The distribution of glutamate-like immunoreactivity in the thoracic and abdominal ganglia of the locust (Schistocerca gregaria)

The distribution of glutamate-like immunore-activity in the thoracic and abdominal ganglia of the locust was studied using two polyclonal antibodies against glutamate. Because glutamate is a precursor of the inhibitory transmitter -amino butyric acid (GABA) the distribution of immunostaining by antibodies against glutamate and GABA was closely compared in adjacent serial sections. When the antibodies were used at optimal dilutions there was no overlap in the distribution of immunostaining for glutamate and GABA. In the pro- and mesothoracic ganglia 360–400 somata are immunoreactive for glutamate, while in the metathoracic ganglion about 600 somata were stained. These range in diameter from 10–100 m in diameter and include the majority of the large somata in these ganglia. Bundles of primary neurites emerging from these large somata can be traced through the neuropile. Most of the bundles correspond to the known paths of motor neurone primary neurites. In addition the T-tract is also immunolabelled. The free abdominal ganglia each contain 80–100 somata ranging in size from 10–45 m while the terminal ganglion contains about 250 somata, 10–60 m in diameter.     

The distribution of GABA-like immunoreactivity in relation to ganglion structure in the abdominal nerve cord of the locust (Schistocerca gregaria)

Summary An antiserum raised against GABA was used to stain the abdominal nervous system of the locust. To interpret the results, however, it was first necessary to describe the structure of the free abdominal and terminal ganglia. This was done on the basis of ethyl-gallate staining. The free abdominal ganglia are similar in structure to the abdominal neuromeres of the metathoracic ganglia. The terminal ganglion is composed of four neuromeres (representing ganglia 8–11), but only three can be distinguished in the adult on morphological grounds. The eighth neuromere resembles the free ganglia, but the ninth lacks DCI (dorsal commissure I) and the T tracts. In the tenth, only DCII and III are recognisable of the commissures, but two more posterior ones of uncertain homology are also present. Immunocytochemistry reveals three populations of somata in each abdominal ganglion. Of these only one, the medial posterior group, is found in the thoracic ganglia. DCIV and the supra-median commissure are composed of stained neurites, DCII and V contain both unstained neurites and DCI, III and VI are unstained. With the exception of the median ventral tract, all the longitudinal tracts contain some stained axons.     

GABA-like immunoreactivity in the suboesophageal ganglion of the locust Schistocerca gregaria

Summary Neurones in the suboesophageal ganglion of the locust Schistocerca gregaria were stained with an antiserum raised against gamma amino butyric acid (GABA). This ganglion consists of the fused mandibular, maxillary and labial neuromeres. Immunoreactive cell bodies of similar size and distribution occur in the lateral, ventral and middorsal regions of all three neuromeres. Approximately 200 cell bodies stain in both the mandibular and maxillary neuromeres and 270 in the labial neuromere. A few distinctly larger cells occur in the ventral groups and one large pair occurs in the lateral group of the maxillary neuromere. Dorsal commissures DCIV and DCV are composed mainly of stained fibres, while DCI–DCIII are largely unstained. A ventral commissure also stains in the maxillary neuromere. All longitudinal tracts contain both stained and unstained fibres. Many processes within the neuropil are also immunoreactive. A stained axon is found in the posterior tritocerebral commissure which enters the anterior dorsal region of the mandibular neuromere. The salivary branch of the 7th nerve contains one stained axon and two axons stain in nerve 8 which innervates neck muscles.     

External sensilla of the locust abdomen provide the central nervous system with an interganglionic network

External mechanoreceptors and contact chemoreceptors on the cuticle of the sixth abdominal segment of locusts have divergent primary projections of their sensory neurons that form arbours in the segmental and anterior abdominal ganglia. Homologous interganglionic projections from adjacent segments converge in the neuropile of each abdominal ganglion. Of the contributing types of sensilla, three were previously unknown for locust pregenital segments: tactile mechanosensory hairs with dual innervation, external proprioceptors of the hairplate type covered by intersegmental membranes and single campaniform sensilla that monitor cuticular strain in sternites and tergites. In general, interdependence of motor coordination in the abdominal segments is based on a neural network that relies heavily on intersegmental primary afferents that cooperate to identify the location, parameters and strength of external stimuli.     

The development of GABA-like immunoreactivity in the thoracic ganglia of the locust Schistocerca gregaria

Summary The development of GABA-like immunoreactivity was investigated in embryonic and juvenile locusts using an antibody raised against GABA-protein conjugates. GABA-like immunoreactivity was first detectable in the neuropile of embryonic ganglia at 55% development, and in neuronal somata at 62% development. The total number of immunoreactive somata increased between 62% and 85% embryonic development, and followed an anterio-posterior pattern of expression. At 85% development, the number of immunoreactive somata reached adult levels and no change in number was then seen. In embryonic stages and first and second juvenile instars two dorsal and four ventral groups of somata were labeled in all three thoracic ganglia, whilst in later juvenile instars one of the dorsal groups was visible as a separate entity only in the metathoracic ganglion. These early patterns were modified by alterations in the positions of some of the groups during late embryogenesis and during juvenile development to produce the adult pattern. The results show that the development of GABA expression is similar to that of other neurotransmitters. The characteristics of the development of immunoreactivity indicate that some of these immunoreactive clusters may be derived from clonally related neurones. Finally, we demonstrate the presence of immunoreactive somata and processes in embryos, which correspond to those of identified local and intersegmental interneurones studied in the adult.Abbreviations Ab1–3 first-third abdominal ganglion - CON connective - CI _1–3 common inhibitors 1–3 - CTC tract - DC I–VII dorsal commissures I–VII - DIT dorsal intermediate tract - DMT dorsal median tract - LDT lateral dorsal tract - LF lateral fibres - o, iLVT outer and inner lateral ventral tract - MVT median ventral tract - N1–5 nerves 1–5 - aPT anterior perpendicular tract - PT perpendicular tract - aRT anterior ring tract - R1–5 nerve roots 1–5 - PVC posterior ventral commissure - SMC supra-median commissure - T3 metathoracic neuromere - TT T tract - aVAC anterior ventral association centre - VC I ventral commissure I - d,vVCII dorsal and ventral parts of ventral commissure II - VF ventral fibres - VIT ventral intermediate tract - VLT ventral lateral tract - VMT ventral median tract - (d,v)LAG (dorsal and ventral) lateral anterior group - LDG lateral dorsal group - LVG lateral ventral group - MDG medial dorsal group - MPG medial posterior group - MVG medial ventral group     

The distribution of GABA-like immunoreactivity in the thoracic nervous system of the locust Schistocerca gregaria

Summary An antiserum raised against gamma aminobuyric acid (GABA) was used to stain the thoracic nervous system of the locust. It stained both neuronal somata and processes within the neuropile. Among the stained somata, those of the three pairs of common inhibitory motor neurones could be identified in each of the three thoracic ganglia. In the pro- and mesothoracic ganglia five discrete groups of somata are stained, four ventral and one dorsal. In the metathoracic neuromere, an additional second dorsal group can be identified. In the abdominal neuromeres of the metathoracic ganglion both dorsal and ventral somata are stained but the latter cannot be divided into discrete populations. In each ganglion, dorsal commissures (DC) IV and V are composed of stained neurites, DCVII, the supramedian commissure, the perpendicular tract, and all the longitudinal tracts contain both stained and unstained neurites. DCI, II, III and VI, the T and I tracts are unstained. An abundance of GABA-like immunoreactive processes is found throughout the neuropile except for the anterior ventral association centre where stained processes are sparser. Some of the stained cell groups contain neurones that have been studied physiologically. The function of these neurones is discussed.Beit Memorial Fellow     

Central projections of sensory cells of the midleg of the locust,Schistocerca gregaria

The central projections of trichoid hairs and of some scolopidial organs of the mesothoracic leg of the locust Schistocerca gregaria were studied by using nickel chloride backfilling and single cell recording. Trichoid hair sensilla on different parts of the legs project somatotopically in the ventral part of the ipsilateral neuropile of the mesothoracic ganglion. Generally, distally located receptors have their terminal arborizations in ventro-lateral areas of the neuropile, and proximally located receptors in ventro-medial areas. The axons of the subgenual organ and tarsal chordotonal organs project into the intermediate neuropile.     

GABA and glutamate-like immunoreactivity at synapses received by dorsal unpaired median neurones in the abdominal nerve cord of the locust

Dorsal unpaired median (DUM) neurones in the abdominal ganglia of the locust were impaled with microelectrodes and some were injected intracellularly with horseradish peroxidase so that their synapses could be identified in the electron microscope. Simultaneous recordings from DUM neurones in different abdominal ganglia revealed that they received common postsynaptic potentials from descending interneurones. Post-embedding immunocytochemistry using antibodies against GABA and glutamate was carried out on ganglia containing HRP-stained neurones. GABA-like immunoreactivity was found in 39% (n=82) of processes presynaptic to abdominal DUM neurones and glutamate-like immunoreactivity in 21% (n=42) of presynaptic processes. Output synapses from the DUM neurites were rarely observed within the neuropile. Structures resembling presynaptic dense bars but not associated with synaptic vesicles, were seen in some large diameter neurites.     

Allatostatin-like immunoreactivity in the abdomen of the locust Schistocerca gregaria

A polyclonal antibody against allatostatin 1 (AST-1) of cockroach Diploptera punctata was used to investigate the distribution of AST-like immunoreactivity within the abdomen of the locust, Schistocerca gregaria. In the abdominal ganglia, AST-like immunoreactivity was found in both cell bodies and neuropile. In ganglia 6 and 7, staining was found in serial homologous cell bodies in anterior dorsolateral and dorsomedial, and posterior ventrolateral and ventromedial locations. In the terminal ganglion, the numerous immunoreactive somata were smaller in size than those in the unfused ganglia. The combination of backfill experiments with immunocytochemistry showed that, in abdominal ganglion 7, one neuron of the ventromedian cell body cluster and two of the ventrolateral cluster innervated the oviduct, which itself was covered with a dense mesh of AST-immunoreactive varicosities. Combining electron microscopy with immunohistochemistry revealed AST-like immunoreactivity in dense-core vesicles located in neurohaemal-like terminals lacking structures normally found in synapses. A mesh of AST-immunoreactive varicosities was also found on the muscles of the spermatheca and the spermathecal duct. In addition, a mesh of strongly stained varicosities was present in the distal perisympathetic organs (neurohaemal organs in the abdomen) and on the lateral heart nerves (a putative neurohaemal release zone). These data indicate that AST is an important neuroactive substance that is probably involved in multiple tasks in the control of the locust abdomen.     

Dendritic projections of different types of octopaminergic unpaired median neurons in the locust metathoracic ganglion

Octopaminergic dorsal unpaired median (DUM) neurons of locust thoracic ganglia are important components of motor networks and are divided into various sub-populations. We have examined individually stained metathoracic DUM neurons, their dendritic projection patterns, and their relationship to specific architectural features of the metathoracic ganglion, such as longitudinal tracts, transverse commissures, and well-defined sensory neuropils. The detailed branching patterns of individually characterized DUM neurons of various types were analyzed in vibratome sections in which architectural features were revealed by using antibodies against tubulin and synapsin. Whereas DUM3,4,5 and DUM5 neurons (the group innervating leg and "non-wing-power" muscles) had many ventral and dorsal branches, DUM1 and DUM3,4 neurons (innervating "wing-power" muscles) branched extensively only in dorsal areas. The structure of DUM3 neurons differed markedly from that of the other DUM neurons examined in that they sent branches into dorsal areas and had differently structured side branches that mostly extended laterally. The differences between the branching patterns of these neurons were quantified by using currently available new reconstruction algorithms. These structural differences between the various classes of DUM neurons corresponded to differences in their function and biophysical properties.     

Structure of isolated sensilla and sensory neurons in vitro: Observations on developing silkmoth antennae

Summary By combined enzymatic and mechanical treatment, it was possible to dissociate the sensory epithelium of developing antennae of male Antheraea polyphemus and A. pernyi silkmoths from the stage of separation of the antennal branches up to the early stages of cuticle deposition. Large numbers of entire developing trichoid sensilla were isolated. These are characterized by a large trichogen cell with a long apical, hair-forming process and a large nucleus. A cluster of 2–3 sensory neurons, enclosed by the thecogen cell, is situated in the basal region. The dendrites run past the nucleus of the trichogen cell into the apical process from which they protrude laterally. The nuclei of the tormogen and a 4^th enveloping cell can be distinguished near the base of the prospective hair. After further dissociation, only the neuron clusters remain, still enclosed by their thecogen cell and often attached to the antennal branch nerve via their axons. It is finally possible to disrupt the thecogen cells and the axons, leaving the sensory neurons with inner dendritic segments and axon stumps. The majority of these neurons can be expected to be olfactory.     

Standardized atlas of the brain of the desert locust, Schistocerca gregaria

In order to understand the connectivity of neuronal networks, their constituent neurons should ideally be studied in a common framework. Since morphological data from physiologically characterized and stained neurons usually arise from different individual brains, this can only be performed in a virtual standardized brain that compensates for interindividual variability. The desert locust, Schistocerca gregaria, is an insect species used widely for the analysis of olfactory and visual signal processing, endocrine functions, and neural networks controlling motor output. To provide a common multi-user platform for neural circuit analysis in the brain of this species, we have generated a standardized three-dimensional brain of this locust. Serial confocal images from whole-mount locust brains were used to reconstruct 34 neuropil areas in ten brains. For standardization, we compared two different methods: an iterative shape-averaging (ISA) procedure by using affine transformations followed by iterative nonrigid registrations, and the Virtual Insect Brain (VIB) protocol by using global and local rigid transformations followed by local nonrigid transformations. Both methods generated a standard brain, but for different applications. Whereas the VIB technique was designed to visualize anatomical variability between the input brains, the purpose of the ISA method was the opposite, i.e., to remove this variability. A novel individually labeled neuron, connecting the lobula to the midbrain and deutocerebrum, has been registered into the ISA atlas and demonstrates its usefulness and accuracy for future analysis of neural networks. The locust standard brain is accessible at http://www.3d-insectbrain.com .     

Ultrastructure and immunocytochemistry of endocrine cells in the midgut of the desert locust,Schistocerca gregaria (Forskal)

Summary The endocrine cells of the midgut epithelium of the desert locust are found dispersed among the digestive cells and are similar to those of the vertebrate gut. According to their reactivity to silver impregnation techniques and the ultrastructural features of the secretory granules (shape, electron-density, size, and structure) 10 types of endocrine cell have been identified, of which seven are located in the main segment of the midgut or in the enteric caeca, and the other three seem to be present only in the ampullae through which the Malpighian tubules drain into the gut. The endocrine cells have a slender cytoplasmic process that reaches the gut lumen, a feature that supports the receptosecretory nature postulated for this cellular type in insects as well as vertebrates. Antisera directed against mammalian gastrin, CCK, insulin, pancreatic polypeptide and bombesin reacted with some of the endocrine cells. This is the first time that insulin- and bombesin-like immunoreactive cells have been described in the midgut of an insect.     

Preparation-dependent distribution of intramembranous particles in freeze-fracture replicas of the neuromuscular junction of the locust Schistocerca gregaria

Summary Freeze-fracture replicas of the neuromuscular junction were prepared from untreated retractor unguis muscles of the locust Schistocerca gregaria that were rapidly frozen by contact with a copper block cooled by liquid helium. These replicas were compared with others prepared from tissue following fixation with glutaraldehyde and cryoprotection in glycerol. Freeze-fracture of rapidly frozen tissue produced replicas of high quality with little evidence of tissue damage by ice crystals in the superficial layers. The gross fracturing characteristics of the neuromuscular junction were consistent with replicas from fixed and cryoprotected tissue; all of the membrane specializations were recognisable but with some alterations in infrastructure. In tissue replicas prepared using either method intramembranous particles in the presynaptic membrane were arranged in a bar-like array. The intramembranous particles of this presynaptic bar array of the rapidly frozen material were large and found on the E-face of the cleaved membrane. This contrasts with the P-face distribution of the comparable particles in muscles fixed in glutaraldehyde and cryoprotected in glycerol, in which they are also smaller and more numerous. This difference in partitioning between rapidly frozen, and fixed, cryoprotected nerve terminals is not found at cholinergic synapses and thus may reflect functional differences between the two types of junction.Indentations of the nerve-terminal membrane occur in replicas from rapidly frozen muscle as well as fixed and cryoprotected muscle suggesting they are not fixation or glycerol-induced artifacts. It is suggested from their position and size that these indentations are more likely to be part of a membrane retrieval system than exocytotic figures.This work was supported by an S.E.R.C. project grant to I.R.D.     

Fine structure and primary sensory projections of sensilla located in the sacculus of the antenna of Drosophila melanogaster

Sensilla lining the inner walls of the sacculus on the third antennal segment of Drosophila melanogaster were studied by light and transmission electron microscopy. The sacculus consists of three chambers: I, II and III. Inside each chamber morphologically distinct groups of sensilla having inflexible sockets were observed. Chamber I contains no-pore sensilla basiconica (np-SB). The lumen of all np-SB are innervated by two neurons, both resembling hygroreceptors. However, a few np-SB contain one additional neuron, presumed to be thermoreceptive. Chamber II houses no-pore sensilla coeloconica (np-SC). All np-SC are innervated by three neurons. The outer dendritic segments of two of these neurons fit tightly to the wall of the lumen and resemble hygroreceptor neurons. A third, more electron-dense sensory neuron, terminates at the base of the sensillum and resembles a thermoreceptor cell. Chamber III of the sacculus is divided into ventral and dorsal compartments, each housing morphologically distinct grooved sensilla (GS). The ventral compartment contains thick GS₁, and the dorsal compartment has slender sensilla GS₂. Ultrastructurally, both GS₁ and GS₂ are doublewalled sensilla with a longitudinal slit-channel system and are innervated by two neurons. The dendritic outer segment of one ofthe two neurons innervates the lumen of the GS and branches. On morphological criteria, we infer this neuron to be olfactory. The other sensory neuron is probably thermoreceptive. Thus, the sacculus in Drosophila has sensilla that are predominantly involved in hygroreception, thermoreception, and olfaction. We have traced the sensory projections of the neurons innervating the sacculus sensilla of chamber III using cobaltous lysine or ethanolic cobalt (II) chloride. The fibres project to the antennal lobes, and at least four glomeruli (VM₃, DA₃ and DL_2-3) are projection areas of sensory neurons from these sensilla. glomerulus DL₂ is a common target for the afferent fibres of the surface sensilla coeloconica and GS, whereas the VM₃, DA₃ and DL₃ glomeruli receive sensory fibres only from the GS.     

An identified dorsal unpaired median neurone and bilaterally projecting neurones exhibiting bovine pancreactic polypeptide-like/FMRFamide-like immunoreactivity in abdominal ganglia of the migratory locust

Summary Three antisera were used to study the distribution and anatomy of bovine pancreatic polypeptide (BPP)-like/FMRFamide-like immunoreactive neurones within the unfused abdominal ganglia of the migratory locust, Locusta migratoria. All the antisera used stained two or more clusters of perikarya, localized anteriorly and posteriorly near the midline within each unfused abdominal ganglion. Double labelling experiments with intracellular dye injection, or differential backfilling, combined with subsequent immunostaining were carried out to identify these neurones. Two of the antisera (antisera 1 and 2, both raised against FMRFamide) stained three groups of midline neurones, located anterior dorsal, anterior ventral and posterior dorsal within the ganglion. Neurones of the former of these two clusters projected via the anterior median nerve to a neurohaemal organ. The posterior cluster of midline cells comprised immunopositive perikarya all but one of which also projected via the anterior median nerve to innervate the neurohaemal organ. Double labelling with Lucifer yellow and antisera 1 and 2 showed that the remaining neurone was the previously identified doral unpaired median (DUM)heart1 neurone. The third antiserum (AK141), also raised against FMRFamide, stained neurones within an anterior dorsal cluster, and in a posterior cluster. Double labelling with differential Co²⁺/Ni²⁺-backfilling and the antiserum 3 (AK141) demonstrated that the large neurones of both clusters belonged to the population of bilaterally projecting neurones (BPNs), including the DUMheart1 neurone. Since the antisera cross-react with BPP and fail to label neurones when preadsorped with BPP or FMRFamide, we conclude that the labelled neurones contain polypeptides of the FMRFamide/BPP-family.     

Immunocytochemistry of histamine in the brain of the locust<Emphasis Type="Italic"> Schistocerca gregaria</Emphasis>

Histamine serves a neurotransmitter role in arthropod photoreceptor neurons, but is also present in a small number of interneurons throughout the nervous system. In search of a suitable model system for the analysis of histaminergic neurotransmission in insects, we mapped the distribution of histamine in the brain of the desert locust Schistocerca gregaria by immunocytochemistry. In the optic lobe, apparently all photoreceptor cells of the compound eye with projections to the lamina and medulla showed intense immunostaining. Photoreceptors of the dorsal rim area of the eye had particularly large fiber diameters and gave rise to uniform varicose immunostaining throughout dorsal rim areas of the lamina and medulla. In the locust midbrain 21 bilateral pairs of histamine-immunoreactive interneurons were found, and 13 of these were reconstructed in detail. While most neuropil areas contained a dense meshwork of immunoreactive processes, immunostaining in the antennal lobe and in the calyces of the mushroom body was sparse and no staining occurred in the pedunculus and lobes of the mushroom body, in the protocerebral bridge, and in the lower division of the central body. A prominent group of four immunostained neurons had large cell bodies near the median ocellar nerve root and descending axonal fibers. These neurons are probably identical to previously identified primary commissure pioneer neurons of the locust brain. The apparent lack in the desert locust of certain histamine-immunoreactive neurons which were reported in the migratory locust may be responsible for differences in the physiological role of histamine between both species.The study was supported by the Deutsche Forschungsgemeinschaft, grants Ho 950/13 and 950/14     

Distribution of SchistoFLRFamide-like immunoreactivity in the adult ventral nervous system of the locust,Schistocerca gregaria

SchistoFLRFamide (PDVDHVFLRF-NH₂) is one of the major endogenous neuropeptides of the FMRF-amide family found in the nervous system of the locust,Schistocerca gregaria. To gain insights into the potential physiological roles of this neuropeptide we have examined the distribution of SchistoFLRFamide-like immunoreactivity in the ventral nervous system of adult locusts by use of a newly developed N-terminally specific antibody. SchistoFLRFamide-like immunoreactivity in the ventral nerve cord is found in a subgroup of the neurones that are immunoreactive to an antiserum raised against bovine pancreatic polypeptide (BPP). In the suboesophageal ganglion three groups of cells stain, including one pair of large posterior ventral cells. These cells are the same size, in the same location in the ganglion and have the same branching pattern as a pair of BPP immunoreactive cells known to innervate the heart and retrocerebral glandular complex of the locust. In the thoracic and abdominal ganglia two and three sets of cells, respectively, stain with both the SchistoFLRFamide and BPP antisera. In the abdominal ganglia the immunoreactive cells project via the median nerves to the intensely immunoreactive neurohaemal organs.     

Metamorphic changes in the central projects of hair sensilla inTenebrio molitor L. (Insecta: Coleoptera)

Summary This paper describes the afferent projections of hair sensilla of the pro- and mesothoracic legs and the lateral thoracic sclerites of larval and adultTenebrio molitor and the corresponding set of pupal hair sensilla. The sensory neurons that innervate the hair sensilla of larval or adult insects project somatotopically into the thoracic neuropil. Different types of sensilla on the same region of the body surface project to the same zone of the ipsilateral thoracic ventral neuropil but exhibit different arborization patterns. Although there is a profound reorganization of body surface sensilla, the basic somatotopic layout of the larva is maintained in the adult. The sensory neurons that innervate the pupal hair sensilla possess central projections similar to those of the corresponding adult sensory neurons. The central projections of pupal sensory neurons are somatotopically oriented. Their projection pattern is serially homologous in the thoracic and the abdominal ganglia. The central projection pattern of the described pupal sensory neurons is constant throughout pupation. MAb 22C10 immunoreactivity allows an estimate of the timing of the early differentiation of the imaginal sensory neurons originating during pupation. Ablation experiments indicate that pupal sensory neurons influence the central projection pattern of the differentiating imaginal sensory neurons.