Brain Research, 163 (1979) 207-222 © Elsevier/North-Holland BiomedicalPress
207
L9 M O D U L A T I O N OF LT'S ELICITED G I L L W I T H D R A W A L RESPONSE IN APLYSIA
KEN LUKOWIAK* Department of Physiology, McGill University, Montreal, Quebec, H3G 1 Y6 (Canada)
(Accepted June 22nd, 1978)
SUMMARY The effects of induced L9 activity on gill motor neuron LT's ability to elicit a gill withdrawal response were studied. It was found that L9 was a modulator of LT's effectiveness to elicit gill movements. Following L9 activity, L7's ability to elicit a gill withdrawal response was significantly potentiated by up to 240~o of control. L9 activity potentiated L7's elicited gill withdrawal response when frequencies of L7 activity were used which would result in decrement of the response or when frequencies were used in which decrement does not occur. Induced L9 activity may also have a minor potentiating effect on LDGt's ability to elicit gill movements. L9 was the only neuron found to possess these modulatory abilities. The interposition of activity in the other gill motor neurons failed to have any effect. Lg's modulatory role is separate and independent from its role as a gill motor neuron. L9's modulation of L7 is mediated peripherally in the gill and not in the CNS.
INTRODUCTION Recent studies have shown that the central (CNS) and peripheral (PNS) nervous systems interact and are parts of an integrated system which normally mediates adaptive behaviour of the gill withdrawal reflex in Aplysia 10. During the course of that study, it was found that when a central gill motor neuron, L7, was depolarized to produce 6-14 action potentials (AP's) per sec once every 30 sec its elicited gill withdrawal response habituated and that the interposition of a tactile stimulus to the gill or siphon resulted in dishabituation or facilitation of LT's ability to elicit a gill movement. It was further shown that the facilitatory effect of the tactile stimulus on LT's efficacy to elicit gill movements occurred peripherally in the gill and not centrally * Present address: Faculty of Medicine, Division of PhysiologicalSciences, University of Calgary., Calgary, Alberta, Canada, T2N 1N4.
208 in the abdominal ganglion. This and a subsequent study 8 thus showed that the efficacy of Lv's peripheral terminations could be altered by both its own activity and activity evoked in another pathway. Lukowiak and Peretz further examined whether the interposition of activity in the other major gill motor neurons LDG1 or LDG2 could also effect the efficacy of L7's peripheral terminations. They found that the interposition of LDG1 or LDG2 activity had little if any effect on L7's ability to elicit a gill withdrawal response. Swann, et al. 14 found, however, that they could potentiate L7's ability to elicit a gill withdrawal response by perfusing dopamine through the gill. In addition, they presented some evidence that Lg, a minor gill motor neuron, may be dopaminergic. It thus seemed logical, in light of the Swann et al. study and the previous effort of Lukowiak and Peretz, to determine if the interposition of induced activity in L9 results in potentiation of L7's ability to elicit a gill withdrawal response. If potentiation of L7's ability to elicit a gill withdrawal response were observed following the interposition of L9 activity, it would show that a central gill motor neuron could also act to modulate the effectiveness of another gill motor neuron to elicit gill movements. I now report that the interposition of L9 activity significantly potentiated the ability of L7 to elicit a gill withdrawal response, both at low frequencies (6-12 AP's/sec) of stimulation where decrement is normally observed with repeated L7 activation and at higher frequencies (15-20 AP's/sec) where decrement is not observed s. In addition, L9 may also have a potentiating effect on LDG1 ability to elicit gill movements but this effect is minor. These data then, along with the results obtained in a number of different preparations indicate that the modulation of the effectiveness of a neuron by the activity of another neuron is a feature of the neurons system which is both widespread and important across phyla 2,~1,1~,a~. METHODS
Aplysia californica, obtained from Pacific Biomarine Laboratories (Venice, Calif.) and weighing between 150 and 250 g were used. The animals were kept before use in artificial seawater (Instant Ocean) at 15-16 °C and pH 7.9. The preparation used in these studies was similar to that described previously 1°. It consisted of the siphon, mantle, gill and the abdominal ganglion. The branchial, ctenidial and siphon nerves by which the abdominal ganglion innervates the gill and siphon were left intact. Particular attention was paid to a small nerve branch that leaves the siphon nerve and innervates the gill. It has been shown that it is this small branch of the siphon nerve which carries LCs axon to the gill 7. As a control, in some preparations this small branch of the siphon nerve will be severed after L9 modulatory effect has been demonstrated in order to show that Lg's effect occurs in the periphery and not in the abdominal ganglion. All preparations were maintained at l 5-16 °C in a 1000 ml chamber of artificial seawater during the course of the experiment. The preparations remained viable for at least 24 h as evidenced by the continued observance of the periodic spontaneous gill respiratory movements. Twenty preparations were used in this present study and data from all 20 have been used and statistically analyzed.
209 The amplitude of the gill withdrawal response was measured using a force transducer (Grass Ft .03) connected to a single gill pinnule by fine surgical thread and recorded on an oscilloscope and a polygraph from which the measurements were made. Recording gill movements in this manner did not result in any observable damage to the gill and gill movements could continue to be elicited 24 h after dissection. Neuron identification was based on a number of criteria and is consistent with that of Koester and Kandel 5. For L7 and LDG1 the criteria included: location, size, synaptic activity during spontaneous gill respiratory movements, the appearance of one-for-one activity recorded in the gill pinnule by an extracellular suction electrode (referred to as a pinnule potential) and the type of gill movement elicited by depolarization of the neuron. Similar criteria were also used to identify L0. L0 receives similar synaptic input as Lv during periodic spontaneous gill contractions and causes a similar type of gill movement as L7 when it's depolarized. However, L0 must be depolarized to produce a higher rate of AP's for a longer duration than L7 to produce a similar amplitude contraction. Moreover, pinnule potentials were not observed which correlated with L9 activity and L9 innervates the gill via a small branch of the siphon nerve and not via the branchial or ctenidial nerves. In this regard L9 is unique. No distinction has been made between the two L0's, both gave similar results. Normally the larger of the two (when there were two) was recorded from. Micropipettes filled with 3 M KC1 and having a resistance between 15 and 30 M ~ were used. A bridge circuit in the electrometer (Getting M-5) allowed simultaneous recording and stimulation. When identified central motor neurons were depolarized, constant current (1000 msec duration) was passed into the neuron through the recording electrode. The rate of AP's on each trial was held constant and for L7 two frequencies were used: 6-12 AP's/sec and 15-20 AP's/sec. To determine if induced activity in L9 effected the gill withdrawal response elicited by L7 activity, the following experimental protocols were followed. L7 was depolarized to produce a fixed number of AP's (6-12) for one sec; L0 was then depolarized to produce 5-15 AP's/sec for 4-6 sec and then L7 was depolarized again to produce the same number of AP's as on the first trial. The interval between L7 depolarization was 30 sec. This series of LT, Lg, L7 stimulation constitute a test. Preparations were always rested at least 10 min between tests. Data were normalized, 100 ~ being the amplitude of the initial gill withdrawal response in the individual test. When pooled data are presented the mean 4- S.E.M. are given. Similar tests were also performed when L7 was depolarized to produce 15-20 AP's/sec as well as when LDG1 was depolarized to produce 6-10 AP's/sec on each trial. In one set of control experiments the interposition of L9 activity was omitted; while in another set LDGa activity was interposed between L7 stimulations or L7 activity was interposed between LDG1 stimulations. L7 or LDG1 activity was also interposed between L9 stimulation. Finally, in some preparations after L9's potentiating effect was demonstrated, the small branch of the siphon nerve by which L9 innervates the gill was severed and following a 1 h rest the experiment was repeated in order to determine the locus of La's effect.
210
A
B
250
+ ~200
;[ O (n '" r~
1SO
rt uJ )-
_~
!oo
I--
50
L71
I'1 L72
I-71
L72
Fig. 1. Pooled normalized data showing the effect of the interposition of L9 activity on the ability of L7 to elicit a gill withdrawal response. A: L7 was initially depolarized to produce between 6-14 AP's per sec (L71) and this response was taken as 100~. Thirty sec later L7 was again depolarized to produce the same number of AP's as on the initial trial (L72). As can be seen, the amplitude of the contraction (mean ± S.E.M.) produced on the second L7 trial (L72) was reduced by 35 ~ . These data are from 14 preparations with 45 separate tests. B : the data are from the same 14 preparations (n = 92, individual tests). L7 was initially depolarized to produce 6-14 AP's per sec (L71) and again this response was taken as 100 ~ . L9 activity was then interposed. Following the interposition of L9 activity and 30 sec after the initial depolarization, L7 was again depolarized to produce the same number of AP's as on the initial trial. As can be seen (L72) , the amplitude of the response following L9 activity was much larger than the initial response. It was found that the amplitude of the response following L9 activity was significantly larger than the initial response (P < 0.001) and the expected (P < 0.001) had not L,3 activity been interposed (L72 in A).
Fig. 2. Potentiation of L7's elicited gill withdrawal response by induced L9 activity. A: L7 was depolarized to produce 6 AP's for 1 sec and this produced a gill withdrawal response of 8 mm. Fourteen seconds later, L9 was depolarized to produce 12 AP's/sec for approximately 4 sec and this produced a gill withdrawal response of about 12 mm. Thirty seconds after the initial depolarization of LT, L7 was again stimulated to produce 6 AP's for 1 sec and this resulted in a potentiated response of 15 ram. B: in another preparation, as in A, only the interstimulus interval (ISI) was shortened to 17 sec. The initial L7 response (6 AP's) was 13 ram, the response elicited by L9 depolarization (5 AP's/sec for 5 sec) was 5 mm and the response elicited by L7 stimulation (6 AP's) following L9 stimulation was approximately 27 mm. C: an oscilloscope recording taken from a third preparation which again shows the potentiation of L7's gill withdrawal response following L9 activity. GT1 was the initial gill withdrawal response elicited by L7 stimulation (6 AP's) while GT2 was the gill withdrawal response elicited by L7 stimulation (6 AP's) following the interposition of L9 activity (not shown, 8 AP's/sec for 4 sec. The ISI for L7 stimulation was 30 sec and L9 was depolarized 18 sec following the initial L7 depolarization. The base line for GT~ was slightly adjusted upward so as to make a clearer presentation. Scales: A, B : L7 -- 40 mV, L• 30 mV, 1 sec: 7 mm = 100 mg, C: 20 mV, 100 msec.
211
|
15
6
13
2?
L?
L L9
C
m
212 RESULTS Previously it had been shown that when Lv was depolarized to produce between 6-14 AP's per trial, the amplitude of the elicited gill withdrawal response decremented with repeated activation (I/30 sec) 8,10. In the 20 preparations tested in this series of experiments, similar results were obtained and these data are shown in Fig. 1A. On the average, the response amplitude on the 2rid L7 trial (L72) was reduced by 35'},~, compared to the initial response. In these same 20 preparations, when L9 activity was interposed between the two Lv trials, the response amplitude on the 2nd L7 trial (L72) was not reduced but rather it was potentiated by an average of 127 ?/,, (Fig. I B). At least 3 separate experimental tests were performed in each of the 20 preparations and the results of 92 individual tests are represented in the bar graph in Fig. lB. A t-test was performed to statistically determine whether the interposition of L9 activity significantly effected the effectiveness of L7 to elicit gill movements. It was found that there was a significant difference (P 0.2). Thus L9 m a y have a minor m o d u l a t o r y effect on L D G I ' S ability to elicit a gill withdrawal response. L9 could be exerting its m o d u l a t o r y action on L7 either in the abdominal ganglion or out in the periphery, in the gill itself. Because it had previously been shown that infusion o f dopamine t h r o u g h the gill potentiated Lv's elicited gill withdrawal response, it was t h o u g h t that Lg's action might also occur peripherally. To test this
218
A
T
1S
11
irr
L_
LI Fig. 7. A: Induced activity in L9 does not have any observable effect on activity in L7. L9 was depolarized to produce 8 AP's/sec for 3 sec and as can be seen in the higher gain recording from L 7 , this did not result in any observable synaptic input to L7 nor did it affect the spontaneous ongoing synaptic activity in L7. Scale: L9 = 50 mV; L7 ~ 5 m V , 1000 msec. B, C: Lg's potentiating effect only occurs with the siphon nerve intact. In B, induced activity in L9 potentiates the gill withdrawal response elicited by L7 stimulation. However, following cutting of the siphon nerve through which L~ innervates the gill, the interposition of L9 activity does not result in potentiation of LT's response (B, C). This has been a consistent finding. Scale: L7 ~ 40 mV; L9 = 30 mV; 1 see.
219 notion the following experiments were perfomed. In all preparations tested so far, induced activity in L9 never resulted in any observable synaptic input to L7 nor did it effect LT's normal ongoing activity. An example of this is shown in Fig. 7A. More convincing evidence for a peripheral site of action was obtained from experiments performed in 3 preparations. L9's modulatory effect on L7's ability to elicit a gill withdrawal response was first demonstrated (Fig. 7B) and then the small branch of the siphon nerve which carries L9's axon to the gill was severed. Following this, induced activity in L9 no longer resulted in potentiation of LT'S ability to elicit gill movements (Fig. 7C). This was a consistent finding in all 3 of the preparations tested in this manner. The data shown in Fig. 7B also show that induced activity in L9 did not necessarily have to elicit a gill movement in order to have its modulatory action on Lv. However, in the vast majority of cases (91 ~), induced activity in L9 did result in an observable gill movement. DISCUSSION The results reported here clearly show that L9 is a modulatory neuron, in addition to its previous known role of a minor central gill motor neuron n. Lg's activity was found to significantly effect the ability of L7 to elicit a gill withdrawal response. L9 may also have a similar modulatory action on LDG1, but its effect is minor compared to its effect on LT. In addition, Lg's modulatory action on L7 is not dependent on it eliciting a gill withdrawal response, but it is dependent on its innervation to the gill being intact. Thus L9's modulatory action on L7 is not exerted centrally but rather peripherally in the gill. It was demonstrated that LT'S ability to elicit a gill withdrawal response was potentiated following the interposition of L9 activity. With low frequency activity in L7 (6-14 AP's/sec), the gill withdrawal response elicited normally decrements with repeated stimulation; however, as soon as L9 activity was interposed, the response elicited by L7 activation was significantly potentiated both when compared to the initial response amplitude and to the expected amplitude had L9 activity not been interposed (Figs. 1 and 2). It was further shown that the L9 modulatory action was not dependent on a response that had decremented. The interposition of L9 activity also significantly potentiated LT's ability to elicit a gill withdrawal response when frequencies of L7 activation were used (15-20 AP's/sec) which do not result in response decrement (Fig. 3). However, the percent potentiation produced by the interposition of L9 activity on the average was less when the higher frequencies of Lv activity were used even though the range of potentiation produced by L9 activity were similar for both the high and low frequency groups. This is most likely due to a ceiling effect since the amplitudes of the gill withdrawal responses produced by higher frequency L7 stimulation were significantly larger than those produced by the lower frequencies. In addition, L9 may also modulate LDGI's ability to elicit gill withdrawal movements but its modulatory effect is much less than that observed for L7 (Fig. 6). This may in part be due to the fact that repetitive activation of LDG1 leads to response potentiation 10. Thus L0 can certainly be considered as a modulatory neuron in that its activity effects the efficacy of another neuron's ability to elicit a gill withdrawal response.
220 Although L9 is a central gill motor neuron, its modulatory properties on L7 are not dependent upon its ability to elicit an observable gill movement. As was seen (Fig. 7B), L9 activity did not necessarily have to elicit a gill movement in order for it to modulate Lv's ability to elicit a gill response. All that was necessary was that the small branch of the siphon nerve, through which L9 innervates the gill, be intact (Fig. 7C). All central gill motor neurons, however, do not possess the ability to modulate the effectiveness of another gill motor neuron to elicit a gill withdrawal response as does Lg. It was found (Fig. 5B) that the interposition of L7 activity did not modulate Lg's effectiveness to elicit a gill movement nor did the interposition of activity in LDG1 effect L7's ability to elicit a gill withdrawal response (Fig. 5C). It was further found, although not shown here, that the interposition of Lv activity failed to have any modulatory action on LDG1. Of all the central motor neurons tested, including LDG2, only L9 was found to be able to modulate the ability of another gill motor neuron to elicit a gill movement. While L9 was the only central gill motor neuron found to exert modulatory activity on other gill motor neurons, the other gill motor neurons do possess the ability to modulate other gill withdrawal responses. For example, Lukowiak and Peretz 10 showed that induced low level activity in L7 or LDG1 (1-3 AP's/sec), which does not itself elicit an observable gill movement, resulted in a significantly larger reflex amplitude than when the neuron was not artificially depolarized to spike at this rate. In addition, the interposition of L7 or LDG1 activity was also shown to result in dishabituation of the gill withdrawal reflex habituated to repeated tactile stimulation of the siphon or gill lo. Finally, Sinback 13 has shown that induced activity in central gill motor neurons can modulate the periodicity of the spontaneous rhythmic gill respiration movements. It thus appears that in addition to being gill motor neurons, these cells also possess modulatory actions of different types. At this time, it is not known for certain the mechanisms or the actual sites of these modulatory effects. However, it is known that Lg's modulatory effect occurs peripherally in the gill (Fig. 7B, C), as does L7's and LDGI's modulatory action on the amplitude of the gill withdrawal reflex, while LT's and LDGI's modulation of gill rhythmicity occurs centrally. The duration of Lg's modulatory action was not systematically examined in any one of the 20 preparations tested. However, in a number of the preparations, the time interval between the termination of L9 activity and the onset of L7 activity was varied. It was found that Lg's effect persisted for at least 60-90 sec but that it appeared to be strongest in the 15-30 sec interval following the termination of its induced activity. In one preparation, it was found that with an interval of 3 rain, there was no effect. It would thus appear that Lg's modulatory action does not persist much longer than l or 2 min. However, it remains to be seen how variations in the frequency and duration of L9 activity affect not only the persistence of the modulatory effect, but also the modulatory effect itself. It may well be that with higher frequencies of activity for longer duration, the modulatory effect will be larger and longer lasting. This demonstration of a modulatory effect of a neuron on another neuron is not the first to be reported - - similar results have been reported both in Aplysia and in other preparations 2,1m5,16. Thus, this effect may be a widespread phenomenon in the
221 nervous system. Weiss et al, described central modulatory neurons in the cerebral ganglia of Aplysia, the metacerebral cells (MCC's) 15,16. These neurons exelt a modulatory influence on the buccal musculature. They exert their potentiating action both centrally at the level of buccal ganglion motor neurons and peripherally at the level of the buccal muscle. Peripherally the MCC's enhanced muscle contraction by two different mechanisms. They enhanced the size of the EJP's produced by stimulation of the central buccal motor neurons and they had a direct effect on the excitation-contraction coupling process in the muscle itself. The direct effect on the excitation-contraction coupling probably accounted for the greatest proportion of the potentiation observed. Whether La produces its modulatory effect on L7 in a similar manner is not known and remains to be determined. However, it is known that the effect occurs peripherally and not centrally. A certain amount of evidence, however, favours a direct effect on the muscle itself. The amplitude of the pinnule potentials produced by L7 activity does not change following La activity. If La were effecting the amplitude of the EJP's, the size of the pinnule potentials would be expected to change. Additionally, although La has not definitely been shown to be dopaminergic, evidence is accumulating which supports this notion and the infusion of dopamine into the gill results in an increase in cyclic AMP levels4. The increase in c-AMP may modulate the contractability of the muscle fibers in the gill16. An increase in c-AMP synthesis by serotonin has been postulated to account for the direct effect of the MCC's on buccal muscle and for facilitation in the abdominal ganglion in Aplysial, 16. However, further research is necessary to determine just how La's modulatory effect is mediated. La may normally function to amplify the activity of LT. Both motor neurons receive similar synaptic input during spontaneous periodic and tactilely evoked gill withdrawal movements. However, L9's activity does not significantly effect the amplitude of the gill withdrawal reflex evoked by tactile stimulation of the siphon or gill and for this reason it has been described as a motor neuron which makes a relatively minor contribution to the gill withdrawal reflex 8. The above data show that La's major importance may not be in itself eliciting gill movements, but in potentiating LT's ability to cause gill movements. Not only is the amplitude of LT's gill withdrawal response potentiated following La activity, but in the vast majority of instances, the latency of the contraction following the initiation of L7 activity was decreased. It must also be noted that L7 must reach a higher frequency of activity to produce a contraction of similar amplitude as that produced by activity induced in LDG1. Thus it would appear that L7 plays a more minor role in the mediation of gill behaviours. However, if La were active along with L7 or just prior to it, L7 would produce a larger contraction for a given rate of AP's. Thus L7 would play a more major role in gill reflex behaviours. This is indeed what seems to occur with tactile stimulation of the siphon (Lukowiak, in preparation). Preliminary experiments have also indicated that La may play a major role in the CNS control of the gill withdrawal reflex and its subsequent habituation evoked by repeated tactile stimulation of the siphona. It was found that with induced low level La activity 0-3 AP's/sec), repeated tactile stimulation of the siphon no longer resulted in gill reflex habituation. In addition, in an already habituated preparation, this induced low level activity brought about a reversal of
222 h a b i t u a t i o n with repeated stimulation. L9 may thus act as a major central control n e u r o n of gill reflex behaviours by both m o d u l a t i n g the efficacy of other m o t o r n e u r o n s to elicit gill m o v e m e n t s and by m o d u l a t i o n of the peripheral sites of gill reflex habituation. Once the sites a n d mechanisms of L9 m o d u l a t i o n are k n o w n , we will be closer to a n u n d e r s t a n d i n g of how a relatively simple nervous system with relatively few n e u r o n s mediates adaptive behaviours which resemble those of higher organisms where they are mediated by m u c h more complicated nervous systems.
REFERENCES 1 Brunelli, M., Castellucci, V. and Kandel, E., Synaptic facilitation and behavioural sensitization in Aplysia: possible role of serotonin and cyclic AMP, Science, 194 (1976) 1178-1180. 2 Evans, P. D., Talamo, B. R. and Kravitz, E. A., Octopamine neurons: morphology, release of octopamine and possible physiological role, Brain Research, 90 (1975) 340-347. 3 Jacklet, J. and Rine, J., Facilitation at neuromuscularjunctions: contribution to habituation of the Aplysia gill withdrawal reflex, Proc. nat. Acad. Sci. (Wash.), 74 (1977) 1267-1271. 4 Kebabian, P., Kebabian, J., Swann, J. and Carpenter, D., Cyclic AMP in Aplysia gill: increase by putative neurotransmitters, Neurosci. Abstr., 7 (1977) 557. 5 Koester, J. and Kandel, E., Further identification of neurons in the abdominal ganglion of Aplysia using behavioural criteria, Brain Research, 121 (1977) 1-20. 6 Kupfermann, I., Carew, T. and Kandel, E., Local, reflex and central commands controlling gill and siphon movements in Aplysia, J. Neurophysiol., 37 (1974) 996-10t9. 7 Lukowiak, K., CNS control of the PNS-mediated gill withdrawal reflex and its habituation, Canad. J. Physiol. Pharmacol., 55 (1977) 1252-1262. 8 Lukowiak, K., Facilitation, habituation and the retardation of habituation of LT'S elicited gill withdrawal response in Aplysia, Brain Research, 134 (1977) 387-392. 9 Lukowiak, K., Induced activity in L 9 prevents habituation of the gill withdrawal reflex in A plysia, submitted to Soc. Neurosci., 1978 meeting. 10 Lukowiak, K. and Peretz, B., The interaction between the central and peripheral nervous systems in the mediation of gill withdrawal reflex behaviour in Aplysia, J. comp. PhysioL, 117 (1977) 219-244. 11 O'Shea, M. and Evans, P., Synaptic modulation by an identified octopaminergic neurons in the locust, Neurosci. Abstr., 7 (1977) 583. 12 Peretz, B. and Estes, J., Histology and histochemistry of the peripheral plexus in the Aplysia gill, J. NeurobioL, 5 (1974) 3-19. 13 Sinback, C. N., A Convergent Network of ldentified Neurons Modulating Respiratory Rhythms in Aplysia californica. Ph.D. Thesis, University of Kentucky, 1975. 14 Swarm, J., Sinback, C. N. and Carpenter, D., Lg-induced gill contractions in Aplysia antagonized by the dopamine receptor blockers fluphenazine and ergometrine, Neurosci. Abstr., 7 (1977) 592. 15 Weiss, K., Cohen, J. and Kupfermann, I., Potentiation of muscle contraction: a possible modulatory function of an identified serotonergic cell in Aplysia, Brain Research, 99 (1975) 381-386. 16 Weiss, K. R., Cohen, J. L. and Kupfermann, I., Modulatory control of buccal musculature by a serotonergic neuron (metacerebrat cell) in Aplysia, J. NeurophysioL, 41 (1977) 181-203.
207
L9 M O D U L A T I O N OF LT'S ELICITED G I L L W I T H D R A W A L RESPONSE IN APLYSIA
KEN LUKOWIAK* Department of Physiology, McGill University, Montreal, Quebec, H3G 1 Y6 (Canada)
(Accepted June 22nd, 1978)
SUMMARY The effects of induced L9 activity on gill motor neuron LT's ability to elicit a gill withdrawal response were studied. It was found that L9 was a modulator of LT's effectiveness to elicit gill movements. Following L9 activity, L7's ability to elicit a gill withdrawal response was significantly potentiated by up to 240~o of control. L9 activity potentiated L7's elicited gill withdrawal response when frequencies of L7 activity were used which would result in decrement of the response or when frequencies were used in which decrement does not occur. Induced L9 activity may also have a minor potentiating effect on LDGt's ability to elicit gill movements. L9 was the only neuron found to possess these modulatory abilities. The interposition of activity in the other gill motor neurons failed to have any effect. Lg's modulatory role is separate and independent from its role as a gill motor neuron. L9's modulation of L7 is mediated peripherally in the gill and not in the CNS.
INTRODUCTION Recent studies have shown that the central (CNS) and peripheral (PNS) nervous systems interact and are parts of an integrated system which normally mediates adaptive behaviour of the gill withdrawal reflex in Aplysia 10. During the course of that study, it was found that when a central gill motor neuron, L7, was depolarized to produce 6-14 action potentials (AP's) per sec once every 30 sec its elicited gill withdrawal response habituated and that the interposition of a tactile stimulus to the gill or siphon resulted in dishabituation or facilitation of LT's ability to elicit a gill movement. It was further shown that the facilitatory effect of the tactile stimulus on LT's efficacy to elicit gill movements occurred peripherally in the gill and not centrally * Present address: Faculty of Medicine, Division of PhysiologicalSciences, University of Calgary., Calgary, Alberta, Canada, T2N 1N4.
208 in the abdominal ganglion. This and a subsequent study 8 thus showed that the efficacy of Lv's peripheral terminations could be altered by both its own activity and activity evoked in another pathway. Lukowiak and Peretz further examined whether the interposition of activity in the other major gill motor neurons LDG1 or LDG2 could also effect the efficacy of L7's peripheral terminations. They found that the interposition of LDG1 or LDG2 activity had little if any effect on L7's ability to elicit a gill withdrawal response. Swann, et al. 14 found, however, that they could potentiate L7's ability to elicit a gill withdrawal response by perfusing dopamine through the gill. In addition, they presented some evidence that Lg, a minor gill motor neuron, may be dopaminergic. It thus seemed logical, in light of the Swann et al. study and the previous effort of Lukowiak and Peretz, to determine if the interposition of induced activity in L9 results in potentiation of L7's ability to elicit a gill withdrawal response. If potentiation of L7's ability to elicit a gill withdrawal response were observed following the interposition of L9 activity, it would show that a central gill motor neuron could also act to modulate the effectiveness of another gill motor neuron to elicit gill movements. I now report that the interposition of L9 activity significantly potentiated the ability of L7 to elicit a gill withdrawal response, both at low frequencies (6-12 AP's/sec) of stimulation where decrement is normally observed with repeated L7 activation and at higher frequencies (15-20 AP's/sec) where decrement is not observed s. In addition, L9 may also have a potentiating effect on LDG1 ability to elicit gill movements but this effect is minor. These data then, along with the results obtained in a number of different preparations indicate that the modulation of the effectiveness of a neuron by the activity of another neuron is a feature of the neurons system which is both widespread and important across phyla 2,~1,1~,a~. METHODS
Aplysia californica, obtained from Pacific Biomarine Laboratories (Venice, Calif.) and weighing between 150 and 250 g were used. The animals were kept before use in artificial seawater (Instant Ocean) at 15-16 °C and pH 7.9. The preparation used in these studies was similar to that described previously 1°. It consisted of the siphon, mantle, gill and the abdominal ganglion. The branchial, ctenidial and siphon nerves by which the abdominal ganglion innervates the gill and siphon were left intact. Particular attention was paid to a small nerve branch that leaves the siphon nerve and innervates the gill. It has been shown that it is this small branch of the siphon nerve which carries LCs axon to the gill 7. As a control, in some preparations this small branch of the siphon nerve will be severed after L9 modulatory effect has been demonstrated in order to show that Lg's effect occurs in the periphery and not in the abdominal ganglion. All preparations were maintained at l 5-16 °C in a 1000 ml chamber of artificial seawater during the course of the experiment. The preparations remained viable for at least 24 h as evidenced by the continued observance of the periodic spontaneous gill respiratory movements. Twenty preparations were used in this present study and data from all 20 have been used and statistically analyzed.
209 The amplitude of the gill withdrawal response was measured using a force transducer (Grass Ft .03) connected to a single gill pinnule by fine surgical thread and recorded on an oscilloscope and a polygraph from which the measurements were made. Recording gill movements in this manner did not result in any observable damage to the gill and gill movements could continue to be elicited 24 h after dissection. Neuron identification was based on a number of criteria and is consistent with that of Koester and Kandel 5. For L7 and LDG1 the criteria included: location, size, synaptic activity during spontaneous gill respiratory movements, the appearance of one-for-one activity recorded in the gill pinnule by an extracellular suction electrode (referred to as a pinnule potential) and the type of gill movement elicited by depolarization of the neuron. Similar criteria were also used to identify L0. L0 receives similar synaptic input as Lv during periodic spontaneous gill contractions and causes a similar type of gill movement as L7 when it's depolarized. However, L0 must be depolarized to produce a higher rate of AP's for a longer duration than L7 to produce a similar amplitude contraction. Moreover, pinnule potentials were not observed which correlated with L9 activity and L9 innervates the gill via a small branch of the siphon nerve and not via the branchial or ctenidial nerves. In this regard L9 is unique. No distinction has been made between the two L0's, both gave similar results. Normally the larger of the two (when there were two) was recorded from. Micropipettes filled with 3 M KC1 and having a resistance between 15 and 30 M ~ were used. A bridge circuit in the electrometer (Getting M-5) allowed simultaneous recording and stimulation. When identified central motor neurons were depolarized, constant current (1000 msec duration) was passed into the neuron through the recording electrode. The rate of AP's on each trial was held constant and for L7 two frequencies were used: 6-12 AP's/sec and 15-20 AP's/sec. To determine if induced activity in L9 effected the gill withdrawal response elicited by L7 activity, the following experimental protocols were followed. L7 was depolarized to produce a fixed number of AP's (6-12) for one sec; L0 was then depolarized to produce 5-15 AP's/sec for 4-6 sec and then L7 was depolarized again to produce the same number of AP's as on the first trial. The interval between L7 depolarization was 30 sec. This series of LT, Lg, L7 stimulation constitute a test. Preparations were always rested at least 10 min between tests. Data were normalized, 100 ~ being the amplitude of the initial gill withdrawal response in the individual test. When pooled data are presented the mean 4- S.E.M. are given. Similar tests were also performed when L7 was depolarized to produce 15-20 AP's/sec as well as when LDG1 was depolarized to produce 6-10 AP's/sec on each trial. In one set of control experiments the interposition of L9 activity was omitted; while in another set LDGa activity was interposed between L7 stimulations or L7 activity was interposed between LDG1 stimulations. L7 or LDG1 activity was also interposed between L9 stimulation. Finally, in some preparations after L9's potentiating effect was demonstrated, the small branch of the siphon nerve by which L9 innervates the gill was severed and following a 1 h rest the experiment was repeated in order to determine the locus of La's effect.
210
A
B
250
+ ~200
;[ O (n '" r~
1SO
rt uJ )-
_~
!oo
I--
50
L71
I'1 L72
I-71
L72
Fig. 1. Pooled normalized data showing the effect of the interposition of L9 activity on the ability of L7 to elicit a gill withdrawal response. A: L7 was initially depolarized to produce between 6-14 AP's per sec (L71) and this response was taken as 100~. Thirty sec later L7 was again depolarized to produce the same number of AP's as on the initial trial (L72). As can be seen, the amplitude of the contraction (mean ± S.E.M.) produced on the second L7 trial (L72) was reduced by 35 ~ . These data are from 14 preparations with 45 separate tests. B : the data are from the same 14 preparations (n = 92, individual tests). L7 was initially depolarized to produce 6-14 AP's per sec (L71) and again this response was taken as 100 ~ . L9 activity was then interposed. Following the interposition of L9 activity and 30 sec after the initial depolarization, L7 was again depolarized to produce the same number of AP's as on the initial trial. As can be seen (L72) , the amplitude of the response following L9 activity was much larger than the initial response. It was found that the amplitude of the response following L9 activity was significantly larger than the initial response (P < 0.001) and the expected (P < 0.001) had not L,3 activity been interposed (L72 in A).
Fig. 2. Potentiation of L7's elicited gill withdrawal response by induced L9 activity. A: L7 was depolarized to produce 6 AP's for 1 sec and this produced a gill withdrawal response of 8 mm. Fourteen seconds later, L9 was depolarized to produce 12 AP's/sec for approximately 4 sec and this produced a gill withdrawal response of about 12 mm. Thirty seconds after the initial depolarization of LT, L7 was again stimulated to produce 6 AP's for 1 sec and this resulted in a potentiated response of 15 ram. B: in another preparation, as in A, only the interstimulus interval (ISI) was shortened to 17 sec. The initial L7 response (6 AP's) was 13 ram, the response elicited by L9 depolarization (5 AP's/sec for 5 sec) was 5 mm and the response elicited by L7 stimulation (6 AP's) following L9 stimulation was approximately 27 mm. C: an oscilloscope recording taken from a third preparation which again shows the potentiation of L7's gill withdrawal response following L9 activity. GT1 was the initial gill withdrawal response elicited by L7 stimulation (6 AP's) while GT2 was the gill withdrawal response elicited by L7 stimulation (6 AP's) following the interposition of L9 activity (not shown, 8 AP's/sec for 4 sec. The ISI for L7 stimulation was 30 sec and L9 was depolarized 18 sec following the initial L7 depolarization. The base line for GT~ was slightly adjusted upward so as to make a clearer presentation. Scales: A, B : L7 -- 40 mV, L• 30 mV, 1 sec: 7 mm = 100 mg, C: 20 mV, 100 msec.
211
|
15
6
13
2?
L?
L L9
C
m
212 RESULTS Previously it had been shown that when Lv was depolarized to produce between 6-14 AP's per trial, the amplitude of the elicited gill withdrawal response decremented with repeated activation (I/30 sec) 8,10. In the 20 preparations tested in this series of experiments, similar results were obtained and these data are shown in Fig. 1A. On the average, the response amplitude on the 2rid L7 trial (L72) was reduced by 35'},~, compared to the initial response. In these same 20 preparations, when L9 activity was interposed between the two Lv trials, the response amplitude on the 2nd L7 trial (L72) was not reduced but rather it was potentiated by an average of 127 ?/,, (Fig. I B). At least 3 separate experimental tests were performed in each of the 20 preparations and the results of 92 individual tests are represented in the bar graph in Fig. lB. A t-test was performed to statistically determine whether the interposition of L9 activity significantly effected the effectiveness of L7 to elicit gill movements. It was found that there was a significant difference (P 0.2). Thus L9 m a y have a minor m o d u l a t o r y effect on L D G I ' S ability to elicit a gill withdrawal response. L9 could be exerting its m o d u l a t o r y action on L7 either in the abdominal ganglion or out in the periphery, in the gill itself. Because it had previously been shown that infusion o f dopamine t h r o u g h the gill potentiated Lv's elicited gill withdrawal response, it was t h o u g h t that Lg's action might also occur peripherally. To test this
218
A
T
1S
11
irr
L_
LI Fig. 7. A: Induced activity in L9 does not have any observable effect on activity in L7. L9 was depolarized to produce 8 AP's/sec for 3 sec and as can be seen in the higher gain recording from L 7 , this did not result in any observable synaptic input to L7 nor did it affect the spontaneous ongoing synaptic activity in L7. Scale: L9 = 50 mV; L7 ~ 5 m V , 1000 msec. B, C: Lg's potentiating effect only occurs with the siphon nerve intact. In B, induced activity in L9 potentiates the gill withdrawal response elicited by L7 stimulation. However, following cutting of the siphon nerve through which L~ innervates the gill, the interposition of L9 activity does not result in potentiation of LT's response (B, C). This has been a consistent finding. Scale: L7 ~ 40 mV; L9 = 30 mV; 1 see.
219 notion the following experiments were perfomed. In all preparations tested so far, induced activity in L9 never resulted in any observable synaptic input to L7 nor did it effect LT's normal ongoing activity. An example of this is shown in Fig. 7A. More convincing evidence for a peripheral site of action was obtained from experiments performed in 3 preparations. L9's modulatory effect on L7's ability to elicit a gill withdrawal response was first demonstrated (Fig. 7B) and then the small branch of the siphon nerve which carries L9's axon to the gill was severed. Following this, induced activity in L9 no longer resulted in potentiation of LT'S ability to elicit gill movements (Fig. 7C). This was a consistent finding in all 3 of the preparations tested in this manner. The data shown in Fig. 7B also show that induced activity in L9 did not necessarily have to elicit a gill movement in order to have its modulatory action on Lv. However, in the vast majority of cases (91 ~), induced activity in L9 did result in an observable gill movement. DISCUSSION The results reported here clearly show that L9 is a modulatory neuron, in addition to its previous known role of a minor central gill motor neuron n. Lg's activity was found to significantly effect the ability of L7 to elicit a gill withdrawal response. L9 may also have a similar modulatory action on LDG1, but its effect is minor compared to its effect on LT. In addition, Lg's modulatory action on L7 is not dependent on it eliciting a gill withdrawal response, but it is dependent on its innervation to the gill being intact. Thus L9's modulatory action on L7 is not exerted centrally but rather peripherally in the gill. It was demonstrated that LT'S ability to elicit a gill withdrawal response was potentiated following the interposition of L9 activity. With low frequency activity in L7 (6-14 AP's/sec), the gill withdrawal response elicited normally decrements with repeated stimulation; however, as soon as L9 activity was interposed, the response elicited by L7 activation was significantly potentiated both when compared to the initial response amplitude and to the expected amplitude had L9 activity not been interposed (Figs. 1 and 2). It was further shown that the L9 modulatory action was not dependent on a response that had decremented. The interposition of L9 activity also significantly potentiated LT's ability to elicit a gill withdrawal response when frequencies of L7 activation were used (15-20 AP's/sec) which do not result in response decrement (Fig. 3). However, the percent potentiation produced by the interposition of L9 activity on the average was less when the higher frequencies of Lv activity were used even though the range of potentiation produced by L9 activity were similar for both the high and low frequency groups. This is most likely due to a ceiling effect since the amplitudes of the gill withdrawal responses produced by higher frequency L7 stimulation were significantly larger than those produced by the lower frequencies. In addition, L9 may also modulate LDGI's ability to elicit gill withdrawal movements but its modulatory effect is much less than that observed for L7 (Fig. 6). This may in part be due to the fact that repetitive activation of LDG1 leads to response potentiation 10. Thus L0 can certainly be considered as a modulatory neuron in that its activity effects the efficacy of another neuron's ability to elicit a gill withdrawal response.
220 Although L9 is a central gill motor neuron, its modulatory properties on L7 are not dependent upon its ability to elicit an observable gill movement. As was seen (Fig. 7B), L9 activity did not necessarily have to elicit a gill movement in order for it to modulate Lv's ability to elicit a gill response. All that was necessary was that the small branch of the siphon nerve, through which L9 innervates the gill, be intact (Fig. 7C). All central gill motor neurons, however, do not possess the ability to modulate the effectiveness of another gill motor neuron to elicit a gill withdrawal response as does Lg. It was found (Fig. 5B) that the interposition of L7 activity did not modulate Lg's effectiveness to elicit a gill movement nor did the interposition of activity in LDG1 effect L7's ability to elicit a gill withdrawal response (Fig. 5C). It was further found, although not shown here, that the interposition of Lv activity failed to have any modulatory action on LDG1. Of all the central motor neurons tested, including LDG2, only L9 was found to be able to modulate the ability of another gill motor neuron to elicit a gill movement. While L9 was the only central gill motor neuron found to exert modulatory activity on other gill motor neurons, the other gill motor neurons do possess the ability to modulate other gill withdrawal responses. For example, Lukowiak and Peretz 10 showed that induced low level activity in L7 or LDG1 (1-3 AP's/sec), which does not itself elicit an observable gill movement, resulted in a significantly larger reflex amplitude than when the neuron was not artificially depolarized to spike at this rate. In addition, the interposition of L7 or LDG1 activity was also shown to result in dishabituation of the gill withdrawal reflex habituated to repeated tactile stimulation of the siphon or gill lo. Finally, Sinback 13 has shown that induced activity in central gill motor neurons can modulate the periodicity of the spontaneous rhythmic gill respiration movements. It thus appears that in addition to being gill motor neurons, these cells also possess modulatory actions of different types. At this time, it is not known for certain the mechanisms or the actual sites of these modulatory effects. However, it is known that Lg's modulatory effect occurs peripherally in the gill (Fig. 7B, C), as does L7's and LDGI's modulatory action on the amplitude of the gill withdrawal reflex, while LT's and LDGI's modulation of gill rhythmicity occurs centrally. The duration of Lg's modulatory action was not systematically examined in any one of the 20 preparations tested. However, in a number of the preparations, the time interval between the termination of L9 activity and the onset of L7 activity was varied. It was found that Lg's effect persisted for at least 60-90 sec but that it appeared to be strongest in the 15-30 sec interval following the termination of its induced activity. In one preparation, it was found that with an interval of 3 rain, there was no effect. It would thus appear that Lg's modulatory action does not persist much longer than l or 2 min. However, it remains to be seen how variations in the frequency and duration of L9 activity affect not only the persistence of the modulatory effect, but also the modulatory effect itself. It may well be that with higher frequencies of activity for longer duration, the modulatory effect will be larger and longer lasting. This demonstration of a modulatory effect of a neuron on another neuron is not the first to be reported - - similar results have been reported both in Aplysia and in other preparations 2,1m5,16. Thus, this effect may be a widespread phenomenon in the
221 nervous system. Weiss et al, described central modulatory neurons in the cerebral ganglia of Aplysia, the metacerebral cells (MCC's) 15,16. These neurons exelt a modulatory influence on the buccal musculature. They exert their potentiating action both centrally at the level of buccal ganglion motor neurons and peripherally at the level of the buccal muscle. Peripherally the MCC's enhanced muscle contraction by two different mechanisms. They enhanced the size of the EJP's produced by stimulation of the central buccal motor neurons and they had a direct effect on the excitation-contraction coupling process in the muscle itself. The direct effect on the excitation-contraction coupling probably accounted for the greatest proportion of the potentiation observed. Whether La produces its modulatory effect on L7 in a similar manner is not known and remains to be determined. However, it is known that the effect occurs peripherally and not centrally. A certain amount of evidence, however, favours a direct effect on the muscle itself. The amplitude of the pinnule potentials produced by L7 activity does not change following La activity. If La were effecting the amplitude of the EJP's, the size of the pinnule potentials would be expected to change. Additionally, although La has not definitely been shown to be dopaminergic, evidence is accumulating which supports this notion and the infusion of dopamine into the gill results in an increase in cyclic AMP levels4. The increase in c-AMP may modulate the contractability of the muscle fibers in the gill16. An increase in c-AMP synthesis by serotonin has been postulated to account for the direct effect of the MCC's on buccal muscle and for facilitation in the abdominal ganglion in Aplysial, 16. However, further research is necessary to determine just how La's modulatory effect is mediated. La may normally function to amplify the activity of LT. Both motor neurons receive similar synaptic input during spontaneous periodic and tactilely evoked gill withdrawal movements. However, L9's activity does not significantly effect the amplitude of the gill withdrawal reflex evoked by tactile stimulation of the siphon or gill and for this reason it has been described as a motor neuron which makes a relatively minor contribution to the gill withdrawal reflex 8. The above data show that La's major importance may not be in itself eliciting gill movements, but in potentiating LT's ability to cause gill movements. Not only is the amplitude of LT's gill withdrawal response potentiated following La activity, but in the vast majority of instances, the latency of the contraction following the initiation of L7 activity was decreased. It must also be noted that L7 must reach a higher frequency of activity to produce a contraction of similar amplitude as that produced by activity induced in LDG1. Thus it would appear that L7 plays a more minor role in the mediation of gill behaviours. However, if La were active along with L7 or just prior to it, L7 would produce a larger contraction for a given rate of AP's. Thus L7 would play a more major role in gill reflex behaviours. This is indeed what seems to occur with tactile stimulation of the siphon (Lukowiak, in preparation). Preliminary experiments have also indicated that La may play a major role in the CNS control of the gill withdrawal reflex and its subsequent habituation evoked by repeated tactile stimulation of the siphona. It was found that with induced low level La activity 0-3 AP's/sec), repeated tactile stimulation of the siphon no longer resulted in gill reflex habituation. In addition, in an already habituated preparation, this induced low level activity brought about a reversal of
222 h a b i t u a t i o n with repeated stimulation. L9 may thus act as a major central control n e u r o n of gill reflex behaviours by both m o d u l a t i n g the efficacy of other m o t o r n e u r o n s to elicit gill m o v e m e n t s and by m o d u l a t i o n of the peripheral sites of gill reflex habituation. Once the sites a n d mechanisms of L9 m o d u l a t i o n are k n o w n , we will be closer to a n u n d e r s t a n d i n g of how a relatively simple nervous system with relatively few n e u r o n s mediates adaptive behaviours which resemble those of higher organisms where they are mediated by m u c h more complicated nervous systems.
REFERENCES 1 Brunelli, M., Castellucci, V. and Kandel, E., Synaptic facilitation and behavioural sensitization in Aplysia: possible role of serotonin and cyclic AMP, Science, 194 (1976) 1178-1180. 2 Evans, P. D., Talamo, B. R. and Kravitz, E. A., Octopamine neurons: morphology, release of octopamine and possible physiological role, Brain Research, 90 (1975) 340-347. 3 Jacklet, J. and Rine, J., Facilitation at neuromuscularjunctions: contribution to habituation of the Aplysia gill withdrawal reflex, Proc. nat. Acad. Sci. (Wash.), 74 (1977) 1267-1271. 4 Kebabian, P., Kebabian, J., Swann, J. and Carpenter, D., Cyclic AMP in Aplysia gill: increase by putative neurotransmitters, Neurosci. Abstr., 7 (1977) 557. 5 Koester, J. and Kandel, E., Further identification of neurons in the abdominal ganglion of Aplysia using behavioural criteria, Brain Research, 121 (1977) 1-20. 6 Kupfermann, I., Carew, T. and Kandel, E., Local, reflex and central commands controlling gill and siphon movements in Aplysia, J. Neurophysiol., 37 (1974) 996-10t9. 7 Lukowiak, K., CNS control of the PNS-mediated gill withdrawal reflex and its habituation, Canad. J. Physiol. Pharmacol., 55 (1977) 1252-1262. 8 Lukowiak, K., Facilitation, habituation and the retardation of habituation of LT'S elicited gill withdrawal response in Aplysia, Brain Research, 134 (1977) 387-392. 9 Lukowiak, K., Induced activity in L 9 prevents habituation of the gill withdrawal reflex in A plysia, submitted to Soc. Neurosci., 1978 meeting. 10 Lukowiak, K. and Peretz, B., The interaction between the central and peripheral nervous systems in the mediation of gill withdrawal reflex behaviour in Aplysia, J. comp. PhysioL, 117 (1977) 219-244. 11 O'Shea, M. and Evans, P., Synaptic modulation by an identified octopaminergic neurons in the locust, Neurosci. Abstr., 7 (1977) 583. 12 Peretz, B. and Estes, J., Histology and histochemistry of the peripheral plexus in the Aplysia gill, J. NeurobioL, 5 (1974) 3-19. 13 Sinback, C. N., A Convergent Network of ldentified Neurons Modulating Respiratory Rhythms in Aplysia californica. Ph.D. Thesis, University of Kentucky, 1975. 14 Swarm, J., Sinback, C. N. and Carpenter, D., Lg-induced gill contractions in Aplysia antagonized by the dopamine receptor blockers fluphenazine and ergometrine, Neurosci. Abstr., 7 (1977) 592. 15 Weiss, K., Cohen, J. and Kupfermann, I., Potentiation of muscle contraction: a possible modulatory function of an identified serotonergic cell in Aplysia, Brain Research, 99 (1975) 381-386. 16 Weiss, K. R., Cohen, J. L. and Kupfermann, I., Modulatory control of buccal musculature by a serotonergic neuron (metacerebrat cell) in Aplysia, J. NeurophysioL, 41 (1977) 181-203.
