Abstract
Peripheral chemoreflex sensitivity is potentiated in clinical and experimental chronic heart failure (CHF). Downregulation of nitric oxide (NO) synthase (NOS) in the carotid body (CB) is involved in this effect. However, it remains poorly understood whether carbon monoxide (CO) also contributes to the altered peripheral chemoreflex sensitivity in CHF. This work highlights the effect of NO and CO on renal sympathetic nerve activity (RSNA) in response to graded hypoxia in conscious rabbits. Renal sympathetic nerve responses to graded hypoxia were enhanced in CHF rabbits compared with sham rabbits. The NO donor S-nitroso-N-acetylpenicillamine (SNAP, 1.2 μg·kg−1·min−1) and the CO-releasing molecule tricarbonyldichlororuthenium (II) dimer {[Ru(CO)3Cl2]2, 3.0 μg·kg−1·min−1} each attenuated hypoxia-induced RSNA increases in CHF rabbits (P < 0.05), but the degree of attenuation of RSNA induced by SNAP or [Ru(CO)3Cl2]2 was smaller than that induced by SNAP + [Ru(CO)3Cl2]2. Conversely, treatment with the NOS inhibitor Nω-nitro-l-arginine (30 mg/kg) + the heme oxygenase (HO) inhibitor Cr (III) mesoporphyrin IX chloride (0.5 mg/kg) augmented the renal sympathetic nerve response to hypoxia in sham rabbits to a greater extent than treatment with either inhibitor alone and was without effect in CHF rabbits. In addition, using immunostaining and Western blot analyses, we found that expression of neuronal NOS, endothelial NOS, and HO-2 protein (expressed as the ratio of NOS or HO-2 expression to β-tubulin protein expression) was lower in CBs from CHF (0.19 ± 0.04, 0.17 ± 0.06, and 0.15 ± 0.02, respectively) than sham (0.63 ± 0.04, 0.56 ± 0.06, and 0.27 ± 0.03, respectively) rabbits (P < 0.05). These results suggest that a deficiency of NO and CO in the CBs augments peripheral chemoreflex sensitivity to hypoxia in CHF.
Keywords: carotid body, renal sympathetic nerve activity, chronic heart failure
enhanced peripheral chemoreflex sensitivity occurs in patients with chronic heart failure (CHF) and in experimental models of CHF (4, 5, 12, 16, 17, 24, 28). Arterial chemoreceptors in the aortic body and carotid body (CB) are the principal sensory receptors that detect changes in arterial O2 (6, 8, 18). Our previous studies documented that an augmented afferent input from CB chemoreceptors is involved in the enhancement of peripheral chemoreflex function in pacing-induced CHF rabbits (28), and this enhanced sensitivity of the peripheral chemoreflex contributes to the sympathetic activation in CHF (27). However, the mechanism of CB hypersensitivity-induced sympathetic activation in CHF is not well resolved.
Several lines of evidence indicate that the endogenous production of nitric oxide (NO) plays an important role in modulating CB chemoreceptor activity to hypoxia. Previous work from our laboratory demonstrated that decreased NO production in the CB was involved in the enhanced chemoreceptor activity in CHF (10, 27). Neuronal NO synthase (nNOS) is downregulated in the CB in CHF (10), and administration of an NO donor reduces the elevated CB chemoreceptor discharge in CHF rabbits (27). Gene transfer of nNOS to CBs reverses the enhanced chemoreceptor function in CHF rabbits (10). Endogenously generated carbon monoxide (CO) from heme oxygenase (HO)-2 (HO-2) is also a physiological regulator of CB activity (1, 19, 35). However, it is not known whether CO is also involved in the peripheral chemoreflex hypersensitivity in the experimental CHF rabbit.
First, we measured renal sympathetic nerve activity (RSNA) in response to graded hypoxia before and after intravenous administration of a CO donor, tricarbonyldichlororuthenium (II) dimer {[Ru(CO)3Cl2]2}; an HO inhibitor, Cr (III) mesoporphyrin IX chloride (CrMP); an NO donor, S-nitroso-N-acetylpenicillamine (SNAP); or an NOS inhibitor, Nω-nitro-l-arginine (l-NNA), in sham and CHF rabbits. Second, we observed the effects of the combined NO-CO manipulation on chemoreflex function. Third, we investigated protein expression of NOS and HO-2 in CBs from sham and CHF rabbits to determine whether the enhanced peripheral chemoreflex function in CHF rabbits is concomitant with an altered expression of NOS and HO-2 in CBs from these animals.
MATERIALS AND METHODS
Experimental Animals
All experiments were carried out on 56 male New Zealand White rabbits weighing 2.5–3.5 kg. The experimental protocols were approved by the University of Nebraska Medical Center Institutional Animal Care and Use Committee and were carried out in accordance with the guidelines of the National Institutes of Health (NIH Publication No. 85-23, revised 1996) and the American Physiological Society. Rabbits were housed in individual cages under controlled temperature and humidity and a 12:12-h dark-light cycle and fed standard rabbit chow with water available ad libitum. They were assigned to sham-operated and CHF groups.
Pacemaker Implantation and Induction of CHF
Rabbits were anesthetized with a cocktail consisting of 1.2 mg of acepromazine, 5.9 mg of xylazine, and 58.8 mg of ketamine, given as an intramuscular injection. Using sterile technique, we performed a left thoracotomy as previously described (10, 12, 28). Briefly, a pin electrode was attached to the left ventricle (LV) for pacing, the chest was closed, and the rabbits were placed on an antibiotic regimen [enrofloxacin (Baytril), 5 mg/kg im] for 5 days. After the rabbits recovered from the thoracotomy (∼2 wk), pacing was started at 320 beats/min and continued at this rate for 7 days; then the rate was gradually increased to 360 beats/min, with an increment of 20 beats/min each week. The progression of CHF was monitored by weekly echocardiograms (Acuson Sequoia 512C with a 4-MHz probe), with the pacemaker turned off for ≥30 min before the recordings were started. Sham-operated animals were subjected to a similar period of echocardiographic measurements. CHF was characterized by a >40% reduction in ejection fraction and fraction of shortening and dilation of the LV in systole and diastole.
Recording of RSNA
RSNA-recording electrodes were implanted as described previously (10, 12, 28). At the same time, arterial and venous catheters were inserted into the right carotid artery and jugular vein, respectively. The experiments were carried out 3–5 days after surgery. Changes of RSNA in response to stimulation of peripheral chemoreceptors were recorded in sham and CHF rabbits in the conscious state according to our previously described method (10, 12, 28). RSNA was expressed as percentage of maximum RSNA, which was determined in each rabbit by an intravenous bolus injection of sodium nitroprusside (100 μg/kg) (13). Peripheral chemoreceptors were stimulated preferentially in rabbits allowed to breathe graded mixtures of hypoxic gas for 3 min under isocapnic conditions. Because hypoxic stimulation of ventilation induces hyperventilatory hypocapnia, 2–4% CO2 was added to the hypoxic gases to maintain relatively constant arterial Pco2 during hyperventilation (28).
Experimental Protocols for Recording of RSNA
After the pacemaker was turned off for ∼30 min, the mean arterial pressure (MAP), heart rate (HR), and RSNA were measured at normoxia (21% O2) and different levels of hypoxia (15% and 10% O2). Arterial Po2 (PaO2), arterial Pco2, and pH of arterial blood were measured using a blood-gas analyzer (model ABL5, Radiometer, Copenhagen, Denmark). The following protocols were used to investigate the effects of endogenous CO and NO on RSNA in response to hypoxia in sham and CHF rabbits.
Protocol 1: effects of the HO inhibitor CrMP.
In sham and CHF rabbits, the RSNA-PaO2 curve was recorded before and after bolus administration of the HO inhibitor CrMP (0.5 mg/kg iv).
Protocol 2: effects of the CO donor [Ru(CO)3Cl2]2.
In sham and CHF rabbits, the RSNA-PaO2 curve was recorded before and repeated 10 min after administration of [Ru(CO)3Cl2]2 (3.0 μg·kg−1·min−1 iv). [Ru(CO)3Cl2]2 was continuously infused during the hypoxia and recovery intervals to construct the RSNA-PaO2 curve.
Protocol 3: effects of the NOS inhibitor l-NNA.
To test the effects of l-NNA on RSNA in sham and CHF rabbits, protocol 1 was repeated, except l-NNA (30 mg/kg iv) was substituted for CrMP. To minimize the hypertensive effect of l-NNA infusion on the baroreflex, hydralazine (0.01–0.06 mg·kg−1·min−1 iv) was infused simultaneously with l-NNA to keep MAP at the control level.
Protocol 4: effects of the NO donor SNAP.
The NO donor SNAP (1.2 μg·kg−1·min−1) was infused as described for protocol 2. To minimize the hypotensive effect of SNAP on the baroreflex, phenylephrine (0.2–1.0 μg·kg−1·min−1 iv) was infused simultaneously with SNAP to keep MAP at the control level.
Protocol 5: effects of the HO inhibitor + the NOS inhibitor.
To assess the combined effects of HO and NOS blockade, protocol 3 was repeated, except CrMP (0.5 mg/kg) and l-NNA (30 mg/kg) were administered simultaneously intravenously. Similarly, intravenous infusion of hydralazine was used to minimize activation of the baroreflex.
Protocol 6: effects of the CO donor + the NO donor.
To assess the combined effects of CO and NO, [Ru(CO)3Cl2]2 (3.0 μg·kg−1·min−1) and SNAP (1.2 μg·kg−1·min−1) were infused simultaneously intravenously over the time course described in protocols 2 and 4. The other procedures described in protocol 4 were followed.
Doses of drugs were based on established effective doses (14, 15, 27, 29, 32) and the dose-response curves in our preliminary study. All drugs were freshly prepared before perfusion on the experimental day.
Immunofluorescence Detection
The animals were perfused transcardially with 4% paraformaldehyde for 30 min. Both CBs in each rabbit were rapidly removed and postfixed in 4% paraformaldehyde for 12 h at 4°C and then soaked in 30% sucrose for 12 h at 4°C for cryostat protection. The CB was cut into 10-μm-thick sections, which were mounted on precoated glass slides for immunofluorescence for nNOS, endothelial NOS (eNOS), HO-2, and tyrosine hydroxylase (TH) detection. CB sections on the glass slide were incubated with 10% normal donkey or goat serum for 1 h and then with primary anti-nNOS, anti-eNOS, anti-HO-2, and anti-TH antibodies overnight at 4°C. Then the sections were incubated with appropriate secondary antibody for 60 min at room temperature. Slides were observed under a Leica fluorescent microscope with appropriate excitation/emission filters. Images were captured by a digital camera system. No staining was observed when the procedure described above was repeated using PBS instead of the primary antibody.
Western Blot Analysis for Rabbit NOS and HO-2
CBs were rapidly removed and immediately frozen in dry ice and stored at −80°C. Briefly, the protein in CBs was extracted in lysing buffer (10 mM PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, and 1% SDS) plus protease inhibitor cocktail (100 μl/ml). After centrifugation at 12,000 g for 20 min at 4°C, protein concentration in the supernatant was measured using a bicinchoninic acid protein assay kit (Pierce Chemical, Rockford, IL). Samples were adjusted to the same concentrations of protein with equal volumes of loading buffer containing β-mercaptoethanol and boiled for 5 min. Protein (5 μg) was loaded for electrophoresis using a Bio-Rad minigel apparatus at 80 mA for 45 min. Proteins were fractionated in a 10% polyacrylamide gel along with molecular weight standards and electrophoretically transferred onto the polyvinylidene difluoride membrane (Millipore) at 300 mA for 90 min. The membrane was probed with primary antibody (1:1,000 dilutions of anti-nNOS, anti-eNOS, and anti-HO-2 polyclonal antibodies) and secondary antibody (1:5,000 dilutions of IgG-horseradish peroxidase) and then treated with enhanced chemiluminescence substrate for 5 min at room temperature. The bands in the membrane were visualized and analyzed using UVP BioImaging Systems. HO-2 or NOS protein intensity was normalized by β-tubulin.
Drugs and Reagents
CrMP was obtained from Frontier Scientific; goat anti-nNOS, goat anti-eNOS, and chicken anti-HO-2 antibodies from Abcam, R & D Systems, and Alpha Diagnostic, respectively; secondary antibodies for immunofluorescent labeling from Molecular Probes; chemicals for protein extraction, anti-β-tubulin antibody, secondary antibodies, IgG-horseradish peroxidase, and chemiluminescence substrate for Western blotting from Pierce Chemical; and mouse anti-TH, [Ru(CO)3Cl2]2, SNAP, l-NNA, and the other chemicals from Sigma-Aldrich (St. Louis, MO).
Data Analysis
Values are means ± SE. RSNA is expressed as percentage of maximal RSNA. Peripheral chemoreflex function curves were analyzed by plotting data points averaged over 30 s for RSNA against the corresponding PaO2. Statistical significance was determined by Student's paired t-test for hemodynamics and Student's unpaired t-test for protein expression between sham and CHF rabbits. Two-way ANOVA, with Bonferroni's procedure for post hoc analysis, was used for comparisons of the changes of RSNA in response to graded levels of hypoxia. All data were confirmed to fit reasonably within normal distribution, and equality of variances was confirmed by Levene's test. Statistical significance was accepted when P < 0.05.
RESULTS
Characteristics of CHF
CHF was induced by the 3rd or 4th wk of rapid ventricular pacing, consistent with our previous studies (10, 12, 28). CHF was characterized by an enlarged heart and weak contractile function, as evaluated by the changes of heart weight, LV chamber diameter, and shortening and ejection fractions (Table 1). The paced rabbits also exhibited a slight decrease in MAP and a significant increase in HR compared with the unpaced state.
Table 1.
Body weight, ventricular weight, and cardiac function in sham and CHF rabbits
| Sham (n = 30) |
CHF (n = 26) | |||
|---|---|---|---|---|
| Baseline | 3–4 wk | Baseline | 3–4 wk | |
| BW, kg | 2.68±0.10 | 3.18±0.08* | 2.64±0.11 | 3.35±0.06* |
| VW, g | 5.99±0.12 | 7.95±0.36† | ||
| VW/BW, g/kg | 1.89±0.05 | 2.36±0.06† | ||
| LVEDD, mm | 12.96±0.62 | 13.71±0.54 | 12.56±0.48 | 15.71±0.51*† |
| LVESD, mm | 8.52±0.48 | 8.40±0.54 | 8.34±0.35 | 12.01±0.46*† |
| FS, % | 41.06±1.98 | 40.21±2.68 | 42.53±2.31 | 21.73±1.69*† |
| EF, % | 75.01±2.65 | 74.95±2.83 | 74.84±1.96 | 48.59±2.52*† |
Values are means ± SE; n, number of rabbits. CHF, chronic heart failure; BW, body weight; VW, weight of both ventricles; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; FS, fraction of shortening; EF, ejection fraction.
P < 0.05 vs. baseline.
P < 0.05 vs. sham at 3–4 wk.
Effect of HO Inhibitor, CO Donor, NOS Inhibitor, and NO Donor on Baseline MAP, HR, and RSNA
Injection of the HO inhibitor CrMP (0.5 mg/kg iv) induced a transient slight MAP increase of 5–8 mmHg in sham and CHF rabbits that returned to control levels. HR was not significantly different from baseline levels in sham and CHF rabbits. Baseline RSNA was increased in sham rabbits and not changed in CHF rabbits (Table 2). Infusion of the CO donor [Ru(CO)3Cl2]2 (3.0 μg·kg−1·min−1) did not alter baseline MAP and HR in sham and CHF rabbits. CO donor infusion did not change the baseline RSNA in sham rabbits but decreased RSNA in CHF rabbits (Table 2).
Table 2.
Effects of CrMP and [Ru(CO)3Cl2]2 on baseline MAP, HR, and RSNA in sham and CHF rabbits
| Sham |
CHF | |||||
|---|---|---|---|---|---|---|
| MAP, mmHg | HR, beats/min | RSNA, %max | MAP, mmHg | HR, beats/min | RSNA, %max | |
| Control | 85.3±1.8 (8) | 243±13 (8) | 17.8±1.9 (8) | 76.4±2.2 (6) | 264±10 (6) | 24.4±2.8† (6) |
| CrMP | 86.4±2.4 (8) | 235±9 (8) | 23.6±2.3* (8) | 78.9±2.0 (6) | 257±13 (6) | 24.8±2.6 (6) |
| Control | 84.8±1.5 (7) | 238±10 (7) | 20.8±3.2 (7) | 77.4±1.6 (6) | 267±12 (6) | 25.3±1.8† (6) |
| [Ru(CO)3Cl2]2 | 83.9±2.2 (7) | 241±12 (7) | 19.7±2.7 (7) | 76.7±1.4 (6) | 274±11 (6) | 18.8±1.6* (6) |
Values are means ± SE of number of rabbits in parentheses. CrMP, Cr (III) mesoporphyrin IX chloride; [Ru(CO)3Cl2]2, tricarbonyldichlororuthenium (II) dimer; MAP, mean arterial blood pressure; HR, heart rate; RSNA, renal sympathetic nerve activity (expressed as percentage of maximum).
P < 0.05 vs. control.
P < 0.05 vs. sham.
Infusion of SNAP (1.2 μg·kg−1·min−1) caused a decrease in MAP and an increase in HR and RSNA in sham and CHF rabbits. Phenylephrine (0.2–1.0 μg·kg−1·min−1) returned MAP and HR to control levels in sham and CHF rabbits. After MAP stabilization, the baseline RSNA in sham rabbits was returned to control level, but the baseline RSNA in CHF rabbits remained decreased below control level (Table 3). l-NNA (30 mg/kg iv) increased MAP and lowered HR and RSNA in sham and CHF rabbits. Hydralazine (0.01–0.06 mg·kg−1·min−1 iv) returned MAP and HR to control levels. After MAP stabilization, the baseline RSNA returned to control level in CHF rabbits but remained above control level in sham rabbits (Table 3).
Table 3.
Effects of l-NNA and SNAP on baseline MAP, HR, and RSNA in sham and CHF rabbits and effects of hydralazine or phenylephrine infusion
| Sham |
CHF | |||||
|---|---|---|---|---|---|---|
| MAP, mmHg | HR, beats/min | RSNA, %max | MAP, mmHg | HR, beats/min | RSNA, %max | |
| Control | 84.5±1.7 (7) | 243±9 (7) | 17.8±2.2 (7) | 75.2±2.1 (6) | 264±12 (6) | 23.2±2.8† (6) |
| l-NNA | 93.3±1.7* (7) | 223±10* (7) | 10.2±2.8* (7) | 85.3±1.9* (6) | 235±9* (6) | 15.2±1.6* (6) |
| l-NNA + HY | 85.1±1.6 (7) | 245±14 (7) | 22.9±2.3* (7) | 75.6±1.6 (6) | 267±13 (6) | 24.8±3.4 (6) |
| Control | 83.9±1.3 (6) | 236±11 (6) | 16.9±2.5 (6) | 76.3±1.4 (7) | 267±10 (7) | 25.2±2.3† (7) |
| SNAP | 74.6±1.2 (6) | 267±14* (6) | 24.3±1.7* (6) | 66.2±1.3* (7) | 289±15* (7) | 32.4±1.6* (7) |
| SNAP + PE | 84.6±2.1 (6) | 239±12 (6) | 15.9±2.4 (6) | 75.8±1.2 (7) | 264±13 (7) | 20.2±2.8* (7) |
Values are means ± SE of number of rabbits in parentheses. l-NNA, Nω-nitro-l-arginine; SNAP, S-nitroso-N-acetylpenicillamine; HY, hydralazine; PE, phenylephrine.
P < 0.05 vs. control.
P < 0.05 vs. sham.
Effect of NO Donor and NOS Inhibitor on the Peripheral Chemoreflex in Sham and CHF Rabbits
Baseline RSNA (normoxia) and the response of RSNA to hypoxia were enhanced in CHF rabbits compared with sham rabbits (Figs. 1–3), consistent with our previous studies (10, 12, 28).
Fig. 1.
A and C: renal sympathetic nerve activity (RSNA)-arterial Po2 (PaO2) relationships before and after intravenous administration of the nitric oxide (NO) donor S-nitroso-N-acetylpenicillamine (SNAP) in sham rabbits and rabbits with chronic heart failure (CHF). B and D: RSNA-PaO2 relationships before and after intravenous administration of the NO synthase (NOS) inhibitor Nω-nitro-l-arginine (l-NNA). Hydralazine (0.01–0.06 mg·kg−1·min−1) or phenylephrine (0.2–1.0 μg·kg−1·min−1) was infused as appropriate to maintain mean arterial pressure (MAP) at control levels and minimize baroreflex effects (see Table 3). Values are means ± SE; n, number of animals. % of max, percentage of maximum. *P < 0.05 vs. control (before drug administration).
Fig. 3.
A and C: RSNA-PaO2 relationships before and after intravenous administration of SNAP + [Ru(CO)3Cl2]2 in sham and CHF rabbits. B and D: RSNA-PaO2 relationships before and after intravenous administration of l-NNA + CrMP. Hydralazine (0.01–0.06 mg·kg−1·min−1) or phenylephrine (0.2–1.0 μg·kg−1·min−1) was infused as appropriate to normalize MAP (see Table 3). Values are means ± SE; n, number of animals. *P < 0.05 vs. control.
In sham rabbits, SNAP (1.2 μg·kg−1·min−1 iv) did not affect the baseline RSNA during normoxia or the response of RSNA to graded hypoxia with MAP held at the control level by phenylephrine infusion (Fig. 1A); l-NNA increased RSNA during normoxia and enhanced the responses of RSNA to isocapnic hypoxia with MAP held constant by hydralazine infusion (Fig. 1B).
In CHF rabbits (Fig. 1, C and D), l-NNA did not change the baseline RSNA or the response of RSNA to hypoxia while maintaining control MAP. However, the NO donor SNAP markedly decreased RSNA responses to hypoxia in CHF rabbits at constant MAP.
Effect of CO Donor and HO Blocker on the Peripheral Chemoreflex in Sham and CHF Rabbits
In sham rabbits, [Ru(CO)3Cl2]2 (3.0 μg·kg−1·min−1 iv) did not affect RSNA during normoxia or in response to isocapnic hypoxia at constant MAP (Fig. 2A). CrMP increased RSNA during normoxia and enhanced the responses of RSNA to isocapnic hypoxia (Fig. 2B).
Fig. 2.
A and C: RSNA-PaO2 relationships before and after intravenous administration of the carbon monoxide (CO)-releasing molecule tricarbonyldichlororuthenium (II) dimer {[Ru(CO)3Cl2]2} in sham and CHF rabbits. B and D: RSNA-PaO2 relationships before and after intravenous administration of the heme oxygenase (HO) inhibitor Cr (III) mesoporphyrin IX chloride (CrMP) in sham and CHF rabbits. Values are means ± SE; n, number of animals. *P < 0.05 vs. control.
In CHF rabbits, the same dose of [Ru(CO)3Cl2]2 significantly reduced RSNA during normoxia and attenuated the response to isocapnic hypoxia in CHF rabbits at constant MAP (Fig. 2C). CrMP did not change RNSA during normoxia (baseline) or in response to isocapnic hypoxia in CHF rabbits (Fig. 2D).
Effect of CO Donor + NO Donor and HO Inhibitor + NOS Inhibitor on the Peripheral Chemoreflex in Sham and CHF Rabbits
The CO donor [Ru(CO)3Cl2]2 + the NO donor SNAP showed no effect on RSNA during normoxia or the response of RSNA to isocapnic hypoxia in the sham rabbits at constant MAP (Fig. 3A) but significantly blunted the enhanced RSNA during normoxia and hypoxia in CHF rabbits at constant MAP (Fig. 3C). Infusion of the HO inhibitor CrMP + the NOS inhibitor l-NNA increased RSNA during normoxia and hypoxia in sham rabbits at constant MAP (Fig. 3B). However, RSNA during normoxia and hypoxia was not altered by CrMP + l-NNA at constant MAP in CHF rabbits (Fig. 3D).
Comparison of NO and CO Pathways in the Peripheral Chemoreflex in Sham and CHF Rabbits
In sham rabbits, the HO inhibitor CrMP, the NOS inhibitor l-NNA, and CrMP + l-NNA enhanced the renal sympathetic nerve response to hypoxia compared with the renal sympathetic nerve response before administration of the drugs (P < 0.05; Figs. 1B, 2B, and 3B). However, the magnitude of the increased renal sympathetic nerve response to hypoxia induced by CrMP + l-NNA was greater than that induced by CrMP or l-NNA alone (Fig. 4A). In CHF rabbits, the CO donor [Ru(CO)3Cl2]2, the NO donor SNAP, and [Ru(CO)3Cl2]2 + SNAP decreased the augmented RSNA during normoxia and in response to hypoxia (P < 0.05; Figs. 1C, 2C, and 3C). The magnitude of the blunted RSNA in response to hypoxia induced by [Ru(CO)3Cl2]2 + SNAP was greater than that induced by SNAP or [Ru(CO)3Cl2]2 alone (Fig. 4B).
Fig. 4.
A: changes in RSNA response to hypoxia after manipulation with l-NNA (30 mg/kg), CrMP (0.5 mg/kg), or l-NNA + CrMP in sham rabbits. B: changes in RSNA response to hypoxia after manipulation with SNAP (1.2 μg·kg−1·min−1), Ru(CO)3Cl2]2 (3.0 μg·kg−1·min−1), or SNAP + Ru(CO)3Cl2]2 in CHF rabbits. Hydralazine (0.01–0.06 mg·kg−1·min−1) was infused after l-NNA injection to maintain MAP in sham rabbits. Normoxia (∼90 Torr PaO2), mild hypoxia (∼65 Torr PaO2), and severe hypoxia (∼42 Torr PaO2) as illustrated in Figs. 1–3. Values are means ± SE; n, number of animals. *P < 0.05 vs. l-NNA. #P < 0.05 vs. CrMP. $P < 0.05 vs. SNAP. &P < 0.05 vs. Ru(CO)3Cl2]2.
Protein Expression of HO-2 and NOS in CBs From Sham and CHF Rabbits
Western blot analysis showed that CHF markedly decreased the protein expression of endogenous HO-2 in the CBs compared with sham rabbits (Fig. 5, A and B). We also found that distinct immunostaining of HO-2 was predominantly localized to the glomus cell clusters in CB tissue and that immunostaining for HO-2 was decreased in the CB glomus cell clusters from CHF rabbits compared with sham rabbits (Fig. 5C). The protein expression of nNOS and eNOS also was lower in CBs from CHF rabbits than sham rabbits (Figs. 6, A and B, and 7, A and B). Similarly, Figs. 6C and 7C exemplify immunolabeling of nNOS and eNOS in CB tissue, respectively, with decreased immunostaining of both NOS isoforms in CBs from CHF rabbits within the region of glomus cells and in other areas of the CB.
Fig. 5.
Protein expression of heme oxygenase-2 (HO-2) in carotid bodies (CBs) from sham and CHF rabbits. A: representative bands of HO-2 and β-tubulin proteins. B: relative HO-2 protein expression in sham and CHF CBs. Values are means ± SE; n = 5 in each group. *P < 0.05 vs. sham. C: colocalization of tyrosine hydroxylase (TH) and HO-2 in CBs from a sham (a–c) and a CHF (d–f) rabbit. a and d: Green immunofluorescent image for TH; b and e: red immunofluorescent image for HO-2; c and f: merged image (yellow) for overlap of TH and HO-2.
Fig. 6.
Protein expression of neuronal NO synthase (nNOS) in CBs from sham and CHF rabbits. A: representative bands of nNOS and β-tubulin proteins. B: relative nNOS protein expression in sham and CHF CBs. Values are means ± SE; n = 5 in each group. *P < 0.05 vs. sham. C: colocalization of TH and nNOS in CBs from a sham (a–c) and a CHF (d–f) rabbit. a and d: Green immunofluorescent image for TH; b and e: red immunofluorescent image for nNOS; c and f: merged image (yellow) for overlap of TH and nNOS.
Fig. 7.
Protein expression of endothelial NO synthase (eNOS) in CBs from sham and CHF rabbits. A: representative bands of eNOS and β-tubulin proteins. B: relative eNOS protein expression in sham and CHF CBs. Values are means ± SE; n = 5 in each group. *P < 0.05 vs. sham. C: colocalization of TH and eNOS in CBs from a sham (a–c) and a CHF (d–f) rabbit. a and d: Green immunofluorescent image for TH; b and e: red immunofluorescent image for eNOS; c and f: merged image (yellow) for overlap of TH and eNOS.
DISCUSSION
The major findings in the present study are that pharmacological suppression of CO and NO signaling pathways enhances CB chemoreflex sensitivity to hypoxia in a cumulative manner in normal rabbits. The CO and NO pathways appear to be endogenously suppressed in CHF, which contribute, in a similar collective manner, to the enhanced chemoreflex function observed in these animals. These alterations are consistent with the measured downregulation of HO-2, nNOS, and eNOS in CBs from CHF rabbits.
Studies from our laboratory have documented that peripheral chemoreflex sensitivity is increased in conscious rabbits with pacing-induced CHF compared with sham rabbits (10, 12, 28). Inasmuch as rabbits lack functional aortic chemoreceptors, the peripheral chemoreflex is attributable mainly to the CBs in this species (3, 33). We showed previously that inhibition of the CB reduces RSNA in CHF rabbits during normoxia (28). These results support the notion that enhanced basal activity from the CB chemoreceptors contributes to the sympathetic activation in CHF.
Although numerous factors may affect peripheral chemoreceptor function, NO is thought to play an important role in this alteration. We previously documented that a decrease of NO production in the CBs contributes to an exaggerated CB chemoreceptor activity and CB function in CHF rabbits (10, 27), but we have not investigated the effects of NO compared with CO on the functional chemoreflex in conscious rabbits. The two constitutive isomers of NOS, nNOS (or NOS-1) and eNOS (or NOS-3), are normally abundant in the CB (19, 23), as illustrated in the present immunohistochemical analysis of the CB in normal rabbits. The present findings of enhanced chemoreflex function with NOS inhibition in normal rabbits and with decreased nNOS expression in the CB in CHF are consistent with our previous findings (10, 11, 27). Kline et al. (9) observed greater chemoreflex responses to hypoxia in nNOS-deficient than in wild-type mice. We previously confirmed that the adenoviral transfer of the nNOS gene to the CB elevates nNOS protein expression and NO production in CHF rabbits to levels found in sham rabbits and reverses the enhanced chemoreceptor function in CHF rabbits (10). The specific nNOS inhibitor S-methyl-l-thiocitrulline enhanced CB chemoreceptor activity in sham rabbits and failed to increase CB chemoreceptor activity in CHF rabbits (10). These results, taken together, demonstrate that a marked downregulation of endogenous nNOS in the CB plays an important role in the enhanced CB chemoreflex in CHF rabbits.
A downregulation of eNOS, as well as nNOS, was also found in the CB of CHF rabbits. Decreased eNOS along with decreased nNOS is likely to exacerbate depletion of functional NO in the CB and contribute to enhancement of the CB chemoreflex in CHF (31). Suppression of eNOS expression in the CB raises an intriguing but unresolved question: do changes in CB microvascular function contribute to altered chemoreceptor sensitivity in CHF?
CO is also known to be an important signaling molecule in the CB (19, 20, 22). The present study documents for the first time that CO, similar to NO, plays a functional role in restraining hypoxic sensitivity of the CB in the normal conscious rabbit and that a CO deficiency in the CB contributes to enhanced peripheral chemoreflex function and sympathetic activation in CHF. At least three isoforms of HO have been identified, the inducible HO-1 and the constitutive HO-2 and HO-3 isoforms (15, 22), but HO-2 is highly and constitutively expressed in neuronal and chemosensing tissues, including CB glomus cells, whereas HO-1 is not (19, 35). Several studies have assessed the functional significance of HO-2 in the CB (19, 20, 22). Zn-protoporphyrin-9, another inhibitor of HO, augments the CB sensory discharge in vitro, and exogenous administration of CO can reverse the augmentation of sensory discharge induced by Zn-protoporphyrin-9 (20). In the anesthetized rat, HO inhibition enhances respiratory responses to hypoxia but not to CO2, and the site of action is on the CB, since the effects of HO inhibition are abolished by bilateral section of carotid sinus nerves (19). Similarly, the respiratory responses to hypoxia are greater in mutant mice lacking the HO-2 isoform than in wild-type mice (19). These data strongly support our notion that an attenuated HO-2 activity in the CB of CHF rabbits contributes to the enhanced peripheral reflex response to hypoxia in CHF.
Williams et al. (35) suggest that HO-2 functions as an O2 sensor by regulating K+ channel activity during O2 deprivation in CB cells, primarily through CO production (35). However, we observed that pharmacological inhibition of CO production enhances, rather than abolishes or suppresses, chemoreflex responsiveness to hypoxia. Thus it seems that, in the rabbit at least, CO, similar to NO, serves as a modulator of afferent activity, rather than as a mediator of O2 sensing. Our present results also show that HO-2 protein expression is markedly decreased in the CB glomus cell cluster, whereas reflex responses to hypoxia are enhanced in CHF rabbits compared with sham rabbits. Enhanced sensitivity to hypoxia in the setting of HO-2 downregulation would be inconsistent with HO-2 serving as a primary O2 sensor.
Our experiments further document a functional interaction between CO and NO pathways in the RSNA response to hypoxia. Administration of CO + NO donors blunted the exaggerated hypoxia-induced chemoreflex responses in CHF rabbits to a greater extent than administration of either donor alone. Conversely, in sham rabbits, HO + NOS inhibition induced a greater enhancement of the RSNA activation to hypoxia than either inhibitor alone. Thus the CO and NO pathways in the CB have cumulative effects to enhance peripheral chemoreflex function under normoxic and hypoxic conditions in pacing-induced CHF rabbits.
The commonality of the downstream signaling pathways for CO and NO synthesized in rabbit CBs need to be further defined. Importantly, NO and CO share some common properties with O2. 1) They bind to heme with greater affinity than O2 and are coupled to activation of heme-containing proteins (19). 2) Recent studies have demonstrated that NO and CO are involved in regulating intracellular Ca2+ concentration (19) and the Ca2+-sensitive K+ channel in glomus cells (11, 35). 3) Since NOS and HO-2 activity are sensitive to O2 concentration and operate over a wide range of O2, both are capable of contributing to the O2 sensitivity of the CB (11, 19, 35). 4) NO and CO responses are mediated by stimulation of cGMP production in many cells (11, 19, 26, 35). Importantly, several studies have shown that NO elevates cGMP levels and enhances Ca2+-dependent K+ channel activity via a cGMP-dependent mechanism in glomus cells (25, 34). Our previous results confirmed that the effect of NO on outward K+ currents (IK) in rabbit CB glomus cells is cGMP dependent, because the guanylate cyclase inhibitor 1H-[1,2,4]oxadiazole[4,3-a]quinoxalin-1-one inhibited the effect of SNAP on IK and the cGMP agonist can mimic the effect of SNAP on IK in sham and CHF rabbits (11). However, further studies are needed to determine whether CO shares a similar effect.
Limitations
Pharmacological manipulation in the whole animal does not allow one to easily define the target of the drug effects. In the present study, our aim was to demonstrate that NO and CO play a functional role in CB chemoreflex sensitivity in the conscious rabbit and that these signaling pathways are altered in CHF. We are not able to define the specific locations in the reflex arc where these molecules are operating. However, CrMP does not cross the blood-brain barrier in the rat (2), and l-NNA inhibition of NOS in the brain is delayed ≥30 min after systemic administration in cat, pig, and dog (30), which would suggest that the effect of these inhibitors to enhance chemoreflex control of RSNA as observed in the present study is not due to a central action. In addition, from companion studies in our laboratory in which we used the same rabbit model of pacing-induced CHF, we have shown an enhanced afferent input from CB chemoreceptors in CHF (10, 12, 27). Furthermore, the enhanced afferent input from the CB in CHF was shown to be mediated in part by a downregulation of nNOS in the CB (10). In the present study, we found that protein levels of HO-2 and eNOS, in addition to nNOS, are depressed in the CB of CHF rabbits. From these results, we can infer that l-NNA, SNAP, CrMP, and [Ru(CO)3Cl2]2 act on the chemoreflex, at least in part, by modulating CB chemoreceptor activity, but we cannot completely exclude possible effects on central or efferent sympathetic neural sites.
Changes in arterial pressure due to the vasoactive effects of NO/CO or their inhibitors were minimized in the present study by intravenous infusion of hydralazine or phenylephrine to prevent confounding influences of the arterial baroreflex on RSNA. These agents, themselves, are not likely to have contributed to alterations in chemoreflex function observed in our experiments. First, qualitatively similar changes in chemoreflex effects on RSNA were observed in a few animals in which pressure was not controlled by infusion. Second, infusion of these drugs alone at doses used in the present study had no effect on RSNA responses to hypoxia in sham or CHF rabbits.
Perspective
It is generally accepted that the CB chemoreflex plays a relatively minor role in the overall control of tonic sympathetic outflow under normal resting conditions. Consequently, the CB chemoreflex is often overlooked or dismissed as a factor that contributes to altered sympathetic function in cardiovascular disease not associated with hypoxemia. We demonstrate from our present and previous results that a marked increase in CB chemoreflex sensitivity occurs in CHF as a result of alterations in multiple signaling pathways in the CB and central autonomic sites (7, 10–12, 21, 23, 24). This shift in sensitivity enhances CB chemoreflex control of sympathetic outflow, even under normoxic conditions in CHF (Fig. 3) (23, 28). Thus the concept of the CB chemoreflex as a “hypoxia” reflex needs to be redefined to accommodate its enhanced functionality under more normoxic eucapnic conditions in cardiovascular diseases. This shift in functionality of the chemoreflex elucidates its significance in the development and progression of sympathetic dysfunction in CHF.
The present study is the first to demonstrate a role for CO in altered CB chemoreflex function in a disease state. Our results demonstrate a cooperative role of the endogenous CO and NO pathways in the tonic restraint of CB chemoreflex sensitivity and sympathetic outflow in normal rabbits. Cosuppression of endogenous CO and NO acts cumulatively to enhance peripheral chemoreflex function in CHF. This alteration contributes to the increased sympathetic outflow under normoxic and hypoxic conditions in this disease state. The significance of CO and CO-NO interactions in other neuronal mechanisms of integrative control of sympathetic outflow in health and disease deserves further study.
GRANTS
This study was supported by National Institutes of Health Program Project Grant PO1-HL-62222 and Grant P20-RR-17675. Y. Ding was supported by American Heart Association, Heartland Affiliate, Postdoctoral Fellowship 0725749Z.
Acknowledgments
The authors thank Dr. Kurtis Cornish and Kaye Talbitzer for surgical assistance and management of the heart failure animal core at the University of Nebraska Medical Center.
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.
REFERENCES
- 1.Barbe C, Al-Hashem F, Conway AF, Dubuis E, Vandier C, Kumar P. A possible dual site of action for carbon monoxide-mediated chemoexcitation in the rat carotid body. J Physiol 543: 933–945, 2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bundock EA, Drummond GS, Kappas A. Tissue distribution of synthetic heme analogues: studies with tin, chromium, and zinc mesoporphyrins. Pharmacology 52: 187–198, 1996. [DOI] [PubMed] [Google Scholar]
- 3.Chalmers JP, Korner PI, White SW. The relative roles of the aortic and carotid sinus nerves in the rabbit in the control of respiration and circulation during arterial hypoxia and hypercapnia. J Physiol 188: 36–50, 1967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chua TP, Ponikowski P, Webb-Peploe K, Harrington D, Anker SD, Piepoli M, Coats AJ. Clinical characteristics of chronic heart failure patients with an augmented peripheral chemoreflex. Eur Heart J 18: 480–486, 1997. [DOI] [PubMed] [Google Scholar]
- 5.Chugh SS, Chua TP, Coats AJS. Peripheral chemoreflex in chronic heart failure: friend and foe. Am Heart J 132: 900–904, 1996. [DOI] [PubMed] [Google Scholar]
- 6.Fione S, Gonzalez C. Initiation and control of chemoreceptor activity in the carotid body. In: Handbook of Physiology. The Respiratory System. Control of Breathing. Bethesda, MD: Am. Physiol. Soc., 1986, sect. 3, vol. II, pt. 2, chapt. 9, p. 247–312.
- 7.Gao L, Pan YX, Wang WZ, Li YL, Schultz HD, Zucker IH, Wang W. Cardiac sympathetic afferent stimulation augments the arterial chemoreceptor reflex in anesthetized rats. J Appl Physiol 102: 37–43, 2007. [DOI] [PubMed] [Google Scholar]
- 8.Gonzalez C, Almaraz L, Obeso A, Rigual R. Carotid body chemoreceptors: from natural stimuli to sensory discharges. Physiol Rev 74: 829–898, 1994. [DOI] [PubMed] [Google Scholar]
- 9.Kline DD, Yang T, Huang PL, Prabhakar NR. Altered respiratory responses to hypoxia in mutant mice deficient in neuronal nitric oxide synthase. J Physiol 511: 273–287, 1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Li YL, Li YF, Liu D, Cornish KG, Patel KP, Zucker IH, Channon KM, Schultz HD. Gene transfer of neuronal nitric oxide synthase to carotid body reverses enhanced chemoreceptor function in heart failure rabbits. Circ Res 97: 260–267, 2005. [DOI] [PubMed] [Google Scholar]
- 11.Li YL, Sun SY, Overholt JL, Prabhakar NR, Rozanski GJ, Zucker IH, Schultz HD. Attenuated outward potassium currents in carotid body glomus cells of heart failure rabbit: involvement of nitric oxide. J Physiol 555: 219–229, 2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Li YL, Xia XH, Zheng H, Gao L, Li YF, Liu D, Patel KP, Wang W, Schultz HD. Angiotensin II enhances carotid body chemoreflex control of sympathetic outflow in chronic heart failure rabbits. Cardiovasc Res 71: 129–138, 2006. [DOI] [PubMed] [Google Scholar]
- 13.Liu JL, Irvine S, Reid IA, Patel KP, Zucker IH. Chronic exercise reduces sympathetic nerve activity in rabbits with pacing-induced heart failure: a role for angiotensin II. Circulation 102: 1854–1862, 2000. [DOI] [PubMed] [Google Scholar]
- 14.Liu JL, Murakami H, Zucker IH. Angiotensin II-nitric oxide interaction on sympathetic outflow in conscious rabbits. Circ Res 82: 496–502, 1998. [DOI] [PubMed] [Google Scholar]
- 15.Motterlini R, Sawle P, Bains S, Hammad J, Alberto R, Foresti R, Green CJ. CORM-A1: a new pharmacologically active carbon monoxide-releasing molecule. FASEB J 19: 284–286, 2005. [DOI] [PubMed] [Google Scholar]
- 16.Ponikowski P, Chua TP, Piepoli M, Banasiak W, Anker SD, Szelemej R, Molenda W, Wrabec K, Capucci A, Coats AJS. Ventilatory response to exercise correlates with impaired heart rate variability in patients with chronic congestive heart failure. Am J Cardiol 82: 338–344, 1998. [DOI] [PubMed] [Google Scholar]
- 17.Ponikowski P, Chua TP, Piepoli M, Ondusova D, Webb-Peploe K, Harrington D, Anker SD, Volterrani M, Colombo R, Mazzuero G, Giordano A, Coats AJS. Augmented peripheral chemosensitivity as a potential input to baroreflex impairment and autonomic imbalance in chronic heart failure. Circulation 96: 2586–2594, 1997. [DOI] [PubMed] [Google Scholar]
- 18.Prabhakar NR Neurotransmitters in the carotid body. Adv Exp Med Biol 360: 57–69, 1994. [DOI] [PubMed] [Google Scholar]
- 19.Prabhakar NR NO and CO as second messengers in oxygen sensing in the carotid body. Respir Physiol 115: 161–168, 1999. [DOI] [PubMed] [Google Scholar]
- 20.Prabhakar NR, Dinerman JL, Agani FH, Snyder SH. Carbon monoxide: a role in carotid body chemoreception. Proc Natl Acad Sci USA 92: 1994–1997, 1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Reddy MK, Schultz HD, Zheng H, Patel KP. Altered nitric oxide mechanism within the paraventricular nucleus contributes to the augmented carotid body chemoreflex in heart failure. Am J Physiol Heart Circ Physiol 292: H149–H157, 2007. [DOI] [PubMed] [Google Scholar]
- 22.Ryter SW, Otterbein LE. Carbon monoxide in biology and medicine. Bioessays 26: 270–280, 2004. [DOI] [PubMed] [Google Scholar]
- 23.Schultz HD, Li YL. Carotid body function in heart failure. Respir Physiol Neurobiol 157: 171–185, 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Schultz HD, Sun SY. Chemoreflex function in heart failure. Heart Fail Rev 5: 45–56, 2000. [DOI] [PubMed] [Google Scholar]
- 25.Silva JM, Lewis DL. Nitric oxide enhances Ca2+-dependent K+ channel activity in rat carotid body cells. Pflügers Arch 443: 671–675, 2002. [DOI] [PubMed] [Google Scholar]
- 26.Snyder SH Nitric oxide: first in a new class of neurotransmitters. Science 257: 494–496, 1992. [DOI] [PubMed] [Google Scholar]
- 27.Sun SY, Wang W, Zucker IH, Schultz HD. Enhanced activity of carotid body chemoreceptors in rabbits with heart failure: role of nitric oxide. J Appl Physiol 86: 1273–1282, 1999. [DOI] [PubMed] [Google Scholar]
- 28.Sun SY, Wang W, Zucker IH, Schultz HD. Enhanced peripheral chemoreflex function in conscious rabbits with pacing-induced heart failure. J Appl Physiol 86: 1264–1272, 1999. [DOI] [PubMed] [Google Scholar]
- 29.Szalay L, Shimizu T, Schwacha MG, Choudhry MA, Rue LW 3rd, Bland KI, Chaudry IH. Mechanism of salutary effects of estradiol on organ function after trauma-hemorrhage: upregulation of heme oxygenase. Am J Physiol Heart Circ Physiol 289: H92–H98, 2005. [DOI] [PubMed] [Google Scholar]
- 30.Traystman RJ, Moore LE, Helfaer MA, Davis S, Banasiak K, Williams M, Hurn PD. Nitro-l-arginine analogues: dose- and time-related nitric oxide synthase inhibition in brain. Stroke 26: 864–869, 1995. [DOI] [PubMed] [Google Scholar]
- 31.Valdes V, Mosqueira M, Rey S, Del Rio R, Iturriaga R. Inhibitory effects of NO on carotid body: contribution of neural and endothelial nitric oxide synthase isoforms. Am J Physiol Lung Cell Mol Physiol 284: L57–L68, 2003. [DOI] [PubMed] [Google Scholar]
- 32.Vera T, Henegar JR, Drummond HA, Rimoldi JM, Stec DE. Protective effect of carbon monoxide-releasing compounds in ischemia-induced acute renal failure. J Am Soc Nephrol 16: 950–958, 2005. [DOI] [PubMed] [Google Scholar]
- 33.Verna A, Roumy M, Leitner LM. Loss of chemoreceptive properties of the rabbit carotid body after destruction of the glomus cells. Brain Res 100: 13–23, 1975. [DOI] [PubMed] [Google Scholar]
- 34.Wang ZZ, Stensaas LJ, Bredt DS, Dinger B, Fidone SJ. Localization and actions of nitric oxide in the cat carotid body. Neuroscience 60: 275–286, 1994. [DOI] [PubMed] [Google Scholar]
- 35.Williams SEJ, Wootton P, Mason HS, Bould J, Iles DE, Riccardi D, Peers C, Kemp PJ. Hemoxygenase-2 is an oxygen sensor for a calcium-sensitive potassium channel. Science 306: 2093–2097, 2004. [DOI] [PubMed] [Google Scholar]