© 2001 by The Society for Integrative and Comparative Biology
The Neuroanatomy of Nitric Oxide–Cyclic GMP Signaling in the Locust: Functional Implications for Sensory Systems1
1 School of Biological Sciences, Queen Mary, University of London, Mile End Road, London E1 4NS, UK
2 Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONCOMPARATIVE ANATOMY OF NO-CYCLIC...FUNCTIONAL IMPLICATIONSReferences |
|---|
Recent studies have investigated the source and target neurons for the diffusible neuronal messenger molecule nitric oxide (NO) in the nervous system of the locust. Here we compare the neuroarchitecture of NO signaling between different sensory systems. The available neuroanatomical data implicate NO in sensory processing for modalities as diverse as mechanoreception, vision, olfaction, gustation and hearing. All respective first-order sensory neuropils are innervated by NOS-containing interneurons. The corresponding sensory receptor neurons lack NOS but seem to express soluble guanylyl cyclase (sGC), the main receptor molecule for NO in the nervous system. The axonal projections of sensory neurons must therefore be considered the primary target of NO in these sensory neuropils. An exception is the antennal olfactory system where sGC is apparently expressed in interneurons, in partial colocalization with NOS.
We discuss these anatomical findings in relation to the spatiotemporal characteristics of NO signaling. Many sensory neuropils are organized into maps that reflect neuronal response properties (i.e., tuning or receptive fields). A local release of NO within such maps will therefore most strongly affect neurons with similar coding properties. If sensory receptor activity triggers NO synthesis locally in the map, this mechanism could link groups of similarly tuned receptors dynamically according to stimulus intensity. Furthermore, we explore the functional implications of differences between sensory systems in the anatomy of NOS-expressing interneurons, using the compound eye and the thoracic tactile system as examples.
|
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONCOMPARATIVE ANATOMY OF NO-CYCLIC...FUNCTIONAL IMPLICATIONSReferences |
|---|
Unlike "conventional" neurotransmitters, the intercellular signaling molecule nitric oxide (NO) diffuses through cell membranes and can thus exert effects intracellularly without interacting with membrane-associated receptors. Probably the most important effector protein for NO in the nervous system is the cytosolic enzyme known as soluble guanylyl cyclase (sGC) which generates the second messenger cyclic GMP (cGMP). It is important to keep in mind, however, that NO can interact with a large number of other biomolecules. Such non-sGC-mediated effector pathways (Stamler, 1994
; Stamler et al., 1997) are to be expected and may add further complexity to NO signaling in insects.
The downstream effects of the NO-cGMP signaling pathway range from changes in ion channel properties to changes in gene expression. In this article we will focus on NO-cGMP signaling in sensory systems and most relevant here are effects of NO on synaptic transmission (reviewed in Schuman and Madison, 1994), a role recently confirmed for invertebrates in physiological experiments. In Drosophila larvae, NO donors and cGMP analogues induce vesicle release at the neuromuscular junction (Wildemann and Bicker, 1999). In the crayfish terminal ganglion, drugs acting on the NO-cGMP pathway modulate connections between mechanosensory afferents and their postsynaptic interneurons. Intriguingly, in this system NO either enhances or decreases the postsynaptic potentials, depending on the type of interneuron in question (Aonuma and Newland, 2001).
Here we compare the neuroarchitecture of NO-cGMP signaling in different sensory systems of the locust Schistocerca gregaria and explore the functional implications of architectural similarities and differences. Our analysis of the cellular sources of NO is based upon a series of studies which used the NADPH-diaphorase (NADPHd) histochemical technique to localize NO synthase (NOS) in the locust brain (Müller and Bicker, 1994; Bicker and Hähnlein, 1995; Elphick et al., 1995, 1996; Bicker et al., 1996; Bicker and Schmachtenberg, 1997) and ventral nerve cord (Müller and Bicker, 1994; Ott and Burrows, 1998, 1999). Biochemical studies have shown a close correlation between NADPHd activity and NOS enzyme activity in different regions of the locust brain (Müller and Bicker, 1994; Elphick et al., 1995). Moreover, the tissue distribution of NADPHd and NOS immunoreactivity are virtually identical after paraformaldehyde fixation (Ogunshola, 1997). Together, these results suggest that NADPHd is a robust and reliable marker for NOS in the locust nervous system, an experimental advantage that does not necessarily apply to other invertebrate species (cf., Ott and Burrows, 1999; Ott et al., 2001). More recently, putative cellular targets of NO in the locust nervous system have been identified by immunohistolocalization of the enzyme sGC (Elphick and Jones, 1998; Jones and Elphick, 1999; Ott et al., 2000) and/or of NO-induced cGMP formation (Bicker et al., 1996; Bicker and Schmachtenberg, 1997; Ott et al., 2000). However, it should be recognised that neurons revealed by these techniques may not necessarily be exposed to NO in vivo. For this reason, it is important to compare the distribution of NOS with sGC and/or NO-induced cGMP throughout the nervous system.
The neuroanatomical studies indicate that NO-cGMP signaling is involved in modalities as diverse as vision, olfaction, taste, hearing, and tactile exteroception. To explain this abundance of NO-cGMP signaling in sensory processing, we discuss in depth the hypothesis of Ott and Burrows (1998) that NO released locally within sensory maps is used to couple sensory information channels that code similar stimulus properties.
|
COMPARATIVE ANATOMY OF NO-CYCLIC GMP SIGNALING IN LOCUST SENSORY SYSTEMS |
|---|
|
TOP
SYNOPSISINTRODUCTIONCOMPARATIVE ANATOMY OF NO-CYCLIC...FUNCTIONAL IMPLICATIONSReferences |
|---|
Mechanosensory systems
The processing of mechanosensory information is best understood in the thorax where it plays a key role in leg motor control (for a detailed review, see Burrows, 1996
). The thorax and its appendages are abundantly equipped with two types of tactile receptor: tactile hairs or trichoid sensilla, and shorter, peg-like basiconic sensilla (Burrows and Newland, 1994; Newland and Burrows, 1994). The multitude of other mechanoreceptors act mainly as proprioceptors although some encode both proprioceptive and exteroceptive information (reviewed in Burrows, 1996). Within the ventral nerve cord, the axon projections of tactile receptors and proprioceptors are strictly separated (Pflüger et al., 1988). Tactile projections form a somatotopic map in neuropil known as ventral association centres (VACs; Newland, 1991), whereas proprioceptors project into the lateral association centers and/or the medial VAC, a neuropil distinct from the tactile VACs.
The tactile projection neuropils are densely innervated by NOS-expressing interneurons (Ott and Burrows, 1998, 1999). Figure 1A schematically summarizes the neuroarchitecture of NO-cGMP signaling in the tactile system. The meshwork of NOS-expressing fibers in the VACs is derived from the collaterals of only a few intersegmental axons (red in Fig. 1A; cf. Fig. 2 in Ott and Burrows, 1998). This innervation pattern is repeated in all three thoracic ganglia, and it appears that each of these intersegmental axons supplies the VACs of all three ganglia with NOS-containing collaterals. Proprioceptive neuropils also contain NOS-expressing fibers, but these stem from local interneurons not shown in Figure 1. Interestingly, therefore, the nitrergic innervation of the thoracic sensory neuropils follows the separation of tactile and proprioceptive pathways found in early mechanosensory processing.
|
The afferent axons from thoracic sense organs do not contain NOS. As shown in Ott and Burrows (1999)
, NADPHd-staining in cricket afferent fibers (Schürmann et al., 1997) is most likely unrelated to NOS. This interpretation is supported by results obtained with a "universal" NOS antibody which in locusts stains an interneuronal fiber meshwork but no afferents (S. R. Ott and M. Burrows, unpublished).
The cellular distribution of sGC in the thoracic mechanosensory system has been analyzed by using two complementary immunohistochemical approaches (Ott et al., 2000): first, direct localization of the sGC 
-subunit, and second, localization of cGMP after exposure of the tissue to NO donors. Immunoreactivity for sGC is strongly expressed in mechanosensory neurons including those of the tactile hairs (blue in Fig. 1A; cf. Figs. 2 and 3 in Ott et al., 2000) and various proprioceptors. Immunoreactivity is present in the peripheral cell bodies, axons, and central axon terminals. In the axons and axon terminals, the presence of functional sGC has been confirmed by NO-induced cGMP immunoreactivity. In contrast, the first-order interneurons that act as primary integrators of mechanosensory information (e.g., midline spiking local interneurons, reviewed in Burrows, 1996; grey in Fig. 1A) show low levels of sGC expression. The axonal projections of sensory neurons are therefore a primary target for NO released within the mechanosensory neuropils.
In the locust brain, strong sGC staining is present in afferent axons of antennal mechanoreceptors (Elphick and Jones, 1998). These afferents bypass the olfactory neuropil in the antennal lobe to project into the antennal mechanosensory and motor centers. This finding is consistent with the expression of sGC in the thorax and suggests that mechanosensory afferent terminals are targets for interneuronally released NO both in the ventral nerve cord and in the brain.
Visual system
The visual system of the locust comprises a pair of large, complex compound eyes and three comparatively simple ocelli. The retina of the compound eye is organized into approximately 3000 functional units or ommatidia, each containing eight photoreceptor cells. Information from these receptors is processed in the optic lobes of the brain which are divided into three subsequent retinotopic neuropils named respectively the lamina, medulla, and lobula. NADPHd-positive neurons are abundant in the optic lobes (Elphick et al., 1996). Within the distal part, NADPHd is strongly expressed in a subset of lamina monopolar cells (red in Fig. 1B; Elphick et al., 1996; Bicker and Schmachtenberg, 1997), first-order interneurons that receive direct synaptic inputs from the photoreceptors. NADPHd is absent, however, from the photoreceptor cells themselves, and also from all other cells in the retina. In the more proximal parts of the optic lobes, NADPHd-positive fibers are arranged in a complex pattern the significance of which is as yet not understood (Elphick et al., 1996).
Nitric oxide donors induce cGMP immunostaining in the photoreceptor cells (blue in Fig. 1B), but not in lamina monopolar cells (Bicker and Schmachtenberg, 1997). Within the photoreceptor cells, cGMP staining is induced in the retinular segment, the cell body, and over the entire length of the axon down to the axon terminals which are located in the lamina or medulla, depending on the receptor cell type. The role of cGMP in insect phototransduction is still unclear. Although the primary transduction machinery of insect photoreceptors does not seem to involve cGMP (reviewed in Ranganathan et al., 1995), a cGMP-gated ion channel is expressed in the eye of Drosophila (Baumann et al., 1994), and application of cGMP enhances the light-induced excitation in Drosophila photoreceptors (Bacigalupo et al., 1995). Furthermore, NO donors and cGMP analogues increase the light response of photoreceptor cells (Schmachtenberg and Bicker, 1999).
NO could therefore be released from NOS-expressing lamina monopolar cells as a retrograde messenger activating sGC in the photoreceptor cells. Schmachtenberg and Bicker (1999) suggested both receptor cell adaptation and changes in the gain of the output synapses as potential effects. It is important, however, to discriminate clearly between these two possibilities. Adaptation would have to take place in the distal, retinular part of the receptor cell where the machinery for phototransduction resides. NO synthase is absent from the retina and the nearest potential sources of NO are lamina monopolar cells. To modulate phototransduction, NO would thus have to diffuse over a considerable distance. For a modulation of synaptic transmission, on the other hand, we would expect sGC expression in the receptor axon terminals presynaptic to the lamina monopolar cells.
It is therefore interesting to compare the distribution of NO-induced cGMP with that of sGC in the locust retina and lamina (Elphick and Jones, 1998; Jones and Elphick, 1999). Immunoreactivity for sGC was highly concentrated in the rhabdomeres within the distal, retinular compartments of the photoreceptor cells. Hardly any sGC was found in the photoreceptor axons and terminals. In the light of this, the pronounced axonal cGMP-immunostaining (Bicker and Schmachtenberg, 1997) is puzzling although cGMP might diffuse along the receptor axons while the preparation is exposed to the NO donor. Alternatively, sGC in the axons might have escaped detection because Bouin fixation was used which gives a comparatively low detection sensitivity (Ott et al., 2000).
Either way, the massive concentration of sGC in the rhabdomeres is striking. The present evidence suggests that sGC plays a dual role in photoreceptor cells. In the receptor terminals in the lamina, sGC may mediate a modulation of synaptic transmission by NO released from lamina monopolar cells (Fig. 1B). In the rhabdomeric part of the receptor cells, it may be involved in phototransduction and/or receptor adaptation. In the absence of NADPHd in the retina it has been suggested that the primary endogenous activator might be carbon monoxide (CO) rather than NO (Jones and Elphick, 1999). Moreover, the activity of sGC in the retina may not be solely determined by NO or CO. When locust eyes are incubated with an NO donor, an increase in cGMP is observed only in dark-adapted eyes (Jones and Elphick, 1999). This inhibitory effect of light on NO-dependent cGMP production appears to be Ca2+-dependent because dark adaptation can be mimicked by incubation of light-adapted eyes in Ca2+-free saline. Collectively these data indicate that sGC and cGMP may participate in photoreceptor adaptation under dark conditions which is consistent with observations of Schmachtenberg and Bicker (1999) who showed that NO donors and cGMP analogues increase the light response of dark-adapted locust photoreceptor cells.
Complementing the large compound eyes, locusts have three much smaller simple eyes or ocelli characterized by rapid signal transmission and high photic sensitivity but very poor spatial resolution (reviewed in L. J. Goodman, 1974; Toh and Tateda, 1991; Mizunami, 1994). Each ocellus comprises a common lens, a few hundred photoreceptor cells in groups of two to seven with a central rhabdom complex, and a synaptic neuropil where the receptors synapse onto large and small ocellar interneurons (C. S. Goodman, 1976; C. S. Goodman and Williams, 1976). The axons of both classes of interneuron enter the brain via the ocellar nerve, which also carries the axons of efferent neurons (L. J. Goodman, 1974). We have recently discovered fine NADPHd-positive axons in the ocellar nerve and globular NADPHd-positive fiber meshworks in the ocellar neuropil (Ott and Elphick, unpublished). The ocellar photoreceptors are immunoreactive for sGC, like their counterparts in the compound eyes (Elphick and Jones, 1998). In both cases strong sGC-immunoreactivity is concentrated in the central rhabdom complex of the photoreceptor cell groups. This important observation suggests that rhabdomeric sGC-expression is a fundamental property of locust photoreceptor cells, perhaps linked to photoreceptor dark adaptation in both kinds of eye (see above).
Intriguingly, sGC-immunoreactivity is also present in the axons and cell bodies of the large ocellar interneurons (Elphick and Jones, 1998). The ocellar system thus presents an unusual situation where sGC is located both in pre- and postsynaptic cells. Therefore, if the photoreceptor terminals contain significant amounts of sGC, NOS-expressing ocellar neurons may modulate transmission both pre- and postsynaptically.
Chemosensory systems
The primary olfactory integration centers in the insect brain are the antennal lobes. Each lobe receives the axonal projections of the 100,000–200,000 olfactory receptor neurons from the ipsilateral antenna. This information is processed by less than 750 local (intrinsic) interneurons and a few hundred output interneurons, the latter also passing the olfactory information to other parts of the brain. The antennal lobe neuropil is organized into spherical glomeruli (approximately 1,000 in the locust; Leitch and Laurent, 1996); intriguingly, a similar glomerular organization is found in the primary olfactory center of vertebrates, the olfactory bulb (reviewed in Hildebrand and Shepherd, 1997).
The antennal lobes show the highest concentration of NADPHd and NOS activity in the locust brain (Müller and Bicker, 1994; Elphick et al., 1995). Approximately 50 NADPHd-positive intrinsic interneurons supply the olfactory glomeruli with a dense meshwork of strongly NADPHd-positive fibers. Neither the antennal nerve nor the tracts that carry the axons of projection neurons show diaphorase activity (Elphick et al., 1995; Bicker and Hähnlein, 1995; Bicker et al., 1996). All NADPHd-positive fibers in the olfactory neuropil are thus derived from intrinsic antennal lobe interneurons. The olfactory receptor afferents do not seem to contain sGC. About 20 antennal lobe interneurons display sGC immunoreactivity (Elphick and Jones, 1998), and NO induces cGMP-staining in antennal lobe interneurons (Bicker et al., 1996) which are presumably identical to those that stain for sGC. Interestingly, the cGMP-immunostaining co-localizes with NADPHd, although some antennal lobe interneurons stain for either NADPHd or cGMP alone. In combination, these results suggest that many antennal lobe interneurons express both NOS and sGC.
It has been proposed that the glomeruli serve as diffusion compartments for NO (Breer and Shepherd, 1993; Bicker et al., 1997) but the detailed neuroarchitecture of NO signaling in the antennal lobes is still poorly understood. Can afferent inputs to one glomerulus effect the release of NO in another glomerulus that is innervated by the same NOS-expressing interneuron? Do NOS interneurons differ from other intrinsic interneurons in the number of olfactory glomeruli they supply? Is every glomerulus innervated by a single NOS neuron or does multiple innervation occur? Such questions will have to be answered before detailed functional conclusions can be drawn.
Chemoreceptors are not restricted to the antennae; basiconic sensilla, which occur on many parts of the body surface and the legs, are innervated by mechanosensory and chemosensory neurons (Burrows and Newland, 1994; Newland and Burrows, 1994). Intriguingly, while sGC appears to be absent from antennal chemoreceptors (Bicker et al., 1996; Elphick and Jones, 1998), chemosensory neurons of basiconic sensilla express sGC-immunoreactivity in their cell bodies (Ott et al., 2000). The large number of axons staining for sGC in the sensory roots of thoracic nerves suggests that sGC expression extends into their central projections. Therefore, thoracic chemosensory neurons have to be considered as targets for NO.
Auditory system
From an apparent lack of NOS-expressing fibers in the primary auditory neuropil it has been concluded that NO is unlikely to play a role in early auditory processing (Ott and Burrows, 1998). This interpretation, however, needs to be reconsidered in the light of recent findings that suggest a novel role for NO-cGMP signaling in the insect auditory system. The locust ears are derived from a pair of lateral (pleural) chordotonal organs in the first abdominal segment. The auditory axons project into the posterior region of the metathoracic mVAC where they form a tonotopic map (reviewed in Jacobs et al., 1999). The anterior mVAC accommodates afferent terminals from other chordotonal organs that have retained their proprioceptive function. Whereas the anterior mVAC is innervated by a local group of NADPHd-positive interneurons, the main bulk of the auditory mVAC lacks NADPHd-positive fibers (Ott and Burrows, 1998).
In this apparent absence of sources for NO in the auditory neuropil it was surprising to find strong sGC immunoreactivity in auditory receptor neuron cell bodies, axons, and axon terminals; moreover, strong NO-induced cGMP-immunoreactivity has been demonstrated in auditory afferents (Ott et al., 2000). It is therefore very likely that auditory receptor neurons express high levels of sGC and are potential targets for NO. This prompted us to re-examine the NADPHd staining. It showed that NADPHd-positive fibers in a "small median neuropil area immediately posterior to mVAC" (Ott and Burrows, 1998) might actually be within the posteriormost portion of mVAC although it is hard to say whether they overlap with auditory afferent projections. What is clear, however, is that the extent of the NADPHd-positive fibers in this region is small compared to the volume occupied by the projections of the auditory receptors (cf., Jacobs et al., 1999). While it will be helpful to determine more precisely the spatial relationship between NO source and target cells in the auditory pathway, the biggest current obstacle in interpreting these anatomical data is our ignorance of the range of NO action in vivo.
Bullerjahn and Pflüger (1999; A. Bullerjahn, personal communication) have investigated the distribution of NADPHd in the locust Locusta migratoria during postembryonic development. The thoracic staining was very similar to that in Schistocerca (Ott and Burrows, 1998), and NADPHd-positive innervation of the tactile VACs was found throughout postembryonic life. Intriguingly, however, the staining in the posteriormost mVAC was present only in last instars and adults, thus preceding known changes in the tuning of those receptors that project into this region: only after the final moult, these receptors acquire their tuning to high frequencies (presumably due to changes in the tympanal mechanics; Petersen et al., 1982). Tonotopy and the frequency analysis by postsynaptic interneurones becomes thus effective only in adults, which may require a rewiring of the synaptic connections. Here, like in other sensory systems (Truman et al., 1996; Gibbs and Truman, 1998), NO signaling might mediate both a developmental specification of synaptic connections and a modulation of their efficacy in the mature system.
|
FUNCTIONAL IMPLICATIONS |
|---|
|
TOP
SYNOPSISINTRODUCTIONCOMPARATIVE ANATOMY OF NO-CYCLIC...FUNCTIONAL IMPLICATIONSReferences |
|---|
The above comparison shows that throughout the locust nervous system, primary sensory neuropils are innervated by NOS-expressing interneurons. Sensory neurons investigated so far do not express NOS. Rather, their terminals appear to be the primary target for interneuronally released NO in their projection neuropils. A notable exception is the antennal olfactory system, where NO appears to act primarily onto intrinsic interneurons (cf., Bicker et al., 1997
).
Previous studies have suggested a role for NO-cGMP signaling in the olfactory and visual systems of vertebrates (reviewed in Breer and Shepherd, 1993; Cudeiro and Rivadulla, 1999). The evidence reviewed here allows this hypothesis to be extended to insect sensory systems, and to modalities other than vision and olfaction.
Sensory maps and diffusible messengers
At this point it is helpful to recall the properties that characterize NO signaling. NO synthesis is linked to the activity of NOS-expressing neurons via the Ca2+/calmodulin-dependency of neuronal NOS. The diffusion of NO is rapid and unhindered by biological membranes, but the signaling range is limited by the radicalic nature, and hence instability, of NO. Within a set of parallel sensory channels, therefore, NO could relay information about channel activity to neighboring channels in the absence of lateral synaptic connections.
While this principle could act even in simple sensory systems consisting of only a few sensory channels, the spatial characteristics of NO have further implications within larger arrays of sensory receptors or interneurons. Such arrays are usually organized into maps so that neurons with similar coding properties project into adjacent neuropil areas. To explain the ubiquity of NO signaling in sensory systems, we have thus proposed a link between the properties of NO and the organization of sensory systems into maps (Ott and Burrows, 1998; Ott et al., 2000). This hypothesis is exemplified in Figure 1 for the thoracic mechanosensory system (A) and the visual system (B). Local activation of receptor neurons in the sensory arrays (arrowheads in Fig. 1Ai and B) results in local neural activity within the sensory map and hence in a local gradient of NO (yellow in Fig. 1Ai and B). In a topologically organized sensory system, the physical distance between two sensory channels in the map correlates with their distance in the coding space. As a consequence of this and the short half life of NO, the strength of coupling between sensory channels by NO will be proportional to their similarity in tuning.
This hypothesis also implies that the radius of influence can scale both with the number of activated receptor cells and with their level of activation. Stronger receptor activation and a larger number of active receptors would result in a higher rate of NO synthesis and thus in a wider spread of coupling. Simple activity-dependent local NO release may therefore provide plasticity in processing that could otherwise be achieved only with extensive lateral synaptic networking. The actual sphere of influence will further depend on several factors including the diffusional spread of NO, the electrical properties of the NOS-containing neurons and the spatiotemporal dynamics of intracellular Ca2+.
Implications of architectural differences
The idea that local gradients of NO act as a correlator of local activity within neural maps provides a leitmotif for the abundance of NO signaling in sensory systems. The anatomy of the NOS-expressing interneurons, however, differs significantly between different sensory systems. Such differences set constraints for the potential mode of action of NO at the network level, as exemplified in Figure 1 for the tactile and the visual system. In these examples, the anatomical differences between the NOS-expressing interneurons reflect their different roles in "conventional" neural transmission in the respective sensory networks. In the optic lobes, NOS is expressed in a large number of "classical" first-order interneurons (lamina monopolar cells; red in Fig. 1B), each of which has a very small field of projections. In the thoracic tactile system, NOS is expressed in a few intersegmental neurons with extensive arborizations in the projection neuropils of the afferents (red in Fig. 1A); the "classical" first-order interneurons that act as primary integrators of the sensory information (e.g., midline spiking local interneurons; reviewed in Burrows, 1996; grey in Fig. 1A) do not express NOS. Assuming that the NOS-expressing neurons receive synaptic inputs from afferents in the tactile neuropil, this architecture allows two modes of action for NO.
(i) The activation of a sensory neuron (Fig. 1Ai, arrowhead) could trigger a local increase in intracellular Ca2+ in branches of the NOS-containing neuron (dark red). The increased intracellular Ca2+ would activate NOS via calmodulin and cause a local release of NO. The resultant NO-gradient in the map (yellow) could affect synaptic transmission in the terminals of the active sensory neuron (dark blue) but also in the adjacent terminals of sensory neurons that originate from the same region of the body surface (medium blue). Transmission would be unaffected for receptors further away from the active receptor (light blue).
(ii) Alternatively, or in addition, action potentials in the NOS-containing neuron might propagate throughout its entire arbor (Fig. 1Aii, dark red) and effect a release of NO throughout the tactile projection neuropils (yellow). Such a global release would affect synaptic transmission for receptors from large regions on the body surface (blue). Action potentials in the NOS-containing neuron could either be triggered locally by a summation of afferent inputs within the map, or could originate in other ganglia in the nerve cord. This would permit an NO-mediated modulation of tactile information from all tactile receptors within a segment, and even across segments, in addition to regional plasticity. The architecture might therefore reflect a mechanism that can modulate or override local events in order to integrate tactile information from different parts of the body.
In contrast to the tactile system, the architecture of the visual system does not support a global control of NO release, because the lateral arbors of the NOS-containing interneurons are very small (Fig. 1B, red). All release must be due to local neuronal activity in the map. Photoexcitation of the photoreceptors in one ommatidium (arrowhead) would trigger the release of NO (yellow) from monopolar cells in the corresponding cartridge in the lamina. This could strengthen or weaken the synaptic output of the photoreceptor cells within this cartridge (dark blue). In addition, however, NO could diffuse into adjacent cartridges and affect the synaptic output of photoreceptors from adjacent ommatidia (medium blue). As in the tactile system, the output of receptors that are further away from the active receptor would be unaffected (light blue). These considerations re-emphasize the importance of determining the radius of action of NO in vivo.
Intermodal integration by NO
Basiconic sensilla serve a dual tactile and chemosensory function. There are at least two aspects in which this extra-antennal chemosensory system resembles the mechanosensory system more closely than the antennal chemosensory system. First, it seems that thoracic chemosensory receptor neurons express sGC but antennal olfactory neurons do not (Ott et al., 2000; Elphick and Jones, 1998). Second, and more intriguingly, their projections lack the glomerular organization characteristic for both vertebrate and insect olfactory centers. Rather, the chemosensory afferents from thoracic basiconic sensilla project into the VACs, a non-glomerular neuropil, to form a somatotopic map together with the tactile afferents (Newland, 1991; Newland et al., 2000).
An important implication of this somatotopic organization is that NO released locally within the VACs would act on the terminals of chemosensory and mechanosensory receptors from the same region of the body surface. By this mechanism, NO might participate in the intermodal integration of chemosensory and mechanosensory information. Both the chemosensory and the mechanosensory system in the thorax are designed to preserve information about the stimulus position in order to direct leg movements. The chemical coding space is comparatively simple, and an essentially two-dimensional map in the VAC could accommodate a spatially continuous representation of both tactile and chemical stimuli. In the antennal lobes, in contrast, somatotopic information is discarded in order to construct a multidimensional olfactory coding space that cannot be accommodated in a continuous map but must be broken up into a discontinuous spatial representation. Because free diffusion is spatially continuous, the necessity would arise for a compartmentalization of the neuropil into glomeruli. These considerations suggest that the neuroanatomy of sensory systems is intimately linked to the diffusion of NO.
Physiology operates within the constraints of structure. A neuroanatomical analysis can reliably exclude certain physiological mechanisms, but for those that are compatible with the structural constraints it can only suggest their existence. The neuroanatomical data reviewed here open up an exciting field for physiological experiments, and the accessibility of the locust to electrophysiological recordings makes it an excellent model to study the roles of NO in sensory systems.
|
ACKNOWLEDGMENTS |
|---|
We would like to acknowledge support to: S.R.O. from the Balfour funds (Department of Zoology, Cambridge, U.K.) and the Fürst Dietrichstein'sche Stiftung (Austria), M.R.E. from BBSRC grants S03858 and S11816 [GenBank] and M.B. from NIH grant NS 16058.
|
FOOTNOTES |
|---|
1 From the Symposium Nitric Oxide in the Invertebrates: Comparative Physiology and Diverse Functions presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 4–8 January 2000, at Atlanta, Georgia.
2 E-mail: S.R.Ott{at}qmw.ac.uk
|
References |
|---|
|
TOP
SYNOPSISINTRODUCTIONCOMPARATIVE ANATOMY OF NO-CYCLIC...FUNCTIONAL IMPLICATIONSReferences |
|---|
Aonuma, H., and P. L. Newland. 2001. Opposing actions of nitric oxide on synaptic inputs of identified interneurones in the central nervous system of the crayfish. J. Exp. Biol, .
Bacigalupo, J., D. M. Bautista, D. L. Brink, J. F. Hetzer, and P. M. O'Day. 1995. Cyclic-GMP enhances light-induced excitation and induces membrane currents in Drosophila retinal photoreceptors. J. Neurosci, 15:7196-7200.[Abstract]
Baumann, A., S. Frings, M. Godde, R. Seifert, and U. B. Kaupp. 1994. Primary structure and functional expression of a Drosophila cyclic nucleotide-gated channel present in eyes and antennae. EMBO J, 13:5040-5050.[Web of Science][Medline]
Bicker, G., and I. Hähnlein. 1995. NADPH diaphorase expression in neurons and glial cells of the locust brain. NeuroReport, 6:325-328.[Web of Science][Medline]
Bicker, G., and O. Schmachtenberg. 1997. Cytochemical evidence for nitric oxide/cyclic GMP signal transmission in the visual system of the locust. Eur. J. Neurosci, 9:189-193.[CrossRef][Web of Science][Medline]
Bicker, G., O. Schmachtenberg, and J. De Vente. 1997. Geometric considerations of nitric oxide-cyclic GMP signalling in the glomerular neuropil of the locust antennal lobe. Proc. R. Soc. London B, 264:1177-1181.
Bicker, G., O. Schmachtenberg, and J. De Vente. 1996. The nitric oxide/cyclic GMP messenger system in olfactory pathways of the locust brain. Eur. J. Neurosci, 8:2635-2643.[CrossRef][Web of Science][Medline]
Breer, H., and G. M. Shepherd. 1993. Implications of the NO/cGMP system for olfaction. Trends Neurosci, 16:5-9.[CrossRef][Web of Science][Medline]
Bullerjahn, A., and H.-J. Pflüger. 1999. Nitric oxide synthase (NOS) in the metathoracic ganglion and abdominal ganglia of the locust during postembryonic development. In N. Elsner and U. Eysel (eds.), Göttingen Neurobiology Report 1999, pp. 598. Georg Thieme Verlag, Stuttgart.
Burrows, M. 1996. The neurobiology of an insect brain. Oxford University Press, New York.
Burrows, M., and P. L. Newland. 1994. Convergence of mechanosensory afferents from different classes of exteroceptors onto spiking local interneurones in the locust. J. Neurosci, 14:3341-3350.[Abstract]
Cudeiro, J., and C. Rivadulla. 1999. Sight and insight—on the physiological role of nitric oxide in the visual system. Trends Neurosci, 22:109-116.[Web of Science][Medline]
Elphick, M. R., and I. W. Jones. 1998. Localization of soluble guanylyl cyclase -subunit in identified insect neurons. Brain Res, 800:174-179.[CrossRef][Web of Science][Medline]
Elphick, M. R., R. C. Rayne, V. Riveros-Moreno, S. Moncada, and M. O'Shea. 1995. Nitric oxide synthesis in locust olfactory interneurons. J. Exp. Biol, 198:821-829.[Web of Science][Medline]
Elphick, M. R., J. L. D. Williams, and M. O'Shea. 1996. New features of the locust optic lobe—evidence of a role for nitric oxide in insect vision. J. Exp. Biol, 199:2395-2407.[Abstract]
Gibbs, S. M., and J. W. Truman. 1998. Nitric oxide and cyclic GMP regulate retinal patterning in the optic lobe of Drosophila. Neuron, 20:83-93.[CrossRef][Web of Science][Medline]
Goodman, C. S. 1976. Anatomy of the ocellar interneurons of acridid grasshoppers. I. The large interneurons. Cell Tissue Res, 175:183-202.[Web of Science][Medline]
Goodman, C. S., and J. L. D. Williams. 1976. Anatomy of the ocellar interneurons of acridid grasshoppers. II. The small interneurons. Cell Tissue Res, 175:203-225.[Web of Science][Medline]
Goodman, L. J. 1974. The neural organization and physiology of the insect dorsal ocellus. In G. A. Horridge (ed.), The compound eye and vision of insects, pp. 515–548. Clarendon Press, Oxford.
Hildebrand, J. G., and G. M. Shepherd. 1997. Mechanisms of olfactory discrimination: Converging evidence for common principles across phyla. Annu. Rev. Neurosci, 20:595-631.[CrossRef][Web of Science][Medline]
Jacobs, K., B. Otte, and R. Lakes-Harlan. 1999. Tympanal receptor cells of Schistocerca gregaria: Correlation of soma positions and dendrite attachment sites, central projections and physiologies. J. Exp. Zool, 283:270-285.[CrossRef][Web of Science]
Jones, I. W., and M. R. Elphick. 1999. Dark-dependent soluble guanylyl cyclase activity in locust photoreceptor cells. Proc. R. Soc. London B, 226:413-419.
Leitch, B., and G. J. Laurent. 1996. GABAergic synapses in the antennal lobe and mushroom body of the locust olfactory system. J. Comp. Neurol, 372:487-514.[CrossRef][Web of Science][Medline]
Mizunami, M. 1994. Functional diversity of neural organization in insect ocellar systems. Vision Research, 35:443-452.[CrossRef][Web of Science]
Müller, U., and G. Bicker. 1994. Calcium-activated release of nitric oxide and cellular distribution of nitric oxide-synthesizing neurons in the nervous system of the locust. J. Neurosci, 14:7521-7528.[Abstract]
Newland, P. L. 1991. Morphology and somatotopic organisation of the central projections of afferents from tactile hairs on the hind leg of the locust. J. Comp. Neurol, 312:193-509.[CrossRef][Web of Science][Medline]
Newland, P. L., and M. Burrows. 1994. Processing of mechanosensory information from gustatory receptors on a hind leg of the locust. J. Comp. Physiol. A, 174:399-410.[Medline]
Newland, P. L., S. M. Rogers, I. Gaaboub, and T. Matheson. 2000. Parallel somatotopic maps of gustatory and mechanosensory neurons in the central nervous system of an insect. J. Comp. Neurol, 425:82-96.[CrossRef][Web of Science][Medline]
Ogunshola, O. O. 1997. Molecular studies of the NO-cGMP signalling pathway in the desert locust Schistocerca gregaria. Ph.D. Diss., University of Sussex, UK.
Ott, S. R., and M. Burrows. 1998. Nitric oxide synthase in the thoracic ganglia of the locust: Distribution in the neuropiles and morphology of neurones. J. Comp. Neurol, 395:217-230.[CrossRef][Web of Science][Medline]
Ott, S. R., and M. Burrows. 1999. NADPH diaphorase histochemistry in the thoracic ganglia of locusts, crickets and cockroaches: Species differences and the impact of fixation. J. Comp. Neurol, 410:387-397.[CrossRef][Web of Science][Medline]
Ott, S. R., I. W. Jones, M. Burrows, and M. R. Elphick. 2000. Sensory afferents and motor neurons as targets for nitric oxide in the locust. J. Comp. Neurol, 422:521-532.[CrossRef][Web of Science][Medline]
Ott, S. R., H. Aonuma, P. L. Newland, and M. R. Elphick. 2001. NADPH diaphorase in invertebrate nervous systems: New techniques and a note of caution. In N. Elsner and G. W. Kreutzberg (eds.), Göttingen Neurobiology Report 2001. Georg Thieme Verlag, Stuttgart. (In press).
Petersen, M., K. Kalmring, and A. Cokl. 1982. The auditory system in larvae of the migratory locust. Physiol. Entomology, 7:43-54.
Pflüger, H.-J., P. Bräunig, and R. Hustert. 1988. The organization of mechanosensory neuropiles in locust thoracic ganglia. Phil. Trans. R. Soc. London B, 321:1-26.
Ranganathan, R., D. M. Malicki, and C. S. Zucker. 1995. Signal transduction in Drosophila photoreceptors. Annu. Rev. Neurosci, 18:283-317.[CrossRef][Web of Science][Medline]
Schmachtenberg, O., and G. Bicker. 1999. Nitric oxide and cyclic GMP modulate photoreceptor cell responses in the visual system of the locust. J. Exp. Biol, 202:13-20.[Abstract]
Schuman, E. M., and D. V. Madison. 1994. Nitric oxide and synaptic function. Annu. Rev. Neurosci, 17:153-183.[CrossRef][Web of Science][Medline]
Schürmann, F.-W., J. Helle, M. Knieriem-Grenzebach, M. Pauls, and U. Spörhase-Eichmann. 1997. Identified neuronal cells and sensory neuropiles in the ventral nerve cord of an insect stained by NADPH-diaphorase histochemistry. ZACS, 100:98-109.
Stamler, J. S. 1994. Redox signaling: Nitrosylation and related target interactions of nitric oxide. Cell, 78:931-936.[CrossRef][Web of Science][Medline]
Stamler, J. S., E. J. Toone, S. A. Lipton, and N. J. Sucher. 1997. (S)NO signals: Translocation, regulation, and a consensus motif. Neuron, 18:691-696.[CrossRef][Web of Science][Medline]
Toh, Y., and H. Tateda. 1991. Structure and function of the insect ocellus. Zool. Sci, 8:395-413.
Truman, J. W., J. De Vente, and E. E. Ball. 1996. Nitric oxide-sensitive guanylate cyclase activity is associated with the maturational phase of neuronal development in insects. Development, 122:3949-3958.[Abstract]
Wildemann, B., and G. Bicker. 1999. Nitric oxide and cyclic GMP induce vesicle release at Drosophila neuromuscular junction. J. Neurobiol, 39:337-346.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||