to prevent the occurrence of detrimental effect as a result of
drying. In addition, since stabilization of internal phospholipid layer was not required, it is
possible that the inner part of liposomes was not dehydrated at all. The glassy matrix formed
upon dehydration may in the outer side of liposomes may cease the diffusion of water
molecule from the inner part of liposomes.
Spray drying of probiotic bacteria 134
The optimal sugar/liposome ratio (Fig. 21) could be taken as a theoretical basis in predicting
the critical sugar concentration required to efficiently protect bacterial membranes, for
instance lactic acid bacteria. Taking a usual bacterial cell concentration of 109 cFU mL-1 into
account, which corresponds to a dry matter of 10 mg mL-1, the minimal amount of exogenous
sugar required to confer protection was estimated to be 20%. This value correlates with
usual sugar concentration used to protect bacteria during spray drying (Tab. 1). However,
higher amounts of sugar was sometimes required. This can be explained by the fact that on
bacteria not only the membrane should be protected from but also functional proteins
embedded in bacterial membranes. The critical sugar concentration to maintain the integrity
of liposome upon drying could therefore be effectively used to give rough estimation about
the amount of protectant required to minimize drying induced damage on other organisms.
With help of this simple model system labour-intensive microbiological analysis to evaluate
the performance of a certain protectant as well as to determine its critical concentration could
be significantly reduced.
Stabilizing effect of sugars in the evaluated spray drying media on the integrity of
liposomes during drying
Having been establishing the analysis protocol using flow cytometry to investigate drying
effect on liposomes, the sugars present in the spray drying media was evaluated on their
stabilizing effect on liposomes during dehydration. Figure 23a shows the size distributions of
EPC liposomes before dehydration as well as after de- and rehydration in the presence of
lactose, polydextrose, Raftilose®P95 or in absence of sugar. In the case of drying without
sugar, an increase in liposome size was found, as evidenced by a broadening of the
histogram in the direction of higher liposome sizes. This behaviour was thought to be highly
related to fusion and/or aggregation of liposomes. In contrast, using an optimized sugar :
liposome ratio of 20:1, as previously determined (Fig. 21), the size distributions of liposomes
dried in the presence of all evaluated sugars were nearly identical to the one before
dehydration. This indicates that the addition of sugars could effectively prevent fusion and/or
aggregation (Fig. 23a).
Spray drying of probiotic bacteria 135
0 200 400 600 800 1000 1200
0
20
40
60
80
100
120
140
Number of events (-)
Rel. cF-fluorescence (-)
Dried w/o sugar
Prior to drying
Raftilose P95
Lactose
Polydextrose
0 200 400 600 800 1000 1200
-20
0
20
40
60
80
100
120
140
160
180
200
220
240
260
280
Dried w/o sugar
Prior to drying
Raftilose P95
Lactose
Polydextrose
Number of events (-)
Rel. particle size (-)
ab
Figure 23
Frequency histograms obtained from flow cytometric analysis showing the effect of drying with and
without different types of saccharides on important characteristics of liposomes, i.e. on particle size
distribution (a) and on distribution of fluorescence intensity (b). For this experiment, liposomes (5 mg
mL-1) were dried at ambient temperature in the presence of solution of each sugar type (100 mg mL-1)
prior to rehydration and measurement.
On the other hand, not all sugars tested had the good capacity in preventing leakage of cF
out of liposomes (Fig. 23b). In the absence of sugar there was only a small fraction of
liposomes which still had fluorescence intensity values equivalent to that before dehydration.
It is obvious that drying increased the permeability of the phospholipid bilayer, thus allowing
cF to leak from the liposomes. The numerical values for the drying experiments with sugars
present in spray drying media (Concentration 100 mg mL-1) are presented in Table 7. The
gating procedure described in the previous sub-section was also used to assess drying effect
on the distribution of liposome size and fluorescence intensity.
Table 7
Effect of sugar in spray drying media on liposome characteristics after drying
Sugar type Rel. regain of initial
fluorescence distribution ± SD
Rel. regain of initial liposome size
distribution ± SD
Without sugar 0.572 ± 0.0452 0.657 ± 0.0075
Lactose 0.686 ± 0.1618 0.994 ± 0.0093
Raftilose®P95 0.379 ± 0.0977 0.993 ± 0.0086
Polydextrose 0.821 ± 0.1467 0.989 ± 0.0141
SD: standard deviations of three or more replicates
The capacity of the sugars evaluated here to prevent cF leakage could be clearly
differentiated. Raftilose®P95 was found to be incapable of preventing loss of entrapped cF
Spray drying of probiotic bacteria 136
from liposomes. The loss of cF is reflected by the apparent shift of the fluorescence
distribution towards lower fluorescence values. In contrast, the stabilizing effect of either
lactose or polydextrose was better than that of Raftilose®P95 in terms of cF retention.
Despite good stabilizing properties the fluorescence distributions of liposomes dried together
with lactose or polydextrose shifted to values lower than the one before drying. This
behaviour indicates that – other than suppression of fusion/aggregation – leakage of cF
could not be fully prevented by the sugars evaluated in this study in the applied
concentration.
The positive effect of different types of disaccharides in minimizing drying induced damage
has already been substantiated [20, 23, 54, 59]. Lactose has not been investigated yet, but
the present result gives sound evidence about the positive effect of lactose to dried
liposomes. In contrast, the effect of oligo- or polysaccharides is not clearly understood yet.
Generally, polysaccharide, for instance hydroxyethyl starch, has elevated Tg. These sugars
are thus capable of inhibiting fusion between liposomes during drying but it does not depress
Tm in the dry phospholipids so that leakage was not prevented [54]. Polysaccharides are
thought to be sterically hindered from penetrating the bilayer in the dry state so that the gel to
liquid crystalline phase transition temperature was not depressed and leakage of entrapped
aqueous solution was not prevented [54]. However, other work reported that neither fusion
nor cF leakage could be prevented by hydroxyethyl starch during freeze drying [45].
Surprisingly, other polysaccharides such as inulins of a DP ranging between 10 and 30 from
chicory and dahlia could stabilize liposomes during freeze-drying and the stabilization is
mediated by a direct interaction of the polysaccharides with membrane lipids despite of the
proposed problem with steric hindrance [45] and even a high molecular mass bacterial levan
(DP > 25000) was able to directly interact with membranes [60]. Consequently, oligo- or
polymeric sugars may not be ruled out from being applied as drying protectant. It was found
that with increasing chain length, fructo-oligosaccharides were more effective than gluco-
oligosaccharides in stabilizing dried liposomes against leakage of aqueous content after
rehydration [61]. According to FTIR spectroscopy data it was observed that the ability of
gluco-oligosaccharides to hydrogen bond to the head group of dry lipids decreased
dramatically with increasing DP, whereas chain length hardly affected the ability of fructo-
oligosaccharides to interact. In contrast to the conclusion made by Hincha et al (2002), data
compiled in this study revealed that polydextrose (gluco-oligosaccharides, DP = 12 or more)
was more effective than Raftilose®P95 (oligofructose, DP = 2-8) in sufficiently protecting
dried liposomes against fusion and cF leakage. The role of impurities, i.e. mono- or
disaccharides, in polydextrose preparation, which may contribute to stabilization, could be
neglected, since according to the information provided by the manufacturer the degree of
Spray drying of probiotic bacteria 137
impurity is less than 10% (Annex 4 and 5), and it was shown before that high sugar to lipid
ratio was needed to render liposomes stable against damage during drying.
3.4.7 The role of milk constituents in the protection
It was shown that polydextrose and lactose were capable of adequately protecting liposomes
against drying-induced leakage of cF and fusion/aggregation. In contrast, Raftilose®P95
could only prevent fusion/aggregation but not cF leakage, as evidenced by the shift of the
distribution of cF fluorescence towards lower fluorescence values. It was noted before that
sugar may protect membranes by direct interaction with polar head-groups of phospholipids
[21, 43]. The maintenance of hydrated spacing upon sugar insertion leads to reduction of
membrane phase transition temperature; thus preventing phase transitions and leakage
during rehydration at room temperature.
The fact, that the degree of direct interaction and thus the protection performance of
polydextrose, a highly branched polysaccharide, and lactose, a simple disaccharide, are
similar, indicates that general concerns about the inefficiency of large sugar molecules in
protecting membranes due to the steric hindrance is unfounded. In the light of this result one
would expect that partial substitution of milk solids with polydextrose results in a protection
capacity equivalent to RSM or at least does not bring about any detrimental effect on
survivability during spray drying and subsequent storage. However, as already indicated by
Figure 17, it was evident that the performance of RSM/polydextrose in protecting LGG was
poorer than the one of RSM alone, especially during storage at 37°C. This result leads to
further consideration, whether the reduced protection performance of RSM/polydextrose was
caused by the exclusion of other milk components other than lactose. Typically the major
constituents of skim milk powder are lactose at ca. 52 %, w/w, followed by milk proteins with
ca. 37%, w/w [150]. The reduced availability of milk protein in RSM/polydextrose matrix was
suspected as the source of the decreasing protective capacity.
To evaluate this hypothesis, RSM was treated with a proteolytic enzyme preparation
Corolase®PP (ABEnyzmes, Darmstadt, Germany). The freshly prepared RSM was incubated
overnight with Corolase®PP at optimal temperature of 50°C, decontaminated at 90°C and
used as spray drying medium. Similarly, it was attempted to investigate the impact of
enzymatically degrading lactose in RSM with ß-galactosidase (G-3665, Sigma, St. Louis,
MO, USA) into glucose and galactose on protection performance of this modified RSM. It
was known that hydrolysis of lactose results in a marked decrease of Tg [114, 124].
Spray drying of probiotic bacteria 138
0123456
-4
-3
-2
-1
0
log N/N0 (-)
Storage time (week)
Figure 24
Effect of degrading skim milk components on the storage stability of L. rhamnosus GG spray dried
with the enzymatically modified protective media: native RSM (S), RSM pre-treated with 0.1%
Corolase®PP (Annex 6) overnight at 50°C () or with ß-galactosidase (Annex 7) at ambient
temperature for 2 h (z). Solids content was 20% (w/w). Spray drying experiment was conducted at an
outlet temperature of 80°C and the dried powders were subsequently stored at 37°C and 11% relative
humidity. Results are means of two independent drying and storage experiments.
It was shown in Figure 24 that the inactivation of LGG dried in proteolytic-treated RSM took
place with a higher rate during storage compared to native RSM. This result gives evidence
about the influence of protein degradation on the protection capacity of RSM.
A previous work had pointed out the importance of milk proteins as protective coating in
stabilizing dried yeast [151]. Similarly, it was reported that milk proteins might be involved in
protecting staphylococcal strains against thermal damage [115]. Positive role of milk proteins
was not only demonstrated on biological materials but also on volatile aroma compound.
Among the components of milk, it was reported that milk proteins seemed to have the
strongest positive effect, more than lactose, on retention of diacetyl during spray drying [152].
Moreover, as already discussed before, lipid oxidation is deemed responsible for death of
dried cells during storage [11, 12, 108]. Thus, the contribution of milk proteins to reduction of
cellular damage of dried bacteria during storage might involve inhibition of lipid oxidation.
The protective effect of milk protein was even more evident upon spray drying of LGG with
trehalose (20%, w/w) as drying medium, in which compared to RSM no proteins were
present. Trehalose is known to be a good drying protectant for different microorganisms [13,
71]. However survival of spray dried LGG during storage was very poor when trehalose as
drying medium. After 4 weeks of storage at identical conditions, i.e. at 37°C with 11% relative
humidity, a reduction by more than 5 log cycles was obtained (data not shown). From this
Spray drying of probiotic bacteria 139
result it could be concluded that the application of sugar alone is most likely not sufficient to
achieve an equivalent protection as obtained when skim milk is used. In addition, this
findings also emphasizes the superiority of skim milk as drying protectant for lactic acid
bacteria [141] and milk proteins contribute to this protective effect in a still unknown way.
Furthermore, compared to drying with native RSM the application of RSM treated with ß-
galactosidase (or RSM with hydrolyzed lactose) as spray drying medium led to a increased
rate of inactivation during storage. Due to the enzymatic degradation of lactose into glucose
and galactose, the anhydrous Tg of the RSM was reduced from 101°C to 49°C [91, 124]. At
5% water content, which is typical residual moisture for powder obtained during spray drying
at 80°C, the Tg of RSM with hydrolyzed lactose was –8°C [124], whereas the Tg of native
RSM was around 50°C (Fig. 19). Consequently, during storage at 37°C, the RSM with
hydrolyzed lactose was in rubbery state, while using native RSM the bacteria were entrapped
in a glassy matrix. It was concluded in sub-section 3.4.5 that the presence of bacteria in an
external glassy matrix was not sufficient to ensure good storage stability (Fig. 17 and 19).
However, the results of storage test of LGG in native RSM and RSM with hydrolyzed lactose
(Fig. 24) demonstrated that indeed the formation of glassy state which surrounded bacteria
facilitated higher survival rates. Furthermore, the higher amounts of reducing groups in sugar
molecules (glucose and galactose) in RSM with hydrolyzed lactose compared to native RSM
(lactose) could be regarded as an additional factor which also led to reduced stabilization of
LGG in RSM with hydrolyzed lactose.
123
Figure 25
Extent of browning occurring on spray dried RSM, which were subsequently oven dried overnight at
102°C. RSM were enzymatically treated with ß-galactosidase (1), or with Corolase®PP (2), prior to
spray drying at 80°C or not pre-treated (3). Solids concentration of all drying media was 20% (w/w).
It was reported the substantial decrease of the activity of dried alkaline phosphatase can be
ascribed to the low Tg of the used sugars in combination with the occurrence of the Maillard
reaction [32]. Reducing groups can react with the amine groups of the protein and this
reaction is the first of a cascade of reactions also known as the Maillard reaction. Since the
enzyme is the main component containing proteins, the reducing groups in sugars used to
protect the enzyme reacted with the enzyme, thus destabilizing instead of stabilizing it.
Spray drying of probiotic bacteria 140
Figure 25 documents the visual evidence of Maillard reaction in RSM treated with ß-
galactosidase, which was oven-dried at 102°C overnight. It was obvious that in samples
containing less reducing sugar groups (native RSM and RSM treated with Corolase®PP) the
browning intensity was far lower. When bacteria are dried and stored in the presence of
RSM, the reducing groups of sugar may most likely react either with milk proteins, thereby
modifying the physical properties of the matrix, thus destabilizing it, or with proteins
embedded on cell envelope, thereby inducing cellular injury. It can be speculated that both
events could lead to loss of viability during storage. The accelerated rate of inactivation of
LGG in RSM treated with ß-galactosidase could therefore be attributed to the greater extent
of this deteriorative reaction owing to higher amounts of reducing groups. Consequently the
absence of reducing groups is another requirement to be met by sugar molecules, when they
are to be applied as bacterial protective ingredient.
3.4.8 Role of milk constituents in conferring stability against low pH and bile acids
For probiotic bacteria in foods to be beneficial in the host, they should be able to survive
gastric transit, reach the small intestine in sufficient numbers and persist in this environment
to be effective [153]. The harsh environment of the gastro-intestinal tract is mainly attributed
to the low pH conditions of the stomach, in addition to the presence of bile in the small
intestine. In this study, the viability of spray dried LGG was assessed following exposure to in
vitro at 37°C. As already indicated above skim milk is superior in conferring protection
against adverse conditions encountered during spray drying and subsequent storage. The
protective effect of milk compounds was found to be related to effective direct interaction of
lactose with phospholipid bilayers, high Tg, reduced tendency for non-enzymatic browning
reaction and unknown protective action of milk.
Likewise, the specific degradation strategy was also applied to systematically evaluate the
role played by skim milk constituents in conferring protection against harsh environmental
conditions during passage of gastro-intestinal tract. Accordingly, RSM treated with
Corolase®PP was used as spray drying medium to elucidate the role of milk proteins,
whereas RSM treated with ß-galactosidase might give indication about the role of lactose.
The protocol for acid and bile stability test was adapted from the work of Saarela et al (2004)
[154]. As a model for the performance of the dried probiotic preparations during gastric
transit, the survival of the strain in each powder was investigated when exposed to PBS
adjusted to pH 2.0 for 2 h. The effect of drying media on bile tolerance was examined by
incubating spray dried LGG in PBS pH 7.0, which was supplemented with bile acid in an end
concentration of 1%. It was found that the highest resistance of LGG to acid and bile stress
was obtained when they were spray dried in native RSM (Fig. 26).
Spray drying of probiotic bacteria 141
-4
-3
-2
-1
0
1
log reduction (-)
AB CD
Figure 26
Survival rates of L. rhamnosus spray dried with different drying media following incubation at pH 2.0
for 2 h at 37°C (white bars) or following incubation in the presence of 1% bile acids for 3 h at 37°C
(grey bars). The spray drying media evaluated were native RSM (A), RSM previously treated with
Corolase®PP (C), RSM previously treated with ß-galactosidase (C), and trehalose (D). Bacteria
suspended in media with 20% solids content (w/w) were spray dried at an outlet temperature of 80°C.
Data are means of two independent spray drying and acid or bile challenge experiments.
When milk proteins were degraded, the acid and bile resistance were decreasing. Thus, the
presence of milk proteins exerted a major effect on survival in low pH conditions and in the
presence of bile. Similarly, when trehalose was used as drying medium, the bile resistance
was low, pointing out the important role of milk proteins in minimizing the toxic effect of bile
acids. In conclusion, it is evident that the contribution of milk proteins to acid and bile
resistance was far more crucial than the one of sugar. Data from literature suggest that
survival of lactic acid bacteria in human gastric juice adjusted to low pH was enhanced by the
addition of skim milk [155]. Other authors proposed that milk proteins function as buffering
agents in vivo, thereby protecting ingested bacterial strains during upper gastrointestinal
transit [153].
Furthermore, there is a slight decrease of cell count (ca. 1 log cycle reduction) following acid
or bile challenge when RSM with hydrolyzed lactose was used as drying medium. This
observation supports the aforementioned conclusion that in RSM primarily the milk proteins
and not the type of sugars rendered the bacteria more stable to acid and bile stress. With this
regard it was expected that upon bile or acid challenge the survival characteristics of LGG
dried in RSM with hydrolyzed lactose should be equal or even better than the one dried in
native RSM due to the presence of native proteins in both media. In terms of acid tolerance,
it is known, that a prerequisite for a good survival in acid environment is the presence of
Spray drying of probiotic bacteria 142
fermentable sugar in the medium (R.P. Ross, personal communication). It was reported that
F0F1-ATPase is involved in proton extrusion through ATP hydrolysis. Since LGG is not
capable of metabolizing lactose (Annex 1) but can utilize glucose [156], it was thought that
the enzyme degradation of lactose would improve the availability of fermentable sugar for
LGG, resulting in maintenance of a pH and a less viability loss in the acidic environment.
However, in native RSM, where no fermentable sugar was present, a higher acid tolerance
was obtained. Thus, this pH-homeostasis mechanism related to F0F1-ATPase activity was
most likely not involved in acid protection of LGG dried with native RSM or RSM with
hydrolyzed lactose. It seems probable to conclude that in these media, native milk proteins
which were still present, served as the buffering agents [153]. The viability loss of LGG dried
with RSM with hydrolyzed lactose during acid or bile challenge might therefore be most likely
explained by the higher amounts of reducing sugars compared to native RSM, which may
undergo unexpected reaction with proteins embedded in cell wall or cell membrane of LGG
during spray drying. As a result, some cells experienced cellular damage, which made them
more sensitive towards acid or bile stress.
3.5 Conclusion
This study demonstrated the possibility of producing dry probiotic bacterial preparation using
spray drying. Using reconstituted skim milk (RSM) as the drying medium, a bacterial survival
rate ≥ 50% was achievable at an outlet temperature of 80°C. The powder contained more
than 109 cfu g-1. Using flow cytometry, bacterial membranes were identified as the main site
of injury during spray drying. The incorporation of commercial prebiotic substances such as
Raftilose®P95 or polydextrose in the skim milk powder did not exert any adverse effect on
bacterial survival upon spray drying. However, stability of bacteria during long term storage
was impaired by partial substitution of skim milk with either of the prebiotic substances
evaluated. The presence of the dried medium in the glassy state appeared to have had only
little effect on bacterial stability in the spray dried powder. Although the glass transition
temperatures of all media were well above the applied storage temperatures, bacteria
inactivation still took place during storage; indicating the insufficiency of entrapment in glassy
state in inhibiting deteriorative events involved in cell death. The decreased protection
capacity of prebiotic containing media could be resulted from the reduced amount of
protective compounds in skim milk solids, which could not adequately substituted by
prebiotics. As a result, due to the presence of oxygen in the storage atmosphere applied in
this study, related deteriorative reaction, most likely lipid oxidation, may take place at a
higher rate.
Flow cytometric analysis was conducted on model membranes containing
carboxyfluorescein, in order to evaluate the contribution of sugar molecule to stabilization
Spray drying of probiotic bacteria 143
during drying. It was observed that oligosaccharides present in Raftilose®P95 was not
capable of directly interacting with cytoplasmic membranes in dehydrated state, whereas
polydextrose and lactose could effectively prevented drying induced membrane leakage.
Moreover, the implication of milk proteins in stabilization of dried bacteria during storage was
observed. When milk proteins were enzymatically degraded, the performance of such treated
RSM-based media in conferring protection during storage was considerably reduced. In
conclusion, the superiority of skim milk over the prebiotics in stabilizing dried bacteria is most
likely based on the direct interaction of lactose with bacterial membranes as well as proteins
and on protective effect of milk proteins. The higher susceptibility of bacteria dried in
prebiotic containing matrix to deteriorative events could be reduced by storing them at low
temperatures.
Apart from spray drying, another study was also conducted on the utilization of spray
generation-assisted processing in subzero atmosphere. Similar to spray drying, with this
approach it was envisaged to generate small droplets of bacteria suspension, where
extensive heat and mass transfer occur due to large surface to volume ratio. In contrast, the
process does not involve heated air to evaporate water, but due to running the process in a
subzero environment the formation of small ice crystals could be achieved. These are
collected in a container placed in bottom part of the freezer. A schematic drawing of the
experimental unit is shown in Figure 27. The pathway of the sprayed suspension is
furthermore protected with a “wave-breaker” in order to reduce turbulences resulted from the
air-blast of the freezer, so that the generated droplets can freely move down instead of being
circulated. Although not readily optimal, this construction already allowed reproducible
production of small ice crystals, pointing out that the crystallization heat could be sufficiently
removed by the cold air during the passage from the nozzle to the bottom part of the freezer.
More works are required in the construction of a “freezing zone”, in analogy to the drying
chamber in spray dryer, so as to realize forced, close-to-laminar flow of cold air along with
the generated droplets.
Spray drying of probiotic bacteria 144
Freezer
Cryostat
Feed pump
Feed solution
Silicon oil
Spray nozzle
Product collector
Pressure gauge
Air pressure
Thermometer
Thermometer
Display
Figure 27
Schematic view of the spray-freezing installation used for experiments on L. rhamnosus GG [157]. The
two-fluid nozzle from Büchi B-191 spray dryer is used to generate the spray together with pressurized
air (6 bar). The nozzle was heated (T > 10°C) using silicon oil flowing through heating jacket to prevent
frozen-induced clogging. The temperature of the freezer was set at –35°C.
First data on spray freezing of L. rhamnosus GG with the aforementioned installation have
already been complied. This rapid freezing procedure was found to be less detrimental to the
viability (as determined by plate count methods methods) or to the membrane integrity (as
measured by flow cytometry) of probiotic bacteria than freezing an equivalent volume of
bacterial suspension (data not shown). It seems that the improved survivability using spray
freezing process may result from higher freezing rate [158, 159] , thus reducing the size of
ice crystals and intracellular concentration effect owing to water removal from the cell [160,
161].
Based on the beneficial effects of high freezing rates on microbial viability spray freezing is
therefore a good candidate to be evaluated on its feasibility as the freezing method involved
in freeze-drying. In addition, it is known that in the conventional practice of freeze-drying a
cake is formed, which needs to be milled in order to obtain a free-flowing dried powder. With
help of the generation of small-sized frozen ice crystals the milling step may be fully
circumvented or applied with a low intensity. On the other hand, it is also possible to use
spray freezing to facilitate homogeneous incorporation of readily frozen probiotic bacteria into
ice cream matrix. The production of probiotic ice cream typically involves addition of
Spray drying of probiotic bacteria 145
suspension of probiotic bacteria in the ice cream mix prior to freezing them altogether [162,
163]. As already mentioned above the latter procedure – depending on the freezing
temperature – would not allow rapid freezing to be achieved, thus increase the probability of
losing a considerable amount of viable bacteria during the freezing step. These suggestions
have to be investigated and substantiated more intensively in further works.
3.6 References
1. Johnson, J.A.C. and Etzel, M.R. 1995. Properties of Lactobacillus helveticus CNRZ-32 attenuated by
spray-drying, freeze-drying, or freezing. Journal of Dairy Science. 78: 761-768.
2. Paul, E., Fages, J., Blanc, P., Goma, G., and Pareilleux, A. 1993. Survival of alginate-entrapped cells
of Azospirillum lipoferum during dehydration and storage in relations to water properties. Applied
Microbiology and Biotechnology. 40: 34-39.
3. Teixeira, P.M., Castro, H.P., and Kirby, R. 1995. Death kinetics of Lactobacillus bulgaricus in a spray
drying process. Journal of Food Protection. 57: 934-936.
4. Teixeira, P.M., Castro, H.P., and Kirby, R. 1995. Spray drying as a method for preparing concentrated
cultures of Lactobacillus bulgaricus. Journal of Applied Bacteriology. 78: 456-462.
5. Brennan, M., Wanismail, B., Johnson, M.C., and Ray, B. 1986. Cellular damage in dried Lactobacillus
acidophilus. Journal of Food Protection. 49: 47-53.
6. Gardiner, G.E., O'Sullivan, E., Kelly, J., Auty, M.A.E., Fitzgerald, G.F., Collins, J.K., Ross, R.P., and
Stanton, C. 2000. Comparative survival rates of human-derived probiotic Lactobacillus paracasei and L.
salivarius strains during heat treatment and spray drying. Applied and Environmental Microbiology. 66:
2605-2612.
7. Corcoran, B.M., Ross, R.P., Fitzgerald, G., Stanton, C. 2004. Comparative survival of probiotic
lactobacilli spray dried in the presence of prebiotic substances. Journal of Applied Microbiology. 96:
1024–1039.
8. Lievense, L.C., Verbeek, M.A.M., Noomen, A., and van't Riet, K. 1994. Mechanism of dehydration
inactivation of Lactobacillus plantarum. Applied Microbiology and Biotechnology. 41: 90-94.
9. Mauriello, G., Aponte, M., Andolfi, R., Moschetti, G., and Villani, F. 1999. Spray-drying of bacteriocin-
producing lactic acid bacteria. Journal of Food Protection. 62: 773-777.
10. Silva, J., Carvalho, A.S., Teixeira, P., and Gibbs, P.A. 2002. Bacteriocin production by spray-dried
lactic acid bacteria. Letters in Applied Microbiology. 34: 77-81.
11. Teixeira, P.M., Castro, H.P., and Kirby, R. 1996. Evidence of membrane lipid oxidation of spray-dried
Lactobacillus bulgaricus during storage. Letters in Applied Microbiology. 22: 34-38.
12. Castro, H.P., Teixeira, P.M., and Kirby, R. 1996. Changes in the cell membrane of Lactobacillus
bulgaricus during storage following freeze drying. Biotechnology Letters. 18: 99-104.
13. Leslie, S., Israeli, E., Lighthart, B., Crowe, J., and Crowe, L. 1995. Trehalose and sucrose protect
both membranes and proteins in intact bacteria during drying. Applied and Environmental Microbiology.
61: 3592-3597.
14. Castro, H.P., Teixeira, P.M., and Kirby, R. 1997. Evidence of membrane damage in Lactobacillus
bulgaricus following freeze drying. Journal of Applied Microbiology. 82: 87-94.
15. Crowe, L.M., Mouradian, R., Crowe, J.H., Jackson, S.A., and Womersley, C. 1984. Effects of
carbohydrates on membrane stability at low water activities. Biochimica et Biophysica Acta. 769: 141-
150.
Spray drying of probiotic bacteria 146
16. Crowe, J.H., Crowe, L.M., Carpenter, J.F., Rudolph, A.S., Wistrom, C.A., Spargo, B.J., and
Anchordoguy, T.J. 1988. Interactions of sugars with membranes. Biochimica Biophysica Acta. 947:
367-384.
17. Frezard, F. 1999. Liposomes: from biophysics to the design of peptide vaccines. Brazilian Journal of
Medical and Biological Research. 32: 181-189.
18. New, R.R.C., Liposomes - a practical approach. 1990, Oxford: Oxford University Press.
19. Beattie, G.M., Crowe, J.H., Lopez, A.D., Cirulli, V., Ricordi, C., and Hayek, A. 1997. Trehalose: A
cryoprotectant that enhances recovery and preserves function of human pancreatic islets after long term
storage. Diabetes. 46: 519-523.
20. Crowe, L.M., Crowe, J.H., Rudoph, A., Womersley, C., and Appel, L. 1985. Preservation of freeze-
dried liposomes by trehalose. Archives in Biochemistry and Biophysics. 242: 240-247.
21. Crowe, J.H., Crowe, L.M., and Carpenter, J.F. 1993. Preserving dry biomaterials: The water
replacement hypothesis, Part 1. BioPharm. 6: 28-33.
22. Crowe, J.H., Crowe, L.M., Carpenter, J.F., and Aurell-Wistrom, C. 1987. Stabilization of dry
phospholipid bilayers and proteins by sugars. Biochemical Journal. 242: 1-10.
23. Oliver, A.E., Crowe, L.M., Crowe,J.H. 1998. Methods for dehydration-tolerance: Depression of the
phase transition temperature in dry membranes and carbohydrate vitrification. Seed Science Research.
8: 211-221.
24. Crowe, L.M. and Crowe, J.H. 1982. Hydration-dependent hexagonal phase lipid in a biological
membrane. Archives of Biochemistry and Biophysics. 217: 582–587.
25. Winter, R. 2002. Synchrotron X-ray and neutron small-angle scattering of lyotropic lipid mesophases,
model biomembranes and proteins in solution at high pressure. Biochimica et Biophysica Acta. 1595:
160-184.
26. Mazzobre, M.F., Buera, M.P., and Chirife, J. 1997. Glass transition temperature and thermal stability of
lactase in low-moisture amorphous polymeric matrices. Biotechnology Progress. 13: 195-199.
27. Pikal-Cleland, K.A. and Carpenter, J.F. 2001. Lyophilization-induced protein denaturation in phosphate
buffer systems: Monomeric and tetrameric ß-Galactosidase. Journal of Pharmaceutical Sciences. 90:
1255-1268.
28. Burin, L., Jouppila, K., Roos, Y.H., J., K., and Buera, M.P. 2004. Retention of ß-galactosidase activity
as related to Maillard reaction, lactose crystallization, collapse and glass transition in low moisture whey
systems. International Dairy Journal. 14: 517-525.
29. Tzannis, S.T. and Prestrelski, S.J. 1999. Activity-stability considerations of trypsinogen during spray
drying: effects of sucrose. Journal of Pharmaceutical Sciences. 88: 351-359.
30. Lee, J.C. and Timasheff, S.N. 1981. The stabilization of proteins by sucrose. Journal Biological
Chemistry. 256: 7193-7201.
31. Lopez-Diez, E.C. and Bone, S. 2004. The interaction of trypsin with trehalose: an investigation of
protein preservation mechanisms. Biochimica et Biophysica Acta. 1673: 139-148.
32. Hinrichs, W.L.J., Prinsen, M.G., and Frijlink, H.W. 2001. Inulin glasses for the stabilization of
therapeutic proteins. International Journal of Pharmaceutics. 215: 163-174.
33. Terebiznik, M.R., Buera, M.P., and Pilosof, A.M.R. 1997. Thermal stability of dehydrated α-amylase in
trehalose matrices in relation to its phase transitions. Lebensmittel-Wissenschaft und -Technologie (lwt).
30: 513-518.
34. Colaco, C., Sen, S., Thangavelu, M., Pinder, S., and Roser, B. 1992. Extraordinary stability of
enzymes dried in trehalose: simplified molecular biology. Biotechnology. 10.
35. Rossi, S., Buera, M.P., Moreno, S., and Chirife, J. 1997. Stabilization of the restriction enzyme EcoRI
dried with trehalose and other selected glass-forming solutes. Biotechnology Progress. 13: 609-616.
Spray drying of probiotic bacteria 147
36. Buera, M.P., Rossi, S., Moreno, S., and Chirife, J. 1999. DSC confirmation that vitrification is not
necessary for stabilization of the restriction enzyme EcoRI dried with saccharides. Biotechnology
Progress. 15: 577-579.
37. Cardona, S., Schebor, C., Buera, M.P., Karel, M., and Chirife, J. 1997. Thermal stability of invertase in
reduced-moisture amorphous matrices in relation to glassy state and trehalose crystallization. Journal
Food Science. 62: 105-112.
38. Sola-Penna, M. and Meyer-Fernandes, J.R. 1998. Stabilization against thermal inactivation promoted
by sugars on enzyme structure and function: why is trehalose more effective than other sugars. Archives
of Biochemistry and Biophysics. 360: 10-14.
39. Carpenter, J.F. and Crowe, J.H. 1989. An infrared spectroscopy study of the interactions of
carbohydrates with dried proteins. Biochemistry. 28: 3916-3922.
40. Imamura, K., Ogawa, T., Sakiyama, T., and Nakanishi, K. 2003. Effects of types of sugar on the
stabilization of protein in the dried state. Journal of Pharmaceutical Sciences. 92: 266-274.
41. Auh, J.H., Kim, Y.R., Cornillon, P., Yoon, J., Yoo, S.H., and Park, K.H. 2003. Cryoprotection of
protein by highly concentrated branched oligosaccharides. International Journal of Food Science and
Technology. 38: 553-563.
42. Prestrelski, S.J., Tedeschi, N., Arakawa, T., and Carpenter, J.F. 1993. Dehydration-induced
conformational transitions in proteins and their inhibition by stabilizers. Biophysical Journal. 65: 661-671.
43. Crowe, J.H., Crowe, L.M., and Carpenter, J.F. 1993. Preserving dry biomaterials: The water
replacement hypothesis, Part 2. BioPharm. 6: 40-43.
44. Crowe, J.H. and Crowe, L.M. 2000. Preservation of mammalian cells - learning nature's tricks. Nature
America. 18: 145-146.
45. Hincha, D.K., Hellwege, E.M., Heyer, A.G., and Crowe, J.H. 2000. Plant fructans stabilize
phosphatidylcholine liposomes during freeze drying. European Journal of Biochemistry. 267: 535-540.
46. Crowe, J.H., Hoekstra, F.A. , Nguyen, K.H.N., Crowe, L.M. 1996. Is vitrification involved in depression
of the phase transition temperature in dry phospholipids? Biochimica et Biophysica Acta. 1280: 187-196.
47. Sun, W.Q., Leopold, A.C., Crowe, L.M., and Crowe, J.H. 1996. Stability of dry liposomes in sugar
glasses. Biophysical Journal. 70: 1769-1776.
48. Sun, W.Q. and Leopold, C. 1997. Cytoplasmic vitrification and survival of anhydrobiotic organisms.
Comparative biochemistry and physiology. 117A: 327-333.
49. Champion, D., le Meste, M., and Simatos, D. 2000. Towards an improved understanding of glass
transition and relaxation in foods: molecular mobility in the glass transition range. Trends in Food
Science and Technology. 11: 41-55.
50. Roos, Y. and Karel, M. 1990. Differential scanning calorimetry study of phase transitions affecting the
quality of dehydrated materials. Biotechnology Progress. 6: 159-163.
51. Franks, F. 1993. Solid aqueous solutions. Pure and Applied Chemistry. 65: 2527-2537.
52. Roos, Y.H. 1995. Glass-transition related physicochemical changes in foods. Food Technology. 48: 97-
102.
53. Crowe, J.H., Leslie, S.B., and Crowe, L.M. 1994. Is vitrification sufficient to preserve liposomes during
freeze-drying? Cryobiology. 31: 355-366.
54. Crowe, J.H., Oliver, A.E., Hoekstra, F.A., and Crowe, L.M. 1997. Stabilization of dry membranes by
mixtures of hydroxyethyl starch and glucose: the role of vitrification. Cryobiology. 35: 20-30.
55. Harrigan, P.R., Madden, T.D., and Cullis, P.R. 1990. Protection of liposomes during dehydration or
freezing. Chemistry and Physics of Lipids. 52: 139-149.
56. Crowe, J.H., Carpenter, J.P., and Crowe, L.M. 1998. The role of vitrification in anhydrobiosis. Annual
Review of Physiology. 60: 73-103.
Spray drying of probiotic bacteria 148
57. Roos, Y. 1993. Melting and glass transitions of low molecular weight carbohydrates. Carbohydrate
Research. 238: 39-48.
58. Crowe, L.M., Reid, D.S., and Crowe, J.H. 1996. Is trehalose special for preserving dry biomaterials?
Biophysical Journal. 71: 2087-2093.
59. Suzuki, T., Komatsu, H., and Miyajima, K. 1996. Effects of glucose and its oligomers on the stability of
freeze-dried liposomes. Biochimica et Biophysica Acta. 1278: 176-182.
60. Vereyken, I.J., Chupin, V., Demel, R.A., Smeekens, S.C.M., and de Kruijff, B. 2001. Fructans insert
between the headgroups of phospholipids. Biochimica et Biophysica Acta. 1510: 307-320.
61. Hincha, D.K., Zuther, E., Hellwege, E.M., and Heyer, A.G. 2002. Specific effects of fructo- and gluco-
oligosaccharides in the preservation of liposomes during drying. Glycobiology. 12: 103-110.
62. Tanaka, K., Takeda, T., Fujii, K., and Miyajima, K. 1992. Cryoprotective mechanism of saccharides on
freeze-drying of liposomes. Chemistry of Pharmaceutics Bulletin. 401: 1-5.
63. Wolkers, W.F., Oldenhof, H., Tablin, F., and Crowe, J.H. 2004. Preservation of dried liposomes in the
presence of sugar and phosphate. Biocimica et Biophysica Acta. 1661: 125-134.
64. Crowe, J.H., Carpenter, J.F., Crowe, L.M., and Anchordoguy, T.J. 1990. Are freezing and
dehydration similar stress vectors? A comparison of modes of interaction of stabilizing solutes with
biomolecules. Cryobiology. 27: 219-231.
65. Crowe, J.H., Crowe, L.M., Oliver, A.E., Tsvetkova, N., Wolkers, W., and Tablin, F. 2001. The
trehalose myth revisited: introduction to a symposium on stabilization of cells in the dry state.
Cryobiology. 43: 89-105.
66. Allison, S.D., Chang, B., Randolph, T.W., and Carpenter, J.F. 1999. Hydrogen bonding between
sugar and protein is responsible for inhibition of dehydration-induced protein unfolding. Archives of
Biochemistry and Biophysics. 365: 289-298.
67. O' Brien, J. 1996. Stability of trehalose, sucrose and glucose to nonenyzmatic browning in model
systems. Journal of Food Science. 61: 679-682.
68. Levine, H. and Slade, L. 1992. Another view of trehalose for drying and stabilizing biological materials.
BioPharm. 5: 36-40.
69. Mazzobre, M.F. and Buera, M.P. 1999. Combined effect of trehalose and cations on the thermal
resistance of ß-galactosidase in freeze-dried systems. Biochimica et Biophysica Acta. 1473: 337-343.
70. Schebor, C., Burin, L., Buera, M.P., and Chirife, J. 1999. Stability to hydrolysis and browning of
trehalose, sucrose and raffinose in low-moisture systems in relation to their use as protectants of dry
biomaterials. Lebensmittel-Wissenschaft und -Technologie (lwt). 32: 481-485.
71. Leslie, S.B., Teter, S.A., Crowe, L.M., and Crowe, J.H. 1994. Trehalose lowers membranes phase
transitions in dry yeast cells. Biochimica et Biophysica Acta. 1192: 7-13.
72. Carpenter, J.F., Crowe, J.H., and Arakawa, T. 1990. Comparison of solute-induced protein stabilization
in aqueous solution and in the frozen and dried states. Journal of Dairy Science. 73: 3627-3636.
73. Linders, L.J.M., Wolkers, W.F., Hoekstra, F.A., and van't Riet, K. 1997. Effect of added carbohydrates
on membrane phase behavior and survival of dried Lactobacillus plantarum. Cryobiology. 35: 31-40.
74. Andersen, A.B., Fog-Petersen, M.S., Larsen, H., and Skibsted, L.H. 1999. Storage stability of freeze-
dried starter cultures (Streptococcus thermophilus) as related to physical state of freezing matrix.
Lebensmittel-Wissenschaft und -Technologie (lwt). 32: 540-547.
75. Diniz-Mendes, L., Bernardes, E., de Araujo, P.S., Panek, A.D., and Paschoalin, V.M.F. 1999.
Preservation of frozen yeast cells by trehalose. Biotechnology and Bioengineering. 65: 572-578.
76. Chen, T., Acker, J.P., Eroglu, A., Cheley, S., Bayley, H., Fowler, A., and Toner, M. 2001. Beneficial
effect of intracellular trehalose on the membrane integrity of dried mammalian cells. Cryobiology. 43:
168-181.
Spray drying of probiotic bacteria 149
77. Wolkers, W.F., Walker, N.J., Tablin, F., and Crowe, J.H. 2001. Human platelets loaded with trehalose
survive freeze-drying. Cryobiology. 42: 79–87.
78. Hirasawa, R., Yokoigawa, K., Isobe, Y., and Kawai, H. 2001. Improving the freeze tolerance of baker's
yeast by loading with trehalose. Bioscience Biothechnology Biochemistry. 65: 522-526.
79. Welsh, D.T. and Herbert, R.A. 1999. Osmotically induced intracellular trehalose, but not glycine betaine
accumulation promotes desiccation tolerance in Escherichia coli. FEMS Microbiology Letters. 174: 57-
63.
80. de Castro, A., Bredholt, H., Strøm, A.R., and Tunnacliffe, A. 2000. Anhydrobiotic engineering of
gram-negative bacteria. Applied and Environmental Microbiology. 66: 4142-4144.
81. Bayley, H. 1997. Building doors into cells. Scientific American. 277: 42-47.
82. Russo, M.J., Bayley, H., and Toner, M. 1997. Reversible permeabilization of plasma membranes with
an engineered switchable pore. Nature Biotechnology. 15: 278-282.
83. Guo, N., Puhlev, I., Brown, D.R., Mansbridge, J., and Levine, F. 2000. Trehalose expression confers
desiccation tolerance on human cells. Nature Biotechnology. 18: 168–171.
84. Mussauer, H., Sukhorukov, V.L., and Zimmermann, U. 2001. Trehalose improves survival of
electrotransfected mammalian cells. Cytometry. 45: 161-169.
85. Hoekstra, F., Golovina, E.A., and Buitink, J. 2001. Mechanisms of plant desiccation tolerance. Trends
in Plant Science. 6: 431-438.
86. Schebor, C., Galvagno, M., del Pilar-Buera, M., and Chirife, J. 2000. Glass transition temperatures
and fermentative activity of heat-treated commercial active dry yeast. Biotechnology Progress. 16: 163-
168.
87. Mazzobre, M.F., Hough, G., and Buera, M.P. 2003. Phase transitions and functionality of enzymes and
yeasts in dehydrated matrices. Food Science and Technology International. 9: 163-172.
88. Cerrutti, P., de Huergo, M.S., Galvagno, M., Schebor, C., and Buera, M. 2000. Commercial baker´s
yeast stability as affected by intracellular content of trehalose, dehydration procedure and the physical
properties of external matrices. Applied Microbiology Biotechnology. 54: 575-580.
89. Masters, K., Spray Drying. 1991, Essex, UK: Longman Scientific & Technical and John Wiley & Sons Inc.
90. Mermelstein, N.H. 2001. Spray drying. Food Technology. 55: 92-95.
91. Roos, Y.H. 2002. Importance of glass transition and water activity to spray drying and stability of dairy
powders. Lait. 82: 475-484.
92. Knorr, D. 1998. Technology aspects related to microorganisms in functional foods. Trends in Food
Science and Technology. 9: 295-306.
93. Marcotte, M. 2001. Dehydration?- It's not so dry as all that! Le Monde alimentaire. 5: 20-22.
94. Bimbenet, J.-J., Schuck, P., Roignant, M., Brule, G., and Mejean, S. 2002. Heat balance of a
multistage spray-dryer: principles and example of application. Lait. 82: 541-551.
95. Kilara, A., Shahani, K.M., and Das, N.K. 1976. Effect of cryoprotective agents on freeze-drying and
storage of lactic cultures. Cultured Dairy Products Journal. 11.
96. Prajapati, J.B., Shah, R.K., and Dave, J.M. 1987. Survival of Lactobacillus acidophilus in blended-spray
dried acidophilus preparations. Australian Journal of Dairy Technology. 42: 17-21.
97. Kim, S.S. and Bhowmik, S.R. 1990. Survival of lactic acid bacteria during spray drying of plain yogurt.
Journal of Food Science. 55: 1008-1010,1048.
98. Espina, F. and Packard, V.S. 1979. Survival of Lactobacillus acidophilus in a spray-drying process.
Journal of Food Protection. 42: 149-152.
99. Johnson, J.A.C. and Etzel, M.R. 1993. Inactivation of lactic acid bacteria during spray drying, in Food
Dehydration, Barbosa-Canovas, G. and Okos, M.R., Editors. American Institute of Chemical
Engineering: New York. p. 98-107.
Spray drying of probiotic bacteria 150
100. To, B.C.S. and Etzel, M.R. 1997. Spray drying, freeze drying, or freezing of three different lactic acid
bacteria species. J. Food Sci. 62: 576-578.
101. Bielecka, M. and Majkowska, A. 2000. Effect of spray drying temperature of yoghurt on the survival of
starter cultures, moisture content and sensoric properties of yoghurt powder. Nahrung/Food. 44: 257-
260.
102. Desmond, C., Stanton, C., Fitzgerald, G.F., Collins, K., and Ross, R.P. 2001. Environmental
adaptation of probiotic lactobacilli towards improvement of performance during spray drying. International
Dairy Journal. 11: 801-808.
103. O'Riordan, K., Andrews, D., Buckle, K., and Conway, P. 2001. Evaluation of microencapsulation of a
Bifidobacterium strain with starch as anapproach to prolonging viability during storage. Journal of Applied
Microbiology. 91: 1059-1066.
104. Lian, W.-C., Hsiao, H.-C., and Chou, C.-C. 2002. Survival of bifidobacteria after spray-drying.
International Journal of Food Microbiology. 74: 79-86.
105. Desmond, C., Ross, R.P., O´Callaghan, E., Fitzgerald, G., and Stanton, C. 2002. Improved survival of
Lactobacillus paracasei NFCB 338 in spray-dried powders containing gum acacia. Journal of Applied
Microbiology. 93: 1003-1011.
106. Fávaro-Trindade, C.S. and Grosso, C.R.F. 2002. Microencapsulation of L. acidophilus (La-05) and B.
lactis (Bb-12) and evaluation of their survival at the pH values of the stomach and in bile. Journal of
Microencapsulation. 19: 485 - 494.
107. LMBG. 1981. Bestimmung des Wassergehalts von Magermilchpulver - L.02.06. Amtliche Sammlung von
Untersuchungsverfahren nach §35 LMBG.
108. Castro, H.P., Teixeira, P.M., and Kirby, R. 1995. Storage of lyophilized cultures of Lactobacillus
bulgaricus under different relative humidities and atmospheres. Applied Microbiology and Biotechnology.
44: 172-176.
109. Robinson, R.A. and Stokes, R.H., Electrolyte Solutions. 1959, London: Butterworth Scientific
Publications. 510.
110. Roos, Y. and Karel, M. 1991. Phase transitions of mixtures of amorphous polysaccharides and sugars.
Biotechnol. Prog. 7: 49-53.
111. Aguilera, J.M., Levi, G., and Karel, M. 1993. Effect of water content on the glass transition and caking
of fish protein hydrolyzates. Biotechnology Progress. 9: 651-654.
112. Vuataz, G. 2002. The phase diagram of milk: a new tool for optimising the drying process. Lait. 82: 485–
500.
113. Palzer, S. and Zürcher, U. 2004. Verfestigung im Griff - Berechnung des Glasübergangs komplexer
amorpher Lebensmittel - Teil 1. Lebensmitteltechnik. 36: 61-63.
114. Bhandari, B.R. and Howes, T. 1999. Implication of glass transition for the drying and stability of dried
foods. J. Food Engineering. 40: 71-79.
115. Daemen, A.L.H. and van der Stege, H.J. 1982. The destruction of enzymes and bacteria during spray
drying of milk and whey. 2. The effect of the drying conditions. Netherlands milk and dairy journal. 36:
211-229.
116. Re, M.I. 1998. Microencapsulation by spray-drying. Drying Technology. 16: 413-425.
117. LiCari, J.J. and Potter, N.N. 1970. Salmonella survival during spray drying and subsequent handling of
skim milk powders. II. Effect of drying conditions. Journal of Dairy Science. 53: 871-876.
118. Corry, J.E.L. 1975. The effect of water activity on the heat resistance of bacteria, in Water relations of
foods, Duckworth, R.B., Editor. Academic Press: London. p. 325-338.
119. Härnulv, B.G., Johansson, M., and Snygg, B.G. 1977. Heat resistance of Bacillus stearothermophilus
spores at different water activities. Journal of Food Science. 42: 91-93.
Spray drying of probiotic bacteria 151
120. Cerny, G. and Fink, A. 1986. Untersuchungen zur Abhängigkeit der thermischen Abtötung von
Mikroorganismen von Viskosität und Wasseraktivität der Erhitzungsmedien. Zeitschrift für Lebensmittel-
und Verfahrenstechnik. 2: 110-115.
121. Hardy, J., Scher, J., and Banon, S. 2002. Water activity and hydration of dairy powders. Lait. 82: 441–
452.
122. Masters, K. 1985. Analytical methods and properties of dried dairy products, in Evaporation, membrane
filtration and spray drying in milk powder and cheese production, Hansen, R., Editor. North European
Dairy Journal: Vanlose, Denmark. p. 393-403.
123. Kosanke, J.W., Osburn, R.M., Shuppe, G.I., and Smith, R.S. 1992. Slow rehydration improves the
recovery of dried bacterial populations. Canadian Journal of Microbiology. 38: 520–525.
124. Jouppila, K. and Roos, Y.H. 1994. Glass transitions and crystallization in milk powders. Journal of Dairy
Science. 77: 2907-2915.
125. Jouppila, K. and Roos, Y.H. 1994. Water sorption and time dependent phenomena of milk powders. J.
Dairy Science. 77: 1798-1808.
126. Peri, C. and De Cesari, L. 1974. Thermodynamics of water sorption on Sacc. cerevisae and cell viability
during spray-drying. Lebensmittel-Wissenschaft und -Technologie (lwt). 7: 76-81.
127. Gunasekera, T.S., Attfield, P.V., and Veal, D.A. 2000. A flow cytometry method for rapid detection and
enumeration of total bacteria in milk. Applied and Environmental Microbiology. 66: 1228-1232.
128. Holm, C., Mathiasen, T., and Jespersen, L. 2004. A flow cytometric technique for quantification and
differentiation of bacteria in bulk tank milk. Journal of Applied Microbiology. 97: 935-941.
129. Bunthof, C.J. and Abee, T. 2002. Development of a flow cytometric method to analyze subpopulations
of bacteria in probiotic products and dairy starters. Applied and Environmental Microbiology. 68: 2934-
2942.
130. Parthuisot, N., Catala, P., Lebaron, P., Clermont, D., and Bizet, C. 2003. A sensitive and rapid
method to determine the viability of freeze-dried bacterial cells. Letters in Applied Microbiology. 36: 412-
417.
131. Attfield, P.V., Kletsas, S., Veal, D.A., van Rooijen, R., and Bell, P.J.L. 2000. Use of flow cytometry to
monitor cell damage and predict fermentation activity of dried yeast. Journal of Applied Microbiology. 89:
207-214.
132. de Valdez, G.F., de Giori, G.S., de Ruiz Holgado, A.P., and Oliver, G. 1983. Comparative study of the
efficiency of some additives in protecting lactic acid bacteria against freeze-drying. Cryobiology. 20: 560-
566.
133. de Valdez, G.F., de Giori, G.S., de Ruiz Holgado, A.P., and Oliver, G. 1985. Effect of drying medium
on residual moisture content and viability of freeze-dried lactic acid bacteria. Applied and Environmental
Microbiology. 49: 413-415.
134. Champagne, C.P., Gardner, N., Brochu, E., and Beaulieu, Y. 1991. The freeze drying of lactic acid
bacteria. A review. Canadian Institute for Science and Technology Journal. 24: 118-125.
135. Gibson, G.R. and Roberfroid, M.B. 1995. Dietary modulation of the human colonic microbiota:
introducing the concept of prebiotics. Journal of Nutrition. 125: 1401-1412.
136. Roberfroid, M.B. 1998. Prebiotics and synbiotics: concepts and nutritional properties. British Journal of
Nutrition. 80: 197-202.
137. Karatas, S. and Esin, A. 1990. A laboratory scraped surface drying chamber for spray drying of tomato
paste. Lebensmittel-Wissenschaft und -Technologie (lwt). 23: 354-357.
138. Teixeira, P.C., Castro, M.H., Malcata, F.X., and Kirby, R.M. 1995. Survival of Lactobacillus delbrueckii
ssp. bulgaricus following spray-drying. Journal of Dairy Science. 78: 1025-1031.
139. SLMB. 1991. Wasseraktivität. Schweizer Lebensmittelbuch. Kapitel 64.
Spray drying of probiotic bacteria 152
140. Marnett, L.J., Hurd, H., Hollstein, M., Levin, D.E., Esterbauer, H., and Ames, B.N. 1985. Naturally
occurring carbonyl compounds are mutagens in Salmonella tester strain TA104. Mutation Research.
148: 25-34.
141. Carvalho, A.S., Silva, J., Ho, P., Teixeira, P., Malcata, F.X., and Gibbs, P. 2004. Relevant factors for
the preparation of freeze-dried lactic acid bacteria. International Dairy Journal. 14: 835-847.
142. Bruni, F. and Leopold, A.C. 1991. Glass transitions in soybean seed. Relevance to anhydrous biology.
Plant Physiology. 90: 660-663.
143. Chen, T., Fowler, A., and Toner, M. 2000. Literature review: Supplemented phase diagram of the
trehalose–water binary mixture. Cryobiology. 40: 277-282.
144. Walstra, P. and Jenness, R., Dairy Chemistry and Physics. 1984, New York: John Wiley & Sons.
145. Lodato, P., Segovia de Huergo, M., and Buera, M.P. 1999. Viability and thermal stability of a strain of
Saccharomyces cerevisae freeze-dried in different sugar and polymer matrices. Applied Microbiology
and Biotechnology. 52: 215-220.
146. Noel, T.R., Parker, R., and Ring, S.G. 1995. Kinetic processes in highly viscous, aqueus carbohydrate
liquids, in Food Macromolecules and Colloids, Dickinson, E. and Lorient, D., Editors. Royal Society of
Chemistry: Cambridge, U.K. p. 543-551.
147. Hancock, N. and Zografi, G. 1997. Characteristics and significance of the amorphous state in
pharmaceutical systems. Journal of Pharmaceutical Science. 86: 1-12.
148. Tanaka, R., Takeda, T., and Miyajima, K. 1991. Cryoprotective effect of saccharides on denaturation of
catalase by freeze drying. Chemical & Pharmaceutical Bulletin. 30.
149. Hincha, D.K., Zuther, E., and Heyer, A.G. 2003. The preservation of liposomes by raffinose family
oligosaccharides during drying is mediated by effects on fusion and lipid phase transitions. Biochimica et
Biophysica Acta. 1612: 172-177.
150. Anon. 2005. The Canadian Dairy Commission. www.milkingredients.ca.
151. Abadias, M., Benabarre, A., Teixido, N., Usall, J., and Vinas, I. 2001. Effect of freeze drying and
protectants on viability of the biocontrol yeast Candida sake. International Journal of Food Microbiology.
65: 173-182.
152. Senoussi, A., Dumoulin, E.D., and Berk, Z. 1995. Retention of diacetyl in milk during spray drying and
storage. J. Food Science. 60: 894-897.
153. Charteris, W.P., Kelly, P.M., Morelli, L., and Collins, J.K. 1998. Development and application of an in
vitro methodology to determine the transit tolerance of potentially probiotic Lactobacillus and
Bifidobacterium species in the upper human gastrointestinal tract. Journal of Applied Microbiology. 84.
154. Saarela, M., Rantala, M., Hallamaa, K., Nohynek, L., Virkajärvi, I., and Mättö, J. 2004. Stationary-
phase acid and heat treatments for improvement of the viability of probiotic lactobacilli and bifidobacteria.
Journal of Applied Microbiology. 96: 1205-1214.
155. Conway, P.L., Gorbach, S.L., and Goldin, B.R. 1987. Survival of lactic acid bacteria in the human
stomach and adhesion to intestinal cells. Journal of Dairy Science. 70: 1-12.
156. Saxelin, M. 1997. Lactobacillus GG - A human probiotic strain with thorough clinical documentation.
Food Reviews International. 13: 293-313.
157. Volkert, M. 2004. Konservierung von probiotischen Bakterien durch Sprühverfahren am Beispiel von
Lactobacillus rhamnosus, in Department of Food Biotechnology and Food Process Engineering. Berlin
University of Technology: Berlin. p. 73.
158. Tsvetkov, T. and Shishkova, I. 1982. Studies on the effects of low temperatures on lactic acid bacteria.
Cryobiology. 19: 211-214.
Spray drying of probiotic bacteria 153
159. Ryhänen, E.-L. 1991. Über den Einfluss der Gefriergeschwindigkeit auf Lebensfähigkeit und
Stoffwechselaktivität gefrorener und gefriergetrockneter Lactobacillus acidophilus Kulturen. Finnish
Journal of Dairy Science. 49: 14-36.
160. Karlsson, J.O.M. and Toner, M. 1996. Long-term storage of tissues by cryopreservation: critical issues.
Biomaterials. 17: 243-256.
161. Darvall, J.G.L. 2000. Preservation of microorganisms. Culture. 21: 1-5.
162. Hagen, M. and Narvhus, J.A. 1999. Production of ice cream containing probiotic bacteria.
Milchwissenschaft. 54: 265-268.
163. Godward, G. and Kailasapathy, K. 2003. Viability and survival of free, encapsulated and co-
encapsulated probiotic bacteria in ice cream. Milchwissenschaft. 58: 161-164.
154
4 PRESSURE INDUCED STRESS RESPONSE
Cross-adaptive stress response of pressure pre-treatment on probiotic bacteria:
Characterization and importance for production processes
Pressure induced response in probiotic bacteria 155
4.1 Introduction
Foods containing probiotic bacteria gain growing acceptance in broad communities. This rise
is stimulated by increasing health consciousness of people in industrial countries. On the
other hand due to published scientific evidences about the positive effects of probiotic
bacteria on human health the consumer was becoming more convinced. This tendency was
well documented by increasing market potential/market share of probiotic products among all
other functional foods [1, 2].
Beneficial effects of probiotics on human health are suggested to be mainly related to the
presence and activity of a high number of viable cells in the intestine. Therefore,
maintenance of high viability level and retention of physiological activity during processing
and storage are some of the proposed technological criteria, which need to be fulfilled by
potential bacteria with probiotic traits, when they are aimed to be applied in food products [3].
Moreover it is also demanded, that the consumed bacteria persist adverse conditions
throughout GI-tract without losing associated probiotic properties [4, 5].
Technologically relevant bacterial stress response
Stress-sensing system and defense mechanism of microorganisms were utilized to prepare
themselves in withstanding either harsh conditions or sudden environmental changes. In
response to these external disturbances specific metabolic processes of the treated cells
(transcription rates, translation products, metabolic pathway, etc.) are altered, resulting in
increased production of certain stress metabolites, which are involved in counteracting such
abnormalities in their environment; thus help the bacteria survive the deleterious conditions.
These survival mechanisms exhibited by bacteria when confronted to stress are generally
referred to as the stress response. The exploration of bacterial stress response to adverse
environmental conditions is motivated by basic scientific reasons but also by industrial and
safety aspect in food microbiology [6-8].
In particular, bacterial response against stress conditions was found to be related to
coordinated expression of genes which alter different cellular processes (cell division, DNA
metabolism, housekeeping, membrane composition, transport, etc.) and act in concert to
improve tolerance [9]. Regarding the type of external stimuli applied to induce stress
response, one survival mechanism is the adaptive response, that is, when cells are exposed
to a moderate level of stress, they acquire increased resistance to a subsequent exposure to
a more severe level of the same stress at lethal dose (homologous agents). When cells are
exposed to one stress they develop resistance, not only to that stress, but to other unrelated
stresses (heterologous agents). This mechanism is known as cross-protection [10].
Pressure induced response in probiotic bacteria 156
A specific stress that has been extensively examined is heat stress. A major group of stress
metabolites which are frequently associated with bacterial response to sub-lethal dose of
heat stress is referred to as heat shock proteins. An induction of genes corresponding to
these proteins on transcriptional level [11, 12] as well as an up-regulation in their synthesis
were observed, when the bacteria were incubated at a temperature significantly higher than
their normal growth temperature [13-20]. For example, heat shock proteins in Lactococcus
lactis subsp. lactis – immunologically related to DnaK and GroEL of E. coli – were induced
when temperature was shifted from 30 to 40 or 42°C [13, 14]. Generally, these stress
proteins maintain quality control of cellular proteins. They bind substrate protein in a transient
non-covalent manner and function by preventing denaturation of proteins (cleaving misfolded
and partially folded proteins), correctly refolding denatured proteins or removing denatured
proteins before they cause damage to the cells [6, 10, 21]. In terms of viability improvement
this heat inducible stress response resulted in higher survivability compared to untreated
population after subsequent exposure to lethal temperatures [10, 12, 13, 16, 19, 22-24].
Because a large number of other stress conditions induce the heat shock proteins or at least
the most abundant ones, this response is often termed the universal stress response [25]
The versatility of applying heat stress to exert cross-protective action against unrelated
stresses has been evaluated [10, 12, 19]. With help of inducible cross-protection the stress
induction procedure must not be necessarily the same as the lethal condition the bacteria
have to face afterwards – a fact that allows a higher degree of freedom in selecting the type
of shock inducer to be used. Although the induction of heat shock proteins by sub-lethal
stresses not related to heat was regarded to be less effective as by heat [26], there is a
growing body of evidence which substantiated the importance of cross-protection
phenomena. For instance, a brief pre-conditioning heat shock at 55°C can trigger increased
chaperone production in Lactobacillus johnsonii and provide significant cross-protection from
the stresses imposed during the production of frozen culture concentrates [11]. Heat
treatment before freezing of commercial baker’s yeast in dough could also improve their
freeze tolerance [27]. Heat adapted cells of Lactobacillus plantarum also showed increased
growth at pH 5 and in the presence of 6% NaCl [19].
On the other hand, upon exposure to UV irradiation, ethanol, certain heavy metals, hydrogen
peroxide and both alkaline and acidic pH conditions heat shock proteins could be induced as
well [6, 28]. Preincubation at a low temperature of 8°C led also to thermotolerance to a 52°C
challenge in L. lactis [29]. This latter study demonstrated a complex relationship between
cold and heat shock. Survival characteristics of Bifidobacterium adolescentis during heat
treatment at 55°C can not only be improved by previous adaptation at 47°C but also by pre-
incubation in 1.5% NaCl [12]. Furthermore, mild osmotic stress using 0.4 M NaCl could
Pressure induced response in probiotic bacteria 157
increase thermotolerance in Lactobacillus delbrueckii [23]. Pre-exposure to bile could not
only help Lactobacillus acidophilus cells in tolerating lethal bile concentration but also against
heat treatment [10]. Physiological linkage between unrelated stresses was shown by
proteomic study on L. lactis, which revealed an overlap between the heat and salt stress
responses, demonstrating that all salt stress-induced proteins, such as DnaK, GroEL and
GroES were the ones, which were rapidly induced (within 10 min) by sub-lethal heat stress at
43°C [15].
The stresses associated with the production, storage, and distribution of frozen, lyophilized,
or spray-dried bacterial culture concentrates can dramatically reduce their viability and
activity. The triggering of bacterial stress response was proposed to be one feasible
approach to maintain high viability level and retention of physiological activity during these
production stages [11, 14, 18, 22, 24, 30-33].
It was reported that when exponential phase cells of Lactobacillus bulgaricus were heat
shocked at 42°C prior to spray drying with an air outlet temperature of 80°C a significant
increase of their resistance to the process as compared to control cells was observed [30];
however, it was further indicated that the survival of heat shocked cells from exponential
growth phase did not reach the normal levels found with unshocked stationary phase cells.
Heat pre-treated Lactobacillus paracasei cells were reported to possess up to 18 fold greater
thermotolerance compared with control, when they were spray-dried at 95-105°C, while salt-
adapted cultures exhibited 16 fold greater viability than control [24]. When a pre-adaptation
step either with heat (50°C) or salt (0.6 M NaCl) was conducted prior to fluid bed drying of
Lactobacillus rhamnosus the viability was found to be significantly improved compared with
the unadapted control culture after storage at 30°C over a period of 14 weeks [18].
When frozen storage of bacteria was considered, data on pre-incubation of L. lactis at 8°C
led to development of cryotolerance, i.e. enhanced capacity to survive exposure to freezing
temperature of –20°C [29]. Cold shock at 10°C could significantly improve cryotolerance of L.
lactis for short periods of frozen storage at –20°C, but the protective effect was less marked
following longer storage period [32]. The improved cryotolerance was attributed to the
presence of a cold shock protein with a molecular weight of 6.3 kDa in the pre-stressed
sample. Adaptation of Streptococcus thermophilus at 20°C resulted in a 1000-fold increase
of survival after four freeze-thawing cycles compared to control group [34]. In this organism
several 7-kDa cold-induced proteins were identified as the major induced proteins after a
shift to 20°C.
Pressure induced response in probiotic bacteria 158
Pressure stress response on bacteria
Pressure effects on physical and biochemical processes are based on the fundamental
principle of Le Chatelier (Eq. 1),
RT
V
p
k
T
*
ln ∆
−=
∂
∂ Equation 1
where k is the rate constant, p is the pressure (bar), T is the absolute temperature (Kelvin), R
is the gas constant (mL bar K-1mol-1) and ∆V* is the apparent volume change of activation
(activation volume) and represents the difference in volume between the reactants and the
transition state.
Analogue to Arrhenius equation which describes the temperature-dependency of reaction
rate, a specific biochemical reaction might be accelerated, retarded or left unaffected
depending on how the system volume changes under exposure to elevated pressure. When
a reaction is accompanied by volume increase of activated complexes or end products, it is
inhibited by elevated pressure. When a reaction is accompanied by a volume decrease, it is
enhanced by elevated pressure. For instance, if a reaction is accompanied by a volume
decrease of 300 mL mol-1, it is enhanced more than 200 000-fold by applying a pressure of
1000 bar.
Volume changes can occur due to changes in water structure around proteins, nucleic acids,
ions and enzyme substrates [35]. In particular, protein subunit dissociation (multimeric
enzymes, ribosomes, cytoskeleton proteins, and proteins acting in signal transduction
pathway) is facilitated by elevated pressure since upon hydration of a protein, water
molecules are arranged in its vicinity more densely than in bulk water, leading to a volume
reduction of the system [36, 37]. Similar to the deteriorative effect of heat, the destruction of
hydrogen bonding in macromolecules was caused by decreased viscosity due to high
pressure [38].
However, the overall effects of pressure on metabolic processes in living organisms are
thought to be very complex. Even in the case of a well-known metabolic pathway such as
glycolysis, elevated hydrostatic pressure might result in enhanced, neutral or inhibitory
effects as a result of the variation in sign and size of the volume changes at each step. Thus,
even when the value of ∆V* for biochemical reactions is known, it is still difficult to predict
how elevated hydrostatic pressure would affect metabolic pathways or alter the pool sizes of
metabolites in living organisms [39].
Most of the studies on the impact of high hydrostatic pressure on changes in microbial
physiology are dedicated towards understanding of the molecular bases of life in deep-sea
high pressure environments, but there is an increasing interest in the exploitation of
biotechnological potential of the piezophiles (microbes preferring high pressure conditions)
Pressure induced response in probiotic bacteria 159
and on the stress response of microorganisms not coming from high pressure environments
[35, 37, 39, 40]. In terms of stress response, elevated hydrostatic pressure can strongly
influence gene and protein expression in both 1 atmosphere adapted and high pressure
adapted microorganism [40].
Continuous exposure of Escherichia coli to 55 MPa inhibited cell division but allowed the OD
of the culture to increase owing to cell filamentation and the doubling of the biomass
production [41]. It was speculated from the observed increase in OD following initial lag
phase – in absence of an increase in viable cell number – that an overlap exists between
inhibitory aspects of high pressure and other environmental stresses, for which E. coli has
evolved adaptive response. Long filamentous E. coli cells with segregated nucleoids were
observed after treatment at 75 MPa as resulted either by pressure induction of a putative cell
division inhibitor protein or by pressure induced denaturation of FtsZ molecule, which is
involved in septum formation [42]. In contrast, study performed with Lactobacillus
sanfransciscencis showed no filament formation during incubation under elevated pressure
[43].
With respect to the fact that OD can increase without any apparent increase in cell count it is
also interesting to evaluate, whether under these circumstances E. coli might enter the
viable-but-not-culturable state at elevated pressure, which is regarded as a survival strategy
in coping with harsh environmental conditions [44].
The study of Welch et al (1997) also revealed an increase in the relative rates of a set of heat
shock proteins on cells growing at elevated pressure. Proteins synthesized at increased rate
with higher pressure are defined as pressure induced proteins (PIP). The alteration in the
pattern of protein synthesis seemed to be important during adaptation and growth at a sub-
lethal pressure level. The magnitude of PIP induction correlated with the magnitude of
pressure shift (up to 100 MPa) in a barometer-like fashion. In particular, the PIP spots on 2D
electrophoretic gels were identified as classical heat shock proteins acting as chaperones
including GroEL, DnaK as well as proteins related to cold shock response and an unknown
protein of 15.6 kDa. One unique characteristic of the proteomic profile of cells grown under
elevated pressure is that high pressure induced more HSPs than most other conditions
outside of those which precisely mimic a heat shock response, while also inducing more
CSPs than most conditions outside of those which precisely mimic a cold shock response.
In a similar study on continuous exposure of L. sanfranciscencis to high pressure cold shock
proteins could be identified [43]. Compared to the work of Welch et al (1997) pressures
beyond 100 MPa were evaluated as well. Although some PIP were gradually increased with
rising pressure, the induction of several other PIP occurred only at a certain pressure level. A
striking difference to the pressure induced proteomic profile of E. coli is that the classical
Pressure induced response in probiotic bacteria 160
stress proteins (DnaK, DnaJ, GroES) did not belong to the major PIP found in L.
sanfranciscencis. Furthermore, the fact that the protein expression effect had a maximum
after a pressure treatment of 150 MPa 1h, and was continuously decreasing upon application
of higher pressures, indicated that this process was not governed by Le Chatelier principle,
otherwise the effect would reach a steady-state due to the preferred volume decrease under
pressure. In this respect, the implication of signal cascades which turn on the expression of
proteins need to be investigated [43].
Since the alteration in the pattern of protein synthesis seemed to be important during
adaptation and growth at a sub-lethal pressure level, works have been performed in order to
unravel the contribution of the overexpressed proteins to pressure resistance as well as to
the acquisition of cross-protection.
Molecular characterization of pressure-resistant mutants of E. coli and Listeria
monocytogenes emphasized the importance of protein management, especially the role of
heat shock proteins, in withstanding extremely high pressures [45, 46]. In E. coli, pressure
resistance was found to correlate with level of dnaK expression and the heat shock proteins
were suggested to prevent cellular damage and/or aid cell recovery [46].
Related study on yeast indicated that molecular chaperons hsp 104 and hsc 70 were found
to confer tolerance on S. cerevisae towards the damage caused by hydrostatic pressure and
heat, although the accumulation of trehalose was found to be more important for the
acquisition of barotolerance [47-51].
In the contrary, it is reported that heat shock proteins are not considered to cause
barotolerance at 300 MPa in L. sanfranciscencis, since all kinds of stress inducers (high
pressure, acidic, osmotic, cold, starvation) except for heat could induce barotolerance [52].
This observation revealed that adaptive mechanism different than the general stress
response was responsible for increased barotolerance.
The contradictory findings could be explained by the fact that in comparison to the
application of sub-lethal heat treatment, the signal for heat shock induction is generated only
slowly by exposure to high pressure [41]. Slow or weak induction of heat shock proteins
during brief exposure to growth-inhibiting level of pressure could lead to failure to induce heat
or pressure resistance in S. cerevisae and E. coli by pressure shock [46, 53].
On the other hand, high hydrostatic pressure of 50-75 MPa induced synthesis of heat shock
protein (hsp104) and tolerance against various stresses such as high temperature, high
pressure and high concentration of ethanol [54]. Similarly, high pressure pre-treatment at 80
MPa could induce tolerance to lethal dose of heat and high pressure in L. sanfranciscencis
[52].
Pressure induced response in probiotic bacteria 161
Apart from the synthesis of heat shock proteins, some cold shock proteins were also induced
by high pressure [41, 55]. Both low temperature and high pressure inhibit an early step of
translation, and the resulted cold shock response was principally an adaptive response to
facilitate gene expression in translation initiation [56]. According to the data from
Wemekamp-Kamphuis et al (2002) cold shock proteins with a molecular size ranged at 7-
kDa might be involved in adaptation to both low temperature and pressure treatment. Owing
to the induction of CSPs during a cold-shock treatment at 10°C for 4 h cells of L.
monocytogenes can better survive subsequent high pressure treatment. Furthermore,
survivors of high pressure treated Staphylococcus aureus were approximately two log cycles
higher when cells were cold shocked at a temperature of –20°C for 15 min [57].
Furthermore, the adaptive response of microorganism towards high pressure was similar to
the one taking place during cold adaptation, since both high pressure and low temperature
reduce the membrane fluidity, which perturbs membrane associated processes, including
transmembrane ion and nutrient flux as well as DNA replication [58, 59]. To maintain proper
fluidity a general shift in the lipid composition of bacterial membrane from saturated to
unsaturated fatty acid was observed in response to an increase in growth pressure [60]
Genomic studies have been also conducted to identify changes in gene transcription of
Saccharomyces cerevisae after hydrostatic pressure treatment by whole genome DNA
microarray hybridization [61, 62]. The result of the hierarchical clustering analysis of genome
wide expression profiles indicated that pressure shock response (180 MPa for 4 min) was
highly similar to that caused by detergents, oils and freezing/thawing cycle. These kinds of
stress caused damage to the membrane structure and/or cell organelles [61]. Genes being
upregulated in response to the sensed damage were proposed to be involved in repair of
cellular organelles or in the degradation and removal of damaged proteins. Functional
classification of 274 genes affected by pressure treatment for 30 min showed that the
upregulated genes were involved in stress defense and carbohydrate metabolism while most
of the repressed ones were in cell cycle progression and protein synthesis categories [62].
However the upregulation of some uncharacterized genes clearly demonstrated a pressure-
specific stress response pattern.
With regards to signalling pathway leading to pressure stress response Figure 1 shows
schematically the commonalities of between high pressure effect and both the effect of
decreases and increases in temperatures. This shared effect is of utmost importance in the
signalling pathway, through which high pressure could induce the synthesis of heat shock
proteins. Either high pressure, high or low temperature can destabilize the tertiary and
quaternary structure of proteins. Thus, pressure-induced increases in the proportion of
Pressure induced response in probiotic bacteria 162
misfolded proteins or dissociated subunits could induce a σ32 factor dependent (in gram
negative bacteria) or a CIRCE/HrcA-regulon dependent (in gram positive bacteria) heat
shock response and trigger the synthesis of heat shock proteins [63, 64].
Figure 1
Schematic representation of the elliptic phase diagram of proteins. The arrows marked by the letters p,
h, c show the specific denaturation ways known as pressure, heat and cold denaturation, respectively
[65].
Alternatively, the signalling pathway was proposed to be based on its inhibitory effects on
ribosome assembly or function [66]. It is known, that the prokaryotic ribosome has been
implicated to act as a sensor for both heat and cold shock response networks [25]. Indeed,
since cell death upon exposure to ultra high pressure
Pressure induced response in probiotic bacteria 163
was associated with ribosome damage [67], the fact that stresses the implication of ribosome
as temperature sensors, may lead to believe that cold shock and heat shock proteins play a
role in stress response under exposure to elevated pressures [6].
Furthermore, it was reported that although the induction of several heat shock promoters can
be induced by pressure, the genetic response of E. coli upon pressure treatment did not
appear to be a DNA damage response usually known as SOS response [46, 68]. This is
strongly different to the typical SOS response as induced by to DNA-damaging treatment.
4.2 Objective
This work aims to evaluate, whether and to which extent heat tolerance of Lactobacillus
rhamnosus GG is affected by mild pressure treatments prior to exposure to lethal
temperature so as to give evidence to the presence of technologically important cross-
protective stress response of high pressure. Cross-protective action of pressure especially
against heat needs to be extensively investigated to allow better justification regarding its
applicability in assisting probiotic production, where in case of implementation of spray-drying
the lethal effect of high temperature needs to be overcome.
The effect of incubation at elevated pressure was evaluated by means of monitoring their
post-pressure growth characteristics. Kinetic analysis of the heat inactivation curves at 60°C
of variously pre-treated cells was performed in order to identify the optimal combination of
pressure and temperature required in the pre-treatment phase. The role of de novo protein
synthesis on acquisition of pressure induced heat tolerance was evaluated. Furthermore,
flow cytometric analysis combined with multiple staining strategy was applied so as to have
additional insights in the physiological changes affected by pressure pre-treatment. Other
cross-protective characteristics of pressure treated cells in tolerating bile acids, nisin and
spray drying conditions was assessed as well.
4.3 Material and methods
4.3.1 Test organism
Lactobacillus rhamnosus GG (ATCC 53103) – thereafter abbreviated with LGG – was
obtained from Valio R&D, Helsinki, FL. The culture was sent in freeze-dried form in glass
ampoule. For long-term maintenance LGG was stored as glass bead cultures (Roti-Store,
Carl-Roth, Karlsruhe, D) in a -80°C freezer (U101, New Brunswick Scientific, Nürtingen, D).
4.3.2 Preparation of bacterial suspension
A deep-frozen culture of LGG was transferred into MRS broth (Oxoid, Basingstoke, UK) and
incubated overnight at 37°C. A second MRS broth was then inoculated from the overnight
Pressure induced response in probiotic bacteria 164
culture under adjustment of the optical density value to 0.1. determined at 600 nm
(Graphicord UV-240, Shimadzu, JPN). The culture was incubated at 37°C up to until an OD
value of 0.5, which corresponded to a growth stadium in the exponential phase with a cell
concentration in the range of 2⋅107 – 3⋅107 cfu/mL (Fig. 2).
0 5 10 15 20 25
0
1
2
3
4
5
6
7
1E7
1E8
1E9
OD600nm, (-); pH,(-)
Cultivation time (h)
Cell count, (CFU/mL)
Figure 2
Growth behavior of L. rhamnosus GG as determined by 3 independent measurement methods : cell
count on MRS agar, optical density (OD) of cell suspension at 600 nm, and pH of cell suspension.
Cells were grown at 37°C and the initial optical density value was adjusted at 0.1.
4.3.3 High pressure treatment
The high pressure unit (U111, Unipress, Warsaw, PL), as schematically drawn in Figure 3,
was developed to meet the need of conducting kinetic studies up to pressures of 700 MPa
and a wide temperature range between –40°C and 100°C [69]. This unit consists of five
pressure chambers, which are separated from each other via high pressure valves. The
chambers are immersed in a water bath equipped with a thermostat. This design allows a
simultaneous treatment of five different samples in one pressure build-up step at close to
isothermal conditions. Each chamber is equipped with a K-type thermocouple and a pressure
sensor to monitor the temperature and pressure history of each sample during the treatment
cycle.
The exponentially grown culture of LGG (OD 0.5) was filled into sample containers (Nunc
Cryo Tubes Nr. 375299, Nunc A/S, Roskilde, DK) and subjected to pressure treatments at
different temperatures (25 to 50°C) and various pressure level (100 to 300 MPa).
Pressure induced response in probiotic bacteria 165
Figure 3
Schematic hydraulic diagram of multivessel high pressure apparatus U111. The intensifier is
connected with the pressure vessels through high pressure valves (1-5). The multiplication factor
(~11) of the intensifier leads to a maximum pressure of 700 MPa. Valves 6-11 are used for loading
and unloading the pressure medium (silicon oil).
4.3.4 Assessment of growth behavior after pressure treatment
Cells were grown according to aforementioned standard protocol to OD 0.5. The suspension
was then subjected to pressure treatments for 5 min at 50, 100 and 200 MPa as well as at
100 MPa for 10 and 20 min. Only treatment temperature of 37°C was evaluated. Following
pressure treatments 1 mL of treated suspension was inoculated into 50 mL of fresh MRS
broth. The bottles were then incubated at 37°C. Continuous measurement of OD was
performed until stationary growth phase was reached.
4.3.5 Lethal heat challenge at 60°C
A pressure, temperature and time combination in the range of 100-200 MPa, 37-50°C, 5-10
min, respectively, was investigated in order to determine optimal pre-treatment conditions for
the induction of heat tolerance. 0.5 mL of pre-treated and untreated sample were distributed
into previously sterilized glass tubes. Cells were exposed to a temperature of 60°C in water
bath for 1.5, 3, 5, 7 and 10 min. Following the heat treatment the samples were immediately
stored in ice bad.
To determine the role of protein synthesis in pressure induced thermotolerance, 10 µg mL-1
chloramphenicol was added into the treatment medium prior to pressure treatment and
challenge to 60°C. In this concentration chloramphenicol is known to be not lethal to
lactobacilli [23]
Pressure induced response in probiotic bacteria 166
The pre-treatment condition was further assessed in terms of minimizing the duration of
pressure treatment, so as to evaluate the possibility of pressurizing without any holding time
in the manner of high pressure homogenization. Subsequent to pressure pre-treatment at
with various pressure holding time (1 to 10 min) LGG cells were exposed to heat challenge at
60°C for 3 min.
4.3.6 Plate enumeration method
Samples from pressure and heat inactivation experiments were serially diluted in Ringer’s
solution (No. 15525, MERCK, Darmstadt, D) and plated in duplicate on MRS agar without
additional surface layer (Oxoid, Basingstoke, UK). Plates were placed in an anaerobic jar
under anaerobic atmosphere, which was produced by an anaerobic kit (AnaerocultA, Merck,
Darmstadt, D). The viable cell numbers were determined after 48h of incubation at 37°C.
The heat inactivation data were expressed as logarithmic value of relative survivor fraction
(log N/N0), which reflected the magnitude of thermal death of either pressure-treated or
untreated cells at 60°C. N refers to the bacterial count following heat challenge at a particular
holding time, whereas N0 represents the initial count prior to the exposure to heat. Heat
inactivation experiments were performed at least in two replicates. Data were analyzed and
plotted with use of Origin7 software package (OriginLab, Northhampton, MA, USA).
4.3.7 Mathematical description of heat inactivation kinetics
First order inactivation kinetics was used to describe the linear part of inactivation curves of
heat treated LGG. The onset of the linear inactivation phase was fixed 1.5 min after the
treatment was started, since inactivation-free phase during the first 1.5 min was observed.
The calculated negative slope of the linear survivor curve (k) represented the contribution of
a particular combination of pressure and temperature pre-treatment conditions to induction of
heat tolerance. All calculations were performed using Origin7 software package (OriginLab,
Northhampton, MA, USA). A three-dimensional grid of derived inactivation rate (k) for the 3D
graph was constructed using Plot It software package (Scientific Programming Enterprise,
Haslett, MI, USA). The final 3D plot describing the simultaneous dependence of inactivation
rate on applied pressure levels and temperatures during pre-treatment step was drawn using
TableCurve 3D software (Systat Software Inc, Richmond, CA, USA).
4.3.8 Staining procedure with LIVE/DEAD® BacLight™ Bacterial Viability Kit
The LIVE/DEADBacLight bacterial viability kit (Molecular Probes Europe BV, Leiden, NL)
consisted of two separate stock solutions of SYTO9 and PI. According to the information
Pressure induced response in probiotic bacteria 167
given by manufacturer, both of them were dissolved in dimethyl sulfoxide at high
concentrations: 3,34 mM SYTO9 and 20 mM PI.
Bacterial sample (15 to 20 µL portions) were added into 1 mL of distilled and previously filter-
sterilized water. This bacteria suspension was then incubated for 10 min with 1,5 µL of each
of SYTO9 and PI in the dark prior to the measurement.
4.3.9 Flow cytometric measurement and data analysis
Analysis was performed on a Coulter®EPICS®XL-MCL flow cytometer (BeckmanCoulter Inc.,
Miami-FL, USA) equipped with a 488 nm laser. Cell was delivered at the low flow rate,
corresponding to 400 to 600 events per s. Forward scatter (FS), sideward scatter (SS), green
(FL1) and red fluorescence (FL3) of each single cell were measured, amplified, and
converted into digital signals for further analysis. SYTO9 emits green fluorescence at 530
nm following excitation with laser light at 488 nm, whereas red fluorescence at 635 nm is
emitted by PI-stained cells.
A set of band pass filters of 525 ± 20 nm and 620 ± 15 nm was used to collect green
fluorescence (FL1) and red fluorescence (FL3), respectively. All registered signals were
logarithmically amplified. A gate created in the dot-plot of FS vs SS was preset to
discriminate bacteria from artefacts. Data were analysed with the software package Expo32
ADC (BeckmanCoulter Inc., Miami-FL, USA).
Dot plot analysis of FL1 vs FL3 was applied to resolve the fluorescence properties of the
population measured by flow cytometer (Fig. 4). With this graph the population was able to
be graphically differentiated and gated according to their fluorescence behaviours.
Two regions were created in this plot for gating cells with intact membrane and the ones with
ruptured membrane.
The designation of gates according to the properties of cellular membranes was performed
by means of measuring fluorescence dot plot signals of untreated cells, which were located
in gate LIVE. On the other hand, cells heat treated at 95°C for 10 min were entirely
encountered in the area surrounded by gate DEAD.
4.3.10 Statistical analysis
The Student’s t-test was applied to evaluate the impact of pressure treatment on viability.
Differences were considered significant at the p < 0.05 level of probability. Statistical
significance of the effect of different pressure treatments on post-pressure growth
characteristics was examined using ANOVA-test. Statistical analysis was performed with
Origin7 software package (OriginLab, Northhampton, MA, USA).
Pressure induced response in probiotic bacteria 168
4.4 Results and discussion
4.4.1 Heat inducible thermotolerance of L. rhamnosus GG
Prior to the trials related to pressure induced thermotolerance on LGG the heat shock
response of the organism was evaluated so as to ensure that this inducible cellular defensive
mechanism could be detected and quantified with the described heat challenge procedure.
The machinery of heat inducible thermotolerance on LGG was investigated on exponential
and stationary growth phase using MRS broth as treatment medium in order to assess its
growth-phase specificity.
When cells from stationary growth phase were used, no significant differences were
observed in the inactivation kinetics of pre-incubated and control cells (Fig. 4A). It is a well-
known phenomena that cells in stationary growth phase have already possessed greater
inherent stability against severe treatment conditions owing to either accumulation of
protective compounds or exposure to various stresses in growth medium including
starvation, low pH, etc. [7, 70, 71]. In this state only negligible additional heat resistance
could be evoked.
With an incubation step of cells from exponential growth phase at 50°C for 10 min prior to
exposure to lethal temperature at 60°C for 10 min the thermal stability of LGG could be
increased in contrast to control sample (Fig. 4B). The lethal temperature of 60°C for heat
challenge was adapted from studies made with other probiotic lactobacilli [10, 20, 24]. Since
thermal inactivation kinetics at 60°C was followed it could be further observed, that the
protective effect resulting from heat shock diminished at prolonged exposure to 60°C. No
difference can be obtained in the inactivation level of pre-treated and untreated cells. This
behaviour can be attributed to predominating thermal damage that was exceeding the
magnitude that can be overcome by the acquired cellular repair mechanism, in which the
synthesized heat shock proteins may play a major role.
A lethal temperature of 60°C seemed to be suitable to be applied for the substantiation of
pressure induced thermotolerance and was applied throughout the study due to the clear
tendency in differentiating untreated and pre-treated population as well as due to a sufficient
reduction of cell count within 10 min of treatment.
Pressure induced response in probiotic bacteria 169
0 5 10 15 20 25 30
-6
-5
-4
-3
-2
-1
0
0 5 10 15 20 25 30
-6
-5
-4
-3
-2
-1
0B
Heat treatment time (min)
A
log N/N0 (-)
Figure 4.
Effect of heat shock treatment at 50°C for 10 min on the survival of L. rhamnosus GG from stationary
(A) and exponential growth phase (B) against subsequent thermal challenge at 60°C. Treatment
media were MRS broth. Heat shocked cells and untreated cells are represented by circles and
squares, respectively. Data are means of at least three independent trials and error bars represent
standard deviations.
4.4.2 Identification of non-lethal pre-treatment condition
Cells from the exponential growth phase were pressure treated to determine the suitable pre-
treatment conditions, in which stress-related thermotolerance could be evaluated. In general,
the applied pre-treatment step may not exert any lethal effect on the adapted cells.
Obviously, in excess of 200 MPa, inactivation of the cells occurred (Fig.5). Beneath this
critical point, no significant loss (p > 0.05) of viability following pressurization was observed.A
maximal pressure-temperature combination of 200 MPa and 50°C was then identified as
upper limit. Beneath this critical point, no significant loss (p > 0.05) of viability following
pressurization was observed. Within the range of conditions investigated the duration of
pressurization (at pressures lower than 200 MPa) did not impair the viability.
Pressure induced response in probiotic bacteria 170
0 15304560
-7
-6
-5
-4
-3
-2
-1
0
1
0 15304560 0 15304560
50°C
37°C
log N/N0 (-)
Pressure holding time (min)
25°C
Figure 5
High pressure inactivation kinetics of exponentially grown L. rhamnosus GG cells at 100 MPa (), 200
MPa (z), and 300 MPa (▲). For this experiment cells were growth at 37°C until OD 0.5 was reached.
Treatment media were MRS broth. Initial cell count was approximately 107 CFU/mL.
Pressure inactivation data were expressed as logarithmic value of relative survivor fraction (log N/N0)
at increasing pressure holding time. N refers to the bacterial count following pressurization at
corresponding holding time, whereas N0 represents the initial count prior to the exposure to pressure.
The temperature levels (25, 37 and 50°C) as indicated in the figure refer to the applied temperatures
during pressure treatment. Data shown are means of at least two independent measurements.
4.4.3 Post treatment growth behaviour of L. rhamnosus GG
To investigate post-pressure physiological activity, the growth behaviour of pressure treated
cells was monitored by means of the measurement of the changes in the optical density of
the cell suspension (Fig. 6). Pressure treatments were conducted at a constant temperature
of 37°C. The effect of different pressure levels (50, 100, 200 MPa) at a constant holding time
on growth behaviour was evaluated. Moreover, the impact of different pressure holding times
(5, 10, 20 min) at a constant pressure level (100 MPa) on the growth behavior was
characterized.
According to Fig. 6b, at low pressure level (p ≤ 100 MPa) and short holding times (t ≤ 10 min)
the OD-values of the cells at these intensities could not be significantly differentiated from
untreated population.
Pressure induced response in probiotic bacteria 171
10-2
10-1
100
101
OD at 600 nm [-]
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0 5 10 15 20 25 30 0 1 2 3 4 5 6
Control
50 MPa - 5 min
100 MPa - 5 min
200 MPa - 5 min
100 MPa - 10 min
100 MPa - 20 min
Time [h]
a
b
Figure 6
Monitoring of the growth behaviour of LGG after pressure treatment at various intensities. Parameters
for pressure treatment are indicated in the legend. All treatments were performed at 37°C.
Fig. 6a showed the increase of the optical density (OD) of the pressure treated cells. In this figure the
y-axis is logarithmically scaled. Fig. 6b resulted from the same experiment, but only exhibited the OD
changes within the first 6 h of incubation at 37°C. The y-axis in Fig 6b is arithmetically scaled.
Detailed experimental procedure is explained in the section Material and Methods. Data shown are
means of at least two independent measurements.
Apparently, at higher intensities of pressure treatments (higher pressure levels, i.e. 200 MPa,
5 min and longer duration, i.e 100 MPa, 20 min) a significant difference in the rate of OD-
change to the less severely treated groups (p < 0.05) was obvious. Although such treated
populations were not lethally affected by pressure (Fig. 5) and indeed able to resume
growing and reached the stationary growth phase as the control group did, they showed a
rather retarded growth at initial growth phase (t < 6 h).
The sigmoidal wth curves was fitted with modified Gompertz equation, which is expressed by
Equation 1. With this model the characteristic parameters of the growth curves, i.e duration
of lag phase and the µmax (maximum change in the OD600 nm ) could be derived.
(
)()
+−⋅
⋅
−⋅= 1
1exp
expexp max t
A
Ay
λ
µ
Equation 1
y: optical density at 600 nm; A: max. optical density; µmax: max. change in the optical density, λ:
duration of lag phase
The growth parameters derived from the development of OD-values after pressure
treatments using Eq.1 are listed in Table 1. From this table it is obvious that with increasing
Pressure induced response in probiotic bacteria 172
pressure the cells required longer time to recover from the pressure-induced sub-lethal injury
prior to entering the exponential growth phase. Moreover, the growth rates of pressure
treated cells were generally reduced compared to the untreated ones.
Table 1
Growth parameters derived from modified Gompertz equation
Pressure (MPa) Time (min) µmax (h-1) λ (h) Fit Standard Error
0 0 0.853 4.19 0.267
50 5 0.862 4.47 0.288
100 5 0.824 4.64 0.345
100 10 0.82 4.76 0.15
100 20 0.794 5.26 0.254
200 5 0.791 5.62 0.324
By means of flow cytometric assessment combined with LIVE/DEAD®BacLight™ Bacterial
Viability Kit (Molecular Probes, Leiden, NL) it was possible to obtain insights in the nature of
the cellular injury occurred during pressure treatment, especially when higher treatment
intensities were applied. Following a pressure treatment at 200 MPa the occurrence of cell
population with a higher degree of membrane damage was observed (data not shown). From
these results it could be concluded that a prolonged exposure to sub-lethal stress conditions
as well high pressure level may not be beneficial for the fitness of the culture. Therefore, the
treatment conditions have to be optimized in a way, that increased stress tolerance could be
achieved; on the other hand overprocessing which leads to higher cellular damage, has to be
avoided.
4.4.4 Heat treatment at 60°C
Figure 7 shows heat inactivation kinetics at 60°C of exponentially grown LGG cells. Pressure
level and temperature during pre-treatment step were varied to allow better insight of the role
of each process parameter on induction of thermotolerance. Upon exposure to 60°C for 5
min, cells survived better by 1.5 log-cycles in comparison to control group, when they were
previously pressure treated at 100 MPa and 37°C for 10 min (Fig 7A). Induction of
thermotolerance could therefore be achieved at normal growth temperature by merely
elevating system pressure for 10 min. Taken the data from previous observation on post-
pressure growth of cells exposed to 100 MPa for 20 min into account, it is not necessary to
subject the LGG cells for a long period since a relatively short pre-treatment time can
definitely confer protection against lethal effect of heat.
Pressure induced response in probiotic bacteria 173
0246810
-5
-4
-3
-2
-1
0
0246810 0246810
B
log N/N0 (-)
AC
Time (min)
Figure 7
Heat inactivation curves at 60°C of LGG, which were previously pre-treated for 10 min at 100 MPa (z)
and 200 MPa ({) in comparison with pressure-untreated population ().
The denotation of the three figures with A, B, and C corresponds to the temperature levels (37, 43,
and 50°C, respectively) applied during pressure pre-treatment.
In case of pressure-treated cells N0 was cell count after pressure treatment. Data shown are means of
at least two independent measurements.
Such improvement against lethal effect of heat could also be observed using pre-treatment at
higher temperatures; however thermal reduction of bacterial load was more pronounced the
higher the applied pre-treatment temperatures were (Fig. 7B and 7C). Doubling the pressure
level at the adaptation step to 200 MPa was found to be less effective in provoking pressure
induced thermotolerance (Fig. 7A to 7C).
To allow a better assessment of the effect of selected pre-treatment processing parameters,
a mathematical modelling of the linear part of inactivation curves of heat treated LGG was
performed. Figure 8a and 8b show the simultaneous dependency of heat inactivation rate (k
in min-1) at 60°C on applied pressure and temperature of adaptation step at two different
pressure holding times, i.e. 5 and 10 min, respectively. The k-values were calculated
according to the analysis method described in the section Material and Methods and are
characteristic for each heat inactivation curves (Regression parameters in Annex 8).
Furthermore, a second mathematical model based on Weibull’s distribution was also used to
fit the experimental inactivation data (Annex 8).
A 3D diagram was constructed from the derived inactivation rates. This plot enables a better
comparison of heat protection efficacy of different pre-treatment processing parameters.
Pressure induced response in probiotic bacteria 174
From this plot, optimal pre-treatment process conditions, which ensure minimal loss of
viability during heat exposure, could be directly identified.
100
120
140
160
180
Pressure (MPa)
37
40.25
43.5
46.75
Temperature (°C)
0.1
0.1
0.2 0.2
0.3 0.3
0.4 0.4
0.5
0.5
0.6 0.6
0.7 0.7
0.8 0.8
0.9 0.9
11
1.1 1.1
1.2 1.2
Inactivation rate (1/min)
Inactivation rate (1/min)
ab
100
120
140
160
180
Pressure (MPa)
37
40.25
43.5
46.75
Temperature (°C)
0.1
0.1
0.2 0.2
0.3 0.3
0.4 0.4
0.5
0.5
0.6 0.6
0.7 0.7
0.8 0.8
0.9 0.9
11
1.1 1.1
1.2 1.2
Inactivation rate (1/min)
Inactivation rate (1/min)
Figure 8.
Simultaneous dependency of heat inactivation rate of L. rhamnosus GG at 60°C (k in min-1) on applied
pressure and temperature in adaptation step. In Fig. 8a and 8b. the holding time during pressure
application was fixed to 5 and 10 min, respectively.
Inactivation rates were obtained by calculating the slope of the linear part of the corresponding
inactivation curve (see Fig 7. and explanation in the section Material and Methods).
According to the aforementioned curve analysis, the calculated k-value of the control sample
was 0.996 min-1, whereas the k-values of pre-treated samples in the applied p,T-conditions
ranged between 0.199 and 1.154 min-1. A commonality that was shared by both evaluated
pressure holding times is that a local minima of heat inactivation rate was observed at
pressures and temperatures as high as 100-125 MPa and 42-43°C, respectively. It was also
evident from both plots, that at 200 MPa and 50°C, the protective effect of pressure
treatment was at its minimum.
From this work it could be concluded that incubation of LGG at elevated pressure of 100
MPa for 5 - 10 min prior to exposure to lethal heat at 60°C led to an increasing heat
resistance as compared to untreated population. Optimal working range to induce
thermotolerance properly was determined at 100 MPa and 37 - 43°C. Regarding the duration
of pressure holding time, pressurization up to 10 min could effectively trigger
thermotolerance mechanism, when operating at these optimal conditions. These apparent
optima at both pressure holding time could also be identified in the 3D plot generated under
application of Weibull distribution (Annex 8).
Working at higher temperature and pressure, despite of the demonstrated possibility to
induce thermotolerance, was less effective; presumably due to higher extent of injury. This
Pressure induced response in probiotic bacteria 175
might in turn exceed the magnitude, which could still be tolerated by the cells in order to
instantly regenerate after pressure release and trigger thermotolerance-conferring reactions.
0246810
-2.0
-1.8
-1.6
-1.4
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
log N/N0 (-)
Pressure holding time (min)
Figure 9
Effect of pressure holding time in the pre-treatment phase on the survival of L. rhamnosus GG during
subsequent thermal challenge at 60°C for 3 min. Pre-treatment step was conducted at 100 MPa and
37°C. N and N0 refer to cell count before and after heat treatment, respectively. Data represents
means of two independent experiments.
The acquisition of thermotolerance can still be achieved by exposing the cells to high
pressures for 5 min. In terms of optimizing pre-treatment step further trials were focused on
the reduction of pressure pre-treatment time. In particular it is of huge interest to check
whether extremely short exposure to pressure, or “flash adaptation,” could still induce
tolerance in bacteria. For this evaluation the thermal challenge at lethal temperature of 60°C
was not performed in form of a kinetic. Instead the cells were exposed for a fixed treatment
time of 3 min, since after 3 min of holding time at 60°C good differentiation in the effect of
pre-treatment on survival rate could be made (Fig. 7).
Figure 9 shows the effect of variation of pressure holding time on the resulted thermal
survival rate at 60°C for 3 min. It was obvious, that there exists a threshold treatment time of
5 min, below of which the pressure-induced thermotolerance gradually decreased. Pressure-
shock response, which is ultimately manifested in increase of tolerance, ranged in minutes
and was relatively quick. This is similar to data on the time required to initiate heat shock
response [64]. In contrast it took hours to initiate cold shock response [72]. As reviewed by
Yura et al (2000) a modest temperature upshift (∆T~ 12°C) elevated the synthesis of heat
shock proteins almost immediately and reached a maximum induction (10- to 15-fold) within
5 min, where heat shock proteins represented over 20% of total proteins synthesized [26].
Pressure induced response in probiotic bacteria 176
Furthermore, the rapid induction is followed by a gradual decrease, during the adaptation
phase, to attain a new steady-state level (2- to 3-fold of the pre-shift level) by 20 to 30 min.
With regards to the onset of pressure shock response a proteomic study on heat shock
response of L. lactis documented an elevated rate of synthesis of fast-induced proteins (such
as DnaK, GroEL, GroES) during the first 10 min of exposure to sub-lethal heat stress [15].
Walker et al (1999) also reported that the maximum groESL transcription activity was
increased following exposure to sub-lethal temperature and a 15 to 30 min exposure of log-
phase cells to this temperature increased the recovery of freeze-thawed L. johnsonii [11]. All
of these studies indicated the necessity to expose the cells to stress condition for a sufficient
holding time in order to allow the cellular process of protein synthesis to be accomplished. In
this context, the determination of optimum sub-lethal stress conditions (temperature, time,
etc.) for RNA expression over a stress operon can further facilitate monitoring the heat shock
induction of the groESL chaperone operon, since stress protection was found to correlate
with the timing and the level of expression of the chaperone operon [11]. The idea of
applying rapid pressure processing, in which cells were only subjected to a single pulse of
hydrostatic pressure without any holding time, seems to be inappropriate to allow sufficient
protection against heat.
50 75 100
-1.6
-1.4
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
log N/N0 (-)
Pressure (MPa)
Figure 10
Influence of adjusted pressure levels on the survival of L. rhamnosus GG during high pressure
homogenization process (MicronLab 40, APV Gaulin GmbH, Lübeck, Germany). Bacteria were grown
to an OD of 0.5 in MRS broth and subjected to one cycle of homogenization. N0 and N refer to
bacterial count before and after homogenization step, respectively. Data represents means of two
independent experiments.
In contrast, an extremely short exposure to sub-lethal of bile salt induced tolerance in
Enterococcus faecalis [73]. This is the first work reporting the kinetics of induction of
Pressure induced response in probiotic bacteria 177
tolerance against lethal concentration of bile, which could already be observed only after 5 s
exposure to 0.08% bile and reached its maximum after 30 min incubation. Interestingly, the
development of bile salt tolerance is not at all compromised by the blockage of protein
synthesis prior to challenge in adapted E. faecalis cells.
Trials made on a similar rapid process, i.e. high pressure homogenization show that even
during the homogenization step a reduction by one log cycle was obtained (Fig. 10). This
process is commonly used for cell disruption of concentrated microbial cultures, and the
subsequent recovery of intracellular metabolites. Cell death during passage through high
pressure homogenizer is mostly related to turbulence, shear, cavitation – which are also the
main physical causes for fat globule disruption –, and heat generation, raising the
temperature by 2.5°C per 10 MPa [74].
4.4.5 Determination of the role of de novo protein synthesis in inducible heat tolerance
Works dealing with the application of mild heat shock also demonstrated that the expression
of certain proteins at higher levels were required for the acquisition of thermotolerance [16,
19, 23]. The role of protein synthesis was elucidated by incorporating a bacteriostatic
concentration of chloramphenicol during adaptation, in consequence of which the tolerance
against lethal heat decreased markedly. Determination of the N-terminal sequence of a
series of these proteins reveals that these proteins are involved in a variety of cellular
processes [16]. The majority of these proteins are homologous or immunologically related to
stress proteins in other microorganisms [13, 14]. Several of the proteins identified belong to
the group of molecular chaperones (e.g. DnaK, GroEL, etc.) or ATP-dependent proteases
(ClpP). The induction of this type of proteins forms a highly conserved response to heat
stress. These proteins are termed heat shock proteins and are involved in protein folding,
assembly, and repair and prevention of aggregation under stress and normal conditions [64].
2D-PAGE combined with densitometric analysis revealed that the chaperone protein GroEL
was among the most strongly expressed proteins in the cell under heat adaptation conditions
[20] and after heat shock the GroEL synthesis was reported to be 15-fold higher than the pre-
shift level [18]. However, tolerance of GroESL-overproducing strains against heat, spray
drying and freeze drying was not necessarily better than heat-adapted parent strain [20],
since heat adaptation response that leads to an elevated state of cell resistance does not
only rely on the overexpression of heat stress proteins, but may be due to the mechanisms
associated with stress proteins, changes in the synthesis of some glycolytic enzymes and
other stationary-phase-related proteins and regulatory factors [18].
According to Figure 11, the magnitude of cell death as a result of exposure to lethal heat
could be reduced when pressure adaptation was applied on the cells prior to heat challenge.
Pressure induced response in probiotic bacteria 178
Pressure treated cells (log N/N0 ~ 0) were significantly more heat resistant than untreated
populations (log N/N0 ~ -1.5) after 3 min of exposure to 60°C. The inactivation-free phase
was longer in the pressure pre-treated sample (3 min) compared to the control one (1.5 min).
The survival curves started with a shoulder followed by a linear decline curve. Mechanistic
approaches had been made to explain factors accounting for the lag phase, in which the
concentration of microorganisms remained the same. Either the existence of clumps of
microorganisms, the ability of the cells to resynthesize a vital component, cumulative thermal
inactivation rather than instantly lethal, or multiple target sites for thermal inactivation were
made responsible for the generation of lag phase [75]. Only after this limiting factor had been
eradicated, the inactivation of microorganisms could follow first order kinetics in the linear
phase. When the distribution of resistance within the population to be treated was
considered, a distributive function combined with a first-order reaction describing subsequent
linear inactivation kinetics can explain the initial transition of bacterial cells into an
inactivation free, metastable intermediate state, which is reached after endogenous
homeostatic mechanisms balancing displacement of equilibrium can no longer be maintained
[76].
0246
-5.0
-4.5
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
8
log N/N0 (-)
Heat treatment time (min)
Figure 11
Effect of heat treatment at 60°C on the residual survival fraction (log N/N0) of exponential-phase
control cells () and that of cells pretreated for 10 min at 37°C 100 MPa in the absence (S) and
presence of 10 µg/mL chloramphenicol (z). Data are means of at least three independent
experiments.
Moreover, the heat inactivation curve of pressure pre-treated LGG with chloramphenicol
being incorporated in the medium resembled the one of control population (Fig. 11). This
result points out that thermotolerance acquired after a mild pressure shock highly depends
Pressure induced response in probiotic bacteria 179
on protein synthesis. The addition of chloramphenicol in the cell suspension prior to pressure
adaptation blocked protein synthesis so that increased heat tolerance disappeared. Similar to
heat shock response, which implicates the synthesis heat shock proteins in the induction of
heat tolerance [16, 19, 23], a brief adaptation period under elevated pressure could enhance
their heat tolerance as well. Indeed, it has been shown that heat shock proteins were among
the ones overexpressed upon brief or continuous exposure of bacterial cells to moderate
pressure levels [41, 43, 46]. This result underlines once more the universal importance of
heat shock proteins in coping with various adverse conditions.
Taking the lag phase of pressure pre-treated sample prior to the linear inactivation phase
was longer than that of untreated one (Fig. 11), it is most likely that the newly synthesized
proteins were involved in cell repair mechanism, which temporarily led to a higher stability
against heat. The induced tolerance against heat vanished after 3 min of exposure to 60°C;
presumably because the rate of destruction of crucial bio-molecules exceeded the rate of
repair/resynthesis of vital components [75].
4.4.6 Flow cytometric assessment of damaged on cellular membrane as affected by heat
The applicability of the commercially available LIVE/DEADBacLight bacterial viability kit
has been evaluated on a wide spectrum of bacteria [77-82]. This kit was developed to
differentiate live and dead bacteria based on plasma membrane permeability. The staining
mechanism using LIVE/DEAD kit on bacterial cells is based on the attachment of the non-
fluorescent agents on nucleic acids. Once the DNA-dye complex is built fluorescence could
be measured. This kit is constituted of two fluorochromes, which have distinct fluorescent
behaviour in terms of emission wavelengths and membrane permeability. The first
component is the membrane-permeant stain SYTO9®, which fluoresces green at 530 nm
upon excitation at 488 nm. It stains all cells, thus acting as total cell stain. During cell death,
accompanied with membrane damage, the second dye, the membrane-impermeant dye
propidium iodide (PI) penetrates into cells and quenches the green SYTO9® fluorescence. PI
is able to be excited at 488 nm as well and emits red fluorescence at 620 nm. When used in
combination, intact cells are labeled green and cells with damaged membranes are labeled
red.
According to flow cytometry data, exposure to 100 MPa did not affect fluorescence behavior
of LGG (Fig. 12B), when compared to the one of control group (Fig. 12A). The presence of
chloramphenicol in the media during pressure adaptation step did not cause detrimental
effect on cellular membranes either (Fig. 12C).
Pressure induced response in probiotic bacteria 180
A BC
D EF
Figure 12
Flow cytometry dot plots of FL1 (fluorescence collected at 525 nm) vs FL3 (fluorescence collected at
620 nm) of LGG to assess the effect of heat treatment on the integrity of cellular membranes.
Two gates were fixed for discrimination of two distinct extreme states of membrane conditions, i.e.
intact and completely ruptured (gate LIVE and DEAD, respectively).
Figure 12A, 12B and 12C show fluorescence behaviour of control population and population
pressurized at 37°C 100 MPa for 10 min with and without chloramphenicol, respectively.
Heat treatment at 60°C for 3 min was applied to control population (D), and pressure pretreated cells
(E). Pressure pretreatments at 37°C 100 MPa for 10 min were conducted in the absence and
presence of chloramphenicol (E and F, respectively).
Once the cells were subjected to 60°C for 3 min a higher degree of membrane rupture was
able to be detected in the dot plot FL1-FL3 (Fig. 12D to F). The onset of this cellular damage
was marked by characteristic population shift from gate LIVE, where cells with intact
membranes and stained by SYTO9® were located, towards gate DEAD, in which cells with
completely disintegrated membranes and stained by PI were encountered. During exposure
to heat structural changes in proteins and membranes may lead to cell death. Heat treatment
at temperatures in the vicinity of 60°C was reported to cause damage in the cytoplasmic
membrane of L. bulgaricus, whereas for temperature of 65°C and immediately above,
ribosomes and/or proteins denaturation as well as cell wall damage may be responsible for
thermal death [83]. Using fluorescence marker aimed on cells with compromised membrane
the flow cytometric fluorescence data presented in this study underlined the findings of the
work of Teixeira et al (1997) about the deteriorative action of heat at 60°C on cytoplasmic
membrane. Furthermore, when Figure 12D and E are compared it is obvious that the extent
Pressure induced response in probiotic bacteria 181
of the shift of SYTO9®-labeled, viable cells toward the population labeled with PI was
reduced, when prior to exposure to lethal heat at 60°C for 3 min, the cells were incubated at
100 MPa and 37°C for 10 min (Fig 12E). Pre-treatment at elevated pressure level seemed to
confer protection against destructive effect of heat on cellular membrane. The acquired
membrane stabilization was documented by a less pronounced shift of the cells towards gate
DEAD, which may be related to a reduced influx of PI across cytopasmic membrane. On the
other hand, the presence of chloramphenicol diminished the membrane stabilization effect of
pressure (Fig. 12F). The pattern of population shift from gate LIVE towards gate DEAD was
found to be similar to the one of untreated population despite of application of pressure pre-
treatment. Similar to the results obtained in the previous sub-section this observation is
indicative for the decisive role of de novo protein synthesis as a consequence of pressure
pre-treatment in withstanding degradative events on cell membrane upon heat treatment at
60°C.
Some works dealing with the preferential localization of heat shock proteins in cellular
membrane could give some hints about their functionality. It was reported that after a sub-
lethal heat treatment of cyanobacterial cells, an increase of the membrane-associated GroEL
fraction was observed concomitantly with an increase in the heat stability of the
photosynthetic electron transport machinery [84]. In particular, the soluble chaperonin GroEL
from E. coli has high affinity for model lipid membranes, and the conserved C terminus of
GroEL is involved in membrane binding. Apart from the role in membrane stabilization,
GroEL may also function as lipochaperonin that can prevent the irreversible thermal
aggregation and assist the refolding of membrane proteins. Moreover, water-soluble proteins
could also be rescued by lipochaperonins under stress conditions. This suggests that, during
stress, chaperonins can assume the functions of assisting the folding of both soluble and
membrane-associated proteins while concomitantly stabilizing lipid membranes [85].
Similarly, it was found from a study on cellular localization of GroEL in Clostridium difficile
using immunoelectron microscopy that after heat shock at 48 °C GroEL was distributed in a
relatively uniform fashion over the bacterial surface and was partially also found to be
localized within the extracellular space [86]. Furthermore, genomic expression pattern of S.
cerevisae exposed to 200 MPa for 30 min showed the expression of genes coding proteins
with molecular mass less than 20 kDa that present putative transmembrane domains [62]. An
increase in the expression in the expression of small membrane binding proteins was
assumed to be involved in the protection against membrane damage.
4.4.7 Pressure induced tolerance against nisin and bile acid
As already mentioned in the previous sub-sections, the contribution of pressure-induced
protein biosynthesis in the enhancement of bacterial heat tolerance was found to be crucial,
Pressure induced response in probiotic bacteria 182
since the presence of chloramphenicol in the treatment medium hampered the effect of
pressure adaptation on inducible tolerance against heat. Furthermore, the role of proteins
expressed during pressure adaptation in withstanding the thermal degradation on cell
membrane had been characterized by flow cytometric assessment.
These facts led to further question, whether pressure adapted cells also showed improved
tolerance against other membrane-degrading agents, such as bile acid (B8381, Sigma-
Aldrich, Munich, DE) or nisin (Nisaplin®, Danisco, Copenhagen, DK).
Bile is a digestive secretion that plays a major role in the emulsification and solubilization of
lipids. These ‘biological detergents’ are synthesized in the liver from cholesterol, conjugated
to either glycine or taurine, and then secreted as amino acid conjugates into the intestine
where they facilitate fat absorption. Bile is primarily composed of bile acids (12% by weight),
which are found as sodium salts under physiological conditions. Forming part of the body’s
physicochemical defense system, bile salts possess potent antimicrobial activity and have
the ability to dissolve the phospholipids, cholesterol, and proteins of cell membranes. They
disorganize the lipid bilayer structure of the cellular membranes, thus causing cells to lyse
[87]. Analysis of Enterococcus faecalis susceptibility towards the bile salts gave evidence for
an extremely rapid killing effect which is attributed to the solubilization of membrane
components [88]. An improved bile tolerance is essential for probiotic bacteria since they
need not only to survive processing conditions but also harsh environmental challenge during
gastrointestinal passage. Especially to achieve high degree of gut colonization, these
bacteria have to interact with inhibitory host-produced substances.
It was observed, that bacteria pressure pre-treated prior to exposure to 1%, w/v bile acid
showed an improved tolerance against this antimicrobial compound (Fig.13a). This result
confirmed the contributive role of protein biosynthesis during pressure adaptation in
protecting cellular membranes. With regards to induced bile tolerance the pressure shock
proteins can presumably be functionally grouped into bile salt hydrolases, which are known
as detergent shock proteins that protect the bacteria that produce this enzyme from the
toxicity of bile acids in the gastrointestinal tract [89].
Nisin is a protein with 34 amino acid residues which is produced by L. lactis subsp. lactis.
This antimicrobial peptide has a broad activity spectrum and is active against a variety of
gram-positive bacteria. It has been shown that nisin permeabilizes the cytoplasmic
membrane, thereby dissipating the membrane potential, inhibiting transport of amino acids,
and causing release of accumulated acids from cells and membrane vesicles from various
bacteria [90]. Models for nisin/membrane interactions propose that the peptide forms
poration complexes in the membrane through a multi-step process of binding, insertion, and
pore formation [91].
Pressure induced response in probiotic bacteria 183
A higher survival in the presence of 12.5 mg L-1 nisin was obtained, when LGG were
previously pressure pre-treated (Fig 13b). The applied nisin concentration was slightly lower
than the one recommended by manufacturer for typical use in food. This result emphasizes
once more about the structural/functional modifications in the cellular membrane of pressure
adapted bacteria, which allow them to counteract the pore-forming activity of nisin and thus
reduce its lethality on bacteria.
01234567891011
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
log N/N0 (-)
Incubation time at 37°C (min)
0 102030405060
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
log N/N0 (-)
Incubation time at 37°C (min)
ab
Figure 13
Effect of pressure pre-treatment on the survival of L. rhamnosus GG in the presence of bile acid (a) or
nisin (b). Pressure pre-treated (▲,z) and untreated cells () were incubated in growth medium
supplemented with bile acid (1%, w/v) or nisin (12.5 mg L-1 ). Pressure pre-treatment condition was
100 MPa, 37°C, 10 min. Data were means of at least three independent trials and error bars
represented the standard deviations of the means.
4.4.8 Spray drying of pressure pre-treated bacteria
The potential of pressure pre-treatment at 100 MPa was assessed on its impact on the
survival of LGG during spray-drying. Pressure adaptation on LGG was done in their own
growth media, whereas reconstituted skim milk (20%, w/v) was used as the drying medium.
In a related study the drying medium was the one also used to grow the bacteria, i.e.
reconstituted skim milk [24]. This was not possible for LGG due to their inability to ferment
lactose [92]. Preliminary results showed that pressure adaptation improved the survival of
LGG during spray-drying at outlet temperatures of 80 and 90°C (Fig. 14). The survival rate of
pressure-treated cells was as high as 32%, whereas spray-drying of untreated population
resulted in only 14% survival (unpublished data). Due to the predominant heat damage which
exceeded the capacity of pressure induced proteins in conferring protection the pressure
induced increase of survival rate could not be quantified during spray drying at an outlet
temperature of 100°C. The level of improvement achieved in the present study was in the
same range as obtained in the work of Teixeira et al (1995), where they showed that heat
Pressure induced response in probiotic bacteria 184
shock at 50°C for 30 min on exponentially growing cells of L. bulgaricus (grown in MRS broth
and heat shocked in skim milk) increased the survival by 10% during spray drying at an
outlet temperature of 80°C [30]. However, the magnitude of viability enhancement during
spray drying can be markedly enhanced by growing and performing heat adaptation in the
final drying medium. It was reported, that the viability of heat adapted L. paracasei in
reconstituted skim milk was enhanced 6-fold and 18-fold during spray drying at outlet
temperatures of 95-100°C and 100-105°C, respectively [24]. Similar improvement of
tolerance against spray drying was achieved using salt adaptation with 0.3 M NaCl. Although
this osmotic pre-treatment was efficient but the presence of salt in the product could be a
problem during the preservation process and/or subsequent cell uses [23]. Furthermore, heat
and osmotic adaptation on both stationary-phase and log-phase were found to be also
effective in improving the storage stability of dried L. rhamnosus [18].
In conclusion, data from this and other related works on induction of thermotolerance by
physical or chemical stresses give evidence to the potential of utilizing this particular cellular
auxiliary mechanism in protecting probiotic bacteria from multiple environmental stresses
(including heat, osmotic and oxidative stress) encountered during spray drying as well as
during subsequent storage.
80 90 100
0
5
10
15
20
25
30
35
40
Survival rate (%)
Outlet temperature (°C)
Figure 14
Effect of pressure pre-treatment on the survival rate of L. rhamnosus GG during spray drying at
various outlet temperatures. Drying carrier was 20% (w/v) reconstituted skim milk (RSM). Initial cell
count of the feed solution was ~ 107 CFU/mL. Bacteria were grown in MRS broth until an OD value of
0.5 was reached. Pressure pre-treatment conditions were 100 MPa and 43°C for 10 min. Pressure
pre-treatments were conducted on bacteria in their own growth medium. Data are means of two, or
more, spray drying experiments.
Pressure induced response in probiotic bacteria 185
4.4.9 Pressure induced thermotolerance on other L. rhamnosus strain
The intrinsic thermal resistance of L. rhamnosus E800 was found to be higher than the one
of L. rhamnosus GG. According to Figure 7 a reduction by almost 2 log cycles was achieved
on untreated population of LGG from exponential growth phase upon exposure to 60°C for 3
min, whereas the strain E800 from the same growth stage could resist this challenge
condition without loss of viability (Fig. 15). This tendency has also been confirmed by other
research group that compared the thermal resistance of various probiotic lactobacilli
suspended in reconstituted skim milk [93]. The thermal resistance is therefore highly strain-
specific and genetically determined [94].
-2 0 2 4 6 8 10 12 14 16
-4
-3
-2
-1
0
log N/N0 (-)
Heat treatment time (min)
Figure 15
Effect of high pressure pre-treatment on the inactivation kinetics of L. rhamnosus strain E800 (VTT,
Espoo, FI) upon exposure to a lethal temperature of 60°C. Similar to trials made with LGG, cells were
grown until an OD value of 0.5 was reached. Afterwards, cells were either pre-treated with 100 MPa,
37°C for 10 min ({) or left untreated (z) prior to heat treatment at 60°C.
4.5 Conclusion
The effect of mild pressure pre-treatments on technological behaviour of LGG was
evaluated. With respect to heat tolerance pressure pre-treated cells (at 100 MPa and 37°C)
showed higher survivability than untreated ones when both were exposed to heat treatment
at 60°C. Further investigation with flow cytometric analysis indicated that the acquisition of
pressure-induced heat tolerance was related to membrane stabilization and protein
biosynthesis. Pressure induced thermotolerance occurred as a consequence of stabilization
of cellular membranes – presumably by incorporation of heat shock proteins into cytoplasmic
membrane – which in turn led to an enhanced transient protection against degradative
effects of heat on cell membrane. The absence of induced thermal tolerance upon addition of
Chloramphenicol suggested that the proteins expressed during pressure adaptation was
involved in the prevention of thermal degradation on cell membrane. Pressure pre-treated
Pressure induced response in probiotic bacteria 186
LGG showed also an improved tolerance against chemical compounds which are able to
degrade cytoplasmic membranes, such as bile acid and nisin. This result confirmed the
positive role of protein synthesis during pressure adaptation in protecting this vital cellular
component. The potential of utilizing pressure shock response in increasing the survivability
of LGG during spray drying, which poses multiple environmental stresses (including heat,
osmotic and oxidative stress) was also demonstrated.
In the production of lactic acid bacteria, where they are exposed to different types of
environmental stresses, cross-protection induced by expression of adaptive response could
be regarded as an effective approach in enhancing stability. Cross-protective action of
pressure especially against heat was investigated in this work to evaluate its possible
application in assisting probiotic production, in particular to facilitate higher bacterial survival
upon spray-drying. The improvement of tolerance against various lethal stresses not related
to the stress inducers is resulted from an involvement of common regulators. One advantage
rising from this cross adaptive stress response is that there is a higher degree of freedom in
selecting the type of stress to be applied for the induction of for instance heat tolerance.
Generally, technologically relevant stress imposed to bacteria can be classified in two major
groups: physical and chemical means. Use of physical stresses such as heat, UV, irradiation
and pressure for the induction step would be more attractive compared to the use of
chemical additives, including osmotic agents (salt or sugar), acid, bile, etc., which in many
cases need to be removed from the product. With regards to acid stress response, Saarela
et al (2004), who had conducted stress induction of probiotic bacteria in fermenter, found that
in the fermenter it took 45 min to reach pH 4 [33]. Unfortunately, a feasible approach to
drastically reduce pH by using a stronger acid might have had detrimental effects on cell
viability because of the developing acid concentration gradient.
Due to these major drawbacks of chemical stress inducers physical methods appear to be a
better choice. Exposure to sub-lethal heat could easily be done with existing equipment;
however temperature increase to 47°C for heat pre-treatment took 10 min [33]. In order to
accelerate heat transfer and achieve thermal equilibrium temperature gradient could easily
be increased. Yet, the presence of a radial temperature field with higher temperature at the
fermenter’s wall would lead to an overprocessing of bacterial population at this site.
With help of the application of pressure to induce beneficial stress response on bacteria the
limitations of heat transfer into product with small ratio of surface to volume could be
compensated. Provided that the medium does not show pronounced inhomogeneity in
compressibility the effect of pressure is uniform because of the instantaneous transmission
which affects all volume elements of the confined fluid [95]. Due to adiabatic compression
Pressure induced response in probiotic bacteria 187
pressure generation is accompanied by temperature increase of 2 to 3°C per MPa [96],
which can be instantly equilibrated by temperature control.
The trials conducted in this PhD thesis to induce thermotolerance were principally performed
on cells from exponential growth stage. Cells from stationary growth phase were ruled out,
since bacteria that enter into stationary phase had already developed resistance against
various types of environmental stress (including subsequent down-stream processing and
storage). It is likely that for this reason stress treatment studies on probiotic cultures usually
have been performed with log-phase cultures instead of stationary phase cultures. There are
very few studies where the sublethal and lethal stress responses of stationary-phase
probiotic cultures have been investigated in well-controlled manner. However, data on stress
response studies with culture from stationary growth phase revealed the potential of pre-
treatment of cells in this particular growth stage to improve survival during subsequent
treatment [18, 33, 97, 98].
02468101214161820
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
1E7
1E8
1E9
1E10
1E11
log N/N0 (-)
Cultivation time (h)
Cell count, (CFU/mL)
Figure 16
The survival rate of L. rhamnosus GG heat treated at 60°C for 3 min subsequent to pressure pre-
treatment () or without any pre-treatment () as a function of cultivation time. Cell count (S) is
included in order to establish a correlation between growth-stage and acquired or intrinsic heat
tolerance. The conditions for pressure pre-treatment: 100 MPa, 37°C for 10 min.
Own data compiled on the influence of growth phase on the level of pressure inducible
thermotolerance of heat treated LGG revealed that the heat tolerance of cells from stationary
growth phase could still be improved by this adaptive response mechanism (Fig 16).
According to Figure 16 cells entered stationary growth phase after 5 to 6 h of cultivation in
MRS broth. It was demonstrated further that cells harvested until 8 h of fermentation still
showed inducible stress response leading to higher tolerance against heat. In contrast,
Pressure induced response in probiotic bacteria 188
pressure pre-treatment on cells from mid- or late-stationary phase (after 12 or 18 h of
cultivation, respectively) did not increase their heat tolerance. These data indicated the
importance of further assessing the adaptive or cross-protective behaviours of cells from
stationary growth phase exposed to sub-lethal stress. As already stressed by Saarela et al
(2004) stationary-phase cells had to be studied instead of log-phase cells as during the
industrial production of cultures high enough cell densities have to be reached before
harvesting the cells.
4.8
5.0
5.2
5.4
5.6
5.8
6.0
6.2
6.4
6.6
6.8
pH
36
37
38
39
40
41
42
Temperature [°C]
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
Biomass [g/ml]
90
92
94
96
98
100
102
Stirrer [1/min]
0
20
40
60
Base pump [ml]
8
9
10
11
12
13
Optical density (620 nm)
0
2
4
6
8
10
12
14
cfu [ E+008/ml]
0
5
10
15
20
25
30
Survival rate [%]
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460
Fermentation time [min]
pH
Biomass
Optical density cfu after F/T with -20°C
Survival rate
Medium 5f03
Temperature
cfu
Stirrer
Base pump
Figure 17
Full fermentation profile of Lactobacillus bulgaricus showing values measured online (bottom) and
offline (upper). The fermentation was performed using a 2-liter bioreactor (Biostat M, Braun Biotech
International GmbH, Melsungen, D).
Further arguments on inducing adaptive or cross-protective stress response during stationary
growth phase were substantiated by Figure 17, which basically shows a profile of pH-
controlled fermentation of L. bulgaricus [99]. Taken into account the measured offline values
(i.e. cell count, biomass, optical density) which indicate the progress of the growth, it is
obvious that the cells entered the stationary phase after 4 to 6 h of cultivation. However, the
survival rates of the cells upon challenging them to freeze-thaw step were very poor at the
first hours in the stationary phase and increased drastically after 80 min of further residence
in this growth phase. There was also a certain fermentation time where the survival rate
reached the highest value. This behaviour points out that there is still considerable
differences in the resistance of the bacteria within the stationary growth phase. This
conclusion was supported by works on L. lactis, in which the authors found that stationary
phase is not a uniform state of cell properties [100]. They also confirmed the existence of
Pressure induced response in probiotic bacteria 189
different “states” in the course of stationary phase. Taken together, these facts emphasize
the necessity to revisit the inducible stress response within stationary phase. In particular, it
is of utmost importance to evaluate whether the induction of stress response to improve the
technological suitability of bacteria is possible at the beginning of stationary growth phase.
Taking the scenario described in Figure 17 into consideration – when improvement in
freeze/thaw survival within the first hours of entry in stationary growth phase can be achieved
using sub-lethal stress, then the fermentation time in order to get viable and stable culture
could considerably be reduced.
Additionaly, the results of this study on pressure induced cross adaptation point out also
some precautional consideration for the use of high pressure technology for pasteurisation
purposes. Due to the possibility for cells to gain protective stress response at lower pressure
levels the residence time at these critical pressure levels (up to 200 MPa) has to be kept as
minimal as possible, in order to minimize the acquisition of multi-stress tolerance of certain
pathogens. It can be extracted from this work that a proper acquisition of thermotolerance
required at least 5 min of holding time at 100 MPa. Based on this consideration a rapid
compression step bears a beneficial side effect, since it is unlikely that adaptive stress
response can be triggered efficiently in this highly changing environmental conditions during
pressurization. Currently there is not many processing units available that utilize rapid
compression step. A commercial unit of LAB50 single pressure processor (SIG Simonazzi,I)
allows high compression and decompression rates: a processing pressure level of 600 MPa
is generated in about 3-4 seconds, while complete release of pressure is realized in less than
1 second [101].
To ensure the safety of the processing which employ the effect of mild pressure in conferring
cross-protection against heat and other membrane-destabilizing stresses one has to take the
worst case scenario into account, particularly when problematic microbial contaminants such
spoilage or pathogenic microbes incidentally pre-stressed and acquired tolerance against
multiple stresses. Data on the effect of pressure treatment on pathogenic bacteria E. coli
showed that sub-lethally injured bacteria could not survive in acidic environment, i.e. in fruit
juices (pH 3 to 4) during cold storage [102], whereas probiotic bacteria might have better
survival characteristics due to their higher acid tolerance. However, the aforementioned
safety concerns emphasized that upon integration of this innovative processing concept into
industrial application it is necessary to utilize existing genomic or proteomic tools in order to
identify pre-treatment conditions, with help of which the beneficial shock response can be
specifically aimed on microorganisms to be produced and not on bio-contaminants. The risk
of the aforementioned incidence could also be controlled by applying aseptic processing line.
Pressure induced response in probiotic bacteria 190
4.6 References
1. Stanton, C., Gardiner, G., Meehan, H., Collins, K., Fitzgerald, G., Lynch, P.B., and Ross, R.P. 2001.
Market potential for probiotics. American Journal of Clinical Nutrition. 73(Suppl.): 476S-483S.
2. Hollingsworth, P. 2001. Culture wars. Food Technology. 55: 43-46.
3. Mattila-Sandholm, T., Myllärinen, P., Crittenden, R., Mogensen, G., Fonden, R., and Saarela, M.
2002. Technological challenge for future probiotic foods. International Dairy Journal. 12: 173-182.
4. Gomes, A.M.P. and Malcata, F.X. 1999. Bifidobacterium spp. and Lactobacillus acidophilus: Biological,
biochemical, technological and therapeutical properties relevant for use as probiotics. Trends in Food
Science and Technology. 10: 139-157.
5. Saarela, M., Mogensen, G., Fondén, R., Mättö, J., and Mattila-Sandholm, T. 2000. Probiotic bacteria :
safety, functional and technological properties. Journal of Biotechnology. 84: 197-215.
6. Abee, T. and Wouters, J.A. 1999. Microbial stress response in minimal processing. International
Journal of Food Microbiology. 50: 65-91.
7. van de Guchte, M., Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S.D., and Maguin, E. 2002.
Stress responses in lactic acid bacteria. Antonie van Leeuwenhoek. 82: 187-216.
8. Beales, N. 2004. Adaptation of microorganisms to cold temperatures, weak acid preservatives, low pH,
and osmotic stress: A review. Comprehensive Reviews in Food Science and Food Safety. 3: 1-20.
9. Stortz, G. and Hengge-Aronis, R., eds. Bacterial stress responses. . 2000, ASM Press: Washington DC.
485.
10. Kim, W.-S., Perl, L., Park, J.-H., Tandianus, J.E., and Dunn, N.W. 2001. Assessment of stress
response of the probiotic Lactobacillus acidophilus. Current Microbiology. 43: 346-350.
11. Walker, D.C., Girgis, H.S., and Klaenhammer, T.R. 1999. The groESL chaperone operon of
Lactobacillus johnsonii. Applied and Environmental Microbiology: 3033-3041.
12. Schmidt, G. and Zink, R. 2000. Basic features of stress response in three species of bifidobacteria: B.
longum, B. adolescentis, and B. breve. International Journal of Food Microbiology. 55: 41-45.
13. Whitaker, R.D. and Batt, C.A. 1991. Characterization of the heat shock response in Lactococcus lactis
subsp. lactis. Applied and Environmental Microbiology. 57: 1408-1412.
14. Auffray, Y., Gansel, X., Thammavongs, B., and Boutibonnes, P. 1992. Heat shock-induced protein
synthesis in Lactococcus lactis subsp. lactis. Current Microbiology. 24: 281-284.
15. Kilstrup, M., Jacobsen, S., Hammer, K., and Vogensen, F.K. 1997. Induction of heat shock proteins
DnaK, GroEL, and GroES by salt stress in Lactococcus lactis. Applied and Environmental Microbiology.
63: 1826-1837.
16. Periago, P.M., Schaik, W.v., Abee, T., and Wouters, J.A. 2002. Identification of proteins involved in the
heat stress response of Bacillus cereus ATCC 14579. Applied and Environmental Microbiology. 68:
3486-3495.
17. Gouesbet, G., Jan, G., and Boyaval, P. 2002. Two-dimensional electrophoresis study of Lactobacillus
delbrueckii subsp. bulgaricus thermotolerance. Applied and Environmental Microbiology. 68: 1055-1063.
18. Prasad, J., McJarrow, P., and Gopal, P. 2003. Heat and osmotic stress responses of probiotic
Lactobacillus rhamnosus HN001 (DR20) in relation to viability after drying. Applied and Environmental
Microbiology. 69: 917-925.
19. De Angelis, M., Di Cagno, R., Huet, C., Crecchio, C., Fox, P.F., and Gobbett, M. 2004. Heat shock
response in Lactobacillus plantarum. Applied and Environmental Microbiology. 70: 1336-1346.
20. Desmond, C., Fitzgerald, G.F., Stanton, C., and Ross, R.P. 2004. Improved stress tolerance of
GroESL-overproducing Lactococcus lactis and probiotic Lactobacillus paracasei NFBC 338. Applied and
Environmental Microbiology. 10: 5929–5936.
Pressure induced response in probiotic bacteria 191
21. Hendrick, J.P. and Hartl, F.-U. 1993. Molecular chaperone functions of heat-shock proteins. Annual
Reviews in Biochemistry. 62: 349-384.
22. Teixeira, P.M., Castro, H.P., and Kirby, R. 1994. Inducible thermotolerance in Lactobacillus bulgaricus.
Letters in Applied Microbiology. 18: 218-221.
23. Gouesbet, G., Jan, G., and Boyaval, P. 2001. Lactobacillus delbrueckii ssp. bulgaricus
thermotolerance. Lait. 81: 301-309.
24. Desmond, C., Stanton, C., Fitzgerald, G.F., Collins, K., and Ross, R.P. 2001. Environmental
adaptation of probiotic lactobacilli towards improvement of performance during spray drying. International
Dairy Journal. 11: 801-808.
25. VanBogelen, R.A. and Neidhardt, F.C. 1990. Ribosomes as sensors of heat and cold shock in
Escherichia coli. Proceedings of the National Academy of Sciences of the United States of America. 87:
5589-5593.
26. Yura, T., Kanemori, M., and Morita, M.T. 2000. The heat shock response: Regulation and function, in
Bacterial Stress Responses, Storz, G. and Hengge-Aronis, R., Editors. ASM Press: Washington, D.C. p.
3-18.
27. Nakagawa, S. and Ouchi, K. 1994. Improvement of freeze tolerance of commercial baker's yeast in
dough by heat treatment before freezing. Bioscience Biotechnology Biochemistry. 58: 2077-2079.
28. Hartke, A., Bouche, S., Laplace, J.M., Benachour, A., Boutibonnes, P., and Auffray, Y. 1995. UV-
inducible proteins and UV-induced cross-protection against acid, ethanol, H2O2 or heat treatments in
Lactococcus lactis subsp. lactis. Archives in Microbiology. 163: 329-336.
29. Panoff, J.-M., Thammavongs, B., Laplace, J.-M., Hartke, A., Boutibonnes, P., and Auffray, Y. 1995.
Cryotolerance and cold adaptation in Lactococcus lactis subsp. lactis IL1403. Cryobiology. 32: 516-520.
30. Teixeira, P.M., Castro, H.P., and Kirby, R. 1995. Spray drying as a method for preparing concentrated
cultures of Lactobacillus bulgaricus. Journal of Applied Bacteriology. 78: 456-462.
31. de Urraza, P. and de Antoni, G. 1997. Induced cryotolerance of Lactobacillus delbrueckii subsp.
bulgaricus LBB by preincubation at suboptimal temperatures with fermentable sugar. Cryobiology. 35:
159-164.
32. Kim, W.S., Khunajakr, N., and Dunn, N.W. 1998. Effect of cold shock protein synthesis and on
cryotolerance of cells frozen for long periods in Lactococcus lactis. Cryobiology. 37: 86-91.
33. Saarela, M., Rantala, M., Hallamaa, K., Nohynek, L., Virkajärvi, I., and Mättö, J. 2004. Stationary-
phase acid and heat treatments for improvement of the viability of probiotic lactobacilli and bifidobacteria.
Journal of Applied Microbiology. 96: 1205-1214.
34. Wouters, J.A., Rombouts, F.M., de Vos, W.M., Kuipers, O.P., and Abee, T. 1999. Cold shock proteins
and low-temperature response of Streptococcus thermophilus CNRZ302. Applied and Environmental
Microbiology. 65: 4436-4442.
35. Somero, G.N. 1992. Adaptations to high hydrostatic pressure. Annual Reviews in Physiology. 54: 557-
577.
36. Gross, M. and Jaenicke, R. 1994. Proteins under pressure: The influence of high hydrostatic pressure
on structure, function and assembly of proteins and protein complexes. European Journal of
Biochemistry. 221: 617 - 630.
37. Abe, F. and Horikoshi, K. 2001. The biotechnological potential of piezophiles. Trends in Biotechnology.
19: 102-108.
38. Bett, K.E. and Cappi, J.B. 1965. Effect of pressure on the viscosity of water. Nature. 207: 620-621.
39. Abe, F., Kato, C., and Horikoshi, K. 1999. Pressure-regulated metabolism in microorganisms. Trends
in Microbiology. 7: 447-453.
Pressure induced response in probiotic bacteria 192
40. Bartlett, D.H., Kato, C., and Horikoshi, K. 1995. High pressure influences on gene and protein
expression. Research in Microbiology. 146: 697-706.
41. Welch, T.J., Farewell, A., Neidhardt, F.C., and Bartlett, D.H. 1993. Stress response of Escherichia coli
to elevated hydrostatic pressure. Journal of Biotechnology.
42. Kawarai, T., Wachi, M., Ogino, H., Furukawa, S., Suzuki, K., Ogihara, H., and Yamasaki, M. 2004.
SulA-independent filamentation of Escherichia coli during growth after release from high hydrostatic
pressure treatment. Applied Microbiology and Biotechnology. 64: 255 - 262.
43. Drews, O., Weiss, W., Reil, G., Parlar, H., Wait, R., and Görg, A. 2002. High pressure effects step-
wise altered protein expression in Lacbacillus sanfraciscensis. Proteomics. 2: 765-774.
44. Rowan, N.J. 2004. Viable but non-culturable forms of food and waterborne bacteria: Quo Vadis? Trends
in Food Science and Technology. 15: 462-467.
45. Karatzas, K.A.G., Wouters, J.A., Gahan, C.G.M., Hill, C., Abee, T., and Bennik, M.H.J. 2003. The
CtsR regulator of Listeria monocytogenes contains a variant glycine repeat region that affects
piezotolerance, stress resistance, motility and virulence. Molecular Microbiology. 49: 1227-1238.
46. Aertsen, A., Vanoirbeek, K., De Spiegeleer, P., Sermon, J., Hauben, K., Farewell, A., Nyström, T.,
and Michiels, C.W. 2004. Heat shock protein-mediated resistance to high hydrostatic pressure in
Escherichia coli. Applied and Environmental Microbiology. 70: 2660–2666.
47. Fujii, S., Iwahashi, H., Obuchi, K., and Komatsu, Y. 1996. Characterization of a barotolerant mutant of
the yeast Saccharomyces cerevisiae: importance of trehalose content and membrane fluidity. FEMS
Microbiological Letters. 141: 97–101.
48. Iwahashi, H., Obuchi, K., Fujii, S., and Komatsu, Y. 1997. Barotolerance is dependent on both
trehalose and heat shock protein 104 but is essentially different from thermotolerance in Saccharomyces
cerevisiae. Letters in Applied Microbiology. 25: 43-47.
49. Iwahashi, H., Kaul, S.C., Obuchi, K., and Komatsu, Y. 1991. Induction of barotolerance by heat shock
treatment in yeast. FEMS Microbiology Letters. 64: 325-328.
50. Iwahashi, H., Obuchi, K., Fujii, S., and Komatsu, Y. 1997. Effect of temperature on the role of Hsp104
and trehalose in barotolerance of Saccharomyces cerevisiae. FEBS Letters. 416: 1-5.
51. Iwahashi, H., Nwaka, S., and Obuchi, K. 2001. Contribution of Hsc70 to barotolerance in the yeast
Saccharomyces cerevisiae. Extremophiles. 5: 417 - 421.
52. Scheyhing, C.H., Hörmann, S., Ehrmann, M.A., and Vogel, R.F. 2004. Barotolerance is inducible by
preincubation under hydrostatic pressure, cold-, osmotic- and acid-stress conditions in Lactobacillus
sanfranciscensis DSM 20451. Letters in Applied Microbiology. 39: 284–289.
53. Fernandes, P.M., Panek, A.D., and Kurtenbach, E. 1997. Effect of hydrostatic pressure of a mutant of
Saccharomyces cerevisiae deleted in the trehalose-6-phosphate synthase gene. FEMS Microbiology
Letters. 152: 17–21.
54. Tamura, K., Miyashita, M., and Iwahashi, H. 1998. Stress tolerance of pressure-shocked
Saccharomyces cerevisiae. Biotechnology Letters. 20: 1167-1169.
55. Wemekamp-Kamphuis, H.H., Karatzas, A.K., Wouters, J.A., and Abee, T. 2002. Enhanced levels of
cold shock proteins in Listeria monocytogenes LO28 upon exposure to low temperature and high
hydrostatic pressure. Applied and Environmental Microbiology. 68: 456-463.
56. Jones, P.G., VanBogelen, R.A., and Neidhardt, F.C. 1987. Induction of proteins in response to low
temperature in Escherichia coli. Journal of Bacteriology. 169: 2092-2095.
57. Noma, S. and Hayakawa, I. 2003. Barotolerance of Staphylococcus aureus is increased by incubation
at below 0 °C prior to hydrostatic pressure treatment. International Journal of Food Microbiology. 80:
261-264.
Pressure induced response in probiotic bacteria 193
58. Russell, N.J., Evans, R.I., ter Steeg, P.F., Hellemons, J., Verheul, A., and Abee, T. 1995. Membranes
as a target for stress adaptation. International Journal of Food Microbiology. 28: 255-261.
59. Russell, N.J. 2002. Bacterial membranes: the effects of chill storage and food processing. An overview.
International Journal of Food Microbiology. 79: 27-34.
60. Yano, Y., Nakayama, A., Ishihara, K., and Saito, H. 1998. Adaptive changes in membrane lipids of
barophilic bacteria in response to changes in growth pressure. Applied and Environmental Microbiology.
64: 479-485.
61. Iwahashi, H., Shimizu, H., Odani, M., and Komatsu, Y. 2003. Piezophysiology of genome wide gene
expression levels in the yeast Saccharomyces cerevisiae. Extremophiles. 7: 291-298.
62. Fernandes, P.M.B., Dimitrovic, T., Kao, C.M., and Kurtenbach, E. 2004. Genomic expression pattern
in Saccharomyces cerevisiae cells in response to high hydrostatic pressure. FEBS Letters. 556: 153-
160.
63. Craig, E.A. and Gross, C.A. 1991. Is hsp70 the cellular thermometer? Trends in Biochemical Science.
16: 135-140.
64. Yura, T. and Nakahigashi, K. 1999. Regulation of the heat-shock response. Current Opinions in
Microbiology. 2: 153-158.
65. Smeller, L. 2002. Pressure-temperature phase diagrams of biomolecules. Biochimica et Biophysica
Acta. 1595: 11-29.
66. Gross, M., Lehle, K., Jaenicke, R., and Nierhaus, K.H. 1993. Pressure-induced dissociation of
ribosomes and elongation cycle intermediates. Stabilizing conditions and identification of the most
sensitive functional state. Journal of Biochemistry. 218: 463-468.
67. Niven, G.W., Miles, C.A., and Mackey, B.M. 1999. The effects of hydrostatic pressure on ribosome
conformation in Escherichia coli: an in vivo study using differential scanning calorimetry. Microbiology.
145: 419–425.
68. Aertsen, A., Van Houdt, R., Vanoirbeek, K., and Michiels, C.W. 2004. An SOS response induced by
high pressure in Escherichia coli. Journal of Bacteriology. 186: 6133-6141.
69. Arabas, J., Szczepek, J., Dmowski, L., Heinz, V., and Fonberg-Broczek, M. 1999. New technique for
kinetic studies of pressure-temperature induced changes of biological materials, in Advances in high
pressure bioscience and biotechnology, Ludwig, H., Editor. Springer-Verlag: Berlin. p. 537-540.
70. Brashears, M.M. and Gilliland, S.E. 1995. Survival during frozen and subsequent refrigerated storage
of Lactobacillus acidophilus cells as influenced by the growth phase. Journal of Dairy Science. 78: 2326-
2335.
71. Lorca, G.L. and de Valdez, G.F. 1999. The effect of suboptimal growth temperature and growth phase
on resistance of Lactobacillus acidophilusto environmental stress. Cryobiology. 39: 144-149.
72. Kim, W.S. and Dunn, N.W. 1997. Identification of a cold shock gene in lactic acid bacteria and the effect
of cold shock on cryotolerance. Current Microbiology. 35: 59-63.
73. Flahaut, S., Frere, J., Boutibonnes, P., and Auffray, Y. 1996. Comparison of the bile salts and sodium
dodecyl sulfate stress responses in Enterococcus faecalis. Applied and Environmental Microbiology. 62:
2416–2420.
74. Thiebaud, M., Dumay, E., Picart, L., Guiraud, J.P., and Cheftel, J.C. 2003. High-pressure
homogenisation of rawbovine milk. Effects on fat globule size distribution and microbial inactivation.
International Dairy Journal. 13: 427–439.
75. Xiong, R., Xie, G., Edmonson, A.E., and Sheard, M.A. 1999. A mathematical model for bacterial
inactivation. International Journal of Food Microbiology. 46: 45-55.
76. Heinz, V. and Knorr, D. 1996. High pressure inactivation kinetics of Bacillus subtilis cells by a three-
state-model considering distribution resistance mechanisms. Food Biotechnology. 10: 149-161.
Pressure induced response in probiotic bacteria 194
77. Terzieva, S., Donnelly, J., Ulevicius, V., Grinshpun, S.A., Willeke, K., Stelma, G.N., and Brenner,
K.P. 1996. Comparison of methods for detection and enumeration of airborne microorganisms collected
by liquid impingement. Applied and Environmental Microbiology. 62: 2264-2272.
78. Jacobsen, C.N., Rasmussen, J., and Jakobsen, M. 1997. Viability staining and flow cytometric
detection of Listeria monocytogenes. Journal of Microbiological Methods. 28: 35-43.
79. Boulos, L., Prevost, M., Barbeau, B., Coallier, J., and Desjardins, R. 1999. LIVE/DEAD BacLight :
application of a new rapid staining method for direct enumeration of viable and total bacteria in drinking
water. Journal of Microbiological Methods. 37: 77-86.
80. Auty, M.A.E., Gardiner, G.E., McBrearty, S.J., O'Sullivan, E.O., Mulvihill, D.M., Collins, J.K.,
Fitzgerald, G.F., Stanton, C., and Ross, R.P. 2001. Direct in situ viability assessment of bacteria in
probiotic dairy products using viability staining in conjunction with confocal scanning laser microscopy.
Applied and Environmental Microbiology. 67: 420-425.
81. Bunthof, C.J., Schalkwijk, S.v., Meijer, W., Abee, T., and Hugenholtz, J. 2001. Fluorescent method
for monitoring cheese starter permeabilization and lysis. Applied and Environmental Microbiology. 67:
4264-4271.
82. Alonso, J.L., Mascellaro, S., Moreno, Y., Ferrús, M.A., and Hernández, J. 2002. Double-staining
method for differentiation of morphological changes and membrane integrity of Campylobacter coli cells.
Applied and Environmental Microbiology. 68: 5151-5154.
83. Teixeira, P., Castro, H., Mohácsi-Farkas, C., and Kirby, R. 1997. Identification of sites of injury in
Lactobacillus bulgaricus during heat stress. Journal of Applied Microbiology. 83: 219-226.
84. Kovacs, E., Török, Z., Horvath, I., and Vigh, L. 1994. Heat stress induces association of the GroEL-
analog chaperonin with thylakooid membranes in cyanobacterium, Synechocystis PCC 68 03. Plant
Physiology and Biochemistry. 32: 285–293.
85. Török, Z., Horváth, I., Goloubinoff, P., Kovács, E., Glatz, A., Balogh, G., and Vígh, L. 1997.
Evidence for a lipochaperonin: Association of active proteinfolding GroESL oligomers with lipids can
stabilize membranes under heat shock conditions. Proceedings of the National Academy of Sciences of
the United States of America. 94: 2192-2197.
86. Hennequin, C., Porcheray, F., Waligora-Dupriet, A., Collignon, A., Barc, M., Bourlioux, P., and
Karjalainen, T. 2001. GroEL (Hsp60) of Clostridium difficile is involved in cell adherence. Microbiology.
147: 87–96.
87. Hofmann, A.F. 1994. Bile acids, in The liver: biology and pathobiology., I. M. Arias, J.L.B., N. Fausto, W.
B. Jackoby, D. A. Schachter, and D. A. Shafritz, Editor. Raven Press: New York. p. 677–718.
88. Rince, A., Le Breton, Y., Verneuil, N., Giard, J.C., Hartk, A., and Auffray, Y. 2003. Physiological and
molecular aspects of bile salt response in Enterococcus faecalis. International Journal of Food
Microbiology. 88: 207– 213.
89. Moser, S.A. and Savage, D.C. 2001. Bile salt hydrolase activity and resistance to toxicity of conjugated
bile salts are unrelated properties in lactobacilli. Appl. Environ. Microbiol. 67: 3476-3480.
90. Abee, T., Rombouts, F.M., Hugenholtz, J., Guihard, G., and Letellier, L. 1994. Mode of action of nisin
Z against Listeria monocytogenes Scott A grown at high and low temperatures. Applied and
Environmental Microbiology. 60: 1962–1968.
91. Montville, T.J. and Chen, Y. 1998. Mechanistic action of pediocin and nisin: recent progress and
unresolved questions. Applied Microbiology and Biotechnology. 50: 511±519.
92. Saxelin, M., Grenov, B., Svensson, U., Fonden, R., Reniero, R., and Mattila-Sandholm, T. 1999. The
technology of probiotics. Trends in Food Science and Technology. 10: 387-392.
Pressure induced response in probiotic bacteria 195
93. Corcoran, B.M., Ross, R.P., Fitzgerald, G., Stanton, C. 2004. Comparative survival of probiotic
lactobacilli spray dried in the presence of prebiotic substances. Journal of Applied Microbiology. 96:
1024–1039.
94. De Angelis, M. and Gobetti, M. 2004. Environmental stress responses in Lactobacillus: A review.
Proteomics. 4: 106–122.
95. Knorr, D. and Heinz, V. 2001. Development of nonthermal methods for microbial control, in Disinfection,
sterilization, and preservation, Block, S.S., Editor. Lippincott Williams&Wilkins: Philadelphia. p. 853-877.
96. Cheftel, J.C. 1995. Review: High pressure, microbial inactivation and food preservation. Food Science
and Technology International. 1: 75-90.
97. Bâati, L., Fabre Gea, C., Auriol, D., and Blanc, P.J. 2000. Study of the cryotolerance of Lactobacillus
acidophilus: effect of culture and freezing conditions on the viability and cellular protein levels.
International Journal of Food Microbiology. 59: 241-247.
98. Lorca, G.L. and G.F., d.V. 2001. A low-pH-inducible, stationary-phase acid tolerance response in
Lactobacillus acidophilus CRL 639. Current Microbiology. 42: 21-25.
99. Mathys, A. 2004. Beurteilung der Überlebensrate von Lactobacillus bulgaricus nach Gefrier-Tau-
Prozess während der Fermentation und durch Messung der Membranpermeabilisierung mit dem Flow
Cytometer und der elektro-optischen Methode, in Department of Food Biotechnology and Food Process
Engineering. Berlin University of Technology: Berlin. p. 95.
100. Duwat, P., Cesselin, B., Sourice, S., and Gruss, A. 2000. Lactococcus lactis, a bacterial model for
stress response and survival. International Journal of Food Microbiology. 55: 83-86.
101. Ardia, A. 2004. Process considerations on the application of high pressure treatment at elevated
temperature levels for food preservation, in Department of Food Biotechnology and Food Process
Engineering. Berlin University of Technology: Berlin. p. 102.
102. Garcia-Graells, C., Hauben, K.J.A., and Michiels, C.W. 1998. High-pressure inactivation and sublethal
injury of pressure-resistant Escherichia coli mutants in fruit juices. Applied and Environmental
Microbiology. 64: 1566-1568.
196
5 SUMMARY AND OUTLOOK
Summary and Outlook 197
Probiotic products gain growing interest in the global food markets, as documented by
increasing market revenues in Europe and the USA [1]. A key factor responsible for this
development is the improved awareness of consumers about the relationship between diet
and health. This consumer’s perception together with sound scientific evidences about the
effectiveness of probiotic therapy and innovative marketing strategies promote the
willingness of the consumer to buy probiotic products. Apart from using fermented milk as
food vehicle for probiotic consumption there is an increased interest in the diversification of
probiotic containing food products, which eventually require processing and storage
conditions far beyond the ones currently applied and optimized for maintaining good survival
behaviour in fermented dairy products. Regardless of the type of food vehicle the ultimate
requirement for declaring probiotic products as such remains the same, i.e. that the product
should contain living bacteria in a sufficient number until the end of its shelf life. The
technological challenge constantly addressed to scientists working in this field is to identify
critical environmental factors in a given product, which may lead to cell injury or loss of
viability during processing and storage as well as to develop innovative but feasible solution
to sufficiently protect probiotic bacteria in order to maintain the viability levels above the dose
required for eliciting health effects.
The present work is primarily focused on the exploration of spray drying as an alternative
processing method to produce dried probiotic preparations in milk based media. Most of
dried bacterial preparations are currently produced by freeze-drying due to the possibility to
operate at mild conditions, so that the degree of injury could be minimized. However, some
drawbacks of freeze drying process, such as long processing time and high energy
consumption, led to efforts in evaluating alternative drying processes [2]. Spray drying is one
of the promising process for production of dry probiotic preparations, since under optimized
conditions it allows high processing rates, low energy requirements and thus lower operating
costs.
In this context investigations were performed to study the mechanisms leading to cell
damage during drying and the role of the physical state of the drying matrix as well as the
interactions of protective compounds with cellular membranes as related to dehydration
tolerance. Moreover, a pre-adaptation step under sub lethal high pressure conditions was
assessed regarding its potential in increasing tolerance against heat, which is considered a
viability-determining hurdle for probiotic bacteria during spray drying.
To allow identification of cellular injuries occurring during spray drying of L. rhamnosus GG
(LGG) a microbiological analysis method involving flow cytometric analysis in combination
with a multiple staining strategy was prepared and established. The application of this
technique to evaluate the mechanism of microbial inactivation with LGG as model organism
Summary and Outlook 198
by means of physical treatments was discussed in Chapter 2. Physiological fluorescence
dyes carboxyfluoresceindiacetate (cFDA) and propidium iodide (PI) were applied to examine
process-induced changes in cellular integrity or metabolic activities, which were not explicitly
assessable by culture techniques. As a result, it was possible to differentiate the
mechanisms of microbial inactivation occurring during different physical treatments, in order
to allow problematic contaminants to be injured or inactivated more effectively as well as to
effectively combine different treatments, which have different cellular target sites. It was
found that although the survival rates according to the plate counts result were in the same
range, different treatments led to different responses of the cell to cFDA/PI labelling. This
may indicate that the treatments applied differed in the cellular sites being primarily affected.
Regarding heat induced damage on LGG, it was demonstrated that the target sites of heat
inactivation may differ depending on the temperature level used. Based on these results, it is
recommended to expose vegetative cells to temperatures above 60°C, when the inactivation
of intracellular esterase and membrane damage is aimed. In contrast the major population of
high pressure inactivated cells of LGG could accumulate fluorescent molecule
carboxyfluorescein (cF), which indicated that some of the dead cells were still enzymatically
active and not severely membrane compromised. The fact, that pressure inactivated bacteria
could perform enzymatic conversion of cFDA into cF needs further attention, since the
presence of such metabolically active, but dead bacteria in food might be critical in terms of
their potential activity on excreting toxic or food spoiling metabolites. Moreover, it was shown
that the lethal effect of pressures higher than 400 MPa was related to the irreversible
perturbation of dye extrusion machinery, which is most likely mediated by an ATP-driven
transport system. Accordingly, dead cells, which lost the capacity to reproduce themselves
and grow on agar, are the ones which were not able to extrude cF, although the membrane
was still intact and esterase activity remained. These findings underline the results from
previous works on pressure induced damage on other ATP-dependent, membrane bound
enzymes, which are crucial in maintaining viability [3-7]. Likewise, it seems that high intensity
ultrasound did not considerably affect the cytoplasmic membrane, although according to
plate count results viability loss occurred. It could be concluded that, cell death which was
observed upon applying high-intensity ultrasound seemed to result from non-membrane
related degradation.
The importance of the ingestion of viable bacteria in eliciting health effect is sometimes
questioned, since non-viable bacteria were reported to be not only as effective as viable
ones; they are of interest due to easy-handling and longer shelf life [8]. However, systematic
studies on the relationship between probiotic effect and the type of inactivation treatments
used to produce non-viable bacteria are still lacking. In this context, high pressure killed cells
might be one of the promising candidate to be investigated, since the fluorescence pattern of
Summary and Outlook 199
pressure inactivated cells – which is indicative for a lower extent of damage on metabolic
activity and on membrane – is quite similar to the one of viable cells. Moreover, probiotic
bacteria may become dormant during storage, i.e. they retain a functional cell membrane
typical for viable cells, but were not culturable [9]. The transformation into a non-culturable
state seems to be a universal adaptive response in coping with adverse environmental
conditions. Supported by this findings and taking into account that the greater amount of
bacteria in the gastrointestinal tract is not necessarily culturable, the current practice in
putting bacterial culturability as an absolute measure of probiotic effectiveness seems to be
not necessarily appropriate and would exclude many functional but non-culturable strains,
which might be more effective than culturable probiotics in eliciting health effect upon
consumption.
Chapter 2 deals with the comprehensive evaluation of the spray drying process in the
manufacture of probiotic powders. The investigation of bacterial stability during subsequent
storage is included as well. It was found that when reconstituted skim milk (RSM) was used
as the drying medium, a bacterial survival rate ≥ 50% was achievable at an outlet
temperature of 80°C. The powder contained more than 109 cfu g-1 of LGG. Using flow
cytometric analysis, bacterial membranes were identified as the main site of injury during
spray drying. The protective media used as drying medium should therefore be targeted on
protection of membrane against deteriorative effects of water removal and thermal stresses
on membranes occurring during spray drying.
The incorporation of commercial prebiotic substances such as Raftilose®P95 (oligofructose)
or polydextrose in the skim milk powder with the aim to produce synbiotic powder did not
exert any adverse effect on bacterial survival upon spray drying. However, stability of
bacteria during long term storage was impaired by partial substitution of skim milk with either
of the prebiotic substances evaluated. The entrapment of bacteria in an extracellular glassy
state appeared to have only little effect on their stability. Although the glass transition
temperatures of all media were well above the storage temperatures applied, bacterial
inactivation still took place during storage; indicating the insufficiency of entrapment in glassy
state in inhibiting deteriorative events involved in cell death. The decreased protection
capacity of prebiotic containing media could be resulted from the reduced amount of
protective compounds in skim milk solids, which could not adequately substituted by
prebiotics. As a result, due to the presence of oxygen in the storage atmosphere applied in
this study, related deteriorative reaction, most likely lipid oxidation, may take place at a
higher rate.
The assessment of the contribution of sugar molecule to stabilization during drying was
conducted with help of flow cytometric analysis on dried model membrane. It was observed
Summary and Outlook 200
that oligosaccharides present in Raftilose®P95 were not capable of directly interacting with
cytoplasmic membranes in the dehydrated state, whereas polydextrose and lactose could
effectively prevent drying induced leakage on membranes. Moreover, the effect of milk
proteins in stabilization of dried bacteria during storage was observed. When milk proteins
were enzymatically degraded, the performance of such treated RSM-based media in
conferring protection during storage was considerably reduced. In conclusion, data compiled
in this study suggest that the superiority of skim milk over prebiotics in stabilizing dried
bacteria is most likely based on the direct interaction of lactose with bacterial membranes as
well as proteins and on protective effect of milk proteins. The higher susceptibility of bacteria
dried in prebiotic containing matrices to deteriorative events could be reduced by storing
them at refrigerated temperatures.
Furthermore, in Chapter 4 the application of sub lethal high pressure pretreatment was
assessed on its effectiveness in inducing adaptive responses, which may ultimately lead to
an increase of the bacterial tolerance against stress related to harsh conditions encountered
during spray drying. It was shown that pressure pre-treated cells showed higher survivability
than untreated ones when both were exposed to heat treatment at 60°C. Further
investigation with flow cytometric analysis indicated that the acquisition of pressure-induced
heat tolerance was related to membrane stabilization and protein biosynthesis. Pressure
induced thermotolerance occurred as a consequence of stabilization of cellular membranes –
presumably by incorporation of heat shock proteins into cytoplasmic membranes – which in
turn led to an enhanced transient protection against degradative effects of heat on cell
membranes. The absence of induced thermal tolerance upon addition of chloramphenicol, an
inhibitor of protein synthesis, suggested that the proteins expressed during pressure
adaptation was involved in the prevention of thermal degradation on cell membrane.
Pressure pre-treated LGG also showed an improved tolerance against chemical compounds
which are able to degrade cytoplasmic membranes, such as bile acid and nisin. This result
confirmed the positive role of protein synthesis during pressure adaptation in protecting this
vital cellular component. The potential of utilizing pressure shock response in increasing the
survivability of LGG during spray drying, which poses multiple environmental stresses
(including heat, osmotic and oxidative stress) was also demonstrated. In relation to this
observation it was proposed that physical methods appear to be a better choice for the
induction of adaptive response. Exposure to sub-lethal heat could easily be done with
existing equipment; however temperature increase to 47°C for heat pre-treatment took 10
min [10]. Yet, the presence of a radial temperature field with higher temperature at the
fermenter’s wall would lead to an overprocessing of bacterial population at this site. The
limitations of heat transfer into product with small ratio of surface to volume could be
Summary and Outlook 201
compensated by application of pressure, which is transmitted instantaneously throughout all
volume elements of the confined fluid [11]. More studies are required to investigate the
pressure stress response of bacteria from different sub-phases in the stationary growth
stage, since cultures from stationary phase have higher biomass, which makes them more
attractive for industrial production.
In conclusion, the maintenance of high viability level of a given probiotic bacteria in the
dehydrated state is determined by both extrinsic and intrinsic factors, which create a dynamic
environment in extra- or intracellular space (Fig. 1).
As shown in Figure 1, the processing condition used to spray dry bacteria not only has a
direct influence on the bacterial viability but also affects the properties of drying media. With
regards to the latter issue, overprocessing would result in thermal degradation of protective
compounds, such as sugar (due to browning). Low drying rate would facilitate the sugar
crystallization rather than entrapment of bacteria in glassy state.
Drying media
Processing
conditions
Bacterial viability
and stability
Storage
Intrinsic factors
T, t, patomization
T, t, Drying rate
T, gas
composition,
RH
Dynamic microenvironment
Accumulation of
•osmolytes,
• internal sugars,
• heat shock proteins
Growth
conditions &
pre-adaptation
extra- and intracellular
Chemical inertness
Porosity
Encapsulation
Antioxidant
T, gas composition, RH
Glassy state
Hydrogen bonding
Cellular injuries
Figure 1
Interplay among extrinsic and intrinsic factors, which govern the viability and stability of spray dried
bacteria
In terms of the drying media used to protect the bacteria, the thermophysical properties,
chemical inertness, antioxidative properties, possibility of hydrogen bonding with cellular
components upon water removal, etc. should be evaluated. It is expected that physico-
chemical changes occurred in the protective matrix due to changes in the environmental
conditions (humidity, temperature, atmosphere) or as a result of modification in the matrix
composition would lead to a considerable alteration of the microstructure in the vicinity or
inside the dried bacteria. This may eventually lead to higher susceptibility of protected cells
towards reactive molecules which increase the potential of irreversible cellular injuries. This
Summary and Outlook 202
situation necessitates the identification of the critical alterations and how to effectively
prevent them.
Furthermore, the efforts in improving understanding on the nature of cellular injuries
affected by industrial process need to be reinforced. The knowledge acquired from these
microbiological analyses would allow scientific-based solutions, in terms of alteration of
process or environmental parameters or selection of protective media, to be applied in order
to prevent or minimize cellular injury.
It is also of interest to take advantage from the defense mechanism of the cells themselves
in responding to the changes in their environment during the first stages of desiccation (by
accumulation of compatible solutes, synthesis of stress proteins, modified metabolic
pathways, etc.) need to be studied explicitly to effectively utilize this cellular response and
identify the most important stress metabolites involved in the dehydration tolerance. It is also
possible to control the growth of the probiotic bacteria in such a way that increased
accumulation of protective metabolites can be achieved.
References
1. de Jong, L. 2004. Probiotics:strain-specific behaviour - Real challenge for product developers. Food
Engineering & Ingredients. 10: 38-41.
2. Marcotte, M. 2001. Dehydration?- It's not so dry as all that! Le Monde alimentaire. 5: 20-22.
3. Wouters, P.C., Glaasker, E., and Smelt, J.P.P.M. 1998. Effects of high pressure on inactivation kinetics
and events related to proton efflux in Lactobacillus plantarum. Applied and Environmental Microbiology. 64: 509-
514.
4. Ulmer, H.M., Gänzle, M.G., and Vogel, R.F. 2000. Effects of high pressure on survival and metabolic
activity of Lactobacillus plantarum TMW1.460. Applied and Environmental Microbiology. 66: 3966-3973.
5. Ulmer, H.M., Herberhold, H., Fahsel, S., Gänzle, M.G., Winter, R., and Vogel, R.F. 2002. Effects of
pressure-induced membrane phase transitions on inactivation of HorA, an ATP-dependent multidrug resistance
transporter, in Lactobacillus plantarum. Applied and Environmental Microbiology. 68: 1088-1095.
6. Molina-Gutierrez, A., Stippl, V., Delgado, A., Gänzle, M.G., and Vogel, R.F. 2002. In situ
determination of the intracellular pH of Lactococcus lactis and Lactobacillus plantarum during pressure treatment.
Applied and Environmental Microbiology. 68: 4399-4406.
7. Molina-Höppner, A., Doster, W., Vogel, R.F., and Gänzle, M.G. 2004. Protective effect of sucrose and
sodium chloride for Lactococcus lactis during sublethal and lethal high-pressure treatments. Applied and
Environmental Microbiology. 70: 2013-2020.
8. Ouwehand, A.C., Kirjavainen, P.V., Shortt, C., and Salminen, S. 1999. Probiotics: mechanisms and
established effects. International Dairy Journal. 9: 43-52.
9. Lahtinen, S.J., Gueimonde, M., Ouwehand, A.C., Reinikainen, J.P., and Salminen, S.J. 2005.
Probiotic bacteria may become dormant during storage. Applied and Environmental Microbiology. 71: 1662-1663.
10. Saarela, M., Rantala, M., Hallamaa, K., Nohynek, L., Virkajärvi, I., and Mättö, J. 2004. Stationary-
phase acid and heat treatments for improvement of the viability of probiotic lactobacilli and bifidobacteria. Journal
of Applied Microbiology. 96: 1205-1214.
11. Knorr, D. and Heinz, V. 2001. Development of nonthermal methods for microbial control, in Disinfection,
sterilization, and preservation, Block, S.S., Editor. Lippincott Williams&Wilkins: Philadelphia. p. 853-877.
Annexes 203
6 ANNEXES
6.1 Annex 1 : Fermentation profile of L. rhamnosus GG (API 50 CHL, Bio Merieux,
France)
0 d 1 d 2 d
Carbohydrate Fermentation Carbohydrate Fermentation
Row 1 0 Control - Row 4 30 Melibiose -
1 Glycerol - 31 Saccharose -
2 Erythritol - 32 Trehalose +
3 D-arabinose + 33 Inuline -
4 L-arabinose - 34 Melezitose +
5 Ribose + 35 D-raffinose -
6 D-xylose - 36 Amidon -
7 L-xylose - 37 Glycogene -
8 Adonitol - 38 Xylitol -
9 β-methylxyloside - 39 β-gentiobiose +
Row 2 10 Galactose + Row 5 40 D-turanose -
11 D-glucose + 41 L-lyxose -
12 D-fructose + 42 D-tagatose +
13 D-mannose + 43 D-fucose -
14 L-sorbose - 44 L-fucose +
15 Rhamnose - 45 D-arabitol -
16 Dulcitol + 46 L-arabitol -
17 Inositol + 47 Gluconate +
18 Mannitol + 48 2-cetogluconate -
19 Sorbitol + 49 5-cetogluconate -
Row 3 20 α−methyl-D-mannoside -
21 α−methyl-D-glucoside -
22 N-acetyl-glucosamine +
23 Amygdaline +
24 Arbutine +
25 Esculine +
26 Salicine +
27 Cellobiose +
28 Maltose -
29 Lactose -
Annexes 204
6.2 Annex 2 : Detector’s configurations for flow cytometric analysis
Dye
combination
FL1
(Volt)
FL3
(Volt)
Compensation
FL1 in FL3 (%)
Discriminator Sample
cFDA/PI 807 807 28.2 SS = 0 L. rhamnosus GG
(stationary phase)
LIVE/DEAD®
Bac Light
508 508 Not made FS = 8 L. rhamnosus GG
(exponential phase)
cF 961 Not
relevant
17.8 FS = 3 or 4 Liposomes
Annexes 205
6.3 Annex 3 : Estimated residence time of dried particle in spray dryer
4.0 4.5 5.0 5.5 6.0 6.5 7.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
Estimated residence time (s)
Total volume of dryer (L)
60 m3 h-1
54 m3 h-1
48 m3 h-1
36 m3 h-1
24 m3 h-1
Total volume of Büchi Spray Dryer B-191 was estimated to be 6.8 L.
Standard working condition for flow rate of drying air in Büchi B-191 60 m3 h-1 (100% power
of aspirator)
Annexes 206
6.4 Annex 4 : Technical specifications of Raftilose®P95 (Orafti, Tienen, Belgium)
Description
Raftilose®P95
• Is a powder containing mainly oligofructose produced by partial enzymatic hydrolysis of chicory
inulins;
• Is a good food ingredient composed of oligofructose, fructose, glucose and sucrose
Oligofructose
is a mixture of oligosaccharides which are composed of fructose units linked together by ß(2-1)
linkages. Almost every molecule is terminated by a glucose unit. The total number of fructose or
glucose units (= Degree of Polymerisation or DP) of oligofructose ranges mainly between 2 and 8.
Compositional Specifications
All values expressed on dry matter.
Oligofructose ≥ 93.2 %
Glucose + fructose + sucrose < 6.8%
Dry Matter (d.m.) 97 ± 1.5 %
Carbohydrate content > 99.5 %
Ash (sulphated) < 0.2 %
Conductivity (28 Brix) < 250 µS
Heavy Metals Pb, As each <0.2 mg/kg
Cd, Hg each <0.01 mg/kg
pH (30-50°Brix) 5.0 – 7.0
Other information
Aspect fine white powder
Behaviour hygroscopic
Taste Slightly sweet, without aftertaste
Solubility in water > 750 g/L
Wettability in water Excellent
Dispersability in water Excellent. May require stirring
Density Approx. 600 ± 70 g/L
Optimal storage conditions Cool and dry, in its original airtight packaging
Maximum durability minimum 18 months upon delivery
Annexes 207
6.5 Annex 5 : Technical specifications of Polydextrose (Danisco, Copenhagen,
Denmark)
Annexes 208
6.6 Annex 6 : Technical specification of COROLASE®PP (AB Enzymes, Darmstadt,
Germany)
Annexes 209
6.7 Annex 7 : Kinetic of lactose degradation using ß-galactosidase (G-3665, Sigma,
St. Louis, MO)
Dosage : 0.5 mL ß-galactosidase in 100 mL RSM
0 20 40 60 80 100 120
0
20
40
60
80
100
% Lactose degraded (%)
Treatment time (min)
Annexes 210
6.8 Annex 8 : Regression parameters for heat inactivation curves (4.4.4)
(A) Model : linear regression, partial
Pressure treatment time 5 min
Pressure (Mpa) Temperature (°C) k (min-1) R2
100 37 0.256 0.99
100 43 0.203 0.967
100 50 0.592 0.985
200 37 0.529 0.962
200 43 0.366 0.974
200 50 0.737 0.998
Pressure treatment time 10 min
Pressure (Mpa) Temperature (°C) k (min-1) R2
100 37 0.388 0.926
100 43 0.199 0.878
100 50 0.55 0.997
200 37 0.808 0.989
200 43 0.842 0.964
200 50 1.154 0.971
Annexes 211
(B) Model : Weibull distribution
−
−
⋅=
1
0
4
log b
t
EeN
N
N
NE: lower asymptote, b:model parameter
Pressure treatment time 5 min
Pressure (MPa) Temperature (°C) Model parameter, b (-)
100 37 6.8123
100 43 7.09048
100 50 6.6723
200 37 6.00561
200 43 6.35517
200 50 5.2672
Pressure treatment time 10 min
Pressure (MPa) Temperature (°C) Model parameter, b (-)
100 37 6.3918
100 43 6.5109
100 50 5.7465
200 37 5.2006
200 43 4.5051
200 50 3.3871
200
190
180
170
160
150
140
130
120
110
Pressure (MPa)
35
37.5
40
42.5
45
47.5
Temperature (°C)
3
3
3.5
3.5
4
4
4.5
4.5
5
5
5.5
5.5
6
6
6.5
6.5
7
7
Model parameters (-)
Model parameters (-)
200
190
180
170
160
150
140
130
120
110
Pressure (MPa)
35
37.5
40
42.5
45
47.5
Temperature (°C)
5
5
5.5
5.5
6
6
6.5
6.5
7
7
7.5
7.5
Model parameter (-)
Model parameter (-)
rogram
Files\TableCurve\TableCurve
3D
v3\CLIPBRD
.
PRN
ab
3D plots generated under application of Weibull distribution for identification of optimal pre-
treatment conditions at pressure treatment time of 5 (a) or 10 min (b)
212
7 LIST OF DISSEMINATION ACTIVITIES
Scientific articles
Ananta, E., Heinz, V., Schlüter, O., Knorr, D.
Kinetic studies on high-pressure inactivation of Bacillus stearothermophilus spores
suspended in food matrices
Innovative Food Science & Emerging Technologies 2 (2001) 261-272
Ananta, E. & Knorr, D.
Pressure induced thermotolerance of Lactobacillus rhamnosus GG.
Food Research International 36 (2003) 991-997
Ananta, E., Heinz, V. and Knorr, D.
Assessment of high pressure induced damage on Lactobacillus rhamnosus GG by flow
cytometry
Food Microbiology 21 (2004) 567–577
Luscher, C., Balasa, A., Frohling, A., Ananta, E., and Knorr, D.
Effect of high-pressure-induced ice I-to-ice III phase transitions on inactivation of Listeria
innocua in frozen suspension
Applied and Environmental Microbiology 70 (2004) 4021-4029
Ananta, E. et al
Processing effects on the nutritional advancement of probiotics and prebiotics
Microbial Ecology in Health and Disease 16 (2004) 113-124
Ananta, E. and Knorr, D.
Evidence on the role of protein biosynthesis in the induction of heat tolerance of
Lactobaccilus rhamnosus GG by pressure pre-treatment
International Journal of Food Microbiology 96 (2004) 307-313
Ananta, E., Volkert, M. and Knorr, D.
Cellular injuries and storage stability of spray dried Lactobacillus rhamnosus GG
International Dairy Journal 15 (2005) 399-409
213
Ananta, E., Bauer, B., Volkert, M. und Knorr, D.
Sprühtrocknung von probiotischen Bakterien
Deutsche Molkerei Zeitung – dmz 2 (2005) 52-55
Ananta, E., Voigt, D., Zenker, M., Heinz, V. and Knorr, D.
Cellular injuries upon exposure of Escherichia coli and Lactobacillus rhamnosus to high-
intensity ultrasound
Accepted for publication in Journal of Applied Microbiology
Oral presentations
Angersbach, A., Heinz, V., Schlueter, O., Ananta, E., Knorr, D., Bunin, V.
Sicherung der Reproduzierbarkeit von Populationszuständen bei der Untersuchungen von
Mikroorganismen unter Nutzung einer elektro-optischen Messmethode
Proceedings vom GDL-Kongress Lebensmitteltechnologie 8-10 November 2001, Berlin
GDL eV, Bonn
Ananta, E., Heinz, V., and Knorr, D.
Anwendung von hohem hydrostatischen Druck zur Erhöhung der Hitzeresistenz von
probiotischen Mikroorganismen.
Vortrag anlässlich der Sitzung von GVC-VDI Fachausschuss Lebensmittelverfahrenstechnik,
12-14 März 2003, D-Fresing-Weihenstephan
Ananta, E. and Knorr, D.
Pressure-induced stress response of probiotic bacteria Lactobacillus rhamnosus GG and its
potential towards their industrial production.
Oral presentation at the conference New Functional Ingredients and Foods 2003, 9-11 April
2003, Copenhagen, Denmark
Ananta, E., Volkert, M., Gloyna, D. & Knorr, D.
Untersuchungen zur Anwendbarkeit von Sprühtrocknung in der Herstellungstechnologie von
probiotischen Bakterien.
Vortrag anlässlich des Kongresses der Gesellschaft Deutscher Lebensmitteltechnologen, 25-
27 September 2003, Stuttgart-Hohenheim.
214
Ananta, E. and Knorr, D.
Cross-adaptive stress response of pressure pre-treatment on probiotic bacteria:
Characterization and importance for production processes
Oral presentation at ICEF9 International Congress on Engineering and Food, 7-11 March
2004, Montpellier, France
Ananta, E. und Knorr, D.
Zur Optimierung der Herstellung von probiotischen Mikroorganismen – Verfahren, Neue
Ansätze, Methodik.
Vortrag anlässlich des Minisymposiums vom Biotechnologie Centrum (BTC) – Berlin, 9-10
Juli 2004, Berlin
Ananta, E. and Knorr, D.
Use of flow cytometric analysis for the evaluation of microbial inactivation mechanisms
Oral presentation at FoodMicro 2004, 12-16 September 2004, Portoroz, Slovenia
Ananta, E., Volkert, M. and Knorr, D.
Factors influencing survival rate and storage stability of spray dried Lactobacillus rhamnosus
GG
Oral presentation at International Probiotic Conference, 15-19 September 2004, Kosice,
Slovakia
Ananta, E., Volkert, M., Voigt, D., Gunawan, R., and Knorr, D.
The role of protective media in the viability retention of probiotic bacteria Lactobacillus
rhamnosus GG during spray drying and storage
Oral presentation at EFFoST Conference, Food Innovations for an Expanding Europe, 26-29
October 2004, Warsaw, Poland
Ananta, E. und Knorr, D.
Ansätze zur Verbesserung der Stabilität von probiotischen Bakterien in
Lebensmittelsystemen – Ergebnisse aus dem EU-Projekt PROTECH
Vortrag anlässlich des Professorentreffens vom VdF - Verband der Deutschen
Fruchtsaftindustrie e.V., 3 November 2004, Berlin
215
Poster presentations
Heinz, V., Ananta, E. and Knorr, D.
Spore control in food by pressure assisted heating
Poster presentation at EUROCAFT 2001, 5-7 December 2001, Berlin
Ananta, E., Ponanti, M. and Knorr, D.
Improvement of survival rate of probiotic bacteria during spray-drying using high pressure
and heat pre-treatment.
Poster presentation at IBERDESH 2002 – Symposium Drying : Process, structure and
functionality, 25-27 September 2002, Valencia, Spain.
Ananta, E., Heinz, V. and Knorr, D.
Flow cytometric analysis of high pressure treated microorganism.
Poster presentation at SAFE Consortium meeting: Newly emerging pathogens, 24-25 April
2003, Brussels, Belgium.
Ananta, E., Heinz, V. and Knorr, D.
Impact of high pressure treatment on metabolic activities of lactic acid bacteria as assessed
by flow cytometric analysis.
Poster presentation at European Federation of Food Science and Technology (EFFoST)
meeting – Nonthermal processing workshop, 7-10 September 2003, Wageningen, The
Netherlands.
Ananta, E., Voigt, D., Zenker, M. and Knorr, D.
Applicability of high-intensity ultrasound to inactivate Escherichia coli and Lactobacillus
rhamnosus and to increase sensitivity against nisin
Poster presentation at SAFE Consortium meeting: Novel food-preservation technologies, 22-
23 January 2004, Brussels, Belgium.
Ananta, E., Ardia, A., Heinz, V. and Knorr, D.
High pressure treatment of bacteria and spores – Monitoring pressure-induced changes in
microbial metabolic activities and cellular properties using flow cytometer
Poster presentation at ICEF9 International Congress on Engineering and Food, 7-11 March
2004, Montpellier, France.
216
Ananta, E., Volkert, M. and Knorr, D.
Crucial aspects of the application of spray drying in the production of probiotics and
prebiotics containing preparation
Poster presentation at 3rdPROEUHEALTH workshop, 15-17 March, Sitges, Spain
Luscher, C., Ananta, E. and Knorr, D.
Application of high pressure processing to frozen food systems - Influence on
microorganisms
Poster presentation at FoodMicro 2004, 12-16 September 2004, Portoroz, Slovenia
217
8 CURRICULUM VITAE
PERSONAL INFORMATION
Name
Nationality
Date and place of birth
Marital status
: Edwin Ananta
: Indonesian
: Malang, Indonesia, January 23rd 1975
: Married
EDUCATION
August 1993 – September 1994
Visit of German course and university preparatory school at Berlin University of Technology
October 1994 – May 2000
Dipl.-Ing in Food Technology at Berlin University of Technology
Thesis: Combined action of pressure and temperature for decontamination purposes in
different stages of cocoa processing
INTERNSHIP
August - September 1996
Internship at University Potsdam, Germany
Assisting Prof. G. Muschiolik and co-workers in the production and analysis of multiple
emulsions
August - October 1997
Industrial internship at Storck Schokolade GmbH, Berlin, Gemany
Performing routine chemical analysis in the quality control of chocolate production plant
WORKING EXPERIENCE AS STUDENT
December 1997 – March 2000
Student co-worker in Berlin University of Technology, Department of Food Biotechnology and
Food Process Engineering
218
WORKING EXPERIENCE
June 2000 – May 2005
PhD student and research co-worker at Berlin University of Technology, Department of Food
Biotechnology and Food Process Engineering
PhD Thesis: Identification of environmental factors involved in viability and stability of
probiotic bacteria Lactobacillus rhamnosus GG during production processes
and new approaches for improvement
ORGANIZATIONAL ACTIVITIES
Since 1996
Member of GDL (Society of German Food Technologists)
Since 1999
Member of VDI e.V. (“Verein deutscher Ingenieure” - The German Engineers’ Association)
Danksagung
Die Arbeiten für die vorliegende Dissertation wurden in der Zeit von 2001 bis 2005 am Institut für
Lebensmittelbiotechnologie –und prozesstechnik der TU Berlin durchgeführt.
Meinem Doktorvater, Prof. Dr. Dipl.-Ing. Dietrich Knorr danke ich herzlich für die Betreuung der
Arbeit. Insbesondere bedanke ich mich für zahlreiche wertvolle Diskussionen, die sehr große
Freiheit sowie für die Möglichkeit zur Mitwirkung in dem EU Projekt PROTECH.
Herrn Prof. Dipl.-Ing. Dr. Ulf Stahl danke ich für die Übernahme der Aufgaben als zweiter
Berichter.
Herrn Prof. Dr. Herbert Kunzek bin ich dankbar für die Übernahme des Vorsitzes im
Promotionsausschuss.
Weiterhin gilt mein Dank allen jetzigen und ehemaligen Institutsangehörigen für ihre tatkräftige
Unterstützung und die gute Zusammenarbeit in all den Jahren, die dazu geführt haben, dass ich
mich im Institut wohlfühlen und meine Arbeit zu einem erfolgreichen Abschluss bringen konnte.
An dieser Stelle möchte ich mich bedanken bei Dr.-Ing. Volker Heinz, Dr.-Ing. Alexander &
Natalie Angersbach, Dr.-Ing. Oliver Schlüter, Dr.-Ing. Marco Zenker, Dr.-Ing. Adriano Ardia, Dr.-
Ing. Birgit Bauer, Cornelius Luscher, Stefan Töpfl, Roman Buckow, Manuela Guderjan, Elizabeth
Sunny-Roberts, Marcus Volkert, Ana Balasa, Alexander Mathys, Gabriel Urrutia-Benet, Sybille
Candea, Irene Hemmerich, Gisela Martens, Martin Bunzeit, Stefan Boguslawski, Gabriele Ehrlich,
Tanja Wiehle, Celine Tchatchoua, Marcella Ponanti, Asli Demirel, Daniela Voigt und Rosmewi
Gunawan.
Ich bedanke mich auch bei allen Kollegen aus dem EU Projekt PROTECH, von denen ich
während der gesamten Laufzeit des Projektes (2001 bis 2004) viel lernen konnte.
Ein herzliches Dankeschön geht an meine Eltern und meine Familie für die gesamte
Unterstützung während des Studiums und der Promotion. Insbesondere bedanke ich mich sehr
bei meiner Frau, die mir in der gesamten Zeit immer hilfsreich und verständnisvoll zur Seite
stand.
Schlussendlich möchte ich auch meinem Gott und Herrn Jesus Christus dafür danken, dass er
mir bisher geholfen hat. Rückblickend kann ich diese beiden Erkenntnisse nur bestätigen: Groß
sind die Werke des HERRN; wer sie erforscht, der hat Freude daran (Psalm 111,2).
