Immunohistochemical and electrochemical detection of serotonin in the nervous system of the blood-feeding bug Rhodnius prolixus.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 8187-201 (1988) lmmunohistochemical and Electrochemical Detection of Serotonin in the Nervous System of the Blood-Feeding Bug, Rhodnius prolixus Angela B. Lange, Ian Orchard, and Robert J. Lloyd Department of Zoology, University of Toronto, Toronto, Ontario, Canada The distribution of serotonin throughout the nervous system of the bloodfeeding bug Rhodnius prolixus has been studied using immunohistochemistry and reversed-phase high-performance liquid chromatography coupled to electrochemical detection. Approximately 150 serotonin-like immunoreactive neurons are distributed throughout the central nervous system. These neurons are distributed over both ventral and dorsal surfaces and are found in all ganglia. Several of these neurons appear to be homologous to previously described serotonin-like immunoreactive neurons in insects. These include a large pair of bilaterally symmetrical neurons that project axons out of the suboesophageal ganglion, and some serially homologous neurons which project contralaterally and appear to be interneurons. lmmunoreactive branches and varicosities are found in the neuropile of all ganglia, and immunoreactive axons are found in all interganglionic connectives. Several of the peripheral nerves are covered in a plexus of immunoreactive processes. These processes result in a meshwork of fine varicose branches lying superficially over the peripheral nerves, which resemble neurohemal areas. The central nervous system of Rhodnius contains about 5.5 pmol serotonin unequally distributed between the ganglia. The highest content is found in the brain and optic lobes, and the lowest content is found in the prothoracic ganglion. Substantial amounts of serotonin are present in the corpus cardiacum and in the peripheral nerves possessing the serotonin-like neurohemal areas. Serotonin is released from these latter neurohemal areas in a calcium-dependent manner in response to high-potassium saline. It is concluded that serotonin plays an important central and peripheral role in this insect. Key words: serotonin-like immunoreactivity, serotonin content, central nervous system, neurohemal area, serotonin release, insect Acknowledgments: This work was supported by the Natural Sciences and Engineering Research Council of Canada. We are grateful to Dr. Michael Barrett for advice, encouragement, and provision of Rhodnius throughout this study. Received January 9,1988; accepted April 25,1988. Address reprint requests to Dr. Angela B. Lange, Dept. of Zoology, University of Toronto, Toronto, Ontario, Canda MSS 1Al. 01988 Alan R. Liss, Inc. 188 Lange et al. INTRODUCTION The insect CNS" has been shown to contain the biogenic amines dopamine, norepinephrine, octopamine, and serotonin [l-31. Relatively little, however, is known about the physiological roles for these amines, with the exception of octopamine. The discovery of identified octopaminergic neurons in insects has led to a detailed understanding of many of the functional roles for octopamine [see Ref. 41. Thus, as a prelude to physiological studies of these other amines, it is important to have a detailed map as to their distribution within the nervous system, with the ultimate goal of characterising identified neurons. Serotonin appears to have important regulating functions within insects. It has been identified by biochemical analysis in the CNS of a variety of insects [see Ref. 31 where it is assumed to act as a neuroactive substance. Serotonin also exerts pharmacological effects on a variety of insect peripheral tissues such as the heart [5,6], salivary glands [7-91, Malpighian tubules [lo], epidermal cells [ll],and visceral muscles [V]. Some of these observations suggest that serotonin may act as a neurohormone; others indicate direct delivery of serotonin to its target tissue. Serotonin has been detected with fluorescence histochemistry such as the Falck-Hillarp or glyoxylic acid techniques [see Ref. 131. More recently, the production of highly specific antibodies to serotonin [13,14] has allowed the use of immunohistochemical localisation of serotonin in both vertebrates and invertebrates [3,13-161. This technique is more sensitive and specific than the others for serotonin [3,17l. The use of serotonin antibodies has enabled the mapping of neurons showing serotonin-like immunoreactivity in a variety of insects [3,16,18-231. Of some interest is the conservation of apparently homologous neurons between insect species and the demonstration of an extensive plexus of immunoreactive varicose fibers in the neural sheath of peripheral nerves [16,22,24] and ganglia 1221, indicative of neurohemal areas for the release of serotonin into the hemolymph. Few studies, however, have combined immunohistochemical mapping with direct assays of serotonin content, and no reports have demonstrated the release of endogenous serotonin from the apparent neurohemal areas. The present paper describes a study of serotonin throughout the nervous system of the blood-feeding bug, Rhodnius prolixus. We have chosen Rhodnius because of an earlier report of serotonin-like immunoreactivity in ventral nerve cord of the fifth instars of this insect  and the demonstration of a plexus of immunoreactive fibers in the neural sheath of some abdominal nerves. Our interest lies in the central and peripheral actions of serotonin and the possible involvement of serotonin as a neurohormone released from the apparent neurohemal areas. However, our preliminary immunohisto- *Abbreviations: BSA = bovine serum albumin; CNS = central nervous system; FlTC = fluorescein isothiocyanate: HEPES = N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid; HPLC = high-performance liquid chromatography; NGS = normal goat serum; PBS = phosphate-buffered saline. Serotonin in Rhodnius 189 chemical studies revealed more stained neurons then previously shown by Flanagan , and so we have re-evaluated the distribution of serotonin-like immunoreactivity throughout the nervous system of Rhodnius, including the brain, which had not previously been examined. In addition we have assayed the content of serotonin in the same structures by HPLC coupled with electrochemical detection, and we have confirmed that the serotonin-like immunoreactivity is probably due to endogenous serotonin. Finally, we have demonstrated for the first time in an insect that endogenous serotonin can be released from neurohemal areas, and this release is calcium-dependent. MATERIALS AND METHODS Adults of Rhodnius prolixus Stil, obtained from a long-established colony at the University of Toronto, were used throughout the present study. The bugs were maintained at 25°C in high relative humidity and were used, unfed, within 30-40 days of emergence as adults. Serotonin-Like Immunoreactivity Immunohistochemistry was performed on whole mounts of the central nervous system, corpus cardiacum, and stretches of peripheral nerve, which were fixed for 2 h at room temperature in 2% paraformaldehyde in Millonigs buffer (120 mM NaH2P04buffered with NaOH to pH 7.4, containing 1% Dglucose and 0.005% CaC12).Following fixation the tissues were washed for 2 hr at room temperature in PBS (10 mh4 phosphate buffer (pH 7.2) containing 0.9% NaC1) and then incubated for 1h in PBS containing 2% BSA, 4% Triton X-100, and 2% NGS. The anti-serotonin antiserum (Immuno Nuclear Corp, Stillwater, MN) was diluted 1:1,000 with PBS containing 2% BSA, 0.4% Triton X-100, and 2% NGS, and incubated at 4°C for 18 h. The diluted antiserum was then incubated with the tissues for 48 h at 4°C with constant shaking. The tissues were subsequently washed for 5 h at room temperature in PBS and then incubated in 1:200 FITC-labeled sheep anti-rabbit IgG (Daymar Laboratories, Toronto) in 10% NGS in PBS for 18 h at 4°C. Following a wash of 4-18 h in PBS the tissues were mounted in 5% n-propyl gallate in glycerol, pH 7.3, and observed for serotonin-like immunoreactivity with epiillumination fluorescence microscopy. The specificity of the serotonin-like immunoreactivity was verified by preabsorbing the antiserotonin antiserum with either 1 mgiml serotonin or 100 pglml serotonin-conjugated BSA (Immuno Nuclear Corp.) for 18 h at 4°C. Electrochemical Detection of Serotonin Serotonin was assayed by HPLC using electrochemical detection  according to slight modifications of the procedure of Orchard et al. . Appropriate tissues were dissected under physiological saline, immediately placed into polypropylene tubes containing 25 pl of ice-cold 0.2 N perchloric acid, and subsequently maintained in the dark to avoid UV inactivation . After approximately 15 min, 225 pl of HPLC buffer (see later) was added to the tubes. The mixture was sonicated and, following centrifugation at 30,OOOg 190 Lange et al. for 30 min or filtration through a 0.22-pm nylon filter, 100 pl was injected onto a Brownlee RP-18 Spheri-5 HPLC column (4.6mm X 22cm). The mobile phase, pumped at 0.8 mllmin, contained 62.5 mM NaH2P04,1.5 mM sodium dodecyl sulfate, 1pM EDTA, 15% methanol, and 16.2% acetonitrile and was adjusted to pH 3.3 with concentrated perchloric acid. Detection of eluted compounds was achieved electrochemically using an ESA model 5100 A detection system coupled to an ESA model 5010 dual coulometric detector (ESA, Inc., Bedford, MA). The first detector was set at 0.05 V to act as a screen, and serotonin was detected using the second detector set at 0.35 V. A guard cell inserted before the injector valve was set at 0.4 V to preoxidize possible contaminants in the mobile phase. The output of the second detector was recorded on a Spectra Physics 4270 integrator (Spectra Physics, San Jose, CA), and serotonin was quantified using the external standard method. Tissues were pooled to provide suitable content for quantification (three to five tissues per pool) and spiked with serotonin to confirm the identity of the oxidizable substance and to check for losses. Release of Serotonin Stretches of abdominal nerves were dissected under saline, and their lengths measured with an eye-piece micrometer. Pooled nerves were washed and incubated for 10 min in 20 pl of various experimental salines at room temperature. The salines were individually collected, and 90 pl of HPLC buffer was added. This solution was injected directly onto the HPLC column, and the serotonin content was quantified by reference to an external standard. Under the conditions of the present study, the sensitivity of the assay for serotonin was found to be 3 pg (14 fmol). At the end of the experiment the nerves were collected into 20 pl of 0.2 N perchloric acid, and serotonin was measured as described earlier for the nervous tissue. The composition of the salines was as follows: normal saline, NaCl 150 mM, KC1 8.6 mM, CaC12 2.0 mM, MgC12 8.5 mM, HEPES buffer (pH 7.0) 10 mM, glucose 34 mM; high potassium saline, normal saline in which 100 mM KC1 was substituted for 100 mM NaCl; zero calcium saline, normal saline in which CaC12 was removed and MgC12 was elevated to 17 mM; high potassium zero calcium saline, zero calcium saline in which 100 mM KC1 was substituted for 100 mM NaCl. HPLC grade reagents (American Burdick and Jackson, Canlab, Toronto), were used throughout for the HPLC work. All other chemicals were obtained from Sigma Chemical Co. (St. Louis), unless otherwise stated. RESULTS Serotonin-Like Immunoreactivity The CNS of Rhodnius consists of a series of ganglia comprising the brain, suboesophageal ganglion, prothoracic ganglion, and the mesothoracic ganglionic mass (a fused ganglion composed of the mesothoracic, metathoracic, and abdominal ganglia). The serotonin-antiserum revealed numerous immunoreactive neurons and processes within the nervous system of Xhodnius. Serotonin in Rhodnius 191 The staining observed was reduced, or for most cells eliminated, by preabsorption of the antiserum with serotonin. All staining was eliminated by preabsorption with serotonin conjugated to BSA. Figures 1and 2 summarise the data obtained for cells that stained with high-to-moderate intensity in a consistent manner. Neurons showing serotonin-like immunoreactivity occur on both dorsal and ventral aspects of all ganglia of the CNS and usually exist as bilaterally symmetrical clusters, although a small number in the midline are not so obviously paired. The ventral suface of the CNS is conspicuous for a series of apparently homologous, bilaterally symmetrical neurons lying in a posterior lateral position in each ganglion of the ventral nerve cord. These neurons were originally described by Flanagan . The neurites of these cells characteristically project transversely in the neuropile and then extend as contralateral axons, which in some cases appear to ascend and descend the ventral nerve cord. However, it is difficult to trace the axons with any certainty, and we do not know if all of these ventral contralateral neurons are interganglionic. They do not, however, appear to leave the CNS, and, therefore, are apparently interneurons. The ventral contralateral neurons produce numerous branches and varicosities in the anterior quandrant of the neuropile, extending over both ventral and dorsal surfaces. The suboesophageal ganglion and prothoracic ganglion each contain three bilaterally symmetrical ventral contralateral neurons, although the cell body diameters in the prothoracic ganglion are larger than those in the suboesophageal ganglion (15 pm versus 10 pm diameter). The mesothoracic ganglionic mass has two ventral contralateral neurons (20 pm diameter) on each side of the mesothoracic segmental neuromere, one on each side of the metathoracic neuromere (10 pm diameter), and what appears to be one on each side of the remaining abdominal neuromeres (10 pm cell body diameter). The remaining cells on the ventral surface consist of a pair of large ventral lateral neurons in the prothoracic ganglion (23 pm diameter), a pair of small neurons (7 pm diameter) lying bilaterally in a medial position in both the metathoracic neuromere and prothoracic ganglion (the ventral medial bilateral neurons), and a cluster (four or five) of ventral medial bilateral neurons in the suboesophageal ganglion (15 pm diameter). These latter neurons have been described earlier . Processes are hard to trace from the cell bodies of these neurons. In addition, the suboesophageal ganglion also contains a pair of large bilaterally symmetrical cells whose axons leave the CNS via the ipsilateral anterior peripheral nerve root. These ventral efferent neurons, which have been described earlier , have large cell bodies (20 pm diameter) and are the only neurons in the CNS whose axons can be traced with certainty leaving the CNS (Figs. 1, 2). Their axons project anteriorly into the head region along the peripheral nerve root. The brain has only four pairs of cells on the ventral surface, lying in a medial position: the ventral medial bilateral neurons. The most anterior pair are large neurons (35 pm diameter), while the other three pairs are smaller (10-15 pm diameter). A neurite can be seen leaving each cell body but cannot be traced any great distance. Lying at the base of the optic lobes are a cluster of four or five cells located in a posterior position: the ventral optic neurons. Again, we have been unable to trace the neurites leaving their cell bodies. Lange et al. 192 DORSAL *; DAMB VENTRAL . w vo DPO 3% BRAIN SOG PRO MTGM 200 pln Fig. 1. Serotonin-like irnrnunoreactivity in the central nervous system of Rhodnius. Cornposite of camera lucida drawings of neurons that consistently reacted with the serotonin antiserum. Positively staining neurons and peripheral nerves containing serotonin-like irnrnunoreactive processes are shown as solid. Faintly staining neurons are shown as dotted. Neurons as follows: DAL, dorsal anterior lateral: DAMB, dorsal anterior medial bilateral; DAO, dorsal anterior optic; DL, dorsal lateral; DM, dorsal medial; DPL, dorsal posterior lateral; DPMB, dorsal posterior medial bilateral; DPO, dorsal posterior optic; VC, ventral contralateral; VE, ventral efferent; VMB, ventral medial bilateral; VL, ventral lateral; VO, ventral optic; MTGM, mesothoracic ganglionic mass; PRO, prothoracic ganglion; SOG, suboesophageal ganglion. Serotonin in Rhodnius 193 Fig. 2. Whole-mount preparations showing serotonin-like immunoreactive neurons in Rhodnius nervous system. Ventral surface of selected areas of A, brain and suboesophageal ganglion; B, prothoracic ganglion; and C, mesothoracic ganglionic mass. Labelling as for Figure 1. lmmunoreactive processes associated with peripheral nerves are shown in D-F. D: Two axons from the VE neurons of the suboesophageal ganglion as they project in the anterior peripheral nerve root. E,F: lmmunoreactive branches and varicosities lying superficially on abdominal nerves. Scale bar: A, 40 pm; B,C, 50 pm; D,13 pm; E, 29 pm; F, 50 pm. 194 Lange et al. The dorsal surface of the CNS also contains numerous serotonin-like immunoreactive neurons, although processes are rarely seen leaving their cell bodies. Lying at the posterior dorsal surface of the mesothoracic ganglionic mass are a cluster of bilaterally symmetrical cells, the dorsal posterior lateral neurons, whose number varies from six to nine. Their cell body diameters range from 5 pm to 20 pm. As well, the mesothoracic ganglionic mass contains a cluster (three to four) of bilaterally symmetrical neurons lying in a midlateral position, the dorsal lateral neurons, whose cell body diameters range from 10 pm to 30 pm. Also within this ganglion are some lightly staining cells. A group of dorsal medial neurons with very large cell bodies (40 pm diameter) lie medially in a posterior region and are not obviously paired. Lightly staining bilaterally symmetrical neurons are also consistently observed in an anterior lateral position of both the mesothoracic ganglionic mass and prothoracic ganglion. These dorsal anterior lateral neurons also have large cell bodies (20-30 pm diameter), with a single neuron occurring on each side of both the mesothoracic ganglionic mass and prothoracic ganglion. The suboesophageal ganglion contains bilaterally symmetrical neurons, the dorsal lateral neurons, which lie as triplets on each side of the ganglion. These have neurites passing toward the midline of the ganglion and projecting posteriorly. Numerous serotonin-like immunoreactive neurons lie as bilaterally symmetrical clusters at the posterior dorsal surface of the brain. There are approximately nine to twelve cells in this dorsal posterior medial bilateral cluster. In addition, there are two sets of bilateral pairs of neurons in the brain, the dorsal anterior medial bilateral neurons, and the dorsal anterior lateral neurons. Finally, the dorsal surface of the optic lobes contains two clusters of neurons: the dorsal posterior optic neurons and the dorsal anterior optic neurons. Serotonin-Like Immunoreactive Neurohemal Area The corpus cardiacum, a well-defined neurohemal organ in insects, contains processes that are serotonin-like immunoreactive in Xhodnius. These processes consist of branches and varicosities throughout the corpus cardiacum, while other processes continue through the corpus cardiacum to project along the aorta. Several peripheral nerves associated with the ventral nerve cord contain a plexus of immunoreactive processes lying superficially along their length. These nerves are identified in Figure 1. Most notably the four pairs of thin abdominal nerves projecting posteriorly from the abdominal neuromere of the mesothoracic ganglionic mass contain extensive arborizations of serotonin-like immunoreactive processes, which result in a meshwork of fine, varicose branches over much of their surface (Fig. 2). The neurohemal areas lying on these four pairs of abdominal nerves have been described previously . The large pair of posterior median nerve trunks do not in themselves possess this plexus, although fine abdominal nerves that branch from them more posteriorly do. (It is worth mentioning, however, that we have observed this plexus on the large posterior median nerve trunks in fifth instar Serotonin in Rhodnius 195 larval Rhodnius.) Other nerves noted in Figure 1also contain serotonin-like immunoreactive processes which are located along their surface, although the plexus on each nerve is less extensive than that observed on the thin abdominal nerves. We have not observed serotonin-like immunoreactive processes connecting each plexus with cells within the CNS. Each plexus terminates close to where the peripheral nerve enters the CNS. Electrochemical Detection of Serotonin We have used HPLC coupled with electrochemical detection to quantify the content of serotonin within the nervous system of Rhodnius (Table 1). The CNS of adult males contains about 5.5 pmol serotonin unequally distributed between the ganglia. The highest content is found in the brain and optic lobes, with lower values in the suboesophageal ganglion and mesothoracic ganglionic mass, and the lowest content in the prothoracic ganglion. The content of serotonin throughout the nervous system tends to parallel the distribution and number of serotonin-immunoreactive neurons. Substantial amounts of serotonin are present in the corpus cardiacum and in lengths of peripheral nerves dissected from the four pairs of thin abdominal nerves projecting posteriorly from the mesothoracic ganglionic mass (Table 1).Thus again, content of serotonin parallels the distribution of serotonin-like immunoreactivity. Release of Serotonin The meshwork of serotonin-like immunoreactive processes lying over the abdominal and other peripheral nerves resembles neurohemal tissue. In an attempt to provide evidence for a possible hormonal role for serotonin we examined for the release of serotonin from these structures. Incubation of abdominal nerves containing this meshwork in high-potassium saline resulted in the release of serotonin. During a 10-min incubation in normal saline, 40.4 28.7 fmol serotonin (n = 8) was released compared to 119.4 28.7 fmol serotonin (n = 8) released in high-potassium saline. The tissue content of serotonin after the experiments was 426.4 k 99.6 fmol (n = 8) indicating that 14.5% of the total store of serotonin had been released by elevated potassium. In a different series of experiments, the high-potassium- + TABLE 1. Serotonin Content of Rhodnius Nervous Tissue* Nervous tissuea + Brain optic lobes Suboesophageal ganglion Prothoracic ganglion Mesothoracic ganglionic mass Corpus cardiacum Abdominal nerves (per cm) nb 4 4 4 4 4 4 pmolitissue (mean i SEM) 2.79 & 1.46 & 0.53 & 1.37 & 0.78 0.39 0.05 0.28 0.45 & 0.19 0.77 & 0.31 *Serotonin content determined by HPLC with electrochemical detection. 'Peripheral nerves were removed from the ganglia close to their exit. Tissues were pooled (three to five) for each assay. bNo. of replicates. 196 Lange et al. induced release of serotonin was shown to be calcium-dependent (Fig. 3). The effects of calcium-free saline were reversible, with 11%of the total store of serotonin released by a subsequent 10-min incubation in elevated potassium. DISCUSSION The serotonin antiserum used in the present study has been reported to have high specificity with no serious problems of crossreactivity [17,19,28,29]. We believe that the immunoreactivity observed in the present study is probably due to serotonin since preabsorption of the antiserum with serotonin conjugated to BSA abolishes all staining, and the distribution of serotoninlike immunoreactivity tends to parallel the content of serotonin determined by HPLC. The content of serotonin throughout the CNS of Rhodnius reveals levels of a similar order of magnitude (though lower) to those described for other insects. For example, the cerebral ganglia of honeybee , cockroach , and blowfly  contain 21.4, 23.2, and 6.5 pmol, respectively, while the suboesophageal ganglion of honeybee  contains 5.6 pmol and the thoracic ganglia of blowfly  contains 2.2 pmol. The number of neurons staining with the antiserum in each ganglion correlates fairly well with the content of serotonin. Thus, the brain contains the highest number of immunoreactive I' Zero Ca 2mM Ca m Time (min) Fig. 3. Release of serotonin from abdominal nerves. Pooled abdominal nerves were incubated in calcium-free saline, calcium-free high-potassium saline, normal saline and highpotassium saline for 10 min, and serotonin release into the medium was quantified by HPLC with electrochemical detection. High-potassium saline resulted in the release of serotonin. This release did not occur in the absence of calcium ions. Following the experiment the abdominal nerves were found to contain 450 & 140 fmol serotonin, thereby illustrating that approximately 11 % of the total store was released by high-potassium saline. Serotonin in Rhodnius 197 neurons and the highest content of serotonin, whereas the prothoracic ganglion contains the smallest number and smallest content. One anomaly lies in the almost equal content of serotonin in the suboesophageal ganglion and mesothoracic ganglionic mass in spite of the fact that the mesothoracic ganglionic mass possesses more immunoreactive neurons. However, the serotonin content must reflect not only cell body content but arborisations and terminals within the neuropile. It is possible, therefore, that more serotonin is stored in the neuropile of the suboesophageal ganglion than in the neuropile of the mesothoracic ganglionic mass. The presence of serotonin in the corpus cardiacum of Rhodnius extends earlier observations of its presence in other insect neurohemal organs [see Ref. 231. The precise role of serotonin within this organ is not understood, but it may serve as a neurohormone or as a neuromodulator of peptide hormone release . The present investigation of adult Rhodnius CNS reveals approximately 150 serotonin-like immunoreactive neurons. These neurons are distributed over both ventral and dorsal surfaces and are found in all ganglia. A number of these neurons are large and present the possibility of becoming identified serotoninergic neurons. Our demonstration of serotonin-like immunoreactivity in the ventral nerve cord of Rhodnius extends an earlier report for this insect. Flanagan  examined the distribution of serotonin-like immunoreactivity in the ventral nerve cord of Rhodnius (the brain was not examined) and found only 26 cells. It is possible that the differences in number of stained cells may be genuine biological differences between the batches of animals used. The differences are unlikely to be due to technique since the cells described by Flanagan  were intensely stained. Nor are they due to a difference in stage of insect since Flanagan reported similar results for third instar, fifth instar, and adult Rhodnius, and we have observed no differences between fifth instar and adult, which could account for the lack of cells. We are left with the conclusion that the neurons in Flanagan’s study contained a lower content of serotonin, which resulted in their remaining undetected. It will be interesting to try to identify the physiological conditions of the Rhodnius that led to this lower content. Of some particular interest from the present and previous study  are the neurons which appear to correspond to serotonin-like immunoreactive neurons described in other insects. Intensely staining neurons with neurites extending contralaterally are typically found in the posterolateral margin of ganglia in cockroaches, grasshoppers, locusts, and crickets [16,18,33,35]. These ventral contralateral neurons of Rhodnius project along ganglionic connectives and, in accord with similar neurons mentioned above, do not appear to project out of the CNS. They are apparently interneurons. A second set of serotonin-like immunoreactive neurons that are conserved between insect species include the large bilaterally paired neurons in the suboesophageal ganglion that have axons projecting out of the CNS. These ventral efferent neurons of Rhodnius appear to be homologous to neurons in cockroach and blowfly [16,22,33]. In these latter insects, such neurons give rise to an extensive neurohemal complex lying over the surface of the ganglia andlor peripheral nerves, which innervate the mouthparts. No such complex 198 Lange et al. was evident close to the suboesophageal ganglion of Rhodnius, although it is possible that the axons arborise more anteriorly. The ventral medial bilateral neurons in the suboesophageal ganglion of Rhodnius and the ventral medial bilateral neurons in the prothoracic ganglion and mesothoracic ganglionic mass appear to be similar to ones described in cockroach , although without stained neurites further comparisons are hard to make. The dorsal surface of the ventral nerve cord contains serotoninlike immunoreactive neurons lying in lateral positions, at the base of peripheral nerves, and therefore located in similar positions to neurons in cockroach . The dorsal anterior lateral neurons, though lightly stained, may be homologous to anterior dorsal bilateral neurons of cockroaches , and the dorsal medial neurons may be homologous to a medial group of neurons in the terminal abdominal ganglion of crickets that project to the hindgut . As described for cockroach [19,33], honeybee , blowfly , locust , and dragonfly , numerous serotonin-like immunoreactive neurons are distributed throughout the brain and optic lobes. Dense immunoreactive branches and varicosities are located in the dorsal and ventral neuropile of all ganglia of the ventral nerve cord and brain, and immunoreactive axons are found in all interganglionic connectives. Thus, serotonin must play an important central role within the nervous system of Rhodnius. The presence of serotonin-like immunoreactivity and serotonin throughout the optic lobes, brain, ganglia, and connectives indicates a functionally diverse involvement of serotonin within the CNS. Unfortunately, at the present time, nothing is known about the central role of serotonin or serotoninergic neurons in the insect CNS. In addition to its obvious central role, serotonin appears to have a peripheral function as indicated by the meshwork of immunoreactive processes lying upon peripheral nerves. The morphology of these processes suggests a neurohemal function for these areas. An earlier study  reported the presence of such neurohemal areas on the four pairs of abdominal nerves. However, the present study reveals that these areas are more extensive and lie upon several other peripheral nerves. Although superficial serotonin-like immunoreactive nerve terminals have been found in neurohemal regions of peripheral nerve roots of Rhodnius 1241 and other insect species [16,22,24], serotonin has not previously been localised and quantified in them. This is an important discovery because it lends credence to the notion that the immunoreactive terminals are indeed serotoninergic. The present results reveal substantial amounts of serotonin within the four pairs of thin abdominal nerves that possess the meshwork of serotonin-like immunoreactive branches and varicosities and, even more importantly, demonstrates the release of endogenous serotonin from such areas in response to a depolarising stimulus. This release is calcium dependent, giving some confidence to the actual physiological nature of the release. This is the first report to demonstrate the release of endogenous serotonin from neurohemal areas in insects, although [3H] serotonin has been shown to be sequestered and released by these areas [%I. Thus, serotonin may well possess a neurohormonal function in Rhodnius in addition to its central role. Such is believed to be the case in crustaceans, in which serotonin-like immunoreactive processes Serotonin in Rhodnius 199 are found centrally in the neuropile and superficially on peripheral nerve roots [271. Serotonin also occurs in physiological levels in the blood of crustaceans [see Ref. 271. While serotonin is yet to be demonstrated in the hemolymph of insects in response to a physiological event, it has been shown to be present following injections of high-potassium saline into blowflies . Since high-potassium saline injection results in salivation, a process mimicked by serotonin and inhibited by serotonin antagonists, it has been suggested that serotonin may be a neurohormone stimulating salivation in this insect . Other investigators have suggested that serotonin may be a neurohormone released at feeding to regulate peripheral events required at this time [10,11]. This may also be the case in Rhodnius, in which serotonin has been shown to induce plasticisation of the abdominal cuticle [ll] and to act upon the Malpighian tubules to stimulate diuresis [lo]-events which occur at feeding [lO,ll]. We have recently reported an immunohistochemical map and content of the pentapeptide proctolin in Rhodnius . The distribution of proctolinergic neurons appears quite distinct from the serotoninergic neurons. For example, proctolin and proctolin-like immunoreactive neurons are more concentrated in the ventral nerve cord, whereas serotonin and serotonin-immunoreactive neurons are more concentrated in the brain. There are no proctolin-immunoreactive neurons in the optic lobes, whereas serotonin-immunoreactive neurons are present. Indeed, there appears in general to be little overlap between proctolin and serotonin, i.e., there are few cells that appear to stain with both antibodies. Thus, as with cockroach , proctolin and serotonincontaining neurons are generally distinct neurons. The distribution of serotonin-like immunoreactive neurons within the CNS of Rhodnius constitutes a subpopulation of cells with a distinct biochemical phenotype. We are now continuing our studies in order to establish some of these neurons as identified neurons and to define their physiological role. In particular we are interested in the neurohormonal role of serotonin and possible involvement in the integration of feeding activities. LITERATURE CITED 1. Evans PD: Biogenic amines in the insect nervous system. Adv Insect PhysiolZ5, 317 (1980). 2. Orchard I. Martin RJ, Sloley BD, Downer RGH: The association of 5-hydroxytryptamine, octopamine, and dopamine with the intrinsic (glandular) lobe of the corpus cardiacum of Locusta rnigraforia. Can J Zoo1 64, 271 (1986). 3. Nassel DR: Serotonin and serotonin-immunoreactive neurons in the nervous system of insects. Prog Neurobiol30, 1(1988). 4. Evans PD: Octopamine. In: Comprehensive Insect Physiology Biochemistry and Pharmacology. Kerkut GA, Gilbert LI, eds. Pergamon Press, Oxford, Vol. 11, pp 499-530 (1985). 5. Collins C, Miller T: Studies on the acton of biogenic amines on cockroach heart. J Exp Biol 67, l(1977). 6. Tublitz NJ, Truman JW: Insect cardiactive peptides. 11. Neurohormonal control of heart activity by two cardioacceleratory peptides in the tobacco hawkmoth, Manduca sexfa. J Exp Biol 314, 381 (1985). 7. Berridge MJ, Patel NG: Insect salivary glands: Stimulation of fluid secretion by 5-hydroxytryptamine and adenosine-3, 5-monophosphate. Science 162, 462 (1968). 8. House CR, Smith RK: On the receptors involved in the nervous control of salivary secretion by Nauphoefa cinera Oliver. J Physiol (Lond) 279, 457 (1978) 200 Lange et al. 9. Trimmer BA: Serotonin and the control of salivation in the blowfly Calliphora. J Exp Biol 114, 307 (1985). 10. Maddrell SHP, Pilcher DEM, Gardiner BOC: Pharmacology of the Malpighian tubules of Rhodnius and Carausius: The structure-activity relationship of tryptamine analogues and the role of cyclic AMP. J Exp Biol54, 779 (1971). 11. Reynolds S: Pharmacological induction of plasticization in the abdominal cuticle of Rhodnius. J Exp Biol61, 705 (1974). 12. Brown BE: Neuromuscular transmitter substance in insect visceral muscle. Science 155, 595 (1967). 13. Steinbusch HWM, Verhofstad AAJ, Joosten HWJ: Localization of serotonin in the central nervous system by immunohistochemistry: Description of a specific and sensitive technique and some applications. Neuroscience 3, 811 (1978). 14. Steinbusch HWM, Verhofstad AAJ, Hoosten HWJ: Antibodies to serotonin for neuroimmunocytochemical studies. J Histochem Cytochem 30, 756 (1982). W. Belemtougri G, Andries J-C, Tramu G: Immunohistochemical localization of serotonin containing neurons in the nervous system of Aeshna cyanea (Insects, Odonata). Biogenic Amines 3, 209 (1985). 16. Davis NT: Neurosecretory neurons and their projections to the serotonin neurohemal system of the cockroach Periplanefa americana (L.), and identification of mandibular and maxillary motor neurons associated with this system. J Comp Neurol259, 604 (1987). 17. Tandler CJ, Bursco A, Peressini S, Pecci Saavedra J: Further evidence for the specificity of anti-5-HT (serotonin)-like antisera in immunocytochemisty. Histochemisty 85, 67 (1986). 18. Hustert R, Tope1 U: Location and major postembryonic changes of identified 5-HTimmunoreactive neurones in the terminal ganglion of a cricket (Acheta domesticus). Cell Tissue Res 245, 615 (1986). 19. Klemm N, Steinbusch HWM, Sundler F: Distribution of serotonin-containing neurons and their pathways in the supraoesophageal ganglion of the cockroach Periplaneta americana (L.) as revealed by immunocytochemistry. J Comp Neurol225, 387 (1984). 20. Klemm N, Hustert R, Cantera R, Nassel DR: Neurons reactive to antibodies against serotonin in the stomatogastric nervous system and in the alimentary canal of locust and crickets (Orthoptera, Insecta). Neuroscience 17, 247 (1986). 21. Nassel DR, Elekes K: Ultrastructural demonstration of serotonin-immunoreactivity in the nervous system of an insect (Calliphoraerythrocephala). Neurosci Lett 48, 203 (1984). 22. Nassel DR, Elekes K: Serotonergic terminals in the neural sheath of the blowfly nervous system: Electron microscopical immunocytochemisty and 5,7-dihydroxytryptamine labelling. Neuroscience 15, 293 (1985). 23. Nassel DR, Klemm N: Serotonin-like immunoreactivity in the optic lobes of three insect species. Cell Tissue Res 232, 129 (1983). 24. Flanagan TRJ: Cytological evidence for serotonin-containing fibers in an abdominal neurohemal organ in a Hemipteran. Brain Res 306, 235 (1984). 25. Flanagan TRJ: Serotonin-containing, catecholamine-containing and dopamine-sequestering neurones in the ventral nerve cord of the Hemipteran Rhodnius prolixus. J Insect Physiol 32, 17 (1986). 26, Downer RGH, Bailey BA, Martin RJ: Estimation of biogenic amines in biological tissues. In: Neurobiology. Gilles R, Balthazart J, eds. Springer-Verlag, Berlin, pp 248-263 (1985). 27. Siwicki KK, Beltz BS, Kravitz EA: Proctolin in identified serotonergic, dopaminergic and cholinergic neurons in the lobster Homarus americanus. J Neurosci 7, 522 (1987). 28. Schipper J, Tilders FJH: A new technique for studying specificity of immunocytochemical procedures: Specificity of serotonin immunostaining. J Histochem Cytochem 31, 12 (1983). 29. Steinbusch HWM, Verhofstad AAJ, Joosten HWJ: Antibodies to serotonin for neuroimmunocytochemical studies: Methodological aspects and applications. In: Immunohistochemistry, Cuello AC, ed. Wiley and Sons, New York, pp 193-214 (1983). 30. Mercer AR, Mobbs PG, Davenport AP, Evans PD: Biogenic amines in the brain of the honeybee, Apis mellifera. Cell Tissue Res 234, 655 (1983). 31. Sloley BD, Owen MD: The effects of reserpine on amine concentrations in the nervous system of the cockroach (Periplanefaamericana).Insect Biochem 12, 469 (1982). 32. Orchard I: The role of biogenic amines in the regulation of peptidergic neurosecretory cells. In: Insect Neurochemistry and Neurophysiology. Borkovec AB, Kelly TJ, eds. Plenum Press, New York, pp 115-136 (1984). Serotonin in Rhodnius 201 33. Bishop CA, O’Shea M: Serotonin immunoreactive neurons in the central nervous system of an insect (Periplunefuurnericunu). J Neurobiol 24, 251 (1983). 34. Flanagan TRJ, Berlind A: Serotonin modulation of the release of sequestered [3H] serotonin from nerve terminals in an insect neurohemal organ in vitro. Brain Res 306, 243 (1984). 35. Taghert PH, Goodman CS: Cell determination and differentiation of identified serotoninimmunoreactive neurons in the grasshoper embryo. J Neurosci 4, 989 (1984). 36. Orchard I, Lange AB, Barret FM: The presence and distribution of proctolin in the bloodfeeding bug Rhodnius prolixus. J Insect Physiol34, 379(1988).