Search

Johnston´s organ has been shown to act as an antennal auditory organ across a spectrum of insect species. In the hemimetabolous desert locust Schistocerca gregaria, Johnston´s organ must be functional on hatching and so develops in the pedicellar segment of the antenna during embryogenesis. Here we employ the epithelial cell marker Lachesin to identify the pedicellar domain of the early embryonic antenna, and then triple-label against Lachesin, the mitosis marker phosphohistone-3, and neuron-specific horseradish peroxidase to reveal the sense-organ precursors for Johnston´s organ and their lineages. Beginning with a single progenitor at approximately a third of embryogenesis, additional precursors subsequently appear in both the ventral and dorsal pedicellar domains, each generating a lineage or clone. Lineage locations are remarkably conserved across preparations and ages, consistent with the epithelium possessing an underlying topographic coordinate system which determines the cellular organization of Johnston´s organ. By mid-embryogenesis twelve lineages are arranged circumferentially in the pedicel as in the adult structure. Each sense-organ precursor is associated with a smaller mitotically active cell from which the neuronal complement of each clone may derive. Neuron numbers within a clone increase in discrete steps with age, are invariant between clones and across preparations of a given age. At mid-embryogenesis each clone comprises five cells consolidated into a tightly bound cartridge. A long scolopale extends apically from each cartridge to an insertion point in the epithelium, and bundled axons project basally towards the brain. We review data from Drosophila to suggest mechanisms which might regulate the developmental program of Johnston´s organ in the locust.

Sensory and motor systems in insects with hemimetabolous development must be ready to mediate adaptive behavior directly on hatching from the egg. For the desert locust S. gregaria, cholinergic transmission from antennal sensillae to olfactory or mechanosensory centers in the brain requires that choline acetyltransferase (ChAT) and the vesicular acetylcholine transporter (vAChT) already be present in sensory cells in the first instar. In this study, we used immunolabeling to demonstrate that ChAT and vAChT are both expressed in sensory cells from identifiable sensilla types in the immature antennal nervous system. We observed ChAT expression in dendrites, neurites and somata of putative basiconic-type sensillae at the first instar stage. We also detected vAChT in the sensory axons of these sensillae in a major antennal nerve tract. We then examined whether evidence for cholinergic transmission is present during embryogenesis. Immunolabeling confirms that vAChT is expressed in somata typical of campaniform sensillae, as well as in small sensory cell clusters typically associated with either a large basiconic or coeloconic sensilla, at 99% of embryogenesis. The vAChT is also expressed in the somata of these sensilla types in multiple antennal regions at 90% of embryogenesis, but not at earlier (70%) embryonic stages. Neuromodulators are known to appear late in embryogenesis in neurons of the locust central complex, and the cholinergic system of the antenna may also only reach maturity shortly before hatching.

The tritocerebral commissure giant (TCG) of the grasshopper Schistocerca gregaria is one of the best anatomically and physiologically described arthropod brain neurons. A member of the so-called Ventral Giant cluster of cells, it integrates sensory information from visual, antennal and hair receptors, and synapses with thoracic motor neurons in order to initiate and regulate flight behavior. Its ontogeny, however, remains unclear. In this study, we use bromodeoxyuridine incorporation and cyclin labeling to reveal proliferative neuroblasts in the region of the embryonic brain where the ventral giant cluster is located. Engrailed labeling confirms the deutocerebral identity of this cluster. Comparison of soma locations and initial neurite projections into tracts of the striate deutocerebrum help identify the cells of the ventral cluster in both the embryonic and adult brain. Reconstructions of embryonic cell lineages suggest deutocerebral NB1 as being the putative neuroblast of origin. Intracellular dye injection coupled with immunolabeling against neuron-specific horseradish peroxidase is used to identify the VG1 (TCG) and VG3 neurons from the ventral cluster in embryonic brain slices. Dye injection and backfilling are used to document axogenesis and the progressive expansion of the dendritic arbor of the TCG from mid-embryogenesis up to hatching. Comparative maps of embryonic neuroblasts from several orthopteroid insects suggest equivalent deutocerebral neuroblasts from which the homologous TCG neurons already identified in the adult brain could originate. Our data offer the prospect of identifying further lineage-related neurons from the cluster and so understand a brain connectome from both a developmental and evolutionary perspective.

We have investigated the pattern of apoptosis in the antennal epithelium during embryonic development of the grasshopper Schistocerca gregaria. The molecular labels lachesin and annulin reveal that the antennal epithelium becomes subdivided into segment-like meristal annuli within which sensory cell clusters later differentiate. To determine whether apoptosis is involved in the development of such sensory cell clusters, we examined the expression pattern of the cell death labels acridine orange and TUNEL in the epithelium. We found stereotypic, age-dependent, wave-like patterns of cell death in the antenna. Early in embryogenesis, apoptosis is restricted to the most basal meristal annuli but subsequently spreads to encompass almost the entire antenna. Cell death then declines in more basal annuli and is only found in the tip region later in embryogenesis. Apoptosis is restricted throughout to the midregion of a given annulus and away from its border with neighboring annuli, arguing against a causal role in annular formation. Double-labeling for cell death and sensory cell differentiation reveals apoptosis occurring within bands of differentiating sensory cell clusters, matching the meristal organization of the apical antenna. Examination of the individual epithelial lineages which generate sensory cells reveals that apoptosis begins peripherally within a lineage and with age expands to encompass the differentiated sensory cell at the base. We conclude that complete lineages can undergo apoptosis and that the youngest cells in these lineages appear to die first, with the sensory neuron dying last. Lineage-based death in combination with cell death patterns in different regions of the antenna may contribute to odor-mediated behaviors in the grasshopper.

The antenna is a key sensory organ in insects. Factors which pattern its epithelium and the spacing of sensillae will play an important role in shaping its contribution to adaptive behavior. The antenna of the grasshopper S. gregaria has three major articulations: scape, pedicel, and flagellum. During postembryonic development, the flagellum lengthens as segments (so-called meristal annuli) are added at each molt. However, the five most apical annuli do not subdivide; thus, their epithelial domains must already be defined during embryogenesis. We investigated epithelial compartmentalization and its relationship to the developing primordial nervous system of the antenna by simultaneous immunolabeling against the epithelial cell surface molecule Lachesin, against neuron-specific horseradish peroxidase, and against the mitosis marker phospho-histone 3. We found that Lachesin is initially expressed in a highly ordered pattern of “rings” and a “sock” in the apical antennal epithelium of the early embryo. These expression domains appear in a stereotypic order and prefigure later articulations. Proliferative cells segregate into these developing domains and pioneer- and sensory-cell precursors were molecularly identified. Our study allows pioneer neurons, guidepost cells, and the earliest sensory cell clusters of the primordial nervous system to be allocated to their respective epithelial domain. As the apical-most five domains remain stable through subsequent development, lengthening of the flagellum must originate from more basal regions and is likely to be under the control of factors homologous to those which regulate boundary and joint formation in the antenna of Drosophila.

Panarthropods are typified by disparate grades of neurological organization reflecting a complex evolutionary history. The fossil record offers a unique opportunity to reconstruct early character evolution of the nervous system via exceptional preservation in extinct representatives. Here we describe the neurological architecture of the ventral nerve cord (VNC) in the upper-stem group euarthropod Chengjiangocaris kunmingensis from the early Cambrian Xiaoshiba Lagerstätte (South China). The VNC of C. kunmingensis comprises a homonymous series of condensed ganglia that extend throughout the body, each associated with a pair of biramous limbs. Submillimetric preservation reveals numerous segmental and intersegmental nerve roots emerging from both sides of the VNC, which correspond topologically to the peripheral nerves of extant Priapulida and Onychophora. The fuxianhuiid VNC indicates that ancestral neurological features of Ecdysozoa persisted into derived members of stem-group Euarthropoda but were later lost in crown-group representatives. These findings illuminate the VNC ground pattern in Panarthropoda and suggest the independent secondary loss of cycloneuralian-like neurological characters in Tardigrada and Euarthropoda.

In the early embryonic grasshopper, two pairs of sibling cells near the apex of the antenna pioneer its dorsal and ventral nerve tracts to the brain. En route, the growth cones of these pioneers contact a so-called base pioneer associated with each tract and which acts as a guidepost cell. Both apical and basal pioneers express stereotypic molecular labels allowing them to be uniquely identified. Although their developmental origins are largely understood, the fates of the respective pioneers remain unclear. We therefore employed the established cell death markers acridine orange and TUNEL to determine whether the apical and basal pioneers undergo apoptosis during embryogenesis. Our data reveal that the apical pioneers maintain a consistent molecular profile from their birth up to mid-embryogenesis, at which point the initial antennal nerve tracts to the brain have been established. Shortly after this the apical pioneers undergo apoptosis. Death occurs at a developmental stage similar to that reported elsewhere for pioneers in a leg – an homologous appendage. Base pioneers, by contrast, progressively change their molecular profile and can no longer be unequivocally identified after mid-embryogenesis. At no stage up to then do they exhibit death labels. If they persist, the base pioneers must be assumed to adopt a new role in the developing antennal nervous system.

The antennal nervous system of the grasshopper Schistocerca gregaria features two parallel axon tracts each established early in embryogenesis by discrete pairs of pioneer neurons located at the antennal tip and whose growth cones contact so-called base pioneers en route to the brain. Here we present two antennal phenotypes in which a stereotypic dysregulation of axogenesis in a given tract is observed when only the base pioneer associated with that pathway is missing, consistent with a role for this cell type in guided axogenesis. Dysregulation involves defasciculation and aberrant navigation by pioneer axons resulting in a missing or depleted primordial antennal nerve to the brain. The dysregulated phenotypes reveal that axogenesis in each pathway is regulated independently. Previously unseen discrepancies in the navigational decisions made by pioneer neurons which derive sequentially from the same mother cell demonstrate that these progeny have separate identities. Possible mechanisms for the dysregulated phenotypes are considered.

The twin nerve tracts of the antenna of the grasshopper Schistocerca gregaria are established early in embryogenesis by sibling pairs of pioneers which delaminate from the epithelium into the lumen at the antennal tip. These cells can be uniquely identified via their co-expression of the neuronal labels horseradish peroxidase and the lipocalin Lazarillo. The apical pioneers direct axons toward the antennal base where they encounter guidepost-like cells called base pioneers which transiently express the same molecular labels as the apical pioneers. To what extent the pioneer growth cones then progress into the brain neuropil proper, and what their targets there might be, has remained unclear. In this study, we show that the apical antennal pioneers project centrally beyond the antennal base first into the deutocerebral, and then into the protocerebral brain neuropils. In the protocerebrum, we identify their target circuitry as being identified Lazarillo-positive cells which themselves pioneer the primary axon scaffold of the brain. The apical and base antennal pioneers therefore form part of a molecularly contiguous pathway from the periphery to an identified central circuit of the embryonic grasshopper brain.

The antennae of the grasshopper Schistocerca gregaria possess a pair of nerve pathways which are established by so-called pioneer neurons early in embryonic development. Subsequently, sensory cell clusters mediating olfaction, flight, optomotor responses, and phase changes differentiate from the antennal epithelium at stereotypic locations and direct their axons onto those of the pioneers to then project to the brain. Early in embryonic development, before the antennae become cuticularized, immunolabeling can be used to follow axogenesis in these pioneers and sensory cells. At later stages, immunolabeling becomes problematical as the cuticle is laid down and masks internal antigen sites. In order to immunolabel the nervous system of cuticularized late embryonic and first instar grasshopper antennae, we modified a procedure known as sonication in which the appendage is exposed to ultrasound thereby rendering it porous to antibodies. Comparisons of the immunolabeled nervous system of sectioned and sonicated antennae show that the cellular organization of sensory clusters and their axon projections is intact. The expression patterns of neuron-specific, microtubule-specific, and proliferative cell-specific labels in treated antennae are consistent with those reported for earlier developmental stages where sonication is not necessary, suggesting that these molecular epitopes are also conserved. The method ensures reliable immunolabeling in intact, cuticularized appendages so that the peripheral nervous system can be reconstructed directly via confocal microscopy throughout development.

Neuropils in the cerebral midline of Panarthropoda exhibit a wide spectrum of neuroarchitectures – from rudimentary to highly elaborated – and which at first sight defy a unifying neuroarchitectural principle. Developmental approaches have shown that in model arthropods such as insects, conserved cellular and molecular mechanisms first establish a simple axon scaffold in the brain. However, to be adapted for adult life, this immature ground plan is transformed by a developmental process – known in the grasshopper as “fascicle switching” – in which subsets of neurons systematically redirect their growth cones at stereotypic locations across the brain midline. A topographic system of choice points along the transverse brain axis where axons decussate features in all panarthropods studied even though different modes of neurogenesis and varying degrees of neuropilar elaboration are involved. This suggests that the molecular mechanisms regulating choice point selection may be conserved. In combination with recent cladistic interpretations of arthropod phylogeny based on nuclear protein-coding sequences the data argue for this topographic decussation as having evolved early and being a conserved feature of the Panarthropoda. Differences in elaboration likely reflect both the extent to which neuropilar reorganization has progressed during development and the lifestyle of the individual organism.

The antennal nervous system of the grasshopper Schistocerca gregaria comprises two parallel pathways projecting to the brain, each pioneered early in embryogenesis by a pair of sibling cells located at the antennal tip. En route, the growth cones of pioneers from one pathway have been shown to contact a guidepost-like cell called the base pioneer. Its role in axon guidance remains unclear as do the cellular guidance cues regulating axogenesis in the other pathway supposedly without a base pioneer. Further, while the tip pioneers are known to delaminate from the antennal epithelium into the lumen, the origin of this base pioneer is unknown. Here, we use immunolabeling and immunoblocking methods to clarify these issues. Co-labeling against the neuron-specific marker horseradish peroxidase and the pioneer-specific cell surface glycoprotein Lazarillo identifies not only the tip pioneers but also a base pioneer associated with each of the developing antennal pathways. Both base pioneers co-express the mesodermal label Mes3, consistent with a lumenal origin, whereas the tip pioneers proved Mes3-negative confirming their affiliation with the ectodermal epithelium. Lazarillo antigen expression in the antennal pioneers followed a different temporal dynamic: continuous in the tip pioneers, but in the base pioneers, only at the time their filopodia and those of the tip pioneers first recognize one another. Immunoblocking of Lazarillo expression in cultured embryos disrupts this recognition resulting in misguided axogenesis in both antennal pathways.

The central complex represents one of the most conspicuous neuroarchitectures to be found in the insect brain and regulates a wide repertoire of behaviors including locomotion, stridulation, spatial orientation and spatial memory. In this review, we show that in the grasshopper, a model insect system, the intricate wiring of the fan-shaped body begins early in embryogenesis when axons from the first progeny of four protocerebral stem cells (called W, X, Y, Z respectively) in each brain hemisphere establish a set of tracts to the primary commissural system. Decussation of subsets of commissural neurons at stereotypic locations across the brain midline then establishes a columnar neuroarchitecture in the fan-shaped body which is completed during embryogenesis. Examination of the expression patterns of various neurochemicals in the central complex including neuropeptides, a neurotransmitter, and the gas nitric oxide, show that these appear progressively and in a substance-specific manner during embryogenesis. Each neuroactive substance is expressed by neurons located at stereotypic locations in a given central complex lineage, confirming that the stem cells are biochemically multipotent. The organization of axons expressing the various neurochemicals within the central complex is topologically related to the location, and hence birthdate, of the neurons within the lineages. The neurochemical expression patterns within the fan-shaped body are layered, and so reflect the temporal topology present in the lineages. This principle relates the neuroanatomical to the neurochemical architecture of the central complex and so may provide insights into the development of adaptive behaviors.

The nervous system of the antenna of the grasshopper Schistocerca gregaria consists of two nerve tracts in which sensory cells project their axons to the brain. Each tract is pioneered early in embryogenesis by a pair of identified cells located apically in the antennal lumen. The pioneers are thought to originate in the epithelium of the antenna and then delaminate into the lumen where they commence axogenesis. However, unambiguous molecular identification of these cells in the epithelium, of an identifiable precursor, and of their mode of generation has been lacking. In this study, we have used immunolabeling against neuron-specific horseradish peroxidase and against Lachesin, a marker for differentiating epithelial cells, in combination with the nuclear stain DAPI, to identify the pioneers within the epithelium of the early embryonic antenna. We then track their delamination into the lumen as differentiated neurons. The pioneers are not labeled by the mesodermal/mesectodermal marker Mes3, consistent with an epithelial (ectodermal) origin. Intracellular dye injection, as well as labeling against the mitosis marker phospho-histone 3, identifies precursor cells in the epithelium, each associated with a column of cells. Culturing with the S-phase label 5-ethynyl-2'-deoxyuridine (EdU) shows that both a precursor and its column have incorporated the label, confirming a lineage relationship. Each set of pioneers can be shown to belong to a separate lineage of such epithelial cells, and the precursors remain and are proliferative after generating the pioneers. Analyses of mitotic spindle orientation then enable us to propose a model in which a precursor generates its pioneers asymmetrically via self-renewal.

The central complex of the grasshopper (Schistocerca gregaria) brain comprises a modular set of neuropils, which develops after mid-embryogenesis and is functional on hatching. Early in embryogenesis, Repo-positive glia cells are found intermingled among the commissures of the midbrain, but then redistribute as central complex modules become established and, by the end of embryogenesis, envelop all midbrain neuropils. The predominant glia associated with the central body during embryogenesis are glutamine synthetase-/Repo-positive astrocyte-like glia, which direct extensive processes (gliopodia) into and around midbrain neuropils. We used intracellular dye injection in brain slices to ascertain whether such glia are dye-coupled into a communicating cellular network during embryogenesis. Intracellular staining of individual cells located at any one of four sites around the central body revealed a population of dye-coupled cells whose number and spatial distribution were stereotypic for each site and comparable at both 70 and 100 % of embryogenesis. Subsequent immunolabeling confirmed these dye-coupled cells to be astrocyte-like glia. The addition of n-heptanol to the bathing saline prevented all dye coupling, consistent with gap junctions linking the glia surrounding the central body. Since dye coupling also occurred in the absence of direct intersomal contacts, it might additionally involve the extensive array of gliopodia, which develop after glia are arrayed around the central body. Collating the data from all injection sites suggests that the developing central body is surrounded by a network of dye-coupled glia, which we speculate may function as a positioning system for the developing neuropils of the central complex.

The neurons of the insect brain derive from neuroblasts which delaminate from the neuroectoderm at stereotypic locations during early embryogenesis. In both grasshopper and Drosophila, each developing neuroblast acquires an intrinsic capacity for neuronal proliferation in a cell autonomous manner and generates a specific lineage of neural progeny which is nearly invariant and unique. Maps revealing numbers and distributions of brain neuroblasts now exist for various species, and in both grasshopper and Drosophila four putatively homologous neuroblasts have been identified whose progeny direct axons to the protocerebral bridge and then to the central body via an equivalent set of tracts. Lineage analysis in the grasshopper nervous system reveals that the progeny of a neuroblast maintain their topological position within the lineage throughout embryogenesis. We have taken advantage of this to study the pioneering of the so-called w, x, y, z tracts, to show how fascicle switching generates central body neuroarchitecture, and to evaluate the roles of so-called intermediate progenitors as well as programmed cell death in shaping lineage structure. The novel form of neurogenesis involving intermediate progenitors has been demonstrated in grasshopper, Drosophila and mammalian cortical development and may represent a general strategy for increasing brain size and complexity. An analysis of gap junctional communication involving serotonergic cells reveals an intrinsic cellular organization which may relate to the presence of such transient progenitors in central complex lineages.

In this study we employed the expression of the astrocyte-specific enzyme glutamine synthetase, in addition to the glia-specific marker Repo, to characterize glia cell types associated with the embryonic development of the central complex in the grasshopper Schistocerca gregaria. Double labeling experiments reveal that all glutamine synthetase-positive cells associated with the central complex are also Repo-positive and horseradish peroxidase-negative, confirming they are glia. Early in embryogenesis, prior to development of the central complex, glia form a continuous population extending from the pars intercerebralis into the region of the commissural fascicles. Subsequently, these glia redisperse to envelop each of the modules of the central complex. No glial somata are found within the central complex neuropils themselves. Since glutamine synthetase is expressed cortically in glia, it allows their processes as well as their soma locations to be visualized. Single cell reconstructions reveal one population of glia as directing extensive ensheathing processes around central complex neuropils such as the central body, while another population projects columnar-like arborizations within the central body. Such arborizations are only seen in central complex modules after their neuroarchitecture has been established suggesting that the glial arborizations project onto a prior scaffold of neurons or tracheae.

All eight neuroblasts from the pars intercerebralis of one protocerebral hemisphere whose progeny contribute fibers to the central complex in the embryonic brain of the grasshopper Schistocerca gregaria generate serotonergic cells at stereotypic locations in their lineages. The pattern of dye coupling involving these neuroblasts and their progeny was investigated during embryogenesis by injecting fluorescent dye intracellularly into the neuroblast and/or its progeny in brain slices. The tissue was then processed for anti-serotonin immunohistochemistry. A representative lineage, that of neuroblast 1-3, was selected for detailed study. Stereotypic patterns of dye coupling were observed between progeny of the lineage throughout embryogenesis. Dye injected into the soma of a serotonergic cell consistently spread to a cluster of between five and eight neighboring non-serotonergic cells, but never to other serotonergic cells. Dye injected into a non-serotonergic cell from such a cluster spread to other non-serotonergic cells of the cluster, and to the immediate serotonergic cell, but never to further serotonergic cells. Serotonergic cells tested from different locations within the lineage repeat this pattern of dye coupling. All dye coupling was blocked on addition of an established gap junctional blocker (n-heptanol) to the bathing medium. The lack of coupling among serotonergic cells in the lineage suggests that each, along with its associated cluster of dye-coupled non-serotonergic cells, represents an independent communicating pathway (labeled line) to the developing central complex neuropil. The serotonergic cell may function as the coordinating element in such a projection system.

We have examined the developmental expression of the neuromodulators locustatachykinin, leucokinin-1, allatostatin and serotonin in a subset of lineages (Y, Z) of the central complex in the brain of the grasshopper Schistocerca gregaria. First, we show that all these neuromodulators are expressed in the same lineages during embryogenesis. The neuroblasts generating these lineages are therefore biochemically multipotent. Second, the neurons expressing the different neuromodulators are found clustered at stereotypic locations in their respective lineages. Locustatachykinin and leucokinin-1 map to the apical region of the lineage, allatostatin medially and serotonin to the base of the lineage. Since the location in these lineages translates into their birth order, we have been able ontogenetically to analyse their biochemical expression patterns. The age-profile within a lineage reveals that locustatachykinin- and leucokinin-1-expressing neurons are born first, then allatostatin neurons and finally serotoninergic neurons. Co-expression has been tested for serotonin with locustatachykin, leucokinin-1 or allatostatin and is negative but is positive for locustatachykinin and leucokinin-1, consistent with the stereotypic location of cells in the lineages. The delay between the birth of a neuron and the expression of its neuromodulator is stereotypic for each substance. Combined with a known birth date, this delay translates into a developmental expression pattern for the central complex itself.

We have investigated the ontogenetic basis of locustatachykinin-like expression in a group of cells located in the pars intercerebralis of the grasshopper midbrain. These cells project fibers to the protocerebral bridge and the central body via a characteristic set of fiber bundles called the w, x, y, z tracts. Lineage analyses associate the immunoreactive cells with one of four neuroblasts (termed W, X, Y, Z) in each protocerebral hemisphere of the early embryo. Locustatachykinin is a ubiquitous myotropic peptide among the insects and its expression in the pars intercerebralis begins at approximately 60-65% of embryogenesis. This coincides with the appearance of the columnar neuroarchitecture characteristic of the central body. The number of immunoreactive cells in a given lineage is initially small, increases significantly in later embryogenesis, and attains the adult situation (about 7% of a lineage) in the first larval instar after hatching. Although each neuroblast generates progeny displaying a spectrum of cell body sizes, there is a clear morphological gradient, which reflects birth order within the lineage. Locustatachykinin expressing cells are located stereotypically at or near the tip of their lineage, which an age profile reveals places them amongst the first born progeny of their respective neuroblasts. Although these neuroblasts begin to generate progeny at approximately 25-27% of embryogenesis, their daughter cells remain quiescent with respect to locustatachykinin expression for over 30% of embryogenesis.

The grasshopper antenna is an articulated appendage associated with the deutocerebral segment of the head. In the early embryo, the meristal annuli of the antenna represent segment borders and are also the site of differentiation of pioneer cells which found the dorsal and ventral peripheral nerve tracts to the brain. We report here on another set of cells which appear earlier than the pioneers during development and are later found arrayed along these tracts at the border of epithelium and lumen. These so-called nerve tract associated cells differ morphologically from pioneers in that they are bipolar, have shorter processes, and are not segmentally organized in the antenna. Nerve tract associated cells do not express horseradish peroxidase and so are not classical neurons. They do not express antigens such as repo and annulin which are associated with glia cells in the nervous system. Nerve tract associated cells do, however, express the mesodermal/mesectodermal cell surface marker Mes-3 and putatively derive from the antennal coelom and then migrate to the epithelium/lumen border. Intracellular recordings show that such nerve tract associated cells have resting potentials similar to those of pioneer cells and can be dye coupled to the pioneers. Similar cell types are present in the maxilla, a serially homologous appendage on the head. The nerve tract associated cells are organized into a cellular scaffold which we speculate may be relevant to the navigation of pioneer and sensory axons in the early embryonic antennal nervous system.

The central complex is a brain specific structure involved in multimodal information processing and in coordinating motor behaviour. It possesses a highly organized neuroarchitecture, which is remarkably conserved across insect species. A prominent feature of this neuroarchitecture is the stereotypic projection of axons from clusters of neurons in the pars intercerebralis to the central body via the so-called w, x, y and z tracts. Despite extensive analyses of this neuroarchitecture in adults, little is known about its ontogeny in any insect. In this paper we use the expression pattern of the segment polarity gene engrailed to identify those neuroblasts belonging to the protocerebrum of the early embryonic brain of the grasshopper Schistocerca gregaria. We present a new map for this brain region in which the 95 protocerebral neuroblasts in each hemisphere are organized into seven rows, as they are in the neuromeres of the ventral nerve cord. We then identify a subset of four of these neuroblasts as being the progenitor cells for four clusters of neurons, some of whose axons we show project via discrete tracts (w, x, y, z) into the central complex. These tracts begin to form prior to 39% of embryogenesis. We show further, that the cells from one of these clusters (the Z cluster) are organized according to age, and direct axons topologically according to age into the appropriate z tract. This pattern is repeated in each of the other three clusters, thus establishing a clonally based modular system of fibre tracts consistent with the model proposed for this brain region in the adult.

In this paper, we propose an ontogeny for previously identified cells from the median domain in the midline of the embryonic brain of the grasshopper Schistocerca gregaria. The so-called lateral cells (LCs) are characteristically located laterally within the median domain at its border with the protocerebral hemispheres. The LC occurs singly and can be identified in the early embryo on the basis of their expression of the cell surface lipocalin Lazarillo. Using immunocytochemical, dye injection, electron microscopical and histological methods, we show that these LC are neurons and derive as postmitotic cells directly from the epithelium of the median domain. Further, they and the other identified cells of the median domain such as the protocerebral commissure pioneers (PCP), co-express the Mes-3 antigen, consistent with a derivation from the mesectodermal germ layer of the embryo. Subsequent to axogenesis, electron microscopy reveals that these Mes-3-expressing LC fasciculate with the co-expressing PCPs within the developing protocerebral commissure. We present a model for the origin of all these cells based on histological data and bromodeoxyuridine incorporation. The model suggests a delamination of cells from the mesectoderm followed by a migration to their ultimate sites within the median domain.

The primary axon scaffold of the insect brain is established early in embryogenesis and comprises a preoral protocerebral commissure, a postoral tritocerebral commissure and longitudinal fiber pathways linking the two. In both grasshopper and fly its form is approximately orthogonal and is centered around the stomodeum. We show how pioneer fibers from the protocerebrum and tritocerebrum cross the brain midline directly via their respective commissures. The deutocerebrum, however, lacks its own commissure and we describe how deutocerebral pioneers circumnavigate the gut to cross the midline either via the protocerebral commissure or the tritocerebral commissure. In contrast to all other commissures of the central nervous system, the protocerebral commissure persists, albeit in reduced form, in the commissureless mutation in the fly. Besides the com gene, a further, as yet unidentified, mechanism must regulate this commissure. The formation of the tritocerebral commissure involves labial, a member of the Hox gene group. Genetic rescue experiments in labial mutants reveal that the formation of this commissure can be rescued by all other Hox genes except Abdominal-B. However, only in the labial and Deformed null mutants are the commissures associated with the respective expression domains (tritocerebral, mandibular, respectively) absent. This suggests that the molecular mechanisms regulating postoral brain commissure formation are distinct from those in the neuromeres of the ventral nerve cord.

Our analysis of head segmentation in the locust embryo reveals that the labrum is not apical as often interpreted but constitutes the topologically fused appendicular pair of appendages of the third head metamere. Using molecular, immunocytochemical and retrograde axonal staining methods we show that this metamere, the intercalary segment, is innervated by the third brain neuromere-the tritocerebrum. Evidence for the appendicular nature of the labrum is firstly, the presence of an engrailed stripe within its posterior epithelium as is typical of all appendages in the early embryo. Secondly, the labrum is innervated by a segmental nerve originating from the third brain neuromere (the tritocerebrum). Immunocytochemical staining with Lazarillo and horseradish peroxidase antibodies reveal that sensory neurons on the labrum contribute to the segmental (tritocerebral) nerve via the labral nerve in the same way as for the appendages immediately anterior (antenna) and posterior (mandible) on the head. All but one of the adult and embryonic motoneurons innervating the muscles of the labrum have their cell bodies and dendrites located completely within the tritocerebral neuromere and putatively derive from engrailed expressing tritocerebral neuroblasts. Molecular evidence (repo) suggests the labrum is not only appendicular but also articulated, comprising two jointed elements homologous to the coxa and trochanter of the leg.

The expression pattern of the segment polarity gene engrailed was studied at the single cell level in the protocerebrum of the early embryonic brain of the grasshopper Schistocerca gregaria, the neuromere containing the secondary headspot cells. The engrailed protein is first expressed in the protocerebrum at about 22% of embryogenesis by a group of identified neuroblasts bordering the antennal lobe. The number of immunoreactive neuroblasts increases up to 26% of embryogenesis and then rapidly declines so that by 30% only the three most posterior remain immunoreactive. These three neuroblasts become incorporated into the developing antennal lobe of the deutocerebrum. Subsequently, there is a progressive re-expression of the engrailed protein in the protocerebrum by the so-called six secondary headspot cells. These are the first born sibling progeny of three identified protocerebral neuroblasts which themselves expressed the engrailed protein prior to generating their lineages, and so represents a reacquisition of engrailed expression within identified clones. The secondary headspot cells are neurons which direct axonal processes into the developing optic tract and so contribute to the primary axon scaffold of the brain. From our analysis of their ontogeny, we conclude that the secondary headspot cells do not represent a segmental border in the brain.

The anterior midline of the brain, also known as the pars intercerebralis, contains the largest collection of neurosecretory cells in the central nervous system of the grasshopper. In this study, we use immunocytochemical, intracellular staining, and histological methods to establish the ontogenies of the various cell types in the brain midline, and show how these cells contribute to the pars intercerebralis of the adult brain. We show that the adult pars intercerebralis develops from three distinct embryonic cell groups: (1) the median neurosecretory cells, which derive from a subset of neuroblasts in the protocerebral hemispheres, and which project axons to the corpora cardiaca; (2) the paired primary commissure pioneers, which derive directly from the mesectoderm of the dorsal median domain and whose axons project to the ventral nerve cord via the midline tract; and (3) the six progeny of the median precursor in the dorsal median domain, which share a common axonal projection with the primary commissure pioneers. Since the adult pars intercerebralis is a fusion product of these independent cellular components, it can only be understood in terms of its origins in the embryonic brain. When the expression pattern of the TERM-1 antigen is compared in subsets of median neurosecretory cells in a wide range of insect orders, the results suggests a common organizational Bauplan for the pars intercerebralis. This hypothesis is supported by the identification of putative homologs of the grasshopper primary commissure pioneers in all these insects.

The Publisher regrets that this article is an accidental duplication of an article that has already been published, http://dx.doi.org10.1016/j.asd.2014.11.003. The duplicate article has therefore been withdrawn. The full Elsevier Policy on Article Withdrawal can be found at http://www.elsevier.com/locate/withdrawalpolicy.

The expression pattern of the engrailed protein was studied in neuroblasts which delaminate at the border of the protocerebrum and antennal lobe of the deutocerebrum in the early embryonic brain of the grasshopper. The antennal lobe is a complex structure comprising both glomerular and non-glomerular components, a cellular organization which distinguishes it from the striate-like neuropil comprising the remainder of the deutocerebrum. Early in embryogenesis engrailed expression in the protocerebrum is restricted to a compact block of neuroblasts located at its interface with the antennal lobe. Subsequently engrailed expression in these cells disappears in a stepwise manner from anterior to posterior so that by 37% of embryogenesis only a single row of three engrailed positive neuroblasts and their progeny remains. Contemporaneously engrailed expression reappears in a group of more anterior progeny deriving from neuroblasts which are no longer immunoreactive. The three remaining engrailed positive neuroblasts then become separated from their non-immunoreactive neighbours by an invagination of the perineurium called the lateral cleft and come to lie completely within the developing antennal lobe. These cells then direct columns of immunoreactive progeny centrifugally towards the centre of the lobe. Such a protocerebral contribution to the antennal lobe suggests that the evolution and ontogeny of this brain region need to be reconsidered.

The authors studied the contribution of seven clusters of Lazarillo-expressing cells to the primary axon scaffold of the brain in the grasshopper Schistocerca gregaria from 26% to 43% of embryogenesis. Each cluster, which was numbered according to when Lazarillo expression first appeared, was uniquely identifiable on the basis of its stereotypic position in the brain and the number of Lazarillo-expressing cells it contained. At no time during embryogenesis was Lazarillo expression found in brain neuroblasts: It was found only in progeny. For ease of analysis, axogenesis was followed in a cell cluster that contained only a single Lazarillo-expressing cell (the lateral cell) in the dorsal median domain of the brain midline. Bromodeoxyuridine incorporation revealed the presence of only a single midline precursor cell in this region during embryogenesis. Intracellular injection of Lucifer yellow into the lateral cell at various ages showed that there was no dye coupling to the midline precursor or to the nearby term-1-expressing primary commissure pioneers;The lateral cell is not related lineally to these cells and most likely differentiates directly from the neuroectoderm of the brain midline. Lazarillo expression appears at the onset of axogenesis as the lateral cell projects an axon laterally toward the next Lazarillo-expressing cell cluster. The cells of this target;cluster direct axons into separate brain regions, thereby establishing an orthogonally organized scaffold that the lateral cell axon follows as it navigates away from the brain midline. The primary axon scaffold of the brain results from a stepwise interlinking of discrete brain regions, as exemplified by axons from neighboring Lazarillo-expressing cell. clusters. J. Comp. Neurol. 419:394-405, 2000. (C) 2000 Wiley-Liss, Inc.

The bilaterally paired primary commissure pioneer neurons in the median domain of the grasshopper brain are large, descending interneurons that uniquely express the TERM-1 antigen, even in the adult. After pioneering the primary interhemispheric brain commissure, these neurons extend TERM-1-immunoreactive collaterals into most parts of the brain except the mushroom bodies. In this report, the authors show that the TERM-1 antigen is located in the cell body cytoplasm of these neurons and not on the membranes. Screening with antisera to insect neuropeptides reveals that an antiserum recognizing peptides of the leucokinin family labels the cell body cytoplasm of the primary commissure neurons. Leucokinin-related peptides are known to modulate motility of visceral muscle, play a role in diuresis, and are likely to be neuromodulators in the insect nervous system. The primary commissure neurons differ ultrastructurally from median neurosecretory cells in that their cell body cytoplasm is more extensive, contains high numbers of mitochondria and extensive endoplasmic reticulum, but does not contain neurosecretory granules. In the adult, the cell somata are enveloped by multiple glia membranes and associated trophospongia. According to these ultrastructural characteristics, the primary commissure pioneers are not classical neurosecretory cells.

Embryonic development in the median domain of the brain of the grasshopper Schistocerca gregaria was investigated with immunohistochemical, histological, and intracellular dye injection techniques. The early head midline is divisible into a dorsal median domain and a ventral median domain based on the orientation of cell somata in each region. At 25% of embryogenesis, a single large midline precursor differentiates in the dorsal median domain and produces a lineage of six neuronal progeny before degenerating. No further precursors arise. In addition, the primary commissure pioneers and a pair of lateral neurons differentiate directly from the ectoderm in this region. Lucifer yellow dye injected into the midline precursor stains only this cell and its progeny. Similarly, there is no dye coupling from the primary commissure pioneers to the midline lineage or to neuroblasts of the brain hemispheres. Neurogenesis in the dorsal median domain therefore proceeds separately within each subset of cells, and is not related to development in the brain hemispheres. Beginning at 42% of embryogenesis, the primary commissure pioneers undergo a morphological transformation and concomittantly express the Term-1 antigen. Expression continues throughout embryogenesis and into the adult, where the midline primary commissure pioneer cells are the only ones labeled by Term-1 in the entire brain. The cellular organization of the dorsal median domain therefore remains remarkably conserved throughout embryogenesis, even as the brain undergoes extensive morphological transformation.

Neuron 714 is morphologically one of the most prominent neurons in the central auditory pathway of the grasshopper with arborizations extending from the abdominal neuromeres of the metathoracic ganglion to the brain. The aim of this study is to explore auditory information flow involving neuron 714 at the level of the ventral nerve cord. Paired intracellular recordings were made from neuron 714 in the mesothorax on the one hand, and from candidate presynaptic auditory neurons of the metathorax on the other. Electrical stimulation of the tympanal nerves provides an estimate of the synaptic distance between these interneurons and auditory afferents. Four, including neuron 714, are monosynaptically connected to afferents, the remainder disynaptically. Current-injection and spike-triggered averaging reveal that of nine neurons examined, seven make either monosynaptic, disynaptic or polysynaptic connections onto neuron 714. All connections are excitatory. Paired recordings show that response duration and response amplitude in synaptically linked cells vary according to the frequency of the stimulus. Measurements of the latency of the first excitatory post-synaptic potential evoked in neuron 714 by afferents and by metathoracic interneurons show how the synaptic drive from these sources shapes the auditory response of neuron 714.

Distortion-product otoacoustic emissions were recorded from the tympanum of the grasshopper, Locusta migratoria. The hearing organ of this insect is in direct contact with the tympanum and does not contain sensory hair cells. 2f1−f2 distortions were measured for stimulus frequencies between 2–70 kHz. For frequencies between 3–9 kHz, the level of 2f1–f2 was 30–50 dB below the stimulus level. 2f1−f2 threshold curves calculated from distortion growth functions at different f2 frequencies are most sensitive between 3–9 kHz. These thresholds match the auditory sensitivity of low frequency receptor neurons in the ear [Romer, J. Comp. Physiol. 109, 101–122 (1976)]. In contrast to vertebrates, the dependence of the 2f1−f2 level on the frequency ratio f2/f1 did not show distinct maxima for most f2 frequencies. The distortion levels were largest for small ratios close to 1. The behavior of 2f1−f2 was significantly different for stimulus frequencies below and above 10 kHz. Below 10 kHz, the thresholds were more sens...

We have studied the embryonic development of the pars intercerebralis/central complex in the brain of the grasshopper using immunocytochemical and histochemical techniques. Expression of the cell-surface antigen lachesin reveals that the neuroblasts of the pars intercerebralis first differentiate from the neuroectoderm at around 26% of embryogenesis. Differentiation of medial and lateral neuroblasts occurs first. By the 28% stage a more or less uniform sheet of 20 neuroblasts has formed. As a result of both cell proliferation and cell translocation, the pars intercerebralis proliferative cluster in each hemisphere expands so that at 30% the most medial neuroblasts lie apposed at the midline. We followed the further development of the pars intercerebralis of each brain hemisphere using bromo-deoxy-uridine incorporation and osmium-ethyl-gallate staining. Within the pars intercerebralis itself, the neuroblasts redistribute into discrete subsets. The neuroblasts of each subset generate clusters of progeny which extend in a stereotypic, subset-specific direction in the brain. We have used this feature to identify one subset of four neuroblasts as being the likely progenitor cells for four clusters of embryonic neurons (W, X, Y, Z) which develop at around 55% of embryogenesis. We show that these progeny project axons via four discrete fascicles (w, x, y, z) into the embryonic central complex. At the single cell level, Golgi impregnation reveals that the axons from these neighbouring cell clusters remain discrete, and those from the same cluster tightly fasciculated, as they project into the central complex, consistent with a modular organization for this brain region.

Injection of fluorochromes such as Alexa Fluor(®) 568 into single cells in brain slices reveals a network of dye-coupled cells to be associated with the central complex. Subsequent immunolabeling shows these cells to be repo positive/glutamine synthetase positive/horseradish peroxidase negative, thus identifying them as astrocyte-like glia. Dye coupling fails in the presence of n-heptanol indicating that dye spreads from cell to cell via gap junctions. A cellular network of dye-coupled, astrocyte-like, glia surrounds and infiltrates developing central complex neuropils. Intracellular dye injection techniques complement current molecular approaches in analyzing the functional properties of such networks.

Injection of fluorochromes such as Alexa Fluor(®) 568 into single cells in brain slices reveals a network of dye-coupled cells to be associated with the central complex. Subsequent immunolabeling shows these cells to be repo positive/glutamine synthetase positive/horseradish peroxidase negative, thus identifying them as astrocyte-like glia. Dye coupling fails in the presence of n-heptanol indicating that dye spreads from cell to cell via gap junctions. A cellular network of dye-coupled, astrocyte-like, glia surrounds and infiltrates developing central complex neuropils. Intracellular dye injection techniques complement current molecular approaches in analyzing the functional properties of such networks.

This study employs labels for cell proliferation and cell death, as well as classical histology to examine the fates of all eight neural stem cells (neuroblasts) whose progeny generate the central complex of the grasshopper brain during embryogenesis. These neuroblasts delaminate from the neuroectoderm between 25 and 30 % of embryogenesis and form a linear array running from ventral (neuroblasts Z, Y, X, and W) to dorsal (neuroblasts 1-2, 1-3, 1-4, and 1-5) along the medial border of each protocerebral hemisphere. Their stereotypic location within the array, characteristic size, and nuclear morphologies, identify these neuroblasts up to about 70 % of embryogenesis after which cell shrinkage and shape changes render progressively more cells histologically unrecognizable. Molecular labels show all neuroblasts in the array are proliferative up to 70 % of embryogenesis, but subsequently first the more ventral cells (72-75 %), and then the dorsal ones (77-80 %), cease proliferation. By contrast, neuroblasts elsewhere in the brain and optic lobe remain proliferative. Apoptosis markers label the more ventral neuroblasts first (70-72 %), then the dorsal cells (77 %), and the absence of any labeling thereafter confirms that central complex neuroblasts have exited the cell cycle via programmed cell death. Our data reveal appearance, proliferation, and cell death proceeding as successive waves from ventral to dorsal along the array of neuroblasts. The resulting timelines offer a temporal blueprint for building the neuroarchitecture of the various modules of the central complex.

We have studied the morphogenetic reorganization that occurs in the grasshopper brain during embryogenesis. We find that morphogenetic movements occur at three organizational levels during brain development. First, the entire developing brain changes its orientation with respect to the segmental chain of ventral ganglia. A 90 degrees shift in the attitude of the brain neuraxis occurs during embryogenesis due to a gradual upward movement of the cerebral structures in the head. Second, the clusters of proliferating neuroblasts and progeny that generate the neuroarchitecture of the mature brain move relative to one another and to nonneural structures such as the stomodeum. This is especially pronounced for the pars intercerebralis and for the tritocerebrum, as shown by annulin and engrailed immunoreactivity. Third, individual neuroblasts within a given proliferative cluster undergo positional reorganization during embryogenesis. Identified neuroblasts of the tritocerebrum and the pars intercerebralis are displaced within the brain. We conclude that the transformation of the simple sheet-like structure of the early embryonic brain into the highly differentiated structure of the mature brain involves a series of morphogenetic movements that occur in virtually all parts of the brain.

We have studied a group of midline cells in the embryonic brain of the grasshopper by using immunocytochemical and intracellular dye injection techniques. This cluster of midline cells differentiates between the pars intercerebralis lobes of the protocerebrum during early embryogenesis, and is composed of putative midline progenitors as well as neuronal and glial cells. Annulin immunoreactive glial processes surround the borders of the midline cell cluster and also form a network of processes extending from there to the borders of proliferative clusters in the brain hemispheres. Among the cells that derive from the midline cluster are two bilaterally symmetrical pairs of identified primary commissure pioneer neurons. By navigating along the glial bound borders of the midline proliferative cluster, the axons of these pioneers establish an initial axonal bridge across the brain midline. This analysis identifies a glial-bound midline proliferative cluster in the brain and shows that neuronal and glial cells of this cluster are closely associated with neurons pioneering the primary brain commissure. Comparable features of midline cells in the ventral ganglia and similarities to other proliferative clusters in the brain hemispheres are discussed.

Axogenesis in the embryonic brain was studied at the single cell level in the grasshopper Schistocerca gregaria. A small set of individually identifiable pioneer neurons establishes a primary axon scaffold during early embryogenesis. At the beginning of scaffold formation, pioneering axons navigate along and between glial borders that surround clusters of proliferating neuroblasts. In each brain hemisphere, an axonal outgrowth cascade involving a series of pioneer neurons establishes a pathway from the optic ganglia to the brain midline. At the midline the primary preoral commissural interconnection in the embryonic brain is pioneered by a pair of midline-derived pioneer neurons. A second preoral commissural connection is pioneered by two pairs of pars intercerebralis pioneer neurons. Descending tracts are pioneered by the progeny of identified neuroblasts in the pars intercerebralis, deutocerebrum and tritocerebrum; the postoral tritocerebral commissure is pioneered by a pair of tritocerebral neurons. All of the pioneering brain neurons express the cell adhesion molecule fasciclin I during initial axon outgrowth and fasciculation. Once established, the primary axon scaffold of the brain is used for fasciculation by subsequently differentiating neurons and, by the 40% stage of embryogenesis, axonal projections that characterize the mature brain become evident. The single cell analysis of grasshopper brain development presented here sets the stage for manipulative cell biological experiments and provides the basis for comparative molecular genetic studies of embryonic brain development in Drosophila.

We have investigated the pattern of glia associated with central complex lineages in the embryonic brain of the grasshopper Schistocerca gregaria. Using the glia-specific marker Repo, we identified glia associated externally with such lineages, termed lineage-extrinsic glia, and glia located internally within the lineages, termed lineage-intrinsic glia. Populations of both glial types increase up to 60 % of embryogenesis, and thereafter decrease. Extrinsic glia change their locations over time, while intrinsic ones are consistently found in the more apical part of a lineage. Apoptosis is not observed for either glial type, suggesting migration is a likely mechanism accounting for changes in glial number. Proliferative glia are present both within and without individual lineages and two glial clusters associated with the lineages, one apically and the other basally, may represent sources of glia.

Central complex modules in the postembryonic brain of the grasshopper Schistocerca gregaria are enveloped by Repo-positive/glutamine-synthetase-positive astrocyte-like glia. Such cells constitute Rind-Neuropil Interface glia. We have investigated the postembryonic development of these glia and their anatomical relationship to axons originating from the w, x, y, z tract system of the pars intercerebralis. Based on glutamine synthetase immunolabeling, we have identified four morphological types of cells: bipolar type 1 glia delimit the central body but only innervate its neuropil superficially; monopolar type 2 glia have a more columnar morphology and direct numerous gliopodia into the neuropil where they arborize extensively; monopolar type 3 glia are found predominantly in the region between the noduli and the central body and have a dendritic morphology and their gliopodia project deeply into the central body neuropil where they arborize extensively; multipolar type 4 glia link the central body neuropil with neighboring neuropils of the protocerebrum. These glia occupy type-specific distributions around the central body. Their gliopodia develop late in embryogenesis, elongate and generally become denser during subsequent postembryonic development. Gliopodia from putatively type 3 glia within the central body have been shown to lie closely apposed to individual axons of identified columnar fiber bundles from the w, x, y, z tract system of the central complex. This anatomical association might offer a substrate for neuron/glia interactions mediating postembryonic maturation of the central complex.

The Indian antHarpegnathos saltator may be unique among insects in using its jumping capacity not only as an escape mechanism but also as a normal means of locomotion, and for catching its prey in flight. High-speed cinematography used to analyse the various phases of the jump suggests thatHarpegnathos employs a novel jumping mechanism to mediate these behaviours: namely the synchronous activation of its middle and hindlegs. Electrophysiological recordings from muscles or nerves in pairs of middle and hindlegs show remarkably synchronous activity during fictive jumping, supporting the synchronous activation hypothesis.Harpegnathos is not the only ant to jump, and a cladistic analysis suggests that jumping behaviour evolved independently three times during ant evolutionary history.

The central complex of the insect brain is a remarkably miniaturized but highly complex multimodal information-processing network. Recent work on central complex development in Drosophila and grasshopper reveals that the cells comprising its complex circuitry are generated by a surprisingly small number of primary progenitors. Of these, four identified neural stem cells generate a large number of neurons through a novel mode of neurogenesis that involves self-renewing intermediate progenitor cells. Interestingly, a comparable mode of amplification of proliferation also operates in the developing mammalian cortex; this could be a general strategy for increasing brain size and complexity. Although this type of proliferation generates a large number of progeny, it is also prone to dysregulation, resulting in brain tumors. Thus, furthering our knowledge of the development of the central complex is likely to be valuable not only for understanding brain complexity but could also have important implications for identifying developmental pathways that go awry during tumor formation.

The brain (supraoesophageal ganglion) is the most complex of the segmental ganglia composing the nerve cord of the locusts Schistocerca gregaria and Locusta migratoria. In this paper, we describe the ground plan of the commissures crossing the midline of the brain and propose a nomenclature with the aim of making a complex neuropil more understandable at the level of individual neurons. For developmental and comparative reasons the neuroarchitecture of the brain is related to the neural axis, not to the body axis. We have identified 73 commissural fiber bundles belonging to the adult brain, and these are named according to their location (ventral, dorsal, anterior, posterior, medial) with respect to the central complex as reference point. Reconstructions of identified neurons from intracellular stainings, cobalt backfills, or immunohistochemical studies demonstrate the various configurations in which fibers cross the brain.

The groundplan of longitudinal fiber tracts and commissures comprising the thoracic CNS is remarkably similar throughout a diverse group of holo- and hemimetabolous insects. Among tympanate insects, the axons of tympanal (auditory) receptors project to equivalent tracts and association areas in the central nervous system, even when the ears are located on different parts of the body. In segments without tympanal receptors, mechanoreceptive pleural chordotonal afferents are found to project to these neuropilar areas. Consistent with this, the tympanal receptors of the grasshopper have been shown to be serially homologous with those of the pleural chordotonal organs arrayed along the body wall. In atympanate insects, mechanosensitive receptors project to the equivalent neuropilar areas, via the same routes, as do tympanal receptors in insects with ears. Again, the auditory receptors of tympanate insects such as the grasshopper have been shown to be homologous with the chordotonal receptors of atympanate insects such as Drosophila. Identified first-order interneurons which have their dendrites in areas of neuropil occupied by tympanal afferents have been shown to be serially homologous along the CNS of the grasshopper. Putative homologs to these interneurons have also been identified in a range of other tympanate and atympanate insects. It is proposed that the tympanal receptors are a special part of a general chordotonal system in which homologous peripheral receptors project to sets of homologous interneurons.

The central complex is a major integrative region within the insect brain with demonstrated roles in spatial orientation, the regulation of locomotor behavior, and sound production. In the hemimetabolous grasshopper, the central complex comprises the protocerebral bridge, central body (CB), ellipsoid body, noduli, and accessory lobes, and this modular organization develops entirely during embryogenesis. From a biochemical perspective, a range of neuroactive substances has been demonstrated in these modules of the adult central complex, but little is known about their developmental expression. In this study, we use matrix-assisted laser desorption/ionization-imaging mass spectrometry on single brain slices to confirm the presence of several peptide families (tachykinin, allatostatin, periviscerokinin/pyrokinin, FLRFamide, and neuropeptide F) in the adult central complex and then use immunohistochemistry and histology to examine their developmental expression, together with that of the indolamin serotonin, and the endogenous messenger nitric oxide (NO; via its synthesizing enzyme). We find that each neuromodulator is expressed according to a unique, stereotypic, pattern within the various modules making up the central complex. Neuropeptides such as tachykinin (55%) and allatostatin (65%), and the NO-synthesizing enzyme diaphorase (70%), are expressed earlier during embryonic development than the biogenic amine serotonin (80%), whereas periviscerokinin-like peptides and FLRFamide-like peptides begin to be expressed only postembryonically. Within the CB, these neuroactive substances are present in tangential projection neurons before they appear in columnar neurons. There is also no colocalization of serotonin-positive and peptide-positive projections up to the third larval instar during development, consistent with the clear dorsoventral layering of the neuropil we observe. Our results provide the first neurochemical fingerprint of the developing central complex in an hemimetabolous insect.