Thermal and chemical stimuli known to promote ingestive behaviours in the medicinal leech Hirudo medicinalis were tested for their physiological effects on Retzius neurones and for their biochemical effects on serotonin levels in the central nervous system, pharynx and body wall. Retzius neurones throughout the leech nerve cord receive excitatory synaptic input during thermal or chemical stimulation of the prostomial lip. These neurones respond to the rate of change of temperature as well as to absolute temperature at the lip. Exposure of the lip to sodium chloride excites Retzius neurones, whereas exposure to arginine has little effect. Thermal stimulation of the lip elicits a more rapid but less prolonged excitation of Retzius neurones than does chemical stimulation.

Stimulation of the prostomial lip is associated with afferent activities in the cephalic nerves D1, D2 and V1–2. Thermal stimulation of the prostomial lip results in depletion of serotonin from midbody ganglia, whereas chemical stimulation has no effect. Conversely, chemical stimulation of the lip results in depletion of serotonin from the body wall, whereas thermal stimulation does not. Pharyngeal serotonin content is decreased with either modality. These data distinguish two important feeding-related sensory input pathways to central serotonergic effector neurones in Hirudo medicinalis.

The medicinal leech Hirudo medicinalis expresses feeding-related behaviours alternating between hunger and satiation with an annual cycle (Lent, 1985; Groome et al. 1993). Neuronal serotonin orchestrates a number of behaviours in this organism, including specific appetitive and consummatory elements of the behavioural repertoire of ingestion. This monoamine is sequestered by several interneurones of the leech central nervous system (CNS) as well as by the Retzius (RZ) and large lateral (LL) effector neurones, whose serotonergic arborizations are prominent in both pharyngeal and body wall musculatures (Lent and Dickinson, 1984; Leake et al. 1985).

The consummatory phase of the ingestive process is correlated with decreased serotonin levels in the leech CNS (Lent et al. 1991). Depletion of ganglionic serotonin following neurotoxic lesion with 5,7-dihydroxytryptamine is correlated with alterations in the feeding-related behavioural state (Groome et al. 1993). Exogenous application of serotonin, or the stimulation of serotonergic effector neurones, activates several peripheral tissues involved in feeding (Lent and Dickinson, 1984; Marshall and Lent, 1988). These findings suggest that effector neurones release serotonin within the CNS, or in peripheral tissues, during the course of ingestion.

Leech ingestive behaviour is expressed through specific sensory modalities. Appetitive behaviours, including prey detection, are dependent upon photic, mechanical or chemical stimuli (Mann, 1962; Dickinson and Lent, 1984; Sawyer, 1986). Local warming of the lip, which has been shown to excite RZ and LL neurones of rostral ganglia (Lent and Dickinson, 1984), or the presentation of chemical cues (Elliot, 1986) constitute two such feeding-related sensory cues. The sensory input pathways that initiate and orchestrate feeding in H. medicinalis have not been characterized. We decided to determine the relative effects of these two feeding-related stimuli upon the excitability of serotonergic RZ neurones of the leech CNS and upon levels of serotonin. An abstract of parts of this work has been reported elsewhere (Groome and Lent, 1991).

Physiology

Leeches (Hirudo medicinalis L., Leeches USA, Westbury, NY, USA) were maintained in glass aquaria in artifical pond water (Muller et al. 1981) at room temperature (20–23 °C). Leeches were chilled and dissected in cold leech saline (115 mmol l-1 NaCl, 4 mmol l-1 KCl, 2 mmol l-1 CaCl2, 10 mmol l-1 Hepes buffer, pH 7.4) leaving intact the CNS, cephalic nerves and prostomial lip. The lip and ventral nerve cord were pinned to the bottom of a 5 ml resin-lined recording chamber and separated with a Vaseline bridge (Fig. 1).

Fig. 1.

Diagram of the leech prostomial lip and central nervous system (CNS) preparation. The CNS is enlarged here for clarity. Intracellular recordings were made from RZ neurones throughout the ventral nerve cord, and extracellular recordings were made from the cephalic nerves D1, D2 and the paired V1–2 during responses to thermal or chemical stimulation of the lip. In some preparations, recordings were made from the isolated lip and cephalic nerves.

Fig. 1.

Diagram of the leech prostomial lip and central nervous system (CNS) preparation. The CNS is enlarged here for clarity. Intracellular recordings were made from RZ neurones throughout the ventral nerve cord, and extracellular recordings were made from the cephalic nerves D1, D2 and the paired V1–2 during responses to thermal or chemical stimulation of the lip. In some preparations, recordings were made from the isolated lip and cephalic nerves.

Glass microelectrodes containing 4 mol l-1 potassium acetate were inserted into midbody (M) RZ neurones. Membrane potentials were monitored with Getting model 5 preamplifiers and bridge circuits (Iowa City, IA). Spike activities in the cephalic nerves D1, D2 and V1–2 were monitored with suction electrodes placed en passant or onto the cut ends of these nerves. Extracellular activities were monitored with Grass P-15 preamplifiers (Grass Instruments, Quincy, MA). Signals were recorded onto magnetic tape (Toshiba 200 T, A. R. Vetterer, Rebersburg, PA), and analyzed following A/D conversion. Alterations in RZ neurone spike frequency were calculated by counting spikes for 1 min immediately prior to stimulation of the prostomial lip and comparing this value with the spike count for 1 min during peak excitation. Statistical analyses were performed using Student’s t-tests.

Thermal stimulation of the lip was accomplished by passing direct current through a 2.2 kil resistor covered with heat-shrink tubing and positioned next to the dorsal lip of the preparation. Temperature increases at the lip were monitored continuously with a thermistor. A mixture of sodium chloride (NaCl, 150 mmol l-1) and arginine (Arg, 1 mmol l-1) at pH 7.4 was used for chemical stimulation of the prostomial lip. These two chemicals are necessary and sufficient to promote ingestion in intact H. medicinalis (Elliot, 1986). The prostomial lip was superfused with artificial pond water, or leech saline with a reduced NaCl concentration (50 mmol l-1), for 1–2 h prior to exposure to NaCl plus Arg. The lip was located 10 mm from an inlet port through which superfusion, at 5 ml min-1, was continuous to minimize sudden mechanical stimulation of the lip. The osmolalities of control and test solutions at the lip were monitored using a Wescor 5500 vapour pressure osmometer (Logan, UT). With this technique, the rates of perturbations of temperature or osmolality at the lip were similar.

Biochemistry

Leeches were matched both for mass (within 0.02 g) and for the frequency with which they bit a warm (37 °C) surface. This procedure minimized the observed range of variation in ganglionic serotonin levels between individual leeches (Lent et al. 1991). Leeches were chilled in artificial pond water at 4 °C and pinned to the bottom of a wax platform. The prostomial lip of each leech was bathed with pond water at room temperature for 1–2 h. Leeches were then subjected to thermal stimulation, chemical stimulation or sham treatment.

The thermal stimulation paradigm consisted of a 15 °C increase in temperature at the lip, with a maintained lip temperature of 37 °C for 30 min. The sham treatment consisted of placing the resistor near the dorsal lip without passage of current. Chemical stimulation consisted of three 10 min treatments with NaCl and Arg at room temperature, each followed by 10 min of superfusion with pond water. The sham treatment consisted of superfusing the prostomial lip with pond water.

Immediately after the last stimulation period, each leech was immobilized with isotonic MgCl2 and dissected in cold leech saline containing 20 mmol l-1 MgCl2, in order to minimize synaptic release of serotonin (Muller et al. 1981).

Six samples were taken from the CNS: the cerebral ganglia (containing the supraoesophageal ganglion and the fused rostral ganglia R1–R4); the midbody segmental ganglia M2–M4, M7–M10, M12–M15 and M17–M20; and the caudal ganglia C1–C7. In addition, the pharynx and a segment of the ventrolateral body wall within the segmental region of M10–M12 were dissected, briefly blotted and weighed. All tissues were placed in 200 μl of cold 0.1 mol l-1 HCl and then kept at -20 °C overnight.

Serotonin levels were measured by high pressure liquid chromatography with electrochemical detection (HPLC-EC). Acid extracts were injected onto a 5 μ,m C18 reverse-phase HPLC column (Phenomenex, Ranchos Palos Verdes, CA). The mobile phase (HPLC buffer) contained 0.1 mol l-1 monosodium phosphate, 0.13 mmol l-1 octyl sodium sulphate, 0.1 mmol l-1 EDTA and 10 % methanol, at pH 2.6. Serotonin was detected (Bioanalytical Systems, Indianapolis, IN), and sample analyses were performed as described by Lent et al. (1991).

Effects of lip stimulation on Retzius neurones of the leech central nervous system

Thermal stimulation of the prostomial lip elicited a depolarization of midbody RZ neurones concomitant with an increase in spike frequency (Fig. 2A), as has been reported previously for RZ neurones of the rostral ganglia (Lent and Dickinson, 1984). Chemical stimulation of the prostomial lip also resulted in depolarization of these neurones and increased action potential frequency (Fig. 2B). Spike frequencies were increased significantly, and to an equivalent degree, in anterior and posterior ganglia, after stimulation of the lip with either of these sensory cues (Fig. 3).

Fig. 2.

Stimulation of the prostomial lip provides excitatory input to RZ neurones. (A) In this recording, RZ neurones of segmental ganglia M8 and M13 respond with an intense barrage of impulses during the initial change in temperature and with bursts of lesser intensity as the lip temperature is further increased over a stimulus period of 1 min. (B) A 2 min exposure of the lip to a mixture of 150 mmol l-1 NaCl and 1 mmol l-1 arginine elicits common, excitatory input to RZ neurones of both anterior (M4) and posterior (M18) segmental ganglia.

Fig. 2.

Stimulation of the prostomial lip provides excitatory input to RZ neurones. (A) In this recording, RZ neurones of segmental ganglia M8 and M13 respond with an intense barrage of impulses during the initial change in temperature and with bursts of lesser intensity as the lip temperature is further increased over a stimulus period of 1 min. (B) A 2 min exposure of the lip to a mixture of 150 mmol l-1 NaCl and 1 mmol l-1 arginine elicits common, excitatory input to RZ neurones of both anterior (M4) and posterior (M18) segmental ganglia.

Fig. 3.

Distribution of excitatory effects of thermal (A) and chemical (B) sensory inputs to RZ neurones. Significant increases in spike frequency were produced by either type of input, in each region of the leech CNS examined. Bars indicate RZ neurone spike frequencies (+ S.E.M., N=9) calculated for 1 min immediately prior to stimulation of the prostomial lip (spontaneous), and for 1 min at the time of peak response (thermal or chemical input). Levels of significant difference are indicated for stimulation treatments in each region.

Fig. 3.

Distribution of excitatory effects of thermal (A) and chemical (B) sensory inputs to RZ neurones. Significant increases in spike frequency were produced by either type of input, in each region of the leech CNS examined. Bars indicate RZ neurone spike frequencies (+ S.E.M., N=9) calculated for 1 min immediately prior to stimulation of the prostomial lip (spontaneous), and for 1 min at the time of peak response (thermal or chemical input). Levels of significant difference are indicated for stimulation treatments in each region.

Increases in RZ neurone spike frequency following thermal or chemical stimulation of the lip outlasted the stimulus period (Fig. 2). However, the time course of the excitatory response elicited by thermal stimulation of the lip differed from that elicited by chemical stimulation. Thermal stimulation typically resulted in a more rapid excitation of RZ neurones than did chemical stimulation. During the first 20–60 s, thermal stimulation of the lip produced a significantly greater effect on the spike frequency than did chemical stimulation (P⩽0.05). However, chemical stimulation elicited a more prolonged excitation of the RZ neurones than did thermal input, as shown by a comparison of spike frequency increases for 1–3 min after cessation of lip stimulation (P⩽0.05).

Characteristics of thermal and chemical excitation of Retzius neurones

The response of RZ neurones to thermal stimulation of the prostomial lip depended on both the rate of increase of lip temperature and the absolute lip temperature (Fig. 4). Equivalent 10 °C increases at the lip most often caused significant increases in RZ neurone spike frequency when the rate of temperature increase was 0.25 °C s-1 or greater. Typically, RZ neurones responded to such thermal input with a prolonged burst of action potentials during the initial increase in temperature and with an elevated spiking frequency and shorter bursts when the temperature at the lip was maintained at 30–35 °C.

Fig. 4.

Time course of the response of RZ neurones to thermal (A) or chemical (B) stimulation of the prostomial lip. The mean RZ neurone spike frequencies (+ S.E.M.) are shown for 19 experiments with thermal stimulation of the lip (A) and for 20 experiments with chemical stimulation of the lip (B). Stimulus intensities are shown by dotted lines. The stimulation period lasted from time 0 to 120 s.

Fig. 4.

Time course of the response of RZ neurones to thermal (A) or chemical (B) stimulation of the prostomial lip. The mean RZ neurone spike frequencies (+ S.E.M.) are shown for 19 experiments with thermal stimulation of the lip (A) and for 20 experiments with chemical stimulation of the lip (B). Stimulus intensities are shown by dotted lines. The stimulation period lasted from time 0 to 120 s.

We examined the central effects of NaCl or Arg applied at the prostomial lip (Fig. 5). In ten preparations, NaCl (150 mmol l-1) with Arg (1 mmol l-1) produced an increase in RZ neurone spike frequency 8.42±3.08 times that of the baseline (P⩽0.0001). NaCl alone elicited an increase 8.94±2.36 times that of the baseline frequency (P⩽0.0001). However, exposure of the lip to Arg alone (in pond water or in isotonic sucrose) had little effect on the activity of RZ neurones (1.14±0.44 times baseline).

Fig. 5.

Excitatory input to RZ neurones during chemical stimulation of the prostomial lip is dependent upon salt detection. Increased RZ neurone spike frequency in response to application of 150 mmol l-1 NaCl and 1 mmol l-1 arginine to the lip (A) is mimicked by NaCl alone (C), but arginine alone has little effect (B). Arrows indicate times of addition and removal of stimulus.

Fig. 5.

Excitatory input to RZ neurones during chemical stimulation of the prostomial lip is dependent upon salt detection. Increased RZ neurone spike frequency in response to application of 150 mmol l-1 NaCl and 1 mmol l-1 arginine to the lip (A) is mimicked by NaCl alone (C), but arginine alone has little effect (B). Arrows indicate times of addition and removal of stimulus.

Cephalic nerve activities during lip stimulation

Thermal and chemoreceptive elements of the leech prostomial lip communicate with the CNS, presumably via the cephalic nerves (Sawyer, 1986). Two of these nerves, D1 and D2, exit the supraoesophageal ganglion and innervate the dorsal lip while two other nerves exit the rostral ganglia as the paired connective V1–2 and innervate both dorsal and ventral regions of the lip (E. J. Elliot, personal communication). We monitored the activities of these nerves during excitation of RZ neurones induced by lip stimulation using thermal or chemical cues. In some experiments, we recorded from the cut ends of the cephalic nerves attached to the prostomial lip region to distinguish afferent from efferent activities.

We observed increased activity in several units in recordings from the cephalic nerves in response to thermal (Fig. 6) or chemical (Fig. 7) stimulation of the lip. After examination of the increased cephalic nerve activity in response to either sensory cue, units originating from the mechanosensory (T) neurones were clearly recognizable. These units were identical to those elicited with tactile stimulation of the lip (data not shown). Impulses of these and other, unidentified, units were typically observed in bursts which were coincident for each of the nerves in some, but not all, cases. Typically, activity in RZ neurones was increased during or just after maximal activity in the cephalic nerves. However, a 1:1 correspondence between activation of any of the units in these nerves and excitation of RZ neurones was not apparent. It was not unusual to observe bursts in the cephalic nerves and in RZ neurones separated by several seconds.

Fig. 6.

(A) Activities in supraoesophageal (D1, D2) as well as rostral (V1–2) cephalic nerves increase in response to thermal stimulation of the lip. Multiple-unit bursts in these nerves are observed prior to the onset of excitation of RZ neurones in segmental ganglion M8. (B) Increased afferent activities of cephalic nerves in the isolated lip preparation of Hirudo medicinalis in response to thermal stimulation.

Fig. 6.

(A) Activities in supraoesophageal (D1, D2) as well as rostral (V1–2) cephalic nerves increase in response to thermal stimulation of the lip. Multiple-unit bursts in these nerves are observed prior to the onset of excitation of RZ neurones in segmental ganglion M8. (B) Increased afferent activities of cephalic nerves in the isolated lip preparation of Hirudo medicinalis in response to thermal stimulation.

Fig. 7.

(A) Activities in each of the cephalic nerves and in RZ neurones of segmental ganglion M7 are increased after 2 min of superfusion of the lip with 150 mmol l-1 NaCl and 1 mmol l-1 arginine. (B) Afferent activities in each of the cephalic nerves in the isolated lip preparation are increased following chemical stimulation.

Fig. 7.

(A) Activities in each of the cephalic nerves and in RZ neurones of segmental ganglion M7 are increased after 2 min of superfusion of the lip with 150 mmol l-1 NaCl and 1 mmol l-1 arginine. (B) Afferent activities in each of the cephalic nerves in the isolated lip preparation are increased following chemical stimulation.

In isolated lip experiments, spontaneous activities in each of these nerves were reduced slightly. Increased spike frequencies and burst activities were observed in D1, D2 and V1–2 during thermal stimulation of the lip (Fig. 6B). Chemical stimulation of the lip also markedly enhanced the afferent activities of each of these nerves (Fig. 7B). Once again, mechanosensory units were clearly evident in the responses of each of the nerves to thermal or chemical stimulation of the lip, along with unidentified units.

Biochemistry

Leeches were subjected to thermal or chemical stimulation of the prostomial lip and the resulting alterations of serotonin levels in the CNS and peripheral tissues were quantified with HPLC-EC. Thermal stimulation of the lip produced significant depletion of serotonin content in all regions of the leech CNS with the exception of the caudal ganglia (Fig. 8). Depletion of serotonin were 2.41 pmol ganglion-1 (15.6 %) in cerebral ganglia, 3.90 pmol ganglion-1 (20.2 %) in M2–M4, 3.42 pmol ganglion-1 (20.3 %) in M7–M10, 3.74 pmol ganglion-1 (22.9 %) in M12–M15 and 2.94 pmol ganglion-1 (24.5 %) in M17–M20. This pattern is similar to the depletion observed following ingestion of warm blood (Lent et al. 1991).

Fig. 8.

Thermal stimulation of the lip depletes central stores of serotonin (5-HT) in Hirudo medicinalis. (A) Electrochromatograms of standard serotonin (top) and of extracts of cerebral ganglia from control (middle) and stimulated (bottom) leeches. (B) Regional distribution of the effect of thermal stimulation of the lip on levels of CNS serotonin. Values represent the mean + S.E.M. from 10 experiments. Levels of significant difference from control are given above each histogram.

Fig. 8.

Thermal stimulation of the lip depletes central stores of serotonin (5-HT) in Hirudo medicinalis. (A) Electrochromatograms of standard serotonin (top) and of extracts of cerebral ganglia from control (middle) and stimulated (bottom) leeches. (B) Regional distribution of the effect of thermal stimulation of the lip on levels of CNS serotonin. Values represent the mean + S.E.M. from 10 experiments. Levels of significant difference from control are given above each histogram.

Chemical stimulation of the lip did not promote depletion of serotonin content from cerebral, midbody or caudal ganglia (Fig. 9). Three of the leeches exposed to a chemical stimulus ingested a volume of NaCl and Arg solution sufficient to increase body mass approximately threefold. We compared the results from HPLC-EC in these and the four leeches that did not ingest and found no alteration of serotonin levels in either group. Therefore, central stores of serotonin are depleted by the activation of thermal input pathways, but not by the activation of chemical input pathways, nor apparently by ingestion per se.

Fig. 9.

Chemical stimulation of the lip does not deplete central stores of serotonin (5-HT) in Hirudo medicinalis. (A) Electrochromatograms of standard serotonin (top) and of extracts of segmental ganglia M2–M4 from control (middle) and stimulated (bottom) leeches. (B) Chemical stimulation of the lip is without significant effect on serotonin levels in any region of the leech CNS. Values represent the mean + S.E.M. from seven experiments.

Fig. 9.

Chemical stimulation of the lip does not deplete central stores of serotonin (5-HT) in Hirudo medicinalis. (A) Electrochromatograms of standard serotonin (top) and of extracts of segmental ganglia M2–M4 from control (middle) and stimulated (bottom) leeches. (B) Chemical stimulation of the lip is without significant effect on serotonin levels in any region of the leech CNS. Values represent the mean + S.E.M. from seven experiments.

We also examined the effect of feeding-related stimuli on the serotonin content of the peripheral effector organs of feeding. Thermal stimulation of the prostomial lip decreased levels of pharyngeal serotonin by 18.5 %, from 14.1±1.6 to 11.5±1.6 pmol mg-1 tissue (N=10, P⩽0.05). Chemical stimulation decreased pharyngeal serotonin by 14.3 %, from 9.0±1.0 to 7.7±1.3 pmol mg-1 tissue (N=7, P⩽0.05). In these experiments, levels of serotonin in the body wall were unaffected by thermal stimulation (from 5.0±0.8 to 4.8±0.8 pmol mg-1 tissue). In contrast, body wall serotonin levels were reduced by 38.1 % from 3.9±0.3 to 2.4±0.3 pmol mg-1 tissue (P:s;0.02) following activation of chemical input pathways to the CNS.

We have begun to characterize two sensory pathways to the feeding effector RZ neurones of the leech CNS. Thermal stimulation of the prostomial lip, or exposure to NaCl and Arg, share the common effects of activation of the cephalic nerves and long-lasting excitation of midbody RZ neurones. However, these two sensory pathways are distinguished by their relative time courses of neuronal excitation, as well as by their effects on serotonin content in the leech CNS and periphery. We hypothesize that several behaviours in the process of leech ingestion are facilitated by these excitatory inputs to the serotonergic RZ neurones.

Thermosensory pathway

Thermal cues have been shown to play a role in the expression of feeding behaviour by hungry Hirudo medicinalis. Leeches bite a warm surface with maximum frequency as that surface approaches mammalian skin temperature (Dickinson and Lent, 1984). Thermal stimulation of the prostomial lip in semi-intact preparations is capable of eliciting several consummatory elements of ingestive behaviour, including jaw movements, pharyngeal contractions and salivary secretion (Lent and Dickinson, 1984). Although leeches do not ingest without chemical stimuli (Elliot, 1986), thermal stimulation during chemosensory-dependent ingestion increases pharyngeal contraction rate (J. R. Groome, unpublished observations). Rapid temperature changes, such as those expected to be encountered by the lip of a leech attached by its anterior sucker to the skin of a mammal, most effectively promote burst activity in RZ neurones. Taken together, these observations suggest that the thermal pathway might facilitate the initial phases of ingestion, with lesser involvement in the maintenance of ingestive behaviour over a period of 30 min.

Chemosensory pathway

The behavioural study of Elliot (1986) indicates that, of the constituents of blood, only two compounds, NaCl and Arg, are essential elements in the ingestive process. Exclusion of either compound from a feeding solution prohibits ingestion. Two types of chemosensory sensilla have been found within the dorsal prostomial lip (Elliot, 1987), and behavioural analyses of leech feeding have suggested an interactive process underlying the ingestive requirement for both salt and amino acid (Elliot, 1986). We have demonstrated that one central target of feeding-related chemoreceptive input in Hirudo medicinalis is the serotonergic RZ effector neurone. Our data indicate that RZ neurones of midbody ganglia are sensitive to the presence of salt at the lip. Responses from the cephalic nerves during local, chemosensory stimulation of the isolated lip are similarly dependent on the detection of salt (E. J. Elliot, personal communication).

The central targets of arginine as a feeding stimulus remain unknown. We found that midbody RZ neurones are excited by NaCl, a stimulus which by itself is insufficient to promote ingestion in intact leeches. It should be pointed out that our findings imply that a cation-dependent response of RZ neurones to application of the amino acid at the lip cannot be discounted. Alternatively, arginine may have effects on rostral RZ neurones or on other feeding-related neurones. Clearly, a more detailed investigation of the effect of arginine on receptors of the dorsal lip is needed before the central role of this amino acid during leech ingestion may be determined at the cellular level.

Serotonin depletion

Central and pharyngeal stores of serotonin were depleted after thermal input, in the absence of ingestion. Although the magnitude of depletion from the CNS was less than that reported earlier for leeches that had fed on warm mammalian blood (Lent et al. 1991), these data indicate that thermal cues are an important component of feeding-related depletion of ganglionic serotonin. In contrast, exposure of the lip to NaCl and Arg did not cause depletion of serotonin in any ganglionic sample. In addition, the process of ingestion itself does not appear to cause depletion of serotonin from the leech CNS. A physiological role for central depletion of serotonin in the overall repertoire of ingestive behaviour remains to be determined. However, depolarization of RZ neurones is correlated with the release of ganglionic serotonin (Glover and Lent, 1991), and impulse activity in RZ neurones does release sufficient serotonin into the vascular sinus to elicit swimming behaviour (Willard, 1981).

Our data indicate that chemosensory stimulation of the lip, as would occur during ingestion of blood, provides a long-lasting excitation of the serotonergic RZ neurones and depletes stores of body wall serotonin. The body wall is the major reservoir of serotonin in the leech (Lent et al. 1991). Our finding that chemical, but not thermal, stimulation of the lip caused depletion of serotonin within this tissue may be related to the prolonged period of RZ neurone excitation with chemical input. These findings suggest that one role of prolonged chemosensory input to the CNS may be the gradual relaxation of body wall musculature to permit the prominent distension characteristic of a full blood meal (Lent et al. 1988). Consistent with this hypothesis is the finding that electrical stimulation of RZ neurones decreases baseline tension of leech body wall muscles (Mason et al. 1979). A similar role for serotonin as the plasticizing agent in feeding-related distension has been proposed in the blood-sucking insect Rhodnius prolixus (Orchard et al. 1988; Lange et al. 1989). Chemosensory excitation of RZ neurones in Hirudo medicinalis may thus be essential for ingestion of a large, distensive blood meal.

Possible sensory pathways to RZ neurones

In our initial characterization of the feeding-related input pathways to the serotonergic RZ neurones, we noted similarites between the thermal and chemical pathways. It is apparent that mechanosensory afferents are active during the response of the lip to either of these cues (Figs 6, 7 and E. J. Elliot, personal communication). Retzius neurones throughout the ventral nerve cord of the leech receive excitatory input from mechanosensory stimulation (Lent and Dickinson, 1984; Brodfuehrer and Friesen, 1986a, b; J. R. Groome, unpublished observations). We also observed increased activity of other, unidentified, units in the cephalic nerves during thermal or chemical stimulation of the lip. It will be interesting to determine the activities of thermosensory, chemosensory and mechanosensory neurones during excitation of the RZ neurones following presentation of feeding cues to the prostomial lip.

While mechanosensory activity is common to the response to either thermal or chemical stimulation of the lip, our data indicate that these two pathways are distinct in other respects. Pharmacological data suggest the involvement of distinct neurotransmitters during thermal versus chemical stimulation of the lip (Groome and Lent, 1991). A comparison of the relative kinetics of RZ neurone excitation following presentation of these cues to the lip is intriguing. The gradual onset of RZ neurone excitation following activation of the chemical pathway might elicit release of neuromodulatory substances with slow actions. Possible candidates include FMRFamide-like peptides, which have been shown to elicit burst activity in RZ neurones (Sahley et al. 1993). We plan to determine the neural circuitry underlying these input pathways in order to clarify the behavioural significance of the effects of these two feeding stimuli on RZ neurones during leech ingestion.

The authors thank Dr A. Kleinhaus (New York Medical College, Valhalla, NY) for comments on the manuscript and Dr E. J. Elliot (Veterans Administrative Medical Center, Baltimore, MD) for helpful discussions regarding our results. This work was supported by a grant from the Willard L. Eccles Charitable Foundation to C.M.L. and by NSF grant IBN 9113118 to C.M.L. and D.K.V.

Brodfuehrer
,
P. D.
and
Friesen
,
W. O.
(
1986a
).
Initiation of swimming activity by trigger neurons in the leech subesophageal ganglion. I. Output connections of Tr1 and Tr2
.
J. comp. Physiol. A
159
,
489
502
.
Brodfuehrer
,
P. D.
and
Friesen
,
W. O.
(
1986b
).
Initiation of swimming activity by trigger neurons in the leech subesophageal ganglion. III. Sensory inputs to Tr1 and Tr2
J. comp. Physiol. A
159
,
511
519
.
Dickinson
,
M. H.
and
Lent
,
C. M.
(
1984
).
Feeding behavior of the medicinal leech, Hirudo medicinalis
.
J. comp. Physiol. A
154
,
450
456
.
Elliot
,
E. J.
(
1986
).
Chemosensory stimuli in feeding behavior of the leech Hirudo medicinalis
.
J. comp. Physiol. A
159
,
391
401
.
Elliot
,
E. J.
(
1987
).
Morphology of chemosensory organs required for feeding in the leech Hirudo medicinalis
.
J. Morph
.
192
,
181
187
.
Glover
,
J. C.
and
Lent
,
C. M.
(
1991
).
Serotonin is released from isolated leech ganglia by potassium-induced depolarization
.
Comp. Biochem. Physiol.
99
,
437
443
.
Groome
,
J. R.
,
Clark
,
M. C.
and
Lent
,
C. M.
(
1993
).
The behavioural state of satiation in the leech is regulated by body distension and mimicked by serotonin depletion
.
J. exp. Biol.
182
,
265
270
.
Groome
,
J. R.
and
Lent
,
C. M.
(
1991
).
Thermal and chemical inputs differentially affect serotonin neurons that express feeding by the medicinal leech
.
Soc. Neurosci. Abstr.
17
,
198
.
Lange
,
A. M.
,
Orchard
,
I.
and
Barrett
,
M.
(
1989
).
Changes in haemolymph serotonin levels associated with feeding in the blood-sucking bug, Rhodnius prolixus
.
J. Insect Physiol
.
35
,
393
399
.
Leake
,
L. D.
,
Griffith
,
S. G.
and
Burnstock
,
G.
(
1985
).
5-Hydroxytryptamine-like immunoreactivity in the peripheral and central nervous systems of the leech Hirudo medicinalis
.
Cell Tissue Res
.
239
,
123
130
.
Lent
,
C. M.
(
1985
).
Serotonergic modulation of the feeding behavior of the medicinal leech
.
Brain Res. Bull
.
14
,
643
655
.
Lent
,
C. M.
and
Dickinson
,
M. H.
(
1984
).
Serotonin integrates the feeding behavior of the medicinal leech
.
J. comp. Physiol. A
154
,
457
471
.
Lent
,
C. M.
,
Fliegner
,
K. H.
,
Freedman
,
E.
and
Dickinson
,
M. H.
(
1988
).
Ingestive behaviour and physiology of the medicinal leech
.
J. exp. Biol.
137
,
513
527
.
Lent
,
C. M.
,
Zundel
,
D.
,
Freedman
,
E.
and
Groome
,
J. R.
(
1991
).
Serotonin in the leech central nervous system: anatomical correlates and behavioral effects
.
J. comp. Physiol. A
168
,
191
200
.
Mann
,
K. M.
(
1962
).
Leeches (Hirudinea). Their Structure, Physiology, Ecology and Embryology
.
London
:
Pergamon Press
.
Marshall
,
C. G.
and
Lent
,
C. M.
(
1988
).
Excitability and secretory activity in the salivary gland cells of jawed leeches (Hirudinea: Gnathobdellida)
.
J. exp. Biol.
113
,
367
380
.
Mason
,
A.
,
Sunderlan
,
A. J.
and
Leake
,
L. D.
(
1979
).
Effects of leech Retzius cells on body wall muscles
.
Comp. Biochem. Physiol.
63
,
359
361
.
Muller
,
K. J.
,
Nicholls
,
J. G.
and
Stent
,
G. S.
(
1981
).
Neurobiology of the Leech
.
New York
:
Cold Spring Harbor Laboratory Press
.
Orchard
,
I.
,
Lange
,
A. B.
and
Barrett
,
M. F.
(
1988
).
Serotonergic supply to the epidermis of Rhodnius prolixus: evidence for serotonin as the plasticising factor
.
J. Insect Physiol.
34
,
873
879
.
Sahley
,
C. L.
,
Strong
,
J. A.
and
Kleinhaus
,
A. L.
(
1993
).
FMRF-amide modulates the electrical activity of the leech Retzius cell
.
Neurosci. Lett.
164
,
37
40
.
Sawyer
,
R. T.
(
1986
).
Leech Biology and Behaviour
, vol.
I
,
Anatomy, Physiology and Behaviour. Swansea, UK: Biopharm (UK) Limited
.
Willard
,
A. L.
(
1981
).
Effects of serotonin on the generation of the motor program for swimming by the medicinal leech
.
J. Neurosci.
1
,
936
944
.