Implications of Modifying Membrane Fatty Acid Composition on Membrane

Oxidation, Integrity, and Storage Viability of Freeze-Dried Probiotic,

Lactobacillus acidophilus La-5
Marie-Louise R. W. Hansen, Mikael A. Petersen, and Jens Risbo
Dept. of Food Science, Faculty of Science, University of Copenhagen, Denmark
Magdalena H€
ummer
Chr. Hansen A/S, Hørsholm, Denmark
Dept. of Horticulture and Food Technology, University of Applied Sciences Weihenstephan-Triesdorf, Germany
Anders Clausen
Chr. Hansen A/S, Hørsholm, Denmark

DOI 10.1002/btpr.2074
Published online April 23, 2015 in Wiley Online Library (wileyonlinelibrary.com)

The aim of this study was to investigate the effect of altering the fatty acid profile of the
lipid membrane on storage survival of freeze-dried probiotic, Lactobacillus acidophilus La-
5, as well as study the membrane integrity and lipid oxidation. The fatty acid composition of
the lipid membrane of L. acidophilus La-5 was significantly different upon growth in MRS
(containing Tween 80, an oleic acid source), or in MRS with Tween 20 (containing C12:0
and C14:0), linoleic, or linolenic acid supplemented. Bacteria grown in MRS showed the
highest storage survival rates. No indications of loss of membrane integrity could be found,
and membrane integrity could therefore not be connected with loss of viability. Survival of
bacteria grown with linoleic or linolenic acid was more negatively affected by the presence
of oxygen, than bacteria grown in MRS or with Tween 20 supplemented. A small, but signifi-
cant, loss of linolenic acid during storage could be identified, and an increase of volatile
secondary oxidation products during storage was found for bacteria grown in MRS, or with
linoleic, or linolenic acid supplemented, but not for bacteria grown with Tween 20. Overall,
the results indicate that lipid oxidation and loss of membrane integrity are not the only or
most important detrimental reactions which can occur during storage. By altering the fatty
acid composition, it was also found that properties of oleic acid gave rise to more robust
bacteria than more saturated or unsaturated fatty acids did. V C 2015 American Institute of

Chemical Engineers Biotechnol. Prog., 31:799–807, 2015

Keywords: Lactic acid bacteria, probiotics, fatty acid composition, lipid membrane, freeze-
drying, storage, viability, survival

Introduction meet the demands of the final product. Several investigations

Lactic acid bacteria are used in a wide range of fermented
dairy and meat products. Several strains of lactic acid bacte-
ria are associated with health benefits upon digestion and are
known as probiotics.1,2 To extend the shelf-life of the probi-
otic product, as well as to reduce costs in storage and trans-
portation, most probiotic bacteria are freeze-dried. Storage,
even in the dry state, can, however, have a negative effect
on the viability of probiotics. The loss of viability is
unwanted as probiotics must be alive and able to replicate to
Additional Supporting Information may be found in the online ver-
sion of this article.
Current address of Magdalena H€ ummer: Department of Chemical and
Biological Engineering, Faculty of Engineering, University of Erlangen-
N€urnberg, Germany
Correspondence concerning this article should be addressed to Jens
Risbo at jri@food.ku.dk.
have looked into which factors affect survival, and how to
increase survival of the bacteria during storage in the dry
state.3,4 The factors that influence survival during storage in
the dry state include: composition of the cryo-protectant as
well as water activity, temperature, and availability of
oxygen.4
Oxidation of the lipid membrane has been proposed to be
a cause of loss of viability during storage in the dry state.
Unsaturated fatty acids are prone to oxidation, which can
change the physical and chemical properties of the mem-
brane and hereby decrease bacterial viability.5,6 Oxidation of
unsaturated membrane fatty acids of lactobacilli has previ-
ously been demonstrated by calculating the change in the
ratio between unsaturated and saturated fatty acids (U/S).
This ratio was found to decrease dramatically during storage
of Lactobacillus delbrueckii subsp. bulgaricus, simultane-
ously with a decrease of viable cells.7,8 Similarly the amount

C 2015 American Institute of Chemical Engineers

V 799

800
of polyunsaturated fatty acids (PUFAs), compared to the
amount of saturated palmitic acid (C16:0), decreased during
storage of Weissella paramesenteroides LC11, as the primary
oxidation products of polyunsaturated fatty acids (PUFAs)
increased.9 Thus, there are indications of a link between
membrane oxidation and loss of viability of lactic acid
bacteria.
Membrane damage can also be caused by physical
changes in the lipid membrane. In the process of drying,
phospholipids can undergo phase transition from the liquid
crystalline phase to a gel phase. Upon rehydration the lipids
return to the liquid crystalline phase, but not simultaneously.
The coexistence of both phases can cause lateral lipid phase
separation leading to packing defects and leakage from the
intracellular space.10
Leakage of the intracellular enzyme lactate dehydrogenase
(LDH) as a measure of membrane integrity has been corre-
lated to survival after drying.11 Cells with low degree of via-
bility loss during drying (L. casei subsp. casei and L.
plantarum) gave rise to a small amount of leakage of LDH,
whereas cells with high viability loss upon drying (L. helve-
ticus and L. delbr€ uckii subsp. bulgaricus-12) showed larger
degree of leakage.11 Survival upon freezing and drying can
be improved by altering the lipid membrane to contain a
higher percentage of unsaturated and cyclopropane fatty
acids.12–14 In all investigations the lipid membrane was
altered by changing the fermentation conditions to, for exam-
ple, suboptimal temperature or pH settings. Other cellular
functions will, however, also be activated by this procedure,
such as the production of chaperone proteins, which can also
affect survival in a positive manner.15 This means it is not
possible to distinguish whether the improved survival was
due to the membrane fatty acid composition or due to other
modifications of the cells.
Another way of altering the lipid membrane is by supple-
mentation of fatty acids through the fermentation medium.
Supplementation of fatty acids is, furthermore, vital for the
growth of L. acidophilus, because it lacks the enzymes nec-
essary for synthesis of fatty acids de novo.16 The unsaturated
fatty acid, oleic acid, in particular has been found to promote
growth of Lactobacillus species, and is therefore (in the
form of Tween 80) added to de Man, Rogosa and Sharpe
(MRS) medium, which was developed for lactobacilli,17,18
and utilized in the present study.
Studies18–22 have found that lactobacilli alter the fatty acid
composition of the lipid membrane according to the fatty
acids supplemented in the fermentation medium independ-
ently of their ability to synthesise fatty acids de novo.
Hereby, a more unambiguous set-up can be created, in which
it is possible to investigate the effect of an altered membrane
fatty acid composition, without applying other interfering
chemical stress (e.g., pH or ionic strength) to the cells. Inter-
estingly, no studies have, to our knowledge, investigated the
effect of supplementing different fatty acids through the fer-
mentation medium on the survival during dry storage after
freeze-drying. Nor have such an investigation been coupled
with examinations of membrane integrity or membrane fatty
acid oxidation.
The present study takes its starting point in the potential
links between, on one side, oxidation of membrane fatty
acids, membrane integrity and leakage, and on the other
side, loss of viability during storage in dry state. These phe-
nomena were evaluated as potential general main causes of
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Biotechnol. Prog., 2015, Vol. 31, No. 3
loss of viability. By supplementation of different fatty acids
four versions of Lactobacillus acidophilus La-5 were made,
with varying degree of unsaturation and thus different pro-
pensity for membrane leakage and oxidation. Along with
storage of dry bacteria, the integrity of the lipid membrane
was assessed by determining the activity of leaked LDH, the
degree of oxidation of membrane fatty acids was investi-
gated by analysis of the fatty acid composition, and by anal-
ysis of the volatile lipid oxidation products.
Materials and Methods
Bacterial strain and fermentation media
Stock cultures of the strain, Lactobacillus acidophilus La-
5, was obtained from Chr. Hansen A/S, Hørsholm, Denmark,
and stored at 250 C before utilisation. A portion of the
stock culture was thawed at room temperature before
inoculation.
The inoculation material for the fermentations was grown
with 1% (v/v) stock culture in de Man, Rogosa and Sharpe
medium (MRS broth, DifcoTM, Lactobacilli MRS broth),
which includes 1 g/L of Tween 80, for 17 h at 37 C. The
fermentation medium was also MRS broth from Difco, MRS
with 0.15 g/L Tween 20 (a polysorbate surfactant like Tween
80, but primarily containing C12:0 and C14:0 rather than
C18:1, Sigma-Aldrich), 10 mg/L linoleic acid (C18:2cis9,12,
Sigma-Aldrich), or 10 mg/L a-linolenic acid
(C18:3cis9,12,15, Sigma-Aldrich) supplemented.
Fermentation
Fermentation of L. acidophilus La-5 was conducted anae-
robically in 35 L fermenters (Chr. Hansen A/S), at 37 C and
pH 6.0, with a constant agitation of 370 rpm; one fermenta-
tion for each of the four types of media. The pH was con-
trolled by the addition of 13.4% ammonium hydroxide in
water (Brenntag Nordic A/S). Bacterial growth throughout
the fermentations was monitored by measuring the optical
density at 600 nm (OD600), and the bacterial culture was har-
vested when it had reached the stationary growth phase, as
determined by the ammonium hydroxide consumption rate.
Growth was unaffected by the supplementation of exogenous
fatty acids, and the OD600/cell counts (CFU) of the bacterial
cultures at the end of fermentation was 13.62/5.0 3 108
(grown in MRS), 15.36/6.05 3 108 (grown with Tween 20),
12.66/5.45 3 108 (grown with C18:2), and 13.96/7.60 3 108
(grown with C18:3).
Harvest, formulation, and pellet freezing
Bacterial cells from each fermentation culture were har-
vested in a desludging disc centrifuge (Westfalia, CSC6-06-
476), and subsequently washed by mixing one part concen-
trated culture to nine parts 20% (w/w) sucrose solution. The
resuspended cells were then centrifuged in a batch centrifuge
(Heraeus Cryofuge 5500i) to remove medium and wash solu-
tion leftovers. For cryo-protection the cells in the pellet were
mixed with a 30% (w/w) sucrose solution (0.3 g solution per
g fermentation product), before they were frozen in liquid
nitrogen in pellets of approximately 5 mm in diameter. The
frozen pellets were stored at 250 C before freeze-drying
later the same day. Freeze-drying was performed at a cham-
ber pressure of 0.3 mbar with temperature increasing from
242 to 32 C with 1.5 C/min (Hetosicc freeze dryer, CD-10-

Biotechnol. Prog., 2015, Vol. 31, No. 3
1, Heto Lab equipment, Heto-Holten A/S). The freeze-drying
was ended when the weight of the product had been stable
for at least 2 h. The dry product was stored in alu-PE bags
at 250 C until further use.
Experimental design of storage experiment
The freeze-dried bacterial cultures grown in MRS, MRS
with 0.15 g/L Tween 20, MRS with 10 mg/L C18:2, or
MRS with 10 mg/L C18:3, were stored at two different con-
ditions: 30 C, aw 5 0.32 (saturated solution of MgCl2•H2O),
nitrogen atmosphere (0% O2), or 30 C, aw 5 0.32
(MgCl2•H2O), atmospheric air (21% O2). Freeze-dried pellets
were added to individual plastic boxes (480 mL, 500/50
Series, Hofst€atter & Ebbesen A/S) dependent on storage con-
ditions and time of storage. For each freeze-dried bacterial
culture, storage condition, and time of storage there was one
box. Boxes were divided into two compartments; one for the
saturated salt solution and one for the freeze-dried pellets.
Each box was placed in an impermeable alu-PE bag and
using a vacuum chamber packing machine (Multivac A 300/
41/42) either atmospheric air or nitrogen was added as the
atmosphere and the bag sealed.
The oxygen concentration in the alu-PE bags was deter-
mined using a Checkmate 9900, O2 (PBI-Dansensor A/S) for
each sample taken out for analysis, and the nitrogenous atmos-
phere never contained more than 0.13% O2 (data not shown).
Determination of cell counts
Samples were taken out for analysis immediately after
freeze-drying, after 1, 2, 3, 6, 10, and 15 weeks of storage,
and the viability was determined in duplicate by counting
colony forming units (CFU). The freeze-dried bacteria were
diluted 100 times in Maximum Recovery Diluent (1.0 g/L
peptone, 8.5 g/L NaCl, pH 7.0, Oxoid) and the cell counts
were determined according to the procedure of.3
Determination of lipid membrane integrity
The activity assay of lactate dehydrogenase (LDH) was
based on the change in absorbance at 340 nm as NADH was
oxidized to NAD1 by pyruvate as catalysed by LDH. The
results are given as the negative change in absorbance per
gram dry weight bacterial culture per unit time, dA340nm/dt
and expressed in units [(g min)21]. All analyses were carried
out in minimum duplicate.
A 100 mg sample of freeze-dried bacteria was reconsti-
tuted in 1 mL 0.05 M phosphate buffer, pH 7 and separated
into two samples: One for measuring the possible leakage of
LDH from the intracellular space and another for measuring
the total LDH activity within the cells. To determine the
amount of LDH leaked from the cells, the first sample was
centrifuged to separate cells and buffer, and the supernatant
was collected for analysis. To determine the total amount of
LDH activity in the cells the second sample was kept in sus-
pension and the cells lysed. Lysis was performed in the
phosphate buffer in a 96 well plate with 200 mL bacteria sus-
pension and 80 mg glass beads (150–212 mm, acid-washed,
Sigma-Aldrich) in each well. The cells and glass beads were
mixed and sonicated (Sonicator 3000, Misonix) at 42W for
30 min consisting of 30 sec pulses, and 10 sec breaks. The
glass beads and cell debris were separated from the buffer
by centrifugation and the supernatant transferred to a fresh
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
801
96 well plate before analysis. The reaction mix contained the
following (all from Sigma-Aldrich): 96 mM, Tris-HCl, pH
7.2; 5 mM MgCl2•6 H2O; 3 mM fructose 1,6-phosphate;
10 mM pyruvate; 0.6 mM NADH, water and 20% (v/v) sam-
ple solution. Immediately after mixing, the plate was trans-
ferred for absorbance reading at 340 nm every second
minute for 30 min (SPECTROstar Omega).
Membrane fatty acids determined by gas chromatography
Membrane fatty acids were directly converted to fatty acid
methyl esters (FAMEs) by trans-esterification.23 All analyses
were carried out in triplicate.
For the analysis of membrane fatty acids about 0.05 g
freeze-dried bacteria pellets were used. Before methylation
the freeze-dried cells were washed with methanol (Sigma-
Aldrich) to remove residual medium components. Fatty acid
analysis of MRS medium and Tween 20 was performed
directly on the liquid solutions. For the direct methylation a
1 M solution of sodium methoxide (30% sodium methylate
in methanol, Merck) in methanol (Sigma-Aldrich) was used.
The trans-esterification proceeded during 30 min of overhead
stirring (Heto Mastermix, Heto-Holten A/S). The fatty acid
methyl esters were extracted with n-heptane (Pro analysis,
Merck) and analysed by gas chromatography (HP 6890,
Hewlett Packard) equipped with a capillary column (Omega-
wax 320, 30 m 3 0.320 mm 3 0.25 mm, Supelco). The
oven temperature programme was: 50 C for 1 min; from
50 C to 180 C at 15 C/min; from 180 to 240 C at 3 C/min,
and held at 240 C for 10 min. Injection volume was 1 ml
and split ratio was 1:25. Hydrogen was used as carrier gas
and the flow was 30 mL/min. The results were analysed with
Chemstation software (Agilent Technologies) and fatty acid
methyl esters were identified by comparing retention times
with those of known standards. The areas of the individual
peaks were used to determine the relative percentage of each
fatty acid present.
The fatty acids detected were: C12:0, C14:0, C16:0,
C16:1cis9, C18:0, C18:1cis9, C18:1cis11, C18:2cis9,12,
C18;3cis9,12,15, C19d9,10. To determine if the membrane
fatty acid composition of the four freeze-dried bacterial cul-
tures was significantly different, the ratio between unsatu-
rated (C16:1, C18:1, C18:2, C18:3, and C19d9,10) and
saturated (C12:0, C14:0, C16:0, and C18:0) fatty acids was
calculated.
To evaluate if the amount of polyunsaturated fatty acids
(PUFAs) changed during storage, the ratio between the rela-
tive percentages of linoleic acid and palmitic acid (C18:2/
C16:0) and between linolenic acid and palmitic acid (C18:3/
C16:0) was calculated.
Analysis of volatile oxidation products
All analyses were carried out in triplicate. One gram of
freeze-dried bacteria pellet was crushed to a powder before
extraction of volatiles was started. The crushed freeze-dried
bacteria were equilibrated to 50 6 1 C in a circulating water
bath and then purged with nitrogen (100 mL/min) for 60
min. Volatile compounds were collected on Tenax-TA traps.
The traps contained 250 mg of Tenax-TA with mesh size
60/80 and a density of 0.37 g/mL (Buchem by, Apeldoorn,
The Netherlands).
The trapped volatiles were desorbed using an automatic
thermal desorption unit (ATD 400, Perkin Elmer, Norwalk).
 15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
802 Biotechnol. Prog., 2015, Vol. 31, No. 3

Table 1. Relative Fatty Acid Composition in Percentage of MRS Used Agilent Technologies, Palo Alto, California), was used for
in All Fermentations and of Tween 20 Used for Supplementation data analysis. Concentrations are presented as peak areas.
Fatty Acid Designation and Name MRS Tween 20
C12:0 (Lauric) 0.1 55.9
C14:0 (Myristic) 1.1 19.1 Statistical analysis
C16:0 (Palmitic) 11.3 8.3
C16:1 (Palmitoleic) 0.2 4.1 Univariate analyses were carried out by Student’s t-test
C18:0 (Stearic) 14.0 2.8 (2-tail paired with significance measured at a probability
C18:1cis9 (Oleic) 68.9 6.1 level of P  0.05) using Excel (Microsoft Office, 2010) for
C18:1cis11 (Vaccenic) 3.1 1.3 the fatty acid analyses and activity assay for LDH, and JMP
C18:2cis9,12 (Linoleic) 0 ND
C18:3cis9,12,15 (Linolenic) ND ND
for the analysis of hexanol levels (JMP 9.0.2, SAS Institute,
C19d9,10 (Dihydrosterculic) 0.1 2.3 Cary NC). Multivariate analyses were done using Principal
Component Analysis (PCA) (Latentix 2.11, Latent5 Aps,
ND, not detected. Copenhagen).

Primary desorption was carried out by heating the trap to Results

250 C with a flow (60 mL/min) of carrier gas (H2) for 15.0
min. The stripped volatiles were trapped in a Tenax TA cold
trap (30 mg held at 5 C), which was subsequently heated at
300 C for 4 min (secondary desorption, outlet split 1:10).
This allowed for rapid transfer of volatiles to a gas
chromatograph-mass spectrometer (GC-MS, 7890A GC-
system interfaced with a 5975C VL MSD with Triple-Axis
detector from Agilent Technologies, Palo Alto, California)
through a heated (225 C) transfer line.
Separation of volatiles was carried out on a DB-Wax cap-
illary column 30 m long 3 0.25 mm internal diameter, 0.50
mm film thickness. The column pressure was held constant at
2.4 psi resulting in an initial flow rate of approximately
1.2 mL/min using hydrogen as carrier gas. The column tem-
perature programme was: 10 min at 30 C, from 30 C to
240 C at 8 C/min, and finally 5 min at 240 C. The mass
spectrometer was operating in the electron ionisation mode
at 70 eV. Mass-to-charge ratios between 15 and 300 were
scanned. Volatile compounds were identified by probability
based matching of their mass spectra with those of a com-
mercial database (Wiley275.L, HP product no. G1035A).
The software program, MSDChemstation (Version E.02.00,
Fatty acid composition of the lipid membrane
To investigate the effect of an altered fatty acid profile on
storage survival, L. acidophilus La-5 was grown in MRS
(hereafter denoted MRS), or in MRS with Tween 20 (con-
taining C12:0 and C14:0, hereafter denoted Tween 20), fatty
acid C18:2 (hereafter denoted C18:2) or fatty acid C18:3
(hereafter denoted C18:3) supplemented. The fatty acid com-
position of the MRS medium with Tween 80, used as the
basic medium in all fermentations, as well as Tween 20 used
for supplementation, is shown in Table 1. It can be observed
from the table that MRS medium contains a large amount of
C18:1cis9, and some C18:0 and C16:0, and Tween 20 pri-
marily consists of C12:0 and C14:0. By addition of fatty
acid supplements rather than applying suboptimal fermenta-
tion conditions, it was possible to alter the fatty acid compo-
sition of the lipid membrane without otherwise changing the
fermentation condition for the bacteria and thereby minimiz-
ing other responses of the bacteria.
The membrane fatty acid composition was modified
according to the fatty acids supplied in the fermentation
medium and can be seen for the freeze-dried bacteria in

Figure 1. The membrane fatty acid profile and ratio of unsaturated to saturated fatty acids (U/S) after freeze-drying of L. acidophilus
La-5 grown in MRS medium or grown in MRS medium with Tween 20, C18:2, or C18:3 supplemented.
Each fatty acid is given as the percentage of the total amount of fatty acids. Error bars indicate standard deviation (n 5 3). * P  0.05. P values
were estimated by paired Student’s t-test. Comparisons were made for MRS and the three other samples for the U/S ratios.

Biotechnol. Prog., 2015, Vol. 31, No. 3
Figure 2. Survival (Log(CFU/g)) of L. acidophilus La-5 grown
in MRS medium or grown in MRS medium with
Tween 20, C18:2, or C18:3 supplemented and stored
with no oxygen present (a) or in atmospheric air (b).
Error bars indicate standard deviation (n 5 2).
Figure 1. The fatty acid results are also available in table
format in the Supporting Information I, Table 1. The results
were comparable to the outcome of preliminary studies,
which are presented in the Supporting Information II. The
fatty acids C12:0 and C14:0 were only present in bacteria
grown with the supplement Tween 20 and similarly C18:2 or
C18:3 was only present in bacteria grown with these fatty
acids, respectively. The differences in the membrane fatty
acid composition of the four fermentation samples can be
seen from the ratio of unsaturated to saturated fatty acids (U/
S) after freeze-drying (Figure 1). By comparison to the MRS
sample the Tween 20 sample had a significantly lower U/S
ratio, whereas both the C18:2 and C18:3 samples had signifi-
cantly greater U/S ratios.
Survival of L. acidophilus La-5 during storage
The storage conditions for the freeze-dried fermentation
products were 30 C, a water activity of 0.32, and with either
oxygen replaced by nitrogen (about 0.1% O2) or in regular
atmospheric air (21% O2). The amount of viable cells
decreased for all four samples during storage (Figure 2). The
decrease in viable cells was dependent on both the mem-
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
803
Figure 3. Difference in survival (Log(CFU/g)) between storage
with no oxygen present and in atmospheric air.
To create the graph the loss of viable cells during storage in
atmospheric air was subtracted from the loss during storage
with no oxygen present. The greater the negative results, the
greater the loss during storage in atmospheric air was compared
to the loss during storage with no oxygen present.
brane fatty acid profile of the bacteria and the storage
atmosphere.
Overall the MRS sample, which had the highest amount of
C18:1 fatty acid in the lipid membrane, showed the best sur-
vival throughout both storage experiments. After 15 weeks
of storage without oxygen present the survival of the MRS
sample had decreased by a factor of around 105, whereas the
other three samples had decreased by a factor of minimum
107. The Tween 20 sample showed the lowest survival dur-
ing the first 6 weeks of storage without oxygen present, after
which the loss of CFU/g was similar to that of the C18:2
and C18:3 samples.
Oxygen had a negative effect on storage survival for all
four samples. With oxygen present the loss of CFU was sim-
ilar for bacteria grown with the supplements Tween 20,
C18:2, or C18:3. After 6 weeks of storage viable cells could
be detected only in the MRS sample. Furthermore, survival
of bacteria grown with polyunsaturated fatty acids (PUFAs)
was more negatively affected by the presence of oxygen,
than the MRS and Tween 20 samples (Figure 3). After 2
weeks of storage with oxygen, bacteria grown with PUFAs
had lost about 2.7 log units of viable cells more than after 2
weeks of storage without oxygen. After 2 weeks of storage
with oxygen, the MRS and Tween 20 samples had lost about
1.7 log units of viable cells more than after 2 weeks of stor-
age without oxygen. These results indicate that bacteria with
PUFAs in the lipid membrane are more sensitive to oxygen
and suffer more loss of viability than bacteria with only
monounsaturated and saturated fatty acids in the lipid mem-
brane. The proportionally high storage stability in the dry
state of MRS samples indicates that a high share of C18:1
in the membrane is somehow optimal for dry storage of L.
acidophilus La-5.
Investigation of membrane integrity
To investigate the membrane integrity after freeze-drying
and during storage the release of lactate dehydrogenase

804
Figure 4. Activity of released LDH after freeze-drying and
during storage as an expression of membrane leak-
age from L. acidophilus La-5 grown in MRS
medium or grown in MRS medium with Tween 20,
C18:2, or C18:3 supplemented and stored with no
oxygen present (a) or in atmospheric air (b).
The results are given as the negative change in absorbance of
NADH per gram dry weight bacterial culture per unit time,
dA340nm/dt and expressed in units (g min)21. Error bars indi-
cate standard deviation (n 5 2).
(LDH) from the intracellular space was measured as an indi-
cator for leakage (Figure 4). The total activity of LDH of
the bacteria at the beginning of the storage experiment was
415(g min)21 for the MRS sample, 524 for the Tween 20
sample, 529 for the C18:2 sample, and 386 for the C18:3
sample. The total activity of LDH in the bacteria decreased
during storage with 10 to 60% of the initial activity (data
not shown).
Compared with the total activity of LDH only between 0.4
and 1% of the total LDH amount had leaked out of the bac-
teria after freeze-drying, independent of the membrane fatty
acid composition. Furthermore, there were no significant
changes in the activity of leaked LDH during storage. The
amount of leakage from the Tween 20 sample was signifi-
cantly (P < 0.023) higher after freeze-drying compared with
the three other samples, and remained higher throughout
both storage experiments. For all samples the amount of
leakage was low when considering the actual total loss of
viability during the same period and thus only a minority of
the dead cells could be considered leaky.
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Biotechnol. Prog., 2015, Vol. 31, No. 3
Figure 5. Development of the ratio of linoleic/palmitic (C18:2/
C16:0) acid (a) and linolenic/palmitic (C18:3/C16:0)
acid (b) during storage with no oxygen present or in
atmospheric air of L. acidophilus La-5 grown in
MRS with either C18:2 or C18:3 supplemented.
Points marked with an asterisk (*) are significantly different
from the value obtained immediately after freeze-drying,
P < 0.05, Student’s t-test. Error bars indicate standard deviation
(n 5 3).
Evaluation of PUFA degradation
The changes in the fatty acid composition during storage
were found by inspection to be less than 1% for each of the
fatty acids (data not shown). The survival results indicated,
however, that bacteria with PUFAs in the lipid membrane
were more sensitive to oxygen, than bacteria with only
monounsaturated and saturated fatty acids in the lipid mem-
brane. To evaluate whether the sensitivity to oxygen was
caused by oxidative degradation of PUFAs, the level of these
during storage was investigated in greater detail. More spe-
cifically, the ratio between linoleic acid and palmitic acid
(C18:2/C16:0) and between linolenic acid and palmitic acid
(C18:3/C16:0) was investigated. Only the samples grown
with C18:2 or C18:3 fatty acids contained these in the lipid
membrane, and therefore data for the MRS and Tween 20
samples are not shown.
Changes in the ratios compared with the initial ratio are
shown in Figure 5. The initial relative percentages were 14%
for C18:2 and 12% for C18:3 (Figure 1) and the initial ratios
were 1.01 for C18:2/C16:0 and 0.90 for C18:3/C16:0.
The ratios decreased more during storage with oxygen,
than during storage without oxygen. After 15 weeks of stor-
age without or with oxygen the C18:2/C16:0 ratio had
decreased with 5% or 11% from the initial value, however

Biotechnol. Prog., 2015, Vol. 31, No. 3
Figure 6. Concentration (GC peak area, mean of triplicates) of
volatile compounds in freeze-dried L. acidophilus La-
5 during 15 weeks of storage.
Results of a principal component analysis: Storage samples of
freeze-dried bacteria (score plot) (a) and volatile compounds
(loadings) (b). L. acidophilus La-5 was grown in MRS medium
(filled squares) or grown in MRS medium with Tween 20
(filled triangles), C18:2 (filled circles), or C18:3 (open squares)
supplemented. The samples used were collected after freeze-
drying (W0), and after 6 and 15 weeks (W6 and W15) of stor-
age with no oxygen present (0%) or in atmospheric air (21%).
The three time points are marked with circles.
not statistically significantly (P > 0.05). After 15 weeks of
storage without or with oxygen the C18:3/C16:0 ratio had
decreased with 2% (P < 0.046) or 9% (P < 0.000) from the
initial value. Although an indication of degradation of C18:2
could be identified, nothing statistically conclusive can be
said on degradation of C18:2 based on the fatty acid
analysis.
Development of volatile oxidation products during storage
To supplement the results from the fatty acid analysis the
volatile lipid oxidation products were analysed using
dynamic headspace GC-MS. The composition and quantity
of volatile compounds were assessed at three time points
during storage: Week 0, 6, and 15. The results of the analy-
sis of volatile components are shown as principal component
analysis in Figure 6. The first two principal components
accounted for 67% of the total variability. The first principal
component illustrates changes in concentration of volatile
compounds dependent on the time of storage. The second
principal component illustrates changes in concentration of
volatile compounds dependent on the presence of oxygen
during storage.
After freeze-drying the dominant volatile compounds
included two isomers of methylbutanal, diacetyl, acetic acid,
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
805
1-butanol, pentanal, hexanal, and heptanal. During storage,
and going from left to right in the loadings plot (Figure 6b),
the concentration of the initially dominant compounds
decreased whereas the alcohols, hexanol and heptanol,
increased in concentration.
For the fatty acid C18:1 the most abundant volatile oxida-
tion products expected were the aldehydes heptanal, octanal
and nonanal. For C18:2 primarily the aldehydes hexanal and
heptanal were expected, and for C18:3 primarily the alde-
hyde (E,Z)-2,4-heptadienal was expected.24 Of these hexanal
and heptanal were detected, indicating oxidation of C18:1
and C18:2, but no oxidation product specific for C18:3 was
detected. The amount of hexanal and heptanal was expected
to increase with time, but instead it decreased. The changes
in concentration during storage of the previously mentioned
volatile compounds were, nevertheless, significant and it was
possible to fit a model to predict time of storage using partial
least squares regression (data not shown, R2 5 0.78).
Going from the bottom to the top of the scores plot (Fig-
ure 6a) it was possible to distinguish the samples dependent
on the presence of oxygen during storage. It should, how-
ever, be noted that the Tween 20 sample could not be sepa-
rated dependent on the presence of oxygen as the other
samples could.
The concentration of the aldehydes hexanal and heptanal
decreased during storage, but the corresponding alcohols of
hexanal and heptanal (hexanol and heptanol) increased. The
alcohols probably originated from chemical reduction of
aldehydes. The increase of hexanol during storage was fur-
ther investigated (Figure 7).
The concentration of hexanol increased from week 0 to
week 6, and kept stable from week 6 to week 15. The
increase of hexanol was also dependent on the presence of
oxygen during storage. For the MRS and the C18:3 samples
the concentration of hexanol was about 11 times greater after
15 weeks of storage when oxygen was present. For the
C18:2 sample the concentration of hexanol was about 7
times greater, and for the Tween 20 sample it was about 6
times greater. The increase in the hexanol concentration for
the Tween 20 sample stored with oxygen, however, corre-
sponded to less than one tenth of the increase observed for
the MRS sample. The lower concentration of hexanol in the
Tween 20 sample agrees with the results in Figure 6a, where
the Tween 20 samples cannot be separated dependent on the
presence of oxygen as the other samples can.
Discussion
In order to investigate the proposed relation between loss
of viability of lactic acid bacteria in dry state and oxidation
of membrane lipids or loss of membrane integrity, a set of
four versions of L. acidophilus La-5 having different degree
of membrane fatty acid unsaturation was produced. The idea
was that such set of bacteria also could be expected to show
different degree of propensity for oxidation and leakage.
Due to the incorporation of the supplied fatty acids in the
lipid membrane the Tween 20, C18:2, and C18:3 samples
contained a smaller percentage of the fatty acid C18:1 and a
smaller percentage of the cyclopropane fatty acid, dihydros-
terculic acid, C19d9,10. This is consistent with the fact that
cyclopropane fatty acid is derived from mostly octadecenoic
acids in lactic acid bacteria and specifically from C18:1cis9
to C19d9,10.25 Furthermore, the formation of cyclopropane

806
Figure 7. Hexanol concentration in freeze-dried L. acidophilus La-5 grown in MRS medium (a) or grown in MRS medium with Tween
20 (b), C18:2 (c), or C18:3 (d) supplemented.
The samples used were collected after freeze-drying (W0), and after 6 and 15 weeks (W6 and W15) of storage with no oxygen present or in atmos-
pheric air. For each fermentation sample, bars marked with different letters represent values that are significantly different, P  0.05, Student’s t-test.
Error bars indicate standard deviation (n 5 3).
fatty acid begins as bacteria enter the stationary phase, but
can also increase upon stress.13,26,27 As the proportions
between C18:1 and C19d9,10 were the same for all four
samples this could indicate that the bacteria supplemented
with Tween 20, C18:2, or C18:3 did not show other
responses leading to unproportionally high degree of synthe-
sis of C19d9,10. This is consistent with the approach that
supplementation of fatty acids can modify membrane fatty
acid composition without applying other interfering stress on
the bacteria. The addition of exogenous fatty acids, could,
however affect the bacterial cell in other perspectives. The
altered membrane fatty acid composition might affect the
phospholipid content of the membrane, however no literature
has been found to support this notion. The membrane fluidity
would also be altered as a result of the membrane fatty acid
composition, which could affect the activity of membrane
proteins.28 Further studies are needed to investigate whether
there is change in activity of membrane protein due to fatty
acid supplementation and the impact on bacterial viability.
The total LDH activity was rather constant during storage
in dry state, indicating a high degree of stability of this
enzyme, and therefore it can be considered a good indicator
of membrane integrity. The activity of extracellular LDH,
due to loss of membrane integrity, was low (0.4–1%), com-
pared to the total activity of LDH and did not develop dur-
ing the time of storage. The relative amount of leaked LDH
was significantly higher for the Tween 20 sample, but was
not dependent on membrane fatty acid composition of the
other three L. acidophilus La-5 versions. For all four sam-
ples, however, the relative amount of leaked LDH was also
low compared to the loss of vitality as detected by CFU
(several orders of magnitude). This disproportion indicates
that loss of membrane integrity is unlikely to be the main
cause of loss of viability in dry state, but rather loss of mem-
brane integrity can be related to loss of viability in the dry-
ing step, where the lipid membrane possibly undergoes
phase transitions as seen in10. Another investigation of mem-
brane integrity on six Lactobacillus species, using LDH as
the indicator, supports this relation by measuring an increase
in leaked LDH upon drying, but either no or only a small
increase in leakage after a 100 days of storage at 3 C.11 It
should be noted that the high molecular weight of LDH
might lead to undetected leakage and the results could have
been different if smaller molecules, like electrolytes, had
been used as indicator of leakage. It is, however, still inter-
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Biotechnol. Prog., 2015, Vol. 31, No. 3
esting that almost no leakage of LDH is detected even when
the population of dead cells in reality is 100%.
The degradation of unsaturated fatty acids has been
observed during storage of freeze dried bacteria as a large
change of the ratio between unsaturated and saturated fatty
acids. In the present study the ratio of C18:2/C16:0 and
C18:3/C16:0 was evaluated to determine if the PUFAs were
oxidized. In a previous study on freeze-dried Weissella para-
mesenteroides LC11 the change in C18:2/C16:0 and C18:3/
C16:0 during storage at 20 C showed a large decrease
between 30% (in vacuum-sealed aluminium foiled bags) and
80% (ambient atmosphere and an aw of 0.23).9 In the present
study, the ratio changed by considerably smaller amounts
during storage, as only the ratio of C18:3/C16:0 decreased
significantly during storage with oxygen. Therefore less
impact on membrane function must be expected due to alter-
ation of lipid composition, which correlates well with the
high membrane integrity found for all L. acidophilus La-5
versions. When comparing the two studies, it should be
noted that study9 differs from the present one by neither
washing the media leftovers off the cells, nor formulating
culture with any cryo-protectant before freeze drying. This is
interesting, as the media leftovers could contain radicals pro-
duced by lipid oxidation (e.g., peroxy, alkoxyl, or alkyl radi-
cals), which can initiate and accelerate oxidation of the
lipids in the cell membrane.24 Furthermore, cryo-protectants,
primarily disaccharides, are usually added before drying a
bacterial culture to protect the cells in the frozen and in the
dry state by stabilizing membrane and proteins.29
The imposed alteration of lipid composition by supple-
mentation of fatty acids showed that bacteria with more
unsaturated membrane lipids were more sensitive to the pres-
ence of oxygen than the more saturated membranes. This
indicates that some oxidative processes are taking place dur-
ing storage as also observed by analysis of volatile compo-
nents. The loss of viability and lipid oxidation do, however,
not correlate, as the level of lipid oxidation products
increases with the same magnitude for the MRS, C18:2, and
C18:3 samples, whereas the loss of viability does not
increase with the same magnitude.
It has previously been found that supplementation of
C18:1 fatty acid is important for growth of lactic acid bacte-
ria,22 can increase survival in gastric juice,29 and increase
resistance to freezing and frozen storage.12 The present study

Biotechnol. Prog., 2015, Vol. 31, No. 3
adds to this by showing that the sample with the highest
content of C18:1 fatty acid (the MRS sample) also showed
the best viability during storage in dry state. Therefore it is
recommended to add C18:1 to fermentation medium used for
bacteria intended for freeze-drying and dry storage. The
mechanism behind the favourable properties of C18:1 is cur-
rently unknown, but is unlikely to be related to membrane
oxidation as a more saturated membrane lead to less viable
cells.
From the present study, it can be concluded that the prop-
erties of the fatty acid C18:1 are favourable for producing
freeze-dried L. acidophilus La-5, which are stable during
storage under dry conditions. By altering the fatty acid com-
position of the lipid membrane the survival of L. acidophilus
La-5 during dry storage is clearly affected. No connection
between the integrity of the lipid membrane and storage sur-
vival can be found. Analysis of volatile compounds indicates
formations of lipid oxidations products, but the fatty acid
composition of the bacterial membranes are not significantly
changed as a result of lipid oxidation. Furthermore, no con-
nection between the formation of lipid oxidation products
and storage survival can be found and so lipid oxidation
should not be regarded as the main mechanism for loss of
viability in dry state, rather research should be directed
toward finding alternative mechanisms.
Acknowledgments
Chr. Hansen A/S, Hørsholm, Denmark is thanked for pro-
viding the probiotic strain and technical assistance with the
production of the freeze-dried bacterial product. This work
was supported by Chr. Hansen A/S, Hørsholm, Denmark,
The European Regional Development Fund and Vækstforum
Hovedstaden via the Øresund Food Networks projekt” Sund
Vækst,” and the University of Copenhagen, Denmark.
Literature cited
1. Santosa S, Farnworth E, Jones P. Probiotics and their potential
health claims. Nutr Rev. 2006;64:265–274.
2. Naidu AS, Bidlack WR, Clemens RA. Probiotic spectra of lactic
acid bacteria (LAB). Crit Rev Food Sci Nutr. 1999;39:13–126.
3. Kurtmann L, Carlsen CU, Risbo J, Skibsted LH. Storage stabil-
ity of freeze–dried lactobacillus acidophilus (la-5) in relation to
water activity and presence of oxygen and ascorbate. Cryobiol-
ogy. 2009;58:175–180.
4. Carvalho AS, Silva J, Ho P, Teixeira P, Malcata FX, Gibbs P.
Relevant factors for the preparation of freeze-dried lactic acid
bacteria. Int Dairy J. 2004;14:835–847.
5. Borst JW, Visser NV, Kouptsova O, Visser AJWG. Oxidation
of unsaturated phospholipids in membrane bilayer mixtures is
accompanied by membrane fluidity changes. Biochim Biophys
Acta. 2000;1487:61–73.
6. Santivarangkna C, Kulozik U, Foerst P. Inactivation mecha-
nisms of lactic acid starter cultures preserved by drying proc-
esses. J Appl Microbiol. 2008;105:1–13.
7. Castro HP, Teixeira PM, Kirby R. Changes in the cell mem-
brane of lactobacillus bulgaricus during storage following
freeze-drying. Biotechnol Lett. 1996;18:99–104.
8. Teixeira P, Castro H, Kirby R. Evidence of membrane lipid oxi-
dation of spray-dried lactobacillus bulgaricus during storage.
Lett Appl Microbiol. 1996;22:34–38.
9. Yao A, Coulibaly I, Lognay G, Fauconnier M, Thonart P. Impact
of polyunsaturated fatty acid degradation on survival and acidifi-
cation activity of freeze-dried weissella paramesenteroides lc11
during storage. Appl Microbiol Biotechnol. 2008;79:1045–1052.
15206033, 2015, 3, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/btpr.2074 by Roskilde University, Wiley Online Library on [20/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
807
10. Leslie SB, Israeli E, Lighthart B, Crowe JH, Crowe LM. Treha-
lose and sucrose protect both membranes and proteins in intact
bacteria during drying. Appl Environ Microbiol. 1995;61:3592.
11. Selmer-Olsen E, Birkeland S, Sørhaug T. Effect of protective
solutes on leakage from and survival of immobilized lactobacil-
lus subjected to drying, storage and rehydration. J Appl Micro-
biol. 1999;87:429–437.
12. Beal C, Fonseca F, Corrieu G. Resistance to freezing and frozen
storage of streptococcus thermophilus is related to membrane
fatty acid composition. J Dairy Sci. 2001;84:2374–2356.
13. Li C, Zhao J, Wang Y, Han X, Liu N. Synthesis of cyclopro-
pane fatty acid and its effect on freeze-drying survival of lacto-
bacillus bulgaricus l2 at different growth conditions. World J
Microbiol Biotechnol. 2009;25:1659–1665.
14. Murga MLF, Cabrera GM, De Valdez GF, Disalvo A, Seldes
AM. Influence of growth temperature on cryotolerance and lipid
composition of lactobacillus acidophilus. J Appl Microbiol.
2000;88:342–348.
15. Corcoran BA, Ross RP, Fitzgerald GF, Dockery P, Stanton C.
Enhanced survival of GroESL-overproducing lactobacillus para-
casei NFBC 338 under stressful conditions induced by drying.
Appl Environ Microbiol. 2006;72:5104–5107.
16. Altermann E, Russell W, Azcarate-Peril M, Barrangou R, Buck
B, McAuliffe O, Souther N, Dobson A, Duong T, Callanan M,
Lick S, Hamrick A, Cano R, Klaenhammer T. Complete
genome sequence of the probiotic lactic acid bacterium lactoba-
cillus acidophilus NCFM. Proc Nat Acad Sci USA. 2005;102:
3906–3912.
17. De Man J, Rogosa d, Sharpe ME. A medium for the cultivation
of lactobacilli. J Appl Bacteriol. 1960;23:130–135.
18. Partanen L, Marttinen N, Alatossava T. Fats and fatty acids as
growth factors for lactobacillus delbrueckii. Syst Appl Microbiol.
2001;24:500–506.
19. Jenkins J, Courtney P. Lactobacillus growth and membrane
composition in the presence of linoleic or conjugated linoleic
acid. Can J Microbiol. 2003;49:51–57.
20. Kankaanpaa P, Salminen S, Isolauri E, Lee Y. The influence of
polyunsaturated fatty acids on probiotic growth and adhesion.
FEMS Microbiol Lett. 2001;194:149–153.
21. Kankaanpaa P, Yang B, Kallio H, Isolauri E, Salminen S.
Effects of polyunsaturated fatty acids in growth medium on lipid
composition and on physicochemical surface properties of lacto-
bacilli. Appl Environ Microbiol. 2004;70:129–136.
22. Muller JA, Ross RP, Sybesma WFH, Fitzgerald GF, Stanton C.
Modification of the technical properties of lactobacillus johnso-
nii NCC 533 by supplementing the growth medium with unsatu-
rated fatty acids. Appl Environ Microbiol. 2011;77:6889–6898.
23. Rozes N, Garbay S, Denayrolles M, Lonvaudfunel A. A rapid
method for the determination of bacterial fatty-acid composition.
Lett Appl Microbiol. 1993;17:126–131.
24. Belitz HD, Grosch W, Schieberle P. Changes in acyl lipids of
food. In: Grosch W, Schieberle P, eds. Food Chemistry. Vol1.
4th ed. Berlin, Heidelberg: Springer-Verlag; 2009:187–225.
25. Polacheck JW, Tropp BE, Law JH, Mccloske JA. Biosynthesis
of cyclopropane compounds. VIII conversion of oleate to dihy-
drosterculate. J Biol Chem. 1966;241:3362–3364.
26. Wang A, Cronan JE. The growth phase-dependent synthesis of
cyclopropane fatty acids in escherichia coli is the result of an
RpoS(KatF)-dependent promoter plus enzyme instability. Mol
Microbiol. 1994;11:1009–1017.
27. Wang Y, Corrieu G, Beal C. Fermentation pH and temperature
influence the cryotolerance of lactobacillus acidophilus rd758.
J Dairy Sci. 2005;88:21–29.
28. Mouritsen OG. Lipids, curvature, and nano-medicine. Eur J
Lipid Sci Tech. 2011;113:1174–1187.
29. Corcoran BM, Stanton C, Fitzgerald GF, Ross RP. Growth of pro-
biotic lactobacilli in the presence of oleic acid enhances subse-
quent survival in gastric juice. Microbiology. 2007;153:291–299.
Manuscript received Sept. 16, 2014, and revision received Mar. 21,
2015.