Morphometry of a peptidergic transmitter system: dynorphin B-like immunoreactivity in the rat hippocampal mossy fiber pathway before and after seizures.

While the morphometry of classical transmitter systems has been extensively studied, relatively little quantitative information is available on the subcellular distribution of peptidergic dense core vesicles (DCVs) within axonal arbors and terminals, and how distribution patterns change in response to neural activity. This study used correlated quantitative light and electron microscopic immunohistochemistry to examine dynorphin B-like immunoreactivity (dyn B-LI) in the rat hippocampal mossy fiber pathway before and after seizures. Forty-eight hours after seizures induced by two pentylenetetrazol injections, light microscopic dyn B-LI was decreased dorsally and increased ventrally. Ultrastructural examination indicated that, in the hilus of the dentate gyrus, these alterations resulted from changes that were almost entirely restricted to the profiles of the large mossy-like terminals formed by mossy fiber collaterals (which primarily contact spines), compared to the profiles of the smaller, less-convoluted terminals found on the same collaterals (which primarily contact aspiny dendritic shafts). Dorsally, mossy terminal profile labeled DCV (/DCV) density dropped substantially, while ventrally, both mossy terminal profile perimeter and /DCV density increased. In all terminal profile examined, /DCVs also were closely associated with the plasma membrane. Following seizures, there was a reorientation of /DCVs along the inner surface of mossy terminal profile membranes, in relation to the types of profiles adjacent to the membrane: in both the dorsal and ventral hilus, significantly fewer /DCVs were observed at sites apposed to dendrites, and significantly more were observed at sites apposed to spines. Thus, after seizures, changes specific to: (1) the dorsoventral level of the hippocampal formation, (2) the type of terminal, and (3) the type of profile in apposition to the portion of the terminal membrane examined were all observed. An explanation of these complex, interdependent alterations will probably require evoking multiple interrelated mechanisms, including selective prodynorphin synthesis, transport, and release.

Selective changes in hippocampal neuropeptide Y neurons following removal of the cholinergic septal inputs.

The number and distribution of subpopulations of hilar interneurons containing neuropeptide Y (NPY), somatostatin (SOM), or gamma-aminobutyric acid (GABA) immunoreactivities were examined in the hilus of the dentate gyrus following removal of the cholinergic septal inputs. One, 2, 4, 8, 12, and 24 weeks after intracerebroventricular injections of immunotoxin, consisting of antibody to the low-affinity nerve growth factor receptor conjugated to saporin (192 IgG-saporin), lesioned rats were processed simultaneously with controls for NPY, SOM, or GABA immunolabeling. Across all time points, the number of NPY-labeled neurons was reduced to a statistically significant level (paired t-test, P = 0.001) in the injected rats (73% of control values, on average). The decrease in the number of NPY-labeled neurons was not limited to any particular subregion rostrally but appeared greater in the central region caudally. The size of NPY-labeled neurons did not differ statistically between control and immunolesioned rats examined at 1, 2, and 24 week time points. In contrast, the number of both SOM- and GABA-immunoreactive neurons in injected rats did not appear to be affected in any consistent manner. Examination of the hilus in adjacent Nissl-stained sections with the optical dissector revealed that although the total number of small nonprincipal cells (5-15 microm in diameter) did not appear affected at the 4-week time point, there was a statistically significant (P = 0.03) reduction across the 8-24-week time points (to 80% of control values, on average). Dual-labeling studies on separate rats showed that a small subpopulation of the NPY- and SOM-labeled neurons, primarily in the infragranular hilus, were colocalized with neurons containing GABA immunoreactivity (18% and 5%, respectively). These studies demonstrate that removal of the cholinergic septal inputs (1) can cause relatively rapid, selective decreases in the number of NPY-immunoreactive hippocampal interneurons and (2) appears to lead to the death of hippocampal interneurons over a longer time course. The changes in NPY immunoreactivity seem to occur in the portion of interneurons that probably does not contain either SOM or GABA immunoreactivity.

Nucleus tractus solitarius efferent terminals synapse on neurons in the caudal ventrolateral medulla that project to the rostral ventrolateral medulla.

The caudal ventrolateral medulla (CVL) contains neurons that are vasodepressor and are a critical component of the baroreceptor reflex pathway. While electrophysiological studies suggest that CVL neurons are intercalated in the baroreceptor pathway between the nucleus tractus solitarius (NTS) and the rostral ventrolateral medulla (RVL), there is no direct evidence for this projection. Therefore, we identified CVL neurons that project to RVL by retrogradely labelling them with wheat germ agglutinin-apo-horseradish peroxidase conjugated to colloidal gold (WAHG) injected into the RVL. Retrogradely labelled neurons were seen in previously identified vasodepressor areas of the rostral CVL that are critical for the baroreceptor reflex. Double labelling for WAHG and tyrosine hydroxylase (TH) immunocytochemistry indicated that CVL neurons that project to the RVL (CVL --> RVL neurons) are distinct from the noradrenergic neurons of the A1 cell group. To establish the presence of a direct projection from the NTS to CVL --> RVL neurons, the retrograde tracer WAHG was pressure injected into the RVL and the anterograde tracer biocytin was iontophoresed into the NTS of anesthetized rats. After 4-6 h, anesthetized rats were perfused transcardially with 3.75% acrolein in 2% paraformaldehyde and sections through the CVL were processed for both markers. By light microscopy, numerous biocytin-labelled varicose processes overlapped neurons containing WAHG in the CVL. By electron microscopy, biocytin was found in myelinated and unmyelinated axons and in axon terminals (0.9 + 0.02 microns) that contained primarily small clear vesicles. These terminals formed predominantly asymmetric synapses on large (1.5-6.0 microns in diameter) dendrites within the CVL. Some of the post-synaptic perikarya and large dendrites contained WAHG associated with lysosomes and multivesicular bodies, indicating that they belong to neurons which project to the RVL. We conclude that CVL --> RVL neurons are (a) distinct from A1 noradrenergic cells; (b) receive direct synaptic contacts from NTS efferent terminals; (c) are potently and monosynaptically excited (asymmetric synapses) by NTS efferent terminals. These data support the hypothesis that CVL neurons are intercalated between the NTS and the RVL in the baroreceptor reflex pathway.

Septohippocampal neurons in the rat septal complex have substantial glial coverage and receive direct contacts from noradrenaline terminals.

The ultrastructure of septohippocampal neurons in the septal complex and their relations with catecholamine, principally noradrenaline, terminals were examined in single thin sections. Projection neurons were identified by retrograde transport of wheat-germ agglutinated apo-horseradish peroxidase conjugated to colloidal gold particles (WAHG) following an injection into the hippocampal formation of anesthetized adult rats. After a 1 day survival, sections through the septal complex were labeled with antibodies to tyrosine hydroxylase (TH) or dopamine-beta-hydroxylase (DBH). By light microscopy, numerous processes with TH- and DBH-immunoreactivity were near neurons containing retrogradely transported WAHG. By electron microscopy, most WAHG was associated with lysosomes, multivesicular and 'sequestration' bodies in the cytoplasm of perikarya and large dendrites. WAHG-labeled perikarya (n = 114) had a large amount of astrocytic coverage (> 60% of surface) and a low amount of terminal coverage (< 25%). WAHG-labeled perikarya and dendrites were either directly contacted by TH- or DBH-labeled terminals or abutted glial processes apposed to TH- or DBH-labeled terminals. Immunoreactivity for TH and DBH was found primarily in axons and axon terminals. The morphology and synaptic associations of TH-labeled terminals was similar to that reported previously. DBH-labeled terminals (n = 314; 0.5 +/- 0.2 microns in diameter) contained numerous small clear vesicles and from 0-4 large, dense-core vesicles. DBH-containing terminals: (1) contacted perikarya and dendrites (58%), 10% of which contained WAHG; (2) were closely apposed to other terminals (7%); or (3) were separated by glial processes (35%). DBH-labeled terminals formed chiefly symmetric synapses on perikarya. However, most DBH-containing terminals formed both asymmetric and symmetric synapses on the shafts of small dendrites, suggesting both excitatory and inhibitory functions for noradrenaline terminals on septal neurons. The results demonstrate that septohippocampal neurons (1) are mostly engulfed by astrocytes and have very little terminal coverage; (2) are both directly contacted (synapses) and indirectly contacted (appositions to apposing astrocytes or axon terminals) by catecholamine, particularly noradrenaline, terminals.