Comp. Biochem. Physiol. Vol. 118A, No. 1, pp. 9–14, 1997 ISSN 0300-9629/97/$17.00Copyright 1997 Elsevier Science Inc. PII S0300-9629(96)00370-2

Oxygen Sensors in the Organism: Examples ofRegulation Under Altitude Hypoxia in Mammals

J.- P. RichaletLaboratoire de Physiologie, Association pour la Recherche

en Physiologie de l’Environnement, U.F.R. Medecine, 93012 Bobigny, France

ABSTRACT. Oxygen sensing is a determinant function of mammals, especially humans, to maintain theiractivity under acute or chronic exposure to hypoxia. True O2 sensors (chemoreceptors, erythropoietin secretingcells) are involved in regulation loops, which aim to restore O2 availability to the cells. Pseudo O2 sensors arecells activated by the lack of oxygen but not clearly involved in regulation processes. Potassium channels inthe carotid bodies have been suspected to be O2 sensitive and could mediate the chemosensitivity to hypoxia.Na,K-ATPase related ion transport in alveolar pneumocytes could be sensitive to O2 availability and regulatethe flux of water and sodium in the alveolar space. Signal transduction in G-protein–dependent receptor systemsis modified in hypoxia, such as in cardiac β-receptors and adenosinergic and muscarinic receptors. Recent studieshave provided some evidence to the possible role of hypoxia inducible factors (HIF-1) in the regulation ofprotein synthesis at the transcriptional level. Similarities between O2-sensing mechanisms in erythropoiesis andin the synthesis of vascular endothelial growth factor were recently evidenced. Both genes are upregulated inhypoxia. However, the precise structure (heme-linked enzyme?) of all these O2-sensitive sites is not known,either in the erythropoietic system or in the chemoreceptor function. An adequate balance between hypoxia-induced upregulation and downregulation processes is necessary for optimal survival in a hypoxic environment.comp biochem physiol 118A;1:9–14, 1997. 1997 Elsevier Science Inc.

KEY WORDS. Hypoxia, altitude, oxygen sensing, humans, rats, erythropoiesis, pneumocytes, downregulation,beta-receptors

INTRODUCTION this article, which is not designed to be an exhaustive re-view on oxygen-dependent regulation mechanisms. Other

All cells within aerobic organisms, and especially mammals,important systems, such as heat shock proteins in hypoxic

are concerned with oxygen availability and will be affectedconditioning or potassium channels in hypoxic pulmonary

by oxygen lack. Aerobic and anaerobic metabolism will bevasoconstriction, will not be addressed here.

modified under acute and chronic hypoxia. As a result, inhumans, maximal O2 consumption is progressively reducedat high altitude and is not restored by prolonged exposure

O2 SENSORS AND ‘‘PSEUDO O2 SENSORS’’and acclimatization to chronic hypoxia (10). Hypoxia willalso have a specific effect on certain cells called ‘‘O2 sen- It is important to differentiate O2 sensors involved in a regu-

lation loop concerning O2 availability from ‘‘pseudo O2 sen-sors.’’ These cells are capable of a specific response to hyp-oxic stress. Studies have been developed in our laboratory sors’’ not involved in such a regulation loop (Fig. 1). To

date, two ‘‘true’’ O2 sensors are known: chemoreceptors in-on the physiological processes of acclimatization to altitudehypoxia, especially in humans. Some of the results with re- volved in the regulation of ventilation and circulation, and

erythropoietin secreting cells. These two sensors are at onespect to O2 sensing and preservation of life and physicalactivity in spite of reduced O2 availability will be presented end of regulation loops, the activation of which is aimed to

increasing O2 availability to the cells. Hyperventilationhere.Examples of oxygen-sensing systems will be described in leads to decreased arterial CO2 pressure and concomitant

increase in arterial O2 pressure. Increased adrenergic activ-This paper was presented at the European Society of Comparative Physiol- ity leads to an increase in cardiac output. The stimulationogy and Biochemistry Symposium on ‘‘Life in Extreme Environments’’ un-

of erythropoiesis leads to an increase in hemoglobin and O2der the theme of ‘‘Hypoxia,’’ held in June 1995 at La Seyne-sur-Mer,France. availability at the peripheral capillary level (12). Thus,

Address reprint requests to: Dr. J.- P. Richalet, Laboratoire de Physiologie, these two systems are clearly dedicated to restoring O2 avail-Association pour la Recherche en Physiologle de l’Environnement, U.F.R.

ability in the face of hypoxic stress.Medecine, 93012 Bobigny, France. Fax 33-1-4838-7777.Received 14 December 1995; accepted 6 September 1996. Numerous other systems are sensitive to O2 lack but ap-

10 J.- P. Richalet

FIG. 1. Major systems involved in O2 sensing. True and pseudo O2 sensors.

parently are not involved in regulation loops directed to maintaining arterial O2 content. The increase in adrenergicactivity tends to restore arterial O2 transport. The formationrestoring O2 availability. Smooth muscle cells are relaxed

(peripheral vessels) or constricted (pulmonary vessels) of new vessels maintains tissue O2 pressure. Vasodilatationin some organs contributes to maintaining tissue PO2,when exposed to hypoxia. Endothelial cells release media-

tors following a hypoxic stress; these mediators will induce whereas vasoconstriction of pulmonary vessels diverts bloodfrom nonoxygenated areas. Here we present some exampleschanges in vasomotricity, hemostasis, cell proliferation, and

capillary permeability. Hormonal release by adrenal cortex of the effect of hypoxia on the organism through varioustypes of O2 sensing.and kidney is modified by hypoxia. Hypoxia induces a

downregulation of β-receptors (5) and probably of other re-ceptors where G-proteins are involved in the regulation of

CAROTID BODIES AND HYPOXEMIAsignal transduction. Pneumocyte PII regulation function ofalveolar hypophase is modified by hypoxia (9). All these The first example of O2 sensing is sensory transduction in

the carotid body. The different steps of signal transductionactions induced by hypoxia do not have a clear regulatorypurpose directed toward restoration of O2 availability. Thus, in the carotid body include: (1) oxygen detection, (2) Na1

and Ca21 action potentials resulting in Ca21 influx, (3) risethey are called ‘‘pseudo O2 sensors.’’ However, this classifi-cation may be transitory until a regulation mechanism in- in cytosolic Ca21, (4) release of transmitter, and (5) increase

of firing in afferent fibers (8). An O2-sensitive sensor couldvolving one of these pseudo sensors is revealed.Strategies of mammals to cope with hypoxia are deter- be coupled to a potassium channel following various possi-

ble modalities: (1) the channel and the sensor are the samemined by the specific action of the hypoxic stimulus on vari-ous systems. The stimulation of chemoreceptors and its ac- entity, and O2 could act directly on the tetrameric structure

of the channel; (2) the sensor sends a signal to the channeltion on ventilation contribute to maintaining arterial O2

pressure. The activation of erythropoiesis contributes to via a mediator; (3) the sensor is separate but coupled to the

Oxygen Sensors 11

channel via a direct interaction, without any mediator (2). reabsorption tended to decrease in hypoxia; however, somesubjects showed an increase in proximal reabsorption.The O2-sensitive potassium channel represents the first

known example of an ion channel directly regulated by O2. Whatever the sense of variation with altitude exposure, thisparameter was well related to EPO production (Fig. 3). Thisclose relationship is in favor of the presence, inside or in

PNEUMOCYTES AND ALVEOLAR HYPOXIA the neighborhood of the proximal tubule, of O2-sensitivecells, the stimulation of which determines EPO production.There are other examples of ionic pumps, the function ofThe rate of EPO production depends on the balance be-which is oxygen-dependent via an Na,K-ATPase. Thetween O2 supply (blood O2 content and renal blood flow)pneumocyte PII, which lines the alveolar wall, has beenand O2 demand (O2 consumption in the tubular cells) inshown recently to be sensitive to ambient hypoxia. The so-the kidneys, which determines PO2

in the EPO producingdium transport by alveolar epithelium represents an impor-cells. The last factor is directly determined by the intensitytant mechanism for airspace fluid clearance in acute lungof proximal tubular reabsorption, which could therefore beinjury such as high-altitude pulmonary edema. The effectan important stimulus for EPO production (1). Kidney func-of hypoxia on Na,K-ATPase activity was studied in alveolartion may play an important role in modulating one of theepithelial cells (AEC). SV40 virus transformed rat type IImajor components of acclimatization of humans to a hyp-AEC were cultured on collagen-coated plastic dishes andoxic environment. A normal acclimatization process wouldexposed to either hypoxic (5% O2) (H) or normoxic (21%include a brisk bone marrow response to EPO, followed byO2) (N) for increasing times (up to 48 hr) in the absencean increase in arterial O2 content and a reduction in tissueor presence of 1025 M nifedipine. Na,K-ATPase activityhypoxia. Occurrence of acute mountain sickness (AMS)was determined using ouabain-sensitive 86rubidium influxhas been linked to a low ventilatory response to hypoxia,(OsRb). Exposure to H induced a time-dependent decreaseprobably responsible for a lower tissue PO2

. Clinical signs ofof OsRb. Incubation of N cells with supernatant of H cellsgood acclimatization would follow the restoration of tissueresulted in a 45% decrease of OsRb within 1 hr (9). Nifedi-oxygenation. As EPO is mainly triggered by tissue hypoxia,pine prevented hypoxia-induced decrease of OsRb (9).one would expect a parallel decrease in EPO and AMS.These results indicate that: (1) hypoxia induces a time-de-However, in subjects with subchronic AMS, EPO wouldpendent decrease of Na,K-ATPase activity in alveolar typeremain elevated, an index of chronic tissue hypoxia. Thus,II cells; (2) this effect is most likely due to the release of apersistent high EPO can be considered as a good marker ofsoluble factor; and (3) it is prevented by nifedipine. Alter-uncompleted acclimatization to high altitude (12).ation of AEC Na,K-ATPase activity during hypoxia may

From the molecular side, the link between O2 sensing andreduce airspace fluid clearance and contribute to the forma-EPO mRNA accumulation is not via cAMP, cGMP, proteintion and/or maintenance of alveolar edema. The altitude-kinase C, and a Ca-dependent reaction. Furthermore, a crit-induced alveolar edema may thus result not only from pul-ical fall in ATP levels fails to stimulate EPO production.monary arterial hypertension inducing interstitial edema,The only pharmacological compound that can mimic thebut also from weakness of the alveolar wall (11).effect of hypoxia on EPO formation in vivo is cobalt. A hemeprotein has been hypothesized to be involved in the O2-ERYTHROPOIESIS AND sensing process (3).

ERYTHROPOIETIN REGULATION

Erythropoiesis is another example of O2 sensing, the regula-DIFFERENTIAL REGULATIONtion of which could be summarized as in Fig. 2. Factors ofOF CARDIAC RECEPTORS IN HYPOXIA

erythropoiesis were studied in humans during a 3-week ex-posure to severe altitude hypoxia (6542 m). Erythropoietin Prolonged exposure to high-altitude hypoxia induces a de-

crease in maximal exercise heart rate and chronotropicconcentration increased from 3- to 134-fold after 1 week ofexposure to hypoxia, then decreased but remained higher response to endogenous (exercise) and exogenous (isopro-

terenol) adrenergic activation (13). Desensitization of β-than its sea-level value. The main determinants of EPO pro-duction were arterial O2 content, renal O2 demand and nu- receptors (β-AR) could be one of the mechanisms involved.

A hypoxia-induced decrease in density of β-AR was foundtritional factors. Renal blood flow decreased during the stayat 6542m, but it was not correlated to serum EPO concen- in human lymphocytes and in rat myocardium. Chronic

hypoxia also induced, in the left ventricle (LV) of rats, atration. In fact, renal blood flow is not a determinant factorof EPO secretion. When renal blood flow is lowered, the downregulation of adenosine receptors and an upregulation

of muscarinic receptors (6) (Fig. 4). All these effects couldsodium load to the proximal tubule and the reabsorptionrate decrease. Because the reabsorption of sodium is the be linked to a modification in the synthesis and/or activity

of Gs and Gi2 proteins. We examined the effect on genemain energy-consuming transport process within the renaltubule, a reduction in renal blood flow may finally result in a expression and level and function of G proteins in the heart

of rats exposed for 30 days to hypobaric hypoxia (HX,reduction in oxygen demand of the kidney. Mean proximal

12 J.- P. Richalet

FIG. 2. Factors involved in the regulation of erythropoietin secretion.

380 mmHg) compared to normoxic rats (N) (5). Myocar-dial Gs and Gi2 mRNAs were identified by Northern blotand quantified by slot blot hybridizations. In right ventricles(RVs) that were hypertrophied by hypoxia-induced pulmo-nary hypertension, mRNA levels of Gi2 were increased by40% (P , 0.05) without changes in LVs. No change wasobserved in Gs mRNA levels of both RV and LV. Func-tional activity of Gs, measured by reconstitution usingmouse lymphoma S49 cyc-cells, deficient in Gs, was signifi-cantly decreased by 18% in HX LV and by 21% in HX RV.In conclusion, gene expression, protein level and activityof Gs and Gi2 do not show parallel variations in HX. Desen-sitization in HX appears to be related to decreased func-tional activity of Gs in spite of normal transcriptional activ-ity. In HX RV, although the mRNA level of Gi2 increased,the protein level was unchanged. Thus, downregulation ofβ-adrenergic and adenosinergic receptors and upregulationof muscarinic receptors could be explained by O2-sensedmodifications in a G-protein signal transduction system.The overall effect of these modifications is a reduction inmaximal chronotropic function and a limitation of O2 de-

FIG. 3. Relation between proximal tubular renal reabsorp- mand in the myocardium exposed to chronic decrease intion (APRH/APRN 5 relative increase in APR after 2 weeks

O2 availability.at 6542 m) and EPO (EPOH/EPON 5 relative increase inEPO after 2 weeks at 6542 m) in nine sea-level natives.[From (12).]

PARATHORMONE ANDCALCIUM REGULATION IN HYPOXIA

Another possible example of hypoxia-induced desensitiza-tion process in which the G-proteins could be involved is

Oxygen Sensors 13

FIG. 4. Differential regulationof cardiac receptors (left ven-tricles of rats exposed for 3weeks to 0.5 atm) in chronichypoxia. bAR 5 b-receptors;M2 5 muscarinic receptors,A1 5 adenosinergic receptors.[From (6).]

the role of parathormone in calcium regulation mecha- O2-related alterations in G-protein expression or activity(15).nisms. The effects of altitude hypoxia on bone markers, se-

rum and urinary parameters of calcium metabolism, and thebiological response to a 200 U 1-34 PTH injection were

A COMMON CONTROL SYSTEMstudied in men during a 5-day stay at 4350 m (15). TheOF HYPOXIA-SENSITIVE MECHANISMS?

magnitude of the response to exogenous PTH was similarin normoxia and hypoxia for serum Ca21, PO4, calcitriol, Hypoxia has been shown to induce an increased expression

of several genes (4), coding for EPO, platelet-derivedand PO4 reabsorption rate but significantly lower in hypoxiafor urinary cAMP, which is a marker of renal PTH receptor growth factor, endothelin (7), interleukin-1α, ornithine de-

carboxylase and vascular endothelial growth factor (VEGF)stimulation. Thus, in these conditions of hypoxia, a relativeresistance to PTH occurred, which could be mediated by (14).

FIG. 5. Adequate acclimatization to altitude hypoxia is determined by a good balance between activated and resistance states.

14 J.- P. Richalet

2. Ganfornina, M.D.; Lopez-Barneo, J. Potassium channel typesIn fact, there are some similarities between the O2-sens-in arterial chemoreceptor cells and their selective modulationing mechanisms regulating the expression of VEGF andby oxygen. J. Gen. Physiol. 100:401–426;1992.

EPO (4): (1) the expression of both mRNAs is upregulated 3. Goldberg, M.A.; Dunning, S.P.; Bunn, H.F. Regulation of theby hypoxia and cobalt chloride; (2) half life of both mRNAs erythropoietin gene: Evidence that the oxygen sensor is a

heme protein. Science 242:1412–1415;1988.is markedly prolonged by cycloheximide (protein synthesis4. Goldberg, M.A.; Schneider, T.J. Similarities between the oxy-inhibitor); and (3) hypoxic induction of both is inhibited

gen-sensing mechanisms regulating the expression of vascularby CO. Is it possible to suppose from these observations thatendothelial growth factor and erythropoietin. J. Biol. Chem.

hypoxia-responsive genes share a common regulatory 269:4355–4359;1994.pathway? 5. Kacimi, R.; Moalic, J.M.; Aldashev, A.; Vatner, D.E.; Richa-

let, J.-P.; Crozatier, B. Differential regulation of G protein ex-Then, two questions arise from this: (1) Is there a wide-pression in rat hearts exposed to chronic hypoxia. Am. J.spread O2-signalling mechanism that may be triggered inPhysiol. 269(Heart Circ. Physiol. 38):H1865–H1873;1995.hypoxia, other than solely controlling EPO? (2) Are genes

6. Kacimi, R.; Richalet, J.P.; Crozatier, B. Hypoxia-induced dif-known to be regulated by O2 availability, under a common ferential modulation of adenosinergic and muscarinic recep-control system where hypoxia-responsible DNA binding tors in rat heart. J. Appl. Physiol. 75:1123–1128;1993.

7. Kourembanas, S.; Marsden, P.A.; McQuillan, L.P.; Faller,proteins (HIF-1) may play a major role?D.V. J. Clin. Invest. 88:1054–1057;1991.Hypoxia inducible factor (HIF-1) is a nuclear factor

8. Lopez-Barneo, J.; Benot, A.R.; Urena, J. Oxygen sensing andwhose DNA binding activity is induced by hypoxia. It bindsthe electrophysiology of arterial chemoreceptor cells. N.I.P.S.

to an enhancer fragment on the human EPO gene. Many 8:191–195;1993.mammalian cell types, not only EPO-producing cells, can 9. Planes, C.; Richalet, J.-P.; Friedlander, G.; Amiel, C.; Clerici,

C. Nifedipine prevents hypoxia-induced decrease of Na,K-sense O2 tension and respond to hypoxia by changes in geneATPase activity in alveolar type II cells. In: Sutton, J.R.;expression. The mechanism of this response could includeHouston, C.S.; Coates, G. (eds). Hypoxia and the Brain: Pro-the activation of HIF-1 (16).ceedings of the Ninth Hypoxia Symposium. Lake Louise: Can-ada; 1995:335.

10. Pugh, L.G.C.E.; Gill, M.B.; Lahiri, S.; Milledge, J.S.; Ward,CONCLUSIONSM.P.; West, J.B. Muscular exercise at great altitudes. J. Appl.Physiol. 19:431–440;1964.Hypoxia triggers upregulation processes resulting in the ac-

11. Richalet, J.-P. High-altitude pulmonary oedema: Still a placetivation of several systems such as adrenergic activity, eryth-for controversy? (Editorial). Thorax 50:923–929;1995.

ropoiesis and carotid receptors. The activation of these hor- 12. Richalet, J.-P.; Souberbielle, J.-C.; Antezana, A.-M.; De-mones and reflex loops results in a reaction against hypoxia, chaux, M.; Le Trong, J.-L.; Bienvenu, A.; Daniel, F.; Blan-

chot, C.; Zittoun, J. Control of erythropoiesis in humans dur-realizing an ‘‘activated state.’’ On the other side, hypoxiaing prolonged exposure to the altitude of 6542 m. Am. J.triggers downregulation processes such as the desensitiza-Physiol. 266(Reg. Int. Comp. Phys. 35):R756–R764;1994.tion of β-adrenergic, adenosinergic or parathormone recep-

13. Richalet, J.-P.; Le Trong, J.-L.; Rathat, C.; Merlet, P.; Bouis-tors, which may protect the organism against the effects of sou, P.; Keromes, A.; Veyrac, P. Reversal of hypoxia-inducedan overall hyperactivation, leading to a ‘‘resistance state.’’ decrease in human cardiac response to isoproterenol infusion.

J. Appl. Physiol. 67:523–527;1989.The quality of survival in the extreme environment of alti-14. Shweiki, D.; Itin, A.; Soffer, D.; Keshet, E. Vascular endothe-tude hypoxia depends on the balance between these activa-

lial growth factor induced by hypoxia may mediate hypoxia-tion and resistance processes induced by hypoxia (Fig. 5).initiated angiogenesis. Nature 359:843–845;1992.

Hypoxia-inducible factors could be the common pathway 15. Souberbielle, J.-C.; Richalet, J.-P.; Garabedian, M.; Sachs, C.;in the regulation of genes involved in O2 transport and con- Dechaux, M. High altitude hypoxia and calcium metabolism.

In: Sutton, J.R.; Houston, C.S.; Coates, G. (eds). Hypoxia andsumption.the Brain: Proceedings of the Ninth Hypoxia Symposium.Lake Louise: Canada; 1995:336.

16. Wang, G.L.; Semanza, G.L. General involvement of hypoxia-References1. Eckardt, K.-U.; Kurtz, A.; Bauer, C. Regulation of erythropoi- inducible factor 1 in transcriptional response to hypoxia. Proc.

Natl. Acad. Sci. USA 90:4304–4308;1993.etin production is related to proximal tubular function. Am. J.Physiol. 256(Renal Fluid Electrolyte Physiol. 25):F942–F947;1989.