Oxygen Sensing and Homeostasis

Abstract

The discovery of carotid bodies as sensory receptors for detecting arterial blood oxygen levels, and the identification and elucidation of the roles of hypoxia-inducible factors (HIFs) in oxygen homeostasis have propelled the field of oxygen biology. This review highlights the gas-messenger signaling mechanisms associated with oxygen sensing, as well as transcriptional and non-transcriptional mechanisms underlying the maintenance of oxygen homeostasis by HIFs and their relevance to physiology and pathology.

As early as 1868, Pflüger reported that hypoxia stimulates breathing (58), a finding that spurred investigation to identify the structures that “sense” systemic O₂ levels and trigger physiological responses. Nearly 50 years later, Fernando de Castro, along with Jean-Francois Heymans and Corneille Heymans, independently identified carotid bodies as the “sensory organs” for monitoring arterial blood O₂ levels (10, 24). Heymans et al. further demonstrated that hypoxia-induced stimulation of breathing is due to a chemosensory reflex arising from the carotid body (24). A later study by von Euler et al. (89) firmly established the sensory nature of the carotid body by directly recording the sensory nerve responses to arterial hypoxemia (i.e., decreased arterial blood O₂ levels). Since identification of the carotid body as an O₂-sensing organ, much attention has been focused on delineating the mechanisms underlying the transduction process. Given that the carotid body responds to hypoxemia within a few seconds, it is likely that the transduction process involves changes in existing proteins rather than de novo protein synthesis (37).

Prolonged hypoxia initiates a series of physiological responses over a time scale of hours to days to maintain O₂ homeostasis. Examples include increased red blood cell production, formation of new blood vessels, and metabolic re-programming of cells, to name a few (72). The increased number of red blood cells improves O₂ carrying capacity, whereas angiogenesis facilitates the transport of oxygenated blood to the tissues, and re-organization of metabolic machinery optimizes O₂ utilization under prolonged O₂ deprivation. While all these biological consequences of prolonged hypoxia have been well documented for over a century (91), the underlying molecular mechanisms remained elusive until the last quarter century. The discovery of hypoxia-inducible factor 1 (HIF-1) in 1992 (76) and the subsequent identification of other members of the HIF family of transcriptional activators revolutionized the current understanding of the molecular mechanisms underlying O₂ homeostasis (63).

O₂ sensing by the carotid body impacts systemic cardiovascular and respiratory homeostasis in contrast to local effects, such as pulmonary vasoconstriction in response to alveolar hypoxia. The goal of this brief review is to highlight recently identified mechanisms of O₂ sensing by the carotid body and maintenance of O₂ homeostasis by HIF-dependent transcriptional programming during continuous exposure to hypoxia. The role of the carotid body and HIFs in the pathological response to intermittent hypoxia has been reviewed recently (70) and will not be discussed here.

O₂ Sensing

Sensing Elements and Characteristics of the Sensory Response to Hypoxia

The carotid bodies are located at the bifurcation of the common carotid artery into external and internal carotid arteries. The carotid bodies receive the highest blood flow per tissue weight of any organ in the body (1, 8, 9). The sensory innervation to the chemoreceptor tissue is provided by a branch of the glossopharyngeal nerve called the carotid sinus nerve (CSN). Under normoxia (arterial Po₂ of ∼100 Torr), the CSN activity (i.e., the frequency of nerve impulses) is low but increases dramatically even with a modest drop in arterial Po₂ (e.g., from 100 to 60-80 Torr) (4, 14, 27, 88). The response is remarkably fast, occurring within 0.2–0.4 s after onset of the stimulus in intact animals (5). The chemoreceptor tissue is a conglomeration of two cell types: type I or glomus cells, which are of neuronal phenotype, and the glia-like type II or sustentacular cells. A substantial body of evidence indicates that type I cells are the primary site of O₂ sensing and that they work in concert with the nearby afferent nerve ending as a sensory unit (37).

Signaling Mechanisms

Sensory receptors detecting diverse external stimuli such as tactile, thermal, visual, olfactory, and gustatory sensations utilize two categories of signaling molecules to initiate the sensory transduction. Ion channels initiate the transduction of touch and temperature, whereas G-protein-coupled receptors (GPCRs) trigger the transduction cascade at visual, olfactory, and taste receptors (32). In contrast, emerging evidence suggests that hypoxic sensing by the carotid body chemoreceptor utilizes biochemical mechanisms involving O₂-dependent interplay between two gaseous messengers: carbon monoxide (CO) and hydrogen sulfide (H₂S).

O₂-dependent CO generation by heme oxygenase 2.

CO is generated during the enzymatic degradation of heme by heme oxygenase 1 (HO-1) and HO-2, which require O₂ as a reaction substrate (42). Although HO-1 is an inducible isoform, HO-2 is constitutively expressed in various tissues, including carotid body type I cells of cat, rat (60), mouse (52), and humans (49). HO-2 is the primary enzyme contributing to CO generation in the carotid body (95). Hypoxia leads to a graded reduction of CO levels in the carotid body (57), and CO is absent in HO-2-null carotid bodies (95). The O₂-dependent modulation of CO generation was also observed in human embryonic kidney (HEK)-293 cells with forced expression of HO-2 (95). These findings demonstrate that CO generation from HO-2 is sensitive to changes in O₂.

Mechanisms of O₂ sensitivity of HO-2.

The inherent O₂ sensitivity of HO-2 requires the presence of two cysteine residues (Cys²⁶⁵ and Cys²⁸²) in the heme regulatory motif, and substitution of alanine for Cys²⁶⁵ and Cys²⁸² prevents the inhibition of CO generation under hypoxic conditions (95). With intact Cys²⁶⁵ and Cys²⁸² residues, HO-2 exhibits low affinity for O₂, with an apparent K_m = 65 ± 5 Torr, whereas mutation of Cys²⁶⁵ and Cys²⁸² dramatically increases O₂ affinity (apparent K_m = 25 ± 3 Torr). Physiological studies showed that CO signaling during normoxia inhibits CSN activity (57, 60). Under hypoxic conditions, CO-dependent inhibition is released, leading to carotid body activation. Emerging evidence indicates that CO mediates carotid body activation indirectly by regulating the synthesis of H₂S, another gaseous messenger.

H₂S synthesis by cystathionine-γ-lyase mediates carotid body activation.

Type I cells express cystathionine-γ-lyase (CSE), an enzyme that catalyzes H₂S generation (40, 56). Hypoxia increases H₂S generation in the carotid body in a graded manner, and this response is markedly attenuated or absent following pharmacological blockade or genetic ablation of CSE activity (56). However, the increased H₂S generation is not due to a direct effect of hypoxia on CSE, because in HEK-293 cells with forced expression of CSE, H₂S generation was unaffected by low O₂ (95)_. Recent studies have demonstrated that CO suppresses H₂S levels by inhibiting CSE activity in the carotid body (56, 57) as well as in HEK-293 cells engineered to express both HO-2 and CSE (95). CO inhibits CSE by stimulating soluble guanylate cyclase, which produces cyclic GMP that activates protein kinase G (PKG). PKG-dependent phosphorylation of CSE at serine 377 renders the enzyme inactive, resulting in reduced H₂S generation (95) (FIGURE 1).

FIGURE 1.

The O₂-sensing and -signaling pathway in the carotid body

Schematic presentation of CO-regulated H₂S generation in type I cells of the carotid body and its impact on sensory nerve activity. HO-2, heme oxygenase 2; sGC, soluble guanylate cyclase; cGMP, cyclic guanosine monophosphate; PKG- II, protein kinase G-II; CSE, cystathionine-ϒ-lyase; ETC, electron transport chain; NT, neurotransmitter.

The following findings demonstrate that CSE-derived H₂S mediates increased CSN activity in response to hypoxia: 1) CSE-deficient mice exhibit a remarkable absence of CSN excitation by hypoxia and near absence of stimulation of breathing by low O₂, a hallmark of the carotid body chemosensory reflex (56); 2) a similar reduction or absence of CSN responses to hypoxia was also seen following pharmacological inhibition of CSE (40, 56); and 3) exogenous application of H₂S donors, like hypoxia, stimulate the CSN activity of several mammalian species, including mice, rats, rabbits, and cats (33, 40, 56).

Cellular basis of H₂S-dependent activation of the carotid body by hypoxia.

The current consensus regarding the cellular basis of carotid body O₂ sensing is that hypoxia depolarizes type I cells leading to Ca²⁺-dependent release of excitatory neurotransmitters, which, by stimulating the afferent nerve ending, increase CSN activity (15, 20, 37, 51, 62). Two theories have been proposed to explain how hypoxia facilitates Ca²⁺-dependent neurotransmitter release from type I cells. According to the membrane hypothesis, low O₂ depolarizes type I cells by inhibiting K⁺ conductance, causing an influx of Ca²⁺ through voltage-gated Ca²⁺ channels, leading to neurotransmitter release. The metabolic or mitochondrial hypothesis proposes that hypoxia, by reducing a mitochondrial cytochrome, leads to depolarization of mitochondria that initiate sensory transduction (37).

The following findings suggest that H₂S affects ion channel conductance in type I cells. First, H₂S, like hypoxia, inhibits Ca²⁺-activated K⁺ channel activity (40, 85), as well as TASK-like K⁺ conductance, and depolarizes type I cells (6). Second, H₂S increases [Ca²⁺]_i in type I cells, and this response is not observed in the absence of extracellular Ca²⁺ (6, 43). Third, preventing the depolarization by voltage-clamping the type I cell at the resting membrane potential also abolishes H₂S-induced [Ca²⁺]_i elevation (6). Fourth, nifedipine, a blocker of L-type voltage-gated Ca²⁺ channels, as well as TTA-2, a T-type voltage-gated Ca²⁺ channel blocker, prevent H₂S-mediated [Ca²⁺]_i elevation in hypoxic type I cells (43, 44). Fifth, perhaps most important is the finding that CSE-null type I cells exhibit near absence of hypoxia-induced [Ca²⁺]_i elevation and transmitter release (43).

Based on spectral analysis, it was proposed that carotid bodies express a unique cytochrome with low affinity for O₂, which is half-reduced at Po₂ levels of 60–80 Torr (39, 81). It is quite possible that the spectral changes in the carotid body previously attributed to a low-affinity cytochrome might in fact be arising from the heme-HO-2 complex because it exhibits similar redox-dependent spectral changes (47). Furthermore, the apparent K_m of HO-2 for O₂ is 65 ± 5 Torr (95). The finding that H₂S increases NADH autofluorescence in type I cells (6) suggests that H₂S mediates its actions in part by affecting the mitochondrial electron transport chain (ETC). Taken together, these findings suggest that H₂S transduces the hypoxic stimulus by altering ion channel function, and perhaps also mitochondrial electron transport chain activity, in type I cells to elicit sensory nerve excitation.

O₂ sensing in HO-2-null mice: evidence for a backup signaling mechanism.

The inherent O₂ sensitivity of HO-2 and the ability of CO to regulate generation of the effector molecule H₂S suggests that HO-2-dependent CO generation is obligatory for carotid body O₂ sensing. Consistent with this possibility, HO-2-null mice exhibited increased basal CSN activity and H₂S levels (95). However, hypoxia still stimulated the carotid body and increased H₂S levels. Further analysis revealed that loss of HO-2 in type I cells resulted in a compensatory increase of another O₂-sensitive gaseous messenger-generating enzyme, neuronal nitric oxide synthase (nNOS), which catalyzes generation of nitric oxide (NO). Remarkably, inhibition of nNOS eliminated sensory nerve excitation and H₂S generation in response to hypoxia in HO-2-null carotid bodies (95). The O₂ affinity of nNOS is much lower (apparent K_m = 250 Torr) (95) than HO-2 (apparent K_m = 65 Torr), suggesting that even a modest reduction in Po₂ will lead to decreased NO production by nNOS. Furthermore, NO, like CO, inhibits CSN activity (50, 61) by binding to soluble guanylate cyclase and stimulating PKG-dependent phosphorylation of CSE at Ser³⁷⁷ (95). These findings suggest that, in the absence of HO-2 activity, the nNOS/NO system provides an alternative mechanism by which H₂S generation can be regulated according to O₂ availability, thereby providing an important fail-safe redundancy for a vital homeostatic process (FIGURE 1).

Physiological implications of gaseous messenger-mediated carotid body O₂ sensing.

The chemosensory reflex is a critical regulator of autonomic functions including breathing, sympathetic tone, and blood pressure (16, 37). However, the chemosensory reflex exhibits substantial variation in healthy human subjects (90) and in rodents (25, 82). For instance, the hypoxic ventilatory response (HVR), a hallmark of the chemosensory reflex, is augmented in spontaneous hypertensive (SH) rats and reduced in Brown Norway (BN) rats compared with Sprague-Dawley (SD) rats (23, 25, 79).

SH rats, even before development of hypertension, exhibited heightened carotid body responses to hypoxia, and this effect was associated with increased basal and hypoxia-induced H₂S levels and reduced CO levels in the chemoreceptor tissue compared with SD rats (57). It was further shown that the altered gas messenger levels were not due to changes in CSE or HO-2 protein levels. Instead, they appear due in part to decreased affinity of HO-2 for its substrate hemin, resulting in reduced CO generation and impaired inhibition of CSE, leading to high levels of H₂S. Treatment with L-propargylglycine (L-PAG), an inhibitor of CSE, or with CORM-2, a CO donor, eliminated hypersensitivity of the SH carotid body to hypoxia (57).

An exaggerated carotid body chemosensory reflex has been implicated in the development of essential hypertension in SH rats (55, 64, 86). Five-week-old SH rats were treated with L-PAG every day for 5 wk. L-PAG-treated rats exhibited a pronounced reduction in blood pressure compared with vehicle-treated controls. Ablation of the carotid bodies from 5-wk-old SH rats also attenuated age-dependent hypertension to the same extent as L-PAG treatment (57). However, treating carotid body-ablated rats with L-PAG caused no further decline in blood pressure. These findings indicate that elevated H₂S signaling in the carotid body contributes to the pathogenesis of hypertension in SH rats.

In striking contrast to SH rats, BN rats exhibited a severely impaired carotid body response to hypoxia, which was associated with elevated CO and reduced H₂S levels compared with SD rats. The chemosensory reflex was nearly absent in BN rats, as evidenced by the lack of ventilatory and sympathetic nerve stimulation by low O₂. Chronic exposure to high-altitude hypoxia leads to a carotid body-mediated increase in breathing, a phenomenon known as ventilatory adaptation to hypoxia (VAH) (11). Insufficient VAH leads to pulmonary edema (22, 26, 46). BN rats exposed to hypobaric hypoxia (0.35 atm; simulating 8,500-m altitude) for 16 h exhibited an absence of VAH and developed pulmonary edema, as evidenced by elevated protein levels in bronchoalveolar lavage fluid and an increased wet-to-dry lung weight ratio compared with SD rats (57). Treating BN rats with an HO inhibitor, which lowers CO levels, restored the carotid body response to hypoxia comparable to that seen in SD rats, improved VAH, and prevented pulmonary edema in response to hypobaric hypoxia (57). Taken together, these findings suggest that dysregulation of CO-H₂S signaling in the carotid body leads to pathological consequences.

O₂ Homeostasis

Maintenance of cellular O₂ levels is critical because either insufficient or excess O₂ leads to increased levels of reactive oxygen species (ROS), which oxidize lipids, proteins, and nucleic acids, leading to cell dysfunction and death. Thus both the delivery and the consumption of O₂ are precisely regulated. Many different molecular mechanisms are utilized to maintain oxygen homeostasis. Here, we will focus on adaptive responses to continuous hypoxia that are mediated by HIFs.

HIFs

At the transcriptional level, HIFs function as master regulators of oxygen homeostasis by controlling O₂ supply and demand. The HIFs are heterodimeric proteins that consist of an O₂-regulated HIF-α subunit (HIF-1α, HIF-2α, or HIF-3α) and a constitutively expressed HIF-1 subunit. HIFs are subject to transcriptional, posttranscriptional, translational, and posttranslational regulation (63). In particular, the dioxygenases PHD2 and FIH-1 use O₂ and α-ketoglutarate as substrates to hydroxylate proline and asparagine residues in the HIF-α subunits, which inhibit protein stability and transcriptional activity, respectively. Prolyl hydroxylation by PHD2 leads to binding of the VHL protein, which recruits an ubiquitin protein ligase that targets HIF-α subunits for proteasomal degradation, whereas asparagine hydroxylation by FIH-1 blocks binding of the co-activator protein p300 (63). Under hypoxic conditions, dioxygenase activity is inhibited, leading to HIF-α protein accumulation and induction of transcriptional activity, providing a molecular mechanism for the transduction of changes in O₂ availability to changes in gene expression (FIGURE 2).

FIGURE 2.

HIF-1 regulates O₂ supply and demand

Pathways that are active under hypoxic and nonhypoxic conditions are indicated by blue and red arrows, respectively.

Over 1,500 target genes are known to be transactivated by binding of HIFs to a cis-acting hypoxia response element (HRE) containing the sequence 5′-RCGTG-3′ (R = A or G), located either within the target gene or its flanking sequences (75). Within any given cell, hypoxia increases the expression of hundreds of mRNAs and decreases the expression of a similar number of genes in a HIF-dependent manner (45). HIFs indirectly repress gene expression by transactivation of genes encoding transcriptional repressors and micro-RNAs. HIFs also have global epigenetic effects on transcription by transactivating genes encoding chromatin-modifying enzymes, such as histone demethylases and histone deacetylases (41, 74, 92, 93).

In addition to their canonical role in the regulation of gene transcription as components of heterodimeric DNA-binding proteins, isolated HIF-α subunits also play critical roles in hypoxic adaptation that are independent of HIF-1β. The HIF-1α subunit interacts with other DNA-binding transcription factors to function as either a co-activator or a co-repressor (74). HIF-1α and HIF-2α also regulate responses to hypoxia via non-transcriptional mechanisms: HIF-1α interacts with protein components of the pre-replication complex in hypoxic cells to regulate DNA synthesis (28, 29), and HIF-2α interacts with translation initiation factors to regulate protein synthesis under hypoxic conditions (87). Thus HIFs contribute to oxygen homeostasis by multiple molecular mechanisms and regulate the synthesis of DNA, mRNA, microRNA, and protein. The utilization of these molecular mechanisms results in physiological responses to continuous hypoxia that either increase O₂ delivery or decrease O₂ consumption, as illustrated by several examples below (FIGURE 2).

Regulation of O₂ Delivery

Erythropoiesis.

Erythropoietin is a glycoprotein hormone produced by the kidney that binds to receptors on erythroid progenitor cells in the bone marrow and stimulates their survival, proliferation, and differentiation (30, 36, 80). EPO levels increase in response to anemia or systemic hypoxia, leading to increased blood O₂-carrying capacity. HIF-1 was originally identified as a nuclear DNA-binding protein that was induced by hypoxia and bound to an HRE in the human EPO gene, which encodes erythropoietin (76). Subsequent studies have revealed that HIF-2 plays a critical role in regulating erythropoiesis by activating transcription of EPO and genes encoding proteins that are required for the absorption and delivery of iron to the bone marrow, such as ceruloplasmin, transferrin, transferrin receptor, and the iron membrane transport protein DMT1 (46, 53, 69, 74, 77, 94). Individuals with hereditary erythrocytosis (also known as congenital polycythemia) have excess red blood cell production due to missense mutations in the gene encoding PHD2, VHL, or HIF-2α, providing genetic proof for the key role of the HIF pathway in regulating erythropoiesis in humans (73, 94). EPO production is impaired in patients with chronic renal failure who are currently treated by injections of recombinant human erythropoietin (13). Prolyl hydroxylase inhibitors, which induce HIF activity, provide an alternative pharmacological approach to stimulating erythropoiesis by induction of hepatic EPO production (3, 12, 48, 67, 83).

Angiogenesis.

Whereas anemia results in systemic hypoxia, vascular disease results in local tissue hypoxia in the distribution of the affected vessel in the heart or limb muscle of patients with coronary or peripheral vascular disease, respectively. A further difference is that anemia results in reduced O₂ availability (i.e., hypoxia), whereas vascular disease results in ischemia, a condition in which reduced tissue perfusion leads to reduced availability of O₂ and energy substrates (e.g., glucose) as well as increased levels of toxic metabolites (e.g., CO₂, H⁺, K⁺, and NH₃). Remarkably, however, hypoxia appears to represent the critical stimulus that evokes an adaptive response to ischemia, principally through the HIF-dependent production of multiple angiogenic cytokines and growth factors, including vascular endothelial growth factor, angiopoietins, placental growth factor, platelet-derived growth factor B, stem cell factor, and stromal-derived factor 1, which stimulate angiogenesis, which is the formation of new capillaries, as well as arteriogenesis, which is the remodeling of existing vessels to increase luminal diameter and blood flow (66). Aging impairs vascular responses to ischemia by inhibiting HIF activity and thereby inhibiting the production of angiogenic factors as well as the ability of bone marrow-derived angiogenic cells to contribute to vascular responses (65, 66).

Regulation of O₂ Consumption

Metabolic re-programming.

The erythropoietic and angiogenic responses to hypoxia do not increase O₂ delivery immediately, since it may take days to weeks to increase red blood cell counts or generate new blood vessels that are sufficient to improve tissue oxygenation. In contrast, responses designed to reduce O₂ consumption occur over the course of hours to days. The vast majority of O₂ is consumed by the mitochondria during oxidative phosphorylation, which utilizes reducing equivalents generated during the oxidation of acetyl coenzyme A (AcCoA) to form an electrochemical gradient that is used to synthesize ATP. The electron transport chain (ETC), which is responsible for forming the gradient, is designed to operate most efficiently at physiological Po₂, and deviations above or below result in premature loss of electrons, leading to reactive oxygen species formation. Thus the principal metabolic response to hypoxia is a switch from oxidative to glycolytic metabolism.

HIF-1 controls the expression of multiple gene products that mediate the metabolic switch. First, HIF-1 activates transcription of the PDK1 gene, encoding pyruvate dehydrogenase (PDH) kinase 1, which phosphorylates and inactivates the catalytic subunit of PDH, the enzyme that converts glucose-derived pyruvate into AcCoA for entry into the mitochondrial tricarboxylic acid (TCA) cycle (33, 54). Second, HIF-1 transactivates the LDHA gene, which encodes lactate dehydrogenase, the enzyme that converts pyruvate to lactate (74). Third, HIF-1 mediates a subunit switch in cytochrome c oxidase (ETC complex IV) by transactivation of the LONP gene, which encodes a protease that degrades COX4-1 and the COX4I2 gene, which encodes the alternative COX4-2 subunit (17). Under hypoxic conditions, yeast cells undergo a homologous subunit switch, which serves to increase the efficiency of electron transfer (38). However, yeast lack HIF-1, and the subunit switch occurs by a different molecular mechanism, suggesting that subunit switching arose independently but represents a universal response of unicellular and multicellular eukaryotes to hypoxia. Fourth, HIF-1 mediates the suppression of acyl CoA dehydrogenases that generate AcCoA by oxidation of fatty acids (96). Fifth, HIF-1 mediates the expression of miR-210, a micro-RNA that inhibits the expression of ISCU1 and ISCU2, which are iron-sulfur complex assembly factors that are required for activity of the TCA enzyme aconitase and ETC complex I (7). Sixth, HIF-1 activates expression of BNIP3 (96) and BNIP3L (2), which are mitochondrial proteins that trigger mitochondrial-selective autophagy, thereby reducing cellular oxidation of both glucose and fatty acids. Thus multiple mechanisms are available to modulate metabolism, with different tissues using different strategies to maintain redox homeostasis during hypoxia (17).

Cell cycle arrest.

Just as more mitochondria within a single cell consume more O₂, more cells within a tissue also consume more O₂. Thus a fundamental physiological response to hypoxia is cell cycle arrest. HIF-1α is necessary for hypoxia-induced cell cycle arrest (19, 28, 29, 35, 84), and forced expression of HIF-1α is sufficient to induce G₁-phase cell cycle arrest (21) due to inhibition of Myc activity (35) and direct interaction of HIF-1α with protein components of the pre-replication complex that blocks origin firing and DNA replication (28, 29).

Some cell types, such as endothelial cells stimulated by hypoxia-induced angiogenic growth factors, must proliferate under hypoxic conditions. But to do so, they must counteract the inhibitory effect of HIF-1α on DNA replication. The cyclin-dependent kinase CDK2, which is active from late G₁ through S phase and G₂, binds to HIF-1α and triggers lysosomal degradation of the protein by chaperone-mediated autophagy, whereas CDK1, which is active from late G₂ through M phase, binds to HIF-1α and protects it from lysosomal degradation (28). Thus the cell cycle-specific regulation of HIF-1α by CDK1 and CDK2 allows HIF-1α to perform its role (with HIF-1β) as a transcription factor to mediate adaptive responses to hypoxia while preventing HIF-1α from performing its non-transcriptional role as an inhibitor of DNA replication.

Perspectives

O₂ sensing by the carotid body and the chemosensory reflex play a critical role in regulation of breathing and blood pressure in various physiological conditions, including exercise, high altitude, and pregnancy, to name a few (37). How these physiological situations affect the gas-messenger signaling in the carotid body and the underlying mechanisms remains to be studied. Enhanced carotid body sensitivity to hypoxia has been implicated in a variety of cardiovascular diseases with increased sympathetic nerve activity, including sleep apnea and neurogenic hypertension (37, 59). Recent studies from Schultz and coworkers (68) have shown that disrupted gaseous messenger signaling also mediates carotid body hyperactivity, which contributes to increased sympathetic activation in an experimental model of congestive heart failure. Surgical ablation of the carotid body was proposed as a potential therapeutic intervention for reducing blood pressure in patients with essential hypertension (55). However, carotid body ablation would abolish adaptive systemic cardiovascular and ventilatory responses to hypoxia (57). Pharmacological manipulation of the gaseous messenger-signaling pathway may represent a safer and more effective approach for treating carotid body-related morbidity.

In addition to the essential involvement of HIFs in development and physiology, in many common disease states, HIFs play either protective roles (anemia, inflammatory bowel disease, ischemic cardiovascular disease) or pathogenic roles (cancer, ocular neovascularization, pulmonary hypertension, sleep apnea, transplant rejection) (18, 31, 63, 71, 78). In addition, the ventilatory acclimatization to high-altitude hypoxia, which is mediated by the carotid body, and responses of the isolated carotid body to acute hypoxia are also dependent on HIF-1α expression (34), providing a direct and unifying link between the systemic and molecular O₂-sensing and response pathways.