Abstract The brittle star Ophiopteris papillosa exhibits an escape response to the predatory asteroid Pycnopodia. Stimulation of Ophiopteris with a tube foot of Pycnopodia evokes action potentials in the radial nerve chord (R.N.C,), Furthermore, recordings of these action potentials are similar both in velocity and amplitude to activity evoked by electrical stimuli of the R.N.C. itself. Also, the axons mediating these action potentials appear to be independant of external Ca' ions and to run uninterupted by chemical synapses for at least several millimeters, These axons are probably transmitting sensory information, as tube foot stimulus is a sensory stimulation. In addition, the velocity of action potentials in these axons are about 17 times faster than propogation of the apparent motor output. Introduction The ophiuroid Ophiopteris papillosa is a natural for neuroethological studies. Ophiopteris is an extremely mobile ophiuroid which exhibits a variety of interesting behavior. The R.N.C. is fairly easy to dissect and animals are easily maintained, The greatest problem encountered in the study of ophiuroid nervous systems is the size of the nervous units. Brehm(1977), and Cobb & Stubbs (1981) have found axons up to 10 - 20 u in diameter. These are much larg¬ er than most echinoderm axons, yet they are still quite small, Therefore, extracellular suction electrodes have been used to record neryous actiy¬ ity, and analysis of activity in single neurons has not yet been possible Work by Tuft & Gilly (1984) and Yee (1984) has shown that some of these "giant" axons are Ca independant, rapidly conducting axons, Tuft & Gilly refered to these axons as "class 1" axons, and showed them to have a conduction velocity of 1.40'm/sec at 19°c. Yee found a velocity of 0,5 m/sec at 14°c. This paper concerns the general integratory mechanisms under¬ lying an escape response in Ophiopteris. This escape response is exhib¬ ited when the ophiuroid is stimulated by the touch of Pycnopodia. Methods Extracellular recording methods were employed for the neurophysio- logical part of this study. This was done with glass suction electrodes drawn to a tip diameter of about 0,2 mm. These glass electrode tips were fitted to the end of a syringe. One of the Ag:AgCl electrode wires was placed inside the syringe, the other was wrapped around the outside of the glass electrode tip. Differential recordings were made with a Textronix type F 122 preamplifier with hi/low pass filters set at 8 Hz, 1 KHz, The tube foot stimulator consisted of a tube foot placed at the end of a bar which was attached to the movable breaker of a D.C, relay, At the click of a switch, a sweep on the ocilloscope was triggered, and the relay was simultaneously activated to pull down the bar upon which a tube foot hadobèen placed, The tube foot would then (in the ideal situation) strike the tip of the leg which was being recorded from, A refrigeration unit ranpbèneath the preparation to maintain a temperature of 13 - 15°0. This helped keep the preparation fresh, and standardized temperatures, Ca free experiments used the same set up; however,the portion of the leg which was being recorded from was placed in an isolation chamber. The isolation chamber consisted simply of a vial cap, and plasticine clay which made a relatively water tight seal around the leg. (See fig. 1.) The animal was placed upside down and restrained with pins that went around the leg and into a substrate of Sylgard gel. The ophiuroid was then dissected simply by removal of segmental arm plates on the oral surface. Behavioral observations were made by stimulating the tip of the leg with a Pycnopodia tube foot. The Ophiopteris in question was usually in a horizontal position in a glass or plastic container, a natural habitat, Ophiopterus is frequently found in a crevice in a vertical position, with only the leg tips protruding out. To study the time course of the "leg jerk" response, the animal was videotaped as it was being stimulated by a Pycnopodia tube foot, The results were then drawn onto an acetate sheet, directly off the video screen. (See fig, 3.) The time course was followèd in a frame by frame analysis of the videotape, The velocity of the behavior was followed by comparing each drawing to the initial drawing, and map¬ ping out the evident displacement of the leg. The delay time to first observable movement was calculated by averaging 20 experiments. To test the difference between tactile and tube foot stimulation, the animal was first placed in the middle of a rectangular container, Following stimulation, the ophiuroid would locomote to the side of the dish, The time taken to locomote to the side of the dish was recorded for 10 trials using a metal probe as a stimulus, and for 10 trials using a Pycnopodia tube foot as a stimulus. Similar trials were made using tube feet from other starfish (Patiria miniata and Pisaster ochraceous), but these did not result in escape responses. Results Ophiopteris shows a response to Pycnopodia tube feet on virtually every body surface. The leg tip is the most sensitive spot, When the tip of the leg is stimulated with a tube foot, the ophiopteris will almost always exhibit a "leg jerk" response. (fig. 2. leg jerk) This response does not appear to habituate, even after more than 100 stimulations, given about ten seconds between stimulations. A single leg severed from the body still exhibits the "leg jerk" response, The "leg jerk" is propogated down the leg in the form of a wave of behaviour, (fig. 3.) This behavior pulls the leg tip away from the stimulus quite rapidly, Figure 3 shows the change in position of the leg relative to the initial leg position following tube foot stimulation. The behavior is followed through 1630 msec. Twenty such responses were videotaped and an attempt was made to discover the average time before an initial response was detected, there was a good range of uncertainty as to when the leg was stimulated, and as to when the first movement occured, The-average response time was 242 msec, - 94 msec (S.D., N=20), Figure 4 shows the velocity at which the leg jerk response in fig 8, propogates down the leg. The behavioral propogation is fairly con¬ stant at about 0.059 m/sec for the first 3.5 cm. It then decreases greatly. Following the "leg jerk", the ophiuroid will exhibit a fairly stereotyped pattern of whole body responses, In this, the two legs opposite to the one stimulated pull back, dragging the whole animal away from the site of stimulation. (fig. 2, locomotion) These two legs are reextended and may pull back for one or two more cycles. At this point the animal will use a great variety of forms of locomotion, seemingly with no clear preferance. In experiments comparing escapes from a tactile stimulus (probe) to escapes from tube foot, the average time spent in locomotion follow¬ ing stimulation was 52 seconds for tactile vs, 31.7 seconds for tube foot. The escape speed from tube foot stimulus was 1,6 times faster than it was for tactile stimuli. At times the Ophiopteris will feed on the surface film of the water, with all tube feet on the feeding leg extended, Stimulation of a single tube foot with a Pycnopodia tube foot results in contraction of the four tube feet adjacent to the one stimulated, and the five tube feet on the opposite side of the leg. The rest of the tube feet remain extended, Poking a single tube foot with a metal probe resulted in only that tube foot contracting, This shows another difference between tactile and tube foot stimul:. Slow R.N.C. Electrical Activity Due Lo Chemical Stimulation Figure 6 shows the results from an experiment in which Pycnopodia homogenate was dropped onto the tip of a leg which was being recorded from. This was at time 0, From the beginning of the trace to time 0, very little spontaneous activity is evident. There is a brief period of noise lasting about 1 second which ocurred during stimulation. Fol¬ lowing stimulation, there is a general increase in nervous activity which becomes most dramatic at 5 - 7 seconds, The activity then tapers off until the second stimulation, After this stimulus there is, once again, a brief period of noise, immidiatly followed by intense nervous activity. Following both stimulations, behavior, in the form of violent thrashing motions of the legs occured. The cause of the delay in activ¬ ity in the second case is unclear and will be discussed later, Fast Activity In The R.N.C. Due To Chemo/Tactile Vs. Electrical Stimu¬ lation Stimulation of Ophiopteris leg by a tube foot or a stainless steel needle ("chemo" vs "tactile" stimulation) results in electrical activity very similar to that seen in response to brief electrical shocks applied directly to the radial nerve chord. In the experiment illustrated in fig. 5, electrical activity was recorded at a single side 21 mm proximal to the site of electrical stimulation, and about 29 mm proximal to the site of tube foot stimulation, near the tip of the leg. Three record¬ ings of each type of stimuli were obtained. The conduction velocities are as follows (electrical vs, tube foot in each case): 0.43 m/sec vs. 0.36 m/sec (fig, 54), 0.43 m/sec vs. 0.52 m/sec (fig. 5B), and 0.41 m/sec vs. 0.63 m/sec (fig. 50). The mean value for electrical stimulation is 0.42 m/sec; that for tube foot stimulation is 0,50 m/sec, As one can see, the variability in conduction velocity is much larger for the tube foot stimulus than for electrical shocks, However, on the average, itis very fast, and nearly the same as for electrical stimulas tion. The most rapidly propagating activity due to electrical stimula¬ tion of Ophiopteris nerve has been reported to be about 1.40 m/sec at 19°0 (Tuft & Gilly, 1984) and 0,5 m/sec at 14°C. These authors identify this activity with large class 1 axons. The conduction velocity given above for tube foot (0.5 m/sec) and electrical (0.42 m/sec) stimulation at 13 - 15°0,suggest that class 1 axons underly activity recorded in my experiments due to both modes of stimulation, East R.N.C. Electrical Activity Due To Needle Vs, Tube Foot Stimulation Yigure 7 shows results of two experiments testing the difference between fast nervous activity generated by a purely tactile (syringe needle) stimulus and by a stimulus using Pycnopodia tube foot. In figure 7A the top trace shows the result from a stimulus using a syringe needle, The next trace shows a control experiment in which the tube foot stim¬ ulator was set to hit the water close to,but not on the leg of the Ophiopteris. The third trace shows the result of tube foot stimulation; the recorded activity is similar to that seen with needle stimulation, From the bottom experiment, (fig. 7B) one finds the conduction velocity of the tube foot stimulation to be 0.41 m/sec. The conduction velocity of both needle stimuli is 0.33 m/sec. These conduction velocities were calculated using a 17 mm distance from stimulation to recording site. This could have varied by as much as + 3 mm due to difficulty in stim¬ ulating the leg tip in exactly the same place from trial to trial. Although the conduction velocity for the tactile stimulation seems to be slower than that for tube foot stimulation, this must be regarded as tentative. Velocities above were calculated for the traces in figure 7B only, The distance from stimulus to recording site was not measured for the experiment in figure 7A, making calculationsof velocities impossible. However, by inspection, one can see that tactile and tube foot stimuli evoke action potentials with nearly the same velocity in this case. Cat Free Experiments - Persistance of Fast Electrical Activity Following Tube Foot Stimulation In light of the possible role of involvement of the "giant" class 1 axons in tube foot stimulation results, several experiments were car¬ ried out using a Car free medium. Activity of class 1 axons is not abolished by Ca" removal, whereas that of slower axons, and chemical synapsis is. (Brehm,1977; Tuft & Gilly, 1984; Yee,1984) My first Ca' free experiment was a behavioral one. One of the legs of a large Ophiuroid was placed in a dish of Car free sea water, The rest of the animal was in a seperate dish of normal sea water. Once the anesthetized leg had stopped spontaneous movement, It was stimulated with a tube foot, There was no reaction. The other legs, when individ¬ ually stimulated, caused the whole animal (except the anesthetized leg) to react by moving violently. Next, I set up an isolation chamber experiment for electrical re¬ cording as pictured in figure 1, The dissection had two ganglia exposed distal to the recording site, and the isolation chamber was filled with Ca“t free sea water. In figure 84, traces 1 - 3 show recordings of electrical activity following a tube foot stimulation taken prior to application of Catt free sea water. Trace 4 is a control. Traces 5 - 9 show stimulated recordings made at the times indicated following the change to Ca free sea water. Traces 10 and 11 were taken after the isolation chamber had been refilled with normal sea water, It is un¬ clear why the activity in traces 6 - 8 is reduced, but activity is clearly present in 9. These results show that the very rapid trans¬ mission of information down the ophiuroid R.N.C. following chemo/ tactile stimulation can persist for a distance of at least two ganglia, and probably much farther. This finding is consistent with the proposed involvement of class 1 axons. Conclusions 1. There is an escape response in Ophiopteris papillosa which results from stimulation by the tube feet of Pycnopodia. 2. The escape resonse does not habituate, 3. The first visible response by Ophiopteris occurs after about 240 msec. 4. An initial wave of behavior travels down the leg at a velocity of about 0.059 m/sec following stimulation with Pycnopodia tube feet. 5. The "leg jerk" does not require integration in the circumoral nerve, 6. An escape ersponse can be evoked by purely chemical stimulation, a weaker response is evoked through tactile stimulation. 7, Action potentials are evoked by both tactile and chemosensory components. 8. The nervous units conducting these action potentials appear to be the same as the class 1 axons studied by Tuft & Gilly (1984) and Yee (1984). That is, they are the most rapidly conducting axons and are probably uninterupted by chemical synapses for distances of several centimeters, Discussion Similarities between action potentials generated ty tube foot stim¬ ulation, and electrical stimulation indicate that similar, if not the same, axons are being used to rapidly transmit information down the R.N.C, in both cases. These axons are the fastest, and probably the largest of Ophiopteris axons. Work done by Yee (1984) using electrical stimulation indicates that these axons may run on the dorsal edge of the ectoneural tissue of the nerve chord, and probably transmit sensory data. My studies are in agreement with this, as I was definitely stimulating activity via sensory pathways, and recording from the ectoneural portion of the nerve chord. Therefore, giant class 1 axons mediate the rapid conduction of sensory information, and are sensory interneurones. Other data concerning the similarity of conduction pathways for sensory and electrical stimuli involve the Ca free experiments. Ca free media did not block transmission of sensory data, or fast electrical activity (Yee, 1984; Tuft & Gilly, 1984) At this point it is unclear whether or not Ca+ is required to get sensory data into these hypothesised sensory interneurones; however, behavioral data indicate that it is. In other words, chemical synapses may be involved in the transmission of sensory cell data to the sensory interneurone. This hypothesis may add specialized sensory cells to the list of fine structures in echinoderms. According to Pentreath and Cobb (1972), The evidence so far available confirms the hypothesis that epi¬ thelial cells are in part sensory in function, only rarely showing even the simplest specialization, and that they are receptive to most stimuli (i.e. mechanical, chemical, light, etc.). I believe a chemosensitive cell specialized to react to some chemical component of Pycnopodia must exist, These cells may be receptive to other sensory input as well. In particular they may also be mechanoreceptors. Other starfish including Pisaster ochraceas, and Patira miniata were not able to elicit an escape response. A major question which now exists is why bother to have rapidly transmitting axons when behavior is propogated down the arm at a rate : nearly twenty times slower? Also, how is this slow propogation ac¬ complished? The most logical theories seem to me to be ones involving either chained reflexes or sensory feedback - böth being quite unusual for an escape response. The mechanics of leg movement may require such unusual pathways. The need for rapid conducting axons maybe to synchronize the other arms or to ready the animal for an escape. The differences in delay times preceeding increased nervous activity, following a chemical stimulation (fig. 2.) may be accounted for by hypothesizing that the first stimulation "primed" the animal for the second one, Finally, work by Smith (1950) and Kerkut (1954) has shown that "pacemaker" centers within the nervous system are important in the coordination of behavioral responses. These studies also suggested that there are discrete centers which control locomotorygbehavior, (cited by Pentreath & Cobb, 1972) With more work and newer techniques, I think some of these "discrete centers" may be discovered and in particular, a center controlling the coordination of escape locomotion may be found. Indeed, one element of this "escape center" would appear to be the class 1 axons extending down the nerve chord. 11 Acknowledgements T would like to thank Gilly for spending so much of his time expertise, and sanity helping me with my project. T would also like to thank Freya for doing a wonderful job of collecting giant brittle stars. References Brehm, P. (1977). Electrophysiology and luminescence of an ophiuroid radial nerve. J. exp. Biol. 71, 213-227. Kerkut, G. A. (1954) The mechanism of coordination of the starfish tube feet. Behaviour 6, 206-32. Pentreath, V.W. & Cobb J. L. S. (1972) Neuro jology of echino¬ dermata. Biol Rev. 47, 363-392. Smith, J. E. (1950). Some observations on the nervous mechanisms underlying the behaviour of starfishes, Symp. Soc. exp. Biol, 4, 196- 220. Tuft, P. J, & Gilly, M. F. (1984). Ionic basis of action potential propogation along two classes of "giant" axons in the ophiuroid, Ophiop-- teris papillosa, J. exp. Biol, In press. Yee A. (1984). Unpublished report, Biology 175H, Hopkins Marine Station of Stanford University, Mapping of electrical activity along the radial nerve cord of Ophiopteris Papillosa. Figure legend Figure 1. Experimental set up for recording from isolated portions of the ophiuroid leg which can placed in Ca+ free sea water, Figure 2. Initial escape response following stimulation by tube foot. Figure 3. Leg jerk behavior Figure 4. Behavioral latency of leg jerk, The line on the graph was fit by eye. Figure 5. Tube foot vs, electrical stimulation, The experiment was done at 13 - 15°C, the electrical stimulus was a 30V 0.5 msec shock, The electrical traces have a 7amsec delay, while the t 'foot traces have an 8 msec delay imposed nearly entirely by inertia of the relay, Figure 6. Chemical stimulation using Pycnopodia homogenate. Figure 7. Needle va, Tube foot stimulation, Two experiments are represented. In each case, activity was recorded from a single site, Temperature was 13 - 15°c. Figure 8. Ca+t free results. The Ca+ free medium had a ph of 7.78, and osmolarity of 981. 0 32 2 O L — — J O LI 0 2 O 0 L O L O J 0 L 8 2 39 " ---. 1 . 40 TEMPORAL PAT RESPONSE ESCAPE 3 360 ERNO EG JERK) 280 6 msec 4 480 630 960 —— 8. Oo GOOO O — TUBE FOOT ELECTRICAL TUBE FOOT ELECTRICAL ELECTRICAL TUBE FOOT 100 uvolt 100 uvolt 50 msec T i 100 msec 100 uvolt needle control tube foot OO msec tube foot needle needle control STIMULUS COMPARISON NEEDLE VS TUBE FOOT 0 P 30 58 — o —- - --