1 s2.0 S0022030220300965 Main
1 s2.0 S0022030220300965 Main
1 s2.0 S0022030220300965 Main
103:3066–3075
https://doi.org/10.3168/jds.2019-17685
© American Dairy Science Association®, 2020.
ABSTRACT INTRODUCTION
Although freeze-drying is an excellent method for Lactic acid bacteria (LAB) are commonly used as
preserving microorganisms, it inevitably reduces cell fermentation starters and probiotics, and are thus high-
activity and function. Moreover, probiotic strains differ ly important to the food and dairy industries (Velly et
in terms of their sensitivity to the freeze-drying process. al., 2015). They are also associated with various poten-
Therefore, it is necessary to optimize the variables rel- tial health benefits (Damaskos and Kolios, 2008; Lee et
evant to this process. The pre-freezing temperature is al., 2018) and have been studied widely to explore their
a critical parameter of the freeze-drying process, but it potential effects on various digestive diseases (Gareau
remains unclear whether the optimal pre-freezing tem- et al., 2010; Fan et al., 2019), chronic kidney disease
perature differs among strains and protectants. This (Koppe et al., 2015), immunomodulation (Ohtsuka et
study explored the effects of 4 different pre-freezing al., 2012), and blood cholesterol reduction (Nguyen et
temperatures on the survival rates of different Lactoba- al., 2007; Korcz et al., 2018). However, the wide applica-
cillus plantarum strains after freeze-drying in the pres- tion of Lactobacillus spp. is dependent on the biological
ence of different protectants. Using phosphate-buffered stability of these probiotics during long-term storage.
saline solution and sorbitol as protectants, pre-freezing Freeze-drying (i.e., lyophilization), which is widely used
at −196°C, −40°C, and −20°C ensured the highest in the industrial production of LAB powder, is cur-
survival rates after freeze-drying for AR113, AR307, rently the most effective method for maintaining the
and WCFS1, respectively. Using trehalose, pre-freezing stability of Lactobacillus (Li et al., 2011).
at −20°C ensured the best survival rate for AR113, Freeze-drying is a form of cold-drying preservation in
and −60°C was the best pre-freezing temperature for which a substance is dehydrated by sublimation. This
AR307 and WCFS1. These results indicate that the process involves 3 phases: freezing, primary drying, and
pre-freezing temperature can be changed to improve secondary drying (Fissore et al., 2019). Freeze-drying
the survival rate of L. plantarum, and that this effect is the preferred long-term preservation method used at
is strain-specific. Further studies have demonstrated microbial resource centers and in the industrial pro-
that pre-freezing temperature affected viability via duction of bacterial starters (Prakash et al., 2013). In
changes in cell membrane integrity, membrane perme- addition to its lower storage and transportation costs
ability, and lactate dehydrogenase activity. In sum- and easier handling than freezing (Santivarangkna et
mary, pre-freezing temperature is a crucial factor in L. al., 2008; Velly et al., 2015), freeze-drying also helps to
plantarum survival after freeze-drying, and the choice maintain the stability of products such as medications
of pre-freezing temperature depends on the strain and during production and extends the shelf life of finished
the protectant. products.
Key words: freeze-drying, pre-freezing temperature, Freeze-drying has many advantages and is among the
protectants, Lactobacillus plantarum best methods for microorganism preservation. However,
some aspects of this method require improvement. The
long-term exposure of cells to extreme environments
during freeze-drying has various effects on cell mem-
brane integrity (Dianawati et al., 2016; Wang et al.,
2019) and fluidity (Oldenhof et al., 2005; Schwab et al.,
2007), and on the structures of sensitive proteins (Li et
Received September 30, 2019.
al., 2011; Kandil and El Soda, 2015). Inevitably, these
Accepted November 28, 2019. changes induce physiological damage and reduce the
*Corresponding author: ailianzhong@hotmail.com activities and functions of cells. Moreover, the process-
3066
Wang et al.: STRAIN-SPECIFIC EFFECTS OF PRE-FREEZING TEMPERATURE 3067
Determination of Cell Membrane Integrity tion per minute. The specific activity is expressed per
and Permeability milligrams of protein.
The effect of freeze-drying on L. plantarum cell
Assay of Pyruvate Kinase
membrane integrity was determined using a double
fluorescent staining process with fluorescein diacetate Pyruvate kinase activity in the samples was de-
(FDA) and propidium iodide (PI). Samples obtained termined using a pyruvate kinase assay kit (Nanjing
before and after lyophilization were washed with 1 mL Jiancheng Bioengineering Institute, Nanjing, China).
of sterile PBS and resuspended in 850 μL of PBS. Next, The specific activity was expressed per milligram of
100 μL of FDA solution was added to yield a final con- protein.
centration of 100 μg/mL. The sample was mixed, and
50 μL of PI was added to yield a final concentration of
Statistical Analysis
50 μg/mL. The suspension was shaken and incubated
for 30 min at room temperature while protected from Statistical analyses were performed using Origin 8.5
light. After washing the stained sample 3 times with (OriginLab, Northampton, MA) and SPSS (IBM Corp.,
PBS buffer, the cells were precipitated and resuspended Armonk, NY). We examined significant differences
in 500 μL of PBS. The resulting bacterial solution was between the data using single-factor and multi-factor
dropped on a glass slide and observed under a fluores- methods. A P-value < 0.05 indicated a statistically
cence microscope. significant difference.
The effect of freeze-drying on L. plantarum cell mem-
brane permeability was determined by single staining
RESULTS
with fluorescein diacetate. Samples collected before and
after lyophilization were washed with 1 mL of sterile Survival Rates of Different Strains Using
PBS and resuspended in 1 mL of fresh PBS. Next, 150 Sorbitol as a Protectant
μL of each bacterial suspension and 50 μL of PI were
mixed in the wells of a 96-well plate. A microplate Three L. plantarum strains were freeze-dried after ex-
reader was used to determine the level of fluorescence posure to various pre-freezing temperatures in the pres-
in each well. ence of sorbitol, and the survival rates after freezing,
drying, and freeze-drying were determined (Figures 1,
Preparation of Cell-Free Extract 2, and 3). The effects of the 4 pre-freezing temperatures
varied widely among the strains, but the effect of this
Each lyophilized sample was washed with 1 mL of variable also varied among the phases of freeze-drying
PBS and resuspended in 1 mL of fresh PBS. Next, this for each strain.
bacterial suspension was disrupted ultrasonically in an As shown in Figure 1, strains AR113, AR307, and
ice water bath for 20 min. Each sonication cycle was WCFS1 all exhibited higher freezing survival rates
set to 5 s on/5 s off for 240 cycles. The cell debris was at −20°C and −40°C. In all 3 strains, the maximum
then removed by centrifugation at 7,378 × g and 4°C survival rate was obtained at −20°C (70.6 ± 2.37,
for 5 min, and the supernatant was stored at 4°C before 85.8 ± 1.11, and 82.2 ± 2.63%, respectively). How-
measurement of relevant enzyme activity levels. ever, the 3 strains of L. plantarum differed in terms
of the pre-freezing temperature associated with the
Lactate Dehydrogenase Assay lowest survival rate. We found that AR113 exhibited
a worse survival response to freezing at −60°C (24.71
Lactate dehydrogenase (LDH) reduces pyruvate to ± 2.14%, or one-third of the maximum) compared to
lactic acid, and its activity can be determined by the AR307 and WCFS1 (73.98 ± 5.06 and 79.07 ± 2.33%,
reduced absorption of NADH at 340 nm. In this experi- respectively). The AR307 strain exhibited the lowest
ment, the total reaction volume of 0.31 mL contained survival rate at −196°C (10.29 ± 0.38%, or less than
0.29 mL of sodium pyruvate solution, 0.01 mL of NADH one-eighth the maximum), and WCFS1 exhibited the
solution, and 0.01 mL of cell-free extract. The absorp- lowest survival rate at −196°C, although its relative
tion value of each reaction at 340 nm was measured decrease in survival rate was much less severe (half the
using a microplate reader after shaking. Measurements maximum). These results suggest that AR307 is less
were recorded every minute during 5-min periods. By resistant to low temperatures.
plotting time as the abscissa and absorption as the As shown in Figure 2, AR307 exhibited dry survival
ordinate, the initial linear relationship was determined rates of less than 10%, regardless of the pre-freezing
and used to calculate the reduction in NADH absorp- temperature. For AR113 and WCFS1, the highest dry
survival rates were obtained at −60°C and −196°C, 50%. In summary, most of the damage associated with
respectively, in contrast to the survival outcomes af- freeze-drying may occur during the drying stage.
ter freezing. These rates were 3 and 4 times higher, Figure 3 reflects the freeze-drying survival rates as-
respectively, than the corresponding dry survival rates sociated with 4 pre-freezing temperatures. The AR113
at −40°C. During the drying stage, WCFS1 exhibited and WCFS1 strains achieved maximum survival rates
a survival rate of 76.55 ± 1.01% at −196°C, whereas at −20°C and −196°C, respectively, and these rates
AR113 and AR307 exhibited survival rates of less than were 2.3 to 2.5 times higher than these observed at
−40°C. The AR307 strain exhibited low survival rates
at all 4 pre-freezing temperatures. In AR307, the high-
est survival rate was achieved at −40°C, approximately
4.3 times higher than the lowest survival rate achieved
at −20°C. Although the pre-freezing temperature could
be changed to improve the freeze-drying survival rate
of all 3 L. plantarum strains in the presence of sorbitol,
the overall outcomes were poor. Therefore, we explored
various protectants to identify conditions better suited
to survival.
Figure 5. Drying survival rates of Lactobacillus plantarum (A) Figure 6. Freeze-drying survival rates of Lactobacillus plantarum
AR113, (B) AR307, and (C) WCFS1 exposed to various protectants (A) AR113, (B) AR307, and (C) WCFS1 exposed to various protec-
and freezing temperatures. Data are presented as means of triplicates tants and freezing temperatures. Data are presented as means of tripli-
± SD. Means with different letters differ significantly (P < 0.05). cates ± SD. Means with different letters differ significantly (P < 0.05).
Figure 7. Effect of pre-freezing temperature on the integrity of the Lactobacillus plantarum (strains AR113, AR307, and WCFS1) cell
membrane after freeze-drying. Bacteria with intact cell membranes (viable cells) or damaged cell membranes (dead cells) emit green and red
fluorescence, respectively. (a) AR113 trehalose −20°C; (b) AR113 trehalose −196°C; (c) AR307 trehalose −20°C; (d) AR307 trehalose −196°C;
(e) WCFS1 trehalose −20°C; (f) WCFS1 trehalose −196°C.
Mechanism Underlying the Effects of Pre-Freezing damage occurred at −196°C than at −20°C, consistent
Temperature on Survival with the results of the cell membrane integrity analysis.
Lactate dehydrogenase catalyzes the conversion of
Figure 7 presents fluorescence images of the 3 L. lactate to pyruvate, an important step in energy pro-
plantarum strains stained with FDA and PI after duction in LAB cells. The activity of this enzyme tends
exposure to various pre-freezing temperatures. In the to decrease after freezing or freeze-drying. Therefore,
images, bacteria with intact cell membranes (i.e., viable LDH is used as a model protein for distinguishing
cells) emit green fluorescence, and those with damaged the protective effects of solutes against freezing and
cell membranes (i.e., dead cells) emit red fluorescence. drying-induced denaturation (Carpenter et al., 1993;
Exposure to a pre-freezing temperature of −196°C led
to significant increases in the density of cell membrane
damage relative to the damage observed at −20°C in
all 3 L. plantarum strains (Figure 7), especially AR307.
These data suggest that cell membrane damage is an
important cause of death during freeze-drying, and
that this damage could be reduced by changing the
pre-freezing temperature. In particular, an appropriate
pre-freezing temperature would enable the formation
of more uniform ice crystals and reduce damage to the
cell membrane.
The 3 L. plantarum strains were also stained with
FDA alone, and we used the fluorescence of these cells
after freeze-drying as an indicator of cell membrane
permeability. Specifically, high fluorescence indicated
minor cell membrane damage, and low fluorescence in-
dicated more serious damage. We then determined the
permeability of 3 L. plantarum species exposed to vari-
ous pre-freezing temperatures. As shown in Figure 8,
we observed better membrane permeability −20°C than
at −196°C, consistent with survival rates. For example,
the fluorescence of AR113 at −20°C was 44% (Figure Figure 8. Membrane permeability of Lactobacillus plantarum
AR113, AR307, and WCFS1 after freezing. Data are presented as
6A), nearly 1.3 times higher than that at −196°C. means of triplicates ± SD. Means with different letters differ signifi-
These results suggest that more serious cell membrane cantly (P < 0.05).
Jiang and Nail, 1998). In the present study, the LDH in high internal phase emulsions stabilized with whey
activity levels in all strains were higher at −20°C than protein isolate microgels. Notably, the viability of en-
at −196°C, and this phenomenon was more significant capsulated L. plantarum after pasteurization increased
in AR307 (Figure 9a). These findings were consistent by a factor of approximately 1.8 (Su et al., 2018).
with the results of the cell membrane integrity and sur- Wang et al. (2019) reported that optimization of the
vival analyses. Lactate dehydrogenase is also a marker freezing-thawing conditions improved the survival rate
of plasma membrane integrity. High LDH activity indi-
cates an intact cell membrane and ensures the survival
of the strain. Therefore, appropriate pre-freezing tem-
perature can increase the activity of LDH and ensure
the viability of L. plantarum.
Pyruvate kinase is the key rate-limiting enzyme in
glycolysis. In the present study, the activity of this
enzyme was not significantly affected by differences in
pre-freezing temperatures in any of the 3 strains (Fig-
ure 9b). In other words, the pre-freezing temperature
did not affect pyruvate kinase activity in L. plantarum.
DISCUSSION