C. Michael Foley

Immunohistochemical localization of GAD67‐expressing neurons and processes in the rat brainstem: Subregional distribution in the nucleus tractus solitarius

The role of γ‐aminobutyric acid (GABA) in homeostatic control in the brainstem, in particular, in the nucleus tractus solitarius (NTS), is well established. However, to date, there is no detailed description of the distribution of GABAergic neurons within the NTS. The goal of the current study was to reexamine the efficacy of immunohistochemical localization of glutamic acid decarboxylase (GAD) protein, specifically the 67‐kDa isoform (GAD67), as a marker for GABAergic neurons in the medulla and to provide a detailed map of GAD67‐immunoreactive (‐ir) cells within rat NTS by using a recently developed mouse monoclonal antibody. We describe a distribution of GAD67‐ir cells in the medulla similar to that reported previously from in situ hybridization study. GAD67‐ir cells were localized in regions known to contain high GABA content, including the ventrolateral medulla, raphe nuclei, and area postrema, but were absent from all motor nuclei, although dense terminal labeling was discerned in these regions. In the NTS, GAD67‐ir was localized in all subregions. Semiquantitative analysis of the GAD67‐ir distribution in the NTS revealed greater numbers of GAD67‐ir cells medial to the solitary tract. Finally, dense GAD67 terminal labeling was found in the medial, central, intermediate, commissural, and subpostremal subregions, whereas sparse labeling was observed in the ventral subregion. Our findings support the use of immunohistochemistry for GAD67 as a marker for the localization of GABAergic cells and terminal processes in the rat brainstem. Furthermore, the reported heterogeneous distribution of GAD67‐ir in the NTS suggests differential inhibitory modulation of sensory processing. J. Comp. Neurol. 493:274–290, 2005.

Glutamate in the nucleus of the solitary tract activates both ionotropic and metabotropic glutamate receptors

Glutamate is the proposed neurotransmitter of baroreceptor afferents at the level of the nucleus of the solitary tract (NTS). Blockade of ionotropic glutamate receptors with kynurenic acid blocks the arterial baroreflex but, paradoxically, does not abolish the response to exogenous glutamate. This study tested the hypothesis that exogenous glutamate in the NTS activates both ionotropic and metabotropic glutamate receptors (mGluRs). In urethan-anesthetized rats, unilateral microinjections of glutamate into the NTS decreased mean arterial pressure, heart rate, and lumbar sympathetic nerve activity. The cardiovascular response to injection of glutamate was not altered by NTS blockade of mGluRs with α-methyl-4-carboxyphenylglycine (MCPG). Blockade of ionotropic glutamate receptors with kynurenic acid attenuated the response to glutamate injection. After combined NTS injection of MCPG and kynurenic acid, the response to glutamate was blocked. These data suggest that exogenous glutamate microinjected into the NTS acts at both ionotropic glutamate receptors and mGluRs. In addition, blockade of both classes of glutamate receptors is required to block the cardiovascular response to microinjection of glutamate in the NTS.

Regulation of arterial pressure by the paraventricular nucleus in conscious rats: interactions among glutamate, GABA, and nitric oxide.

The paraventricular nucleus (PVN) of the hypothalamus is an important site for autonomic and neuroendocrine regulation. Experiments in anesthetized animals and in vitro indicate an interaction among gamma-aminobutyric acid (GABA), nitric oxide (NO), and glutamate in the PVN. The cardiovascular role of the PVN and interactions of these neurotransmitters in conscious animals have not been evaluated fully. In chronically instrumented conscious rats, mean arterial pressure (MAP) and heart rate (HR) responses to microinjections (100 nl) in the region of the PVN were tested. Bilateral blockade of ionotropic excitatory amino acid (EAA) receptors (kynurenic acid, Kyn) in the PVN produced small but significant decreases in MAP and HR. GABAA receptor blockade (bicuculline, Bic), and inhibition of NO synthase [(NOS), N-(G)-monomethyl-L-arginine, L-NMMA] each increased MAP and HR. The NO donor sodium nitroprusside (SNP) produced depressor responses that were attenuated by Bic. NOS inhibition potentiated both pressor responses to the selective EAA agonist, N-methyl-D-aspartic acid (NMDA), and depressor responses to Kyn. Increases in MAP and HR due to Bic were blunted by prior blockade of EAA receptors. Thus, pressor responses to GABA blockade require EAA receptors and GABA neurotransmission contributes to NO inhibition. Tonic excitatory effects of glutamate in the PVN are tonically attenuated by NO. These data demonstrate that, in the PVN of conscious rats, GABA, glutamate, and NO interact in a complex fashion to regulate arterial pressure and HR under normal conditions.

Cardiovascular response to group I metabotropic glutamate receptor activation in NTS

Glutamate is the proposed neurotransmitter of baroreceptor afferents at the level of the nucleus tractus solitarius (NTS). Exogenous glutamate in the NTS activates neurons through ionotropic and metabotropic glutamate receptors (mGluRs). This study tested the hypothesis that group I mGluRs in the NTS produce depressor, bradycardic, and sympathoinhibitory responses. In urethan-anesthetized rats, unilateral 30-nl microinjections of the group I-selective mGluR agonist 3,5-dihydroxyphenylglycine (DHPG) into the NTS decreased mean arterial pressure, heart rate, and lumbar sympathetic nerve activity. The dose of drug that produced 50% of the maximal response (ED50) was 50-100 μM. The response to microinjection of equal concentrations of DHPG or the general mGluR agonist 1-aminocyclopentane-1 S,3 R-dicarboxylic acid (ACPD) produced similar cardiovascular effects. The cardiovascular response to injection of DHPG or ACPD was abolished by NTS blockade of mGluRs with α-methyl-4-carboxyphenylglycine (MCPG). Blockade of ionotropic glutamate receptors with kynurenic acid did not attenuate the response to DHPG or ACPD injection. These data suggest that DHPG and ACPD activate mGluRs in the NTS and do not require ionotropic glutamate receptors to produce their cardiovascular response. In the NTS the group I mGluRs produce responses that are consistent with excitation of neurons involved in reducing sympathetic outflow, heart rate, and arterial pressure.Glutamate is the proposed neurotransmitter of baroreceptor afferents at the level of the nucleus tractus solitarius (NTS). Exogenous glutamate in the NTS activates neurons through ionotropic and metabotropic glutamate receptors (mGluRs). This study tested the hypothesis that group I mGluRs in the NTS produce depressor, bradycardic, and sympathoinhibitory responses. In urethan-anesthetized rats, unilateral 30-nl microinjections of the group I-selective mGluR agonist 3,5-dihydroxyphenylglycine (DHPG) into the NTS decreased mean arterial pressure, heart rate, and lumbar sympathetic nerve activity. The dose of drug that produced 50% of the maximal response (ED50) was 50-100 microM. The response to microinjection of equal concentrations of DHPG or the general mGluR agonist 1-aminocyclopentane-1S,3R-dicarboxylic acid (ACPD) produced similar cardiovascular effects. The cardiovascular response to injection of DHPG or ACPD was abolished by NTS blockade of mGluRs with alpha-methyl-4-carboxyphenylglycine (MCPG). Blockade of ionotropic glutamate receptors with kynurenic acid did not attenuate the response to DHPG or ACPD injection. These data suggest that DHPG and ACPD activate mGluRs in the NTS and do not require ionotropic glutamate receptors to produce their cardiovascular response. In the NTS the group I mGluRs produce responses that are consistent with excitation of neurons involved in reducing sympathetic outflow, heart rate, and arterial pressure.

Nitric oxide synthase activity and expression are decreased in the paraventricular nucleus of pregnant rats

Pregnancy is characterized by elevated heart rate and decreased total peripheral resistance and arterial blood pressure. Plasma volume is expanded and plasma osmolality is decreased, yet vasopressin secretion in pregnant animals, including humans, is no different than levels in the nonpregnant state. Although reflex compensatory sympathoexcitation is suppressed, baseline sympathetic nerve activity to the heart and vasculature is well maintained or slightly elevated in pregnancy. Clearly there are central nervous system (CNS) adaptations in systems for regulation of cardiovascular and body fluid homeostasis in pregnant animals. The paraventricular nucleus (PVN) and supraoptic nucleus (SON) of the hypothalamus are important CNS sites for control of sympathetic nerve activity and vasopressin secretion. Nitric oxide (NO), an important neuromodulator in these hypothalamic nuclei, contributes to tonic inhibition of neurosecretory and pre-autonomic neurons. Alterations in NO within the PVN and SON could contribute to changes in regulation of vasopressin and sympathetic nerve activity in pregnancy. In the present study, nitric oxide synthase (NOS) activity (NADPH-diaphorase staining), neuronal NOS (nNOS) protein, and nNOS mRNA were assessed in nonpregnant estrus stage and near-term pregnant rats. nNOS mRNA, protein, and activity were greater in the PVN than in the SON. In the PVN only, pregnancy was associated with significant decreases in all three measurements for assessment of nNOS. Thus decreased NO production and relative disinhibition of the PVN may contribute to maintenance of baseline vasopressin secretion and baseline sympathetic nerve activity in the pregnant state.

CNS effects of ovarian hormones and metabolites on neural control of circulation.

Abstract: Pregnant women often experience orthostatic hypotension, and pregnancy is associated with increased susceptibility to hemorrhagic hypotension. Experiments evaluating arterial baroreflex control of efferent sympathetic nerve activity in virgin and term‐pregnant rats revealed that arterial baroreflex sympathoexcitation is attenuated, while sympathoinhibitory responses are well‐maintained or potentiated. Following a hypotensive challenge, pregnant animals exhibit attenuated Fos expression in the rostral ventrolateral medulla (RVLM), suggesting that unloading of arterial baroreceptors results in less excitation of presympathetic neurons in the brain stem. Other experiments, in which afferent baroreceptor discharge was recorded, suggest that this was not due to differences in afferent baoreceptor function. GABAergic mechanisms are responsible for tonic inhibition of sympathoexcitatory neurons in the RVLM and the major metabolite of progesterone, 3α‐OH‐dihydroprogesterone (3α‐OH‐DHP), which is elevated in pregnancy, is the most potent endogenous positive modulator of CNS GABAA receptor function. Additional experiments revealed that acutely administered 3α‐OH‐DHP, either intravenously or directly into the RVLM, mimicked the effects of pregnancy on baroreflex control of efferent sympathetic nerve activity and potentiated pressure sensitivity of spinally projecting RVLM neurons. Preliminary experiments using semiquantitative RT‐PCR, evaluated the relative expression of three subunits (α1–3) of the GABAA receptor, and suggest that chronic exposure to elevated levels of ovarian hormones can result to changes in GABAA receptor subunit composition. It is likely that changes in control of sympathetic outflow in pregnancy are related to complex interactions between genomic and nongenomic actions of ovarian hormones and metabolites.

GABAA α1 and α2 receptor subunit expression in rostral ventrolateral medulla in nonpregnant and pregnant rats

Abstract Pregnancy results in attenuated baroreflex mediated sympathoexcitatory responses which may be due to potentiation of γ-aminobutyric acid (GABA) inhibition in the rostral ventrolateral medulla (RVLM). The major metabolite of progesterone, 3α-hydroxy-dihydroprogesterone (3α-OH-DHP), which is elevated in pregnancy, is a potent neurosteroid positive modulator of GABA A receptors, and sensitivity of GABA A receptors to 3α-OH-DHP is dependent on the receptor subunit composition. The purpose of this study was to evaluate the GABA A α 1 and α 2 receptor subunit mRNA and protein expression in the RVLM of nonpregnant and late term pregnant rats. Micropunches of RVLM were collected from nonpregnant and late term pregnant rats and the expression levels of GABA A α 1 and α 2 receptor subunits were analyzed using quantitative competitive reverse transcriptase polymerase chain reaction (RT-PCR) and immunoblot techniques. The competitive RT-PCR analysis allows comparison of expression levels between different mRNA, and the mRNA expression level of GABA A α 1 was several hundred fold greater than GABA A α 2 in both groups. However, this relative distribution of GABA A α 1 and α 2 receptor subunits protein or mRNA expression was not altered in late term pregnant compared to nonpregnant rats. These data demonstrate, that within the RVLM of both nonpregnant and late term pregnant rats, the relative expression levels of GABA A α 1,2 receptor subunits favor GABA A receptors susceptible to positive modulation by progesterone metabolites.

Increased nitric oxide synthase activity and expression in the hypothalamus of hindlimb unloaded rats

Upon return from spaceflight or resumption of normal posture after bed rest, individuals often exhibit cardiovascular deconditioning. Although the mechanisms responsible for cardiovascular deconditioning have yet to be fully elucidated, alterations within the central nervous system have been postulated to be involved. The paraventricular nucleus (PVN) and supraoptic nucleus (SON) of the hypothalamus are important brain regions in control of sympathetic outflow and body fluid homeostasis. Nitric oxide (NO) modulates the activity of PVN and SON neurons, and alterations in NO transmission within these brain regions may contribute to symptoms of cardiovascular deconditioning. The purpose of the present study was to examine nitric oxide synthase (NOS) activity and expression in the PVN and SON of control and hindlimb unloaded (HU) rats, an animal model of cardiovascular deconditioning. The number of neurons exhibiting NOS activity as assessed by NADPH-diaphorase staining was significantly greater in the PVN but not SON of HU rats. Western blot analysis revealed that neuronal NOS (nNOS) but not endothelial NOS (eNOS) protein expression was higher in the PVN of HU rats. In the SON, there was a strong trend for an increase in nNOS (p=0.052) and a significant increase in eNOS expression in HU rats. Our results suggest that increased nNOS in the PVN contributes to autonomic and humoral alterations following cardiovascular deconditioning. In contrast, the functional significance of increases in nNOS and eNOS protein in the SON may be related to alterations in vasopressin release observed previously in HU rats.

Intravenous solid tip ECG lead placement in telemetry implanted dogs: Part 2: High quality telemetry signals yield high sensitivity to drug-induced changes.

INTRODUCTION Dogs are commonly used in cardiovascular drug safety assessment, and implanted telemetry models include subcutaneous or epicardial electrocardiogram (ECG) electrode placements. The purpose of this study was to determine the sensitivity of a canine telemetry model with intravenous ECG lead placement: the negative ECG lead (solid tip) inserted into the jugular vein and the positive lead sutured to the diaphragm. Reference drugs were administered to test the sensitivity to drug-induced changes. METHODS Twenty-four dogs were implanted with PCT or PCTP transmitters [Data Sciences International (DSI)]. Three reference drugs were administered: sotalol to eight PCT and milrinone to eight PCTP transmitter-implanted dogs. Twenty-four dogs received moxifloxacin (12 dogs/transmitter type). Telemetry data were collected for 25h and analyzed using double Latin squares for sotalol and milrinone data or a 4×4 or 3×6 parallel design for moxifloxacin data. Evaluated parameters were PR, QT, corrected QT (QTc), QRS, heart rate, left ventricular function, and hemodynamic data. Various correction factors for QTc interval were tested. Retrospective power analysis was performed to detect minimal absolute changes comparing a single to a double Latin square or the two parallel designs. RESULTS Expected changes on ECG and hemodynamic parameters were observed after administration of all reference drugs. The individual animal corrected QT (QTci) interval provided the optimal correction factor. Retrospective power analysis confirmed detection of smaller differences in double versus single Latin squares. Minimal detectable differences were smaller in both Latin squares compared to parallel designs, with smaller detectable differences in a 3×6 compared to a 4×4 parallel design. DISCUSSION The solid tip intravenous ECG lead configuration in dogs is a viable radiotelemetry model to detect drug-induced changes with high sensitivity. This model yields comparable signal quality and represents a refinement over epicardial ECG leads and allows for possible reduction in the number of animals if study design and size are selected based on needed assay sensitivity.

Evaluation of cardiovascular changes in dogs administered three positive controls using jacketed external telemetry-blood pressure (JET-BP).

INTRODUCTION Nonclinical safety studies are increasingly incorporating cardiac safety endpoints to discover potential cardiovascular liabilities. This trend for more thorough cardiovascular nonclinical safety evaluation is prudent given the high attrition rate of potential therapeutics due to unexpected cardiovascular liabilities discovered in late-stage clinical trials or post-market approval. In particular, the causal relationship of blood pressure changes that lead to risk of major adverse cardiac events suggests hemodynamic changes should be critically evaluated in preclinical studies of novel therapeutics. METHODS Jacketed external telemetry with an implanted miniature blood pressure transmitter (JET-BP) was used to characterize the tolerability, functionality, and sensitivity of this study design in dogs. Thirty-six male or female beagles (n=6 dogs/sex/group) were administered vehicle control (reverse osmosis water) or etilefrine (1, 10mg/kg), sotalol (3, 30mg/kg), and hydralazine (1, 10mg/kg) on separate days. Telemetry data were evaluated for positive control article-related changes and retrospective power analysis was also completed. Animals were evaluated for instrumentation-related changes in clinical and anatomic pathology endpoints. RESULTS All three positive controls elicited the expected pharmacologic responses that were statistically different at high and low doses. Retrospective power analysis confirmed this study design was able to statistically differentiate minor (approximately 5 to 15%) changes in electrocardiography and blood pressure values. This study also demonstrated the potential advantages of combining cardiovascular data across sex when the test article exposure and pharmacodynamics were consistent. Data collection using miniature telemetry blood pressure transmitters did not result in anatomic or clinical pathology findings that would prevent their use in general toxicology studies. DISCUSSION This characterization study indicates that JET-BP in dogs offers a scientifically-robust method to evaluate novel therapeutics for potential cardiovascular liabilities.