ABSTRACT Organisms occupying rocky intertidal habitats are exposed to extreme temperature fluctuations. Given the pervasive effects of temperature on physiological function, large changes in gene expression would be expected in these organisms as temperature varies. Gene expression at elevated temperatures was investigated in vitro using gill tissue from the mussel Mytilus californianus. Tissues were exposed to temperatures between 20.1 and 36.2° C. and differential expression was evaluated using a 4224 cDNA Mytilus microarray. At 33° C. mRNA levels for hsp70 were induced 60 to 110-fold compared to the control temperature. Several other genes coding for proteins involved in protein folding (hsp20), degradation of denatured polypeptides (sequestosome 1), apoptosis (fos-related antigen 2), and other activities appeared to be induced with elevated temperature. mRNA levels for most genes were greatly reduced at 36.2° C, suggesting a deleterious effect of temperature on gene expression (either retardation of induction or greater degradation of mRNA) at this temperature. Simulated mussel body temperatures in excess of 41° C were recorded at the collection site. Thus, habitat temperatures may occasionally exceed the thermal tolerance limits of the animal. This is likely to affect the ecology and biogeography of the species. INTRODUCTION Temperature strongly impacts many important physiological processes. Extreme temperatures can have negative effects on protein stability, protein synthesis, heart function. action potential generation, mitochondrial respiration, and membrane static order (Somero. 2002). Given the impacts of thermal stress, organisms that are subject to widely varying body temperatures must have adaptations to safeguard biological function over the range of temperatures to which they are exposed. This paper focuses on the exploitation of phenotypic plasticity through changes in gene expression as a strategy for dealing with elevated temperature in the mussel Mytilus californianus. M. californianus is a fitting experimental organism for studying temperature adaptation. This species occurs in the mid-intertidal zone of wave-swept rocky shores along the west coast of North America from Alaska to Baja California (Seed and Suchanek, 1992; Roberts et al., 1997). Due to the effects of tidal immersion and emersion, as well as daily weather patterns, temperatures in this habitat zone can vary as much as 18° C in a single day (Helmuth and Hofmann, 2001). Because mussels are sessile ectotherms with limited potential for behavioral thermal regulation, their body temperatures fluctuate on the same scale.' While the lower vertical limit of M. californianus is generally determined by interactions with other species (Paine, 1966; Paine, 1974), the species’ upper limit is primarily governed by abiotic factors (Seed and Suchanek, 1992). Temperature is likely to be of primary importance among these factors, which also include desiccation stress and hypoxia. Coping with these environmental stresses may carry a significant energetic cost, as individuals from sites higher in the intertidal It is possible that mussels may be able to cool themselves slightly at high temperatures by gaping (opening the shell slightly) to allow for evaporative cooling. However, Helmuth and Hofmann (2001) argue that this is unlikely due to the accompanying risk of desiccation. have been found to have smaller maximum size and slower growth rates than lower-occurring individuals (Harger, 1970). In light of temperature’s pervasive effects, one would expect large changes in gene expression as temperature changes. Until recently, investigations of these shifts have been limited to targeted studies of individual proteins. This type of work has yielded a large amount of information on a small number of proteins that, taken together, constitute what is known as the heat shock response. With the exception of some Antarctic fishes (Hofmann et al., 2000), this suite of proteins has been found to be conserved in every examined species. Heat shock protein 70 (hsp70) and heat shock protein 20 (hsp20) are high- and low-molecular weight inducible isoforms of chaperone proteins that bind to polypeptides denatured by heating, cooling, or other stresses and aid in refolding. Heat shock proteins also prevent the formation of dangerous aggregations of unfolded proteins in the cell (Ellis and van der Vies, 1991). Another central element of the heat shock response is the protein ubiquitin, which tags irreversibly denatured proteins for degradation in the proteasome (Hershko and Ciechanover, 1992). The temperatures at which the genes involved in the heat shock response are induced have been shown to be subject to acclimation and acclimatization (Roberts et al., 1997). There is also a maximum temperature of heat shock protein synthesis, above which the proteins are no longer induced (Tomanek and Somero, 1999). Many studies have described the threshold, peak, and maximal induction temperatures, as well as the potential for acclimation/acclimatization of the heat shock response, in various intertidal organisms, including species of the genus Mytilus (Roberts et al., 1997; Hofmann and Somero, 1996a; Hofmann and Somero, 1996b; Buckley et al., 2001) and Tegula (Tomanek and Somero, 1999; Tomanek and Somero; 2000). In 13° C laboratory acclimated M. californianus, Buckley et al. (2001) found a threshold induction temperature of 26° C for heat shock proteins. Targeted protein studies are only able to address a small number of changes at a time. In contrast, the application of cDNA microarray technology allows for a global view of gene expression. This technique makes it possible to understand broad changes and subtle shifts in mRNA levels across the genome. For example, using this technique, Podrabsky and Somero (2004) found that several genes involved in cell growth and proliferation were differentially expressed in fish of the species Austrofundulus limnaeus that were exposed to different temperature regimes. Because they allow for genome-wide analysis, DNA microarray approaches have great promise for identifying genes induced in response to environmental change (Gracey and Cossins, 2003; Podrabsky and Somero, 2004). While mRNA levels do not perfectly predict the concentration or activity of proteins in the cell (due to RNA degradation or post-translational modification of enzymes), there is good quantitative evidence that transcript amounts accurately reflect protein levels in many cases (Gracey and Cossins, 2003). The connection between transcription and higher order physiological processes is especially solid during periods of stress or environmental change (Gracey and Cossins, 2003, Gasch et al., 2000). Applications of microarray technology to non- model organisms are currently largely exploratory. Descriptive studies are required to move the field forward, especially in light of the large number of sequenced genes whose functions have not been characterized (Gracey and Cossins, 2003). While the approach is still in its infancy, microarray studies offer an avenue for linking gene regulation with the ecology and biogeography of M. californianus and other species. MATERIALS AND METHODS Collection and Care of Specimens Mussels were collected from a moderately wave-protected area of the Hopkins Marine Life Refüge (Monterey, CA) on March 14, 2004. They were held in 14° C filtered seawater tanks for two weeks prior to the heat shock experiment. Heat Shock Experiment Gills from single animals were grouped together for the heating experiments to eliminate the effects of interindividual variation on the data. Three animals were used in the study. Each mussel's gills were sectioned into 8 pieces. Dissections were performed on plastic trays floating in 14° C water to minimize heat exposure during this step. One gill sample from each animal was frozen on dry ice immediately after dissection. A second control sample was held at 14° C throughout the course of the experiment. The remaining 6 samples from each individual were exposed to temperatures of 20.1° C, 23.0° C, 25.9° C, 28.9° C, 33.0° C, or 36.2° C for 30 minutes. Following the heat shock, samples were allowed to recover at 14° C for 1 hour. To prevent a confounding effect of hypoxia, samples were aerated during the heat shock and recovery by blowing a concentrated stream of air into the tube via a Pasteur pipette. At the end of the recovery period, samples were frozen on dry ice for later processing RNA isolation and microarray hybridization RNA was extracted from the gill samples using Trizol according to manufacturer's instructions (Invitrogen, San Diego). A reference RNA was prepared by extracting RNA from the pooled gill and adductor muscle tissues of 5 individual animals that had been acclimated to 14° C for 3 weeks. Treatment and reference RNAs were reverse transcribed into cDNA and labeled with fluorescent dyes using standard protocols [www.microarrays.orgl. Briefly, 10 ug of total RNA were mixed with 5 ug of oligo-dT primer in a volume of 14 ul water, denatured at 70° C for 5 mins, and annealed on ice for 10 mins. Then 2 ul of O.1M dithiothreitol, 1 ul of deoxyribonucleoside triphosphate (dNTP's) [10 mM dATP, dCTP, dGTP, 5 mM dTTP. 5 mM amino-allyl dUTP), 2ul of lOx reverse transcriptase reaction buffer, and 1 ul of Stratascript reverse transcriptase (Stratagene, Austin) were added. The reaction was terminated by addition of 5 ul 0.8 MNaOH and heating to 65° C for 15 mins. 5 ul of 0.8 MHCl were added to neutralize the reaction. Resulting cDNA was purified using MinElute purification columns [Qiagen, Valencia, CAJ. Purified CDNA was brought to 10 ul in 0.1 M NaHCOz. The cDNA was coupled to Cy¬ dyes (Amersham Biosciences, Piscataway, NJ) using standard protocols [www.microarrays.orgl. Treatment samples were coupled to Cy-5 and reference to Cy-3. Treatment and reference samples were competitively hybridized onto a 4,224 cDNA Mytilus microarray in 20 ul of 3x sodium chloride/sodium citrate solution (SSC), 20 mM HEPES pH 7, 0.2% SDS, 10 ug of synthetic poly-A RNA at 65° C for 18 hours. Slides were washed and scanned according to standard procedures www.microarrays.org). Spot intensities were assayed using an Axon scanner (Axon, Foster City, CA) and GenePix software (Axon, Foster City, CA). Sequencing Ninety-six genes that showed induction with temperature were selected for sequencing. Products were sequenced from the 5’ direction using Big Dye terminator mix (Applied Biosystems, Foster City, CA). A putative homology was assigned to genes based on BlastX searches against the NCBI database [http://www.ncbi.nlm.nih.gov/1. Data Normalization and Analysis Data analysis was completed using GeneSpring software (GeneSpring, Liverpool). Per¬ spot normalizations were used to normalize the treatment samples to the reference by dividing the intensity of the Cy-5 signal at each spot by the intensity of the Cy-3 signal at that spot. Per- chip intensity-dependent Lowess normalization was also applied. This procedure divides the intensity of the signal for each gene by the median intensity for the chip, in order to reduce the effects of experimental variation across the chip. Following these normalizations, the "Normalize to Specific Samples" function in GeneSpring was used to normalize the elevated temperature samples (20.1 -36.2° C) for each animal to the median of the two control chips for that animal. This normalization divided the intensity of a given spot on the treatment array by the intensity of the same spot on the control arrays. Environmental Temperature Monitoring The temperature records used in this study were measured with iButton temperature recorders (Dallas Semiconductor, Dallas). A single iButton was epoxied on the rock next to the mussel bed where experimental animals were collected. iButtons were placed inside of two müssel shells filled with silicone sealant, which has been shown to approximate the heat capacity of mussel tissue (Helmuth and Hofmann, 2001). These shells were epoxied next to the mussel bed to track simulated body temperatures. The iButtons were in the field between 4/23/04 and 5/13/04. RESULTS AND DISCUSSION As this was the first experiment to use the Mytilus microarray, and because the RNA yield from each tissue sample was limited by the small mass of tissue used (1/8" of a gill), some of the data were of low quality. Problems during the array production process resulted in slide- by-slide variation in the degree of hybridization success. Many slides showed significant background fluorescence. In general, the slides for Animal A showed less background fluorescence and higher spot intensities than those from Animals B and C. Differential Gene Expression Hsp70 and hsp20 were strongly induced with temperature in all three animals (figs. 1 and 2). The expression profiles of these genes fit a priori expectations given their molecular chaperone function. This result increases our confidence in the validity of our technique. The strongest induction of these genes was seen at 33.0° C in two animals and at 28.9° C in the third Hsp70 was more highly induced at all temperatures than hsp20. Hsp70 mRNA induction in the treatment samples at 33° C was 60 - 110 times greater than in the control samples. This very strong induction of molecular chaperone mRNA suggests that they are a central part of the organism’s response to high temperatures. Consistent inter-individual variation in induction of hsp70 and hsp20 was apparent. Animals A and B showed similar induction at most temperatures for both genes, while the mRNA levels in animal C were lower than in the other animals at all temperatures for both genes. The observed induction peak for animal C came at a lower temperature than for A and B in both genes. These consistent patterns suggest that there may be some difference in the gene regulatory strategies of these particular animals in response to heat. Variation of this type could be caused by prior acclimatization (although all 3 mussels were acclimated at 14° C for two weeks prior to the experiment), genetic differences between individuals, or some other unknown source. 2 This is with the exception of the frozen 14e control, 20.19, and 28. 90 C, where the Cy-S dye failed to couple to the treatment samples from Animal A, resulting in missing data at these temperature points for this individual, The inducible isoform of hsp70 exists side-by-side in the cell with the constitutively expressed heat shock cognate protein 70 (hsc70). Like hsp70, hsc70 is a molecular chaperone protein (Dworniczak and Mirault, 1987). It also plays a role in the trafficking of proteins within the cell by working with auxilin to remove clathrin from coated vesicles (Ungewickell et al., 1995). Because it is expressed constitutively, cellular concentrations of hsc70 would not be expected to increase with heat shock. This has been confirmed by Clark and Menoret (2001). In animal A, we observed an approximately 6-fold induction at 33° C at two spots on the array corresponding to hsc70 (the gene is represented twice on the array)(fig. 3). This probably represents improper binding of hsp70 cDNA to the hsc70 probe. The failure of the array to discriminate between hsp70 and hsc70 in this case may be related to the extreme abundance of hsp70 mRNA. In animal A at 33° C, hsp70 mRNA levels were 100-fold greater than at the controls. Apparent Hsc70 induction was not observed in animals B and C. Some denatured proteins cannot be refolded even with the help of molecular chaperones like the hsp's. Sequestosome 1 is a ubiquitin-binding protein that helps to aggregate irreversibly denatured proteins within the cell that have been tagged for degradation (Thompson et al., 2003). As seen in figure 4, this gene also showed a trend towards induction at higher temperatures in the three animals. This is an intriguing result that deserves further investigation given the known function of this gene. Induction of sequestosome 1 may signal a cellular response to increasing levels of denatured proteins as temperature increases. In addition to these changes in hsp’s and peptide degradation proteins, we also found possible evidence of programmed cell death in response to temperature damage. Fos-related antigen 2 (FÖSL2) codes for a nuclear protein that dimerizes with proteins of the jun family to form the transcription factor AP-1. This complex, and especially the similar gene c-Fos, has 10 been implicated in regulation of the cell cycle and in programmed cell death in mice (Smeyne et al., 1993). FÖSL2 showed a general pattern of induction at higher temperatures (fig. 3), although the trend was not as clear as with the molecular chaperones (fig. 1 and 2). It is possible that the induction of FOSL2 at elevated temperatures is an indication of an apoptotic response to heat stress. Previous studies have shown that FÖSL2 is induced in response to injury and environmental stress in mice and rats (Butler and Pennypacker, 2004; Pompeiano et al., 2002). For the two animals with data for both control points, the induction of FÖSL2 was slightly lower in the frozen control sample than in the control that was kept at 14° C during the experiment, This suggests that there may be some effect of dissection or handling on the expression of FÖSL2. Due to the higher quality of the array data from Animal A compared to B and C, trends may be clearer in the results from this animal than from the other two. Table 1 lists several genes in Animal A that showed expression patterns similar to hsp70 and hsp20. Since many of these genes only showed differential expression in one animal, it would be premature to conclude that they are part of a temperature response. However, these genes would be worth examining in future experiments. DNA double strand repair enzyme might be induced in response to potential thermal damage to DNA. MAP kinase, adenosine A2b receptor, and wnt¬ 7b are all involved in cellular signaling. Histone deacetylase II is a transcriptional repressor Across all three animals, genes were chosen for sequencing that showed induction at elevated temperature. Because the genomes of non-model organisms such as M. californianus have not been as extensively described as those of model organisms such as D. melanogaster, M. musculus, or C. elegans, many of the genes that we sequenced could not be identified in the genetic databases. Further characterization of the Mytilus genome will advance understanding of the heat stress response by elucidating the function of currently unidentified genes. Environmental temperatures may exceed the maximum induction temperature Many genes, including hsp70 and hsp20, showed reduced induction at 36.2° C in all 3 animals (figs 1, 2; table 1). This is suggestive either of a deleterious effect of heat on transcription or of elevated degradation of mRNA at this temperature. This may represent a threshold above which the cellular machinery involved in mRNA induction cannot function. The temperature data in figure 5 show that simulated body temperatures exceeded 37° C for over 120 minutes on several days in March. While these data are not representative of typical habitat temperatures (they are from an unusually warm episode), they indicate that the peak body temperatures of mussels may occasionally exceed the animal’s ability to mount a stress response. If there is a maximum induction threshold at 36° C, the mussels would have been unable to transcribe genes for heat shock and other proteins for about 90 minutes on the hottest days. This type of extreme temperature exposure would be expected to cause severe damage to the organism. While no large-scale mussel death was noted in the mussel population at Hopkins Marine Station during this hot event, anecdotal reports indicate a notable mussel die-off at Bodega Bay, further north of Monterey. In addition, even if the acute heating were a sublethal stress, it could still have significant cost-of-living effects on the organism that could be deleterious over longer time scales (Somero, 2002). Simulated body temperatures were found to track environmental temperatures closely. Figure 6 shows the environmental temperatures recorded by an exposed iButton attached to the rock next to the mussel bed where the simulated body temperatures were recorded. Figure 7 shows the close relationship between habitat temperature and body temperature for a single day 12 (4/25/04). Body temperatures exceed habitat temperatures at the hottest point. This figure also shows the high and low tides on this particular day, Limitations of the study The large number of missing data points as well as the limited number of replicates in the study preclude rigorous statistical analysis of gene expression. Increased replication is a priority for future experiments. An increased number of temperature points would also allow for a more accurate understanding of where induction of various genes begins, peaks, and ends. This study characterizes transcript responses to heat stress in a laboratory setting. Gene induction under natural conditions is also influenced by other factors such as hypoxia and desiccation stress. Expression patterns in the field may differ from those in the lab. Nevertheless, this study can be viewed as a necessary baseline for a field study of gene expression in Mytilus californianus. CONCLUSION Despite the limitations just discussed, this study identifies several important aspects of gene expression in M. californianus in response to elevated temperature. The clear expression profiles of hsp70 and hsp20 indicate that the microarray can accurately detect induction of genes as temperature increases. The consistent inter-individual variation observed in the expression profiles of hsp70 and hsp20 suggests that there may be some difference (possibly due to genetic differences or thermal history) between individuals that affects their response to temperature. Several other genes showed induction with heat. Most of these genes code for proteins with functions that would be likely to be induced with increased thermal stress. A number of unidentifiable genes also showed strong induction. These results all deserve further exploration. Environmental temperatures occasionally exceed what appears to be a maximum temperature for induction of transcription in this species. This aspect of the relationship between temperature and gene expression might have a strong impact on the ecology and biogeography of Mytilus californianus. ACKNOWLEDGEMENTS Andy Gracey, George Somero, Brad Buckley, and Will Tyburczy have all been invaluable sources of aid and advice. Andy especially has spent a great deal of time with me on this project and 1 am grateful for it. Jim Watanabe provided general research advice and encouragement. REFERENCES Buckley, B. A., M-E. Owen, and G.E. Hofmann. (2001). Adjusting the thermostat: The threshold induction temperature for the heat-shock response in intertidal mussels (genus Mytilus) changes as a function of thermal history. Journal of Experimental Biology 204, no.20; 3571-3579. Butler, T. L. and K. R. Pennypacker (2004). Temporal and regional expression of Fos-related proteins in response to ischemic injury. Brain Research Bulletin 63, no.1: 65-73. Clark, P. R. and A. Menoret (2001). The inducible Hsp70 as a marker of tumor immunogenicity, Cell Stress & Chaperones 6, no.2: 121-125. Dworniczak, B. and M. Mirault (1987). Structure and expression of a human gene coding for a 71 kd heat shock 'cognate' protein. Nucleic Acids Research 15, no.13: 5181-5197. Ellis, R. J. and S. M. Van Der Vies (1991). Molecular chaperones. Annual Review of Biochemistry 60: 321-348. Gasch, A. P., P. T. Spellman, C.M. Kao, O. Carmel-Harel, M.B. Eisen, G. Storz, D. Botstein., and P.O. Brown. (2000). Genomic expression programs in the response of yeast cells to environmental changes. Molecular Biology of the Cell 11, no.12: 4241-4257. Gracey, A. Y. and A. R. Cossins (2003). Application of microarray technology in environmental and comparative physiology. Annual Review of Physiology 65: 231-259. Harger, J. R. E. (1970). The effect of wave impact on some aspects of the biology of sea mussels. Veliger 12, no.4: 401-414. Helmuth, B. S. T. and G. E. Hofmann (2001). Microhabitats, thermal heterogeneity, and patterns of physiological stress in the rocky intertidal zone. Biological Bulletin 201, no.3: 374- 384. Hershko, A. and A. Ciechanover (1992). The ubiquitin system for protein degradation. Annual Review of Biochemistry 61: 761-807. Hofmann, G. E., B. A. Buckley, S. Airaksinen, J.E. Keen, and G.N. Somero. (2000). Heat-shock protein expression is absent in the Antarctic fish Trematomus bernacchii (Family Nototheniidae). Journal of Experimental Biology 203, no.15: 2331-2339. Hofmann, G. E. and G. N. Somero (1996a). Interspecific variation in thermal denaturation of proteins in the congeneric mussels Mytilus trossulus and M. galloprovincialis: Evidence from the heat-shock response and protein ubiquitination. Marine Biology 126, no.1: 65- 75. Hofmann, G. E. and G. N. Somero (1996b). Protein ubiquitination and stress protein synthesis in Mytilus trossulus occurs during recovery from tidal emersion. Molecular Marine Biology and Biotechnology 5, no.3: 175-184. Paine, R. T. (1966). Food web complexity and species diversity. American Naturalist 118: 65- 75. Paine, R. T. (1974). Inter tidal community structure experimental studies on the relationship between a dominant competitor and its principal predator. Oecologia 15, no.2: 93-120. Podrabsky, J. E. and G. N. Somero (2004). Changes in gene expression associated with acclimation to constant temperatures and fluctuating daily temperatures in an annual killifish Austrofundulus limnaeus. Journal of Experimental Biology 207, no.13: 2237- Pompeiano, M., P. D’Ascanio, C. Centini, O. Pompeiano, and E. Balaban. (2002). Short-term (Fos) and long-term (FRA) protein expression in rat locus coeruleus neurons during the Neurolab mission: Contribution of altered gravitational fields, stress, and other factors. Neuroscience 115, no.1: 111-123. Roberts, D.A., G.E. Hofmann, and G.N. Somero. (1997). Heat-shock protein expression in Mytilus californianus: Acclimatization (seasonal and tidal-height comparisons) and acclimation effects. Biological Bulletin 192, no.2: 309-320. Seed, R. and T.H. Suchanek (1992). Population and community ecology of Mytilus. Developments in Aquaculture and Fisheries Science 25: 87-169. Smeyne, R.J., M. Vendrell, M. Hayward, S.J. Baker, G.G. Miao, K. Schilling, L.M. Robertson. T. Curran, J.I. Morgan. (1993). Continuous c-fos expression precedes programmed cell death in vivo. Nature 363, no.6425: 166-169. Somero, G. N. (2002). Thermal physiology and vertical zonation of intertidal animals: Optima, limits, and costs of living. Integrative and Comparative Biology 42, no.4: 780 - 789. Thompson HG, H. J., Wold BJ, Lin F, Brody JP (2003). p62 overexpression in breast tumors and regulation by prostate-derived Ets factor in breast cells. Oncogene 22, no.15: 2322-33. Tomanek, L. and G. N. Somero (1999).Evolutionary and acclimation-induced variation in the heat shock responses of congeneric marine snails (genus Tegula) from different thermal habitats: Implications for limits of thermotolerance and biogeography. Journal of Experimental Biology 202: 2925 -36. Tomanek, L. and G. N. Somero (2000). Time course and magnitude of synthesis of heat-shock proteins in congeneric marine snails (genus Tegula) from different tidal heights. Physiological and Biochemical Zoology 73, no.2: 249-256. 16 Ungewickell, E., H. Ungewickell, S.E.H. Holstein, R. Lindner, K. Prasad, W. Barouch, B. Martin, L.E. Greene, E. Eisenberg. (1995). Role of auxilin in uncoating clathrin-coated vesicles. Nature 378, no.6557: 632-635. Table 1: Genes showing differential transcription at elevated temperatures in Animal A. Array data for this individual were of higher quality than animals B and C, so these data are presented here alone. The possible roles of these genes in response to temperature stress deserve further study. Gene Function Sequence Fold¬ Fold¬ Fold¬ Identity/ Induction at Induction at Induction at Confidence 28.90 C 33.0° C 36.2° C Hsp70 2e 25.4 Molecular 116.3 1.2 chaperone 31.9 Hsp20 Molecular 2e 3.9 1.2 chaperone fos-related Transcription antigen 2 factor; stress 8e 6.4 7.0 0.3 response Sequestosome Sequestration le? 4.0 7.9 0.81 ubiquitinated proteins DNA double 0.78 DNA repair 1.76 1.82 1.05 strand repair Wnt-7b Signal transduction; 4.02 2e 1.65 1.36 development Cell cycle Histone 2e 1.29 1.38 0.92 Deacetylase II regulation Figure Legends Fig. 1: Hsp70 mRNA levels in the three animals at each temperature point. Fig. 2: Hsp20 mRNA levels in the three animals. Consistent within-animal patterns are apparent between hsp70 (figure 1) and hsp20. Fig. 3: False induction signal for Hsc70. 04B06 and 05D17 refer to the two different Hsc70 spots on the array. Fig. 4: Sequestosome 1 expression profile. This gene is involved in aggregating ubiquitinated polypeptides. Fig. 5: FÖSL2 expression profiles for the three animals. Fig. 6: Simulated body temperatures recorded by iButtons placed inside silicone-filled mussel shells between 4/23/04 and 5/13/04 at Hopkins Marine Station. Fig. 7: Environmental temperatures recorded by an exposed iButton attached to the rock next to the mussel bed where simulated body temperatures were recorded. Fig. 8: Habitat temperatures and simulated body temperatures from April 25, 2004. High and low tides are given in distance (feet) from MLW and are represented by red dots. 140 120 100 80 5 60 5 40 20 Figure 1. —Animal A -H-Animal B Animal C 4 23.0 25.9 14.0 14.0 20.1 28.9 33.0 36.2 Temperature (C) 70 § 60 3 50 1 40 5 30 20 + 10 Figure 2 —-Animal A H-Animal B 4 Animal C — —1 14.0 14.0 20.1 23.0 25.9 28.9 33.0 36.2 Temperature (C) Figure 3 — Animal A - 04B06 -Animal B - 04B06 Animal C - 04B06 — Animal A- 05017 — Animal B- 05017 Animal C-05D17 25.9 28.9 33 36.2 14 23 14 20.1 Temperature (C) Figure 4 —Animal A Animal B — Animal C 8 22 14 14.0 14.0 20.1 23.0 25.9 28.9 33.0 36.2 Temperature (C) 9 Figure 5 Animal A E Animal B A Animal C 12 10 E 6 — 3 2 14.0 14.0 20.1 23.0 25.9 28.9 33.0 36.2 Temperature (C) Figure 6 ------------------- —---- —------— -Simulated Mussel 1 Simulated Mussel 2 — 45.00 40.00 35.00 4 S 30.00 + 25.00 5 20.00 15.00 t t 10.00 5.00 0.00 S S8 qà odo H o S o J 9O. P Op; Op; S Foo Date/Time 25 Figure 7 40.00 35.00 30.00 25.00 20.00 150 o 10.00 5.00 0.00 — aaaaaaa- oo o 0 o Date/Time 47.00 42.00 37.00 32.00 27.00 22.00 17.00 12.00 2.00 -3.00 Figure 8 April 25, 2004 . S •4.80 3.703.60 0.20— Vg §9 gg g GOES O. So. o s . oodoooddod Time —Simulated Mussel 1 —2 Simulated Mussel 2 — Exposed button — Tidal Extremes