Freeze-thaw tolerance and clues to the winter survival of a soil community

Abstract

Although efforts have been made to sample microorganisms from polar regions and to investigate a few of the properties that facilitate survival at freezing or subzero temperatures, soil communities that overwinter in areas exposed to alternate freezing and thawing caused by Foehn or Chinook winds have been largely overlooked. We designed and constructed a cryocycler to automatically subject soil cultures to alternating freeze-thaw cycles. After 48 freeze-thaw cycles, control Escherichia coli and Pseudomonas chlororaphis isolates were no longer viable. Mixed cultures derived from soil samples collected from a Chinook zone showed that the population complexity and viability were reduced after 48 cycles. However, when bacteria that were still viable after the freeze-thaw treatments were used to obtain selected cultures, these cultures proved to be >1,000-fold more freeze-thaw tolerant than the original consortium. Single-colony isolates obtained from survivors after an additional 48 freeze-thaw cycles were putatively identified by 16S RNA gene fragment sequencing. Five different genera were recognized, and one of the cultures, Chryseobacterium sp. strain C14, inhibited ice recrystallization, a property characteristic of antifreeze proteins that prevents the growth of large, potentially damaging ice crystals at temperatures close to the melting temperature. This strain was also notable since cell-free medium derived from cultures of it appeared to enhance the multiple freeze-thaw survival of another isolate, Enterococcus sp. strain C8. The results of this study and the development of a cryocycler should allow further investigations into the biochemical and soil community adaptations to the rigors of a Chinook environment.

FIG. 1.

Design of the cryocycler apparatus that was used for automatically subjecting bacterial cultures to freeze-thaw cycles. (A) Ethylene glycol circulation through the jacketed sample chamber was controlled by a programmable timer and a panel of four valves, two of which were normally open (N.O.) and two of which were normally closed (N.C.), as described in the text. (B) Temperature profile of the cryocycler sample chamber, programmed as described in Materials and Methods, during the warming cycle (boxes) and the cooling cycle (triangles).

FIG. 2.

Decreases in viability of bacterial cultures as a function of the number of freeze-thaw cycles in the cryocycler. Cultures derived from single colonies of P. chlororaphis or E. coli were compared to a mixed culture derived from soil samples collected from a Chinook zone (Calgary soil). Bacteria from the Calgary soil culture that survived 48 freeze-thaw cycles were regrown and again subjected to 48 cycles (Calgary selected). The error bars indicate standard deviations.

FIG. 3.

Colony phenotypes before and after freeze-thaw cycles. The upper plate contained a typical sample (diluted 10⁶-fold) with diverse colony morphologies in the mixed culture prior to freeze-thaw treatment. The lower left plate shows that after 48 freeze-thaw cycles the number of viable bacteria had decreased (sample diluted 10²-fold; Calgary soil [Fig. 2]) and the phenotypic complexity was reduced. The lower right plate shows that when these bacteria were used to initiate freeze-thaw-resistant cultures (Calgary selected [Fig. 2]) and were subjected to 48 additional freeze-thaw cycles, the colony phenotypic diversity was further reduced (sample diluted 10⁵-fold).

FIG. 4.

Decreases in viability of single-colony isolates obtained after 48 serial freeze-thaw cycles (Calgary selected [Fig. 2]) compared to the viability of E. coli cultures. The Chinook isolates were named after the closest relatives in the database (Table 2). The error bars indicate standard deviations.

FIG. 5.

Decreases in viability of single-colony isolates of selected Chinook strains after serial freeze-thaw cycles. Individual isolates that survived 48 freeze-thaw cycles (Calgary selected [Fig. 2]) were used to initiate cultures, and these cultures were subjected to further freeze-thaw treatments (Enterococcus sp. strain C8) or separated from the culture medium and resuspended in medium derived from Chryseobacterium sp. strain C14 cultures (Enterococcus sp. C8 +). Control cultures (E. coli) were also resuspended in medium derived from Chyseobacterium sp. strain C8 cultures (E. coli +). The viability of Chryseobacterium sp. strain C8 cultures (cells and medium) is shown, and the values were similar to the values for Chryseobacterium sp. strain C8 cells resuspended in E. coli medium (not shown). The error bars indicate standard deviations.

FIG. 6.

Inhibition of ice recrystallization in protein samples and bacterial cultures. Samples of protein or bacterial cultures were placed in microcapillaries (diameter, 740 μm), flash frozen, and incubated at −6°C overnight in the jacketed sample holder of the cryocycler (Fig. 1), and they were subsequently examined between crossed polarizing filters. The images are typical of the inhibition of ice recrystallization assay samples. (A) Assays with control proteins. The samples included water (microcapillaries 1 and 2), sample buffer (microcapillaries 3 and 4), 10-fold serial dilutions of fish AFP in buffer (microcapillary 5, 20 mg/ml; microcapillary 6, 2 mg/ml; microcapillary 7, 0.2 mg/ml; microcapillary 8, 0.02 mg/ml; microcapillary 9, 0.002 mg/ml; microcapillary 10, 0.0002 mg/ml), and 10-fold serial dilutions of BSA in buffer (microcapillary 11, 20 mg/ml; microcapillary 12, 2 mg/ml; microcapillary 13, 0.2 mg/ml; microcapillary 14, 0.02 mg/ml; microcapillary 15, 0.002 mg/ml; microcapillary 16, 0.0002 mg/ml). (B) Assays with various Chinook strain and control cultures. Microcapillaries 1 and 2 contained E. coli cultures (∼2 × 10⁻⁸ CFU/ml). Prior to treatment in the cryocycler, the other samples were characterized. Microcapillaries 3 and 4 contained ∼2 × 10⁻⁸ CFU/ml Chyseobacterium sp. strain C14, microcapillaries 5 and 6 contained samples from microcapillaries 3 and 4 diluted 50% with 10% TSB, microcapillaries 7 and 8 contained ∼5 × 10⁻⁸ CFU/ml Acinetobacter sp. strain C3, microcapillaries 9 and 10 contained ∼2 × 10⁻⁸ CFU/ml Buttiauxella sp. strain C, microcapillaries 11 and 12 contained ∼5 × 10⁻⁷ CFU/ml Enterococcus sp. strain C8, and microcapillaries 13 and 14 contained ∼5 × 10⁻⁷ CFU/ml Carnobacterium sp. strain C4. (C) Assays with bacterial cultures and protease. The samples included buffer or medium (capillaries 1 and 2), 0.02 mg/ml fish AFP in medium (capillaries 3 and 4), in proteinase K buffer (capillaries 5 and 6), or in proteinase K buffer and incubated at 37°C with buffer alone (capillaries 7 and 8) or with 2 mg/ml proteinase K (capillaries 9 and 10) before the assay, Chyseobacterium sp. strain C14 cultures (∼5 × 10⁻⁷ CFU/ml) incubated at 25°C with buffer (microcapillaries 11 and 12) or with 2 mg/ml proteinase K (microcapillaries 13 and 14), and the same Chyseobacterium sp. strain C14 cultures incubated at 37°C with buffer (microcapillaries 15 and 16) or with 2 mg/ml proteinase K (microcapillaries 17 and 18).