ABSTRACT Few observations have been made on Atolla, a bioluminescent deep-sea coronate scyphomedusa. With the use of a remotely operated vehicle and a specially designed plankton-kriesel, Atolla in the Monterey Bay have been videoed in situ, captured, and observed in lab to learn more about their behavior. A characteristic response to stimulation with light is described called the light reflex. The light reflex consists of a number of quick pulses and/or the medusa turning belly-up. Once in the belly-up position, Atolla rarely reacts to any further stimulus. In addition, scanning electron microscopy and light microscopy were used to take a preliminary look at the nematocyst composition of the trailer tentacle compared to the many short tentacles. e. INTRODUCTION In the past, because Atolla have generally been found between 500 and 800 meters depth, they have been relatively inaccessible for behavioral studies. Even early anatomical descriptions of Atolla were incomplete because the animals were caught in trawl nets which ripped off their tentacles. In more recent studies of the bioluminescent responses of Atolla, Herring noted that: the observed physiological responses of oceanic animals that have been caught in trawl nets rarely reflect the true capabilities of these animals in situ. Better methods of capture, maintenance and in situ observation are needed in order to appreciate the full range of responses which oceanic animals can employ and the circumstances in which they are normally elicited. (Herring 1990) Therefore, this study was conducted with a remotely operated vehicle (Etchemendy and Davis 1991) both to view Atolla of Monterey Bay in situ and to carefully capture it alive and intact for further lab studies. These live animals were maintained in a plankton-kriesel. This study has revealed that Atolla behave differently in different types of light and that they share a similar reflex, described as the "off-response" (Singla 1974; Arkett and Spencer 1986), with many hydromedusa. MATERIALS AND METHODS The Monterey Bay Aquarium Research Institute's remotely operated vehicle (ROV) was used to view Atolla in situ. The ROV provided video footage of Atolla under bright white lights and in turbulent waters. Thirteen Atolla wyvillei Haekel, 1880 and Atolla vanhoeffeni, Russell, 1957 were brought back. One was collected with the suction sampler on the ROV, the other twelve were collected in detritus samplers on the ROV. These animals were collected at depths ranging from 350 meters to 595 meters. Because of the delicate structure of these creatures, species identification was difficult and all specimens were referred to as Atolla. All specimens were kept in à specially designed plankton-kriesel with a constant current, at 6.5 dégrees Celsius (+ 5 degrees), and with a salinity concentration maintained at 34.5 parts per thousand (+- 1 part per thousand) Red Verses White Light The first two specimens were observed and studied in white light. The remaining eleven were observed in a room lit with six one-hundred watt red lightbulbs. The red light was adjusted with a rheostat and all observations were recorded under two different red- light conditions; either full intensity or half intensity. When not being studied, these animals were kept in darkness except when other people were working in the coldroom. The last eight Atolla were kept in the same conditions except black plastic was used to prevent any white light from reaching them unless it was an intentional part of an experiment. Thin White Beam With Red Background To observe reactions of Atolla to white light, specific body parts were stimulated. These include the trailer tentacle, short tentacles, bell, and bell margin. This beam originated from two different flashlights; one averaging 8.86 mE/m2/s and the other averaging 9.48 mE/m2/s when measured twelve centimeters away from the light source. Each was taped up with many layers of duct tape, allowing only a pinhole of light to shine through the tape. Full Flashlight With Red Background A full white flashlight was sometimes used both directly and indirectly to observe the behavior of Atolla when moving shadows were created in the kriesel and to observe how Atolla react to a sudden, full beam of white light. Unlit Room A night vision monocular (Litton M982) was used to gather information on Atolla kept in the darkness. Mechanical Stimulation Mechanical stimulation was used to catalog reactions of Atolla to stimuli other than light. Indirect mechanical stimulation: involved altering water conditions around Atolla. Eddies were created by stirring up the water in the kriesel. More subtle current changes were generated by using thin rubber tubing to gently blow bubbles near an animal. Direct mechanical stimulation; involved using a glass rod to manipulate Atolla directly. Observations were recorded for all experiments. Feeding Experiments A variety of potential prey items were introduced into the plankton-kriesel and any interactions were recorded over two to seven hour intervals spanned over one to four days. Adult and larval Artemia, Beroe, Aurelia, Nanomia, Pleurobrachia and gelatin were all items used. For nearly all the feeding experiments, only one type of possible food was offered at a time. However, one experiment was conducted on four Atolla in which both Pleurobrachia and larval Artemia were allowed in the kriesel simultaneously. Twelve hours later a tiny Beroe was added. Twenty-seven hours after the Beroe entered, clear gelatin was added to the kriesel. Nearly another day passed, and then more larval Artemia were added. After a few hours, ten adult Artemia were added. Live animals were used as prey and they were placed in the kriesel at least seven hours after Atolla were caught. In some cases, Atolla would remain in captivity many days before ever having an opportunity to feed. Scanning Electron Microscopy (SEM) Preparation of tentacle for SEM: Small portions of tentacles (about 1 cm long) were cut from a live, newly captured Atolla. These specimens were preserved in three percent gluteraldehyde and kept refrigerated. Further processing took place within thirty-six hours. The tentacles were then washed with seawater for five minutes, after which they were placed in 1% osmium tetroxide for one hour. No buffers were used. Dehydration with progressively increasing concentrations of ethanol began by doing two rinses at 30% for five minutes each and then continuing with washes of 40%, 50%, 60%, up to 70% at which point the specimens could be refrigerated for up to twenty-four hours. Dehydration continued at 80%, 90%, 95% and ending with two 100% washes. The tentacle samples were mounted on stubs, coated with gold, and then viewed with the Scanning Electron Microscope (Hitachi S450, 15,000 volts). Nematocyst Isolation Protocol (derived from Weber 1987): Whole tentacles or large portions of tentacles were severed from live jellyfish and washed in cold distilled water. Keeping each tentacle in separate containers, excess water was removed from them before they were frozen. Further processing occurred within thirty-six hours. Each tentacle was then thawed and homogenized with a pasteur pipette in a cold solution of distilled water containing fifty percent Percoll. These homogenates were placed on ice for thirty minutes and then centrifuged for ten minutes at 5,000 rpm. The liquid, debris and fragments were removed from each container and the nematocyst pellets (which were often too small to see with the naked eye) were frozen for less than thirty-six hours before being resuspended with a drop or two of distilled water. Light Microscopy Squash preparations of the tentacles, nematocyst isolations, and debris from the isolations were examined with an inverted compound microscope (Axiovert 10) at 100, 200, and 400 times magnification. RESULTS Video Footage When viewed by video with the ROV, Atolla generally seemed to pulse rapidly, often actively swimming away from the ROV. Atolla was also documented to stop pumping, and would roll over into what is designated as the "belly-up position. In this position the jellyfish rarely pumps, but rather drifts with the current and allows its tentacles to pile on top of its belly (Figure 1). Collections Out of the thirteen Atolla captured, seven were red and six were white. The red jellyfish tended to be more active and to react more vigorously to different stimuli. In captivity, the red jellyfish lived longer than the white jellyfish (Figure 2). Also, the red jellyfish tended to lose their pigment gradually from the moment of capture until death. =-Lab Observatioms=- White Light When viewed in constant white light, Atolla consistently cycles around the kriesel in the belly-up position (Figure 3). If kept in the full white light long enough (about twenty minutes) Atolla may partially spread out its tentacles (Figure 4). Red Light When viewed in red light Atolla tend to stay upright, alternating between actively swimming and drifting in circles around the tank (Figure 5). Only animals in extremely poor condition, which have been in captivity many days (over seven days), drift in the belly-up position when viewed in red light. Unlit Room When viewing the jellyfish in an unlit room with a night-vision monocular, Atolla behaved in the same way as it did under the red light. Thin White Beam Shining a thin white beam on different parts of Atolla produced varying responses. When the thin beam was directed at the bell margin and tentacle-base region of an up-right Atolla, the jellyfish would tend to turn belly-up (Figure 6) . Prior to assuming this belly up position, Atolla would perform a series of actions such as rapidly pumping, swimming away from the light source, darting all over the kriesel, and finally rolling over. I have referred to this series as "the light reflex, and have summarized it as "a number of quick pulses and/or turns belly-up." Two out of the twenty-nine times stimulated, a ring around the margin of Atolla bioluminesced before turning belly-up. The light reflex occurs at different intensities for different individuals. The red Atolla was more vigorous in all its responses compared to the white. The white jellyfish was never observed to bioluminesce and rarely darted around the kriesel. When the beam was directed at the tentacles, the tentacles were pulled in towards the bell and/or Atolla turned belly-up, twelve out of twenty-four times (Figure 7). A number of quick pulses were never associated with the thin beam directed at the tentacles. The red Atolla reacted in this way more than twice as often as the white Atolla. Shining the thin beam on any of the other body parts did not seem to result in any consistent reactions. Full Flashlight When Atolla were flashed with bright white light for a few seconds, they would often perform the light reflex; either turn belly¬ up or dart away from the light source and then turn belly-up (Figure 8). This response was recorded so often and was so predictable that Atolla could be "steered" to any specific area in the kriesel by "chasing" it with the beam. Moving Shadows When Atolla was subjected to shadows created by white light, it nearly always responded by performing the "light reflex" (Figure 9). Sudden "Lights Out' Many of the times that Atolla did not react to the flashlight shining on them, they would react when the white flashlight was turned off again. The animals would portray the characteristic "light reflex," when the light turned off rather than when it was turned on (Figure 10). Three out of the thirteen stimulations resulted in Atolla bioluminescing around its margin. Manipulations Once Animal Is Belly-up If Atolla was already in the belly-up position it would not alter its behavior unless stimulated to an unusually high degree. Stirring the water in the kriesel was almost invariably followed by more active swimming unless Atolla was already belly-up before the disturbance began. Sometimes the jellyfish would turn belly-up in response to stirred waters. Blowing bubbles through thin tubing also resulted in more active swimming and pumping by Atolla, but was never associated with the belly-up position. Touching the animal often produced no noticeable response, unless the rod was touching the area at the base of the tentacles by the bell margin. In this case, three out of three times the jellyfish would pump away quickly and turn belly-up. Furthermore, using the rod to move Atolla through the water ended with the jellyfish turning belly-up. None of these mechanical experiments resulted in bioluminesence. Recovery back to active, up-right swimming took a variable amount of time and required an end to all extra stimuli. This recovery occurred more rapidly for Atolla in good condition than for Atolla in poor condition. In fact, once these jellyfish have been in captivity for five or six days and are beginning to appear unhealthy, they tend to remain in the belly-up position more often than the up-right position. Feeding Atolla was never seen definitely using its tentacles to feed. Two adult Artemia did drift into the mouth of one Atolla who proceeded to swallow them. Additionally, at least three different Atolla were seen putting one to four tentacles in their mouths when larval Artemia were also in the kriesel. It could not be determined, however, whether there were any larval Artemia on the tentacles being used. Atolla were never seen placing their tentacles in their mouths when larval Artemia were not in the kriesel. Other than these observations, the jellyfish did not react noticeably to contact with the other items offered. Scanning Electron Microscopy Differences could be discerned between the trailer tentacle and the short tentacles when viewed under 400 to 10,000 times magnification. A fine connective net tissue covered both types of tentacle. Nematocysts were intertwined within that tissue, one of which was found to have discharged (Figure 11). The connective tissue and nematocysts on the trailer tentacle were finer and smaller than the short tentacles, requiring more magnification to discern the same amount of detail Nematocyst Isolation Both discharged and non-discharged nematocysts were examined after isolation. Nematocysts averaging 10.5 to 11.0 microns diameter occurred in the long tentacle (Figure 12). In contrast, nematocysts from the short tentacle were in much higher concentration and consisted of both the 10.5 to 11.0 micron size range and the 23 micron size range (Figure 13). They can be identified as anacrophores and holotrichous isorhizas, respectively (Purcell 1984). DISCUSSION These results demonstrate that Atolla has a stereotypical response to both light and mechanical stimulation, namely pulsing quickly a number of times and/or turning belly-up. Possible reasons for behaving this way may be associated with interactions between other deep-sea bioluminescing animals. For instance, pushing a non-bioluminescing Beroe into the tentacles of Atolla produced no response in Atolla three out of three times. However, on the fourth manipulation, Beroe bioluminesced as it contacted Atolla, and Atolla responded very vigorously with the entire series of the characteristic light reflex, which included bioluminescing. It is possible that mechanical manipulation of Beroe induced its bioluminescent response as an alarm reflex, which in turn caused Atolla to also perform an alarm reflex. Thus, the bioluminescence of one creature may serve as a signal or warning to other light- sensitive creatures that danger may be near. Further studies in this area are needed before any solid conclusions can be drawn. Results from the thin beam experiment show that the bell¬ margin region and the tentacles were the most sensitive to light. This correlates well with Arkett and Spencer's study on hydromedusae which suggests that the ocelli, structures located near the bell margin at the base of each tentacle, are the primary photoreceptors (Arkett and Spencer 1986). The results from the sudden "lights off" experiments demonstrate that Atolla responds predictably by quickly pulsing a number of times and/or turning belly-up. Whenever there was a reaction, more active pumping was involved; however, only a small percent of the reactions involved the belly-up position. This response to rapidly diminishing light intensity suggests that deep¬ sea scyphomedusae, such as Atolla, share a similar reflex with many shallow water hydromedusae, such as Polyorchis penicillatus , which have a predictable "off-response" (Singla 1974). This "off-response consists of an initial swimming contraction, followed by a series of rapid successive tentacle contractions, finishing with several more swimming contractions (Arkett and Spencer 1986). Such behavior in shallow water medusae can be interpretted as a shadow reflex. These medusae are normally exposed to sunlight but when an animal swims above them a shadow is created. This type of shadow could be caused by approaching predators and may be why the shallow water jellyfish respond to rapidly diminishing light intensity. However, this behavior is not so easily explained for deep-sea medusae. Generally, only minimal amounts of light penetrate to 500 meters. Still, that may be enough light for the ocelli of Atolla to detect. In that case, Atolla would be sensitive to very subtle changes in light intensity. Possibly, the reasons for an off-response in deep- sea jellyfish is based on different principals than the shallow water jellyfish. Atolla may be reacting to alarm signals given by other bioluminescent animals. Further research on this topic could lead to a better understanding of the off-response in deep-sea medusae. All these experiments combined, reveal that once in the belly up position, Atolla rarely reacts with any further stimulus. This suggests that the belly-up position may be a stress or alarm response. By tucking in its tentacles and pulsing very infrequently, Atolla avoids disturbing the surrounding waters, thus supplying near-by creatures with fewer clues of its location. Additionally, dead animals and debris are constantly drifting down through the water column. Atolla may appear more like a dead animal or sinking debris when assuming the belly-up position. This could deter possible predators from choosing to eat Atolla. The condition of the animal has been shown to be important in that unhealthy animals or animals stressed by remaining in captivity too long respond less often and less predictably than animals in good condition. Because red Atolla respond more often and more predictably to stimuli than the white Atolla, there may be a correlation between the color of the jellyfish and its health. Atolla which were red upon capture were noted to gradually lose their pigment while in captivity. Further research in this area would be very helpful. Closer studies of the tentacles and of isolated nematocysts with SEM and light microscopy revealed differences in the nematocyst populations of the trailing tentacle as compared to the short tentacles. Further research distinguishing differences between the two tentacles and the significance of the trailer are much needed. Though special care was taken to minimize stress to these animals and to simulate a deep-sea environment as well as possible methods could still be improved. When observing and collecting Atolla with an ROV, red lights should be tried. Their sensitivity to white light and their fragility suggest the importance of transfering jellyfish into their kriesel with as little handling and disturbance as possible. ACKNOWLEDGMENTS First and foremost I must thank George Matsumoto for taking me under his wing: spending hours of his time guiding me in lab techniques, loaning source material, and advising me on my presentation. Warm thanks to Annie Reese, my jelly-twin, who shivered by me as we observed our animals in the coldroom and kept me up when Raison and Pickle were getting me down. My advisors Chuck Baxter and Bruce Robison, have been a tremendous support and source of inspiration. Hey, Robie, what more must I do to go down in Deep Rover? I also need to thank Kim Reisenbichler for putting up with us in the coldroom he designed. Thanks to Chris Patton and Tom Schroeder for help with SEM. Gil VanDykheusen and Fraya Sommer, I appreciate your gelatinous gifts. Thanks to Lynn Lewis and the entire staff at MBARI. Additional thanks to Alan Baldridge, Mark Denny and the entire Hopkins staff. This project could not have happened without the help and humor of Brian Ackerman, Dave Bracher, Jon Consiglio, Craig "1.C. Darve, Steve Etchemendy, Chris Grech, Roger Hayes, Greg Maudlin, Jim McFarlane, Mark VandenBerg and Bill Wardle. Moral support from my housemates (all six of them) was fuel for this project. LITERATURE CITED Arkett SA, Spencer AN (1986) Neuronal mechanisms of a hydromedusan shadow reflex. I. Identified reflex components and sequence events. J Comp Physiol A 159: 201-213 Etchemendy S, Davis D (1991) Designing an ROV for oceanographic research. Monterey Bay Aquarium Research Institute converted an oil-field ROV into a scientifically usable tool. Sea Technology: 21-24 Herring PJ, (1990) Bioluminescent Responses of Atolla. Mar Biol 106 (3): 413-417 Purcell JE, (1984) The functions of nematocysts in prey capture by epipelagic siphonophores (Coelenterata, Hydrozoa). Biol Bull 166: 310-327 Singla CL (1974) Ocelli of hydromedusae. Cell Tissue Res 149: 413- 429 Weber (1987) Some physical and chemical properties of purified nematocysts of Hydra attenuata Pall. (Hydrozoa, Cnidaria). Comp Biochem and Physiol 88B: 855-862 THE BELLYUP POSTTION FIGURE1 I E O OOON QC O L S I 55 0 5 5 0 -. 55 8 a 8 58 52 V THE BELLYUP POSTTION AFTER SOME TIME FIGURE 4 + E C I 5 . 5c a 5 a8 5 o- I C O O 0 2 a S C I + E 3 88 D 0 g o a 2 L — I C 2 o- 0 O C — + I C 2. 5 5 a 9 25 55 0 - C I C 5- DO 95 2 0 0 0 — —: 32 35 d O - — - NORO 9g N O 3 o 8 3 2o 9 o o — 5 NNNNNC ONOOOROOONRoo — — Q — S