Relationship of Critical Temperature to Macromolecular Synthesis and Growth Yield in Psychrobacter cryopegella

The isolation and characterization of P. cryopegella (previously referred to as Psychrobacter sp. 5) from Siberian permafrost has been described (4). S. oneidensis MR-1 was provided by K. Nealson. P. cryopegella was grown in R2A broth (4) plus 3% NaCl (R3) while S. oneidensis was grown in Luria-Bertani Medium (LB) at various temperatures in a VWR model 2005 low-temperature incubator with shaking at 100 rpm (4). To determine growth rate, absorbance at 600 nm was monitored over time using a Spectronic 20D spectrophotometer (Thermo Spectronic, Rochester, N.Y.).

For macromolecular synthesis experiments conducted at temperatures at or above 0°C, P. cryopegella or S. oneidensis was first grown at the experimental temperature for 1 to 7 days, diluted (1:10) into 18 ml of fresh R3 in 30-ml polycarbonate straight-sided jars (Nalgene), and incubated for an additional 1.5 to 24 h prior to the addition of [³H]leucine or [³H]adenine. For experiments conducted at −4 and −7°C, P. cryopegella was grown overnight at 22°C, diluted (1:20) into fresh R3, and incubated for an additional 11 days at −4°C or 6 days at −7°C. Prior to the addition of [³H]leucine or [³H]adenine, 4-ml subsamples were removed to determine optical density at 600 nm and cell numbers by dilution plating. For killed controls, 0.16 g of trichloroacetic acid (TCA) was added.

Triplicate cultures of P. cryopegella were grown in 50 ml of LB3 (10 g of glucose, 30 g of NaCl, and 5 g of yeast extract [all per liter] [pH 7.0]) in 250-ml flasks at 20, 16, 10, 4, 0, −4, and −10 °C with shaking at 100 rpm. The flasks were inoculated with a loopful of cells, and cultures were collected during exponential phase. Cultures grown at 0, −4, and −10°C were inoculated with 5 ml of an overnight culture grown at 16°C and incubated for 5, 6, and 20 days, respectively. Cell weights and glucose concentrations of inocula were also measured. Cells were collected by centrifugation for 10 min at 10,000 × g and 4°C. Cell pellets were washed with 5 ml of phosphate buffer, recollected by centrifugation, resuspended in 0.5 ml of phosphate buffer, transferred to preweighed Al trays, dried at 65°C for 3 days, and weighed. Supernatants were stored at −20°C and passed through 0.2-μm-pore-size filters prior to analysis using Sigma's glucose (GO) assay kit (GAGO-20).

[³H]adenine incorporation rates into DNA and RNA were determined by standard procedures with minor modifications (16). Briefly, [2,8-³H]adenine (specific activity, 24.2 Ci/mmol; Perkin-Elmer) was added to a final concentration of 30 nM and samples were incubated for at least 10% of the expected doubling time at the experimental temperature. Subsamples were taken in duplicate at six or seven time points within this incubation period. At the time points, 1 ml of culture was filtered through a 25-mm-diameter Whatman GF/F filter in a glass filter tower, placed into a 1.5-ml Eppendorf tube, and stored at −80°C. Filters were removed from the freezer, and 1 ml of ice-cold 5% TCA, 1 mg of reagent RNA (baker's yeast; Sigma), and 1 mg of DNA (calf thymus; Sigma) were added. Samples were mixed thoroughly, incubated on ice for 1 h, and spun at 12,000 × g for 1 min, and the supernatant was discarded. Filters were washed three times with 1 ml of ice-cold 5% TCA and three times with 1 ml of ice-cold 95% ethanol. Residual ethanol was evaporated by gentle heating. Filters were mixed with 1 ml of 1 M NaOH, incubated at 37°C for 1 h, incubated on ice for 15 min, mixed with 0.25 ml of 9 M HCl-100% TCA-H₂O (2:1:1), incubated on ice for 15 min, and spun at 12,000 × g for 1 min. For [³H]RNA analysis, 0.625 ml of supernatant was transferred to a clean scintillation vial while the remaining supernatant was discarded. Subsequently, filters were washed two times with 1 ml of ice-cold 5% TCA and two times with ice-cold 95% ethanol. Residual ethanol was evaporated by gentle heating. Filters were mixed with 0.5 ml of 5% TCA, incubated at 95 to 100°C for 30 min, cooled on ice, mixed with 0.5 ml of 1 M NaOH, and spun at 12,000 × g for 1 min. For [³H]DNA analysis, 0.5 ml of supernatant was transferred to a clean scintillation vial containing 0.5 ml of 2 M HCl. Five milliliters of BetaMax Scintillation cocktail (ICN) was added to vials followed by analysis in a Beckman LS6000 scintillation counter for ³H.

[³H]leucine incorporation rates were determined by standard procedures (17). Briefly, [3,4,5-³H(N)]leucine (specific activity, 170 Ci/mmol; Perkin-Elmer) was added to a final concentration of 4.2 nM. Samples were incubated and subsampled as above. At time points, 1 ml of culture was filtered through a 25-mm-diameter Whatman GF/F filter in a glass filter tower, rinsed twice with 3 ml of ice-cold 5% TCA, and rinsed twice with 3 ml of ice-cold 80% ethanol. Filters were placed into clean scintillation vials and air dried. Five milliliters of scintillation cocktail was added to the vials, incubated at room temperature for 2 days to maximize dispersion into the cocktail, and analyzed in a Beckman LS6000 scintillation counter as described above.

Specific growth rates (μ) were calculated by determining the slope of the straight-line portion of exponential growth (on a semilog plot). At least six time points over approximately three doublings were used to determine the slope of the line of best fit. The minimum acceptable r² value of the lines was 0.8 (average r² = 0.959). The average growth rate and standard deviation for each temperature were calculated by standard methods. For normalization purposes (see below), the growth rate at −7°C for P. cryopegella was determined from the line of best fit (μ = 0.308 e^{0.271 T}, r² = 0.987) for data from 4 to −10°C.

Growth yield was calculated as follows: Y_Glc = (m_cells × MW_Glc)/[V × ([Glc]_initial⁻ [Glc]_final)], where m_cells is the grams (dry weight) of cells produced, MW_Glc is the molecular weight of glucose in grams/mole, [Glc] is the concentration of glucose in grams/liter, and V is the volume of the culture in liters. The average growth yield and standard deviation at each temperature were calculated by standard methods. Standard t tests were performed to verify the significance of differences in Y_Glc between temperatures.

Synthesis rates (k) were calculated from the linear portion of the ³H uptake versus time curve (duplicate samples minus killed controls). The slope and r² value for these lines were determined (r² ranged from 0.56 to 0.99, with an average of 0.86, for 21 slopes). The standard error of each slope (SE_b) was calculated as follows: SE_b = SE_Y|X/√SS_XX, where SE_Y|X is the standard error of the regression and SS_XX is the sum of squares of X. The standard error of regression, SE_Y|X, is equal to √[(SS_YY − b × SS_XY)/(n − 2)], where b is the slope, SS_YY is the sum of squares of Y, and SS_XY is the sum of the cross-products of X and Y. The sum of squares of X is equal to \({\sum}(\mathit{X}_{\mathit{i}}{-}\mathit{X}_{mean})^{2}\)(analogous calculation for SS_YY). The sum of the cross-products of X and Y is equal to \({\sum}(\mathit{X}_{\mathit{i}}{-}\mathit{X}_{mean})(\mathit{Y}_{\mathit{i}}{-}\mathit{Y}_{mean})\).

Normalized rates, k^N, were calculated as follows: k^N = k/μ, where k is the ³H uptake rate in zeptomoles/cell/day and μ is the specific growth rate in divisions/day. The standard deviation of the normalized rate was calculated as follows: σ = k^N × √[(SE_b/k)² + (σ_μ/μ)²], where SE_b is the standard error of k and σ_μ is the standard deviation of the growth rate. Standard t tests were performed to verify the significance of differences in k^N between temperatures.

When graphed on a semilog plot (Arrhenius relationship), the temperature dependence of growth rate was linear with two distinct slopes below T_opt (Fig. 1). The slope of the linear relationship changed at 4°C, which is identified as the T_critical. These two slopes describe two domains of growth: low temperature (−10 to 4°C) and high temperature (4 to 22°C). The slopes of the lines of best fit for growth rate were 0.271 (r² = 0.987) for the low-temperature domain and 0.117 (r² = 0.954) for the high-temperature domain. In addition, the temperature dependence of growth yield (grams [dry weight] per mole of glucose) was determined (Fig. 2). Growth yield peaked at 0 to 4°C, or approximately T_critical, and declined rapidly as temperature decreased further. t tests confirmed that the reduced yields measured at 16 and −10°C were significantly different from the yields measured at 0 and 4°C (P < 0.05).

DNA, RNA, and protein synthesis rates (k_DNA, k_RNA, and k_p, respectively) of P. cryopegella were measured from −7 to 16°C. These macromolecular synthesis rates generally followed the Arrhenius relationship (Fig. 3A). However, a deviation from the linear relationship occurred for k_DNA at T_critical, similar to growth rate. The slopes of the lines of best fit for k_RNA and k_p were 0.186 (r² = 0.979) and 0.194 (r² = 0.991), respectively. The slopes of the lines of best fit for k_DNA were 0.248 (r² = 0.975) for the low-temperature domain and 0.120 (r² = 0.979) for the high-temperature domain.

In order to reveal other temperature-dependent processes, synthesis rates were normalized (divided by growth rate), placing synthesis on a per-cell-division basis and removing time from the analysis. Normalized synthesis rates, k^N, in zeptomoles (1 zmol = 10⁻²¹ mol) of [³H]adenine or [³H]leucine incorporated per cell division are shown in Fig. 3B. There is no significant change in kDNAN with temperature (verified via t test, P > 0.25). The temperature dependencies of kRNAN and kPN were more complex (Fig. 3B). Both kRNAN and kPN reached a minimum at approximately 4°C and increased above and below 4°C. t tests indicated that increases in both kRNAN and kPN at the two temperature extremes were significant (P < 0.05) with respect to values at 4°C. However, values of both kRNAN and kPN at 0 and 7°C were not always significantly different from those at 4°C.

For comparison, the DNA and RNA synthesis rates (k_DNA and k_RNA, respectively) and growth rates (μ) of the mesophilic S. oneidensis MR-1 were measured from 4 to 22°C (Fig. 4A). Both the DNA synthesis rate and growth rate generally followed the Arrhenius relationship (k_DNA = 24.8 e^{0.083 T}; r² = 0.925; and μ = 0.690 e^{0.087 T}; r² = 0.971) Normalized respiration rates indicated that T_critical was about 12°C for S. oneidensis (data not shown); however, no obvious deviations from the linear relationship for k_DNA and μ occurred at T_critical. Below 18°C, RNA synthesis rates remained constant with temperature.

Normalized DNA and RNA synthesis rates are shown in Fig. 4B. As seen for P. cryopegella, there was no significant change in kDNANwith temperature (verified via t test, P > 0.10). The temperature dependence of k_RNA^N was more complex (Fig. 4B): kRNAN was approximately constant from 22 to 12°C and increased dramatically below 12°C, T_critical. t tests indicated that the increase in kRNAN below T_critical was significant (P < 0.05) with respect to values at temperatures above T_critical.

The yield of S. oneidensis MR-1 was estimated as the number of cells present at early stationary phase during growth in LB at various temperatures (Fig. 4C). Yield was highest and approximately constant from 14 to 22°C; t tests showed no significant difference between cell yields at these temperatures (P > 0.15). Below 14°C, yield decreased significantly (P < 0.05) as temperature decreased.