ABSTRACT 1 examined the changes in levels of endogenous heat shock protein 70 (Hsp 72 and Hsp 74) in the Tegula congeners T. brunnea, T. pulligo, and T.funebralis in response to cold shock. I also monitored behavioral changes caused by cold exposure. To examine Hsp levels, snails were first placed in either 1° or 4° C (40.5° C) seawater for either 2 or 10 h, then placed back in ambient temperature (13° C) seawater. Levels of Hsp expression 2 and 10 h after the end of the shock were quantified via western analysis. These levels were compared to those of snails not exposed to cold stress. In individuals facing 10 h cold exposures, levels of Hsp 72 after 2 h of recovery in general did not vary substantially from pre-exposure levels, but decreased significantly after 10 h of recovery. These data suggest that different exposure lengths may confer different signals that activate distinct physiological responses to cold shock. The behavioral study found that following cold (1-3° C) exposure, the subtidal congeners T. brunnea and T. pulligo were more likely to enter a chill coma (foot retracted, operculum firmly closed) than T. funebralis. These behavioral differences could indicate that the subtidal species would be less capable than T. funebralis of responding to environmental cues and conditions following cold shock. INTRODUCTION In an age of increasingly varying weather patterns and global climate change, the mechanisms by which organisms cope with thermal stress merit study. While much work has been devoted to the ability of organisms to withstand heat stress, relatively little has looked at the lower thermal tolerances of species. In this study, I examine the cold stress response of the Tegula congeners T. funebralis, T, brunnea, and T. pulligo. The Tegula species used provide an ideal study system for examining important aspects of thermal tolerance and cold stress. As they are closely related (though there are some concerns about T. pulligo, see Hellberg 1998), any comparisons between their thermal biology are unlikely to be confounded by differences in phylogeny. Also, one would expect them to experience different ranges of temperatures as they live at different tidal heights and have different northern distributions. T. funebralis inhabits the low- to mid- intertidal zone, and has a northern distribution that extends to Vancouver Island, British Columbia (Abbott and Haderlie 1980). T. brunnea lives in the low intertidal to subtidal zone, and ranges northward to Cape Arago, Oregon (Abbott and Haderlie 1980). T. pulligo lives exclusively in the subtidal zone, and ranges as far north as Sitka, Alaska (Abbott and Haderlie 1980). T. funebralis has been shown previously to be more capable of handling heat stress than T. brunnea (Tomanek 2002). However, at the northern extreme of its range T. funebralis should also experience colder temperatures than either T. brunnea or T. pulligo during emersion. To date, little research has been conducted to explain this aspect of the thermal biology of the Tegula congeners. This study looks at the patterns of Hsp 70 expression in response to different cold stresses in the different species, as well as the variation in behavioral response of the species to cold shock. MATERIALS AND METHODS Chill Coma Experiment This study subjected snails to four different temperature treatments. In two of the treatments snails were placed in containers filled with ambient temperature (13° + 1° C) seawater. These containers were placed in either an ice bath (1" treatment) or a cold room (air temperature 4° C, 2“ treatment), and the seawater was allowed to equilibrate thermally with its surroundings for 12 hours. Seawater in the cold room reached temperatures of 8° C(+ 1° C) and 6° C(+1° C) after 6 and 12 hours, respectively. Seawater in the ice bath reached temperatures of 6°C (+1° C) and 3° C (+ 1° C) after 6 and 12 hours, respectively. In the other two treatments, snails were exposed directly to either 1° C (+1° C) or 4° C (40.1° C) air for 6 hours. Five large adults of each species were used for each treatment. Snails were returned to ambient temperature seawater at the end of the treatments. Individuals were observed 20 minutes and 24 hours following the end of the cold exposures. During each observation, individuals were categorized as being dead, active, or in a chill coma. For the purposes of the experiment, animals were dead if they did not exhibit a foot withdrawal response, active if their operculum was not closed and they exhibited a foot withdrawal response, and in a chill coma if their operculum was closed firmly. Analysis of Hsp 70 Expression All snails to be used for the experiment were placed in an acclimation tank with ambient temperature (13° C + 1° C) seawater for five days prior to cold exposure. Individuals were subjected to four different exposure treatments: 1° C seawater (40.5° C) for 2 hours. 1° C seawater (40.5° C) for 10 hours, 4° C seawater (40.5° C) for 2 hours, and 4° C seawater (40.5° C) for 10 hours. Ten large adults of each species were used for each treatment. Following exposure, the animals were placed in ambient temperature seawater. To ensure that individuals stayed fully immersed in water during the entire experiment, snails were placed inside small mesh-enclosed containers inside larger seawater tanks for all of the incubations. The gill tissue of five individuals of each species was removed 2 hours and 10 hours following the end of each cold exposure treatment. For a control sample, the gill tissue of five animals of each species was taken before any cold exposure. The gill tissues obtained by this method were used to quantify Hsp72 and Hsp74 expression via solid phase immunochemical analysis (western analysis). The analysis was run according to the protocol described by Tomanek and Somero (2002) with the following exceptions and additions: 300ul of homogenization buffer was added to the gill tissue of T. pulligo for homogenization, and total protein concentration was determined using the BCA protein (Pierce) assay rather than the micro-BCA (Pierce) protein assay. The resulting data for each species and heat-shock protein were analyzed separately using an asymmetrical, 3-factor ANOVA model with exposure temperature, exposure period, and recovery period as the three factors. Each factor was treated as being fixed and orthogonal to the other factors. Significant results from the models were investigated using the Student Newmann-Keuls (SNK) test. RESULTS Chill Coma Experiment All snails were active one day following their exposure to cold treatments. However, 20 minutes following their exposure, all individuals of T. brunnea in both the 1° C air and the seawater in the ice bath were inactive. All of the T. pulligo exposed to 1° C air were also inactive at the 20 minute time point. All T. funebralis individuals in each treatment group were active after 20 minutes. Analysis of Hsp70 Expression The ANOVAs for levels of Hsp 72 in T. brunnea and T. funebralis and levels of Hsp 74 in T. pulligo and T. funebralis all showed highly significant (p«0.01) three way interactions between the factors. The ANÖVA for Hsp 74 in T. brunnea displayed significant interactions between exposure length and exposure temperature as well as between recovery period and exposure temperature, and the ANÖVA for Hsp 72 in T. pulligo showed significant two way interaction between all pairs of factors. The most consistent pattern in the data was a significant reduction in Hsp 72 levels relative to control values after 10 hours of recovery in groups with 10-hour cold exposure times (Figs. 1-3). This pattern was present in all 10-hour exposure treatments. In all of these groups except for T. pulligo exposed to 4° C water, Hsp 72 levels after 2 hours of recovery were similar to control values and significantly greater than the levels after 10 hours of recovery. Groups subjected to 2-hour exposures did not show any difference in Hsp 72 levels after 10 hours of recovery relative to control, with the exception of T. pulligo exposed to 4° C water (Fig. 3). Of these 2-hour exposure treatments, only T. pulligo exposed to 1° C and 4° C water and T. brunnea exposed to 4° C water showed levels of Hsp 72 after 2 hours of recovery that were significantly different from pre-exposure levels (Figs. 2, 3). Expression of Hsp 74 proved difficult to quantify accurately, as the bands on the gel were generally weak and often overrun by the Hsp 72 signal. No consistent patterns of expression were found either within or among species. As previous studies have indicated that Hsp 74 is constitutively expressed and does not usually respond strongly to thermal stress (Tomanek 2002), we assume the same is occurring in this case. DISCUSSION The highly significant interaction terms from the ANÖVAs suggest that none of the factors are acting independently of each other. In particular, it seems that a combination of a 10- hour exposure time and a 10-hour recovery time alone cause a significant down regulation of Hsp 72, and to a lesser extent Hsp 74. This would seem to indicate that snails exposed to prolonged cold stress exhibit a different response profile than those experiencing only transient exposure to cold. These results differ from studies of cold shock responses of other organisms, which generally noted transient increases in levels of inducible Hsp 70 or its mRNA precursor (Ju et al. 2002; Martinez et al. 2001). In their study of gene expression in catfish, Ju et al. (2002) noted a transient increase of inducible Hsp 70 2 hours following initial exposure to reduced temperature. This increase in the level of inducible Hsp 70 had subsided after 24 hours of exposure (Ju et al. 2002). A study of Hsp production in nematodes also showed a transient increase in inducible Hsp 70 (Martinez et al. 2001). The increased levels were noticeable after 2 hours, but had subsided after 24 hours (Martinez et al. 2001). Both of these studies maintained their experimental organisms at the cold exposure temperature and did not bring animals back to non¬ stress inducing temperature (Ju et al. 2002; Martinez et al. 2001). In our experiment, no samples were taken until the organisms had been in ambient temperature seawater for at least 2 hours. These differences in sampling procedures could explain the discrepancies between the experiments. Any increase in endogenous levels of inducible Hsp 70 during cold exposure in this experiment could have already subsided during the 2-hour recovery period after which the first sample was taken. Similarly, the decrease in Hsp 72 seen in the 10-hour exposure treatments in this experiment might not have occurred in the other studies because the animals were not allowed a recovery period. Alternatively, the differences between the results of the studies might simply reflect taxonomic differences of the study animals chosen for each experiment. Although there was relatively little variation among the different species in patterns of Hsp 70 expression following cold shock, evidence from the chill coma experiment indicates that T. funebralis may be less severely stressed by exposure to cold than its two subtidal congeners. While cold exposure may not directly kill T. brunnea or T. pulligo, it may render them incapable of responding to other cues and conditions in the environment, putting them at a significant disadvantage relative to T. funebralis. Transplant studies similar to those done for heat stress (Tomanek 2002) could indicate whether the observed differences have any effect in situ. Since T. funebralis seem more capable of tolerating cold treatment, but no interspecific variation of patterns of Hsp 70 expression were found, T. funebralis may have some other method of coping with cold stress. This could be investigated more directly in further studies by tracing radioactively labeled amino acids or using DNA micro-array technology. CONCLUSIONS All species exhibited a significant decrease in endogenous Hsp 72 10 h following the end of a 10 h cold exposure. This suggests that different lengths of exposure to cold elicit different physiological responses to the cold shock. While patterns of Hsp 72 expression were similar among species, there were differences in behavior of individuals exposed to 1° C air and 3° C water. The chill comas exhibited by T. brunnea in the 1° C air and 3° C water and T. pulligo in the 1° C air indicate that these species could be less capable than T. funebralis of responding to changes in the environment following cold shock. Since patterns of Hsp 72 expression were similar among species, it may be that T. funebralis has some other mechanism for dealing with cold stress not possessed by its subtidal congeners. ACKNOWLEDGEMENTS The author wishes to thank George Somero and Lars Tomanek for their continual guidance on all aspects of the project. Thanks also to Jim Watanabe for his aid with statistical analysis of the data, as well as James Lopez and Anne Schwedt for their assistance with lab techniques and protocols. LITERATURE CITED Ju, Z., R. A. Dunham, R. A., and Liu, Z. 2002. Differential gene expression in the brain of channel catfish (Ictalurus punctatus) in response to cold acclimation. Molecular Genetics and Genomics. 268:87-95 Martinez, J., J. Perez-Serrano, W. E. Bernadina, and F. Rodriguez-Caabeiro 2001. Stress response to cold in Trichinella species. Cryobiology. 43:293-302. Morris, R. H., D. P. Abbott, and E. C. Haderlie. Intertidal Invertebrates of California. Stanford University Press: Stanford, 1980. Tomanek, L. 2002. The heat shock response: its variation, regulation, and ecological importance in intertidal gastropods (genus Tegula). Integrative and Comparative Biology :797-807. Tomanek, L. and G. Somero. 2002. Interspecific- and acclimation-induced variation in levels of heat-shock proteins 70 (Hsp 70) and 90 (Hsp 90) and heat-shock transcription factor 1 (HSF 1) in congeneric marine snails (genus Tegula): implications for regulation of hsp gene expression. Journal of Experimental Biology. 205:667-685. FIGURE LEGENDS Fig. 1. Levels of Hsp 72 in T. funebralis relative to bovine Hsc 70 standard. Error bars are standard errors of the mean (n=5). Asterisks denote significant difference from pre-exposure values. Fig. 2. Levels of Hsp 72 in T. brunnea relative to bovine Hsc 70 standard. Error bars are standard errors of the mean (n=5). Asterisks denote significant difference from pre-exposure values. Fig. 3. Levels of Hsp 72 in T. pulligo relative to bovine Hsc 70 standard. Error bars are standard errors of the mean(n=5 for control, 2-hour exposures, n—4 for 10-hour exposures). Asterisks denote significant difference from pre-exposure values. 0.1 001 10 10h — pre-exposure 2h recovery Oh recovery Fig. 1 40 10h OC 2h Exposure Regime 40 2h 10 0.1 0.01 1C 10h — Pre-exposure 2h recovery 1Oh recovery Fig. 2 L 40 10h OC 2h Exposure Regime 402h 0.1 0.01 — pre-exposure 2h recovery Oh recover 4 10 10h Fig. 3 40 10h OC 2h Exposure Regime 40 2h