Physicochemical Parameters for Growth of the Sea Ice Bacteria Glaciecola punicea ACAM 611^T and Gelidibacter sp. Strain IC158

Abstract

The water activity and pH ranges for growth of Glaciecola punicea (a psychrophile) were extended when this organism was grown at suboptimal rather than optimal temperatures. No such extension was observed for Gelidibacter sp. strain IC158 (a psychrotolerant bacterium) at analogous temperatures. Salinity and pH may be primary physicochemical parameters controlling bacterial community development in sea ice.

The Antarctic sea ice environment, in comparison to that of the underlying water column, is dominated by psychrophilic bacteria (those with growth temperature optima of ≤15°C) (3, 15). Theories concerning the dominance of psychrophiles in Antarctic sea ice have ranged from the effect of high organic nutrient levels (12) to the association of psychrophilic bacteria with sea ice microalgae (3, 12). Nichols et al. (25) discussed the shortcomings of these arguments and the potential role of the temperature-salinity regimen in the selection of psychrophilic bacteria in the sea ice environment. However, few physiological investigations have been conducted on psychrophilic bacteria from sea ice, although a number of unique psychrophilic taxa from this environment have been described to date (4–7, 11, 24).

Variations in the environmental temperature-salinity regimen are recognized to be of direct importance when considering the survival and viability of psychrophilic marine bacteria (14, 16, 22, 31). Concurrently, the salinity regimen within sea ice is known to vary widely on both temporal and spatial scales during the ice formation process (1). The pH of sea ice microbial habitats also varies due to biological activity and precipitation of carbonates in highly concentrated brines (9, 10, 12). These phenomena coincide with the establishment within sea ice of a unique bacterial community dominated by psychrophilic organisms (3, 15). The physiological response of psychrophilic bacteria to combined temperature, salinity, and pH stress is therefore of central importance to the understanding of the bacterial sea ice community.

In this study, the growth responses of two bacterial species from Antarctic sea ice, Glaciecola punicea ACAM 611^T (psychrophilic) (7) and Gelidibacter sp. strain IC158 (psychrotolerant) (3), are described in terms of temperature, water activity (a_w), and pH.

Temperature.

The effect of temperature on the rate of growth was determined by using a temperature gradient incubator (TGI; Toyo Kagaku Sangyo, Tokyo, Japan). Incubations (20-ml volumes) were conducted over the range of −2 to 40°C at approximately 1°C intervals (27), utilizing Zobell’s broth (ZB) (33) inoculated with an actively growing culture. Growth was monitored by determining percent transmittance at a wavelength of 540 nm, and growth rates at each temperature were calculated by fitting a modified Gompertz function (19). The model of Ratkowsky et al. (27) was then fitted to the growth rate data by using UltraFit software. The plots of the inverse of the square root of the generation time versus temperature for Gelidibacter sp. strain IC158 and G. punicea are shown in Fig. 1, with derived cardinal temperatures being noted in Table 1. G. punicea can be defined as psychrophilic while Gelidibacter sp. strain IC158 can be classified as psychrotolerant by the criteria of Morita (21). The observation that the psychrophilic species G. punicea retained a higher growth rate at a low (suboptimal) temperature is consistent with the general trend of the response of psychrophilic or psychrotolerant organisms to temperature (14) and suggests that the dominance of psychrophilic organisms within sea ice is a result of their higher growth rates at low temperatures. However, Ferroni and Kaminski (8) failed to find a correlation between the numerical predominance of psychrophilic or psychrotolerant populations and growth rate (influenced by temperature) in a freshwater lake. Further, the fact that Antarctic seawater and the bottom ice region, where the majority of the bacterial biomass resides (32), experience a similar temperature regimen (ca. −2°C) argues against temperature being the sole selective pressure.

FIG. 1.

Plot of the inverse of the square root of the generation time (in minutes) versus temperature for G. punicea ACAM 611^T (a) and Gelidibacter sp. strain IC158 (b).

TABLE 1.

Cardinal growth values and theoretical growth ranges for the sea ice bacteria investigated

Parameter	Value ± SE for:
	G. punicea ACAM 611T	Gelidibacter sp. strain IC158
Cardinal tempa
Tmin	−12.1 ± 2.2°C	−6.2 ± 1.9°C
Toptd	15.4°C	24.1°C
Tmax	26.4 ± 1.5°C	38.1 ± 0.8°C
TRG tempe	38.7 ± 3.7°C	44.3 ± 2.7°C
aws, pHs, suboptimal tempb
awmin	0.932 ± 0.001	0.930 ± 0.003
awoptd	0.980	0.993
awmax	0.995 ± 0.000	1.002 ± 0.006
TGR awe	0.064 ± 0.015	0.073 ± 0.010
pHmin	5.90 ± 0.11	6.16 ± 0.03
pHoptf	7.3–8.1	7.2–8.0
pHmax	9.56 ± 0.12	9.11 ± 0.04
TRG pHe	3.66 ± 0.23	2.95 ± 0.07
aws, pHs, optimal tempc
awmin	0.958 ± 0.003	0.931 ± 0.005
awoptd	0.981	0.995
awmax	1.000 ± 0.001	1.001 ± 0.001
TGR awe	0.042 ± 0.004	0.070 ± 0.006
pHmin	6.16 ± 0.07	6.12 ± 0.03
pHoptf	7.4–8.1	7.4–8.2
pHmax	9.35 ± 0.10	9.45 ± 0.03
TRG pHe	3.19 ± 0.17	3.33 ± 0.06

^aCardinal temperatures from cultures in ZB (a_w, 0.983; pH 7.0). 

^bCardinal a_w and pH values from cultures at suboptimal temperature (8.0°C) in ZB with various SWS concentrations. 

^cCardinal a_w and pH values from cultures at optimal temperatures (15.0°C, G. punicea; 25.0°C, Gelidibacter sp. strain IC158) in ZB with various SWS concentrations. 

^dDerived value. 

^eTGR, theoretical growth range (T_max − T_min, a_wmax − a_wmin, or pH_max − pH_min). 

^fEstimated range. 

Water activity.

The effect of the concentration of artificial seawater salts (SWS) (30) on the growth rates of G. punicea and Gelidibacter sp. strain IC158 was investigated at suboptimal (8.0°C) and optimal (15.0 and 25.0°C, respectively) growth temperatures. The media used for both bacteria consisted of combinations of three modified ZBs (0, 100, and 200 ppt SWS; for the 200-ppt SWS, insoluble components were removed by filtration prior to use). Various ratios of these broths were mixed to produce SWS concentrations within the range of 0 to 130 ppt. The a_ws of resultant broths were determined by using an Aqualab CX2 dew point instrument (Decagon Devices). Inoculation, measurements, and data analysis for both 8 and 25°C Gelidibacter sp. strain IC158 experiments were conducted as described above with the TGI held isothermal at the incubation temperature. The 15.0°C experiment for G. punicea was conducted as described above apart from the use of 40-ml cultures grown in 125-ml side-arm flasks which were incubated in water baths. The 8.0°C G. punicea data set is a combination of data from two experiments, the first using eight side-arm flasks as for the 15.0°C determinations and the second being a TGI experiment as described previously. Growth rate data were fitted to the modified square root-type model of Miles et al. (20) for a_w. The significance of changes in theoretical minimum a_ws for growth (a_wmin) and maximum a_ws for growth (a_wmax) was determined by using an approximate Z test.

At optimal and suboptimal growth temperatures, the psychrophilic species exhibited smaller a_w ranges for growth than did the psychrotolerant species (Fig. 2, Table 1). G. punicea required the presence of SWS for growth at 8°C but not at 25°C, representing a significant (P < 0.01) increase in a_wmax. In contrast, Gelidibacter sp. strain IC158 did not require SWS for growth at either temperature and demonstrated no significant change (P > 0.05) in a_w range between growth temperatures. The maximum concentration of SWS at which G. punicea grew was also a function of the growth temperature, with a significant (P < 0.001) decrease in a_wmin being evident at low temperatures. The interactive effects of temperature and a_w on microbial growth are well documented for many food spoilage-causing and pathogenic bacteria. In the majority of such cases, less-stringent a_ws may be tolerated with decreasing temperature, while increased tolerance of a_w extremes is usually apparent over the optimum growth temperature (T_opt) region (18). While observed values of growth parameters are known to vary with different combinations of environmental constraints (18), the theoretical minimum temperature (T_min), minimum pH (pH_min), and a_wmin values are held to be independent of culture conditions (2, 18, 19, 27). This is not the case for G. punicea (Fig. 2), implying that there is an adaptive effect of this bacterium to low a_w at low temperatures.

FIG. 2.

Plot of the inverse of the generation time (in minutes) versus a_w for G. punicea ACAM 611^T cultured at suboptimal (8.0°C [+]) and optimal (15.0°C [○]) growth temperatures (a) and Gelidibacter sp. strain IC158 cultured at suboptimal (8.0°C [+]) and optimal (25.0°C [○]) growth temperatures (b). For medium details, refer to the text.

The interaction of low temperature and low a_w is recognized as the controlling factor for microbial growth or no growth in sea ice (25). The influence of salinity in controlling the species composition of diatom assemblages in sea ice is also well established (28, 29). Studies of the temperature-salinity interactions of psychrophilic bacteria have concentrated primarily on the effect of low-a_w conditions on maximum temperature (T_max) (13, 21, 23, 31). Such studies hold little relevance to the sea ice environment, in which bacteria are not subjected to temperatures approaching T_max values. Further, the majority of such studies have utilized sodium chloride as the sole humectant. Many marine bacteria, such as G. punicea and Gelidibacter sp. strain IC158, require sea salt mixtures for optimal growth (3, 7). The present study employed SWS for the adjustment of a_w in recognition of this fact. While the a_w ranges for growth at the suboptimal growth temperature were similar (Fig. 2), G. punicea exhibited a lower optimal a_w (a_wopt) and maintained a higher growth rate over the suboptimal a_w region. It is implied that this is an effect of a_w adaptation in G. punicea rather than a comparative effect of temperature, since the two bacteria exhibited similar growth rates at 8°C (when grown at an a_w of 0.980 [Fig. 1]). However, the possibility that this effect was due to the more-favorable a_w conditions for G. punicea in the temperature-variable growth experiment cannot be excluded since Gelidibacter sp. strain IC158 was subsequently found to have a consistently higher a_wopt than G. punicea.

A significant decrease in the a_wmin of Escherichia coli (from ca. 0.967 to ca. 0.955) with the addition of the compatible solute betaine to cultures grown in minimal medium has been reported (17). This demonstrates a physiological adaptation (the uptake of a compatible solute) directly affecting the observed and theoretical a_w limits for growth of a bacterium. It is plausible that such an effect can explain the results observed in this study. For example, it is possible that G. punicea contains a cold-activated compatible solute transport system, allowing the bacterium to extend its a_w growth range at low temperatures.

pH.

The effect of pH on the growth rate of G. punicea was determined in ZB modified by the addition of 0.088 M HCl (for acidic conditions) or 0.25 M NaOH (for alkaline conditions) over a pH range of 6.2 to 9.1 (in high-pH broths, precipitated salts were removed by filtration prior to use). Cultures (40 ml) were grown in 125-ml side-arm flasks incubated in water baths. Inoculation, measurements, and data analysis were conducted as described above with the exception that growth rate data were fitted to a model derived from that of Presser et al. (26). The significance of changes in theoretical pH_min and in theoretical maximum pHs for growth (pH_max) was determined by using an approximate Z test.

Both species can be classed as neutralophilic (Fig. 3; Table 1). While the pH_min of Gelidibacter sp. strain IC158 was unaffected by the growth temperature, there was a significant (P < 0.05) increase in pH_min for G. punicea between 8.0 and 15.0°C. This was similar to the noted effect on a_wmin, suggesting a higher tolerance of low pH along with lower a_w by the psychrophilic species at low temperatures. Conversely, there was no significant (P < 0.05) increase in the pH_max of G. punicea with increased growth temperature, but the pH_max of Gelidibacter sp. strain IC158 did increase significantly (P < 0.001) between 8.0 and 25.0°C.

FIG. 3.

Plot of the inverse of the generation time (in minutes) versus pH for G. punicea ACAM 611^T cultured at suboptimal (8.0°C [+]) and optimal (15.0°C [○]) growth temperature (a) and Gelidibacter sp. strain IC158 grown at suboptimal (8.0°C [+]) and optimal (25.0°C [○]) growth temperatures (b). For medium details, refer to the text.

Further studies are required to elucidate the mechanistic basis for the observations of this study. The present data suggest that psychrophilic sea ice bacteria may have adapted to endure wider salinity and pH ranges for growth at suboptimal temperatures than psychrotolerant species from the same environment. The implication is that the salinity and pH within sea ice may be primary physicochemical parameters controlling bacterial community development.