- Email: [email protected]
Glycine betaine as a cryoprotectant for prokaryotes
Journal of Microbiological Methods 58 (2004) 31 – 38 www.elsevier.com/locate/jmicmeth
Glycine betaine as a cryoprotectant for prokaryotes David Cleland, Paul Krader, Coral McCree, Jane Tang, David Emerson * Bacteriology Program, American Type Culture Collection, 10801 University Boulevard, Manassas, VA 20110, USA Received 17 February 2004; received in revised form 27 February 2004; accepted 27 February 2004 Available online 9 April 2004
Abstract Osmoprotectants are low molecular weight, hydrophilic, nontoxic molecules that assist a cell under osmotic stress to stabilize its concentration of internal solutes. These properties are similar to compounds used as cryoprotectants for the preservation of prokaryotic cells during freezing. This study tested the ability of a common compatible solute, glycine betaine (GB), to act as a cryoprotectant. In a series of freeze-drying studies using a variety of prokaryotes, GB performed as well, or better than, two commonly used cryoprotectants, sucrose/bovine serum albumin (S/BSA) and trehalose/dextran (T/D). GB did especially well maintaining cell viability after long-term storage (simulated equivalent of 20 years) for microorganisms like Neisseria gonorrhoeae and Streptococcus pneumoniae. GB was tested for its ability to preserve members of the genus Acidothiobacillus, a difficult genus to preserve. For two strains of Acidithiobacillus ferrooxidans that were preserved using liquid drying, GB performed as well as S/BSA. Results were more mixed for two strains of Acidithiobacillus thiooxidans; one strain could be preserved with S/BSA but not GB, the other strain gave low recoveries with both cryoprotectants. GB also proved to be a useful cryoprotectant for liquid nitrogen preservation yielding equivalent results to the cryopreservative, glycerol for halophilic archaea, and neutrophilic Fe-oxidizing bacteria. These results indicate that GB is a simple and useful cryoprotectant that works for a wide range of prokaryotic organisms under different cryopreservation regimens. D 2004 Elsevier B.V. All rights reserved. Keywords: Glycine betaine; Cryoprotectant; Prokaryotes
1. Introduction A unique and practical aspect of prokaryotes is that, with few exceptions, they can be preserved indefinitely in either freeze-dried or frozen form and then recovered. To facilitate this kind of long-term storage, a freeze medium containing a cryoprotectant is required. In general, cryoprotectants have excellent colligative * Corresponding author. Tel.: +1-703-365-2804; fax: +1-703365-2790. E-mail address: [email protected] (D. Emerson). 0167-7012/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2004.02.015
properties with water, which both retard the formation of intracellular ice crystals and reduce the potential for osmotic injury during the freezing and thawing processes. A wide variety of compounds have been used with varying degrees of success as cryoprotection agents, including alcohols, carbohydrates, amino acids, as well as complex reagents such as yeast extract and skimmed milk (Hubalek, 2003). For the frozen storage of prokaryotic cells, the two most commonly used cryoprotectants are glycerol and dimethylsulfoxide (DMSO). For freeze drying, commonly used reagents are skim milk, combinations of trehalose
32
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
1993); furthermore, it is inexpensive and readily available.
and dextran, or sucrose and bovine serum albumin (Simione and Brown, 1991; Hubalek, 2003). All of these agents are quite effective, although none of them may be universal cryoprotectants that work across all genera and for different cryopreservation procedures. Most living organisms respond to osmotic stress by producing osmoprotectants or compatible solutes that are designed to help the cell’s cytoplasm maintain an equivalent osmotic pressure with the external environment without building up high intracellular salt concentrations (Csonka and Epstein, 1996). Good osmoprotectants have properties that include being very hydrophilic, nontoxic, and are usually uncharged. There are a large number of organic osmoprotectants, including trehalose, ectoine, and glycine betaine (GB) (Lengeler et al., 1999). Functionally osmoprotectants and cryoprotectants share important similarities. To our knowledge, relatively few osmoprotectants have been tested for their efficacy as cryoprotectants; two examples include trehalose and proline. Given the functional equivalence of cryoprotection and osmoadaptation, we chose to evaluate glycine betaine, or N,N,N-trimethyl glycine, as a cryoprotectant. Glycine betaine is a very common osmoprotectant that is used by many bacteria, as well as some halophilic archaea (Galinski and Tru¨per, 1994). It is also a common osmoprotectant in animals, plants, and other eukaryotes (Rhodes and Hanson,
2. Materials and methods 2.1. Organisms and growth conditions All prokaryotes used in this study were from the American Type Culture Collection (ATCC) and were grown on the appropriate medium and under the conditions prescribed in Table 1. 2.2. Cryo-reagents Glycine betaine was made up as a 12% (w/v) stock solution in deionized water (d-H2O), unless otherwise noted, and sterilized by filtration through a 0.2-Am filter. The sucrose/bovine serum albumin (S/ BSA) reagent consisted of 10% (w/v) bovine serum albumin fraction V and 20% sucrose in d-H2O that was filter-sterilized. The dextran/trehalose (D/T) reagent consisted of a 10% (w/v) dextran solution that was autoclaved and a 5% trehalose solution that was filter-sterilized. The two solutions were combined 1:1 to make the D/T reagent. Monosodium glutamate (MSG) was made up at the appropriate final concentration in buffer. For studies on preser-
Table 1 List of organisms used in this study ATCC #
Genus and species
ATCC medium#a
Temperature (jC)
Phylotype
Organism type
BAA-724 BAA-590 27886 29096 14119 19859 53983 15494 8085 29605 33170 14990 49619 10150 43504
Psychromonas marina Silicibacter pomeroyi Methylobacterium organophilum Methanothermobacter thermautotrophicus Acidithiobacillus ferrooxidans Acidithiobacillus ferrooxidans Acidithiobacillus ferrooxidans Acidithiobacillus thiooxidans Acidithiobacillus thiooxidans Haloferax volcanii Halobacterium salinarum Staphylococcus epidermidis Streptococcus pneumoniae Neisseria gonorrhoeae Helicobacter pylori
2 2 784 2133 2039 2039 2039 125 125 974 217 3 18 814 260
15 30 30 65 26 26 26 26 26 30 35 37 37 37 37
g-Proteobacteria a-Proteobacteria a-Proteobacteria Euryarchaeota g-Proteobacteria g-Proteobacteria g-Proteobacteria g-Proteobacteria g-Proteobacteria Euryarchaeota Euryarchaeota Gram (+) low G + C Gram (+) low G + C h-Proteobacteria q-Proteobacteria
Psychrophilic marine bacteriium Marine bacterium Methylotroph Methanogen Acidophilic iron oxidizer Acidophilic iron oxidizer Acidophilic iron oxidizer Acidophilic sulfur oxidizer Acidophilic sulfur oxidizer Extreme Halophile Extreme Halophile Bio-medical Bio-medical Bio-medical Bio-medical
a
Medium recipes are available from the ATCC website (www.atcc.org).
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
vation in liquid nitrogen (LN), glycerol was prepared as a 20% (v/v) stock solution that was sterilized by autoclaving. For the work with extreme halophiles, cryopreservation solutions were made up in a 20% NaCl solution. GB, trehalose, and dextran were purchased from Sigma; glycerol and BSA (fraction V) were from ICN Biomedical; MSG was from Fisher Scientific; sucrose was from Difco, and DMSO was from EM Science. 2.3. Freeze-dry studies For each of the cultures of Psychromonas marina, Silicibacter pomeroyi, Methylobacterium organophilum, Staphlyococcus epidermidis, and Streptococcus pneumoniae, cells were grown to a confluent lawn on solid medium in Kolle flasks. The cells were washed off the agar surface with the appropriate medium containing one of the three cryoprotectants, GB, S/ BSA, or D/T. Haloferax volcanii and Halobacterium salinarum were each grown in the broth medium while shaking (200 rpm). The cells were concentrated by centrifugation at 8000 rpm for 20 min and resuspended in growth medium. The cells were then diluted with an equal volume of sterile cryoprotectant. For the methanogenic culture of Methanothermobacter thermautotrophicus, the cells were first concentrated by centrifugation and resuspended in the appropriate pre-reduced cryoprotectant. All cell manipulations were done under anaerobic conditions. For FD and LD (see below), 0.2 ml of a cell suspension was added to a sterile glass ampoule which was closed with a sterile cotton plug. The cotton plugs were cut off even with the top of the vial and the vials were placed at 4 jC until they could be processed. Special precautions were taken with M. thermautotrophicus to keep the culture anaerobic by gassing the vials under a stream of N2/CO2 (80:20) during filling. Each vial was immediately flash frozen in liquid nitrogen. The ATCC’s standard double vial FD protocol was used for P. marina, S. pomeroyi, M. organophilum, M. thermautotrophicus, H. volcanii, and H. salinarum. The vials are treated in a commercial freeze-dryer and then the freeze-dried vial is placed inside a second glass vial that is heat-sealed under vacuum (Simione and Brown, 1991). S. epidermidis and S. pneumoniae were preserved using the precep-
33
trol method, in which the cells are freeze-dried in a single glass serum vial with a rubber stopper (Simione and Brown, 1991). 2.4. Liquid drying of Acidothiobacillus spp. For LD studies of Fe- and S-oxidizing bacteria. 300 ml of cells was grown for 21 days (Acidithioobacillus ferrooxidans) or 28 days (Acidithiobacillus thiooxidans). The cultures were vigorously hand shaken for 1 min and allowed to sit for 10 min at room temperature to allow the particulates of iron oxides or S particles to settle. The culture fluid was poured off (leaving large particulates behind) and centrifuged at 8000 g for 30 min at 4 jC. We found that centrifugation even at much higher speeds often resulted in poor recovery of cells, and so the supernatant was passed through a 0.2-Am filter. The cells were recovered from the filter using sterile medium and combined with the cell pellet, which was washed again by centrifugation at 27,000 g for 20 min at 4 jC (during which the cells pelleted readily). The cells were resuspended in sterile medium, distributed into screw cap test tubes, and centrifuged in a table top centrifuge at 2000 g at room temperature for 30 min. The supernatant was again discarded and each cell pellet was resuspended in the appropriate cryoprotectant and added to ampoules as described above. A detailed description of the L-drying procedure is given in the paper by Malik (1990). In brief, a stainless steel pan containing the glass ampoules was covered with a flat plastic disk. The plastic disk had a stainless steel tube through the center that attached to a rubber tube which was connected to a condenser and then to a vacuum pump. The ampoules were subjected to approximately 5-mbar vacuum for 30 min at room temperature. After 30 min, the vacuum was increased from 0.1 to 0.05 mbar for 2 h at room temperature to complete the drying procedure. The ampoules were transferred to glass tubes containing silica gel and heat-sealed under a vacuum. 2.5. Liquid nitrogen studies H. volcanii and H. salinarum were each grown and prepared as described above. The cell pellet was
34
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
resuspended in a small amount of a sterile 20% NaCl solution and then mixed 1:1 with stocks of GB or glycerol. The Fe-oxidizing bacterium (FeOB) strain ES-1 was grown according to published procedures and harvested by centrifugation (Emerson and Moyer, 1997). The cell pellet was resuspended in the appropriate cryopreservative. The cell suspensions (0.5 ml each) were added to sterile plastic cryo-vials. These were placed in a 80 jC freezer prior to being placed in LN. 2.6. Accelerated storage It is recognized that organisms preserved by freeze-drying methods may undergo a slow deterioration with time. It is impractical to wait for years to do long-term viability studies of FD or LD material. A previous study by Sakane and Kuroshima compared simulated and natural survival rates for 59 FD cultures of a variety of gram negative and gram positive bacteria. Their accelerated storage protocol simply maintained freshly prepared FD vials of the microbes for 2 weeks at 37 jC. Their results showed that storage of FD vials at 37 jC was quite effective at mimicking the die-off caused by longterm storage at the normal maintenance temperature for FD vials of 4 jC for z 20 years (Sakane and Kuroshima, 1997). Therefore, in the present study, accelerated storage experiments were done by placing FD and LD vials in a 37 jC incubator in the dark for 2 weeks. This simulated the temporal changes that occurred during at least 20 years of storage at 4 jC. 2.7. Quantification of cell recovery For each experiment, the number of viable cells was determined for pre-preservation (Pre-P), post-preservation (Post-P) and after accelerated storage. For the organisms that could be grown on solid medium, cell numbers were quantified by performing serial dilutions and plating onto the appropriate medium to determine colony forming units (CFUs). Enumeration
Post-P and after accelerated storage was done on three separate vials for each organism. The plates were incubated at the appropriate temperature and time (24 –96 h) until colonies could be counted. The viability of organisms that could not be plated, M. thermautotrophicus, strains of A. ferrooxidans and A. thiooxidans, and Fe-oxidizing bacteria was determined by most probable number (MPN) using a threetube MPN assay. An appropriate medium (see Table 1) was used for each organism. The lyophilized or frozen sample was subjected to serial dilution. A three-tube MPN series was inoculated from each dilution tube. Positive tubes were scored for growth by turbidity and/or light microscopy. Cell numbers were then determined using a standard MPN Table (de Man, 1975).
3. Results and discussion 3.1. Glycine betaine Glycine betaine is a quaternary ammonium compound and is a common compatible solute of prokaryotes, plants, and animals. GB is cited as one of the most common compatible solutes produced by bacteria, synthesized both by halophiles that are adapted to living at high salt concentrations and non-halophilic bacteria that are subjected to osmotic stress (Galinski and Tru¨per, 1994). Halophilic archaea, especially halophilic methanogens, have also been shown to synthesize glycine betaine as an osmoprotectant (Martin et al., 1999). Furthermore, GB is considered to be one of the most common compatible solutes that is synthesized by a wide variety of plants in response to environmental stresses including drought (McNeil et al., 1999). There have been several studies that investigated GB’s properties as a cryoprotectant for plant cells during freezing as an environmental stress (Rhodes and Hanson, 1993; Xing and Rajashekar, 2001). Among bacteria, GB has been shown to enhance the physiological activity of Listeria monocytogenes and some cyanobacteria at cold temperatures (Ko et al., 1994; Deshnium et al.,
Fig. 1. Freeze-drying results. For Streptococcus and Helicobacter the symbol * denotes no growth. The error bars denote one standard deviation from the mean for triplicate samples. For Halobacterium and Haloferax, D/T was not tested, and accelerated storage was not done. For each data series, the first bar denotes pre-preservation, the second bar post-preservation, and the third bar accelerated storage values for viable cell numbers. Note that the values for the Y-axes of the graphs are not all the same.
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
35
36
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
The results from a series of lyophilization experiments comparing GB with S/BSA and D/T for preservation of a variety of prokaryotes are shown in Fig. 1. Both S/BSA and D/T are standard cryoprotectants that have been used for freeze-drying hundreds of different type strains of prokaryotes at the ATCC. As the results show, GB works as well, or better than either S/BSA or D/T for all the organisms tested. For all three reagents, there was typically a 1 –2-log loss of cell viability immediately after freeze-drying. In most cases, cell recoveries following accelerated storage showed an additional 1 – 2-log decline in viability. D/T consistently had the greatest loss of viability from initial cell numbers to post-storage recovery. Three of the genera tested proved to be important exceptions. S. pneumoniae (49619) exhibited complete die-off following accelerated storage for both S/ BSA and D/T, while stocks preserved with GB only had a 2-log decrease in viability. This strain of S. pneumoniae has always been difficult to preserve in long-term FD storage, so GB will prove quite useful for this organism. Neisseria gonorrhoeae also had a dramatic die-off following accelerated storage with both S/BSA (4 –5-log drop) and D/T (5 –6-log drop). With GB, there was only a 2– 3-log drop in viability after accelerated storage. Finally, H. pylori had a 2 – 3log drop in viability for all the regents following lyophilization; however, it did not survive under any condition following simulated storage. Therefore, we recommend that Helicobacter spp. be preserved only by LN freezing (Fig. 1).
preserve by FD methods. The ATCC has always maintained growing stocks of these cultures due to inconsistent results with cryo-preservation. A. ferrooxidans has successfully been preserved by L-drying. Therefore, we chose to compare GB and S/BSA as cryoprotectants for L-drying two strains of A. ferrooxidans (14119 and 53983) and two strains of A. thiooxidans (8085 and 15494). The results are shown in Fig. 2A. GB worked adequately for preserving A. ferrooxidans. There was a 3 –4-log loss of viability after accelerated storage; however, as long as a dense cell suspension is used, the chances of recovery are excellent. GB compared quite favorably to S/BSA (Fig. 2A) in L-drying of A. ferrooxidans. Experiments were also carried out using MSG, another reagent that has been used in Ldrying of A. ferrooxidans previously ( K. Suzuki, personal communication). In our hands, MSG did not work as well as GB or S/BSA; it yielded post-storage recoveries that were 2 –3 logs below those of the other cryopreservatives (results not shown). For A. thiooxidans, the results were more mixed. For strain 15494, GB did not work effectively as a cryoprotectant yielding no recovery after L-drying (Fig. 2B). S/BSA did yield a very low recovery of cells following simulated long-term storage. For the other strain, 8085, GB had a much greater loss of viability after L-drying than S/BSA; however, after accelerated storage, GB maintained a low level of viability that matched the other reagent (Fig. 2B). MSG also failed to preserve strain 15494 after accelerated storage, and yielded similar results to GB and S/BSA for strain 8085 (results not shown). Based on these results, it is recommended that A. thiooxidans be preserved only by freezing in LN. GB was also capable of preserving A. ferrooxidans under the regular freeze-drying protocol. However, the recoveries after long-term preservation were poor yielding only 10 –100 viable cells per vial (Fig. 2C). Based on the comparison of L-drying and FD for the two strains of A. ferrooxidans using GB as a cryopreservative, L-drying would be the preferred method of preservation.
3.3. Lyophilization of Acidothiobacillus
3.4. Liquid nitrogen preservation
The acidophilic and lithotrophic Fe-oxidizing bacterium A. ferrooxidans and sulfur-oxidizing bacterium A. thiooxidans have always proven challenging to
We tested GBs effectiveness as a cryopreservative when freezing halophilic archaea and neutrophilic Feoxidizing bacteria. The results are shown in Table 2.
1997; Angelidis and Smith, 2003). We are not aware of its previous use as a cryo-reagent for the preservation of prokaryotes (Hubalek, 2003). During preservation studies with A. ferrooxidans (see below), different concentrations of GB (0.3% to 10%) were tested and it was found that a 6% (w/v) solution gave the best recoveries (results not shown). This concentration was used for all additional work. 3.2. Freeze-drying
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
37
Fig. 2. Lyophilization results for Acidothiobacillus. (A) L-drying results for two strains of A. ferrooxidans. (B) L-drying results for two strains of A. thiooxidans; the symbol * denotes no viable cell recovery. (C) FD results for two strains of A. ferroxidans using GB as a cryoprotectant. Note the values for the Y-axes of the graphs are not all the same.
38
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
Table 2 LN Results for halophiles and FeOB; values in parentheses are standard deviation from triplicate samples Organism
GB
Glycerol
Pre-freeze H. volcanii H. salinarium FeOB ES-1
10
Post Freeze 10
6.1 10 (1.4 10 ) 1.4 1010 (1.1 1010) 6.7 107
9
Pre-freeze 9
8.8 10 (7.2 10 ) 3.3 109 (7.9 108) 1.1 106
GB was just as effective as glycerol as a preservative for the halophiles and nearly equivalent at protecting the FeOB. GB has also been used to preserve other strains of halophiles with excellent results, and we have tested it nonquantitatively with a number of bacteria and it appears effective. These results suggest that GB has the potential as a simple, universal cryopreservative that will work with a broad range of prokaryotes.
4. Conclusions Glycine betaine was proven to be an effective cryoprotectant for freeze-drying, liquid-drying, and LN freezing. It is a cheap, simple reagent to make and use. It is nontoxic and, in most cases, will not serve as a growth substrate for prokaryotes. It was more effective at preserving freeze-dried S. pneumoniae after simulated long-term storage than two other common cryopreservatives. As the case with A. thiooxidans and H. pylori illustrates, its efficacy should not be taken for granted, but like all cryo-preservatives, should be checked empirically before being used to preserve a novel group of microorganisms. Acknowledgements We thank Ken Jones of the ATCC operations group for assistance with freeze-drying. This work was supported in part by a grant from the National Science Foundation DBI-0090224. References Angelidis, A.S., Smith, G.M., 2003. Role of the glycine betaine and carnitine transporters in adaptation of Listeria monocytogenes to
10
Post-freeze 9
4.4 10 (3 10 ) 2.5 1010 (3.7 109) 6.7 107
3.2 109 (6.2 108) 4.2 109 (2.2 108) 6 106
chill stress in defined medium. Appl. Environ. Microbiol. 69, 7492 – 7498. Csonka, L.N., Epstein, W., 1996. Osmoregulation. In: Neihardt, F.C., Curtiss R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., (Eds.), Escherichia Coli and Salmonella: Cellular and Molecular Biology. ASM Press, Washington, DC, pp. 1210 – 1224. de Man, J.C., 1975. The probability of most probable numbers. Eur. J. Appl. Microbiol. 1, 67 – 78. Deshnium, P., Gombos, Z., Nishiyama, Y., Murata, N., 1997. The action in vivo of glycine betaine in enhancement of tolerance of Synechococcus sp. strain PCC 7942 to low temperature. J. Bacteriol. 179, 339 – 344. Emerson, D., Moyer, C., 1997. Isolation and characterization of novel iron-oxidizing bacteria that grow at circumneutral pH. Appl. Environ. Microbiol. 63, 4784 – 4792. Galinski, E.A., Tru¨per, H.G., 1994. Microbial behaviour in saltstressed ecosystems. FEMS Microbiol. Rev. 15, 95 – 108. Hubalek, Z., 2003. Protectants used in the cryopreservation of microorganisms. Cryobiology 46, 205 – 229. Ko, R., Smith, L.T., Smith, G.M., 1994. Glycine betaine confers enhanced osmotolerance and cryotolerance on Listeria monocytogenes. J. Bacteriol. 176, 426 – 431. Lengeler, J.W., Drews, G., Schlegel, H.G., 1999. Biology of Prokaryotes. Blackwell, New York. Malik, K.A., 1990. A simplified liquid-drying method for the preservation of microorganisms sensitive to freezing and freezedrying. J. Microbiol. Methods 12, 125 – 132. Martin, D.D., Ciulla, R.A., Roberts, M.F., 1999. Osmoadaptation in Archaea. Appl. Environ. Microbiol. 65, 1815 – 1825. McNeil, S.D., Nuccio, M.L., Hanson, A.D., 1999. Betaines and related osmoprotectants: targets for metabolic engineering of stress resistance. Plant Physiol. 120, 945 – 949. Rhodes, D., Hanson, A.D., 1993. Quaternary ammonium and tertiary sulfonium compounds in higher plants. Annu. Rev. Plant Physiol. 44, 357 – 384. Sakane, T., Kuroshima, K., 1997. Viabilities of dried cultures of various bacteria after preservation for over 20 years and their prediction by the accelerated storage test. Microbiol. Cult. Collect. 13, 1 – 7. Simione, F.P., Brown, E.M. (Eds.), 1991. ATCC Preservation Methods: Freezing and Freeze-Drying, American Type Culture Collection, Rockville, MD, pp. 7 – 13. Xing, W., Rajashekar, C.B., 2001. Glycine betaine involvement in freezing tolerance and water stress in Arabidopsis thaliana. Environ. Exp. Bot. 46, 21 – 28.
Glycine betaine as a cryoprotectant for prokaryotes David Cleland, Paul Krader, Coral McCree, Jane Tang, David Emerson * Bacteriology Program, American Type Culture Collection, 10801 University Boulevard, Manassas, VA 20110, USA Received 17 February 2004; received in revised form 27 February 2004; accepted 27 February 2004 Available online 9 April 2004
Abstract Osmoprotectants are low molecular weight, hydrophilic, nontoxic molecules that assist a cell under osmotic stress to stabilize its concentration of internal solutes. These properties are similar to compounds used as cryoprotectants for the preservation of prokaryotic cells during freezing. This study tested the ability of a common compatible solute, glycine betaine (GB), to act as a cryoprotectant. In a series of freeze-drying studies using a variety of prokaryotes, GB performed as well, or better than, two commonly used cryoprotectants, sucrose/bovine serum albumin (S/BSA) and trehalose/dextran (T/D). GB did especially well maintaining cell viability after long-term storage (simulated equivalent of 20 years) for microorganisms like Neisseria gonorrhoeae and Streptococcus pneumoniae. GB was tested for its ability to preserve members of the genus Acidothiobacillus, a difficult genus to preserve. For two strains of Acidithiobacillus ferrooxidans that were preserved using liquid drying, GB performed as well as S/BSA. Results were more mixed for two strains of Acidithiobacillus thiooxidans; one strain could be preserved with S/BSA but not GB, the other strain gave low recoveries with both cryoprotectants. GB also proved to be a useful cryoprotectant for liquid nitrogen preservation yielding equivalent results to the cryopreservative, glycerol for halophilic archaea, and neutrophilic Fe-oxidizing bacteria. These results indicate that GB is a simple and useful cryoprotectant that works for a wide range of prokaryotic organisms under different cryopreservation regimens. D 2004 Elsevier B.V. All rights reserved. Keywords: Glycine betaine; Cryoprotectant; Prokaryotes
1. Introduction A unique and practical aspect of prokaryotes is that, with few exceptions, they can be preserved indefinitely in either freeze-dried or frozen form and then recovered. To facilitate this kind of long-term storage, a freeze medium containing a cryoprotectant is required. In general, cryoprotectants have excellent colligative * Corresponding author. Tel.: +1-703-365-2804; fax: +1-703365-2790. E-mail address: [email protected] (D. Emerson). 0167-7012/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2004.02.015
properties with water, which both retard the formation of intracellular ice crystals and reduce the potential for osmotic injury during the freezing and thawing processes. A wide variety of compounds have been used with varying degrees of success as cryoprotection agents, including alcohols, carbohydrates, amino acids, as well as complex reagents such as yeast extract and skimmed milk (Hubalek, 2003). For the frozen storage of prokaryotic cells, the two most commonly used cryoprotectants are glycerol and dimethylsulfoxide (DMSO). For freeze drying, commonly used reagents are skim milk, combinations of trehalose
32
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
1993); furthermore, it is inexpensive and readily available.
and dextran, or sucrose and bovine serum albumin (Simione and Brown, 1991; Hubalek, 2003). All of these agents are quite effective, although none of them may be universal cryoprotectants that work across all genera and for different cryopreservation procedures. Most living organisms respond to osmotic stress by producing osmoprotectants or compatible solutes that are designed to help the cell’s cytoplasm maintain an equivalent osmotic pressure with the external environment without building up high intracellular salt concentrations (Csonka and Epstein, 1996). Good osmoprotectants have properties that include being very hydrophilic, nontoxic, and are usually uncharged. There are a large number of organic osmoprotectants, including trehalose, ectoine, and glycine betaine (GB) (Lengeler et al., 1999). Functionally osmoprotectants and cryoprotectants share important similarities. To our knowledge, relatively few osmoprotectants have been tested for their efficacy as cryoprotectants; two examples include trehalose and proline. Given the functional equivalence of cryoprotection and osmoadaptation, we chose to evaluate glycine betaine, or N,N,N-trimethyl glycine, as a cryoprotectant. Glycine betaine is a very common osmoprotectant that is used by many bacteria, as well as some halophilic archaea (Galinski and Tru¨per, 1994). It is also a common osmoprotectant in animals, plants, and other eukaryotes (Rhodes and Hanson,
2. Materials and methods 2.1. Organisms and growth conditions All prokaryotes used in this study were from the American Type Culture Collection (ATCC) and were grown on the appropriate medium and under the conditions prescribed in Table 1. 2.2. Cryo-reagents Glycine betaine was made up as a 12% (w/v) stock solution in deionized water (d-H2O), unless otherwise noted, and sterilized by filtration through a 0.2-Am filter. The sucrose/bovine serum albumin (S/ BSA) reagent consisted of 10% (w/v) bovine serum albumin fraction V and 20% sucrose in d-H2O that was filter-sterilized. The dextran/trehalose (D/T) reagent consisted of a 10% (w/v) dextran solution that was autoclaved and a 5% trehalose solution that was filter-sterilized. The two solutions were combined 1:1 to make the D/T reagent. Monosodium glutamate (MSG) was made up at the appropriate final concentration in buffer. For studies on preser-
Table 1 List of organisms used in this study ATCC #
Genus and species
ATCC medium#a
Temperature (jC)
Phylotype
Organism type
BAA-724 BAA-590 27886 29096 14119 19859 53983 15494 8085 29605 33170 14990 49619 10150 43504
Psychromonas marina Silicibacter pomeroyi Methylobacterium organophilum Methanothermobacter thermautotrophicus Acidithiobacillus ferrooxidans Acidithiobacillus ferrooxidans Acidithiobacillus ferrooxidans Acidithiobacillus thiooxidans Acidithiobacillus thiooxidans Haloferax volcanii Halobacterium salinarum Staphylococcus epidermidis Streptococcus pneumoniae Neisseria gonorrhoeae Helicobacter pylori
2 2 784 2133 2039 2039 2039 125 125 974 217 3 18 814 260
15 30 30 65 26 26 26 26 26 30 35 37 37 37 37
g-Proteobacteria a-Proteobacteria a-Proteobacteria Euryarchaeota g-Proteobacteria g-Proteobacteria g-Proteobacteria g-Proteobacteria g-Proteobacteria Euryarchaeota Euryarchaeota Gram (+) low G + C Gram (+) low G + C h-Proteobacteria q-Proteobacteria
Psychrophilic marine bacteriium Marine bacterium Methylotroph Methanogen Acidophilic iron oxidizer Acidophilic iron oxidizer Acidophilic iron oxidizer Acidophilic sulfur oxidizer Acidophilic sulfur oxidizer Extreme Halophile Extreme Halophile Bio-medical Bio-medical Bio-medical Bio-medical
a
Medium recipes are available from the ATCC website (www.atcc.org).
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
vation in liquid nitrogen (LN), glycerol was prepared as a 20% (v/v) stock solution that was sterilized by autoclaving. For the work with extreme halophiles, cryopreservation solutions were made up in a 20% NaCl solution. GB, trehalose, and dextran were purchased from Sigma; glycerol and BSA (fraction V) were from ICN Biomedical; MSG was from Fisher Scientific; sucrose was from Difco, and DMSO was from EM Science. 2.3. Freeze-dry studies For each of the cultures of Psychromonas marina, Silicibacter pomeroyi, Methylobacterium organophilum, Staphlyococcus epidermidis, and Streptococcus pneumoniae, cells were grown to a confluent lawn on solid medium in Kolle flasks. The cells were washed off the agar surface with the appropriate medium containing one of the three cryoprotectants, GB, S/ BSA, or D/T. Haloferax volcanii and Halobacterium salinarum were each grown in the broth medium while shaking (200 rpm). The cells were concentrated by centrifugation at 8000 rpm for 20 min and resuspended in growth medium. The cells were then diluted with an equal volume of sterile cryoprotectant. For the methanogenic culture of Methanothermobacter thermautotrophicus, the cells were first concentrated by centrifugation and resuspended in the appropriate pre-reduced cryoprotectant. All cell manipulations were done under anaerobic conditions. For FD and LD (see below), 0.2 ml of a cell suspension was added to a sterile glass ampoule which was closed with a sterile cotton plug. The cotton plugs were cut off even with the top of the vial and the vials were placed at 4 jC until they could be processed. Special precautions were taken with M. thermautotrophicus to keep the culture anaerobic by gassing the vials under a stream of N2/CO2 (80:20) during filling. Each vial was immediately flash frozen in liquid nitrogen. The ATCC’s standard double vial FD protocol was used for P. marina, S. pomeroyi, M. organophilum, M. thermautotrophicus, H. volcanii, and H. salinarum. The vials are treated in a commercial freeze-dryer and then the freeze-dried vial is placed inside a second glass vial that is heat-sealed under vacuum (Simione and Brown, 1991). S. epidermidis and S. pneumoniae were preserved using the precep-
33
trol method, in which the cells are freeze-dried in a single glass serum vial with a rubber stopper (Simione and Brown, 1991). 2.4. Liquid drying of Acidothiobacillus spp. For LD studies of Fe- and S-oxidizing bacteria. 300 ml of cells was grown for 21 days (Acidithioobacillus ferrooxidans) or 28 days (Acidithiobacillus thiooxidans). The cultures were vigorously hand shaken for 1 min and allowed to sit for 10 min at room temperature to allow the particulates of iron oxides or S particles to settle. The culture fluid was poured off (leaving large particulates behind) and centrifuged at 8000 g for 30 min at 4 jC. We found that centrifugation even at much higher speeds often resulted in poor recovery of cells, and so the supernatant was passed through a 0.2-Am filter. The cells were recovered from the filter using sterile medium and combined with the cell pellet, which was washed again by centrifugation at 27,000 g for 20 min at 4 jC (during which the cells pelleted readily). The cells were resuspended in sterile medium, distributed into screw cap test tubes, and centrifuged in a table top centrifuge at 2000 g at room temperature for 30 min. The supernatant was again discarded and each cell pellet was resuspended in the appropriate cryoprotectant and added to ampoules as described above. A detailed description of the L-drying procedure is given in the paper by Malik (1990). In brief, a stainless steel pan containing the glass ampoules was covered with a flat plastic disk. The plastic disk had a stainless steel tube through the center that attached to a rubber tube which was connected to a condenser and then to a vacuum pump. The ampoules were subjected to approximately 5-mbar vacuum for 30 min at room temperature. After 30 min, the vacuum was increased from 0.1 to 0.05 mbar for 2 h at room temperature to complete the drying procedure. The ampoules were transferred to glass tubes containing silica gel and heat-sealed under a vacuum. 2.5. Liquid nitrogen studies H. volcanii and H. salinarum were each grown and prepared as described above. The cell pellet was
34
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
resuspended in a small amount of a sterile 20% NaCl solution and then mixed 1:1 with stocks of GB or glycerol. The Fe-oxidizing bacterium (FeOB) strain ES-1 was grown according to published procedures and harvested by centrifugation (Emerson and Moyer, 1997). The cell pellet was resuspended in the appropriate cryopreservative. The cell suspensions (0.5 ml each) were added to sterile plastic cryo-vials. These were placed in a 80 jC freezer prior to being placed in LN. 2.6. Accelerated storage It is recognized that organisms preserved by freeze-drying methods may undergo a slow deterioration with time. It is impractical to wait for years to do long-term viability studies of FD or LD material. A previous study by Sakane and Kuroshima compared simulated and natural survival rates for 59 FD cultures of a variety of gram negative and gram positive bacteria. Their accelerated storage protocol simply maintained freshly prepared FD vials of the microbes for 2 weeks at 37 jC. Their results showed that storage of FD vials at 37 jC was quite effective at mimicking the die-off caused by longterm storage at the normal maintenance temperature for FD vials of 4 jC for z 20 years (Sakane and Kuroshima, 1997). Therefore, in the present study, accelerated storage experiments were done by placing FD and LD vials in a 37 jC incubator in the dark for 2 weeks. This simulated the temporal changes that occurred during at least 20 years of storage at 4 jC. 2.7. Quantification of cell recovery For each experiment, the number of viable cells was determined for pre-preservation (Pre-P), post-preservation (Post-P) and after accelerated storage. For the organisms that could be grown on solid medium, cell numbers were quantified by performing serial dilutions and plating onto the appropriate medium to determine colony forming units (CFUs). Enumeration
Post-P and after accelerated storage was done on three separate vials for each organism. The plates were incubated at the appropriate temperature and time (24 –96 h) until colonies could be counted. The viability of organisms that could not be plated, M. thermautotrophicus, strains of A. ferrooxidans and A. thiooxidans, and Fe-oxidizing bacteria was determined by most probable number (MPN) using a threetube MPN assay. An appropriate medium (see Table 1) was used for each organism. The lyophilized or frozen sample was subjected to serial dilution. A three-tube MPN series was inoculated from each dilution tube. Positive tubes were scored for growth by turbidity and/or light microscopy. Cell numbers were then determined using a standard MPN Table (de Man, 1975).
3. Results and discussion 3.1. Glycine betaine Glycine betaine is a quaternary ammonium compound and is a common compatible solute of prokaryotes, plants, and animals. GB is cited as one of the most common compatible solutes produced by bacteria, synthesized both by halophiles that are adapted to living at high salt concentrations and non-halophilic bacteria that are subjected to osmotic stress (Galinski and Tru¨per, 1994). Halophilic archaea, especially halophilic methanogens, have also been shown to synthesize glycine betaine as an osmoprotectant (Martin et al., 1999). Furthermore, GB is considered to be one of the most common compatible solutes that is synthesized by a wide variety of plants in response to environmental stresses including drought (McNeil et al., 1999). There have been several studies that investigated GB’s properties as a cryoprotectant for plant cells during freezing as an environmental stress (Rhodes and Hanson, 1993; Xing and Rajashekar, 2001). Among bacteria, GB has been shown to enhance the physiological activity of Listeria monocytogenes and some cyanobacteria at cold temperatures (Ko et al., 1994; Deshnium et al.,
Fig. 1. Freeze-drying results. For Streptococcus and Helicobacter the symbol * denotes no growth. The error bars denote one standard deviation from the mean for triplicate samples. For Halobacterium and Haloferax, D/T was not tested, and accelerated storage was not done. For each data series, the first bar denotes pre-preservation, the second bar post-preservation, and the third bar accelerated storage values for viable cell numbers. Note that the values for the Y-axes of the graphs are not all the same.
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
35
36
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
The results from a series of lyophilization experiments comparing GB with S/BSA and D/T for preservation of a variety of prokaryotes are shown in Fig. 1. Both S/BSA and D/T are standard cryoprotectants that have been used for freeze-drying hundreds of different type strains of prokaryotes at the ATCC. As the results show, GB works as well, or better than either S/BSA or D/T for all the organisms tested. For all three reagents, there was typically a 1 –2-log loss of cell viability immediately after freeze-drying. In most cases, cell recoveries following accelerated storage showed an additional 1 – 2-log decline in viability. D/T consistently had the greatest loss of viability from initial cell numbers to post-storage recovery. Three of the genera tested proved to be important exceptions. S. pneumoniae (49619) exhibited complete die-off following accelerated storage for both S/ BSA and D/T, while stocks preserved with GB only had a 2-log decrease in viability. This strain of S. pneumoniae has always been difficult to preserve in long-term FD storage, so GB will prove quite useful for this organism. Neisseria gonorrhoeae also had a dramatic die-off following accelerated storage with both S/BSA (4 –5-log drop) and D/T (5 –6-log drop). With GB, there was only a 2– 3-log drop in viability after accelerated storage. Finally, H. pylori had a 2 – 3log drop in viability for all the regents following lyophilization; however, it did not survive under any condition following simulated storage. Therefore, we recommend that Helicobacter spp. be preserved only by LN freezing (Fig. 1).
preserve by FD methods. The ATCC has always maintained growing stocks of these cultures due to inconsistent results with cryo-preservation. A. ferrooxidans has successfully been preserved by L-drying. Therefore, we chose to compare GB and S/BSA as cryoprotectants for L-drying two strains of A. ferrooxidans (14119 and 53983) and two strains of A. thiooxidans (8085 and 15494). The results are shown in Fig. 2A. GB worked adequately for preserving A. ferrooxidans. There was a 3 –4-log loss of viability after accelerated storage; however, as long as a dense cell suspension is used, the chances of recovery are excellent. GB compared quite favorably to S/BSA (Fig. 2A) in L-drying of A. ferrooxidans. Experiments were also carried out using MSG, another reagent that has been used in Ldrying of A. ferrooxidans previously ( K. Suzuki, personal communication). In our hands, MSG did not work as well as GB or S/BSA; it yielded post-storage recoveries that were 2 –3 logs below those of the other cryopreservatives (results not shown). For A. thiooxidans, the results were more mixed. For strain 15494, GB did not work effectively as a cryoprotectant yielding no recovery after L-drying (Fig. 2B). S/BSA did yield a very low recovery of cells following simulated long-term storage. For the other strain, 8085, GB had a much greater loss of viability after L-drying than S/BSA; however, after accelerated storage, GB maintained a low level of viability that matched the other reagent (Fig. 2B). MSG also failed to preserve strain 15494 after accelerated storage, and yielded similar results to GB and S/BSA for strain 8085 (results not shown). Based on these results, it is recommended that A. thiooxidans be preserved only by freezing in LN. GB was also capable of preserving A. ferrooxidans under the regular freeze-drying protocol. However, the recoveries after long-term preservation were poor yielding only 10 –100 viable cells per vial (Fig. 2C). Based on the comparison of L-drying and FD for the two strains of A. ferrooxidans using GB as a cryopreservative, L-drying would be the preferred method of preservation.
3.3. Lyophilization of Acidothiobacillus
3.4. Liquid nitrogen preservation
The acidophilic and lithotrophic Fe-oxidizing bacterium A. ferrooxidans and sulfur-oxidizing bacterium A. thiooxidans have always proven challenging to
We tested GBs effectiveness as a cryopreservative when freezing halophilic archaea and neutrophilic Feoxidizing bacteria. The results are shown in Table 2.
1997; Angelidis and Smith, 2003). We are not aware of its previous use as a cryo-reagent for the preservation of prokaryotes (Hubalek, 2003). During preservation studies with A. ferrooxidans (see below), different concentrations of GB (0.3% to 10%) were tested and it was found that a 6% (w/v) solution gave the best recoveries (results not shown). This concentration was used for all additional work. 3.2. Freeze-drying
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
37
Fig. 2. Lyophilization results for Acidothiobacillus. (A) L-drying results for two strains of A. ferrooxidans. (B) L-drying results for two strains of A. thiooxidans; the symbol * denotes no viable cell recovery. (C) FD results for two strains of A. ferroxidans using GB as a cryoprotectant. Note the values for the Y-axes of the graphs are not all the same.
38
D. Cleland et al. / Journal of Microbiological Methods 58 (2004) 31–38
Table 2 LN Results for halophiles and FeOB; values in parentheses are standard deviation from triplicate samples Organism
GB
Glycerol
Pre-freeze H. volcanii H. salinarium FeOB ES-1
10
Post Freeze 10
6.1 10 (1.4 10 ) 1.4 1010 (1.1 1010) 6.7 107
9
Pre-freeze 9
8.8 10 (7.2 10 ) 3.3 109 (7.9 108) 1.1 106
GB was just as effective as glycerol as a preservative for the halophiles and nearly equivalent at protecting the FeOB. GB has also been used to preserve other strains of halophiles with excellent results, and we have tested it nonquantitatively with a number of bacteria and it appears effective. These results suggest that GB has the potential as a simple, universal cryopreservative that will work with a broad range of prokaryotes.
4. Conclusions Glycine betaine was proven to be an effective cryoprotectant for freeze-drying, liquid-drying, and LN freezing. It is a cheap, simple reagent to make and use. It is nontoxic and, in most cases, will not serve as a growth substrate for prokaryotes. It was more effective at preserving freeze-dried S. pneumoniae after simulated long-term storage than two other common cryopreservatives. As the case with A. thiooxidans and H. pylori illustrates, its efficacy should not be taken for granted, but like all cryo-preservatives, should be checked empirically before being used to preserve a novel group of microorganisms. Acknowledgements We thank Ken Jones of the ATCC operations group for assistance with freeze-drying. This work was supported in part by a grant from the National Science Foundation DBI-0090224. References Angelidis, A.S., Smith, G.M., 2003. Role of the glycine betaine and carnitine transporters in adaptation of Listeria monocytogenes to
10
Post-freeze 9
4.4 10 (3 10 ) 2.5 1010 (3.7 109) 6.7 107
3.2 109 (6.2 108) 4.2 109 (2.2 108) 6 106
chill stress in defined medium. Appl. Environ. Microbiol. 69, 7492 – 7498. Csonka, L.N., Epstein, W., 1996. Osmoregulation. In: Neihardt, F.C., Curtiss R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., (Eds.), Escherichia Coli and Salmonella: Cellular and Molecular Biology. ASM Press, Washington, DC, pp. 1210 – 1224. de Man, J.C., 1975. The probability of most probable numbers. Eur. J. Appl. Microbiol. 1, 67 – 78. Deshnium, P., Gombos, Z., Nishiyama, Y., Murata, N., 1997. The action in vivo of glycine betaine in enhancement of tolerance of Synechococcus sp. strain PCC 7942 to low temperature. J. Bacteriol. 179, 339 – 344. Emerson, D., Moyer, C., 1997. Isolation and characterization of novel iron-oxidizing bacteria that grow at circumneutral pH. Appl. Environ. Microbiol. 63, 4784 – 4792. Galinski, E.A., Tru¨per, H.G., 1994. Microbial behaviour in saltstressed ecosystems. FEMS Microbiol. Rev. 15, 95 – 108. Hubalek, Z., 2003. Protectants used in the cryopreservation of microorganisms. Cryobiology 46, 205 – 229. Ko, R., Smith, L.T., Smith, G.M., 1994. Glycine betaine confers enhanced osmotolerance and cryotolerance on Listeria monocytogenes. J. Bacteriol. 176, 426 – 431. Lengeler, J.W., Drews, G., Schlegel, H.G., 1999. Biology of Prokaryotes. Blackwell, New York. Malik, K.A., 1990. A simplified liquid-drying method for the preservation of microorganisms sensitive to freezing and freezedrying. J. Microbiol. Methods 12, 125 – 132. Martin, D.D., Ciulla, R.A., Roberts, M.F., 1999. Osmoadaptation in Archaea. Appl. Environ. Microbiol. 65, 1815 – 1825. McNeil, S.D., Nuccio, M.L., Hanson, A.D., 1999. Betaines and related osmoprotectants: targets for metabolic engineering of stress resistance. Plant Physiol. 120, 945 – 949. Rhodes, D., Hanson, A.D., 1993. Quaternary ammonium and tertiary sulfonium compounds in higher plants. Annu. Rev. Plant Physiol. 44, 357 – 384. Sakane, T., Kuroshima, K., 1997. Viabilities of dried cultures of various bacteria after preservation for over 20 years and their prediction by the accelerated storage test. Microbiol. Cult. Collect. 13, 1 – 7. Simione, F.P., Brown, E.M. (Eds.), 1991. ATCC Preservation Methods: Freezing and Freeze-Drying, American Type Culture Collection, Rockville, MD, pp. 7 – 13. Xing, W., Rajashekar, C.B., 2001. Glycine betaine involvement in freezing tolerance and water stress in Arabidopsis thaliana. Environ. Exp. Bot. 46, 21 – 28.