Rostrocaudal differences in morphology and neurotransmitter content of cells in the subretrofacial vasomotor nucleus

Abstract

The rostral ventrolateral medulla (RVLM) contains sympathoexcitatory neurons that exert a powerful control over the sympathetic outflow to the cardiovascular system. In the cat there is a concentration of such neurons (but not neurons subserving other functions) within a narrow longitudinal column in the RVLM termed the subretrofacial (SRF) nucleus. Furthermore, it has been suggested that there are subgroups of cells, located at different rostrocaudal levels of the SRF nucleus, that preferentially or exclusively control different vascular beds (e.g. in the kidney and hindlimb). The aim of this study was to map quantitatively the rostrocaudal distribution within the nucleus of different cell types, defined according to morphological and/or chemical criteria, and to correlate this with the regional vasomotor effects (in hindlimb and kidney) evoked by stimulation of SRF cells at the corresponding rostrocaudal levels. SRF cells were highly heterogeneous with respect to both their morphology and chemical properties. They varied greatly in size (equivalent diameter ranging from 10–40 μm) as well as in shape and orientation. An immunohistochemical examination using the avidin-biotin procedure revealed that many SRF cells (estimated 57% of all SRF cells) were immunoreactive for tyrosine hydroxylase (TH, a marker of catecholamine cells). In addition, there were SRF cells immunoreactive for neuropeptide Y (NPY, 11% of total), enkephalin (ENK, 16% of total), and serotonin (5HT, 10% of total), but not for substance P, galanin or somatostatin. Different cell types, defined according to their morphology and/or chemical properties, were unevenly distributed throughout the nucleus. In the most caudal part of the SRF nucleus, virtually all cells were TH-positive, and the large majority (estimated 80%) were NPY-positive, suggesting that many cells at this level contained both TH and NPY. In contrast, in the most rostral part of the SRF nucleus, only 30% of cells were TH-positive, and no NPY-positive cells were observed. Both 5HT- and ENK-positive cells were found throughout the rostrocaudal extent of the nucleus, but predominantly within its rostral part. Furthermore, TH-positive cells in the rostral SRF nucleus were on average significantly larger (mean equivalent diameter 18–43% greater) than TH/NPY-positive cells in the caudal part of the nucleus, but smaller than 5HT- or ENK-positive cells at the same level. Overall, rostral cells (regardless of their chemical type) were larger than caudal cells within the SRF nucleus (mean equivalent diameter 13–28% greater). Talking into account both morphological and chemical properties, a minimum of four distinct cell subgroups in the SRF nucleus could be defined: caudal TH/NPY-positive cells, rostral TH-positive (but NPY-negative) cells, 5HT-positive cells (possibly also containing ENK), and cells that were not immunoreactive for any of these compounds. Large (> 100%) increases in blood pressure and iliac vascular resistance were elicited by microinjection of picomolar amounts of sodium glutamate into sites throughout the entire rostrocaudal extent of the SRF nucleus. In contrast, large or moderate (> 50%) increases in renal sympathetic nerve activity were evoked only from sites within the most rostral part of the nucleus. Correlation of the functional and immunohistochemical data indicates that NPY/TH-positive cells in the caudal SRF nucleus contribute to the regulation of the hindlimb but not the renal vascular bed. Secondly, they suggest that both catecholamine