Influence of growth phase on the phospholipidic fatty acid composition of two marine bacterial strains in pure and mixed cultures.

Abstract

This in vitro study was conducted in order to determine the effects of hydrocarbons and growth phase on the phospholipid ester-linked fatty acid composition of two marine sedimentary hydrocarbon-degrading bacteria. These two strains, namely Corynebacterium sp. and Sphingomonas sp. 2MPII, were cultivated on either a simple soluble substrate (ammonium acetate) or a hydrocarbon (respectively n-eicosane and phenanthrene). The incubations were stopped at different times corresponding to point of lag (2 days), exponential (7 days) and stationary phases (21 and 56 days). The effects of growth phase and hydrophobic substrates were successfully demonstrated by a simple index, given as the sum of saturated fatty acids divided by the sum of unsaturated fatty acids ( summation operatorSFA/ summation operatorMUFA), ranging from 1.4 to 3, 0.3 to 0.6, and 0.5 to 1.0 for Corynebacterium sp., Sphingomonas sp. 2MPII, and mixed cultures, respectively. This result was validated by a principal component analysis. In pure cultures, the phospholipid fatty acid (PLFA) composition was strongly influenced by both the carbon source and the growth phase. Nevertheless, the two strains showed different "behaviors". For 2MPII, the main PLFA composition changes were observed at 2 days while they were progressive as a function of time for Corynebacterium sp. These differences could explain the evolution of PLFAs of mixed cultures.

Full text

Page 1

Research in Microbiology 157 (2006) 479–486
www.elsevier.com/locate/resmic
Influence of growth phase on the phospholipidic fatty acid composition
of two marine bacterial strains in pure and mixed cultures
Agung Dhamar Syaktia, Nicolas Mazzellab, Franck Torrec, Monique Acquavivad,
Michele Gilewiczd, Michel Guilianob, Jean-Claude Bertrandd, Pierre Doumenqb,∗
aFishery and Marine Science Program, General SOEDIRMAN University, JI HR. Boenyamin 708 Case 15, 53122 Purwokerto, Indonesia
bLaboratoire de Chimie Analytique de l’Environnement, UMR 6171, IFR PMSE 112, Université d’Aix Marseille III, Europôle de l’Arbois,
BP 80, 13545 Aix-en-Provence Cedex 4, France
cInstitut Méditerranéen d’Ecologie et de Paléoécologie, UMR 6116, Université d’Aix Marseille III, Av. Escadrille Normandie Niemen,
F-13397 Marseille Cedex 02, France
dLaboratoire de Microbiologie, Biochimie et Ecologie Marine, UMR 6117, Case 901, Université de la Méditerranée, 163 avenue de Luminy,
13228 Marseille Cedex 09, France
Received 22 June 2005; accepted 2 November 2005
Available online 1 December 2005
Abstract
This in vitro study was conducted in order to determine the effects of hydrocarbons and growth phase on the phospholipid ester-linked
fatty acid composition of two marine sedimentary hydrocarbon-degrading bacteria. These two strains, namely Corynebacterium sp. and Sphin-
gomonas sp. 2MPII, were cultivated on either a simple soluble substrate (ammonium acetate) or a hydrocarbon (respectively n-eicosane and
phenanthrene). The incubations were stopped at different times corresponding to point of lag (2 days), exponential (7 days) and stationary phases
(21 and 56 days). The effects of growth phase and hydrophobic substrates were successfully demonstrated by a simple index, given as the sum
of saturated fatty acids divided by the sum of unsaturated fatty acids (?SFA/?MUFA), ranging from 1.4 to 3, 0.3 to 0.6, and 0.5 to 1.0 for
pure cultures, the phospholipid fatty acid (PLFA) composition was strongly influenced by both the carbon source and the growth phase. Never-
theless, the two strains showed different “behaviors”. For 2MPII, the main PLFA composition changes were observed at 2 days while they were
progressive as a function of time for Corynebacterium sp. These differences could explain the evolution of PLFAs of mixed cultures.
© 2005 Elsevier SAS. All rights reserved.
Corynebacterium sp., Sphingomonas sp. 2MPII, and mixed cultures, respectively. This result was validated by a principal component analysis. In
Keywords: Fatty acids; Hydrocarbons; Hydrocarbonoclastic bacteria; GC/MS
1. Introduction
The phospholipid fatty acid (PLFA) composition of bac-
terial membranes is of great value in the understanding of
bacterial phylogenic and taxonomic classifications and in es-
timation of microbial biomass. PLFA analyses also provide
information on microorganism nutritional states and stress con-
ditions [15,24,31]. These bacterial membrane constituents have
been reported to be markedly affected by growth conditions,
i.e. pH [23], growth temperature [6,7], carbon source [7,28],
*Corresponding author.
E-mail address: pierre.doumenq@univ.u-3mrs.fr (P. Doumenq).
salinity [29] and growth phase (age of cultures) [1,19]. In most
of the previous works, PLFAs of hydrocarbon-grown bacteria
were obtained from the exponential growth phase. However,
there is a lack of information related to the modification of fatty
acid composition during the different growth phases. In spite
of environmental changes, microorganisms modify their fatty
acid and lipid composition to maintain optimal fluidity, stabil-
ity and permeability of cytoplasmic membranes (homeoviscous
adaptation) [30]. Therefore, PLFA analysis can be considered
as a tool to assess the effects of these environmental alterations
on bacteria. The use of bacterial PLFA profiles is based on the
fact that phospholipids are found in the membrane of living
cells. Additionally, bacteria contain a relatively constant pro-
0923-2508/$ – see front matter © 2005 Elsevier SAS. All rights reserved.
doi:10.1016/j.resmic.2005.11.001

Page 2

480
A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
portion of phospholipids [31]. The first aim of this work was to
carry out the effects of the growth phase and hydrocarbons on
bacterial PLFA composition.
Two marine hydrocarbonoclastic strains, Corynebacterium
sp. (COR) and Sphingomonas sp. 2MPII (2MPII), previously
reported to possess degradation capabilities toward saturated
and polycyclic aromatic hydrocarbons, respectively [10,27],
were chosen for this in vitro study. These two strains were cul-
tivated on a simple soluble substrate (acetate) and on hydrocar-
bons (n-eicosane and phenanthrene). Incubations were stopped
at different times corresponding to the point of lag (2 days), ex-
ponential (7 days) and stationary (21 and 56 day) phases. The
second objective was to determine an index to obtain good dis-
crimination of the PLFA profiles. In addition, through the use
of this index, we sought to improve the description of the adap-
tive responses of microorganisms during growth in the presence
of hydrocarbons in pure and mixed cultures.
2. Materials and methods
2.1. Bacterial strains and culture conditions
The marine hydrocarbonoclastic bacteria used in this study
were Corynebacterium sp. and Sphingomonas sp. 2MPII
(DSMZ 11572) [10], Gram-positive and Gram-negative, re-
spectively. These strains are able to grow under aerobic con-
ditions with hydrocarbons as sole carbon and energy source.
Corynebacterium sp. was able to grow on aliphatic hydrocar-
bons, but failed to grow on polyaromatic hydrocarbons [27].
Inversely, Sphingomonas sp. 2MPII could grow on aromatic
hydrocarbons but not on n-alkanes [10]. These strains were
originally isolated from a polluted marine sediment collected
in the vicinity of a petroleum refinery of the Gulf of Fos, on the
French Mediterranean coast [10].
The experiments were carried out in the dark at 20◦C in
T-shaped flasks containing 50 ml of synthetic seawater, in
areciprocalshaker(1.5Hz).Sterilesyntheticseawaterwasused
as culture medium, and its pH was adjusted to 7.8 with droplets
of 2 M HCl [27]. After autoclaving, FeSO4(0.1 mM), K2HPO4
(0.33 mM), and a V7 vitamin solution (0.5 mll−1) were added
to the culture. Each strain was precultured with 3 gl−1ammo-
nium acetate as carbon and energy source to ensure exponential
growth. Then, cultures were inoculated with these log-phase
bacterial precultures and with the same protein content (for
each strain and each condition). The first set of experiments
used 1 gl−1of n-eicosane (nC20) with Corynebacterium sp. as
inoculum (COR-C20). The second set, inoculated with Sphin-
gomonas sp. 2MPII, was supplemented with 0.4 gl−1of
phenanthrene (2MPII-PHE). In the third set, mixed cultures of
both strains were used to test the biodegradation of a mixture
of both n-C20and PHE, present in the same concentrations as
in pure cultures. A last set of experiments was conducted in or-
der to compare exponential growth-phase fatty acids of the two
tested strains cultured on ammonium acetate (3 gl−1) with fatty
acids obtained from the corresponding hydrocarbon cultures.
This experiment set (ammonium acetate cultures) was designed
to serve as a reference. Microbial growth was determined by
measuring turbidity at 450 nm for strain COR and 610 nm for
strain 2MPII with a Shimadzu spectrophotometer (UV 240 Ky-
oto, Japan) [27]. Experiments were executed in triplicate for all
incubation times (2, 7, 21, and 56 days).
2.2. Chemicals
Phenanthrene (97%) was obtained from Fluka AG, Buch,
Switzerland. n-Eicosane (99%) was from Sigma Aldrich-
Chimie, Steinheim, Germany. Acetone, dichloromethane, hep-
tane, methanol (chromasolve grade), diethyl ether (puriss p.a.),
and boron trifluoride in methanolic solution (10%, w/v) were
purchased from Fluka (Germany). Thin-layer chromatogra-
phy plates Si60F254, silica gel (0.063–0.200 mm), and alu-
minum (0.063–0.200 mm) were purchased from Merck (Ger-
many). GF/F filters (47 mm ∅) and silica gel plus Sep-pak
cartridges were purchased, respectively, from Whatman (Eng-
land) and Waters (Ireland). Dimethyl disulfide (99%), iodine
(99.99%), and pyrrolidine (99%) were purchased from Aldrich
(Germany). Bacto-yeast extract and bacto-peptone (Difco)
were provided by BD Biosciences (San Jose, CA, USA).
The synthetic sea water components, namely, NaCl, Tris (hy-
droxymethyl) aminomethane, KCl, NH4Cl, MgSO4·7H20 and
MgCl2were purchased from Fluka (Germany).
2.3. Extraction of lipids
Bacterial cells were extracted as wet pellets using a modi-
fied Bligh and Dyer method [3]. Cell pellets (0.5–1.5 g) were
extracted at 4◦C for 24 h by agitation with a magnetic stirrer in
a monophasic mixture of chloroform/methanol/water (10/20/8,
v/v/v). After centrifugation at 6500 g for 5 min, the extract was
filtered and divided into two phases by adding 16 ml of chloro-
form and 20 ml of water. After settling (24 h), the lower phase
was collected and the aqueous phase was re-extracted with
26 ml chloroform. The two organic phases were then grouped
and concentrated using rotary evaporation followed by a blow-
down under a gentle stream of nitrogen.
2.4. Fractionation/separation
Lipid extracts from sedimentary samples were fractionated
on open tubular columns (1.5 cm i.d., 30 cm length) packed
with6gofsilicagel-60deactivatedwith5%distilledwater.The
sample (50–100 mg) was loaded onto the top of the column and
elutedsuccessivelywithchloroform(100ml),acetone(100ml),
and methanol (300 ml). The last fraction, corresponding to pu-
rified phospholipids, was concentrated using rotary evaporation
followed by a blow-down under a gentle stream of nitrogen and
then frozen until further analysis.
2.5. Fatty acid analysis
Ester-linked PLFAs were transmethylated (BF3/MeOH) to
produce fatty acid methyl esters (FAMEs) [2,20]. The double-
bond position, the geometry of monounsaturated fatty acids
and elucidation of methyl branching positions were determined

Page 3

A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
481
by forming dimethyl disulfide and N-acylpyrolidide adducts
[2,7,21]. The N-acylpyrrolidine derivatives were prepared by
treatment of FAMEs and purified by thin-layer chromatogra-
phy (heptane/diethyl ether, 80/20, v/v) before gas chromatogra-
phy/mass spectrometry (GC/MS) analysis.
GC/MS analyses of FAMEs and derivatives were performed
with a Hewlett–Packard 5890 series II gas chromatograph (HP,
Geneva, Switzerland) equipped with a AT-5 MS (Alltech),
60 m × 0.25 mm × 0.25 µm capillary column and coupled to
a Hewlett–Packard 5898A MS engine mass spectrometer. The
transfer line was held at 298◦C and the source at 240◦C. Elec-
tron impact mass spectra were acquired at 70 eV. For the whole
compounds, total ion currents (Full Scan) were acquired from
40 to 600 Da. A volume of 1 µl was injected in splitless mode
with a constant flow rate of 1 mlmin−1of helium as carrier
gas. For FAME and DMDS derivative analyses, the oven pro-
gram was 30◦C for 1 min, and then 50◦Cmin−1up to 70◦C,
10◦Cmin−1up to 120◦C, 2◦Cmin−1up to 290◦C, and finally
held for 10 min. For pyrrolidine derivative analyses, the column
was held at 30◦C for 1 min, ramped to 100◦C at 50◦Cmin−1,
ramped to 200◦C at 20◦Cmin−1, then to 290◦C at 2◦Cmin−1
and finally held at 295◦C for 20 min.
2.6. FAME nomenclature
Fatty acid terminology used “X:YωZ” where “X” indicates
the total number of carbon atoms, “Y” the number of unsatura-
tions, and “ω” precedes “Z”, the number of carbon atoms be-
tweentheclosestunsaturationandthealiphaticendofthemole-
cule (e.g. oleic acid is 18:1ω9). The cis- or trans-configuration
is given after the position of the double bond. AMe refers to
a methyl group on the A carbon from the carboxylic end of the
fatty acid.
2.7. Statistical analyses
Principal component analysis (PCA) was used to compare
PLFA profiles between NC and C microcosm sediments and
was performed with Ade4™ for Windows™.
3. Results
3.1. Cultures on ammonium acetate
ThemainPLFAsofCorynebacteriumsp.andSphingomonas
sp. 2MPII grown on ammonium acetate were even-numbered
saturated acids (SFAs) and monounsaturated acids (MUFAs),
in the range of C12–C20(Table 1). High amounts of both SFAs
and MUFAs were detected with respective amounts of 49.4 and
50.6% for Corynebacterium sp. and 28.9 and 69.9% for Sphin-
gomonas sp. 2MPII. Branched fatty acids were also present in
significant quantities (respectively 8.3 and 1.2%).
As shown in Table 1, we observed for Corynebacterium sp.
a major contribution of 16:0 (33.4%), 18:1ω9 (48%), 10
Me18:0 (8.3%), and coeluted 18:0+9Me18:1ω8 (5.7%). Some
acidswerealsodetectedasminorcompoundssuchas14:0(2%)
and 16:1ω9c (2.6%). For Sphingomonas sp. 2MPII, the main
fatty acids were 16:0 (21%), 18:1ω9 (59.7%), the coeluted
group [16:1ω7c + 16:1ω9t] (6.5%) and 18:0 (6.6%). Some
other acids were detected, with abundances of less than 4%,
namely, 12:0, 14:0, 15:0, 16:1ω9c, and br19:1.
3.2. Fatty acid composition of Corynebacterium sp.
grown on n-eicosane
The PLFA composition of Corynebacterium sp. membrane
was determined after growth on n-eicosane as sole carbon
and energy source (COR-C20). Even-numbered saturated fatty
acids (SFAs) and even-MUFAs were the main fatty acids of
Corynebacterium sp. SFA amounts accounted for from 58.6 to
74.9%, while MUFAs were from 25.1 to 41.4%.
As shown in Table 1 and Fig. 1, and as with acetate-grown
cells, we observed a major contribution of 16:0 and 18:1ω9
ranging from 35.2 to 40.5% and from 24.3 to 40.5%, respec-
tively. For branched fatty acids, we found substantial amounts
of 10Me18:0, from 8.3 to 13.9%. Some fatty acids were de-
tected as minor compounds, mainly 15:0, 17:0, and 20:0. As for
acetate-grown cells, we found that 9Me18:1ω8 co-eluted with
18:0, from 5.4 to 17.3%. Moreover, previous works have re-
ported that misidentification between 9Me18:1ω8 and 18:0
might occur [22]. Fortunately, this problem is solved by us-
ing judicious derivatization techniques such as pyrrolidide or
DMDS adducts.
3.3. Fatty acid composition of Sphingomonas sp. 2MPII
grown on phenanthrene
The PLFAs of Sphingomonas sp. 2MPII grown on phenan-
threne are shown in Table 1 and Fig. 2. We found large amounts
of even-MUFAs ranging from 61.2 to 76.9%, mainly composed
of ω9 16-C and 19-C MUFAs. A large predominance of even-
numbered SFAs was observed (from 22.9 to 37.7%). We also
found small amounts of odd-numbered straight chain SFAs,
varying from 0.5 to 1.3%. Major PLFAs of Sphingomonas sp.
2MPII grown on phenanthrene (2MPII-PHE) were 18:1ω9c
(46.1–49.6%); 16:0 (16.1–19.2%); 16:1ω9c (10.3–24.6%), and
18:0 (5.5–19.7%) (see Table 1 and Fig. 2).
3.4. Mixed cultures
The PLFAs of the mixed cultures of Corynebacterium sp.
and Sphingomonas sp. 2MPII grown on both nC20and phenan-
threne led to large amounts of even-MUFAs (49.6–68.4%).
SFAs were mainly represented by even-numbered acids rang-
ing from 28 to 40.6%. Branched SFAs ranged from 3.4
to 7.8%. Major PLFAs were as follows (decreasing abun-
dances): 18:1ω9c (44.6–55.5%), 16:0 (21.3–22.4%), [18:0 +
9Me18:1ω8] (3.5–16.5%), 16.1ω9c (3.3–9.5%), and 10Me18:0
(3.4–7.8%). Except for a 2-day incubation time, 20:0 was de-
tected (0.2–0.6%).
3.5. PCA results
The importance of each culture condition and/or physiolog-
ical state relative to changes in fatty acids could be assessed by

Page 4

482
A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
Table 1
Effect of growth phase on fatty acid composition of two marine bacteria in pure and mixed cultures
Peak
number
PLFA PLFA percentage areas
Ace
COR
COR-C20culture
T2
Ace
2MPII
2MPII-PHE culturesMixed cultures
T7T21 T56 T2T7T21T56 T2T7T21 T56
1
2
3
4
5,6
7
8
9
10
11,12
13
14
15
12:0
14:0
15:0
16:1ω9c
16:1ω7c + 16:1ω9t
16:0
17:0
18:1ω9c
18:1ω9t
18:0+9Me18:1ω8*
10Me18:0
br19:1
20:0
–
2
nd
2.6
–
33.4
–
48
–
6.1
nd
0.9
–
38.4
–
40.5
–
5.4
8.4
–
0.3
–
4.1
0.1
0.4
–
35.2
0.1
34.6
–
13.0
11.6
–
0.9
–
4.2
0.1
0.8
–
36.6
0.1
30.0
–
13.8
13.9
–
0.5
–
2.2
0.3
0.8
–
41.7
0.1
24.3
–
17.3
12.7
–
0.7
0.3
0.9
0.1
3.7
6.5
21
–
59.7
–
6.6
–
1.2
–
0.1
0.8
0.3
24.6
1.6
16.1
0.1
49.6
1.2
5.5
–
0.1
–
–
0.7
0.2
18.9
1.6
18.0
0.7
47.4
1.5
10.2
–
0.8
–
–
0.5
0.8
14.3
1.2
19.2
0.6
46.1
1.1
15.1
–
1.3
–
–
0.4
0.4
10.3
0.7
16.4
0.8
48.9
1.3
19.7
–
1.1
–
–
2.1
0.1
9.5
2.5
22.4
0.2
55.5
1.0
3.5
3.4
nd
nd
–
2.1
0.1
8.3
1.6
22.3
0.2
47.7
0.9
9.2
6.9
0.3
0.4
–
1.9
0.3
6.0
0.8
21.3
0.4
47.7
0.9
11.9
7.4
1.0
0.2
–
1.7
0.2
3.3
0.6
21.8
0.3
44.6
1.1
16.5
7.8
1.5
0.6
5.7
8.3
–
–
SFAs
Even
Odd
Branched
MUFAs
Even
Branched
49.4
41.1
–
8.3
50.6
50.6
–
58.6
50.2
–
8.4
41.4
41.4
–
65.0
53.1
0.3
11.6
35.1
35.1
–
69.2
55.1
0.2
13.9
30.9
30.9
–
74.9
61.9
0.3
12.7
25.1
25.1
–
28.9
28.8
0.1
0.0
69.9
69.9
1.2
22.9
22.4
0.5
–
76.9
76.9
0.1
29.8
28.9
0.9
–
69.4
69.4
0.8
36.1
34.8
1.3
–
62.6
62.6
1.3
37.7
36.5
1.2
–
61.2
61.2
1.1
31.6
28.0
0.2
3.4
68.4
68.4
0.0
41.2
34.0
0.3
6.9
58.5
58.5
0.3
43.4
35.3
0.7
7.4
55.5
55.5
1.0
48.9
40.6
0.4
7.8
49.6
49.6
1.5
MUFAs cis
MUFAs trans
50.6
–
41.4
–
35.0
–
30.9
–
25.1
–
63.474.2
1.2
66.3
1.5
60.3
1.1
59.2
1.3
64.9
1.0
56.0
0.9
53.7
0.9
47.9
1.1
Index
SFA/MUFA1.01.41.52.230.40.30.40.60.60.50.70.81.0
Fatty acid phospholipid identification is based on GC retention time data, GC/MS confirmation and derivatization by DMDS adducts of monounsaturated fatty acid
as well as pyrrolidide adduct of branching fatty acid.
*9Me18:1ω8 and 18:0 correspond to unresolved peak containing these two fatty acids in the case of Corynebacterium sp. grown in acetate and eicosane cultures
(pure and mixed). br = branched. nd = not detected. Ace COR and Ace 2MPII were the cultures of Corynebacterium sp. and Sphingomonas sp. 2MPII in their
exponential growth phase, respectively.
Fig. 1. Example of GC/MS total ion current of PLFAs (as methyl esters) of Corynebacterium sp. grown on n-eicosane in pure cultures using an HP-5 capillary
column. * Indicates that identified peak was in the limits of quantification. See Table 1 for peak identification.

Page 5

A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
483
Fig. 2. Examples of GC/MS total ion current of PLFAs (as methyl esters) of Sphingomonas sp. 2MPII grown on phenanthrene in pure culture. * Indicates that
identified peak was in the limits of quantification. See Table 1 for peak identification.
PCA. In the same way, the importance of each fatty acid in the
differences between culture conditions could also be assessed.
Figs. 4 and 5 display the principal components: first, the dif-
ferent cultures (PC1; COR-C20-, mixed-, 2MPII-PHE cultures,
from positive to negative value of correlation by this principal
component respectively); second, the time series (PC2; dura-
tion of the cultures). 16:0 and 10Me18:0 correlate positively
with the first axis (PC1, Fig. 4) while 16:1ω9c and 18:1ω9c are
negatively correlated with PC1.
The relationship between PC1 (which explained 74% of the
total inherent variance) and PC2 (which explained a further
13%) is also shown in Fig. 5. Several distinct clusters were
obtained, representing the contributions of the original vari-
ables. The scatter plot confirms the good replication, since
triplicates are gathered by circles. The numbers (1–3), (4–6),
(7–9), (10–12) indicate, respectively, the different incubation
times of COR-C20cultures (T2, T7, T21, and T56). In the same
manner, (13–15), (16–18), (19–21), and (22–24) indicate the
different incubation times, respectively (T2, T7, T21, and T56),
of 2MPII-PHE cultures. For mixed cultures, T2, T7, T21, and
T56 are given, respectively, by the following clusters (25–27),
(28–30), (31–33), and (34–36). AceCOR and Ace2MPII repre-
sent the reference cultures of Corynebacterium sp. and Sphin-
gomonas sp. 2MPII on ammonium acetate, respectively. The
last value of the two cultures is the mean of at least six experi-
ments. For each culture condition, the clusters (incubation time
triplicate) were ordered on the basis of the increasing times.
4. Discussion
4.1. Adaptive response to hydrocarbons
The respective responses of the two strains (i.e. Corynebac-
terium sp. and Sphingomonas sp. 2MPII) to either n-eicosane
or phenanthrene exposure were investigated. First, we observed
higher generation times on hydrocarbons than on ammonium
acetate for both strains. Indeed, the Corynebacterium sp. gen-
eration time on ammonium acetate was about 3.5 h, while it
was about 16 h on n-eicosane. The same trend was observed for
2MPII (4.5 h for acetate versus 24 h for phenanthrene). This ef-
fect couldbe dueto the nutrientassimilationrate,the accumula-
tion of toxic metabolic products and changes in ion equilibrium
[5]. The fatty acid composition of Corynebacterium sp., when
grownonahydrocarbonsubstrate,comparedtoacetate,showed
a decrease in 16-C MUFAs (from 2.6 to ≈0.7%), while the SFA
increase was about 17%. For 2MPII, we observed a slight in-
crease in SFAs, from 28.9 to about 32%, while MUFAs re-
mained constant. In the present work, acetate cultures of both
strains were stopped after 3 days because all growth phases
were reached. For neither strain did we observe any signif-
icant changes in fatty acid composition during course time.
This could be explained by the fact that ammonium acetate,
as a single carbon and energy source, is easily metabolized and
assimilated.
4.2. Effects of growth phase on the fatty acid
composition of COR-C20cultures
The comparison of PLFA profiles of COR-C20 over time
showed a substantial increase in SFAs, mainly due to the en-
hanced contribution of both 18:0 and 10Me18:0. The correlated
decrease in MUFAs, mainly due to 18:1ω9c, suggests a de-
crease in membrane fluidity. This change may be due to the
induction and expression of alk genes which are responsible
for a change in bilayer fluidity [4]. Concerning 10Me18:0, it
is interesting to note that Tsitsko et al., in 1999 [28], sug-
gested that this fatty acid may participate in the adaptation of
Rhodococcus opacus to lipophilic aromatic compounds.

Page 6

484
A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
It is interesting to note the absence of 20:0 in ammonium
acetate cultures, since it was observed in n-eicosane cultures,
ranging from 0.3 to 0.9%. This trend is in good agreement
with previous works which showed the direct incorporation in
cellular lipids of monoterminal oxidation products of hydrocar-
bons [17].
Concerning our study, the relative abundance (%) ratio of
SFAs/MUFAs led to good discrimination between the differ-
ent physiological states. In fact, this ratio gave distinct values
for each physiological state. For example, the value for sta-
tionary phase (2.2) is about twofold higher than in early log
phase.Drici-Cachonetal.[8]andValderramaetal.[29]showed
asimilarresultwhencomparingLeuconostocoenoscellsgrown
from early log to stationary phase. Our results are also in agree-
ment with previous studies on Pseudomonas nautica [7].
4.3. Effects of growth phase on the fatty acid
composition of 2MPII-PHE cultures
Concerning 2MPII-PHE cultures, we observed PLFA
changes close to those observed above for COR-C20. We found
an increase in SFAs, counteracted by a general decrease in
MUFAs. For this strain, 16-C MUFAs seemed responsible for
the regulation, probably as a response to the toxicity or hy-
drophobicity of the substrate [6], while the abundance of 18-C
MUFAs was barely constant. Hence, such compounds might
be useful as potential biomarkers for the detection or quan-
tification of Sphingomonas sp. in bacterial communities. This
finding is in agreement with previous works [9,14]. During
growth, the SFA/MUFA index increased progressively and was
twofold higher at the end of the experiments. A similar re-
sult on a MUFA decrease was observed for a major fatty acid,
9-cis-hexadecenoic acid (16:1), of a psychrophilic Vibrio sp.,
as a response to temperature increase and growth phase [12].
A recent study on a dinoflagellate [16] reported a similar trend
(e.g. a lower abundance of MUFAs in the stationary phase).
These results indicated that the growth phase induced signifi-
cant fatty acid composition modifications possibly involved in
protection of cells against aging that may lead to disruption of
the membrane cell structure.
In this experiment, we found an increase in the trans/cis
ratio for unsaturated fatty acids (Table 1). In several works,
the trans/cis ratio was used as an indicator for starvation or
stress, resulting from xenobiotic exposure [11,25]. Further-
more, depending on their hydrophobicity, hydrocarbons accu-
mulate more substantially and can interact with the fatty acid
acyl chains in membrane lipid bilayers [26]. However, the aro-
matic hydrocarbons are expected to partition in the acyl chains
of the phospholipids, and consequently would induce transfor-
mation of a lamellar lipid phase into a hexagonal phase [18].
Moreover, this phenomenon is often associated with an increase
in trans-MUFAs. On the other hand, in 2MPII-PHE cultures,
we did not observe any increase in trans-MUFAs. The observed
change in the ratio trans/cis is due to a decrease in cis MUFAs,
probably correlated with enhancement of membrane rigidity.
4.4. Influence of growth phase in mixed culture
The PLFA composition in mixed cultures is shown in Ta-
ble 1 and Fig. 3. An increase in branched fatty acids, partic-
ularly those from Corynebacterium sp. such as 9Me18:1ω8
and 10Me18:0, led to an increase in cell membrane rigidity.
Someauthorshavesuggestedthatthesechangesareaprotective
mechanism against lipid oxidation [18], due to culture aging.
Fig. 3. Examples of GC/MS total ion current of PLFAs (as methyl esters) of Corynebacterium sp. and Sphingomonas sp. 2MPII grown on a mixture of n-eicosane
and phenanthrene in mixed cultures. * Indicates that identified peak was within the limits of quantification. See Table 1 for peak identification.

Page 7

A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
485
Moreover, such regulation could be valuable for cell adaptation
to environmental variations [23]. Heipieper et al. [13] showed
a similar response of microorganisms to ethanol exposure.
In our experiments, a MUFA decrease (about 1.5-fold) was
counteracted by an increase in SFAs. This result was in good
agreement with previous works [6] reporting an increase in
SFAs/MUFAs when cells entered the stationary growth phase.
4.5. PCA interpretation
Concerning the PCA results, 10Me18:0, which is con-
sidered as a Corynebacterium sp. signature, correlated pos-
itively with the first axis (PC1, Fig. 4). The corresponding
scatter plot (Fig. 5) showed that this fatty acid mainly de-
scribed COR-C20 cultures. On the other hand, 16:1ω9c and
18:1ω9c were negatively correlated with PC1 and therefore
better describe 2MPII-PHE cultures. Concerning PC2, Fig. 5
shows clearly the influence of growth phase on fatty acid com-
position. One of the most striking results was the strong control
of the co-eluted fatty acids 18:0 and 9Me18:1ω8c by the bacte-
rial growth phase. Nevertheless, as shown by the corresponding
eigen vector, they are closer to the COR-C20culture.
We should point out two results shown in Figs. 4 and 5. First,
the ammonium acetate cultures are distinct from these hydro-
carbon cultures, especially for Sphingomonas sp. 2MPII, illus-
trating strong control of the carbon source in the bacterial PLFA
composition. For this strain, changes induced by the carbon
source are more substantial than those resulting from course
time. These findings are in good agreement with previous
works [7]. Second, the fatty acid composition of COR-C20cul-
ture appeared to be more strongly influenced by the growth
phase than 2MPII-PHE fatty acids.
Lastly, for mixed cultures, we also observed a marked effect
of growth phase. At the beginning of the mixed culture experi-
Fig. 4. Vector plots of the PLFA profiles. Annotated fatty acids correspond to
those that most strongly contributed to the variability of the PCA. Numbers
indicate peak numbers reported in Table 1.
ment, the fatty acid composition seemed to be closer to the lipid
composition of Corynebacterium sp. Thus, the changes were
progressive as a function of time, as for Corynebacterium sp.
Nevertheless, for further incubation times, it is difficult to dis-
tinguish similarity/dissimilarity from pure cultures.
In conclusion, the effect of the hydrophobic substrate and
age of culture was successfully demonstrated by PCA and
a simple index, given as the sum of saturated fatty acids divided
by the sum unsaturated (?SFA/?MUFA). We demonstrated,
influenced by two different dominant parameters: the carbon
source for Sphingomonas sp. 2MPII and the growth phase for
Corynebacterium sp.
Lastbutnotleast,mostenvironmentalsampleinterpretations
based on fatty acid patterns are established based on in vitro
cultures. In such cultures, we therefore suggest that the growth
phase must be carefully taken into consideration. Sedimentary
microcosm experiments are currently in progress to check the
influence of culture aging.
for the two studied strains, that the fatty acid compositions were
Acknowledgements
This work was supported by funds from the CNRS and Elf
Aquitaine within the GDR “HYCAR”1123 projects (Origin,
effects and fate of petroleum hydrocarbons in marine environ-
ment). We gratefully acknowledge Mr. A.D. Syakti’s scholar-
ship provided by The French Foreign Ministry.
Fig. 5. Scatter plot of the PLFA profiles of the PCA (87% of variability). Cir-
cles indicate the formed clusters representing distribution of each experiment
(triplicate). The numbers (1–3), (4–6), (7–9), (10–12) indicate the time courses
of Corynebacterium sp. on n-eicosane cultures (T2, T7, T21, and T56), respec-
tively. The numbers (13–15), (16–18), (19–21), and (22–24) indicate the times
courses (T2, T7, T21, and T56) of Sphingomonas sp. 2MPII on phenanthrene
cultures respectively. The mixed culture time courses (T2, T7, T21, and T56)
are given as (25–27), (28–30), (31–33), and (34–36), respectively. AceCOR and
Ace2MPIIrepresentthecultures of Corynebacterium sp.andSphingomonas sp.
2MPII on ammonium acetate, respectively. For each of the two cultures, the last
value is the mean of at least six experiments.

Page 8

486
A.D. Syakti et al. / Research in Microbiology 157 (2006) 479–486
References
[1] B.A. Annous, M.F. Kozempel, M.J. Kurantz, Change in membrane fatty
acid composition of Pediococcus sp. strain NRRL B-2354 in response to
growth conditions and its effects on thermal resistance, Appl. Environ.
Microbiol. 65 (1999) 2857–2862.
[2] E. Aries, P. Doumenq, J. Artaud, M. Acquaviva, J.C. Bertrand, Effects
of petroleum hydrocarbons on the phospholipid fatty acid composition of
a consortium composed of marine hydrocarbon-degrading bacteria, Org.
Geochem. 32 (2001) 891–903.
[3] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purifi-
cation, Can. J. Biochem. Physiol. 35 (1959) 911–917.
[4] Q. Chen, A. Nijenhuis, H. Preusting, J. Dolfing, D.B. Janssen, B. Witholt,
Effects of octane on the fatty acid composition and transition temperature
ofPseudomonasoleovoransmembranelipidsduringgrowthintwo-liquid-
phase continuous cultures, Enz. Microbial. Tech. 17 (1995) 647–652.
[5] T.J. Denich, L.A. Beaudette, H. Lee, J.T. Trevors, Effect of selected en-
vironmental and physico-chemical factors on bacterial cytoplasmic mem-
branes, J. Microbiol. Meth. 52 (2003) 149–182.
[6] R. Diefenbach, H.J. Heipieper, H. Keweloh, The conversion of cis into
trans unsaturated fatty acids in Pseudomonas putida P8: Evidence for
a role in regulation of the membrane fluidity, Appl. Microbiol. Biotech-
nol. 38 (1992) 382–387.
[7] P. Doumenq, M. Acquaviva, L. Asia, J.P. Durbec, Y. Le Dréau, G. Mille,
J.C. Bertrand, Change in fatty acids of Pseudomonas nautica, a marine
denitrifying bacterium, in response to n-eicosane as carbon source and
various culture conditions, FEMS Microbiol. Ecol. 28 (1999) 151–161.
[8] Z. Drici-Cachon, J.F. Cavin, C. Divies, Effect of pH and age of culture on
cellular fatty acid composition of Leuconostoc oenos, Lett. Appl. Micro-
biol. 22 (1996) 331–334.
[9] J.K. Fredrickson, D.L. Balkwill, G.R. Drake, M.F. Romine, D.B. Ringel-
berg, D.C. White, Aromatic-degrading Sphingomonas isolates from the
deep subsurface, Appl. Environ. Microbiol. 61 (1995) 1917–1922.
[10] M. Gilewicz, Ni’matuzahroh, T. Nadalig, H. Budzinski, P. Doumenq,
V. Michotey, J.C. Bertrand, Isolation and characterization of a marine
bacterium capable of utilizing 2-methylphenanthrene, Appl. Microbiol.
Biotechnol. 48 (1997) 528–533.
[11] J.B. Guckert, M.A.Hood, D.C. White, Phospholipid ester-linked fatty acid
profilechangesduringnutrientdeprivationofVibriocholeraeincreasesthe
trans/cis ratio and proportions of cyclopropyl fatty acids, Appl. Environ.
Microbiol. 52 (1986) 794–801.
[12] T.Hamamoto, N.Takata,T.Kudo, K.Horikoshi, Effectof temperatureand
growth phase on fatty acids composition of the psychrophilic Vibrio sp.
strain no. 5710, FEMS Microbiol. Let. 119 (1994) 77–91.
[13] H.J.Heipieper, S. Isken, M. Saliola,Ethanol tolerance and membrane fatty
acid adaptation in adh multiple and null mutants of Kluyveromyces lactis,
Res. Microbiol. 151 (2000) 777–784.
[14] A.R. Johnsen, A. Winding, U. Karlson, P. Roslev, Linking of microorga-
nisms to phenanthrene metabolism in soil by analysis of13C-labeled cell
lipids, Appl. Environ. Microbiol. 68 (2002) 6106–6113.
[15] H. Lechevalier, M.P. Lechevalier, Chemotaxonomic use of lipids: An
overview, in: C. Ratledge, S.G. Wilkinson (Eds.), Microbial Lipids, Aca-
demic Press, London, 1988, pp. 869–902.
[16] M.P. Mansour, J.K. Volkman, S.I. Blackburn, The effect of growth on the
lipid class, fatty acid and sterol composition in the marine dinoflagellate
Gymnodinium sp. in batch culture, Phytochemistry 63 (2003) 145–153.
[17] R. Makula, W.R. Finnerty, Microbial assimilation of hydrocarbons: Fatty
acids derived from normal alkanes, J. Bacteriol. 95 (1968) 2002–2107.
[18] R.V. McDaniel, T.J. McIntosh, S.A. Simon, Non-electrolyte substitu-
tion for water in phosphatidylcholine bilayers, BBA-Biomembranes 731
(1983) 79–108.
[19] J.T. McGarrity, J.B. Armstrong, The effect of temperature and other
growth conditions on the fatty acid composition of Escherichia coli,
Can. J. Microbiol. 27 (1981) 835–840.
[20] W.R. Morrison, L.M. Smith, Preparation of fatty acids methyl ester and
dimethyl acetals from lipids with boron trifluoride-methanol, J. Lipid
Res. 5 (1964) 600–608.
[21] P.D. Nichols, J.B. Guckert, D.C. White, Determination of monounsatu-
rated fatty acid double bond position and geometry for microbial mono-
cultures and complex consortia by capillary GC-MS of their dimethyl
disulfide adducts, J. Microbiol. Meth. 5 (1986) 49–55.
[22] T. Niepel, H. Meyer, V. Wray, W.R. Abraham, Intraspecific variation
of unusual phospholipids from Corynebacterium spp. containing a novel
fatty acid, J. Bacteriol. 80 (1998) 4650–4657.
[23] N.J. Russell, R.I. Evan, P.F. Ter Steeg, J. Hellemons, A. Verheul, T. Abee,
Membrane as a target stress adaptation, Int. J. Food Microbiol. 28 (1995)
255–261.
[24] H. Rutters, H. Sass, H. Cypionka, J. Rullkotter, Phospholipid analysis as
atooltostudycomplexmicrobialcommunitiesinmarinesediments,J.Mi-
crobiol. Meth. 48 (2002) 149–160.
[25] S. Sardessai, S. Bhosle, Tolerance of bacteria to organic solvents, Res.
Microbiol. 153 (2002) 263–268.
[26] J. Sikkema, J.A.M. de Bont, B. Poolman, Mechanisms of membrane toxi-
city of hydrocarbons, Microbiol. Rev. 59 (1995) 201–222.
[27] A.D.Syakti,M.Acquaviva,M.Gilewizc,P.Doumenq,J.C.Bertand,Com-
parison of n-eicosane and phenanthrene degradation by pure and mixed
cultured of two marine bacteria, Environ. Res. 96 (2004) 206–218.
[28] I.V. Tsitko, G.M. Zaitsev, A.G. Lobanok, M.S. Salkinoja-Salonen, Effect
of aromatic compounds on fatty acid composition of Rhodococcus opacus,
App. Environ. Microbiol. 65 (1999) 853–855.
[29] M.J. Valderrama, M. Monteoliva-sanchez, E. Quesada, A. Ramos-Cor-
menzana, Influence of salt concentration on the cellular fatty acid com-
position of the moderately halophilic bacterium Halomonas salina, Res.
Microbiol. 149 (1998) 675–679.
[30] F.J. Weber, J.A.M. De Bont, Adaptation mechanisms of microorganisms
to the toxic effects of organic solvents on membranes, Biochim. Biophys.
Acta 1286 (1996) 225–245.
[31] D.C. White, R.H. Findlay, Biogeochemical markers for measurement of
predation effect on the biomass, community structure, nutritional status,
and metabolic activity of microbial biofilm, Hydrobiologia 159 (1988)
119–123.