Abstract During development, organisms must establish properly functioning and well coordinated sensory and motor systems in order to survive. Pacific bonito (Sarda chiliensis), pelagic fishes belonging to the tuna family, undergo a relatively rapid developmental transformation in the X24¬ hour period after hatching. The two major morphological changes that occur are enlargement of the jaw and pigmentation of the eye (retinal epithelium). In addition, dramatic behavioral changes take place. For example, early (stage LI) animals frequently exhibit fast start escape responses while in the later stage (L2), more controlled sustained swimming movements are present. Li larvae are also more sensitive to touch and vibration stimuli than L2. To test the hypothesis that the emerging visual system changes the activity of descending motor systems, I recorded field potentials in the hindbrain and spinal cord during this developmental period. Recordings of spontaneous activity showed that there was essentially no difference between LI and L2 larvae. either in the light or dark. The same was true for L2 larvae reared in the dark. Looking at the strong visual stimuli, larvae reared in dark exhibit a movement response but recover more slowly than those reared under normal (daylight) conditions. This finding suggests that dark rearing negatively affects the sensitivity of the developing visual system, however appropriate connections with the hindbrain still appear to be made. Based on these observations, it appears that visual inputs might be sufficient to modify the activity of an identified reflex circuit, and more specific experiments are warranted to identify the physiological basis for this change. Introduction During development, organisms need to establish properly functioning sensory and motor systems in order to survive. These systems also must be well coordinated to changing environmental conditions. Pacific Bonito (Sarda chiliensis), pelagic fishes belonging to the tuna family, undergo a relatively rapid developmental transformation early in their larval life history, Some of the morphological changes that occur during this period include jaw enlargement, pigmentation of the retinal epithelium, and lateral migration of the neuromasts. These changes take place about one day after hatching, are complete within 24 hours and are accompanied by a change in behavior. The behavioral changes include less sensitivity to vibration and touch. horizontal orientation as opposed to vertical, and slower more sustained movement. Most fishes and amphibians possess a well conserved stimulus/response pathway that underlies sensation of tactile stimuli. The receptor for the sensory limb of this pathway is the neuromast, which is anatomically and physiologically similar to hair cells. Neuromasts transmit somatosensory information to the CNS via the lateral line nerve to motor centers in the hindbrain. For most fish, signals converge at the Mauthner cell (M-cell) which is a large neuron found in the medulla oblongata. In goldfish, the M-cell has two main dendrites. The lateral dendrite receives input from the ear and the ventral dendrite receives input from the eye through the optic tectum (Zottoli et al., p. 153). This neuron coordinates muscular activity by propagating the signal down a centralized pathway. This pathway is thought to be responsible for generating the large accelerations that occur during the escape response. Stimulation of a single M-cell produces a forceful flip of the truck and tail to the side opposite of the M-cell. In addition, sudden movement of both eyeballs, both operculi, and the lower jaw is synchronous with the tail movement (Diamond, p. 266). These experiments looked at the possible influence the emerging visual system may gain in the developing larvae. Because a main source of sensory input to the M-cell is from the eve changes in behavior associated with the pigmentation of the retinal epithelium may be a sign that the visual system is now influencing escape behavior mediated by the M-cell to a greater extent (fig. 1). Animals in larval stage 1 (LI) and larval stage 2 (L2), morphologically and behaviorally, are very different. LI begins right after hatching. The head is small and in relative proportion to the rest of the body. Blaxter (1986) found that the eyes of many teleost species do not have pigmentation and "are almost certainly non-functional" at hatching. Retinal epithelium pigmentation is absent in LI larvae suggesting that their visual system is not fully functional. Blaxter (1986) also found that by first feeding all species had pigmented eyes. After hatching, LI larvae, who are generally not very active rely on the yolk sac for nutrients and do not need to feed. However as the yolk sac is depleted, the larvae need to be able to catch prey. This requires having a jaw that is large enough to capture prey, mastering feeding behavior, and having a functional visual system that allows the larvae to locate prey. In contrast to LI, L2 occurs about the time when the yolk sac is beginning to be depleted, larvae have an enlarged jaw, which opens and closes in a manner characteristic of adult feeding behavior, and retinal epithelium pigmentation. LI and L2 larvae also exhibit differences in responsiveness to tactile and visual stimuli. LI larvae, though not generally very active, are extremely responsive to vibration or touch. When the tank is tapped or the larvae are mechanically stimulated, LI larvae typically exhibit a fast start escape response. In contrast, L2 larvae are much more active, move in a more sustained swimming manner, and have lost much of their sensitivity to stimuli especially vibration. Not observing a predator, the visual system may send an inhibitory signal to the command cell (M-cell) causing an L2 larva not to respond to simple vibrational stimuli. An underdeveloped visual system, not having the same influence in the sensory-motor pathway would not be able to inhibit this response and the larvae would be extremely responsive as is the case with LI larvae. In order to study the developing visual system, an electrophysiological setup was used to look at both spontaneous and light stimulated activity in the hindbrain and the spinal cord. LI, L2 dark reared (L2D), and L2 ambient light reared (L2L) larvae were compared. In respect to behavior, L2D larvae were generally more similar to LI larvae than L2L larvae. Though spontaneous activity in the light and dark were similar for all three groups, L2L larvae tended to have a broader range of movement frequencies with a greater amplitude of higher frequency movement in the dark and lower frequency movement in the light. In addition, a comparison between light and dark reared L2 larvae's behavioral response to a sequence of photic stimuli showed a difference in refractory time with L2D larvae needing a longer time to recover. Methods Larvae were obtained and hatched at Monterey Bay Aquarium (Monterey, California) They were left undisturbed in seawater kept at room temperature until after hatching, at which time they were moved to the lab. Some larvae were allowed to develop in ambient light and others were placed in a dark room. Eugenol was used to anesthetize the larvae because it works quickly and is reversible. The anesthetized larvae were then embedded in 1.2% agar (made with tuna Hanks) on its side and then allowed to recover. An electrophysiological setup was used to record the activity in the hind brain and spinal cord. Two glass electrodes were pulled using 1.5mm glass with a filament and filled with O.5 M Nacl. The two AC coupled differential recording electrodes were inserted into the hindbrain and spinal cord. Neural signals and movement artifacts were visualized on using an oscilloscope and recorded at 4 kHz using Axoscope (Axon). Only movement artifacts were analyzed. The following procedure was carried out for LI, L2 dark reared, and L2 ambient light reared larvae. Each larva was allowed to "dark adapt" for five minutes in a cage covered with tin foil on all sides except the front which had a dark tarp over the front of it. I recorded activity starting at minute four. At minute five, I administered a sequence of eight light flashes (-O.5 sec duration) with decreasing latency between them (fig. 2) The first flash being at 5 min, followed by one at 5:30, 5:45, 5:55, 6.00, 6.05, 6:10, and 6:15. The larva was then allowed to "dark adapt" again for 2 min at which point the microscope light was turned on. The larva remained on the setup for five minutes with the light off and the procedure was repeated, In order to look at the influence of the emerging visual system on the response pathway, two comparisons were made. Spontaneous activity (movement) in the light (recording when the light was on after the sequence of light flashes) and dark (recording at minute 4 after the larvae were allowed to dark adapt) were analyzed by plotting the power density spectra for both light and dark for all three developmental conditions. Power density spectra was used in order to look at differences in movement frequencies. The second comparison was a qualitative difference in response between L2 dark reared and L2 ambient light reared larvae to a sequence of photic stimuli. Due to a small sample size (n- 2), statistical analysis was not performed. Results Because eugenol wears off eventually, larvae were able to move although embedded in agar. Movement events (amplitude of electrode deflection) were fairly large (about 40x) in comparison to neural events and were therefore easier to separate out from the background noise. In general they tended to be long events, with multiple peaks that had both positive and negative deflections. Neural activity was not looked at in as much detail as movement because due to sampling frequencies and smaller amplitudes it was harder to distinguish from noise From the recordings, I was able to extract signals from the nosie. The Power Density Spectra (PDS) shows the relative abundance or amplitude of a given frequency for a given trace In contrast to a fast Fourier transformation, it uses a sliding box to calculate the relative abundances. Because the PDS is relative for each trace, a difference in the shape of the curves rather than in the magnitude of power shows the differences in the amplitude of a given frequency. Figure 3 shows a difference between movement traces and baseline noise between 10 and 60Hz. plotted the PDS for spontaneous activity in the light versus activity in the dark (fig. 4). LI and L2 dark reared larvae are shown to have similar movement frequencies. L2 ambient light reared larvae also has these movement frequencies as well as a greater abundance of higher frequency movements in the dark and lower frequency movements in the light. In general however, activity levels are consistent in the light versus the dark for all three rearing conditions. In response to photic stimulation, there is a difference between ambient light and dark reared larvae. The response to the initial flash and the one 30 seconds later in both rearing conditions was similar. It tended to be long (40-60 ms) and was large in amplitude with multiple positive and negative deflections. The response to the short latency flashes (10 sec and 5 sec after previous flash) also was similar between the rearing conditions. It was small with usually only one or two positive deflections. The difference between the two rearing conditions came with flash 15 sec after the previous one (fig. 5). Ambient light reared L2 had a large response like that for the initial and 30 sec flash. However, the dark reared L2 had a response characteristic of the shorter latency flashes. In addition, a few trials were completed in which there was an initial flash followed by one 5 sec later in which both ambient light and dark reared larva responded in a similar manner to the short latency flashes of above (fig. 5). Discussion The results from the spontaneous activity in light versus dark experiment was interesting in that it showed that L2D larvae, although appearing the same as L2L larvae morphologically, were more similar to LI larvae behaviorally. It appears that activity levels are fairly consistent among all three groups in both the light and dark for mid range frequencies, but diverges at the higher and lower frequencies for L2L larvae. A possible explanation for the greater amplitude of lower frequency events in the light for L2L larvae is that these larvae tend to be more active in the light. Also being embedded on their side, which is not the natural orientation, may cause the larvae to struggle to orient towards the light. The LI larvae and L2D larvae, not being as responsive to light because of their less developed visual system, and the L2 larvae in the dark without the light cue are not struggling as much to reorient. The photic stimulation results suggest that dark rearing affects properties of the visual system. The difference in response could be due to either slower refractory time or faster adaptation to the stimuli by dark reared larvae. Because a few trials were done where an initial light flash was followed by a second one 5 sec later, the difference is most likely due to a longer refractory time. This assumes that refractoriness recovers aster a time greater than 5 sec and less than 30 sec and adaptation does not. Without out the photic stimulation to the visual system, it may have not developed to the same complexity or sensitivity as in the light reared larvae. There are other complications with dark rearing. Growth and development does not occur in the dark in the wild and therefore due to the abnormal growth environment some of the differences are most certainly due to impacts of the dark on systems other than the visual system. Other methods of testing the absence or underdevelopment of the visual system would be to dissect out the eye, cut the optic nerve, or inactivating the optic tectum by injecting with lidocaine. Each of these techniques would look at different parts of the visual pathway. By looking at them all, we may be able to see which part of the emerging visual system is developing at a given time in growth. In fürther studies, there should be better control of temperature and light. It is possible that the dark reared larvae were slightly cooler than the ambient light reared larvae which could have an effect on the rate of development. Ambient light in Monterey is extremely variable with some days being very sunny and others (the majority) being mostly foggy. Therefore it would be helpful to put them on a 12:12 light:dark cycle. It would also be good to work in darkness while setting the larvae and inserting electrodes as opposed to letting them light adapt and then dark adapt again. Other possible experiments include recording in the optic tectum to find out what type of activity it modulates. Because input from the eye travels through the optic tectum to the M-cell. recording in the optic tectum could possibly narrow down where less developed areas in the visual system between LI and L2 larvae are. If activity is found in the increased activity is found in the optic tectum between the 2 stages, then it would suggest that the difference in development is before that point (eg having to due with lack or presence of eye pigmentation). However, if activity levels are the same, this would suggest that the difference is after the optic tectum and the link between the optic tectum and the M-cell could be what is less developed between the 2 stages. In order to better look at action potentials and neural activity, it would be helpful to use curare in future experiments in order to suppress movement in the larvae Acknowledgments Special thanks to: Matthew McFarlane who spent endless hours in lab with me making sure everything worked and for teaching me a love for bonito. My advisor Stuart Thompson who always had so much energy and excitement for the project. And Christian Reilly who provided help not only on the project, but also sanity breaks by talking to me about everything except bonito. 10 Literature Cited Blaxter, J. H. S. 1986. Development of sense organs and behaviour of teleost larvae with special reference to feeding and predator avoidance. Transactions of the American Fisheries Society. 115:98-114. Diamond, J. 1971. The Mauthner cell. pp. 265-345 in Fish Physiology, vol.5, Academic Press, New York. Zottoli, S. J., A. Bentley, B. Prendergast, and H. Rieff. 1995. Comparative studies in the Mauthner cell of teleost fish in relation to sensory input. Brain Behav. Evol. 46.151-164. Figure Legends Fig. 1. Fast start escape response pathway. Influence of the emerging visual system on M-cell mediated response is unknown. Fig. 2. Photic stimuli sequence. Larvae allowed to dark adapt for 5 min before the administering of 8 light flashes. Fig. 3. Power Density Spectra (PDS) of movement and baseline noise. Lower frequencies are movement whereas higher frequencies are noise. Movement traces were determined by watching the larvae under the microscope and correlating when they moved and the activity on the oscilloscope. Fig. 4. PDS activity in the light vs. dark for LI, L2D, and L2L. LI and L2D are very similar. LZL has a greater amplitude of higher frequency movement in the dark and lower frequency movement in the light. Fig. 5. Activity traces for photic stimuli after differing latencies between flashes. Shows an increased refractory time for L2D. 12 0 Ked eue5 -------- 8 5 : 30. Gt.C Og:g 00g 8 S (silun Keligie) IeMo O veg sod 9