See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/273638476

Improving survival and storage stability of bacteria recalcitrant to freeze-

drying: a coordinated study by European culture collections

Article in Applied Microbiology and Biotechnology · March 2015

DOI: 10.1007/s00253-015-6476-6

CITATIONS READS

45 5,009

12 authors, including:

Jindrich Peiren Paul De Vos

Ghent University Ghent University
6 PUBLICATIONS 249 CITATIONS 630 PUBLICATIONS 28,792 CITATIONS

SEE PROFILE SEE PROFILE

Dominique Clermont Evélyne Bégaud

Institut Pasteur International Network Institut Pasteur International Network
107 PUBLICATIONS 3,636 CITATIONS 58 PUBLICATIONS 570 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Evélyne Bégaud on 17 July 2017.

The user has requested enhancement of the downloaded file.

Appl Microbiol Biotechnol
DOI 10.1007/s00253-015-6476-6

METHODS AND PROTOCOLS

Improving survival and storage stability of bacteria recalcitrant

to freeze-drying: a coordinated study by European culture
collections
Jindrich Peiren & Joke Buyse & Paul De Vos & Elke Lang & Dominique Clermont &
Sylviane Hamon & Evelyne Bégaud & Chantal Bizet & Javier Pascual &
María A. Ruvira & M. Carmen Macián & David R. Arahal

Received: 14 November 2014 / Revised: 9 February 2015 / Accepted: 12 February 2015

# Springer-Verlag Berlin Heidelberg 2015

Abstract The objective of this study is to improve the viabil- protocols. Furthermore, strains were exchanged between part-
ity after freeze-drying and during storage of delicate or recal- ners to perform experiments with different freeze-dryer-
citrant strains safeguarded at biological resource centers. To dependent parameters. Of all tested variables, choice of the
achieve this objective, a joint experimental strategy was lyoprotectant had the biggest impact on viability after freeze-
established among the different involved partner collections drying and during storage. For nearly all tested strains, skim
of the EMbaRC project (www.embarc.eu). Five bacterial milk as lyoprotectant resulted in lowest viability after freeze-
strains considered as recalcitrant to freeze-drying were sub- drying and storage. On the other hand, best freeze-drying and
jected to a standardized freeze-drying protocol and to seven storage conditions were strain and device dependent. For
agreed protocol variants. Viability of these strains was deter- Aeromonas salmonicida CECT 894T, best survival was ob-
mined before and after freeze-drying (within 1 week, after 6 tained when horse serum supplemented with trehalose was
and 12 months, and after accelerated storage) for each of the used as lyoprotectant, while Aliivibrio fischeri LMG 4414T
should be freeze-dried in skim milk supplemented with ma-
rine broth in a 1:1 ratio. Freeze-drying Campylobacter fetus
Electronic supplementary material The online version of this article CIP 53.96T using skim milk supplemented with trehalose as
(doi:10.1007/s00253-015-6476-6) contains supplementary material, lyoprotectant resulted in best recovery. Xanthomonas
which is available to authorized users. fragariae DSM 3587T expressed high viability after freeze-
J. Peiren (*) : P. De Vos drying and storage for all tested lyoprotectants and could not
Belgian Coordinated Collections of Microorganisms/Bacteria be considered as recalcitrant. In contrary, Flavobacterium
Collection (BCCM/LMG), Ghent University, K.L. Ledeganckstraat
35, 9000 Ghent, Belgium
columnare LMG 10406T did not survive the freeze-drying
e-mail: [email protected] process under all tested conditions.

J. Buyse : P. De Vos
Laboratory of Microbiology, Ghent University, K.L. Ledeganckstraat Keywords Freeze-drying . Bacteria . Lyoprotectant .
35, 9000 Ghent, Belgium Viability . Biological resource centers . Residual moisture
content
E. Lang
Leibniz-Institut DSMZ-Deutsche Sammlung von Mikroorganismen
und Zellkulturen GmbH (DSMZ), Inhoffenstraße 7 B,
38124 Braunschweig, Germany Introduction
D. Clermont : S. Hamon : E. Bégaud : C. Bizet
Centre de Ressources Biologiques de l’Institut Pasteur (CRBIP), For biological resource centers (BRCs) maintaining and pro-
Institut Pasteur, 75724 Paris, Cedex 15, France viding prokaryotes, freeze-drying or lyophilization is general-
ly preferred over cryopreservation as long-term preservation
J. Pascual : M. A. Ruvira : M. C. Macián : D. R. Arahal
method. Although cryopreservation of bacterial strains at
Colección Española de Cultivos Tipo (CECT) & Dpto.
Microbiología y Ecología, Universitat de València, Campus cryogenic temperatures (<−150 °C) generally results in higher
Burjassot-Paterna, Valencia, Spain survival compared to freeze-drying (Heylen et al. 2012;
 Appl Microbiol Biotechnol

Hoefman et al. 2012), a continuous supply of liquid nitrogen to map bacterial strains recalcitrant to freeze-drying among
or a permanent cooling is mandatory. A properly freeze-dried participant BRCs.
bacterial strain can be stored for 30 years and even beyond that
(personal observations at participating collections), without
any high cooling expenses during storage or transport. Materials and methods
Moreover, freeze-dried bacterial cells can immediately be
used after rehydration without any wash steps, in contrary to Biological material
cryopreservation. However, freeze-drying is a very complex
physical process affected by many parameters and variables A first objective of this study is to map all bacterial
such as growth medium, cell concentration, freezing rate, strains considered as recalcitrant to freeze-drying among
lyoprotectant, reconstitution medium, and time (Carvalho partner culture collections. Therefore, strains were listed
et al. 2003a, b, 2004; Hoefman et al. 2012; Morgan et al. based on consideration of being recalcitrant by three or
2006). Therefore, bacteria are generally more sensitive to more culture collections. The amount of recalcitrant spe-
freeze-drying compared to cryopreservation, where possible cies and strains per reported genus is represented by
damage is mainly related to freezing and thawing injuries Fig. 1a. Because of limitations in time and funding, five
(Prakash et al. 2013). Long-term experience with freeze- genera were chosen from different physiological origins.
drying of a wide variety of strains within BRCs reveals that The distribution of species of these genera is shown in
certain bacterial species are vulnerable to freeze-drying and Fig. 1b. Out of these mapped species, five representative
are often not viable after rehydration. As BRCs are reference type strains were chosen. Bacterial strains used in this
institutions for microbial diversity, fundamental aspects of study, their optimal growth media, and growth time as
freeze-drying should be well documented and experience well as their physiological origin are represented in
should be shared. Despite there are many studies on freeze- Table 1. Bacterial cells were harvested from solid agar
drying of bacteria, the focus lies mainly on the optimization of media with an inoculation loop at three different growth
industrially important micro-organisms such as probiotic times: optimal growth time (OGT), elongated growth time
strains, which are not considered recalcitrant to freeze-drying (OGT+1/3×OGT) and shortened growth time (OGT–1/3×
at BRCs (Broadbent and Lin 1999; Carvalho et al. 2002, OGT). For most effective recovery after freeze-drying,
2004; Conrad et al. 2000; de Valdez et al. 1985a, b; Miao start cell concentrations should be at least 1×108 cfu/ml
et al. 2008; Ming et al. 2009; Pehkonen et al. 2008; Saarela (Morgan et al. 2006; Palmfeldt et al. 2003). Accordingly,
et al. 2005; Schoug et al. 2006; Shao et al. 2014; Tymczyszyn after harvesting, cells were weighed on a balance (Mettler-
et al. 2007; Zavaglia et al. 2003; Zhao and Zhang 2005). Toledo) and suspended in 1 ml of the respective
Currently, the amount of bacterial strains recalcitrant to lyoprotectant (Table 2) to become a mother suspension
freeze-drying at BRCs is not well mapped. The few reported of 10 mg cells/ml: corresponding to at least 1×108 cfu/
bacterial strains recalcitrant to freeze-drying are often associ- ml. Start cell concentrations of each investigated strain
ated with slow growth or the need of microaerophilic or an- prior to freeze-drying are shown in Supplementary
aerobic growth conditions and include species from the genera Table S1 and S2. Another variable during growth was
Campylobacter (Malik and Lang 1996; Portner et al. 2007) the application of a cold shock: the mother suspension is
and Helicobacter (Spengler et al. 1992). Motile genera, pro- placed for 2 h at 7 °C. A cold shock triggers production
ducing peritrichous flagella such as Vibrio and Aeromonas of small compatible compounds (Broadbent and Lin 1999;
show as well low survival rates after freeze-drying Morgan et al. 2006; Shao et al. 2014). In this way, cells
(Miyamoto-Shinohara et al. 2008). However, a single freeze- were prepared to the upcoming aggressive dehydration.
drying protocol is used in most studies excluding the effect of
different freeze-dryer-dependent parameters such as drying Product formulation
temperature and drying pressure on survival of bacteria. The
main aim of this study is to develop an optimal freeze-drying Four different lyoprotectants were formulated in this study
protocol for five selected bacterial strains considered recalci- (Table 2): 20 % skim milk (Sigma-Aldrich), 20 % skim
trant to freeze-drying at participant BRCs, in order to enhance milk supplemented with 10 % trehalose (final concentra-
their viability after freeze-drying as well as during storage. tion, Sigma-Aldrich), 20 % skim milk + liquid growth
Variables during growth and variations in lyoprotectant are medium (1:1) and horse serum + 10 % trehalose (final
considered as well as freeze-dryer-dependent parameters. concentration, Sigma-Aldrich). Skim milk formulations
Bacterial strains used in this study belong on one hand to were autoclaved for 13 min at 115 °C, horse serum +
genera reported to be recalcitrant to freeze-drying, such as trehalose was filter sterilized. All lyoprotectants were
Campylobacter, Aeromonas, and Vibrio. On the other hand, checked for sterility, by plating a few drops on a
due to few reported studies, a preliminary aim of this work is trypticase soy agar plate, followed by incubation at
Appl Microbiol Biotechnol

Fig. 1 Mapping of bacterial strains recalcitrant to freeze-drying Distribution of species of five chosen recalcitrant genera. The number of
according to five participant biological resource centers. (a) Amount of recalcitrant strains per species is shown inside the cake part. The chosen
strains (light gray bars) and species (dark gray bars) per recalcitrant species are shown in bold
genus. Asterisk refers to chosen genus for species selection (b)

28 °C. Viable cells in mother suspensions of 10 mg cells/ (Supplementary Table S1 and S2). The suspension was
ml lyoprotectant were counted (see BViable cell counting^) aliquoted in sterile glass ampules with a filling volume
and expressed as colony-forming units per milliliter of 200 μl.

Table 1 Bacterial strains with their optimal growth medium and temperature, optimal and variable growth time, and physiological group

Strains Growth medium T (°C) OGT OGT−1/3 OGT+1/3 Physiological group

Aeromonas salmonicida ssp. salmonicida CECT 894T Nutrient agar I 26 24 h 16 h 32 h Freshwater bacteria
Aliivibrio fischeri LMG 4414T Marine agar 20 3d 2d 4d Marine bacteria
Flavobacterium columnare LMG 10406T Modified Shieh agar 25 4d 3d 5d Fish pathogen
Campylobacter fetus ssp. fetus CIP 53.96T Trypticase soy agar 37 3d 2d 4d Opportunistic human pathogen
Xanthomonas fragariae DSM 3587T R2A agar 28 22 h 15 h 29 h Plant pathogen

OGT optimal growth time, h hours, d days

 Appl Microbiol Biotechnol

Table 2 Overview of variables

for freeze-drying Test Incubation Growth medium supplemented Cold shock Lyoprotectant
condition time with 10 % trehalose (2 h at 7ºC)

V1 OGT No No Skim milk

V2 OGT Yes No Skim milk
V3 OGT No Yes Skim milk
V4 OGT−1/3 No No Skim milk
V1 is considered as a reference
method. Other variables are V5 OGT+1/3 No No Skim milk
alterations (marked in italics) of V6 OGT No No Skim milk+10 % trehalose
the reference method V7 OGT No No Skim milk+growth medium (1:1)
V variant, OGT optimal growth V8 OGT No No Horse serum+10 % trehalose
time

Freeze-drying process Storage and viability assay

During freeze-drying, three main phases can be distinguished: Heat-sealed ampules were stored in the dark at 4 °C. After
freezing phase, primary drying phase, and secondary drying 1 week and 6 and 12 months storage, three ampules of each
phase. During the freezing phase, water is converted into ice, variable (Table 2) were opened and reconstituted with 200 μl
usually entrapped in the lyoprotectant. Filled ampules were liquid growth medium. The rehydrated cells were allowed to
frozen for 2 h at −80 °C in a freezer and transferred to the equilibrate for 30 min at room temperature. Viability was de-
freeze-dryer. During primary drying, ice is removed through termined as described in BViable cell counting^. Additionally,
sublimation by lowering the pressure and adding heat to the an accelerated storage test was performed by placing heat-
product, leaving channels in the dried product, thereby creat- sealed ampules at 37 °C for 14 days, corresponding to a stor-
ing a porous structure. During secondary drying, residual un- age time of 20 years at room temperature (Sakane 1997). The
frozen water entrapped in the glassy matrix of the change in viability after freeze-drying and storage is expressed
lyoprotectant is removed by isothermal desorption. After com- as BSurvival factor,^ according to the following formula:
pletion of secondary drying, glass ampules were constricted
and placed on a manifold freeze-dryer under vacuum. After at
least 18 min on the manifold freeze-dryer (Table 3), ampules
were gas-tight sealed with a hand torch. Gas tightness of the Survival factor ð%Þ


seal was checked with a high-frequency Tesla coil spark tester LOG10 ðcfu=mlÞ after freeze‐drying; base
¼  100;
(Vander Heyden, Belgium). Storage under vacuum was per- LOG10 ðcfu=mlÞ before freeze‐drying
formed to limit any diffusion and oxidative degeneration of
bacterial cells. base ¼ 1st week; 6 months; 12 months or accelerated storage
In this study, four freeze-dryers with different parameter
settings were used to freeze-dry bacterial cell suspensions
(see Table 3). A typical freeze-drying plot, with indicated pa-
Process survival (PS) is defined as survival factor within
rameters of Table 3 is shown in Fig. 2.
first week of storage. Death rate is defined as the reduction of
viable cells during storage and is expressed as
Viable cell counting Death rate=LOG10 (cfu/ml 1st week)–LOG10 (cfu/ml ba-
se); base=6 months, 12 months or accelerated storage.
Viable cell concentrations, expressed as colony-forming All viability data are mean values of three viability counts.
units per milliliter, were determined before and after
freeze-drying, using the plate count method. Serial dilu-
tions were made ranging from 10−1 to 10−8: 200 μl Residual moisture content determination
mother suspension was transferred to 1.8 ml of fresh
growth medium to obtain a dilution of 10−1. Further Determination of residual moisture content (RMC) was
dilutions were made by transferring 500 μl of the pre- performed with a coulometric Karl-Fischer titrator (831
vious dilution to 4.5 ml growth medium. Of each dilu- KF Coulometer, Metrohm) with oven (832 KF
tion, 100 μl was transferred and spread to solid growth Thermoprep, Metrohm). The oven was used at a tem-
media in triplicate. Inoculated dishes were incubated perature of 150 °C to evaporate residual water from the
according to their optimal growth temperature and time product without decomposition. Generated water vapor
(see Table 1). was transferred to the titration vessel by a dried air
Appl Microbiol Biotechnol

Table 3 Overview of variations in the freeze-drying process parameters among European partner BRCs

Freeze-drying process parameters LP1 (BCCM/LMG) LP2 (CIP) LP3 (CECT) LP4 (DSMZ)

Glass ampules Single vial (100×7 mm) Single vial (155×7 mm) Single vial (125×9 mm) Double vial (outer 135×15 mm,
inner 43×11 mm)
Inner cotton plug Yes No Yes Yes
Freezing shelf T (ºC) −80 −80 −80 −80
Freezing place External freezer External freezer External freezer External freezer
Freezing time (hours) 2h 2h 2h 2h
Primary drying shelf T (ºC) −18 −48 −48 −20
Freeze-drying device FTS dura-top and dura-dry SMH 15 VirTis/benchtop Martin Christ alpha 1–4
6KBTEL
Primary drying pressure (mbar) 0.53 0.001 <0.13 <0.1
Primary drying time (hours) 13.5 20 16 16
Secondary drying shelf T (ºC) 20 RT RT RT
Secondary drying pressure (mbar) ≤0.013 NA <0.13 <0.1
Secondary drying time (hours) 18.5 NA 18.5 2–4
Vacuum breaking gas Nitrogen Air Air Air
Sealing freeze-dryer (manifold) Manifold freeze-dryer SMH 15 VirTis/Benchtop Martin Christ alpha 1–4
(TCPS, Belgium) 6KBTEL
Manifold drying time (hours) ≥0.5 NA 0.3 4
Manifold drying T (ºC) RT NA RT RT
Sealing vacuum (mbar) ≤0.13 NA 0.001 0.038

LP1-4 lyophilization protocols, T temperature, RT room temperature, NA not applicable

flow. The titration cell contained an anolyte solution the cake was weighed using an analytical balance
(Hydranal coulomat A for oven, Sigma-Aldrich). (Mettler-Toledo). Vials were capped with an aluminum-
Background titration drifts stayed below 10 μg water/ encapsulated rubber septum (Metrohm) and placed in to
min. Ampules were opened and the freeze-dried cake the oven of the Karl-Fischer system. End-point was de-
was transferred to a 6-ml vial (Metrohm). The mass of termined automatically when titration drift dropped

Fig. 2 Typical freeze-drying plot

at BCCM/LMG with indication
of different freeze-drying
parameters mentioned in Table 2:
1 freezing shelf temperature; 2
primary drying shelf temperature;
3 primary drying pressure; 4
secondary drying shelf
temperature; 5 secondary drying
pressure
 Appl Microbiol Biotechnol

below 10 μg water/min. The RMC is expressed as mass storage slowed down (1.52 and 1.89, respectively). Adding
h i
mass water
fraction: mass freeze‐dried  100 ð%m=mÞ. Blanks growth medium to skim milk (V7) resulted in best process
product
survival (81 %) and lowest death rate after accelerated storage
(vials filled with air) were inserted into the oven to (0.83) and was considered as the optimal condition. Cold
measure background moisture content, which was shock (V3) seemed effective for better survival of the freeze-
subtracted from sample measurements. The system was drying process compared to the reference condition, resulting
calibrated using a water standard (Hydranal water stan- in 77 % process survival and a death rate of 2.81 after accel-
dard KF oven 140−160 °C, Sigma-Aldrich). erated storage. Altering growth time resulted in higher process
survival, resp. 59 and 57 % for OGT−1/3 (V4) and OGT+1/3
(V5). Nevertheless, no recovery was observed after the accel-
Results erated storage test. Adding trehalose to the growth medium
(V2) did not result in better process survival and yielded
For freeze-drying bacterial strains, eight variables were highest death rate of all conditions (4.10).
included, which are shown in Table 2. Condition V1 is
considered as a reference condition, while V2 till V8 Flavobacterium columnare LMG 10406T
are single variants of V1: conditions V2–V5 are varia-
tions during growth (supplementation of growth medi- F. columnare was freeze-dried according to LP1 and did not
um, growth time, cold shock) while V6–V8 are varia- survive freeze-drying for all conditions tested, except for the
tions of lyoprotectant. Viability results are displayed longer incubation time OGT+1/3 (V5) (no graphical data
graphically in Fig. 3. Numeric values are given in shown). However, no recovery was seen after accelerated stor-
Supplementary Table S1. age. None of the lyoprotectants used seem to be useful.
Combinations of longer growth time (V5) with addition of
Aeromonas salmonicida CECT 894T lyoprotectant (conditions V6, V7, and V8) gave no recovery
after 1 week and after accelerated storage (data not shown).
A. salmonicida was freeze-dried according to LP3. Changing
lyoprotectant composition had the biggest effect on process Campylobacter fetus CIP 53.96T
survival and death rate during storage. Reference
lyoprotectant skim milk (V1) was not suitable since process C. fetus was freeze-dried according to LP2. Skim milk (V1) is
survival was low (55 %) and no single growth variation with not a good lyoprotectant since resulting process survival was
skim milk as lyoprotectant surpassed the accelerated storage low (54 %) and no single growth variation with skim milk as
test. However, by adding trehalose (V6), process survival lyoprotectant passed the accelerated storage test, except the
raised to 75 % and death rate after accelerated storage slowed longer incubation time (V5). However, when adding trehalose
down to 4.22. Best process survival (83 %) was achieved by to skim milk (V6), process survival reached its highest value
using horse serum supplemented with trehalose (V8). This (84 %) and death rate after accelerated storage decreased to
condition resulted also in lowest death rate after accelerated 0.49. Low death rates after accelerated storage were also ob-
storage (2.00) and could be considered as the optimal condi- served for skim milk + growth medium (V7) and horse serum
tion. Shifting up or down the growth time resulted in lower + trehalose (V8): 2.20 and 1.27, respectively. High process
process survival compared to the reference growth condition. survival (77 %) was also obtained when shortening growth
When older cells were used (OGT+1/3, V5), no recovery was time to 2 days (OGT−1/3, V4). However, death rate after
seen after 1 week storage. Adding trehalose to the growth accelerated storage was high (6.74) resulting in no recovery.
medium (V2) resulted in low process survival (24 %) and even On the other hand, extended growth (V5) resulted in an ac-
no recovery after 6 months storage. Cold shock (V3) had no ceptable process survival (69 %) and overall lowest death rate
beneficial impact on process and storage survival compared to after accelerated storage (0.00). However, absolute recovery
the reference condition. after accelerated storage is lower than condition V6. Cold
shock (V3) and adding trehalose to growth medium (V2) did
Aliivibrio fischeri LMG 4414T not clearly result in any advantage compared to the reference
condition.
A. fischeri was freeze-dried according to LP1. Skim milk (V1)
yielded lowest process survival (48 %) and a high death rate Xanthomonas fragariae DSM 3587 T
(3.81), showing no recovery after accelerated storage.
However, by adding trehalose to a protein formulation (V6, X. fragariae was freeze-dried according to LP4. In contrast to
V8), process survival raised to 73 % for skim milk and 77 % other examined strains, X. fragariae yielded very good pro-
for horse serum, respectively, and death rate after accelerated cess survival values with skim milk as lyoprotectant (V1, V2,
Appl Microbiol Biotechnol

Fig. 3 Process survival and

survival curve of examined
strains as function of different
variables
 Appl Microbiol Biotechnol

and V3). However, death rate after accelerated storage was skim milk + growth medium (resp. 1.01 and 1.87 %) com-
high compared to other lyoprotectants (V6, V7, V8) used. pared to skim milk + trehalose and horse serum + trehalose
Changing growth time affected death rate after accelerated (resp. 0.61 and 0.56 %).
storage positively (2.42 for OGT−1/3 and 1.67 for OGT+1/
3), although a loss of more viable cells during freeze-drying Influence of the freeze-drying process on viability
was also observed (96 and 89 % survival) compared to the after freeze-drying
reference condition. Addition of growth medium to skim milk
(V7) resulted in lowest death rate after accelerated storage As freeze-drying consists of a freezing, primary and secondary
(1.25), but also impaired loss of viability during freeze- drying phase, settings of temperature, time, and pressure is
drying (82 %). Skim milk supplemented with trehalose (V6) operator and device dependent. In this study, four different
and horse serum+trehalose (V8), performed very well yield- freeze-dryers were used, each with different process parame-
ing survival values of 96 and 97 %, respectively, and low ters and named as LP1 to LP4 (Table 3). As a consequence,
death rates after accelerated storage (2.74 and 2.24, respec- freezing and drying damage to bacterial cells can differ depen-
tively). Cold shock (V3) resulted in highest death rate (4.79) dent on the used freeze-drying procedure. In order to investi-
and did not seem favorable for preserving X. fragariae. gate this dependency, strains were exchanged among partner
BRCs (Table 5) to be freeze-dried using the reference condi-
tion (V1) and the best condition for each strain as shown in
Residual moisture content Supplementary Table S1. Due to import restrictions of quar-
antine organisms, X. fragariae DSM 3587T was not ex-
Residual moisture after freeze-drying is mainly dependent on changed and instead strain CECT 549T was used with LP3.
freeze-drying parameters (drying temperature, pressure, and Results are represented graphically in Fig. 5. Numeric values
time) of the secondary drying. Apart from these parameters, are shown in Supplementary Table S2.
RMC is also dependent on the type of lyoprotectant used and
even the kind of micro-organism. Therefore, for all A. salmonicida CECT 894T For all three freeze-drying proto-
lyoprotectants and bacterial strains used in this study, RMC cols used (LP1, LP3, and LP4) conditions with highest and
was determined. As predicted, the RMC of freeze-dried ma- lowest process survival could be reproduced, although abso-
terial varied with the freeze-dried strain as well as lute values are different. Lowest death rate could not be
lyoprotectant. For skim milk, lowest value is 3.51 % and reproduced by all freeze-drying protocols since LP4 resulted
highest value 5.65 % (Fig. 4, Table 4). Highest averaged in overall lowest death rate (1.48) for skim milk (V1).
RMC is observed when formulations without trehalose are However, highest process survival (92 %) and survival factor
used as lyoprotectant: 4.64 % for skim milk and 4.47 % for after accelerated storage (75 %) is retained with LP1 for horse
skim milk + growth medium. However, when adding treha- serum + trehalose (V8).
lose to skim milk, RMC is lowered to 2.19 %. Horse serum +
trehalose had also a low average RMC value after freeze- A. fischeri LMG 4414T In contrary to LP1, survival factors
drying (2.53 %). Moreover, variability of RMC values, repre- for conditions V1 and V7 with LP2 are similar. In accordance
sented by the standard deviation, are higher for skim milk and with LP1, death rate after accelerated storage is lowest for
skim milk + growth medium (V7). However, it should be
emphasized that process survival and survival factor after ac-
celerated storage for V7 is remarkably lower when LP2 (resp.
68 and 49 %) is used compared to LP1 (resp 81 and 71 %).

F. columnare LMG 10406T No reproducible process survival

values could be obtained by two freeze-drying protocols (LP1
and LP2). Death rate was reproducible since no recovery was
seen for three variables used in two freeze-drying protocols
after accelerated storage.

C. fetus CIP 53.96T For all three freeze-drying protocols used

(LP1, LP2, and LP3), conditions with highest and lowest pro-
cess survival could be reproduced except with LP1 where all
conditions yielded similar and highest process survival factor
Fig. 4 Boxplot of residual moisture content (RMC) values for different values (V1, 94 %; V5, 96 %; V6, 96 %). Lowest death rate
lyoprotectants could not be reproduced by all three freeze-drying protocols
Appl Microbiol Biotechnol

Table 4 Residual moisture content (RMC, %)

Strain V1 (skim milk) V6 (skim milk + trehalose) V7 (skim milk + growth medium) V8 (horse serum + trehalose)

A. salmonicida CECT 894T 5.31 (0.69) 3.26 (0.70) 6.12 (0.46) 2.65 (0.69)
A. fischeri LMG 4414T 3.57 (0.15) 1.68 (0.11) 2.08 (0.34) 2.93 (0.08)
F. columnare LMG 10406T 3.51 (0.17) 1.61 (0.09) 3.01 (0.47) 3.17 (0.27)
C. fetus CIP 53.96T 4.16 (0.64) 1.93 (0.15) 6.92 (0.38) 1.93 (0.16)
X. fragariae DSM 3587T 5.65 (1.02) 2.28 (0.09) 5.04 (1.40) 1.98 (0.92)
Average 4.64 (1.01) 2.19 (0.61) 4.47 (1.87) 2.53 (0.56)

Standard deviation (SD) is shown between brackets

since LP2 resulted in overall lowest death rate (no starvation) Altering growth conditions to improve survival after
for longer incubated cells (OGT+1/3, V5). However, highest freeze-drying did not seem to be the major factor influencing
absolute survival (78 %) after accelerated storage is retained cell survival. Accordingly, no beneficial effects could be ob-
with LP2 for skim milk + trehalose (V6). served when growth time was altered for A. salmonicida
(CECT 894T), A. fischeri (LMG 4414T), and X. fragariae
X. fragariae DSM 3587T, CECT 549T Best and worst condi- (DSM 3587T). Better process survival was seen for C. fetus
tions could be reproduced by two tested freeze-drying proto- (CIP 53.96 T ) for both OGT + 1/3 and OGT − 1/3 and
cols (LP3 and LP4), but absolute survival factor within first F. columnare (LMG 10406T) for OGT+1/3. However, this
week of storage and after accelerated storage were obviously beneficial effect was lost during storage in the dried state.
higher with LP4. Highest process survival and lowest death Best survival rates were obtained when cells are grown to
rate are obtained with LP4, using skim milk (V1, 109 %) and the late exponential phase. These results are in accordance
skim milk + growth medium (V7, 1.25), respectively. with the theory that older cells grown to early stationary phase
trigger starvation responses protecting them also during des-
iccation (Morgan et al. 2006).
Supplementation of growth media with sugars has been
Discussion shown to be beneficial for survival after freeze-drying for
Enterococcus faecalis, Enterococcus durans, and
Bacterial strains recalcitrant to freeze-drying were first sub- Lactobacillus delbrueckii ssp. bulgaricus (Carvalho et al.
jected to different device-independent variables of the freeze- 2003a, 2004). Trehalose has been described as a sugar with
drying process in order to enhance survival after freeze-drying very good lyoprotective properties towards membranes and
and during storage. In a second aspect, freeze-dryer- proteins in bacteria (Leslie et al. 1995). When cells have the
dependent parameters were examined to determine their role ability to take up trehalose intracellulary, critical intracellular
in improving and reproducing viability after freeze-drying and macromolecules (DNA, ribosomes,…) could possibly be
during storage. protected against freezing and drying injuries from the inside,

Table 5 Exchange of strains among partner culture collections

Strain From To Reference condition Best condition(s)

A. salmonicida CECT 894T CECT (LP3) BCCM/LMG (LP1), DSMZ (LP4) Skim milk (V1) Horse serum + trehalose (V8)
T
A. fischeri LMG 4414 BCCM/LMG (LP1) CIP (LP2) Skim milk (V1) Skim milk + growth medium (1:1) (V7)
F. columnare LMG 10406T BCCM/LMG (LP1) CIP (LP2) Skim milk (V1) 1) Longer growth time (V5)
2) Skim milk + 10 % trehalose (V6)
C. fetus CIP 53.96T CIP (LP2) BCCM/LMG (LP1), CECT (LP3) Skim milk (V1) 1) Longer growth time (V5)
2) Skim milk + 10 % trehalose (V6)
X. fragariae DSM 3587T DSMZ (LP4) CECT (LP3) Skim milk (V1) Skim milk + growth medium (1:1) (V7)
(CECT 549T)

The freeze-drying protocol used is indicated between brackets. Strains are tested on reproducibility of the reference condition (V1) and the best freeze-
drying conditions as shown in Supplementary Table S1. Due to import restrictions of quarantine organisms, strain DSM 3587T was not sent, instead
CECT 549T was used for testing LP3
 Appl Microbiol Biotechnol

Fig. 5 Survival factor after

1 week storage (= process
survival) and after accelerated
storage of exchanged strains as
function of the reference (V1)
condition and the best
condition(s) according to Table 5.
An additional variable is the
freeze-drying process

while the cell wall is protected by the lyoprotectant used in the increased in all cases. Trehalose could either not be imported
freeze-drying formulation. Moreover, addition of trehalose to intracellularly or was possibly used as carbon source, resulting
the growth medium provides a better protection against des- in a more acidic growth medium and starvation of cells during
iccation for Bradyrhizobium japonicum (Streeter 2003). The growth.
use of 1 % trehalose in trypticase soy broth for cultivation of Some studies suggest a cold shock to trigger production of
methane oxidizing bacteria was shown to be very effective for small compatible compounds before freeze-drying (Broadbent
recovery after freeze-drying (Hoefman et al. 2012). Therefore, and Lin 1999; Morgan et al. 2006; Shao et al. 2014). In this
bacteria were grown on their optimal growth medium supple- way, cells are prepared to the upcoming aggressive dehydra-
mented with trehalose to a final concentration of 10 %. This tion. In our study, cold shock was only effective for A. fischeri
study shows that for none of the strains, adding trehalose to (LMG 4414T), resulting in higher survival after freeze-drying
the growth medium had a beneficial effect on process survival and even after accelerated storage compared to the reference
and survival rate during storage. In contrary, death rate condition.
Appl Microbiol Biotechnol

This study clearly demonstrates that process and storage dried in skim milk + growth medium with LP1, which is far
survival are mainly dependent on the lyoprotectant used dur- below the average value of 4.47 %. This could explain the low
ing freeze-drying. Usage of skim milk without any additives death rate of A. fischeri (LMG 4414T) during storage when skim
was deleterious for survival during storage for nearly all milk + growth medium is used. However, this hypothesis as-
strains tested in this study, except X. fragariae (DSM sumes that trehalose formulations are possibly less sensitive to-
3587T). Particularly, trehalose had a remarkable beneficial ef- wards fluctuating RMC values caused by variations in the
fect: a mixture of 10 % trehalose-skim milk raised process and freeze-drying process parameters.
storage survival compared to skim milk alone for all strains In order to investigate the impact of freeze-dryer-dependent
included in this study, except F. columnare (LMG 10406T). parameters, strains were exchanged among partner collections
For C. fetus (CIP 53.96T) and X. fragariae (DSM 3587T), and subjected to freeze-drying with different freeze-drying
using trehalose-skim milk resulted in best storage survival. settings. In most cases, the reference condition (V1) and best
These results confirm previous research (Portner et al. condition (Table 5) could be reproduced, although absolute
2007). Moreover, using trehalose-horse serum as recovery values differ clearly among different freeze-drying
lyoprotectant resulted in best process and storage survival protocols. This indicates that freeze-dryer-dependent parame-
for A. salmonicida (CECT 894T). The general beneficial effect ters have a major influence on the survival after freeze-drying
of trehalose in this study is in accordance with other studies and during storage. More stringently controlled processes like
(Conrad et al. 2000; Costa et al. 2000; Miao et al. 2008; Ming LP1 and LP4 give better results regarding viability after
et al. 2009; Pehkonen et al. 2008; Portner et al. 2007; freeze-drying and during storage compared to less controlled
Tymczyszyn et al. 2007; Zavaglia et al. 2003; Zayed and processes like LP2 and LP3. A separate primary and second-
Roos 2004). ary drying is favorable for product integrity since collapse
Besides trehalose, addition of growth medium to skim milk events can be prevented. The collapse temperature (Tc) is
had also an advantageous effect on process and storage sur- generally defined as the maximum temperature where dried
vival for all strains except X. fragariae (DSM 3587T). products can support their own weight during primary drying.
Moreover, this condition resulted in best process and storage As a consequence, the product temperature should stay below
survival for A. fischeri (LMG 4414T). This finding is surpris- Tc during primary drying to avoid collapse. Collapsed prod-
ing because the high salt concentration in the growth media ucts lose their stable amorphous glassy state by viscous flow
will crystallize during freezing, possibly damaging the cell and are considered to have a higher moisture content, a longer
wall. The inhibitory effect of sugar mixtures present in the reconstitution time, and a possible loss of functionality
product formulation on crystallization could be an explanation (Fonseca et al. 2004). Further research on freeze-drying pro-
for our results (Morgan et al. 2006). cess parameters should reveal their importance on viability of
The residual moisture content (RMC) present in the freeze- freeze-dried strains after reconstitution.
dried material strongly affects viability or activity during storage. Considering all tested freeze-drying protocols and variables,
Several studies showed that for freeze-dried bacterial solutions, a best recovery after freeze-drying for the bacteria
high residual water content negatively affects viability during A. salmonicida (CECT 894T) and A. fischeri (LMG 4414T)
storage (de Valdez et al. 1985a; Santivarangkna et al. 2011; was obtained when LP1 was used with horse serum + trehalose
Zayed and Roos 2004). In our study, RMC is mainly dependent and skim milk + growth medium as respective lyoprotective
on the used lyoprotectant, particularly on the presence of treha- agents. C. fetus (CIP 53.96T) should be freeze-dried according
lose. Skim milk and skim milk + growth medium contained to LP2 with skim milk + trehalose as lyoprotective agent. For
highest RMC compared to trehalose formulations, although the X. fragariae (DSM 3587T), skim milk + growth medium
bulking agent of the latter can either be skim milk or horse seems to be an acceptable lyoprotectant when LP4 is used as
serum. These results show that a small concentration of trehalose freeze-drying protocol. F. columnare (LMG 10406T) did not
can alter the complete physical properties of the lyoprotectant survive the freeze-drying process for all tested conditions and
towards water, resulting in less residual water content after protocols. This result is contradictional to previously published
freeze-drying. Moreover, RMC values of lyoprotectants contain- work (Desolme and Bernardet 1996), where survival was ob-
ing trehalose show a lower variability among the freeze-dried served one week after freeze-drying when horse serum +
bacterial strains compared to lyoprotectants without trehalose. Brucella broth was used as lyoprotectant. Further effort should
The higher RMC for skim milk and skim milk + growth medium find a way to properly freeze-dry F. columnare, for instance by
could explain the high death rate of all tested bacteria during figuring out what part of the process is most critical for survival
storage after freeze-drying. The higher variability of RMC of this species (start cell concentration, growth condition,
values of skim milk and skim milk +growth medium could be freezing, drying, and reconstitution).
explained by the variations in the freeze-drying process param- A joint experimental setup between bacterial culture col-
eters between partner collections. For instance, a RMC value of lections resulted in development of successful freeze-drying
2.08 % is obtained for A. fischeri (LMG 4414T) when freeze- protocols for four out of five recalcitrant strains. In this study,
 Appl Microbiol Biotechnol

choice of the lyoprotectant had a major impact on process and de Valdez GF, de Giori GS, de Ruiz Holgado AP, Oliver G (1985b) Effect
of the rehydration medium on the recovery of freeze-dried lactic acid
storage survival. Moreover, lyoprotectants yielding high re-
bacteria. Appl Environ Microbiol 50(5):1339–1341
covery rates shortly after freeze-drying are often not suitable Desolme B, Bernardet JF (1996) Freeze-drying of Flavobacterium
for protection during long-term storage in the dried state. columnare, Flavobacterium psychrophilum and Flexibacter
Therefore, the lyoprotectant should be chosen carefully. This maritimus. Dis Aquat Org 27(1):77–80
study clearly shows that optimal freeze-drying conditions are Fonseca F, Passot S, Cunin O, Marin M (2004) Collapse temperature of
freeze-dried Lactobacillus bulgaricus suspensions and protective
strain specific: a single optimal protocol suitable for all strains media. Biotechnol Prog 20(1):229–238
in a culture collection seems therefore utopian. On the other Heylen K, Hoefman S, Vekeman B, Peiren J, De Vos P (2012)
hand, for institutions safeguarding thousands of micro-organ- Safeguarding bacterial resources promotes biotechnological innova-
isms, strain specific protocols are practically and economical- tion. Appl Microbiol Biotechnol 94(3):565–574. doi:10.1007/
s00253-011-3797-y
ly impossible. Based on the mapping of recalcitrant strains in
Hoefman S, Van Hoorde K, Boon N, Vandamme P, De Vos P, Heylen K
this study, the investigated strains are considered as represen- (2012) Survival or revival: Long-term preservation induces a revers-
tatives of their genus. Therefore, culture collections could ible viable but non-culturable state in methane-oxidizing bacteria.
adapt their preservation protocols to certain genera or groups Plos One 7(4). doi:10.1371/journal.pone.0034196
of micro-organisms. Furthermore, BRCs should optimize Leslie SB, Israeli E, Lighthart B, Crowe JH, Crowe LM (1995) Trehalose
and sucrose protect both membranes and proteins in intact bacteria
their freeze-drying protocols to evolve to better controlled during drying. Appl Environ Microbiol 61(10):3592–3597
processes, resulting in freeze-dried products of reproducible Malik KA, Lang E (1996) Successful preservation of
viability during long-term storage. Campylobacteraceae and related bacteria by liquid-drying under
anaerobic conditions. J Microbiol Methods 25(1):37–42
Acknowledgments The research leading to these results has received Miao S, Mills S, Stanton C, Fitzgerald GF, Roos Y, Ross RP (2008) Effect
funding from the European Community’s Seventh Framework Pro- of disaccharides on survival during storage of freeze dried
gramme (FP7, 2007–2013), Research Infrastructures action, under the probiotics. Dairy Sci Technol 88(1):19–30
grant agreement No. FP7-228310 (EMbaRC project). Ming LC, Rahim RA, Wan HY, Ariff AB (2009) Formulation of protec-
tive agents for improvement of Lactobacillus salivarius I 24 survival
rate subjected to freeze drying for production of live cells in
powderized form. Food Bioprocess Technol 2(4):431–436
References Miyamoto-Shinohara Y, Sukenobe J, Imaizumi T, Nakahara T (2008)
Survival of freeze-dried bacteria. J Gen Appl Microbiol 54(1):9–24
Morgan CA, Herman N, White PA, Vesey G (2006) Preservation of
Broadbent JR, Lin C (1999) Effect of heat shock or cold shock treatment micro-organisms by drying; a review. J Microbiol Methods 66(2):
on the resistance of Lactococcus lactis to freezing and lyophiliza- 183–193
tion. Cryobiology 39(1):88–102 Palmfeldt J, Radstrom P, Hahn-Hagerdal B (2003) Optimisation of initial
Carvalho AS, Silva J, Ho P, Teixeira P, Malcata FX, Gibbs P (2002) cell concentration enhances freeze-drying tolerance of
Survival of freeze-dried Lactobacillus plantarum and Pseudomonas chlororaphis. Cryobiology 47(1):21–29. doi:10.
Lactobacillus rhamnosus during storage in the presence of protec- 1016/S0011-2240(03)00065-8
tants. Biotechnol Lett 24(19):1587–1591 Pehkonen KS, Roos YH, Miao S, Ross RP, Stanton C (2008) State tran-
Carvalho AS, Silva J, Ho P, Teixeira P, Malcata FX, Gibbs P (2003a) sitions and physicochemical aspects of cryoprotection and stabiliza-
Effect of various growth media upon survival during storage of tion in freeze-drying of Lactobacillus rhamnosus GG (LGG). J Appl
freeze-dried Enterococcus faecalis and Enterococcus durans. J Microbiol 104(6):1732–1743
Appl Microbiol 94(6):947–952 Portner DC, Leuschner RGK, Murray BS (2007) Optimising the viability
Carvalho AS, Silva J, Ho P, Teixeira P, Malcata FX, Gibbs P (2003b) during storage of freeze-dried cell preparations of Campylobacter
Effects of addition of sucrose and salt, and of starvation jejuni. Cryobiology 54(3):265–270
upon thermotolerance and survival during storage of freeze- Prakash O, Nimonkar Y, Shouche YS (2013) Practice and prospects of
dried Lactobacillus delbrueckii ssp bulgaricus. J Food Sci microbial preservation. FEMS Microbiol Lett 339(1):1–9. doi:10.
68(8):2538–2541 1111/1574-6968.12034
Carvalho AS, Silva J, Ho P, Teixeira P, Malcata FX, Gibbs P (2004) Saarela M, Virkajarvi I, Alakomi HL, Mattila-Sandholm T, Vaari A,
Effects of various sugars added to growth and drying media upon Suomalainen T, Matto J (2005) Influence of fermentation time, cryo-
thermotolerance and survival throughout storage of freeze-dried protectant and neutralization of cell concentrate on freeze-drying
Lactobacillus delbrueckii ssp bulgaricus. Biotechnol Prog 20(1): survival, storage stability, and acid and bile exposure of
248–254 Bifidobacterium animalis ssp. lactis cells produced without milk-
Conrad PB, Miller DP, Cielenski PR, de Pablo JJ (2000) Stabilization and based ingredients. J Appl Microbiol 99(6):1330–1339
preservation of Lactobacillus acidophilus in saccharide matrices. Sakane T (1997) Viabilities of dried cultures of various bacteria after
Cryobiology 41(1):17–24 preservation for over 20 years and their prediction by the accelerated
Costa E, Usall J, Teixido N, Garcia N, Vinas I (2000) Effect of protective storage test. Microbiol Cult Coll 13(1):1–7
agents, rehydration media and initial cell concentration on viability Santivarangkna C, Aschenbrenner M, Kulozik U, Foerst P (2011) Role of
of Pantoea agglomerans strain CPA-2 subjected to freeze-drying. J glassy state on stabilities of freeze-dried probiotics. J Food Sci 76(8):
Appl Microbiol 89(5):793–800 R152–R156. doi:10.1111/j.1750-3841.2011.02347.x
de Valdez GF, de Giori GS, de Ruiz Holgado AP, Oliver G (1985a) Effect Schoug A, Olsson J, Carlfors J, Schnurer J, Hakansson S (2006) Freeze-
of drying medium on residual moisture content and viability of drying of Lactobacillus coryniformis Si3—effects of sucrose con-
freeze-dried lactic acid bacteria. Appl Environ Microbiol 49(2): centration, cell density, and freezing rate on cell survival and
413–415 thermophysical properties. Cryobiology 53(1):119–127
Appl Microbiol Biotechnol

Shao YY, Gao SR, Guo HL, Zhang HP (2014) Influence of culture con- Zavaglia AG, Tymczyszyn E, De Antoni G, Disalvo EA (2003) Action of
ditions and preconditioning on survival of Lactobacillus delbrueckii trehalose on the preservation of Lactobacillus delbrueckii ssp
subspecies bulgaricus ND02 during lyophilization. J Dairy Sci bulgaricus by heat and osmotic dehydration. J Appl Microbiol
97(3):1270–1280. doi:10.3168/jds. 2013-7536 95(6):1315–1320
Spengler A, Gross A, Kaltwasser H (1992) Successful freeze storage and Zayed G, Roos YH (2004) Influence of trehalose and moisture content on
lyophilization for preservation of Helicobacter pylori. J Clin Pathol survival of Lactobacillus salivarius subjected to freeze-drying and
45(8):737–737 storage. Process Biochem 39(9):1081–1086
Streeter JG (2003) Effect of trehalose on survival of Bradyrhizobium Zhao G, Zhang G (2005) Effect of protective agents, freezing tempera-
japonicum during desiccation. J Appl Microbiol 95(3):484–491. ture, rehydration media on viability of malolactic bacteria subjected
doi:10.1046/j.1365-2672.2003.02017.x to freeze-drying. J Appl Microbiol 99(2):333–338
Tymczyszyn EE, Gomez-Zavaglia A, Disalvo EA (2007) Effect of sugars
and growth media on the dehydration of Lactobacillus delbrueckii
ssp. bulgaricus. J Appl Microbiol 102(3):845–851

View publication stats