Search

Physiological studies of odor-elicited responses from the olfactory epithelium and bulb in the tiger salamander, Ambystoma tigrinum, have elucidated a number of features of olfactory coding that appear to be conserved across several vertebrate species. This animal model has provided an accessible in vivo system for observing individual and ensemble olfactory responses to odorant stimulation using biochemical, neurophysiological, and behavioral assays. In this paper we have complemented these studies by characterizing 35 candidate odorant receptor genes. These receptor sequences are similar to those of the large families of olfactory receptors found in mammals and fish. In situ hybridization, using RNA probes to 20 of these sequences, demonstrates differential distributions of labeled cells across the extent and within the depth of the olfactory epithelium. The distributions of cells labeled with probes to different receptors show spatially restricted patterns that are generally localized to different degrees in medial-lateral and anterior-posterior directions. The patterns of receptor expression in the ventral olfactory epithelium (OE) are mirrored in the dorsal OE. We present a hypothesis as to how the sensory neuron populations expressing different receptor types responding to a particular odorant may relate to the distribution patterns of epithelial and bulbar responses previously characterized using single-unit and voltage-sensitive dye recording methods.

During posttranslational processing to generate CCK 8, pro-cholecystokinin (CCK) undergoes endoproteolytic cleavage at three sites. Several studies using endocrine and neuronal tumor cells in culture and recombinant enzymes and synthetic substrates in vitro have pointed to the subtilisin/kexin-like enzymes prohormone convertase (PC) 1, PC2, and PC5 as potential candidates for these endoproteolytic cleavages. In these experimental models, they all appear to be able to cleave pro-CCK to make the correct products. One rodent model has provided information about the role of PC2. PC2 knockout mouse brains had less CCK 8 than wild-type, although a substantial amount of CCK was still present. The degree to which CCK levels were reduced in these mice was regionally specific. These data indicated that PC2 is important for normal production of CCK but that it is not the only endoprotease that is involved in CCK processing. To evaluate whether PC1 and PC5 are possible candidates for the other enzymes involved in CCK processing, the distribution of PC1, PC2, and PC5 mRNA was studied in rat brain. Their colocalization with CCK mRNA was examined using double-label in situ hybridization. PC2 was the most abundant of these enzymes in terms of the intensity and number of cells labeled. It was widely colocalized with CCK. PC1 and PC5 mRNA-positive cells were less abundant, but they were also widely distributed and strongly colocalized with CCK in the cerebral cortex, hippocampus, amygdala, ventral tegmental area, and substantia nigra zona compacta. The degree of colocalization of the enzymes with CCK was regionally specific. It is clear that PC1 and PC5 are extensively colocalized with CCK and could be participating in CCK processing in the rat brain and may be able to substitute for PC2 in its absence. These three enzymes may represent a redundant system to ensure production of biologically active CCK.

The organization of intrinsic axonal projections of principal neurons in the main olfactory bulb (MOB) was studied in hamsters by using wheat germ agglutinin-horseradish peroxidase (WGA-HRP) and fluorescent dyes. Punctate injections of either WGA-HRP or fast blue (FB) that are restricted to small sectors on one side of the MOB produce comparably restricted fields of retrograde labeling on the opposite side. Label is found predominantly in superficially situated (middle and external) tufted cells that lie near and at the border between the external plexiform and glomerular layers. Few of the deeper middle tufted, internal tufted, or mitral cells and no external tufted cells that lie in the superficial two-thirds of the glomerular layer are labeled in regions remote to the injection site. Anterograde transport of WGA-HRP from the injection site labels axons that travel dorsally and ventrally in restricted bands through the internal plexiform layer and then terminate within this layer in the punctate sector on the opposite side that contains retrogradely labeled neurons. Such reciprocal projections between opposing regions of the medial and lateral sides of the MOB are found at all rostrocaudal and dorsoventral levels. When punctate injections of FB into the MOB are paired with restricted injections of a second fluorescent tracer (nuclear yellow or diamidino yellow dihydrochloride) into the appropriate sector of pars externa (pE) of the anterior olfactory nucleus, the punctate region of remote retrogradely labeled principal neurons is embedded within a topographically restricted longitudinal wedge of retrogradely labeled mitral and tufted cells that project extrinsically to or through pE. However, extremely few of these neurons are double-retrogradely labeled. The results reveal the existence of an intrabulbar associational system in which principal neurons engage in point-to-point, reciprocal projections between opposing regions of the medial and lateral MOB. Moreover, the results indicate that this associational system largely arises from superficially situated tufted cells distinct from those that support bulbofugal projections into the topographically organized interbulbar commissural system via pE.