In the present study we investigated the involvement of the hypothalamic paraventricular nucleus (PVN) in the modulation of sympathoexcitatory reflex activated by peripheral and central chemoreceptors. We measured mean arterial blood pressure (MAP), heart rate (HR), renal sympathetic nerve activity (RSNA), and phrenic nerve activity (PNA) before and after blocking neurotransmission within the PVN by bilateral microinjection of 2% lidocaine (100 nl) during specific stimulation of peripheral chemoreceptors by potassium cyanide (KCN, 75 μg/kg iv, bolus dose) or stimulation of central chemoreceptors with hypercapnia (10% CO2). Typically stimulation of peripheral chemoreceptors evoked a reflex response characterized by an increase in MAP, RSNA, and PNA and a decrease in HR. Bilateral microinjection of 2% lidocaine into the PVN had no effect on basal sympathetic and cardiorespiratory variables; however, the RSNA and PNA responses evoked by peripheral chemoreceptor stimulation were attenuated (P < 0.05). Bilateral microinjection of bicuculline (50 pmol/50 nl, n = 5) into the PVN augmented the RSNA and PNA response to peripheral chemoreceptor stimulation (P < 0.05). Conversely, the GABA agonist muscimol (0.2 nmol/50 nl, n = 5) injected into the PVN attenuated these reflex responses (P < 0.05). Blocking neurotransmission within the PVN had no effect on the hypercapnia-induced central chemoreflex responses in carotid body denervated animals. These results suggest a selective role of the PVN in processing the sympathoexcitatory and ventilatory component of the peripheral, but not central, chemoreflex.
- renal sympathetic nerve activity
- phrenic nerve activity
- carotid body
the peripheral and central chemoreflexes play an important role in regulation of ventilatory and circulatory responses to changes in arterial Po2 and Pco2. They are both linked to increases in sympathetic activity, blood pressure, and minute ventilation and are mediated via peripheral chemoreceptors, which are predominately stimulated by hypoxia, and central chemoreceptors, which are predominately stimulated by hypercapnia. There is considerable evidence suggesting enhanced chemoreceptor reflexes in disease states such as congestive heart failure (18, 26) and hypertension (6).
Given the significance of the chemoreceptors in cardiorespiratory reflex regulation in normal and disease conditions, it is important to understand the neural anatomical pathways and neurotransmitters involved in generation, processing, and modulation of complex cardiovascular, respiratory, and behavioral responses to chemoreceptor activation. Several studies have identified central neural structures involved in the sympathoexcitatory component of chemoreflexes; however, most have focused on the pons and the medulla (7), which are the primary projection areas of chemosensitive afferents and the primary sites for respiratory and sympathetic integration. Apart from these areas of the brain, other studies suggest an involvement of the hypothalamus in modulating the respiratory and cardiovascular responses to hypoxia and hypercapnia (9, 14, 19). Furthermore, it was also reported that hypoxia and hypercapnia stimulate neurons in various hypothalamic nuclei, including the paraventricular nucleus (PVN) (2).
Because the PVN is known to influence sympathetic outflow (4, 11, 12, 28) and also because of the reported anatomical projections between the PVN and other areas of brain involved in cardiorespiratory regulation, such as the nucleus tractus solitarius (NTS) (24), as well as projections from the PVN to the rostral ventrolateral medulla (RVLM) (23) and to the spinal cord (10), it is reasonable to speculate that the PVN plays an important role in the modulation of sympathoexcitation induced by the activation of the chemoreflexes in response to changes in arterial blood gases. To further support this notion, Kubo et al. (14) reported that the excitatory amino acid antagonist kynurenate injected into the PVN inhibited the chemoreceptor reflex-induced pressor response in anesthetized rats. Later, Olivan et al. (19) demonstrate bilateral electrolytic lesion of the PVN in awake rats produced a significant reduction in the magnitude and duration of the pressor response induced by chemoreflex activation, suggesting that the PVN is involved in the regulation of cardiovascular responses to peripheral chemoreflex activation. However, the functional involvement of PVN neurons and their neurotransmitters in regulating symapthoexcitatory responses to chemical drive is not well understood. Therefore, the present study was designed to evaluate the contribution of the PVN in modulating the sympathoexcitatory component of the peripheral and/or central chemoreflex before and after blocking neurotransmission within the PVN by bilateral microinjection of 2% lidocaine. To further confirm the involvement of the PVN neurons in mediating the chemoreflex effects, we also tested the hypothesis that activation/inhibition of the PVN neurons by microinjection of bicuculline methiodide, a γ-aminobutyric acid (GABA)A receptor antagonist, and muscimol, a GABAA receptor agonist, respectively, contributes to the chemoreceptor-mediated reflex response.
MATERIALS AND METHODS
Experiments were conducted on male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) weighing between 250 and 300 g. These studies were performed in accordance with the American Physiological Society’s “Guiding Principles for Research Involving Animals and Human Beings.” The experimental procedures were approved by the University of Nebraska Institutional Animal Care and Use Committee.
o2 and arterial Pco2] and pH (ABL-500; Radiometer, Copenhagen, Denmark). Arterial blood gases were maintained at normal levels by adjusting the ventilation rate (60–70 breaths/min) and/or tidal volume (2.0–3.0 ml), and metabolic acidosis was corrected by intravenous injection of an appropriate amount of warm sodium bicarbonate [0.2 × body wt × base excess (meq)]. The body temperature of the animal was maintained at 37°C with a water circulating pad.
Microinjections into the PVN
The rat was placed in a stereotaxic apparatus (Davis Kopf Instruments), and small burr holes were placed on either side of midline of the skull. The coordinates for each side (right and left) of the PVN were determined from the Paxinos and Watson Rat Atlas (21), which were 1.8 mm posterior, 0.4 mm lateral to the bregma, and 7.8 mm ventral to the dura. A cannula (outer diameter, 0.5 mm; inner diameter, 0.1 mm) connected to a microsyringe (0.5 μl, model 7000.5, Hamilton microsyringe) was advanced into the PVN with a manipulator (Narishige Z-1). The cannula was flushed with the experimental injectate before each insertion. After microinjection on one side of the PVN, the cannula was withdrawn and then reinserted using the same coordinates on the opposite side to perform a bilateral microinjection of equal volume. The cannula was withdrawn immediately after each microinjection.
At the end of the experiments, monastral blue dye (100 nl) was injected through the cannula for histological verification of the site of injection. After the rat was killed, the brain was removed and fixed in 10% formaldehyde for at least 24 h. The brain was then frozen, and serial transverse sections (30 μm) were cut using a cryostat (IEC, model CT, International-Harris Cryostat) at −20°C. The sections were thaw mounted on microscope slides and stained with 1% aqueous neutral red staining procedures. Presence of the blue dye within the PVN was verified microscopically.
Recording of Efferent Renal Sympathetic Nerve Activity
Recording of Efferent Phrenic Nerve Activity
These experiments were performed to examine the involvement of the PVN in processing the sympathoexcitatory component of the peripheral chemoreflex response. Initially artificial cerebrospinal fluid (aCSF) was injected bilaterally into the PVN (100 nl on each side) over a 2-min period as described above. Mean arterial blood pressure (MAP), HR, RSNA, and PNA were monitored over the subsequent 20–30 min. After recording the baseline, we stimulated peripheral chemoreceptors by intravenous bolus injection of potassium cyanide (KCN, 75 μg/kg), and responses were recorded over 5–10 min. After all parameters returned to near baseline, 2% lidocaine was microinjected bilaterally into the PVN (100 nl each side) over 2 min, and the above measured variables were monitored over the subsequent 20–30 min. We repeated the same procedure of KCN injection to stimulate the peripheral chemoreceptors after blocking the neurotransmission within the PVN by bilateral microinjection of lidocaine.
In this set of experiments we verified that the cardiorespiratory reflex responses induced by KCN injection were evoked by stimulation of carotid body chemoreceptors, which are the primary source of peripheral chemoreceptor afferents in the rat (13), by carotid sinus denervation. The carotid bifurcation was bilaterally exposed, and sinus nerves were identified and cut. After denervation, animals were allowed to recover, and once all the cardiorespiratory parameters stabilized, the same procedure of KCN injection was repeated as above.
These experiments were performed to assess whether GABA receptors within the PVN influence the reflex sympathoexcitatory component elicited by peripheral chemoreceptor stimulation. Either bicuculline (50 pmol/50 nl), a GABAA receptor antagonist, or muscimol (0.2 nmol/50 nl), GABAA receptor agonist, was injected bilaterally (50 nl each side) into the PVN. Each microinjection was made over a period of 1 min. As a control, aCSF alone was microinjected into the PVN (50 nl each side), followed by stimulation of arterial chemoreceptors with bolus intravenous injection of KCN (75 μg/kg), and reflex responses were recorded for 5 to 10 min. Animals were allowed to recover for 1 h, and the effect of either bicuculline or muscimol into the PVN on the sympathetic and cardiorespiratory parameters was monitored for 5–10 min. Once all parameters stabilized, we evoked the peripheral chemoreflex, and the reflex responses were recorded for 5–10 min.
The response of MAP, HR, RSNA, and PNA were collected in groups of rats in which the bilateral injection sites were not within the PVN area and were considered as anatomical control groups (lidocaine, bicuculline, and muscimol microinjections). These experiments were carried out with the same procedures as described above, except the sites of microinjection were placed outside, but adjacent to the PVN bilaterally.
These experiments were performed to examine the role of the PVN in modulation of the central chemoreflex. The carotid bodies were denervated to eliminate peripheral chemoreceptor input. Central chemoreceptors were stimulated by breathing hypercapnic air (10% CO2 in 50% O2 and with balance of N2), and responses were recorded over 5–10 min. After all parameters returned to near baseline, 2% lidocaine (100 nl) was microinjected bilaterally into the PVN over 2 min, and the above measured variables were monitored over the subsequent 20–30 min. We repeated the same procedure of hypercapnia to stimulate the central chemoreceptors after blocking the neurotransmission within the PVN by bilateral microinjection of lidocaine (100 nl).
All data were analyzed off-line. The baseline values were averaged over 2 min before the experimental stimulus. Changes in MAP and HR during stimulation of chemoreflexes were calculated as the difference between the peak deviation of MAP and HR (5-s average) from the baseline. Levels of RSNA were normalized as a percentage of the maximum level of activity recorded during SNP (100 μg/kg) administration, and peak changes in RSNA during chemoreflex stimulation were calculated as the difference between the peak deviation of RSNA (5-s average) discharge from baseline measured before each stimulus. Similarly, PNA (burst amplitude) during chemoreflex activation was expressed as percent change from the immediately preceding baseline. The data were analyzed by (GB-Stat 6.0 software) by the Student’s t-test or one-way ANOVA, followed by Newman-Keuls post hoc test for comparisons between groups. All results were presented as means ± SE. Differences were considered significant when P < 0.05.
The average resting MAP, HR, and RSNA of adult male rats used in this study in the main experimental groups are presented in Table 1. The locations of microinjection sites in the PVN region for all groups are shown in Fig. 1. Data from experiments with sites targeted outside of the PVN region were analyzed separately as anatomical control groups.
Effect of Microinjection of 2% Lidocaine into PVN on Reflex Responses to KCN
Upon stimulation of arterial chemoreceptors (75 μg/kg iv KCN), the onset of the peripheral chemoreflex response was characterized by an increase in MAP and slight fall in HR. It was accompanied by an increase in the activity of both RSNA and PNA, typically beginning within 5–10 s following the bolus injection of KCN (Fig. 2). The peak response of MAP, HR, and RSNA during the stimulation arterial chemoreceptors lasted for 15–30 s and slowly decreased back to preinjection levels within 1–5 min following KCN injections. However, the peak PNA response lasted slightly longer and returned to baseline within 5 min following onset of reflex response. At the peak response, there was an increase in RSNA (27 ± 3% Δ) and PNA (107 ± 3% Δ), respectively. Similarly there was an increase in MAP (18 ± 2 mmHg) and decrease in HR (10 ± 2 beats/min), respectively, from the baseline (Fig. 3).
Blood gases and pH were not influenced during activation of peripheral chemoreceptors by KCN injection since respiration was mechanically controlled (Table 2). Baseline blood gas and pH values did not deviate among the various experimental groups nor were they affected by KCN injection in any group of the study (data not shown).
Figure 2 illustrates a typical tracing of the changes in RSNA, PNA and ABP in response to peripheral chemoreflex activation before and after bilateral microinjection of lidocaine into the PVN. Lidocaine (100 nl each side, n = 8) was microinjected into the PVN, which produced no significant effect on baseline arterial pressure, HR, renal sympathetic activity, and phrenic nerve response (Table 1). Similarly we observed no difference in above parameters from the baseline when we microinjected similar volume of aCSF into the PVN.
After neurotransmission in the PVN (n = 8) was blocked with microinjection of 2% of lidocaine, the arterial chemoreceptor-mediated increases in RSNA (17 ± 3% Δ), PNA (61 ± 6% Δ), and MAP (7.7 ± 2.5 mmHg) were significantly attenuated, but no significant changes occurred in the HR response (Fig. 3).
Microinjections Targeted Outside the PVN
Figure 3 also summarizes the results where microinjections were outside but adjacent to the PVN (n = 5). There was no significant difference in peripheral chemoreflex-mediated responses to KCN injection (75 μg/kg iv) before and after bilateral PVN microinjection of lidocaine outside the PVN, suggesting that the attenuated chemoreflex response after bilateral microinjection of lidocaine in a previous set of experiments was indeed due to the blocking of neurotransmission within the PVN regions.
Reflex Response to KCN in Sinus Nerve-Denervated Animals
To confirm that the cardiorespiratory responses described above are indeed associated with stimulation of arterial chemoreceptors, in five animals, reflex responses to KCN (75 μg/kg iv) were recorded following bilateral sectioning of the carotid sinus nerves. ABP, HR, and RSNA transiently increased following carotid sinus denervation but returned to baseline levels within 30 min (Table 1). As shown in Fig. 4, carotid sinus denervation eliminated both cardiovascular and respiratory responses to bolus injection of KCN.
Effect of Microinjection of Bicuculline into the PVN on Reflex Responses to KCN
Figure 5 illustrates typical tracings of the changes in RSNA, PNA, and ABP in response to peripheral chemoreflex activation before and after bilateral microinjection of bicuculline (50 pmol in 50 nl) into the PVN. In rats microinjected with vehicle into the PVN, KCN injection (75 μg/kg iv) elicited an increase in RSNA (24 ± 1.5% Δ) and PNA (96 ± 7.5% Δ), accompanied by increase in MAP (16 ± 2 mmHg) and decrease in HR (12 beats/min) (Fig. 6). Bilateral microinjection of bicuculline into the PVN (n = 5) produced an increase in peripheral chemoreflex-evoked changes in RSNA (37 ± 1.4% Δ), PNA (143 ± 6% Δ), and MAP (26 ± 1 mmHg), respectively, compared with control reflex responses above (Fig. 6). However, there was no significant effect of bicuculline on the HR response to chemoreflex activation. Injection of bicuculline outside of the PVN did not affect reflex responses to peripheral chemoreflex activation (Fig. 6, n = 3).
Effect of Microinjection of Muscimol into the PVN on Reflex Responses to KCN
Figure 7 is a typical tracing illustrating the changes in RSNA, PNA, and ABP in response to peripheral chemoreflex activation before and after bilateral microinjection of muscimol into the PVN. Bilateral microinjection of muscimol into the PVN (0.2 nmol in 50 nl, n = 5) attenuated the reflex responses evoked by KCN injection (75 μg/kg iv), producing a significant smaller change in RSNA (11.5 ± 1% Δ), PNA (59 ± 9% Δ), and MAP (9 ± 1 mmHg) compared with control reflex responses (23.8 ± 1.5% Δ, 96 ± 7.5% Δ, 16 ± 2 mmHg), respectively, but had no effect on the HR response (Fig. 8). Injection of muscimol outside of the PVN did not affect reflex responses to peripheral chemoreflex activation (Fig. 8, n = 3).
Effect of Bilateral Microinjection of 2% Lidocaine into the PVN on Reflex Responses to Hypercapnia (10% CO2)
Blood gases and pH under control conditions and during activation of central chemoreceptors by hypercapnia are given in Table 2. Breathing 10% CO2 increased arterial Pco2 by similar levels before and after bilateral microinjection of 2% lidocaine into the PVN. To eliminate peripheral chemoreceptor input during hypercapnia, animals were subjected to carotid sinus denervation before the experimental procedures. Figure 9 summarizes the changes in sympathetic and cardiorespiratory responses to hypercapnia before and after lidocaine into the PVN (n = 7). Exposure to hypercapnia resulted in increases in MAP (16.7 ± 2.4 mmHg), HR (10 ± 4.5 beats/min), RSNA (28 ± 4% Δ), and PNA (189 ± 17% Δ) (Fig. 9). Interestingly, unlike the peripheral chemoreflex, there was no difference in reflex responses evoked by hypercapnia before and after lidocaine injection into the PVN, suggesting that the PVN was not involved in the modulation of reflex response elicited by the central chemoreflex. These data suggest a differential role of PVN in modulating the peripheral vs. central chemoreflex.
The major findings of the present study are as follows. First, blockade of neurotransmission in the region of the PVN with bilateral microinjection of lidocaine attenuated the sympathetic, ventilatory, and pressor responses elicited by the selective stimulation of peripheral (carotid body) chemoreceptors. Second, GABAA receptor inhibition by microinjection of bicuculline in the PVN augmented, whereas GABAA receptor activation by microinjection of muscimol attenuated, these reflex responses elicited by the activation of peripheral chemoreceptors. Third, neither blockade of neurotransmission in the PVN nor activation of GABAA receptors in the PVN altered HR responses to stimulation of peripheral chemoreceptors. 4) In contrast to the peripheral chemoreflex, blockade of neurotransmission in the PVN did not affect any of the reflex responses elicited by selective stimulation of central chemoreceptors. From these results, we conclude that the PVN is involved in the modulation of sympathoexcitatory and ventilatory components of the peripheral chemoreflex but not the central chemoreflex. A GABAergic mechanism within the PVN appears to be involved in the modulation of the peripheral chemoreflex responses.
In the present study, KCN was used to briefly stimulate arterial chemoreceptors. Intravenous administration of KCN (75 μg/kg) elicited typically increased MAP, RSNA, and PNA and smaller decreases in HR. This response was similar to that reported by other investigators using KCN to elicit chemoreflex responses in urethane-anesthetized rats (8). Moreover, the specificity of the cardiovascular and respiratory components of the responses to KCN was confirmed by complete elimination of the responses following bilateral denervation of the carotid sinus nerve and supports the suggestion that the chemoreflex responses in the rat are primarily mediated through carotid chemoreceptor afferents (13).
We observed that bilateral microinjection of lidocaine, a local anesthetic, into the PVN produced significant attenuation of the arterial pressure, RSNA, and phrenic nerve response evoked by the stimulation of arterial chemoreceptors, indicating the significance of this region in the coordination of the reflex responses induced by the activation of peripheral chemoreflex. However, it could be argued that this attenuation of reflex responses may be due to the diffusion of lidocaine into the surrounding hypothalamic regions. Based upon previous studies (20), where lidocaine was used in the PVN to block synaptic transmission, the volume of lidocaine (100 nl) microinjected in the present study was not likely to diffuse into surrounding regions outside of the PVN. However, we did test this possibility by analyzing data from animals where the microinjections of lidocaine were targeted outside but adjacent to the PVN. In this group we did not find any significant changes in reflex responses before and after lidocaine injection, supporting the view that the area of the PVN is the major site of action.
Our data are in consensus with previous studies indicating that peripheral chemoreflex pathways are not restricted to the medulla and pons but also involve mesencephalic and hypothalamic regions (2). Furthermore, the involvement of PVN neurons in the generation of the sympathoexcitatory and ventilatory responses to peripheral chemoreflex activation is consistent with previous studies showing increased arterial pressure and hyperventilation upon electrical stimulation of the PVN (5). The observation that this response pattern is also evoked by chemical stimulation of the same region (16) indicates that it is mediated by cell bodies in the PVN rather than axons of passage. However, the important question that remains unanswered is defining the neural pathways and neurotransmitters involved in processing the sympathoexcitatory component of the peripheral chemoreflex within the PVN. Known alterations in the function of sympathetic pathways within the PVN during heart failure (15) and hypertension (1) underscore the importance of understanding the neural pathways and neurotransmitters that implicate the PVN in altering peripheral chemoreflex function in various disease states.
It is very well known that the inhibitory neurotransmitter GABA plays a pivotal role in the central regulation of cardiorespiratory function. One site where GABA has a tonic influence on ABP, HR, and sympathetic nerve activity is the PVN (17). Indeed, studies conducted in the recent past have documented that acute blockade of GABAA receptors in the PVN leads to increases in renal sympathetic and cardiorespiratory variables (25, 27). In the present study microinjection of bicuculline into the PVN augmented the increase in renal sympathetic activity and ventilatory response typically elicited by stimulation of peripheral chemoreceptors. Conversely, microinjection of muscimol into the PVN attenuated the reflex responses evoked by the chemoreceptor stimulation. Microinjections of these compounds targeted adjacent but outside the PVN had no effect on these reflex responses. These results clearly implicate the contribution of GABAergic mechanisms in the PVN in modulating the sympathetic and ventilatory responses to the peripheral chemoreflex.
However, the important question that arises from the present study is the source of excitatory input to the PVN that provides the necessary excitation to mediate sympathetic and ventilatory responses to the activation of peripheral chemoreceptors. The study of Kubo et al. (14) demonstrated that blockade of both N-methyl-d-aspartate (NMDA) and non-NMDA excitatory amino acid receptors in the PVN attenuated the pressor response evoked by the stimulation of carotid body chemoreceptors; however, little is known concerning the specific neuronal pathways that drive excitatory glutamatergic activity to autonomic neurons of the PVN.
Although we cannot tell from the present study the neural pathways implicating the PVN in the modulation of the peripheral chemoreflex, it is reasonable to speculate that there may be an excitatory input from the NTS (which is the primary synapse for peripheral chemoreceptors) that provides the necessary excitation to the PVN to augment sympathetic and ventilatory responses during the activation of peripheral chemoreceptors. Supporting this notion are previous studies that showed direct projections from the NTS to the PVN (3). Moreover, the anatomical and electrophysiological evidence of projections between PVN and other important nuclei involved in autonomic regulation of cardiorespiratory function and sympathetic activity provides the PVN with the anatomical substrates necessary to influence sympathetic and respiratory adjustments during peripheral chemoreceptor activation. It is plausible that the postsynaptic neurons in the NTS project into the PVN to excite sympathoexcitatory projections from the PVN to the RVLM or from the PVN directly to the intermediate lateral column of the spinal cord.
Another important finding in our present study is the lack of involvement of the PVN in the modulation of reflex responses elicited by central chemoreceptors, indicating a selective role of the PVN in modulating the peripheral chemoreflex. An opposite type of selective response of the hypothalamus was documented in a previous study, which demonstrated that the caudal hypothalamus was selectively involved in the modulation of respiratory responses to hypercapnia but not hypoxia (22). To reconcile our results with this previous study, it seems that different parts of the hypothalamus are selectively involved in the modulation of peripheral and central chemoreflex responses; however, this notion remains to be further evaluated.
In contrast to our results, which did not show any evidence of involvement of the PVN in the modulation of reflex responses elicited by hypercapnia, a previous study demonstrated an increased c-fos expression in the PVN in response to both hypoxia and hypercapnia (2). An important difference between these studies, however, lies with the fact that peripheral chemoreceptor input was eliminated in our study but not in the c-fos study. It is well known that hypercapnia stimulates peripheral as well as central chemoreceptors. Our results, therefore, provide a more clear-cut differentiation between functional central and peripheral chemoreceptor inputs to the PVN.
In conclusion, the present study brings to light the involvement of PVN neurons in modulating the sympathetic and ventilatory responses to the selective stimulation of peripheral chemoreceptors in anesthetized rats. Our data suggest that a GABAergic mechanism within the PVN is involved in the modulation of reflex responses to peripheral chemoreceptor stimulation. The present results also make evident that the PVN is not involved in the modulation of the central chemoreceptor-mediated responses.
National Heart, Lung, and Blood Institute Grant PO-1 HL-62222
The authors thank Denise Arrick for technical assistance and Drs. Xuifei Hong and Yi-Fan Li for help during experiments.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society