RESEARCH REVIEW
Summar y . Infectious diseases currently kill more than 15 million people annu-
ally , and the WHO estimates that every year 1.6 million people die from pneumo-
coccal diseases. S tr eptococcus pneumoniae (pneumococcus), a bacterium with a
long biological pedigree, best illustrates the rapid evolution of antibiotic resistance,
which has led to major public health concern. This article discusses the molecular
basis of the two main virulence factors of pneumococcus, the capsule and cell-wall
hydrolases, as well as new approaches to developing medicinal weapons for pre-
venting pneumococcal infections. In addition, current knowledge regarding pneu-
mococcal phages as potential contributors to virulence and the use of lytic enzymes
encoded by these phages as therapeutic tools is reviewed. [ Int Micr obiol 2004;
7(3):163–171]
K ey w ords : S tr eptococcus pneumoniae · pneumococcus · capsule · pneumococc a l
bacteriophages · lytic enzymes
Streptococcus pneumoniae
and its bacteriophages:
one long argument
Intr oduction
Pneumococcus ( S tr eptococcus pneumoniae ) has been a bac-
terium of utmost clinical significance since Louis Pasteur
(1822–1895) and Geor ge Miller Sternber g (1838–1915) dis-
covered it independently in 1881. Its pedigree is indeed
among the most notorious in the microbial world. Until the
1940s, when the discovery of antibiotics brought much-needed
relief to humanity , pneumococcus killed more human beings
than cancer and heart diseases together . At that time, this reg-
ular colonizer of the human pharynx, then called Diplo-
coccus pneumoniae because the cells are often in pairs, had
been used as an experimental vehicle to demonstrate that
DNA was the carrying material of genetic inheritance. Oswald
T . A very (1877–1955), Colin M. Macleod (1909–1972) and
Maclyn McCarty (born 191 1), based their studies of genetic
transformation on a unique phenotypic characteristic of this
microor ganism—the polysaccharide capsule that surrounds the
bacterial cell [2].
Over the last decade, S. pneumoniae has joined the ranks
of so-called reemer ging infectious agents, due in large part to
its genetic plasticity , which endows it with resistance to many
antibiotics thereby making it a major cause of death through-
out the world. Currently , S. pneumoniae infects around 100
million people every year , with a fatality rate of more than
10%. In addition, pneumococcus is responsible for 10–25%
of pneumonia cases, 10% of meningitis infections (with 30%
mortality), and over 50% of cases of otitis media. It has also
been reported to be the cause of disease in 10–20 × 10 5 inhab-
itants/year in four European countries (Denmark, Finland,
Spain and the United Kingdom). As surveillance of the dis-
ease has been limited to hospitalized patients, this figure is
surely an underestimation of the actual incidence [21].
Rubens López
Department of Molecular
Microbiology , Biological Research
Center , CSIC, Madrid, Spain
I NTERNA TIONAL M ICROBIOLOGY (2004) 7:163–171
www .im.microbios.org
Received 4 May 2004
Accepted 30 May 2004
Address for correspondence:
Departamento de Microbiología Molecular
Centro de Investigaciones Biológicas, CSIC
Ramiro de Maeztu, 9
28040 Madrid, Spain
T el. +34-9183731 12. Fax +34-915360432
E-mail: [email protected]
This article is based on the Opening Lecture of the 19th National Congress of the
Spanish Society for Microbiology (SEM), given by the author in Santiago de
Compostela, Spain, on September 22, 2003.
Historical periods ar e dominated by distinct sets of ideas which,
taken together , form a well defined Zeitgeist . Gr eek philosophy ,
Christianity , Renaissance thought, the Scientific Revolution, and
the Enlightenment ar e examples of sets of ideas that dominated
their historical period. The changes fr om one period to the next ar e
usually rather gradual: other changes — mor e abrupt — ar e often
r eferr ed to as r evolutions .
Ernst Mayr , One Long Ar gument , 1991
164 I NT . M ICROBIOL . V ol. 7, 2004
This review focuses on three main aspects of S. pneumo-
niae : (i) molecular analysis of its polysaccharide capsule, (ii)
the lytic enzymes of the bacterium and its phages, and (iii)
the characteristics of pneumococcal bacteriophages.
The capsule
The polysaccharide capsule that completely envelops S. pneu-
moniae acts as a protective layer , isolating the bacterial cell
from its environment, and is the main virulence factor of this
bacterium. Genes located within the capsular cluster encode
enzymes for the synthesis of more than 90 structural and
serological capsular polysaccharides (serotypes) that con-
tribute to the clinical importance of S. pneumoniae . Although
the biochemistry of some serotypes has long been known, it
was not until the early 1990s that the molecular basis of cap-
sule formation in several bacterial species was unraveled. In
1993, the location and isolation of a gene required for the
synthesis of serotype-3 capsule was reported by my laborato-
ry [9]. That genes involved in pneumococcal capsule synthe-
sis form a cluster (Fig. 1) has been known since the late
1950s. The localization and isolation of those genes ( cap / cps )
showed—as biochemists had suggested—that the greater the
complexity of the capsule’ s composition, the more genes
were involved in its formation, and that the cluster was located
between dexB and aliA , which do not participate in capsular
formation (Fig. 1). Genes coding specifically for a given
serotype or serogroup are found in the central region of the
cluster and are flanked by genes encoding functions related
to the regulation and transport of the capsule’ s components.
Dif ferent capsule types may also have common regions in
their clusters [1].
Knowing the location of the cluster for serotype-3 cap-
sule and having observed similarities among the genes from
the region bordering dexB , we were able to prepare molecu-
lar probes for locating other capsular types, including
serotypes 1, 8 and 33F . Currently , the organization of 16
serotypes is known. Figure 1 shows the two dif ferent arche-
types of functional clusters for all the known serotypes, as
demonstrated by type 3 and type 19, the latter having the
most common form of or ganization. The figure also depicts
the peculiar case of capsule formation of type 37.
In the central region of the serotype 19 cluster , genes that
code for the synthesis of each serotype or serogroup are
flanked by genes involved in the regulation and transport of
the capsule’ s components.
The modular structure of the pneumococcal capsule facilitates, i n
a bacterium that under goes natural genetic transformation,
the exchange of specific genes between serotypes by means
of recombination between flanking homologous regions.
This exchange, evidenced by Cof fey et al. [3], is of great
LÓPEZ
Fig. 1. Genetic organization of the S tr eptococcus pneumoniae region containing the cap/cps gene cluster of three
representative capsular types. Large arrows , complete ORFs; thin arr ows , interrupted ORFs. Regions with identical
color and shading have more than 90% identical nucleotides. Inverted matchsticks , putative transcription terminators;
bent arrows , location of functional promoters. In type 3, the four initial ORFs of the capsular operon are not involved
in capsular polysaccharide biosynthesis and are not expressed. T ype 19F represents the most common capsular gene
cluster organization. T ype 37 capsulation is exceptional, since it is due to the presence of a single copy of a gene ( tts )
located far from the cap cluster .
Int. Microbiol.
165
I NT . M ICROBIOL . V ol. 7, 2004
clinical importance, especially when it occurs between
antibiotic-resistant strains. Indeed, such exchanges provoke
an epidemiological challenge to control the universal expan-
sion of some pneumococcal strains. Moreover , the fact
that changes in capsular type by recombination might be
relatively frequent among pneumococci has implications
for the long-term ef ficacy of conjugate pneumococcal
vaccines, which protect only against a limited number of
serotypes [3].
The complexity of pneumococcus biology is evidenced in
the case of serotype 37, in which a unique gene ( tts ) is located
far from the capsular cluster and enables the bacterium to
produce the homopolysaccharide capsule characteristic of
this serotype. This gene, first described in my laboratory , is
involved in the elaboration of the capsule homopolysaccha-
ride sophorose, which is a characteristic component of this
serotype [18]. As a matter of fact, serotype 37 is genetically
binary in that it has both an active gene ( tts ) and an inactive
cluster almost identical to that of serotype 33. The versatile
expression of tts (coding for a glucosyltransferase) was
demonstrated by introducing the gene into a series of gram-
positive species which subsequently became type-37-capsu-
lated [19]. An understanding of these genetically binary
strains is therefore of great clinical importance, since this
phenomenon further facilitates the perpetuation of pneumo-
coccal pathogenic strains in humans, who are their natural
and almost exclusive hosts.
Interestingly , sequencing of the locus galU also revealed
a gene implicated in the synthesis of all known pneumococ-
cal serotypes. galU codes for uridine diphosphoglucose
(UDP-Glc) pyrophosphorylase, which has turned out to be an
ideal metabolic tar get for future clinical approaches aimed at
blocking the formation of S. pneumoniae ’ s capsule using a
new generation of conjugate vaccines [26].
L ytic enzymes
Gram-positive bacteria are surrounded by peptidoglycan
layers (the glycanic chain made of repetitive units of N -
acetylglucosamine and N -acetylmuramic acid linked by pep-
tide bonds), which give them their special shape. For the bac-
terial cell to expand, this rigid sacculum must continuously
adapt. Cellular restructuring requires the action of murein hy-
drolases, which are endogenous enzymes capable of degrad-
ing peptidoglycan by cleaving covalent bonds of the cell
wall. Cell-wall hydrolases (CWH), or lytic enzymes, have
been found in all eubacteria studied so far , and are considered
to play major roles in the biology of bacteria, including cel-
lular expansion, division, and separation of daughter cells.
CWH are crucial in microbial chemotherapy since they are
responsible for the irreversible ef fects caused by β -lactam
antibiotics. Based on their bond specificity , lytic enzymes are
classified as: (i) glycosidases (lysozymes or muramidases and
glucosaminidases), (ii) endopeptidases, and (iii) amidases.
Figure 2 shows the bonds cleaved by dif ferent pneumococcal
CWH. The membrane lipoteichoic acid of pneu ococci seems to
constrain the activity of some pneumoco ccal lytic enzymes. Most
interestingly , teichoic acids, linked to muramic acids, contain
choline in their structures. Höltje and T omasz [16] showed
that this amino alcohol plays a crucial role in the physiology
of the pneumococcal cell wall. In fact, it is an essential lig-
S. PNEUMONIAE AND ITS BACTERIOPHAGES
Fig. 2. The pneumococcal cell wall, showing the
bonds cleaved by different pneumococcal cell-wall
hydrolases (CWH). The upper part of the diagram
indicates the different components of the cell wall.
For simplicity , only two layers are represented.
T ransglycosylases and endopeptidases have not
been found in S. pneumoniae . The small black cir-
cles attached to teichoic acids represent choline
molecules; GlcNAc, N -acetylglucosamine; MurNAc,
N -acetylmuramic acid; L ytA, amidase; L ytB, glu-
cosaminidase; L ytC, lysozyme; Pce, phosphoryl
choline esterase.
Int. Microbiol.
166 I NT . M ICROBIOL . V ol. 7, 2004
and for the activity of the major murein hydrolase L ytA ami-
dase, which carries out 99% of the lytic activity in the cell.
The gene coding for this fundamental enzyme ( lytA ) was
cloned and sequence in 1985 [7], providing the first report of
the molecular manipulation of a bacterial autolysin. When
pneumococci are grown in the presence of the choline ana-
logue ethanolamine, it is incorporated into the cell wall
instead of choline, and the cell under goes dramatic biochem-
ical and morphological changes [16].
It has been also observed that phages release their progeny
using their own lytic enzymes, which, in the pneumococcal
system, have features common to the lytic enzymes of the
bacterium, such as the dependence of the lytic enzymes of
both the host and its phage on the presence of choline in tei-
choic acids for activity . By using lytA as a probe in hybridiza-
tion experiments, we detected lytic genes in phage that are
similar to lytA . Sequence comparison of lytA and the gene
coding for a choline-dependent lysozyme (Cpl-1) from the
pneumococcal bacteriophage Cp-1 revealed that the C-termi-
nal region of L ytA and Cpl-1 are almost identical and share a
modular or ganization [8,1 1]. The N-terminal moieties, how-
ever , are completely dif ferent (Fig. 3) while the C-terminal
region consists of a series of repeated motifs. Using a molecu-
la r approach, we found six motifs, made up of 20 amino
acids, for L ytA and Cpl-1. Further genetic experiments con-
firmed that the C-terminal domain of both proteins with
repeated motifs was involved in recognizing the substrate to
be degraded, whereas the N-terminal domain harbored the
region responsible for catalytic activity [21]. Biochemical
studies revealed that the mechanisms of anchorage and expo-
sure of these and other proteins to the pneumococcal cell sur-
face are highly varied and unusual [22]. Nevertheless, this
structural or ganization was later found in many modular pro-
teins in several bacterial systems.
Crystal structure analyses revealed that the pneumococcal
C-terminal L ytA domain (C-L ytA) is a solenoid structure
consisting exclusively of multiple β -hairpins that together
form a left-handed superhelix [5]. Every hairpin corresponds
to the motif that we had defined by analyzing the protein’ s
primary structure (Fig. 3). For the enzyme to achieve its
structure, each choline molecule must locate in the hydro-
phobic interphase formed by the consecutive hairpins. In
addition, the active form of the enzyme requires the forma-
tion of a boomerang-shaped dimer .
Preparation of deleted L ytA mutants lacking the main
lytic enzyme (M31) of pneumococcus revealed that this en-
zyme was crucial for lysis of cultures at the end of stationary
phase (autolysis) (Fig. 4A). Bacteria that lacked L ytA were
clinically tolerant because the bacteriolytic ef fect induced by
β -lactam antibiotics was eliminated, leading to a bacteriosta-
tic response to the drug [32].
The coordinated use of immunomarking techniques with
colloidal gold and electron microscopy showed that L ytA is
located mainly in the equatorial region of the cell (Fig. 4A).
The lack in M31 of the major amidase unmasked the pres-
ence of other minor proteins that are strongly bound—prob-
ably through choline—to peripheral structures globally
called choline-binding proteins (ChBPs). Identification of the
LÓPEZ
Fig. 3. The structure of the pneumo-
coccal CWH L ytA, L ytB, L ytC, and
Pce and of two other choline-binding
proteins (PspA and PspC). The
domain containing the active center
of the enzymes is shown in light
green (amidase), yellow (lysozyme),
pink (glucosaminidase), or violet
(phosphorylcholine esterase). The
choline-binding domain (blue) and
the choline-binding repeats are also
shown.
Int. Microbiol.
167
I NT . M ICROBIOL . V ol. 7, 2004
N-terminal domains of these purified proteins along with the
availability of the pneumococcus genome sequence enabled
us to isolate and clone the genes that code for two other pneu-
mococcal murein hydrolases, L ytC and L ytB. L ytC and L ytB
have similar structural or ganizations, in which, unlike L ytA,
the region for choline recognition is located at the N-terminal
domain and consists of 1 1 and 18 repeated motifs, respective-
ly . Furthermore, in contrast to L ytA, both enzymes also have
a signal peptide (see Fig. 3). Inactivation of lytB insertion, by
using suicidal plasmids, led to the formation of long cell
chains (Fig. 4C) [4].
L ytC, the first pneumococcal lysozyme described, can be
easily isolated on a DEAE-cellulose column, since DEAE is
a choline analogue. This technique, developed by my labora-
tory , selectively retains ChBPs proteins [34]. Biologically ,
L ytC functions as an autolysin (Fig. 4B) when cultures are
incubated at 30ºC [12]. As the carrier state of pneumococcus
is located in the upper respiratory tract, usually a well-venti-
lated region of the body (ca. 34ºC), this enzyme might play a
role in the natural transformation processes that are carried
out in this location. Likewise, we have observed that cells
lacking this enzyme tend to form clusters.
W e have purified L ytB and identified it as a glucosamini-
dase. Addition of the purified enzyme to cultures of pneumo-
coccus mutants R6B or M31B, which form characteristic
long chains, promoted their dispersion into diplococci or
S. PNEUMONIAE AND ITS BACTERIOPHAGES
Fig. 4. Biological roles of the pneumococcal L ytA (amidase), L ytC (lysozyme) and L ytB (glucosaminidase). ( A ) Curves of
growth of the pneumococcal lytA mutants (M31 and M32), M51 (a strain that overexpresses L ytA), and R6, the wild-type strain
of S. pneumoniae . The inset at the upper left shows the morphology of pneumococcus in the absence (M31) or presence (M51)
of L ytA. The inset at the right shows the immunochemical localization, using gold-labeled antibodies and electron microscopy ,
of L ytA in intact cells. ( B ) Growth curves of the pneumococcal lytA mutants incubated at 30 and 37ºC. The activity responsi-
ble for lysis at this temperature has been ascribed to L ytC. ( C ) Dispersion of M31B chains by the activity of purified L ytB. At
the top left the morphological alteration of a S. pneumoniae strain that lacks lytA (M31) and of the lytA , lytB mutants as exam-
ined by phase-contrast microscopy (bar = 25 µ m). Black arrows , morphology of M31 lytB mutants examined by transmission
electron microscopy (bar = 2 µ m). The bottom right inset shows thin sectioning and electron microscopy of R6B cells untreat-
ed (bottom, bar = 1 µ m) or treated (top, bar = 0.5 µ m) with purified L ytB. Note that neighboring cells in the chain are separat-
ed by cross walls of normal dimensions and normal depth that are disrupted upon treatment with L ytB ( white arr ow ).
Int. Microbiol.
168 I NT . M ICROBIOL . V ol. 7, 2004
short chains, the typical morphology of R6 and M31 strains,
respectively (Fig. 4C). In addition, the preparation of chime-
ric enzymes by means of a translational fusion between the
gene coding for green fluorescent protein (GFP), gfp , and
lytB showed that L ytB accumulates at the cell poles, where it
might very selectively lyse the cell wall (not shown).
V ariations in the composition of choline motifs might
account for the selective recognition of L ytB; for example,
there could be specific receptors for the enzyme at the polar
region of the cell surface, where peptidoglycan hydrolysis
takes place. This could explain why L ytB, unlike L ytA and
L ytC, does not behave as an autolytic enzyme. Since cellular
dispersion might be a major factor in virulence, the lack of
L ytB should make pneumoccal dissemination during the
infectious process more dif ficult.
Pneumococcal bacteriophages
The lack of a simple culture medium and reproducible tech-
niques to visualize phage plaques on pneumococcal lawns
delayed the isolation and identification of phages in this sys-
tem. In fact, it was not until 1975 that a phage was first iso-
lated, independently by C. Ronda and M. McDonnell (Dp-1),
in A. T omasz’ s laboratory [24 ] , and by J.G . T iraby , in M.
Fox’ s laboratory ( ω 1) [37]. Later on, in 1981, our laboratory
isolated the Cp (Complutense phage) family [30 ] . Since then
we have analyzed a series of lytic and temperate phages [21].
Their morphologies are shown in Fig. 5.
Using this system, we also carried out transfection exper-
iments employing pneumococcal phage Cp-1 DNA as donor
LÓPEZ
Int. Microbiol.
Fig. 5. Electron micrographs of nega-
tively stained preparations of purified
pneumococcal bacteriophages. ( A ) Dp-1;
( B ) Cp-1; ( C ) HB-3; ( D ) EJ-1, ( E ) MM1.
169
I NT . M ICROBIOL . V ol. 7, 2004
material and intact competent cells as receptors. While trans-
fecting activity disappeared when Cp-1 DNA was treated
with proteolytic enzymes, transfection with bacteriophage
Dp-4 DNA was not affected [31]. Salas and V iñuela had pre-
viously shown that the genomes of several small Bacillus
phages are associated with proteins [33]. It was also reported
that a protein covalently linked to the 5´ end of phage DNA,
through a phosphodiester bond between serine and dAMP ,
was essential for replication of φ 29 phage DNA in vivo [14].
As in the case of Bacillus subtilis φ 29 phage, Cp-1 DNA has
a protein covalently bound to DNA (in this case through a
phosphothreonine bond) [10,13] that serves as the initiator of
its DNA replication by a “sliding-back” mechanism, similar
to that described for φ 29 phage [25]. These observations
explain the loss of transfecting ability of the Cp-1 DNA-pro-
tein complex treated with proteolytic enzymes.
Our laboratory also showed that the presence of choline
in cell-wall receptors was essential for the adherence of cer-
tain phages, e.g. Dp-1. In fact, when choline was replaced by
ethanolamine, pneumococcal cells could not adsorb to the
cell wall. In addition, the cells were resistant to lysis by bac-
terial or phage-encoded lysins [21]. More recently , sequenc-
ing of the complete genomes of two lytic and two temperate
phages [23,27–29], Cp-1, Dp-1, MM1 and EJ-1, respective-
ly , enabled us to determine their functional or ganization.
Several of the genes in these clusters are the focus of our cur-
rent research, for example, a gene in the genome of Dp-1
( orf55, coding for the anti-receptor) that codes for a protein
exhibiting motifs similar to those in ChBPs for recognition of
choline in the cell wall. This observation explains the choline
requirement of Dp-1 phage receptors for adsorption of the
phage Dp-1 [21]. W e have also described the structural
or ganization of four ChBPs, identified as CWHs, that are
responsible for the specific recognition of choline units and
which liberate progeny from infected S. pneumoniae. These
pneumococcal lytic enzymes have great intrinsic flexibility ,
which enables them to change the recognition units from the
C-terminal the N-terminal region (Fig. 4). This ability fulfills
the functional exchanging properties that Doolittle attributed
to a well-defined domain [22] . W e also observed that the
murein hydrolase of phage Cp-7, which is very similar to Cp-
1, was an exception to the choline dependency of phage lytic
enzymes, since the enzyme degrades the pneumococcal wall
in the absence of the amino alcohol. These features are
reflected in the primary structure of the protein, as the motifs
for choline recognition are replaced by three lar ge, identical
48-amino-acid motifs [1 1] .
T o obtain direct experimental evidence for the modular
or ganization of pneumococcal enzymes, we constructed
chimeric functional phage-bacterial lytic enzymes [6]. The
functional characteristics of these proteins dif fered from
those of the parental enzymes. Furthermore, we produced
inter generic chimeric enzymes with Clostridium aceto-
butylicum [21] . The results allowed us to postulate that this
kind of exchange can endow the pneumococcal system with
great plasticity , in that new enzymatic combinations, by
means of simple genetic recombination, are possible.
Experiments based on the isolation, cloning and purifica-
tion of the Dp-1 phage lytic enzyme Pal supported this
working hypothesis. This enzyme, which is an amidase, has
an N-terminal region very similar to that of a lactococcus
phage amidase, whereas the C-terminal domain is highly
similar to those recognizing choline-containing substrates.
This means that Pal is a natural inter generic chimera [36]
and suggests that the catalytic ef ficiency of a primigenic
enzyme lacking a recognition unit can be improved by join-
ing it with such a unit.
Since 1927 it has been known that temperate phages pos-
sess genes that, in the lysogenic state, are capable of express-
ing toxins associated with bacterial virulence. For example,
in the case of scarlet fever , it was previously observed that
filtered supernatants of toxigenic streptococcal cultures con-
tained scarlatinal toxin. Although poorly understood at the
time, this phenomenon was suggested t o be due to transduc-
tion—the transfer of genetic material to a bacterial cell via phage
infection [38]. The hypothesis that bacteria acquire virulence
properties from phages has since been widely accepted. In fact, it
has been shown that many virulence genes are transferred among
bacteria by phages (via transduction) and other mobile genetic ele-
ments, such as plasmids (via conjugation), as well as by incorpo-
ration of the phage genome into the bacterial chromosome. Th i s
type of mechanism was later ascribed to cholera, diphtheria
and several other bacterial diseases, and is most likely the
case in S. pneumoniae . In fact, 70% of pneumoccocal geno-
mes from clinical isolates have temperate phages. How these
phages contribute to pneumococcal virulence is under inves-
tigation in our laboratory . It is probable that the causes of vir-
ulence also follow other patterns.
Since their discovery by d’Herelle and T wort some 90
years ago, phages have been used as an antibacterial therapy
in Eastern Europe [21]. A brilliant experimental variant to
phage therapy was developed by Loef fler et al. [20], who
used lytic enzymes; in the case of pneumococcus, these
assays included the Pal amidase [20] . More recently , this
experimental approach was extended to Bacillus anthracis ,
with the aim of fighting anthrax spread by bioterrorism [35],
and to S tr eptococcus pyogenes .
The lytic enzymes used in such experimental approaches
have been designed as enzybiotics , and instillations of these
purified enzymes have been tested in a murine model of
S. PNEUMONIAE AND ITS BACTERIOPHAGES
170 I NT . M ICROBIOL . V ol. 7, 2004
pharynx infection. In addition, a murine septicemia model of
S. pneumoniae was developed in collaborative research
between my laboratory and the Instituto Carlos III, in
Madrid. Our results showed that a single dose of Cpl-1
lysozyme or Pal amidase could protect experimental animals
from fatal infection by virulent pneumococcus obtained from
a clinical isolate. So far , there have been no adverse reactions
arising from the application of these murein hydrolases [17].
Thus, enzybiotics of fer a promising new therapeutic
approach to lessen the alarming resistances evolved by many
bacterial species, especially pneumococcus, against a wide
array of antibiotics due to the genetic plasticity of bacteria
and to the misuse of these drugs.
Conclusion
The capsular polysaccharide is an absolute requirement for
pneumococcal infections and thus serves as the basis for the
development of the polysaccharide and conjugate vaccines
that are currently available. Elucidation of the wide genetic
variability underlying capsular formation requires detailed
knowledge of the molecular mechanisms involved in pro-
ducing the 90 serotypes already known. Moreover , detailed
analysis of the unique characteristics of the lytic enzymes
of pneumococcus and its phages has provided valuable
information on their biological roles while also leading
to the therapeutical use of these enzymes in treating infec-
tions.
Phages are the most abundant living entities in nature:
there are an estimated 10 31 phage particles on Earth, and
they represent a pool of biological agents whose potential
for the control of human pathogens has not yet been
explored [21]. The detailed knowledge provided by analy-
sis of the crystalline structure of phage lytic enzymes—
including Cpl-1 lysozyme (see Cover of this issue), recent-
ly established by Hermoso et al. [15]—will surely con-
tribute to establishing a wide range of clinical applications
of these proteins.
Seneca, the Roman philosopher born in the Iberian
Peninsula, stated that: “Peoples that for get their history are
sentenced to repeat it.” Although microorganisms have
caused more deaths than all wars throughout history (more
than 500 million in the twentieth century alone), they have
also contributed to the enormous progress in molecular biology
and to the improved social welfare of developed countries.
Hopefully , microbial-based therapies will also contribute to
fighting infectious diseases in less fortunate countries, in which
they remain the main cause of mortality .
Ac knowledgements . My friends and collaborators E. García, C.
Ronda, P . García and J. L. García, who have dedicated many years of effort
and commitment to work in pneumococcus and its phages, and A. T omasz,
my mentor at Rockefeller University , deserve special recognition and my
long-lasting gratitude. I sincerely thank both my former and current students
and the many colleagues who have contributed to enriching the more than 40
years that I have carried out scientific research. I especially thank Mercè
Piqueras and Marta León, from the Publication Board of I NTERNA TIONAL
M ICROBIOLOGY , for their efficient collaboration in the final form of this arti-
cle, based on the inaugural lecture of the 19th Congress of the Spanish
Society for Microbiology .
References
1. Arrecubieta C, López R, García E (1996) T ype 3-specific synthase of
S tr eptococcus pneumoniae (Cap3B) directs type 3 polysaccharide
biosynthesis in Escherichia coli and in pneumococcal strains of differ-
ent serotypes. J Exp Med 184:449–455
2. A very OT , MacLeod CM, McCarty M (1944) Studies on the chemical
nature of the substance inducing transformation of pneumococcal types.
Induction of transformation by a desoxyribonucleic acid fraction isolat-
ed from pneumococcus type III. J Exp Med 76:127–158
3. Coffey TJ, Enright MC, Daniels M, Morona JK, Morona R, Hryniewicz
W , Paton JC, Spratt BG (1998) Recombinational exchanges at the cap-
sular polysaccharide biosynthetic locus lead to frequent serotype
changes among natural isolates of S tr eptococcus pneumoniae . Mol
Microbiol 27:73–83
4. De las Rivas B, García JL, López R, García P (2002) Purification and
polar localization of pneumococcal L ytB, a putative endo- β - N -acetilglu-
cosaminidase: the chain dispersing murein hydrolase. J Bacteriol
184:4988–5000
5. Fernández-T ornero C, López R, García E, Giménez-Gallego G , Romero A
(2001) A novel solenoid fold in the cell wall anchoring domain of the
pneumococcal virulence factor L ytA. Nat S truct Biol 8:1020–1024
6. Díaz E, López R, García JL (1990) Chimeric phage-bacterial enzymes: A
clue to the modular evolution of genes. Proc Natl Acad Sci USA
87:8125–8129
7. García E, García JL, Ronda C, García P , López R (1985) Cloning and
expression of the pneumococcal autolysin gene in Escherichia coli . Mol
Gen Genet 201:225–230
8. García E, García JL, García P , Arrarás A, Sánchez-Puelles JM, López R
(1988) Molecular evolution of lytic enzymes of S tr eptococcus pneumo-
niae and its bacteriophages. Proc Natl Acad Sci USA 85:914–918
9. García E, García P , López R (1993) Cloning and sequencing of a gene
involved in the synthesis of the capsular polysaccharide of S tr epto-
coccus pneumoniae type 3. Mol Gen Genet 239:188–195
10. García E, Gómez A, Ronda C, Escarmís C, López R (1983)
Pneumococcal bacteriophage Cp-1 contains a protein bound to the 5´
termini of its DNA.V irology 128:92–104
1 1. García P , García JL, García E, Sánchez-Puelles JM, López R (1990)
Modular organization of the lytic enzymes of S tr eptococcus pneumo-
niae and its bacteriophages. Gene 86:81–88
12. García P , González MP , García E, García JL, López R. (1999) The
molecular characterization of the first autolytic lysozyme of
S tr eptococcus pneumoniae reveals evolutionary mobile domains. Mol
Microbiol 33:128–138
13. García P , Hermoso JM, García JA, García E, García JL, López R, Salas
M (1986) Formation of a covalent complex between the terminal pro-
tein of pneumococcal bacteriophage Cp-1 and 5´-dAMP . J. V irol.
58:31–35
14. Hermoso JM, Salas M (1980) Protein p3 is linked to the DNA of phage
φ 29 through a phosphoester bond between serine and 5´-dAMP . Proc
Natl Acad Sci USA 77:6425–6428
LÓPEZ
171
I NT . M ICROBIOL . V ol. 7, 2004
15. Hermoso JA, Monterroso B, Albert A, Galan B, Ahrazem O, García P ,
Martínez-Ripoll M, García JL, Menéndez M (2003) Structural basis for
selective recognition of the pneumococcal cell wall by modular
endolysin from phage Cp-1. Structure 1 1:1239–1249
16. Höltje JV , T omasz A (1975) Specific recognition of choline residues in
the cell wall teichoic acid by the N -acetylmuramyl- L -alanine amidase of
pneumococcus. J Biol Chem 250:6072–6076
17. Jado I, López R, García E, Fenoll A, Casal J, García P , Spanish
Pneumococcal Infection Study Network (2003) Phage lytic enzymes as
therapy for antibiotic-resistant S tr eptococcus pneumoniae infection in a
murine sepsis model. J Antimicrob Chemother 52:967–973
18. Llull D, Muñoz R, López R, García E (1999) A single gene ( tts ) located
outside the cap locus directs the formation of S tr eptococcus pneumoniae
type 37 capsular polysaccharide. T ype 37 pneumococci are natural,
genetically binary strains. J Exp Med 190:241–252
19. Llull D, García E, López R (2001) Tts, a processive β -glucosyltrans-
ferase of S tr eptococcus pneumoniae , directs the synthesis of the
branched type 37 capsular polysaccharide in pneumococcus and other
Gram-positive species. J Biol Chem 276:21053–21061
20. Loeffler J, Nelson D, Fischetti V A (2001) Rapid killing of S tr eptococcus
pneumoniae with a bacteriophage cell wall hydrolase. Science
294:2170–2172
21. López R, García E (2004) Recent trends on the molecular biology of
pneumococcal capsules, lytic enzymes, and bacteriophage. FEMS
Microbiol Rev doi:10.1016/j.femsre.2004.05.002
22. López R, García E, García P , García JL (2004) Cell wall hydrolases. In:
T uomanen EI (ed) The Pneumococcus. ASM Press W ashington, DC, pp
75-89
23. Martín AC, López R, García P (1996) Analysis of the complete
nucleotide sequence and functional organization of the genome of
S tr eptococcus pneumoniae bacteriophage Cp-1. J V irol 70:3678-3687
24. McDonnell M, Ronda C, T omasz A (1975) “Diplophage”: a bacterio-
phage of Diplococcus pneumoniae . V irology 63:577–582
25. Meijar WJ, Horcajadas JA, Salas M (2001) φ 29 Family phages.
Microbiol Mol Biol Rev 65:261–287
26. Mollerach M, López R, García E (1998) Characterization of the galU
gene of S tr eptococcus pneumoniae encoding a uridine diphosphoglu-
cose pyrophosphorylase: a gene essential for capsular polysaccharide
biosynthesis. J Exp Med 188:2047–2056
27. Obregón V , García JL, García E, López R, García P (2003) Genome
organization and molecular analysis of the temperate bacteriophage
MM1 of S tr eptococcus pneumoniae . J Bacteriol 185:2362–2368
28. Obregón V , García JL, García E, López R, García P (2004) Peculiarities
of the MM1, a temperate phage of S tr eptococcus pneumoniae . Int
Microbiol 7:133–137
29. Romero P , López R, García E (2004) Genomic organization and molec-
ular analysis of the inducible prophage EJ-1, a mosaic myovirus from an
atypical pneumococcus. V irology 322:239–252
30. Ronda C, López R, García E (1981) Isolation and characterization of a
new bacteriophage, Cp-1, infecting S tr eptococcus pneumoniae . J V irol
40:551–559
31. Ronda C, López R, Gómez A, García E (1983) Protease-sensitive trans-
fection of S tr eptococcus pneumoniae with bacteriophage Cp-1 DNA.
J V irol 48:721–730
32. Sánchez-Puelles JM, Ronda C, García JL, García P , López R, García
E (1986) Searching for autolysin functions. Characterization of a
pneumococcal mutant deleted in the lytA gene. Eur J Biochem
158:289-293
33. Salas M, V iñuela E (1980) Proteins covalently linked to viral nucleic
acid. T rends Biochem Sci 5:191–193.
34. Sanz J, López R, García JL (1988) Structural requirement of choline
derivatives for convertion of pneumococcal amidase. FEBS Lett
232:308-312
35. Schuch R, Nelson D, Fischetti V A (2002) A bacteriolytic agent that
detects and kills Bacillus anthracis (2002) Nature 418:884–889
36. Sheehan M, García JL, López R, García P (1997) The lytic enzyme of
the pneumococcal phage Dp-1: a chimeric lysin of intergenic origin.
Mol Microbiol 25:717–725
37. T iraby JG , T iraby E, Fox MS (1975) Pneumococcal bacteriophages.
V irology 68:566–569
38. W agner PL, W aldor MK (2002) Bacteriophage control of bacterial viru-
lence. Infect Immun 70:3985–3993
S. PNEUMONIAE AND ITS BACTERIOPHAGES
S treptococcus pneumoniae y sus bacteriófagos: una pr o-
longada contr oversia
Resumen. Las enfermedades infecciosas matan anualmente a unos 15
millones de personas y la OMS estima que 1,6 millones de esas muertes se
deben a infecciones neumocócicas. S tr eptococcus pneumoniae (neumoco-
co), una bacteria con una notable contribución histórica a la biología, es el
mejor ejemplo que ilustra el rápido desarrollo de la resistencia a los anti-
bióticos, lo que puede originar un grave problema sanitario. Esta revisión
analiza las bases moleculares de los dos factores principales de virulencia en
el neumococo, la cápsula y las hidrolasas de la pared celular y describe
nuevos enfoques para el desarrollo de nuevas herramientas médicas para
prevenir las infecciones neumocócicas. T ambién se analizan el conocimien-
to actual de la posible contribución de los fagos de neumococo a la virulen-
cia de esta bacteria y el uso como arma terapéutica de las enzimas líticas
codificadas por estos fagos. [ Int Micr obiol 2004; 7(3):163–171]
P alabras clav e: S tr eptococcus pneumoniae · neumococo · cápsula ·
bacteriófagos de neumococo · enzimas líticas
S treptococcus pneumoniae e seus bacteriófagos: uma pr o-
longada contr ovérsia
Resumo. As enfermidades infecciosas matam anualmente cêrca de 15 mi-
lhões de pessoas. A OMS estima que 1,6 milhões dessas mortes são decor-
rentes de infeccões pneumocócicas. S tr eptococcus pneumoniae (pneumoco-
co), uma bactéria com uma notável contribuição histórica para a biologia, é
o melhor exemplo que ilustra o rápido desenvolvimento da resistência aos
antibióticos, o que pode originar um grave problema sanitário. Esta revisão
analisa as bases moleculares dos fatores principais de virulência no pneumo-
coco, a cápsula e as hidrolases da parede celular e descreve novos enfoques
para o desenvolvimento de ferramentas médicas inovadoras para prevenir as
infecções pneumocócicas. T ambém foi analisado o atual conhecimento da
possível contribuição dos fagos de pneumococo para a virulência desta bac-
téria e o uso como arma terapéutica das enzimas líticas codificadas por estes
fagos. [ Int Micr obiol 2004; 7(3):163–171]
P alavras c have: S tr eptococcus pneumoniae · pneumococo · cápsula ·
bacteriófagos de pneumococo · enzimas líticas
