Abstract Ectothermic animals in the intertidal zone perform complex behaviors over a large range of temperatures. Based on their ability to function at such different temperatures, their nervous systems must compensate to accommodate for environmental changes, either in order to keep behavior the same or to change the animal’s behavior in favor of homeostasis. To look at how a nervous system behaves at different temperatures, isolated ganglia from the yellow dorid nudibranch, Doriopsilla albopunctata, were exposed to temperature ranges simulating those of the intertidal zone (10 to 25 degrees Celsius). Single electrode intracellular recordings were taken to study the behavior of individual neurons. Variation in spontaneous activity, firing frequency, utilization time and synaptic potentials revealed individualistic, non-uniform responses to temperature on the neural level. The data suggest that neurons in Doriopsilla albopunctata are not individually temperature compensated, and therefore temperature compensation does not happen on the neuronal level. Nor do the data show that all nerve cells respond to temperature the same way. Instead these results suggest that the overall maintenance of neural function across a temperature gradient arises from properties of the system as a whole rather than individual neurons. Introduction It is remarkable that ectothermic animals can live in places such as the intertidal zone. They cannot regulate their body temperatures to keep functioning properly, yet they perform complex behaviors in a wide range of temperatures as the weather and the tides change around them. Presumably, their nervous systems have ways to compensate for such dramatically changing conditions. Some work has been done in the past on the effects of temperature on the nervous system. Knape (1999) presented evidence that the spontaneous action potential frequencies in sensory nerve bundles in the walking legs of porcelain crabs varies more with temperature for low intertidal species than high intertidal species, indicating a nervous system compensation to temperature for the high intertidal species. Johnson. Smith, and Thompson (1986) suggested that temperature compensation in bursting pacemaker cells in molluscs may be accomplished by a temperature - dependent gradation in the relative contribution of different ionic currents. Anderson (1976) found that the rate at which bursts are produced is relatively insensitive to temperature, again suggesting temperature compensation. The question then is whether their nervous systems are in fact temperature compensated. And if they are, how? One hypothesis is that temperature compensation happens within each individual neuron. If this were true, each neuron would be able to adapt such that its behavior did not change at all within the temperature range the animals experience. Another hypothesis is that the neurons do change with temperature, but they do so uniformly. If this were the case, the whole integrated system could work the same way at every temperature, just at a different base level. A third hypothesis is that the neurons are individualistically temperature sensitive. The nervous system as a whole, individual circuits or intercellular interactions would compensate for temperature changes by integrating the information in different ways at different temperatures in order to maintain constant behavior over the whole range of temperatures. Fourthly, the neurons and the nervous system as a whole might vary with temperature such that the animal’s behavior changes in a way that is adaptive to the changing environment and helps it maintain relative homeostasis. To begin to test the first two of these hypotheses and obtain preliminary data for the second two, 1 used Doriopsilla albopunctata as the study organism because it is an ectothermic, subtidal and intertidal organism that has large, easy to work with neurons. Materials and Methods Doriopsilla albopunctata were collected intertidally at the Great Tide Pool in Pacific Grove, California. They were stored in flowing natural seawater tanks at approximately 13°C with tunicates and sponges to eat. Ganglia were surgically removed and treated with dispase, a neutral protease, for one hour at 4°C followed by a one to 12 hour soak in sea water (again at 4°C) to facilitate removal of the epineural sheath, which was then removed using forceps. It has been well documented that mulluscan ganglia show the same types of record activity, including well-coordinated activities, when isolated as they do inside an intact animal (for example, Treistman 1985) The ganglia were left intact and pinned to a dish using cactus spines. Single electrode intracellular recordings were taken from individual neurons in the pleural ganglia using electrodes with resistances between 1 and 6 megohms and filled with 3M KCl. The dish was placed on a Peltier heating and cooling device that increased and decreased the temperatures over periods of two to four hours between 10 and 20 or 14 and 25°C to simulate more or less what the animals might see in the intertidal zone (Somero and Tomanek 1999). Voltages were recorded continuously throughout each temperature ramp using Clampex 8.0. In the case of neurons that showed no spontaneous activity, a depolarizing stimulus (constant throughout an entire trace but not between experiments) was used. Results Pacemaker neurons Two neurons exhibited spontaneous activity at some time during the temperature ramp. One was in a bath that was heated from 10° to 20°C over one hour then cooled from 20° to 10° over the same time interval (Fig. 1). It showed bursting pacemaker activity at the beginning of the experiment. As it was heated, the frequency of the action potentials increased, and it changed from firing bursts of action potentials to constant action potentials. At 20°, it fired a few last, high frequency bursts then ceased to fire. During the cooling to 10°, it did not regain its firing behavior nor did it start again even after two hours in a 10° bath after the experiment. Throughout this time, it could be stimulated to fire action potentials with a depolarizing current. 1 observed a second neuron as it was heated from 14° to 25°C over two hours then cooled from 25° to 14° over the same time interval (Fig. 2). This cell did not show any spontaneous or endogenous activity at 14° or for the first hour and a half while it was heated. At 19.5°, it began firing spontaneous action potentials which continued and increased in frequency as the temperature increased to 25°. During the two hours while the temperature returned to 14°, the frequency of the action potentials decreased and started to show a bursting pattern. This behavior continued for the next two hours at 14 instead of returning to the silent state it was in before the ramp. Neurons with no spontaneous activity-firing frequency Trecorded from two cells that showed no spontaneous activity at any point during the two hour temperature ramp from 10° to 20°C and back. To compare the frequency at which they fired action potentials in response to a stimulus, they were given 5 second depolarizing stimuli periodically throughout the temperature ramp to elicit trains of action potentials (Fig. 3). The depolarizing current was constant throughout each experiment, but was not the same for the two different neurons. The average firing frequency during the 5 second stimuli of the first neuron increased with temperature, and then decreased to a lower frequency than it had before the ramp (Fig. 4). The firing frequency of the second neuron also increased with temperature. However, it remained high even after the bath was cooled to 10° again (Fig. 5). Neurons with no spontaneous activity - utilization time I also looked at how the utilization time (the interval between the onset of the depolarizing stimulus and half peak of the first action potential) changed with temperature in the two neurons I discussed in the section on firing frequency. Paralleling the changes in firing frequency, the utilization time of the first neuron decreased with temperature then increased again while the temperature returned to 10°C (Fig. 6). The utilization time of the second neuron decreased with increasing temperature then continued to decrease even while the temperature returned to 10° (Fig. 7). Synaptic potential Tlooked at the synaptic input received by one neuron within an intact ganglion during a two hour temperature ramp from 10° to 20°C and back to 10° (Fig. 8). Looking at the half time to rise and half time to fall of individual excitatory post-synaptic potentials at different points along the temperature ramp, little change in the half time to rise was apparent but half time to fall increased dramatically at the end of the temperature ramp. Photo response In order to record a poly synaptic response in a pleural ganglion neuron, I recorded its activity in response to turning off the light. The ganglion was left intact so it could sense changes in light. The photo response of the intact ganglion caused an excitatory post-synaptic potential in the pleural ganglion neuron at 10°C and an action potential at 20°C (Fig. 9). Discussion Pacemaker neurons Pacemaker neurons generate patterned electrical activity in the absence of synaptic input from other cells. Neurons of this type control a wide range of physiological processes including respiration, peristalsis, locomotion and sensory perception - all necessary for the animal’s survival. I would expect that some pacemaker neurons would have a way to compensate for temperature changes to maintain a constant output in all the temperatures in the animal’s environment. Work by Johnson, Smith and Thompson suggested differential temperature sensitivities of potassium currents that regulate the bursting frequency of bursting pacemakers to allow the animals to behave the same at different temperatures. In the case of certain pacemaker neurons, this may be the case. In the two I observed, however, their behavior changed dramatically with the changes in temperature. Because 1 do not know what processes these two particular neurons control and could not see what effects they have on the behavior of the animal, I can only guess about the meaning of such obvious changes in behavior on the part of the two individual neurons. Fletcher and Ram (1991) presented similarly interesting results working with R15 bursting pacemaker neuron in Aplysia, noting that high temperatures reversibly silenced the cells and that a return to low temperatures allowed most R15 neurons to resume bursting activity but with evident changes from pre-heating activity. Moffett and Wachtel (1976) demonstrated that at really low temperatures, some bursting pacemakers are not active as well as that whole animal behavior could be elicited only at temperatures at which the nervous system showed a sufficiently high general level of activity and bursting pacemakers were active. Perhaps the two neurons I observed change their behavior according to temperature in order to change the behavior of the animal in a way that is adapted to the environment in which it finds itself. One neuron could send the signal for the animal to seek shade to get back to a more livable temperature. Or, it could tell the animal that the tide pool in which it finds itself is too hot so it should not lay eggs there. I could not directly test these ideas, but they would be consistent with my fourth hypothesis (that the nervous system changes in response to temperature so that the animal’s behavior changes in response to temperature) and be an exciting area for further research. Another possibility is that the behavior of these two neurons changed in order to play their part to keep the whole integrated nervous system working the same even as each individual cell changed. Again, I did not test this hypothesis, but it may be possible. My results did directly refute the first two hypotheses proposed in the introduction, however. Both pacemaker neurons changed, and they changed in different ways, disproving the hypothesis that neurons do not change with temperature and that neurons change uniformly with temperature. The next four examples provide further evidence against these two hypotheses. Neurons with no spontaneous activity-firing frequency The firing frequency of action potentials of both neurons changed during the temperature ramp when stimulated with constant depolarizing current stimuli. The firing frequency of the first neuron increased with temperature then over-compensated by returning to a lower frequency at the end of ramp compared to the beginning (both at 10°C). The second neuron demonstrated strong hysteresis by remaining at the high frequency even after the temperature returned to 10 Not only do these two neurons change with temperature, they change hysteretically and differently, further disproving hypotheses 1 and 2. Neurons with no spontaneous activity - utilization time This interval is largely determined by an outward, inactivating, Shaker-like potassium current called A current. The balance between it and several inward currents determines the time it takes for an action potential to occur after the onset of a depolarizing stimulus (Connor and Stevens 1970 and Smith and Thompson 1987). Temperature evidently changed this balance in both neurons I observed. The first cell behaves in a way one might expect - the interval decreases with an increase in temperature then increases when the temperature returns to lower temperatures again. I did not test the mechanism in this experiment, but maybe the potassium current inactivates more quickly at higher temperatures or maybe the potassium current decreases at higher temperatures. Either way, the kinetics of the action potential change with temperature. The other neuron shows a different and in some ways more interesting and complex response to temperature. In its case, the utilization time decreases during the increase in temperature as it does in the previous example, but then continues to decrease even after the temperature decreases again. For this nerve cell, the temperature change has a large and long-lasting effect on the kinetics of the currents that determine action potentials that causes the utilization time to continue to shorten even after the temperatures return to those at the beginning of the experiment. Synaptic potential The behavior of individual nerve cells in response to temperature changes is important, but so are neural networks and intercellular interactions if we are to know the nature of how the whole nervous system responds to temperature. Here, I tried to gain some insight into how the kinetics of the neurotransmitters that neurons use to communicate with each other change with temperature. Recording the half time to rise of excitatory post-synaptic inputs in one cell as it received information from neighboring cells, it appeared that the half time to rise - or the time between the onset of the input and half of the maximum depolarization - increased to a small degree throughout the temperature ramp. The half time to fall - or the time between the peak depolarization and half way back to baseline - increased substantially at the end of the temperature ramp. Again, I did not look into the mechanism, but I can make guesses and propose areas where fürther research could tell us more about how the nervous systems of these ectothermic, intertidal animals deal with temperature gradients. Perhaps when the temperature increased, the pre-synaptic neuron started releasing more neurotransmitters to compensate for some other temperature-related changes in the system. Then when the temperature returned to 10°, the kinetics returned to how they had been before the ramp but the pre-synaptic neuron was still releasing an elevated amount of neurotransmitter. The small increase in half time to rise and the large increase in half time to fall then could be due to the neuron’s ability to deal with the extra 11 neurotransmitter. Alternatively, the high temperatures may have altered the conformation of the channels or the integrity of the enzymes that remove the neurotransmitters. Regardless of the mechanism, the temperature apparently caused a change in these neurons that had long lasting effects and that again showed a complicated temperature response. Photo response Another synaptic phenomenon I observed as it changed with temperature was photosensitivity. When I turned the light off at 10°C, it caused a depolarization in the neuron from which 1 was recording but that depolarization did not cause an action potential. At 20°, the same stimulus (turning off the light) did cause the neuron to fire an action potential. It appears that the action potential threshold was lowered at the higher temperature, again due to a change in the balance of inward and outward currents. When the neuron was 10° warmer, either the depolarization it experienced or its reaction to that stimulus was different. Conclusions The first mechanism 1 proposed for temperature compensation in these animals was that of individual neurons. All of the nerve cells I studied change with temperature and disprove this hypothesis. The second mechanism I proposed was that the individual nerve cells of the whole nervous system change uniformly with temperature. The hysteresis and obvious non-uniformity of my results disprove this hypothesis. A third possibility is that the individual neurons are very temperature sensitive and change in their own individualistic ways to changes in temperature. My results are consistent with this hypothesis, and further study is needed to determine the nature of such a complicated and integrated mechanism of temperature compensation if behavior does indeed remain the same. The fourth explanation is that the changes in the nervous system result in changes in the animal’s behavior. My experiment did not test this hypothesis directly, but my results are consistent with such a mechanism. I would like to see behavioral studies that connect the activity of the nervous system with how the animals behave at different temperatures. Overall, my results point to an integrated temperature compensation on the circuit and cell interaction level that is more complicated and integrative than temperature compensation in each neuron or a simple change in baseline as all the neurons change uniformly with temperature. Complex integration of the nervous system would allow these animals to maintain complex behaviors over the temperature range of their habitat and change their behavior adaptively. Acknowledgments 1 would like to thank Stuart Thompson for his time, energy and enthusiasm this quarter, John Lee for simulating the temperatures of the intertidal zone in the laboratory, and Freya Sommer for helping me collect Doriopsilla albopunctata and document their habitat. Thanks also to Rob DeConde his support, Christian Reilly for help making sense of my project, Ryan Laponis for introducing me to the world of electrophysiology, and Tori Arch for keeping me company in Fisher. Literature Cited Anderson, W. W. "Endogenous Bursting Tritonia Neurons at Low Temperature." Brain Research,1976. v. 103, p. 407-411. Connor, J. A and C. F. Stevens. "Inward and Delayed Outward Membrane Currents in Isolated Neural Somata under Voltage Clamp." Journal of Physiology, 1971. v. 213, p. 1-20. Fletcher, Stephen D. and Jeffrey L. Ram. "High Temperature Induces Reversible Silence in Aplysia R15 Bursting Pacemaker Neuron." Comparative Biochemical Physiology, 1991. v. 98, no. 3-4, p. 399-405. Johnson, J. W., Smith, S. J., and Thompson, S. “Slow Outward Tail Currents in Mulluscan Bursting Pacemaker Neurons: Two Compenents Differing in Temperature Sensitivity." The Journal of Neuroscience, November 1986. v. 6(11), p. 3169-3176. Knape, Jessica Elizabeth. "Temperature Adaptation of Neural Function among Intertidal Porcelain Crab Congeners from Different Thermal Habitats." Unpublished honors thesis. Stanford University, 1999. Moffett, S and Wachtel, H. “Correlations between temperature effects on behavior in Aplysia and firing patterns of identified neurons." Marine Behavior and Physiology, 1976. v. 4, no. 1, p. 61-74. Smith, Stephen J and Stuart T. Thompson. “Slow Membrane Currents in Bursting Pace- maker Neurones of Tritonia." Journal of Physiology, 1987. v. 382, p. 425-448. Somero, George N. and Lars Tomanek. "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, 1999. v. 202, p. 2925-2936. Treistman, S. N. “Effects of Sea Water Temperature on Bursting Pacemaker Activity in Cell R-15 in the intact Aplysia." Brain Research, 1985. v. 364, no. 1, p. 155-159. 15 Figure Legend Fig. 1. Change in the behavior of one pacemaker neuron during a two hour temperature ramp from 10° to 20 and back Voltage in millivolts on the y-axes, Time in milliseconds on the x-axes. One minute traces. Times in parentheses are times after the start of the temperature ramp. Fig. 2. Change in the behavior of a second pacemaker neuron during a four hour temperature ramp from 14° to 25 C and back Voltage in millivolts on the y-axes, Time in milliseconds on the x-axes. One minute traces. Times in parentheses are times after the start of the temperature ramp. Fig. 3. A train of action potentials during a 5 second depolarizing stimulus. Fig. 4. Average firing frequency during a 5 second depolarizing stimulus given to a neuron with no spontaneous activity during a two hour temperature ramp from 10° to 20 C and back. Fig. 5. Average firing frequency during a 5 second depolarizing stimulus given to a second neuron with no spontaneous activity during a two hour temperature ramp from 10°to 20 C and back. Fig. 6. Utilization time between a depolarizing stimulus and the first action potential during a two hour temperature ramp from 10° to 20 and back. Fig. 7. Utilization time between a depolarizing stimulus and the first action potential of a second neuron during a two hour temperature ramp from 10° to 20 ° and back. Fig. 8. The kinetics of unitary post-synaptic potentials during a two hour temperature ramp from 10° to 20 and back. Fig. 9. The post-synaptic potential in one neuron from a photo receptor after a light is turned off at 10° and 20 . 16 la. 10°C (0 minutes) O odten 1c. 20° (60 minutes) 2 1b. 15°C (30 minutes) vsnttttitttfttft r 1d. 10°C (120 minutes) n en en een e ene een Im Fig. 1 poo 2a. 10°C (0 minutes) ts mtn 2c. 25°C (120 minutes) neneaeuaeeeaa 2b. 19.5°C (90 minutes) stesenetoes datt Ere 2d. 10°C (240 minutes) tttttttt 18 Fig. 2 60 40 20 -20 -40 -60 -80 -100 e 5001 1001 2001 3001 Time (ms) 4001 Fig. 3 19 4.5 Fig. 4 — 10 to 20 degrees L 20 to 10 degrees Temperature (degrees Celsius) 20 8.5 — 5.5 5 10 11 Fig. 5 20 to 10 degrees —10 to 20 degrees 12 13 14 15 16 17 18 19 20 Temperature (degrees Celsius) 21 a Fig. 6 m 1 51 101 151 201 251 301 351 401 451 Time (ms -10 degrees (after ramp) — 15 degrees -20 degrees (top of ramp) - 15 degrees — 10 degrees (before ramp) 22 — Fig. 7 f o O N Time (ms) 10 degrees (after ramp) 15 degrees 20 degrees (top of ramp) - 15 degrees 10 degrees (before ramp) 23 400 250 200 Fig. 8 EHalf Time to Rise EHalf Time to Fall a- Temperature (degrees Celsius) 5 aern 100 a Time (ms) 24 40 -20 -120 L Fig. 9 10 degrees C 20 degrees C prrtndedaptete 301 401 101 201 501 601 701 Time (ms)