Regulation of hypoxia-inducible factor-α isoforms and redox state by carotid body neural activity in rats

Abstract

Previous studies reported that chronic intermittent hypoxia (CIH) results in an imbalanced expression of hypoxia-inducible factor-α (HIF-α) isoforms and oxidative stress in rodents, which may be due either to the direct effect of CIH or indirectly via hitherto uncharacterized mechanism(s). As neural activity is a potent regulator of gene transcription, we hypothesized that carotid body (CB) neural activity contributes to CIH-induced HIF-α isoform expression and oxidative stress in the chemoreflex pathway. Experiments were performed on adult rats exposed to CIH for 10 days. Rats exposed to CIH exhibited: increased HIF-1α and decreased HIF-2α expression; increased NADPH oxidase 2 and decreased superoxide dismutase 2 expression; and oxidative stress in the nucleus tractus solitarius and rostral ventrolateral medulla as well as in the adrenal medulla (AM), a major end organ of the sympathetic nervous system. Selective ablation of the CB abolished these effects. In the AM, sympathetic activation by the CB chemoreflex mediates CIH-induced HIF-α isoform imbalance via muscarinic acetylcholine receptor-mediated Ca(2+) influx, and the resultant activation of mammalian target of rapamycin pathway and calpain proteases. Rats exposed to CIH presented with hypertension, elevated sympathetic activity and increased circulating catecholamines. Selective ablation of either the CB (afferent pathway) or sympathetic innervation to the AM (efferent pathway) abolished these effects. These observations uncover CB neural activity-dependent regulation of HIF-α isoforms and the redox state by CIH in the central and peripheral nervous systems associated with the chemoreflex.

Figure 1. CBA selectively impairs chemoreflex but not baroreflex function

A, rats were subjected to sham surgery or CBA by cryocoagulation. Sections of the carotid body were stained with 4,6′-diamino-2-phenylindole (blue) to detect nuclei and antityrosine hydroxylase antibody (red) to detect type I glomus cells (scale bar, 20 μm). B, representative tracings of phrenic nerve activity during Hx (12% O₂) in sham-operated and CBA rats are shown. The black bar indicates the duration of hypoxic challenge. C, minute neural breathing (in AU min^–1; means ± s.e.m.) was determined during Nx and Hx in sham-operated (n = 6) and CBA (n = 6) rats. **P < 0.01; n.s., P > 0.05 compared to sham-operated rats. D, representative tracings of SNA responses to increased BP induced by PE (3 μg kg⁻¹ i.v. at arrows) in sham-operated and CBA rats are shown. E, SNA baroreflex response (means ± s.e.m.) of sham-operated (n = 6) and CBA (n = 6) rats; n.s., P > 0.05. AP, action potential; BP, blood pressure; CBA, carotid body ablation; Hx, hypoxia; Nx, normoxia; PE, phenylephrine; ∫Phr, integrated phrenic nerve activity; ∫SNA, integrated splanchnic nerve activity. Note the absence of hypoxic ventilatory response in CBA rats with a fully preserved SNA baroreflex response.

Figure 2. CBA prevents CIH-induced changes in HIF-α isoform expression in brainstem regions

A, representative example of simultaneous measurements of ambient O₂ and blood O₂ saturation in a conscious sham-operated rat during exposure to 8 h of intermittent hypoxia. B, three groups of rats were studied: sham-operated and exposed to normoxia (C); sham-operated and exposed to CIH; and CIH-exposed CBA rats (CIH + CBA). Anatomical localization of nTS (green), RVLM (red) and a brainstem region that neither receives carotid body sensory input nor regulates sympathetic nerve activity (CON, blue) based on adult rat brain atlas (Paxinos & Watson, 2006) are shown (Ba). Representative immunoblots (Bb) and densitometric analysis (Bc) of HIF-1α, and HIF-2α in the micro-punches from nTS, RVLM and CON regions are shown. Changes in HIF-α isoforms were normalized to HIF-1β (HIF-α/HIF-1β) and presented as means ± s.e.m. relative to sham-operated, normoxia-exposed control rats. n = 6–7 rats each, **P < 0.01; n.s., P > 0.05. C, control; CBA, carotid body ablation; CIH, chronic intermittent hypoxia; HIF, hypoxia-inducible factor; nTS, nucleus tractus solitarious; RVLM, rostral ventrolateral medulla.

Figure 3. CBA prevents CIH-induced changes in HIF-α isoform expression in nTS and RVLM neurons

Three groups of rats were studied: sham-operated and exposed to normoxia (C); sham-operated and exposed to CIH; and CIH-exposed CBA rats (CIH + CBA). Brainstem sections containing either nTS or RVLM were obtained based on coordinates from adult rat brain atlas (Paxinos & Watson, ; nTS, Bregma-14.04 mm; RVLM, Bregma-12.36 mm). A and B, low magnification (2×) of sections stained with NeuN (red), a neuronal marker. C–F, high magnifications of sections containing nTS (C and D) and RVLM (E and F) double stained with either NeuN and HIF-1α or NeuN and HIF-2α are shown scale bar, 20 μm). Amb, nucleus ambiguus; AP, area postrema; CBA, carotid body ablation; CC, central canal; CIH, chronic intermittent hypoxia; HIF, hypoxia-inducible factor; nTS, nucleus tractus solitarious; RVLM, rostral ventrolateral medulla.

Figure 4. Effect of CBA on CIH-induced changes in the redox state in brainstem regions

Nox2 mRNA levels (A), Nox enzyme activity (B), Sod2 mRNA levels (C), Sod2 enzyme activity (D) and MDA levels (E) were determined in nTS, RVLM and control region micro-punches from the following groups of rats: sham-operated and exposed to normoxia (C); sham-operated and exposed to CIH; and CIH-exposed CBA rats (CIH + CBA). Data in bar graphs are presented as means ± s.e.m., n = 6–7 rats per group; *P < 0.05; **P < 0.01; and n.s., P > 0.05. CBA, carotid body ablation; CIH, chronic intermittent hypoxia; HIF, hypoxia-inducible factor; MDA, malondialdehyde; Nox, NADPH oxidase; nTS, nucleus tractus solitarious; RVLM, rostral ventrolateral medulla; Sod, superoxide dismutase.

Figure 5. Effects of CBA on CIH-induced changes in HIF-α isoform and redox state in the AM

Three groups of rats were studied: sham-operated and exposed to normoxia (C); sham-operated and exposed to CIH; and CIH-exposed CBA rats (CIH + CBA). A, representative immunoblots (Aa) and densitometric analysis (Ab) of HIF-1α and HIF-2α in the AMs of sham-operated, normoxia-exposed control rats (C) presented as ratio of HIF-α/HIF-1β. B–F, Nox2 mRNA levels (B), Nox enzyme activity (C), Sod2 mRNA levels (D), Sod2 enzyme activity (E) and cytosolic and mitochondrial aconitase activities (F). Data in bar graphs are presented as means ± s.e.m., n = 6–7 rats per group. **P < 0.01; n.s., P > 0.05 compared to sham-operated, normoxia-exposed rats (C). CBA, carotid body ablation; CIH, chronic intermittent hypoxia; Cyto, cytosolic; HIF, hypoxia-inducible factor; Mito, mitochondrial; Nox, NADPH oxidase; Sod, superoxide dismutase.

Figure 6. CBA prevents CIH-induced ASN

A, three groups of rats were studied: sham-operated and Nx-exposed (control); sham-operated and exposed to CIH; and CIH-exposed CBA rats (CIH + CBA). Representative tracings of ASN responses to Nx (21% O₂) and Hx (12% O₂). B, ASN response to Hx was quantified (means ± s.e.m., n = 6–7 rats each); **P < 0.01 compared to control and ##P < 0.01 compared to CIH-exposed rats. AP, action potential; ∫ASN, integrated adrenal sympathetic nerve activity; CBA, carotid body ablation; CIH, chronic intermittent hypoxia; Hx, hypoxia; Nx, normoxia.

Figure 7. ASA abolishes CIH-induced changes in HIF-α isoform and redox state in the AM

Three groups of rats were studied: sham-operated and exposed to normoxia (C); sham-operated and exposed to CIH; and unilateral ASA and exposed to CIH (CIH + ASA). A, immunofluorescence of choline acetyltransferase expression (stained in green) in sections from the AM of rats subjected to sham surgery (Aa) or ipsilateral ASA (Ab). B, representative immunoblots (Ba) and densitometric analysis (Bb) of HIF-1α and HIF-2α in the AM presented as HIF-α/HIF-1β relative to sham-operated, normoxia-exposed control rats (C). C–G, Nox2 mRNA levels (C), Nox enzyme activity (D), Sod2 mRNA levels (E), Sod2 enzyme activity (F) and cytosolic and mitochondrial aconitase activities (G) in the AM are shown. Data in bar graphs are presented as means ± s.e.m., n = 6–7 rats per group; **P < 0.01; n.s., P > 0.05 compared to vehicle-treated, normoxia-exposed rats (C). AM, adrenal medulla; ASA, adrenal sympathetic nerve ablation; CIH, chronic intermittent hypoxia; Cyto, cytosolic; HIF, hypoxia-inducible factor; Mito, mitochondrial; Nox, NADPH oxidase; Sod, superoxide dismutase.

Figure 8. Effects of ATR on CIH-induced autonomic responses and functional changes in the AM

Three groups of rats were studied: vehicle-treated and exposed to Nx (C); vehicle-treated and exposed to CIH; and CIH-exposed rats ATR-treated (10 mg kg⁻¹ day⁻¹ i.p., CIH + ATR). A–E, effects of ATR on heart rate (A), Hx-induced increase in CB sensory activity (B and C), and ASN in Nx and Hx (D and E). F, representative immunoblots (Fa) and densitometric analysis (Fb) of HIF-1α and HIF-2α in the AM presented as a ratio of HIF-α/HIF-1β are shown. G–I, Nox, Sod2 and aconitase enzyme activities in the AM are shown. Data in bar graphs are presented as means ± s.e.m., n = 6–7 rats per group; *P < 0.05; **P < 0.01; and n.s., P > 0.05 compared to vehicle-treated, Nx-exposed rats (C). AM, adrenal medulla; ASN, adrenal sympathetic nerve activity; ATR, atropine; CB, carotid body; CIH, chronic intermittent hypoxia; Cyto, cytosolic; HIF, hypoxia-inducible factor; Hx, hypoxia; imp, impulse; Mito, mitochondrial; Nox, NADPH oxidase; Nx, normoxia; Sod, superoxide dismutase.

Figure 9. Effects of Mus and nicotine on HIF-α isoform expression in pheochromocytoma 12 cells

A, Mus, at indicated concentration, was applied to pheochromocytoma 12 cell cultures repetitively (Mus for 5 min followed by vehicle for 10 min, repeated for three cycles). Representative immunoblots (Aa) and densitometric analysis (Ab) of HIF-1α and HIF-2α are shown. B, effects of ATR on repetitive Mus-induced (Mus + ATR) changes in HIF-α isoform expression are shown. C–E, effects of continuous Mus treatment (C), repetitive (nicotine for 5 min followed by vehicle for 10 min, three cycles; D) and continuous nicotine treatment (for 15 min; E) on HIF-α isoform expression are shown in representative immunoblots (Ca–Ea) and densitometric analysis (Cb–Eb) of HIF-1α and HIF-2α. Changes in HIF-1α and HIF-2α are analysed as HIF-α/HIF-1β. Data are means ± s.e.m., n = 6–7 independent experiments. **P < 0.01; n.s., P > 0.05 compared to control cells (C). ATR, atropine; C, control; HIF, hypoxia-inducible factor; Mus, muscarine.

Figure 10. Signalling mechanisms associated with repetitive Mus-mediated regulation of HIF-α isoform in pheochromocytoma 12 cells

A and B, following repetitive Mus treatment (10 μm, three cycles), cells were treated with either vehicle (A) or CHX (20 nm; B) for up to 90 min. Representative immunoblots (Aa and Ba) and densitometric analysis (Ab and Bb) of HIF-1α presented as HIF-1α/HIF-1β relative to control (C) cells are shown. C, effect of ATR (5 μm) on repetitive Mus-induced changes in [Ca²⁺]_i levels. D, representative immunoblots (Da) and densitometric analysis (Db) of p-mTOR protein levels presented as ratio of p-mTOR/mTOR in cells treated repetitively with Mus (10 μm, three cycles) in the absence or presence of either ATR (5 μm), calcium chelator (BAPTA, 10 μm), or Rap (100 nm) are shown. E, representative immunoblots (Ea) and densitometric analysis (Eb) of HIF-1α presented as HIF-1α/HIF-1β following repetitive Mus (10 μm, three cycles) treatment in the absence or presence of Rap (100 nm). F, calpain activity in cells treated repetitively with Mus (10 μm, three cycles) in the absence or presence of ATR (5 μm) or ALLM (10 μm) is shown. G, representative immunoblots (Ga) and densitometric analysis (Gb) of HIF-2α presented as HIF-2α/HIF-1β protein levels following repetitive Mus (10 μm, three cycles) treatment in the absence or presence of calpain inhibitor (ALLM, 10 μm). Data in bar graphs are presented as means ± s.e.m., n = 6–7 independent experiments; *P < 0.05; **P < 0.01; n.s., P > 0.05 compared to control cells (C). ALLM, N-acetyl-leucine-leucine-methionine-aldehyde; ATR, atropine; CHX, cycloheximide; HIF, hypoxia-inducible factor; mTOR, mammalian target of rapamycin; Mus, muscarine; p-mTOR, phosphorylated form of mTOR; Rap, rapamycin.

Figure 11. Effects of CIH on mTOR and calpain activation in the adrenal medulla

Five groups of rats were studied: vehicle-treated and exposed to normoxia (C); vehicle-treated and CIH-exposed (CIH); ATR-treated (10 mg kg⁻¹ day⁻¹ i.p.) and CIH-exposed (CIH + ATR); CIH-exposed CBA (CIH + CBA); and CIH-exposed bilateral ASA (CIH + ASA). Representative immunoblots (Aa) and densitometric analysis (Ab) of p-mTOR/total mTOR protein (A) and calpain activity (B). Data in bar graphs are presented as means ± s.e.m., n = 6–7 rats per group; *P < 0.05; **P < 0.01; n.s., P > 0.05. ASA, adrenal sympathetic nerve ablation; ATR, atropine; CBA, carotid body ablation; CIH, chronic intermittent hypoxia; mTOR, mammalian target of rapamycin; p-mTOR, phosphorylated mTOR.

Figure 12. Carotid body chemoreflex mediates CIH-induced changes in CA secretion from adrenal medulla, plasma CA and blood pressure

Four groups of rats were studied: sham-operated and normoxia-exposed (C); sham-operated and CIH-exposed (CIH); carotid body ablated (CBA) and CIH exposed (CBA + CIH); and bilateral ASA and CIH-exposed (CIH + ASA). A, Hx ( = 40 ± 3 mmHg, 5 min)-evoked NA and A secretion from adrenal medulla slices is presented as the percentage of basal release. B and C, plasma CA levels of NA and A (B) and MBP (C) are shown. D and E, effect of AIH (15 s at 12% O₂, followed by 5 min at 21% O₂) on ABP. Representative changes in ABP during pre-AIH, first and 10th AIH, and post-AIH periods (D) and MBP (E) are shown. Data in bar graphs are presented as means ± s.e.m., n = 6–7 rats; *P < 0.05; ***P < 0.001; n.s., P > 0.05. A, adrenaline; ABP, arterial blood pressure; AIH, acute intermittent hypoxia; ASA, adrenal sympathetic nerve ablation; C, control; CA, catecholamine; CBA, carotid body ablation; CIH, chronic intermittent hypoxia; Hx, hypoxia; MBP, mean blood pressure; NA, noradrenaline.