ma ine d ugs
A icle
Op imiza ion o G ow h and Ca o enoid P oduc ion
by Halo e ax medi e anei Using Response
Su ace Me hodology
Zaida Mon e o-Loba o 1, Ad ián Ramos-Me chan e 2, Juan Luis Fuen es 1, Ana Sayago 3,
Ángeles Fe nández-Recamales 3, Rosa Ma ía Ma ínez-Espinosa 4, JoséMa ía Vega 5,
Ca los Vílchez 1,* and Inés Ga bayo 1
1Algal Bio echnology G oup, CIDERTA and RENSMA, Uni e si y o Huel a, 21071 Huel a, Spain;
[email p o ec ed] (Z.M.-L.); [email p o ec ed] (J.L.F.); [email p o ec ed] (I.G.)
2Depa men o In eg a ed Sciences, Facul y o Expe imen al Sciences, Uni e si y o Huel a,
21007 Huel a, Spain; [email p o ec ed]
3Depa men o Chemis y, Facul y o Expe imen al Sciences, Uni e si y o Huel a, 21007 Huel a, Spain;
[email p o ec ed] (A.S.); [email p o ec ed] (A.F.-R.)
4Biochemis y and Molecula Biology Di ision, Ag ochemis y and Biochemis y Depa men ,
Facul y o Sciences, Uni e si y o Alican e, E-03080 Alican e, Spain; [email p o ec ed]
5Depa men o Plan Biochemis y and Molecula Biology, Facul y o Chemis y, Uni e si y o Se ille,
41012 Se ille, Spain; [email p o ec ed]
*Co espondence: [email p o ec ed]; Tel.: +34-959-218-442
Recei ed: 11 Sep embe 2018; Accep ed: 3 Oc obe 2018; Published: 9 Oc obe 2018


Abs ac :
Halo e ax medi e anei p oduces C50 ca o enoids ha ha e s ong an ioxidan p ope ies.
The esponse su ace me hodology (RSM) ool helps o accu a ely analyze he mos sui able condi ions
o maximize C50 ca o enoids p oduc ion by haloa chaea. The e ec s o empe a u e (15–50
◦
C),
pH (4
−
10), and salini y (5–28% NaCl (w/ )) on he g ow h and ca o enoid con en o H. medi e anei
we e analyzed using he RSM app oach. G ow h was de e mined by measu ing he u bidi y
a 600 nm. To de e mine he ca o enoid con en , ha es ed cells we e lysed by eeze/ hawing,
hen e-suspended in ace one and he o al ca o enoid con en de e mined by measu ing he
abso bance a 494 nm. The analysis o ca o enoids was pe o med by an HPLC sys em coupled wi h
mass spec ome y. The esul s indica ed he heo e ical op imal condi ions o 36.51 o 36.81
◦
C, pH o
8.20 o 8.96, and 15.01% o 12.03% (w/ ) salini y o he g ow h o haloa chaea (
OD600 = 12.5 ±0.64
)
and p oduc ion o o al ca o enoids (3.34
±
0.29 mg/L), espec i ely. These condi ions we e alida ed
expe imen ally o g ow h (OD600 = 13.72
±
0.98) and ca o enoid p oduc ion (3.74
±
0.20 mg/L).
The ca o enoid p o ile showed ou isome s o bac e io ube in (89.13%). Ou indings sugges ha
he RSM app oach is highly use ul o de e mining op imal condi ions o la ge-scale p oduc ion o
bac e io ube in by haloa chaea.
Keywo ds:
bac e io ube in; Halo e ax medi e anei; esponse su ace me hodology (RSM); cen al
composi e design (CCD)
1. In oduc ion
Ca o enoids (ca o enes and xan hophylls) a e pigmen s p esen in all li ing o ganisms; howe e ,
hey a e syn hesized only by bac e ia, algae, ungi, and plan s. They comp ise a la ge amily o
o e 700 na u ally-occu ing pigmen s cha ac e is ically p esen in lea es, lowe s, and ui s o
plan s, whe e hey play a ious oles. In plan s and algae, hey u ilize ligh ene gy o suppo he
chlo ophyll-dependen pho osyn he ic elec on low inside he chlo oplas s. In addi ion, ca o enoids
Ma . D ugs 2018,16, 372; doi:10.3390/md16100372 www.mdpi.com/jou nal/ma ined ugs
Ma . D ugs 2018,16, 372 2 o 12
dissipa e excess ligh ene gy and owing o hei an ioxidan ac i i y, p o ec he pho osyn he ic
machine y agains pho oinhibi ion caused by ee oxygen adicals [1].
Ca o enoids also play an impo an ole in human heal h by ac ing as p o i amin A, which p o ec s
agains macula degene a i e disease and cance . These e ec s, coupled wi h he ac ha humans
use ca o enoids om hei die , make hese pigmen s highly aluable o oods, pha maceu ics,
and cosme ics indus ies. Al hough hey a e usually comme cially p oduced by chemical syn hesis,
mic oo ganisms can also be impo an al e na i e sou ces o ca o enoids and hei ac i e isome s.
β
-Ca o ene, as axan hin, lu ein, and can haxan hin a e C40 ca o enoids, which a e highly aluable o
bio echnological pu poses [2,3].
Halophilic a chaea include mic oo ganisms ha g ow op imally in cul u e media wi h high sal
concen a ions o up o 4 M. The amily Halo e acaceae comp ise non-pho osyn he ic and la gely
ae obic he e o ophs, which p oduce ca o enoids as componen s o hei cy oplasmic memb anes,
especially unde condi ions o low salini y in he medium [
4
]. Apa om ca o enoids, haloa chaea
also p oduce high-added- alue p oduc s o bio echnological in e es , such as enzymes capable
o being ac i e a high empe a u e and high ionic s eng h, polysaccha ides, polyalkanoa es,
and polyhyd oxybu y a e [
5
]. In addi ion, Halo e ax medi e anei exc e es halocins capable o killing
o he a chaea. Halocin H4 is a p o ein o mass 34.9 kDa ha a ge s he plasma memb ane o
mic oo ganisms, e ec ing change in pe meabili y and causing ionic imbalance [6].
Usually, he C50 ca o enoid bac e io ube in and i s de i a i es monoanhyd obac e io ube in
and bisanhyd obac e io ube in a e he majo ca o enoids p oduced by halophilic a chaea.
These ca o enoids may be ound as ans and cis isome s [
7
]. They imp o e he igidi y and luidi y
o he cell memb ane [
8
], and, owing o hei s ong an ioxidan p ope ies, p o ec he cells om he
ha m ul e ec s o adia ion ene gy as well as om osmo ic s ess p oduced by low salini y in he
medium [
9
,
10
]. Se e al halophilic bac e ia also p oduce o he ca o enoids such as
β
-ca o ene, lycopene,
and can haxan hin [
11
,
12
]. C50 ca o enoids p oduced by haloa chaea possess highe an ioxidan
capaci y han C40 ca o enoids p oduced by mos pho osyn he ic o ganisms, due o he highe numbe
o pai s o conjuga ed double bonds. C50 ca o enoids a e he e o e in e es ing in ood applica ions and
o he pha maceu ical indus y. The ela i e p opo ion o bac e io ube in con en in cells depends on
he s ain o haloa chaea and he cul u e condi ions used, pa icula ly empe a u e, pH, and salini y.
O he ac o s such as he addi ion o selec ed o ganic compounds o he cul u e medium also in luences
he ca o enoid p oduc ion o halophilic a chaea [
12
]. The cul u e condi ions should be se be o ehand
o maximize biomass yield and ca o enoid p oduc ion, he eby imp o ing yield and educing cos s [
5
].
S udies on he bio echnological use o halophilic a chaea a e sca ce, despi e he widesp ead
in e es in C50 ca o enoids, and H. medi e anei can be a good candida e due o i s abili y o g ow h in a
wide ange o empe a u es, pH, and salini y. Gene ally, he app oach o s anda dize and op imize he
condi ions o g ow h and ca o enoid p oduc ion simul aneously, pa icula ly a la ge-scale p oduc ion,
is complica ed [
7
,
13
]. S a is ical expe imen al me hods such as cen al composi e design (CCD) and
esponse su ace me hodology (RSM) can be used in mic obial p ocesses o de e mine he condi ions
o op imal p oduc i i y [
14
]. We demons a e ha RSM is use ul o op imiza ion o condi ions o
g ow h a e and ca o enoid p oduc ion by H. medi e anei a he labo a o y scale. This app oach will be
aluable o ca o enoid p oduc ion a he indus ial scale.
2. Resul s and Discussion
2.1. E ec o Ai Volume inside he Cul u e Flasks and Speed o Agi a ion on he G ow h Ra e o
H. medi e anei
Oxygen supply is essen ial o op imal g ow h and ca o enoid p oduc ion in haloa chaea [
15
].
The in luence o ai olume inside he cul u e lasks on he g ow h a e o H. medi e anei was e alua ed
while all he o he cul u e condi ions we e kep cons an . This s udy was no aimed a de e mining he
op imal ai phase olume o p oduc ion o H. medi e anei as i depends on he cul i a ion sys em
design and pa ame e s speci ically selec ed in each p oduc ion p ocess. Howe e , an ai phase olume
Ma . D ugs 2018,16, 372 3 o 12
was used in his s udy so ha he oxygen a ailabili y allowed H. medi e anei o comple e g ow h
un il he s a iona y phase. Ai occupying 20%, 40%, o 60% o he cul u e lask olume was kep in
con ac wi h he cell cul u e in liquid medium and agi a ed a 100 pm. In he g ow h condi ions
and cul i a ion sys em used in his s udy (see Ma e ials and Me hods), he op imal g ow h a e o
H. medi e anei was obse ed in he cul u e wi h 60% ai phase in he lask, hus emphasizing he
impo ance o oxygen o his haloa chaeon (Figu e 1A). Expe imen s wi h 80% ai phase in he lask
did no suppose a signi ican imp o emen o he ha chaeal p oduc i i y (da a no shown).
The e ec o cul u e agi a ion speed on he g ow h a e o H. medi e anei was also e alua ed
using h ee di e en condi ions: no agi a ion (0 pm), 100 pm, o 150 pm, a 60% ai phase. Figu e 1B
shows ha an op imal g ow h a e o he haloa chaeon was ob ained in cul u es agi a ed a 150 pm,
bu signi ican g ow h was also obse ed a 100 pm. No g ow h was obse ed in non-gi a ed cul u es
(0 pm), hus con i ming ha good ae a ion is absolu ely essen ial o sus ain he haloa chaeal g ow h.
These indings a e consis en wi h hose epo ed o H. alexand inus, he op imal g ow h o which was
ob ained in 100 mL cul u e medium in 500 mL lasks (80% ai phase) [
16
], and Halo ub um sp. SHI,
which equi ed agi a ion a 550 pm o op imal g ow h in 50 mL o cul u e medium in 250 mL lasks
(80% ai phase) [17].
Ma . D ugs 2018, 16, x 3 o 12
an ai phase olume was used in his s udy so ha he oxygen a ailabili y allowed H. medi e anei o
comple e g ow h un il he s a iona y phase. Ai occupying 20%, 40%, o 60% o he cul u e lask
olume was kep in con ac wi h he cell cul u e in liquid medium and agi a ed a 100 pm. In he
g ow h condi ions and cul i a ion sys em used in his s udy (see Ma e ials and Me hods), he op imal
g ow h a e o H. medi e anei was obse ed in he cul u e wi h 60% ai phase in he lask, hus
emphasizing he impo ance o oxygen o his haloa chaeon (Figu e 1A). Expe imen s wi h 80% ai
phase in he lask did no suppose a signi ican imp o emen o he ha chaeal p oduc i i y (da a no
shown).
The e ec o cul u e agi a ion speed on he g ow h a e o H. medi e anei was also e alua ed
using h ee di e en condi ions: no agi a ion (0 pm), 100 pm, o 150 pm, a 60% ai phase. Figu e
1B shows ha an op imal g ow h a e o he haloa chaeon was ob ained in cul u es agi a ed a 150
pm, bu signi ican g ow h was also obse ed a 100 pm. No g ow h was obse ed in non-gi a ed
cul u es (0 pm), hus con i ming ha good ae a ion is absolu ely essen ial o sus ain he haloa chaeal
g ow h. These indings a e consis en wi h hose epo ed o H. alexand inus, he op imal g ow h o
which was ob ained in 100 mL cul u e medium in 500 mL lasks (80% ai phase) [16], and Halo ub um
sp. SHI, which equi ed agi a ion a 550 pm o op imal g ow h in 50 mL o cul u e medium in 250
mL lasks (80% ai phase) [17].
Figu e 1. E ec o ai phase (A) and shake speed (B) o cul u es on he g ow h o H. medi e anei.
Cells we e g own unde s anda d condi ions, as s a ed in Ma e ials and Me hods, and using he
indica ed ai phase and shake speed. When indica ed, he u bidi y o he cul u e was de e mined a
600 nm.
Fo he ope a ion, we used condi ions o agi a ion speed o 150 pm and 60% ai phase in he
cul u e lasks. Unde hese condi ions, H. medi e anei showed a gene a ion ime o 33.6 h and
maximum p oduc i i y o 22.16 g d y weigh /L. This yield is signi ican ly be e han ha ob ained
p e iously, no only wi h H. medi e anei [13], bu also o he ae obic haloa chaea [14]. Apa om he
a ailabili y o oxygen, g ow h o H. medi e anei also depends on salini y, pH, and empe a u e o he
cul u e medium [18]. Acco ding o Schneegu (2012) [19], a 10% inc ease in salini y educes oxygen
solubili y in he cul u e by app oxima ely 50%, which migh ha e an impac on he a ailabili y o
oxygen o he haloa chaea cells. Oxygen solubili y also dec eases as he empe a u e inc eases o pH
dec eases, hus indica ing ha salini y, empe a u e, and pH ha e a complex in luence on he g ow h
a e o H. medi e anei cul u es.
Ligh is usually impo an o he egula ion o ca o enoid syn hesis in many ypes o
mic oo ganisms. Howe e , we ound no di e ences ei he in pigmen a ion o in biomass
concen a ion o H. medi e anei when cul i a ed in absence o p esence o ligh (da a no shown). The
e ec o ligh on pigmen a ion o halophilic a chaea g ea ly depends on species and s ains. Fo
ins ance, s ains Hb . salina um ATCC 33170, Hb . salina um ATCC 43214 and H x. alexand inus TM
(JCM 10717T), showed no di e ence in pigmen a ion when cul i a ed in he absence o in he
Figu e 1.
E ec o ai phase (
A
) and shake speed (
B
) o cul u es on he g ow h o H. medi e anei.
Cells we e g own unde s anda d condi ions, as s a ed in Ma e ials and Me hods, and using he
indica ed ai phase and shake speed. When indica ed, he u bidi y o he cul u e was de e mined a
600 nm.
Fo he ope a ion, we used condi ions o agi a ion speed o 150 pm and 60% ai phase in
he cul u e lasks. Unde hese condi ions, H. medi e anei showed a gene a ion ime o 33.6 h and
maximum p oduc i i y o 22.16 g d y weigh /L. This yield is signi ican ly be e han ha ob ained
p e iously, no only wi h H. medi e anei [
13
], bu also o he ae obic haloa chaea [
14
]. Apa om he
a ailabili y o oxygen, g ow h o H. medi e anei also depends on salini y, pH, and empe a u e o he
cul u e medium [
18
]. Acco ding o Schneegu (2012) [
19
], a 10% inc ease in salini y educes oxygen
solubili y in he cul u e by app oxima ely 50%, which migh ha e an impac on he a ailabili y o
oxygen o he haloa chaea cells. Oxygen solubili y also dec eases as he empe a u e inc eases o pH
dec eases, hus indica ing ha salini y, empe a u e, and pH ha e a complex in luence on he g ow h
a e o H. medi e anei cul u es.
Ligh is usually impo an o he egula ion o ca o enoid syn hesis in many ypes o
mic oo ganisms. Howe e , we ound no di e ences ei he in pigmen a ion o in biomass concen a ion
o H. medi e anei when cul i a ed in absence o p esence o ligh (da a no shown). The e ec o
ligh on pigmen a ion o halophilic a chaea g ea ly depends on species and s ains. Fo ins ance,
s ains Hb . salina um ATCC 33170, Hb . salina um ATCC 43214 and H x. alexand inus TM (JCM 10717T),
showed no di e ence in pigmen a ion when cul i a ed in he absence o in he p esence o ligh .
Howe e , he pigmen composi ion o Hb . salina um JCM 10927 al e s acco ding o ligh condi ions,
Ma . D ugs 2018,16, 372 4 o 12
pa icula ly by inc easing he bac e io ube in con en and dec easing he con en o C40 ca o enoids [
7
].
Thus, he e ec o ligh should be s udied o each speci ic halophilic mic oo ganism.
2.2. Use o RSM o Op imize Cul u e Condi ions o G ow h and Ca o enoid P oduc ion by H. medi e anei
Cen al composi e design (CCD) was used o de ine he expe imen al g ow h condi ions,
which should ob ain he p edic i e model o op imal g ow h and ca o enoids p oduc ion by
H. medi e anei. As can be seen in Figu e 1B, he cul u e a 60% ai phase and an agi a ion a e
o 150 pm is a he la e loga i hmic phase on day 4, which seems adequa e o measu emen s in he
CCD expe imen s. Acco dingly, 20 cul i a ion expe imen s we e un on an o bi al shake using he
pa ame e s o agi a ion speed and ai olume ac ion de ined abo e (150 pm and 60%, espec i ely).
The CCD and yield pa ame e s a e summa ized in Table 1, which de e mined he esponse models in
h ee-dimensional su aces o he a iables o haloa chaea g ow h (Figu e 2A–C) and o al ca o enoid
con en (Figu e 2D–F). Acco ding o he model, op imal g ow h o H. medi e anei should be ob ained
a 36.51 ◦C, pH o 8.20, and 15.01% (w/ ) NaCl.
Table 1.
Cen al composi e design (CCD) ma ix and he esponses o g ow h and o al ca o enoid
con en a di e en empe a u e, pH, and salini y le els. S d O de : S anda d O de .
Independen Va iables Responses
Coded Le els
S d O de Tempe a u e (◦C) pH Salini y
(NaCl% w/ )
Tu bidi y
(O.D.600 nm)
To al Ca o enoids
(mg/L)
1−1 23.8 −1 5.5 −1 9.8 2.53 0.28
2 1 23.8 −1 8.5 −1 9.8 3.09 1.61
3−1 41.3 1 5.5 −1 9.8 7.88 0.80
4 1 41.3 1 8.5 −1 9.8 11.79 3.12
5−1 23.8 −1 5.5 1 23.3 0.88 1.34
6 1 23.8 −1 8.5 1 23.3 5.31 1.05
7−1 41.3 1 5.5 1 23.3 2.04 0.24
8 1 41.3 1 8.5 1 23.3 7.25 0.78
9−1.68 32.5 0 4.5 0 16.5 1.01 0.19
10 1.68 32.5 0 9.5 0 16.5 11.51 2.85
11 0 17.8 1.68 7.0 0 16.5 0.29 0.13
12 0 47.2 1.68 7.0 0 16.5 5.14 0.85
13 0 32.5 0 7.0 −1.68 5.1 5.22 0.66
14 0 32.5 0 7.0 1.68 27.9 3.82 0.37
15 * 0 32.5 0 7.0 0 16.5 10.93 3.34
16 * 0 32.5 0 7.0 0 16.5 11.25 3.13
17 * 0 32.5 0 7.0 0 16.5 10.78 2.40
18 * 0 32.5 0 7.0 0 16.5 10.62 2.71
19 * 0 32.5 0 7.0 0 16.5 11.21 2.50
20 * 0 32.5 0 7.0 0 16.5 12.34 3.07
* Cen al poin alues con ibu ing o he deg ee o eedom o pu e e o calcula ion.
Ma . D ugs 2018,16, 372 5 o 12
Ma . D ugs 2018, 16, x 5 o 12
Figu e 2. The 3-D-su ace and con ou esponse plo s gene a ed om a quad a ic model ep esen ing
he combined e ec s o empe a u e, pH, and salini y on he g ow h a e (A–C) and ca o enoids
con en (D–F) by liquid cul u es o H x. medi e anei. The in e ac ions be ween salini y and
empe a u e (A) and (D); pH and empe a u e (B) and (E), and pH and salini y (C) and (F) we e
analyzed. O he de ails o expe imen al condi ions a e s a ed in he Ma e ials and Me hods sec ion.
The ollowing equa ion could be used o p edic he O.D. a 600 nm unde di e en condi ions:
𝑂.𝐷.600 𝑛𝑚 =−85.1 + 8.74 · 𝑋+2.681·𝑋+1.729·𝑋−0.681·𝑋
 –0.03635·𝑋

− 0.04710 𝑋
 + 0.0394 𝑋·𝑋 + 0.0639 𝑋·𝑋− 0.02318 𝑋·𝑋 (1)
whe e X
1
, X
2
, X
3
deno e empe a u e, pH, and salini y, espec i ely (see Table 3).
On he o he hand, he maximum o al ca o enoid con en in cul u es o H. medi e anei cells
should be obse ed a 36.81 °C, pH o 8.96, and 12.03% o NaCl. The ca o enoid con en a any poin
du ing he cul u e and di e en condi ions could be p edic ed acco ding o he ollowing equa ion:
Figu e 2.
The 3-D-su ace and con ou esponse plo s gene a ed om a quad a ic model ep esen ing
he combined e ec s o empe a u e, pH, and salini y on he g ow h a e (
A
–
C
) and ca o enoids con en
(
D
–
F
) by liquid cul u es o H x. medi e anei. The in e ac ions be ween salini y and empe a u e (
A
)
and (
D
); pH and empe a u e (
B
) and (
E
), and pH and salini y (
C
) and (
F
) we e analyzed. O he de ails
o expe imen al condi ions a e s a ed in he Ma e ials and Me hods sec ion.
The ollowing equa ion could be used o p edic he O.D. a 600 nm unde di e en condi ions:
O.D.600 nm =−85.1 +8.74·X2+2.681·X1+1.729·X3−0.681·X2
2−0.03635·X2
1
−0.04710 X2
3+0.0394 X2·X1+0.0639 X2·X3−0.02318 X1·X3
(1)
whe e X1,X2,X3deno e empe a u e, pH, and salini y, espec i ely (see Table 3).

Ma . D ugs 2018,16, 372 6 o 12
On he o he hand, he maximum o al ca o enoid con en in cul u es o H. medi e anei cells
should be obse ed a 36.81
◦
C, pH o 8.96, and 12.03% o NaCl. The ca o enoid con en a any poin
du ing he cul u e and di e en condi ions could be p edic ed acco ding o he ollowing equa ion:
Ca o enoidsmg
L=−27.78 +2.913·X2+0.647·X1+1.027·X3
−0.1692·X2
2−0.00974·X2
1−0.01612·X2
3+0.0171 X2·X1−0.0419 X2−X3−0.00718 X1−X3(2)
Using he one- ac o -a -a- ime app oach, op imal condi ions o p oduce ca o enoids (2.06 mg/g
d y weigh o cells) by H. alexand inus we e 37
◦
C, pH o 7.2, and 25% NaCl [
16
], which a e conside ably
di e en o his s udy, hus ei e a ing he impo ance o he haloa chaea species used and he
in e ac ions be ween ac o s.
2.3. Valida ion o he Op imal Condi ions o G ow h and To al Ca o enoid P oduc ion by H. medi e anei
The accu acy o he model was e i ied by analyzing he p edic ed and obse ed expe imen al
esul s. Th ee expe imen s we e ca ied ou o de e mine he eliabili y o op imal condi ions p edic ed
by he models o Equa ions (1) and (2), using he da a ob ained o biomass and o al ca o enoid
con en , espec i ely. Figu e 3shows ha high alues o R
2
(93.1%) and adjus ed R
2
(92.7%) highligh
he ag eemen be ween p edic ed and obse ed expe imen al alues. Hence, an accep able ela ionship
be ween independen a iables ( empe a u e, pH, and salini y) and esponse a iables (g ow h and
o al ca o enoids) was p o ed. The highes biomass p oduc ion (21.95
±
1.57 g d y weigh /L) and
o al ca o enoid con en (3.74
±
0.20 mg/L) we e e y close o he alues es ima ed using RSM a he
op imal condi ions (20.18
±
1.02 g d y weigh /L, and 3.34
±
0.29 mg/L, espec i ely) indica ing ha
RSM is e ec i e in ixing cul u e condi ions whe e se e al a iables could in luence he inal esul .
I can also p edic esul s o o he po en ial cul u e condi ions o he haloa chaea, as epo ed by he
au ho s o Re e ence [14], in simila s udies on Halo ub um sp. TZB126.
Ma . D ugs 2018, 16, x 6 o 12
𝐶𝑎𝑟𝑜𝑡𝑒𝑛𝑜𝑖𝑑𝑠󰇡𝑚𝑔
𝐿󰇢= −27.78 + 2.913 · 𝑋+ 0.647 · 𝑋 + 1.027 · 𝑋
−0.1692·𝑋
− 0.00974 · 𝑋
−0.01612·𝑋

+ 0.0171 𝑋·𝑋
− 0.0419 𝑋−𝑋− 0.00718 𝑋−𝑋
(2)
Using he one- ac o -a -a- ime app oach, op imal condi ions o p oduce ca o enoids (2.06 mg/g
d y weigh o cells) by H. alexand inus we e 37 °C, pH o 7.2, and 25% NaCl [16], which a e
conside ably di e en o his s udy, hus ei e a ing he impo ance o he haloa chaea species used
and he in e ac ions be ween ac o s.
2.3. Valida ion o he Op imal Condi ions o G ow h and To al Ca o enoid P oduc ion by H. medi e anei
The accu acy o he model was e i ied by analyzing he p edic ed and obse ed expe imen al
esul s. Th ee expe imen s we e ca ied ou o de e mine he eliabili y o op imal condi ions
p edic ed by he models o Equa ions (1) and (2), using he da a ob ained o biomass and o al
ca o enoid con en , espec i ely. Figu e 3 shows ha high alues o R2 (93.1%) and adjus ed R2 (92.7%)
highligh he ag eemen be ween p edic ed and obse ed expe imen al alues. Hence, an accep able
ela ionship be ween independen a iables ( empe a u e, pH, and salini y) and esponse a iables
(g ow h and o al ca o enoids) was p o ed. The highes biomass p oduc ion (21.95 ± 1.57 g d y
weigh /L) and o al ca o enoid con en (3.74 ± 0.20 mg/L) we e e y close o he alues es ima ed
using RSM a he op imal condi ions (20.18 ± 1.02 g d y weigh /L, and 3.34 ± 0.29 mg/L, espec i ely)
indica ing ha RSM is e ec i e in ixing cul u e condi ions whe e se e al a iables could in luence
he inal esul . I can also p edic esul s o o he po en ial cul u e condi ions o he haloa chaea, as
epo ed by he au ho s o Re e ence [14], in simila s udies on Halo ub um sp. TZB126.
Figu e 3. Theo e ical alues o esponse a iables p edic ed om he espec i e models and obse ed
alues o he expe imen al design wi h a p- alue < 0.05 o he g ow h a e and o al in acellula
ca o enoids by H x. medi e anei. The g ow h and ca o enoids con en a e as desc ibed in Ma e ials
and Me hods. CI = eliable in e al and PI = p edic ed in e al.
The da a demons a es o he i s ime in H. medi e anei ha he RSM app oach migh be used
o p edic op imal condi ions o la ge scale ca o enoid p oduc ion.
2.4. Ca o enoid P o ile Ob ained om H. medi e anei
Bac e io ube in is he majo C50 ca o enoid in all he a chaeal s ains, howe e , β-ca o ene,
lycopene, as axan hin, and can haxan hin we e also ound in hese o ganisms [3,20]. The p o ile o
he ca o enoid ob ained om H. medi e anei was analyzed using HPLC, and he esul s e ealed a
ch oma og am wi h 10 peaks whe e bac e io ube in (89.13%) was he majo compound p oduced
unde he op imal condi ions used o ca o enoid p oduc ion (Figu e 4A). Peaks 1–4 had he same
molecula weigh (Table 2). Figu e 4B shows a simila 3- inge ype abso p ion spec um o hese
ca o enoid ac ions wi h ypical bac e io ube in abso p ion maxima a 468, 495, and 530 nm, he eby
indica ing ha hey we e isome s o he main ca o enoid, p obably 13-cis-bac e io ube in and 9-cis-
Figu e 3.
Theo e ical alues o esponse a iables p edic ed om he espec i e models and obse ed
alues o he expe imen al design wi h a p- alue < 0.05 o he g ow h a e and o al in acellula
ca o enoids by H x. medi e anei. The g ow h and ca o enoids con en a e as desc ibed in Ma e ials and
Me hods. CI = eliable in e al and PI = p edic ed in e al.
The da a demons a es o he i s ime in H. medi e anei ha he RSM app oach migh be used
o p edic op imal condi ions o la ge scale ca o enoid p oduc ion.
2.4. Ca o enoid P o ile Ob ained om H. medi e anei
Bac e io ube in is he majo C
50
ca o enoid in all he a chaeal s ains, howe e ,
β
-ca o ene,
lycopene, as axan hin, and can haxan hin we e also ound in hese o ganisms [
3
,
20
]. The p o ile o
he ca o enoid ob ained om H. medi e anei was analyzed using HPLC, and he esul s e ealed a
Ma . D ugs 2018,16, 372 7 o 12
ch oma og am wi h 10 peaks whe e bac e io ube in (89.13%) was he majo compound p oduced
unde he op imal condi ions used o ca o enoid p oduc ion (Figu e 4A). Peaks 1–4 had he same
molecula weigh (Table 2). Figu e 4B shows a simila 3- inge ype abso p ion spec um o hese
ca o enoid ac ions wi h ypical bac e io ube in abso p ion maxima a 468, 495, and 530 nm, he eby
indica ing ha hey we e isome s o he main ca o enoid, p obably 13-cis-bac e io ube in and
9-cis-bac e io ube in, espec i ely [
9
]. Howe e , he de ini i e s uc u es o peaks 1-4 canno be
elucida ed un il u he s udies a e done.
Ma . D ugs 2018, 16, x 7 o 12
bac e io ube in, espec i ely [9]. Howe e , he de ini i e s uc u es o peaks 1-4 canno be elucida ed
un il u he s udies a e done.
Figu e 4. HPLC analysis o he ca o enoids p esen in H x. medi e anei (A) and he abso p ion
spec um o isola ed bac e io ube in (B). Peaks 1–4 a e he isome s o bac e io ube in, peaks 5, 6, 7, 8,
9, and 10 a e C50 ca o enoid de i a i es om bac e io ube in. Fu he expe imen al condi ions a e
indica ed in he Ma e ials and Me hods.
Figu e 4A also e eals o he mino peaks co esponding o chemically modi ied bac e io ube in-
de i ed compounds such as monoanhyd obac e io ube in and bisanhyd obac e io ube in [13], he
molecula weigh s di e ed om bac e io ube in (Table 2). The o he peaks obse ed in he
ch oma og am co esponded o unknown ca o enoids. Calo e al. (1995) [21] epo ed 3-
hyd oxyechinenone as a majo ca o enoid o H. medi e anei. Howe e , he exis ence o 3-
hyd oxyechinenone in H. medi e anei has no been e e ed o in any o he pape a e 1995. O he
s udies on H. medi e anei epo ed 70% and 52.4% bac e io ube in in he ca o enoid ac ion [8,13],
espec i ely, indica ing he in luence o he cul u e condi ions on he yields o ca o enoids and
composi ion o he haloa chaea. By inc easing he amoun o magnesium sul a e in he medium, he
ela i e a io o bac e io ube in was inc eased, eaching a cons an le el a 8% (w/ ) o magnesium
[13]. In ou s udy, 2% magnesium sul a e was used in he cul u e medium o H. medi e anei. The
bac e io ube in con en ob ained om o he haloa chaea is highly a iable, as shown in Re e ence
[14], (98.1% in Halo ub um sp.); [10], (68.1% in Haloa cula japonica); and [22], (49.2% in Halobac e ium
SP–2 and 55.3% in Halo ub um SP–4). These da a show ha H. medi e anei g own unde he
condi ions s a ed in his wo k con ains high le els o bac e io ube in compa ed wi h o he
haloa chaea.
Table 2. Ten a i e iden i ica ion o ca o enoids p esen in Halo e ax medi e anei. BR: bac e io ube in;
MABR: monoanhyd obac e io ube in; BABR: bisanhyd obac e io ube in.
Peak Ca o enoid Re en ion
Time (min) λmax (nm) Molecula Ion
(m/z) M+ F agmen s P o ile
1 BR 2.325 468, 496, 530 740.7 723.7, 705.7, 687.7, 666.7, 561.5, 515,1
2 BR 2.553 468, 494, 528 740.7 723.7, 705.7, 681,6, 666.8, 655.6, 627.6
3 BR 2.740 468, 496, 528 740.7 723.7, 705.7, 682.6, 669.6, 665.6
4 BR 2.816 464, 494, 524 740.7 723.7, 705.7, 682.6, 665.6
5 MABR 3.021 470, 500, 534 737.7 725.6, 709.6, 699.7
6 BABR 3.168 460, 488, 520 705.7 681.6, 669.7, 579.7, 522.7
7 BABR 3.233 456, 485, 526 705.7 699.7, 671.7, 668.7, 647.6, 579.6
8 BABR 3.508 472, 498, 532 705.7 699.7, 687.7, 671.7, 653.8, 607.6
9 BABR 3.620 468, 490, 520 705.7 699.7, 671.7, 653.8, 550.6
Figu e 4.
HPLC analysis o he ca o enoids p esen in H x. medi e anei (
A
) and he abso p ion spec um
o isola ed bac e io ube in (
B
). Peaks 1–4 a e he isome s o bac e io ube in, peaks 5, 6, 7, 8, 9, and 10
a e C
50
ca o enoid de i a i es om bac e io ube in. Fu he expe imen al condi ions a e indica ed in
he Ma e ials and Me hods.
Figu e 4A also e eals o he mino peaks co esponding o chemically modi ied
bac e io ube in-de i ed compounds such as monoanhyd obac e io ube in and
bisanhyd obac e io ube in [
13
], he molecula weigh s di e ed om bac e io ube in (Table 2).
The o he peaks obse ed in he ch oma og am co esponded o unknown ca o enoids.
Calo e al. (1995) [21]
epo ed 3-hyd oxyechinenone as a majo ca o enoid o H. medi e anei.
Howe e , he exis ence o 3-hyd oxyechinenone in H. medi e anei has no been e e ed o in any
o he pape a e 1995. O he s udies on H. medi e anei epo ed 70% and 52.4% bac e io ube in in
he ca o enoid ac ion [
8
,
13
], espec i ely, indica ing he in luence o he cul u e condi ions on he
yields o ca o enoids and composi ion o he haloa chaea. By inc easing he amoun o magnesium
sul a e in he medium, he ela i e a io o bac e io ube in was inc eased, eaching a cons an le el a
8% (w/ ) o magnesium [
13
]. In ou s udy, 2% magnesium sul a e was used in he cul u e medium
o H. medi e anei. The bac e io ube in con en ob ained om o he haloa chaea is highly a iable,
as shown in Re e ence [
14
], (98.1% in Halo ub um sp.); [
10
], (68.1% in Haloa cula japonica); and [
22
],
(49.2% in Halobac e ium SP–2 and 55.3% in Halo ub um SP–4). These da a show ha H. medi e anei
g own unde he condi ions s a ed in his wo k con ains high le els o bac e io ube in compa ed wi h
o he haloa chaea.
Ma . D ugs 2018,16, 372 8 o 12
Table 2.
Ten a i e iden i ica ion o ca o enoids p esen in Halo e ax medi e anei. BR: bac e io ube in;
MABR: monoanhyd obac e io ube in; BABR: bisanhyd obac e io ube in.
Peak Ca o enoid Re en ion
Time (min) λmax (nm) Molecula Ion
(m/z) M+F agmen s P o ile
1 BR 2.325 468, 496, 530 740.7
723.7, 705.7, 687.7, 666.7, 561.5, 515,1
2 BR 2.553 468, 494, 528 740.7
723.7, 705.7, 681,6, 666.8, 655.6, 627.6
3 BR 2.740 468, 496, 528 740.7 723.7, 705.7, 682.6, 669.6, 665.6
4 BR 2.816 464, 494, 524 740.7 723.7, 705.7, 682.6, 665.6
5 MABR 3.021 470, 500, 534 737.7 725.6, 709.6, 699.7
6 BABR 3.168 460, 488, 520 705.7 681.6, 669.7, 579.7, 522.7
7 BABR 3.233 456, 485, 526 705.7 699.7, 671.7, 668.7, 647.6, 579.6
8 BABR 3.508 472, 498, 532 705.7 699.7, 687.7, 671.7, 653.8, 607.6
9 BABR 3.620 468, 490, 520 705.7 699.7, 671.7, 653.8, 550.6
2.5. Bac e io ube in P oduc ion by H. medi e anei
The maximum ca o enoid yield in ou expe imen s was 3.74 mg/L (equi alen o 23.51 mg/g
d y weigh ), which is di e en om ha epo ed in o he haloa chaea. The yield o ca o enoids in
haloa chaea mainly depended on he s ain and on he cul u e condi ions used. H. alexand inus
accumula es 2.6 mg/g d y weigh [
16
]; Halobac e ium salina um, 45
µ
g/g d y weigh and
Halococcus mo huae, 89
µ
g/g d y weigh [
9
], Halobac e ium halobium, 7.63 mg/L [
23
], Halo ub um sp.,
10.78 mg/L [
14
], H. medi e anei, 125 mg/L [
8
], Haloa cula japonica, 335
µ
g/g d y weigh [
10
];
Halo ub um sp. SH1, 25 mg/L [
17
], and Halo e igena u kmenica, 32
µ
g/g d y weigh [
24
]. Howe e ,
in mos cases, he co esponding in o ma ion conce ning biomass p oduc ion and/o cell iabili y,
unde he condi ions used o ca o enoid p oduc ion a e absen , which makes i di icul o selec one
s ain o haloa chaea o la ge-scale p oduc ion o ca o enoids.
The chosen s a egy signi ican ly a ec s he inal cos s, as he ollowing op ions indica e:
(i) one-s ep p oduc ion unde op imal g ow h condi ions, in which he ca o enoid p oduc ion is
di ec ly linked o he biomass p oduc ion o he cul u es, o (ii) a wo-s ep sys em, whe eby he i s
s ep o biomass p oduc ion unde op imal g ow h condi ions is ollowed by he second phase o
cul i a ion unde s ess o p omo e biosyn hesis and accumula ion o ca o enoids. In ou s udy,
he sal con en in he cul u e medium seemed o es ablish he bes condi ions o ca o enoid
p oduc ion. Acco ding o li e a u e, haloa chaea equi e high sal concen a ion o op imal g ow h,
while maximum ca o enoid p oduc ion is achie ed when cells a e unde s ess p oduced by low
ex e nal salini y.
Chen e al. (2015)
[
8
] showed ha H. medi e anei g owing a 40 S/m conduc i i y
(a measu emen o sal concen a ion) in saline medium accumula ed 125 mg/L o o al ca o enoids;
howe e , i he conduc i i y o he medium was dec eased o 25 S/m, he pigmen s could be inc eased
o a maximum alue o 555.6 mg/L. F om Equa ions (2) and (3), we es ima ed ha H. medi e anei can
p oduce 3.34 mg/L o ca o enoids, while he heo e ical alue o he g ow h o haloa chaea unde
such condi ions is 18.51 g d y weigh /L, which co esponds o a loss o abou 7.5% o he biomass
p oduc i i y. Thus, he op ion o a one-s ep p ocess is adequa e o he high-scale bac e io ube in
p oduc ion by H. medi e anei. Fixing he op imal condi ions o ca o enoid p oduc ion inc eases he
bio echnological alue o his halophilic mic oo ganism.
Calega i-San os e al. (2016) [
7
] e iewed he e ec o di e en s ess condi ions on ca o enoids
p oduc ion in halophilic a chaea. In addi ion o he a iables conside ed in his wo k, he C-sou ce
and he p esence o absence o me als is also ele an . Howe e , he e ec o N-s a a ion and o he
nu i ional s ess ac o s emain o be examined.
3. Ma e ials and Me hods
3.1. Mic oo ganism
The highly halophilic a chaeon Halo e ax medi e anei, s ain R4 (ATCC 33500T), used in his
s udy, was p o ided by D . Rosa Ma ía Ma ínez om he Depa men o Ag ochemis y and
Ma . D ugs 2018,16, 372 9 o 12
Biochemis y, Uni e si y o Alican e, Spain. This a chaeon was i s isola ed and epo ed by he
au ho s o Re e ence [25], om saline wa e a San a Pola in Alican e (Spain).
3.2. G ow h Condi ions and Biomass Quan i ica ion
The haloa chaea we e g own in a basal cul u e medium as o mula ed in Re e ence [
13
], con aining
(pe li e ): Glucose, 10 g; NaCl, 156 g; MgCl
2·
6H
2
O, 13 g; MgSO
4·
7H
2
O, 20 g; CaCl
2·
6H
2
O, 1 g; KCl,
4 g; NaHCO
3
, 0.2 g; NaB , 0.5 g; yeas ex ac , 5 g; and he pH was adjus ed o 7.0 by addi ion o
dilu ed KOH o HCl. The mo he cul u e was p epa ed in 100 mL o liquid medium con ained in a
250 mL lask and incuba ed a 37
◦
C and 150 pm on an o bi al shake un il he exponen ial phase o
g ow h was achie ed (s anda d condi ions). This cul u e was used as inoculum a 10% ( / ), in all he
expe imen s. The g ow h was de e mined by measu ing he u bidi y o he cul u e a 600 nm using a
UV-Vis spec opho ome e (The mo Spec onic, Genesis, Wal ham, MA, USA). The d y weigh was
de e mined using 1 mL sample o he co esponding cul u e, which was il e ed h ough a p e-weighed
memb ane (
φ
= 0.2
µ
m) and he e ained cells we e washed on he il e using 5 mL o 1% NaCl (w/ )
solu ion. The memb ane was hen d ied a 80
◦
C un il a cons an weigh was eached. A con ol wi h
1 mL o uninocula ed cul u e medium was un in pa allel. The weigh was la e deduc ed om he
sample. Cul u e wi h OD o 1.0 a 600 nm had a d y weigh o 1.60 g/L.
3.3. Ex ac ion, Quan i ica ion, and Analysis o Pigmen s
Fo ex ac ion o ca o enoids, he cul u e samples (10 mL) we e cen i uged a 3500
×
g o 45 min,
he ha es ed cells we e lysed by eeze/ hawing, and inally, he biological ma e ial was esuspended
in 1 mL o pu e ace one and kep o e nigh a 4
◦
C. The suspension was cen i uged a 3500
×
g o
5 min. The o al ca o enoid con en o he supe na an was de e mined by measu ing he abso bance
a 494 nm and calcula ed using an ex inc ion coe icien ,
ε
(1%), o 2540, acco ding o he ollowing
exp ession: mg/L = (OD494/2540) ×104.
The HPLC analysis o ca o enoids in ace one was pe o med using a Po oshell 120-C18
column (Agilen , San a Cla a, CA, USA) (3
×
50 mm, 2.7
µ
m) on an Agilen 1200 se ies sys em
(
San a Cla a, CA, USA
) equipped wi h a diode a ay de ec o scanning om 400 o 690 nm.
To de e mine he mass spec a o he di e en compounds, a 6410 T iple Quad LC/MS sys em (Agilen ,
San a Cla a, CA, USA) was used equipped wi h an elec osp ay ioniza ion sou ce (ESI) ope a ing
in posi i e scan mode (m/z ange o 300–900), wi h
±
0.1 u.m.a. p ecision, and con olled by Mass
Hun e Wo ks a ion So wa e (Agilen , B.05.00, San a Cla a, CA, USA). The ollowing speci ic wo king
condi ions we e used: capilla y ol age 4000 V, gas low a e 10 L m
−1
, gas empe a u e 300
◦
C,
and nebulize p essu e 35 psi [17].
3.4. Response Su ace Me hodology Expe imen al Design
The one- ac o -a -a- ime app oach used o analyze a p oblem based on h ee o mo e pa ame e s
o e looks he in e ac ions be ween di e en ac o s [
26
]. To add ess hese issues, RSM was used o
iden i y he op imal alue o be applied in o de o de e mine he main e ec as well as any signi ican
in e ac ions be ween ac o s ha may exe impo an e ec s on esponse a iables [
14
,
27
,
28
]. A cen al
and axial poin s design (CCD) app oach was used o op imize he cul u e condi ions o bo h cell
g ow h (O.D. a 600 nm) and o al ca o enoid con en (mg/L) by H. medi e anei. In his s udy,
empe a u e, salini y, and pH we e conside ed o he CCD analysis. They we e in es iga ed a
i e di e en le els wi hin he ollowing anges: empe a u e (15–50
◦
C), pH (4
−
10), and NaCl
concen a ion (5–28%, w/ ) in o de o deduce he op imum alues o g ow h and ca o enoid con en .
The code and ac ual alues o he a iables a e p esen ed in Table 3.