Collection Sites, Preacclimation, and Photoperiod

Cr. rosae v. psychrophila strains CU 204 (UTEX SNO 11) and CU 479A (UTEX SNO 56) were collected from Whiteface Mountain, NY, and CU 381F (UTEX SNO 50) and CU 381G (UTEX SNO 51) from the White Mountains, Arizona. Cr. tughillensis strains CU 581A (UTEX SNO 86), CU 581C (UTEX SNO 88), CU 581D (UTEX SNO 89), CU 582A (UTEX SNO 90), CU 582C (UTEX SNO 91), and CU 582D (UTEX SNO 92) were collected from Whetstone Gulf State Park, Tughill Plateau, NY. Cr. chenangoensis strain CU 722B (UTEX SNO 147) was collected from the Chenango Valley, Hamilton, NY. The sampling network of these taxa were restricted because Chloromonas chenangoensis and Chloromonas tughillensis are known from only one site each in NY (Hoham et al., 2006). Chloromonas rosae v. psychrophila is worldwide in its distribution, but cultures of it exist only from western and eastern North America and two cultures from each region were included in this study. For more information on collection sites, which include dates and elevations, see the UTEX web site at  http://www.bio.utexas.edu/research/utex/.

Prior to the experiments, strains were maintained on modified M-1 agar medium (Hoham et al., 2006) for 4 weeks at 4°C. For acclimation to temperatures used in the experiments, see the EXPERIMENTAL SETUP section below. Each temperature regime was established in Percival® Model 1–30 BLL Growth Chambers. For Cr. rosae v. psychrophila, GE Cool White fluorescent lighting was at an irradiance level of 75–140 µmol photons m⁻² s⁻¹ with a 16∶8 hour (light∶dark) photoperiod, which followed procedures of Hoham and Mohn (1985) and Hoham et al. (1998) for Whiteface Mountain, NY. For Cr. tughillensis and Cr. chenangoensis, blue light (GE F20T12.B-Blue fluorescent tubes) was used at irradiance levels of ∼95 and ∼115 µmol photons m⁻² s⁻¹, respectively, with a 14∶10 photoperiod for both species, which followed procedures of Hoham et al. (2006).

Growth Chamber Setups, Irradiance, and Cell Counts

Glassware and media were autoclaved and preacclimated along with sterile pipettes for each experiment. Using sterile technique, cells were washed from agar plates with M-1 medium, and aliquots were placed into Erlenmeyer flasks. Cell concentrations were enumerated using hemacytometers to start experiments at 8 × 10³ for Cr. rosae v. psychrophila and 10 × 10³ cells mL⁻¹ for Cr. tughillensis and Cr. chenangoensis, respectively. A 15-mL cell suspension was poured into each test tube, caps were loosely tightened for gas exchange, and replicate test tubes for each strain were arranged along the periphery inside 500 mL beakers. Empty test tubes were placed in the center of the beaker and in equally spaced intervals along the outside to keep experimental tubes vertical and to minimize light variations. Replicates were in triplicate for Cr. rosae v. psychrophila and quadruplicate for Cr. tughillensis and Cr. chenangoensis. Beakers were arranged on white blotters for maximum reflectivity of light.

A Licor LI-1935a 3-D “Spherical” sensor attached to a LI-1000 “Datalogger” was used to measure µmol photons m⁻² s⁻¹ irradiance prior to experiments and before and after counts were taken. The sensor was placed equidistant from and in the middle of the beakers to obtain the irradiance level within each chamber. During measurements the growth chamber door was closed until the light reading stabilized, and the shelf was raised or lowered to give the approximate irradiance level needed.

Cells were observed through Zeiss® phase-contrast microscopes and enumerated using hemacytometers weekly for four weeks or the maximum time that populations may exist in field samples. Prior to each enumeration, cells were resuspended five times with a 1-mL sterile pipette to achieve an equal suspension of cells, and this was performed in each chamber with the door open quickly to minimize any changes in temperature or lighting. Four people enumerated separate hemacytometer fields and an average was taken to avoid sampling bias. Absorbance data at 440 nm (not shown) were compiled for all experiments and the results were similar to cell count data.

Experimental Setup: Temperature Optima, Acclimation, and Nutrients

Three replicate cultures of each of the four strains of Cr. rosae v. psychrophila were grown at 4, 10, 15, and 20°C. Four replicate cultures of each of the six strains of Cr. tughillensis were grown at 5, 10, 15, and 20°C. In a separate experiment, four replicate cultures of each of these six strains were grown at 2.5, 5, and 7.5°C. Four replicate cultures of the Cr. chenangoensis strain were grown at 2.5, 5, 7.5, and 10°C. All of these cultures were preacclimated for two weeks at each temperature prior to beginning experiments. For all three taxa, 0°C was not tested due to the limits of the growth chambers employed. For each data set, a series of two-way ANOVAs and one-way ANOVAs with post-hoc Tukey tests were used where appropriate to investigate variation in average cell counts at four weeks among strains, growth temperatures, and combinations of strains and growth temperatures.

Three replicate cultures of Cr. rosae v. psychrophila CU 204 and CU 479A were grown at 4, 10, 15, and 20°C from two different starting conditions. In the first, cells were moved from 4°C to the other temperatures (nonacclimated treatments). In the second, cells were acclimated for two weeks at the other temperatures prior to beginning experiments. Because of significant among-strain variation (see below), these data were analyzed separately for each strain using two-way ANOVAs with growth temperature and acclimation treatment as fixed factors. Four replicate cultures of each of the six strains of Cr. tughillensis were grown under four different acclimation-growth temperature treatments. Cells were acclimated at either 2.5 or 7.5°C for five months prior to experiments and then grown at either 2.5 or 7.5°C for four weeks. Nutrients (nitrate, ammonium, and sulfate) were analyzed in these experiments using Dionex Ion Chromatography after five weeks or one week after final cell counts because there was not enough time to analyze both after four weeks. Due to the significant among-strain variation (see below), these data were analyzed using a series of two-way ANOVAs with acclimation and growth temperature as fixed factors, and then a series of one-way ANOVAs with post-hoc Tukey tests were used to investigate differences among the four acclimation-growth temperature combinations separately for each strain.

Temperature Optima

Separate two-way ANOVAs were performed for Cr. rosae v. psychrophila and Cr. tughillensis with cell count as the dependent variable and strain (Cr. rosae v. psychrophila, four levels; Cr. tughillensis, six levels) and growth temperature (four levels) as fixed factors. Each analysis showed that there was a significant main effect of strain for each species (Table 1); therefore, the data were analyzed for each strain of a particular species separately using a series of separate one-way ANOVAs.

TABLE 1

Results of two-way ANOVAs for Chloromonas rosae v. psychrophila (A) and two studies of Chloromonas tughillensis (B and C) with cell count as the dependent variable and strain and growth temperature as fixed factors.

For Cr. rosae v. psychrophila, average cell counts significantly varied among temperature groups for each strain (CU 204: F_3,11  =  20.81, p < 0.001; CU 479A: F_3,11  =  29.09, p < 0.001; CU 381F: F_3,11  =  28.01, p < 0.001; CU 381G: F_3,11  =  6.13, p < 0.050). As suggested by the two-way ANOVA, strains varied with respect to the temperature that yielded the highest average cell counts (Fig. 1A). Post-hoc Tukey tests showed that average cell counts at 20°C were significantly lower than all other temperature groups for all strains, but did not identify a statistically significant temperature optimum for any strain.

FIGURE 1

Mean cell counts (+1 SE) of growth after four weeks for different strains of Cr. rosae v. psychrophila (A), Cr. tughillensis (B and C), and Cr. chenangoensis (D) at different temperatures (°C).

In the first experiment with Cr. tughillensis, average cell counts significantly varied among temperature groups for each strain (CU 581A: F_3,15  =  261.06, p < 0.001; CU 581C: F_3,15  =  433.96, p < 0.001; CU 581D: F_3,15  =  45.27, p < 0.001; CU 582A: F_3,15  =  266.10, p < 0.001; CU 582C: F_3,15  =  175.50, p < 0.001; CU 582D: F_3,15  =  92.49, p < 0.001). Although strains varied with respect to overall average cell counts (Fig. 1B), post-hoc Tukey tests showed that average cell counts at 5°C were significantly higher than all other temperature groups for each strain. In the second experiment with Cr. tughillensis, average cell counts again significantly varied among temperature groups for each strain (CU 581A: F_2,11  =  64.44, p < 0.001; CU 581C: F_2,11  =  41.43, p < 0.001; CU 581D: F_2,11  =  12.28, p < 0.010; CU 582A: F_2,11  =  63.49, p < 0.001; CU 582C: F_2,11  =  7.61, p < 0.05; CU 582D: F_2,11  =  59.85, p < 0.001). Post-hoc Tukey tests showed that different strains had different optimum temperatures (Fig. 1C); the optimum temperature for strains CU 581A and CU 581D was 5°C, whereas the optimum temperature for the other four strains was 2.5°C.

For Cr. chenangoensis, average cell counts significantly varied among temperature groups (CU 722B: F_3,15  =  526.50, p < 0.001). A post-hoc Tukey test showed that average cell counts at 2.5 and 5°C were not different from each other, but were significantly higher than average cell counts at 7.5 and 10°C (Fig. 1D).

Acclimation

Separate two-way ANOVAs with growth temperature (four levels) and acclimation status (two levels) for Cr. rosae v. psychrophila strains CU 204 and CU 479A showed that average cell counts significantly varied among temperature groups for both strains (CU 204: F_2,12  =  78.13, p < 0.001; CU 479A: F_2,12  =  10.71, p < 0.010), but there was no significant main effect of acclimation (CU 204: F_1,12  =  0.525, p > 0.100; CU 479A: F_1,12  =  2.66, p > 0.100) or interaction between growth and acclimation temperature (CU 204: F_2,12  =  0.714, p > 0.100; CU 479A: F_2,12  =  0.373, p > 0.100). Within temperature groups, average cell counts between the acclimated and not-acclimated groups were strikingly similar (Figs. 2A and 2B).

FIGURE 2

Mean cell counts (+1 SE) of growth after four weeks for Cr. rosae v. psychrophila strain CU 204 (A) and strain CU 479A (B) at different temperatures (°C). Closed bars denote cultures that were acclimated at each temperature; open bars denote cultures that were not acclimated.

In the experiment with Cr. tughillensis, separate two-way ANOVAs for each strain with growth temperature (two levels) and acclimation temperature (two levels) as fixed factors showed significant effects of growth temperature on average cell counts, but mixed results of acclimation temperature on cell counts (Table 2). There was substantial variation among strains with respect to how growth and acclimation temperatures affected cell counts (Fig. 3A). To investigate this further, a coded variable was created for each strain with four levels to represent the combination of growth and acclimation temperature. Then, a series of separate one-way ANOVAs were used for each strain and post-hoc Tukey tests were performed to identify homogeneous subsets. For strains CU 581A, 581C, 582A, and 582C, growth at 2.5°C following acclimation at 7.5°C yielded significantly higher average cell counts than all other treatment combinations. There was no significant difference between average cell counts at 2.5°C following acclimation at either 2.5 or 7.5°C for strain CU 581D. Growth at 7.5°C following acclimation at 2.5°C was significantly lower than all other groups for strain CU 582D, but average cell counts did not differ among the other three treatment combinations.

TABLE 2

Results of two-way ANOVAs for Chloromonas tughillensis strains with cell count as the dependent variable and growth temperature and acclimation temperature as fixed factors.

FIGURE 3

Mean cell counts (+1 SE) of growth after four weeks for different strains of Cr. tughillensis under combinations of growth (first temperature) and acclimation (second temperature) (°C) (A) and after five weeks for nitrate (NO₃), ammonium (NH₃), and sulfate (SO₄) content (+1 SE) of strains grown at 2.5°C following acclimation at 7.5°C, and of strains grown at 7.5°C following acclimation at 2.5°C (B).

Differences in nitrate, ammonium, and sulfate concentrations for Cr. tughillensis cultures grown at 2.5°C following acclimation at 7.5°C and cultures grown at 7.5°C following acclimation at 2.5°C were assessed using separate two-way ANOVAs for each nutrient with temperature treatment (two levels) and strain (six levels) as fixed effects. There were significant temperature and strain main effects and interaction effects for nitrate (temperature: F_1,36  =  1837.61, p < 0.001, strain: F_5,36  =  85.31, p < 0.001; interaction: F_5,36  =  41.50, p < 0.001), ammonium (temperature: F_1,35  =  385.94, p < 0.001; strain: F_5,35  =  42.68, p < 0.001; interaction: F_5,35  =  27.03, p < 0.001), and sulfate (temperature: F_1,36  =  1012.95, p < 0.001, strain: F_5,36  =  66.04, p < 0.001, interaction: F_5,36  =  45.70, p < 0.001). Separate one-way ANOVAs with post-hoc Tukey tests showed that nitrate concentrations were lower for each strain grown at 2.5°C following acclimation at 7.5°C than at 7.5°C following acclimation at 2.5°C (p < 0.001 in all cases; Fig. 3B), and that the magnitude of the difference varied among strains (significant interaction term in two-way ANOVA). These analyses showed similar results for ammonium concentration for all strains (p < 0.001; Fig. 3B) except CU 582D, for which there was no difference (p > 0.100). Sulfate concentrations were similar to those of nitrate concentrations (p < 0.001 in all cases; Fig. 3B).

Temperature Optima

Phylogenetic studies using 18S rDNA and rbcL gene sequence analysis revealed that snow algal species of Chlamydomonas (Cd.) and Chloromonas probably invaded the snow habitat two or three times (Hoham et al., 2002). These snow species were likely derived evolutionarily from temperate species and align with other cold tolerant species of these genera that live in alpine soil, peat, and lakes. Little is known, however, about the origins and evolutionary history of other algal genera found in snow.

Temperature optima studies of snow algae have been conducted qualitatively and quantitatively (Table 3). Chlainomonas rubra and Chlainomonas kolii shed their flagella at temperatures above 4°C when observed on a laboratory cooling stage and their protoplasts were damaged when cells were maintained at just below 0°C for a few hours (Hoham, 1975a). Lab growth experiments indicate that temperature optima of snow algae vary from 1 to 10°C with several species not being able to survive at temperatures above 10°C, which are termed true or obligate cryophiles.

TABLE 3

Temperature ranges and optima for snow algal taxa.

Red cells of Chlamydomonas were more freeze tolerant than green cells because of their higher monounsaturated fat contents (Bidigare et al., 1993). In red cells 80% of fatty acids were monounsaturated, which allowed them to stay intact during a −70°C freezing period, whereas green cells, which had 72% saturated fatty acids, suffered extensive damage. Near-freezing temperature optima in snow algae are reliant upon their ability to maintain lipid fluidity. The snow alga, Cd. nivalis, was more tolerant to cold stress (freezing injury, shrinkage, and rehydration) than were three temperate species of Chlamydomonas (Morris et al., 1979), and the suggested mechanism was the greater number of unsaturated fatty acids in cells of Cd. nivalis that allowed for greater membrane fluidity (Morris et al., 1979, 1981).

In this study, strains of Cr. rosae v. psychrophila were grown from 4 to 20°C with best growth at 4, 10, and 15°C with no significant differences in growth at these temperatures (Fig. 1A), and at 20°C strains grew significantly less than at the other temperatures. Strains of Raphidonema nivale grew from 1 to 15°C and optimally at 5°C and of Cylindrocystis brébissonii from 1 to 20°C and optimally at 10°C (Hoham, 1975a). However, none of these taxa were considered obligate snow algae or cryophiles because of their ability to grow at temperatures above 10°C, which might allow them to grow on wet soil, peat, or in lakes after snowmelt.

In the first experiment with Cr. tughillensis (Fig. 1B), all strains had significantly higher cell counts at 5°C vs. 10, 15, or 20°C, and cultures did not survive at the two higher temperatures after a few weeks. In the second experiment with Cr. tughillensis (Fig. 1C), all strains except CU 581D had cell counts that were significantly higher for growth at 2.5 vs. 7.5°C. However, strains CU 581C, 582A, 582C, and 582D showed significantly higher cell concentrations at 2.5 vs. 5.0°C where strains CU 581A and CU 581D were significantly higher at 5.0°C vs. 2.5 and 7.5°C. These results indicate subtle differences in temperature optima among strains of Cr. tughillensis with optimum growth at either 2.5 or 5.0°C. For Cr. chenangoensis strain CU722B, cell counts were not significantly different at 2.5 and 5.0°C but were significantly higher than at 7.5 or 10°C (Fig. 1D), and cultures died after a few weeks at 10°C (not shown in Fig. 1D).

Previous studies of snow algae showed an optimum temperature of 5°C for Chromulina chionophilia (Stein, 1963); 4°C for Raphidonema tatrae (Hindák and Komárek, 1968); 3°C for Cr. polyptera, Cr. rubroleosa, and Chlorosarcina antarctica (Ling, 1996); 2 to 10°C for Cryptomonas frigoris (Javornický and Hindák, 1970), Desmotetra antarctica and Desmotetra aureospora (Ling, 2001); 1 to 4°C for Chlainomonas rubra and Chlainomonas kolii (Hoham, 1975a); and 1°C for Cr. pichinchae (Hoham, 1975a) (Table 3). Cells of Cr. pichinchae that survived at 10°C were abnormal and in large non-motile clumps (Hoham, 1975a, 1975b). All of these taxa are obligate cryophiles because they have optimal growth at temperatures below 10°C and do not grow above that temperature (Hoham, 1975a, 1980). Other taxa listed in Table 3 may be obligate cryophiles such as Cd. nivalis where cells were cultivated at 4°C, but other temperatures for growth were not tested (Czygan, 1970).

In temperature studies of non-cryophilic algae showing psychrophilic character, two strains of an Antarctic Chlamydomonas isolated from floating ice, Cd. sp. ICE-L and Cd. sp. ICE-W, grew between 0 and 10°C, propagated in the ice, and did not survive at ambient temperatures above 15°C (Liu et al., 2006), which is similar to non-obligate snow algae. A similar temperature range for growth occurred in Cd. raudensis isolated from beneath ice in Lake Bonney, Antarctica (Pocock et al., 2004). The marine euglenoid, Tetreutreptia pomquetensis isolated from the Gulf of St. Lawrence, grew between 0 and 7°C, but did not survive at 10°C (McLachlan et al., 1999), which is similar to obligate snow algae. Several unsaturated fatty acids were identified from this alga, some of which were unusual, and their high content supported that the degree of unsaturation was indicative of low temperature and membrane integrity as reported for snow algae. Growth of two marine diatoms, Thalassiosira antarctica and Nitzschia frigida, at temperatures below 0°C was enhanced by increased salinity levels (Aletsee and Jahnke, 1992).

Acclimation

For Cr. rosae v. psychrophila, there were no significant differences acclimating cultures of Adirondack, NY, strains CU 204 and CU 479A for two weeks prior to experiments vs. using non-acclimated strains that were moved from the 4°C growth chamber directly to the temperatures tested (4, 10, 15, and 20°C) (Figs. 2A, 2B). Consequently strains CU 381A and CU 381B of Cr. rosae v. psychrophila from AZ were not tested for acclimation. When comparing strains of Cr. tughillensis in acclimation experiments (Fig. 3A), four strains (CU 581A, 581C, 582A, and 582C) had significantly higher cell counts at 2.5 (7.5)°C when compared to 2.5 (2.5), 7.5 (2.5), and 7.5 (7.5)°C. When the strains were switched to the lower temperature, their growth rates greatly increased, which supports that optimal growth of Cr. tughillensis is closer to 2.5°C than to 7.5°C; however, population sizes varied among strains. After four weeks at 2.5 (7.5)°C, strain CU 582A reached a population size of 8.2 × 10⁵ cells mL⁻¹ while CU 582D was at 3 × 10⁵ cells mL⁻¹. For strain CU 581D, there was no significant difference between average cell counts at 2.5°C following acclimation at either 2.5 or 7.5°C, which had significantly higher cells counts than the other two treatments. For strain CU 582D, cell counts at 7.5°C following acclimation at 2.5°C were significantly lower than the other three treatments, which did not differ from one another. In a different study, populations of Cd. nivalis showed optimal uptake of ¹⁴HCO₃⁻ at −3, 0, 10, or 20°C, which was attributed to responses of different temperature strains (Mosser et al., 1977).

In the nitrate, ammonium, and sulfate experiments, there were significantly lower cell counts among strains of Cr. tughillensis grown in the acclimation experiments at 2.5 (7.5)°C vs. 7.5 (2.5)°C except for strain CU 582D in the ammonium experiments, which reflects the higher cell densities at 2.5 (7.5)°C (Fig. 3B). Snow algae deplete nutrients as their cell concentrations increase, and nutrient depletion is a limiting factor for growth. Even though concentrations of the nutrients differed in the two temperature regimes, they were not limiting at either temperature after four weeks. Chloromonas pichinchae grew optimally in snow meltwater containing coniferous leaves, bark, and pollen, which contributed nitrogen and phosphorus (Hoham, 1976, 1980). Populations and strains of Cd. nivalis and Cr. pichinchae probably have different nutrient requirements because concentrations of the same elements differed at three snow collection sites where they were found (Hoham and Duval, 2001), which supports the varying concentrations of nitrate, ammonium, and sulfate among strains of Cr. tughillensis observed here.