The rostral ventrolateral medulla (RVLM) contains neurons critical for cardiovascular, respiratory, metabolic, and motor control. The activity of these neurons is controlled by inputs from multiple identified brain regions; however, the neurochemistry of these inputs is largely unknown. Gamma-aminobutyric acid (GABA) and enkephalin tonically inhibit neurons within the RVLM. The aim of this study was to identify all brain regions that provide GABAergic or enkephalinergic input to the rat RVLM. Neurons immunoreactive for cholera toxin B (CTB-ir), retrogradely transported from the RVLM, were assessed for expression of glutamic acid decarboxylase (GAD67) or preproenkephalin (PPE) mRNA using in situ hybridization. GAD67 mRNA was expressed in CTB-ir neurons in the following regions: the nucleus of the solitary tract (NTS, 6% of CTB-ir neurons), area postrema (AP, 8%), caudal ventrolateral medulla (17%), midline raphe (40%), ventrolateral periaqueductal gray (VLPAG, 15%), lateral hypothalamic area (LHA, 25%), central nucleus of the amygdala (CeA, 77%), sublenticular extended amygdala (SLEA, 86%), interstitial nucleus of the posterior limb of the anterior commissure (IPAC, 56%), bed nucleus of the stria terminals (BNST, 59%), and medial preoptic area (MPA, 53%). PPE mRNA was expressed in CTB-ir neurons in the following regions: the NTS (14% of CTB-ir neurons), midline raphe (26%), LHA (22%), zona incerta (ZI, 15%), CeA (5%), paraventricular nucleus (PVN, 13%), SLEA (66%), and MPA (26%). Thus, limited brain regions contribute GABAergic and/or enkephalinergic input to the RVLM. Multiple neurochemically distinct pathways originate from these brain regions projecting to the RVLM.

Heart rate and blood pressure oscillate in phase with respiratory activity. A component of these oscillations is generated centrally, with respiratory neurons entraining the activity of pre-sympathetic and parasympathetic cardiovascular neurons. Using a combination of optogenetic inhibition and excitation in vivo and in situ in rats, as well as neuronal tracing, we demonstrate that preBötzinger Complex (preBötC) neurons, which form the kernel for inspiratory rhythm generation, directly modulate cardiovascular activity. Specifically, inhibitory preBötC neurons modulate cardiac parasympathetic neuron activity whilst excitatory preBötC neurons modulate sympathetic vasomotor neuron activity, generating heart rate and blood pressure oscillations in phase with respiration. Our data reveal yet more functions entrained to the activity of the preBötC, with a role in generating cardiorespiratory oscillations. The findings have implications for cardiovascular pathologies, such as hypertension and heart failure, where respiratory entrainment of heart rate is diminished and respiratory entrainment of blood pressure exaggerated.

Previous studies have demonstrated that a range of stimuli activate neurons, including catecholaminergic neurons, in the ventrolateral medulla. Not all catecholaminergic neurons are activated and other neurochemical content is largely unknown hence whether stimulus specific populations exist is unclear. Here we determine the neurochemistry (using in situ hybridization) of catecholaminergic and noncatecholaminergic neurons which express c-Fos immunoreactivity throughout the rostrocaudal extent of the ventrolateral medulla, in Sprague Dawley rats treated with hydralazine or saline. Distinct neuronal populations containing PPCART, PPPACAP, and PPNPY mRNAs, which were largely catecholaminergic, were activated by hydralazine but not saline. Both catecholaminergic and noncatecholaminergic neurons containing preprotachykinin and prepro-enkephalin (PPE) mRNAs were also activated, with the noncatecholaminergic population located in the rostral C1 region. Few GlyT2 neurons were activated. A subset of these data was then used to compare the neuronal populations activated by 2-deoxyglucose evoked glucoprivation (Brain Structure and Function (2015) 220:117). Hydralazine activated more neurons than 2-deoxyglucose but similar numbers of catecholaminergic neurons. Commonly activated populations expressing PPNPY and PPE mRNAs were defined. These likely include PPNPY expressing catecholaminergic neurons projecting to vasopressinergic and corticotrophin releasing factor neurons in the paraventricular nucleus, which when activated result in elevated plasma vasopressin and corticosterone. Stimulus specific neurons included noncatecholaminergic neurons and a few PPE positive catecholaminergic neuron but neurochemical codes were largely unidentified. Reasons for the lack of identification of stimulus specific neurons, readily detectable using electrophysiology in anaesthetized preparations and for which neural circuits can be defined, are discussed.

Glucoprivation or hypoglycemia induces a range of counterregulatory responses, including glucose mobilization, reduced glucose utilization, and de novo glucose synthesis. These responses are mediated in part by the sympathetic nervous system. The aim of this study was to determine the chemical codes of sympathetic preganglionic neurons (SPN) activated by glucoprivation, induced by 2-deoxy-D-glucose (2DG). SPN controlling the adrenal glands and celiac ganglia, which ultimately can innervate the liver and pancreas, were targeted together with the superior cervical ganglia (control). 23.9% ± 1.3% of SPN in the T4-T11 region contained c-Fos immunoreactivity following 2DG; 70.3% ± 1.8% of SPN innervating the adrenal glands and 37.4% ± 3% of SPN innervating celiac ganglia were activated. 14.8% ± 3.5% of SPN (C8-T3) innervating superior cervical ganglia were activated. In the C8-T3 region 55% ± 10% of SPN activated contained PPCART, with only 12% ± 3% expressing PPE mRNA, whereas, in the T4-T11 region, 78% ± 4% contained PPE, with only 6.0% ± 0.6% expressing PPCART mRNA. Thus CART is not involved in glucose mobilization. Two chemically distinct populations of SPN (PPE⁺ 57.4% ± 5%, PPE⁻ ∼40%) were identified to regulate adrenaline release in response to glucoprivation. Multiple chemically distinct SPN populations innervating a specific target could suggest their graded recruitment. The two distinct populations of SPN (PPE⁺ 67.6% ± 9%, PPE⁻ ∼30%) projecting to celiac ganglia activated by glucoprivation could direct pancreatic and hepatic or other counterregulatory responses. Nearly all SPN that expressed PPE mRNA and projected to the adrenal glands or celiac ganglia were activated, suggesting a role for the inhibitory peptide enkephalin in responses evoked by glucoprivation.

Microinjection of somatostatin (SST) causes site-specific effects on respiratory phase transition, frequency, and amplitude when microinjected into the ventrolateral medulla (VLM) of the anesthetized rat, suggesting selective expression of SST receptors on different functional classes of respiratory neurons. Of the six subtypes of SST receptor, somatostatin 2a (sst2a ) is the most prevalent in the VLM, and other investigators have suggested that glutamatergic neurons in the preBötzinger Complex (preBötC) that coexpress neurokinin-1 receptor (NK1R), SST, and sst2a are critical for the generation of respiratory rhythm. However, quantitative data describing the distribution of sst2a in respiratory compartments other than preBötC, or on functionally identified respiratory neurons, is absent. Here we examine the medullary expression of sst2a with particular reference to glycinergic/expiratory neurons in the Bötzinger Complex (BötC) and NK1R-immunoreactive/inspiratory neurons in the preBötC. We found robust sst2a expression at all rostrocaudal levels of the VLM, including a large proportion of catecholaminergic neurons, but no colocalization of sst2a and glycine transporter 2 mRNA in the BötC. In the preBötC 54% of sst2a -immunoreactive neurons were also positive for NK1R. sst2a was not observed in any of 52 dye-labeled respiratory interneurons, including seven BötC expiratory-decrementing and 11 preBötC preinspiratory neurons. We conclude that sst2a is not expressed on BötC respiratory neurons and that phasic respiratory activity is a poor predictor of sst2a expression in the preBötC. Therefore, sst2a is unlikely to underlie responses to BötC SST injection, and is sparse or absent on respiratory neurons identified by classical functional criteria. J. Comp. Neurol. 524:1384-1398, 2016. © 2015 Wiley Periodicals, Inc.

Long-term hypercapnia is associated with respiratory conditions including obstructive sleep apnea, chronic obstructive pulmonary disease and obesity hypoventilation syndrome. Animal studies have demonstrated an initial (within hours) increase in ventilatory drive followed by a decrease in this response over the long-term (days-weeks) in response hypercapnia. Little is known about whether changes in the central respiratory chemoreflex are involved. Here we investigated whether central respiratory chemoreceptor neurons of the retrotrapezoid nucleus (RTN), which project to the respiratory pattern generator within the ventral respiratory column (VRC) have a role in the mechanism of neuroplasticity associated with long-term hypercapnia. Adult male C57BL/6 mice (n = 5/group) were used. Our aims were (1) to determine if galanin, neuromedin B and gastrin-releasing peptide gene expression is altered in the RTN after long-term hypercapnia. This was achieved using qPCR to measure mRNA expression changes of neuropeptides in the RTN after short-term hypercapnia (6 or 8 h, 5 or 8% CO2) or long-term hypercapnia exposure (10 day, 5 or 8% CO2), (2) in the mouse brainstem, to determine the distribution of preprogalanin in chemoreceptors, and the co-occurrence of the galanin receptor 1 (GalR1:Gi-coupled receptor) with inhibitory GlyT2 ventral respiratory column neurons using in situ hybridization (ISH) to better characterize galaninergic RTN-VRC circuitry, (3) to investigate whether long-term hypercapnia causes changes to recruitment (detected by cFos immunohistochemistry) of respiratory related neural populations including the RTN neurons and their galaninergic subset, in vivo. Collectively, we found that hypercapnia decreases neuropeptide expression in the RTN in the short-term and has the opposite effect over the long-term. Following long term hypercapnia, the number of RTN galanin neurons remains unchanged, and their responsiveness to acute chemoreflex is sustained; in contrast, we identified multiple respiratory related sites that exhibit blunted chemoreflex activation. GalR1 was distributed in 11% of preBötC and 30% of BötC glycinergic neurons. Our working hypothesis is that during long-term hypercapnia, galanin co-release from RTN neurons may counterbalance glutamatergic inputs to respiratory centers to downscale energetically wasteful hyperventilation, thereby having a role in neuroplasticity by contributing to a decrease in ventilation, through the inhibitory effects of galanin.