PHD2 inactivation in Type I cells drives HIF-2α-dependent multilineage hyperplasia and the formation of paraganglioma-like carotid bodies.

KEY POINTS: The carotid body is a peripheral arterial chemoreceptor that regulates ventilation in response to both acute and sustained hypoxia. Type I cells in this organ respond to low oxygen both acutely by depolarization and dense core vesicle secretion and, over the longer term, via cellular proliferation and enhanced ventilatory responses. Using lineage analysis, the present study shows that the Type I cell lineage itself proliferates and expands in response to sustained hypoxia. Inactivation of HIF-2α in Type I cells impairs the ventilatory, proliferative and cell intrinsic (dense core vesicle) responses to hypoxia. Inactivation of PHD2 in Type I cells induces multilineage hyperplasia and ultrastructural changes in dense core vesicles to form paraganglioma-like carotid bodies. These changes, similar to those observed in hypoxia, are dependent on HIF-2α. Taken together, these findings demonstrate a key role for the PHD2-HIF-2α couple in Type I cells with respect to the oxygen sensing functions of the carotid body. ABSTRACT: The carotid body is a peripheral chemoreceptor that plays a central role in mammalian oxygen homeostasis. In response to sustained hypoxia, it manifests a rapid cellular proliferation and an associated increase in responsiveness to hypoxia. Understanding the cellular and molecular mechanisms underlying these processes is of interest both to specialized chemoreceptive functions of that organ and, potentially, to the general physiology and pathophysiology of cellular hypoxia. We have combined cell lineage tracing technology and conditionally inactivated alleles in recombinant mice to examine the role of components of the HIF hydroxylase pathway in specific cell types within the carotid body. We show that exposure to sustained hypoxia (10% oxygen) drives rapid expansion of the Type I, tyrosine hydroxylase expressing cell lineage, with little transdifferentiation to (or from) that lineage. Inactivation of a specific HIF isoform, HIF-2α, in the Type I cells was associated with a greatly reduced proliferation of Type I cells and hypoxic ventilatory responses, with ultrastructural evidence of an abnormality in the action of hypoxia on dense core secretory vesicles. We also show that inactivation of the principal HIF prolyl hydroxylase PHD2 within the Type I cell lineage is sufficient to cause multilineage expansion of the carotid body, with characteristics resembling paragangliomas. These morphological changes were dependent on the integrity of HIF-2α. These findings implicate specific components of the HIF hydroxylase pathway (PHD2 and HIF-2α) within Type I cells of the carotid body with respect to the oxygen sensing and adaptive functions of that organ.

Regulation of ventilatory sensitivity and carotid body proliferation in hypoxia by the PHD2/HIF-2 pathway.

Ventilatory sensitivity to hypoxia increases in response to continued hypoxic exposure as part of acute acclimatisation. Although this process is incompletely understood, insights have been gained through studies of the hypoxia-inducible factor (HIF) hydroxylase system. Genetic studies implicate these pathways widely in the integrated physiology of hypoxia, through effects on developmental or adaptive processes. In keeping with this, mice that are heterozygous for the principal HIF prolyl hydroxylase, PHD2, show enhanced ventilatory sensitivity to hypoxia and carotid body hyperplasia. Here we have sought to understand this process better through comparative analysis of inducible and constitutive inactivation of PHD2 and its principal targets HIF-1α and HIF-2α. We demonstrate that general inducible inactivation of PHD2 in tamoxifen-treated Phd2(f/f);Rosa26(+/CreERT2) mice, like constitutive, heterozygous PHD2 deficiency, enhances hypoxic ventilatory responses (HVRs: 7.2 ± 0.6 vs. 4.4 ± 0.4 ml min(-1) g(-1) in controls, P < 0.01). The ventilatory phenotypes associated with both inducible and constitutive inactivation of PHD2 were strongly compensated for by concomitant inactivation of HIF-2α, but not HIF-1α. Furthermore, inducible inactivation of HIF-2α strikingly impaired ventilatory acclimatisation to chronic hypoxia (HVRs: 4.1 ± 0.5 vs. 8.6 ± 0.5 ml min(-1) g(-1) in controls, P < 0.0001), as well as carotid body cell proliferation (400 ± 81 vs. 2630 ± 390 bromodeoxyuridine-positive cells mm(-2) in controls, P < 0.0001). The findings demonstrate the importance of the PHD2/HIF-2α enzyme-substrate couple in modulating ventilatory sensitivity to hypoxia.

Carotid body hyperplasia and enhanced ventilatory responses to hypoxia in mice with heterozygous deficiency of PHD2.

Oxygen-dependent prolyl hydroxylation of hypoxia-inducible factor (HIF) by a set of closely related prolyl hydroxylase domain enzymes (PHD1, 2 and 3) regulates a range of transcriptional responses to hypoxia. This raises important questions about the role of these oxygen-sensing enzymes in integrative physiology. We investigated the effect of both genetic deficiency and pharmacological inhibition on the change in ventilation in response to acute hypoxic stimulation in mice. Mice exposed to chronic hypoxia for 7 days manifest an exaggerated hypoxic ventilatory response (HVR) (10.8 ± 0.3 versus 4.1 ± 0.7 ml min(-1) g(-1) in controls; P < 0.01). HVR was similarly exaggerated in PHD2(+/-) animals compared to littermate controls (8.4 ± 0.7 versus 5.0 ± 0.8 ml min(-1) g(-1); P < 0.01). Carotid body volume increased (0.0025 ± 0.00017 in PHD2(+/-) animals versus 0.0015 ± 0.00019 mm(3) in controls; P < 0.01). In contrast, HVR in PHD1(-/-) and PHD3(-/-) mice was similar to littermate controls. Acute exposure to a small molecule PHD inhibitor (PHI) (2-(1-chloro-4-hydroxyisoquinoline-3-carboxamido) acetic acid) did not mimic the ventilatory response to hypoxia. Further, 7 day administration of the PHI induced only modest increases in HVR and carotid body cell proliferation, despite marked stimulation of erythropoiesis. This was in contrast with chronic hypoxia, which elicited both exaggerated HVR and cellular proliferation. The findings demonstrate that PHD enzymes modulate ventilatory sensitivity to hypoxia and identify PHD2 as the most important enzyme in this response. They also reveal differences between genetic inactivation of PHDs, responses to hypoxia and responses to a pharmacological inhibitor, demonstrating the need for caution in predicting the effects of therapeutic modulation of the HIF hydroxylase system on different physiological responses.

Developmental role of PHD2 in the pathogenesis of pseudohypoxic pheochromocytoma.

Despite a general role for the HIF hydroxylase system in cellular oxygen sensing and tumour hypoxia, cancer-associated mutations of genes in this pathway, including PHD2, PHD1, EPAS1 (encoding HIF-2α) are highly tissue-restricted, being observed in pseudohypoxic pheochromocytoma and paraganglioma (PPGL) but rarely, if ever, in other tumours. In an effort to understand that paradox and gain insights into the pathogenesis of pseudohypoxic PPGL, we constructed mice in which the principal HIF prolyl hydroxylase, Phd2, is inactivated in the adrenal medulla using TH-restricted Cre recombinase. Investigation of these animals revealed a gene expression pattern closely mimicking that of pseudohypoxic PPGL. Spatially resolved analyses demonstrated a binary distribution of two contrasting patterns of gene expression among adrenal medullary cells. Phd2 inactivation resulted in a marked shift in this distribution towards a Pnmt-/Hif-2α+/Rgs5+ population. This was associated with morphological abnormalities of adrenal development, including ectopic TH+ cells within the adrenal cortex and external to the adrenal gland. These changes were ablated by combined inactivation of Phd2 with Hif-2α, but not Hif-1α. However, they could not be reproduced by inactivation of Phd2 in adult life, suggesting that they arise from dysregulation of this pathway during adrenal development. Together with the clinical observation that pseudohypoxic PPGL manifests remarkably high heritability, our findings suggest that this type of tumour likely arises from dysregulation of a tissue-restricted action of the PHD2/HIF-2α pathway affecting adrenal development in early life and provides a model for the study of the relevant processes.

Marked and rapid effects of pharmacological HIF-2α antagonism on hypoxic ventilatory control.

Hypoxia-inducible factor (HIF) is strikingly upregulated in many types of cancer, and there is great interest in applying inhibitors of HIF as anticancer therapeutics. The most advanced of these are small molecules that target the HIF-2 isoform through binding the PAS-B domain of HIF-2α. These molecules are undergoing clinical trials with promising results in renal and other cancers where HIF-2 is considered to be driving growth. Nevertheless, a central question remains as to whether such inhibitors affect physiological responses to hypoxia at relevant doses. Here, we show that pharmacological HIF-2α inhibition with PT2385, at doses similar to those reported to inhibit tumor growth, rapidly impaired ventilatory responses to hypoxia, abrogating both ventilatory acclimatization and carotid body cell proliferative responses to sustained hypoxia. Mice carrying a HIF-2α PAS-B S305M mutation that disrupts PT2385 binding, but not dimerization with HIF-1ß, did not respond to PT2385, indicating that these effects are on-target. Furthermore, the finding of a hypomorphic ventilatory phenotype in untreated HIF-2α S305M mutant mice suggests a function for the HIF-2α PAS-B domain beyond heterodimerization with HIF-1ß. Although PT2385 was well tolerated, the findings indicate the need for caution in patients who are dependent on hypoxic ventilatory drive.