- Email: [email protected]
Behavioral hierarchy in the medicinal leech, Hirudo medicinalis: feeding as a dominant behavior
Behavioural Brain Research 90 (1998) 13 – 21
Research report
Behavioral hierarchy in the medicinal leech, Hirudo medicinalis: feeding as a dominant behavior Lisa M. Misell, Brian K. Shaw, William B. Kristan Jr. * Department of Biology 0357, Uni6ersity of California, San Diego, La Jolla, CA 92093 -0357, USA Received 13 December 1996; received in revised form 25 April 1997; accepted 26 April 1997
Abstract The effect of feeding behavior on other behaviors (swimming, crawling and shortening) was investigated in the leech, Hirudo medicinalis. The stimulus locations and intensities required to produce mechanically elicited behaviors were first determined in the non-feeding leech. Stimuli were delivered while the leech was in various body positions to determine whether stimulus location affected behavioral response. Response thresholds were determined for the mechanically elicited behaviors. The same stimuli were then applied to feeding leeches to determine if response thresholds had changed. A solution with NaCl and arginine was used to elicit feeding. The same sets of stimuli were applied at intervals for an hour after feeding, to determine the duration of feeding-induced changes in behavior. Depending on the body position and stimulus location, stimuli produced different combinations of behaviors that included shortening, swimming and crawling. Anterior stimuli generally elicited shortening, whereas posterior stimuli generally elicited crawling and swimming, with swimming more likely to ventral stimulation than to dorsal stimulation. Having the front sucker attached changed these behavioral patterns. During feeding, the response thresholds changed dramatically, from 3–5 V to greater than 9 V. This increase in threshold began with the start of feeding, even before ingestion commenced. Suppression of the behaviors lasted up to 1 h after the end of feeding, with the effect on swimming being the most pronounced and longest lasting. © 1998 Elsevier Science B.V. Keywords: Behavioral hierarchy; Behavioral choice; Decision-making; Feeding; Medicinal leech
1. Introduction To behave adaptively, animals often give higher priority to some behaviors than to others. An early proposal was that components of behaviors are elicited by ‘sign stimuli’ and that responses to these stimuli are arranged in a hierarchical cascade [25]. Later authors showed that a strict hierarchical arrangement was too simple to explain all behavioral interactions [18]. For instance, one must also consider such ideas as ‘drive’ and ‘internal state’ to predict many behaviors [26]. A pioneering effort to track down neuronal mechanisms by which animals * Corresponding author. Tel: + 1 619 5344760; fax: + 1 619 5347309; e-mail [email protected] 0166-4328/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 6 - 4 3 2 8 ( 9 7 ) 0 0 0 7 2 - 7
choose among behaviors [4] concluded that the marine mollusc Pleurobranchaea showed a behavioral hierarchy: it fed rather than righting itself, mating, or withdrawing from a weak mechanical stimulus, and laid eggs rather than feeding. Further study showed that this hierarchy was flexible; for instance, feeding would drop well down the hierarchy in a well-fed animal [5]. Similar studies of behavioral hierarchies have since been carried out in a variety of molluscs, including Aplysia [29], Helix [1] and Hermissenda [22]. Ideally, this kind of information about behavior can provide a foundation for explorations of the neural underpinnings of behavioral choice. An example of this is the study of the neural mechanisms underlying the suppression of withdrawal during feeding in Pleurobranchaea [10].
14
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
In the present study, we concerned ourselves with aspects of the behavioral hierarchy of the medicinal leech, Hirudo medicinalis. In particular, we were interested in what priority feeding has relative to other behaviors. The medicinal leech is a bloodsucking carnivore which feeds infrequently on amphibians or mammals; a leech may go for weeks or months between blood meals [23]. Therefore, it would make sense that feeding would have a high priority for the leech: on those rare occasions when a meal was available, it might not want to be distracted. In line with this, leeches with a variety of incisions in their bodies will feed normally [12,28]. We compared feeding to mechanically elicited reflex or locomotor behaviors. Mechanical stimulation can elicit a range of such behaviors in leeches, including local bending, whole-body shortening, swimming, and crawling, depending upon the location and intensity of the stimulus [11]. The neural basis of these behaviors is understood to varying degrees. The circuits for local bending and swimming are quite well-characterized [15,16,3]; the circuit for the wholebody shortening reflex is partially understood [24]; and the outline of the circuit for crawling has been established [2,6]. These circuits provide a rich opportunity for studying the neural basis of behavioral choice and interactions between behaviors. In contrast to the other behaviors, very little is known at present about the neuronal basis of feeding, except for the possible role of modulatory neurons [14,28], although the establishment of a semi-intact preparation that feeds [28] has made such a characterization feasible. The experimental approach we employed was first to determine the stimulus locations and intensities that reliably produce the mechanically elicited behaviors of interest in leeches that were not feeding. We then applied the same stimuli to feeding leeches to determine if the response thresholds changed. We found a dramatic increase in the thresholds for the mechanically elicited behaviors during feeding; in fact, even stimuli that were far above the behavioral thresholds in nonfeeding leeches could not produce whole-body behaviors in feeding leeches. Thus it appears that feeding occupies a high position in the leech’s behavioral hierarchy.
2. Methods Experiments were performed from May to July on leeches that had not been fed a blood meal for at least 2 months. Animals were kept in artificial pond water (38 mg/liter Instant Ocean), at 15°C, with 12:12 h light:dark cycles. Experiments were performed in a 10 gallon aquarium which was filled with 3 – 4 cm of artificial pond water at an ambient temperature of 20 – 22°C. The leeches were acclimatized to room tem-
perature water for 30 min prior to starting an experiment. Several of the experiments were videotaped using a Sanyo VDC 3825 video camera and a Panasonic AG-1730 video cassette recorder for frame-by-frame analysis. The stimulus used to elicit behaviors was a train of electrical shocks delivered with a hand-held electrode. These electrical stimuli mimic mechanical stimuli—they activate the terminals of touch (T) and pressure (P) mechanoreceptors —but are more easily controlled [11]. The terminals of the electrode were 1 mm apart and were made from 0.008 in. diameter silver wire. They were glued to the end of a wooden dowel and fully insulated except for the terminal 1 mm. A Grass SD9 stimulator was used to deliver 1 msec pulses in trains at 10 Hz. The tips of the wires were placed lightly on the skin and at least 1 s was allowed to elapse before the electrical stimulus was delivered, to be sure that the animal was not responding to mechanical contact. The electrical stimulus was applied until the animal responded, which generally occurred within 1 s. In cases when no response was produced, the stimulus was withdrawn after approximately 2 s. To avoid habituation, trials were separated by at least 3 min.
2.1. Responses to stimuli at different body locations To establish a baseline for subsequent experiments, responses were scored to stimuli given at three different locations on the leech body. Stimuli of 8 V intensity, suprathreshold for eliciting behaviors [11], were delivered to the anterior dorsal, posterior ventral, and posterior dorsal surfaces of the leech (Fig. 1A). The anterior stimulus was applied to the midline of segment 2 or 3, and the posterior stimulus was delivered to midline of the last few segments, 18–21. These three stimulus locations were tested with the leech in three different body positions: with just the back sucker attached (back sucker attached), just the front sucker attached (front sucker attached), and with both suckers attached (both suckers attached). Each leech was stimulated approximately 15 times.
2.2. Measurement of response thresholds Thresholds for shortening, crawling and swimming were determined by delivering stimuli of varying intensity, ranging from 2–8 V in 1 V increments. Trials were performed in random order (a ‘trial’ is a single stimulus presentation for one animal). The threshold for shortening was defined as the weakest stimulus needed to cause an observable movement of the whole body. The threshold for crawling was defined as the stimulus needed to cause at least one step, consisting of the release of the back sucker, forward movement, and replacement of the back sucker. The threshold for
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
swimming was defined as the stimulus needed to produce the characteristic undulating swimming movements of the leech. For these experiments, only the anterior dorsal and posterior ventral stimulus sites were used, and the leech was stimulated only when in a both suckers attached body position. This was done for comparison with subsequent feeding trials when the leech would have both suckers attached to the feeding apparatus.
2.3. Response thresholds during feeding The feeding apparatus consisted of a plastic 50 ml centrifuge tube with a diameter of 3 cm, cut and smoothed so that the tube was open on both ends (Fig. 1B). One end was covered with sausage casing (processed bovine intestine) that had been rinsed thoroughly in distilled water. The casing was stretched over the open end of the tube and taped to the outside of it. The tube was suspended in the aquarium, with the covered end just below the surface of the water. To determine thresholds for swimming, shortening, and crawling while feeding, leeches were presented with
Fig. 1. Techniques used to deliver stimuli and induce feeding. A. A diagram of an adult medicinal leech, showing the three stimulus locations used in these experiments. Anterior dorsal stimuli were delivered on the dorsal midline of segment 2 or 3. Posterior stimuli were delivered on the midline of the dorsal or ventral surface of segments 18 – 20. B. The feeding apparatus used for feeding trials. A 50 ml polyethelene (3 cm diameter) centrifuge tube, with the closed end sawed off and smoothed, was covered by a stretched layer of sausage casing. The casing was taped to the side of the tube. The apparatus was suspended in a 10 gallon aquarium with the covered end just below the surface of the water.
15
a solution of 150 mM NaCl and 5 mM L-arginine, which has been shown to elicit feeding as effectively as does whole blood [7]. The solution was warmed to 38°C, to make it more likely to elicit feeding [7], and approximately 30 ml was added to the centrifuge tube. As soon as the leeches were within 1–2 cm of the feeding tube, they attended to it and began to probe the surface of the sausage casing with their anterior sucker. Typically, within seconds, the leeches would attach their front sucker to the sausage casing and bite through the surface. Both posterior ventral and anterior dorsal stimuli were delivered while the leech was actively feeding (pharyngeal peristalsis could be observed) with both suckers attached to the feeding apparatus.
2.4. Responses to suprathreshold stimuli before and after feeding Behavioral responses to stimuli were measured before and after feeding, using 8 V stimulus pulses. Before food was presented, each leech was tested once to be sure that it responded like the animals in the earlier experiments. A series of stages was operationally defined: pre-feeding, the period of probing and sucker attachment that occurred before the leech bit through the sausage casing; feeding, starting when rapid peristalsis of the anterior segments began; and post-feeding, beginning when the leech detached its front sucker from the feeding surface. Because leeches usually attach their posterior sucker as well as their anterior sucker while feeding, most stimulus trials were in the both suckers attached position. One group of leeches were tested using the previously described NaCl/arginine solution. A smaller group was tested with cow blood, to be sure that responses seen were due to feeding per se, and not due to receiving the NaCl/arginine solution instead of blood. The solutions were presented as previously described. In pre-feeding trials, stimuli were delivered while the leech was probing and attaching to the feeding apparatus. Leeches fed for approximately 20–25 min before detaching from the surface, similar to the feeding durations observed in other studies [13]. After feeding, stimuli were delivered immediately upon detachment and every 3 min thereafter. At each 3 min interval, both an anterior dorsal stimulus and a posterior ventral stimulus were given, with about a 30 second interval between them.
3. Results
3.1. Beha6ioral responses to skin stimulation Skin stimulation produced a range of behaviors in the leech (Fig. 2). Anterior stimuli most commonly
16
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
Fig. 2. Behavioral responses to an 8 V stimulus at the three different body locations, with the leeches in different body positions. The attachment positions were: back sucker attached (BSA), front sucker attached (FSA), and back and front sucker attached (BFSA). Ten leeches were tested approximately 15 times each, for a total of 156 trials. The n values listed on each graph represent the number of trials performed under each set of conditions. The values of n differ because some positions were more frequently assumed by the leeches than others. Each leech received one to five trials per body position, except for the front sucker attached position. Several leeches did not assume this position and thus did not have any front sucker attached trials. The most frequently assumed position was back sucker attached, and so this category includes the most trials. The most frequently observed behaviors were swimming, crawling, and shortening; behaviors in the ‘other’ category include local bending. ‘No resp.’ refers to no response observed. A. Responses to anterior dorsal stimuli. B. Responses to posterior dorsal stimuli. C. Responses to posterior ventral stimuli.
caused a whole-body shortening reflex (when the leech was in the back sucker attached or both suckers attached position): the leech contracted its body, releasing its front sucker if it had been attached and keeping the back sucker attached, pulling the body backward away from the stimulus. Posterior stimuli most often caused
swimming or a crawling step, which also served to get the animal away from the stimulus. The beginning of a crawling step sometimes resembled a whole-body shortening reflex when the front sucker was attached: the rear sucker was released (if it had been attached) and the animal shortened, pulling the body forward toward
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
17
Fig. 3. Response thresholds for shortening, crawling and swimming in non-feeding leeches. Seven leeches were tested, using both the anterior dorsal and posterior ventral stimulus locations. A. Thresholds for shortening to anterior dorsal stimuli. A total of 69 trials were performed. Each voltage was tested seven to ten times, in random order. B. Crawling and swimming thresholds to posterior ventral stimuli. A total of 63 trials were performed, with each voltage tested seven to 12 times.
the attached front sucker. These responses were counted as crawling if the rear sucker re-attached after the contraction and the front sucker subsequently released and the animal extended; otherwise, the responses were counted as shortenings. When the back sucker only was attached, crawling steps began with an extension, followed by an attachment of the front sucker. The ‘other’ category consisted largely of weak responses, such as local bending. Leeches were rarely found in the front sucker attached position—usually only when they were in the middle of a crawling step, with the posterior sucker moving forward to reattach — thus accounting for the lower numbers of front sucker attached trials for each stimulus site. In contrast to the results for the other body positions, leeches in the front sucker attached position usually didn’t shorten in response to an anterior stimulus, but showed a range of ‘other’ behaviors such as local bends, wriggling, and front-to-back extensions. This result highlights the importance of the leech’s body position and the state of its suckers in regulating the expression of its behavioral responses; it would make no sense for a leech to shorten to an anterior stimulus when its back sucker was unattached because it would pull itself forward, toward the stimulus. Posterior ventral trials were not performed in the front sucker attached position because the leech was usually moving its posterior end forward, making delivery of the stimulus difficult. The present results show some resemblances to those previously obtained in de-brained leeches [11]. In both cases, anterior stimuli most often caused shortening. However, in de-brained leeches, posterior dorsal stimuli usually caused swimming, whereas in the present study
they usually caused crawling (de-brained leeches generally do not crawl [2]). Posterior ventral stimuli, on the other hand, caused both swimming and crawling in this study. Because we were interested in the effects on both swimming and crawling behaviors, the posterior ventral site rather than the posterior dorsal was used in subsequent experiments.
3.2. Response thresholds in non-feeding and feeding leeches The thresholds for shortening, crawling and swimming in non-feeding leeches were roughly in the range of 3–5 V (Fig. 3), similar to those in de-brained leeches [11,24]. Behaviors in the ‘other’ category for these experiments were mostly local bends or no response. It was obvious from initial pilot experiments that the thresholds for shortening, swimming and crawling were higher during feeding. Thus, to illustrate the shift in thresholds, only voltages from 6–9 V were used in feeding experiments. Four leeches were tested (with the NaCl/arginine solution), using the anterior dorsal and posterior ventral stimulus locations. For the anterior dorsal site, 13 trials total were performed, with each voltage tested three to four times. For the posterior ventral site, 12 trials total were performed, with each voltage tested two to four times. The results were dramatic. During feeding, 6–9 V stimuli failed completely to elicit shortening, swimming, or crawling behaviors. One trial was performed with a stimulus of 14 V and this produced merely a local bend. The only reactions observed during these trials were small local bends at the stimulus site or no response at all. In particular, the stimuli did not cause a termination of feeding.
18
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
3.3. Responses to suprathreshold stimuli before and after feeding Different responses were observed in leeches to stimuli delivered before and after feeding (Fig. 4). The control trials demonstrate that the leeches behaved normally prior to food presentation: they produced shortening to anterior stimuli, swimming and crawling to posterior stimuli. However, when stimuli were delivered during the pre-feeding stage (probing and attachment), the animals did not perform any of these behaviors. The only behaviors elicited during pre-feeding were local bends, no response, or, in a few cases, a release and reattachment of the front sucker (these responses make up the ‘other’ category for Figs. 4 and 5). Some decrease in probability of shortening, crawling and swimming remained for 30 – 60 min after feeding terminated. Swimming was not elicited even 1 h after feeding. Similar experiments were performed on leeches that were fed on blood, their normal food, rather than the NaCl/arginine solution (Fig. 5). During pre-feeding, all whole-body behaviors (i.e. shortening, crawling and swimming) were suppressed and the suppression lingered for some time after feeding, although less so for the crawling behavior. All the swimming responses observed in the 30 min after feeding were due solely to one leech which had been swimming frequently before the feeding stimulus was introduced. These results are very similar to those of the previous experiment, indicating that the feeding behavior produced by the artificial NaCl/arginine was normal. While stimulus-elicited swimming and crawling were suppressed for 30– 60 min after feeding terminated, spontaneous crawling was observed intermittently during this period. Spontaneous swimming, however, was almost never seen during this period; it was observed only in the one leech from the blood-fed group which had been swimming frequently prior to feeding, and continued to show some spontaneous swimming after feeding.
4. Discussion This study demonstrates that feeding occupies a high position in the behavioral hierarchy of the medicinal leech: it dominated all the mechanically elicited behaviors that were tested. Feeding appears to be such a strong priority that a feeding leech will ignore powerful mechanical stimuli-which would normally cause vigorous escape responses like whole-body shortening or swimming-to continue feeding. This implies that the circuits that mediate the behaviors studied here must be strongly inhibited during feeding. Given the extensive understanding of at least some of these circuits, this
represents a ripe opportunity for investigations of the neural basis of behavioral choice. There do exist complexities in the interactions between these behaviors that are not always consistent with a simple ‘behavioral hierarchy’ model. While stimulus-elicited swimming is suppressed during feeding, feeding leeches sometimes make undulatory movements that resemble swimming [13]. Thus the interaction between these behaviors may be more complicated than a complete suppression of swimming. In a similar vein, stimulus-elicited swimming and crawling show a longterm suppression after feeding, but spontaneous crawling can be observed during this time. Spontaneous swimming was not observed after feeding, except in one leech, for the entire length of the experiment (1 h). Behavioral studies in other animals have often shown that feeding has a high priority, particularly in carnivores. Feeding suppresses most other behaviors, including mating, in Pleurobranchaea [4]. Similar findings have been obtained in other carnivorous molluscs: in Hermissenda, feeding dominates withdrawal reflexes and rolling-over behavior [22], and in Clione, feeding dominates escape and withdrawal responses [19,20]. Feeding tends to be less dominant in herbivores: mating dominates feeding in Helix [1], and feeding can occur along with mating and other behaviors in Aplysia [29]. Given the infrequent feeding opportunities of the leech, a carnivore that may go for weeks or months between meals, it is not surprising that feeding has such a high priority. It remains an open question, however, whether or not feeding dominates mating, a leech behavior that is difficult to observe [23]. Although not much is known about the neural basis of feeding in leeches, the role of serotonin in this behavior has been studied to some degree. Application of serotonin increases feeding tendencies, while pharmacologically lesioning serotonin-containing neurons decreases the tendency to feed [12]. Stimuli that elicit biting, such as touching a warm surface or feeding-inducing chemicals applied to the leech’s oral area, activate certain serotonin-containing neurons [9]; during the consummatory phase of feeding, the activity of these neurons decreases [28]. After feeding, serotonin is depleted from the nervous system [9]. These sorts of findings have led to the hypothesis that serotonin is effectively a feeding hormone: increases in its blood level induce food-seeking and feeding behaviors (i.e. the animal acts hungry), whereas feeding lowers the serotonin supply for many days and the animal ignores possible food sources (i.e. the animal appears sated). Our observation that swimming cannot be elicited for at least an hour after feeding is consistent with this idea: serotonin increases the tendency to swim [27], and leeches whose serotonin has been lowered pharmacologically will not swim [8,21]. It is unlikely, however, that decreasing serotonin levels could entirely explain
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
19
Fig. 4. Behavioral responses to an 8 V stimulus at different times relative to the feeding behavior (using the NaCl/arginine solution). Nine leeches were tested, using the anterior dorsal and posterior dorsal stimulus locations. The n values represent the number of trials performed at each timepoint. Variability in the value of n in different columns was because some leeches did not assume the both suckers attached position during all time periods. A. Responses to anterior dorsal stimuli; n =105. B. Responses to posterior ventral stimuli; n =103.
20
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
Fig. 5. Behavioral responses to an 8 V stimulus at different times relative to the feeding behavior (using cow blood). Five leeches were tested, using the anterior dorsal and posterior dorsal stimulus locations. A. Response to anterior dorsal stimuli; n =70. B. Response to posterior ventral stimuli; n=70.
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
the shutdown of other behaviors during feeding, because the suppression is observed immediately when the leech has begun to probe and attach to the feeding apparatus, prior to the onset of feeding itself, a time when the firing rate of serotonin-containing neurons can still be high [28]. It is of interest that the Leydig cells, whose activity suppresses the local bending response [17], are active during feeding [28]. This observation suggests that the Leydig neurons may help to turn off the responses to mechanical stimuli during feeding. With the establishment of a semi-intact leech preparation that will feed [28], this becomes a testable hypothesis.
Acknowledgements Supported by NIMH research grant MH43396 (W.B.K.) and NIH training grant GM08107 (B.K.S.). We thank R. Wilson for comments on portions of the manuscript.
References [1] S.A. Adamo, R. Chase, The interactions of courtship, feeding, and locomotion in the behavioral hierarchy of the snail Helix aspersa, Behav. Neural Biol. 55 (1991) 1–18. [2] A.P. Baader, W.B. Kristan Jr., Parallel pathways coordinate crawling in the medicinal leech, Hirudo medicinalis, J. Comp. Physiol. A 176 (1995) 715–726. [3] P.D. Brodfuehrer, E.A. Debski, B.A. O’Gara, W.O. Friesen, Neuronal control of leech swimming, J. Neurobiol. 27 (1995) 403 – 418. [4] W.J. Davis, G.J. Mpitsos, J.M. Pinneo, The behavioral hierarchy of the mollusk Pleurobranchaea. I. The dominant position of the feeding behavior, J. Comp. Physiol. 90 (1974) 207–224. [5] W.J. Davis, G.J. Mpitsos, J.M. Pinneo, The behavioral hierarchy of the mollusk Pleurobranchaea. II. Hormonal suppression of feeding associated with egg-laying, J. Comp. Physiol. 90 (1974) 225 – 243. [6] F.J. Eisenhart, T.W. Cacciatore, W.B. Kristan Jr., R.F. Wessel, Crawling in the medicinal leech is produced by a central pattern generator, Soc. Neurosci. Abs. 21 (1995) 149. [7] E.J. Elliott, Chemosensory stimuli in feeding behavior of the leech Hirudo medicinalis, J. Comp. Physiol. A 159 (1986) 391 – 401. [8] J.C. Glover, A.P. Kramer, Serotonin analog selectively ablates identified neurons in the leech embryo, Science 216 (1982) 317 – 319. [9] J.R. Groome, M. Clark, C.M. Lent, The behavioral state of satiation in the leech is regulated by body distention and mimicked by serotonin depletion, J. Exp. Biol. 182 (1993) 265 – 270.
.
21
[10] M.P. Kovac, W.J. Davis, Neural mechanism underlying behavioral choice in Pleurobranchaea, J. Neurophysiol. 43 (1980) 469 – 487. [11] W.B. Kristan Jr., S.J. McGirr, G.V. Simpson, Behavioural and mechanosensory neurone responses to skin stimulation in leeches, J. Exp. Biol. 96 (1982) 143 – 160. [12] C.M. Lent, M.H. Dickinson, Serotonin integrates the feeding behavior of the medicinal leech, J. Comp. Physiol. A 154 (1984) 457 – 471. [13] C.M. Lent, K.H. Fliegner, E. Freedman, M.H. Dickinson, Ingestive behavior and physiology of the medicinal leech, J. Exp. Biol. 137 (1988) 513 – 527. [14] C.M. Lent, D. Zundel, E. Freedman, J.R. Groome, Serotonin in the leech central nervous system: Anatomical correlates and behavioral effects, J Comp. Physiol. A 168 (1991) 191 –200. [15] S.R. Lockery, W.B. Kristan Jr., Distributed processing of sensory information in the leech. I. Input-output relations of the local bending reflex, J. Neurosci. 10 (1990) 1811 – 1815. [16] S.R. Lockery, W.B. Kristan Jr., Distributed processing of sensory information in the leech. II. Identification of interneurons contributing to the local bending reflex, J. Neurosci. 10 (1990) 1816 – 1829. [17] S.R. Lockery, W.B. Kristan Jr., Two forms of sensitization of the local bending reflex of the medicinal leech, J. Comp. Physiol. A 168 (1991) 165 – 177. [18] D.J. McFarland, R.M. Sibly, The behavioral final common path, Phil. Trans. R. Soc. B 270 (1975) 265 – 293. [19] T.P. Norekian, R.A. Satterlie, An identified cerebral interneuron initiates different elements of prey capture behavior in the pteropod mollusc, Clione limacina, Invert. Neurosci. 1 (1995) 235 – 248. [20] T.P. Norekian, R.A. Satterlie, Whole body withdrawal circuit and its involvement in the behavioral hierarchy of the mollusk Clione limacina, J. Neurophysiol. 75 (1996) 529 – 537. [21] B.A. O’Gara, H. Chae, L.B. Latham, W.O. Friesen, Modification of leech behavioral patterns by reserpine-induced amine depletion, J. Neurosci. 11 (1991) 96 – 110. [22] J.L. Ram, N. Gilbert, S. Waddell, M.A. Anderson, Singleness of action in the interactions of feeding with other behaviors in Hermissenda crassicornis, Behav. Neural Biol. 49 (1988) 97–111. [23] R.T. Sawyer, Leech Biology and Behaviour (3 volumes), Clarendon Press, Oxford, 1986. [24] B.K. Shaw, W.B. Kristan Jr., The whole body shortening reflex of the medicinal leech: Motor pattern, sensory basis, and interneuronal pathways, J. Comp. Physiol. A 177 (1995) 667–681. [25] N. Tinbergen, The Study of Instinct, Oxford University Press, Oxford, 1951. [26] F. Toates, Motivational Systems, Cambridge University Press, Cambridge, 1986. [27] A.L. Willard, Effects of serotonin on the generation of the motor program for swimming by the medicinal leech, J. Neurosci. 1 (1981) 936 – 944. [28] R.J. Wilson, W.B. Kristan Jr., A.L. Kleinhaus, An increase in activity of serotonergic Retzius neurones may not be necessary for the consummatory phase of feeding in the leech Hirudo medicinalis, J. Exp. Biol. 199 (1996) 1405 – 1414. [29] I. Ziv, S. Markovich, C. Lustig, A.J. Susswein, Effects of food and mates on time budget in Aplysia fasciata: Integration of feeding, reproduction and locomotion, Behav. Neural Biol. 55 (1991) 68 – 85.
Research report
Behavioral hierarchy in the medicinal leech, Hirudo medicinalis: feeding as a dominant behavior Lisa M. Misell, Brian K. Shaw, William B. Kristan Jr. * Department of Biology 0357, Uni6ersity of California, San Diego, La Jolla, CA 92093 -0357, USA Received 13 December 1996; received in revised form 25 April 1997; accepted 26 April 1997
Abstract The effect of feeding behavior on other behaviors (swimming, crawling and shortening) was investigated in the leech, Hirudo medicinalis. The stimulus locations and intensities required to produce mechanically elicited behaviors were first determined in the non-feeding leech. Stimuli were delivered while the leech was in various body positions to determine whether stimulus location affected behavioral response. Response thresholds were determined for the mechanically elicited behaviors. The same stimuli were then applied to feeding leeches to determine if response thresholds had changed. A solution with NaCl and arginine was used to elicit feeding. The same sets of stimuli were applied at intervals for an hour after feeding, to determine the duration of feeding-induced changes in behavior. Depending on the body position and stimulus location, stimuli produced different combinations of behaviors that included shortening, swimming and crawling. Anterior stimuli generally elicited shortening, whereas posterior stimuli generally elicited crawling and swimming, with swimming more likely to ventral stimulation than to dorsal stimulation. Having the front sucker attached changed these behavioral patterns. During feeding, the response thresholds changed dramatically, from 3–5 V to greater than 9 V. This increase in threshold began with the start of feeding, even before ingestion commenced. Suppression of the behaviors lasted up to 1 h after the end of feeding, with the effect on swimming being the most pronounced and longest lasting. © 1998 Elsevier Science B.V. Keywords: Behavioral hierarchy; Behavioral choice; Decision-making; Feeding; Medicinal leech
1. Introduction To behave adaptively, animals often give higher priority to some behaviors than to others. An early proposal was that components of behaviors are elicited by ‘sign stimuli’ and that responses to these stimuli are arranged in a hierarchical cascade [25]. Later authors showed that a strict hierarchical arrangement was too simple to explain all behavioral interactions [18]. For instance, one must also consider such ideas as ‘drive’ and ‘internal state’ to predict many behaviors [26]. A pioneering effort to track down neuronal mechanisms by which animals * Corresponding author. Tel: + 1 619 5344760; fax: + 1 619 5347309; e-mail [email protected] 0166-4328/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 6 - 4 3 2 8 ( 9 7 ) 0 0 0 7 2 - 7
choose among behaviors [4] concluded that the marine mollusc Pleurobranchaea showed a behavioral hierarchy: it fed rather than righting itself, mating, or withdrawing from a weak mechanical stimulus, and laid eggs rather than feeding. Further study showed that this hierarchy was flexible; for instance, feeding would drop well down the hierarchy in a well-fed animal [5]. Similar studies of behavioral hierarchies have since been carried out in a variety of molluscs, including Aplysia [29], Helix [1] and Hermissenda [22]. Ideally, this kind of information about behavior can provide a foundation for explorations of the neural underpinnings of behavioral choice. An example of this is the study of the neural mechanisms underlying the suppression of withdrawal during feeding in Pleurobranchaea [10].
14
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
In the present study, we concerned ourselves with aspects of the behavioral hierarchy of the medicinal leech, Hirudo medicinalis. In particular, we were interested in what priority feeding has relative to other behaviors. The medicinal leech is a bloodsucking carnivore which feeds infrequently on amphibians or mammals; a leech may go for weeks or months between blood meals [23]. Therefore, it would make sense that feeding would have a high priority for the leech: on those rare occasions when a meal was available, it might not want to be distracted. In line with this, leeches with a variety of incisions in their bodies will feed normally [12,28]. We compared feeding to mechanically elicited reflex or locomotor behaviors. Mechanical stimulation can elicit a range of such behaviors in leeches, including local bending, whole-body shortening, swimming, and crawling, depending upon the location and intensity of the stimulus [11]. The neural basis of these behaviors is understood to varying degrees. The circuits for local bending and swimming are quite well-characterized [15,16,3]; the circuit for the wholebody shortening reflex is partially understood [24]; and the outline of the circuit for crawling has been established [2,6]. These circuits provide a rich opportunity for studying the neural basis of behavioral choice and interactions between behaviors. In contrast to the other behaviors, very little is known at present about the neuronal basis of feeding, except for the possible role of modulatory neurons [14,28], although the establishment of a semi-intact preparation that feeds [28] has made such a characterization feasible. The experimental approach we employed was first to determine the stimulus locations and intensities that reliably produce the mechanically elicited behaviors of interest in leeches that were not feeding. We then applied the same stimuli to feeding leeches to determine if the response thresholds changed. We found a dramatic increase in the thresholds for the mechanically elicited behaviors during feeding; in fact, even stimuli that were far above the behavioral thresholds in nonfeeding leeches could not produce whole-body behaviors in feeding leeches. Thus it appears that feeding occupies a high position in the leech’s behavioral hierarchy.
2. Methods Experiments were performed from May to July on leeches that had not been fed a blood meal for at least 2 months. Animals were kept in artificial pond water (38 mg/liter Instant Ocean), at 15°C, with 12:12 h light:dark cycles. Experiments were performed in a 10 gallon aquarium which was filled with 3 – 4 cm of artificial pond water at an ambient temperature of 20 – 22°C. The leeches were acclimatized to room tem-
perature water for 30 min prior to starting an experiment. Several of the experiments were videotaped using a Sanyo VDC 3825 video camera and a Panasonic AG-1730 video cassette recorder for frame-by-frame analysis. The stimulus used to elicit behaviors was a train of electrical shocks delivered with a hand-held electrode. These electrical stimuli mimic mechanical stimuli—they activate the terminals of touch (T) and pressure (P) mechanoreceptors —but are more easily controlled [11]. The terminals of the electrode were 1 mm apart and were made from 0.008 in. diameter silver wire. They were glued to the end of a wooden dowel and fully insulated except for the terminal 1 mm. A Grass SD9 stimulator was used to deliver 1 msec pulses in trains at 10 Hz. The tips of the wires were placed lightly on the skin and at least 1 s was allowed to elapse before the electrical stimulus was delivered, to be sure that the animal was not responding to mechanical contact. The electrical stimulus was applied until the animal responded, which generally occurred within 1 s. In cases when no response was produced, the stimulus was withdrawn after approximately 2 s. To avoid habituation, trials were separated by at least 3 min.
2.1. Responses to stimuli at different body locations To establish a baseline for subsequent experiments, responses were scored to stimuli given at three different locations on the leech body. Stimuli of 8 V intensity, suprathreshold for eliciting behaviors [11], were delivered to the anterior dorsal, posterior ventral, and posterior dorsal surfaces of the leech (Fig. 1A). The anterior stimulus was applied to the midline of segment 2 or 3, and the posterior stimulus was delivered to midline of the last few segments, 18–21. These three stimulus locations were tested with the leech in three different body positions: with just the back sucker attached (back sucker attached), just the front sucker attached (front sucker attached), and with both suckers attached (both suckers attached). Each leech was stimulated approximately 15 times.
2.2. Measurement of response thresholds Thresholds for shortening, crawling and swimming were determined by delivering stimuli of varying intensity, ranging from 2–8 V in 1 V increments. Trials were performed in random order (a ‘trial’ is a single stimulus presentation for one animal). The threshold for shortening was defined as the weakest stimulus needed to cause an observable movement of the whole body. The threshold for crawling was defined as the stimulus needed to cause at least one step, consisting of the release of the back sucker, forward movement, and replacement of the back sucker. The threshold for
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
swimming was defined as the stimulus needed to produce the characteristic undulating swimming movements of the leech. For these experiments, only the anterior dorsal and posterior ventral stimulus sites were used, and the leech was stimulated only when in a both suckers attached body position. This was done for comparison with subsequent feeding trials when the leech would have both suckers attached to the feeding apparatus.
2.3. Response thresholds during feeding The feeding apparatus consisted of a plastic 50 ml centrifuge tube with a diameter of 3 cm, cut and smoothed so that the tube was open on both ends (Fig. 1B). One end was covered with sausage casing (processed bovine intestine) that had been rinsed thoroughly in distilled water. The casing was stretched over the open end of the tube and taped to the outside of it. The tube was suspended in the aquarium, with the covered end just below the surface of the water. To determine thresholds for swimming, shortening, and crawling while feeding, leeches were presented with
Fig. 1. Techniques used to deliver stimuli and induce feeding. A. A diagram of an adult medicinal leech, showing the three stimulus locations used in these experiments. Anterior dorsal stimuli were delivered on the dorsal midline of segment 2 or 3. Posterior stimuli were delivered on the midline of the dorsal or ventral surface of segments 18 – 20. B. The feeding apparatus used for feeding trials. A 50 ml polyethelene (3 cm diameter) centrifuge tube, with the closed end sawed off and smoothed, was covered by a stretched layer of sausage casing. The casing was taped to the side of the tube. The apparatus was suspended in a 10 gallon aquarium with the covered end just below the surface of the water.
15
a solution of 150 mM NaCl and 5 mM L-arginine, which has been shown to elicit feeding as effectively as does whole blood [7]. The solution was warmed to 38°C, to make it more likely to elicit feeding [7], and approximately 30 ml was added to the centrifuge tube. As soon as the leeches were within 1–2 cm of the feeding tube, they attended to it and began to probe the surface of the sausage casing with their anterior sucker. Typically, within seconds, the leeches would attach their front sucker to the sausage casing and bite through the surface. Both posterior ventral and anterior dorsal stimuli were delivered while the leech was actively feeding (pharyngeal peristalsis could be observed) with both suckers attached to the feeding apparatus.
2.4. Responses to suprathreshold stimuli before and after feeding Behavioral responses to stimuli were measured before and after feeding, using 8 V stimulus pulses. Before food was presented, each leech was tested once to be sure that it responded like the animals in the earlier experiments. A series of stages was operationally defined: pre-feeding, the period of probing and sucker attachment that occurred before the leech bit through the sausage casing; feeding, starting when rapid peristalsis of the anterior segments began; and post-feeding, beginning when the leech detached its front sucker from the feeding surface. Because leeches usually attach their posterior sucker as well as their anterior sucker while feeding, most stimulus trials were in the both suckers attached position. One group of leeches were tested using the previously described NaCl/arginine solution. A smaller group was tested with cow blood, to be sure that responses seen were due to feeding per se, and not due to receiving the NaCl/arginine solution instead of blood. The solutions were presented as previously described. In pre-feeding trials, stimuli were delivered while the leech was probing and attaching to the feeding apparatus. Leeches fed for approximately 20–25 min before detaching from the surface, similar to the feeding durations observed in other studies [13]. After feeding, stimuli were delivered immediately upon detachment and every 3 min thereafter. At each 3 min interval, both an anterior dorsal stimulus and a posterior ventral stimulus were given, with about a 30 second interval between them.
3. Results
3.1. Beha6ioral responses to skin stimulation Skin stimulation produced a range of behaviors in the leech (Fig. 2). Anterior stimuli most commonly
16
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
Fig. 2. Behavioral responses to an 8 V stimulus at the three different body locations, with the leeches in different body positions. The attachment positions were: back sucker attached (BSA), front sucker attached (FSA), and back and front sucker attached (BFSA). Ten leeches were tested approximately 15 times each, for a total of 156 trials. The n values listed on each graph represent the number of trials performed under each set of conditions. The values of n differ because some positions were more frequently assumed by the leeches than others. Each leech received one to five trials per body position, except for the front sucker attached position. Several leeches did not assume this position and thus did not have any front sucker attached trials. The most frequently assumed position was back sucker attached, and so this category includes the most trials. The most frequently observed behaviors were swimming, crawling, and shortening; behaviors in the ‘other’ category include local bending. ‘No resp.’ refers to no response observed. A. Responses to anterior dorsal stimuli. B. Responses to posterior dorsal stimuli. C. Responses to posterior ventral stimuli.
caused a whole-body shortening reflex (when the leech was in the back sucker attached or both suckers attached position): the leech contracted its body, releasing its front sucker if it had been attached and keeping the back sucker attached, pulling the body backward away from the stimulus. Posterior stimuli most often caused
swimming or a crawling step, which also served to get the animal away from the stimulus. The beginning of a crawling step sometimes resembled a whole-body shortening reflex when the front sucker was attached: the rear sucker was released (if it had been attached) and the animal shortened, pulling the body forward toward
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
17
Fig. 3. Response thresholds for shortening, crawling and swimming in non-feeding leeches. Seven leeches were tested, using both the anterior dorsal and posterior ventral stimulus locations. A. Thresholds for shortening to anterior dorsal stimuli. A total of 69 trials were performed. Each voltage was tested seven to ten times, in random order. B. Crawling and swimming thresholds to posterior ventral stimuli. A total of 63 trials were performed, with each voltage tested seven to 12 times.
the attached front sucker. These responses were counted as crawling if the rear sucker re-attached after the contraction and the front sucker subsequently released and the animal extended; otherwise, the responses were counted as shortenings. When the back sucker only was attached, crawling steps began with an extension, followed by an attachment of the front sucker. The ‘other’ category consisted largely of weak responses, such as local bending. Leeches were rarely found in the front sucker attached position—usually only when they were in the middle of a crawling step, with the posterior sucker moving forward to reattach — thus accounting for the lower numbers of front sucker attached trials for each stimulus site. In contrast to the results for the other body positions, leeches in the front sucker attached position usually didn’t shorten in response to an anterior stimulus, but showed a range of ‘other’ behaviors such as local bends, wriggling, and front-to-back extensions. This result highlights the importance of the leech’s body position and the state of its suckers in regulating the expression of its behavioral responses; it would make no sense for a leech to shorten to an anterior stimulus when its back sucker was unattached because it would pull itself forward, toward the stimulus. Posterior ventral trials were not performed in the front sucker attached position because the leech was usually moving its posterior end forward, making delivery of the stimulus difficult. The present results show some resemblances to those previously obtained in de-brained leeches [11]. In both cases, anterior stimuli most often caused shortening. However, in de-brained leeches, posterior dorsal stimuli usually caused swimming, whereas in the present study
they usually caused crawling (de-brained leeches generally do not crawl [2]). Posterior ventral stimuli, on the other hand, caused both swimming and crawling in this study. Because we were interested in the effects on both swimming and crawling behaviors, the posterior ventral site rather than the posterior dorsal was used in subsequent experiments.
3.2. Response thresholds in non-feeding and feeding leeches The thresholds for shortening, crawling and swimming in non-feeding leeches were roughly in the range of 3–5 V (Fig. 3), similar to those in de-brained leeches [11,24]. Behaviors in the ‘other’ category for these experiments were mostly local bends or no response. It was obvious from initial pilot experiments that the thresholds for shortening, swimming and crawling were higher during feeding. Thus, to illustrate the shift in thresholds, only voltages from 6–9 V were used in feeding experiments. Four leeches were tested (with the NaCl/arginine solution), using the anterior dorsal and posterior ventral stimulus locations. For the anterior dorsal site, 13 trials total were performed, with each voltage tested three to four times. For the posterior ventral site, 12 trials total were performed, with each voltage tested two to four times. The results were dramatic. During feeding, 6–9 V stimuli failed completely to elicit shortening, swimming, or crawling behaviors. One trial was performed with a stimulus of 14 V and this produced merely a local bend. The only reactions observed during these trials were small local bends at the stimulus site or no response at all. In particular, the stimuli did not cause a termination of feeding.
18
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
3.3. Responses to suprathreshold stimuli before and after feeding Different responses were observed in leeches to stimuli delivered before and after feeding (Fig. 4). The control trials demonstrate that the leeches behaved normally prior to food presentation: they produced shortening to anterior stimuli, swimming and crawling to posterior stimuli. However, when stimuli were delivered during the pre-feeding stage (probing and attachment), the animals did not perform any of these behaviors. The only behaviors elicited during pre-feeding were local bends, no response, or, in a few cases, a release and reattachment of the front sucker (these responses make up the ‘other’ category for Figs. 4 and 5). Some decrease in probability of shortening, crawling and swimming remained for 30 – 60 min after feeding terminated. Swimming was not elicited even 1 h after feeding. Similar experiments were performed on leeches that were fed on blood, their normal food, rather than the NaCl/arginine solution (Fig. 5). During pre-feeding, all whole-body behaviors (i.e. shortening, crawling and swimming) were suppressed and the suppression lingered for some time after feeding, although less so for the crawling behavior. All the swimming responses observed in the 30 min after feeding were due solely to one leech which had been swimming frequently before the feeding stimulus was introduced. These results are very similar to those of the previous experiment, indicating that the feeding behavior produced by the artificial NaCl/arginine was normal. While stimulus-elicited swimming and crawling were suppressed for 30– 60 min after feeding terminated, spontaneous crawling was observed intermittently during this period. Spontaneous swimming, however, was almost never seen during this period; it was observed only in the one leech from the blood-fed group which had been swimming frequently prior to feeding, and continued to show some spontaneous swimming after feeding.
4. Discussion This study demonstrates that feeding occupies a high position in the behavioral hierarchy of the medicinal leech: it dominated all the mechanically elicited behaviors that were tested. Feeding appears to be such a strong priority that a feeding leech will ignore powerful mechanical stimuli-which would normally cause vigorous escape responses like whole-body shortening or swimming-to continue feeding. This implies that the circuits that mediate the behaviors studied here must be strongly inhibited during feeding. Given the extensive understanding of at least some of these circuits, this
represents a ripe opportunity for investigations of the neural basis of behavioral choice. There do exist complexities in the interactions between these behaviors that are not always consistent with a simple ‘behavioral hierarchy’ model. While stimulus-elicited swimming is suppressed during feeding, feeding leeches sometimes make undulatory movements that resemble swimming [13]. Thus the interaction between these behaviors may be more complicated than a complete suppression of swimming. In a similar vein, stimulus-elicited swimming and crawling show a longterm suppression after feeding, but spontaneous crawling can be observed during this time. Spontaneous swimming was not observed after feeding, except in one leech, for the entire length of the experiment (1 h). Behavioral studies in other animals have often shown that feeding has a high priority, particularly in carnivores. Feeding suppresses most other behaviors, including mating, in Pleurobranchaea [4]. Similar findings have been obtained in other carnivorous molluscs: in Hermissenda, feeding dominates withdrawal reflexes and rolling-over behavior [22], and in Clione, feeding dominates escape and withdrawal responses [19,20]. Feeding tends to be less dominant in herbivores: mating dominates feeding in Helix [1], and feeding can occur along with mating and other behaviors in Aplysia [29]. Given the infrequent feeding opportunities of the leech, a carnivore that may go for weeks or months between meals, it is not surprising that feeding has such a high priority. It remains an open question, however, whether or not feeding dominates mating, a leech behavior that is difficult to observe [23]. Although not much is known about the neural basis of feeding in leeches, the role of serotonin in this behavior has been studied to some degree. Application of serotonin increases feeding tendencies, while pharmacologically lesioning serotonin-containing neurons decreases the tendency to feed [12]. Stimuli that elicit biting, such as touching a warm surface or feeding-inducing chemicals applied to the leech’s oral area, activate certain serotonin-containing neurons [9]; during the consummatory phase of feeding, the activity of these neurons decreases [28]. After feeding, serotonin is depleted from the nervous system [9]. These sorts of findings have led to the hypothesis that serotonin is effectively a feeding hormone: increases in its blood level induce food-seeking and feeding behaviors (i.e. the animal acts hungry), whereas feeding lowers the serotonin supply for many days and the animal ignores possible food sources (i.e. the animal appears sated). Our observation that swimming cannot be elicited for at least an hour after feeding is consistent with this idea: serotonin increases the tendency to swim [27], and leeches whose serotonin has been lowered pharmacologically will not swim [8,21]. It is unlikely, however, that decreasing serotonin levels could entirely explain
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
19
Fig. 4. Behavioral responses to an 8 V stimulus at different times relative to the feeding behavior (using the NaCl/arginine solution). Nine leeches were tested, using the anterior dorsal and posterior dorsal stimulus locations. The n values represent the number of trials performed at each timepoint. Variability in the value of n in different columns was because some leeches did not assume the both suckers attached position during all time periods. A. Responses to anterior dorsal stimuli; n =105. B. Responses to posterior ventral stimuli; n =103.
20
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
Fig. 5. Behavioral responses to an 8 V stimulus at different times relative to the feeding behavior (using cow blood). Five leeches were tested, using the anterior dorsal and posterior dorsal stimulus locations. A. Response to anterior dorsal stimuli; n =70. B. Response to posterior ventral stimuli; n=70.
L.M. Misell et al. / Beha6ioural Brain Research 90 (1998) 13–21
the shutdown of other behaviors during feeding, because the suppression is observed immediately when the leech has begun to probe and attach to the feeding apparatus, prior to the onset of feeding itself, a time when the firing rate of serotonin-containing neurons can still be high [28]. It is of interest that the Leydig cells, whose activity suppresses the local bending response [17], are active during feeding [28]. This observation suggests that the Leydig neurons may help to turn off the responses to mechanical stimuli during feeding. With the establishment of a semi-intact leech preparation that will feed [28], this becomes a testable hypothesis.
Acknowledgements Supported by NIMH research grant MH43396 (W.B.K.) and NIH training grant GM08107 (B.K.S.). We thank R. Wilson for comments on portions of the manuscript.
References [1] S.A. Adamo, R. Chase, The interactions of courtship, feeding, and locomotion in the behavioral hierarchy of the snail Helix aspersa, Behav. Neural Biol. 55 (1991) 1–18. [2] A.P. Baader, W.B. Kristan Jr., Parallel pathways coordinate crawling in the medicinal leech, Hirudo medicinalis, J. Comp. Physiol. A 176 (1995) 715–726. [3] P.D. Brodfuehrer, E.A. Debski, B.A. O’Gara, W.O. Friesen, Neuronal control of leech swimming, J. Neurobiol. 27 (1995) 403 – 418. [4] W.J. Davis, G.J. Mpitsos, J.M. Pinneo, The behavioral hierarchy of the mollusk Pleurobranchaea. I. The dominant position of the feeding behavior, J. Comp. Physiol. 90 (1974) 207–224. [5] W.J. Davis, G.J. Mpitsos, J.M. Pinneo, The behavioral hierarchy of the mollusk Pleurobranchaea. II. Hormonal suppression of feeding associated with egg-laying, J. Comp. Physiol. 90 (1974) 225 – 243. [6] F.J. Eisenhart, T.W. Cacciatore, W.B. Kristan Jr., R.F. Wessel, Crawling in the medicinal leech is produced by a central pattern generator, Soc. Neurosci. Abs. 21 (1995) 149. [7] E.J. Elliott, Chemosensory stimuli in feeding behavior of the leech Hirudo medicinalis, J. Comp. Physiol. A 159 (1986) 391 – 401. [8] J.C. Glover, A.P. Kramer, Serotonin analog selectively ablates identified neurons in the leech embryo, Science 216 (1982) 317 – 319. [9] J.R. Groome, M. Clark, C.M. Lent, The behavioral state of satiation in the leech is regulated by body distention and mimicked by serotonin depletion, J. Exp. Biol. 182 (1993) 265 – 270.
.
21
[10] M.P. Kovac, W.J. Davis, Neural mechanism underlying behavioral choice in Pleurobranchaea, J. Neurophysiol. 43 (1980) 469 – 487. [11] W.B. Kristan Jr., S.J. McGirr, G.V. Simpson, Behavioural and mechanosensory neurone responses to skin stimulation in leeches, J. Exp. Biol. 96 (1982) 143 – 160. [12] C.M. Lent, M.H. Dickinson, Serotonin integrates the feeding behavior of the medicinal leech, J. Comp. Physiol. A 154 (1984) 457 – 471. [13] C.M. Lent, K.H. Fliegner, E. Freedman, M.H. Dickinson, Ingestive behavior and physiology of the medicinal leech, J. Exp. Biol. 137 (1988) 513 – 527. [14] C.M. Lent, D. Zundel, E. Freedman, J.R. Groome, Serotonin in the leech central nervous system: Anatomical correlates and behavioral effects, J Comp. Physiol. A 168 (1991) 191 –200. [15] S.R. Lockery, W.B. Kristan Jr., Distributed processing of sensory information in the leech. I. Input-output relations of the local bending reflex, J. Neurosci. 10 (1990) 1811 – 1815. [16] S.R. Lockery, W.B. Kristan Jr., Distributed processing of sensory information in the leech. II. Identification of interneurons contributing to the local bending reflex, J. Neurosci. 10 (1990) 1816 – 1829. [17] S.R. Lockery, W.B. Kristan Jr., Two forms of sensitization of the local bending reflex of the medicinal leech, J. Comp. Physiol. A 168 (1991) 165 – 177. [18] D.J. McFarland, R.M. Sibly, The behavioral final common path, Phil. Trans. R. Soc. B 270 (1975) 265 – 293. [19] T.P. Norekian, R.A. Satterlie, An identified cerebral interneuron initiates different elements of prey capture behavior in the pteropod mollusc, Clione limacina, Invert. Neurosci. 1 (1995) 235 – 248. [20] T.P. Norekian, R.A. Satterlie, Whole body withdrawal circuit and its involvement in the behavioral hierarchy of the mollusk Clione limacina, J. Neurophysiol. 75 (1996) 529 – 537. [21] B.A. O’Gara, H. Chae, L.B. Latham, W.O. Friesen, Modification of leech behavioral patterns by reserpine-induced amine depletion, J. Neurosci. 11 (1991) 96 – 110. [22] J.L. Ram, N. Gilbert, S. Waddell, M.A. Anderson, Singleness of action in the interactions of feeding with other behaviors in Hermissenda crassicornis, Behav. Neural Biol. 49 (1988) 97–111. [23] R.T. Sawyer, Leech Biology and Behaviour (3 volumes), Clarendon Press, Oxford, 1986. [24] B.K. Shaw, W.B. Kristan Jr., The whole body shortening reflex of the medicinal leech: Motor pattern, sensory basis, and interneuronal pathways, J. Comp. Physiol. A 177 (1995) 667–681. [25] N. Tinbergen, The Study of Instinct, Oxford University Press, Oxford, 1951. [26] F. Toates, Motivational Systems, Cambridge University Press, Cambridge, 1986. [27] A.L. Willard, Effects of serotonin on the generation of the motor program for swimming by the medicinal leech, J. Neurosci. 1 (1981) 936 – 944. [28] R.J. Wilson, W.B. Kristan Jr., A.L. Kleinhaus, An increase in activity of serotonergic Retzius neurones may not be necessary for the consummatory phase of feeding in the leech Hirudo medicinalis, J. Exp. Biol. 199 (1996) 1405 – 1414. [29] I. Ziv, S. Markovich, C. Lustig, A.J. Susswein, Effects of food and mates on time budget in Aplysia fasciata: Integration of feeding, reproduction and locomotion, Behav. Neural Biol. 55 (1991) 68 – 85.