Inhalation of irritating substances leads to activation of the trigeminal nerve, triggering protective reflexes that include apnea or sneezing. Receptors for trigeminal irritants are generally assumed to be located exclusively on free nerve endings within the nasal epithelium, requiring that trigeminal irritants diffuse through the junctional barrier at the epithelial surface to activate receptors. We find, in both rats and mice, an extensive population of chemosensory cells that reach the surface of the nasal epithelium and form synaptic contacts with trigeminal afferent nerve fibers. These chemosensory cells express T2R ''bitter-taste'' receptors and ␣-gustducin, a G protein involved in chemosensory transduction. Functional studies indicate that bitter substances applied to the nasal epithelium activate the trigeminal nerve and evoke changes in respiratory rate. By extending to the surface of the nasal epithelium, these chemosensory cells serve to expand the repertoire of compounds that can activate trigeminal protective reflexes. The trigeminal chemoreceptor cells are likely to be remnants of the phylogenetically ancient population of solitary chemoreceptor cells found in the epithelium of all anamniote aquatic vertebrates.

Taste buds contain a variety of morphological and histochemical types of elongate cells. Serotonin, neuron-specific enolase (NSE), ubiquitin carboxyl terminal hydrolase (PGP 9.5), and neural cell adhesion molecule (N-CAM) all have been described as being present in the morphologically defined Type III taste cells in rats. In order to determine whether these substances coexist in a single cell, we undertook immunohistochemical and ultrastructural analysis of taste buds in rats. Double-label studies show that PGP 9.5 and NSE always colocalize. In contrast, PGP 9.5 and serotonin seldom colocalize. Further, whereas the serotonin-immunoreactive cells are always slender and elongate, the PGP 9.5/NSE population comprise two morphological types--one slender, the other broader and pyriform. Although gustducin-immunoreactive taste cells appear similar in overall shape to the pyriform PGP 9.5/NSE population, gustducin never colocalizes with PGP 9.5 or NSE. The serotonin-immunoreactive taste cells have an invaginated nucleus, synaptic contacts with nerve fibers, and taper apically to a single, large microvillus. These are all characteristics of Type III taste cells described previously in rabbits (Murray [1973] Ultrastructure of Sensory Organs I. Amsterdam: North Holland. p 1-81). PGP 9.5-immunoreactive taste cells exhibit two morphological varieties. One type is similar to the serotonin-immunoreactive population, containing an invaginated nucleus, synapses with nerve fibers, and a single large microvillus. The other type of PGP 9.5-immunoreactive taste cell has a large round nucleus and the apical end of the cell tapers to a tuft of short microvilli, which are characteristics of Type II taste cells. Thus, in rats, some Type III cells accumulate serotonin but do not express PGP 9.5, whereas others express PGP 9.5 but do not accumulate amines. Similarly, Type II taste cells come in at least two varieties: those immunoreactive for gustducin and those immunoreactive for PGP 9.5.

The nasal epithelium is richly invested with peptidergic (substance P and calcitonin gene-related peptide [CGRP]) trigeminal polymodal nociceptors, which respond to numerous odorants as well as irritants. Peptidergic trigeminal sensory fibers also enter the glomerular layer of the olfactory bulb. To test whether the trigeminal fibers in the olfactory bulb are collaterals of the epithelial trigeminal fibers, we utilized dual retrograde labeling techniques in rats to identify the trigeminal ganglion cells innervating each of these territories. Nuclear Yellow was injected into the dorsal nasal epithelium, and True Blue was injected into the olfactory bulb of the same side. Following a survival period of 3-7 days, the trigeminal ganglion contained double-labeled, small (11.8 x 8.0 microm), ellipsoid ganglion cells within the ethmoid nerve region of the ganglion. Tracer injections into the spinal trigeminal complex established that these branched trigeminal ganglion cells also extended an axon into the brainstem. These results indicate that some trigeminal ganglion cells with sensory endings in the nasal epithelium also have branches reaching directly into both the olfactory bulb and the spinal trigeminal complex. These trigeminal ganglion cells are unique among primary sensory neurons in having two branches entering the central nervous system at widely distant points. Furthermore, the collateral innervation of the epithelium and bulb may provide an avenue whereby nasal irritants could affect processing of coincident olfactory stimuli.

Axons immunoreactive for calcitonin gene-related peptide (CGRP) and substance P are present in the olfactory nerve, although few, if any, olfactory receptor cells contain immunocytochemically detectable levels of these peptides. The possible trigeminal origin of these fibers was tested by performing unilateral stereotaxic lesions of the ophthalmic division of the trigeminal nerve, followed 2-25 days later by immunocytochemistry for CGRP and substance P. As reported previously, free nerve endings immunoreactive for both peptides were found transversing the nasal epithelium on the unlesioned side. Also on the unlesioned side, peptidergic axons, immunoreactive for both CGRP and substance P, could be traced from the olfactory nerve into the glomerular layer throughout the olfactory bulb, but especially into its rostral third. Ipsilateral to the trigeminal ganglion lesion, such peptide-immunoreactive fibers were absent or markedly reduced in the bulb, nerve, and epithelium. These results indicate that the peripheral branches of the ophthalmic branch of the trigeminal nerve enter the olfactory bulb along with the olfactory nerve and terminate in the glomerular layer along with the olfactory axons. Ultrastructural analysis of the CGRP-immunoreactive terminals in the glomeruli reveal vesicle-filled axonal processes terminating in the absence of obvious pre- or postsynaptic specializations. Whether the trigeminal fibers in the bulb are functional, e.g., convey information to the olfactory bulb via an axon reflex, or relay information from the olfactory bulb to the brainstem trigeminal nuclei is unclear.

The taste system of catfish, having distinct taste receptor sites for L-alanine and L-arginine, is highly sensitive to amino acids. A previously described monoclonal antibody (G-10), which inhibits L-alanine binding to a partial membrane fraction (P2) derived from catfish (Ictalurus punctatus) taste epithelium, was found in Western blots to recognize a single band, at apparent MW of 113,000 D. This MW differs from the apparent MW for the presumed arginine receptor identified previously by PHA-E lectin affinity. In order to test whether PHA-E lectin actually reacts with the arginine-receptor, reconstituted membrane proteins partially purified by PHA-E affinity were used in artificial lipid bilayers. These reconstituted channels exhibited L-arginine-activated activity similar to that found in taste cell membranes. Accordingly, we utilized the PHA-E lectin and G-10 antibody as probes to differentially localize the L-alanine and L-arginine binding sites on the apical surface of catfish taste buds. Each probe labels numerous, small (0.5-1.0 micron) patches within the taste pore of each taste bud. This observation suggests that each bud is not tuned to a single taste substance, but contains putative receptor sites for both L-arginine and L-alanine. Further, analysis of double-labeled tissue reveals that the PHA-E and G-10 sites tend to be separate within each taste pore. These findings imply that in catfish, individual taste cells preferentially express receptors to either L-arginine or L-alanine. In addition, PHA-E binds to the apices of solitary chemoreceptor cells in the epithelium, indicating that this independent chemoreceptor system may utilize some receptor sites similar to those in taste buds.

Gustatory afferent fibers of the vagus nerve that innervate taste buds of the oropharynx of the goldfish, Carassius auratus, project to the vagal lobe, which is a laminated gustatory nucleus in the dorsal medulla. As in the mammalian gustatory system, responses by second-order cells in the goldfish medulla are mediated by N-methyl-D-aspartate (NMDA) and non-NMDA ionotropic glutamate receptors. We utilized a cobalt uptake technique to label vagal lobe neurons that possess cobalt-permeable ionotropic glutamate receptors. Vagal lobe slices were bathed in kainate (40 microM) or glutamate (0.5 or 1 mM) in the presence of CoCl(2), which can pass into cells through the ligand-gated cation channels of non-NMDA receptors made up of certain subunit combinations. Cobalt-filled cells and dendrites were observed in slices that were activated by kainate or glutamate, but not in control slices that were bathed in CoCl(2) alone, nor in slices that were bathed with the non-NMDA receptor antagonist 6,7-dinitroquinoxaline-2,3-dione (10 microM) in addition to an agonist. Likewise, simple depolarization of the cells with KCl failed to induce cobalt loading. Cobalt-filled round unipolar cells, elongate or globular bipolar cells, and multipolar cells with elongate or polygonal perikarya were distributed throughout the cell layers in the sensory zone of the vagal lobe. Numerous labeled neurons had dendrites spanning layers IV and VI, the two principal layers of primary afferent input. Apical and basal dendrites often extended radially through neighboring laminae, but many cells also extended dendrites tangential to the lamination of the sensory zone. In the motor layer, cell bodies and proximal dendrites of small, multipolar neurons, and large motoneurons were regularly loaded with cobalt.

In transgenic neurotrophin-3 lacZ-neo (NT-3(lacZneo)) mice, in which the coding region for NT-3 is replaced by Eschericia coli lacZ, the expression of beta-galactosidase faithfully mimics the expression of NT-3 (Vigers AJ, Baquet ZC, Jones KR [2000], J Comp Neurol 416:398-416). During embryonic and early postnatal development, beta-galactosidase is detected in the olfactory system, beginning at embryonic day 11.5 in the nasal epithelium and at embryonic day 16.5 in the olfactory bulb. Levels of beta-galactosidase rise with age, reaching a peak during the second postnatal week, when beta-galactosidase reactivity is visible in up to 50% of the glomeruli. As the animal matures, the beta-galactosidase levels decline, but staining remains present in axons and cell bodies of a specific subset of olfactory receptor neurons (ORNs) projecting to a limited subset of glomeruli. The heavily labeled ORNs do not follow the typical OR expression zones in the epithelium but appear similar to the "patch" expression pattern of mOR37 receptors. The most heavily reactive glomeruli exhibit a striking reproducible pattern in the ventral olfactory bulb (OB). Some glomeruli of the OB contain calcitonin gene-related peptide (CGRP)-immunoreactive fibers of the trigeminal nerve. However, double-label immunocytochemistry for CGRP and beta-galactosidase rendered no correlation between trigeminal innervation and the degree of innervation by NT-3-expressing ORNs. Thus, the timing and presence of beta-galactosidase in a subset of ORNs suggests that NT-3 plays a role in synaptogenesis and/or synapse function in a specific subset of ORNs within the olfactory bulb.

In order to study the pattern of innervation of taste buds and the surrounding epithelium, the carbocyanine dye diI was applied to the nerve stump in isolated, paraformaldehyde-fixed barbels obtained from channel catfish, Ictalurus punctatus. After a diffusion period of 7-41 days, the barbels were sectioned on a vibratome and examined with epifluorescence. Labeled axons were observed up to 1 cm from the site of application. Frequently, a fascicle of labeled axons turned outward toward the epithelium to innervate taste buds or to end apparently as free endings within the epithelium. Within 2-3 mm of the dye-application site, many taste buds contained one or at most 5-10, labeled spindle-shaped, presumed receptor, cells. In taste buds containing multiple labeled cells, the cells usually were arranged as intertwined pairs or triplets rather than being homogeneously distributed within the taste bud. In a few cases, labeled basal cells could be discerned among the labeled axons of the basal plexus. The cells of the taste bud apparently were labeled by transcellular passage of the dye from the nerve fibers into the cells. The limited number of labeled cells within each taste bud may indicate a special relationship between these cells and the nerve fibers innervating them.