Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Influence of lyophilization, fluidized bed drying, addition of

protectants, and storage on the viability of lactic acid
bacteria
S. Strasser, M. Neureiter, M. Geppl, R. Braun and H. Danner
Institute for Environmental Biotechnology, Department for Agrobiotechnology, IFA-Tulln, University of Natural Resources and Applied Life Sciences,
Vienna, Tulln, Austria

Keywords Abstract
fluidized bed drying, industrial starter cultures,
lactic acid bacteria, lyophilization, protectants, Aims: The present study focuses on the impact of two different drying technol-
shelf life. ogies and the influence of protectants on process survival and storage stability
of the two lactic acid bacterial strains Enterococcus faecium and Lactobacillus
Correspondence plantarum.
Markus Neureiter, Institute for Environmental
Methods and Results: After incubation with the protectants glucose, sucrose,
Biotechnology, Department for
Agrobiotechnology, IFA-Tulln, University of
trehalose, and maltodextrin the concentrated bacterial suspensions were sub-
Natural Resources and Applied Life Sciences, jected to fluidized bed drying and lyophilization and subsequently stored at 4,
Vienna, Konrad Lorenz Str. 20, 3430 Tulln, 22, and 35C for half a year. Lactobacillus plantarum turned out to be more
Austria. E-mail: [email protected] sensitive to both drying methods than Ent. faecium. Without the addition of a
protectant cells of both strains suffered higher losses during fluidized bed
2008 ⁄ 0277: received 18 February 2008, drying. Elevated storage temperatures correlate with a higher decline of viable
revised 24 November 2008 and accepted
bacterial cells.
1 December 2008
Conclusions: Although survival rates varied between the strains, the nonredu-
doi:10.1111/j.1365-2672.2009.04192.x cing disaccharides revealed overall best protection for both investigated lactic
acid bacteria during processing and storage. The addition of protective carbo-
hydrates can prevent the decline in viability during fluidized bed drying.
Significance and Impact of the Study: The influence of protectants proved to
be species specific and therefore needs to be determined on a case-to-case basis.
Survival rates, duration, and energy consumption appear to be the crucial
parameters to evaluate the economy of production processes for industrial
starter cultures.

the main obstacles in the commercialization of lactic

Introduction
acid bacteria is the development of storable formulated
Due to their broad range of application lactic acid bac- products that retain the viability and the activity of the
teria are of commercial significance for the fermentation initial population.
industry. Next to fermentation and transformation of For many years lactic acid bacteria have been mainly
foods the production of lactic acid is of economic preserved by freeze drying, which is a gentle, but time-
importance. In recent years new applications of lactic consuming and expensive drying method (Ratti 2001;
acid bacteria such as probiotic dietary food and feed Morgan et al. 2006; Santivarangkna et al. 2007). Spray
supplements (Gomes and Malcata 1999; Saxelin et al. drying would allow inexpensive production of large
1999) and silage inoculants have increased the needs for amounts, as the costs of spray drying can be six times
research and production focusing on stabilization of lower per kilogram water removed than the cost of
starter cultures. All these industrial applications require freeze drying (Knorr 1998). High operating tempera-
the maintenance of high viabilities of bacterial popula- tures, however, result in poor survival of micro-organ-
tions during preservation and storage, therefore one of isms subjected to spray drying (Johnson and Etzel

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177 167
Freeze vs fluid bed drying of LAB S. Strasser et al.

1995). By contrast the optimal heat and mass transport tion processes and storage. Furthermore this study com-
as well as the equal temperature distribution in a prises a direct comparison of the drying methods
fluidized bed dryer allow gentle drying of sensitive lyophilization and fluidized bed drying in terms of
compounds. In addition the fluidized bed process con- survival of lactic acid bacteria, which is often impossible
sumes less time and energy than freeze drying (Chua due to distinct experimental setups and strains applied. In
and Chou 2003; Barbosa-Canovas and Uliano 2004) and general, literature on fluidized bed drying of lactic acid
is therefore a cost-effective alternative for preserving bacteria – especially on pilot scale as done in this study –
bioactive compounds like heat-labile micro-organisms is limited (Santivarangkna et al. 2007). The effect taken
(Bucio et al. 2005). by protective carbohydrate compounds on bacterial cells
Desiccation brings about different types of cell dam- was investigated because the preconditioning of bacterial
age, primarily due to changes in the physical state of cells prior to the drying process appears to be of particu-
membrane lipids and in the structure of sensitive lar importance.
proteins that lead to an often severe loss in bacterial
viability (Leslie et al. 1995). In many cases sugars and
Material and methods
sugar derivatives were found to be effective toward
protection of various lactic acid bacteria: due to the
Bacterial strains
presence of hydroxyl groups carbohydrates replace water
thereby stabilizing cell membrane lipids and proteins; Two lactic acid bacteria strains, Enterococcus faecium
polyols were found to prevent oxidative damage by strain IFA No.045 and Lactobacillus plantarum strain IFA
scavenging of free radicals (Smirnoff and Cumbes No. 278 (Department of Agrobiotechnology, Institute for
1989). It was previously demonstrated that the protec- Environmental Biotechnology, IFA-Tulln, Austria) were
tive effect does not depend on whether the sugars can used in this study. The bacteria were grown in de Man-
be metabolized or not and therefore a physicochemical Rogosa-Sharpe (MRS) broth (Oxoid, Basingstoke, UK) at
nature of the effect was postulated (Carvalho et al. 37C overnight and maintained as glycerol (20% v ⁄ v, JT
2002). A few reports describe the stabilizing effects of Baker, Phillipsburg, USA) stocks at )80C. For experi-
carbohydrates or their derivatives (Linders et al. ments, the bacteria were precultured and then grown with
1997a,b; Efiuvwevwere et al. 1999) and other osmopro- pH-regulation but without aeration in a 2 m3 fermenter.
tective molecules (Selmer-Olsen et al. 1999a) on survival Because of confidentiality arrangements growth
of lactic acid bacteria subjected to convective drying. conditions, which were kept constant for all production
Several studies dealt with the influence of cryoprotective processes to minimize effects on process survival, cannot
sugars that exhibited a good preservative effect on lactic be disclosed in detail, however, special effects due to the
acid bacteria during the well established freeze drying fermentation conditions are not expected since these were
method (Carcoba and Rodriguez 2000; De Giulio et al. similar to the standard conditions used for strain prepara-
2005) and during subsequent storage (Carvalho et al. tion and no measures known to enhance survival during
2002; Zarate and Nader-Macias 2006). An enhanced drying were taken. Due to the schedule of pilot scale pro-
stabilization of freeze dried Lactobacillus species could duction processes Ent. faecium cells were harvested 8 h
be achieved either by cross linking a carbohydrate com- after the end of the acidifying activity, whereas Lact. plan-
pound with salt ions (Conrad et al. 2000) or by apply- tarum cells were separated at the end of the exponential
ing a mixture of sugars with complex media like skim growth phase. Either bacterial strain was concentrated by
milk (Zayed and Roos 2004; Zarate and Nader-Macias separation (Type NA 7-06-076, Westfalia Separator AG,
2006). Further investigations revealed that the survival Oelde, Germany). The resulting bacterial solutions still
of freeze dried Lactobacillus bulgaricus during storage is contained growth medium components that passed the
dependent on the drying medium and the growth separation step.
medium (Carvalho et al. 2004a). A study on mutant
Lactobacillus acidophilus revealed that the internalization
Incubation with protective carbohydrates
of the compatible solute trehalose plays an important
role in cryoprotection (Duong et al. 2006). The carbohydrates and the applied concentrations used
In order to meet the requirements of product quality in this study were chosen because of their reported pro-
in terms of long-term stability and viability of lactic acid tective capacity during freeze drying of lactic acid bacte-
bacteria starter cultures as well as economic feasibility, ria (De Giulio et al. 2005). Thirty-two percent w ⁄ v of
the focus of the present study is on the stabilization of the following protective carbohydrates: glucose (Avebe,
two strains of lactic acid bacteria, Enterococcus faecium Veendam, The Netherlands), sucrose (Agrana, Tulln,
and Lactobacillus plantarum, during pilot scale preserva- Austria), trehalose (Cargill, Mechelen, Belgium), or

ª 2009 The Authors

168 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177
S. Strasser et al. Freeze vs fluid bed drying of LAB

maltodextrin (DE 19, Syral, Marckolsheim, France) was product temperature was rising from 0 to 30C. Finally
directly suspended in the concentrated bacterial solu- 17 h after the generation of vacuum it was raised to
tions. Per trial 600 ml of concentrated bacterial solution 0Æ1 Pa. Drying took place below glass transition tempera-
and 192 g of protectant were used for the freeze drying ture being glucose the protectant having the lowest glass
experiments, while 2 l of concentrate mixed with 640 g transition temperature (31C) of all protectants applied
of protectant were applied to fluidized bed drying trials, in this study (Bhandari and Howes 1999; Chen et al.
which were carried out in duplicates, consecutively and 2000).
randomized. For control purposes, a blank trial without
addition of any compound was carried along with each
Storage and rehydration conditions
set of trials. An incubation time of 1 h was chosen as
on one hand being representative for industrial applica- One gram samples of each experiment were weighed into
tions (Conrad et al. 2000) and on the other hand to a 15-ml-plastic tube immediately after drying, retaining
allow for equilibration between the cells and the added residual oxygen and humidity of the packaging environ-
protectant (Carvalho et al. 2002, 2004a). After incuba- ment. As atmosphere, relative humidity (Castro et al.
tion at room temperature the solutions were subjected 1995), and exposure to light appear to be important
to either drying method. Nonprocessed bacterial suspen- factors for the recovery of freeze dried cells, tubes con-
sion was stored at 4C for maximum 30 h before taining samples that were to be analysed at the same
drying. point of time were hermetically sealed in aluminium
sachets and exposed to darkness. Since previous studies
(Champagne et al. 1996; Wang et al. 2004) showed that
Fluidized bed bottom-spray drying
temperature is a critical parameter for microbial survival
The bacterial solution was atomized by a two-fluid nozzle during storage, samples of each experiment were stored at
in bottom-spray position using a peristaltic pump 4, 22, and 35C. The rehydration conditions were held
(Watson Marlow, Falmouth, UK) applying a spraying air constant for all samples as temperature, volume, and the
pressure of 250 kPa. Each trial was sprayed onto 5 kg of rehydration solution itself may significantly affect the rate
powdered cellulose carrier material Arbocel G350 (JRS, of recovery to the viable state, and thus influence survival
Rosenberg, Germany) previously loaded into a pilot scale rates (Selmer-Olsen et al. 1999a).
fluid bed dryer (AMMAG, Gunskirchen, Austria) of a
maximum loading capacity of 50 l depending upon pow-
Determination of CFU
der density. Inlet air temperature was set to 45C which
resulted in a maximal bed temperature of 37C depending CFU were determined by the pour plate method in MRS
on the spraying rate. The spraying rate ranging between agar (Oxoid). The CFU assay was carried out in
50 and 100 g min)1 was adjusted so that agglomeration triplicates and after 48 h of incubation at 35C the plates
of particles was avoided. The fluidizing air flow was held containing 20–200 CFU were taken to determine the
constant at 800 m3 h)1. Drying was conducted at a bed relative bacterial viability by the following equation [%]:

Population of the powder (CFU g1 Þ  amount of powder (g)

 100
Population of the suspension (CFU ml1 Þ  amount of solution (ml)

temperature of 37C; finally the powder was cooled to The resulting bacterial concentrates after separation
room temperature. contained 1Æ9 · 1011 and 4Æ2 · 1010 CFU ml)1 of Ent. fae-
cium and Lact. plantarum, respectively. To observe the
shelf life of the different formulations, cell viability was
Lyophilization
determined over a period of half a year after drying. The
The incubated suspensions were filled into 30 · 40 cm results are given as means ± SD. Comparisons between
plastic bags, put flattened in a deep freeze and frozen to means of relative viabilities of either dried bacterial
)80C. The frozen mixtures were then desiccated under species and at certain storage temperatures were
vacuum (31 Pa) in the freeze dryer (Martin Christ, Oste- performed by one-way analysis of variance (anova), and
rode, Germany) for 2 h while the product temperature by Tukey’s multiple comparison test. P-values of 0Æ05 or
was rising from )30 to 0C and further 9 h while the less were considered significant.

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177 169
Freeze vs fluid bed drying of LAB S. Strasser et al.

addition of trehalose and sucrose showing relative viabili-

Determination of dry matter content
ties of 36Æ9 ± 2Æ8 and 36Æ4 ± 6Æ8%, respectively. Viabilities
Dry matter contents were determined with an infrared of cells formulated with glucose and maltodextrin were
balance (Sartorius MA30, Goettingen, Germany). A cer- both significantly lower (17Æ5 ± 2Æ9 and 14Æ2 ± 6Æ8%,
tain amount (about 2 g) of sample is dried to equilibrium respectively), but compared to the untreated sample still
moisture and then the dry matter content is automatically higher by a factor of 100 (Fig. 1b). Without the addition
calculated. of protectants the fluidized bed process was recognized to
cause more damage to the survival rates of Lact. planta-
rum cells than lyophilization, resulting in relative viabili-
Results
ties of 0Æ2 ± 0Æ2 compared to 16Æ3 ± 1Æ1%, respectively.
No significant differences were observed between the
Effect of protective carbohydrates on drying methods
freeze dried and the fluidized bed dried formulations con-
In this study the monosaccharide glucose (Glc), the two taining the protectants.
disaccharides trehalose (Tre) and sucrose (Suc), and the
polysaccharide maltodextrin (MD) were tested with
Effect of protective carbohydrates on storage stability of
respect to their ability to protect bacterial cells against
fluidized bed dried Ent. faecium
drying. After either lyophilization or fluidized bed drying,
cell viabilities of Ent. faecium in the absence of the pro- The shelf life of the different carbohydrate formulations
tective agents were significantly lower (39Æ7 ± 2Æ2 and was investigated at refrigerated storage (4C) and at the
11Æ0 ± 6Æ4%, respectively) than with the addition of any temperatures of 22 and 35C for the time interval of half
compounds tested (Fig. 1a). In general, survival of cells a year. At 4C storage temperature no significant loss of
was higher after lyophilization than after fluidized bed initial viability of fluidized bed dried Ent. faecium cells
drying being statistically significant (P < 0Æ05) for the without any protectant and formulated with the four pro-
untreated freeze and fluidized bed dried cells, and the tectants was observed throughout the whole period of
protectants glucose. All compounds tested were found to investigation (Fig. 2a, Table 1). At an elevated storage
be equally effective in protecting Ent. faecium during temperature of 22C no significant loss of initial viability
freeze drying (>83Æ4%) and fluidized bed drying was demonstrated by the trehalose and sucrose formula-
(>57Æ3%) compared to the untreated sample (Fig. 1a). tions showing relative viabilities of 57Æ9 ± 6Æ8 and
The highest viability of Lact. plantarum cells after freeze 56Æ0 ± 6Æ6%, respectively, after 6 months (Fig. 2b,
drying was obtained by the addition of trehalose Table 1). Formulations of glucose and maltodextrin
(40Æ1 ± 2Æ7%) followed by sucrose (34Æ1 ± 1Æ8%) as cryo- decreased in viability after 6 months of storage declining
protectants (Fig. 1b). Formulating cells with glucose and to 36Æ9 ± 6Æ6 and 42Æ7 ± 5Æ4% viability. Without any pro-
maltodextrin before lyophilization did not enhance the tective compound viability was considerably lower but no
survival rates (10Æ1 ± 1Æ3% and 23Æ0 ± 2Æ9%, respectively) statistically significant changes were detected throughout
significantly compared to the untreated sample 6 months of storage. Glucose was found to be ineffective
(16Æ3 ± 1Æ1%). The best results protecting Lact. plantarum in the protection of Ent. faecium at a storage temperature
cells during fluidized bed drying were obtained by the of 35C as viabilities dropped to <1% after 1 month,

(a) (b)
d d 100
100 d
Relative viability (%)

d c,d
Relative viability (%)

c,d
80 c,d 80
b,c
60 60
b d d
40 40 d d
c
a b,c b,c b,c
20 20 b
a
0 0
w/o Glc Tre Suc MD w/o Glc Tre Suc MD
Protectant Protectant

Figure 1 Observed relative viabilities of (a) Enterococcus faecium and (b) Lactobacillus plantarum cells in the absence (w ⁄ o) and in the presence
of the protectants glucose (Glc), trehalose (Tre), sucrose (Suc), and maltodextrin (MD) at a concentration of 32% (w ⁄ v) either after freeze drying
(FD or fluidized bed drying (FBD ). Different letters within one figure indicate that means are significantly different (P < 0Æ05).

ª 2009 The Authors

170 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177
S. Strasser et al. Freeze vs fluid bed drying of LAB

(a) 100 (d) 100

Relative viability (%)

Relative viability (%)

80 80

60 60

40 40

20 20

0 0
w/o Glc Tre Suc MD w/o Glc Tre Suc MD
Protectant Protectant

(b) 100 (e) 100

Relative viability (%)

Relative viability (%)

80 80

60 60

40 40

20 20

0 0
w/o Glc Tre Suc MD w/o Glc Tre Suc MD
Protectant Protectant

(c) 100 (f) 100

Relative viability (%)

Relative viability (%)

80 80

60 60

40 40

20 20

0 0
w/o Glc Tre Suc MD w/o Glc Tre Suc MD
Protectant Protectant

Figure 2 Relative viabilities of fluidized bed dried (a, b, c) and freeze dried (d, e, f) Enterococcus faecium cells in the absence (w ⁄ o) and in the
presence of the protectants glucose (Glc), trehalose (Tre), sucrose (Suc), and maltodextrin (MD) after 0 ( ), 1 ( ), 3 ( ), 4 ( ), and 6 (h) months
of storage at different temperatures. (a, d) Cells stored at 4C. (b, e) Storage at 22C. (c, f) Storage at 35C.

while without any protectant this level was reached after (Fig. 2d, Table 1). After 6 months and with 41Æ1 ± 1Æ9%
6 months of storage (Fig. 2c, Table 1). Trehalose and relative viability glucose formulated cells did not differ
sucrose proved to be the best protectants at 35C storage any longer from untreated cells, whose viabilities did not
temperature losing viability significantly after 3 months decrease significantly during the same period of time
and leading to final viabilities of 35Æ8 ± 2Æ5 and showing a viability of 22Æ8 ± 1Æ8% after half a year.
39Æ3 ± 5Æ5%, respectively, after 6 months (Fig. 2c). Using Furthermore nontreated freeze dried Ent. faecium cells
maltodextrin as protective agent caused a significant loss exhibited a better overall performance than nontreated
in viability already after 1 month, but still a relative fluidized bed dried cells (Fig. 2a,d). A decline of viability
viability of 31Æ0 ± 5Æ2% could be preserved after after the first month of storage at 22C could be observed
6 months of storage at 35C. in all formulated freeze dried cells. While trehalose and
sucrose were able to stabilize cells from the first month,
glucose and maltodextrin formulations did not further
Effect of protective carbohydrates on storage stability of
protect cells during storage as no significant difference to
freeze dried Ent. faecium
untreated cells could be revealed after 6 months (Fig. 2e,
The long-term stability of freeze dried samples formulated Table 1). Freeze dried cells without being protected by
with trehalose and stored at 4C showed no significant any compound suffered damage after the fourth month
decrease in viability after half a year of storage of storage, from this point in time no difference to non-
(70Æ3 ± 12Æ3%), while neither sucrose nor maltodextrin treated fluidized bed dried Ent. faecium cells could be
could prevent a drop during storage ending up with rela- observed (Fig. 2b,e). A constant decrease of viability was
tive viabilities of 61Æ3 ± 2Æ4 and 56Æ8 ± 4Æ1%, respectively the trend for sucrose, trehalose and maltodextrin treated

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177 171
Freeze vs fluid bed drying of LAB S. Strasser et al.

Table 1 Average CFU per g of the resulting freeze dried (FD) and fluidized bed dried (FBD) powders of all formulations of both Enterococcus
faecium and Lactobacillus plantarum at the beginning (FD ⁄ FBD) and after 6 months of storage at different temperatures (4, 22, and 35C)

Ent. faecium

(CFU g)1) FD 4C 22C 35C FBD 4C 22C 35C

w⁄o 3Æ60E+11 2Æ06E+11 1Æ47E+11 5Æ23E+09 1Æ00E+10 4Æ45E+09 2Æ51E+09 1Æ05E+08

Glc 4Æ00E+11 1Æ97E+11 1Æ65E+11 1Æ36E+06 4Æ93E+10 3Æ87E+10 3Æ18E+10 5Æ07E+07
Tre 4Æ57E+11 3Æ67E+11 3Æ43E+11 1Æ12E+11 3Æ92E+10 3Æ65E+10 3Æ20E+10 1Æ99E+10
Suc 4Æ30E+11 2Æ97E+11 2Æ60E+11 2Æ37E+11 5Æ17E+10 4Æ43E+10 4Æ03E+10 2Æ85E+10
MD 4Æ33E+11 2Æ87E+11 1Æ60E+11 6Æ77E+10 5Æ82E+10 4Æ42E+10 3Æ52E+10 2Æ55E+10
Lact. plantarum

(CFUg)1) FD 4C 22C FBD 4C 22C

w⁄o 4Æ60E+10 5Æ83E+09 1Æ81E+09 3Æ43E+07 – –

Glc 1Æ38E+10 8Æ37E+09 4Æ17E+09 2Æ88E+09 1Æ35E+09 2Æ17E+08
Tre 4Æ57E+10 3Æ53E+10 1Æ80E+10 5Æ87E+09 1Æ62E+09 4Æ30E+08
Suc 3Æ70E+10 3Æ43E+10 1Æ08E+10 6Æ02E+09 2Æ27E+09 5Æ65E+08
MD 2Æ57E+10 5Æ50E+09 5Æ43E+08 2Æ13E+09 3Æ02E+08 4Æ55E+07

w ⁄ o: without protectant, Glc: glucose, Tre: trehalose, Suc: sucrose, MD: maltodextrin.

cells stored at 35C, nevertheless sucrose turned out to be rides showed the same trend as at 4C, not being signifi-
capable of stabilizing Ent. faecium at this temperature cantly more efficient in the protection of Lact. plantarum
significantly better than trehalose and maltodextrin, end- compared to glucose and even maltodextrin formulations
ing at a final viability of 48Æ9 ± 4Æ3% after 6 months of after the third month of storage. No compound among
storage (Fig. 2f, Table 1). As already observed in fluidized the carbohydrates tested turned out to be effective in the
bed dried samples glucose was not effective in protecting protection of Lact. plantarum cells at a storage tempera-
Ent. faecium at this elevated temperature (1Æ5 ± 0Æ6% ture of 35C dropping on average more than 1 log cycle
after 1 month) being sevenfold worse than without the after each month of storage (data not shown).
addition of a protective agent.
Effect of protective carbohydrates on storage stability of
Effect of protective carbohydrates on storage stability of freeze dried Lact. plantarum
fluidized bed dried Lact. plantarum
At 4C sucrose mediated best storage stability for Lact.
As less than 0Æ2 ± 0Æ2% of the initial Lact. plantarum plantarum cells without significant losses in viability over
population survived the fluidized bed drying process the entire period of investigation and a final viability of
without any protection by carbohydrates, this approach 31Æ6 ± 4Æ7% (Fig. 3c, Table 1). In contrast, trehalose for-
was not subjected to the storage trials. At 4C storage mulated cells suffered a significant decline in viability
temperature the best results in long-term stability were after the fourth month of storage, nevertheless single
initially obtained when cells were formulated with the monthly viabilities did not differ from sucrose treated
two disaccharides, nevertheless after the third month of cells. Furthermore glucose and maltodextrin formulations
storage no significant difference in stabilization compared were already reported not to promote process survival
to glucose formulated cells could be observed. The three during freeze drying of Lact. plantarum (Fig. 1b) nor did
formulations secured viabilities in the range between they exhibit a stabilizing effect during storage compared
8Æ2 ± 2Æ6 (glucose), 10Æ2 ± 0Æ9 (trehalose), and to nontreated cells at the storage temperatures of 4 and
13Æ8 ± 2Æ0% (sucrose) after 6 months of storage (Fig. 3a, 22C. Cells without being protected retained viabilities of
Table 1). Maltodextrin displayed significantly lower 2Æ1 ± 0Æ2% and 0Æ6 ± 0Æ04% at 4 and 22C, respectively,
relative viabilities than the disaccharides over the whole after 6 months of storage (data not shown). After the
period of investigation (except for trehalose after third month at 22C sucrose could not prevent an earlier
6 months) ending with 2Æ0 ± 1Æ3% residual viability. At decline of viability than trehalose, while no significant dif-
22C storage temperature all formulations revealed a ference in stabilization of freeze dried Lact. plantarum
decline of more than a half of the relative viability, in the during storage between the two disaccharides could be
first place after 1 month and secondly from the first to observed from the fourth month (Fig. 3d). Despite being
the third month of storage (Fig. 3b). The two disaccha- reduced to more than a half and more than a third,

ª 2009 The Authors

172 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177
S. Strasser et al. Freeze vs fluid bed drying of LAB

(a) (c)
50 50

Relative viability (%)

Relative viability (%)
40 40

30 30

20 20

10 10

0 0
Glc Tre Suc MD Glc Tre Suc MD
Protectant Protectant

(b) (d)
50 50

Relative viability (%)

Relative viability (%)

40 40

30 30

20 20

10 10

0 0
Glc Tre Suc MD Glc Tre Suc MD
Protectant Protectant

Figure 3 Relative viabilities of fluidized bed dried (a, b) and freeze dried (c, d) Lactobacillus plantarum cells in the absence (w ⁄ o) and in the pres-
ence of the protectants glucose (Glc), trehalose (Tre), sucrose (Suc), and maltodextrin (MD) after 0 ( ), 1 ( ), 3 ( ), 4 ( ), and 6 (h) months of
storage at different temperatures. (a, c) Cells stored at 4C. (b, d) Storage at 22C.

respectively, the two disaccharides trehalose and sucrose Table 2 Average residual moisture contents of the resulting formula-
maintained highest viabilities of 15Æ8 ± 0Æ7% and tions of both Enterococcus faecium and Lactobacillus plantarum after
9Æ9 ± 0Æ7%, respectively, among all compounds tested. freeze drying (FD) and fluidized bed drying (FBD)
With a focus on trehalose and sucrose formulations freeze
Residual moisture
dried cells were overall more stable than fluidized bed content (%) Ent. faecium Lact. plantarum
dried cells at both storage temperatures 4 and 22C. Simi-
larly to fluidized bed dried cells storage at 35C led to Protectant FBD FD FBD FD
extensive losses in viability and neither of the compounds Without (w ⁄ o) 3Æ9 3Æ3 3Æ8 4Æ1
used could prevent those (data not shown). Glucose (Glu) 6Æ1 2Æ4 5Æ2 2Æ8
Trehalose (Tre) 5Æ7 8Æ0 4Æ9 4Æ3
Sucrose (Suc) 4Æ5 3Æ1 4Æ6 4Æ7
Dry matter contents of bacterial concentrates Maltodextrin (MD) 4Æ7 2Æ5 4Æ0 1Æ9
and resulting powders
The concentrating step yielded an average dry mass of tively, than glucose formulated fluidized bed dried cells.
17Æ8 and 7Æ3% for the bacterial solutions of Ent. faecium The disaccharide formulations of Lact. plantarum did not
and Lact. plantarum, respectively. In the case of Ent. fae- differ in particular, while trehalose formulated Ent. fae-
cium a ratio between protectant and cell mass of 1Æ8 was cium showed higher residual moisture contents than
achieved, while the lower cell density of Lact. plantarum sucrose formulated Ent. faecium cells.
caused a ratio of 4Æ4.
Residual moisture contents of fluidized bed dried sam-
Discussion
ples of both strains lay in the range between 3Æ8% and
6Æ1%, in contrast freeze dried samples varied to a higher This study provides a comparison of lyophilization and
extent from 1Æ9% to 8Æ0% (Table 2). Cells of both strains fluidized bed drying in terms of survival of two lactic acid
without any protectant added resulted in dried products bacteria strains. Both industrial drying processes are
of a narrow moisture range (3Æ3–4Æ1%). Maltodextrin for- employed for preserving sensitive bioactive compounds in
mulated cells were mostly amongst the driest products order to achieve the preservation of high viability during
while only glucose formulated freeze dried cells showed the drying process and subsequent storage of bacterial
smaller residual moisture contents of 2Æ4 and 2Æ8, respec- starter cultures. Several previous surveys focused on the

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177 173
Freeze vs fluid bed drying of LAB S. Strasser et al.

survival of lactic acid bacteria after lyophilization and encountered during spray drying (Corcoran et al. 2004)
subsequent storage (Carvalho et al. 2004b), while previ- can be minimized by applying gentle process conditions,
ously published data related to fluidized bed drying of thus dehydration inactivation remains the main problem
lactic acid bacteria often refers to drying of cells granu- in convective drying (Linders et al. 1997a). Furthermore
lated to pellets (Linders et al. 1997c), homogenized into the consequences of oxidation are very harmful especially
powders (Mille et al. 2004) or entrapped in alginate beads to the membrane structure, thus oxygen-sensitive bacte-
(Selmer-Olsen et al. 1999b). In all the mentioned surveys rial species suffer severe damage during convective air
bacteria are encapsulated in a certain matrix before being drying (Mille et al. 2004).
dried in contrast to spraying concentrated liquid cell sus- According to the literature mentioned above it can be
pensions onto fluidized carrier material as described in summarized that in addition to dehydration inactivation
this study. Thus the technique applied is spray drying but microbial cells are mainly affected by ice crystal formation
in contrast to conventional spray drying lower tempera- occurring during freeze drying or by oxidative stress
tures can be applied in the fluidized bed preventing encountered during fluidized bed drying. Our results cor-
damage by heat. relate to a certain extent with those previous findings,
Desiccation imposes severe stress on micro-organisms nevertheless apart from the action of protectants there are
as the removal of water induces conformational changes several more factors being responsible for the poorer sur-
in proteins and cell membranes. It is apparent that induc- vival of fluidized bed dried cells that need to be consi-
tion of desiccation tolerance takes place under certain dered in further studies. In the present experiments only
culture and predrying conditions. The exact mode of one established processing condition per drying method
action of protective agents is indeed complex, and not was investigated, therefore variation of the processing
fully understood to date being water replacement the parameters is likely to yield different results and should
essential mechanism underlying the protection mediated be used for further optimization.
by certain protectants (Potts 2001). Fluidized bed dried The results of the current study point out that the
Lact. plantarum cells turned out to be particularly suscep- damage caused by either drying method may be reduced
tible to desiccation induced damage resulting in a residual by incubating bacterial cells with certain protective carbo-
viability of <0Æ2%, thus highlighting the need for formu- hydrates as their presence during fluidized bed and freeze
lations containing protective compounds that enhance drying resulted at least in equal, in most cases in higher
process survival. Due to bad performance of the polyol survival rates than those of nontreated cells. Especially
sorbitol at a concentration of 32% (w ⁄ v) in the fluidized trehalose was reported to be capable of stabilizing biologi-
bed drying process and incomplete drying during lyophil- cal material like cells and proteins (Schiraldi et al. 2002).
ization (data not shown), no further investigations were Several physical principles underlying the mechanisms of
carried out with this compound despite positive reports stabilization of cells and proteins in the dry state were
(Carvalho et al. 2002). In contrast to previous reports mentioned: the prevention of fusion and phase transition
(Selmer-Olsen et al. 1999a; Carcoba and Rodriguez 2000) of phospholipid bilayers, direct interaction by hydrogen
results obtained with the addition of skim milk at 10% bonding of OH-groups to polar residues and furthermore
(w ⁄ v) and skim milk in combination with trehalose and vitrification forming glassy amorphous solids (Crowe
sorbitol were not significantly better than the above et al. 1992). Previous observations reported the ability of
mentioned protectants (data not shown) and therefore trehalose and sucrose to stabilize cell membranes by pre-
not considered in shelf life studies. The poor solubility of venting the phospholipids’ phase transition during drying
lactose triggered clogging of the spray nozzle in the fluid- and rehydration which in turn would be the cause of
ized bed drying process and was excluded in this study. membrane leakage (Leslie et al. 1995; Hoekstra et al.
Process survival rates after both drying methods 1997). Despite obvious variations in residual moisture
revealed that without any protective compound added contents each of the four carbohydrate formulations
fluidized bed drying is more harmful to both bacterial showed similar protective effects on Ent. faecium both
strains than freeze drying. Similarly, previous observations during freeze drying and during fluidized bed drying. In
comparing freeze and spray drying showed that higher contrast, the two disaccharides exhibited the best perfor-
survival rates of lactic acid bacteria were achieved by the mance during fluidized bed drying of Lact. plantarum and
former drying method (To and Etzel 1997; Wang et al. furthermore they were the only compounds capable of
2004; Corcoran et al. 2006). Another comparative study enhancing survival during lyophilization of Lact. planta-
on this topic revealed contrary results, proving that sus- rum. Appropriate water binding capacities may explain
ceptibility to certain drying methods is strongly depen- why the two disaccharides cause residual moisture con-
dent on the strain under evaluation (Zamora et al. 2006). tents that are particularly favourable for survival of Lact.
Focusing on fluidized bed drying thermal inactivation plantarum. The low residual moisture contents of glucose

ª 2009 The Authors

174 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177
S. Strasser et al. Freeze vs fluid bed drying of LAB

and maltodextrin formulated freeze dried Lact. plantarum Ent. faecium while it should be recommended for Lact.
indicate the removal of structural water causing damage plantarum starter cultures. As sucrose is more cost-effective
of important cell molecules. A certain amount of water than trehalose, conventional sugar should be favoured if it
must remain in the dehydrated state and the optimum exerts the same protective effects as trehalose.
moisture content needs to be determined for a satisfac- Several facts explain why Lact. plantarum turned out to
tory survival rate of the strain of investigation, e.g. for be more sensitive to both drying methods and subsequent
freeze dried Lactobacillus salivarius the optimum moisture storage than Ent. faecium. First, previous observations
content was reported to range from about 2Æ8% to 5Æ6% showed that Enterococci (i.e. small spherical cells) are
(Zayed and Roos 2004). apparently more resistant to freezing and freeze drying
Among the carbohydrates studied trehalose and sucrose than rod-shaped Lactobacilli (Fonseca et al. 2000, 2004),
furthermore proved to mediate equal if not the longest last- which are more susceptible to membrane damage due to
ing protection during storage of both lactic acid bacterial their higher surface area. Secondly, Lact. plantarum cells
strains and regardless of the drying method. While all com- were harvested earlier than Ent. faecium, namely on the
pounds tested stabilized fluidized bed dried Ent. faecium at onset of stationary phase, which could impact negatively
4C throughout the whole period of investigation, only the on resistance of Lact. plantarum to desiccation. Optimal
disaccharides secured initial viabilities from significant survival of spray drying was reported of Lactobacillus cul-
losses at an elevated storage temperature of 22C. As tures in the stationary phase of growth (Corcoran et al.
observed previously the stability of dried samples decreased 2004), which additionally proved to be more resistant to
during storage showing higher losses in viability at elevated stress conditions (Kim et al. 2001). Finally, the 4Æ5-fold
storage temperatures (Gardiner et al. 2000; Corcoran et al. higher initial cell concentration could be made responsi-
2004) which could correspond to a higher rate of fatty acid ble for the better performance of Ent. faecium in both
oxidation (Castro et al. 1995). In particular glucose must drying methods. The chosen experimental setup, however,
not be used in the drying medium at a storage temperature does not allow a direct comparison between the two
of 35C as its glass transition temperature is exceeded strains. Optimization of all process steps is expected to
(Bhandari and Howes 1999) exerting negative effects on result in improved survival rates for Lact. plantarum.
storage stability. Storage at 35C led to significant losses of To sum up, further measures to enhance bacterial
bacterial viability after 1 month, whereas the addition of survival during drying processes need to be taken: Among
the disaccharides resulted in a longer-lasting stabilization those are optimizing culture conditions (Saxelin et al.
in comparison to the other two protectants tested. The 1999; Palmfeldt and Hahn-Hagerdal 2000) and cell density
residual moisture content was reported to be directly (Morgan et al. 2006), and testing of different protective
related to the type of drying medium (Zayed and Roos compounds, concentrations and times of incubation
2004). Since the residual moisture content of the final thereof to strengthen bacterial desiccation tolerance. By
product may have a considerable impact on the survival exhibiting stress conditions (pH, temperature, osmotic
rate after drying and during storage, it is conceivable that pressure etc) adaptive response can be provoked resulting
the protective properties found in the present study are at in more desiccation-resistant bacterial cells (Abee and
least partially due to indirect effects, caused by the fact that Wouters 1999; Baati et al. 2000; Conrad et al. 2000; Kim
the addition of different types of carbohydrates results in et al. 2001; Saarela et al. 2004; Morgan et al. 2006).
different moisture contents after the drying process. In Response surface methodologies were reported to be par-
addition to the influence of certain protectants on storage ticularly suitable to detect interactions and co-dependen-
stability the optimum moisture content of the individual cies between the individual parameters mentioned
strain therefore needs to be determined in further studies. (Schoug et al. 2006). Furthermore the findings underline
In general, dried Lact. plantarum showed poorer survival the need of strain selection with regard to the performance
during storage after fluidized bed drying than after freeze during fermentation and downstream processing. In addi-
drying, which is comparable to previous findings revealing tion the fact that fluidized bed drying produces less than a
higher survival of freeze dried than of spray dried Lactoba- fifth of the fixed and manufacturing costs estimated for
cillus sakei (Ferreira et al. 2005) and Streptococcus thermo- freeze drying (Santivarangkna et al. 2007) needs to be con-
philus (Wang et al. 2004). Interestingly, after 3 months of sidered when choosing a drying method on a large scale.
storage at 4 and 22C both disaccharide formulations of
fluidized bed dried Lact. plantarum lost in viability to such
Acknowledgements
an extent that no difference to other formulations could be
observed. With respect to a shelf life of half a year and The authors wish to acknowledge ecoplus, the business
depending on product requirements the more costly refrig- agency of Lower Austria, for the financial support to
erated storage could be avoided for optimally formulated conduct this research.

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177 175
Freeze vs fluid bed drying of LAB S. Strasser et al.

Corcoran, B.M., Ross, R.P., Fitzgerald, G.F. and Stanton, C.

References
(2004) Comparative survival of probiotic Lactobacilli
Abee, T. and Wouters, J.A. (1999) Microbial stress response in spray-dried in the presence of prebiotic substances. J Appl
minimal processing. Int J Food Microbiol 50, 65–91. Microbiol 96, 1024–1039.
Baati, L., Fabre-Gea, C., Auriol, D. and Blanc, P.J. (2000) Corcoran, B.M., Ross, R.P., Fitzgerald, G.F., Dockery, P. and
Study of the cryotolerance of Lactobacillus acidophilus: Stanton, C. (2006) Enhanced survival of GroESL-overpro-
effect of culture and freezing conditions on the viability ducing Lactobacillus paracasei NFBC 338 under stressful
and cellular protein levels. Int J Food Microbiol 59, conditions induced by drying. Appl Environ Microbiol 72,
241–247. 5104–5107.
Barbosa-Canovas, G.V. and Uliano, P.J. (2004) Adaptation of Crowe, J.H., Hoekstra, F.A. and Crowe, L.M. (1992) Anhydro-
classical processes to new technical developments and biosis Annu Rev Physiol 54, 579–599.
quality requirements. J Food Sci 69, E240–E250. De Giulio, B., Orlando, P., Barba, G., Coppola, R., Rosa,
Bhandari, B.R. and Howes, T. (1999) Implication of glass tran- M.D., Sada, A., Prisco, P.P.D. and Nazzaro, F. (2005) Use
sition for the drying and stability of dried foods. J Food of alginate and cryo-protective sugars to improve the
Eng 40, 71–79. viability of lactic acid bacteria after freezing and freeze-
Bucio, A., Hartemink, R., Schrama, J.W., Verreth, J. and drying. World J Microb Biot 21, 739–746.
Rombouts, F.M. (2005) Survival of Lactobacillus plantarum Duong, T., Barrangou, R., Russell, W.M. and Klaenhammer,
44a after spraying and drying in feed and during exposure T.R. (2006) Characterization of the tre locus and analysis
to gastrointestinal tract fluids in vitro. J Gen Appl Micro- of trehalose cryoprotection in Lactobacillus acidophilus
biol 51, 221–227. NCFM. Appl Environ Microbiol 72, 1218–1225.
Carcoba, R. and Rodriguez, A. (2000) Influence of cryoprotec- Efiuvwevwere, B.J.O., Gorris, L.G.M., Smid, E.J. and Kets,
tants on the viability and acidifying activity of frozen and E.P.W. (1999) Mannitol-enhanced survival of Lactococcus
freeze-dried cells of the novel starter strain Lactococcus lactis subjected to drying. Appl Microbiol Biotechnol 51,
lactis ssp. lactis CECT 5180. Eur Food Res Technol 211, 100–104.
433–437. Ferreira, V., Soares, V., Santos, C., Silva, J., Gibbs, P.A. and
Carvalho, A.S., Silva, J., Ho, P., Teixeira, P., Malcata, F.X. and Teixeira, P. (2005) Survival of Lactobacillus sakei during
Gibbs, P. (2002) Survival of freeze-dried Lactobacillus plan- heating, drying and storage in the dried state when growth
tarum and Lactobacillus rhamnosus during storage in the has occurred in the presence of sucrose or monosodium
presence of protectants. Biotechnol Lett 24, 1587–1591. glutamate. Biotechnol Lett 27, 249–252.
Carvalho, A.S., Silva, J., Ho, P., Teixeira, P., Malcata, F.X. and Fonseca, F., Beal, C. and Corrieu, G. (2000) Method of quanti-
Gibbs, P. (2004a) Effects of various sugars added to fying the loss of acidification activity of lactic acid starters
growth and drying media upon thermotolerance and during freezing and frozen storage. J Dairy Res 67, 83–90.
survival throughout storage of freeze-dried Lactobacillus Fonseca, F., Passot, S., Lieben, P. and Marin, M. (2004) Col-
delbrueckii ssp. bulgaricus. Biotechnol Prog 20, 248–254. lapse temperature of bacterial suspensions: the effect of cell
Carvalho, A.S., Silva, J., Ho, P., Teixeira, P., Malcata, F.X. and type and concentration. Cryoletters 25, 425–434.
Gibbs, P. (2004b) Relevant factors for the preparation of Gardiner, G.E., O’Sullivan, E., Kelly, J., Auty, M.A.E., Fitzger-
freeze-dried lactic acid bacteria. Int Dairy J 14, 835–847. ald, G.F., Collins, J.K., Ross, R.P. and Stanton, C. (2000)
Castro, H.P., Teixeira, P.M. and Kirby, R. (1995) Storage of Comparative survival rates of human-derived probiotic
lyophilized cultures of Lactobacillus bulgaricus under differ- Lactobacillus paracasei and L. salivarius strains during heat
ent relative humidities and atmospheres. Appl Microbiol treatment and spray drying. Appl Environ Microbiol 66,
Biotechnol 44, 172–176. 2605–2612.
Champagne, C.P., Mondou, F., Raymond, Y. and Roy, D. Gomes, A.M.P. and Malcata, F.X. (1999) Bifidobacterium spp.
(1996) Effect of polymers and storage temperature on the and Lactobacillus acidophilus: biological, biochemical, tech-
stability of freeze-dried lactic acid bacteria. Food Res Int nological and therapeutical properties relevant for use as
29, 555–562. probiotics. Trends Food Sci Technol 10, 139–157.
Chen, T., Fowler, A. and Toner, M. (2000) Literature review: Hoekstra, F.A., Wolkers, W.F., Buitink, J., Golovina, E.A.,
supplemented phase diagram of the trehalose-water binary Crowe, J.H. and Crowe, L.M. (1997) Membrane stabiliza-
mixture. Cryobiology 40, 277–282. tion in the dry state. Comp Biochem Phys A 117, 335–341.
Chua, K.J. and Chou, S.K. (2003) Low-cost drying methods Johnson, J.A.C. and Etzel, M.R. (1995) Properties of Lactoba-
for developing countries. Trends Food Sci Technol 14, cillus helveticus CNRZ-32 attenuated by spray-drying,
519–528. freeze-drying, or freezing. J Dairy Sci 78, 761–768.
Conrad, P.B., Miller, D.P., Cielenski, P.R. and de Pablo, J.J. Kim, W.S., Perl, L., Park, J.H., Tandianus, J.E. and Dunn,
(2000) Stabilization and preservation of Lactobacillus aci- N.W. (2001) Assessment of stress response of the probiotic
dophilus in saccharide matrices. Cryobiology 41, 17–24. Lactobacillus acidophilus. Curr Microbiol 43, 346–350.

ª 2009 The Authors

176 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177
S. Strasser et al. Freeze vs fluid bed drying of LAB

Knorr, D. (1998) Technology aspects related to microorgan- Saxelin, M., Grenov, B., Svensson, U., Fonden, R., Reniero, R.
isms in functional foods. Trends Food Sci Technol 9, 295– and Mattila-Sandholm, T. (1999) The technology of
306. probiotics. Trends Food Sci Technol 10, 387–392.
Leslie, S.B., Israeli, E., Lighthart, B., Crowe, J.H. and Crowe, Schiraldi, C., Di Lernia, I. and De Rosa, M. (2002) Trehalose
L.M. (1995) Trehalose and sucrose protect both mem- production: exploiting novel approaches. Trends Biotechnol
branes and proteins in intact bacteria during drying. Appl 20, 420–425.
Environ Microbiol 61, 3592–3597. Schoug, A., Olsson, J., Carlfors, J., Schnürer, J. and Håkansson,
Linders, L.J.M., de Jong, G.I.W., Meerdink, G. and van’t Riet, S. (2006) Freeze-drying of Lactobacillus coryniformis
K. (1997a) Carbohydrates and the dehydration inactivation Si3-effects of sucrose concentration, cell density, and
of Lactobacillus plantarum: the role of moisture distribu- freezing rate on cell survival and thermophysical
tion and water activity. J Food Eng 31, 237–250. properties. Cryobiology 53, 119–127.
Linders, L.J.M., Wolkers, W.F., Hoekstra, F.A. and van’t Riet, Selmer-Olsen, E., Birkeland, S.E. and Sørhaug, T. (1999a)
K. (1997b) Effect of added carbohydrates on membrane Effect of protective solutes on leakage from and survival of
phase behavior and survival of dried Lactobacillus planta- immobilized Lactobacillus subjected to drying, storage and
rum. Cryobiology 35, 31–40. rehydration. J Appl Microbiol 87, 429–437.
Linders, L.J.M., Meerdink, G. and van’t Riet, K. (1997c) Effect Selmer-Olsen, E., Sørhaug, T., Birkeland, S.E. and Pehrson, R.
of growth parameters on the residual activity of Lactobacil- (1999b) Survival of Lactobacillus helveticus entrapped in
lus plantarum after drying. J Appl Microbiol 82, 683–688. Ca-alginate in relation to water content, storage and rehy-
Mille, Y., Obert, J.P., Beney, L. and Gervais, P. (2004) New dration. J Ind Microbiol Biot 23, 79–85.
drying process for lactic bacteria based on their dehydra- Smirnoff, N. and Cumbes, Q.J. (1989) Hydroxyl radical
tion behavior in liquid medium. Biotechnol Bioeng 88, scavenging activity of compatible solutes. Phytochemistry
71–76. 28, 1057–1060.
Morgan, C.A., Herman, N., White, P.A. and Vesey, G. (2006) To, B.C.S. and Etzel, M.R. (1997) Spray drying, freeze drying,
Preservation of micro-organisms by drying; A review. or freezing of three different lactic acid bacteria species.
J Microbiol Meth 66, 183–193. J Food Sci 62, 576–578.
Palmfeldt, J. and Hahn-Hagerdal, B. (2000) Influence of cul- Wang, Y.C., Yu, R.C. and Chou, C.C. (2004) Viability of lactic
ture pH on survival of Lactobacillus reuteri subjected to acid bacteria and bifidobacteria in fermented soymilk after
freeze-drying. Int J Food Microbiol 55, 235–238. drying, subsequent rehydration and storage. Int J Food
Potts, M. (2001) Desiccation tolerance: a simple process? Microbiol 93, 209–217.
Trends Microbiol 9, 553–559. Zamora, L.M., Carretero, C. and Pares, D. (2006) Comparative
Ratti, C. (2001) Hot air and freeze-drying of high-value foods: survival rates of lactic acid bacteria isolated from blood,
a review. J Food Eng 49, 311–319. following spray-drying and freeze-drying. Food Sci Technol
Saarela, M., Rantala, M., Hallamaa, K., Nohynek, L., Int 12, 77–84.
Virkajarvi, I. and Matto, J. (2004) Stationary-phase acid Zarate, G. and Nader-Macias, M.E. (2006) Viability and bio-
and heat treatments for improvement of the viability of logical properties of probiotic vaginal Lactobacilli after
probiotic lactobacilli and bifidobacteria. J Appl Microbiol lyophilization and refrigerated storage into gelatin capsules.
96, 1205–1214. Process Biochem 41, 1779–1785.
Santivarangkna, C., Kulozik, U. and Foerst, P. (2007) Zayed, G. and Roos, Y.H. (2004) Influence of trehalose and
Alternative drying processes for the industrial moisture content on survival of Lactobacillus salivarius
preservation of lactic acid starter cultures. Biotechnol subjected to freeze-drying and storage. Process Biochem 39,
Prog 23, 302–315. 1081–1086.

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 167–177 177