Sedentary conditions and enhanced responses to GABA in the RVLM: role of the contralateral RVLM

Abstract

A sedentary lifestyle is a major risk factor for cardiovascular disease, and both conditions are associated with overactivity of the sympathetic nervous system. Ongoing discharge of sympathetic nerves is regulated by the rostral ventrolateral medulla (RVLM), which in turn is modulated by the primary excitatory and inhibitory neurotransmitters glutamate and γ-amino-butyric acid (GABA), respectively. We reported previously that sedentary conditions enhance GABAergic modulation of sympathoexcitation in the RVLM, despite overall increased sympathoexcitation. Thus the purpose of this study was to test the hypothesis that sedentary conditions increase responsiveness to GABA in RVLM. Male Sprague-Dawley rats performed either chronic wheeling running or remained sedentary for 12–15 wk. Animals were instrumented to perform RVLM microinjections under Inactin anesthesia while mean arterial pressure (MAP) and splanchnic sympathetic nerve activity (SSNA) were recorded. Unilateral microinjections of GABA (30 nl, 0.3–600 mM) into the RVLM produced dose-dependent decreases in MAP and SSNA; however, no group differences were observed. Inhibition of the contralateral RVLM (muscimol, 2 mM, 90 nl) caused decreases in MAP and SSNA that were not different between groups but enhanced decreases in SSNA to GABA in sedentary rats only. In sinoaortic denervated rats, GABA microinjections before or after inhibition of the contralateral RVLM caused decreases in MAP and SSNA that were not different between groups. Our results suggest that the contralateral RVLM plays an important role in buffering responses to inhibition of the ipsilateral RVLM under sedentary but not physically active conditions. Based on these studies and others, sedentary conditions appear to enhance both sympathoinhibitory and sympathoexcitatory mechanisms in the RVLM. Enhanced sympathoinhibition may act to reduce already elevated sympathetic nervous system activity following sedentary conditions.

a sedentary lifestyle is a major risk factor for developing cardiovascular disease (4, 46, 60). Cardiovascular disease has been linked to elevated sympathetic nervous system (SNS) activity (29, 55). Sympathetic nerve activity (SNA) contributes significantly to resting blood pressure and is regulated in the brain by the rostral ventrolateral medulla (RVLM) (21). Recently, overactivity of neurons in the RVLM has been proposed to lead to increased SNA and contribute to cardiovascular disease (1, 21, 25, 35, 38, 59, 62). The mechanisms by which the activity of neurons in the RVLM is altered under conditions that contribute to cardiovascular disease, including a sedentary lifestyle, are still not completely understood.

Although there are several neurotransmitters in the RVLM (47), the majority of the activity of the RVLM is regulated by the primary excitatory and inhibitory neurotransmitters glutamate (27, 37) and γ-aminobutyric acid (GABA) (25, 32), respectively. A primary source of GABAergic inhibition of RVLM neurons originates in the caudal ventrolateral medulla (CVLM) (32, 42, 49, 50, 63–66), which receives tonic excitatory stimulation from the nucleus tractus solitarius (NTS) as part of the arterial baroreceptor reflex (49). Because tonic activation of GABAergic neurons in the CVLM is due to both baroreceptor-dependent and baroreceptor-independent inputs, GABA release in the RVLM maintains a reduced level of resting activity of bulbospinal RVLM neurons (36, 50). It has also been suggested that the RVLM receives inhibitory input from the contralateral RVLM (30) along with other brain regions (5, 9).

A reduction in tonic inhibitory GABAergic input to the RVLM from the CVLM has been reported in certain models of hypertension and obesity (25, 56). Similarly, it has been shown that GABA neurotransmission, whether it is due to changes in receptor expression or GABA release, is altered under different physiological states (12, 23, 31, 38, 40). For example, increased salt intake leads to enhanced sensitivity to both glutamate and GABA in the RVLM (1). Collectively, these data suggest that decreased GABAergic signaling in the RVLM likely contributes to physiological and pathophysiological changes in sympathetic outflow and arterial pressure regulation.

We reported previously that sedentary conditions enhance GABAergic modulation of sympathoexcitation in the RVLM, despite overall increased sympathoexcitation (41). We also suggested that sedentary conditions may enhance both excitatory and inhibitory mechanisms in the RVLM similar to other conditions, including a high-salt diet (1). However, the effect of sedentary conditions on GABA sensitivity in the RVLM is unknown. Therefore, the primary purpose of the current study was to determine whether sedentary conditions lead to enhanced GABA sensitivity when compared with active conditions. Since the RVLM is a bilateral structure and microinjections of short-acting agents tend to be done unilaterally, it is possible that responses to GABAergic inhibition of one RVLM may be buffered by the intact contralateral RVLM. In a similar type of study done in the NTS, Catelli et al. (7) demonstrated that blood pressure and heart rate (HR) responses to unilateral microinjection of the GABA_A agonist muscimol were enhanced following contralateral NTS lesions. These data support the concept that contralateral inputs involved in the baroreflex pathway may buffer responses to unilateral microinjection of agonists. Therefore, a secondary purpose of this study was to examine the potential buffering influence of the contralateral RVLM, both in the presence and absence of intact baroreceptor afferent input. We hypothesized that sedentary conditions would lead to enhanced GABA sensitivity in the RVLM compared with active rats and would be most prominent in the absence of buffering by the contralateral RVLM and arterial baroreceptor input.

METHODS

Animals

All experiments were performed in accordance with National Institutes of Health Guide for Care and Use of Laboratory Animals and with the approval of the Wayne State University Institutional Animal Care and Use Committee. Male Sprague-Dawley rats (Harlan Sprague Dawley, Indianapolis, IN; 75–100 g upon arrival, n = 41 total) were used. Animals were housed under standard 12-h dark-light cycle. Physically active rats were housed with a running wheel (Tecniplast No. 2154F0105, 34 cm in diameter, stainless steel) for 12–15 wk. Sedentary rats were housed without a running wheel for an equivalent period. This model of exercise has been used by our laboratory and others in several previous studies (8, 35, 39, 44). Running distances were obtained by bicycle computers (Sigma Sport, Olney, IL) and recorded daily.

General Surgical Preparation

Baroreceptor intact animal studies.

We used methods published previously that allowed instrumentation of animals for acute RVLM microinjections while recording arterial pressure and SNA (35, 38). Briefly, two groups of sedentary (n = 11) or physically active rats (n = 10) were anesthetized initially with isoflurane (5% induction, 2% maintenance; Henry Schein, Melville, NY) and femoral arterial and venous catheters were implanted to record arterial blood pressure and infuse drugs, respectively. A tracheostomy was performed for artificial ventilation, an electrode was implanted on the splanchnic nerve, and the brainstem was exposed for microinjections into the RVLM. Splanchnic sympathetic nerve activity (SSNA) was monitored with an audio monitor and oscilloscope. The animals were placed in a stereotaxic apparatus, and the long-acting anesthetic Inactin (100 mg/kg iv; Sigma Aldrich, St. Louis, MO) was infused. Supplements of Inactin (5–10 mg iv) were given as needed to maintain lack of response to paw pinch and corneal stimulation. Isoflurane was then withdrawn. Body temperature and blood gases were monitored and maintained at normal physiological levels (37°C); Pco₂: 35–40 mmHg; Po₂: >100 mmHg) by a heating pad and by ventilating the animals between 60 and 80 breaths/min, respectively. Animals were neither paralyzed nor vagotomized in these studies.

Brainstem Microinjections

All drugs used for microinjections were dissolved in artificial cerebrospinal fluid with the pH adjusted to 7.4 (35, 38). The animal’s head was placed into a position to allow the caudal tip of area postrema to be oriented 2.4 mm posterior to the interaural line. To localize the RVLM, the following ranges of coordinates relative to calamus scriptorius were used: 0.9–1.1 mm rostral and 1.7–2.2 mm lateral to calamus scriptorius and 3.2–3.6 mm ventral to the dorsal surface of the medulla. Similar to our previous studies and others, the RVLM was identified bilaterally with 10 mM glutamate (30-nl injections) (2, 34, 41). A change in blood pressure of 10–15 mmHg indicated that the pipette was positioned in the region of the RVLM containing bulbospinal neurons involved in control of arterial pressure. In addition to functional identification, injection sites were confirmed postmortem by histological verification (see below). Once the RVLM was identified functionally, triple barrel pipettes containing different concentrations of GABA (0.3, 3, 30, 300, or 600 mM, equivalent to 9, 90, 900, 9,000, or 1,800 pmol) were placed into the left RVLM to perform GABA injections of varying concentrations at a fixed volume (30 nl) in a random order. When given exogenously, GABA produces rapid but brief changes in MAP and SSNA with peak responses occurring within 60 s and only lasting 2 min overall (see raw tracings in Fig. 1). To minimize any potential tachyphylaxis, at least 5 min elapsed between injections ensuring a return to baseline parameters before the next injection.

Fig. 1.

Arterial blood pressure (AP) and splanchnic sympathetic nerve activity (SSNA) responses to GABA microinjections into the rostral ventrolateral medulla (RVLM). A: expanded 2-s recording of baseline AP and SSNA just before injection of GABA in the RVLM. Middle: rectified and integrated SSNA. Bottom: averaged rectified and integrated SSNA. B: AP and SSNA response to microinjection of GABA (30 nl, 300 mM; denoted by arrow) into the RVLM from the same experiment shown in A.

Inhibition of contralateral RVLM.

To test whether buffering of SNA and blood pressure responses to GABA occurred via the right RVLM, the long acting GABA agonist muscimol (90 nl, 2 mM) was injected into the right RVLM. Fifteen minutes later GABA injections were repeated in the remaining intact left side, in the same order as before. After completion of the GABA microinjections, muscimol (90 nl, 2 mM) was microinjected into the left (intact) RVLM to inhibit neuronal activity and confirm that inhibition of the right RVLM was maintained throughout the protocol. Verification was provided by the observation of similar levels of SSNA levels after bilateral inhibition of the RVLM compared with SSNA levels following ganglionic blockade at the end of the experiment (see below). To verify histologically that the microinjections were placed in the RVLM, Chicago Sky Blue (2%, 30 nl) was microinjected bilaterally into both RVLMs after completion of all protocols.

At the end of the experiment, background noise was determined by injection of ganglionic blockers (30 µg/kg hexamethonium iv + 0.1 mg/kg iv atropine). Animals were then euthanized with Fatal Plus (0.2 ml iv; Vortech Pharmaceutical Dearborn, MI) and brains were removed and placed into vials that contained 4% phosphate buffered formalin solution. The heart, left and right adrenal glands and right soleus muscle were removed from all animals, cleaned of connective tissue and weighed.

After postfixation for 5–7 days, brains were transferred to 20% sucrose for 24–48 h. and 30% sucrose for 24–48 h. The brains were frozen and cut into 50-µm coronal sections on a cryostat (Thermo Scientific, Walldorf, Germany). The sections were mounted in an alternating fashion onto two sets of gel-coated slides. One set of slides was left unstained, and the other set was stained with neutral red. Both sets were then coverslipped with Permount (Fischer Scientific, Pittsburg, PA) and allowed to dry. With the use of a brightfield microscope, dye spots were located, and the center of the dye spot was determined. The use of a rat brain atlas helped to verify the anatomical location of the dye (45). The center of the dye injection was represented on diagrams labeled according to a standard rat atlas (45).

Sinoaortic Denervation

In another group of sedentary (n = 11) or physically active rats (n = 9), bilateral sinoaortic denervations (SADs) were performed similar to other laboratories (26, 33). After a ventral incision in the neck was made, the carotid artery bifurcation was identified. Starting at the common carotid artery, surrounding nerves were carefully removed. Moving rostrally along the artery, connective tissue and nerves were removed at the carotid bifurcation and along the internal and external carotid arteries. The superior cervical ganglia were also removed, and the superior laryngeal nerves were transected bilaterally. Phenylephrine (1 µg/kg iv) was given before and after SAD to test baroreflex function and verify the completeness of the SAD procedure, respectively (26, 33). A lack of a reflex decrease in SSNA was used as evidence of an effective SAD (15, 51, 52).

Similar to studies in baroreflex intact animals above, animals were instrumented to record MAP, HR, and SSNA during microinjections of GABA into the left RVLM. After the first set of GABA injections, neuronal activity in the contralateral, right RVLM was inhibited using muscimol (2 mM, 90 nl), and GABA injections were repeated in the remaining intact left RVLM.

Data Acquisition, Analysis, and Statistics

All in vivo data collection was acquired and stored using a computer-based data acquisition system (PowerLab; ADInstruments, Colorado Springs, CO) and Chart software (version 6; ADInstruments). Raw, multiunit SSNA was acquired at 10 kHz filtered using a band pass of 30 Hz to 3 KHz and monitored on both an audio monitor and oscilloscope. The raw SSNA signal was rectified and integrated (time constant of 28 ms) and filtered at 0.1 Hz to produce an average level of SSNA before subtraction of noise. Changes in MAP and HR are expressed in absolute terms (ΔmmHg and Δbeats/min). Changes in SSNA are expressed as a percentage of baseline and as absolute voltage (mV·s) after subtraction of background noise which was measured after administration of ganglionic blockers.

Statistical analysis was performed using Sigma Stat 3.5 (Systat Software, Chicago, IL) for the two-way ANOVA with repeated measures and the Student’s t-tests. Statistical Package for the Social Sciences (SPSS; IBM, New York, NY) was used to perform a three-way ANOVA with repeated measures. Depending on conditions compared (GABA dose; intact vs. contralateral RVLM inhibition; sedentary vs. active), we used a two- or three-way ANOVA with repeated measures. An unpaired Student’s t-test was used to test for statistical significance between sedentary and physically active rats for dye spot location, baseline MAP, absolute SSNA, HR, and organ weights. For the responses to unilateral muscimol injections, an unpaired Student’s t-test was performed to test whether there was a significant difference between groups for the peak change in MAP, HR, percent change in SSNA, and absolute change in SSNA. To determine whether GABA dose responses were significantly different between sedentary and physically active rats, a general linear model using SPSS was used to determine statistical significance (MAP and SSNA). Data are expressed as means ± SE, and P < 0.05 was deemed significant.

RESULTS

Baseline blood pressures, HRs, absolute SSNA, and all organ weights, with the exception of the right adrenals in the intact group, were not significantly different between groups (Tables 1 and 2). As expected, the sedentary rats had body weights that were significantly greater than the physically active rats. In addition, the weights of the right adrenal gland were significantly greater in physically active rats compared with the sedentary rats. The physically active rats ran 326 ± 40 km total over the entire 12- to 15-wk period.

Table 1.

Baseline cardiovascular values for sedentary vs. physically active rats under intact or contralateral RVLM-inhibited conditions (i.e., postmuscimol) and associated body and organ weights for each group of animals

	Resting MAP, mmHg	Resting HR, beats/min	Resting SSNA, mV/s	Body Weight, g	Left Ventricle, mg	Right Ventricle, mg	Right Adrenal, mg	Left Adrenal, mg	Soleus Muscle, mg
Intact conditions
Sedentary (n = 11)	114 ± 5	310 ± 7	1.37 ± 0.33	438 ± 7	782 ± 70	253 ± 17	23 ± 5#	44 ± 21	142 ± 10
Physically active (n = 10)	110 ± 4	295 ± 9	1.67 ± 0.34	396 ± 10	813 ± 26	251 ± 19	30 ± 8	24 ± 5	148 ± 9
Postmuscimol
Sedentary (n = 11)	96 ± 5*	287 ± 9*	1.41 ± 0.33
Physically active (n = 10)	79 ± 4*	267 ± 12*	1.74 ± 0.38

Values are means ± SE. RVLM, rostral ventrolateral medulla; MAP, mean arterial pressure; HR, heart rate; SSNA, splanchnic sympathetic nerve activity.

^*P < 0.05 for main effect of intact vs. postmuscimol for MAP and HR; P = 0.073 for main effect of MAP; P = 0.053 for MAP interaction between group and experimental condition; P = 0.108 for HR main effect of sedentary vs. physically active; #P < 0.05 for significant difference between sedentary and physically active groups.

Table 2.

Baseline cardiovascular values for sinoaortic denervated sedentary vs. physically active rats under intact or contralateral RVLM-inhibited conditions (i.e., postmuscimol) and associated body and organ weights for each group of animals

	Resting MAP, mmHg	Resting HR, beats/min	Resting SSNA, mV/s	Body Weight, g	Left Ventricle, mg	Right Ventricle, mg	Right Adrenal, mg	Left Adrenal, mg	Soleus Muscle, mg
SAD conditions
Sedentary (n = 11)	123 ± 4	316 ± 6	1.86 ± 0.51	439 ± 11	818 ± 30	282 ± 24	34 ± 18	33 ± 3	155 ± 5
Physically active (n = 9)	111 ± 9	288 ± 10	1.23 ± 0.20	404 ± 12#	790 ± 38	325 ± 20	38 ± 2	36 ± 3	149 ± 5
Postmuscimol
Sedentary (n = 11)	76 ± 3*	303 ± 6*	1.27 ± 0.27
Physically active (n = 9)	78 ± 6*	275 ± 9*#	1.16 ± 0.25

Values are means ± SE. SAD, sinoaortic denervated.

^*P < 0.05 for main effect of intact vs. postmuscimol for MAP and HR; P = 0.157 for MAP interaction between group and experimental condition;

^#P < 0.05 for main effect of sedentary vs. physically active for HR; P = 0.093 for main effect of intact vs. postmuscimol for SSNA; P = 0.172 for SSNA interaction between group and experimental condition.

Effect of Sedentary Conditions on GABA Sensitivity in RVLM When Compared With Physically Active Conditions

Before microinjections, baseline MAP, HR, and SSNA were not significantly different between the sedentary and physically active rats (Table 1). Unilateral microinjections of GABA into the left RVLM in baroreceptor intact animals (sedentary n = 11, and physically active n = 10) produced decreases in mean arterial pressure (MAP) and SSNA in both groups of animals (Fig. 2A). Changes in MAP, absolute SSNA, and percent change SSNA decreased in a dose-dependent manner in response to increasing doses of GABA (P < 0.05, main effect of dose). Changes in HR in response to GABA microinjections were small (<11 beats/min) and variable (data not shown). Despite the dose-dependent effects of GABA on MAP and SSNA, there were no significant differences between groups (Fig. 2A).

Fig. 2.

Peak changes in mean arterial pressure (MAP), absolute splanchnic sympathetic nerve activity (SSNA), and percent change in SSNA to inhibition of the rostral ventrolateral medulla (RVLM) with γ-aminobutyric acid (GABA) before (A) and after (B) of unilateral RVLM inhibition in baroreceptor intact rats. A: responses to GABA (30 nl, at 0.3, 3, 30, 300, and 600 mM) before inhibition of the contralateral RVLM. B: responses to GABA (same concentrations) after inhibition of the contralateral RVLM with the long acting GABA_A receptor agonist muscimol (2 mM, 90 nl). Sedentary and physically active rats had similar decreases in MAP and absolute SSNA and percent changes in SSNA before inhibition of the contralateral (right) RVLM. After the contralateral RVLM was inhibited, sedentary and physically active rats had similar decreases in MAP and absolute SSNA in response to acute inhibition of RVLM with GABA (P > 0.05). However, sedentary rats had significantly larger decreases in percent change in SSNA in response to GABA when compared with physically active rats (P < 0.05). *P < 0.05 main effect of dose. #P < 0.05 group effect.

After administration of GABA microinjections in baroreceptor intact animals, the right RVLM was inhibited with muscimol. Muscimol microinjections into the right RVLM produced significant decreases in MAP, SSNA, and absolute SSNA in sedentary and physically active rats (data not shown). The new baselines for MAP, HR, and SSNA, after the muscimol microinjection into the right RVLM were not significantly different between sedentary and physically active rats. However, the new baseline levels for MAP and HR were significantly lower than the pre-muscimol baseline (Table 1).

After prolonged inhibition of the right RVLM, GABA microinjections were repeated into the intact, left RVLM. Similar to baroreceptor intact conditions, decreases in MAP, absolute change in SSNA, and percent change in SSNA were all dose dependent (Fig. 2B; P < 0.05, main effect of dose for all parameters). However, in contrast to baroreceptor intact conditions, percent changes in SSNA to GABA were enhanced in sedentary compared with active rats following inhibition of the right RVLM with muscimol (Fig. 2B; P < 0.05). There were no significant differences between sedentary and physically active rats for the peak decreases to GABA in MAP or absolute SSNA (P > 0.05).

Effect of Sedentary Versus Physically Active Conditions on GABA Sensitivity in Absence of Baroreceptor Reflex Function

We investigated responses to GABA administration in the RVLM in the absence of arterial baroreceptor input by performing SADs in additional groups of sedentary and physically active rats. Before the SAD was performed, baseline values for resting MAP, HR, and SSNA were not different between sedentary and physically active rats. SAD conditions produced increases in MAP, HR and SSNA in sedentary but not physically active rats resulting in significant differences in baseline parameters (Table 2). Before GABA microinjections were administered, we verified the loss of baroreflex responsiveness after SAD by examining responses to phenylephrine (1 µg/kg), similar to other laboratories (26, 33). Before SAD, pressor and baroreflex-mediated sympathoinhibitory response to phenylephrine were comparable between sedentary and physically active rats (Fig. 3, open symbols). After SAD, phenylephrine injections led to an increase blood pressure that was accompanied by no change in SSNA in both sedentary and physically active rats (Fig. 3, closed symbols).

Fig. 3.

Verification of sinoaortic denervation using bolus intravenous injections of phenylephrine (1 µg/kg). Changes in MAP and the percent change in SSNA in sinoaortic denervated (SAD) sedentary (n = 11) and physically active rats (n = 9) in response to (1 µg/kg) phenylephrine challenge before and after SAD. The main effect of SAD on the change in SSNA (∆SSNA) in response to phenylephrine approached but did not reach significance (P = 0.059). Responses to phenylephrine were not different between groups under either condition (P > 0.05).

Similar to animals with intact baroreceptor afferents, microinjections of increasing doses of GABA produced a dose-dependent decrease in MAP and percent change and absolute change in SSNA (P < 0.05; main effect of dose; Fig. 4A). The depressor and sympathoinhibitory responses to increasing concentrations of GABA were not different between sedentary and physically active rats (Fig. 4A).

Fig. 4.

Peak changes in MAP, absolute SSNA, and percent SSNA in response to inhibition of RVLM with microinjections of GABA into the left RVLM in sinoaortic denervated animals with either (A) both RVLMs intact or (B) following inhibition of the right RVLM. A: GABA (30 nl, at 0.3, 3, 30, 300, and 600 mM) responses before inhibition of the contralateral RVLM with long acting GABA_A receptor agonist muscimol (2 mM, 90 nl). SAD sedentary and physically active rats had similar decreases in MAP and absolute SSNA and percent change in SSNA before inhibition of the right RVLM (P > 0.05). B: GABA (30 nl, at 0.3, 3, 30, 300, and 600 mM) responses after inhibition of the contralateral RVLM. Following inhibition of the right RVLM, SAD sedentary and physically active rats had similar decreases in MAP, absolute SSNA and percent change in SSNA in response to acute inhibition of RVLM with GABA (P > 0.05). *P < 0.05 main effect of dose.

Subsequent microinjections of muscimol in the right RVLM resulted in decreases in MAP, HR, and SSNA that were not different between sedentary and physically active rats (data not shown). The decreases in MAP, percent SSNA, and absolute SSNA were not different between the sedentary and the physically active rats (Fig. 4). The new baselines for MAP and HR following the microinjection of muscimol into the right RVLM were significantly different from the baseline before the muscimol microinjection (Table 2). However, the SSNA new baseline was not significantly different from the premuscimol baseline (Table 2). Furthermore, there were no significant differences between sedentary and physically active rats for MAP and SSNA but HR was significantly different between the groups (Table 2). Following inhibition of the right RVLM, GABA microinjection into the remaining left RVLM produced little change in MAP, but it did produce a dose-dependent decrease for absolute SSNA and percent change in SSNA (Fig. 4B). There was no difference between sedentary and physically active rats to GABA following inhibition of the right RVLM for MAP, absolute change in SSNA, and percent change for SSNA.

Histology

RVLM microinjection sites were verified histologically by Chicago Sky Blue dye injections (2%, 30 nl) similar to our previous studies (1, 35, 41). Figure 5A represents the locations of microinjection sites in the left and right RVLM (n = 21 rats). On average, microinjection sites were located just rostral to the caudal pole of facial nucleus, +30 ± 22 µm on the right side and +23 ± 27 µm on the left side. The right and left microinjection sites were not statistically different (P = 0.582). Furthermore, the microinjection sites were all located in the region lateral to the pyramidal tract, medial to the spino-trigeminal tract, and ventral to the nucleus ambiguus. Microinjection sites were not different between sedentary (right 39 ± 31 µm and left 45 ± 38 µm) and physically active rats (right 22 ± 21 µm (P = 0.66) and left 0 ± 35 µm (P = 0.251)). Figure 5B contains photomicrographs of both unstained and neutral red-stained hemisections of the brainstem of one rat used in the current study. Dye injections were localized in the unstained sections (Fig. 5B, left) and then were aligned with adjacent neutral red-stained sections (Fig. 5B, right) to determine their location relative to previously published anatomical reference points (see figure legend for details).

Fig. 5.

Histological verification of RVLM microinjection sites. A: diagrams of bilateral microinjection sites from both baroreceptor intact and SAD sedentary and physically active rats labeled according to Paxinos and Watson (45). There was no difference between sedentary and physically active rats or between left and right RVLM injection sites. Open circles represent physically active rats and closed circles represent sedentary rats. Numbers in negative millimeters (mm) represent section distances caudal to bregma. B: photomicrographs of hemisections from the −12.12-mm level of the brainstem shown in A from 1 experimental animal. The center of the dye injection (arrow) in the unstained section (left hemisection, cropped and outlined for clarity) is located just ventral to the compact portion of the nucleus ambiguus (AmbC). RVLM dye location relative to other brain structures was facilitated by adjacent sections stained with neutral red (at right, cropped for clarity). Scale bar = 1,000 µm. Structural abbreviations (shown unilaterally only) include the following: 7, facial nucleus; py, pyramidal tract; sp5 spinal trigeminal tract. The RVLM has been defined as the region bounded ventrally by AmbC; laterally by py and medially by sp5.

DISCUSSION

The purposes of this study were 1) to determine whether sedentary conditions enhance GABA sensitivity in the RVLM compared with physically active conditions; and 2) to examine potential differences in the buffering influences of the contralateral RVLM and arterial baroreceptors in sedentary vs. physically active animals. The most important findings in the present study are 1) direct inhibition of the RVLM using increasing doses of GABA produced dose-dependent decreases in MAP and SSNA in both sedentary and physically active groups, but responses between groups were not different in the intact state; 2) inhibition of the right RVLM with muscimol revealed conditions in which subsequent inhibition of the remaining intact, left RVLM with increasing doses of GABA produced enhanced sympathoinhibitory responses in sedentary but not physically active rats; and 3) denervation of arterial baroreceptor afferents did not reveal differences between groups in the depressor and sympathoinhibitory responses to increasing concentrations of GABA in the left RVLM whether the contralateral (right) RVLM was inhibited or not. From these data we conclude that following sedentary, but not physically active, conditions, the contralateral RVLM compensates for decreases in sympathetic outflow produced by acute inhibition of the other RVLM. In addition, once the buffering capacity of the contralateral RVLM is inhibited, the remaining neurons in the intact RVLM that tonically regulate sympathetic outflow are more sensitive to GABAergic inhibition in sedentary but not physically active rats. We speculate that enhanced buffering in sedentary animals may serve as a compensatory mechanism to counteract enhanced sympathoexcitation we and others have observed in previous studies.

Acute and reversible inhibition of one RVLM with increasing doses of GABA produced dose-dependent decreases in MAP and SSNA in both sedentary and physically active groups; however, responses between groups were not different in the intact state. In isolation, the simplest explanation of these findings is that GABAergic neurotransmission in the RVLM is unaltered by sedentary vs. physically active conditions. From our standpoint, however, this possibility proved unlikely since several previous studies from our laboratory and others suggested that GABAergic inhibition of the RVLM was significantly altered in sedentary vs. physically active animals. For example, DiCarlo and colleagues (10, 13, 43) demonstrated a number of years ago that sedentary animals exhibited enhanced baroreflex mediated sympathoexcitation when compared with animals that had been physically active. Similarly, and more recently, Mischel and Mueller (35) demonstrated that sedentary vs. chronic, spontaneous wheel running rats also exhibited enhanced baroreceptor reflex-mediated sympathoexcitation. Given that removal of tonic GABAergic inhibition of bulbospinal RVLM neurons is thought to mediate baroreflex mediated sympathoexcitation (11, 21, 48), it was logical to hypothesize that sedentary conditions affected GABAergic neurotransmission in the RVLM. Further substantiating this hypothesis, our laboratory reported that microinjection of bicuculline, a GABA_A receptor antagonist into the RVLM, enhanced blood pressure and SNA responses to glutamate-mediated excitation of the RVLM in sedentary but not physically active rats (41). Therefore, we were somewhat surprised initially that cardiovascular responses to unilateral microinjections of GABA did not differ between sedentary and physically active animals in the present study.

Prolonged inhibition of the right RVLM with muscimol revealed conditions in which subsequent acute and reversible inhibition of the remaining left RVLM with increasing doses of GABA produced enhanced sympathoinhibitory responses in sedentary but not physically active rats. From these data, we conclude that influences from the contralateral RVLM prevented us from seeing differences between sedentary and physically active rats in the intact state. Enhanced responses in the absence of the compensatory effect of the contralateral RVLM only in sedentary animals also suggest that sedentary vs. physically activity conditions produce a fundamental alteration in the regulation of sympathetic outflow by GABA at the level of the RVLM. Similar findings of enhanced responses to GABA microinjections in the RVLM have been reported in rats fed a high- versus a low-salt diet (1). Interestingly, differential responsiveness to GABA was observed in the intact state of high-salt-fed animals, i.e., without inhibition of the contralateral RVLM (1). The significance of this potential difference in buffering capacity may be related to disparate mechanisms between a high-salt diet and sedentary conditions. It is possible that the change in baseline parameters following inhibition of the contralateral RVLM could have influenced our results. To address this possibility, we calculated both percent change and absolute change in MAP and SSNA to determine how differences in baseline might affect the results. For MAP, the absolute and percent change in MAP did not reveal any differences between sedentary and physically active groups in the baroreceptor intact protocol. In addition, although there was a significant effect of muscimol treatment to reduce baseline parameters, the absolute change in SSNA was not significantly different between groups.

Given evidence from previous studies, the enhancement in responses to unilateral microinjection responses following inhibition of the contralateral RVLM in sedentary animals is not all that surprising. For example, almost 30 yr ago, Catelli et al. (7) demonstrated that increases in MAP to unilateral microinjections of muscimol into the NTS were enhanced after lesioning of the contralateral NTS. Similarly, other investigators have performed highly complex protocols to account for bilateral structures in the brainstem involved in regulation of sympathetic outflow and blood pressure (22). Collectively, these studies established the importance of contralateral brain regions in compensating against changes produced by manipulation of neuronal activity in only one region of a bilateral structure. In most cases, buffering by the contralateral side was believed to be solely via the arterial baroreceptor reflex, which could potentially offset responses involving either increases or decreases in arterial blood pressure. In the current study, inhibition of the contralateral RVLM alone but not denervation of arterial baroreceptors alone, resulted in enhanced sympathoinhibitory responses in sedentary vs. physically active animals. Thus we propose the possibility that commissural projections between RVLMs may contribute to important compensatory mechanisms previously attributed solely to buffering from arterial baroreceptors.

Multiple inputs have been proposed to play a role in controlling the tonic activity of bulbospinal neurons in the RVLM (12, 58), including a commissural pathway from the contralateral RVLM (18–20, 30, 61). As first described anatomically, and then subsequently in a slice preparation (19, 20), the connections between the left and right RVLM have gained more recent attention by in vivo studies reported by McMullan and colleagues (30, 61). Defined initially as a potential source of tonic inhibitory input (30), a subsequent study has now suggested an sympathoexcitatory role specifically driving splanchnic nerve activity (61), i.e., the specific nerve activity recorded in the current study. Indeed, as discussed in more detail below, these contralateral projections may provide a mechanism by which inhibition of one RVLM augments responses in sedentary animals.

Denervation of arterial baroreceptor afferents did not affect depressor and sympathoinhibitory responses to increasing concentration of GABA whether the contralateral RVLM was intact or inhibited. Before SAD in each animal, the baroreceptor reflex was tested using the vasoconstrictor phenylephrine to produce increases in arterial pressure and reflex decreases in SNA. Similar to previous studies measuring renal SNA responses (13, 43); there was no significant difference in baroreflex-mediated decreases in SSNA between sedentary and physically active rats. There were also no significant differences between groups in their response to phenylephrine for absolute MAP. We also calculated the percent change in MAP before SAD in response to PE and found that there was no significant difference between the groups or within the groups. Following SAD, appropriate denervation was verified by the lack of SSNA responses to a second injection of phenylephrine similar to previous studies (15, 51, 52). Similar to the results before SAD, there were no significant differences between the sedentary and physically active rats in their responses to phenylephrine. However, there was a significant increase in the baseline MAP after the SAD was performed in sedentary but not physically active rats. Based on these findings, it was possible that differences in baseline may have affected the MAP and SSNA response to phenylephrine. To account for this, the percent change in SSNA and MAP was calculated. We did not observe any differences in the MAP and SSNA responses to phenylephrine between the groups. In experiments performed following SADs, we observed no difference between sedentary and physically active rats in response to GABA microinjections. These data suggest that differences in the arterial baroreflex buffering capacity of the contralateral RVLM were not responsible for differences in GABA responses revealed when the contralateral RVLM was inhibited in intact (i.e., non-SAD) animals. To test whether removal of the contralateral RVLM would still reveal differences following SAD, we subsequently inhibited the contralateral RVLM with muscimol before repeating GABA microinjections. Similar to previous studies (14, 24), prolonged inhibition of the contralateral RVLM with muscimol in both groups of SAD animals produced substantially larger decreases in blood pressure and SNA compared with the intact state. Although responses to inhibition of the contralateral RVLM were not different between groups, it provided further evidence of the importance of the arterial baroreceptors in maintaining pressure via the RVLM (14, 24).

Unlike intact baroreceptor reflex conditions, differences in responses to subsequent GABA microinjections were not revealed when the contralateral RVLM was inhibited under SAD conditions. There are several potential explanations for this finding. First, it is possible that the arterial baroreceptor reflex plays a permissive role in the expression of differences between sedentary and physically active animals. As mentioned above, McMullan’s laboratory has recently begun to characterize the commissural projections between RVLMs in vivo and have provided evidence for both excitatory and inhibitory projections (61). Based on these initial studies, we can only speculate as to the potential interaction between the contralateral projections and arterial baroreceptors in contributing to difference observed between groups. Although it appears that contralaterally projecting RVLM neurons are not barosensitive (61), it does not preclude the possibility that contralateral projections provide a level of tonic input that when removed, influence responsiveness to GABA. Under the intact baroreceptor reflex conditions the SSNA response to GABA is enhanced, this finding further supports the possibility that the baroreceptors may be playing a role. Second, another explanation is that the commissural connections are modulated by multiple inputs that are necessary for full expression of the buffering capacity of the contralateral RVLM. Third, the combination of contralateral RVLM inhibition in the presence of SAD could produce a level of activity in the remaining intact RVLM such that addition of exogenous GABA via microinjections is ineffective at producing a significant amount of further inhibition to detect differences between groups. Certainly, additional experiments are needed to understand the role of the commissural connection in both sedentary and physically active animals.

It is important to recognize that the mechanisms by which sedentary conditions enhance responsiveness to GABA are unknown. One of the most straightforward explanations could be changes in GABA receptor expression or function. Based on findings from our laboratory and others, GABA receptor subunit expression appears to be sensitive to changes in the level of physical activity/inactivity. For example, in a study done by done by Hill et al. (23), sedentary compared with physically activity conditions altered mRNA expression of different GABA_A receptor subunits in the forebrain. In addition, a study performed by Subramanian et al. (57), with the use of laser capture microdissection and PCR, demonstrated a trend for increased gene expression of the GABA_Aα2 receptor subunit in bulbospinal RVLM neurons from physically active versus sedentary rats (57). Whether increases in mRNA expression translate proportionally into increases in protein expression remains to be investigated. In addition, increased receptor subunit expression may not be necessary for functionally related changes to occur. Foley et al. (16), for example, provided evidence that RVLM neurons express greater levels of protein and mRNA expression for GABA_Aα1 receptor subunits when compared with the GABA_Aα2 receptor subunits. Although differences in protein or mRNA expression of these same receptor subunits were not observed in pregnant vs. nonpregnant rats, functional differences in sedentary vs. physically active rats could be a result of changes in GABA_A receptor subunit composition or downstream signal transduction mechanisms (17). Clearly, more studies are necessary to determine how changes in GABA receptors contribute to functional differences in GABA responsiveness produced by sedentary conditions.

Perspectives and Significance

A sedentary lifestyle increases the risk of developing cardiovascular disease (53). Similar to a number of other studies, our data suggest a central nervous system component to the development of cardiovascular disease (25, 28, 35, 54), via structural and functional neuroplasticity occurring in regions that directly or indirectly regulate sympathetic outflow (3, 6, 25, 28, 35). Data from the present study as well as our previous study demonstrate that sedentary conditions enhance inhibition of RVLM neurons compared with physically active conditions. Since the majority of inhibition that restrains the activity of RVLM is due to GABA, we hypothesize that GABAergic inhibition and modulation of excitation is enhanced under sedentary conditions. To determine the exact cause of modulation of GABAergic mechanism under sedentary conditions, future studies would need to look at GABA_A receptor expression on RVLM neurons; specifically, whether the number or the composition of the receptors is altered in response to sedentary conditions. The enhanced inhibition may be a compensatory mechanism to offset enhanced sympathoexcitation that occurs via glutamatergic mechanisms as demonstrated in our previous studies (35, 41). By understanding the mechanism behind the development of enhanced sympathoinhibition, we may be able to develop therapeutic intervention that can further increase GABAergic inhibition and help to treat hypertensive patients. We speculate that the net increase in excitation in the RVLM leads to higher sympathetic outflow and blood pressure and ultimately increases the risk for developing cardiovascular disease.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants R01HL096787 to P. J. Mueller; R01HL096787-S1 to M. D. Dombrowski. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.J.M. conceived and designed research; M.D.D. performed experiments; M.D.D. and P.J.M. analyzed data; M.D.D. and P.J.M. interpreted results of experiments; M.D.D. and P.J.M. prepared figures; M.D.D. drafted manuscript; M.D.D. and P.J.M. edited and revised manuscript; M.D.D. and P.J.M. approved final version of manuscript.