Search

Two neuron types contact the Mauthner cell (M cell) in the axon cap, a specialized region of high electrical resistance surrounding the initial segment of the M cell axon. One type produces a mixed electrical and chemical inhibition of the M cell. The second sends axons into the central core of the axon cap, where they spiral around the initial segment making both conventional synapses and gap junction contacts. The origin and synaptic effects of these spiral fibers have not been studied previously. When goldfish M cells were filled with Lucifer yellow, presynaptic spiral fibers were seen in the axon cap. These fibers could be traced back through the medial longitudinal fasciculus to their somata, near the contralateral fifth nerve motor nucleus. The same somata were labeled by horseradish peroxidase injected extracellularly into the axon cap. Recordings were made in the axon cap and the M cell after stimulation of hindbrain areas near the spiral fiber somata and axons. Extracellularly, a negative potential was observed close to the termination of the spiral fibers and termed the spiral fiber potential (SFP). Intracellularly, a graded, short latency depolarization of the M cell corresponded to the SFP and could cause the M cell to spike. This depolarization did not shunt the membrane, indicating that it may be produced through gap junctions. Intracellular responses to hindbrain stimulation also had a chloride-dependent, second component that shunted the membrane during paired-pulse testing. This inhibitory second component was probably evoked by cells other than the spiral fiber cells themselves.

It has been suggested that the principal output cells of the main olfactory bulb, the mitral cells, along with the glomerulus they enter, form an anatomical and functional column. To test the extent to which mitral cell somata lying close together in the mitral cell layer are connected to the same glomerulus, we reconstructed 267 mitral cells labeled by extracellular injections of horseradish peroxidase (HRP) in the external plexiform layer. Results show that apical dendrites tilt rostrally from the somata to the glomeruli and this tilt is gradually greater as the somata are located more and more rostrally in the olfactory bulb. The apical dendrites of most mitral cells in the same region of the bulb are parallel. We analyzed the degree to which dendrites were parallel by measuring the difference in angle of all pairs of apical dendrites in each section, with the restriction that no cells were separated by more than 500 microns. About half of these pairs of dendrites differed in angle by 20 degrees or less and were therefore said to be parallel. The degree of parallelism did not vary with the cell pair location or with intercell distance. Study of glomerular connections of pairs of mitral cells as a function of intercell distance revealed that 96% of mitral cells connected to the same glomerulus were separated by less than 120 microns, while 72% of cells connected to adjacent glomeruli and only 29% of cells connected to distant, i.e., nonadjacent, glomeruli were separated by less than 120 microns.(ABSTRACT TRUNCATED AT 250 WORDS)

Sensory systems rely upon spatially organized projections for the faithful transfer of receptor activity into the central nervous system. While the mammalian olfactory nerve shows a regional topographical organization, data on the sources of variation within this projection are scarce. We evaluated the degree of precision of olfactory bulb glomerular innervation from the olfactory nerve layer (ONL) by making small cuts across the trajectory of the olfactory nerve fibers and assessing the resulting denervation by wheat germ agglutinin-horseradish peroxidase (WGA-HRP) labeling of primary nerve fibers. In control animals (no ONL cut), the ONL and glomerular layer showed intense labeling around the entire circumference of the bulb. Transneuronal labeling of the mitral cell layer and the external plexiform layer was also observed in animals surviving at least 2 days, which allowed us to study the nerve fiber innervation of second order cells as well. ONL cuts on the lateral face of the bulb produced a stripe of denervation, as evidenced by the absence of primary and transneuronal label in a well-defined region. Fascicles of fibers within the ONL were never observed to enter the denervated region from more than one or two glomerular widths away, indicating a relatively tight limit of variation in the spatial termination of olfactory nerve fibers on the lateral bulb surface. This limit of variation does not hold around the circumference of the bulb since similar ONL cuts on the dorsal surface failed to produce a distinct stripe of denervation. Field potential measurements from the ONL after similar cuts supported the anatomic findings. These results show that primary olfactory axons and their terminals lie in parallel along the lateral face of the olfactory bulb. These spatial relationships of olfactory input to the bulb are maintained in the second order connections. This organization allows one to study the spatial aspects of interneuronally mediated synaptic mechanisms involved in olfactory coding.

Several types of short axon cells of the mammalian olfactory bulb have been described after Golgi impregnation. Two of these types have been observed in our material after treatment with the NADPH-diaphorase procedure or after immunohistochemistry for neuropeptide-Y (NPY). The cells stained by the two procedures have similar morphologies and distributions. A less extensive series of observations confirms that similar cells also display somatostatin (SS)-like immunoreactivity. One of these cell types corresponds to the superficial short axon cell of Golgi and electron microscopic studies. The dendrites of this cell lie within the periglomerular region and in the superficial external plexiform layer (EPL), generally lying parallel to the glomerular layer. In some cases the axon has been traced across the EPL into the granule cell layer (GCL). This cell may provide another route of interaction between the periglomerular region and the granule cells in addition to the influences conducted by basal dendrites and axon collaterals of some mitral and tufted cells. A type of deep short axon cell is also visible with these two procedures. It lies deep in the granule cell layer, frequently near the ventricular layer and its dendrites lie parallel to that layer. This deep short axon cell is stained with much greater frequency by the NADPH-diaphorase and NPY procedures than is the superficial short axon cell. It corresponds most closely to the Blanes or Golgi cells of the Golgi impregnation literature, but it appears to differ from these cells in the position and orientation of its dendrites. No spines have been observed on either the superficial or deep cells in this series. Many glomeruli are also stained by the NADPH-diaphorase procedure, but are not NPY or SS immunoreactive. This may provide additional evidence for functional differences between glomeruli in local regions of the olfactory bulb.

The pattern of output of mitral and tufted cells of the rat olfactory bulb (OB) to layer Ia overlying the pars externa (pE) of the anterior olfactory nucleus (AON) has been studied in the rat by iontophoresis of horseradish peroxidase and Phaseolus vulgaris-leucoagglutinin. These agents labeled mitral and tufted cells and at least the proximal portion of their axons. In most cases we observed small branches from axons of the lateral olfactory tract that appear to terminate in the region of the pE AON, while the main axon could often be traced for considerable distances past these branches. These branches are assumed to terminate in the pE AON because they could not be traced to other terminal regions, because they ramify in layer Ia, and because they usually show small swellings characteristic of axons in terminal regions. Although each ramification could be extensive, we found that the positions of these small branches were related to the positions of the injections within the OB. Dorsal medial injections labeled dorsal branches. Ventral medial injections labeled ventral branches. Injections on the lateral face of the OB labeled intermediate branches. The centers of the regions within which branches were labeled were strongly correlated with the positions of the injection around the circumference. Comparison of the anterior-posterior axis of the OB produced no such strong correlation. Reconstructions of axons showed that terminal branches arise from both mitral and tufted cells, although at least some mitral cells are shown not to have such branches in the pE AON. Studies of the patterns of dendrites and terminals in the pE AON indicate that this region has the same pattern of layer Ia and Ib terminals seen in other olfactory cortical regions. The pE AON cell layer is intercalated just below the boundary between layers Ia and Ib. Since dendrites of the underlying pars lateralis of the AON (pL AON) penetrate into layer Ia over much of the pE AON, it is necessary to remember that at least part of the pL AON may also receive topographically organized inputs.

The output cells of the main olfactory bulb, the mitral and tufted cells, can be categorized into subclasses on the basis of their intrabulbar dendritic and axonal characteristics. Their form was studied in rats following labeling by iontophoretic injection of horseradish peroxidase into the external plexiform layer (EPL). The fact that these extracellular injections labeled small numbers of neurons permitted reconstruction of individual cells. The injection depth within the EPL determined the type of cells labeled. Secondary dendrites of each cell type lay in one of three partially overlapping zones in the EPL. The deepest zone contained the secondary dendrites of one group of mitral cells (Type I), which had the deepest and longest dendrites of the output cells. An intermediate zone of the EPL contained the secondary dendrites of middle tufted and a second class of mitral cells (Type II). The superficial zone, adjacent to the glomerular layer, contained the relatively short, asymmetric dendritic fields of external tufted cells. The few labeled internal tufted cells had secondary dendrites in either the intermediate or deep zones. Every cell type, except the Type I mitral cells, had axon collaterals in the internal plexiform and upper granule cell layers. No cell types had axons re-entering the EPL. These results for output cells combined with our previous observations on granule cells point to a functional sublaminar organization of the EPL that has not previously been proposed.

The organization of the projection of olfactory bulb output cells was studied in the rat by injection of horseradish peroxidase (HRP) into the piriform cortex or olfactory tubercle. We made single HRP injections into small cuts in the fiber layer of the projection areas in order to enhance uptake by axons and to confine the region of HRP uptake. Following most of these injections, HRP-labeled axons could be traced in discrete fascicles through the fiber layer of the cortex or tubercle. These observations indicate that axons innervating the piriform cortex do not emit many long collaterals after they leave the lateral olfactory tract. HRP-labeled cells were generally observed throughout the ipsilateral olfactory bulb, but there were regions of greater density of labeled cells that differed in the various brains. The differences among the distributions of labeled mitral and tufted cells were analyzed statistically in 39 brains to test whether they varied systematically with injection site. In these analyses, the olfactory bulb was divided into 30 standard regions, and the labeled cells in each regions were counted. The distributions of labeled cells were similar for brains where injections were made into similar regions of the piriform cortex. The variations in density of labeled cells of the dorsal and anterior regions of the olfactory bulb were most strongly correlated with the positions of cortical injections. In contrast, the posterior medial regions of the bulb were heavily labeled after almost all injections. The ventral portions of the olfactory bulb were most heavily labeled after injections into the olfactory tubercle.

Cytochrome oxidase staining of the rat olfactory bulb external plexiform layer (EPL) produces a darkly stained intermediate zone bordered by lightly stained superficial and deep zones. Similar zonal staining was seen in cats, rabbits, and hamsters. These zones vary in relative thickness around the circumference of the olfactory bulb; the deep zone is proportionally thicker in the most dorsal and ventral parts of the bulb. Tufted cell somata are unevenly distributed within the EPL; the outer part of the EPL has more somata. The distribution of the cytochrome oxidase reaction product shows that the darkly stained intermediate zone is not produced by staining of tufted cell somata. Zones of cytochrome oxidase staining correspond to the sublaminar distribution of mitral and tufted cell basal dendrites. This was demonstrated by labeling mitral and tufted cells with small extracellular horseradish peroxidase injections and processing alternate sections for horseradish peroxidase and for cytochrome oxidase. Because there was cross-reaction of the cytochrome oxidase procedure with horseradish peroxidase, it was possible to trace many neurons through both series of sections. Middle tufted cells of the superficial EPL have basal dendrites confined to the superficial zone of light cytochrome oxidase staining. Internal tufted cells and middle tufted cells of the intermediate zone send their basal dendrites into the intermediate zone. One group of mitral cells (type I) has basal dendrites confined to the deep zone of lighter cytochrome oxidase staining. A second group of mitral cells (type II) and tufted cells of the intermediate EPL has basal dendrites primarily confined to the intermediate zone of dark cytochrome oxidase staining. The correlation of the enzyme staining with the dendritic laminar patterns supports the existence of three relatively distinct sublaminae in the EPL and supports the designation of two types of mitral cells. The staining pattern also provides an independent method for evaluating the sublaminae of the EPL without the necessity of labeling individual groups of cells. Finally, the staining pattern suggests that the intermediate zone of the EPL may be subjected to more tonic synaptic input, causing it to have an increased level of metabolic activity.

Studies on the morphological organization of the main olfactory bulb have indicated that there are subpopulations of granule cells with different dendritic patterns in the external plexiform layer (EPL). Small, extracellular injections of horseradish peroxidase (HRP) were made iontophoretically into superficial and deep parts of the EPL and the granule cell layer (GCL) in adult rats. Superficial EPL injections principally labeled superficial granule cell somata, whereas deep EPL injections labeled both superficial and deep granule cell somata. Injections in the superficial GCL labeled granule cell dendritic processes extending across the entire EPL. However, deep GCL injections labeled few granule cell dendrites in the superficial EPL, but labeled many such processes in the deep EPL. These results were the same in material processed with the Hanker-Yates procedure, where the morphology of individual neurons could be studied, and in the more sensitive tetramethyl benzidine procedure. Serial reconstructions of individual granule cells were made from both HRP and Golgi-Kopsch material. The distal dendrites of deep granule cells reached only as far as the deep EPL, where they branched extensively and had many dendritic spines. The dendrites of superficial granule cells, however, reached the most superficial part of the EPL where they ramified most extensively. The superficial granule cells typically had a higher spine density in the superficial part of the EPL than in the deep part. On the basis of these results, we conclude that the superficial granule cells predominantly innervate the superficial EPL and that the deep granule cells exclusively influences on the bulbar output neurons, the mitral and tufted cells, through reciprocal dendrodendritic synapses. Since the secondary dendrites of the tufted cells ramify in the superficial EPL and the dentrites of most mitral cells ramify in deep EPL, the superficial and deep granule cells may preferentially modulate the responses of tufted and mitral cells, respectively.