Abstract When a mussel is subjected to rapidly increasing temperatures, its heart rate will sharply decrease above a critical temperature (Hert). Previous studies have suggested that this sudden drop in heart rate may be triggered by a behavioral adaptation to environmental stress, valve closure. This behavior would lower the amount of oxygen available for aerobic metabolism, which could necessitate a decrease in cardiac activity. My study examined the temporal patterns of heart rate and relative valve position (open versus closed) in specimens of ribbed mussel (Mytilus californianus) exposed to controlled heating in water. Heart rate data and valve motion were gathered using, respectively, impedance electrodes inserted into the pericardial space and Hall-effect sensors attached to the mussel valves. The mussels were monitored over the course of a 2.7 hour period, during which water temperature was increased from 14° C to 34° C. On average, valves were more closed after Hart than before (P + 0.0001). However, valve closure began prior to Hart, and the mussels reached their most closed position after Herit had been surpassed. Thus there was no evidence that complete valve closure occurred before heart rate decline. Therefore, the fall in heart rate above Hert cannot be ascribed to a simple behavioral response, complete value closure. Future studies should quantify the responses of valve movement to other factors, such as decreasing water salinity and emersion effects, that are also linked to changes in mussel heart rates to gain further insight into the relationship between valve movement and heart function. Introduction When studying the effects of various environmental stressors on organisms, one level of analysis focuses on relatively short time frame experiments to reveal how quickly different factors within the organism respond to changes in the environment (Bayne, 1976). This type of analysis is especially applicable to life in the intertidal zone, where changes in various abiotic factors take place on a rapid temporal scale. The dynamic conditions of this climate play a key role in determining the distribution of many species, such that a species' potential habitat is often constrained by its physiological limits (Hochachka and Somero, 2002). In studying the effects of environmental stressors on intertidal organisms, it is helpful to look at specific physiological or behavioral factors that are directly affected by the stressors and that may be important in determining the distribution of a species. One group of intertidal organisms that has been studied extensively in relation to such physical factors as temperature, salinity, and exposure to air are marine mussels. An aspect of mussel behavior that has often been observed in response to environmental stressors is valve movement. A mussel's valve movements allow it to regulate access to food and oxygen through filtering water, and studies have shown that mussels mainly remain open when under water, in order to sustain their respiration and feeding rates (Curtis et al., 2000). Mussels will often close their valves when subjected to conditions of environmental stress (Manley and Davenport, 1979). This behavioral mechanism has been used increasingly in monitoring systems to detect pollution in the environment, because the mussels will respond quickly to sudden changes in their habitat (Kramer et al., 1989; Gudimov, 2003). In research on mussels and the effect of environmental stressors, one of the most commonly studied physiological factors is heart rate. This physiological factor is relatively stable under normal conditions and responds to rapid fluctuations in environmental conditions; thus, heart rate is a good indicator of the health of mussels in the presence of stressors (Bayne, 1976; Sabourin and Tullis, 1981; Depledge et al., 2000). Previous research has quantified the relationship between mussel heart rates and many environmental factors that affect its function (Bayne, 1971; Bayne et al., 1976; Nicholson, 2002). One such factor is temperature, perhaps the most critical abiotic factor of the intertidal zone. Organisms in this habitat may face sudden changes in temperature within short time periods. It is well known that mussel heart rates will increase with increasing temperature, which is one of the most important factors in establishing a mussel’s metabolic rate (Bayne, 1976). As temperature increases, at a certain temperature, called the critical temperature (Hert), the mussel heart rate will decrease suddenly. In 2006, Braby and Somero conducted a study that documented the critical temperatures of heart function for different species of marine mussels of the genus Mytilus. Although not examined quantitatively, they observed that valve closure appeared to accompany the rapid drop in heart rate at Herit- Based on these preliminary observations, the authors postulated that the heart rate decline in these mussels was caused by a behavioral mechanism, valve closure. This action would isolate the mussels from a source of oxygen, leading to a drop in pO2 within the mantle cavity. Previous studies had shown that a reduction in pO2 of the mantle cavity resulted in a reduced heart rate in mussels (Bayne, 1971; Curtis et al., 2000). The main purpose of my study was to determine whether valve closure causes the decline in heart rate observed in mussels at high temperatures. Because the relationship between valve movement and heart rate in mussels is unclear, another goal of this research was to determine whether there were any patterns of valve movement that were correlated to heart rate. While Braby and Somero (2006) conducted their research on Mytilus trossulus and Mytilus galloprovincialis, this study used specimens of Mytilus californianus, the ribbed mussel. The latter species is the dominant mussel found in the rocky intertidal zone of Central California, so it is a relevant study species for investigating the physiological ecology of this habitat. Materials & Methods Animal Collection and Care I collected fisty specimens of Mytilus californianus from the intertidal zone of China Point at Hopkins Marine Station in Pacific Grove, California on April 24, 2007. These individuals were placed in mesh bags and kept in an outdoor aquarium with a flow-through system supplying ambient seawater. Heart Rate & Valve Movement Measurement The mussels' heart rates were measured by impedance pneumography, following the methods of Braby and Somero (2006). In each mussel, I used a Dremel MultiPro"M Cordless drill to drill holes approximately I mm in diameter on either side of the posterior end of the hinge region, into the pericardial space. Next, fine copper electrodes, with the first millimeter of insulation scraped off, were inserted into these holes and glued into place with Loctite SuperGlueM. Once the glue had set (approximately 15 minutes), the electrodes were connected to an impedance converter (UFI, Impedance Converter Model 2991). From this point, the voltage reading was recorded through a data acquisition device (AD Instruments PowerLab/ 16SP). After completing the heart rate measurement set up, I glued a small magnet onto one valve of the mussel, on the posterior end opposite from the hinge, using Loctite SuperGlueTM. A linear Hall-effect sensor (Allegro Microsystems, Inc.) that was covered in heat shrink wrap was glued onto the other valve across from the magnet (Fig. 1). This sensor gave a voltage reading that depended on the distance between the magnet and the sensor, which was recorded through the PowerLabM data acquisition system. Higher voltage readings corresponded to a smaller distance between the magnet and the sensor, indicating that the mussel's valves were more closed. The mussels were placed in 14°C water for at least one hour to recover from handling stress. Temperature Experiments In order to test the effects of temperature on heart rate and valve movement in water, I placed up to four mussels at a time in an insulated container, either by letting them hang from their glued sensors, which were supported by a metal rack arm, or gluing a small piece of cork onto one of their valves and clipping this cork onto the arm. An air pump (TetraTec Deep Water DW 96-2) supplied oxygen into the container. The mussels were subjected to increasing temperature ramps from 14°C to 34°C over 2.7 hours (approximately 7.4°C per hour), controlled by a temperature regulator (Lauda Brinkman, RC 6 CS, Westbury, NY, USA). Twenty mussels underwent this treatment. Four other mussels were used as biological controls and kept at 14°C for 2.7 hours after heart rate and valve movement measurement preparation. In order to test the temperature dependence of the voltage reading from the Hall-effect sensors, three sensors with magnets attached were placed in the water for the duration of the temperature ramp as an experimental control. Water temperature was recorded continuously throughout all experiments. Data Analysis After obtaining the heart rate tracings, I counted the number of heartbeats per minute once every 5 minutes of the temperature ramp, as described in Braby and Somero (2006). Next, I found the approximate Hert for each mussel and counted heartbeats for each minute in the 10 minutes before and after the approximate Herit. To find the actual value of Hert for each mussel, I took best-fit lines of the heart rates before and after the heart rate decline began, and I found their intersection point (Fig. 2). Since the starting heart rates varied considerably among individuals, I normalized the heart rates so that I could pool the data. The valve data produced a voltage reading every second. In order to compare the relative position of the valves among mussels, I normalized all the valve voltage data, as for the heart rates. Iaveraged the valve voltage reading from before Hait and after Harit and conducted a paired t-test (q-0.05) between these values for all mussels. I also did the same test on the control specimens, using the average time of at which Herit occurred from the temperature ramp mussels to divide the control set data into before and after "Herit": Results Out of 20 mussels used in the water heat ramp experiments, only 12 had heart rates that could be easily read. These mussels exhibited Hert values that ranged from 22.9°C to 29.1°C. On average, the mussels remained relatively open through the first half of the heat ramp, after which they began to close (Fig. 3a). This closure continued while the heart rates were rapidly dropping. Upon inspection of valve traces for individual mussels, it was apparent that some mussels follow this behavior more than others. In almost half of the mussels, the general patterns seen at the beginning of the temperature ramp were not as easily distinguishable, though in most mussels the closure that occurred sometime after Hart was readily observable. This individual variability was quantified in the larger values of standard error before the valves began to close than after (Fig. 3b). The average pre-Hert valve voltage reading was significantly lower than the average post- Hert valve voltage reading (P +0.0001) (Fig. 4). This result indicates that the valves were more open before Hert than after Hert. In contrast, the control specimens showed no significant difference between the voltages before and after the time at which average Herit occurred in the experimental group (P = O.48). The experimental control data showed a trend for the voltage reading from the linear Hall-effect sensors to decrease slightly with increasing temperatures (Fig. 5). It was not possible to determine the magnitude of the effect on each sensor, however, because this value was not simply dependent on starting voltage. When compared to the amount of movement in the mussels' valves over the same period of time, however, the effects of the voltage's dependence on temperature were minimal (less than 10% of the total movement) and would not have accounted for the magnitude of voltage change observed. Discussion The results of this study suggest that the heart rate decline observed in mussels at high temperatures is not simply caused by a behavioral mechanism, valve closure. As seen in the comparison of valve position before and after the critical temperatures, the valves were significantly more closed after Hert than before, indicating that the valves do not reach their most stable closed point until after Hert has occurred. During the first part of the temperature ramp, high variation in behavior among the individual mussels correlated with observations of other researchers that have noted the distinct differences in valve movement among individual mussels (Loosanoff, 1942; Curtis et al., 2000). In contrast, the lower variation observed when the mussels' valves began to close and when they reached their most closed position implies that this valve closure is a programmed response to the stressful conditions of high temperatures. The dependence of the Hall-effect sensor voltage reading on temperature may be a confounding factor in data analysis, because it automatically biases the valve readings. However, because the bias was in the direction of decreasing voltage as temperature rose, it would have in effect counteracted the behavior observed around the critical temperatures, where the voltage from the sensors increased. Thus, it is possible that the valves were even more closed than they appeared to be. These data provide evidence for a complex relationship between valve closure and heart rate. It is clear that there are patterns that arise in both heart rates and valve closure in response to steadily increasing temperatures, but the question that remains is whether the valve movement and heart rates are causally connected in some way, or if they are perhaps independently affected by this environmental factor. Studies on other factors that cause a heart rate response and their relationship to valve movement have found that the correlation between valve closure and heart rate is not easily quantified. For example, Curtis et al. (2000) studied these factors in relation to increased levels of copper in the water. While there was an easily observed connection between long periods of valve closure and reduced heart rate, the rapid fluctuations in valve movement were not reflected in changes in the mussels' heart rates, and the authors concluded that the observed changes in heart rate were a direct result of increased copper concentrations. Similarly, the rapid heart rate decline seen in this study may be a physiological consequence of the increasing temperature, and may not be related to valve closure. These findings offer a basis for future studies on the relationship between valve closure and heart rates. In regards to the dependency of heart rate on valve closure, further research should focus on the differences among various mussel species in valve movement patterns around Herit. Since different species in the Mytilus genus have shown different average values of Herit (Braby and Somero, 2006), it should be determined whether the different species show the same valve movement in situations of increasing temperature. Another area of future research could be the connection of these two factors with other environmental stressors, such as emersion. Studies have shown that heart rates will change when mussels are placed in air (Bayne et al., 1976). The two general responses observed in mussels are either complete valve closure or periodically gaping valves. While the former results in an anaerobic environment for the mussels and the latter allows the mussels to continue aerobic metabolism, valve gaping also correlates with increased threats of desiccation (Byrne et al., 1990; Shick et al., 1988). Thus, this trade-off between loss of water and metabolic efficiency may evoke different patterns in valve movement than those observed in this study. Acknowledgements I am grateful to my advisor, George Somero, for his steady support, interesting ideas, and continual excitement in my project. Several members of the Somero lab were important contributors to this project, especially Brent Lockwood for his helpful advice and guidance, and Jon Sanders for his expanse of technical knowledge and good ideas. I am grateful for John Lee's assistance with making the Hall-effect sensors and excellent technical advice. Finally, I thank Jim Watanabe for his help in making statistical sense of the data in this project. Literature Cited Bayne, B. L. 1971. Ventilation, the heart beat and oxygen uptake by Mytilus edulis L. in declining oxygen tension. Comp. Biochem. Physiol. 40 A: 1065-1085. Bayne, B. L. ed. 1976. Marine Mussels: Their Ecology and Physiology. Cambridge: Cambridge University Press. Bayne, B.L, C. J. Bayne, T. C. Carefoot, and R. J. Thompson. 1976. The Physiological Ecology of Mytilus californianus Conrad. Oecologia. 22: 229-250. Braby, C. E. and G. N. Somero. 2006. Following the heart: temperature and salinity effects on heart rate in native and invasive species of blue mussels (genus Mytilus). J. Exp. Biol. 209: 2554-2566. Byrne, R. A., E. Gnaiger, R. F. McMahon, and T. H. Dietz. 1990. Behavioral and metabolic responses to emersion and subsequent reimmersion in the freshwater bivalve, Corbicula fluminea. Biol. Bull. 178: 251-259. Curtis, T. M., R. Williamson, and M. H. Depledge. 2000. Simultaneous, long-term monitoring of valve and cardiac activity in the blue mussel Mytilus edulis exposed to copper. Mar. Biol. 136: 837-846. Depledge, M. H., A.-K. Lundebye, T. Curtis, A. Aagaard, and B. B. Andersen. 1996. Automated interpulse-duration assessment (AIDA): a new technique for detected disturbances in cardiace activity in selected macroinvertebrates. Mar. Biol. 126: 313-319. Gudimov, A. V. 2003. Elementary behavioral acts of valve movements in mussels (Mytilus edulis L.). Doklady Biol. Sci. 391: 346-348. Kramer, K. J. M., H. A. Jenner and D. de Zwart. 1989. The valve movement response of mussels: a tool in biological monitoring. Hydrobiologia. 188/189: 433-443. Hochachka, P. W. and G. N. Somero. 2002. Biochemical Adaptation: Mechanism and Process in Physiological Evolution. Oxford: Oxford University Press. Loosanoff, V. L. 1942. Shell movements of the edible mussel, Mytilus edulis (L.) in relation to temperature. Ecology. 23 (2): 231-234. Manley, A. R. and J. Davenport. 1979. Behavioural responses of some marine bivalves to heightened seawater copper concentrations. Bull. Environ. Contam. Toxicol. 22: 739-744. Nicholson, S. 2002. Ecological aspects of cardiac activity in the subtropical mussel Perna viridis (L.). J. Exp. Mar. Biol. Ecol. 267: 207-222. 12 Sabourin, T. D. and R. E. Tullis. 1981. Effect of three aromatic hydrocarbons on respiration and heart rates of the mussel, Mytilus californianus. Bull. Environ. Contam. Toxicol. 26: 729-736. Shick, J. M., J. Widdows and E. Gnaiger. 1988. Calorimetric studies of behavior, metabolism and energetics of sessile intertidal animals. Amer. Zool. 28: 161-181. Figure Legends Fig. 1: Schematic of Hall-effect sensor system. Fig. 2: Heart rate versus temperature. Data for one individual, with best-fit lines for the increasing and decreasing segments and equations shown. Hart was calculated by finding the intersection point for these two lines. BPM = beats min". Fig. 3: Heart rate and valve position. (a) Average normalized heart rate (BPM) and average normalized voltage signal plotted against temperature for all mussels in the temperature ramp treatment. Lower voltage corresponds to more open valves, and higher voltage reflects more closed valves. Error bars on heart rate curve represent standard errors. (b) Average heart beat and valve movement for all mussels, with error bars for voltage based on standard errors. Fig. 4: Valve position relative to Herit. Voltage differences between average pre-Hert and post- Hgit values. For the control, a value for the theoretical t" was determined to be the average Herit found among all mussels in the temperature ramp treatment. Thus, for the comparisons of the control specimens, averages of voltages before and after the time at which the theoretical Hert was reached were used. Paired t-tests were used to compare the change in voltage from before to after Herit. * indicates P0.05. Fig. 5: Linear dependence of Hall-effect sensor on temperature. This trend could not be predicted accurately and thus could not be subtracted from the valve voltage readings for the mussels, but le voltage c the magnitude of ge over the temperature ramp was relatively small in comparison to the amount of movement seen in the mussels (less than 10%). Magnet Fig. 1 Hall-effect sensor Powerlab Computer (Voltage Reading) 45 40 35 30 25 20 15 Fig. 2 k y-1.7857x -5.8496 . y =-9.8993x + 295.81 23 17 27 Temperature (C) b. Fig. 3 — Lalve Mæment 0.9 ——Heart Rate 08 07 06 05 04 W 0.3 02 15.39 16.54 19.27 21.96 24.87 27.79 30.72 Temperature (C) Uhhe Mement 0.9 —— Heart Rate 0.8 0.7. os 03 02 15.39 16.64 19.27 21.96 24.87 27.79 30.72 Temperature (C) 9 08 07 06 05 03 02 0.1 0.9 08 07 6 03 02 0.1 0.9 20.8 30.7 0.6 9 0.5 0.4 0.3 0.2 20.1 Fig. 4 Post-Hcrit“ Pre-Hcrit“ Control Pre-Herit Post-Hcrit Temperature Ramp 0.7 0.6 0.5 0.3 0.2 — Sensor Voltage 0.1 —•— Water Temperature 0:00:00 0:33=20 1:06:40 1,40:00 2:13:20 2:46=40 3:20:00 3:53:20 4=26:40 Time (Hours) 20