the ventilatory response to hypoxia in mammals

The Ventilatory Response to Hypoxia in Mammals:Mechanisms, Measurement, and Analysis

LUC J. TEPPEMA AND ALBERT DAHAN

Department of Anesthesiology, Leiden University Medical Center, Leiden, The Netherlands

I. Introduction 676II. The Hypoxic Ventilatory Response in Fetal and Neonatal Mammals 677

A. Fetal breathing movements 677B. Hypoxia inhibits fetal breathing movements 677C. Maturation of the HVR in neonatal animals 679D. Developmental plasticity of the HVR 681E. The HVR in term and preterm babies 684

III. Peripheral and Central Mechanisms Mediating the Hypoxic Ventilatory Response in Adult Mammals 685A. Carotid body mechanisms in oxygen sensing 685B. Hypoxia activates the transcription factor hypoxia-inducible factor-1 690C. Central effects of hypoxia 692D. Hypoxic ventilatory decline in the adult mammal 698

IV. Genetic Influence on the Ventilatory Response to Hypoxia 703A. The genome and HVR 703B. Mutations and HVR 704

V. Pathophysiological Influences on the Hypoxic Ventilatory Response 705A. Biological variability 705B. Ageing 706C. Gender and pregnancy 706D. Influence of CO2 on the HVR: O2-CO2 interaction 707E. Carotid body resection and the HVR 710F. Chronic hypoxia and the HVR 711G. Chronic intermittent hypoxia and the HVR 720

VI. Measuring, Expressing, and Modeling the Hypoxic Ventilatory Response 725A. Measuring the HVR 725B. Expressing the steady-state HVR 731C. Modeling the HVR 731

VII. Comments and Future Challenges 732

Teppema LJ, Dahan A. The Ventilatory Response to Hypoxia in Mammals: Mechanisms, Measurement, andAnalysis. Physiol Rev 90: 675–754, 2010; doi:10.1152/physrev.00012.2009.—The respiratory response to hypoxiain mammals develops from an inhibition of breathing movements in utero into a sustained increase inventilation in the adult. This ventilatory response to hypoxia (HVR) in mammals is the subject of this review.The period immediately after birth contains a critical time window in which environmental factors can causelong-term changes in the structural and functional properties of the respiratory system, resulting in an alteredHVR phenotype. Both neonatal chronic and chronic intermittent hypoxia, but also chronic hyperoxia, caninduce such plastic changes, the nature of which depends on the time pattern and duration of the exposure(acute or chronic, episodic or not, etc.). At adult age, exposure to chronic hypoxic paradigms inducesadjustments in the HVR that seem reversible when the respiratory system is fully matured. These changes areorchestrated by transcription factors of which hypoxia-inducible factor 1 has been identified as the masterregulator. We discuss the mechanisms underlying the HVR and its adaptations to chronic changes in ambientoxygen concentration, with emphasis on the carotid bodies that contain oxygen sensors and initiate theresponse, and on the contribution of central neurotransmitters and brain stem regions. We also brieflysummarize the techniques used in small animals and in humans to measure the HVR and discuss the specificdifficulties encountered in its measurement and analysis.

Physiol Rev 90: 675–754, 2010;doi:10.1152/physrev.00012.2009.

www.prv.org 6750031-9333/10 $18.00 Copyright © 2010 the American Physiological Society

I. INTRODUCTION

The independent discovery of oxygen by the Swedishapothecary Carl Wilhelm Sheele in 1772 and the Englishscientist Joseph Priestley in 1774 is considered one of thegreatest scientific discoveries (for a historical overview,see Ref. 698). To date, 235 years later, it is a well estab-lished fact that aerobic metabolism in mammals relies onsufficient oxygen availability in the mitochondria whereits reduction to water is coupled to the generation of ATPmolecules. To be completed, this process of oxidativephosphorylation requires efficient transport of electronsbetween five major protein complexes (mitochondrialelectron transport chain, attached to the inner mitochon-drial membrane). The total amount of oxygen in the bodyis low (�1.5 liter in total, �50 ml of which is stored in thetissues), so an undisturbed and complete oxidative phos-phorylation requires a continuous fresh supply of oxygenfrom the environment. Thus an insufficient supply of ox-ygen may compromise the electron transport chain result-ing not only in a decrease in ATP production but also in anincreased generation of reactive oxygen species (ROS),such as superoxide anion. ROS can cause structural dam-age to nucleic acids, proteins, and lipids. At the sametime, cytoplasmic signaling-transduction cascades can di-rectly be influenced by ROS. Mobilization of antioxidantdefense mechanisms, such as superoxide dismutase thatcatalyzes the conversion of superoxide anion into hydro-gen peroxide and catalase that accelerates the conversionof H2O2 into water and oxygen, helps to restore the cel-lular redox balance.

Apart from cellular antioxidant enzymes, the bodyhas several means to compensate for an acute decrease inoxygen availability (acute hypoxia). Animals relying onpulmonary ventilation for their oxygen supply increasetheir ventilation to improve the uptake of oxygen. Thisventilatory response to hypoxia (HVR) in mammals, i.e.,the increase in pulmonary ventilation that results from adecrease in the partial pressure of oxygen in the arterialblood (PaO2

) is the subject of this review. Another meansthat mammals can utilize to preserve oxygen homeostasisin acute hypoxia is to optimize and decrease the rate ofaerobic metabolism, thus reducing oxygen demand and atthe same time increasing ATP production from anaerobicmetabolism (glycolysis).

Neonatal and adult animals use the options of chang-ing ventilation and/or metabolism in a different way.Apart from an initial fast and transient rise in ventilation,neonates, when exposed to acute hypoxia lasting 20–30min, display only a minor increase in ventilation but anappreciable reduction in metabolism, thereby optimizingthe ratio ventilation over oxygen consumption. Duringmaturation into adulthood, this composite response grad-ually develops into one with a predominant and sustainedincrease in ventilation (while the biphasic character of the

response is maintained) but without a(n) (appreciable)decrease in metabolism.

Fetal animals respond differently to hypoxia. Theydisplay breathing movements in utero that can be de-tected in an early stage of gestation (for example, �40days in lambs with a term duration of �147 days and �10wk in humans). When the fetus is exposed to a lowmaternal PaO2

or umbilical cord occlusion, breathingmovements are reduced in frequency and can be com-pletely abolished (54, 137, 170). This review describes themechanisms underlying both the fetal and neonatalbreathing responses to hypoxia and summarizes how theneonatal response gradually matures into the adult phe-notype. The influence of the genotype on the HVR is alsobriefly discussed.

Both in neonates and in the adult, the initial increasein ventilation in hypoxia is initiated by the carotid bodies,located at the carotid bifurcations and containing oxygen-sensitive cells that increase their activity in hypoxia (278,280). The carotid bodies send their afferent input to thebrain stem where further processing by the respiratorycenters takes place. During the last two decades, consid-erable progression has been made in elucidating themechanism of oxygen sensing in the carotid bodies, thefactors that modulate it, and the way in which the brainstem processes the afferent traffic from the carotid bod-ies, and in this review we present a state of the art ofthese areas.

Studying the mechanisms underlying respiratory ad-justments (rise in ventilation, increase in the HVR) tochronic hypoxia and chronic intermittent hypoxia in ani-mals (mainly rats) has revealed a fascinating picture ofnumerous plastic morphological, biochemical, and func-tional changes in the carotid bodies and brain stem thatare initiated and maintained by transcription factors thatregulate the expression of many genes. Both during neo-natal development and in adulthood, the respiratory sys-tems of rats and other species display plasticity, i.e., theability to learn from previous experience and to alter thephenotype of several reflex responses. One developmen-tal period is of particular interest: immediately after birth,within the period of postnatal maturation, there is a crit-ical period (“critical time window”) during which environ-mental factors can modulate structural and functionalproperties of the respiratory system. These plasticchanges can be permanent and may be irreversible inadulthood and possibly even in adolescence (115). Anexample is chronic exposure to hyperoxia during thisperiod: it leads to degenerative changes in the carotidbodies and to a reduced HVR that may persist untiladulthood (189, 220, 838). Adaptations of the HVR to(intermittent) chronic hypoxic paradigms in this criti-cal neonatal time window will be compared with thosein adulthood and placed in the context of plasticchanges in the carotid bodies and brain stem that result

676 LUC J. TEPPEMA AND ALBERT DAHAN

Physiol Rev • VOL 90 • APRIL 2010 • www.prv.org

from changes in gene expression. Of critical impor-tance in interpreting these studies is to take into ac-count the specific methods that are used to measure theHVR, the environmental and physiological context inwhich they are preformed, and the analytical methodsthat are applied to quantify and compare data. Webriefly summarize the techniques used in small animalsand in humans and discuss the specific difficulties thatare encountered in measurement and analysis.

II. THE HYPOXIC VENTILATORY RESPONSE IN

FETAL AND NEONATAL MAMMALS

A. Fetal Breathing Movements

Fetal breathing movements (FBM) are coordinatedrhythmical contractions of respiratory muscles that inter-mittently reduce intrathoracic pressure and cause amni-otic fluid movements within the trachea (302). FBM arecrucial for fetal lung growth and maturation and are ob-served in all mammalian species studied thus far (49, 85,302, 357, 391, 651, 725, 814). The most commonly ob-served FBM are rapid irregular movements and can bedetected in sheep (mean term duration 147 days) from�40 days gestation and in humans from �10 wk (88, 137,170, 302). Early in gestation, until about day 115, fetallambs show continuous FBM that are sometimes inter-rupted by apneas; in the third semester, EEG activitystarts to differentiate between states of low- and high-voltage cortical activity associated with rapid eye move-ments (active sleep) and active movements of neck andlimb muscles, respectively. From this developmental stage un-til term, FBM are limited to active sleep (low voltagecortical activity; reviewed in Refs. 85, 170, 302, 317, 357,625, 814). This association of FBM and electrocorticalactivity is disrupted after brain transection rostral fromthe caudal hypothalamus; when this transection is per-formed at upper pontine level, i.e., a region close to theinferior colliculus in the vicinity of the principal sensoryand motor nuclei of the trigeminal nerve, FBM becomealmost continuous (170, 171, 271, 363, 396, 523, 814, 815).Sleep state should be considered as a factor modulatingfetal breathing rather than providing a primary inducingrole: in fetal lambs, periodic FBM precede the electrocor-tical differentiation by �1 wk, while at term the animalstarts to breathe continuously without changes in EEGpattern (651).

B. Hypoxia Inhibits Fetal Breathing Movements

In fetal sheep, hypoxia caused by (gradually) expos-ing ewes to a hypoxic gas mixture (mostly with CO2

added to maintain isocapnia) to achieve fetal saturation

levels of �30% leads to adaptations such as spinal inhibi-tion, a reduction in rapid eye movements, facilitation ofhigh-voltage electrocortical electroencephalogram (EEG)activity, inhibition of FBM, a fall in metabolic rate, and arise in cerebral blood flow (CBF) (85, 88, 137, 170, 317,408, 588). In the lamb, FBM account for �30% of overalloxygen consumption (670). Anemia, reduced uterineblood flow, and/or umbilical cord occlusion also lead toFBM inhibition (74, 87, 328, 830). Hypoxic inhibition ofFBM also occurs in fetal rat (391), rhesus monkeys (488),and given the relative low FBM incidence in growth-retarded fetuses, presumably also in humans (54, 251).The mechanism of hypoxic inhibition of FBM is complexand may be causally related to one or more of the aboveadaptations. Below we discuss some major factors influ-encing hypoxic inhibition of FBM with emphasis on thosethat may still contribute to shaping the HVR in newbornsand even adults.

1. CBF and metabolism and hypoxic inhibition

of FBM

The fetal brain exhibits a relatively large rate ofoxygen consumption that is tightly linked to oxygen de-livery (588). In the fetus, hypoxia not only leads to de-creases in neuronal activity and metabolic rate, but alsoto a rise in CBF, and both these responses have beenlinked to the inhibition of FBM (588).

2. Pontine and thalamic areas and hypoxic inhibition

of FBM

In fetal lambs, brain stem transection at the (mid)collicular level or electrolytic lesions somewhat morecaudally in the upper lateral pons in the vicinity of thetrigeminal sensory nuclei involving parts of the medialparabrachial and Kolliker Fuse nuclei has two importantconsequences: 1) it leads to continuous breathing in nor-moxia and to a dissociation between EEG activity andbreathing, and 2) it converts the inhibitory response tohypoxia in the intact condition into a stimulatory oneafter transection (170, 171, 271, 363, 396, 523, 814, 815).Transection more rostrally through the caudal hypothal-amus did not affect hypoxia-induced inhibition of FBM,indicating that an area caudally from it is able to elicit thisresponse independently from higher centers (170).

In fetal lambs, hypoxia facilitates a high-voltage elec-trocortical EEG pattern (80). FBM and low-voltage elec-trocortical activity are closely linked, so brain areas in-fluencing behavioral state may modulate hypoxic inhibi-tion of FBM. One such area is the parafascicular nucleusof the thalamus. Lesions in this area abolish hypoxia-in-duced inhibition of FBM while local electrical stimulationeffectively decreases respiratory frequency (397–399).

THE VENTILATORY RESPONSE TO HYPOXIA IN MAMMALS 677


3. Peripheral chemoreceptors and hypoxic inhibition

of FBM

In fetal sheep, the carotid bodies are active andrespond to hypoxia with excitation (79). Thus the ques-tion arises whether they may have a counteractive rolein hypoxia or, alternatively, stimulate brain regions thatare directly involved in inhibiting FBM. Carotid sinusnerve and aortic transection do not prevent hypoxicinhibition of FBM but delay the onset of the response,suggesting a supportive but not indispensable influenceof the peripheral chemoreceptors (406). After midcol-licular transection, hypoxia induced an increase inFBM in fetal lambs, and this response still existed aftersubsequent carotid body denervation (171). After ca-rotid sinus nerve transection, some fetal sheep showedcontinuous breathing in normoxia and even FBM inhypoxia; peripheral chemodenervation alone neitherchanged the incidence of FBM in normoxia nor thehypoxia-induced inhibition of breathing (523). In an-other study, hypoxic stimulation of FBM after transec-tion depended on carotid sinus nerve integrity, suggest-ing that in the intact condition (i.e., without any lesion)a stimulatory influence of the carotid bodies may beoverridden (363, 396). In general, the emerging pictureis that at least in fetal lambs intact carotid bodies arenot required to produce hypoxic inhibition of FBM,while their role in stimulating breathing in hypoxiaafter brain stem transection is somewhat controversial.

4. Placental factors and hypoxic inhibition of FBM

By itself, the fact that umbilical cord occlusionleads to hypoxia in the fetus and decreases the inci-dence of FBM (328, 830) does not exclude placentalfactors as the source of FBM inhibition during maternalhypoxemia. Theoretically, a reduced release by the pla-centa of stimulatory agents could contribute to thereduction of FBM, but to date, such factors have notbeen identified. The placenta produces inhibitory sub-stances such as prostaglandin E2 (PGE2), adenosine,and several peptides (16, 704, 815), while indomethacinand adenosine antagonists induce continuous breathingand reverse hypoxic inhibition of FBM, respectively(125, 170, 403, 404, 407). After prolonged reductions inuterine blood flow, plasma PGE2 concentration in fetallambs and the decrease in incidence of FBM showedparallel time courses (328). However, because sub-stances released from the placenta during maternalhypoxemia did not reduce FBM when the fetal PO2 waskept normoxic (413), we suggest that in this conditionhypoxic FBM inhibition is caused by local factors in thehypoxic fetal brain that also inhibit respiratory move-ments during reduced uterine blood flow/umbilical cordocclusion.

5. Neuromodulators and receptors in hypoxic

inhibition of FBM

A) ADRENERGIC MECHANISMS. Hypoxia-induced inhibitionof FBM may be mediated via adrenergic receptors. In fetalsheep, hypoxia activates catecholaminergic neurons insubcoeruleus- and Kolliker-Fuse regions in the dorsolat-eral pons, some of which appear to have direct connec-tions with the phrenic motoneuron pool in the cervicalcord (551). In both intact and brain stem transected fetallambs, systemic clonidine reduced the incidence of FBM,and this effect was blocked by an �2-adrenergic antago-nist (32). Prolonged reduction in uterine blood flow in-creased plasma norepinephrine concentration in fetallambs (328), while ventricular administration of clonidinereduced the incidence of breathing and a specific �2-adrenergic receptor antagonist abolished hypoxic inhibi-tion of FBM (31). Systemic administration of both �1- and�2-adrenergic antagonists had similar effects (270).

B) ADENOSINE A1 AND A2 RECEPTORS. An important mole-cule in the coupling of cerebral blood flow to metabolismis adenosine. Via adenosine A2 receptors, this moleculeinduces �50% of the hypoxia-induced rise in CBF, whileA1 receptors mediate hypoxic depression of cerebral me-tabolism (83, 479, 588).

Experimental evidence indicates that adenosine is animportant player in the hypoxia-induced reduction inFBM incidence. In fetal lambs, systemic adenosine causeda decrease in FBM incidence similar to hypoxia, whileadenosine receptor antagonists had the reverse effects(76, 125, 405). The posteromedial thalamus displays in-creased adenosine production during hypoxia and con-tains a high density of adenosine A2A receptors that me-diate FBM inhibition (402, 404, 830, 862). Intra-arterialinfusion of a specific A2A but not A1 receptor antagonistabolished both the hypoxic inhibition of rapid eye andbreathing movements, while A1 antagonism only elimi-nated the FBM response (403). Anemic hypoxia upregu-lates pontine adenosine A1 receptors in the brain stem offetal lambs, while the same receptor type in the rostroven-trolateral medulla is also involved in the hypoxic FBMresponse (76, 396), possibly by directly depressing localmetabolism (84, 479).

The way in which adenosine causes hypoxic inhibi-tion of FBM may be quite complex. Following lesion ofthe upper lateral pons in the vicinity of the trigeminalsensory nucleus (see above), normoxic respiratory fre-quency substantially decreases (363). Thus this regionmay exert a tonic excitatory effect on ventrolateral med-ullary respiratory neurons in normoxia, possibly by hy-perpolarizing postinspiratory neurons. If during hypoxiaadenosine, by a presynaptic action, inhibits these pontineneurons, postinspiratory neurons may now depolarize,thus inhibiting respiratory motor output. This would beconsistent with in vitro data from neonatal rat and mice



showing hypoxia-induced depolarization of postinspira-tory neurons (see references in Ref. 73).

In summary, ample evidence thus far indicates in-volvement of adrenergic and purinergic, adenosine A1 aswell as A2A mechanisms in the hypoxic inhibition of FBM,but further studies are necessary to further characterizeand localize other transmitters and receptor subtypesinvolved.

6. Some further considerations on hypoxia and

fetal breathing

In most of the above studies, hypoxia was tested infetal lambs in the last gestational semester. In midgesta-tion, hypoxic inhibition of FBM is milder if not absent,indicating that it develops over the second half of preg-nancy (496).

Does the low fetal arterial PO2 (�25 Torr) imply atonic reduction of FBM? This seems unlikely, since nei-ther fetal lambs (125–130 days gestation) nor normal hu-man fetuses show significant increases in the incidence ofFBM during maternal hyperoxia (54, 78, 88, 654 and ref-erences therein), although at least in the lamb it cannot beexcluded that after �130 days of gestation a response ofcontinuous breathing to hyperoxia starts to develop (24).In humans, it is only in growth-retarded fetuses that hy-peroxia results in a higher FBM frequency (54, 251).

In fetal lambs, prolonged hypoxia (�12 h) initiallyleads to a reduction in FBM frequency, but after a periodof 12–20 h it returns to normal (86, 328, 401, 496, 831).How this reversal occurs is unknown, but it has beenrelated to a downregulation of adenosine receptors andplasma glucose levels, respectively (328, 405).

Several alternative mechanisms to explain the hy-poxic inhibition of FBM have been suggested, of whichacidemia (647), increased blood flow through the ductusvenosus (217), and decreased mitochondrial ATP produc-tion (407) warrant further investigation. The cellular basisof the response is unknown.

C. Maturation of the HVR in Neonatal Animals

1. Measuring the HVR in small animals

The most frequently used techniques to study breath-ing in awake mice and rats are modifications of bodyplethysmography described by Drorbaugh and Fenn (193).Their barometric method is based on the principle thatwhen an animal in a closed chamber inhales a tidal vol-ume, this volume is warmed to body temperature, satu-rated with water vapor giving rise to pressure changes inthe chamber. As discussed by others, both closed andopen systems are subject to limitations (e.g., related topressure changes due to warming, humidification, and airleaks) and particularly in neonatal animals the pressure

signal may be contaminated by airway resistance effectsthat may lead to erroneous estimation of tidal volume(218, 219, 356, 477, 751).

The absence of blood gas data (in most cases) com-bined with the impossibility to control end-tidal gasesrenders a reliable comparison of HVRs between treat-ments difficult. By sheer necessity, the ambient O2 frac-tion is often taken as the independent variable, but itsrelationship with the actual stimulus is uncertain and mayvary. If in a chronic paradigm normoxic ventilation in-creases, arterial PO2 will be higher and PCO2 lower for anyinspired PO2 level resulting in less carotid body stimula-tion; thus, in such cases, it is inaccurate to quantify theHVR as the magnitude of the change in ventilation pergiven decrease in ambient PO2. In addition, a lower initialPCO2 may tend to further reduce at least the early phase ofthe HVR, owing to O2-CO2 interaction in the carotid bod-ies (see sect. V). Without addition of CO2, hypoxic testsmay be poikilocapnic depending on changes in ventilationand/or metabolism. In the ideal case, preferably three orfour measurement points well within the curvilinear partof the exponential ventilation-PO2 relationship should bechosen, avoiding levels at which maximal ventilation isreached (39). A further point concerns the influence of thevigilance state on the HVR (421, 472, 729). Particularly inneonates, EEG recordings are challenging and often notperformed, so vigilance state-dependent changes in sen-sitivity cannot be accounted for. On the other hand, theinfluence of the latter factor may be limited becausesleep-wake cycles are not clearly defined in early life.Balbir et al. (29) proposed a noninvasive method forsleep-wake assessment in neonatal mice, providing a po-tential means to account for (changes in) brain state. Theseauthors found a sound correlation between high electromyo-gram (EMG) activity (associated with the awake state) andcoordinated limb/head movements and suggested that thisbehavioral index could be useful in avoiding the stress re-lated to EMG instrumentation in newborn animals (29).These are only a few of the considerations to take intoaccount when comparing HVRs between treatments.

2. Maturation of the HVR in neonatal animals

In neonatal animals, the HVR appears as a biphasicresponse consisting of a short initial rise in ventilation(augmentation phase) rapidly followed by a secondaryroll-off mostly to a level below control (depressing phase;Ref. 73). Postnatal maturation of the HVR was alreadydescribed in older studies and in various species such aslambs (e.g., Ref. 102), cats (89, 90, 301, 487, 653), piglets(212), and rats (208, 453, 454). In all species, the relativemagnitudes of both phases are subject to maturation.Generally, shortly after birth, the initial rise in ventilationis (much) smaller than the secondary roll-off, rapidlyresulting in a lower baseline ventilation compared with



normoxia. However, when maturation proceeds, the ini-tial rise becomes larger, the secondary depression be-comes smaller, and eventually a sustained rise in ventila-tion remains. Age-matched comparisons between speciesare complicated by the different stimulus paradigms em-ployed. For example, deeper hypoxia may result in agreater initial increase and a larger decrease in ventilationin the depressing phase (see Ref. 73 in which data fromhumans, rat, cat, monkey, and sheep are shown). In ad-dition, relatively mature species at birth have a bettersustained neonatal HVR than those born less mature, e.g.,sheep (mean term duration 147 days) are more maturethan rat (term duration 18–23 days) and cat (term dura-tion 58–64 days; Refs. 73, 529). In 2- to 17-day-old piglets(term duration 112–120 days), the secondary roll-offseems to be limited to the phrenic motoneuron pool, sinceupper airways muscle EMGs showed a sustained increasein activity in the presence of a biphasic response of thediaphragm (490).

In most neonates, a vital component of the hypoxicresponse is a decrease in metabolism (hypoxic hypome-tabolism; Refs. 90, 529, 531, 532). Newborns of manyspecies demonstrate a drop in oxygen consumption (VO2)that in relatively mature species tends to be smaller. Allspecies investigated show an increase in the ratio expiredventilation (VE)/VO2 (i.e., the ratio expired ventilation overO2 consumption), indicating a hyperventilatory response,and this occurs irrespective of the magnitude of the HVR(529, 531, 532). Thus ventilation may decrease in hypoxia,but still the outcome will be hyperventilation because ofa larger drop in VO2. The fall in metabolism is due to areduction in nonshivering thermogenesis by brown adi-pose tissue, rather than to the hypoxia-induced (limited)decrease in body temperature (526, 529, 531, 532). Restingmetabolism is an important determinant of the hypoxicdrop in VO2. Generally, animals with higher metabolic rate(e.g., small animals) demonstrate larger decreases in VO2,but there are exceptions (243, 526, 531). In a cold envi-ronment, requiring more thermogenesis to generate heat,a given animal will show much larger hypoxia-induceddecreases in VO2 and ventilation (or smaller rise in venti-lation) than at an ambient temperature close to thermo-neutrality (526, 531). Consequently, measurements of theHVR per se are more meaningful when combined with VO2

measurements and supplemented with end-tidal CO2 val-ues. Furthermore, comparisons between treatmentsshould preferably be made at equal ambient temperature(526, 531).

When maturation proceeds towards the adult stage,hyperpnea replaces hypometabolism as the predominantfeature of the hypoxic response, but in many adult spe-cies, depending on the ambient temperature, the drop inmetabolism can remain an important element of it (243,260, 482, 526, 531, 675). The magnitude of the fall inmetabolism is inversely related to resting VO2 relative to

body weight and therefore could be limited or absent inadults of larger species (243, 531). The processes under-lying maturation of the HVR are briefly reviewed below byconsidering the augmentation and depressing phases sep-arately. In section IIIC, we briefly discuss a potential neu-ronal circuit that may be involved in the thermogeniccomponent of the HVR.

A) AUGMENTATION PHASE OF THE HVR. In newborn lambs,piglets, and rats, the relatively rapid initial rise in ventila-tion in acute hypoxia, designated as the augmentationphase (73), is abolished or greatly attenuated after carotidsinus nerve transection, indicating that the carotid bodiesmediate the augmentation phase (103, 166, 249, 473, 474,533, 740). Shortly after birth, in newborn lambs, the PaO2

threshold for hypoxic activation of the carotid body isreset from the low fetal PO2 (25 Torr) to �55 Torr (79, 82).This is followed by a gradual increase in hypoxic sensi-tivity of the carotid sinus nerve, a shift of the hyperbolicresponse curve to higher PO2 values, a gradual appearanceof O2-CO2 interaction, and an improved ability to maintainelevated discharge levels during prolonged hypoxia (116,188, 487, 533, 601). The time course of these developmen-tal changes (a few days to 3–4 wk) varies between spe-cies, and the relative maturity at birth determines the timepoint at which adultlike responses are reached (33, 103,208, 487, 601, 826). Processes underlying maturation ofthe augmentation phase appear to be a complex finetuning between afferent nerve innervation of the carotidbodies involving trophic factors such as brain-derivedneuorotrophic factor and all elements of the of hypoxicstimulus-transduction cascade including membrane chan-nels, neurotransmitters, receptors, and enzymes (see sect.III) and are reviewed elsewhere (26, 94, 191, 258, 611).

These maturational processes must keep pace withdevelopmental changes in the central nervous system, forexample, in the nuclei of the solitary tracts (NTS), the firstrelay station of incoming carotid body afferent input.These processes involve changes in the activity of tran-scription factors, formation of anatomical connectionswithin the neuronal respiratory circuitry, and alterationsin neurotransmitter contents and receptor densities to setthe balance between excitatory and inhibitory neuro-transmission. For example, during maturation, the densityand expression of the glutamate receptor ionotropic N-methyl-D-aspartate 1 (NR1) subunit of the NMDA (N-methyl-D-aspartic acid) receptor in the rat NTS increases, whichleads to the upregulation of transcription factors andother proteins that are involved in the hypoxic response(564, 702). Another example of a developmental change inthe NTS is the downregulation of platelet-derived growthfactor (PDGF) during early maturation (702; PDGF exertsan inhibitory influence on ion currents through NMDAreceptors, see also sect. IIID). Also in the rat, Wong-Rileyand Liu (847) have shown shifts in the expression ofGABA receptor subunits in the NTS and Pre-Botzinger



complex around postnatal day 12 (P12) that were paral-leled by ventilatory adjustments at the end of the secondpostnatal week. Between P12 and P16, the HVR showed asharp decline to reach a minimum at P13, whereafter itgradually increased to a final magnitude on P20–P21.Body temperature during hypoxia gradually rose duringthe first three postnatal weeks to also reach the normoxicvalue at P20–P21 (847), consistent with a lesser contribu-tion of the thermogenic component to the hypoxic re-sponse at older age (see sect. IIC2). Maturation of themammalian respiratory network is reviewed in Refer-ences 115, 186, 846.

3. Depressing phase of the HVR

The secondary roll-off in neonates may be associatedwith pulmonary mechanics, a time-dependent decrease incarotid body activity or sensitivity, impaired central inte-gration of afferent carotid sinus nerve input, decreasedmetabolism, increased CBF and, similar to the inhibitionof FBM, active inhibition by pontine and mesencephalicstructures involving inhibitory neurotransmitters (re-viewed in Ref. 73). Muscle fatigue is unlikely to play asignificant role, since the phrenic nerve response in anes-thetized neonatal animals does also show a biphasic pat-tern, indicating that the secondary roll-off is a neurogenicmechanism (121, 249, 522).

In 5- to 10-day-old rabbit pups, brain stem transectionat the midbrain-pontine junction converted the hypoxia-induced depression in controls into sustained stimulationin test animals (491). Focal cooling in the locus coeruleusreversibly abolished the hypoxia-induced depression in 3-to 8-day-old anesthetized lambs (522).

Available evidence also points to a role of the carotidbodies in the depressing phase. The temporal response ofthe sinus nerve discharge in kittens differed among stud-ies, declining in some (116, 487) but sustained in others(81). In 2- to 3-day-old lambs, a sudden hyperoxic stimulusduring a hypoxic exposure had a larger effect after 3 minthan after 15 min, suggesting waning peripheral chemore-ceptor activity or impaired central integration (117). Ca-rotid body denervation attenuated but not completelyabolished most of the secondary roll-off in 0- to 13-day-oldrats (249), and in newborn lambs the subcoeruleus areashowed increased activation in hypoxia after carotid bodydenervation only (95).

In neonates, adenosine, serotonin, prostaglandin,GABA, opioids, catecholamines acting at �2-receptors,and other neuromodulators may all be involved in thedepressing phase (e.g., Refs. 212, 340, 400, 462, 740; seefurther references in Refs. 73, 285, 702). The strongestcase was made for adenosine, which, similar to the fetus,in neonatal sheep reduces hypoxic ventilation both viaadenosine A1A, but mainly adenosine A2A receptors (400).In newborn lambs showing a biphasic phrenic hypoxic

response, focal cooling of regions in the locus coeruleusprevented the (secondary) fall in respiratory motor out-put induced by adenosine and hypoxia, respectively (522).In 3-day-old piglets, the hypoxia-induced increase in tidalvolume waned between the second and fifth minute ofhypoxia, but this gradual waning could be prevented bythe adenosine antagonist 8-phenyltheophylline; 3-wk-oldanimals, however, did not show this secondary decreasein tidal volume, whilst the adenosine antagonist did nothave any influence on the sustained rise in ventilation(212). In piglets 2–7 days of age, aminophylline preventedthe secondary decrease in phrenic activity during hypoxia(121). In summary, the secondary roll-off in neonates maybe partly a manifestation of similar mechanisms that in-duce hypoxic inhibition of FBM in the fetus on which apostnatal influence of the peripheral chemoreceptors issuperimposed. Note, however, that in many of the aboveneonatal studies, the end-tidal PCO2 was uncontrolled. If insmall animals the HVR is measured at temperatures belowthermoneutrality (which is often the case), a fall in me-tabolism could well contribute to the secondary roll-off.In these cases, measurements of metabolism and bodytemperature would be very helpful (260). Other factorssuch as the vigilance state, the severity of the hypoxicstimulus, and other experimental conditions will also in-fluence the magnitude of the depressing phase in theneonatal period.

D. Developmental Plasticity of the HVR

A definition of developmental plasticity of respira-tory control was provided by Carroll (115) as “long-termalterations in the structure or function of the respiratorycontrol neural network caused by experience during pre-or postnatal formation of the respiratory control system.”

Overwhelming evidence has identified the neonatalenvironmental PO2 as a major factor influencing normalHVR development. Hypoxia and/or hyperoxia around thetime of birth and the first few weeks thereafter can dis-turb normal development of the HVR that may persistuntil adulthood. The effects of perinatal chronic or inter-mittent hypercapnia were less frequently studied; in therat, neither the neonatal nor the adult HVR seems to beaffected by neonatal hypercapnic exposures (43, 646).

1. Effects of neonatal chronic intermittent hypoxia

Term and preterm infants in particular are repeatedlyexposed to hypoxia due to their irregular breathing pat-tern with many apneas. Animal studies have now madeclear that within a critical time window of neonatal de-velopmental plasticity, chronic intermittent hypoxia caninduce life-long alterations in the control of breathingwith potential pathophysiological consequences (617,633, 827). In this context, the critical time window of



neonatal plasticity was defined as a period within thedevelopmental stage in which the organism is uniquelysensitive to environmental perturbations that may alterthe ultimate configuration of the neuronal network andassociated functions (115, 632). In terms of respiratorycontrol, it may imply that exposure to low or abnormallyhigh oxygen (or CO2) tensions within this window canlead to life-long perturbations in respiratory control.

The effects of repeated exposures to hypoxia can bequite complex and critically depend on the total durationof an intermittent stimulus, its frequency, the overall num-ber of cycles, and postnatal age which all appear to inter-act (633, 827). One reason for this interaction is that theHVR is build up of a series of sequential events withdifferent lags, time constants, gains, and recovery times.For example, hypoxia augments carotid body activity,increases the discharge frequency in the carotid sinusnerve that results in the release of neurotransmitters inthe NTS, in turn leading to the release of neurotransmit-ters in the ventrolateral medulla etc., eventually causing atime-dependent recruitment of motoneurons. When a newstimulus is offered while one of the processes initiated bythe previous stimulus is still in its recovery time or even(partly or fully) active, it is conceivable that the effects ofboth stimuli on motor output will differ (827).

To investigate the effect of chronic intermittent hyp-oxia on developmental plasticity of the HVR, a variety ofparadigms with different stimulus magnitude, length, andtotal duration and number of exposures were used inseveral species such as mice (197), rabbits (789), piglets(828), and rats (reviewed in Ref. 286). In the rat studies,the age of the animals varied from 0 to as many as 30 daysand more, well beyond the critical time window for de-velopmental plasticity (115, 286). The results of thesestudies in different species are equivocal, varying from areduced (45, 497, 633, 635, 828 and references therein),unaltered (197, 676) to an enhanced (367, 585, 598) HVRand/or carotid body sensitivity. In the rat, blunting of theHVR was reported to persist at 1 mo after cessation of thestimulus paradigm (635). As shown in neonatal piglets,the background PCO2 during the hypoxic episodes mayhave an important influence: a poikilocapnic paradigm(10% O2) blunted the HVR, but hypercapnic chronic inter-mittent hypoxia (6% CO2, 10% O2) caused an increase(828, 829). In instrumented anesthetized piglets, 11 peri-ods of recurrent hypoxia blunted the phrenic nerve re-sponse to acute hypoxia in young (2–10 days) but notolder (2–3 and 8–10 wk) animals (509; note however thatin all recovery periods, these animals breathed 100% O2

which may have an effect by itself, see below). In most(but not all) studies, normoxic ventilation increased (599,631–633) and, depending on the animal’s age at the onsetof the hypoxic stimulus, this effect was permanent (thefirst two postnatal weeks were most efficient in this re-spect; Refs. 632, 633, reviewed in Ref. 45).

These distinct effects of various specific neonatalchronic intermittent hypoxia protocols on the augmenta-tion phase of the HVR and mechanisms underlying neo-natal respiratory plasticity are reviewed elsewhere (45,115, 515, 631, 632, 827). The effects on the depressingphase have not been studied extensively, but at least inthe rat, the secondary roll-off seems to be attenuated andrelated to increased nitric oxide synthesis in the caudalbrain stem (286).

We attribute the above divergence in study outcomesto the considerable variations in the applied stimuluslevel, cycle length, the amount of hypoxic episodes, andthe number of exposure days. Differences in methodology(e.g., plethysmography vs. pneumotachography) and de-velopmental critical time windows between species willalso play a role (115). Interpretation problems due tochronic intermittent hypoxia-induced changes in nor-moxic ventilation and the employment of different meth-ods to normalize data are illustrated by the work of Linget al. (449) and Reeves et al. (633) in adult and neonatalanimals, respectively. The poikilocapnic nature of theexperiments is also a confounding factor. As shown ininfants, kittens, and dogs, hypoxic hypometabolism canbe accompanied by a (sustained) fall in PaCO2

in thedepressing phase despite the decrease in ventilation (93,489, 528, 648, 652).

The problem of measuring in a closed loop configu-ration can be overcome by working with anesthetizedanimals or ex vivo carotid body preparations. Rat pupssubjected to chronic intermittent hypoxia soon after birth(15 s in 5% O2 followed by 5 min in 21%, 9 episodes/hduring 16 h) showed an augmented ex vivo hypoxic ca-rotid body sensitivity that persisted until adulthood andwas associated with hyperplasia of glomus cells (585).Carotid bodies from neonates are much more sensitive tochronic intermittent hypoxia than those of adult animalsand the effects in neonatal animals are permanent (585).Recently, Pawar et al. (586) showed that in neonatal ratschronic intermittent hypoxia induced an augmented basalrelease of endothelin-1 (ET-1) in normoxia, upregulationof ETA receptor mRNA, and an enhanced carotid bodysensitivity to ET-1. In another study, a chronic intermit-tent paradigm led to decreases in both the HVR andisocapnic hypoxic phrenic response (635), which wouldbe more in line with the effect of chronic hypoxia. Theparadigm consisted of alternating room air with 10% O2

every 90 s (12 h/day, during 30 days), raising the questionwhether after a short period of low oxygen, 90 s is suffi-cient to completely restore tissue oxygenation to normal,possibly resulting in chronic hypoxia rather than chronicintermittent hypoxia.

In conclusion, recent data from ex vivo carotid bod-ies indicate an increase in sensitivity by neonatal chronicintermittent hypoxia. Technical limitations and method-ological differences are probably responsible for the fact



that as yet these important results could not be consis-tently confirmed at the level of minute ventilation. Stan-dardizing protocols, uniform quantification of HVR data(combined with metabolic data), and blood gas measure-ments may help to reach consensus on the effects ofneonatal chronic intermittent hypoxia. Intermittent expo-sures in which hypoxic episodes follow each other soclosely in time that episodic reoxygenation is incompleteor does not occur at all may be interpreted as chronic (seealso ref 827).

2. Effects of neonatal chronic hypoxia

Cats, sheep, and rats subjected to hypoxia (10–15%O2) from birth until 24 h before they were tested at agesvarying from 5 days to 10 wk showed a markedly reducedHVR, mainly caused by a depression of the augmentationphase (39, 42, 42, 45, 207, 567, 703 and references therein).

In rats and sheep exposed to neonatal hypoxia for 1–2wk, blunting of the HVR persisted for at least 5–12 wk (39,476, 567, 703), which in the rat was sex specific (only ob-served in males; Refs. 42, 476). The effect of neonatal hyp-oxia on normoxic ventilation is variable: in some studies, along-lasting increase was observed (527, 566, 567) while itwas absent in others (42, 476, 703). At the age of 12 wk, bothmale and female rats displayed an impaired ventilatory ac-climatization to chronic hypoxia (2 wk in 10% O2) when theywere raised in hypoxia until postnatal day 10 (476).

Initially, in the first weeks of maturation the bluntedHVR in chronically hypoxic neonates may be due to adelayed carotid body resetting and/or to an attenuation ofthe maturation of the O2-CO2 interaction (301, 303, 437,730, 826; further references in Ref. 115). Single fibers ofthe carotid sinus nerve from rat puppies that were raisedin 9% O2 from the first or second postnatal day until 3–4wk later, showed a 50% lower peak response to severehypoxia than controls (186). Both at the ages of 3 and 8wk, rats exposed to hypoxia for 6 days in their neonatalperiod had increased carotid body dopamine and norepi-nephrine levels (721). Glomus cells from neonatal ratsreared in 12% O2 during 1–3 wk demonstrated a reducedrise in intracellular [Ca2�] in hypoxia compared with cellsfrom normoxic aged-matched controls, and this impairedmaturation partially recovered 1 wk upon return to nor-moxia (731). Rat puppies born and raised in 10% O2 for9–14 days had a substantially reduced HVR and possessedtype I cells that displayed a reduced expression of maxi-Kchannels and failed to depolarize in hypoxia (854). Inneonatal and juvenile type I cells from rats cultured in 6%O2 for 12 days, maxi-K channels were downregulated(354). These studies suggest that the neonatal chronichypoxia-induced HVR reduction is at least partly due to animpaired carotid body function. Not all data, however,support this because changes in sodium and potassium

currents that would tend to increase the excitability oftype I cells were also reported (315, 316, 726, 727).

Neonatal chronic hypoxia will also affect or delaynormal maturation of the central respiratory neuronalnetwork. For example, exposing newborn rat pups to 6days of chronic hypoxia resulted in a decreased norepi-nephrine turnover in the caudal part of the noradrenergicA2 region in the NTS (an important projection area ofcarotid body afferents) at the age of 8 wk, suggestinglong-term effects of the neonatal challenge (721). NMDAreceptors in pons and caudal medulla of rats raised in 10%for 4–5 wk from birth were downregulated (248). Afterthey were raised in 13–15% O2 from birth to 5–10 wkpostnatally, anesthetized rats showed normal (i.e., similarto age-matched normoxic controls) isocapnic hypoxic si-nus nerve responses, but their HVR remained blunted(207). Adult anesthetized rats, kept in 10% O2 during theirfirst postnatal week, had a normal phrenic response toisocapnic hypoxia, but their HVR was reduced (43). Theauthors of the latter two studies suggested that the sus-tained reduction in HVR after neonatal CH may originatedownstream from the phrenic motoneuron pool. Abnor-mal and/or retarded lung growth may then provide apotential explanation (493). Normal maturation of theHVR is also associated with an age-dependent attenuationof the, centrally governed, depressing phase (see above).In rat puppies, chronic hypoxia from birth delays thisdecrease (207). Other adaptations observed after neonatalchronic hypoxia in rats are an increase in hematocrit,decrease in metabolism, and an enlarged alveolar gasexchange surface perhaps tending to compensate for areduced HVR (527, 531 and references therein). Exposingnewborn piglets to hypobaric hypoxia during 3 dayscaused pulmonary hypertension and prevented the rapidremodeling of the cytoskeleton of pulmonary arterial smoothmuscle cells that normally contributes to the postnatal de-crease in pulmonary vascular resistance (298).

Taken together, neonatal chronic hypoxia-induced blunt-ing of the HVR has multifactorial causes involving plasticchanges in the carotid bodies, the central nervous system,and possibly also lung mechanics and pulmonary vessels.

3. Effects of postnatal chronic hyperoxia

Earlier experiments in fetal sheep suggested thatbrief prenatal hyperoxia would hasten resetting of thecarotid bodies explaining the reported increased carotidbody O2 sensitivity in these animals (82). This contrastswith many studies showing a reduced HVR after chronicneonatal hyperoxia. For example, kittens breathing 30%O2 during their first 13 days showed a reduced early-phaseHVR right away after this exposure (301). In recent years,much attention was focused on the effects of immediatepostnatal chronic hyperoxia on the HVR on the longerterm, extending to the adult age (reviewed in Refs. 39, 45,



115). Here we will briefly highlight some main findings.The effect of neonatal hyperoxia on adult normoxic base-line ventilation appears variable, with some studies show-ing an increase and others reporting no change (refer-ences in Ref. 45). The consequences for the HVR andhypoxic responses of the phrenic and carotid sinus nervesat the adult age can be summarized as follows: 1) chronicneonatal hyperoxia causes a decrease in hypoxic sensi-tivity at all these three integration levels that persists intoadulthood (71, 246, 449). 2) The critical time window toinduce this effect, which is dose dependent and can al-ready be induced by 30% O2, lies within the first 2 wk;longer exposures cause no additional depression (40, 41,71). 3) The attenuation of the HVR can be limited byalternating the chronic hyperoxia with periodic hypercap-nia or by replacing chronic by intermittent hyperoxia,suggesting that intermittently increasing carotid body ac-tivity reduces the effect (43). 4) Central (carotid sinus tophrenic nerve) integration remains unimpaired, suggest-ing that the depressing effect originates from the carotidbodies (450). 5) The carotid bodies show degeneration,hypoplasia, changes in membrane properties, and a re-duced number of unmyelinated fibers that altogether leadto reduced O2 sensitivity of glomus cells and the carotidbody in vitro (189, 220, 622, 838). 6) Petrosal ganglionneurons also show degenerative changes (loss of tyrosinehydroxylase expression; Ref. 220).

A recent study focused on the time course of alter-ations in hypoxic responses of glomus cells and carotidsinus nerve afferents from rats exposed to chronic hyper-oxia starting on postnatal day 7 and lasting for 1, 3, 5, 8,and 14 days, respectively. On postnatal day 8 (so after 1day of hyperoxia), the secretory response of glomus cellsand the nerve response to hypoxia were increased, butbeyond exposure days 3–5, both responses were reduced,and this reduced sensitivity persisted until at least 1 wkafter exposure (187). Are these effects reversible onceanimals have reached the adult age? Bisgard et al. (72)reported that sustained hypoxia at adult age (3–5 mo)only sensitized the carotid sinus nerve when the chronicneonatal hyperoxia did not outlast 2 wk. However, an-other study showed recovery from the deleterious effectsof chronic hyperoxia during the first month after birth atthe age of 4 mo (247). An antioxidant rich diet during theneonatal hyperoxia did not prevent an impaired HVR atadult age (44), not supporting ROS-mediated toxicity. Fu-ture studies are warranted to further examine the influ-ence of dose and duration of neonatal chronic hyperoxiaand how an impaired HVR at adult age might be restored.

E. The HVR in Term and Preterm Babies

In most cases, the HVR in young babies is measuredby exposing them to 15% O2, seldom with addition of CO2.

Most recordings are from quiet sleep when breathing ispredominantly under metabolic control and somewhatless unstable than in active sleep, with less frequentarousals. Control of end-tidal gases is unusual so that themagnitude of the HVR reflects the closed-loop propertiesof the respiratory system.

With the exception of very small premature infantsshowing a sustained “fetal-like” decrease in ventilation(15), most term and preterm infants exhibit a biphasicHVR upon exposure to 15% O2 during 5 min, much similarto that in neonatal animals. The secondary fall appears tobe due mainly to a fall in frequency while in most casestidal volume shows a sustained increase, sometimes afteran initial overshoot (138, 139, 489, 552, 637, 648, 651, 652).With some exceptions (e.g., Refs. 139, 723) early in devel-opment the depressing phase of the HVR is associatedwith a sustained decrease in end-tidal PCO2 (489, 637, 648),similar to neonatal animals, and this is possibly related toa fall in metabolism (147, 332). During maturation, thedecrease in frequency gradually becomes less dramatic,and with advancing age, tidal volume tends to remainelevated (489, 648, 651).

The normal developmental time course of the HVR inhumans appears difficult to determine. Breathing in termand especially preterm infants is characterized by irregu-larities and apneas (625) that have been related to imma-ture brain stem function and connections (318) and alsoto a relatively low PaO2

leading to brisk hypoxic respon-ses, hypocapnia, and central apneas (552). Consequently,when occurring within the critical developmental timewindow, both the resulting chronic and intermittent hyp-oxia or a combination may hinder a normal developmentof the HVR. Despite the quite different term durations inrodents and humans, the animal data raise the question asto whether oxygen therapy in the neonatal period mayhave unwanted side effects with regard to the develop-ment of normal O2 and CO2 reflex sensitivities. By itself,preterm birth exposes the newborn to an environmentthat is hyperoxic relative to postconceptional age.Whether this is related to the lower HVR of preterminfants compared with age-matched controls remains tobe investigated. The higher prevalence of the suddeninfant death syndrome (SIDS) in preterm compared withterm babies has been related to relative immaturity of thecardiorespiratory neuronal circuitry resulting in failure ofventilatory and arousal responses (518).

Of practical interest when studying the HVR in bothneonatal animals and infants is its dependency on thevigilance state (317, 319, 472, 648, 651, 729). Most HVRdata were collected in quiet sleep, but potential influencesof arousal are not always taken into consideration (648).In a longitudinal study in healthy term infants, the HVRwas reported to mature between 2–5 wk and 5–6 mo ofage in quiet but not active sleep, especially in childrenthat did not arouse by the stimulus (15% O2 during 5 min



or 85% SpO2, see Fig. 1). Even at the age of 5–6 mo,ventilation did not exceed prehypoxic control after a5-min hypoxic challenge, and this was considered an im-mature HVR. However, taking into account that the pro-tocol was clearly poikilocapnic and that the HVR wasidentical to that after 2–3 mo and very similar to thepoikilocapnic HVR in adult humans (cf. Fig. 1) leads us tosuggest that maturation of the HVR in term infants may becomplete at �2 mo. Direct assessments of carotid bodysensitivity with brief exposures to 100% O2 (Dejours test;Ref. 13) or by offering alternating breaths containingroom air and 16% O2 (108, 843) pointed to full maturationof the peripheral chemoreflex within the first few weeksand possibly even the first 48 h after birth (13; note,however, that data obtained with Dejours tests should beinterpreted with caution because the peripheral chemo-receptors may still display considerable activity in hyper-oxia). That in term infants the O2-CO2 interaction hasbeen reported to emerge at an age somewhere betweenday 2 and week 8 (no testing after �2 wk) is not incon-sistent with this picture (723).

III. PERIPHERAL AND CENTRAL MECHANISMS

MEDIATING THE HYPOXIC VENTILATORY

RESPONSE IN ADULT MAMMALS

A. Carotid Body Mechanisms in Oxygen Sensing

The ventilatory response to hypoxia is initiated bythe carotid bodies, with the aortic bodies playing only asecondary role, perhaps with the exception of conditionswherein the carotid bodies are functionally eliminated(278). The carotid bodies, bilaterally located in the carotidbifurcations at the port of the brain circulation, have the

highest blood flow-to-metabolism ratio in the body and,both anatomically and biochemically, are extremely com-plex organs (278, 280 and references therein). Not only dothey possess cells specialized in oxygen sensing, but infact, they appear to be polymodal and sensitive to avariety of stimuli such as pressure, hypercapnia, acidosis,hyperkalemia, hyposmolality, circulating hormones, hy-perthermia, hypoglycemia, and numerous pharmacologi-cal agents (reviewed in Refs. 416, 558).

The two main cell types in the carotid bodies areglomus (type I) cells, having many similarities with sen-sory neurons and thought to be oxygen sensitive andsustentacular (type II) cells with a glialike appearance(278). Sensory neurons with cell bodies in the petrosalganglia send peripheral axons to the carotid bodies thatrun with the carotid sinus nerve, a side-branch of theglossopharyngeal nerve, and have nerve endings closelyoriented to type I cells. Central axons of these sensoryneurons enter the brain stem with the glossopharyngealnerve and terminate in caudal subnuclei of the NTS thathave extensive connections with the brain stem respira-tory network (references in Refs. 232, 335).

Hypoxia is followed by release of neurotransmittersby type I cells that act on the afferent endings of thecarotid sinus nerve to increase impulse traffic to the brainstem (278). Neurotransmitter release is the result of a risein intracellular [Ca2�] following depolarization of themembrane potential of type I cells. The mechanisms un-derlying the entire signal-transduction cascade in type Icells from hypoxia to the release of neurotransmittershave been the focus of numerous studies over the last twodecades. This has resulted in the acceptance, albeit notwithout controversies, of a framework in which hypoxialeads to inhibition of ion currents through potassiumchannels in the cell membrane resulting in membrane

V

V.*

** ***

FIG. 1. Poikilocapnic hypoxic ventilatory responses in young babies and adults. A: maturation of the ventilatory response to 15% O2 in 15 terminfants between the ages of 2–3 wk (■), 2–3 mo (E), and 5–6 mo (Œ). *Significantly different with levels after 2–3 mo. **Significantly different whencompared with 5–6 mo. Measurements were performed in quiet sleep, and the children did not arouse from the stimulus. B: poikilocapnic hypoxicventilatory responses in 10 healthy young adults upon a step decrease to and end-tidal PO2 of �44 Torr. Note that both after 2 and 5 mo thepoikilocapnic responses are qualitatively similar to the adult. [Adapted from Richardson et al. (648) and Steinback and Poulin (728).]



depolarization, influx of extracellular Ca2� via voltage-gated (mainly L and P/Q type) Ca2� channels, and releaseof neurotransmitters (278, 465, 467, 469, 662, 835, see Fig. 2).Overall, the main focuses of attention were the following:1) the element(s) that initiate(s) the cascade, in otherwords the identity of the intrinsic O2 sensor(s); 2) thecause(s) by which the resting membrane potential chan-ges, the identity of potassium channels involved in this,and the link of these channels with the intrinsic O2 sen-sor(s); and 3) the identity of the neurotransmitters thatare released by type I cells and excite afferent nerveendings. With regard to the identity of the contributingpotassium channels and excitatory neurotransmitters,there seems to be a reasonable consensus, although someaspects remain controversial. These issues are covered byseveral recent reviews and are briefly referred to belowbut not discussed in detail. Recent studies on the identityof the O2 sensor(s), and their link with potassium chan-nels, to be discussed below in more detail, have improved

our insight into the way type I cells couple changes inPaO2

to alterations in potassium channel activity.

1. Hypoxia decreases the open probability of

potassium channels

It is now widely accepted that hypoxia decreases theopen probability of potassium channels. A first report in1998 showing hypoxic inhibition of voltage-gated potas-sium channels in rabbit type I cells (466) was followed bynumerous other studies, but here we only highlight majorfindings. Three main categories of O2-sensitive potassiumchannels in type I cells have been described. 1) Channelsthat produce voltage-gated K� currents (Kv channels), ofwhich several isoforms have been suggested to be O2

sensitive (reviewed in Ref. 470). 2) Large-conductance K�

channels that are activated by both voltage and intracel-lular Ca2� (KCa2�, maxi-K, or BK channels; reviewed inRef. 594). 3) Voltage-independent two-pore domain back-ground (K2P) channels that contribute to the resting mem-brane potential and of which several isoforms are de-tected in glomus cells (861): K2P3.1 (TASK-1), K2P4.1(TRAAK), K2P5.1 (TASK-2), and K2P9.1 (TASK-3 ). TASK-1(TWIK-1-related acid-sensitive K� channel; TWIK channel,tandem pore domain weak inward rectifying potassiumchannel) is the most likely candidate to operate as O2-sensitive background K� channel (reviewed in Refs. 98,196, 788). Different species appear to have distinct typesof oxygen-sensitive potassium channels. For example, inrabbits, Kv isoforms are the predominant types, but in therat carotid body, significant roles of both maxi-K andTASK-1 have been demonstrated (98, 278, 594 for furtherreferences).

2. Carotid body neurotransmitters

The carotid bodies contain a variety of neurotrans-mitters that are either stored in vesicles or appear asby-products of enzymatic processes in the cytosol wherethey can influence the signal-transduction cascade (426,558). The first category comprises the biogenic aminesacetylcholine, dopamine, norepinephrine, serotonin, andhistamine; (neuro)peptides such as substance P, endothe-lins, and angiotensin II; and amino acids such as GABAand the purine ATP (see Refs. 394 and 438 for histamineand Ref. 426 for further references). Nitric oxide (NO) andcarbon monoxide (CO) are important representatives ofthe second category, synthesized with the aid of theheme-containing enzymes NO synthase and heme oxygen-ase, respectively. For an extensive discussion of the roleof all these neurotransmitters and modulators, the readeris referred to several excellent reviews (425, 426, 558, 621,701, 868). With regard to the overall stimulus-transductioncascade, there seems to be consensus, albeit not withoutcontroversy, that hypoxia induces the release of acetyl-choline and ATP and that these neurotransmitters are

FIG. 2. Stimulus transduction cascade in oxygen-sensitive cells inthe carotid body in hypoxia. A decrease in oxygen tension in oxygen-sensitive carotid body type I cells is detected by primary oxygen sensorsthat rapidly communicate with potassium channels (mitochondria couldalso sense changes in PO2 and communicate with potassium channels).This leads to closure of these channels, which results in membranedepolarization, influx of Ca2� via voltage-dependent Ca2� channels, andsubsequently to a release of neurotransmitters, whereby acetylcholineand ATP are thought to excite the afferent nerve endings of fibers thatrun in the carotid sinus nerve. NTS, nucleus of the solitary tract.



responsible for the excitation of the afferent nerve end-ings (350, 559, 644, 645, 799; see Refs. 558 and 868 forfurther references).

3. O2 sensing in the carotid bodies

Over the last decades two main hypothesis of O2

sensing have been at the foreground: 1) ion channels, andpotassium channels in particular, as the initiators of thetransduction cascade, referred to as the “membrane hy-pothesis”; and 2) heme proteins as oxygen sensors, re-ferred to as the “metabolic” or “mitochondrial” hypothesis(278, 279, 430, 620).

The role of potassium channels in type I cell mem-brane depolarization is now well established (see above).In excised membrane patches, this hypoxia-induced de-polarization is still present, but in the course of time asignificant rundown occurs, indicating that O2 sensingmay be a membrane-delimited process requiring cytosolicfactors to be preserved for longer time periods (800, 842).Whether or not potassium channels possess structuralcomponents required for direct oxygen sensing, many ofthem appear to be redox sensitive (392, 409, 468, 603), butit is unlikely that hypoxia inhibits the current throughthese channels by oxidizing or reducing functional chan-nel protein components (reviewed in Ref. 282). Generally,reductants and oxidants decrease and increase, respec-tively, the open probability of rapidly inactivating O2-sensitive potassium channels in particular, but the effects onmaxi-K channels are more variable (references in Ref. 282).

For several decades, mitochondrial cytochromeswere considered potential O2 sensors providing the linkbetween type I cell metabolism and afferent nerve activ-ity, but to date, their exact role is still a subject of discus-sion (recently reviewed in Refs. 122, 295, 296, 415, 465,837). Generally, with a few exceptions, inhibitors of mi-tochondrial complex proteins inhibit currents throughpotassium channels, depolarize type I cells, and can oc-clude the hypoxic response (references below). However,inhibition of the cytosolic heme-containing enzymesNADPH oxidase, heme oxygenase 2, and NO synthasemay also have these effects and may thus also serve ascandidate O2 sensors.

1. Heme oxygenase 2

After demonstrating the presence of heme oxygenase2 (HO-2) in rat and cat carotid bodies, Prabhakar et al.616) showed that blockade of HO-2 increased afferentcarotid sinus nerve discharge in the cat that was revers-ible with CO. Recently, Kemp, Peers, and co-workers(844) showed that HO-2 coimmunoprecipitated with het-erologously expressed maxi-K channels in human embry-onic kidney cells and that blocking the enzyme with smallinterference RNA abolished channel inhibition by low PO2

(844). In rat carotid body cells, maxi-K channels are in-

hibited by low PO2 (593, 853). In carotid body cells iso-lated from neonatal (8–11 day) rats, substrates of HO-2(heme and NADPH) activated native maxi-K channels.Since increased HO-2 activity generates CO, that by itselfactivates maxi-K channels, it was suggested that the en-zyme acts as an O2 sensor (844). How CO specificallyinteracts with maxi-K channels is unknown.

At an age �3 mo, mice deficient in HO-2 (HO-2�/�)show a carotid body phenotype typical for oxidativestress with enlargement of the organ and downregulationof maxi-K channels and a marked upregulation of tyrosinehydroxylase, suggesting that HO-2 could be a potential O2

sensor (574, 854). Type I cells isolated from HO-2�/� micehave a normal catecholamine secretory response to hyp-oxia, which may suggest that HO-2 is not a key O2 sensorand maxi-K channels are not required for a normal secre-tory response to occur (574). Carotid sinus nerve record-ings were not performed in this study, so it remains to beseen whether a normal secretory response also predicts anormal nerve response. Although at the level of type Icells the release of catecholamines is considered a mea-sure of type I cell activation (e.g., Ref. 574), changes intheir release do not always mirror alterations of carotidsinus nerve activity (100, 185, 350, 351 and referencestherein). It is interesting that HO-2�/� mice showed areduced poikilocapnic HVR despite the fact that, com-pared with wild types, they were challenged with a muchmore severe hypoxic stimulus because of their lowerinitial arterial PO2 (4). So it seems reasonable to concludethat to date the identity of HO-2 as an oxygen sensorremains unsettled. Finally, concerning the participation ofmaxi-K channels in oxygen sensing, it is worth noting thatthese channels tend to be activated at a membrane poten-tial more positive than that seen in type I cells. Therefore,in order for the maxi-K channel to be the initial sensor, itwould have to be open at or near the glomus cell restingmembrane potential.

2. NADPH oxidase

Acker and co-workers (2, 3) suggested a role of themembrane-bound extra-mitochondrial enzyme complexNADPH oxidase using NADPH as an electron donor thatproduces O2

� in proportion to PO2 and proposed that ROSwould activate nearby potassium channels in normoxiabut less so in hypoxia when NADPH oxidase activitywould be reduced (reviewed in Ref. 3). Several NADPHoxidase isoforms were identified in carotid body cellsincluding those containing the subunits gp91phox andp47phox (references in Ref. 3). Later, NADPH oxidase 4was identified as one of these isoforms in glomus cells(184, 276). Diphenyliodonium (DPI), an unspecific NADPHoxidase inhibitor, increased basal carotid sinus nerve activ-ity in the rat and inhibited the hypoxia-induced increase innerve activity (146). In an in vitro perfused carotid body



preparation from the rat, hypoxia reduced a cytochromeb558 isoform whilst the specific NADPH oxidase inhibitor4-(-2-aminoethyl)-benzenesulfonyl fluoride (AEBSF) increasedcarotid sinus nerve activity and occluded the hypoxicresponse (424). Transfection of TASK-1 channels in hu-man embryonic kidney (HEK) cells endogenously ex-pressing NADPH oxidase 4 resulted in colocalization ofboth in the plasma membrane; overexpression of theenzyme enhanced the hypoxia-induced TASK-1 currentinhibition in these cells, but eliminating the enzyme abol-ished the hypoxic response (439).

In the HEK cell model, the concept of ROS-mediatedTASK-1 activation was recently challenged because it ap-peared to be determined by O2 binding to the heme moi-ety of the enzyme rather than by the production of ROS(582). Other evidence did not support a crucial role forNADPH oxidase. For example, in both rat and rabbitcarotid body cells, specific inhibitors of the enzyme didnot affect catcholamine release in normoxia or its in-crease in hypoxia; in addition, DPI seemed to have effectsindependent from NADPH oxidase inhibition (561). Com-pared with wild types, adult mice deficient in gp91phox donot show altered responses of ventilation, carotid sinusnerve activity, potassium currents, or intracellular [Ca2�]to hypoxia (306, 310, 667). Very interesting results wereobtained from p47phox�/� mice: 1) type I cells fromknockout mice did not show the hypoxia-induced in-crease in [ROS] that was observed in cells from wildtypes, but displayed an enhanced depression of potassiumcurrents and a substantial greater rise intracellular[Ca2�]; and 2) both the HVR and hypoxic carotid sinusnerve responses in knockout mice were substantiallygreater than in wild types (184, 310, 678). From theseobservations it was concluded that hypoxia induces a risein [ROS] in type I cells with NADPH oxidase as theirsource and that these ROS modulate (increase) the openprobability of (Kv) potassium channels, rather than beingdirectly involved in oxygen sensing.

3. NO synthase

At least two NO synthase (NOS) isoforms have beenidentified in the carotid bodies: 1) NOS-1 (neuronal NOS)located in sensory and autonomic nerve fibers innervatingvessels and lobules of glomus cells and 2) endothelialNOS (NOS-3; references in Refs. 109, 426). Aspecific NOSinhibition increases basal carotid sinus nerve dischargeand enhances hypoxic carotid body sensitivity and theHVR, whilst NO donors do the opposite (290, 619 forfurther references). Whether NOS is activated or inhibitedduring hypoxia is controversial (references in Ref. 426).NOS-1�/� mice show an enhanced hypoxic carotid bodysensitivity and HVR, supporting an inhibitory role of NO(389). NO may modulate carotid body sensitivity by alter-

ing blood flow and/or exerting direct effects on type I cells(references in Refs. 109, 426, 619).

4. Mitochondrial complex proteins and

complex-derived ROS

The role of ROS in oxygen sensing is the subject of along debate with many controversial data and differentviews. Partly, the lack of consensus is caused by thedifficulty of measuring ROS, by local differences in theirproduction in various microdomains of the cell (cell mem-brane, cytosol, nucleus, mitochondria), and by differ-ences between cell types (see Refs. 276, 282 for refer-ences).

Generally, inhibitors and uncouplers of the electrontransport chain inhibit background potassium channelsand cause membrane depolarization, voltage-gated Ca2�

entry, and neurosecretion. They also stimulate the carotidbodies and, provided the inhibition of electron flux iscomplete, occlude the hypoxic response (851). Originallyit was thought that hypoxia would reduce cytochrome c

activity of complex IV of the electron transport chain,which then would cause mitochondrial depolarizationand Ca2� release (194). Later the rise in intracellular[Ca2�] appeared to depend on extracellular Ca2�, and therole of mitochondria as oxygen sensors lost support (99,465, 802). In addition, several blockers of the electrontransport chain (rotenone in particular) have nonspecificeffects on potassium channels (573).

A modulating role of mitochondrial ROS, and in thiscontext the influence of cellular redox state on oxygensensing was investigated by Gonzalez et al. (281). Thecellular redox state was assessed by measuring the ratiosof reduced (GSH) to oxidized glutathione (GSSG), fromwhich the redox potential (EGSH) can be calculated. EGSH

is considered the main cellular buffer of ROS and repre-sentative for the general redox environment of cells (281).Briefly, Gonzalez et al. (281) found the following. 1) Incalf and rabbit carotid body cells, hypoxia did not reduceEGSH. 2) The antioxidant and GSH precursor N-acetylcys-teine increased EGSH in normoxia in these cells withouteliciting an increased catecholamine release. 3) N-acetyl-cysteine also did not alter the enhanced neurotransmitterrelease induced by hypoxia and high extracellular [K�].4) In rat carotid body cells, specific inhibition of mito-chondrial complexes I, II, III, and IV all reduced EGSH,and, with the exception of the irreversible complex IIinhibitor 3-nitropropionate, decreased cellular ATP levelsand increased catecholamine release. N-acetylcysteine pre-vented these changes in redox state without influencingthe effects on [ATP] and neurosecretion. 5) The mito-chondrial uncoupler dinitrophenole activated rat carotidbody cells without affecting [ATP] and EGSH (275, 281,282, 679).



The conclusion from these studies by Gonzalez et al.(281) is that oxygen sensing in carotid body cells occursindependently from changes in cellular redox state. Infact, this was supported at the level of afferent nervedischarges in superfused rat carotid bodies (666). How-ever, given the fact that N-acetylcysteine restores theredox state but does not prevent ROS production per se,the question now arises whether mitochondrial ROS mayplay any (indirect) role, for example, by influencing ATPproduction. As explained in insightful reviews by Gonza-lez and co-workers (276, 282), there is much controversyas to whether hypoxia decreases or increases mitochon-drial [ROS]. Briefly, these authors provided reasonablearguments why, unlike some other cell lines, type I cellsmost likely do not increase mitochondrial ROS produc-tion in acute hypoxia and why the limited observed rise incellular [ROS] in type I cells is most likely due to in-creased NADPH activity (310). Yamamoto et al. (860) alsoobserved an increase in [ROS]. Maintenance of the redoxstate in the face of increased ROS production via thispathway would occur by an increased production ofNADPH (reducing equivalent) along the hexose mono-phosphate pathway (276, 310). Mitochondrial inhibitorswith divergent effects on local ROS production all hadstimulating effects on type I cells and occluded the hyp-oxic response to hypoxia, suggesting no crucial role ofmitochondrial ROS on O2 sensing (851).

An absence of a direct involvement of ROS and thecellular redox state in oxygen sensing does not precludea modulating effect of oxidants and reductants. In themodel proposed by He et al. (310) and Dinger et al. (184),ROS, generated by NADPH oxidase activity, facilitatespotassium channels. In rat type I cells, inhibition of thecytochrome P-450 enzyme system using NAD(P)H as elec-tron donor prevents hypoxia-induced inhibition of potas-sium channels (304). As suggested by Kline et al. (389), arise in [ROS] could decrease the bioavailability of NO,increasing hypoxic sensitivity. Mice deficient in neuronalNOS have an augmented HVR (389).

5. Individual mitochondrial complexes

To date, the identification of one single mitochon-drial complex protein as the oxygen sensor in type I cellshas been elusive. Light absorption photometry has iden-tified a cytochrome, a592, as a unique component of ca-rotid body cytochrome-c oxidase (complex IV) with aredox sensitivity that depends on the PO2 in a nonlinearfashion (3, 733). By inducing a shortcut of electron flowwithin complex IV, this low O2 affinity cytochrome wouldcontribute to mitochondrial depolarization and intracel-lular Ca2� regulation. However, the intracellular [Ca2�]response to hypoxia depends on extracellular, not intra-cellular, calcium stores (99, 465, 802), whilst the ATPsynthase inhibitor oligomycin, despite causing mitochon-

drial hypopolarization (194), has a similar stimulatoryeffect on type I cells as mitochondrial complex inhibitorsand uncouplers which cause mitochondrial depolariza-tion (851). Despite this, the idea of heme-containing oxy-gen sensors with different O2 affinities in distinct mi-crodomains of the cell, enabling it to adapt to changes incellular PO2 profile (“oxygen redirection”), is interesting(837 for references).

In rat PC12 cells, hypoxia produced a selective andfast downregulation of complex I mRNA. This suggestedthat, apart from being involved in the flow of electronsthrough the electron transport chain and as such mayindirectly contribute to rapid changes in [ATP], complex Imay play an important role in adapting mitochondrialrespiration to prolonged periods of hypoxia (606). Inchronic intermittent hypoxia, carotid body complex I ac-tivity also appeared to be reduced (596).

Mice heterozygous for succinate dehydrogenase sub-unit D (SDHD�/�), one of the anchoring proteins of com-plex II (607), show mild carotid body hypertrophy andhyperplasia and an enhanced catecholamine release bytype I cells in normoxia. These features are associatedwith a decrease in K� conductance (due to a a decreasedCa2� sensitivity of maxi-K channels) and a persistentCa

2�

influx; the hypoxia-induced CA release was enhancedin these animals, but not significantly (607). Thus thesemice do not seem to support a crucial role of complex IIin O2 sensing, and this may fit the finding that the irre-versible complex II inhibitor 3-nitropropionate did notalter the catecholamine release and ATP content of ratcarotid body cells (275). However, carotid body-sinusnerve preparations from SDHD�/� mice were not studied.Human carriers of a missense mutation within the SDHD

gene show an altitude-related phenotype (20). Thus itremains to be seen whether there may exist a relationbetween complex II activity and oxygen sensing (for fur-ther discussion and references, see sect. IV).

6. Mitochondria and the link between metabolism and

potassium channel activity

Failure to show an indispensable role of mitochon-drial ROS in O2 sensing, and the fact that it has beenelusive to identify one of the electron transport chaincomplexes as the O2 sensor that initiates the entire trans-duction cascade, does not preclude a mitochondrial rolebecause O2 sensing could be a direct or indirect functionof mitochondrial energy metabolism and ATP synthesis(851). Mitochondrial complex inhibitors and uncouplershave variable effects on [ROS], but two effects that theyhave in common are that they decrease ATP levels andactivate type I cells (276, 851), perhaps with the exceptionof specific complex II blockers (275).

Recently, cellular energy status was linked to cellactivity in type I cells via AMP-activated protein kinase



(AMPK), which is considered the master regulator ofglucose and fatty acid metabolism (417 and 786 for refer-ences). AMPK is a cellular energy sensor and sensitive tothe cellular AMP-to-ATP ratio. Metabolic stress includinghypoxia increases this ratio and activates AMPK, with theresult that catabolic processes are upregulated and ana-bolic processes inhibited (417, 786). Recently, Evans et al.(223) proposed an AMPK-mediated mechanism of O2

sensing in type I cells, and they provided substantialexperimental evidence to support their model. Their mainfindings were as follows. 1) In rat type I cells, they iden-tified an AMPK isoform that is located close enough to thecell membrane to participate in a “membrane delimited” pro-cess. 2) In type I cells, the AMPK mimetic agent AICAR(5-aminoimmidazole-4-carboxamide riboside) inhibitedTASK-1 and maxi-K channels and caused depolarizationand increased Ca2� influx whilst also increasing sensoryafferent discharge from an isolated carotid body prepara-tion. 3) These effects could be inhibited by the AMPKantagonist compound C that also reduced the effect ofhypoxia on type I cells (223, 852, 855). This model isattractive not only because it provides a link betweencellular metabolism and potassium channel inhibition butalso because it may reconcile the “membrane” and “mito-chondrial” hypotheses of O2 sensing (852).

Other, more direct, links between metabolism andTASK-1 channel inhibition are also possible. In rat type Icells, MgATP and other Mg nucleotides have a powerfulstimulatory action on background potassium channels(800). Mitochondrial complex inhibitors cause a rise in[Mg]i reflecting a fall in [MgATP]i, and the channels weresuggested to be coupled to (or possess) a low-affinity Mgnucleotide sensor (800). A direct, depolarizing effect ofadenosine via A2A receptors resulting in a rise in [Ca2�] intype I cells has also been reported (856). Depletion ofMgATP reflects increased production of AMP, providingpathways to increase [adenosine] and activate AMPK,respectively (800).

The idea of carotid bodies as metabolic sensors is notnew. For example, in an in vitro preparation, the catcarotid body is excited by inhibition of glycolysis (560).Recent studies have identified type I cells in isolation,both in coculture with petrosal ganglion neurons and infresh carotid body slices, as direct glucose sensors thatdepolarize and release neurotransmitters upon loweringof glucose levels (254, 580, 870). However, a direct effectof glucose on type I cells is not uncontroversial because inthe rat, low glucose increased spontaneous ventilation, aneffect that depended on the integrity of the carotid sinusnerve, while in isolated carotid body-sinus nerve prepara-tions it failed to have any effect on nerve discharge andthe release of catecholamines and ATP, respectively (64,141). Several studies have clearly established the role ofthe carotid bodies in glucose homeostasis, which can bemodulated by GABAergic mechanisms in the NTS (see

Refs. 440, 870 and references cited therein). Interestingly,in humans, low plasma glucose was reported to doublethe HVR, but this could be indirectly mediated by coun-terregulatory hormones (825).

In summary, various heme proteins, such as NADPHoxidase, HO-2, NOS, and mitochondrial complex proteinsmay operate in parallel or sequentially as oxygen sensors.The oxygen sensing machinery is already operative inmild hypoxia, in a PO2 range in which mitochondrial com-plex proteins, due to their low P50, will not yet be com-promised (427, 621). This suggests that more than one O2

sensor must be operative (380, 621, 800). Across (but alsowithin) animal species, various classes of potassiumchannels play a crucial role in the membrane depolariza-tion of type I cells in hypoxia by reducing their current.There are multiple ways in which these channels can beinhibited. One of these, changes in the cellular AMP/ATPratio, has recently been identified as a very sensitivemanner to couple cell metabolism to potassium channelactivity and appears to operate already well before mito-chondrial complex proteins are compromised (415). TypeI cells have an extremely complex biochemical contentand release numerous substances during hypoxia, ofwhich ACh and ATP are thought to be the principle excita-tory neurotransmitters acting at the afferent endings ofpetrosal ganglion neurons that innervate the carotid bodies.

B. Hypoxia Activates the Transcription Factor

Hypoxia-Inducible Factor-1

Chronic hypoxia induces expression of genes encod-ing factors that 1) promote tissue oxygenation (e.g., an-giogenesis); 2) sensitize oxygen-sensing mechanisms andfacilitate central processing, thus augmenting the HVR;3) increase the blood O2 transport capacity (erythropoie-tin production); and 4) adapt cellular metabolism by fa-cilitating glucose uptake (upregulation of glucose trans-porters) and promoting anaerobic metabolism whilst re-pressing but also optimizing oxidative phosphorylation.This altered expression of these and many other genes isinitiated by increased activity of numerous transcriptionfactors (reviewed in Refs. 101, 133, 149, 381, 661, 693), oneof which appears to orchestrate this master plan of geneexpression: hypoxia-inducible factor-1 (HIF-1), discov-ered by Semenza and co-workers (see Ref. 689).

HIF-1 is a heterodimer consisting of the constitu-tively expressed nuclear subunits HIF-1� and HIF-1�. Thelatter subunit, HIF-1�, undergoes (cytoplasmic) posttran-scriptional modification under the influence of the PO2

(Fig. 3; Ref. 690). HIF-1 induces many target genes encod-ing proteins that influence O2 supply, metabolism, tran-scription factors, and apoptotic and enzymatic processes(101, 133, 371, 693). In euoxic conditions, HIF-1� is bothsynthesized and rapidly degraded, but in hypoxia it is



rescued from proteasomal degradation enabling it totranslocate to the nucleus where it dimerizes with theconstitutively expressed HIF-1� to form the transcriptionfactor HIF-1. The dimer binds to the hypoxia responseelements of target genes where it recruits coactivatorproteins to promote transcription. Structure and proper-ties of HIF-1 and its subunits can be found in References689–691. HIF-1� has two paralogs: 1) HIF-2�, which isalso regulated by O2, and, when dimerized with HIF-1�,promotes the transcription of both overlapping and dis-tinct target genes (references in Ref. 694); and 2) HIF-3�with a largely unexplored function (369, 694). In many ifnot all body cells, HIF-1 is upregulated during hypoxia.

There are many pathways of posttranscriptional modifica-tion of HIF-1� (reviewed in Refs. 133, 369, 692, 694). Here weonly briefly discuss the controversial role of ROS.

1. HIF-1� stabilization via increased [ROS]

HIF-1� contains an oxygen-dependent degradationdomain with highly conserved specific proline residues(Pro402 and Pro563 in mice, Pro402 and Pro564 in humans)that are hydroxylated in normoxic conditions. This hy-droxylation is catalyzed by enzymes containing a prolylhydroxylase domain (PHD). Currently three PHD familymembers are known, PHD1, PHD2, and PHD3, of whichPHD2 is considered the primary HIF prolyl hydroxylase in

normoxia (63, further references in Ref. 369). PHD2 is adioxygenase that uses molecular oxygen and �-ketogluta-rate as substrates, generates CO2 and succinate as by-products, and utilizes Fe2� and ascorbic acid as cofactors.Prolyl hydroxylation of HIF 1-� is required for binding ofthe von Hippel-Lindau tumor suppression protein pVHLwhich targets the complex for ubiquitin-mediated degra-dation by interacting with a multiprotein ubiquitin ligasecomplex thus creating a recognition site for proteasomaldegradation (Fig. 3; Refs. 352, 353, 499). The current viewis that during hypoxia the activity of PHD2 is reduced bya combined effect of substrate (oxygen) limitation andinhibition of the Fe2� catalytic center by ROS possiblyproduced by complex III of the mitochondrial electrontransport chain (55, 122, 123).

2. HIF-1 stabilization via decreased [ROS]

An opposite role of ROS in HIF-1 stabilization wasproposed by Huang et al. (341; see Ref. 276 for furtherreferences). In human Hela cells, they showed increasedstability of the HIF-1� subunit in hypoxia and a rapidbreakdown in normoxia. In hepatoma 3B (Hep3B) cells,hypoxia also induced HIF-1� stabilization and an in-creased expression of erythropoietin mRNA, and botheffects were dose-dependently blocked by preexposure tohydrogen peroxide. Oxidizing HIF-1� inhibited HIF-1

FIG. 3. Influence of prolyl hydroxylases on hypoxia inducible factor (HIF) stability. In normoxia, with sufficient oxygen availability, prolylhydroxylases [PHD; PHD2 is the isoform that is considered the primary HIF prolyl hydroxylase in normoxia and utilizes iron and ascorbic acid (asc)as co-factors] hydroxylate crucial proline residues in the degradation domain of HIF-1�. This enables the von Hippel-Lindau protein (pVHL) to bindto these residues and ubiquitylate (ubi) the protein for proteasomal degradation. As a consequence, the nucleus contains low levels of HIF-1� andthus also of HIF-1, the heterodimer of HIF-1� and HIF-1�. In hypoxia, PHD2 activity is limited by low PO2. Functional activity of HIF-1� is regulatedby factor inhibiting HIF (FIH) which, in an oxygen-dependent way, hydroxylates a crucial asparagine residue in the COOH-terminal domain (notshown). In hypoxia, HIF-1� is stabilized, activated, and translocated to the nucleus. An important additional step in HIF activation in hypoxia,correlated with the induction of target genes, is the activation of the coactivators p300 and cAMP response element binding protein (CREB) whichcombine with the heterodimer. This complex binds to the hypoxia response element (HRE) of target genes while additional coactivators are alsorecruited. �KG, �-ketoglutarate; succ, succinate; ROS, reactive oxygen species.



DNA binding, and this could be reversed by reducingagents, possibly indicating that hypoxia stabilizes HIF-1�by reducing specific cysteine sulfhydryls in the molecule,whilst hypoxia and hydrogen peroxide would reverse this(341 and references therein).

What is the scenario in glomus cells? In carotid bodiesfrom rats exposed to chronic intermittent hypoxia, HIF-1�was upregulated and complex I activity was reduced, whilst[ROS] and hypoxic carotid body sensitivity were increased(597, 600). HIF�/� mice did not display these adaptations,and together these findings would suggest stabilization ofHIF-1� by increased [ROS] resulting in the augmented HVR(600; see also sect. VG). However, with the assumption ofenhanced ROS production by mitochondrial complex inhib-itors, the finding that in type I cells from rat, complexinhibitors significantly reduced HIF-1� upregulation inducedby superfusion with a hypoxic solution for �45 min, wouldrather support the notion of HIF stabilization with reduced[ROS] (22). This model could also provide a potential expla-nation for the increase in HVR in humans after chronictreatment with antioxidants (321, 608). Notably, the antiox-idant N-acetylcysteine not only augmented the HVR in youngsubjects but also increased the production of erythropoietin,indicating upregulation of HIF-1 (321). Reduction of theisocapnic HVR by low-dose anesthetics or the carbonic an-hydrase inhibitor acetazolamide could be reversed by anantioxidant cocktail consisting of �-tocopherol and ascorbicacid (777, 780, 781). This reversal occurred in the setting ofan acute experiment, so presumably it was unlikely to bemediated via increased stabilization of HIF-1�. It was spec-ulated that the reversal could be due to independent, inhib-iting effects of the antioxidants on potassium channels, al-though other explanations are possible (777, 780, 781). Thestimulating effect of chronic ascorbic acid on the HVR ishard to reconcile with a single role of this antioxidant ascofactor of PHD2. An effort to eliminate the other cofactorof PHD2, Fe2�, by infusing the iron chelator desferrioxam-ine (DFO) for 8 h did not alter the HVR in humans (639). Thiscould be due to failure of DFO to penetrate into type I cells,and this is supported by data from in vitro perfused carotidbody from the rat (168). Finally, in humans, iron infusion didnot influence the augmentation of the HVR that is normallyobserved after chronic hypoxia (see sect. VF), and this could bedue to the inability of iron to reach the interior of type I cells(711).

C. Central Effects of Hypoxia

1. General effects of central nervous system hypoxia

on breathing

In the central nervous system (CNS), moderate hyp-oxia causes a generalized reduction in neuronal excitabil-ity and decrease in intracellular [ATP]. Depending on theduration and severity an initial transient alkalosis, an

increase in potassium conductance and hyperpolarizationmay ensue acting to conserve energy. Some neurons,however, depolarize in hypoxia and if some other condi-tions are also met may be classified as central O2 sensors(see sect. IIIC2). Severe, permanent hypoxia is generallyfollowed by intracellular acidosis, influx of Ca2� and Na�,efflux of K�, persisting depolarization, further decreasesin intracellular [ATP], and eventually cell death (435).Hypoxia in the brain (central hypoxia) affects breathing ina way that depends on the integrity and activity of thecarotid bodies. In addition, several brain regions areclaimed to contain oxygen sensors that may play a role inthe HVR in some circumstances.

Shortly after carotid body denervation, cats anesthe-tized with chloralose (urethane) show decreases in ven-tilation and phrenic nerve activity with both hypoxemicand CO hypoxia (314, 361, 507). With severe hypoxia, thephrenic neurogram displays a gasplike pattern (507). An-imals anesthetized with thiamylal, however, failed toshow ventilatory depression immediately after carotidbody excision, even with severe hypoxia (710). Carotidbody-denervated rats anesthetized with urethane do alsoshow a clear-cut hypoxia-induced decrease in respiratorymotor activity (436). A depressant response of CNS hyp-oxia is also manifest in chloralose-urethane anesthetizedcats subjected to isolated brain stem perfusion with hyp-oxic blood, even when the PCO2 of the peripheral blood iskept hypercapnic (794, 823). In anesthetized cats sub-jected to carotid body denervation or artificial brain stemperfusion, hypoxic ventilatory depression may be causedby both a fall in medullary PCO2, the metabolic stress oflactic acidosis and/or decrease in [ATP], as well as inhib-itory neurotransmitters (62, 436, 540, 543). The absence ofa depression in thiamylal-anesthetized animals (710)might suggest involvement of a GABAergic mechanism.Artificial brain stem perfusion via one vertebral artery inthe cat supplies the pons, medulla, and cerebellum with“central blood” (59) so central hypoxic depression in thispreparation must originate from one or more of theseregions, presumably pons and/or medulla.

At least in cat and rat, the effect of (severe) CNShypoxia on breathing differs between the anesthetizedand awake states. Awake carotid body denervated catsrespond to mild hypoxia (FIO2

�14%) with mild to mod-erate hypoventilation (237, 261, 510) or virtually nochange (464). Severe hypoxia, however, results in a pro-nounced tachypnea and decrease in tidal volume resultingin either a reported decrease (261) or increase (510),respectively, in alveolar PCO2. In unanesthetized animals,a transition into unconsciousness during severe hypoxiadid not result in any change in breathing pattern, suggest-ing the diencephalon as the source of the tachypnea (510).Immediately after carotid body denervation, the awakerat may show some hypoxic respiratory depression (492),but the most common response is a limited rise in venti-



lation (570, 664, 665), raising the question whether thecarotid body-denervated rat is a suitable model to studythe isolated effect of central hypoxia. This is illustrated bythe fact that acute hyperoxia leads to an increase inventilation in this model, indicating a depressant effect ofcentral hypoxia (570). With prolonged central hypoxia,awake goats and dogs undergoing selective carotid bodyperfusion with normocapnic-normoxic blood showed anincrease in ventilation (165, 836) and an absence of respi-ratory depression (706), respectively. Sleeping dogs in-creased their ventilation with central hypoxia, but aftercarotid body denervation there was no response (151).Soon after carotid body denervation, ponies showed nohypoxic ventilatory depression; however, this does notpreclude an inhibitory effect of low central PO2 duringchronic hypoxia because acute hyperoxia induced hyper-ventilation in these animals (239).

Finally, note that the depressant effect of centralhypoxia in anesthetized cats with denervated carotid bod-ies is different mechanistically from HVD in intact cats,with the latter being initiated by the carotid bodies (seealso section VC).

A) CENTRAL O2 SENSORS AND HVR. Before a neuron thatdepolarizes in hypoxia can be classified as intrinsicallysensitive to O2, at least four criteria must be met. 1) Thedepolarization should occur independently from synapticevents. 2) Activation of these neurons is independentfrom input from the arterial chemoreceptors. 3) Withinthe range of physiological O2 levels, the sensitivity ofthese neurons should be much greater than that of otherneurons. 4) At least under certain physiological condi-tions, stimulation by low oxygen of these neurons shouldbe able to elicit an appropriate (cardio)respiratory re-sponse. Below we summarize which regions belonging tothe respiratory neuronal network are claimed to possessneurons that are intrinsically sensitive to oxygen.

I) Location of central O2 sensors. Exposing anesthe-tized cats to 10% O2 excited �20% of neurons studied inthe caudal hypothalamus, independently from carotid si-nus nerve integrity (183). Ninety percent of these cellshad a cardiovascular and/or respiratory rhythm, even af-ter carotid body denervation (183). In brain slice prepa-rations, �80% of caudal hypothalamic neurons respondedto hypoxia independently from synaptic blockade (182,329). O2-sensitive neurons in the caudal hypothalamuswere also found in the rat (330, 671). Oxygen-sensitivecells in the hypothalamus may also reside rostrally inparvocellular cells of the paraventricular nucleus (385).

Located in an area from which O2-sensitive neuronsin the caudal hypothalamus can be antidromically acti-vated, the pontine periaqueductal gray (PAG) containsmany neurons intrinsically sensitive to O2 (410, 671). De-pending on the stimulus location, PAG activation elicitscardiovascular responses that are mediated via the ros-troventrolateral medulla (RVLM; references in Ref. 410).

In slice preparations, the locus coeruleus in the dorsolat-eral pons contains many neurons that are inhibited byhypoxia and a smaller percentage that is depolarized, butit is uncertain whether this represents intrinsic O2 sensi-tivity (545, 864).

Both stimulation of the carotid bodies and centralhypoxia activates neurons in the RVLM, and in the lattercase, this activation is resistant to local glutamate recep-tor antagonism and denervation of the carotid bodies. Sothese cells may be intrinsically O2 sensitive, and this no-tion was corroborated by the finding that in slice prepa-rations RVLM neurons retained their depolarizing re-sponse to hypoxia or cyanide even with synaptic block-ade (741, 742, 744, 745). Nolan and Waldrop (553)reported activation of �64% of neurons studied in thesame RVLM area when exposing anesthetized spontane-ously breathing rats to hypoxia. Many of these neuronsdisplayed a temporal discharge pattern related to thecardiac and/or respiratory cycle; a similar percentage wasactivated after bilateral transection of both the vagus andcarotid sinus nerves and cervical sympathetic and aorticdepressor nerves (553). Brain slices including this RVLMarea and from a region projecting to the spinal cordcontained neurons the majority (�70–80%) of which wasexcited by low PO2 even after synaptic blockade; however,a substantial amount (�20%) was inhibited (553). O2 sen-sitivity in the RVLM is not a unique property of spinallyprojecting barosensitive neurons because it seems to bewidely distributed in this area with many neurons eitherexcited or inhibited by hypoxia (378). Taken together, thecaudal hypothalamus, PAG, and RVLM might act as anaccessory circuit to amplify ventilatory and cardiovascu-lar responses to hypoxia (410).

Located just rostral to the C1 region of the RVLM, thepre-Botzinger complex contains oxygen-sensitive cells.Local application of NaCN into this area augmented in-spiratory output including gasping in the phrenic neuro-gram (718). Dissociated cell cultures with neurons fromboth the C1 region and pre-Botzinger complex containedcells of which �60% responded to NaCN with membranedepolarization and an increase in firing frequency, but�40% with inhibition, and these responses were resistantto synaptic blockade (500). Both RVLM regions expressHO-2 (501), an enzyme that in the carotid bodies may beinvolved in oxygen sensing (844; see also sect. IIIC1). Theexpression of HO-2 in these regions is limited to oxygen-sensitive neurons; blocking enzyme activity abolishes the re-sponse of these neurons to hypoxic hypoxia and NaCN (153).

Oxygen-sensitive cells were also found in the NTS. Incoronal slices from the level of the obex, 75% of theneurons tested responded to hypoxia with hyperpolariza-tion, decreased input resistance, and decreased activity,while 25% showed increased activity (583). Hypoxia inthese slices also decreased postsynaptic potentials elic-ited by electric stimulation of afferents in the solitary



tract, suggesting hypoxic gating of incoming afferent in-put (583). O2-sensitive neurons in the NTS were alsofound by Nolan and Waldrop (553).

Finally, the hypoglosssal nucleus also contains neu-rons that depolarize upon exposure to hypoxia (297).

The mechanism of oxygen sensing in brain stem andhypothalamic neurons is largely unknown. Does a uni-form O2 sensor operate that would facilitate identificationand localization of these neurons? Some authors haveproposed a lack or paucity of ATP-sensitive potassiumchannels as a potential basis of oxygen sensing (e.g., Ref.864), which could make sense because hypoxia decreasesintracellular [ATP] in the brain (435, 436). Also, severalother potassium channel species have been suggested tobe involved in central oxygen sensing, as well as variouscalcium and sodium channels (329, 745, 864; reviewed inRef. 544). Recently, D’Agostino et al. (153) reported thatexcitation by hypoxia of neurons in the C1 region andpre-Botzinger complex of rats critically depended on theactivity of HO-2.

2. Role of central O2 sensors in the HVR?

The contribution of central oxygen sensors to theHVR is unknown. Does the absence of an increase inventilation shortly after carotid body denervation as ob-served in cats and dogs imply a trivial contribution ofcentral O2 chemoreceptors to the HVR of an intact organ-ism? Taking into account that the carotid body may pro-vide oxygen-sensitive neurons in the RVLM with a tonicexcitatory input (741, 744), a preparation with denervatedcarotid bodies would not be the most suitable model toget more insight into this issue. Could the increase inventilation with isolated central hypoxia in goats anddogs with intact, separately perfused carotid bodies bedue to a facilitating peripheral chemoreceptor input, ex-plaining, for example, the elimination of the response indogs after carotid body denervation (151)? Selectivelyperfusing the carotid bodies with either hypoxic or hyper-oxic blood to subsequently measure the effects of centralhypoxia could give some clue. In anesthetized cats andawake goats, the reverse paradigm, i.e., varying the pe-ripheral PO2 while keeping central PO2 constant at differ-ent levels, did not reveal an influence of brain stem PO2 onthe ventilatory response induced by carotid body hypoxia(i.e., no peripheral-central O2 interaction; Refs. 165, 794).

In cats, ponies, and rats, the respiratory system dis-plays a remarkable plasticity in restoring the HVR afterdenervation of the carotid bodies (70, 492, 664, 710). Ratsdisplayed a complete recovery of the HVR 90 days aftercarotid body denervation which was temporally related tochanges in tyrosine hydroxylase expression in variousbrain stem areas containing central O2 sensors, e.g., theA2/C2 region in the caudal NTS and the A1/C1 region inthe RVLM (665, 665). Complete HVR recovery on a com-parable time scale was also observed in rats after lesion-ing glutamate receptor-containing neurons in the NTS,initially resulting in a 70% reduction of the response (130).It is tempting to speculate that tyrosine hydroxylase-con-taining central O2 sensors could serve as a “back-up”mechanism to detect low oxygen levels in cases whereperipheral chemoreceptors are functionally eliminated,but further studies are necessary to confirm this.

Diencephalic structures play a role in the severehypoxia-induced tachypnea in awake cats with dener-vated carotid bodies (510). Disinhibition of cortical-hypo-thalamic interactions by low PO2 contribute to this phe-nomenon. Because the caudal hypothalamus containsneurons excited by O2 independently from the carotidbodies and provide the RVLM with an excitatory input,these O2 sensors could contribute to it (331, 813). Directstimulation of the caudal hypothalamus increases respi-ratory frequency (813).

3. CNS neurotransmitters and HVR

A) GLUTAMATE, GABA, GLYCINE, ADENOSINE, AND ATP. Theentire respiratory network participates in the HVR, so it isnot surprising that it can be influenced by numerousneurotransmitters/modulators and receptor agonists/an-tagonists. Respiratory midbrain and brain stem regions,each with a separate contribution to the HVR, contain avariety of neurotransmitters (recently reviewed in Ref.732). In Figure 4, the major medullary and pontine regionsare shown that participate in the HVR. Classical media-tors of the HVR are GABA, glutamate, and adenosine, andtheir temporal profiles and roles were reviewed previ-ously (104, 327, 379, 502, 540, 649, 752). Generally, gluta-mate is considered an excitatory neurotransmitter in theHVR, while GABA, glycine, and adenosine are inhibitory(18, 104, 327, 379, 565). This is probably an oversimplifi-cation, because one particular transmitter or modulator

FIG. 4. Major brain stem regions involved in the ventilatory response to hypoxia. Major cell groups involved in the ventilatory response tohypoxia are shown in coronal sections between rostrocaudal levels of the rat brain stem from Bregma �9.16 to �14.60 mm. 1, A1 noradrenergicregion; 2, caudal ventral respiratory group; 3, A2 noradrenergic region; 4, raphe pallidus; 5, raphe obscurus; 6, medial subnucleus of the nucleus ofthe solitary tract; 7, commissural subnucleus of the nucleus of the solitary tract; 8, rostral ventral respiratory group; 9, C1/A1 adrenergic/noradrenergic region; 10, C2 adrenergic region; 11, C1 adrenergic region; 12, pre-Botzinger complex; 13, Botzinger complex; 14, raphe magnus; 15,retrotrapezoid nucleus; 16, A5 noradrenergic region; 17, locus coeruleus; 18, parabrachial nucleus, medial part; 19, Kolliker-Fuse nucleus. Py,pyramidal tract; scp, superior cerebellar peduncle; Sol, nucleus of the solitary tract; V, motor nucleus of the trigeminal nerve; VII, facial nucleus.[Sections redrawn from Paxinos and Watson (587).]



can have distinct effects on the HVR, depending on thedose, route of administration, and site of application inthe CNS. For example, in rats, at an ambient temperatureof 30°C, the GABA agonist muscimol reduces the HVRwhen microdialyzed into the NTS but augments it whenapplied in raphe magnus (539, 772). Recently, a modula-tory (facilitatory) role for ATP has been described, inwhich it is released in ventral medullary surface areas inthe secondary phase of the HVR and possibly acts in thePre-Botzinger complex to increase respiratory frequencyvia activation of metabotropic (P2Y) receptors (284, 471).Below we summarize the most relevant data, mainly col-lected over the last decade, concerning the important roleof monoamines (with emphasis on serotonin) and NO inthe HVR. Figure 4 can be used as a reference to envisagethe different brain stem regions where these transmittersexert their action.

B) MONOAMINES AND THE HVR. The absence of a signifi-cant effect of catecholamine depletion on the HVR maysuggest that norepinephrine and/or dopamine do notfunction as primary neurotransmitters in hypoxia (503).However, monoamines are widely distributed in the brainand as such may be expected to contribute in a complexway. Mice lacking the gene for the dopamine transporter(DAT�/�) show a reduced poikilocapnic HVR (805), butdata from these mice are difficult to interpret becauseboth carotid body and central dopamine levels will beincreased. The same problem is encountered in thosestudies using dopaminergic blockers that act both periph-erally and centrally. As is also the case for serotonin (seebelow), the influence of central dopamine on the HVR isnot limited to the ventilatory component of the responsebut also concerns the hypometabolic/hypothermic part ofit. For example, infusion of haloperidol in the third cere-bral ventricle of rats (aimed at exerting effects in thehypothalamus) reduced the drop in hypoxia-induced bodytemperature without affecting the ventilatory component(35). Dopamine 1 receptors in the anteroventral preopticregion of the hypothalamus in particular seem to be in-volved in this response (36).

Several regions belonging to the central neuronalrespiratory network contain noradrenergic/adrenergicneurons, notably the A2 noradrenergic cell group in thecaudal NTS where carotid body afferents terminate, the A5noradrenergic cell group in the ventrolateral pons, the A6group in the locus coeruleus in the dorsolateral pons, andthe A1/C1 noradrenergic cell groups in the ventrolateralmedulla (255, 803). All these regions may contribute to themagnitude and/or shaping of the HVR. For example, Coleset al. (140) reported that lesions in the A5 group reduce orabolish the reduction of respiratory frequency followinghypoxia. The A2 group, containing neurons responsive todopamine, is considered to play a modulatory role in theventilatory acclimatization to chronic hypoxia; the in-crease in CNS gain (CNS gain is defined in sect. VF4) in

chronic hypoxia may be related to (temporal) upregula-tion of local tyrosine hydroxylase (references in sect. VD).In the caudal NTS, norepinephrine can influence bothgluatamatergic and GABAergic transmission via presyn-aptic receptors and may thus have variable effects on theHVR (873 and references cited therein). Neurons in the A1and C1 groups are activated in acute hypoxia but do notseem to have an altered catecholamine turnover inchronic hypoxia (221, 722, 782). The contribution of A6neurons will be discussed in the subsection below on therole of central NO.

Serotonergic neurons in the brain stem are mainlyconcentrated in medullary, pontine, and mesencephalicraphe nuclei (references in Ref. 322) and may be of spe-cial interest for the HVR for several reasons. First, theyhave connections with respiratory motoneurons and areable to increase their excitability (322, and referencescited therein). Second, they are crucial for homeostaticthermoregulation. About 50% of 5-HT neurons in raphepallidus/obscurus increase their activity upon environ-mental and/or hypothalamic cooling, and there is growingevidence that raphe neurons modulate the hypothermiccomponent of the HVR (323, 324). Third, they mediateepisodic hypoxia-induced long-term facilitation (LTF) ofbreathing by causing an enhanced release of 5-HT into thephrenic motor nucleus (230, 481; for a discussion on LTF,see sect. VG; note that LTF can also be evoked by electri-cal simulation of raphe obscurus neurons, Ref. 512). Andfinally, many raphe neurons also possess intrinsic CO2

sensitivity (322), so at least in the secondary phase ofpoikilocapnic hypoxia they will certainly have an influ-ence on ventilatory output.

Activation of neurons in the nucleus raphe magnusattenuated the processing of afferent input from the ca-rotid bodies by the NTS in anesthetized rats (602). Micro-dialysis of the GABA receptor agonist muscimol into ra-phe magnus of conscious rats augmented the poikilocap-nic HVR measured at an environmental temperature of30°C (within in the thermoneutral zone for rats) but re-duced it at 24.5–26.5°C; muscimol reduced body temper-ature at 24.5–26.5°C and increased the anapyrexic com-ponent of the HVR without influencing these parametersat 30°C (772). In the same species, lesions of raphe mag-nus neurons with ibotenic acid (unspecific for serotonergicneurons) increased the hypoxia-induced rise in ventilationmeasured at an ambient temperature of 25°C withoutaffecting the anapyrexic response (256). However, micro-injection of a specific 5-HT1A receptor antagonist intoraphe magnus decreased the poikilocapnic HVR (557).Unless it is speculated that the applied antagonist (WAY-100635) acted on presynaptic inhibitory receptors (auto-receptors) on serotonergic neurons, the latter findingseems difficult to reconcile with the stimulatory effects oflocal lesions and muscimol application on the ventilatorycomponent of the HVR.



To address the role of central serotonin in respiratoryand thermoregulatory control, Richerson and co-workers(324) utilized conditional knockout mice almost entirelydevoid of central 5-HT neurons, designated as Lmxlbf/f/p.These mice are deficient in the capacity to generate heat,resulting in a rapid lowering of body temperature whenplaced in a cold environment; ventilatory CO2 sensitivityis also greatly reduced in these animals (324). Althoughthey had the expected temperature-dependent hypoxia-induced changes in ventilation and metabolism (i.e., loweringthe environmental temperature decreased the ventilatorycomponent of the HVR but increased the anapyrexic re-sponse), these animals had a reduced ventilatory andVE/VO2 response compared with wild types when the HVRwas measured at 25°C. At an ambient temperature of30°C, no difference in HVR between both genotypes wasvisible, whilst intracerebroventricular administration ofserotonin restored the impaired ventilatory response toCO2 but did not alter the (normal) HVR (323). WhetherLmxlbf/f/pmice would fail to show episodic hypoxia-in-duced LTF of breathing (see sect. v) remains to be exam-ined. Hodges and co-workers (323, 324) suggested thatserotonergic neurons are not directly involved in the HVRper se but may be critically important to thermogenesis.

Mice showing overwhelming amounts of serotonin inbrain extracellular fluid are represented by a genotypelacking expression of the serotonin transporter (5-HTT).5-HTT knockout mice showed a reduced activity of brainserotonergic neurons and decreased 5-HT1A receptorbinding (444). At room temperature, these animals had alarger ventilation and higher metabolism but normal bodytemperature. When challenged with 10% O2, they re-sponded with a greater increase in ventilation and largerfall in metabolism than wild types, but the change in theVE/VO2 ratio was equal in both genotypes (444).

A role of serotonergic neurons in medullary raphe inthermoregulation is consistent with the finding that dis-inhibition of raphe pallidus neurons (by means of localinjection of the GABA antagonist bicuculline) increasedbrown adipose sympathetic nerve activity, responsible forthermogenesis in cold stress (482). In the rat, an increasedactivity of brown adipose sympathetic nerve activity(thought to underlie the anapyrexic response to hypoxia)can be induced in several ways: cold stress (i.e., skincooling), activation of neurons in preoptic thalamic re-gions, and disinhibition of raphe pallidus neurons andalso of neurons in the commissural subnucleus of the NTS(482). Interestingly, carotid body activation with NaCN orhypoxia completely inhibits the increase in sympatheticoutflow to brown adipose tissue by these interventions.This suggests that the carotid body not only plays a pivotalrole in the hypoxia-induced rise in ventilation but also in theanapyrexic response (further references in Ref. 482).

From all of the above data, it is not easy to distill aclear and detailed picture of the role of central serotoner-

gic neurons in the HVR. A few general conclusions, how-ever, seem justified. First, in rodents, serotonergic neu-rons in medullary raphe are crucial for the homeostaticcontrol of body temperature and participate in both theventilatory and hypometabolic/anapyretic components ofthe HVR. Second, serotonergic neurons in raphe nucleipossessing connections with respiratory (premotor) neu-rons are activated in hypoxia. Third, activation of seroto-nergic neurons in raphe magnus tend to dampen the risein ventilation and to augment the fall in body temperature,but the magnitude of these effects depends on the ambi-ent temperature. And, finally, the role of raphe serotoner-gic neurons in hypoxia-induced LTF of breathing is wellestablished (discussed in sect. VG1).

C) CENTRAL NO AND HVR. I) NTS. Carotid body stimula-tion in hypoxia is followed by release of glutamate andproduction of NO in the caudal (commissural) NTS. Wheninjected in the NTS, the NO donor sodium nitroprussideaugments the HVR while the nonselective NOS inhibitorNG-monomethyl-L-arginine (L-NMMA) reduces it. Theseobservations in unanesthetized rats, together with thefinding that the increased production of NO is inhibited byNMDA receptor antagonism, led to the suggestion that NOand glutamate operate in the NTS in a positive feedbackcycle to increase the HVR (562; see also Ref. 257). Gluta-mate augments the HVR by activating NMDA receptors inprojection areas of carotid body afferents in the NTS thatare rich in NO (130, 286, 288, 446; reviewed in Ref. 287). Inthe conscious rat, injection of both the nonselective NOSinhibitor NG-nitro-L-arginine methyl ester (L-NAME) andthe selective neuronal inhibitor of the enzyme Nw-propyl-L-arginine in the caudal NTS reduced the increase in re-spiratory frequency induced by carotid body stimulationwith KCN (291).

Lipton et al. (451) proposed an involvement of NO inthe HVR via endogenous production of S-nitrothiosolswithin the NTS. Hemoglobin desaturation goes along withformation of S-nitrothiosols from gluthathione and cys-teine and Hb-bound NO producing S-nitroso-gluthathione(GSNO) and S-nitroso-L-cysteine (L-CSNO), respectively.GSNO is converted into CGSNO (S-nitroso-cysteinyl gly-cine), and this requires the enzyme �-glutamylpeptidase(�-GT; Ref. 451 and references therein). Local injection ofall three nitrothiosols into the NTS mimicked the HVR inconscious rats including its recovery phase (short-termpotentiation) after removal of the hypoxic stimulus, andthe same was observed after injection of plasma fromdeoxygenated but not oxygenated blood (451). The effectof CGSNO was inhibited by pretreatment with a blockerof �-GT, suggesting that conversion of GSNO into CGSNOis an essential step in producing the effect of the former.L-CSNO is less stable in vivo, and thus Lipton et al. (451)hypothesized that �-GT would be required for a normalHVR to occur and found the short-term potentiation to beabolished in �-GT-deficient mice. Apparently, however,



the hypoxia-induced rise in ventilation was not eliminatedin �-GT�/� mice, but their report did not show the mag-nitude of this initial increase (451). From the fact that inthe rat denervation of the carotid bodies reduces the HVRwith at least 70% and virtually eliminates it in other spe-cies, the quantitative contribution of S-nitrothiosols canbe estimated to be small or negligible unless carotid body-activated NOS within the NTS would play a major role inproducing these molecules during hypoxia. On the otherhand, the reported increase of the HVR in humans byN-acetylcysteine (321) could be related to generation ofS-nitrothiosols with the aid of Hb-bound NO (578), ratherthan to the antioxidant properties of this molecule.

II) Pons and locus coeruleus. In the pontine reticularformation, NO promotes the release of acetylcholinewhich induces sleep associated with rapid eye move-ments and respiratory depression (442). The dorsolateralpons uses NO to promote the termination of inspirationduring lung inflation (pneumotaxic mechanism). In awakerats, electrolytic bilateral lesions in the locus coeruleussubstantially augmented the HVR (224). In addition,whereas acute intracerebroventricular administration ofthe nonselective NOS inhibitor L-NAME almost com-pletely inhibited the HVR in control animals, it did so byonly �50% in those with the bilateral lesions, suggesting asubstantial contribution of the locus coeruleus to theinhibition in intact animals (224). The locus coeruleuscontains the A6 cell group that is considered the majornoradrenergic cell group in the brain stem (references inRef. 803). Microinjection of L-NAME in the locus coer-uleus had an impressive inhibitory effect on the HVR,indicating that the “permissive” effect of NO may be dueto inhibition of, presumably noradrenergic, neurons inthis area (225). It is unclear why in another study in ratschronic injection (7 days) of small amounts of L-NAMEinto the right cerebral ventricle did not modify the HVRunless it is hypothesized that chronic NO shortage may becompensated by upregulation of NOS (688). Notwithstand-ing this latter study, all data taken together clearly identifythe locus coeruleus as an important inhibitory modulator ofthe HVR, possibly analogous to an active inhibitory role oflateral pontine regions in the hypoxic inhibition of breathingin newborn lambs (522; see also sect. II).

III) Other regions. Microinjection of the nonselectiveNOS inhibitor L-NMMA into the RVLM reduced the HVR inrats but not the hypothermic response (173). The sameinhibitor had a similar effect when injected into the nu-cleus raphe magnus (556). A possible role of brain stemNO in the acclimatization to chronic hypoxia has beenbriefly mentioned in section VF.

In humans, the acute ventilatory response to hypoxiadoes not seem to be influenced by systemic administra-tion of L-NMMA (347). The intravenous dose applied inthis study (�62 �g/kg), however, was much smaller thanthat mostly used for NOS inhibitors in animal studies

(including those applying ventricular administrations), soperhaps inhibitor concentrations at target sites within thebrain were too small to produce significant effects (al-though blood pressure clearly increased; Ref. 347).

D. Hypoxic Ventilatory Decline in the

Adult Mammal

In adult mammals, the ventilatory response to sus-tained (15–30 min) hypoxia (end-tidal PO2 50–55 Torr andSaO2

�80%) is biphasic (Figs. 5 and 6; Refs. 162, 204, 370,383, 464, 752, 806). During moderate isocapnic hypoxia(i.e., arterial PO2 �50 Torr) lasting �3–5 min, ventilationshows a slow decline from its peak (75–150% above nor-moxic resting values) to reach a new steady-state level(25–40% of normoxic values) within 15–20 min (162, 204,236, 383). The slow decline was formerly known as “hyp-oxic ventilatory depression”; current names include hyp-oxic ventilatory decline (HVD), hypoxic ventilatory roll-off, and secondary roll-off of the HVR.

Although consistently found across mammalian spe-cies, dogs and goats are the exception with little or noHVD during exposure to moderate hypoxia (113, 151,546). However, HVD is dependent on various factors, suchas age, sex, CNS arousal state, (type of) anesthesia,breathing route, and hypoxic test (12, 111, 498, 677, 773).For example, HVD is most pronounced under isocapnicexperimental conditions, while the poikilocapnic HVD ismuch smaller in magnitude (12).

In the awake state, the ventilatory response to sus-tained isocapnic hypoxia has specific characteristics.1) The magnitude of HVD is proportionally related to thesize of the initial hyperventilatory response to acute hyp-oxia (162, 263, 266, 383). 2) Bilateral carotid body resec-tion or chemodenervation results in the loss of the VI

(inspired ventilation) response to acute hypoxia and ab-sence of HVD (Fig. 5; Refs. 383, 464). 3) HVD persistsbeyond the initial period of hypoxic exposure. The VI

response to acute hypoxia following 20 min of moderateisocapnic hypoxia and 5 min of air breathing is depressedby �50%. This delayed recovery persists for up to 1 h (162,205). 4) Suppression of the peripheral drive with low-dosedopamine yields no HVD during 20-min hypoxia. And,despite the presence of initial central hypoxia, a subse-quent VI response to acute hypoxia develops fully (Fig. 6;Ref. 162). 5) The magnitude of the fall in ventilation uponthe relief of hypoxia is smaller than the VI response gener-ated at the onset of hypoxia (204, 382). 6) Awake humansand awake cats display similar response characteristics withrespect to characteristics 1–3 and 5 (Fig. 5; Refs. 204, 464).

These observations (most importantly characteristic

4) suggest that in conscious mammals the peripheralchemoreceptors play a pivotal role in the development ofHVD. The issue remains whether HVD is due to adapta-



tion of the peripheral chemoreceptors (i.e., an exclusiveperipheral mechanism) or to the way the afferent inputfrom the carotid bodies is processed by the CNS (657).The hypothesis that HVD is related to peripheral chemo-receptor adaptation is supported by characteristics 1, 2,and 5 and is further corroborated by the observation thathypoxic sensitivity declines during the development ofHVD (37). However, in the anesthetized cat, the periph-eral chemoreceptors do not adapt during hypoxic expo-sure in terms of ventilation and afferent nerve activity,indicating that an exclusive peripheral mechanism is un-likely (17, 58, 807). Furthermore, characteristics 3 and 4

indicate an effect of HVD on HVR beyond hypoxic expo-sure when peripheral chemoreceptor activity in terms ofthe VI response to CO2 has normalized (57, 267, 461). Thecurrent hypothesis is that, at least in the awake mammal,a large part of HVD is due to modulation of glutamatergicexcitatory activity within the NTS (arising from afferentcarotid body input) shifting the excitatory/inhibitory bal-

ance towards a net inhibitory effect on VI (162, 266, 810).At the termination of hypoxia, changes in neurotransmit-ter turnover are not immediate and, therefore, the depres-sant effects of hypoxia wane slowly. In the awake state, asmall part of HVD is related to an increase in brain bloodflow and the resultant washout of acid metabolites/H�

from the brain. Suzuki et al. (746) showed a decrease inthe gradient between jugular venous and arterial PCO2 of�2 Torr during 15 min of moderate isocapnic hypoxia.Assuming that jugular venous PCO2 approximates braintissue PCO2, �10–20% of measured HVD is related to anincrease in brain blood flow in the adult awake human.Finally, at deeper levels of hypoxia, substrate insuffi-ciency and depressed cellular metabolism may becomethe major origin of HVD (435, 540).

Various mechanisms could explain the decrease inexcitatory neuronal output during sustained hypoxia. Forexample, as suggested by Kazemi and Hoop (379), periph-eral chemoreceptor afferents cause the enhancement of

FIG. 5. Ventilatory effects of hypoxia in cats and humans. A: ventilatory responses to sustained (25–30 min) isocapnic hypoxia in the awakecat (mean response of 7 cats) [adapted from Long et al. (461)] and awake chemodenervated cat. B: mean response of 3 cats. [Adapted from Longet al. (464).] C: isocapnic (open symbols) and poikilocapnic (closed symbols) hypoxic ventilatory response (HVR) in 10 young subjects. [Adaptedfrom Steinback and Poulin (728).] D: isocapnic HVR in awake humans after bilateral carotid body resection (mean of 7 subjects). The data arederived from otherwise healthy patients that underwent a bilateral carotid body resection after development of familial carotid body tumors. Asreported in Reference 227, the hypercapnic response in these subjects could be adequately described by a single slow (central) component,indicating a minor or no significant contribution of a fast (peripheral) component to the response.



glutamatergic excitation in areas of the brain stem in-volved in hypoxic ventilatory control. Glutamate serves asthe precursor of GABA, and conversion of glutamate intoGABA may enhance GABAergic activity with conse-quently a net inhibitory effect on VI. In the awake rat,Tabata et al. (752) measured an increased GABA concen-tration in the extracellular fluid in the NTS during HVDdevelopment, which was absent in animals after bilateralcarotid body resection. Both the reduction in O2 availabil-ity and intracellular acidosis increase GABA levels in thebrain with kinetics that are fast enough to account forHVD (increased synthesis from glutamate through theenzyme glutamic acid decarboxylase and decreasedbreakdown by the enzyme GABA-�-oxyglutarate transam-inase; references in Ref. 540). Via this pathway, the NTSitself could be the source of increased GABA during hyp-oxia (540, 752). However, other sources of GABA are alsopossible because the NTS contains many terminals immu-noreactive to GABA that may inhibit glutamatergic inputs(e.g., Ref. 673 and references therein). GABAergic neu-rons in the NTS send inhibitory efferents to the caudalventrolateral medulla and also to the locus coeruleus,providing other possible pathways involved in HVD (7,795). Other regions such as the hypothalamus may alsoparticipate in HVD via a GABAergic mechanism. For ex-ample, Lemus et al. (440) depicted an interesting circuitryparticipating in the glycemic response of carotid bodystimulation involving antidiuretic hormone (ADH)-secret-ing neurons in the supraoptic nucleus and ADH receptorson the membrane of GABAergic neurons in the NTS that

may originate from the paraventricular nucleus (PVN) ofthe hypothalamus. It is known that arginine vasopressin-containing neurons in the PVN and supraoptic nucleus arestimulated by hypoxia (709 and references therein), but itremains to be seen whether NTS neurons could be di-rectly inhibited during hypoxia via this pathway.

Other inhibitory modulators possibly involved inHVD include dopamine, adenosine, serotonin, endogenous opi-oid peptides, lactic acid, NO, and Ca2� (Tables 1 and 2 forreferences). A relatively new observation is that of Gozal et al.(289) who examined the role of the platelet-derived growthfactor (PDGF) isoforms PDGF-AA (with two A chains andacting on the PDGF-� receptor) and PDGF-BB (with two Bchains and acting on the PDGF-� receptor) in the NTS duringhypoxia. The NTS and other brain stem respiratory nuclei showan abundant expression of the PDGF-� but not PDGF-� recep-tor. In contrast to PDGF-AA, the isoform containing two Bchains and its receptor PDGF-� appeared to play a functionalrole in the development of HVD. Awake mice with a heterozy-gous mutation of the PDGF-� receptor had an intact VI re-sponse to acute hypoxia but a diminished HVD, an effect whichwas unaffected by the injection of a PDGF-� receptor antago-nist. However, wild-type animals showed a reduced HVD afterthe antagonist. Many NTS neurons in wild-type mice (and inrats) showed increased expression of PDGF-B chain mRNAand protein (release) in hypoxia. These data suggest an inhib-itory role for PDGF-BB and its receptor on ventilation duringsustained hypoxia, possibly by inhibiting NMDA receptorswithin the NTS (289).

There is strong evidence that HVD development dur-ing anesthesia is uncoupled from peripheral chemorecep-tor activity (657). For example, in anesthetized cats sub-jected to artificial brain stem perfusion, central hypoxiawith the carotid bodies kept normoxic results in a declinein ventilation with dynamics similar to that observed aftercentral changes in PCO2 (823). Comparable observationsare made in carotid body perfusion experiments and anes-thetized chemodenervated animals (361, 538, 543; Table 2).Similarly, in humans exposed to halothane, the initialhyperventilatory response is greatly reduced while theabsolute magnitude of the HVD is similar to that observedin the awake state (160). Furthermore, the ventilatoryresponse to a step out of sustained hypoxia (the off-response) shows a marked undershoot and is similar inmagnitude to the initial hyperventilatory response (theon-response). These symmetric on- and off-responses inhumans exposed to halothane indicate that in this condi-tion peripheral sensitivity, although reduced under of low-dose halothane, remains unchanged during HVD. Similarobservations were made in humans injected with theintravenous anesthetic propofol (549). These data sharplycontrast with data in awake humans and animals that donot show HVD development after carotid body resection orwhen the peripheral drive is reduced by dopamine (Fig. 6;Refs. 162, 383, 464).

FIG. 6. Dopamine and hypoxic ventilatory decline (HVD) in hu-mans. Ventilatory response to 20-min isocapnic hypoxia (SpO2 � 80%),5-min normoxia and subsequently a second hypoxic episode (SpO2 �80%, lasting 5 min). Closed circles, control response showing the HVDfrom early to late hypoxia and a marked reduced response to a secondhypoxic exposure; open circles, low-dose dopamine infusion (gray bar:3 �g �kg�1 �min�1) during sustained hypoxia blunts the initial hyperven-tilatory response and prevents development of HVD. The ventilatoryresponse to the second hypoxic exposure develops fully. Data are themeans of 16 subjects. [Adapted from Dahan et al. (162).]



During anesthesia, when the peripheral drive is eitherabsent or severely reduced, it seems appropriate to as-cribe the development of HVD to an effect arising selec-tively from within the CNS. One possibility is that whileCBF-induced HVD plays only a minor role in the awake

state, it becomes the major contributor during anesthesia.In the anesthetized cat, brain blood flow-related HVDdoes explain the similarities in ventilatory time constantsof central CO2 steps and hypoxic steps, respectively, re-stricted to the brain stem (823). It would also explain the

TABLE 1. Effect of pharmacological interventions on the development of HVD in adult humans

Target Agent Arousal State Effect CommentsReference

Nos.

�2-Adrenergic receptor Clonidine Sedated AHR� HVD� In line with the observation thatthe magnitude of HVD isrelated to AHR

236

Adenosine receptor Aminophylline (adenosineantagonist)

Awake AHR� HVD2 Data suggest a role foradenosine in the developmentof HVD

203

265Dipyridamole (adenosine

reuptake inhibitor)Awake AHR1 HVD1 Pretreatment with aminophylline

abolished the dipyridamoleeffect; data indicate a role foradenosine in HVDdevelopment

859

Calcium channel Verapamil (Ca2� channelblocker)

Awake AHR� HVD� Data suggest no involvement ofcalcium ions in HVDdevelopment

459

D2 receptor Low-dose dopamine Awake AHR2 HVD2 Dopamine acts peripherally atthis dose; the reduced HVDprevents depression of asubsequent VI response toacute hypoxia

162

Awake AHR2 HVD� 38Awake AHR2 HVD2 821

Haloperidol (central andperipheral D2 antagonist)

Awake AHR1 HVD� Effect contrasts findings inawake cats (see Refs. 769 and460)

589

Domperidone (peripheral D2

antagonist)Awake AHR1 HVD� 38, 235

Lactic acid Dichloroacetate (blockslactate production)

Awake AHR� HVD� Data suggest that brain lacticacidosis is unlikely involved inHVD development

264

Opioid receptor Naloxone (opioid receptorantagonist)

Awake AHR� HVD� Data suggest no involvement ofendogenous opioid system inHVD development

370

Alfentanil (�-opioid receptoragonist)

Awake AHR2 HVD� Data suggest that exogenousopioids potentiate HVD

120

Morphine (�-opioid receptoragonist)

Awake AHR2 HVD� 680

Intrathecal morphine Awake AHR2 HVD� 255-HT receptor Methysergide (5-HT receptor

antagonist)Awake AHR� HVD� Data suggest no involvement of

serotonin in HVD development459

Somatostatin receptor Somatostatin Awake AHR2 HVD2 Effect possibly via disinhibitionof GABAergic inhibitorysystem

231

Carotid body Almitrine Awake AHR1 HVD1 In line with the observation thatthe magnitude of HVD isrelated to AHR

266

Central nervous system Low-dose inhalationalanesthetics (halothane,isoflurane, enflurane)

Lightly anesthetized AHR2 HVD� Data suggest that the linkbetween AHR and HVD is lostduring anesthesia

798

535235

Propofol (intravenousanesthetic)

Lightly anesthetized AHR2 HVD1 Propofol may have increasedHVD via its action at theGABA receptor complex;propofol had no effect onintermittent hypoxicventilation

549

AHR, acute hypoxic ventilatory response; HVD, hypoxic ventilatory decline; VI, inspired ventilation.



TA

BLE

2.E

ffect

of

pharm

acolo

gic

al

inte

rven

tion

son

the

develo

pm

en

tof

HV

Din

aw

ake

an

dan

esth

eti

zed

adu

ltan

imals

Tar

get

Age

ntA

nim

alA

rous

alSt

ate

Eff

ect

Com

men

tsR

efer

ence

Nos

.

Ade

nosi

nere

cept

orA

min

ophy

lline

(ade

nosi

nean

tago

nist

)C

atA

nest

heti

zed

(chl

oral

ose-

uret

hane

)H

VD2

The

caro

tid

sinu

sne

rves

and

vagi

ofth

ean

imal

sw

ere

bila

tera

llyde

nerv

ated

361

The

ophy

lline

ivan

dic

v(a

deno

sine

anta

goni

st)

Cat

Ane

sthe

tize

d(s

odiu

mpe

ntot

hal)

iv:

AH

R1

Dat

afr

omth

ese

inte

rven

tion

ssu

gges

tin

volv

emen

tof

aden

osin

ein

deve

lopm

ent

ofH

VD

866

iv:

HV

D2

icv:

AH

R�

icv:

HV

D2

D2

rece

ptor

Hal

oper

idol

(cen

tral

and

peri

pher

alD

2an

tago

nist

)C

atA

wak

eA

HR

�H

VD2

HV

Dw

asab

olis

hed

byha

lope

rido

l76

9

460

463

Cat

Ane

sthe

tize

d(c

hlor

alos

e-ur

etha

ne)

AH

R�

HV

D2

Iden

tica

lre

sult

sas

inth

eaw

ake

cat;

caro

tid

sinu

sne

rve

acti

vity

rem

aine

dco

nsta

ntdu

ring

HV

Dde

velo

pmen

t

769

Dom

peri

done

(per

iphe

ral

D2

anta

goni

st)

Cat

Aw

ake

AH

R�

HV

D�

Dat

afr

omth

ese

inte

rven

tion

ssu

gges

tth

atce

ntra

lbu

tno

tpe

riph

eral

dopa

min

ergi

cm

echa

nism

sar

edi

rect

lyin

volv

edin

HV

D

769

GA

BA

rece

ptor

Bic

ucul

line

(GA

BA

anta

goni

st)

mic

roin

ject

edin

the

NT

SR

atA

wak

eH

VD2

GA

BA

anta

goni

sts

redu

ced

HV

Din

inta

ctan

imal

son

ly;

inca

roti

dbo

dyde

nerv

ated

anim

als,

noH

VD

was

pres

ent

and

noH

VD

deve

lope

dup

onin

ject

ion

ofG

AB

Aan

tago

nist

sdu

ring

hypo

xia

752

Bic

ucul

line

iv(G

AB

Aan

tago

nist

)C

atA

nest

heti

zed

(chl

oral

ose)

HV

D2

Ani

mal

sw

ere

glom

ecto

miz

ed;

phre

nic

nerv

ere

spon

seto

prog

ress

ive

CO

was

mea

sure

d

508

Lact

icac

idD

ichl

oroa

ceta

te(p

reve

nts

lact

ate

prod

ucti

on)

Cat

Ane

sthe

tize

d(c

hlor

alos

e)H

VD2

Ani

mal

sw

ere

peri

pher

ally

chem

oden

erva

ted;

prog

ress

ive

hypo

xia

indu

ced

byin

hala

tion

ofC

O;

data

sugg

est

aro

lefo

rla

ctic

acid

inH

VD

gene

rati

on

543

Nit

ric

oxid

esy

ntha

se(N

OS)

NG

-nit

ro-L

-Arg

inin

e(a

nons

peci

ficN

OS

bloc

ker)

Rat

Aw

ake

HV

D1

Con

trol

anim

als

did

not

disp

lay

HV

D29

0

S-m

ethy

l-L-t

hioc

itrt

rulli

ne(a

sele

ctiv

ene

ural

NO

Sin

hibi

tor)

Opi

oid

rece

ptor

Nal

oxon

e(o

pioi

dre

cept

oran

tago

nist

)C

atA

nest

heti

zed

(chl

oral

ose)

HV

D2

Ani

mal

sw

ere

peri

pher

ally

chem

oden

erva

ted;

vent

ilato

ryre

spon

seto

prog

ress

ive

CO

was

mea

sure

d

541

Mor

phin

e(�

-opi

oid

rece

ptor

agon

ist)

Cat

Ane

sthe

tzie

d(c

hlor

alos

e-ur

etha

ne)

HV

D2

Stea

dy-s

tate

resp

onse

tosu

stai

ned

hypo

xia

rem

aine

dun

alte

red

bym

orph

ine,

sugg

esti

nga

redu

ced

HV

Dpo

ssib

lyby

redu

cing

brai

nbl

ood

flow

resp

onse

tohy

poxi

a

61



symmetry in on- and off-responses observed in halothane-exposed humans. Manipulating brain blood flow eitherhemodynamically (for example, by aortic balloon infla-tion), pharmacologically (for example, by infusion of pa-paverine to blood perfusing the brain stem), or by expo-sure to central hypoxia produce a similar decrease in VI

that is consistent with washout of central CO2 (60, 293,542, 823). Finally, we cannot exclude an effect of modu-lators accumulating during hypoxia in the anesthetizedmammal. Indeed, some enhancement of GABAergic inhi-bition of VI during hypoxia is expected, especially fromanesthetics that have agonistic activity at the GABA re-ceptor. However, at PO2 values �50 Torr, these effects areprobably small with regard to the reduced tissue PCO2 bychanges in BBF (794).

IV. GENETIC INFLUENCE ON THE

VENTILATORY RESPONSE TO HYPOXIA

A. The Genome and HVR

The genotype plays an important role in the magni-tude of the HVR and consequently explains part of thelarge intersubject response variability (734). Measure-ments of the HVR in humans show that normal healthyvolunteers may vary a factor of 20 in the magnitude oftheir response. This range is much larger than expectedfrom just normal physiological variability and technical,environmental, and/or emotional factors. Furthermore,data from high-altitude residents indicate that genotypemay not only influence the immediate response to hyp-oxia but possibly also the ventilatory response to sus-tained hypoxia and ventilatory adaptation to high altitude(519).

1. Human studies

In humans, evidence for a genetic effect on HVRcomes primarily from family and population studies.First-degree healthy family members of patients with pul-monary disease associated with hypoventilation and CO2

retention and nonathletic parents and siblings of endur-ance athletes show a familial clustering of HVR magni-tudes at the low end of the spectrum (see Ref. 832 andreferences cited therein). Similar observations were madein families with a history of SIDS and obstructive sleepapnea (734). Twin studies show greater similarities in themagnitude of HVR among homozygotic twins comparedwith heterozygotic twins (see Ref. 734 and referencescited therein). Most but not all interpopulation compari-sons indicate the existence of heritable HVR phenotypeslinked to specific populations (96, 97, 152, 783). Strongevidence for a hereditary component comes from studiesin Asians and South Americans. Children from TibetanT

AB

LE2.

—C

on

tin

ued

Tar

get

Age

ntA

nim

alA

rous

alSt

ate

Eff

ect

Com

men

tsR

efer

ence

Nos

.

NM

DA

rece

ptor

MK

801

(NM

DA

rece

ptor

anta

goni

st)

ivan

dm

icro

inje

cted

inP

B-K

F

Rat

Ane

sthe

tize

d(h

alot

hane

-ur

etha

ne)

iv:

HV

R2

mic

roin

ject

ion:

HV

R�

No

dist

inct

ion

betw

een

AH

Ran

dH

VD

ism

ade;

data

sugg

est

are

duct

ion

inA

HR

wit

hen

hanc

emen

tof

HV

Daf

ter

syst

emic

MK

801

inje

ctio

n,w

hile

afte

rm

icro

inje

ctio

nin

toth

eP

B-K

FA

HR

ism

aint

aine

dw

hile

HV

Dis

redu

ced

92

Pla

tele

t-de

rive

dgr

owth

fact

or(P

DG

F)-

�re

cept

or

PD

GF

-�an

tago

nist

mic

roin

ject

edin

the

NT

SR

atA

wak

eA

HR2

HV

D2

289

PD

GF

-�an

tago

nist

(giv

enin

trap

erit

onea

lly)

Het

eroz

ygou

sP

DG

F-�

knoc

kout

mic

e

Aw

ake

HV

D2

inw

ild-t

ype

anim

als

only

Con

trol

expe

rim

ents

show

are

duce

dH

VD

inkn

ocko

utan

imal

sco

mpa

red

wit

hth

eir

wild

-typ

elit

term

ates

;da

tasu

gges

ta

role

for

PD

GF

-�in

HV

Dde

velo

pmen

t

289

D2

rece

ptor

,dop

amin

e2

rece

ptor

;PB

-KF

,Par

abra

chia

l/Kol

liker

fuse

com

plex

;AH

R,a

cute

hypo

xic

vent

ilato

ryre

spon

se;H

VD

,hyp

oxic

vent

ilato

ryde

clin

e;H

VR

,hyp

oxic

vent

ilato

ryre

spon

se;

iv,

intr

aven

ous;

icv,

intr

acer

ebro

vent

ricu

lar.



and Han (Chinese) parents tend to be in the range ob-served in the Han population (152). With the use of an-cestry-informative genetic markers, an evolutionary ori-gin was found for the low HVR in Andean high-altitudenatives of Quechua origin (97).

2. Animal studies

There is ample proof that the divergence in ventila-tory behavior between different inbred strains of the samespecies reflects genetic influences more than effects frombody mass, age, or sex. This also applies to the ventilatoryresponse to moderate hypoxia and the survival responseto severe hypoxia (874). In rats, most studies show aneffect of strain on the magnitude of HVR (50, 274, 736, 739,834). Some strains are difficult to compare as underlyingdisease may affect carotid body function. For example,New Zealand hypertensive rats suffer from sclerotic dam-age to the carotid artery and its branches (50). Strainsmay differ in the magnitude of the HVR by up to a factorof seven (834). The differences are not only related to thecarotid body response to hypoxia but also to centralneural processing during and immediately after exposureto short episodes of hypoxia (274, 739). As stated byGolder et al. (274), the choice of rat strain can (critically)influence conclusions concerning the control of breath-ing. This is also true for drug studies. For example, in rat,the peripheral dopamine D2 receptor antagonist domperi-done and the NOS inhibitor L-NAME differentially affectventilation and the HVR depending on the genetic back-ground (737, 738).

To pinpoint the HVR phenotype to a specific ratchromosome, Forster et al. (240) used the technique ofchromosomal substitution in which an entire chromo-some from one inbred rat strain is introgressed into thegenetic background of another to create a consomic rat.Initially, five chromosomes were studied. HVR was notdetermined by any of the genes on these chromosomes,although genes on chromosomes 9 and 18 did contributeto HVD and/or a reduced metabolic rate during hypoxia.The results of a second study indicate that there arecomplex gene-gender interactions on HVR (199).

Like rats, inbred mice strains vary significantly in themagnitude of their HVR. Tankersley et al. (760) examinedthe HVR in eight inbred strains. The DBA/2J strain had thegreatest HVR, while the A/J strain displayed the lowestresponse. Tankersley et al. (760) did not take the brainstate of the animals into account, and in this this regard,it is interesting that A/J mice are more prone to sleep thanDBA/2J mice and have a higher arousal threshold in hyp-oxia (669). So possibly the ventilatory response could bethe reflection of a stress response. However, the differ-ence in frequency of hypoxic events during sleep (largerin the A/J strain) was inversely related to the hypoxicventilatory sensitivity, whilst the arousal threshold upon

tactile/auditory stimuli was not different between bothstrains (669). The difference in HVR was associated withstructural and functional changes in the carotid bodies(858). Compared with the A/J strain, the carotid bodies ofthe DBA/2J strain were larger, and acetylcholine in-creased intracellular Ca2� in more cells. These morpho-logical and functional differences are accompanied bydifferences in gene expression (30). There were differ-ences in expression of genes related to neurotransmittermetabolism, synaptic vesicles, and neural crest cell mi-gration/development with less expression in A/J mice. Animportant difference was the reduced expression of the�-subunit of maxi-K channels in A/J mice. Further studiesfrom Tankersley and co-workers (758, 759, 761) aimed touncover the genes involved in the HVR phenotype. Datafrom an intercross study between C3H/3J and C57BL/6Jmice suggest that HVR in terms of respiratory timing isdependent on only a handful of genes (759). A putativequantitative trait locus on chromosome 9 could be iden-tified with a significant association with several hypoxicventilatory variables (758), while another locus on chro-mosome 3 was significantly associated with hypoxic in-spiratory drive. Both loci explained 26–30% of the pheno-typic variation among F2 offspring. The interaction be-tween hypoxia and hypercapnia on breathing was linkedto two different loci on chromosomes 1 and 5 (761). Thesimplicity of this genetic model is appealing and is muchless complex than in the rat. It is doubtful whether such arestricted model also applies to the human HVR phenotype.

B. Mutations and HVR

Examples of targeted gene deletions in mice to studythe process of oxygen sensing at the carotid bodies and/orto understand the peripheral and central processing of thecarotid body response to hypoxia are discussed in sectionIII. Here we briefly discuss mouse strains developed tostudy specific diseases and that show altered ventilatoryresponses to hypoxia. We further discuss two humanconditions in which one specific mutation is linked to analtered HVR.

Various mouse strains have been engineered to studythe Congenital Central Hypoventilation Syndrome (CCHS),some of which have recently been reviewed (see Ref.259). Particularly genes of the endothelin and c-ret path-way have been the focus of interest, for example, endo-thelin converting enzyme (ECE1), endothelin-1 (EDN1),endothelin-3 (EDN3), endothelin receptor A (Ednra), en-dothelin receptor B (Ednrb), Mash-1, c-ret, Phox2a, Phox2b,brain-derived neurotrophic factor (Bdnf), and orphan ho-meobox (Rnx) genes (169, 259, 849). Knockouts at thesegenes display a variable respiratory phenotype with reducedhypercapnic response and/or HVR in ECE1�/�, EDN1�/�,Ednra�/� and Mash-1�/� newborn mice; however, with the



exception of EDN1�/� mice, these genotypes did not displayabnormalities as adults. This is an important limitation for adisease that in humans persists throughout life (259). Thereare other mice models associated with CCHS such as theNurr1 mutant mouse and the microphtalmia-associated tran-scription factor (Mitf) mutant mouse (555, 754). Nurr1 mu-tant mice show complete agenesis of midbrain dopaminecells. Newborn Nurr1 knockout mice have a severely dis-turbed pattern of breathing with a blunted HVR (555). Mitfplays an essential role in the differentiation of neural crest-derived melanocytes. Mitf mutant mice display an increasedHVR. The authors relate this to a reduction of melanocyte-derived opioids within the CNS (754). This suggests an in-hibitory role of endogenous opioids on the central ventila-tory network and is in agreement with observations in �-opi-oid receptor gene knockout mice (158).

A mouse model associated with the Rett syndrome isrepresented by a genotype with a ubiquitous deletion ofone allele of the X-linked Mecp2 gene, encoding the DNA-binding protein methyl CpG (Mecp2) involved in genesilencing. Female Mecp2�/� mice have a respiratory phe-notype resembling that of the Rett syndrome and that ischaracterized by episodes of hyperventilation and respi-ratory depression (75). The (isocapnic) HVR in heterozy-gous females is larger than in wild types; mice with aselective deletion of the protein in neurons do also displayan enhanced HVR but do not show the typical episodes ofrespiratory depression (75). It is interesting to note thatdifferent respiratory phenotypes also exist between dif-ferent mouse strains with the gene deletion (references inRef. 563).

In humans, carotid body tumors are rare and oftenassociated with chronic hypoxia from altitude, cyanoticheart disease, and pulmonary disease. One hereditaryform of carotid body tumor (as part of the hereditaryfamiliar paraganglioma syndrome) is due to a missensemutation within the gene that encodes succinate dehydro-genase D (SDHD; Refs. 48, 154, 213). SDHD is an essentialenzyme of the tricarboxylic acid cycle and part of cyto-chrome b588 of the mitochondrial respiratory chain com-plex II. All patients are heterozygous: animal studies in-dicate that homozygotes do not survive and die in utero(607). We measured the isocapnic HVR and hypercapnicventilatory responses in normoxia and hypoxia in 12 tu-mor-free carriers of the mutation (as tested by MRI; au-thor’s unpublished observation). There were three mainobservations: 1) hypercapnic ventilatory responses werenot different from age-matched healthy controls, 2) themagnitude of the isocapnic HVR was at the low end of“normal” (mean value 0.3 versus 1.2 l �min�1 �%�1 in con-trols), and 3) in contrast to controls, none of the testedcarriers of the SDHD gene showed any O2-CO2 interactionon the HCVR. These data may link the SDHD gene tocarotid body oxygen sensing and/or to the well-describedmultiplicative action of CO2 on the HVR (see sect. VD).

Robbins and co-workers (712) studied the effect of amutation in the von Hippel-Lindau tumor suppressor pro-tein (VHL) gene. A homozygous missense mutation thatimpairs but does not ablate VHL function causes Chuvashpolycythemia (47, 213). The disease is endemic to theChuvash population of Russia. In patients with Chuvashpolycythemia, binding of VHL to HIF-1� is diminished andhence normal degradation reduced causing the accumu-lation of HIF (in normoxia) and consequent upregulationof certain target genes such as the erythropoietin gene.The three patients studied showed several abnormalitiesin respiratory and pulmonary vascular regulation. Mostimportantly, the isocapnic HVR and pulmonary arterypressure responses to hypoxia were increased comparedwith healthy and polycythemia control subjects (712). Inother words, cardiorespiratory responses to hypoxia wereupregulated in patients with increased HIF levels and up-regulated HIF target genes. These data are in agreementwith animal studies showing downregulated cardiopulmo-nary responses in HIF-1��/� mice and attenuated adaptiveresponses to chronic intermittent hypoxia (see sect. V).

In conclusion, the above examples of the influence ofsingle gene defects (and their inheritance) on the HVR areclear illustrations of a genetic influence. Strain differ-ences in animals and differences in HVR between low-landers and high-altitude natives illustrate the influence ofgenomic factors, i.e., the interaction between differentgenes on the one hand and that between genes and theenvironment on the other hand. Both respiratory andnonrespiratory stimuli are able to affect the developmentof the HVR (developmental plasticity) causing alterationsin response that may persist throughout adulthood. Insection II we discussed the influence of changes in oxygentension in the neonatal period. One clear example of anonrespiratory stimulus leading to plastic changes isstress from neonatal maternal separation that leads to apermanently augmented HVR of male but not female rats(262). Brain stem areas involved in ventilatory control of theneonate have strong plastic capacities depending on specificenvironmental stimuli and genetic background (735).

V. PATHOPHYSIOLOGICAL INFLUENCES ON

THE HYPOXIC VENTILATORY RESPONSE

A. Biological Variability

Apart from a between-subject variability, one has toconsider the appreciable within-subject variation in themagnitude of the HVR. The coefficient of variation ofrepeated HVRs obtained over a 2-h period ranges from 10to 60% among subjects (mean 20%; Ref. 674). Between-dayvariability is 20–50% greater than within-day variability.The cause of the large between-subject variability may liein differences in genetic make-up and race of the different



subjects tested. However, this explains just part of thevariability (see sect. IV). Other causes are related to phys-iological factors (such as age, metabolism, body temper-ature, the circadian rhythm, hormonal status, and preg-nancy), psychological factors (anxiety, full bladder),pathophysiological factors, the use of pharmacologicalagents that may affect HVR, and environmental factors(previous exposure to hypoxia, altitude). To reduce within- andbetween-subject variability, it is important to performhypoxic experiments in subjects that feel comfortable with thesetup, protocol, and laboratory surroundings and have noapparent urges that may influence their responses (e.g., afull bladder). Also the investigator should do everythingnot to influence the subject (for example, whispering mayhave a negative influence on the subject).

B. Ageing

Breathing is affected by ageing due to a decline infunction at multiple levels of the neuromechanical linkbetween chemosensors, brain stem, and ventilatory pump,causing a reduced ability to maintain blood gas homeosta-sis (360). Morphometric studies of rat carotid bodies in-dicate the occurrence of degenerative changes in old age(142, 179, 610). The carotid bodies of aged rats (�23 mo)show an increase in extracellular matrix, a reduction inthe number and volume of type I cells, and the presenceof necrotic cells (142, 179, 610). In the remaining func-tional type I cells, mitochondrial volumes and neurose-cretory fill-up of vesicles are reduced (610). Carotid bodycatecholamine content and turnover time is reduced withage, as is the catecholamine response to hypoxic stimu-lation (142). Conde et al. (142) measured afferent carotidsinus nerve activity during hypoxic and CO2/acidotic stim-ulation in young (3 mo) and old (18 mo) rats. While no ageeffect was observed in the nerve response to CO2/acidosis, in old animals a �50% reduction was observed inthe response to hypoxia. Degenerative changes such asfibrosis and loss of glomus tissue have been observed inthe human carotid body (311). In 15–30% of humans intheir eighth and ninth decade, there may be signs ofchronic carotid glomitis with infiltration of T-cell lympho-cytes (311). These data collectively indicate importantage-dependent morphological and functional changes ofthe carotid bodies.

A minority of human studies observed a reduced HVRin older persons (age range 64–79 yr; Refs. 412, 604).However, the majority of studies observed no differencesin HVR between young (20–30 yr) and old (60–79 yr) (6,609, 610, 714), while one study observed an increase inHVR (124). The overall picture that emerges from thehuman literature is that HVR is marginally affected at oldage and hence suggests that carotid body function re-mains unaltered. This stands in sharp contrast to the

morphometric and functional studies in carotid bodies ofaged rats (see above). However, it may well be that thehuman studies were performed at the older end of normaladult function when no functional decline in carotid bodyfunction is yet apparent (714). Another possible cause forthe difference between morphological appearance of thecarotid body and its function may be found in the exis-tence of redundancy in carotid body-mediated processes.A reduced number of type I cells may be able to sustainthe functional tasks of the carotid bodies (610). Finally, afailing carotid body function in old age may result inplastic changes in brain stem areas involved in the pro-cessing of its afferent output, causing the (partial) resto-ration of the HVR (610).

C. Gender and Pregnancy

Sex hormones (progesterone and the combinationprogesterone/estrogen as well as testosterone) have astimulatory effect on breathing and HVR. Progesteronestimulates breathing via hypothalamic sites through es-trogen-dependent receptors (46). The role of progester-one in causing an increase in HVR is twofold: it increasescarotid body sensitivity to hypoxia, and in combinationwith estrogen, it has a central modulatory effect on itsafferent output (299). Testosterone increases breathingand the HVR probably through an increase in metabolicrate and central modulatory effects (766, 841).

The majority of human studies could not find a dif-ference in HVR between men and women (157, 362, 364,418, 677, 680, 681). However, most of the studies involveddid not control for the phase of the menstrual cycle ortested women in just one phase (often the follicularphase). The number of animal studies on gender effect onHVR is limited, but all, in contrast to the human data,point toward a greater carotid body response to hypoxiain female animals (366, 530, 765). Only part of these sexdifferences is related to the acute effects of sex steroids.Castration of the adult animals does lead to a reducedHVR in males and females, but the sex differences persist(767). This is not surprising taking into account the long-term developmental and organizational effects of sex ste-roids on neuronal systems that occur in prenatal and earlypostnatal life. In this light, the absence of clear sex dif-ferences in the human population seems remarkable.

An effect of the menstrual cycle on the HVR is smalland ranges from an increased HVR in the luteal phase by10–20% compared with the follicular phase (157, 840).Pregnancy is associated with hyperventilation due to anincrease in metabolic rate and the stimulatory effects ofprogesterone (636). Also, the HVR is increased duringpregnancy, an effect that is already apparent at 20 wk andreaches a maximum at 36 wk (521). More than 60% of theincrease in HVR is related to sex hormones, the rest to the



increase in metabolic rate (521, 636). A more extensiveoverview of the effect of sex steroids on ventilatory con-trol during development and in adult life has been pub-lished recently (52).

D. Influence of CO2 on the HVR:

O2-CO2 Interaction

In both humans and animals, there is ample evidencefor a multiplicative effect of O2 and CO2 on the HVR (12,150, 178, 156, 195, 494, 517, 777). It is uncertain if quanti-tatively this O2-CO2 synergism observed at the ventilationlevel is entirely due to a multiplicative interaction in thecarotid bodies (423; further references in Ref. 178).

However, CO2 does not always act as a multiplier ofthe HVR. In conscious cats, O2-CO2 synergism does notseem to exist, due to a negative interactive effect on tidalvolume (575, 576). When anesthetized, however, this spe-cies demonstrates interaction at both carotid body andventilation levels (234, 423). Although displaying multipli-cative interaction in the carotid bodies (601), the anesthe-tized rat only seems to display a positive synergistic hy-percapnic-hypoxic action at the level of ventilation athypocapnic arterial PCO2 values (145, 613). Earlier studiesshowed that in other species O2-CO2 interaction may behypoadditive or at best additive (references in Ref. 145;more recent studies are discussed below).

A modulating effect of CO2 on the HVR could origi-nate from three sources: the carotid bodies, a central-peripheral interaction, and/or central O2-CO2 interaction.While a multiplicative interaction in the carotid bodies isevident, the brief overview below shows that there is noconsensus as to the modality of the peripheral-centralinteraction: hyperadditive, simply additive, or hypoaddi-tive. There is no evidence for central O2-CO2 interaction.

1. Carotid sinus nerve recording: properties of

type I cells

A relatively simple method to study O2-CO2 interac-tion in the carotid bodies is with carotid sinus nerverecording. Anesthetized cats, rats, lambs, and rabbitsdemonstrate O2-CO2 interaction in the carotid sinus nerve(116, 118, 119, 234, 333, 386, 423, 601, 660; further refer-ences in Ref. 233) that most likely can be reduced to theproperties of type I cells. The few studies that have beenperformed in type I cells may indeed suggest an interac-tion between acid/high CO2 and low O2 on intracellular[Ca2�] (167, 668). As outlined by Peers (592), a thoroughquantitative analysis of the relation between neurotrans-mitter release and intracellular [Ca2�] is required to shedmore light on the mechanism of stimulus interaction intype I cells. Other neurotransmitters than dopamine aloneshould be implicated in such analysis because the in-crease in afferent carotid sinus nerve activity by hyper-

capnia and hypoxia is not tightly coupled to dopaminerelease (100, 351, 743). The O2 sensor may behave like ahemoglobin molecule and, analogous to the Bohr shift, O2

binding may be competitively inhibited by CO2/H�, ex-plaining the finding that the maximal effect of a very highPCO2 (�270 Torr) was the same regardless of the PO2 (180and references therein).

2. Separate perfusion of peripheral and

central chemoreceptors

A) DOGS. In a majority of early studies utilizing selec-tive perfusion of the carotid bodies, hypoadditive effectson ventilation of peripheral and central stimuli were re-ported (references in Ref. 5). Multiplicative O2-CO2 inter-action at the level of minute integrated phrenic nerveactivity was also reported (132). More recently, Smith etal. (707) showed a clear multiplicative action of periph-eral O2 and CO2 on ventilation but also a much largerinhibiting effect on the HVR of systemic hypocapnia thanof carotid body hypocapnia alone, thus not supporting ahypoadditive peripheral-central interaction. In the sameanimal model, it was recently shown that isolated carotidbody perfusion with hyperoxic hypocapnic blood led to asubstantial decrease in ventilation despite a considerablerise in PCO2, not only suggesting an important role of thecarotid body in euoxic/normocapnic ventilation but also ahyperadditive peripheral-central interaction (77).

B) GOATS. Awake goats subjected to selective carotidbody perfusion showed O2-CO2 multiplication at the levelof both carotid bodies and ventilation without any periph-eral-central interaction other than simply additive (163164, 165). In awake goats instrumented for ventriculocis-ternal perfusion, carotid body stimulation with NaCN in-creased ventilation more at a relatively alkaline perfusatepH, possibly suggesting negative peripheral-central inter-action. However, in this study an alkaline cerebrospinalfluid pH was associated with much higher PaCO2

levels(708), thus almost inevitably leading to more intense cen-tral stimulation and potentially to simple additive centraleffects consistent with the findings in goats subjected toselective carotid body perfusion (163, 164, 165).

C) CATS. Anesthetized cats subjected to artificial brainstem perfusion did not show any central-peripheral inter-action other than simply additive (314, 793). As in awakegoats and dogs (163, 707), selective manipulation of pe-ripheral gas tensions showed a multiplicative O2-CO2 ef-fect on ventilation (314, 793). These findings might notentirely be compatible with those in the anesthetized catshowing that, with a constant level of hypoxia and a linearCO2 response of the carotid sinus nerve, CO2 had asmaller effect on ventilation below the eucapnic PCO2 thanabove it, suggesting some form of hyperadditive periph-eral-central interaction (422). The disparity between thefindings in anesthetized and awake cats (awake animals



showing a negative peripheral-central interaction of tidalvolume; Refs. 575, 576) cannot be explained by other thanunidentified influences of anesthesia and/or arousal ef-fects of hypoxia during wakefulness.

D) RATS. Utilizing an in situ dual perfusion system inthe rat to separately perfuse peripheral and central che-moreceptors, Day and Wilson (172) found a hypoadditiveeffect of the central PCO2 on the hypoxic phrenic ampli-tude response. Perhaps this may explain the large HVR inthis species at hypocapnic but not hypercapnic PaCO2

levels (145).

3. Dynamic end-tidal forcing

In both animals and humans, peripheral-central inter-action can be studied using end-tidal forcing (DEF), en-abling one to estimate separate gains of the central (Gc)and peripheral (Gp) chemoreflex loops, respectively (175,747; see sect. VI). In the cat, Gc and Gp (assessed withboth DEF and artificial brain stem perfusion) are indepen-dent, indicating no central-peripheral interaction otherthan additive (175, 314, 793). In addition, in the cat, O2-CO2 interaction takes place at the level of the carotidsinus nerve (234, 423), so the increase in ventilatory CO2

sensitivity with hypoxia in this species seems entirely dueto a peripheral O2-CO2 interaction, compatible with anincreased Gp but unaltered Gc in overall hypoxia (e.g.,Refs. 779, 793). In young piglets, hypoxia increased Gp butdid not alter Gc (845). In humans, an unchanged Gc withhypoxia was not always observed, with some studiesshowing a small but significant rise (226, 227) but othersreporting no significant change (56, 156, 591). When Gpwas lowered by inhalation of subanesthetic concentra-tions of volatile anesthetics, Gc did not change, not sup-porting an other-than-additive central-peripheral interac-tion (161, 796, 797).

4. Dynamic end-tidal forcing in carotid body

resected subjects

Bellville (56) reported an �50% lower central chemo-sensitivity in carotid body-resected subjects comparedwith healthy controls, indicating that a peripheral-centralinteraction other than simply additive cannot be ex-cluded. One to 26 six years after bilateral removal of thecarotid bodies due to paraganglioma, subjects displayed alower central CO2 sensitivity than an age- and sex-matched control group (227). Three days after carotidbody resection (due to hereditary paraganglioma), threepatients showed a substantially reduced central CO2 sen-sitivity that gradually recovered in the following 1–3 years(154). In the anesthetized cat, carotid sinus nerve transec-tion did not reduce central CO2 sensitivity in an acutesetting (314). Thus removal of a tonic input from thecarotid bodies could lead to a resetting of central CO2

sensitivity that needs time to develop. Whether such a

relatively slow process may also be responsible for thelower CO2 sensitivity and higher resting arterial PCO2 aftercarotid body denervation in goats and dogs remains to beestablished (579, 663).

5. Other approaches used in humans

In earlier studies, so-called withdrawal tests wereperformed, suddenly removing a central or peripheralstimulus keeping the background peripheral or centralgas tensions, respectively, at constant levels (reviewed inRef. 150). Underlying assumptions such as lack of activityand absence of undershoots in carotid sinus nerve dis-charge in (mild) hyperoxia make these studies difficult tointerpret (e.g., Ref. 325). Other workers compared tran-sient and steady stimuli assuming that responses to theformer originate at the peripheral chemoreceptors andthose to the latter at both peripheral and central sites (seeRef. 206 for earlier references). The role of HVD could notalways be taken into account, as well as the possibleexistence of an interaction component with a fast timeconstant (e.g., Ref. 656).

One way to study peripheral-central interaction is toexpose subjects to peripheral stimuli at different centralPCO2 levels. Robbins (656) compared the effects of a stepdecrease in end-tidal O2 performed just 30 s after impos-ing a step decrease in end-tidal CO2 with those performedat constant steady-state end-tidal CO2. In two of threesubjects, the response was larger under the former con-dition, suggesting a more-than-additive central-peripheralinteraction (alternative explanations were left open; Ref.656). However, St Croix et al. (724) could not reproducethis. Clement et al. (136) found no evidence for differentresponses to isocapnic changes in arterial pH induced byinfusion of acid and bicarbonate performed at quite dif-ferent background PCO2 levels. With the assumption thatwith acid-base changes the peripheral chemoreceptorsrespond uniquely to changes in arterial [H�] (190, 292,334, 687), these findings provided no evidence for aninteraction other than simply additive (136).

Recently, we measured hypoxic sensitivity in hu-mans, defined as the ratio �VI/�log PaO2

, at three differentlevels of constant end-tidal CO2. The rationale of using theratio �VI/�log PaO2

rather than the conventional ratio�VI/�SpO2

is explained in section VIB (see also Ref. 775).As also demonstrated in the cat (650), humans display asteep and sound linear relationship between hypoxic sen-sitivity and arterial [H�], representing a powerful O2-CO2

interaction (Fig. 7; Ref. 775). But is hypoxic sensitivity aunique function of arterial [H�]? This seems unlikely be-cause it would imply an almost infinitely high sensitivityin a metabolic acidosis with a pH of �7.3 ([H�] �50 nM)which, for example, can be induced by the carbonic an-hydrase inhibitor acetazolamide (775, 778). To investigatethis further, bicarbonate was infused isocapnically, and



this yielded two isocapnic and two virtually isohydricsituations with different hypoxic sensitivities as shown inFigure 7. From these data (775), we can draw severalconclusions. First, O2 sensitivity increases with an isocap-nic rise in arterial [H�], and this is presumably due tointeraction at the carotid bodies (cf. points c and f and b

and d in Fig. 7). Second, although with acid-base alter-ations extracellular H� may be the adequate stimulus tothe carotid bodies, there is no unique relationship be-tween hypoxic sensitivity and arterial [H�] (cf. points f

and a in Fig. 7). Thus, in this regard, our data do notsupport earlier conclusions (e.g., Refs. 250, 349), but nei-ther do they provide conclusive evidence for the carotidbodies as the exclusive site of the O2-CO2/H� interaction.The lower hypoxic sensitivity in point a compared with f

in Figure 7 could be due to the lower carotid body PCO2

(peripheral O2-CO2 interaction) combined with a dimin-ished efficacy of afferent carotid body input to increaseventilation due to the also lower central PCO2. Perhaps amore likely explanation is that it is the intracellular envi-ronment (pH?) of carotid body cells that, together withthe extracellular pH, determines O2 sensitivity.

Roberts et al. (660) subjected humans to mechanicalventilation to produce apnea at different end-tidal CO2

levels varying from 2.5 Torr above to 7.5 Torr beloweucapnia. The background end-tidal CO2 dose-depen-dently influenced the magnitude of the EMG response toa “standard” peripheral chemoreceptor stimulus that wascompletely absent at an end-tidal CO2 �7.5 Torr below

eucapnia (the stimulus consisted of a rapid desaturationwithin 30 s with the end-tidal CO2 briefly adjusted to thesubject’s eucapnic value). These data may suggest that anaugmented carotid body output is translated into an in-crease in respiratory motor output only at sufficientlyhigh central chemoreceptor PCO2 levels. Corne et al. (144)used volume cycle-assisted ventilation to produce steady-state mild and moderate hypocapnia while maintainingstable spontaneous breathing. At arterial PCO2 values 5–10Torr below eucapnia, but not in mild hypocapnia or eu-capnia, a ventilatory response to hypoxia was totally ab-sent. Although clearly of clinical significance, a clear pic-ture of peripheral-central interaction cannot be deducedfrom this approach because a separation with O2-CO2

interaction at the carotid bodies cannot be made. In ad-dition, an influence of developing HVD in these subjectscannot be excluded even though the hypoxic tests did notoutlast 5 min: at the end of the hypoxic test in moderatehypocapnia (in the absence of an HVR), subjects venti-lated less than in the prehypoxic period suggesting HVD(144).

Yang et al. (863) introduced a method for estimatingthe dynamic response to simultaneous changes in PCO2

and PO2 in humans, and their data could be well describedwith a two compartment model, one comprising the slowCO2 and the other the fast O2-CO2 interaction component.The interaction could be entirely modeled by the fastcomponent attributed to the peripheral chemoreceptors,thus not supporting a significant central O2-CO2 interac-tion and/or more than simply additive peripheral andcentral effects.

Agents not crossing the blood-brain barrier and notinfluencing central CO2 sensitivity could be used toinvestigate possible selective effects on the carotidbodies. Acetazolamide, a moderately permeable car-bonic anhydrase inhibitor, will not have penetrated intothe brain in significant amounts 1 h after administrationof a low dose. About 1 h after infusion of 4 mg/kg, theO2-CO2 interaction in healthy subjects was completelyabolished, and this was reversible with antioxidants(Fig. 8; Ref. 777). Infusion of 500 mg in humans did notchange hyperoxic CO2 sensitivity, indicating an unal-tered central ventilatory CO2 sensitivity (750). With theassumption of no central effects of acetazolamide,these findings suggest that in humans, the O2-CO2 in-teraction entirely resides in the carotid bodies. A usualclinical oral dose of acetazolamide does not affect theinteraction, indicating that its effect depends on thedose and route of administration (750, 775). Animaldata suggest that analogous to its effect on hypoxicpulmonary vasoconstriction (749), the inhibitory actionof acetazolamide on the HVR may occur independentlyfrom carbonic anhydrase inhibition (776).

FIG. 7. Relationship between hypoxic ventilatory response andarterial H� concentration in humans. Hypoxic sensitivity, defined as�VI/�log PaO2

as a function of arterial [H�]. The right line represents thecondition with normal arterial bicarbonate concentration in the arterialblood. The acute hypoxic response (i.e., the response 3 min after a stepdecrease in end-tidal O2) is measured at three constant levels of arterialPCO2 (numbers are arterial PCO2 values). There is a sound linear rela-tionship between hypoxic sensitivity and arterial H� concentration in-dicating a powerful O2/CO2-H� interaction at ventilation level. Whensubjects are given intravenous bicarbonate (left line), the line shifts tolower values of arterial [H�] in an approximately parallel way. Thusapparently hypoxic sensitivity is not a unique function of the arterial[H�]. Points a and f are virtually isohydric points, and the lower hypoxicsensitivity in point a is presumably due to the lower PCO2 in the carotidbodies. See text for further explanation. Data are means � SE from 8young healthy volunteers. [Adapted from Teppema et al. (775).]



6. Conclusion

In all species investigated, O2 and CO2 have multipli-cative actions on carotid body output both when awakeand anesthetized, but the interaction site within type Icells has not yet been established. The existence of amultiplicative O2-CO2 interaction on ventilation dependson the species, state of consciousness, and the peripheral-central interaction mode. Generally, humans show O2-CO2 multiplication, but it appears difficult to characterizethe mode of peripheral-central interaction, although thereis no conclusive evidence for hypoadditive central andperipheral effects. Direct anatomical connections be-tween the first relay station of the peripheral chemore-ceptors in the NTS and central chemoreceptors in theretrotrapezoid nucleus (RTN) as described for the rat(753), may render a clear separation of central and pe-ripheral contributions elusive. In addition, a subset of CO2

chemoreceptors in the RTN receives an inhibitory inputfrom (mainly slowly adapting) pulmonary stretch recep-tors (525), providing another complicating factor in theinterpretation of data from intact as well as vagotomizedanimals.

E. Carotid Body Resection and the HVR

Resection of the human carotid bodies was per-formed for a variety of reasons. Both unilateral and bilat-eral resections were carried out in patients with severebronchial asthma and chronic obstructive pulmonary dis-ease. Although the main goal for resection in these pa-tients was often achieved (that is, the alleviation ofbreathlessness), the therapy is controversial and cur-rently not practiced on a wide scale (326, 475, 812). An-other reason for carotid body resection is the develop-

ment of local tumors and/or hyperplasia that, for example,may arise from genetic defects (see sect. IV) and fromchronic hypoxemia related to cyanotic heart disease, pul-monary disease, and high altitude (154, 312). Inadvertentloss of carotid body function may arise from surgicalprocedures in the neck region, such as carotid endarteri-ectomy and neck dissection as part of the treatment ofhead and neck cancer (524, 812).

Both animal and human studies show immediate hy-poventilation, respiratory acidosis, and loss of the hyp-oxic drive in response to bilateral carotid body resection(Fig. 5; Refs. 154, 473, 474, 492, 570, 664). In a smallhuman population tested before and after bilateral re-moval of the carotid bodies for tumors due to a mutationin the SDHD gene, end-tidal PCO2 increased by �8 Torrwithin the first day after resection (154). Similar observa-tions were made in awake rats over a 75-day period aftercarotid body excision (570). These data indicate that thecarotid bodies are essential in maintaining normal venti-lation during air breathing. Especially in the neonatalperiod, elimination of the carotid bodies can have pro-found effects on the control of breathing, depending onthe denervation technique used and the developmentalperiod in which it is performed (238 and references citedtherein). Partial restoration of the hypoxic drive is ob-served in some species (rat, goat, cat, pony, piglet) and ismost likely in neonatal animals (69, 473, 474, 492, 710).The rate of recovery is species dependent and rangesfrom weeks (piglet, rat) to months (goat) to years (pony).The compensatory mechanisms of the partial recovery ofthe hypoxic drive in these animal studies remain un-known. Potential mechanisms include reinforcement ofaortic bodies, activation of medullary oxygen-sensingneurons, chemosensory input from secondary glomus tis-sue in the head/neck region (such as the vagal body in the

FIG. 8. Effect of acetazolamide on O2-CO2 interaction in humans. Left: intravenous low-dose (4 mg/kg) acetazolamide (ACTZ) lowers the acutehypoxic response in normocapnia (N) but also abolishes the increase in its magnitude in hypercapnia (H). Thus this agent, which at the applied doseis unlikely to have direct effects on the central nervous system, is able to abolish the O2-CO2 interaction by a peripheral action most likely locatedin the carotid bodies. An antioxidant cocktail (AOX; �-tocopherol and ascorbic acid) reversed the effects of ACTZ. Right: control study in whichafter the ACTZ, subjects received placebo (PLCB) instead of AOX. See text for further explanation. [From Teppema et al. (777).]



neck), regeneration of sensory terminals of the sinusnerve, and/or plastic changes in the central respiratoryneuronal network. In humans, we were unable to observeany recovery in the HVR over a period of 4 years afterbilateral removal of the carotid bodies (154). Similar ob-servations were reported by Wade et al. (812) in patientsup to 8 years postsurgery. Honda et al. (326) did showsome restoration of the HVR in patients two decades afterbilateral carotid body resection, although this effect wasdependent on the specific hypoxic test employed.

F. Chronic Hypoxia and the HVR

1. Ventilatory acclimatization to chronic hypoxia

We define chronic hypoxia as exposure to hypoxia orhigh altitude lasting several hours to days, weeks, ormonths. A hallmark of adaptation to high altitude is agradual rise in ventilation with a time course dependingon the altitude (ventilatory adaptation to high altitude,VAH). In humans, at very high altitudes (8,000 m andhigher), this may take at least 30 days (839). At lessextreme altitudes, complete VAH requires �10 days (seeRef. 68 and references cited therein), although at 3,800 mventilation appeared not to have reached a plateau yetafter 12 days (683). Dogs, goats, and ponies acclimatizemuch faster and reach complete adaptation after 3, 4–6,and 30 h, respectively (references in Ref. 68). When ex-posed to a simulated altitude of �4,600 m, cats show alarge decrease in arterial PCO2 after 48 h (809), but accli-matization may not yet be complete after this period inthis species since at 5,500 m arterial PCO2 continued to fallfor 10–12 days (774). Qualitatively, in the rat, VAH israther similar to that in humans albeit with a somewhatfaster time course, and this species is considered anadequate animal model for the adaptation in humans(569).

Initially it was suggested that VAH could be attrib-uted to changes in arterial and later to brain stem acid-base status, but eventually it became clear that alterationsin acid-base status during VAH should be considered theconsequence of ventilatory changes rather than the cause(178). Here we do not focus on studies aimed at elucidat-ing the role of the central chemoreceptors in VAH, and thereader is referred to excellent reviews on this subject (68,540, 543). Briefly, there is no conclusive evidence for anessential role of the central chemoreceptors, for example,because the time courses of medullary extracellular fluidpH and VAH do not run in parallel (534, 543). Futurestudies are warranted, however, to unmask a possiblerole of central chemoreceptors in VAH for a reason notanticipated before: during hypoxia in the rat, centralchemoreceptors in the retrotrapezoid nucleus arestrongly activated via a glutamatergic pathway fromcarotid body projection areas in the NTS (753). Thus,

during chronic hypoxia, these neurons may receive atime-dependent progressively increasing synaptic inputthat may overcome inhibitory influences of a lowerPCO2 and may thus contribute to VAH.

2. VAH is associated with an increase of the HVR

Studies in the last two decades have convincinglyshown that VAH is associated with a time-dependent aug-mentation of the HVR that at least partly is due to anincreased carotid body sensitivity. Initial studies in dogs,ponies, goats, and rats had already demonstrated thatcarotid body denervation abolished or at least minimizedVAH (reviewed in Ref. 68). Below we summarize the mostimportant studies in goats, cats, rats, and humans show-ing the crucial role of both the carotid bodies and the CNSin mediating an increase in the HVR in chronic hypoxia.

1. Animal studies

In goats (mainly awake) with vascularly isolated ca-rotid bodies, Bisgard, Forster, and co-workers collectedmany data on the mechanism of VAH that can be summa-rized as follows (66, 67, 105, 200, 216, 546, 836; furtherreferences in Ref. 68). 1) Animals exposed to chronichypoxia show complete VAH within 4–6 h. 2) IsolatedCNS hypoxia and/or hypocapnia do not evoke VAH, indi-cating that it must have a peripheral origin. 3) VAH re-quires intact carotid bodies that are perfused by hypox-emic blood; isolated carotid body hypercapnia results in arapid sustained rise in ventilation but does not induce thetime-dependent fall in arterial PCO2 typical for VAH.4) Both chronic poikilocapnic and isocapnic carotid bodyhypoxias increase the isocapnic acute hypoxic response(AHR), indicating that the enhanced sensitivity developsregardless of the PCO2 at the carotid bodies. 5) Duringisocapnic carotid body hypoxia for 6 h, carotid sinusnerve afferents of anesthetized goats show a progressive,time-dependent increase in impulse frequency that quantita-tively can account for VAH.

When exposed to a barometric pressure of 460–440Torr for 2–3 days, cats demonstrated a large time-depen-dent drop in end-tidal CO2, increases in both the HVR, andafferent carotid sinus nerve activity as well as an aug-mented isocapnic carotid body sensitivity to hypoxia(771, 808). Animals exposed to 10% O2 exhibited a higherisocapnic carotid sinus nerve sensitivity to hypoxia after21 days but not during the initial few hours (34). Inanimals kept at a constant end-tidal O2 of 30 Torr for 7 h,nerve activity started to increase after 3 h (181). However,Lahiri et al. (428) were unable to demonstrate an increasein afferent nerve sensitivity to hypoxia after acclimatiza-tion to 10% O2 for 21–24 days. Finally, Tatsumi et al. (768),after keeping cats at 375 Torr for 3–4 wk, reported atime-dependent decrease in end-tidal CO2 along with adecreased HVR and isocapnic hypoxic carotid sinus nerve



sensitivity. An altitude-dependent decrease in HVR alongwith VAH had already been reported in earlier studies(e.g., Ref. 774; further references in Ref. 68). Thus, after afew days of hypoxia, cats appear to have an augmentedHVR and an increased carotid body sensitivity, but aftermore than 2–3 wk, they show both a blunted HVR andcarotid body sensitivity. So, perhaps an increase in ca-rotid body sensitivity is sufficient for VAH in the first fewdays, but other mechanisms manifest after 2–3 wk.

The rat develops an augmented poikilocapnic HVRand ventilation in the initial few days of chronic hypoxiafollowed by a gradual return of the HVR to control or evenslightly below at the end of the first week (51, 91, 252,811). However, a fast “blunting” of the HVR in the rat hasnot been demonstrated very convincingly, for example,because in all these studies, it was measured underpoikilocapnic conditions. Rather, it may be a manifesta-tion of a gradual attenuation of acclimatization as alsooccurs in other species, but without blunting of the HVR(68). Aaron and Powell (1) showed a clear increase inisocapnic HVR in rats after 7 wk at 380 Torr. Note thatafter acclimatization this enhanced HVR was measuredat a constant arterial PCO2 that was lower than in controlby as much as 6–8 Torr, whilst both control and acclima-tized animals showed a positive O2-CO2 interaction (1).These authors also compared isocapnic HVRs of acclima-tized animals with that of controls that were made slightlyhypercapnic to achieve matching of minute ventilation(613). In control animals, the constant arterial PCO2 waskept at a level that was higher than in acclimatized ani-mals by �9 Torr, but despite this, the HVR in the latterwas much higher, suggesting that either the carotid bod-ies had undergone an impressive increase in O2 sensitivityand/or that the translation of afferent carotid body activ-ity into ventilatory output (CNS gain) was greatly facili-tated. It would be interesting to compare HVRs at equalhypercapnic PCO2 levels in control and acclimatized rats,because it cannot be excluded that a negative O2-CO2

interaction (145, 172) may have played a role in the ob-servations of Powell et al. (613).

Dwinell and Powell (201) found a large increase inthe CNS gain of a standard electrical input from thecarotid sinus nerve in acclimatized rats (380 Torr, 1 wk).However, cats exposed to 440 Torr for 48 h developedboth a greater isocapnic HVR and hypoxic carotid sinusnerve response but did not show an increased CNS gain(808). In goats, the effect of NaCN (assumed to causemaximal carotid body stimulation) on ventilation was notdifferent before and after 4 h of isocapnic or poikilocap-nic hypoxia (215), but this does not preclude changes inCNS gain during prolonged hypoxia in this species (615).Thus, in the rat, the most frequently used animal model tostudy VAH, an increased central gain has clearly beendemonstrated, while it cannot be excluded to exist also inhumans, goats, and ponies (references in Refs. 178, 615).

2. Studies in humans

In Table 3, we have collected studies mainly per-formed over the last 15 years (for a review of earlierstudies, see Ref. 68). Despite all the distinct exposureparadigms (stimulus level and duration, hypobaric hyp-oxia, or just lowered inspired O2), techniques and analysismethods that make quantitative comparisons very diffi-cult if not impossible, the outcomes of all these studieshave two things in common: an increase in resting and, ifmeasured, hyperoxic ventilation and an augmented (acute)hypoxic ventilatory response that develops during thefirst 4–8 h of acclimatization.

Robbins (659) made considerable efforts to spell outthe mechanisms of ventilatory adaptation to chronic hyp-oxia. Apart from a consistent increase in hyperoxic ven-tilation and an augmented acute response to hypoxia, hefound the following: 1) in agreement with results obtainedfrom goats, isocapnic and poikilocapnic hypoxias havethe same effect. 2) Neither eucapnic nor hypocapnic pas-sive hyperventilation evokes VAH or an enhanced HVR,indicating that hypocapnia or hyperpnea per se will notplay a significant role. 3) The effect of hypoxia to inducethe adaptations is due to the low PO2 rather than to lowoxygen content. 4) The mechanism underlying the in-crease in the acute response to hypoxia is fast (observ-able within a few hours as in goats), sensitive (end-tidalO2 10 Torr below resting is sufficient to evoke it), inde-pendent of peripheral adrenergic and dopaminergic sys-tems, and insensitive to general autonomic blockade (ref-erences in Table 3).

Chronic isocapnic and poikilocapnic hypoxia alsoalter peripheral ventilatory CO2 sensitivity. Using dy-namic end-tidal forcing (see sect. VI) and sophisticatedanalysis methods, Fatemian and Robbins showed an in-crease in CO2 sensitivity that most likely can be attributedto an increase in sensitivity of the carotid bodies (e.g.,Refs. 228, 229; further references in Table 3). In a study at4,300 m (629), the amount of VAH correlated well with themagnitude of the HVR at sea level, also supporting animportant role of the peripheral chemoreflex loop.

In a more recent study, Herigstad et al. (320) exposedsubjects to 8 h of isocapnic hypoxia prior to an incremen-tal exercise protocol. The exposure resulted in a decreasein end-tidal PCO2 that was essentially maintained at allwork loads until exhaustion. According to the alveolarventilation equation, this means that ventilation musthave increased in proportion to metabolism. The datafrom this study also implied that an altered acid-basebalance (isocapnia was maintained) or long-lasting in-creases in ventilation (days) are not required for thisphenomenon to occur. Thus VAH augments the ventila-tory response to exercise and in this sense resembles theeffect of hypoxia or carotid body stimulants (referencesin Ref. 320). Exercise also augments the stimulatory effect



TABLE 3. Carotid body adaptations to chronic hypoxia

Species Effects and CommentsDoes the Adaptation Increase or

Decrease CB Sensitivity or the HVR? Reference Nos.

Neurochemical adaptations

Dopaminergic system

Rat Upregulation of TH 65, 614 and referencescited therein

Cat TH upregulation 820Cat CH augments HVR after 2 days at 4,600 m

but domperidone fails to furtherincrease it, suggesting dopaminergicdownregulation

Increase 771

Cat Prolonged hypoxia decreases HVR;domperidone restores it to controllevel, suggesting a role of DAupregulation in long-term adaptation

Decrease 770

Cat Dopaminergic system in efferentinhibitory fibers in the CSN isupregulated not supporting disinhibitionof this system in CH

Decrease 429

Rat After 5 and 2–3 wk of CH, domperidonerestored “blunted” poikilocapnic HVR,attributed to increased CB DA content

Decrease 252, 811

Rat 1–2 days of CH attenuate dopaminergicinhibition of the HVR, but after 8 daysit has increased compared to control

Increase and decrease,respectively

51, 342

Rat CB D2R decreased by 59% after 2 days ofCH, but increased by 274 % after �11days

Increase and decrease,respectively

344

Rabbit After 2 days of CH, CB DA decreased,but 2 days after termination of an 8-daylasting exposure it was higher thancontrol

? 343

Mice Male D2R�/� mice failed to show VAHand augmented HVR after 2–8 days ofCH

Increase, but note possiblecentral effects

343

Rat Type I cells of juvenile rats showexaggerated basal DA release

? 355

Goat The peripheral DA antagonistdomperidone had equal effects inacclimatized and control animals, andlocal CB application of DA did notinfluence the time course andmagnitude of VAH

Neither 358, 359

Human Effects of DA and domperidone did notchange after 8 h of CH

Neither 590

NorepinephrineCat CH for 24–36 h reduced the inhibitory

effect of �2-adrenergic antagonism,suggesting downregulation of CBnoradrenergic mechanisms

Increase 110

Variousspecies

Upregulation of CB NE Decrease References in 65, 278, 818

Goat CB sympathectomy did not alter the timecourse of VAH, the acute HVR and theinhibitory effects of NE and DA

Neither 672

Human VAH and CH-induced augmentation of theHVR are not affected by �-adrenergicand autonomic ganglion blockade

Neither 134, 452, 520

AcetylcholineRat In CB preps from animals exposed to CH

for 9–22 days, ACh had a largerexcitatory effect on CSN dischargecompared with control; however, thenAChR antagonist mecamylamine failedto cause any reduction in hypoxia-induced increase in CSN activity

Increase 309



TABLE 3. —Continued



Rat In type I cells cultured in normoxiamecamylamine did not alter basal DArelease, but in those cultured in hypoxiafor �12 days it completely inhibited it,possibly explaining its failure to inhibit thehypoxia-induced rise in the preparation ofHe et al. (309)

Increase 355

ATPRat In superfused CB preps from rats exposed to

CH for 9–16 days, P2X2 purinergic but notnicotinic antagonists retained the ability toblock hypoxia-induced rise in CSN activity,suggesting an altered purinerg/cholinergicbalance; afterdischarge in CSN fromacclimatized animals was blocked bypurinergic blockers suggesting prolongedATP release as its cause and hypotheticallyproviding a mechanism for continuoushyperventilation upon return to normoxiafrom high altitude (deacclimatization)

Increase 307

Nitric oxideRat Upregulated in hypoxia Decrease 65, 865 for reviewRat In CB-CSN preps from rats exposed to CH

for �16 days, L-NAME caused aprogressively greater stimulation and anNO donor a larger inhibition

Decrease 308

Substance PCat CB Substance P downregulated in animals

exposed to 10% O2 for 2 wkDecrease 820

Endothelin-1Rat Upregulation of ET1, and ETA receptors in

type I cells from animals exposed to CHfor �16 days. Selective inhibition of ETA

receptors reduced basal CSN activity onday 0 by 11%, but halved it at day 16

Increase 128

Rat During CH for 16 days, daily administrationof the ETA/B antagonist bosentandiminished the increase in hypoxic CSNactivity and also greatly reduced CBenlargement and mitotic type I cell activity

Increase 129

Human CH for 4 h increased plasma ET-1;continuous infusion of ET-1 during a 4 hbout of CH exaggerated the increase inHVR

Increase, assuming a stimulatoryeffect of ET-1 on the CB

755

Angiotensin IIRat 10% O2 for 4 wk progressively increased CB

angiotensinogen mRNA, upregulated AT1

receptors and ACE, and increased CBsensitivity to ANG II

Increase 431, 432, 443

Proinflammatorycytokines

Rat Upregulation of IL-1, IL-6, and TNF-� andtheir receptors in type I cells from ratexposed to 10% O2 for 3, 7 and 28 days;after CH, the hypoxia-induced rise inCa2�i was enhanced by the cytokines

Increase 433

Rat Rats exposed to a barometric pressure of 380Torr for 4 wk showed increased CB IL-1�and TNF-� after 1 day, well beforemacrophage invasion; cytokine levelsnormalized after 28 days, except IL-6 dueto increased production by the CB;ibuprofen and dexamethasone inhibitedCH-induced increase in hypoxic CSNactivity

Increase 455






MelatoninRat Exposing rats to 10% O2 for 4 wk

increased CB melatonin receptor mRNAIncrease 785

Adaptations of ion channels and membrane currents

Maxi-K channels Rat Decrease in K� current density but notthat caused by maxi-K channels,proposed to result in a relative largercontribution of the latter to the restingmembrane potential and an increasedhypoxia-induced catecholamine release

Increase 114

Sodium current Rat Higher current density in type I cellisolated from 5- to 12-day-old rat andcultured in hypoxia for 1–2 wk

Increase 726, 727

“Oxygen-sensitiveand -insensitive” Kv

channels

Rabbit In type I cells cultured in hypoxia for72 h, potassium channels responsiblefor an O2-insensitive component of theK� current were downregulated andthose contributing to the O2 sensitivecomponent upregulated; thus K�

current inhibition by hypoxia wasincreased

Increase 368

Sodium channelNav1.1

Rat CB from animals exposed to 10-11% O2

for 1 wk contained type I cells withupregulated voltage-gated sodium(Nav1.1) channels; the sodium channelactivator veratridine caused aseveralfold greater increase in CArelease in these cells, compared withthose from control animals

Increase 106

Gap junctions Rat Exposure to an ambient pressure of 380Torr was followed by increasedexpression of the gap junctional proteinconnexin43 in type I cells and petrosalganglion neurons

Increase? 127

Morphologic adaptions of the carotid body

Variousspecies

Organisms living at high altitude haveenlarged CB

Increased in short-term CH butdecreased in prolongedhypoxia

19, 209, 419, 818

Rat CH induces vascular remodeling,endothelial proliferation, andvasodilation

Unknown 135, 419, 420, 784, 818

Rat Hyperplasia of type I cells during the firstweek of CH, no further increasethereafter; newly formed cells cansurvive for 1 mo

Increase 818, 819

Mice Exposing mice to CH for 20 days resultedin cell division in CB after 1 day, butnew type I cells appeared after 5 days.Initially type II cells start to divide andare replaced by “progenitor,” i.e., stemcells that differentiate into type I cellswith normal electrophysiologicalfeatures. Upon return to normoxia, 50%of the type I cells are freshly formedand the total amount gradually returnsto normal

Increase 581

ACh, acetylcholine; ACE, angiotensin converting enzyme; ANG II, angiotensin II; AT1 receptor, angiotensin I receptor; CH, chronic hypoxia;CA, catecholamine; CSN, carotid sinus nerve; CB, carotid body; DA, dopamine; D2R, dopamine 2 receptor; ET, endothelin; L-NAME, NG-nitro-L-arginine methyl ester, an inhibitor of NO synthase; HVR, hypoxic ventilatory response; IL, interleukin; NE, norepinephrine; nAchR; nicotinicacetylcholine receptor; NO, nitric oxide; TH, tyrosine hydroxylase; TNF, tumor necrosis factor; VAH ventilatory adaptation to chronic hypoxia orhigh altitude.



of hypoxia on ventilation, but this interaction is reducedby a prior exposure to chronic hypoxia, and this phenom-enon will be one of the determinants of the magnitude ofthe HVR at high altitude (848).

A decrease in intercept of the hyperoxic CO2 re-sponse curve was not observed by Fatemian and Robbins(228, 229), suggesting that the increase in resting andhyperoxic ventilation is due to a higher (peripheral) CO2

sensitivity. In one study, a leftward shift of the ventilatoryCO2 response curve was found after as long as 75 h ofpoikilocapnic but not isocapnic hypoxia (148). Studyingthe effect of chronic hypoxia with modified rebreathing(see sect. VI) yields a decrease in intercept of both thehypoxic and hyperoxic CO2 response curves at high alti-tude, but, possibly related to variation in altitude andacclimatization period, no consistent increase in periph-eral CO2 sensitivity (9, 719).

Two major mechanisms, to be discussed in the fol-lowing paragraphs, may contribute to the augmented HVRin chronic hypoxia: an augmented carotid body sensitivityand, as particularly studied in the rat, an increase in CNSgain. Animal studies do not provide direct evidence for animportant role of the sympathetic nervous system (e.g.,Ref. 672). In humans, prolonged hypoxia causes a largeincrease in sympathetic activity, but how this may influ-ence the HVR is unknown (107, 300). Also, note that thechronic hypoxia-induced HVR augmentation in humanswas reported to be unaltered after autonomic ganglionblockade (452). Neonates and adults adapt differently tochronic hypoxia. This is schematically shown in Figure 9.

3. Chronic hypoxia increases carotid body sensitivity

Chronic hypoxia elicits morphological, neurochemi-cal, and membrane-electrophysiological adaptations in

the carotid bodies that act together to increase hypoxicsensitivity. These adaptations occur in parallel, so theirseparate contribution to the increase in carotid body gainis difficult to assess. This is especially true for the contri-bution of neurotransmitters and/or modulators, some ofwhich have a controversial role in the carotid bodies. Forexample, the dopaminergic system seems to be upregu-lated in chronic hypoxia, but although dopamine is gen-erally considered an inhibitory neurotransmitter (mainlybased on observations on the effects of exogenous ap-plied agonists and antagonists), the significance of thisadaptation is uncertain because endogenous dopaminemay well have excitatory effects on the carotid bodies(for a discussion on the controversial role of dopamine inthe carotid body, see Refs. 278, 283, 869). Both an inhib-itory and excitatory role has also been suggested fornorepinephrine (references in Refs. 68, 278). In Table 4,we have summarized recent studies on various adapta-tions of the carotid bodies to chronic hypoxia. We haveadded a column containing the suggested effects of singleadaptations on carotid body sensitivity, but we emphasizethat with regard to neurotransmitters these must be con-sidered hypothetical, because what eventually matters isthe total balance between excitatory and inhibitory trans-mitters. In a recent review, Prabhakar (621) has proposedthe simultaneous presence of excitatory and inhibitorytransmitters to result in a “push-pull” mechanism betweenboth that, rather than causing exhaustion in the absenceof inhibitors, produces sustained activation in their pres-ence. This would be especially effective in the adaptationchronic hypoxia (621).

4. Increase of CNS gain during chronic hypoxia

An increased CNS gain as observed in the rat couldresult from neuroplasticity of the brain stem respiratorycircuitry that processes the afferent input from the carotidbodies. Below we discuss some mechanisms potentially un-derlying the increase in CNS gain in chronic hypoxia.

A) CATECHOLAMINERGIC SYSTEM. Rats challenged with 10%O2 for 1 wk showed a reduced brain dopamine turnoverand an increased carotid body dopamine content (571). Inrats exposed to chronic hypoxia for 2 wk, Schmitt et al.(685) found a positive correlation between the rise inventilation and tyrosine hydroxylase (TH) protein level inthe caudal NTS, i.e., the preferential projection area ofchemoafferent fibers. The same area, containing the A2noradrenergic cell group, displayed an upregulation of THafter 1 wk of chronic hypoxia, but this effect was reversedafter 2 and 3 wk of exposure, respectively (722). Whenadministered in the NTS, dopamine has a stimulatoryinfluence on ventilation (for references, see Ref. 345), andthus an increased tyrosine hydroxylase production mayfacilitate respiratory output in chronic hypoxia. In malenormoxic mice systemic droperidol, a dopamine antago-

VI

.VI

.

FIG. 9. Comparison of chronic hypoxia in neonates and adults.Effects of chronic hypoxia in neonates, within the critical time windowfor developmental respiratory plasticity, compared with those in adults.Question marks concern unresolved or controversial issues. VI, inspiredventilation. For explanations see text.



TABLE 4. Chronic hypoxia in humans

Protocol HVR Measurement Effects on HVR, CommentsReference

Nos.

Operation Everest II: subjectsacclimatized to 428 and 305 Torr

Progressive isocapnic rebreathing atPETCO2

of ambient air breathing1 iHVR about equal at both pressures.

Further elevation upon return to sealevel but note that the test PETCO2was �10 Torr higher than at thelowest pressure

686

Exposure to PB 447 Torr for 7 days Progressive isocapnic rebreathing ata PETCO2

of ambient air breathing1 iHVR 2–4 days after exposure;

postacclimatization resting PETCO2,

however, not provided.

273

Stay at 4,300 m for 19 days Progressive isocapnic rebreathing atsea level only; serial ventilationand gas measurements from day 1to 19

Higher HVR at sea level predicts lowerPETCO2

during the entire period629

pH: 5 days at 3,810 m Stepwise decreases in SpO2 to 77%.PETCO2

at hyperoxic value withVI �140 ml �min�1 �kg�1

Gradual increase of iHVR from 1-5days; test PETCO2

varied between 45and 33 Torr

683

pH: 12 days at 3,810 m Modified rebreathing; PETCO2at

hyperoxic value with VI �140ml �min�1 �kg�1

HVR doubled after 12 days; at altitudehyperoxic PETCO2

with VI �140ml �min�1 �kg�1 decreased by 8-10Torr

682

pH: 1 and 2 days at 5,050 m Modified rebreathing; PETCO2at

about resting value at sea level1 iHVR (5- to 7-fold after 1 and 2

days, respectively)348

iH or pH (PETO255 Torr) during 8 h HVR not tested Time-dependent rise of VI in iH, not pH 338

iH or pH (PETO255 Torr) during 8 h DEF, square-wave and saw-tooth

stimuli; analysis of HVR with first-order model; PETCO2

1.5 Torrabove initial resting in all tests

1 iAHR over 8 h; no differencebetween protocols, suggesting noinfluence of the PCO2; both iH andpH increased baseline hyperoxicventilation and peripheral ventilatoryCO2 sensitivity

337

iH or pH (PETO255 Torr) during 8 h Testing the effects of isocapnic step

hyperoxiaTime-dependent rise in hyperoxic

ventilation in both protocols thatcontinued for at least 8 hpostexposure

762

iH or pH (PETO260 Torr) during 48 h Idem 1 iAHR over this period in both

protocols, gradually reversing overthe subsequent recovery period; noinfluence of PCO2

763

iH or pH (PETO255 Torr) during 8 h No testing of HVR, testing overall

CO2 sensitivityIncrease in CO2 sensitivity after both

protocols; no change in x-interceptof the CO2 response curve

228

Normobaric iH or pH (PETO255 Torr)

during 8 hNo HVR testing, testing effect on

peripheral and central CO2

sensitivity with DEF and modelanalysis

Overall increase in CO2 sensitivity dueto increase in peripheral gain.Contribution, however, of centralchemoreceptors could not beexcluded. No change in x-interceptof CO2 response curve

229

8 h of normobaric iH without drug,with dopamine and withdomperidone, respectively

DEF, square-wave stimuli, modelanalysis; PETCO2

1.5 Torr aboveinitial resting in all tests

1 iAHR after 8 h; dopamine anddomperidone elicited equal effects onthe HVR before and after the iH,indicating no carotid bodydopaminergic involvement in theeffect of iH

590

iH (PETO250 Torr) for 8 h with and

without propranololIdem 1 iAHR after 8 h uninfluenced by �-

adrenergic blockade134

Decreasing CaO2 for 8 h with loweredisocapnic PETO2

of 55 Torr,phlebotomy (500 ml) or 10% COHb

Idem 1 iAHR after low PETO2only,

indicating no role of lower oxygencontent

640

6 h of hypocapnic (PETCO2�10 Torr)

or eucapnic hyperventilationIdem No change in AHR, suggesting that the

increase in AHR during pH and iHare caused by low PO2 itself;hypocapnic, but not eucapnichyperventilation shifts the hyperoxicCO2 response curve to the left

641

pH or poikilocapnic hyperoxia, for 5days, respectively by maintainingPETO2

10 Torr below or aboveresting

Idem 1 iAHR after mild hypoxia and 2after mild hyperoxia

192



nist that penetrates into the brain, increased ventilation,suggesting a predominance of the inhibitory carotid bodydopaminergic system on respiratory output in normoxia(572). However, mice kept in 10% O2 for 10 days re-sponded to droperidol with a decrease in ventilation in-dicating a dominance of the central excitatory dopami-nergic system in hypoxia (572). This seems consistentwith data from mice lacking the dopamine D2 receptor(D

2R�/�) that failed to show VAH and an augmented HVR

after 8 days of acclimatization, an effect that was not yetdemonstrable after 2 days (343). These studies suggestthat a shift from peripheral to central dopaminergic dom-inance may contribute to VAH. How this can be recon-ciled with an increase in carotid body dopamine contentin prolonged hypoxia remains to be seen. Upregulationduring chronic hypoxia of TH and norepinephrine wasalso observed in the in the locus coeruleus of rats (684),but its contribution to VAH is difficult to assess becausethis region is the source of a widespread noradrenergicnetwork in the brain.

B) NMDA RECEPTORS. During hypoxia, glutamatergicNMDA and non-NMDA receptors in the caudal NTS play acrucial role in the transmission of afferent impulses fromthe carotid body (287, 516, 565, 801 and referencestherein). Reeves et al. (634) showed an early upregulationof the NR1 subunit of the NMDA receptor in the dorsalcaudal brain stem of rats which had resolved after 30 daysof chronic hypoxia. The caudal NTS of rats exposed

to 10% O2 for 7–8 days contained higher levels of the �-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)receptor subunit glutamate receptor 2 (GluR2), whichwas thought to contribute to an enhanced glutamatergicsynaptic transmission of arterial chemoreceptors inputsinto the NTS (873). NMDA receptors in the medulla ofmice were upregulated following exposure to 10% O2

during 2 wk (210; see also sect. VE). In rats exposed to 10%O2 for 9 days, the NMDA antagonist MK-801 had a largerinhibitory effect on hypoxic ventilation than in unaccli-matized controls (638).

C) PDGF-� RECEPTORS. Administration of PDGF-BB inthe NTS decreased the early HVR in rats, and mutant miceheterozygous for the PDGF-� receptor showed a reducedHVD (see also sect. IIID). This led to the suggestion thatthat downregulation of PDGF-� receptors during chronichypoxia might contribute to VAH (287, 289). During 14days in 10% O2, rats demonstrated a posttranscriptionaldownregulation of PDGF-� receptor protein in the dorso-caudal brain stem after 7 days, significantly correlatedwith VAH, while the attenuation of the HVR by localPDGF-BB application in the NTS decreased over time andactually was absent from day 7 on (14). In addition toPDGF, other growth factors such as brain-derived neuro-trophic factor (BDNF), located in petrosal ganglion neu-rons and possibly acting postsynaptically, could also con-tribute to plastic changes within the NTS during chronichypoxia.


Protocol HVR Measurement Effects on HVR, CommentsReference

Nos.

iH: PETO250 Torr, 8 h, followed by iv

infusion of the ganglion blockertrimethaphan

Idem 1 iHVR not influenced by autonomicblockade

452

iH 3 h in NHC at PETO2�50 Torr Isoxic hypoxic and hyperoxic CO2

response curve with modifiedrebreathing

Decrease in threshold, not sensitivity ofthe peripheral chemoreflex loop atthe early stages of CH

483

pH during 8 h using a facemask Modified rebreathing; PETCO2at

about resting value1 iHVR by 70% but impaired forearm

hypoxic vasodilation268

pH: 8 wk at 3,800 m Stepwise decreases in SpO2 to 90and 80%; PETCO2

�4 Torr aboveresting in ambient conditions

1 iHVR already after 48 h, sustainedat 8 wk. Generally test PETCO2

about5–6 Torr lower than at sea level

346

pH for 8 h using a facemask Modified rebreathing PETCO2at about

resting value at sea level1 iHVR by �90% 757

5 days at 3,480 m Isoxic hypoxic and hyperoxic CO2


Decreases in the x-intercepts of thehyperoxic and hypoxic CO2 responsecurves of 12.8 and 9.8 Torr,respectively, without changes inslope

719

pH: 9-day ascent to �5,000 m andthen immediate testing at 3,840 m

Isoxic hypoxic and hyperoxic CO2


Difference in slope between hyperoxicand hypoxic CO2 rebreathing becamelarger at high altitude, indicatingincreased hypoxic sensitivity

9

Stay at 1,560 m for 12 days Poikilocapnic response to sustained(20 min) hypoxia (12% O2)

Peak but not sustained poikilocapnicHVR response increased

8

CaO2, arterial oxygen content; iH, isocapnic hypoxia; pH, poikilocapnic hypoxia; iAHR, isocapnic acute hypoxic response; CH, chronichypoxia; iHVR, isocapnic hypoxic ventilatory response; DEF, dynamic end-tidal forcing; NHC, normobaric hypoxic chamber; VI, inspired minuteventilation; PB, barometric pressure; PETCO2

, end-tidal PCO2; PETO2, end-tidal PO2; SpO2, arterial hemoglobin oxygen saturation.



D) NO-RELATED MECHANISMS. As discussed in sectionIIIC3, NO is an important central mediator of the HVR witha stimulatory effect on the HVR. Thus upregulation ofneuronal NOS could also enhance CNS gain. Mice ex-posed to 10% O2 for 2 wk had higher medullary neuronalNOS levels, which seemed related to VAH but not to theincreased HVR: the former was reduced while the latterremained unaffected by a specific neuronal NOS inhibitor(210). Upregulation of neuronal NOS (but not the endo-thelial isoform) was also shown in peripheral and centralneurons in rats after 12–24 h in hypobaric hypoxia (618).The hypoxia-induced rise in intracellular NO in the NTSmay be one of the secondary consequences of NMDAreceptor activation (references in Ref. 287). Thus, in theNTS, a feed-forward mechanism could exist betweenNMDA receptor activation and release of NO to enhancethe CNS gain (562). The above-cited work by El-Hasnaouiet al. (210) does not directly support this hypothesis. Ratsexposed to 12% O2 for 2 wk and receiving a continuousintracerebroventricular infusion of the NOS inhibitorL-NAME showed a lower normoxic ventilation than con-trols after this period, indicating a reduced VAH. How-ever, their HVR was indistinguishable from that in accli-matized controls not receiving the enzyme inhibitor (688).These scarce data raise the impression that upregulatedNO in the medulla may contribute to VAH but not to theincrease in HVR.

E) ERYTHROPOIETIN. Mice overexpressing human eryth-ropoietin (Epo) in the brain without elevation of plasmaEpo and hematocrit (Tg21 mice) were reported to showlarger increases in ventilation and HVR in chronic hyp-oxia, even after carotid sinus nerve section (716). Female(but not male) mice with increased peripheral and centralEpo expression (Tg6 mice) were reported to have anaugmented HVR (717). However, another study in Tg6mice (sex not mentioned) reported no difference in HVRcompared with wild types; after acclimatization, VAH andan augmented HVR were clearly present in wild types buttotally absent in the Tg6 genotype, and this was suggestedto be related to upregulated carotid body dopaminergicmechanisms (804). Finally, infusion into the brain of thesoluble Epo receptor, that normally is downregulated inchronic hypoxia thus increasing Epo bioavailability, re-versed VAH in normal mice (715).

F) CENTRAL OXYGEN-SENSING NEURONS. In the initial phaseof VAH (10% O2 for 4–5 days), O2-sensitive cells in theRVLM displayed an increased O2 sensitivity in vitro that inlater stages (9–10 days) of acclimatization was offset,possibly due to an altered neurotransmitter balance orother processes decreasing membrane excitability (554).In both normoxia and hypoxia, the C1 region and thepre-Botzinger complex show a constitutive expression ofHO-2 (501). Placing rats in 10% O2 for 10 days inducedHO-1 but not HO-2 in the ventral medulla. At this stage,however, the significance of this observation is unclear

since the role of the former isoform in O2 sensing bycentral neurons is unknown (501), whereas in unacclima-tized condition, the isoform HO-2 seems to be required foroxygen sensing in neurons of the pre-Botzinger and C1regions (153). These scarce data could suggest a contri-bution of central oxygen sensors to the increased CNSgain in chronic hypoxia. Isolated central hypoxia, how-ever, does not lead to VAH in the goat (836).

G) EXPRESSION PROFILE OF ION CHANNELS IN THE NTS. Neu-rons in slice preparations from rat containing NTS neu-rons receiving monosynaptic input from the carotid bod-ies respond to hypoxia with an outward potassium cur-rent via activation of KATP channels (i.e., ATP-sensitivepotassium channels that close in the presence of high ATPconcentrations; Ref. 872). In slices from rats exposed to10% O2 for 1 wk, this current was reduced, and this waspossibly caused by an alteration of the expression profileof regulatory subunits of these channels. This then mayhave increased both the excitability of these neurons andCNS gain (872).

H) PLASTICITY-INDEPENDENT MECHANISMS TO INCREASE THE CNS

GAIN. The brain contains high concentrations of O2-sensi-tive (human tandem P domain) TREK-1 potassium chan-nels, predominantly in GABAergic inhibitory interneu-rons, that play a key role in setting the resting membranepotential (references in Ref. 511). In human embryonickidney (HEK-293) cells, these background K� channelsare inhibited by low PO2, in a range relevant for braintissue and also, in an additive way, by intracellular alka-losis (511). Inhibition of these channels in GABAergicinhibitory interneurons by mild alkalosis might then be ofphysiological significance especially at the background ofhypoxia-induced alkalosis (511). In the rat, an additionalway of increasing CNS gain could exist in a negativeperipheral-central interaction (145, 172, 613; see sect. VD).The time-dependent fall in PCO2 in the early acclimatiza-tion phase could augment the central response to carotidbody hypoxia, thus increasing ventilation and the HVR.Isocapnia should then be expected to decrease the mag-nitude of these adaptations, but this has not been system-atically investigated in the rat.

Many of the above modulations and adaptations inboth the carotid bodies and CNS are undoubtedly theconsequence of altered gene expression induced byHIF-1. Rats exposed to chronic hypoxia (0.5 atm, 3 wk)displayed accumulation of HIF-1� in neurons, astrocytes,ependymal cells, as well as endothelial cells and alsoshowed upregulation of genes with hypoxia-responsiveelements (126). HIF-1��/� mice lack an increase in ca-rotid sinus nerve activity in response to hypoxia (not toNaCN) but increase their ventilation, which is initiated bythe aortic bodies. Ventilatory adaptation to a 3-day bout ofhypoxia does not occur in these animals (388). Chronichypoxia induces carotid body expression of both HIF-1�,HIF-2�, and HIF-3�, which all contribute to morphologi-



cal and neurochemical adaptations (434, 584, 614, 784 forfurther references). Humans with an impaired degrada-tion rate of HIF-1� in normoxia (Chuvash polycythemia;see sect. IV) have an exaggerated acute HVR and lowresting PCO2 much similar to VAH (712). The contributionof HIF-1 in both the carotid bodies and the CNS to VAH isrecently reviewed by Powell and Fu (614).

Other transcription factors also are involved. Forexample, a role of activator protein-1 (AP-1) is indicatedby the finding that mice lacking the fosB gene (a memberof the fos family of immediate early genes), althoughhaving a normal HVR when kept in normoxia, do notexhibit VAH and an increase in HVR when exposed tochronic hypoxia (486).

Chronic hypoxia may also recruit additional oxygensensing mechanisms that are silent in a normoxic condi-tion.

G. Chronic Intermittent Hypoxia and the HVR

1. Episodic hypoxia induces phrenic and

ventilatory LTF

In awake rats, dogs, and goats, the termination ofepisodic hypoxia, for example, a challenge with repeatedhypoxia of short duration (mostly 1–5 min) interrupted byshort periods of reoxygenation, is followed by a pro-longed increase in ventilatory motor output, a phenome-non that is known as LTF (112, 506, 568, 792). Sustainedhypoxia of short duration does not evoke the phenome-non (27). At the level of phrenic and hypoglossal motoroutputs, LTF can be evoked by intermittent electricalstimulation of the carotid sinus nerve and/or continuousstimulation of medullary raphe obscurus neurons in anes-thetized, (mostly) vagotomized animals (244, 305, 495,512, 513) or by repetitive hypoxia (23, 245, 387, 504). LTFrequires an intact serotonergic system with neurons inmedullary raphe nuclei playing a crucial role (23, 506,514).

We briefly summarize the main underlying mecha-nisms of phrenic LTF as far as they may also be relevantfor (plastic changes of) the HVR in chronic intermittenthypoxia (references in Refs. 230, 447, 480, 481). Duringepisodic hypoxia, raphe serotonergic neurons in the me-dulla are activated resulting in serotonin release in thevicinity of brain stem respiratory neurons and spinal mo-toneurons. Each hypoxic episode triggers a release of5-HT. One of the downstream cascades then activatesseveral protein kinases that may phosphorylate glutamatereceptor proteins, thus increasing synaptic strength byenhancing NMDA receptor currents. Some phosphates,on the other hand, may lay a constraint on LTF (refer-ences in Ref. 481). A key role in LTF is played by BDNF.During episodic hypoxia, the synthesis of this neurotro-phin in the spinal cord increases; application of BDNF in

the spinal cord of rats is sufficient to evoke LTF. Inaddition, when BDNF translation and new protein synthe-sis is prevented, episodic hypoxia does not induce LTF(28). BDNF is thought to act through the tyrosine kinaseB receptor which then can lead to the facilitation ofglutamatergic transmission via several pathways (28). Ki-nases and phosphatases relevant for LTF are sensitive toROS, explaining the inhibitory effects of antioxidants onLTF. An important source of ROS in various cell types(e.g., motoneurons and carotid body cells) is NADPHoxidase. Depending on the cell type, this enzyme can beactivated in hypoxia, upon reoxygenation as well as viamembrane-bound receptors such as 5-HT2A receptors.Phrenic LTF can be inhibited by serotonin receptor block-ers, antioxidants, and inhibitors of NADPH oxidase, re-spectively (480, 481). Finally, repetitive spinal injectionsof 5-HT induce phrenic LTF, even without hypoxia. So theworking model of episodic hypoxia-induced phrenic LTFis that it is mediated by a release of serotonin, the acti-vation of NADPH oxidase, and increased production ofBDNF and ROS with facilitation of glutamatergic synaptictransmission in the phrenic motor nucleus as the eventualresult (27, 28, 230, 447, 480, 481 and references citedtherein). In both artificially ventilated and spontaneouslybreathing rats, LTF is substantially reduced by NMDAantagonists injected in the phrenic motor nucleus (447).

2. Chronic intermittent hypoxia augments LTF,

sensitizes the carotid bodies in animals, and

increases the HVR

Following three 5-min episodes of isocapnic hypoxia,the phrenic neurogram of rats showed a somewhat in-creased baseline amplitude that further increased overtime to become significantly greater than the prestimulusamplitude after 30 and 60 min, respectively, indicatingLTF (448). Exposing rats to chronic continuous alterna-tions of 5 min of hypoxia (11–12% O2) and 5 min of air for12 h/night for 7 days, i.e., chronic intermittent hypoxia,augmented the acute phrenic response to episodic mildhypoxia and, via a serotonergic pathway, increased LTF(448). This suggests that the chronic intermittent hypoxiahad resulted in an augmented carotid body sensitivity tohypoxia and/or increased CNS gain. Electrical carotidsinus nerve stimulation now resulted in an enhanced ac-tivation of the phrenic nerve, indicating an increased CNSgain (448). In conscious rats, however, not all these find-ings could be confirmed at the level of minute ventilationbecause a similar chronic intermittent hypoxia protocolinduced ventilatory LTF but no increase in the HVR (505,506). Similar to episodic hypoxia-induced LTF, the effectof chronic intermittent hypoxia to augment its magnitudeis a NMDA receptor-related mechanism; however, theserotonin subtype involved may be a different one: a5-HT

2subtype in episodic hypoxia, and a 5-HT6 and/or



5-HT7 subtype in the chronic intermittent hypoxia-in-duced augmentation (447).

These chronic protocols with relatively long dura-tions of mild hypoxia in rats were not directly intended tomimic the rapid and more severe hypoxic swings that areexperienced by sleep apnea patients but were mainlyaimed at studying the plasticity of the neuronal networkshaping the HVR. Peng and Prabhakar (597) have ex-tended the above observations on the effects of episodichypoxia by imposing hypoxic episodes lasting secondsrather than minutes, a stimulus pattern more relevant tosleep apnea. They showed that conditioning rats withchronic intermittent hypoxia consisting of 15 s hypoxiaalternated with 5 min of normoxia, 9 cycles/h for 8 h/dayduring 10 consecutive days, augmented the episodic hyp-oxia-induced phrenic LTF, which was absent in animalsexposed to sustained hypoxia with a comparable cumu-lative duration (597). Combining chronic intermittent hyp-oxia conditioning with daily administration of a superox-ide dismutase (SOD) mimetic prevented this LTF augmen-tation (597). Subsequently, they showed in the rat thatchronic intermittent hypoxia: 1) evoked episodic hypoxia-induced LTF in the carotid sinus nerve (“sensory LTF”)that was prevented with a SOD mimetic; 2) reduced bothaconitase activity in the carotid bodies and the activity ofmitochondrial complex I; and 3) increased hypoxic butnot hypercapnic sensitivity of the in vivo and in vitrocarotid body, which was prevented by scavenging super-oxide radicals and could be reversed by subsequent ad-aptation to normoxia (585, 595–597). This functional ca-rotid body plasticity occurred without macroscopic mor-phological changes (585, 596).

Additional studies in mice with a partial deficiencyfor HIF-1� (HIF-1��/�) revealed that, in contrast to wildtypes, these animals did not respond to chronic intermit-tent hypoxia with increases in HVR and carotid bodysensitivity. Also, the heterozygous animals did not accu-mulate ROS and showed no upregulation of HIF 1-� (600).Prabhakar and colleagues (536) proposed a workingmodel of chronic intermittent hypoxia in which oxidativestress leads to ROS generation (presumably by increasedNADPH oxidase activity and/or inhibition of mitochon-drial complex I), upregulation of HIF-1 and induction ofmany genes by this transcription factor that together helpto sensitize the carotid bodies. Other transcription factorsand posttranscriptional phosphorylation of proteins alsoplay an important role (414, 536).

NO could also be involved in chronic intermittenthypoxia-induced LTF and carotid body sensitization. Micedeficient in neuronal NOS have an augmented HVR butshow an attenuated episodic hypoxia-induced ventilatoryLTF; the effect of chronic intermittent hypoxia in thesemice was not studied (387). Mice deficient in endothelialNOS have a blunted O2 response (390); the effect of

chronic intermittent hypoxia in these animals are un-known.

An increase in the sustained, but not acute, HVR wasreported in rats after alternations of 10 and 21% O2 every90 s, continuously applied during 30 days (634). Challeng-ing cats with alternations of 90 s hypoxia with �5 minnormoxia, 8 h/day for 2–4 days enhanced the in vivo basaldischarge and hypoxic sensitivity of the carotid sinusnerve, upregulated carotid body ET-1 by a factor of �10and shifted the hypoxic response curve of their in vitrovascularly perfused carotid bodies to the right (643). Sub-sequently, a chronic intermittent hypoxia-induced upregu-lation of carotid body ETB but not ETA receptors wasshown, and it was proposed that upregulation of theendothelin system would increase carotid body sensitivityby modulating blood flow (642). Recently, it was sug-gested that, similar to its facilitatory effect on synaptictransmission in the NTS, ET-1 may also stimulate thecarotid body by augmenting the excitatory effect of glu-tamate (456). Rats exposed to chronic intermittent hyp-oxia for 3 wk showed upregulation of carotid body NR1,and NR2, NMDA receptor subunits, and in contrast tocontrols, carotid body baseline activity in these animalswas enhanced by infusion of NMDA without altering thesensitivity to hypoxia. The NMDA antagonist MK-801 notonly reversed this increase in baseline activity but alsoinhibited the stimulatory effect of ET-1 (456). So, althoughprobably not directly involved in O2 sensing, NMDA re-ceptors may play a functional role in the adaptation of thecarotid bodies to chronic intermittent hypoxia.

Chronic intermittent hypoxia will not only initiateplastic changes in the carotid bodies. In NTS neuronsreceiving afferent input from the carotid bodies of ratsexposed to chronic intermittent hypoxia, dose-dependentcurrents induced by AMPA were increased whilst NMDA-activated currents were reduced (174). Reeves et al. (634)showed a chronic hypoxia-induced increase in the expres-sion of the NR1, NR2A, and NR2B NMDA receptor subtypesin the dorsocaudal brain stem. Another factor contribut-ing to a centrally mediated HVR plasticity may be platelet-activator factor (PAF). In an attempt to gain insight intothe role of this phospholipid, Reeves and Gozal (630)compared the effects of a chronic (30 day) intermittenthypoxic paradigm in mice deficient for the PAF receptor(PAFR) with those in their wild-type littermates. Theresults were somewhat contradictory. On the one hand,unlike wild-type mice, PAFR�/� mice failed to show anincrease in normoxic ventilation after the challenge, indi-cating a role of PAF in chronic intermittent hypoxia-induced ventilatory LTF. On the other hand, over theperiod of 30 days, both genotypes showed an equallyrobust time-dependent increase in the response to sus-tained (20 min) hypoxia (630). Thus the precise role ofPAF in the chronic intermittent hypoxia-induced augmen-tation of the HVR remains to be determined.



After exposure to chronic intermittent hypoxia,anesthetized rats showed a rise in blood pressure andan augmented preganglionic sympathetic nerve re-sponse to acute hypoxia (294). Conscious rats re-sponded with an increase in the hypoxic response ofthe renal sympathetic nerve (339). Chronic intermittenthypoxia has a profound influence on the oxidative sta-tus of the brain cortex and leads to protein and nucleicacid oxidation, lipid peroxidation, and apoptosis, a pic-ture that is mitigated in mice overexpressing Cu,Zn-SOD (857). ROS induced specific enzymatic changes inthe brain stem of rats exposed to chronic intermittenthypoxia that eventually led to enhanced �-amidation ofneuropeptides such as substance P and neuropeptide Ythat thereby become biologically active (699). Manyneuropeptides are operating in central neuronal net-works, so similar mechanisms may also contribute tochronic intermittent hypoxia-induced changes in theHVR and sympathetic chemoreflex sensitivity.

In summary, in animals and across species, chronicintermittent hypoxia leads to an augmented episodic hyp-oxia-induced LTF at the level of carotid bodies, phrenicnerve, and ventilation. Carotid body sensitivity is in-creased and the HVR is enhanced. These adaptivechanges are mediated by ROS-related HIF-1� stabilizationand gene expression resulting in adaptations in the sig-naling-transduction cascade in the carotid bodies andplastic changes within the central respiratory neural net-work. As was also the case for chronic hypoxia, chronicintermittent hypoxia has different effects in neonates andadults. This is schematically shown in Figure 10.

3. Chronic intermittent hypoxia increases the HVR

in humans

Protocols used in humans to produce chronic inter-mittent hypoxia vary widely in total duration (days toweeks), severity, and length of individual hypoxic expo-sures. In a minority of cases, isocapnia was maintained.Data we have collected from a number of recent studiesare shown in Table 5. The paradigms employed do notexactly mimic the swings in PO2 and PCO2 that are encoun-tered in sleep apnea syndrome wherein hypoxic episodesare shorter, 20 s on average, and accompanied by hyper-capnia (242). Rather, they may represent time courses ofprolonged hypoxic episodes that can occur in patientswith chronic obstructive pulmonary disease (COPD)and/or severe heart failure. With only a few exceptions,paradigms of chronic intermittent hypoxia in humans re-sult in an increase in the HVR, which can be accompaniedby a rise in resting normoxic ventilation and drop inend-tidal CO2. In cases where isocapnic HVRs were mea-sured at resting end-tidal CO2 levels (Table 5), the effectsmay even have been underestimated (postexposure end-tidal CO2 levels mostly being lower).

If it is accepted that the peripheral chemoreceptorsmediate the initial hypoxia-induced rise in ventilation,enhancement of the acute HVR (8, 10, 11, 253, 484) is mostprobably due to an increase in carotid body sensitivity. Ahigher ventilation during hypoxia combined with an alsohigher arterial oxygen saturation is compatible with anincreased slope of the ventilation-SaO2

relationship (e.g.,Ref. 53). It is not easy, however, to assess the precisequantitative contribution of the carotid bodies to the aug-mented HVR: isocapnic progressive hypoxic tests lasting5–9 min are associated with increases in CBF and centralhypocapnia and may also be accompanied by a rapidlydeveloping HVD (see sect. III). The fact that chronic inter-mittent hypoxia induces augmented cerebrovascular O2

and CO2 sensitivities should also be taken into account (8,395).

The mechanisms of a chronic intermittent hypoxia-induced increase in carotid body sensitivity in humansremain elusive, but plastic changes as described for ani-mals could well operate in humans also. Like in animals,the increase in HVR may depend on stimulus intensity, thenumber of deoxygenation/reoxygenation periods, and to-tal exposure length. Exposing humans to a chronic inter-

VI

.

VI

.

FIG. 10. Comparison of chronic intermittent hypoxia in neonatesand adults. Effects of chronic intermittent hypoxia in neonates (withinthe critical time window for developmental respiratory plasticity) and inadults. Question marks concern unresolved or controversial issues. VI,inspired ventilation; LTF, long-term facilitation. In the adult, ventilatoryLTF can be the result of sensory LTF and LTF in major respiratory motornuclei such as the phrenic and/or hypoglossal motor nuclei and inter-costal motor neurons. In the adult, sensory LTF in the carotid bodiesmay tend to augment carotid body sensitivity and thus destabilizebreathing, but LTF in motor nuclei may have a stabilizing effect. Forexample, if the phenomenon would exist in the hypoglossal (andphrenic) nucleus in humans with obstructive apnea, the chronic inter-mittent hypoxia in these patients may tend to limit the occurrence ofupper airway collapse. For further explanation, see text.



TABLE 5. Chronic intermittent hypoxia in humans

Protocol HVR Measurement Effects on HVR, Comments Reference Nos.

5 min isocapnic hypoxia (12%O2) alternated by 5 minnormoxia for 1 h, 12 days(protocol 1) versus 30 minisocapnic 12% O for 12 days(protocol 2); interruption ofprotocols on days 6 and 7

Progressive isocapnic hypoxia toSaO2

�75 %; PETCO2at resting

values

1 iHVR by 37% on day 12,recovery on day 17; nodifference between bothprotocols; no changes inresting VI and PETCO2

241

FIO20.13 for 2 h/day for 12 days Step decrease in FIO2

from 0.3 to0.13; PETCO2

at level causing ahyperoxic VI of 15ml/min�1 � kg�1

Max 1 of iAHR, on day 5 andthen gradually decreasing tocontrol on day12; isocapnicPETCO2

after hypoxia �1.5Torr lower

253

Daily exposure to a barometricpressure of 432 Torr for 1 h,7 days

Progressive isocapnic hypoxia ateucapnic PETCO2

HVR, defined as the constant ofthe hyperbolic VI-PaO2relationship, increases by 73%; no change in resting PETCO2

376

Daily three 6-min bouts ofisocapnic hypoxia with PETO250-35 Torr interspersed withnormoxia within 30 min, for14 days

Progressive isocapnic rebreathingto FIO2

8-7% at constanteucapnic PETCO2

1 HVR by �50%, no change inresting PETCO2

695 (in which many studies in theformer Sovjet Union arereviewed).

Decompression to 15.5 or 12.3% O2 for 1 h/day for 1 wk


�75 %; PETCO2at resting

values

1 iHVR only after 12.3 % O2 by�25%, indicating an influenceof the depth of hypoxia; nochanges in resting PETCO2

andresting VI

374 (More data from the sameauthors in ref, 373, 375–377)

12% O2 for 5 min, poikilocapnic,followed by 5 min normoxia,for 1 h, 10 days

Isocapnic progressive hypoxia toSaO2

�75%; PETCO2at resting

values.

No change in iHVR, restingPETCO2

somewhat lower afterCIH

623

FIO20.12, isocapnic for 5 min,

followed by 5 min 0.21, 4times/day for 10 days, men60–74 yr old


�85%; PETCO2at resting

values

No change in iHVR, neither infit, nor in sedentary men; nochange in resting PETCO2

700

12 times 5 min poikilocapnic12% O2, 5 min normoxia dailyfor 7 days SDIH versus 60min poikilocapnic 12% O2

daily, 7 LDIH in the samesubjects

Isocapnic progressive hypoxiauntil SaO2

�80 %; measured on7 consecutive days; PETCO2kept at resting values in allconditions

1 iHVR by �50% after bothSDIH and LDIH, max effectreached after 3 days; CO2

threshold (from modifiedrebreathing) decreased afterboth protocols

393

5 min 12% O2, 5 min 21% O2 for90 min during 10–12consecutive days

Poikilocapnic response to acuteand sustained (20 min) hypoxia(12% O2)

Acute, but not sustained pHVRresponse increased; nochange in resting PETCO2

8

60 min of isocapnic hypoxia(SaO2

�80%) for 10consecutive days


�75% and thenmaintained for another 20 min;PETCO2

at resting values.

Sustained (20 min) iHVRdoubles, no change in restingVI and resting PETCO2

478

Sleeping for 5 consecutivenights with FIO2

�0.14iAHR determined from six

descending PETO2steps with

DEF technique; PETCO2kept at

�1.5 Torr above resting

Mean iAHR more than doubles,baseline ventilation increased,resting PETCO2

fell by �3.5Torr

10, 11

14 nights (9 h) of sustainedSaO2

�84%Isocapnic progressive hypoxia to

SaO2�80%

iHVR slope increases �2.5-fold,increase in baselineventilation; resting [paco2]lower by 3.3 Torr

269

20 min of isocapnic (eucapnic)hypoxia (FIO2

� 0.10) dailyfor 14 consecutive days

Isocapnic steps in hypoxia daily;modified rebreathing todetermine thresholds andsensitivities both before andafter isocapnic hypoxic studies

AHR but not sustained HVRincreased; augmented AHRascribed to decrease inperipheral CO2 thresholdbecause the threshold of thehypoxic, not the hyperoxicCO2 rebreathing curvedecreased. Sustained HVR notincreased because of HVD,ascribed to an increase inperipheral CO2 threshold

484



mittent hypoxia protocol with hypoxic episodes of 12% O2

led to an increase in oxidative plasma biomarkers and adecrease in antioxidant capacity that were not observedwith 15% O2 (817). Katayama et al. (374) reported anincrease in HVR by a protocol incorporating 12.3% O2 butnot 15.5%. Patients with obstructive sleep apnea show anincrease in plasma biomarkers of oxidative stress andROS production in leukocytes (202, 365). A recent studyin healthy young subjects showed that after 4 days ofintermittent hypoxia, the acute ventilatory response tohypoxia increased, which was correlated with an aug-mented production of ROS (605).

Long-term chronic intermittent hypoxia in humanshas pathophysiological consequences that are recentlyreviewed elsewhere (242). Briefly, it leads to an in-crease in (muscle nerve) sympathetic activity, impairedbaroreflex, increased production of ET-1 by enhancedshear stress, decreased NO bioavailability, impairedvasodilation, and hypertension. Increased productionof ROS causes upregulation of genes encoding factorsthat elicit inflammation, platelet aggregation, and ath-erosclerosis (242). Chronic intermittent hypoxia as ex-perienced by patients with a sleep apnea syndrome willtend to increase their HVR. An increase in chemoreflex

sensitivity elicits higher sympathetic activity and even-tually results in hypertension in many of these patients(372, 537, 548).

5. Different effects of chronic hypoxia and chronic

intermittent hypoxia

The effects of chronic hypoxia and chronic inter-mittent hypoxia may be difficult to compare. For exam-ple, differences in the duration and severity of individ-ual hypoxic episodes, total exposure time to hypoxia,and the large variation in cycle number of intermittenthypoxic paradigms render it hard to define one singleeffect chronic intermittent hypoxia. Some protocolsmay comprise both chronic hypoxia and chronic inter-mittent hypoxia or repetitive cycles with differentlengths. Protocols with long uninterrupted exposures,e.g., during long nights spent in hypoxia (e.g., Refs. 11,269) lead to increased carotid body sensitivity andperiodic breathing and an ensuing additional, chronicintermittent (hypercapnic) hypoxia superimposed onthe experimental one. This may be a confounding factorin studying the effects of “chronic hypoxia” during longstays at high altitude.


Protocol HVR Measurement Effects on HVR, Comments Reference Nos.

20 consecutive nights with FIO20.163 (Pl) versus 4 blocks of5 nights with each blockinterspersed with 2 nights ofnormoxia (P2). “Living high,training low” in athletes


�75%Time-dependent increase in

HVR in both P1 and P2; effectof P1� P2. HVR expressedboth as linear VI-SaO2

andhyperbolic VI-PaO2relationships. Resting PETCO2lower after both P1 and P2

787

4 consecutive days with cyclingbetween 2 min PETO2

�45Torr and 2 min at 88 Torrduring 6 h in awake humans

Acute isocapnic HVR with DEF;hypoxic sensitivity defined asthe ratio delta ventilation overdelta SaO2

Time-dependent increase in theacute hypoxic response thatafter 4 days was �80%. Theincrease was correlated withan increased production ofreactive oxygen species.

605

Subjects slept during 8-9 h/dayin 13% O2 during 2 and 4 wk,respectively, and wereexposed overnight to 30cycles/h of reoxygenation

Progressive isocapnic hypoxictest; HVR defined as the ratiodelta ventilation over deltaSaO2

Doubling of the HVR after 2 wk;in fact, this protocol consistedof two CIH paradigmssuperimposed on each other.Sleeping in a hypoxicenvironment alone alsoincreases the HVR (e.g., seeRefs. 10 and 269) and acontrol group not subjectedto the reoxygenation cycleswas not included. Aninteresting observation wasthat the hypopneas associatedwith reoxygenation werefollowed by arousals.

756

iH, isocapnic hypoxia; pH, poikilocapnic hypoxia; iAHR, isocapnic acute hypoxic response (3-min hypoxic ventilatory response); iHVR,isocapnic hypoxic ventilatory response; pHVR, poikilocapnic hypoxic ventilatory response; HVD, hypoxic ventilatory decline; DEF, dynamicend-tidal forcing; VI, inspired minute ventilation; FIO2

, oxygen fraction of inspired air; PETCO2, end-tidal PCO2; PETO2

, end-tidal PO2; SaO2, arterial

hemoglobin oxygen saturation; CH, chronic hypoxia; CIH, chronic intermittent hypoxia; SDID, short-duration intermittent hypoxia; LDIH, long-duration intermittent hypoxia.



The effect of chronic intermittent hypoxia clearlydepends on the pattern of the hypoxic exposures. Forexample, rats exposed to chronic intermittent hypoxiawith long uninterrupted hypoxic episodes did not man-ifest an augmented carotid body sensitivity, in contrastto those challenged with hypoxia of equal total durationbut with many oxygenation-reoxygenation cycles; theformer animals may have experienced a milder oxida-tive stress compared with the latter (598). The produc-tion of ROS will widely vary not only with the lengthand severity of the hypoxic episodes but also on thenumber of reoxygenation periods. A single bout ofhypoxia and reoxygenation produce ROS, and this ef-fect is amplified by repetitive hypoxia/reoxygenation(857, 867).

Some of the reported different consequences ofchronic hypoxia and chronic intermittent hypoxia canbe listed as follows. 1) In contrast to chronic hypoxia,a specific chronic intermittent paradigm did not lead tomorphological changes in the carotid bodies (585, 596).2) In PC12 cells, the time course and magnitude ofHIF-1� expression differed between the two paradigms:chronic intermittent hypoxia was more potent in in-creasing HIF-1� protein, and the elevated levels weresustained for longer periods of time (references in Ref.536). 3) Another study reported that in contrast tochronic hypoxia, chronic intermittent hypoxia did notenhance HIF-1� mRNA in the rat carotid body, whileboth paradigms augmented the expression of HIF-2�and HIF-3� (434). 4) In the dorsocaudal brain stem ofthe rat, chronic intermittent hypoxia was associatedwith an increase in the expression of the NR2A andNR2B subunits of the NMDA receptor that did not occurin chronic hypoxia (634). The authors of this studyrelated this to the less pronounced presence of HVDafter the chronic intermittent compared with thechronic protocol (634). 5) As discussed in section VG1,in the rat, chronic intermittent but not chronic hypoxiaaugmented the magnitude of LTF (597).

Using a systems biology approach, Wu et al. (850)studied the expression of genes, sets of genes, and geneticnetworks and found different temporal expression pat-terns in chronic- and chronic intermittent hypoxia in thelungs of male rats. The gene encoding the estrogenreceptor 1 (Esr-1) was unmasked as an important reg-ulator of the response to chronic intermittent but not tochronic hypoxia. Increased expression of Esr-1 alsoresulted in upregulation of �200 target genes of theESR-1 protein, for example, the P2X2 purinergic recep-tor. In addition, ESR-1 closely interacted with genenetworks that already were known to play key roles inhypoxia, for example, a hub node around HIF-1� andvascular endothelial growth factor (VEGF). Chronichypoxia acutely induced a different set of genes thatwere overrepresented by those involved in the immune

response. The network activated by chronic hypoxiaincluded the hub nodes with vascular endothelialgrowth factor, nuclear factor-�B and transforminggrowth factor-�, all of which are known to play a role inpulmonary hypertension (850). This interesting ap-proach awaits application in the carotid bodies.

VI. MEASURING, EXPRESSING, AND MODELING

THE HYPOXIC VENTILATORY RESPONSE

A. Measuring the HVR

Techniques used to measure the human HVR can bedivided into five groups: 1) steady-state techniques involv-ing fixed inspired O2 and CO2 concentrations, 2) single ordouble breath tests involving transient hypoxia or hyper-oxia for no more than a few breaths, 3) progressivehypoxia tests involving the rebreathing of a gas mixturefrom a rebreathing bag or circuit with CO2 absorptionallowing control of end-tidal CO2, 4) step hypoxic tests inwhich rapid decreases in end-tidal O2 are performed whilemaintaining end-tidal CO2 constant, and 5) via measure-ment of isoxic hypoxic and hyperoxic ventilatory CO2

responses.The HVR obtained with steady-state hypoxic tests

(test 1) is a mixture of the acute effects of hypoxia andits central depressant effects. Hence, the response is“contaminated” by HVD. These tests were applied in atime and age in which there was little knowledge onHVD. Consequently, steady-state tests are consideredoutdated (871). Dejours et al. (177) developed the sin-gle breath test (test 2) in which after several controlbreaths the subject inhales one vital capacity of a testgas (100% O2 or 100 N2 with or without the addition ofCO2). The change in ventilation during the next fewbreaths is a measure of hypoxic sensitivity. While thistest is not contaminated by HVD, it does not allow thedevelopment of the full HVR, and isocapnia is not main-tained during the test. Furthermore, to avoid randomeffects from the background variation in ventilation,the test has to be repeated frequently (411). Finally,quantification of the small and short-lived change inventilation is difficult (628), and therefore, the testshould be just considered a qualitative measure of pe-ripheral chemoreceptor activity. At present, most pop-ular tests are those that use progressive (test 3) andstep hypoxia (test 4) and those that measure the HVRvia the VI-CO2 responses at hyperoxia and hypoxia(test 5).

1. Progressive hypoxic tests

Over the years, various progressive hypoxic testshave been designed (143, 272, 411, 458, 485, 626–628,



833). Techniques differ in depth of hypoxia, steepness ofthe hypoxic ramp (and linked to this the duration of thetest), and the technique to keep end-tidal CO2 constant.Weil et al. (833) published their method in 1970: subjectsbreathed sequentially three gas mixtures for 3–5 min(room air, 10 and 8.5% O2 in N2) causing a slow (15 min)progressive isocapnic decline in alveolar PO2 to �40 Torr.Isocapnia was maintained by adding CO2 to the inspira-tory limb of the breathing circuit. Godfrey et al. (272)adapted the CO2 rebreathing technique developed byRead (624) in such a way that hypoxia developed progres-sively over 5–8 min. The subjects rebreathed from a cir-cuit that included a CO2 absorber allowing the mainte-nance of isocapnia. The initial gas mixture contains 25–30% O2

in N2, and as the subjects consumes oxygen, the oxygenconcentration of the bag decreases. During the course ofrebreathing, arterial PO2 drops from 130 to �30 Torr.Rebuck et al. (627) and later Rebuck and Campbell (626)described a rebreathing technique similar to that of God-frey et al. (272). The “Rebuck” approach involves re-breathing from a 6-liter bag containing a gas mixture of 7%CO2 and 22–24% O2 in N2. The obvious problem with thistechnique is that the ensuing HVR is due not only to theprogressive decline in PO2 but also to the step increase inPCO2 (658). In a modified approach, the high PCO2 level iskept constant for 6 min before beginning to rebreathefrom the bag (485). As pointed out by Robbins and Zhang(658), the later description of the modified “Rebuck re-breathing technique” by Rebuck and Slutsky (628) doesnot make it clear that the original technique is seriouslyflawed. See Figure 11A for a description of the apparatusused by Rebuck and colleagues. It is important to beaware that rebreathing methods have the potential forsystematic errors due to contamination by HVD whentests last longer than 5 min (411, 612).

2. Step hypoxic tests

A sophisticated method to obtain ventilatory re-sponses to steps into and out of hypoxia is by using thecomputer-controlled DEF technique, developed by Swan-son and Bellville in the late 1960s and early 1970s inStanford, California (56, 747, 748). Originally DEF wasused to measure the VI-CO2 response and quantify thecontribution of the central and peripheral chemoreflexloops to total ventilation (56). Not much later, it wasadopted by several research teams to study step andsinusoidal hypoxic responses (176, 655, 822). DEF (seeFig. 11B) allows manipulation of the inspired PO2 and PCO2

to induce specific waveforms in end-tidal gas concentra-tions independent of the ventilatory response and venousreturn. Feedback/forward loop algorithms are applied toanticipate and compensate for the respiratory response toa desired end-tidal gas concentration on a breath-to-breathbasis. The control of the end-tidal gas concentrations with

DEF is good, with standard deviations of the end-tidalconcentrations over time �0.6 Torr (161). The main ad-vantage of DEF is its ability to obtain an open-loop esti-mation of ventilatory responses within a short period oftime, something that is impossible under closed-loop con-ditions.

The technique has the disadvantage that relativelycomplex apparatus is required. As a result, less sophisti-cated techniques are utilized more frequently using con-tainers with preset hypoxic mixtures and/or adaptationsof the inspired oxygen concentrations by hand to obtainthe desired end-tidal values (204, 370, 683, 705). Oftenisocapnia is maintained by using a rebreathing circuit(683), otherwise by adding CO2 to the inspiratory limb ofthe apparatus (204).

One variant of the step hypoxic test is the applicationof randomly modulated sequences in hypoxia and hyper-capnia (863). This computer-driven technique requiresmathematical modeling to obtain an indication of themagnitude of the HVR and its interaction with CO2. Whileof great value in the characterization of the ventilatorycontrol system (see also sect. VIC), it is of lesser practicalapplicability.

3. Measuring HVR via the VI-CO2 response

Various authors derive the HVR from isoxic hypoxicand hyperoxic VI-CO2 response curves (195, 349, 411, 547,628, 720). Most frequent studied PO2 values for hyperoxiaare 150–250 Torr and for hypoxia are 40–50 Torr. Sever-inghaus defines the hypoxic response as the increase inventilation at an interpolated end-tidal CO2 going fromhyperoxia to hypoxia (�V40 when a 40 Torr hypoxic levelis chosen, Ref. 683). Others use the change in slope goingfrom hyperoxia to hypoxia (�S) or the ratio of the slopes(S40/S200) as well as the change in the position of theresponse curves (195, 628). When the isoxic VI-CO2 re-sponse curve is assessed using steady-state methods at ahypoxic background, an appreciable effect of HVD on theresponse occurs. To overcome this problem, some au-thors use the (modified) CO2 rebreathing procedureaccording to Read (624). Note, however, that the Read-rebreathing technique, although relatively simple in per-formance, contains inherent difficulties that are not al-ways appreciated (155). For example, abolishing themixed venous-to-arterial PCO2 gradient, one of the keyfeatures of the Read-rebreathing technique (624), is notenough to reduce the brain tissue-arterial PCO2 gradientsufficiently to zero to cause equilibration between arterialand brain tissue PCO2 (155).

4. The choice of method of measuring HVR

A) THE CHOICE OF TEST END-TIDAL PCO2. There is noconsensus on defining the test end-tidal or arterial PCO2

(696). Some investigators raise the end-tidal value ar-



FIG. 11. Experimental set-up to measure the HVR. A: setup used by Rebuck and Slutsky to measure the ventilatory response to hypoxia. TheDC pump draws gas from the bag through the soda lime and returns the gas to the end of the bag (arrows). [Adapted from Rebuck and Slutsky (628).]B: experimental setup used in Leiden to apply dynamic end-tidal forcing of oxygen and carbon dioxide. There are two computer subunits, one toapply control signals to the mass flow controllers, which sets the desired inspiration oxygen and carbon dioxide fractions, and one to collect theend-tidal data.



bitrarily (e.g., to 45 Torr), and others use a fixed in-crease relative to baseline end-tidal CO2 (e.g., �5 Torrabove resting) or a value set to achieve equal back-ground central drive throughout the hypoxic experi-ment (for example, by using a end-tidal CO2 value atwhich ventilation in hyperoxia equals 4 l/m2 per min or140 ml � min�1 �kg�1), presumably to prevent a possibleconfounding influence of central-peripheral interaction(195, 696, 720). The method of choice of the end-tidalCO2 is of importance when studying the effect of spe-cific interventions, such as high altitude, acid-base al-terations, or pharmacological agents, on the HVR. Aninsightful and acceptable alternative to the measure-ment of HVR at a fixed end-tidal PCO2 would be theassessment of HVR at two to three PCO2 values (Fig. 12).Alternatives could be �S or S40/S200.

B) THE CHOICE OF BASELINE END-TIDAL PO2. When study-ing HVR, the baseline oxygen level chosen is sometimeshyperoxic (end-tidal PO2 150 –250 Torr; Refs. 195, 696).This is done to “silence the peripheral chemoreflexresponse to CO2” (195). Whether this is achieved atmild hyperoxia is doubtful. Data from Pedersen et al.(591) indicate that at 200 Torr, the fast peripheral com-ponent of the VI-CO2 response remains present in hu-mans (mean peripheral component was 27% of totalCO2 sensitivity). Even at much “deeper” levels of hy-peroxia (end-tidal PO2 �500 Torr), the human carotidbodies remain active with an average 13% contributionof the peripheral component to total CO2 sensitivity(156). Furthermore, high oxygen levels affect centralrespiratory drive via the Haldane effect and a reductionof brain blood flow (this effect is independent of anyCO2 inhalation), causing a rise in brain tissue PCO2

(156). High oxygen may also cause the formation of

excitatory substances in the brain (e.g., NO, glutamate,ROS). Although these components seem of lesser im-portance during mild hyperoxia (140 –250 Torr), wecannot exclude some excitatory effect on breathingfrom even brief increases in O2 (see Fig. 13; Ref. 613).With all of the above taken into account, our firstchoice would be to assess the HVR using a normoxicbaseline level. The choice of a hyperoxic baseline levelhas little or no advantage, and in addition, from aphysiological view point, hyperoxia does not representa realistic ambient state.

C) A PROPOSAL FOR MEASUREMENT OF HVR. Over the years,Severinghaus and others made several attempts toreach consensus on the HVR methods (131, 195, 696,697). Consensus is required to overcome “lack of com-parability between results obtained by various groupsdue to their variety in methods, definitions, choice ofisocapnic PCO2 and duration of hypoxia” (696). In theso-called “proposed Lake Louise HVR Methods,” Sever-inghaus (696) proposes three tests (Table 6): 1) anisocapnic HVR (iHVR) to determine the hypoxic sensi-tivity of the carotid bodies lasting no longer than 5 minand 2) a poikilocapnic HVR (pHVR) lasting 20 min topredict the response to high altitude. In both tests,baseline (prehypoxia) conditions are hyperoxic (150 �end-tidal PO2 � 250 Torr). 3) The iHVR may be pro-longed to get an indication of the isocapnic HVD. Duffin(195) proposes three tests (Table 6): 1) the assessmentof peripheral chemoreflex responsiveness by measure-ment of the ventilatory response to CO2 at hyperoxia(150 Torr) and hypoxia (40 Torr) preferentially per-formed by using the modified Read-rebreathing test (oralternatively performing 5-min isocapnic steps fromhyperoxia to hypoxia); 2) to get an indication of HVD

FIG. 12. Protocol to measure the ventilatory response to hypoxia. A: measuring the isocapnic ventilatory response to hypoxia (iHVR) atthree arterial PCO2 levels using the “Leiden” approach (test 1 in Table 1) in an awake healthy 30-year-old male volunteer. At each CO2 level,the arterial PO2 is lowered from 100 mmHg (closed data points) to 45 mmHg and kept constant for 5 min (open data points). The ventilatoryincrease is depicted by the arrows. The lines through the data are obtained from linear regression and are the post hoc steady-state normoxicand acute hypoxic VI-CO2 response curves. B: the hypoxic response, expressed as �VI/�SpO2, is plotted for each arterial CO2 concentration(open squares and broken line). The O2/CO2 interaction (i.e., the carotid body-induced increase in hypoxic sensitivity at increasing levels ofarterial CO2) is abolished when the experiment is repeated during the inhalation of 0.15 end-tidal percent halothane.



the first procedure is repeated before and after 20-minof hypoxia (or alternatively performing a 20-min iso-capnic hypoxic response); and 3) a 20-min poikilocap-nic step hypoxic response is performed to mimic theresponse to environmental hypoxia.

Our proposal is a modification of the methods out-lined by Severinghaus and Duffin (Table 6: the “Leiden”proposal). Similar to the other proposals, we suggestperforming three distinct tests. In test 1, 5-min step re-sponses into hypoxia (arterial PO2 45–50 Torr) are per-formed at three different arterial PCO2 levels (e.g., �2, �4,and �8 Torr above resting). However, in contrast to theprevious proposals, baseline oxygen levels are normoxic(arterial PO2 � 100 Torr). One is now able to calculate theacute hypoxic response (i.e., hypoxic sensitivity), andsince the responses are obtained at various PCO2 levels,one gets a full characterization of O2-CO2 interaction, aphenomenon most likely originating at the peripheral che-moreceptors (see also sect. VD; Fig. 12). Tests 2 and 3 aresimilar to the ones proposed by Severinghaus but with

normoxic rather than hyperoxic baseline states to preventany excitatory effect of even mild hyperoxia. A transitionfrom hyperoxia to hypoxia will change central ventilatorydrive more than the transition from normoxia to hypoxia(see sect. VIA4B). This will affect the position of the VI-CO2

response causing a leftward shift (Fig. 13). Of equal im-portance is the fact that hyperoxia does not silence theperipheral chemoreceptors. Furthermore, we refrain fromobtaining VI-CO2 responses curves at different PO2 levelsas proposed by Duffin (195) as ion channel and neuro-transmitter responses to low oxygen and high PCO2/lowpH stimulation at the carotid bodies may differ and, as isalso important, arterial PO2 is a separate stimulus at theperipheral chemoreceptors (277). Furthermore, by per-forming step hypoxic tests, one may decide to use peakhypoxic ventilation rather than 5 min of hypoxia as hy-poxic measurement point. This is especially importantwhen the temporal profile of the response indicates thatHVD is apparent at times �5 min of hypoxia (see also Fig.13 for an example).

FIG. 13. Measuring the hypoxic ventilatory response at three constant levels of end-tidal PCO2. A: isocapnic ventilatory responses to hypoxiaand hyperoxia at three arterial PCO2 levels obtained in a healthy 28-year-old female volunteer. Normoxic, hyperoxic (arterial PO2 � 140 mmHg), andhypoxic (arterial PO2 � 45 mmHg) data points are given. The data points form post hoc steady-state VI-CO2 response curves in normoxia, hyperoxia,and acute hypoxia (lines through the data points derived from linear regression). The left shift of the hyperoxic VI-CO2 response curve relative tothe normoxic curve is due to the central stimulatory effects from even mild hyperoxia. Consequently, using the hyperoxic data as a measure of theisocapnic HVR (see Table 1) makes data interpretation difficult if not impossible. B: the data in these graphs are obtained by using a small adaptationto the test 1 of the “Leiden proposal” of Table 6. The top graph shows the imposed end-tidal fractions of O2 (continuous line) and CO2 (dotted line).In between the normoxic and hypoxic exposures, a 7-min isocapnic hyperoxic exposure (PO2 � 140 mmHg, measured in arterial blood) was added.The middle graph shows the measured arterial O2-Hb saturation using pulse-oximetry. The bottom graph shows the measured inspired ventilation.Each data point is one breath.



TABLE 6. Proposed methods to measure the hypoxic ventilatory response

Proposed Methods to Measure Hypoxic Ventilatory Response

“Lake Louise HVR Method proposal” (696)

Test 1: isocapnic HVR (iHVR) to assess the hypoxicsensitivity of the carotid bodies

Baseline oxygen level 150 � end-tidal PO2 �250 Torr for 10 minHypoxic test Reduce the inspired O2 level such that the SpO2 is rapidly lowered and maintained

at 80% (range 75–85%) for 5 minPCO2 Keep the end-tidal PCO2 constant at a level causing hyperoxic ventilation to rise to

a standard, e.g., 140 ml �min�1�kg�1

Expression of response iHVR � �VI/�SpO2 where �VI � the increase in VI from baseline to the 5th min ofhypoxia

Test 2: measuring the isocapnic hypoxic ventilatorydecline (iHVD)

Hypoxic test Expand test 1 to 20 min of isocapnic hypoxiaExpression of iHVD % fall in �VI from the 5th to the 20th min

Test 3: poikilocapnic HVR (pHVR) to predict the HVR ataltitude

Baseline oxygen level 150 � end-tidal PO2 �250 Torr for 10 minHypoxic test Inspiration of a O2/N2 gas mixture (e.g., 12 % O2) toReduce SpO2 to 80% for 20 min pHVR � �VI/�SpO2

Expression of response pHPR � �PETCO2/�SpO2

The “Ontario proposal” (195)

Test 1: measuring peripheral chemoreflex sensitivityHypoxic test* Ventilatory response to CO2 at an isoxic high (150 Torr) and an isoxic low (40

Torr) oxygen tension using the modified Read-rebreathing testExpression of response Differences in hyperoxic and hypoxic VI-CO2 sensitivities and position of the curves

Test 2: measuring the hypoxic ventilatory decline (HVD)Hypoxic test�� Ventilatory response to CO2 at an isoxic high (150 Torr) and an isoxic low (40

Torr) oxygen tension using the modified Read-rebreathing test (lasting no longerthan 5 min) performed before and after 20 min of hypoxia

Expression of response As in test 1Test 3: measuring the poikilocapnic HVR (pHVR)

Hypoxic test Step reduction in end-tidal PO2 from hyperoxia (150 Torr) to hypoxia lasting 20 minExpression of response �VI from baseline to 20th min of hypoxia at a specified O2 level

�CO2 from baseline to 20th min of hypoxia

The “Leiden proposal” (2010)

Test 1: isocapnic HVR (iHVR) to assess the hypoxicsensitivity and O2/CO2 interaction of the carotidbodies

Baseline oxygen level Arterial PO2 � 100 Torr for 10 minHypoxic test Step reduction in arterial PO2 to 45–50 Torr, which is kept constant for 5 minPCO2 The test is performed at three arterial PCO2 levels, e.g., �2, �5 and �12 Torr

above resting. In between tests there is a 15- to 30-min rest periodExpression of response HVR � �VI/�SpO2; an alternative could be HVR � �VI /�log PO2 (units L/min) if in

humans ventilation would appear to be linearly related to log PO2 below 100Torr (but �10 Torr)

where �VI � the increase in VI from baseline to the 5th min of hypoxia. PO2 isarterial (preferred) or end-tidal PO2

Test 2: isocapnic hypoxic ventilatory decline (iHVD)Baseline oxygen level Arterial PO2 � 100 Torr for 5–10 minHypoxic test Step reduction in arterial PO2 to 45–50 Torr, which is kept constant for 20 minPCO2 The test is performed at one constant arterial PCO2 levels, e.g., �2 or �4 Torr

above restingExpression of response �VI from baseline to 20th min of hypoxia at a specified O2 level

Test 3: poikilocapnic HVR (pHVR)Baseline oxygen level Inspired oxygen concentration �20.8% for 5–10 minHypoxic test Step reduction in inspired O2 concentration (e.g. to 10%), which is kept constant

for 20 minPCO2 CO2 concentrations remains not controlled throughout the testExpression of response �VI from baseline to 20th min of hypoxia at a specified O2 level

�CO2 from baseline to min 20 of hypoxia

*An alternative is the measurement of a three-point steady-state VI-CO2 responses performed at hyperoxia and hypoxia. **An alternative testis an isocapnic 20-min step hypoxic test as suggested by Severinghaus (696). iHVR, isocapnic hypoxic ventilatory response; pHVR, poikilocapnichypoxic ventilatory response; pHPR, response of the end-tidal PCO2 to poikilocapnic hypoxia; SpO2, arterial Hb oxygen saturation as determined bypulse oximetry; PETCO2

, end-tidal PCO2; VI, inspired minute ventilation.



B. Expressing the Steady-State HVR

Mathematical expressions of the HVR have to takeinto account the fact that the steady-state relationshipsbetween ventilation and arterial and end-tidal PO2 arenonlinear. The HVR has been quantified using hyperbolicand exponential expressions (411, 697), and there is nophysiological reason to select one over the other.

The hyperbolic expression (457, 697, 833)

VI � V0 � A/(PO2 C) (1)

where V0 is the horizontal asymptote in VI for infinite highPO2 values, A is the hypoxic sensitivity, and C is thevertical asymptote in PO2 for infinite high VI values. Inmost studies, the asymptote C is fixed to 32 Torr, whichcauses a bias in the estimation of the parameter A (214).Bayesian estimation schemes using soft constraints on C areadvocated to improve the estimation of parameter A (214).The hyperbolic expression indicates that the reciprocal ofventilation versus PO2 is a linear relationship, an observationthat has been established experimentally (272). The expo-nential expression (58, 176, 411, 697)

VI � V0 � G � exp�D � PO2� (2)

where V0 is VI for infinite high PO2 values, G is the hypoxicsensitivity, and D is a shape parameter. The shape param-eter is often fixed to a specific value (e.g., to 0.29%�1; Ref.824). This is done when the O2 input is not rich enough toprovide an estimation of D (e.g., when employing singlesteps into hypoxia). This value of D has been estimated byfitting an exponential through the function 100 � Z, whereZ is the saturation calculated from the Hill model of theoxygen dissociation curve (824). The use of an exponen-tial model is equivalent to the empirically linear relationof log dVI to PO2 as used by Kronenberg et al. (411).

Since there is an empirical linear relationship be-tween VI and blood oxygen desaturation (626, 628), amore simple approach would be to express ventilation asa linear function of oxygen saturation

VI � V0 � S/SpO2�0� SpO2 (3)

where V0 and SpO2(0) are prehypoxic ventilation andsaturation, respectively, and S is the hypoxic sensitivity(units, L �min�1 �%�1). There are several reasons, how-ever, why saturation may not be the most appropriate toolto express the HVR. Oxygen saturation is not the stimulusto the oxygen sensors at the carotid bodies (e.g., Ref.313). The VI-SpO2 relationship is not under all circum-stances and protocols linear (159, 336, 384). And, mostimportantly, a change in the position of the oxygen dis-sociation curve (reflected in a change in P50) creates an

artifact with an incorrect calculation of HVR (613). Forexample, a metabolic acidosis increases P50 and changesthe value of �VI/�SpO2 without any change in the rela-tionship between VI and PO2 and hence of the HVR. Andfrom a technical point of view, it is our experience that atlow saturation levels (� 80%), measurements derivedfrom pulse oximetry are a less reliable measure of theactual arterial HbO2 saturation.

In the cat, it has been demonstrated that in thehypoxic range, ventilation is a linear function of log PaO2

,while hypoxic sensitivity, defined as the ratio �VI/�logPaO2

, is linearly related to the arterial H� concentration(650). Recently, we found that in humans relative changesin hypoxic sensitivity induced by alterations in arterialPCO2 are the same regardless of how the sensitivity wasdefined, either as �VI/�log PaO2

or as �VI/�SpO2 (775). Asin cats, humans also showed a linear relationship betweenhypoxic sensitivity (defined as �VI/�log PaO2

) and arterialH� concentration (see Fig. 7; Ref. 775). Future studies inhumans are warranted to see if, as in the cat, ventilationis a linear function of log PaO2

in the hypoxic range. If thiswould appear to be the case, then we would suggest theuse of the expression �VI/�log PO2 to estimate hypoxicsensitivity, with end-tidal O2 but preferentially arterial PO2

as input. It is a simple approach, alike to the use of oxygensaturation but without the confounders that are inher-ently present in the use of SpO2. When performing multi-ple steps into hypoxia (e.g., at three different backgroundPCO2 levels), it further enables the description of O2-CO2

interaction (see Figs. 7 and 12) and the effects that drugsmay have on this important feature of the peripheralchemoreceptors (such as abolishment of O2-CO2 interac-tion by inhalational anesthetics).

C. Modeling the HVR

Several models allow the estimation of physiologi-cally relevant parameters (hypoxic sensitivities/gains andtime constants) from single experimental data sets. Initialmathematical models of the ventilatory response to iso-capnic hypoxia consisted of two additive components: aninitial stimulating component arising from the carotidbodies and a second, slower component that causes adecline in ventilation (176, 824). These models were un-able to fit human responses into and out of hypoxiasimultaneously due to the asymmetry in the magnitude ofthe fast on- and off-transients (824). The model proposedby DeGoede et al. (176) is an exponential expression, inwhich the two components differ with respect to theirdynamics (expressed by a time constant or ), gain values,and input latencies. One of the first human studies (824)yielded the following parameter values: fast component: � 6 s, gain of response into hypoxia 50 l/min, gain ofresponse out of hypoxia 30 l/min; slow component: � 4



min, gain of response into hypoxia �40 l/min (i.e., HVD),gain of response out of hypoxia �20 l/min (relief of HVD).The asymmetry in the gain values suggests modulationof the peripheral component during sustained hypoxia asthe cause of HVD (cf. Ref. 162).

Khamnei and Robbins (382) examined how HVD isbest incorporated in models of the HVR. They concludedthat modulation of the peripheral chemoreflex gain duringsustained hypoxia consistently describes the human datain contrast to models in which hypoxic depression ismodeled as an additive component or as a modulator ofcentral or total CO2 sensitivity (382). In a quantitativemodeling study (by assuming that the peripheral gainterm is a linear function of arterial oxygen saturation),this was confirmed and moreover allowed (the asymme-try of) the ventilatory response into and out of hypoxia tobe analyzed simultaneously (577). However, in some sub-jects, the description of the response out of hypoxia (theoff-transient) remained inadequate. Therefore, to opti-mize the model, a further model expansion was tested inwhich HVD was modeled by modulation of the peripheralchemoreflex gain (577) combined with a component thatis independent of the peripheral chemoreflex loop (445).The extended model yielded significantly better fits in justtwo of six subjects. Both had a chemoreflex-independentHVD component �50%. This contrasts with a value of�10% in the four other subjects examined. Liang et al.(445) concluded that in some subjects, but not others,there might be a component of HVD that is independent ofthe peripheral chemoreflex sensitivity.

Some studies, human and animal, indicate that thefast hyperventilatory response to isocapnic hypoxia isbest described by a fast and a slower component (58, 198,384, 550) with time constants for the fast and slow com-ponents of 2–3 s and 70–100 s, respectively (58, 550).Possibly, the slow peripheral component is a manifesta-tion of central neuronal dynamics, equivalent to the slowcomponent observed when the carotid sinus nerve iselectrically stimulated (211). This then suggests that theHVR must be modeled by at least three components, twoof peripheral origin, one of which (or both?) is modulatedby central hypoxia, and in some subjects a peripheralchemoreflex loop independent HVD component.

Finally, studies that strictly control end-tidal PCO2

levels during HVR measurement may be influenced by aslow increase in ventilation that occurs upon the induc-tion of isohypercapnia. Tansley et al. (764) show a pro-gressive increase in ventilation lasting 2–8 h during expo-sure to isohypercapnia (end-tidal PCO2 � 6.5 Torr aboveresting levels). The increase in ventilation is slow andmost prominent during the first 2 h of exposure (�66ml/min for the first 2 h). This varying effect of baselineventilation may hamper proper assessment of the HVRirrespective of the method chosen to measure the HVR.Although there are various ways to take this effect into

account (e.g., by subtracting the variations in ventilationby performing additional isohypercapnic experiments orperforming hypoxic experiments just above resting end-tidal PCO2 levels), we suggest adding a trend term in theanalysis. Such an approach has been applied successfullyin studies assessing central and peripheral CO2 chemo-sensitivity yielding significant values for a trend in theorder of 70 ml/min2 (156).

VII. COMMENTS AND FUTURE CHALLENGES

At sea level, hypoxia is a relatively rare event inhealthy individuals, except in utero and perhaps during aneonatal period in which irregular breathing is not un-common, or during sleep when wakefulness drive is ab-sent and one may sometimes discontinue breathing. Whenhealthy individuals are exposed to inspired oxygen ten-sions lower or higher than normal, for example, at highaltitude or in an experimental setting, the respiratorysystem responds in a way that critically depends on thestage of maturation (neonates show different responsesthan adults), the physiological background condition(e.g., CO2 and exercise increase the HVR), the intensity ofthe stimulus, and, last but not least, the time pattern andduration of the exposure. Acute, chronic, and episodichypoxia with or without a chronic character all havedifferent effects on ventilatory control. Both in neonatesand adults, the respiratory system displays remarkableplasticity when exposed to intermittent and/or chronichypoxic paradigms that lead to changes in HVR pheno-types. Animal studies show that when hypoxic stimulusparadigms (or environmental factors such as nicotineexposure or maternal separation) are imposed during aneonatal critical time window, plastic changes persist intoadulthood and are difficult to reverse, if at all. An impor-tant challenge for future research is to examine whetherthe underlying mechanisms that are responsible for thesechanges also apply in humans.

An important circumstance that mediates plasticity-induced respiratory changes in rodents is neonatal oxida-tive stress. Rats exposed to chronic intermittent hypoxiadevelop an augmented carotid body sensitivity and anincreased HVR, changes that persist into adulthood (585;note, however, that as explained in section II, the increasein HVR is not without controversy). If translatable tohuman newborns, this raises a number of clinically rele-vant issues. Newborn babies, especially when premature,display irregular breathing that can be accompanied byfrequent apneic events. Does this disturb a normal matu-ration of the HVR in premature babies, and might this leadto sensitization of the carotid bodies thereby destabilizingbreathing just like in animals? Further destabilizingbreathing might lead to a higher frequency of apneicevents, an increase in the number of deoxygenation/



reoxygenation cycles, and finally to an increased produc-tion of ROS that would further aggravate the oxidativestress (in neonatal rats the augmented carotid body sen-sitivity is a ROS-related phenomenon; Ref. 586). And canpossible changes in the HVR be reversed at adult orpossibly at earlier age? An important difference betweennewborn and adult rats is the much higher sensitivity tochronic intermittent hypoxia of the former with respect tothe effects on carotid body sensitization, morphologicalchanges (e.g., hyperplasia), and reversibility (see Fig. 10;Ref. 585).

In neonatal animals, within a critical time window ofplasticity in the period after birth, chronic hypoxia im-pairs carotid body maturation and O2 sensitivity, causesdegenerative changes in glomus cells and sinus nerveafferents, and reduces both the CNS gain and the HVR(Fig. 9). The fact that at the level of ventilation the dimin-ished response to hypoxia persists into adulthood indi-cates that at least some of these adaptations are perma-nent and perhaps difficult to reverse. If similar mecha-nisms are operating in humans, the question arises as towhether chronic hypoxia in newborn babies, when un-treated, may have similar deleterious effects that ulti-mately may lead to a decreased ability to respond to anasphyxic stimulus, long-lasting exposures to hypoxia, andfailure to autoresuscitate. However, at least in the west-ern world, the clinical relevance of this issue may belimited because when newborn babies are hypoxemic, forexample, due to hypoventilation, ventilation perfusionmismatch, intrapulmonary or cardiac shunts, anemia, etc.,they will be rapidly treated with oxygen. On the otherhand however, apart from the well-known detrimentalside effects of oxygen (for example, retinopathy; see Ref.816 for further references), a permanently reduced HVR(from chronic oxygen therapy) in these children that per-sists into adulthood cannot be excluded. In rats, neonatalchronic hyperoxia can cause a reduced HVR at adult agedepending on the level of hyperoxia and the duration ofthe neonatal exposure (41, 71, 72).

An important point to consider when putting datafrom neonatal animals into the context of newborn babiesis relative maturity at birth. If 1- to 5-day-old rats resemblepreterm babies in their development of the HVR (73), thefindings from this species may be less relevant for termbabies. This, however, does not obviate the need to focusattention on possible adverse effects of neonatal oxida-tive stress (both hypoxia and hyperoxia) on the HVR inboth preterm and term babies.

In adult healthy humans, chronic intermittent hyp-oxia results in an augmented HVR, presumably related tooxidative stress (increased production of ROS; Refs. 374,605, 817). Studies in humans on the effects of chronicintermittent hypoxia are often aimed at mimickinghypoxic episodes that occur in patients with obstructivesleep apnea (OSA). This chronic disease is associated

with an enhanced sympathetic activity, systemic hyper-tension, and an augmented HVR. Similar to animals ex-posed to chronic intermittent hypoxia, OSA patients dis-play increased production of ROS (202, 365). The in-creases in both systemic blood pressure and HVR can bereversed with nasal continuous positive airway pressuretreatment (790, 791). To mimic the complex events thatoccur in OSA in a laboratory setting may be elusive in thesense that (awake) healthy subjects do not display theincrease in resting tone of upper airway dilator musclesthat in OSA patients serves to compensate for a decreasedpatency of the upper airways. This may well be one of thereasons why it is difficult if not impossible to evoke LTFof upper airway dilator muscles in healthy sleeping sub-jects but less so in snorers with upper airway flow limi-tations (21). Apart from the absence of obstructive ap-neas, an additional limitation of studies in awake healthysubjects is the lack of arousals that in sleeping OSApatients may play a crucial (albeit not clearly defined) roleand the absence of the influences of sleep itself on thecontrol of breathing. And finally, in OSA patients, apneicevents followed by hyperventilation are associated withalternating episodes of low and high PCO2, which makesthe stimulus paradigm fundamentally different from thatin most laboratory studies in which the hypoxic episodesare hypocapnic or at best isocapnic (Table 5). On theother hand, studies in animals and healthy humans are ofconsiderable interest because they can uncover mecha-nisms by which chronic intermittent hypoxia by itself canlead to maladapations such as inflammation, endothelialdysfunction, atherosclerosis, and other cardiovascular pa-thology.

From studies over the last decade, the pictureemerges that oxygen sensing by the carotid bodies occursvia several parallel pathways. The key role of potassiumchannel species is well established, but identification ofindividual members involved is complicated by species-and even strain-related differences in potassium channelregulation by hypoxia. AMP-activated protein kinase(AMPK), an enzyme considered to play a key role incellular metabolism, has been identified as a link betweenhypoxia-induced changes in mitochondrial metabolismand potassium channel activity. Mice deficient in the �2subunit of AMPK show an attenuated frequency responseto hypoxia in vivo (222). A possible role of AMPK inglucose sensing by the carotid bodies, in the adaptation tohigh altitude, as well as in exercise remain interestingsubjects for further study. To uncover the role of thenumerous individual type I cell neurotransmitters in theintegrated response of the carotid body to hypoxia isanother challenging issue.

A fascinating example of respiratory plasticity atadult age becomes manifest during adaptation to chronichypoxia. In fact, all elements in the carotid bodies andCNS known to participate in the hypoxia signaling trans-



duction cascade undergo plastic changes during chronichypoxia. How does the organism manage to orchestrateso many changes (recalibrations) at the same time with-out losing control over system stability? One of the masterregulators now identified is HIF-1, a transcription factorthat regulates the expression of numerous genes encod-ing elements involved in improving tissue oxygenation(HVR, erythropoiesis, angiogenesis, etc.) and in metabolicadaptations (increased ATP supply from glycolysis, opti-mization, and reduction of oxidative phosphorylation).Systems biology approaches are currently used to mapindividual genes, untangle genetic networks, and identifytheir hubs (441, 850). Systems biology also offers an im-portant avenue to discover cross-talk between hypoxiaand other key signaling mechanisms. For example, it hasbeen shown that angiogenesis in hypoxia is induced bycross-talk with notch signaling (441); notch receptors arehighly conserved transmembrane proteins thought to beinvolved in cell-to-cell communication and cell differenti-ation (references in Ref. 441). Similar mechanisms mayoperate in the carotid bodies, but efforts to examine thishave not yet been undertaken.

How hypoxia influences the stability and activity oftranscription factors is also focus of current interest. Thefact that transcription factors may also undergo posttran-scriptional modification independent of hypoxia is an in-teresting but complicating factor in the interpretation ofthese studies because hypoxia-independent pathwaysmay then autonomously influence the expression of genesinvolved in the adaptation of the HVR to chronic condi-tions. For example, HIF-1� can be stabilized by NO-me-diated S-nitrosylation and by heat shock protein 90, andboth processes are hypoxia independent (692). With re-gard to HIF-1�, targeting its stability independent of thePO2 may appear to open new treatment options of HIF-1-related pathology. For example, an interesting recent ob-servation in humans is the reversal of hypoxia-inducedpulmonary hypertension by iron supplementation (713;further references in Refs. 441, 692).

Animal studies indicate an important effect of geno-type on the HVR phenotype (i.e., the HVR magnitude).With the relatively large effect of environment and pre-frontal cortex on the magnitude and variability of HVRtaken into account, a genetic influence seems hard todiscover from human studies. However, in certain popu-lations and patient groups, the genotype may be the moredominant factor. An important question that so far hasreceived little attention is whether linking of the genotypeto magnitude of HVR is of clinical importance. Does aHVR risk phenotype exist which leads to an increased riskfor adverse respiratory events? And is this phenotypelinked to an increase in morbidity and mortality? This is,for example, of importance for patients with central orobstructive sleep apnea or patients receiving respiratorydepressants in the perioperative setting.

Finally, the method of measuring the HVR, especiallyin humans, remains a topic of intense discussion. Wepropose a standardized method (the Leiden protocol) thatallows measurement of the HVR with as few problems aspossible in its interpretation. Adoption of this protocolwill enable simple comparisons of physiological and phar-macological studies on HVR between laboratories.

ACKNOWLEDGMENTS

We thank Dr. Egbert Lakke for making Figure 4.Address for reprint requests and other correspondence:

L. J. Teppema, Dept. of Anesthesiology, Leiden University Med-ical Center, PO Box 9600, 2300 RC Leiden, The Netherlands(e-mail: [email protected]).

REFERENCES

1. Aaron EA, Powell FL. Effect of chronic hypoxia on hypoxicventilatory response in awake rats. J Appl Physiol 74: 1635–1640,1993.

2. Acker H. Mechanisms and meaning of cellular oxygen sensing inthe organism. Respir Physiol 95: 1–10, 1994.

3. Acker H. The oxygen sensing signal cascade under the influence ofreactive oxygen species. Philos Trans R Soc Lond B Biol Sci 360:2201–2210, 2005.

4. Adachi T, Ishikawa K, Hida W, Matsumoto H, Masuda T, Date

F, Ogawa K, Takeda K, Furuyama K, Zhang Y, Kitamuro T,

Ogawa H, Maruyama Y, Shibahara S. Hypoxemia and bluntedhypoxic ventilatory responses in mice lacking heme oxygenase-2.Biochem Biophys Res Commun 320: 514–522, 2004.

5. Adams JM, Severns ML. Interaction of chemoreceptor effectsand its dependence on the intensity of stimuli. J Appl Physiol 52:602–606, 1982.

6. Ahmed M, Giesbrecht GG, Serrette C, Georgopoulos D, An-

thonisen NR. Ventilatory response to hypoxia in elderly humans.Respir Physiol 83: 343–351, 1991.

7. Aicher SA, Kurucz OS, Reis DJ, Milner TA. Nucleus tractussolitarius efferent terminals synapse on neurons in the caudalventrolateral medulla that project to the rostral ventrolateral me-dulla. Brain Res 693: 51–63, 1995.

8. Ainslie PN, Barach A, Cummings KJ, Murrell C, Hamlin M,

Hellemans J. Cardiorespiratory and cerebrovascular responses toacute poikilocapnic hypoxia following intermittent and continuousexposure to hypoxia in humans. J Appl Physiol 102: 1953–1961,2007.

9. Ainslie PN, Burgess KR. Cardiorespiratory and cerebrovascularresponses to hyperoxic and hypoxic rebreathing: effects of accli-matization to high altitude. Respir Physiol Neurobiol 161: 201–209,2008.

10. Ainslie PN, Kolb JC, Ide K, Poulin MJ. Effects of five nights ofnormobaric hypoxia on the ventilatory responses to acute hypoxiaand hypercapnia. Respir Physiol Neurobiol 138: 193–204, 2003.

11. Ainslie PN, Kolb JC, Ide K, Poulin MJ. Effect of five nights ofnormobaric hypoxia on cardiovascular responses to acute isocap-nic hypoxia in humans: relationship to ventilatory chemosensitiv-ity. Ergonomics 48: 1523–1534, 2005.

12. Ainslie PN, Poulin MJ. Ventilatory, cerebrovascular, and cardio-vascular interactions in acute hypoxia: regulation by carbon diox-ide. J Appl Physiol 97: 149–159, 2004.

13. Aizad T, Bodani J, Cates D, Horvath L, Rigatto H. Effect of asingle breath of 100% oxygen on respiration in neonates duringsleep. J Appl Physiol 57: 1531–1535, 1984.

14. Alea OA, Czapla MA, Lasky JA, Simakajornboon N, Gozal E,

Gozal D. PDGF-beta receptor expression and ventilatory acclima-tization to hypoxia in the rat. Am J Physiol Regul Integr Comp

Physiol 279: R1625–R1633, 2000.15. Alvaro R, Alvarez J, Kwiatkowski K, Cates D, Rigatto H. Small

preterm infants less than or equal to 1500 g have only a sustained



decrease in ventilation in response to hypoxia. Pediatr Res 32:403–406, 1992.

16. Alvaro RE, Robertson M, Al Saedi S, Lemke RP, Cates DB,

Rigatto H. Preliminary characterization of a placental factor in-hibiting breathing in fetal sheep. Reprod Fertil Dev 9: 641–649,1997.

17. Andronikou S, Shirahata M, Mokashi A, Lahiri S. Carotid bodychemoreceptor and ventilatory responses to sustained hypoxia andhypercapnia in the cat. Respir Physiol 72: 361–374, 1988.

18. Ang RC, Hoop B, Kazemi H. Role of glutamate as the centralneurotransmitter in the hypoxic ventilatory response. J Appl

Physiol 72: 1480–1487, 1992.19. Arias-Stella J, Valcarcel J. Chief cell hyperplasia in the human

carotid body at high altitudes: physiologic and pathologic signifi-cance. Hum Pathol 7: 361–373, 1976.

20. Astrom K, Cohen JE, Willett-Brozick JE, Aston CE, Baysal

BE. Altitude is a phenotypic modifier in hereditary paragangliomatype-1: evidence for an oxygen-sensing defect. Hum Genet 113:228–237, 2003.

21. Babcock M, Shkoukani M, Aboubakr SE, Badr MS. Determi-nants of long-term facilitation in humans during NREM sleep.J Appl Physiol 94: 53–59, 2003.

22. Baby S, Roy A, Lahiri S. Role of mitochondria in the regulation ofhypoxia-inducible factor-1� in the rat carotid body glomus cells.Histochemy Cell Biol 124: 69–76, 2005.

23. Bach KB, Mitchell GS. Hypoxia-induced long-term facilitation ofrespiratory activity is serotonin dependent. Respir Physiol 104:251–260, 1996.

24. Baier RJ, Fajardo C, Alvarez J, Cates DB, Nowaczyk B,

Rigatto H. The effects of gestational age and labour on the breath-ing and behaviour response to oxygen and umbilical cord occlusionin the fetal sheep. J Dev Physiol 18: 93–98, 1992.

25. Bailey PL, Lu JK, Pace NL, Orr JA, White JL, Hamber EA,

Slawson MH, Crouch DJ, Rollins DE. Effects of intrathecalmorphine on the ventilatory response to hypoxia. N Engl J Med

343: 1228–1234, 2000.26. Bairam A, Carroll JL. Neurotransmitters in carotid body devel-

opment. Respir Physiol Neurobiol 149: 217–232, 2005.27. Baker TL, Fuller DD, Zabka AG, Mitchell GS. Respiratory

plasticity: differential actions of continuous and episodic hypoxiaand hypercapnia. Respir Physiol 129: 25–35, 2001.

28. Baker-Herman TL, Fuller DD, Bavis RW, Zabka AG, Golder

FJ, Doperalski NJ, Johnson RA, Watters JJ, Mitchell GS.

BDNF is necessary and sufficient for spinal respiratory plasticityfollowing intermittent hypoxia. Nat Neurosci 7: 48–55, 2004.

29. Balbir A, Lande B, Fitzgerald RS, Polotsky V, Mitzner W,

Shirahata M. Behavioral and respiratory characteristics duringsleep in neonatal DBA/2J and A/J mice. Brain Res 1241: 84–91,2008.

30. Balbir A, Lee H, Okumura M, Biswal S, Fitzgerald RS, Shira-

hata M. A search for genes that may confer divergent morphologyand function in the carotid body between two strains of mice. Am J

Physiol Lung Cell Mol Physiol 292: L704–L715, 2007.31. Bamford O, Hawkins RL. Central effects of an alpha 2-adrenergic

antagonist on fetal lambs: a possible mechanism for hypoxic apnea.J Dev Physiol 13: 353–358, 1990.

32. Bamford OS, Dawes GS, Denny R, Ward RA. Effects of the alpha2-adrenergic agonist clonidine and its antagonist idazoxan on thefetal lamb. J Physiol 381: 29–37, 1986.

33. Bamford OS, Sterni LM, Wasicko MJ, Montrose MH, Carroll

JL. Postnatal maturation of carotid body and type I cell chemore-ception in the rat. Am J Physiol Lung Cell Mol Physiol 276:L875–L884, 1999.

34. Barnard P, Andronikou S, Pokorski M, Smatresk N, Mokashi

A, Lahiri S. Time-dependent effect of hypoxia on carotid bodychemosensory function. J Appl Physiol 63: 685–691, 1987.

35. Barros RC, Branco LG. Central dopamine modulates anapyrexiabut not hyperventilation induced by hypoxia. J Appl Physiol 92:975–981, 2002.

36. Barros RC, Branco LG, Carnio EC. Evidence for thermoregula-tion by dopamine D1 and D2 receptors in the anteroventral preop-tic region during normoxia and hypoxia. Brain Res 1030: 165–171,2004.

37. Bascom DA, Clement ID, Cunningham DA, Painter R, Robbins

PA. Changes in peripheral chemoreflex sensitivity during sus-tained, isocapnic hypoxia. Respir Physiol 82: 161–176, 1990.

38. Bascom DA, Clement ID, Dorrington KL, Robbins PA. Effectsof dopamine and domperidone on ventilation during isocapnichypoxia in humans. Respir Physiol 85: 319–328, 1991.

39. Bavis RW. Developmental plasticity of the hypoxic ventilatoryresponse after perinatal hyperoxia and hypoxia. Respir Physiol

Neurobiol 149: 287–299, 2005.40. Bavis RW, Olson EB Jr, Mitchell GS. Critical developmental

period for hyperoxia-induced blunting of hypoxic phrenic re-sponses in rats. J Appl Physiol 92: 1013–1018, 2002.

41. Bavis RW, Olson EB Jr, Vidruk EH, Bisgard GE, Mitchell GS.

Level and duration of developmental hyperoxia influence impair-ment of hypoxic phrenic responses in rats. J Appl Physiol 95:1550–1559, 2003.

42. Bavis RW, Olson EB Jr, Vidruk EH, Fuller DD, Mitchell GS.

Developmental plasticity of the hypoxic ventilatory response inrats induced by neonatal hypoxia. J Physiol 557: 645–660, 2004.

43. Bavis RW, Russell KE, Simons JC, Otis JP. Hypoxic ventilatoryresponses in rats after hypercapnic hyperoxia and intermittenthyperoxia. Respir Physiol Neurobiol 155: 193–202, 2007.

44. Bavis RW, Wenninger JM, Miller BM, Dmitrieff EF, Olson EB

Jr, Mitchell GS, Bisgard GE. Respiratory plasticity after perina-tal hyperoxia is not prevented by antioxidant supplementation.Respir Physiol Neurobiol 160: 301–312, 2008.

45. Bavis RW, Mitchell GS. Long-term effects of the perinatal envi-ronment on respiratory control. J Appl Physiol 104: 1220–1229,2008.

46. Bayliss DA, Millhorn DE. Central neural mechanisms of proges-terone action: application to the respiratory system. J Appl Physiol

73: 393–404, 1992.47. Baysal BE. A phenotypic perspective on mammalian oxygen sen-

sor candidates. Ann NY Acad Sci 1073: 221–233, 2006.48. Baysal BE, Ferrell RE, Willett-Brozick JE, Lawrence EC,

Myssiorek D, Bosch A, van der MA, Taschner PE, Rubinstein

WS, Myers EN, Richard CW III, Cornelisse CJ, Devilee P,

Devlin B. Mutations in SDHD, a mitochondrial complex II gene, inhereditary paraganglioma. Science 287: 848–851, 2000.

49. Becker RF, King JE, Marsh RH, Wyrick AD. Intraauterine res-piration in the rat fetus. 1. Direct observations–Comparison withthe guinea pig. Am J Obstet Gynecol 90: 236–246, 1964.

50. Bee D, Barer G, Wach R, Pallot D, Emery C, Jones S. Structureand function of the carotid body in New Zealand genetically hyper-tensive rats. Q J Exp Physiol 74: 691–701, 1989.

51. Bee D, Pallot DJ. Acute hypoxic ventilation, carotid body celldivision, and dopamine content during early hypoxia in rats. J Appl

Physiol 79: 1504–1511, 1995.52. Behan M, Wenninger JM. Sex steroidal hormones and respiratory

control. Respir Physiol Neurobiol 164: 213–221, 2008.53. Beidleman BA, Muza SR, Fulco CS, Cymerman A, Ditzler D,

Stulz D, Staab JE, Skrinar GS, Lewis SF, Sawka MN. Intermit-tent altitude exposures reduce acute mountain sickness at 4300 m.Clin Sci Lond 106: 321–328, 2004.

54. Bekedam DJ, Mulder EJ, Snijders RJ, Visser GH. The effects ofmaternal hyperoxia on fetal breathing movements, body move-ments and heart rate variation in growth retarded fetuses. Early

Hum Dev 27: 223–232, 1991.55. Bell EL, Chandel NS. Genetics of mitochondrial electron trans-

port chain in regulating oxygen sensing. In: Methods in Enzymol-

ogy. Oxygen Biology and Hypoxia, edited by H. Sies and B. Brune.Orlando, FL: Academic, 2007, p. 447–461.

56. Bellville JW, Whipp BJ, Kaufman RD, Swanson GD, Aqleh KA,

Wiberg DM. Central and peripheral chemoreflex loop gain innormal and carotid body-resected subjects. J Appl Physiol 46:843–853, 1979.

57. Berkenbosch A, Dahan A, DeGoede J, Olievier IC. The venti-latory response to CO2 of the peripheral and central chemoreflexloop before and after sustained hypoxia in man. J Physiol 456:71–83, 1992.

58. Berkenbosch A, DeGoede J, Ward DS, Olievier CN, Van-

Hartevelt J. Dynamic response of the peripheral chemoreflex loopto changes in end-tidal O2. J Appl Physiol 71: 1123–1128, 1991.



59. Berkenbosch A, Heeringa J, Olievier CN, Kruyt EW. Artificialperfusion of the ponto-medullary region of cats. A method forseparation of central and peripheral effects of chemical stimulationof ventilation. Respir Physiol 37: 347–364, 1979.

60. Berkenbosch A, Olievier CN, DeGoede J, Kruyt EW. Effect onventilation of papaverine administered to the brain stem of theanaesthetized cat. J Physiol 443: 457–468, 1991.

61. Berkenbosch A, Teppema LJ, Olievier CN, Dahan A. Influencesof morphine on the ventilatory response to isocapnic hypoxia.Anesthesiology 86: 1342–1349, 1997.

62. Berkenbosch A, van Beek JH, Olievier CN, De Goede J, Quan-

jer PH. Central respiratory CO2 sensitivity at extreme hypocapnia.Respir Physiol 55: 95–102, 1984.

63. Berra E, Benizri E, Ginouves A, Volmat V, Roux D, Pouysse-

gur J. HIF prolyl-hydroxylase 2 is the key oxygen sensor settinglow steady-state levels of HIF-1alpha in normoxia. EMBO J 22:4082–4090, 2003.

64. Bin-Jaliah I, Maskell PD, Kumar P. Indirect sensing of insulin-induced hypoglycaemia by the carotid body in the rat. J Physiol

556: 255–266, 2004.65. Bisgard GE. Carotid body mechanisms in acclimatization to hyp-

oxia. Respir Physiol 121: 237–246, 2000.66. Bisgard GE, Busch MA, Daristotle L, Berssenbrugge AD, For-

ster HV. Carotid body hypercapnia does not elicit ventilatoryacclimatization in goats. Respir Physiol 65: 113–125, 1986.

67. Bisgard GE, Busch MA, Forster HV. Ventilatory acclimatizationto hypoxia is not dependent on cerebral hypocapnic alkalosis.J Appl Physiol 60: 1011–1015, 1986.

68. Bisgard GE, Forster HV. Ventilatory responses to acute andchronic hypoxia. In: Handbook of Physiology. Environmental

Physiology, edited by Frely MJ and Blatteis CM. New York: OxfordUniv. Press, 1996, p. 1207–1239.

69. Bisgard GE, Forster HV, Klein JP. Recovery of peripheral che-moreceptor function after denervation in ponies. J Appl Physiol 49:964–970, 1980.

70. Bisgard GE, Forster HV, Orr JA, Buss DD, Rawlings CA,

Rasmussen B. Hypoventilation in ponies after carotid body dener-vation. J Appl Physiol 40: 184–190, 1976.

71. Bisgard GE, Olson EB Jr, Wang ZY, Bavis RW, Fuller DD,

Mitchell GS. Adult carotid chemoafferent responses to hypoxiaafter 1, 2, and 4 wk of postnatal hyperoxia. J Appl Physiol 95:946–952, 2003.

72. Bisgard GE, OlsonEB Jr, Bavis RW, Wenninger J, Nordheim

EV, Mitchell GS. Carotid chemoafferent plasticity in adult ratsfollowing developmental hyperoxia. Respir Physiol Neurobiol 145:3–11, 2005.

73. Bissonnette JM. Mechanisms regulating hypoxic respiratory de-pression during fetal and postnatal life. Am J Physiol Regul Integr

Comp Physiol 278: R1391–R1400, 2000.74. Bissonnette JM, Hohimer AR. Acute anemic hypoxemia pro-

duces a transient depression in fetal respiratory activity. J Appl

Physiol 63: 1942–1946, 1987.75. Bissonnette JM, Knopp SJ. Separate respiratory phenotypes in

methyl-CpG-binding protein 2 Mecp2 deficient mice. Pediatr Res

59: 513–518, 2006.76. Bissonnette JM, Reddington M. Autoradiographic localization of

adenosine A1 receptors in brainstem of fetal sheep. Brain Res 61:111–115, 1991.

77. Blain GM, Smith CA, Henderson KS, Dempsey JA. Contribu-tion of the carotid body chemoreceptors to eupneic ventilation inthe intact, unanesthetized dog. J Appl Physiol 106: 1564–1573,2009.

78. Blanco CE, Chen V, Maertzdorf W, Bamford OS, Hanson M.

Effect of hyperoxia PaO250–90 mmHg on fetal breathing move-

ments in the unanaesthetized fetal sheep. J Dev Physiol 14: 235–241, 1990.

79. Blanco CE, Dawes GS, Hanson MA, McCooke HB. The re-sponse to hypoxia of arterial chemoreceptors in fetal sheep andnew-born lambs. J Physiol 351: 25–37, 1984.

80. Blanco CE, Dawes GS, Walker DW. Effects of hypoxia onpolysynaptic hind-limb reflexes in new-born lambs before and aftercarotid denervation. J Physiol 339: 467–474, 1983.

81. Blanco CE, Hanson MA, Johnson P, Rigatto H. Breathing pat-tern of kittens during hypoxia. J Appl Physiol 56: 12–17, 1984.

82. Blanco CE, Hanson MA, McCooke HB. Effects on carotid che-moreceptor resetting of pulmonary ventilation in the fetal lamb inutero. J Dev Physiol 10: 167–174, 1988.

83. Blood AB, Hunter CJ, Power GG. The role of adenosine inregulation of cerebral blood flow during hypoxia in the near-termfetal sheep. J Physiol 543: 1015–1023, 2002.

84. Blood AB, Hunter CJ, Power GG. Adenosine mediates decreasedcerebral metabolic rate and increased cerebral blood flow duringacute moderate hypoxia in the near-term fetal sheep. J Physiol 553:935–945, 2003.

85. Bocking AD. Assessment of fetal heart rate and fetal movementsin detecting oxygen deprivation in-utero. Eur J Obstet Gynecol

Reprod Biol 110 Suppl 1: S108–S112, 2003.86. Bocking AD, Gagnon R, Milne KM, White SE. Behavioral activ-

ity during prolonged hypoxemia in fetal sheep. J Appl Physiol 65:2420–2426, 1988.

87. Bocking AD, Harding R. Effects of reduced uterine blood flow onelectrocortical activity, breathing, and skeletal muscle activity infetal sheep. Am J Obstet Gynecol 154: 655–662, 1986.

88. Boddy K, Dawes GS, Fisher R, Pinter S, Robinson JS. Foetalrespiratory movements, electrocortical and cardiovascular re-sponses to hypoxaemia and hypercapnia in sheep. J Physiol 243:599–618, 1974.

89. Bonora M, Gautier H. Maturational changes in body temperatureand ventilation during hypoxia in kittens. Respir Physiol 68: 359–370, 1987.

90. Bonora M, Marlot D, Gautier H, Duron B. Effects of hypoxia onventilation during postnatal development in conscious kittens.J Appl Physiol 56: 1464–1471, 1984.

91. Bonora M, Vizek M. Ventilation, EELV and diaphragmatic activityin rats during the early phase of normobaric hypoxia. Respir

Physiol 128: 131–145, 2001.92. Boon JA, Milsom WK. NMDA receptor-mediated processes in the

Parabrachial/Kolliker fuse complex influence respiratory re-sponses directly and indirectly via changes in cortical activationstate. Respir Physiol Neurobiol 162: 63–72, 2008.

93. Brady JP, Ceruti E. Chemoreceptor reflexes in the new-borninfant: effects of varying degrees of hypoxia on heart rate andventilation in a warm environment. J Physiol 184: 631–645, 1966.

94. Brady R, Zaidi SI, Mayer C, Katz DM. BDNF is a target-derivedsurvival factor for arterial baroreceptor and chemoafferent primarysensory neurons. J Neurosci 19: 2131–2142, 1999.

95. Breen S, Rees S, Walker D. Identification of brainstem neuronsresponding to hypoxia in fetal and newborn sheep. Brain Res 748:107–121, 1997.

96. Brutsaert TD. Population genetic aspects and phenotypic plastic-ity of ventilatory responses in high altitude natives. Respir Physiol

Neurobiol 158: 151–160, 2007.97. Brutsaert TD, Parra EJ, Shriver MD, Gamboa A, Rivera-Ch M,

Leon-Velarde F. Ancestry explains the blunted ventilatory re-sponse to sustained hypoxia and lower exercise ventilation ofQuechua altitude natives. Am J Physiol Regul Integr Comp Physiol

289: R225–R234, 2005.98. Buckler KJ. TASK-like potassium channels and oxygen sensing in

the carotid body. Respir Physiol Neurobiol 157: 55–64, 2007.99. Buckler KJ, Vaughan-Jones RD. Effects of mitochondrial uncou-

plers on intracellular calcium, pH and membrane potential in ratcarotid body type I cells. J Physiol 513: 819–833, 1998.

100. Buerk DG, Osanai S, Mokashi A, Lahiri S. Dopamine, sensorydischarge, and stimulus interaction with CO2 and O2 in cat carotidbody. J Appl Physiol 85: 1719–1726, 1998.

101. Bunn HF, Poyton RO. Oxygen sensing and molecular adaptationto hypoxia. Physiol Rev 76: 839–885, 1996.

102. Bureau MA, Begin R. Postnatal maturation of the respiratoryresponse to O2 in awake newborn lambs. J Appl Physiol 52: 428–433, 1982.

103. Bureau MA, Lamarche J, Foulon P, Dalle D. The ventilatoryresponse to hypoxia in the newborn lamb after carotid body de-nervation. Respir Physiol 60: 109–119, 1985.

104. Burton MD, Kazemi H. Neurotransmitters in central respiratorycontrol. Respir Physiol 122: 111–121, 2000.



105. Busch MA, Bisgard GE, Forster HV. Ventilatory acclimatizationto hypoxia is not dependent on arterial hypoxemia. J Appl Physiol

58: 1874–1880, 1985.106. Caceres AI, Obeso A, Gonzalez C, Rocher A. Molecular identi-

fication and functional role of voltage-gated sodium channels in ratcarotid body chemoreceptor cells. Regulation of expression bychronic hypoxia in vivo. J Neurochem 102: 231–245, 2007.

107. Calbet JA. Chronic hypoxia increases blood pressure and nor-adrenaline spillover in healthy humans. J Physiol 551: 379–386,2003.

108. Calder NA, Williams BA, Kumar P, Hanson MA. The respiratoryresponse of healthy term infants to breath-by-breath alternations ininspired oxygen at two postnatal ages. Pediatr Res 35: 321–324,1994.

109. Campanucci VA, Zhang M, Vollmer C, Nurse CA. Expression ofmultiple P2X receptors by glossopharyngeal neurons projecting torat carotid body O2-chemoreceptors: role in nitric oxide-mediatedefferent inhibition. J Neurosci 26: 9482–9493, 2006.

110. Cao H, Kuo YR, Prabhakar NR. Absence of chemoreceptor inhi-bition by alpha-2 adrenergic receptor agonist in cats exposed tolow PO2 (Abstract). FASEB J 5: A1118, 1991.

111. Cao KY, Sullivan CE, Berthon-Jones M, Zwillich CW. Breath-ing route and resistive loading: influence on ventilatory response tohypoxia in conscious dogs. J Appl Physiol 75: 1247–1255, 1993.

112. Cao KY, Zwillich CW, Berthon-Jones M, Sullivan CE. In-creased normoxic ventilation induced by repetitive hypoxia inconscious dogs. J Appl Physiol 73: 2083–2088, 1992.

113. Cao KY, Zwillich CW, Berthon-Jones M, Sullivan CE. Ventila-tory response to sustained eucapnic hypoxia in the adult consciousdog. Respir Physiol 89: 65–73, 1992.

114. Carpenter E, Bee D, Peers C. Ionic currents in carotid body typeI cells isolated from normoxic and chronically hypoxic adult rats.Brain Res 811: 79–87, 1998.

115. Carroll JL. Developmental plasticity in respiratory control. J Appl

Physiol 94: 375–389, 2003.116. Carroll JL, Bamford OS, Fitzgerald RS. Postnatal maturation of

carotid chemoreceptor responses to O2 and CO2 in the cat. J Appl

Physiol 75: 2383–2391, 1993.117. Carroll JL, Bureau MA. Decline in peripheral chemoreceptor

excitatory stimulation during acute hypoxia in the lamb. J Appl

Physiol 63: 795–802, 1987.118. Carroll JL, Bureau MA. Peripheral chemoreceptor CO2 response

during hyperoxia in the 14-day-old awake lamb. Respir Physiol 73:339–349, 1988.

119. Carroll JL, Canet E, Bureau MA. Dynamic ventilatory responsesto CO2 in the awake lamb: role of the carotid chemoreceptors.J Appl Physiol 71: 2198–2205, 1991.

120. Cartwright CR, Henson LC, Ward DS. Effects of alfentanil onthe ventilatory response to sustained hypoxia. Anesthesiology 89:612–619, 1998.

121. Cattarossi L, Haxhiu-Poskurica B, Haxhiu MA, Litmanovitz I,

Martin RJ, Carlo WA. Carotid bodies and ventilatory response tohypoxia in aminophylline-treated piglets. Pediatr Pulmonol 20:94–100, 1995.

122. Chandel NS, Budinger GR. The cellular basis for diverse re-sponses to oxygen. Free Radic Biol Med 42: 165–174, 2007.

123. Chandel NS, Schumacker PT. Cellular oxygen sensing by mito-chondria: old questions, new insight. J Appl Physiol 88: 1880–1889,2000.

124. Chapman KR, Cherniack NS. Aging effects on the interaction ofhypercapnia and hypoxia as ventilatory stimuli. J Gerontol 42:202–209, 1987.

125. Chau A, Koos BJ. Metabolic and cardiorespiratory responses tohypoxia in fetal sheep: adenosine receptor blockade. Am J Physiol

Regul Integr Comp Physiol 276: R1805–R1811, 1999.126. Chavez JC, Agani F, Pichiule P, LaManna JC. Expression of

hypoxia-inducible factor-1alpha in the brain of rats during chronichypoxia. J Appl Physiol 89: 1937–1942, 2000.

127. Chen J, He L, Dinger B, Stensaas L, Fidone S. Chronic hypoxiaupregulates connexin43 expression in rat carotid body and petrosalganglion. J Appl Physiol 92: 1480–1486, 2002.

128. Chen J, He L, Dinger B, Stensaas L, Fidone S. Role of endo-thelin and endothelin A-type receptor in adaptation of the carotid

body to chronic hypoxia. Am J Physiol Lung Cell Mol Physiol 282:L1314–L1323, 2002.

129. Chen J, He L, Liu X, Dinger B, Stensaas L, Fidone S. Effect ofthe endothelin receptor antagonist bosentan on chronic hypoxia-induced morphological and physiological changes in rat carotidbody. Am J Physiol Lung Cell Mol Physiol 292: L1257–L1262, 2007.

130. Cheng Z, Guo SZ, Lipton AJ, Gozal D. Domoic acid lesions innucleus of the solitary tract: time-dependent recovery of hypoxicventilatory response and peripheral afferent axonal plasticity.J Neurosci 22: 3215–3226, 2002.

131. Cherniack NS, Dempsey J, Fencl V, Fitzgerald RS, Lourenco

RV, Rebuck AS, Rigg J, Severinghaus JW, Weil JW, Whitelaw

WA, Zwilich CW. Workshop on assessment of respiratory controlin humans. I. Methods of measurement of ventilatory responses tohypoxia and hypercapnia. Am Rev Respir Dis 115: 177–181, 1977.

132. Cherniack NS, Edelman NH, Lahiri S. Hypoxia and hypercapniaas respiratory stimulants and depressants. Respir Physiol 11: 113–126, 1970.

133. Chowdhury R, Hardy A, Schofield CJ. The human oxygen sens-ing machinery and its manipulation. Chem Soc Rev 37: 1308–1319,2008.

134. Clar C, Dorrington KL, Robbins PA. Ventilatory effects of 8 h ofisocapnic hypoxia with and without beta-blockade in humans.J Appl Physiol 86: 1897–1904, 1999.

135. Clarke JA, Daly MB, Marshall JM, Ead HW, Hennessy EM.

Quantitative studies of the vasculature of the carotid body in thechronically hypoxic rat. Braz J Med Biol Res 33: 331–340, 2000.

136. Clement ID, Pandit JJ, Bascom DA, Dorrington KL, O’Connor

DF, Robbins PA. An assessment of central-peripheral ventilatorychemoreflex interaction using acid and bicarbonate infusions inhumans. J Physiol 485: 561–570, 1995.

137. Clewlow F, Dawes GS, Johnston BM, Walker DW. Changes inbreathing, electrocortical and muscle activity in unanaesthetizedfetal lambs with age. J Physiol 341: 463–476, 1983.

138. Cohen G, Katz-Salamon M. Development of chemoreceptor re-sponses in infants. Respir Physiol Neurobiol 149: 233–242, 2005.

139. Cohen G, Malcolm G, Henderson-Smart D. Ventilatory responseof the newborn infant to mild hypoxia. Pediatr Pulmonol 24:163–172, 1997.

140. Coles SK, Dick TE. Neurones in the ventrolateral pons are re-quired for post-hypoxic frequency decline in rats. J Physiol 497:79–94, 1996.

141. Conde SV, Obeso A, Gonzalez C. Low glucose effects on ratcarotid body chemoreceptor cells’ secretory responses and actionpotential frequency in the carotid sinus nerve. J Physiol 585: 721–730, 2007.

142. Conde SV, Obeso A, Rigual R, Monteiro EC, Gonzalez C.

Function of the rat carotid body chemoreceptors in ageing. J Neu-

rochem 99: 711–723, 2006.143. Cormack RS, Cunningham DJ, Gee JB. The effect of carbon

dioxide on the respiratory response to want of oxygen in man. Q J

Exp Physiol Cogn Med Sci 42: 303–319, 1957.144. Corne S, Webster K, Younes M. Hypoxic respiratory response

during acute stable hypocapnia. Am J Respir Crit Care Med 167:1193–1199, 2003.

145. Cragg PA, Drysdale DB. Interaction of hypoxia and hypercapniaon ventilation, tidal volume and respiratory frequency in the an-aesthetized rat. J Physiol 341: 477–493, 1983.

146. Cross AR, Henderson L, Jones OT, Delpiano MA, Hentschel J,

Acker H. Involvement of an NADPH oxidase as a PO2 sensorprotein in the rat carotid body. Biochem J 272: 743–747, 1990.

147. Cross KW, Oppe TE. The effect of inhalation of high and lowconcentrations of oxygen on the respiration of the premature in-fant. J Physiol 117: 38–55, 1952.

148. Cruz JC, Reeves JT, Grover RF, Maher JT, McCullough RE,

Cymerman A, Denniston JC. Ventilatory acclimatization to highaltitude is prevented by CO2 breathing. Respiration 39: 121–130,1980.

149. Cummins EP, Taylor CT. Hypoxia-responsive transcription fac-tors. Pflugers Arch 450: 363–371, 2005.

150. Cunningham DJC, Robbins PA, Wolff CB. Integration of respi-ratory response to changes in alveolar partial pressures in CO2 andO2 and in arterial pH. In: Handbook of Physiology. The Respiratory



System. Control of Breathing. Bethesda, MD: Am. Physiol. Soc.,1986, sect. 3, vol. II, pt. 2, chapt. 15, p. 475–528.

151. Curran AK, Rodman JR, Eastwood PR, Henderson KS, Demp-

sey JA, Smith CA. Ventilatory responses to specific CNS hypoxiain sleeping dogs. J Appl Physiol 88: 1840–1852, 2000.

152. Curran LS, Zhuang J, Sun SF, Moore LG. Ventilation andhypoxic ventilatory responsiveness in Chinese-Tibetan residents at3,658 m. J Appl Physiol 83: 2098–2104, 1997.

153. D’Agostino D, Mazza E Jr, Neubauer JA. Heme oxygenase isnecessary for the excitatory response of cultured neonatal ratrostral ventrolateral medulla neurons to hypoxia. Am J Physiol

Regul Integr Comp Physiol 296: R102–R118, 2009.154. Dahan A, Nieuwenhuijs D, Teppema L. Plasticity of central

chemoreceptors: effect of bilateral carotid body resection on cen-tral CO2 sensitivity. PLoS Medicine 4: e239, 2007.

155. Dahan A, Berkenbosch A, DeGoede J, Olievier IC, Bovill JG.

On a pseudo-rebreathing technique to assess the ventilatory sensi-tivity to carbon dioxide in man. J Physiol 423: 615–629, 1990.

156. Dahan A, DeGoede J, Berkenbosch A, Olievier IC. The influ-ence of oxygen on the ventilatory response to carbon dioxide inman. J Physiol 428: 485–499, 1990.

157. Dahan A, Sarton E, Teppema L, Olievier C. Sex-related differ-ences in the influence of morphine on ventilatory control in hu-mans. Anesthesiology 88: 903–913, 1998.

158. Dahan A, Sarton E, Teppema L, Olievier C, Nieuwenhuijs D,

Matthes HW, Kieffer BL. Anesthetic potency and influence ofmorphine and sevoflurane on respiration in mu-opioid receptorknockout mice. Anesthesiology 94: 824–832, 2001.

159. Dahan A, Sarton E, van den Elsen M, van Kleef J, Teppema L,

Berkenbosch A. Ventilatory response to hypoxia in humans. In-fluences of subanesthetic desflurane. Anesthesiology 85: 60–68,1996.

160. Dahan A, van den Elsen MJ, Berkenbosch A, DeGoede J,

Olievier IC, van Kleef JW, Bovill JG. Effects of subanesthetichalothane on the ventilatory responses to hypercapnia and acutehypoxia in healthy volunteers. Anesthesiology 80: 727–738, 1994.

161. Dahan A, van den Elsen MJLJ, de Goede J, Olievier ICW,

Burm AGL, van Kleef JW. Influence of subanesthetic concentra-tion of halothane on the ventilatory response to step changes intoand out of sustained isocapnic hypoxia in healthy-volunteers. An-

esthesiology 81: 850–859, 1994.162. Dahan A, Ward D, van den Elsen M, Temp J, Berkenbosch A.

Influence of reduced carotid body drive during sustained hypoxiaon hypoxic depression of ventilation in humans. J Appl Physiol 81:565–572, 1996.

163. Daristotle L, Berssenbrugge AD, Bisgard GE. Hypoxic-hyper-capnic ventilatory interaction at the carotid body of awake goats.Respir Physiol 70: 63–72, 1987.

164. Daristotle L, Bisgard GE. Central-peripheral chemoreceptor ven-tilatory interaction in awake goats. Respir Physiol 76: 383–391,1989.

165. Daristotle L, Engwall MJ, Niu WZ, Bisgard GE. Ventilatoryeffects and interactions with change in PaO2

in awake goats. J Appl

Physiol 71: 1254–1260, 1991.166. Darnall RA, Green G, Pinto L, Hart N. Effect of acute hypoxia

on respiration and brain stem blood flow in the piglet. J Appl

Physiol 70: 251–259, 1991.167. Dasso LL, Buckler KJ, Vaughan-Jones RD. Interactions be-

tween hypoxia and hypercapnic acidosis on calcium signaling incarotid body type I cells. Am J Physiol Lung Cell Mol Physiol 279:L36–L42, 2000.

168. Daudu PA, Roy A, Rozanov C, Mokashi A, Lahiri S. Extra- andintracellular free iron and the carotid body responses. Respir

Physiol Neurobiol 130: 21–31, 2002.169. Dauger S, Pattyn A, Lofaso F, Gaultier C, Goridis C, Gallego

J, Brunet JF. Phox2b controls the development of peripheralchemoreceptors and afferent visceral pathways. Development 130:6635–6642, 2003.

170. Dawes GS. The central control of fetal breathing and skeletalmuscle movements. J Physiol 346: 1–18, 1984.

171. Dawes GS, Gardner WN, Johnston BM, Walker DW. Breathingin fetal lambs: the effect of brain stem section. J Physiol 335:535–553, 1983.

172. Day TA, Wilson RJ. Brainstem PCO2 modulates phrenic responsesto specific carotid body hypoxia in an in situ dual perfused ratpreparation. J Physiol 578: 843–857, 2007.

173. De Paula PM, Branco LG. Nitric oxide in the rostral ventrolateralmedulla modulates hyperpnea but not anapyrexia induced by hyp-oxia. Brain Res 977: 231–238, 2003.

174. De Paula PM, Tolstykh G, Mifflin S. Chronic intermittent hyp-oxia alters NMDA and AMPA-evoked currents in NTS neuronsreceiving carotid body chemoreceptor inputs. Am J Physiol Regul

Integr Comp Physiol 292: R2259–R2265, 2007.175. DeGoede J, Berkenbosch A, Ward DS, Bellville JW, Olievier

CN. Comparison of chemoreflex gains obtained with two differentmethods in cats. J Appl Physiol 59: 170–179, 1985.

176. DeGoede J, VanderHoeven N, Berkenbosch A, Olievier CN,

and van Beek JHGM. Ventilatory responses to sudden isocapnicchanges in end-tidal O2 in cats. In: Modelling and Control of

Breathing, edited by Whipp BJ and Wiberg DM. New York: ElsevierBiomedical, 1983, p. 37–45.

177. Dejours P, Labrousse Y, Raynoud J, Teillac A. Oxygen che-moreflex stimulus in ventilation at low altitude in man. I. At rest.J Physiol 49: 115–120, 1957.

178. Dempsey JA, Forster HV. Mediation of ventilatory adaptations.Physiol Rev 62: 262–346, 1982.

179. DiGiulio C, Cacchio M, Bianchi G, Rapino C, Di IlioC. Selectedcontribution: carotid body as a model for aging studies: is there alink between oxygen and aging? J Appl Physiol 95: 1755–1758,2003.

180. DiGiulio C, Huang W, Mokashi A, Lahiri S. Further character-ization of stimulus interaction of cat carotid chemoreceptors. J

Auton Nerv Syst 71: 196–200, 1998.181. DiGiulio C, Huang WX, Mokashi A, Roy A, Cacchio M, Macri

MA, Lahiri S. Sustained hypoxia promotes hyperactive responseof carotid body in the cat. Respir Physiol Neurobiol 134: 69–74,2003.

182. Dillon GH, Waldrop TG. In vitro responses of caudal hypotha-lamic neurons to hypoxia and hypercapnia. Neuroscience 51: 941–950, 1992.

183. Dillon GH, Waldrop TG. Responses of feline caudal hypothalamiccardiorespiratory neurons to hypoxia and hypercapnia. Exp Brain

Res 96: 260–272, 1993.184. Dinger B, He L, Chen J, Liu X, Gonzalez C, Obeso A, Sanders

K, Hoidal J, Stensaas L, Fidone S. The role of NADPH oxidasein carotid body arterial chemoreceptors. Respir Physiol Neurobiol

157: 45–54, 2007.185. Donnelly DF. Chemoreceptor nerve excitation may not be pro-

portional to catecholamine secretion. J Appl Physiol 81: 657–664,1996.

186. Donnelly DF. Development of carotid body/petrosal ganglion re-sponse to hypoxia. Respir Physiol Neurobiol 149: 191–199, 2005.

187. Donnelly DF, Bavis RW, Kim I, Dbouk HA, Carroll JL. Timecourse of alterations in pre- and post-synaptic chemoreceptor func-tion during developmental hyperoxia. Respir Physiol Neurobiol

168: 189–197, 2009.188. Donnelly DF, Doyle TP. Developmental changes in hypoxia-in-

duced catecholamine release from rat carotid body, in vitro.J Physiol 475: 267–275, 1994.

189. Donnelly DF, Kim I, Carle C, Carroll JL. Perinatal hyperoxia for14 days increases nerve conduction time and the acute unitaryresponse to hypoxia of rat carotid body chemoreceptors. J Appl

Physiol 99: 114–119, 2005.190. Donnelly DF, Smith E, Dutton RE. Carbon dioxide versus H ion

as a chemoreceptor stimulus. Brain Res 245: 136–138, 1982.191. Donnelly DF. Developmental aspects of oxygen sensing by the

carotid body. J Appl Physiol 88: 2296–2301, 2000.192. Donoghue S, Fatemian M, Balanos GM, Crosby A, Liu C,

O’Connor D, Talbot NP, Robbins PA. Ventilatory acclimatizationin response to very small changes in PO2 in humans. J Appl Physiol

98: 1587–1591, 2005.193. Drorbaugh JE, Fenn WO. A barometric method for measuring

ventilation in newborn infants. Pediatrics 16: 81–87, 1955.194. Duchen MR, Biscoe TJ. Relative mitochondrial membrane poten-

tial and [Ca2�]i in type I cells isolated from the rabbit carotid body.J Physiol 450: 33–61, 1992.



195. Duffin J. Measuring the ventilatory response to hypoxia. J Physiol

584: 285–293, 2007.196. Duprat F, Lauritzen I, Patel A, Honore E. The TASK back-

ground K2P channels: chemo- and nutrient sensors. Trends Neu-

rosci 30: 573–580, 2007.197. Durand E, Lofaso F, Dauger S, Vardon G, Gaultier C, Gallego

J. Intermittent hypoxia induces transient arousal delay in newbornmice. J Appl Physiol 96: 1216–1222, 2004.

198. Dutton RE, Smith EJ, Ghatak PK, Davies DG. Dynamics of therespiratory controller during carotid body hypoxia. J Appl Physiol

35: 844–850, 1973.199. Dwinell MR, Forster HV, Petersen J, Rider A, Kunert MP,

Cowley AW Jr, Jacob HJ. Genetic determinants on rat chromo-some 6 modulate variation in the hypercapnic ventilatory responseusing consomic strains. J Appl Physiol 98: 1630–1638, 2005.

200. Dwinell MR, Janssen PL, Pizarro J, Bisgard GE. Effects ofcarotid body hypocapnia during ventilatory acclimatization to hyp-oxia. J Appl Physiol 82: 118–124, 1997.

201. Dwinell MR, Powell FL. Chronic hypoxia enhances the phrenicnerve response to arterial chemoreceptor stimulation in anesthe-tized rats. J Appl Physiol 87: 817–823, 1999.

202. Dyugovskaya L, Lavie P, Lavie L. Increased adhesion moleculesexpression and production of reactive oxygen species in leuko-cytes of sleep apnea patients. Am J Respir Crit Care Med 165:934–939, 2002.

203. Easton PA, Anthonisen NR. Ventilatory response to sustainedhypoxia after pretreatment with aminophylline. J Appl Physiol 64:1445–1450, 1988.

204. Easton PA, Slykerman LJ, Anthonisen NR. Ventilatory re-sponse to sustained hypoxia in normal adults. J Appl Physiol 61:906–911, 1986.

205. Easton PA, Slykerman LJ, Anthonisen NR. Recovery of theventilatory response to hypoxia in normal adults. J Appl Physiol

64: 521–528, 1988.206. Edelman NH, Epstein PE, Lahiri S, Cherniack NS. Ventilatory

responses to transient hypoxia and hypercapnia in man. Respir

Physiol 17: 302–314, 1973.207. Eden GJ, Hanson MA. Effects of chronic hypoxia from birth on

the ventilatory response to acute hypoxia in the newborn rat.J Physiol 392: 11–19, 1987.

208. Eden GJ, Hanson MA. Maturation of the respiratory response toacute hypoxia in the newborn rat. J Physiol 392: 1–9, 1987.

209. Edwards C, Heath D, Harris P, Castillo Y, Kruger H, Arias-

Stella J. The carotid body in animals at high altitude. J Pathol 104:231–238, 1971.

210. El Hasnaoui-Saadani R, Alayza RC, Launay T, Pichon A,

Quidu P, Beaudry M, Leon-Velarde F, Richalet JP, Duvallet A,

Favret F. Brain stem NO modulates ventilatory acclimatization tohypoxia in mice. J Appl Physiol 103: 1506–1512, 2007.

211. Eldridge FL, Millhorn DE. Oscillation, gating and memory in therespiratory control system. In: Handbook of Physiology. The Re-

spiratory System. Control of Breathing. Bethesda, MD: Am.Physiol. Soc., 1986, sect. 3, vol. II, pt. 1, chapt. 3, p. 93–114.

212. Elnazir B, Marshall JM, Kumar P. Postnatal development of thepattern of respiratory and cardiovascular response to systemichypoxia in the piglet: the roles of adenosine. J Physiol 492: 573–585, 1996.

213. Eng C, Kiuru M, Fernandez MJ, Aaltonen LA. A role for mito-chondrial enzymes in inherited neoplasia and beyond. Nat Rev

Cancer 3: 193–202, 2003.214. Engeman RM, Sherrill DL, Swanson GD. On the estimation of

hypoxic ventilatory response. Comput Biomed Res 18: 553–562,1985.

215. Engwall MJ, Bisgard GE. Ventilatory responses to chemorecep-tor stimulation after hypoxic acclimatization in awake goats. J Appl

Physiol 69: 1236–1243, 1990.216. Engwall MJ, Vidruk EH, Nielsen AM, Bisgard GE. Response of

the goat carotid body to acute and prolonged hypercapnia. Respir

Physiol 74: 335–344, 1988.217. Enhorning G. Inhibition of fetal breathing: a pilot study. Pediatr

Res 34: 834–836, 1993.

218. Enhorning G, van Schaik S, Lundgren C, Vargas I. Whole-bodyplethysmography, does it measure tidal volume of small animals?Can J Physiol Pharmacol 76: 945–951, 1998.

219. Epstein RA, Epstein MA, Haddad GG, Mellins RB. Practicalimplementation of the barometric method for measurement of tidalvolume. J Appl Physiol 49: 1107–1115, 1980.

220. Erickson JT, Mayer C, Jawa A, Ling L, Olson EB Jr, Vidruk

EH, Mitchell GS, Katz DM. Chemoafferent degeneration andcarotid body hypoplasia following chronic hyperoxia in newbornrats. J Physiol 509: 519–526, 1998.

221. Erickson JT, Millhorn DE. Hypoxia and electrical stimulation ofthe carotid sinus nerve induce Fos-like immunoreactivity withincatecholaminergic and serotoninergic neurons of the rat brainstem.J Comp Neurol 348: 161–182, 1994.

222. Evans AM, Hardie DG, Peers C, Wyatt CN, Viollet B, Kumar P,

Dallas ML, Ross F, Ikematsu N, Jordan HL, Barr BL, Rafferty

JN, Ogunbayo O. Ion channel regulation by AMPK: the route ofhypoxia-response coupling in thecarotid body and pulmonary ar-tery. Ann NY Acad Sci 1177: 89–100, 2009.

223. Evans AM, Mustard KJ, Wyatt CN, Peers C, Dipp M, Kumar P,

Kinnear NP, Hardie DG. Does AMP-activated protein kinase cou-ple inhibition of mitochondrial oxidative phosphorylation by hyp-oxia to calcium signaling in O2-sensing cells? J Biol Chem 280:41504–41511, 2005.

224. Fabris G, Anselmo-Franci JA, Branco LG. Role of nitric oxide inhypoxia-induced hyperventilation and hypothermia: participationof the locus coeruleus. Braz J Med Biol Res 32: 1389–1398, 1999.

225. Fabris G, Steiner AA, Anselmo-Franci JA, Branco LG. Role ofnitric oxide in rat locus coeruleus in hypoxia-induced hyperventi-lation and hypothermia. Neuroreport 11: 2991–2995, 2000.

226. Fatemian M, Gamboa A, Leon-Velarde F, Rivera-Ch M, Pala-

cios JA, Robbins PA. Selected contribution: ventilatory responseto CO2 in high-altitude natives and patients with chronic mountainsickness. J Appl Physiol 94: 1279–1287, 2003.

227. Fatemian M, Nieuwenhuijs DJ, Teppema LJ, Meinesz S, van

der Mey AG, Dahan A, Robbins PA. The respiratory response tocarbon dioxide in humans with unilateral and bilateral resectionsof the carotid bodies. J Physiol 549: 965–973, 2003.

228. Fatemian M, Robbins PA. Human ventilatory response to CO2

after 8 h of isocapnic or poikilocapnic hypoxia. J Appl Physiol 85:1922–1928, 1998.

229. Fatemian M, Robbins PA. Selected contribution: chemoreflexresponses to CO2 before and after an 8-h exposure to hypoxia inhumans. J Appl Physiol 90: 1607–1614, 2001.

230. Feldman JL, Mitchell GS, Nattie EE. Breathing: rhythmicityplasticity chemosensitivity. Annu Rev Neurosci 26: 239–266, 2003.

231. Filuk RB, Berezanski DJ, Anthonisen NR. Depression ofhypoxic ventilatory response in humans by somatostatin. J Appl

Physiol 65: 1050–1054, 1988.232. Finley JC, Katz DM. The central organization of carotid body

afferent projections to the brainstem of the rat. Brain Res 572:108–116, 1992.

233. Fitzgerald RS, Lahiri S. Reflex responses to chemoreceptorsstimulation. In: Handbook of Physiology. The Respiratory System.

Control of Breathing. Bethesda, MD: Am. Physiol. Soc., 1986, sect.3, vol. II, pt. 1, chapt. 10, p. 313–362.

234. Fitzgerald RS, Parks DC. Effect of hypoxia on carotid chemore-ceptor response to carbon dioxide in cats. Respir Physiol 12:218–229, 1971.

235. Foo IT, Martin SE, Lee RJ, Drummond GB, Warren PM. Inter-action of the effects of domperidone and sub-anaesthetic concen-trations of isoflurane on the immediate and sustained hypoxicventilatory response in humans. Br J Anaesth 74: 134–140, 1995.

236. Foo IT, Warren PM, Drummond GB. Influence of oral clonidineon the ventilatory response to acute and sustained isocapnic hyp-oxia in human males. Br J Anaesth 76: 214–220, 1996.

237. Fordyce WE. Hypoxic ventilatory control in the awake cat fiveyears after carotid body resection. Respir Physiol 69: 209–225,1987.

238. Forster HV. Plasticity in the control of breathing following sen-sory denervation. J Appl Physiol 94: 784–794, 2003.

239. Forster HV, Bisgard GE, Rasmussen B, Orr JA, Buss DD,

Manohar M. Ventilatory control in peripheral chemoreceptor-den-



ervated ponies during chronic hypoxemia. J Appl Physiol 41: 878–885, 1976.

240. Forster HV, Dwinell MR, Hodges MR, Brozoski D, Hogan GE.

Do genes on rat chromosomes 9, 13, 16, 18, and 20 contribute toregulation of breathing? Respir Physiol Neurobiol 135: 247–261,2003.

241. Foster GE, McKenzie DC, Milsom WK, Sheel AW. Effects oftwo protocols of intermittent hypoxia on human ventilatory, car-diovascular and cerebral responses to hypoxia. J Physiol 567:689–699, 2005.

242. Foster GE, Poulin MJ, Hanly PJ. Intermittent hypoxia and vas-cular function: implications for obstructive sleep apnoea. Exp

Physiol 92: 51–65, 2007.243. Frappell P, Lanthier C, Baudinette RV, Mortola JP. Metabo-

lism and ventilation in acute hypoxia: a comparative analysis insmall mammalian species. Am J Physiol Regul Integr Comp

Physiol 262: R1040–R1046, 1992.244. Fregosi RF, Mitchell GS. Long-term facilitation of inspiratory

intercostal nerve activity following carotid sinus nerve stimulationin cats. J Physiol 477: 469–479, 1994.

245. Fuller DD, Baker TL, Behan M, Mitchell GS. Expression ofhypoglossal long-term facilitation differs between substrains ofSprague-Dawley rat. Physiol Genomics 4: 175–181, 2001.

246. Fuller DD, Bavis RW, Vidruk EH, Wang ZY, Olson EB Jr,

Bisgard GE, Mitchell GS. Life-long impairment of hypoxicphrenic responses in rats following 1 month of developmentalhyperoxia. J Physiol 538: 947–955, 2002.

247. Fuller DD, Wang ZY, Ling L, Olson EB, Bisgard GE, Mitchell

GS. Induced recovery of hypoxic phrenic responses in adult ratsexposed to hyperoxia for the first month of life. J Physiol 536:917–926, 2001.

248. Fung ML, Dong X. Absence of MK801-induced inspiratory prolon-gation in chronically hypoxic rats. Life Sci 69: 2319–2326, 2001.

249. Fung ML, Wang W, Darnall RA, St John WM. Characterizationof ventilatory responses to hypoxia in neonatal rats. Respir Physiol

103: 57–66, 1996.250. Gabel RA, Weiskopf RB. Ventilatory interaction between hypoxia

and [H�] at chemoreceptors of man. J Appl Physiol 39: 292–296,1975.

251. Gagnon R, Hunse C, Vijan S. The effect of maternal hyperoxia onbehavioral activity in growth-retarded human fetuses. Am J Obstet

Gynecol 163: 1894–1899, 1990.252. Gamboa J, Macarlupu JL, Rivera-Chira M, Monge C, Leon-

Velarde F. Effect of domperidone on ventilation and polycythemiaafter 5 weeks of chronic hypoxia in rats. Respir Physiol Neurobiol

135: 1–8, 2003.253. Garcia N, Hopkins SR, Powell FL. Effects of intermittent hyp-

oxia on the isocapnic hypoxic ventilatory response and erythropoi-esis in humans. Respir Physiol 123: 39–49, 2000.

254. Garcia-Fernandez M, Ortega-Saenz P, Castellano A, Lopez-

Barneo J. Mechanisms of low-glucose sensitivity in carotid bodyglomus cells. Diabetes 56: 2893–2900, 2007.

255. Gargaglioni LH, Bicego KC, Branco LG. Brain monoaminergicneurons and ventilatory control in vertebrates. Respir Physiol

Neurobiol 164: 112–122, 2008.256. Gargaglioni LH, Coimbra NC, Branco LG. The nucleus raphe

magnus modulates hypoxia-induced hyperventilation but not ana-pyrexia in rats. Neurosci Lett 347: 121–125, 2003.

257. Garthwaite J. Glutamate, nitric oxide and cell-cell signalling inthe nervous system. Trends Neurosci 14: 60–67, 1991.

258. Gauda EB, Lawson EE. Developmental influences on carotidbody responses to hypoxia. Respir Physiol 121: 199–208, 2000.

259. Gaultier C, Dauger S, Simonneau M, Gallego J. Genes modu-lating chemical breathing control: lessons from mutant animals.Respir Physiol Neurobiol 136: 105–114, 2003.

260. Gautier H. Interactions among metabolic rate, hypoxia, and con-trol of breathing. J Appl Physiol 81: 521–527, 1996.

261. Gautier H, Bonora M. Possible alterations in brain monoaminemetabolism during hypoxia-induced tachypnea in cats. J Appl

Physiol 49: 769–777, 1980.262. Genest SE, Gulemetova R, Laforest S, Drolet G, Kinkead R.

Neonatal maternal separation and sex-specific plasticity of the

hypoxic ventilatory response in awake rat. J Physiol 554: 543–557,2004.

263. Georgopoulos D, Berezanski D, Anthonisen NR. Effects of CO2

breathing on ventilatory response to sustained hypoxia in normaladults. J Appl Physiol 66: 1071–1078, 1989.

264. Georgopoulos D, Berezanski D, Anthonisen NR. Effect of di-chloroacetate on ventilatory response to sustained hypoxia in nor-mal adults. Respir Physiol 82: 115–122, 1990.

265. Georgopoulos D, Holtby SG, Berezanski D, Anthonisen NR.

Aminophylline effects on ventilatory response to hypoxia and hy-peroxia in normal adults. J Appl Physiol 67: 1150–1156, 1989.

266. Georgopoulos D, Walker S, Anthonisen NR. Increased chemo-receptor output and ventilatory response to sustained hypoxia.J Appl Physiol 67: 1157–1163, 1989.

267. Georgopoulos D, Walker S, Anthonisen NR. Effect of sustainedhypoxia on ventilatory response to CO2 in normal adults. J Appl

Physiol 68: 891–896, 1990.268. Gilmartin G, Tamisier R, Anand A, Cunnington D, Weiss JW.

Evidence of impaired hypoxic vasodilation after intermediate-du-ration hypoxic exposure in humans. Am J Physiol Heart Circ

Physiol 291: H2173–H2180, 2006.269. Gilmartin GS, Tamisier R, Curley M, Weiss JW. Ventilatory,

hemodynamic, sympathetic nervous system, and vascular reactiv-ity changes after recurrent nocturnal sustained hypoxia in humans.Am J Physiol Heart Circ Physiol 295: H778–H785, 2008.

270. Giussani DA, Moore PJ, Bennet L, Spencer JA, Hanson MA.

Alpha 1- and alpha 2-adrenoreceptor actions of phentolamine andprazosin on breathing movements in fetal sheep in utero. J Physiol

486: 249–255, 1995.271. Gluckman PD, Johnston BM. Lesions in the upper lateral pons

abolish the hypoxic depression of breathing in unanaesthetizedfetal lambs in utero. J Physiol 382: 373–383, 1987.

272. Godfrey S, Edwards RHT, Copland GM, Gross PL. Chemosen-sitivity in normal subjects, athletes, and patients with chronicairways obstruction. J Appl Physiol 30: 193–199, 1971.

273. Goldberg SV, Schoene RB, Haynor D, Trimble B, Swenson ER,

Morrison JB, Banister EJ. Brain tissue pH and ventilatory accli-matization to high altitude. J Appl Physiol 72: 58–63, 1992.

274. Golder FJ, Zabka AG, Bavis RW, Baker-Herman T, Fuller DD,

Mitchell GS. Differences in time-dependent hypoxic phrenic re-sponses among inbred rat strains. J Appl Physiol 98: 838–844,2005.

275. Gomez-Nino A, Agapito MT, Obeso A, Gonzalez C. Effects ofmitochondrial poisons on glutathione redox potential and carotidbody chemoreceptor activity. Respir Physiol Neurobiol 165: 104–111, 2009.

276. Gonzalez C, Agapito MT, Rocher A, Gonzalez-Martin MC,

Vega-Agapito V, Gomez-Nino A, Rigual R, Castaneda J, Obeso

A. Chemoreception in the context of the general biology of ROS.Respir Physiol Neurobiol 157: 30–44, 2007.

277. Gonzalez C, Almaraz L, Obeso A, Rigual R. Oxygen and acidchemoreception in the carotid body chemoreceptors. Trends Neu-

rosci 15: 146–153, 1992.278. Gonzalez C, Almaraz L, Obeso A, Rigual R. Carotid body che-

moreceptors: from natural stimuli to sensory discharges. Physiol

Rev 74: 829–898, 1994.279. Gonzalez C, Lopez-Lopez JR, Obeso A, Perez-Garcia MT,

Rocher A. Cellular mechanisms of oxygen chemoreception in thecarotid body. Respir Physiol 102: 137–147, 1995.

280. Gonzalez C, Rocher A, Zapata P. Arterial chemoreceptors: cel-lular and molecular mechanisms in the adaptative and homeostaticfunction of the carotid body. Rev Neurol 36: 239–254, 2003.

281. Gonzalez C, Sanz-Alyayate G, Agapito MT, Obeso A. Effects ofreducing agents on glutathione metabolism and the function ofcarotid body chemoreceptor cells. Biol Chem 385: 265–274, 2004.

282. Gonzalez C, Sanz-Alfayate G, Agapito MT, Gomez-Nino A,

Rocher A, Obeso A. Significance of ROS in oxygen sensing in cellsystems with sensitivity to physiological hypoxia. Respir Physiol

Neurobiol 132: 17–41, 2002.283. Gonzalez-Guerrero PR, Rigual R, Gonzalez C. Effects of

chronic hypoxia on opioid peptide and catecholamine levels and onthe release of dopamine in the rabbit carotid body. J Neurochem

60: 1769–1776, 1993.



284. Gourine AV, Llaudet E, Dale N, Spyer KM. Release of ATP inthe ventral medulla during hypoxia in rats: role in hypoxic venti-latory response. J Neurosci 25: 1211–1218, 2005.

285. Gozal D, Gaultier C. Evolving concepts of the maturation ofcentral pathways underlying the hypoxic ventilatory response.Am J Respir Crit Care Med 164: 325–329, 2001.

286. Gozal D, Gozal E. Episodic hypoxia enhances late hypoxic ven-tilation in developing rat: putative role of neuronal NO synthase.Am J Physiol Regul Integr Comp Physiol 276: R17–R22, 1999.

287. Gozal D, Gozal E, Simakajornboon N. Signaling pathways of theacute hypoxic ventilatory response in the nucleus tractus soli-tarius. Respir Physiol 121: 209–221, 2000.

288. Gozal D, Gozal E, Torres JE, Gozal YM, Nuckton TJ, Hornby

PJ. Nitric oxide modulates ventilatory responses to hypoxia in thedeveloping rat. Am J Respir Crit Care Med 155: 1755–1762, 1997.

289. Gozal D, Simakajornboon N, Czapla MA, Xue YD, Gozal E,

Vlasic V, Lasky JA, Liu JY. Brainstem activation of platelet-derived growth factor-beta receptor modulates the late phase of thehypoxic ventilatory response. J Neurochem 74: 310–319, 2000.

290. Gozal D, Torres JE, Gozal YM, Littwin SM. Effect of nitricoxide synthase inhibition on cardiorespiratory responses in theconscious rat. J Appl Physiol 81: 2068–2077, 1996.

291. Granjeiro EM, Machado BH. NO in the caudal NTS modulatesthe increase in respiratory frequency in response to chemoreflexactivation in awake rats. Respir Physiol Neurobiol 166: 32–40,2009.

292. Gray BA. Response of the perfused carotid body to changes in pHand PCO2. Respir Physiol 4: 229–245, 1968.

293. Green GA, Darnall RA, Bierd TB, Adams JM. Effect of aorticballoon inflation on ventilation and brain stem blood flow in piglets.J Appl Physiol 66: 2174–2180, 1989.

294. Greenberg HE, Sica A, Batson D, Scharf SM. Chronic intermit-tent hypoxia increases sympathetic responsiveness to hypoxia andhypercapnia. J Appl Physiol 86: 298–305, 1999.

295. Guzy RD, Schumacker PT. Oxygen sensing by mitochondria atcomplex III: the paradox of increased reactive oxygen speciesduring hypoxia. Exp Physiol 91: 807–819, 2006.

296. Guzy RD, Mack MM, Schumacker PT. Mitochondrial complex IIIis required for hypoxia-induced ROS production and gene tran-scription in yeast. Antioxid Redox Signal 9: 1317–1328, 2007.

297. Haddad GG, Donnelly DF. O2 deprivation induces a major depo-larization in brain stem neurons in the adult but not in the neonatalrat. J Physiol 429: 411–428, 1990.

298. Hall SM, Gorenflo M, Reader J, Lawson D, Haworth SG. Neo-natal pulmonary hypertension prevents reorganisation of the pul-monary arterial smooth muscle cytoskeleton after birth. J Anat

196: 391–403, 2000.299. Hannhart B, Pickett CK, Weil JV, Moore LG. Influence of

pregnancy on ventilatory and carotid body neural output respon-siveness to hypoxia in cats. J Appl Physiol 67: 797–803, 1989.

300. Hansen J, Sander M. Sympathetic neural overactivity in healthyhumans after prolonged exposure to hypobaric hypoxia. J Physiol

546: 921–929, 2003.301. Hanson MA, Kumar P, Williams BA. The effect of chronic hyp-

oxia upon the development of respiratory chemoreflexes in thenewborn kitten. J Physiol 411: 563–574, 1989.

302. Harding R. Fetal pulmonary development: the role of respiratorymovements. Equine Vet J Suppl 32–39, 1997.

303. Hatton CJ, Carpenter E, Pepper DR, Kumar P, Peers C. De-velopmental changes in isolated rat type I carotid body cell K�

currents and their modulation by hypoxia. J Physiol 501: 49–58,1997.

304. Hatton CJ, Peers C. Effects of cytochrome P-450 inhibitors onionic currents in isolated rat type I carotid body cells. Am J Physiol

Cell Physiol 271: C85–C92, 1996.305. Hayashi F, Coles SK, Bach KB, Mitchell GS, McCrimmon DR.

Time-dependent phrenic nerve responses to carotid afferent acti-vation: intact vs. decerebellate rats. Am J Physiol Regul Integr

Comp Physiol 265: R811–R819, 1993.306. He L, Chen J, Dinger B, Sanders K, Sundar K, Hoidal J,

Fidone S. Characteristics of carotid body chemosensitivity inNADPH oxidase-deficient mice. Am J Physiol Cell Physiol 282:C27–C33, 2002.

307. He L, Chen J, Dinger B, Stensaas L, Fidone S. Effect of chronichypoxia on purinergic synaptic transmission in rat carotid body.J Appl Physiol 100: 157–162, 2006.

308. He L, Chen J, Liu X, Dinger B, Fidone S. Enhanced nitricoxide-mediated chemoreceptor inhibition and altered cyclic GMPsignaling in rat carotid body following chronic hypoxia. Am J

Physiol Lung Cell Mol Physiol 293: L1463–L1468, 2007.309. He L, Dinger B, Fidone S. Effect of chronic hypoxia on cholin-

ergic chemotransmission in rat carotid body. J Appl Physiol 98:614–619, 2005.

310. He L, Dinger B, Sanders K, Hoidal J, Obeso A, Stensaas L,

Fidone S, Gonzalez C. Effect of p47phox gene deletion on ROSproduction and oxygen sensing in mouse carotid body chemore-ceptor cells. Am J Physiol Lung Cell Mol Physiol 289: L916–L924,2005.

311. Heath D. The human carotid body in health and disease. J Pathol

164: 1–8, 1991.312. Heath D, Smith P, Fitch R, Harris P. Comparative pathology of

the enlarged carotid body. J Comp Pathol 95: 259–271, 1985.313. Hebbel RP, Kronenberg RS, Eaton JW. Hypoxic ventilatory

response in subjects with normal and high oxygen affinity hemo-globins. J Clin Invest 60: 1211–1215, 1977.

314. Heeringa J, Berkenbosch A, De Goede J, Olievier CN. Relativecontribution of central and peripheral chemoreceptors to the ven-tilatory response to CO2 during hyperoxia. Respir Physiol 37:365–379, 1979.

315. Hempleman SC. Sodium and potassium current in neonatal ratcarotid body cells following chronic in vivo hypoxia. Brain Res

699: 42–50, 1995.316. Hempleman SC. Increased calcium current in carotid body glo-

mus cells following in vivo acclimatization to chronic hypoxia.J Neurophysiol 76: 1880–1886, 1996.

317. Henderson-Smart DJ, Cohen GL. Chemical control of breathingin early life. Ann NY Acad Sci 533: 276–288, 1988.

318. Henderson-Smart DJ, Pettigrew AG, Campbell DJ. Clinicalapnea and brain-stem neural function in preterm infants. N Engl

J Med 308: 353–357, 1983.319. Henderson-Smart DJ, Read DJ. Ventilatory responses to

hypoxaemia during sleep in the newborn. J Dev Physiol 1: 195–208,1979.

320. Herigstad M, Fatemian M, Robbins PA. Respiratory controlduring air-breathing exercise in humans following an 8 h exposureto hypoxia. Respir Physiol Neurobiol 162: 169–175, 2008.

321. Hildebrandt W, Alexander S, Bartsch P, Droge W. Effect ofN-acetyl-cysteine on the hypoxic ventilatory response and erythro-poietin production: linkage between plasma thiol redox state andO2 chemosensitivity. Blood 99: 1552–1555, 2002.

322. Hodges MR, Richerson GB. Contributions of 5-HT neurons torespiratory control: neuromodulatory and trophic effects. Respir

Physiol Neurobiol 164: 222–232, 2008.323. Hodges MR, Richerson GB. Interaction between defects in ven-

tilatory and thermoregulatory control in mice lacking 5-HT neu-rons. Respir Physiol Neurobiol 164: 350–357, 2008.

324. Hodges MR, Tattersall GJ, Harris MB, McEvoy SD, Richerson

DN, Deneris ES, Johnson RL, Chen ZF, Richerson GB. Defectsin breathing and thermoregulation in mice with near-completeabsence of central serotonin neurons. J Neurosci 28: 2495–2505,2008.

325. Honda Y, Hata N, Sakakibara Y, Nishino T, Satomura Y. Cen-tral hypoxic-hypercapnic interaction in mild hypoxia in man.Pflugers Arch 391: 289–295, 1981.

326. Honda Y, Watanabe S, Hashizume I, Satomura Y, Hata N,

Sakakibara Y, Severinghaus JW. Hypoxic chemosensitivity inasthmatic patients two decades after carotid body resection. J Appl

Physiol 46: 632–638, 1979.327. Hoop B, Beagle JL, Maher TJ, Kazemi H. Brainstem amino acid

neurotransmitters and hypoxic ventilatory response. Respir

Physiol 118: 117–129, 1999.328. Hooper SB, Coulter CL, Deayton JM, Harding R, Thorburn

GD. Fetal endocrine responses to prolonged hypoxemia in sheep.Am J Physiol Regul Integr Comp Physiol 259: R703–R708, 1990.



329. Horn EM, Dillon GH, Fan YP, Waldrop TG. Developmentalaspects and mechanisms of rat caudal hypothalamic neuronal re-sponses to hypoxia. J Neurophysiol 81: 1949–1959, 1999.

330. Horn EM, Waldrop TG. Oxygen-sensing neurons in the caudalhypothalamus and their role in cardiorespiratory control. Respir

Physiol 110: 219–228, 1997.331. Horn EM, Waldrop TG. Suprapontine control of respiration. Re-

spir Physiol 114: 201–211, 1998.332. Hornbein TF. The puzzle of periodicity. Pediatrics 50: 183–185,

1972.333. Hornbein TF, Griffo ZJ, Roos A. Quantitation of chemoreceptor

activity: interrelation of hypoxia and hypercapnia. J Neurophysiol

24: 561–568, 1961.334. Hornbein TF, Roos A. Specificity of H ion concentration as a

carotid chemoreceptor stimulus. J Appl Physiol 18: 580–584, 1963.335. Housley GD, Martin-Body RL, Dawson NJ, Sinclair JD. Brain

stem projections of the glossopharyngeal nerve and its carotidsinus branch in the rat. Neuroscience 22: 237–250, 1987.

336. Howard LS, Robbins PA. Problems with determining the hypoxicresponse in humans using stepwise changes in end-tidal PO2. Re-

spir Physiol 98: 241–249, 1994.337. Howard LS, Robbins PA. Alterations in respiratory control during

8 h of isocapnic and poikilocapnic hypoxia in humans. J Appl

Physiol 78: 1098–1107, 1995.338. Howard LS, Robbins PA. Ventilatory response to 8 h of isocapnic

and poikilocapnic hypoxia in humans. J Appl Physiol 78: 1092–1097, 1995.

339. Huang J, Lusina S, Xie T, Ji E, Xiang S, Liu Y, Weiss JW.

Sympathetic response to chemostimulation in conscious rats ex-posed to chronic intermittent hypoxia. Respir Physiol Neurobiol

166: 102–106, 2009.340. Huang J, Suguihara C, Hehre D, Lin J, Bancalari E. Effects of

GABA receptor blockage on the respiratory response to hypoxia insedated newborn piglets. J Appl Physiol 77: 1006–1010, 1994.

341. Huang LE, Arany Z, Livingston DM, Bunn HF. Activation ofhypoxia-inducible transcription factor depends primarily upon re-dox-sensitive stabilization of its alpha subunit. J Biol Chem 271:32253–32259, 1996.

342. Huey KA, Brown IP, Jordan MC, Powell FL. Changes in dopa-mine D2-receptor modulation of the hypoxic ventilatory responsewith chronic hypoxia. Respir Physiol 123: 177–187, 2000.

343. Huey KA, Low MJ, Kelly MA, Juarez R, Szewczak JM, Powell

FL. Ventilatory responses to acute and chronic hypoxia in mice:effects of dopamine D2 receptors. J Appl Physiol 89: 1142–1150,2000.

344. Huey KA, Powell FL. Time-dependent changes in dopamine D2-receptor mRNA in the arterial chemoreflex pathway with chronichypoxia. Brain Res 75: 264–270, 2000.

345. Huey KA, Szewczak JM, Powell FL. Dopaminergic mechanismsof neural plasticity in respiratory control: transgenic approaches.Respir Physiol Neurobiol 135: 133–144, 2003.

346. Hupperets MD, Hopkins SR, Pronk MG, Tiemessen IJ, Garcia

N, Wagner PD, Powell FL. Increased hypoxic ventilatory re-sponse during 8 weeks at 3800 m altitude. Respir Physiol Neuro-

biol 142: 145–152, 2004.347. Ide K, Worthley M, Anderson T, Poulin MJ. Effect of systemic

administration of the nitric oxide synthase inhibitor L-NMMA onthe human ventilatory response to hypoxia. Adv Exp Med Biol 605:41–45, 2008.

348. Insalaco G, Romano S, Salvaggio A, Braghiroli A, Lanfranchi

P, Patruno V, Donner CF, Bonsignore G. Cardiovascular andventilatory response to isocapnic hypoxia at sea level and at 5,050m. J Appl Physiol 80: 1724–1730, 1996.

349. Irsigler GB, Stafford MJ, Severinghaus JW. Relationship ofCSF pH, O2, and CO2 responses in metabolic acidosis and alkalosisin humans. J Appl Physiol 48: 355–361, 1980.

350. Iturriaga R, Alcayaga J. Neurotransmission in the carotid body:transmitters and modulators between glomus cells and petrosalganglion nerve terminals. Brain Res 47: 46–53, 2004.

351. Iturriaga R, Alcayaga J, Zapata P. Dissociation of hypoxia-induced chemosensory responses and catecholamine efflux in catcarotid body superfused in vitro. J Physiol 497: 551–564, 1996.

352. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, Salic A,

Asara JM, Lane WS, Kaelin WG Jr. HIF� targeted for VHL-mediated destruction by proline hydroxylation: implications for O2

sensing. Science 292: 464–468, 2001.353. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell

SJ, Kriegsheim Av Hebestreit HF, Mukherji M, Schofield CJ,

Maxwell PH, Pugh CW, Ratcliffe PJ. Targeting of HIF-alpha tothe von Hippel-Lindau ubiquitylation complex by O2-regulatedprolyl hydroxylation. Science 292: 468–472, 2001.

354. Jackson A, Nurse C. Dopaminergic properties of cultured ratcarotid body chemoreceptors grown in normoxic and hypoxicenvironments. J Neurochem 69: 645–654, 1997.

355. Jackson A, Nurse CA. Role of acetylcholine receptors and dopa-mine transporter in regulation of extracellular dopamine in ratcarotid body cultures grown in chronic hypoxia or nicotine. J Neu-

rochem 70: 653–662, 1998.356. Jacky JP. Barometric measurement of tidal volume: effects of

pattern and nasal temperature. J Appl Physiol 49: 319–325, 1980.357. Jansen AH, Chernick V. Development of respiratory control.

Physiol Rev 63: 437–483, 1983.358. Janssen PL, Dwinell MR, Pizarro J, Bisgard GE. Intracarotid

dopamine infusion does not prevent acclimatization to hypoxia.Respir Physiol 111: 33–43, 1998.

359. Janssen PL, O’Halloran KD, Pizarro J, Dwinell MR, Bisgard

GE. Carotid body dopaminergic mechanisms are functional afteracclimatization to hypoxia in goats. Respir Physiol 111: 25–32,1998.

360. Janssens JP, Pache JC, Nicod LP. Physiological changes inrespiratory function associated with ageing. Eur Respir J 13: 197–205, 1999.

361. Javaheri S, Teppema LJ, Evers JA. Effects of aminophylline onhypoxemia-induced ventilatory depression in the cat. J Appl

Physiol 64: 1837–1843, 1988.362. Jensen D, Wolfe LA, O’Donnell DE, Davies GA. Chemoreflex

control of breathing during wakefulness in healthy men andwomen. J Appl Physiol 98: 822–828, 2005.

363. Johnston BM, Gluckman PD. Peripheral chemoreceptors re-spond to hypoxia in pontine-lesioned fetal lambs in utero. J Appl

Physiol 75: 1027–1034, 1993.364. Jordan AS, Catcheside PG, Orr RS, O’Donoghue FJ, Saunders

NA, McEvoy RD. Ventilatory decline after hypoxia and hypercap-nia is not different between healthy young men and women. J Appl

Physiol 88: 3–9, 2000.365. Jordan W, Cohrs S, Degner D, Meier A, Rodenbeck A, Mayer

G, Pilz J, Ruther E, Kornhuber J, Bleich S. Evaluation ofoxidative stress measurements in obstructive sleep apnea syn-drome. J Neural Transm 113: 239–254, 2006.

366. Joseph V, Soliz J, Pequignot J, Sempore B, Cottet-Emard JM,

Dalmaz Y, Favier R, Spielvogel H, Pequignot JM. Gender dif-ferentiation of the chemoreflex during growth at high altitude:functional and neurochemical studies. Am J Physiol Regul Integr

Comp Physiol 278: R806–R816, 2000.367. Julien C, Bairam A, Joseph V. Chronic intermittent hypoxia

reduces ventilatory long-term facilitation and enhances apnea fre-quency in newborn rats. Am J Physiol Regul Integr Comp Physiol

294: R1356–R1366, 2008.368. Kaab S, Miguel-Velado E, Lopez-Lopez JR, Perez-Garcia MT.

Downregulation of Kv3.4 channels by chronic hypoxia increasesacute oxygen sensitivity in rabbit carotid body. J Physiol 566:395–408, 2005.

369. Kaelin J, Ratcliffe PJ. Oxygen sensing by metazoans: the centralrole of the HIF hydroxylase pathway. Mol Cell 30: 393–402, 2008.

370. Kagawa S, Stafford MJ, Waggener TB, Severinghaus JW. Noeffect of naloxone on hypoxia-induced ventilatory depression inadults. J Appl Physiol 52: 1030–1034, 1982.

371. Kaluz S, Kaluzova M, Stanbridge EJ. Regulation of gene expres-sion by hypoxia: integration of the HIF-transduced hypoxic signalat the hypoxia-responsive element. Clin Chim Acta 395: 6–13,2008.

372. Kara T, Narkiewicz K, Somers VK. Chemoreflexes: physiologyand clinical implications. Acta Physiol Scand 177: 377–384, 2003.



373. Katayama K, Fujita H, Sato K, Ishida K, Iwasaki K, Miyamura

M. Effect of a repeated series of intermittent hypoxic exposures onventilatory response in humans. High Alt Med Biol 6: 50–59, 2005.

374. Katayama K, Sato K, Hotta N, Ishida K, Iwasaki K, Miyamura

M. Intermittent hypoxia does not increase exercise ventilation atsimulated moderate altitude. Int J Sports Med 28: 480–487, 2007.

375. Katayama K, Sato Y, Ishida K, Mori S, Miyamura M. The effectsof intermittent exposure to hypoxia during endurance exercisetraining on the ventilatory responses to hypoxia and hypercapnia inhumans. Eur J Appl Physiol Occup Physiol 78: 189–194, 1998.

376. Katayama K, Sato Y, Morotome Y, Shima N, Ishida K, Mori S,

Miyamura M. Intermittent hypoxia increases ventilation and SaO2during hypoxic exercise and hypoxic chemosensitivity. J Appl

Physiol 90: 1431–1440, 2001.377. Katayama K, Sato Y, Shima N, Qiu JC, Ishida K, Mori S,

Miyamura M. Enhanced chemosensitivity after intermittenthypoxic exposure does not affect exercise ventilation at sea level.Eur J Appl Physiol 87: 187–191, 2002.

378. Kawai Y, Qi J, Comer AM, Gibbons H, Win J, Lipski J. Effectsof cyanide and hypoxia on membrane currents in neurones acutelydissociated from the rostral ventrolateral medulla of the rat. Brain

Res 830: 246–257, 1999.379. Kazemi H, Hoop B. Glutamic acid and gamma-aminobutyric acid

neurotransmitters in central control of breathing. J Appl Physiol

70: 1–7, 1991.380. Kemp PJ. Detecting acute changes in oxygen: will the real sensor

please stand up? Exp Physiol 91: 829–834, 2006.381. Kenneth NS, Rocha S. Regulation of gene expression by hypoxia.

Biochem J 414: 19–29, 2008.382. Khamnei S, Robbins PA. Hypoxic depression of ventilation in

humans: alternative models for the chemoreflexes. Respir Physiol

81: 117–134, 1990.383. Kimura H, Tanaka M, Nagao K, Niijima M, Masuyama S,

Mizoo A, Uruma T, Tatsumi K, Kuriyama T, Masuda A, Koba-

yashi T, Honda Y. A new aspect of the carotid body functioncontrolling hypoxic ventilatory decline in humans. Appl Human

Sci 17: 131–137, 1998.384. Kirby TP, Wraith PK, De Cort SC, Airlie MA, Hill JE, Carson

ER, Flenley DC, Warren PM. Modelling the dynamic ventilatoryresponse to hypoxia in normal subjects. J Theor Biol 166: 135–147,1994.

385. Kita I, Sato-Suzuki I, Oguri M, Arita H. Yawning responsesinduced by local hypoxia in the paraventricular nucleus of the rat.Behav Brain Res 117: 119–126, 2000.

386. Kiwull-Schone H, Kiwull P, Muckenhoff K, Both W. The role ofcarotid chemoreceptors in the regulation of arterial oxygen trans-port under hypoxia with and without hypercapnia. Adv Exp Med

Biol 75: 469–476, 1976.387. Kline DD, Overholt JL, Prabhakar NR. Mutant mice deficient in

NOS-1 exhibit attenuated long-term facilitation and short-term po-tentiation in breathing. J Physiol 539: 309–315, 2002.

388. Kline DD, Peng YJ, Manalo DJ, Semenza GL, Prabhakar NR.

Defective carotid body function and impaired ventilatory re-sponses to chronic hypoxia in mice partially deficient for hypoxia-inducible factor 1 alpha. Proc Natl Acad Sci USA 99: 821–826, 2002.

389. Kline DD, Yang T, Huang PL, Prabhakar NR. Altered respiratoryresponses to hypoxia in mutant mice deficient in neuronal nitricoxide synthase. J Physiol 511: 273–287, 1998.

390. Kline DD, Yang T, Premkumar DR, Thomas AJ, Prabhakar NR.

Blunted respiratory responses to hypoxia in mutant mice deficientin nitric oxide synthase-3. J Appl Physiol 88: 1496–1508, 2000.

391. Kobayashi K, Lemke RP, Greer JJ. Ultrasound measurements offetal breathing movements in the rat. J Appl Physiol 91: 316–320,2001.

392. Kobertz WR, Williams C, Miller C. Hanging gondola structure ofthe T1 domain in a voltage-gated K� channel. Biochemistry 39:10347–10352, 2000.

393. Koehle MS, Sheel AW, Milsom WK, McKenzie DC. Two pat-terns of daily hypoxic exposure and their effects on measures ofchemosensitivity in humans. J Appl Physiol 103: 1973–1978, 2007.

394. Koerner P, Hesslinger C, Schaefermeyer A, Prinz C, Gratzl M.

Evidence for histamine as a transmitter in rat carotid body sensorcells. J Neurochem 91: 493–500, 2004.

395. Kolb JC, Ainslie PN, Ide K, Poulin MJ. Effects of five consec-utive nocturnal hypoxic exposures on the cerebrovascular re-sponses to acute hypoxia and hypercapnia in humans. J Appl

Physiol 96: 1745–1754, 2004.396. Koos BJ, Chao A, Doany W. Adenosine stimulates breathing in

fetal sheep with brain stem section. J Appl Physiol 72: 94–99, 1992.397. Koos BJ, Chau A, Matsuura M, Punla O, Kruger L. Thalamic

locus mediates hypoxic inhibition of breathing in fetal sheep.J Neurophysiol 79: 2383–2393, 1998.

398. Koos BJ, Chau A, Matsuura M, Punla O, Kruger L. Thalamiclesions dissociate breathing inhibition by hypoxia and adenosine infetal sheep. Am J Physiol Regul Integr Comp Physiol 278: R831–R837, 2000.

399. Koos BJ, Kawasaki Y, Hari A, Bohorquez F, Jan C, Roostaeian

J, Wilson CL, Kruger L. Electrical stimulation of the posterome-dial thalamus modulates breathing in unanesthetized fetal sheep.J Appl Physiol 96: 115–123, 2004.

400. Koos BJ, Kawasaki Y, Kim YH, Bohorquez F. Adenosine A2A-receptor blockade abolishes the roll-off respiratory response tohypoxia in awake lambs. Am J Physiol Regul Integr Comp Physiol

288: R1185–R1194, 2005.401. Koos BJ, Kitanaka T, Matsuda K, Gilbert RD, Longo LD. Fetal

breathing adaptation to prolonged hypoxaemia in sheep. J Dev

Physiol 10: 161–166, 1988.402. Koos BJ, Kruger L, Murray TF. Source of extracellular brain

adenosine during hypoxia in fetal sheep. Brain Res 778: 439–442,1997.

403. Koos BJ, Maeda T, Jan C, Lopez G. Adenosine A2A receptorsmediate hypoxic inhibition of fetal breathing in sheep. Am J Obstet

Gynecol 186: 663–668, 2002.404. Koos BJ, Mason BA, Punla O, Adinolfi AM. Hypoxic inhibition

of breathing in fetal sheep: relationship to brain adenosine concen-trations. J Appl Physiol 77: 2734–2739, 1994.

405. Koos BJ, Matsuda K. Fetal breathing, sleep state, and cardiovas-cular responses to adenosine in sheep. J Appl Physiol 68: 489–495,1990.

406. Koos BJ, Sameshima H. Effects of hypoxaemia and hypercapniaon breathing movements and sleep state in sinoaortic-denervatedfetal sheep. J Dev Physiol 10: 131–144, 1988.

407. Koos BJ, Sameshima H, Power GG. Fetal breathing movement,sleep state and cardiovascular responses to an inhibitor of mito-chondrial ATPase in sheep. J Dev Physiol 8: 67–75, 1986.

408. Koos BJ, Sameshima H, Power GG. Fetal breathing, sleep state,and cardiovascular responses to graded anemia in sheep. J Appl

Physiol 63: 1463–1468, 1987.409. Kourie JI. Interaction of reactive oxygen species with ion trans-

port mechanisms. Am J Physiol Cell Physiol 275: C1–C24, 1998.410. Kramer JM, Nolan PC, Waldrop TG. In vitro responses of neu-

rons in the periaqueductal gray to hypoxia and hypercapnia. Brain

Res 835: 197–203, 1999.411. Kronenberg R, Hamilton FN, Gabel R, Hickey R, Read DJ,

Severinghaus J. Comparison of three methods for quantitatingrespiratory response to hypoxia in man. Respir Physiol 16: 109–125, 1972.

412. Kronenberg RS, Drage CW. Attenuation of the ventilatory andheart rate responses to hypoxia and hypercapnia with aging innormal men. J Clin Invest 52: 1812–1819, 1973.

413. Kuipers IM, Maertzdorf WJ, Keunen H, De Jong DS, Hanson

MA, Blanco CE. Effect of maternal hypoxemia on behavior inunanesthetized normoxic or mildly hyperoxic fetal lambs. J Appl

Physiol 76: 2535–2540, 1994.414. Kumar GK, Prabhakar NR. Post-translational modification of

proteins during intermittent hypoxia. Respir Physiol Neurobiol

164: 272–276, 2008.415. Kumar P. Sensing hypoxia in the carotid body: from stimulus to

response. Essays Biochem 43: 43–60, 2007.416. Kumar P, Bin-Jaliah I. Adequate stimuli of the carotid body: more

than an oxygen sensor? Respir Physiol Neurobiol 157: 12–21, 2007.417. Kumar P, Peers C. AMP-activated protein kinase: function and

dysfunction in health and disease. J Physiol 574: 3–6, 2006.418. Kunitomo F, Kimura H, Tatsumi K, Kuriyama T, Watanabe S,

Honda Y. Sex differences in awake ventilatory drive and abnormal



breathing during sleep in eucapnic obesity. Chest 93: 968–976,1988.

419. Kusakabe T, Hirakawa H, Oikawa S, Matsuda H, Kawakami T,

Takenaka T, Hayashida Y. Morphological changes in the ratcarotid body 1, 2, 4, and 8 weeks after the termination of chroni-cally hypocapnic hypoxia. Histol Histopathol 19: 1133–1140, 2004.

420. Kusakabe T, Matsuda H, Hayashida Y. Hypoxic adaptation ofthe rat carotid body. Histol Histopathol 20: 987–997, 2005.

421. Kuwaki T. Orexinergic modulation of breathing across vigilancestates. Respir Physiol Neurobiol 164: 204–212, 2008.

422. Lahiri S, DeLaney RG. Relationship between carotid chemore-ceptor activity and ventilation in the cat. Respir Physiol 24: 267–286, 1975.

423. Lahiri S, DeLaney RG. Stimulus interaction in the responses ofcarotid body chemoreceptor single afferent fibers. Respir Physiol

24: 249–266, 1975.424. Lahiri S, Ehleben W, Acker H. Chemoreceptor discharges and

cytochrome redox changes of the rat carotid body: role of hemeligands. Proc Natl Acad Sci USA 96: 9427–9432, 1999.

425. Lahiri S, Mitchell CH, Reigada D, Roy A, Cherniack NS. Pu-rines, the carotid body and respiration. Respir Physiol Neurobiol

157: 123–129, 2007.426. Lahiri S, Roy A, Baby SM, Hoshi T, Semenza GL, Prabhakar

NR. Oxygen sensing in the body. Prog Biophys Mol Biol 91: 249–286, 2006.

427. Lahiri S, Rumsey WL, Wilson DF, Iturriaga R. Contribution ofin vivo microvascular PO2 in the cat carotid body chemotransduc-tion. J Appl Physiol 75: 1035–1043, 1993.

428. Lahiri S, Smatresk N, Pokorski M, Barnard P, Mokashi A.

Efferent inhibition of carotid body chemoreception in chronicallyhypoxic cats. Am J Physiol Regul Integr Comp Physiol 245: R678–R683, 1983.

429. Lahiri S, Smatresk N, Pokorski M, Barnard P, Mokashi A,

McGregor KH. Dopaminergic efferent inhibition of carotid bodychemoreceptors in chronically hypoxic cats. Am J Physiol Regul

Integr Comp Physiol 247: R24–R28, 1984.430. Lahiri S. Historical perspectives of cellular oxygen sensing and

responses to hypoxia. J Appl Physiol 88: 1467–1473, 2000.431. Lam SY, Fung ML, Leung PS. Regulation of the angiotensin-

converting enzyme activity by a time-course hypoxia in the carotidbody. J Appl Physiol 96: 809–813, 2004.

432. Lam SY, Leung PS. Chronic hypoxia activates a local angiotensin-generating system in rat carotid body. Mol Cell Endocrinol 203:147–153, 2003.

433. Lam SY, Tipoe GL, Liong EC, Fung ML. Chronic hypoxia up-regulates the expression and function of proinflammatory cyto-kines in the rat carotid body. Histochem Cell Biol 130: 549–559,2008.

434. Lam SY, Tipoe GL, Liong EC, Fung ML. Differential expressionsand roles of hypoxia-inducible factor-1alpha, -2alpha and -3alpha inthe rat carotid body during chronic and intermittent hypoxia. His-

tol Histopathol 23: 271–280, 2008.435. LaManna JC, Haxhiu MA, Kutina-Nelson KL, Pundik S,

Erokwu B, Yeh ER, Lust WD, Cherniack NS. Decreased energymetabolism in brain stem during central respiratory depression inresponse to hypoxia. J Appl Physiol 81: 1772–1777, 1996.

436. LaManna JC, Kutina-Nelson KL, Hritz MA, Huang Z, Wong-

Riley MT. Decreased rat brain cytochrome oxidase activity afterprolonged hypoxia. Brain Res 720: 1–6, 1996.

437. Landauer RC, Pepper DR, Kumar P. Effect of chronic hypoxae-mia from birth upon chemosensitivity in the adult rat carotid bodyin vitro. J Physiol 485: 543–550, 1995.

438. Lazarov N, Rozloznik M, Reindl S, Rey-Ares V, Dutschmann

M, Gratzl M. Expression of histamine receptors and effect ofhistamine in the rat carotid body chemoafferent pathway. Eur

J Neurosci 24: 3431–3444, 2006.439. Lee YM, Kim BJ, Chun YS, So I, Choi H, Kim MS, Park JW.

NOX4 as an oxygen sensor to regulate TASK-1 activity. Cell Signal

18: 499–507, 2006.440. Lemus M, Montero S, Cadenas JL, Lara JJ, Tejeda-Chavez

HR, Alvarez-Buylla R, de Alvarez-Buylla ER. GABA-B receptorsactivation in the NTS blocks the glycemic responses induced by

carotid body receptor stimulation. Auton Neurosci 141: 73–82,2008.

441. Lendahl U, Lee KL, Yang H, Poellinger L. Generating specificityand diversity in the transcriptional response to hypoxia. Nat Rev

Genet 10: 821–832, 2009.442. Leonard TO, Lydic R. Pontine nitric oxide modulates acetylcho-

line release, rapid eye movement sleep generation, and respiratoryrate. J Neurosci 17: 774–785, 1997.

443. Leung PS, Fung ML, Tam MS. Renin-angiotensin system in thecarotid body. Int J Biochem Cell Biol 35: 847–854, 2003.

444. Li A, Nattie E. Serotonin transporter knockout mice have a re-duced ventilatory response to hypercapnia predominantly in malesbut not to hypoxia. J Physiol 586: 2321–2329, 2008.

445. Liang PJ, Bascom DA, Robbins PA. Extended models of theventilatory response to sustained isocapnic hypoxia in humans.J Appl Physiol 82: 667–677, 1997.

446. Lin LH, Emson PC, Talman WT. Apposition of neuronal elementscontaining nitric oxide synthase and glutamate in the nucleustractus solitarii of rat: a confocal microscopic analysis. Neuro-

science 96: 341–350, 2000.447. Ling L. Serotonin and NMDA receptors in respiratory long-term

facilitation. Respir Physiol Neurobiol 164: 233–241, 2008.448. Ling L, Fuller DD, Bach KB, Kinkead R, Olson EB Jr, Mitchell

GS. Chronic intermittent hypoxia elicits serotonin-dependent plas-ticity in the central neural control of breathing. J Neurosci 21:5381–5388, 2001.

449. Ling L, Olson EB Jr, Vidruk EH, Mitchell GS. Attenuation of thehypoxic ventilatory response in adult rats following one month ofperinatal hyperoxia. J Physiol 495: 561–571, 1996.

450. Ling L, Olson EB Jr, Vidruk EH, Mitchell GS. Integratedphrenic responses to carotid afferent stimulation in adult ratsfollowing perinatal hyperoxia. J Physiol 500: 787–796, 1997.

451. Lipton AJ, Johnson MA, Macdonald T, Lieberman MW, Gozal

D, Gaston B. S-nitrosothiols signal the ventilatory response tohypoxia. Nature 413: 171–174, 2001.

452. Liu C, Smith TG, Balanos GM, Brooks J, Crosby A, Herigstad

M, Dorrington KL, Robbins PA. Lack of involvement of theautonomic nervous system in early ventilatory and pulmonary vas-cular acclimatization to hypoxia in humans. J Physiol 579: 215–225,2007.

453. Liu J, Boujedaini N, Cazin L, Mallet E, Clabaut M. Develop-mental changes in cardio-respiratory responses to hypoxia andhypercapnia in anesthetized low-birth-weight rats. Respir Physiol

123: 189–199, 2000.454. Liu Q, Lowry TF, Wong-Riley MT. Postnatal changes in ventila-

tion during normoxia and acute hypoxia in the rat: implication fora sensitive period. J Physiol 577: 957–970, 2006.

455. Liu X, He L, Stensaas L, Dinger B, Fidone S. Adaptation tochronic hypoxia involves immune cell invasion and increased ex-pression of inflammatory cytokines in rat carotid body. Am J

Physiol Lung Cell Mol Physiol 296: L158–L166, 2009.456. Liu Y, Ji ES, Xiang S, Tamisier R, Tong J, Huang J, Weiss JW.

Exposure to cyclic intermittent hypoxia increases expression offunctional NMDA receptors in the rat carotid body. J Appl Physiol

106: 259–267, 2009.457. Lloyd BB, Cunningham DJC. A quantitative approach to the

regulation of human respiration. In: The Regulation of Human

Respiration, edited by Cunningham DJC and Lloyd BB. Oxford UK:Blackwell Science, 1963, p. 331–349.

458. Loeschcke HH, Gertz KH. Effect of oxygen pressure in inspiredair on respiratory activity of the human, tested under the constantbehavior of alveolar carbon dioxide pressure. Pflugers Arch 267:460–477, 1958.

459. Long GR, Filuk R, Balakumar M, Easton PA, Anthonisen NR.

Ventilatory response to sustained hypoxia: effect of methysergideand verapamil. Respir Physiol 75: 173–182, 1989.

460. Long W, Anthonisen NR. Dopaminergic influence on the ventila-tory response to sustained hypoxia in the cat. J Appl Physiol 78:1250–1255, 1995.

461. Long W, Lobchuk D, Anthonisen NR. Ventilatory responses toCO2 and hypoxia after sustained hypoxia in awake cats. J Appl

Physiol 76: 2262–2266, 1994.



462. Long WA, Lawson EE. Neurotransmitters and biphasic respira-tory response to hypoxia. J Appl Physiol 57: 213–222, 1984.

463. Long WQ, Anthonisen NR. Aminophylline partially blocks venti-latory depression with hypoxia in the awake cat. Can J Physiol

Pharmacol 72: 673–678, 1994.464. Long WQ, Giesbrecht GG, Anthonisen NR. Ventilatory response

to moderate hypoxia in awake chemodenervated cats. J Appl

Physiol 74: 805–810, 1993.465. Lopez-Barneo J, del Toro R, Levitsky KL, Chiara MD, Ortega-

Saenz P. Regulation of oxygen sensing by ion channels. J Appl

Physiol 96: 1187–1195, 2004.466. Lopez-Barneo J, Lopez-Lopez JR, Urena J, Gonzalez C. Che-

motransduction in the carotid body: K� current modulated by PO2

in type I chemoreceptor cells. Science 241: 580–582, 1988.467. Lopez-Barneo J, Ortega-Saenz P, Pardal R, Pascual A, Piruat

JI. Carotid body oxygen sensing. Eur Respir J 32: 1386–1398, 2008.468. Lopez-Barneo J, Pardal R, Montoro RJ, Smani T, Garcia-

Hirschfeld J, Urena J. K� and Ca2� channel activity and cytosolic[Ca2�] in oxygen-sensing tissues. Respir Physiol 115: 215–227,1999.

469. Lopez-Barneo J, Pardal R, Ortega-Saenz P. Cellular mechanismof oxygen sensing. Annu Rev Physiol 63: 259–287, 2001.

470. Lopez-Lopez JR, Perez-Garcia MT. Oxygen sensitive Kv chan-nels in the carotid body. Respir Physiol Neurobiol 157: 65–74, 2007.

471. Lorier AR, Huxtable AG, Robinson DM, Lipski J, Housley GD,

Funk GD. P2Y1 receptor modulation of the pre-Botzinger complexinspiratory rhythm generating network in vitro. J Neurosci 27:993–1005, 2007.

472. Lovering AT, Dunin-Barkowski WL, Vidruk EH, Orem JM.

Ventilatory response of the cat to hypoxia in sleep and wakeful-ness. J Appl Physiol 95: 545–554, 2003.

473. Lowry TF, Forster HV, Pan LG, Korducki MA, Probst J, Fran-

ciosi RA, Forster M. Effect of carotid body denervation onbreathing in neonatal goats. J Appl Physiol 87: 1026–1034, 1999.

474. Lowry TF, Forster HV, Pan LG, Serra A, Wenninger J, Nash R,

Sheridan D, Franciosi RA. Effects on breathing of carotid bodydenervation in neonatal piglets. J Appl Physiol 87: 2128–2135, 1999.

475. Lugliani R, Whipp BJ, Seard C, Wasserman K. Effect of bilateralcarotid-body resection on ventilatory control at rest and duringexercise in man. N Engl J Med 285: 1105–1111, 1971.

476. Lumbroso D, Joseph V. Impaired acclimatization to chronic hyp-oxia in adult male and female rats following neonatal hypoxia.Am J Physiol Regul Integr Comp Physiol 297: R421–R427, 2009.

477. Lundblad LK, Irvin CG, Adler A, Bates JH. A reevaluation of thevalidity of unrestrained plethysmography in mice. J Appl Physiol

93: 1198–1207, 2002.478. Lusina SJ, Kennedy PM, Inglis JT, McKenzie DC, Ayas NT,

Sheel AW. Long-term intermittent hypoxia increases sympatheticactivity and chemosensitivity during acute hypoxia in humans.J Physiol 575: 961–970, 2006.

479. Lutz PL. Mechanisms for anoxic survival in the vertebrate brain.Annu Rev Physiol 54: 601–618, 1992.

480. Macfarlane PM, Satriotomo I, Windelborn JA, Mitchell GS.

NADPH oxidase activity is necessary for acute intermittenthypoxia-induced phrenic long-term facilitation. J Physiol 587:1931–1942, 2009.

481. Macfarlane PM, Wilkerson JE, Lovett-Barr MR, Mitchell GS.

Reactive oxygen species and respiratory plasticity following inter-mittent hypoxia. Respir Physiol Neurobiol 164: 263–271, 2008.

482. Madden CJ, Morrison SF. Hypoxic activation of arterial chemo-receptors inhibits sympathetic outflow to brown adipose tissue inrats. J Physiol 566: 559–573, 2005.

483. Mahamed S, Cunningham DA, Duffin J. Changes in respiratorycontrol after three hours of isocapnic hypoxia in humans. J Physiol

547: 271–281, 2003.484. Mahamed S, Duffin J. Repeated hypoxic exposures change respi-

ratory chemoreflex control in humans. J Physiol 534: 595–603,2001.

485. Mahutte CK, Rebuck AS. Influence of rate of induction of hyp-oxia on the ventilatory response. J Physiol 284: 219–227, 1978.

486. Malik MT, Peng YJ, Kline DD, Adhikary G, Prabhakar NR.

Impaired ventilatory acclimatization to hypoxia in mice lacking the

immediate early gene fosB. Respir Physiol Neurobiol 145: 23–31,2005.

487. Marchal F, Bairam A, Haouzi P, Crance JP, Di GC, Vert P,

Lahiri S. Carotid chemoreceptor response to natural stimuli in thenewborn kitten. Respir Physiol 87: 183–193, 1992.

488. Martin CB Jr, Murata Y, Petrie RH, Parer JT. Respiratorymovements in fetal rhesus monkeys. Am J Obstet Gynecol 119:939–948, 1974.

489. Martin RJ, Difiore JM, Jana L, Davis RL, Miller MJ, Coles SK,

Dick TE. Persistence of the biphasic ventilatory response to hyp-oxia in preterm infants. J Pediatr 132: 960–964, 1998.

490. Martin RJ, Van LE, Haxhiu MA, Carlo WA. Upper airway mus-cle and diaphragm responses to hypoxia in the piglet. J Appl

Physiol 68: 672–677, 1990.491. Martin-Body RL, Johnston BM. Central origin of the hypoxic

depression of breathing in the newborn. Respir Physiol 71: 25–32,1988.

492. Martin-Body RL, Robson GJ, Sinclair JD. Restoration ofhypoxic respiratory responses in the awake rat after carotid bodydenervation by sinus nerve section. J Physiol 380: 61–73, 1986.

493. Massaro GD, Olivier J, Dzikowski C, Massaro D. Postnataldevelopment of lung alveoli: suppression by 13% O2 and a criticalperiod. Am J Physiol Lung Cell Mol Physiol 258: L321–L327, 1990.

494. Masuda A, Ohyabu Y, Kobayashi T, Yoshino C, Sakakibara Y,

Komatsu T, Honda Y. Lack of positive interaction between CO2

and hypoxic stimulation for PCO2-VAS response slope in humans.Respir Physiol 126: 173–181, 2001.

495. Mateika JH, Fregosi RF. Long-term facilitation of upper airwaymuscle activities in vagotomized and vagally intact cats. J Appl

Physiol 82: 419–425, 1997.496. Matsuda Y, Patrick J, Carmichael L, Fraher L, Richardson B.

Recovery of the ovine fetus from sustained hypoxia: effects onendocrine, cardiovascular, and biophysical activity. Am J Obstet

Gynecol 170: 1433–1441, 1994.497. Matsuoka T, Yoda T, Ushikubo S, Matsuzawa S, Sasano T,

Komiyama A. Repeated acute hypoxia temporarily attenuates theventilatory respiratory response to hypoxia in conscious newbornrats. Pediatr Res 46: 120–125, 1999.

498. Maxova H, Vizek M. Ventilatory response to sustained hypoxia incarotid body denervated rats. Physiol Res 50: 327–331, 2001.

499. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC,

Cockman ME, Wykoff CC, Pugh CW, Maher ER, Ratcliffe PJ.

The tumour suppressor protein VHL targets hypoxia-inducible fac-tors for oxygen-dependent proteolysis. Nature 399: 271–275, 1999.

500. Mazza E Jr, Edelman NH, Neubauer JA. Hypoxic excitation inneurons cultured from the rostral ventrolateral medulla of theneonatal rat. J Appl Physiol 88: 2319–2329, 2000.

501. Mazza E, Thakkar-Varia S, Tozzi CA, Neubauer JA. Expres-sion of heme oxygenase in the oxygen-sensing regions of the rostralventrolateral medulla. J Appl Physiol 91: 379–385, 2001.

502. McCrimmon DR, Dekin MS, Mitchell GS. Glutamate, GABA andserotonin in ventilatory control. In: Regulation of Breathing, editedby Dempsey JA and Pack AI. New York: Dekker, 1995, p. 151–218.

503. McCrimmon DR, Dempsey JA, Olson EB Jr. Effect of catechol-amine depletion on ventilatory control in unanesthetized normoxicand hypoxic rats. J Appl Physiol 55: 522–528, 1983.

504. McGuire M, MacDonald SM, Song G, Poon CS. Phrenic long-term facilitation is robust to hypercapnia and hypocapnia but nothyperventilatory hypotension under PEEP. Respir Physiol Neuro-

biol 158: 107–111, 2007.505. McGuire M, Zhang Y, White DP, Ling L. Chronic intermittent

hypoxia enhances ventilatory long-term facilitation in awake rats.J Appl Physiol 95: 1499–1508, 2003.

506. McGuire M, Zhang Y, White DP, Ling L. Serotonin receptorsubtypes required for ventilatory long-term facilitation and its en-hancement after chronic intermittent hypoxia in awake rats. Am J

Physiol Regul Integr Comp Physiol 286: R334–R341, 2004.507. Melton JE, Chae LO, Neubauer JA, Edelman NH. Extracellular

potassium homeostasis in the cat medulla during progressive brainhypoxia. J Appl Physiol 70: 1477–1482, 1991.

508. Melton JE, Neubauer JA, Edelman NH. GABA antagonism re-verses hypoxic respiratory depression in the cat. J Appl Physiol 69:1296–1301, 1990.



509. Miller MJ, Haxhiu MA, Haxhiu-Poskurica B, Dreshaj IA, Di-

fiore JM, Martin RJ. Recurrent hypoxic exposure and reflexresponses during development in the piglet. Respir Physiol 123:51–61, 2000.

510. Miller MJ, Tenney SM. Hypoxia-induced tachypnea in carotid-deafferented cats. Respir Physiol 23: 31–39, 1975.

511. Miller P, Peers C, Kemp PJ. Polymodal regulation of hTREK1 bypH, arachidonic acid, and hypoxia: physiological impact in acidosisand alkalosis. Am J Physiol Cell Physiol 286: C272–C282, 2004.

512. Millhorn DE. Stimulation of raphe obscurus nucleus causes long-term potentiation of phrenic nerve activity in cat. J Physiol 381:169–179, 1986.

513. Millhorn DE, Eldridge FL, Waldrop TG. Prolonged stimulationof respiration by a new central neural mechanism. Respir Physiol

41: 87–103, 1980.514. Millhorn DE, Eldridge FL, Waldrop TG. Prolonged stimulation

of respiration by endogenous central serotonin. Respir Physiol 42:171–188, 1980.

515. Mitchell GS, Baker TL, Nanda SA, Fuller DD, Zabka AG,

Hodgeman BA, Bavis RW, Mack KJ, Olson EB Jr. Physiologicaland genomic consequences of intermittent hypoxia: invited review:intermittent hypoxia and respiratory plasticity. J Appl Physiol 90:2466–2475, 2001.

516. Mizusawa A, Ogawa H, Kikuchi Y, Hida W, Kurosawa H, Ok-

abe S, Takishima T, Shirato K. In vivo release of glutamate innucleus tractus solitarii of the rat during hypoxia. J Physiol 478:55–66, 1994.

517. Mohan R, Duffin J. The effect of hypoxia on the ventilatoryresponse to carbon dioxide in man. Respir Physiol 108: 101–115,1997.

518. Moon RY, Horne RS, Hauck FR. Sudden infant death syndrome.Lancet 370: 1578–1587, 2007.

519. Moore LG. Comparative human ventilatory adaptation to highaltitude. Respir Physiol 121: 257–276, 2000.

520. Moore LG, Cymerman A, Huang SY, McCullough RE,

McCullough RG, Rock PB, Young A, Young P, Weil JV, Reeves

JT. Propranolol blocks metabolic rate increase but not ventilatoryacclimatization to 4300 m. Respir Physiol 70: 195–204, 1987.

521. Moore LG, McCullough RE, Weil JV. Increased HVR in preg-nancy: relationship to hormonal and metabolic changes. J Appl

Physiol 62: 158–163, 1987.522. Moore PJ, Ackland GL, Hanson MA. Unilateral cooling in the

region of locus coeruleus blocks the fall in respiratory outputduring hypoxia in anaesthetized neonatal sheep. Exp Physiol 81:983–994, 1996.

523. Moore PJ, Parkes MJ, Nijhuis JG, Hanson MA. The incidenceof breathing movements of fetal sheep in normoxia and hypoxiaafter peripheral chemodenervation and brain-stem transection. J

Dev Physiol 11: 147–151, 1989.524. Moorthy SS, Sullivan TY, Fallon JH, Dierdorf SF, Radpour S,

DeAtley RE. Loss of hypoxic ventilatory response following bilat-eral neck dissection. Anesth Analg 76: 791–794, 1993.

525. Moreira TS, Takakura AC, Colombari E, West GH, Guyenet

PG. Inhibitory input from slowly adapting lung stretch receptors toretrotrapezoid nucleus chemoreceptors. J Physiol 580: 285–300,2007.

526. Mortola JP. How newborn mammals cope with hypoxia. Respir

Physiol 116: 95–103, 1999.527. Mortola JP, Morgan CA, Virgona V. Respiratory adaptation to

chronic hypoxia in newborn rats. J Appl Physiol 61: 1329–1336,1986.

528. Mortola JP, Rezzonico R. Metabolic and ventilatory rates innewborn kittens during acute hypoxia. Respir Physiol 73: 55–67,1988.

529. Mortola JP, Rezzonico R, Lanthier C. Ventilation and oxygenconsumption during acute hypoxia in newborn mammals: a com-parative analysis. Respir Physiol 78: 31–43, 1989.

530. Mortola JP, Saiki C. Ventilatory response to hypoxia in rats:gender differences. Respir Physiol 106: 21–34, 1996.

531. Mortola JP. Implications of hypoxic hypometabolism duringmammalian ontogenesis. Respir Physiol Neurobiol 141: 345–356,2004.

532. Mortola JP. Influence of temperature on metabolism and breath-ing during mammalian ontogenesis. Respir Physiol Neurobiol 149:155–164, 2005.

533. Mulligan EM. Discharge properties of carrotid bodies: develop-mental aspects. In: Developmental Neurobiology of Breathing, ed-ited by Haddad GG and Farber JP. New York: Dekker, 1990, p.321–340.

534. Musch TI, Dempsey JA, Smith CA, Mitchell GS, Bateman NT.

Metabolic acids and [H�] regulation in brain tissue during acclima-tization to chronic hypoxia. J Appl Physiol 55: 1486–1495, 1983.

535. Nagyova B, Dorrington KL, Robbins PA. Effect of low-doseenflurane on the ventilatory response to hypoxia in humans. Br J

Anaesth 72: 509–514, 1994.536. Nanduri J, Yuan G, Kumar GK, Semenza GL, Prabhakar NR.

Transcriptional responses to intermittent hypoxia. Respir Physiol

Neurobiol 164: 277–281, 2008.537. Narkiewicz K, van de Borne PJ, Pesek CA, Dyken ME, Mon-

tano N, Somers VK. Selective potentiation of peripheral chemore-flex sensitivity in obstructive sleep apnea. Circulation 99: 1183–1189, 1999.

538. Natsui T, Kuwana S. Central hypoxic depression of respirationdetermined by perfusing a recipient cat’s carotid bodies with bloodfrom a donor cat. Jpn J Physiol 44: 561–574, 1994.

539. Nattie E, Li A. Muscimol dialysis into the caudal aspect of thenucleus tractus solitarii of conscious rats inhibits chemoreception.Respir Physiol Neurobiol 164: 394–400, 2008.

540. Neubauer JA, Melton JE, Edelman NH. Modulation of respira-tion during brain hypoxia. J Appl Physiol 68: 441–451, 1990.

541. Neubauer JA, Posner MA, Santiago TV, Edelman NH. Nalox-one reduces ventilatory depression of brain hypoxia. J Appl

Physiol 63: 699–706, 1987.542. Neubauer JA, Santiago TV, Posner MA, Edelman NH. Ventral

medullary pH and ventilatory responses to hyperperfusion andhypoxia. J Appl Physiol 58: 1659–1668, 1985.

543. Neubauer JA, Simone A, Edelman NH. Role of brain lacticacidosis in hypoxic depression of respiration. J Appl Physiol 65:1324–1331, 1988.

544. Neubauer JA, Sunderram J. Oxygen-sensing neurons in the cen-tral nervous system. J Appl Physiol 96: 367–374, 2004.

545. Nieber K, Sevcik J, Illes P. Hypoxic changes in rat locus coer-uleus neurons in vitro. J Physiol 486: 33–46, 1995.

546. Nielsen AM, Bisgard GE, Vidruk EH. Carotid chemoreceptoractivity during acute and sustained hypoxia in goats. J Appl Physiol

65: 1796–1802, 1988.547. Nielsen M, Smith H. Studies on the regulation of respiration in

acute hypoxia: with an appendix on respiratory control duringprolonged hypoxia. Acta Physiol Scand 24: 293–313, 1952.

548. Nieto FJ, Young TB, Lind BK, Shahar E, Samet JM, Redline S,

D’Agostino RB, Newman AB, Lebowitz MD, Pickering TG.

Association of sleep-disordered breathing, sleep apnea, and hyper-tension in a large community-based study. Sleep Heart HealthStudy. JAMA 283: 1829–1836, 2000.

549. Nieuwenhuijs D, Sarton E, Teppema L, Dahan A. Propofol formonitored anesthesia care: implications on hypoxic control ofcardiorespiratory responses. Anesthesiology 92: 46–54, 2000.

550. Nieuwenhuijs D, Sarton E, Teppema LJ, Kruyt E, Olievier I,

van Kleef J, Dahan A. Respiratory sites of action of propofol:absence of depression of peripheral chemoreflex loop by low-dosepropofol. Anesthesiology 95: 889–895, 2001.

551. Nitsos I, Walker DW. Characterization of pontine neurons whichrespond to hypoxia in fetal sheep. Neurosci Lett 266: 33–36, 1999.

552. Nock ML, Difiore JM, Arko MK, Martin RJ. Relationship of theventilatory response to hypoxia with neonatal apnea in preterminfants. J Pediatr 144: 291–295, 2004.

553. Nolan PC, Waldrop TG. In vivo and in vitro responses of neuronsin the ventrolateral medulla to hypoxia. Brain Res 630: 101–114,1993.

554. Nolan PC, Waldrop TG. In vitro responses of VLM neurons tohypoxia after normobaric hypoxic acclimatization. Respir Physiol

105: 23–33, 1996.555. Nsegbe E, Wallen-Mackenzie A, Dauger S, Roux JC, Shvarev

Y, Lagercrantz H, Perlmann T, Herlenius E. Congenital hy-



poventilation and impaired hypoxic response in Nurr1 mutantmice. J Physiol 556: 43–59, 2004.

556. Nucci TB, Branco LG, Gargaglioni LH. Nitric oxide pathway inthe nucleus raphe magnus modulates hypoxic ventilatory responsebut not anapyrexia in rats. Brain Res 1017: 39–45, 2004.

557. Nucci TB, Branco LG, Gargaglioni LH. 5-HT1A, but not 5-HT2 and5-HT7, receptors in the nucleus raphe magnus modulate hypoxia-induced hyperpnoea. Acta Physiol Oxf 193: 403–414, 2008.

558. Nurse CA. Neurotransmission and neuromodulation in the chemo-sensory carotid body. Auton Neurosci 120: 1–9, 2005.

559. Nurse CA, Zhang M. Synaptic mechanisms during re-innervationof rat arterial chemoreceptors in co-culture. Comp Biochem

Physiol A Mol Integr Physiol 130: 241–251, 2001.560. Obeso A, Almaraz L, Gonzalez C. Effects of 2-deoxy-D-glucose

on in vitro cat carotid body. Brain Res 371: 25–36, 1986.561. Obeso A, Gomez-Nino A, Gonzalez C. NADPH oxidase inhibition

does not interfere with low PO2 transduction in rat and rabbit CBchemoreceptor cells. Am J Physiol Cell Physiol 276: C593–C601,1999.

562. Ogawa H, Mizusawa A, Kikuchi Y, Hida W, Miki H, Shirato K.

Nitric oxide as a retrograde messenger in the nucleus tractussolitarii of rats during hypoxia. J Physiol 486: 495–504, 1995.

563. Ogier M, Katz DM. Breathing dysfunction in Rett syndrome:understanding epigenetic regulation of the respiratory network.Respir Physiol Neurobiol 164: 55–63, 2008.

564. Ohtake PJ, Simakajornboon N, Fehniger MD, Xue YD, Gozal

D. N-methyl-D-aspartate receptor expression in the nucleus tractussolitarii and maturation of hypoxic ventilatory response in the rat.Am J Respir Crit Care Med 162: 1140–1147, 2000.

565. Ohtake PJ, Torres JE, Gozal YM, Graff GR, Gozal D. NMDAreceptors mediate peripheral chemoreceptor afferent input in theconscious rat. J Appl Physiol 84: 853–861, 1998.

566. Okubo S, Mortola JP. Long-term respiratory effects of neonatalhypoxia in the rat. J Appl Physiol 64: 952–958, 1988.

567. Okubo S, Mortola JP. Control of ventilation in adult rats hypoxicin the neonatal period. Am J Physiol Regul Integr Comp Physiol

259: R836–R841, 1990.568. Olson EB Jr, Bohne CJ, Dwinell MR, Podolsky A, Vidruk EH,

Fuller DD, Powell FL, Mitchel GS. Ventilatory long-term facili-tation in unanesthetized rats. J Appl Physiol 91: 709–716, 2001.

569. Olson EB Jr, Dempsey JA. Rat as a model for humanlike venti-latory adaptation to chronic hypoxia. J Appl Physiol 44: 763–769,1978.

570. Olson EB Jr, Vidruk EH, Dempsey JA. Carotid body excisionsignificantly changes ventilatory control in awake rats. J Appl

Physiol 64: 666–671, 1988.571. Olson EB Jr Vidruk EH, McCrimmon DR, Dempsey JA. Mono-

amine neurotransmitter metabolism during acclimatization to hyp-oxia in rats. Respir Physiol 54: 79–96, 1983.

572. Olson LG, Saunders NA. Effect of a dopamine antagonist onventilation during sustained hypoxia in mice. J Appl Physiol 62:1222–1226, 1987.

573. Ortega-Saenz P, Pardal R, Garcia-Fernandez M, Lopez-Bar-

neo J. Rotenone selectively occludes sensitivity to hypoxia in ratcarotid body glomus cells. J Physiol 548: 789–800, 2003.

574. Ortega-Saenz P, Pascual A, Gomez-Diaz R, Lopez-Barneo J.

Acute oxygen sensing in heme oxygenase-2 null mice. J Gen

Physiol 128: 405–411, 2006.575. Ou LC, Tenney SM. Hypoxia and carbon dioxide as separate and

interactive depressants of ventilation. Respir Physiol 28: 347–358,1976.

576. Ou LC, Tenney SM. The role of brief hypocapnia in the ventilatoryresponse to CO2 with hypoxia. Respir Physiol 28: 333–346, 1976.

577. Painter R, Khamnei S, Robbins P. A mathematical model of thehuman ventilatory response to isocapnic hypoxia. J Appl Physiol

74: 2007–2015, 1993.578. Palmer LA, Doctor A, Chhabra P, Sheram ML, Laubach VE,

Karlinsey MZ, Forbes MS, Macdonald T, Gaston B. S-nitroso-thiols signal hypoxia-mimetic vascular pathology. J Clin Invest

117: 2592–2601, 2007.579. Pan LG, Forster HV, Martino P, Strecker PJ, Beales J, Serra

A, Lowry TF, Forster MM, Forster AL. Important role of carotid

afferents in control of breathing. J Appl Physiol 85: 1299–1306,1998.

580. Pardal R, Lopez-Barneo J. Low glucose-sensing cells in the ca-rotid body. Nat Neurosci 5: 197–198, 2002.

581. Pardal R, Ortega-Saenz P, Duran R, Lopez-Barneo J. Glia-likestem cells sustain physiologic neurogenesis in the adult mamma-lian carotid body. Cell 131: 364–377, 2007.

582. Park SJ, Chun YS, Park KS, Kim SJ, Choi SO, Kim HL, Park

JW. Identification of subdomains in NADPH oxidase-4 NOX-4 crit-ical for the oxygen-dependent regulation of TASK-1 K� channels.Am J Physiol Cell Physiol 2009.

583. Pascual O, Morin-Surun MP, Barna B, avit-Saubie M,

Pequignot JM, Champagnat J. Progesterone reverses the neuro-nal responses to hypoxia in rat nucleus tractus solitarius in vitro.J Physiol 544: 511–520, 2002.

584. Paulding WR, Schnell PO, Bauer AL, Striet JB, Nash JA,

Kuznetsova AV, Czyzyk-Krzeska MF. Regulation of gene ex-pression for neurotransmitters during adaptation to hypoxia inoxygen-sensitive neuroendocrine cells. Microsc Res Tech 59: 178–187, 2002.

585. Pawar A, Peng YJ, Jacono FJ, Prabhakar NR. Comparativeanalysis of neonatal and adult rat carotid body responses tochronic intermittent hypoxia. J Appl Physiol 104: 1287–1294, 2008.

586. Pawar A, Nanduri J, Yuan G, Khan SA, Wang N, Kumar GK,

Prabhakar NR. Reactive oxygen species-dependent endothelinsignaling is required for augmented hypoxic sensory response ofthe neonatal carotid body by intermittent hypoxia. Am J Physiol

Regul Integr Comp Physiol 296: R735–R742, 2009.587. Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates.

San Diego, CA: Academic, 1997.588. Pearce W. Hypoxic regulation of the fetal cerebral circulation.

J Appl Physiol 100: 731–738, 2006.589. Pedersen ME, Dorrington KL, Robbins PA. Effects of haloper-

idol on ventilation during isocapnic hypoxia in humans. J Appl

Physiol 83: 1110–1115, 1997.590. Pedersen ME, Dorrington KL, Robbins PA. Effects of dopamine

and domperidone on ventilatory sensitivity to hypoxia after 8 h ofisocapnic hypoxia. J Appl Physiol 86: 222–229, 1999.

591. Pedersen ME, Fatemian M, Robbins PA. Identification of fastand slow ventilatory responses to carbon dioxide under hypoxicand hyperoxic conditions in humans. J Physiol 521: 273–287, 1999.

592. Peers C. Interactions of chemostimuli at the single cell level:studies in a model system. Exp Physiol 89: 60–65, 2004.

593. Peers C, Buckler KJ. Transduction of chemostimuli by the type Icarotid body cell. J Membr Biol 144: 1–9, 1995.

594. Peers C, Wyatt CN. The role of maxiK channels in carotid bodychemotransduction. Respir Physiol Neurobiol 157: 75–82, 2007.

595. Peng Y, Yuan G, Overholt JL, Kumar GK, Prabhakar NR.

Systemic and cellular responses to intermittent hypoxia: evidencefor oxidative stress and mitochondrial dysfunction. Adv Exp Med

Biol 536: 559–564, 2003.596. Peng YJ, Overholt JL, Kline D, Kumar GK, Prabhakar NR.

Induction of sensory long-term facilitation in the carotid body byintermittent hypoxia: implications for recurrent apneas. Proc Natl

Acad Sci USA 100: 10073–10078, 2003.597. Peng YJ, Prabhakar NR. Reactive oxygen species in the plasticity

of respiratory behavior elicited by chronic intermittent hypoxia.J Appl Physiol 94: 2342–2349, 2003.

598. Peng YJ, Prabhakar NR. Effect of two paradigms of chronicintermittent hypoxia on carotid body sensory activity. J Appl

Physiol 96: 1236–1242, 2004.599. Peng YJ, Rennison J, Prabhakar NR. Intermittent hypoxia aug-

ments carotid body and ventilatory response to hypoxia in neonatalrat pups. J Appl Physiol 97: 2020–2025, 2004.

600. Peng YJ, Yuan G, Ramakrishnan D, Sharma SD, Bosch-Marce

M, Kumar GK, Semenza GL, Prabhakar NR. Heterozygous HIF-1alpha deficiency impairs carotid body-mediated systemic re-sponses and reactive oxygen species generation in mice exposed tointermittent hypoxia. J Physiol 577: 705–716, 2006.

601. Pepper DR, Landauer RC, Kumar P. Postnatal development ofCO2-O2 interaction in the rat carotid body in vitro. J Physiol 485:531–541, 1995.



602. Perez H, Ruiz S. Medullary responses to chemoreceptor activa-tion are inhibited by locus coeruleus and nucleus raphe magnus.Neuroreport 6: 1373–1376, 1995.

603. Perez-Garcia MT, Lopez-Lopez JR, Gonzalez C. Kvbeta1.2 sub-unit coexpression in HEK293 cells confers O2 sensitivity to Kv4.2but not to Shaker channels. J Gen Physiol 113: 897–907, 1999.

604. Peterson DD, Pack AI, Silage DA, Fishman AP. Effects of agingon ventilatory and occlusion pressure responses to hypoxia andhypercapnia. Am Rev Respir Dis 124: 387–391, 1981.

605. Pialoux V, Hanly PJ, Foster GE, Brugniaux JV, Beaudin AE,

Hartmann SE, Pun M, Duggan CT, Poulin MJ. Effects of expo-sure to intermittent hypoxia on oxidative stress and acute hypoxicventilatory response in humans. Am J Respir Crit Care Med 180:1002–1009, 2009.

606. Piruat JI, Lopez-Barneo J. Oxygen tension regulates mitochon-drial DNA-encoded complex I gene expression. J Biol Chem 280:42676–42684, 2005.

607. Piruat JI, Pintado CO, Ortega-Saenz P, Roche M, Lopez-Bar-

neo J. The mitochondrial SDHD gene is required for early embry-ogenesis, and its partial deficiency results in persistent carotidbody glomus cell activation with full responsiveness to hypoxia.Mol Cell Biol 24: 10933–10940, 2004.

608. Pokorski M, Marczak M. Ascorbic acid enhances hypoxic venti-latory reactivity in elderly subjects. J Int Med Res 31: 448–457,2003.

609. Pokorski M, Marczak M. Ventilatory response to hypoxia inelderly women. Ann Hum Biol 30: 53–64, 2003.

610. Pokorski M, Walski M, Dymecka A, Marczak M. The agingcarotid body. J Physiol Pharmacol 55 Suppl 3: 107–113, 2004.

611. Porzionato A, Macchi V, Parenti A, Matturri L, De Caro R.

Peripheral chemoreceptors: postnatal development and cytochem-ical findings in Sudden Infant Death Syndrome. Histol Histopathol

23: 351–365, 2008.612. Powell FL. Lake Louise consensus methods for measuring the

hypoxic ventilatory response. Adv Exp Med Biol 588: 271–276,2006.

613. Powell FL, Dwinell MR, Aaron EA. Measuring ventilatory accli-matization to hypoxia: comparative aspects. Respir Physiol 122:271–284, 2000.

614. Powell FL, Fu Z. HIF-1 and ventilatory acclimatization to chronichypoxia. Respir Physiol Neurobiol 164: 282–287, 2008.

615. Powell FL, Huey KA, Dwinell MR. Central nervous system mech-anisms of ventilatory acclimatization to hypoxia. Respir Physiol

121: 223–236, 2000.616. Prabhakar NR, Dinerman JL, Agani FH, Snyder SH. Carbon

monoxide: a role in carotid body chemoreception. Proc Natl Acad

Sci USA 92: 1994–1997, 1995.617. Prabhakar NR, Peng YJ, Kumar GK, Pawar A. Altered carotid

body function by intermittent hypoxia in neonates and adults:relevance to recurrent apneas. Respir Physiol Neurobiol 157: 148–153, 2007.

618. Prabhakar NR, Pieramici SF, Premkumar DR, Kumar GK,

Kalaria RN. Activation of nitric oxide synthase gene expression byhypoxia in central and peripheral neurons. Brain Res 43: 341–346,1996.

619. Prabhakar NR. NO and CO as second messengers in oxygensensing in the carotid body. Respir Physiol 115: 161–168, 1999.

620. Prabhakar NR. Oxygen sensing by the carotid body chemorecep-tors. J Appl Physiol 88: 2287–2295, 2000.

621. Prabhakar NR. O2 sensing at the mammalian carotid body: whymultiple O2 sensors and multiple transmitters? Exp Physiol 91:17–23, 2006.

622. Prieto-Lloret J, Caceres AI, Rigual R, Obeso A, Rocher A,

Bustamante R, Castaneda J, Lopez-Lopez JR, Perez-Garcia T,

Agapito T, Gonzalez C. Effects of perinatal hyperoxia on carotidbody chemoreceptor activity in vitro. Adv Exp Med Biol 536: 517–524, 2003.

623. Querido JS, Godwin JB, Sheel AW. Intermittent hypoxia re-duces cerebrovascular sensitivity to isocapnic hypoxia in humans.Respir Physiol Neurobiol 161: 1–9, 2008.

624. Read DJ. A clinical method for assessing the ventilatory responseto carbon dioxide. Australas Ann Med 16: 20–32, 1967.

625. Read DJ, Henderson-Smart DJ. Regulation of breathing in thenewborn during different behavioral states. Annu Rev Physiol 46:675–685, 1984.

626. Rebuck AS, Campbell EJ. A clinical method for assessing theventilatory response to hypoxia. Am Rev Respir Dis 109: 345–350,1974.

627. Rebuck AS, Kangalee M, Pengelly LD, Campbell EJ. Correla-tion of ventilatory responses to hypoxia and hypercapnia. J Appl

Physiol 35: 173–177, 1973.628. Rebuck AS, Slutsky AS. Measurement of the ventilatory response

to hypercapnia and hypoxia. In: Reulation of Breathing, Part II,edited by Hornbein TF. New York: Dekker, 1981, p. 745–772.

629. Reeves JT, McCullough RE, Moore LG, Cymerman A, Weil JV.

Sea-level PCO2 relates to ventilatory acclimatization at 4,300 m.J Appl Physiol 75: 1117–1122, 1993.

630. Reeves SR, Gozal D. Platelet-activating factor receptor modulatesrespiratory adaptation to long-term intermittent hypoxia in mice.Am J Physiol Regul Integr Comp Physiol 287: R369–R374, 2004.

631. Reeves SR, Gozal D. Respiratory and metabolic responses toearly postnatal chronic intermittent hypoxia and sustained hypoxiain the developing rat. Pediatr Res 60: 680–686, 2006.

632. Reeves SR, Gozal D. Developmental plasticity of respiratory con-trol following intermittent hypoxia. Respir Physiol Neurobiol 149:301–311, 2005.

633. Reeves SR, Gozal D. Changes in ventilatory adaptations associ-ated with long-term intermittent hypoxia across the age spectrumin the rat. Respir Physiol Neurobiol 150: 135–143, 2006.

634. Reeves SR, Gozal E, Guo SZ, Sachleben LR Jr, Brittian KR,

Lipton AJ, Gozal D. Effect of long-term intermittent and sus-tained hypoxia on hypoxic ventilatory and metabolic responses inthe adult rat. J Appl Physiol 95: 1767–1774, 2003.

635. Reeves SR, Mitchell GS, Gozal D. Early postnatal chronic inter-mittent hypoxia modifies hypoxic respiratory responses and long-term phrenic facilitation in adult rats. Am J Physiol Regul Integr

Comp Physiol 290: R1664–R1671, 2006.636. Regensteiner JG, Woodard WD, Hagerman DD, Weil JV, Pick-

ett CK, Bender PR, Moore LG. Combined effects of femalehormones and metabolic rate on ventilatory drives in women.J Appl Physiol 66: 808–813, 1989.

637. Rehan V, Haider AZ, Alvaro RE, Nowaczyk B, Cates DB, Kwi-

atkowski K, Rigatto H. The biphasic ventilatory response tohypoxia in preterm infants is not due to a decrease in metabolism.Pediatr Pulmonol 22: 287–294, 1996.

638. Reid SG, Powell FL. Effects of chronic hypoxia on MK-801-induced changes in the acute hypoxic ventilatory response. J Appl

Physiol 99: 2108–2114, 2005.639. Ren X, Dorrington KL, Maxwell PH, Robbins PA. Effects of

desferrioxamine on serum erythropoietin and ventilatory sensitiv-ity to hypoxia in humans. J Appl Physiol 89: 680–686, 2000.

640. Ren X, Dorrington KL, Robbins PA. Respiratory control in hu-mans after 8 h of lowered arterial PO2, hemodilution, or carboxy-hemoglobinemia. J Appl Physiol 90: 1189–1195, 2001.

641. Ren X, Robbins PA. Ventilatory responses to hypercapnia andhypoxia after 6 h passive hyperventilation in humans. J Physiol

514: 885–894, 1999.642. Rey S, Corthorn J, Chacon C, Iturriaga R. Expression and

immunolocalization of endothelin peptides and its receptors, ETA

and ETB, in the carotid body exposed to chronic intermittenthypoxia. J Histochem Cytochem 55: 167–174, 2007.

643. Rey S, Del Rio R, Iturriaga R. Contribution of endothelin-1 to theenhanced carotid body chemosensory responses induced bychronic intermittent hypoxia. Brain Res 1086: 152–159, 2006.

644. Reyes EP, Fernandez R, Larrain C, Zapata P. Carotid bodychemosensory activity and ventilatory chemoreflexes in cats per-sist after combined cholinergic-purinergic block. Respir Physiol

Neurobiol 156: 23–32, 2007.645. Reyes EP, Fernandez R, Larrain C, Zapata P. Effects of com-

bined cholinergic-purinergic block upon cat carotid body chemo-receptors in vitro. Respir Physiol Neurobiol 156: 17–22, 2007.

646. Rezzonico R, Mortola JP. Respiratory adaptation to chronic hy-percapnia in newborn rats. J Appl Physiol 67: 311–315, 1989.

647. Richardson BS, Carmichael L, Homan J, Patrick JE. Electro-cortical activity, electroocular activity, and breathing movements



in fetal sheep with prolonged and graded hypoxemia. Am J Obstet

Gynecol 167: 553–558, 1992.648. Richardson HL, Parslow PM, Walker AM, Harding R, Horne

RS. Maturation of the initial ventilatory response to hypoxia insleeping infants. J Sleep Res 16: 117–127, 2007.

649. Richter DW, Schmidt-Garcon P, Pierrefiche O, Bischoff AM,

Lalley PM. Neurotransmitters and neuromodulators controllingthe hypoxic respiratory response in anaesthetized cats. J Physiol

514: 567–578, 1999.650. Riedstra JW. Influence of central and peripheral PCO2 pH on the

ventilatory response to hypoxic chemoreceptor stimulation. Acta

Physiol Pharmacol Neerl 12: 407–452, 1963.651. Rigatto H. Control of ventilation in the newborn. Annu Rev

Physiol 46: 661–674, 1984.652. Rigatto H, Brady JP. Periodic breathing and apnea in preterm

infants. II. Hypoxia as a primary event. Pediatrics 50: 219–228,1972.

653. Rigatto H, Wiebe C, Rigatto C, Lee DS, Cates D. Ventilatoryresponse to hypoxia in unanesthetized newborn kittens. J Appl

Physiol 64: 2544–2551, 1988.654. Ritchie K. The fetal response to changes in the composition of

maternal inspired air in human pregnancy. Semin Perinatol 4:295–299, 1980.

655. Robbins PA. The ventilatory response of the human respiratorysystem to sine waves of alveolar carbon dioxide and hypoxia.J Physiol 350: 461–474, 1984.

656. Robbins PA. Evidence for interaction between the contributionsto ventilation from the central and peripheral chemoreceptors inman. J Physiol 401: 503–518, 1988.

657. Robbins PA. Hypoxic ventilatory decline: site of action. J Appl

Physiol 79: 373–374, 1995.658. Robbins PA, Zhang S. Ventilatory response to hypoxia in humans.

Respir Physiol 115: 333–335, 1999.659. Robbins PA. Role of the peripheral chemoreflex in the early stages

of ventilatory acclimatization to altitude. Respir Physiol Neurobiol

158: 237–242, 2007.660. Roberts CA, Corfield DR, Murphy K, Calder NA, Hanson MA,

Adams L, Guz A. Modulation by “central” PCO2 of the response tocarotid body stimulation in man. Respir Physiol 102: 149–161,1995.

661. Rocha S. Gene regulation under low oxygen: holding your breathfor transcription. Trends Biochem Sci 32: 389–397, 2007.

662. Rocher A, Geijo-Barrientos E, Caceres AI, Rigual R, Gonzalez

C, Almaraz L. Role of voltage-dependent calcium channels instimulus-secretion coupling in rabbit carotid body chemoreceptorcells. J Physiol 562: 407–420, 2005.

663. Rodman JR, Curran AK, Henderson KS, Dempsey JA, Smith

CA. Carotid body denervation in dogs: eupnea and the ventilatoryresponse to hyperoxic hypercapnia. J Appl Physiol 91: 328–335,2001.

664. Roux JC, Pequignot JM, Dumas S, Pascual O, Ghilini G,

Pequignot J, Mallet J, Denavit-Saubi M. O2-sensing after ca-rotid chemodenervation: hypoxic ventilatory responsiveness andupregulation of tyrosine hydroxylase mRNA in brainstem cat-echolaminergic cells. Eur J Neurosci 12: 3181–3190, 2000.

665. Roux JC, Peyronnet J, Pascual O, Dalmaz Y, Pequignot JM.

Ventilatory and central neurochemical reorganisation of O2 che-moreflex after carotid sinus nerve transection in rat. J Physiol 522:493–501, 2000.

666. Roy A, Mokashi A, Rozanov C, Daudu PA, Lahiri S. Reducedglutathione, dithiothreitol and cytochrome P-450 inhibitors do notinfluence hypoxic chemosensory responses in the rat carotid body.Brain Res 889: 131–137, 2001.

667. Roy A, Rozanov C, Mokashi A, Daudu P, Al mehdi AB, Shams

H, Lahiri S. Mice lacking in gp91 phox subunit of NADPH oxidaseshowed glomus cell [Ca2�]i and respiratory responses to hypoxia.Brain Res 872: 188–193, 2000.

668. Roy A, Rozanov C, Mokashi A, Lahiri S. PO2-PCO2 stimulusinteraction in [Ca2�]i and CSN activity in the adult rat carotid body.Respir Physiol 122: 15–26, 2000.

669. Rubin AE, Polotsky VY, Balbir A, Krishnan JA, Schwartz AR,

Smith PL, Fitzgerald RS, Tankersley CG, Shirahata M,

O’Donnell CP. Differences in sleep-induced hypoxia between A/J

and DBA/2J mouse strains. Am J Respir Crit Care Med 168: 1520–1527, 2003.

670. Rurak DW, Gruber NC. Increased oxygen consumption associ-ated with breathing activity in fetal lambs. J Appl Physiol 54:701–707, 1983.

671. Ryan JW, Waldrop TG. Hypoxia sensitive neurons in the caudalhypothalamus project to the periaqueductal gray. Respir Physiol

100: 185–194, 1995.672. Ryan ML, Hedrick MS, Pizarro J, Bisgard GE. Effects of carotid

body sympathetic denervation on ventilatory acclimatization tohypoxia in the goat. Respir Physiol 99: 215–224, 1995.

673. Saha S, Batten TF, McWilliam PN. Glutamate, gamma-aminobu-tyric acid and tachykinin-immunoreactive synapses in the cat nu-cleus tractus solitarii. J Neurocytol 24: 55–74, 1995.

674. Sahn SA, Zwillich CW, Dick N, McCullough RE, Lakshmi-

narayan S, Weil JV. Variability of ventilatory responses to hyp-oxia and hypercapnia. J Appl Physiol 43: 1019–1025, 1977.

675. Saiki C, Matsuoka T, Mortola JP. Metabolic-ventilatory interac-tion in conscious rats: effect of hypoxia and ambient temperature.J Appl Physiol 76: 1594–1599, 1994.

676. Saiki C, Mortola JP. Ventilatory control in infant rats after dailyepisodes of anoxia. Pediatr Res 35: 490–493, 1994.

677. Sajkov D, Neill A, Saunders NA, McEvoy RD. Comparison ofeffects of sustained isocapnic hypoxia on ventilation in men andwomen. J Appl Physiol 83: 599–607, 1997.

678. Sanders KA, Sundar KM, He L, Dinger B, Fidone S, Hoidal JR.

Role of components of the phagocytic NADPH oxidase in oxygensensing. J Appl Physiol 93: 1357–1364, 2002.

679. Sanz-Alfayate G, Obeso A, Agapito MT, Gonzalez C. Reducedto oxidized glutathione ratios and oxygen sensing in calf and rabbitcarotid body chemoreceptor cells. J Physiol 537: 209–220, 2001.

680. Sarton E, Teppema L, Dahan A. Sex differences in morphine-induced ventilatory depression reside within the peripheral che-moreflex loop. Anesthesiology 90: 1329–1338, 1999.

681. Sarton E, van der Wal M, Nieuwenhuijs D, Teppema L, Ro-

botham JL, Dahan A. Sevoflurane-induced reduction of hypoxicdrive is sex-independent. Anesthesiology 90: 1288–1293, 1999.

682. Sato M, Severinghaus JW, Bickler P. Time course of augmen-tation and depression of hypoxic ventilatory responses at altitude.J Appl Physiol 77: 313–316, 1994.

683. Sato M, Severinghaus JW, Powell FL, Xu FD, Spellman MJ Jr.

Augmented hypoxic ventilatory response in men at altitude. J Appl

Physiol 73: 101–107, 1992.684. Schmitt P, Pequignot J, Garcia C, Pujol JF, Pequignot JM.

Regional specificity of the long-term regulation of tyrosine hydrox-ylase in some catecholaminergic rat brainstem areas. I. Influence oflong-term hypoxia. Brain Res 611: 53–60, 1993.

685. Schmitt P, Soulier V, Pequignot JM, Pujol JF, Avit-Saubie M.

Ventilatory acclimatization to chronic hypoxia: relationship to nor-adrenaline metabolism in the rat solitary complex. J Physiol 477:331–337, 1994.

686. Schoene RB, Roach RC, Hackett PH, Sutton JR, Cymerman A,

Houston CS. Operation Everest II: ventilatory adaptation duringgradual decompression to extreme altitude. Med Sci Sports Exerc

22: 804–810, 1990.687. Schuitmaker JJ, Berkenbosch A, De Goede J, Olievier CN.

Effects of CO2 and H� on the ventilatory response to peripheralchemoreceptor stimulation. Respir Physiol 64: 69–79, 1986.

688. Schwenke DO, Pearson JT, Kangawa K, Shirai M. Does centralnitric oxide chronically modulate the acute hypoxic ventilatoryresponse in conscious rats? Acta Physiol Oxf 186: 309–318, 2006.

689. Semenza GL. Regulation of mammalian O2 homeostasis byhypoxia-inducible factor 1. Annu Rev Cell Dev Biol 15: 551–578,1999.

690. Semenza GL. HIF-1: mediator of physiological and pathophysio-logical responses to hypoxia. J Appl Physiol 88: 1474–1480, 2000.

691. Semenza GL. O2-regulated gene expression: transcriptional con-trol of cardiorespiratory physiology by HIF-1. J Appl Physiol 96:1173–1177, 2004.

692. Semenza GL. Regulation of oxygen homeostasis by hypoxia-in-ducible factor 1. Physiology 24: 97–106, 2009.

693. Semenza GL, Shimoda LA, Prabhakar NR. Regulation of geneexpression by HIF-1. Novartis Found Symp 272: 2–8, 2006.



694. Semenza GL. Life with Oxygen. Science 318: 62–64, 2007.695. Serebrovskaya TV. Intermittent hypoxia research in the former

Soviet Union and the Commonwealth of Independent States: his-tory and review of the concept and selected applications. High Alt

Med Biol 3: 205–221, 2002.696. Severinghaus J. Draft: 2006 Lake Louise HVR methods. http://

www.hypoxia.net/forum/HVR.pdf. 2006.697. Severinghaus JW. Proposed standard determination of ventila-

tory responses to hypoxia and hypercapnia in man. Chest 70:129–131, 1976.

698. Severinghaus JW. Fire-air and dephlogistication. Revisionisms ofoxygen’s discovery. Adv Exp Med Biol 543: 7–19, 2003.

699. Sharma SD, Raghuraman G, Lee MS, Prabhakar NR, Kumar

GK. Intermittent hypoxia activates peptidylglycine �-amidatingmonooxygenase in rat brain stem via reactive oxygen species-mediated proteolytic processing. J Appl Physiol 106: 12–19, 2009.

700. Shatilo VB, Korkushko OV, Ischuk VA, Downey HF, Sere-

brovskaya TV. Effects of intermittent hypoxia training on exerciseperformance, hemodynamics, and ventilation in healthy seniormen. High Alt Med Biol 9: 43–52, 2008.

701. Shirahata M, Balbir A, Otsubo T, Fitzgerald RS. Role of ace-tylcholine in neurotransmission of the carotid body. Respir Physiol

Neurobiol 157: 93–105, 2007.702. Simakajornboon N, Kuptanon T. Maturational changes in neu-

romodulation of central pathways underlying hypoxic ventilatoryresponse. Respir Physiol Neurobiol 149: 273–286, 2005.

703. Sladek M, Parker RA, Grogaard JB, Sundell HW. Long-lastingeffect of prolonged hypoxemia after birth on the immediate venti-latory response to changes in arterial partial pressure of oxygen inyoung lambs. Pediatr Res 34: 821–828, 1993.

704. Slegel P, Kitagawa H, Maguire MH. Determination of adenosinein fetal perfusates of human placental cotyledons using fluores-cence derivatization and reversed-phase high-performance liquidchromatography. Anal Biochem 171: 124–134, 1988.

705. Slessarev M, Han J, Mardimae A, Prisman E, Preiss D,

Volgyesi G, Ansel C, Duffin J, Fisher JA. Prospective targetingand control of end-tidal CO2 and O2 concentrations. J Physiol 581:1207–1219, 2007.

706. Smith CA, Bisgard GE, Nielsen AM, Daristotle L, Kressin NA,

Forster HV, Dempsey JA. Carotid bodies are required for venti-latory acclimatization to chronic hypoxia. J Appl Physiol 60: 1003–1010, 1986.

707. Smith CA, Harms CA, Henderson KS, Dempsey JA. Ventilatoryeffects of specific carotid body hypocapnia and hypoxia in awakedogs. J Appl Physiol 82: 791–798, 1997.

708. Smith CA, Jameson LC, Mitchell GS, Musch TI, Dempsey JA.

Central-peripheral chemoreceptor interaction in awake cerebrospi-nal fluid-perfused goats. J Appl Physiol 56: 1541–1549, 1984.

709. Smith DW, Buller KM, Day TA. Role of ventrolateral medullacatecholamine cells in hypothalamic neuroendocrine cell re-sponses to systemic hypoxia. J Neurosci 15: 7979–7988, 1995.

710. Smith PG, Mills E. Restoration of reflex ventilatory response tohypoxia after removal of carotid bodies in the cat. Neuroscience 5:573–580, 1980.

711. Smith TG, Balanos GM, Croft QP, Talbot NP, Dorrington KL,

Ratcliffe PJ, Robbins PA. The increase in pulmonary arterialpressure caused by hypoxia depends on iron status. J Physiol 586:5999–6005, 2008.

712. Smith TG, Brooks JT, Balanos GM, Lappin TR, Layton DM,

Leedham DL, Liu C, Maxwell PH, McMullin MF, McNamara

CJ, Percy MJ, Pugh CW, Ratcliffe PJ, Talbot NP, Treacy M,

Robbins PA. Mutation of von Hippel-Lindau tumour suppressorand human cardiopulmonary physiology. PLoS Med 3: e290, 2006.

713. Smith TG, Talbot NP, Privat C, Rivera-Ch M, Nickol AH,

Ratcliffe PJ, Dorrington KL, Leon-Velarde F, Robbins PA.

Effects of iron supplementation and depletion on hypoxic pulmo-nary hypertension: two randomized controlled trials. JAMA 302:1444–1450, 2009.

714. Smith WD, Poulin MJ, Paterson DH, Cunningham DA. Dy-namic ventilatory response to acute isocapnic hypoxia in septua-genarians. Exp Physiol 86: 117–126, 2001.

715. Soliz J, Gassmann M, Joseph V. Soluble erythropoietin receptoris present in the mouse brain and is required for the ventilatoryacclimatization to hypoxia. J Physiol 583: 329–336, 2007.

716. Soliz J, Joseph V, Soulage C, Becskei C, Vogel J, Pequignot

JM, Ogunshola O, Gassmann M. Erythropoietin regulateshypoxic ventilation in mice by interacting with brainstem andcarotid bodies. J Physiol 568: 559–571, 2005.

717. Soliz J, Thomsen JJ, Soulage C, Lundby C, Gassmann M.

Sex-dependent regulation of hypoxic ventilation in mice and hu-mans is mediated by erythropoietin. Am J Physiol Regul Integr

Comp Physiol 296: R1837–R1846, 2009.718. Solomon IC, Edelman NH, Neubauer JA. Pre-Botzinger com-

plex functions as a central hypoxia chemosensor for respiration invivo. J Neurophysiol 83: 2854–2868, 2000.

719. Somogyi RB, Preiss D, Vesely A, Fisher JA, Duffin J. Changesin respiratory control after 5 days at altitude. Respir Physiol Neu-

robiol 145: 41–52, 2005.720. Sorensen SC, Severinghaus JW. Respiratory sensitivity to acute

hypoxia in man born at sea level living at high altitude. J Appl

Physiol 25: 211–216, 1968.721. Soulier V, Dalmaz Y, Cottet-Emard JM, Lagercrantz H,

Pequignot JM. Long-term influence of neonatal hypoxia on cate-cholamine activity in carotid bodies and brainstem cell groups ofthe rat. J Physiol 498: 523–530, 1997.

722. Soulier V, Gestreau C, Borghini N, Dalmaz Y, Cottet-Emard

JM, Pequignot JM. Peripheral chemosensitivity and central inte-gration: neuroplasticity of catecholaminergic cells under hypoxia.Comp Biochem Physiol A Physiol 118: 1–7, 1997.

723. Sovik S, Lossius K. Development of ventilatory response to tran-sient hypercapnia and hypercapnic hypoxia in term infants. Pediatr

Res 55: 302–309, 2004.724. St Croix CM, Cunningham DA, Paterson DH. Nature of the

interaction between central and peripheral chemoreceptor drivesin human subjects. Can J Physiol Pharmacol 74: 640–646, 1996.

725. Stark RI, Daniel SS, Garland M, Jaille-Marti JC, Kim YI,

Leung K, Myers MM, Tropper PJ. Fetal hiccups in the baboon.Am J Physiol Regul Integr Comp Physiol 267: R1479–R1487, 1994.

726. Stea A, Jackson A, Macintyre L, Nurse CA. Long-term modu-lation of inward currents in O2 chemoreceptors by chronic hypoxiaand cyclic AMP in vitro. J Neurosci 15: 2192–2202, 1995.

727. Stea A, Jackson A, Nurse CA. Hypoxia and N6,O2 -dibutyrylad-enosine 3 ,5 -cyclic monophosphate, but not nerve growth factor,induce Na� channels and hypertrophy in chromaffin-like arterialchemoreceptors. Proc Natl Acad Sci USA 89: 9469–9473, 1992.

728. Steinback CD, Poulin MJ. Ventilatory responses to isocapnic andpoikilocapnic hypoxia in humans. Respir Physiol Neurobiol 155:104–113, 2007.

729. Stephenson R, Liao KS, Hamrahi H, Horner RL. Circadianrhythms and sleep have additive effects on respiration in the rat.J Physiol 536: 225–235, 2001.

730. Sterni LM, Bamford OS, Tomares SM, Montrose MH, Carroll

JL. Developmental changes in intracellular Ca2� response of ca-rotid chemoreceptor cells to hypoxia. Am J Physiol Lung Cell Mol

Physiol 268: L801–L808, 1995.731. Sterni LM, Bamford OS, Wasicko MJ, Carroll JL. Chronic

hypoxia abolished the postnatal increase in carotid body type I cellsensitivity to hypoxia. Am J Physiol Lung Cell Mol Physiol 277:L645–L652, 1999.

732. Stornetta RL. Identification of neurotransmitters and co-localiza-tion of transmitters in brainstem respiratory neurons. Respir

Physiol Neurobiol 164: 18–27, 2008.733. Streller T, Huckstorf C, Pfeiffer CAH. Unusual cytochrome

a592 with low PO2 affinity correlates as putative oxygen sensor withrat carotid body chemoreceptor discharge. FASEB J 16: 1277–1279,2002.

734. Strohl KP. Inheritance and ventilatory behavior in animal models.In: Pharmacology and Pathophysiology of the Control of Breath-

ing, edited by Ward DS, Dahan A, and Teppema LJ. New York:Taylor & Francis, 2005, p. 261–291.

735. Strohl KP, Thomas AJ. Neonatal conditioning for adult respira-tory behavior. Respir Physiol 110: 269–275, 1997.



736. Strohl KP, Thomas AJ, St Jean P, Schlenker EH, Koletsky RJ,

Schork NJ. Ventilation and metabolism among rat strains. J Appl

Physiol 82: 317–323, 1997.737. Subramanian S, Dostal J, Erokwu B, Han F, Dick TE, Strohl

KP. Domperidone and ventilatory behavior: Sprague-Dawley ver-sus Brown Norway rats. Respir Physiol Neurobiol 155: 22–28, 2007.

738. Subramanian S, Erokwu B, Han F, Dick TE, Strohl KP.

L-NAME differentially alters ventilatory behavior in Sprague-Daw-ley and Brown Norway rats. J Appl Physiol 93: 984–989, 2002.

739. Subramanian S, Han F, Erokwu BO, Dick TE, Strohl KP. Dogenetic factors influence the Dejours phenomenon? Adv Exp Med

Biol 499: 209–214, 2001.740. Suguihara C, Hehre D, Bancalari E. Effect of dopamine on

hypoxic ventilatory response of sedated piglets with intact anddenervated carotid bodies. J Appl Physiol 77: 285–289, 1994.

741. Sun MK, Jeske IT, Reis DJ. Cyanide excites medullary sympa-thoexcitatory neurons in rats. Am J Physiol Regul Integr Comp

Physiol 262: R182–R189, 1992.742. Sun MK, Reis DJ. Central neural mechanisms mediating excita-

tion of sympathetic neurons by hypoxia. Prog Neurobiol 44: 197–219, 1994.

743. Sun MK, Reis DJ. Dopamine or transmitter release from ratcarotid body may not be essential to hypoxic chemoreception.Am J Physiol Regul Integr Comp Physiol 267: R1632–R1639, 1994.

744. Sun MK, Reis DJ. Hypoxia selectively excites vasomotor neuronsof rostral ventrolateral medulla in rats. Am J Physiol Regul Integr

Comp Physiol 266: R245–R256, 1994.745. Sun MK, Reis DJ. Hypoxia-activated Ca2� currents in pacemaker

neurones of rat rostral ventrolateral medulla in vitro. J Physiol 476:101–116, 1994.

746. Suzuki A, Nishimura M, Yamamoto H, Miyamoto K, Kishi F,

Kawakami Y. No effect of brain blood flow on ventilatory depres-sion during sustained hypoxia. J Appl Physiol 66: 1674–1678, 1989.

747. Swanson GD, Bellville JW. Step changes in end-tidal CO2: meth-ods and implications. J Appl Physiol 39: 377–385, 1975.

748. Swanson GD, Carpenter TM Jr, Snider DE, Bellville JW. Anon-line hybrid computing system for dynamic respiratory responsestudies. Comput Biomed Res 4: 205–215, 1971.

749. Swenson ER. Carbonic anhydrase inhibitors and hypoxic pulmo-nary vasoconstriction. Respir Physiol Neurobiol 151: 209–216,2006.

750. Swenson ER, Hughes JM. Effects of acute and chronic acetazol-amide on resting ventilation and ventilatory responses in men.J Appl Physiol 74: 230–237, 1993.

751. Szewczak JM, Powell FL. Open-flow plethysmography with pres-sure-decay compensation. Respir Physiol Neurobiol 134: 57–67,2003.

752. Tabata M, Kurosawa H, Kikuchi Y, Hida W, Ogawa H, Okabe S,

Tun Y, Hattori T, Shirato K. Role of GABA within the nucleustractus solitarii in the hypoxic ventilatory decline of awake rats.Am J Physiol Regul Integr Comp Physiol 281: R1411–R1419, 2001.

753. Takakura AC, Moreira TS, Colombari E, West GH, Stornetta

RL, Guyenet PG. Peripheral chemoreceptor inputs to retrotrap-ezoid nucleus RTN CO2-sensitive neurons in rats. J Physiol 572:503–523, 2006.

754. Takeda K, Adachi T, Han F, Yokoyama S, Yamamoto H, Hida

W, Shibahara S. Augmented chemosensitivity in black-eyed whiteMitfmi-bw mice, lacking melanocytes. J Biochem 141: 327–333,2007.

755. Talbot NP, Balanos GM, Robbins PA, Dorrington KL. Intrave-nous endothelin-1 and ventilatory sensitivity to hypoxia in humans.Adv Exp Med Biol 605: 57–62, 2008.

756. Tamisier R, Gilmartin GS, Launois SH, Pepin JL, Nespoulet

H, Thomas R, Levy P, Weiss JW. A new model of chronic inter-mittent hypoxia in humans: effect on ventilation, sleep, and bloodpressure. J Appl Physiol 107: 17–24, 2009.

757. Tamisier R, Hunt BE, Gilmartin GS, Curley M, Anand A,

Weiss JW. Hemodynamics and muscle sympathetic nerve activityafter 8 h of sustained hypoxia in healthy humans. Am J Physiol

Heart Circ Physiol 293: H3027–H3035, 2007.758. Tankersley CG. Selected contribution: variation in acute hypoxic

ventilatory response is linked to mouse chromosome 9. J Appl

Physiol 90: 1615–1622, 2001.

759. Tankersley CG, Elston RC, Schnell AH. Genetic determinantsof acute hypoxic ventilation: patterns of inheritance in mice. J Appl

Physiol 88: 2310–2318, 2000.760. Tankersley CG, Fitzgerald RS, Kleeberger SR. Differential con-

trol of ventilation among inbred strains of mice. Am J Physiol

Regul Integr Comp Physiol 267: R1371–R1377, 1994.761. Tankersley CG, Broman KW. Interactions in hypoxic and hyper-

capnic breathing are genetically linked to mouse chromosomes 1and 5. J Appl Physiol 97: 77–84, 2004.

762. Tansley JG, Clar C, Pedersen ME, Robbins PA. Human venti-latory response to acute hyperoxia during and after 8 h of bothisocapnic and poikilocapnic hypoxia. J Appl Physiol 82: 513–519,1997.

763. Tansley JG, Fatemian M, Howard LS, Poulin MJ, Robbins PA.

Changes in respiratory control during and after 48 h of isocapnicand poikilocapnic hypoxia in humans. J Appl Physiol 85: 2125–2134, 1998.

764. Tansley JG, Pedersen ME, Clar C, Robbins PA. Human venti-latory response to 8 h of euoxic hypercapnia. J Appl Physiol 84:431–434, 1998.

765. Tatsumi K, Hannhart B, Pickett CK, Weil JV, Moore LG.

Influences of gender and sex hormones on hypoxic ventilatoryresponse in cats. J Appl Physiol 71: 1746–1751, 1991.

766. Tatsumi K, Hannhart B, Pickett CK, Weil JV, Moore LG.

Effects of testosterone on hypoxic ventilatory and carotid bodyneural responsiveness. Am J Respir Crit Care Med 149: 1248–1253,1994.

767. Tatsumi K, Pickett CK, Jacoby CR, Weil JV, Moore LG. Role ofendogenous female hormones in hypoxic chemosensitivity. J Appl

Physiol 83: 1706–1710, 1997.768. Tatsumi K, Pickett CK, Weil JV. Attenuated carotid body

hypoxic sensitivity after prolonged hypoxic exposure. J Appl

Physiol 70: 748–755, 1991.769. Tatsumi K, Pickett CK, Weil JV. Effects of haloperidol and

domperidone on ventilatory roll off during sustained hypoxia incats. J Appl Physiol 72: 1945–1952, 1992.

770. Tatsumi K, Pickett CK, Weil JV. Decreased carotid bodyhypoxic sensitivity in chronic hypoxia: role of dopamine. Respir

Physiol 101: 47–57, 1995.771. Tatsumi K, Pickett CK, Weil JV. Possible role of dopamine in

ventilatory acclimatization to high altitude. Respir Physiol 99: 63–73, 1995.

772. Taylor NC, Li A, Nattie EE. Ventilatory effects of muscimolmicrodialysis into the rostral medullary raphe region of consciousrats. Respir Physiol Neurobiol 153: 203–216, 2006.

773. Temp JA, Henson LC, Ward DS. Does a subanesthetic concen-tration of isoflurane blunt the ventilatory response to hypoxia?Anesthesiology 77: 1116–1124, 1992.

774. Tenney SM, Ou LC. Hypoxic ventilatory response of cats at highaltitude: an interpretation of “blunting.” Respir Physiol 30: 185–199, 1977.

775. Teppema LJ, van Dorp ELA, Dahan A. Arterial [H�] and theventilatory response to hypoxia in humans: influence of acetazol-amide-induced metabolic acidosis. Am J Physiol Lung Cell Mol

Physiol 298: L89–L95, 2010.776. Teppema LJ, Bijl H, Mousavi GB, Dahan A. The carbonic an-

hydrase inhibitors methazolamide and acetazolamide have differ-ent effects on the hypoxic ventilatory response in the anaesthetizedcat. J Physiol 574: 565–572, 2006.

777. Teppema LJ, Bijl H, Romberg RR, Dahan A. Antioxidants re-verse depression of the hypoxic ventilatory response by acetazol-amide in man. J Physiol 572: 849–856, 2006.

778. Teppema LJ, Dahan A. Acetazolamide and breathing. Does aclinical dose alter peripheral and central CO2 sensitivity? Am J

Respir Crit Care Med 160: 1592–1597, 1999.779. Teppema LJ, Dahan A, Olievier CN. Low-dose acetazolamide

reduces CO2-O2 stimulus interaction within the peripheral chemo-receptors in the anaesthetised cat. J Physiol 537: 221–229, 2001.

780. Teppema LJ, Nieuwenhuijs D, Sarton E, Romberg R, Olievier

CN, Ward DS, Dahan A. Antioxidants prevent depression of theacute hypoxic ventilatory response by subanaesthetic halothane inmen. J Physiol 544: 931–938, 2002.



781. Teppema LJ, Romberg RR, Dahan A. Antioxidants reverse re-duction of the human hypoxic ventilatory response by subanes-thetic isoflurane. Anesthesiology 102: 747–753, 2005.

782. Teppema LJ, Veening JG, Kranenburg A, Dahan A, Berken-

bosch A, Olievier C. Expression of c-fos in the rat brainstem afterexposure to hypoxia and to normoxic and hyperoxic hypercapnia.J Comp Neurol 388: 169–190, 1997.

783. Terblanche JS, Tolley KA, Fahlman A, Myburgh KH, Jackson

S. The acute hypoxic ventilatory response: testing the adaptivesignificance in human populations. Comp Biochem Physiol A Mol

Integr Physiol 140: 349–362, 2005.784. Tipoe GL, Fung ML. Expression of HIF-1alpha, VEGF and VEGF

receptors in the carotid body of chronically hypoxic rat. Respir

Physiol Neurobiol 138: 143–154, 2003.785. Tjong YW, Chen Y, Liong EC, Tipoe GL, Fung ML. Chronic

hypoxia modulates the function and expression of melatonin re-ceptors in the rat carotid body. J Pineal Res 40: 125–134, 2006.

786. Towler MC, Hardie DG. AMP-activated protein kinase in meta-bolic control and insulin signaling. Circ Res 100: 328–341, 2007.

787. Townsend NE, Gore CJ, Hahn AG, McKenna MJ, Aughey RJ,

Clark SA, Kinsman T, Hawley JA, Chow CM. Living high-train-ing low increases hypoxic ventilatory response of well-trainedendurance athletes. J Appl Physiol 93: 1498–1505, 2002.

788. Trapp S, Aller MI, Wisden W, Gourine AV. A role for TASK-1KCNK3 channels in the chemosensory control of breathing. J Neu-

rosci 28: 8844–8850, 2008.789. Trippenbach T. Ventilatory and metabolic effects of repeated

hypoxia in conscious newborn rabbits. Am J Physiol Regul Integr

Comp Physiol 266: R1584–R1590, 1994.790. Tun Y, Hida W, Okabe S, Kikuchi Y, Kurosawa H, Tabata M,

Shirato K. Effects of nasal continuous positive airway pressure onawake ventilatory responses to hypoxia and hypercapnia in pa-tients with obstructive sleep apnea. Tohoku J Exp Med 190: 157–168, 2000.

791. Tun Y, Hida W, Okabe S, Ogawa H, Kikuchi Y, Oikawa M,

Kitamuro T, Kunio S. Can nasal continuous positive airway pres-sure decrease clinic blood pressure in patients with obstructivesleep apnea? Tohoku J Exp Med 201: 181–190, 2003.

792. Turner DL, Mitchell GS. Long-term facilitation of ventilationfollowing repeated hypoxic episodes in awake goats. J Physiol 499:543–550, 1997.

793. Van Beek JH, Berkenbosch A, De Goede J, Olievier CN. Influ-ence of peripheral O2 tension on the ventilatory response to CO2 incats. Respir Physiol 51: 379–390, 1983.

794. Van Beek JH, Berkenbosch A, De Goede J, Olievier CN. Ef-fects of brain stem hypoxaemia on the regulation of breathing.Respir Physiol 57: 171–188, 1984.

795. Van Bockstaele EJ, Peoples J, Telegan P. Efferent projectionsof the nucleus of the solitary tract to peri-locus coeruleus dendritesin rat brain: evidence for a monosynaptic pathway. J Comp Neurol

412: 410–428, 1999.796. Van den Elsen M, Dahan A, DeGoede J, Berkenbosch A, van

Kleef J. Influences of subanesthetic isoflurane on ventilatory con-trol in humans. Anesthesiology 83: 478–490, 1995.

797. Van den Elsen M, Sarton E, Teppema L, Berkenbosch A,

Dahan A. Influence of 0.1 minimum alveolar concentration ofsevoflurane, desflurane and isoflurane on dynamic ventilatory re-sponse to hypercapnia in humans. Br J Anaesth 80: 174–182, 1998.

798. Van den Elsen MJ, Dahan A, Berkenbosch A, DeGoede J, van

Kleef JW, Olievier IC. Does subanesthetic isoflurane affect theventilatory response to acute isocapnic hypoxia in healthy volun-teers? Anesthesiology 81: 860–867, 1994.

799. Varas R, Alcayaga J, Iturriaga R. ACh and ATP mediate excita-tory transmission in cat carotid identified chemoreceptor units invitro. Brain Res 988: 154–163, 2003.

800. Varas R, Wyatt CN, Buckler KJ. Modulation of TASK-like back-ground potassium channels in rat arterial chemoreceptor cells byintracellular ATP and other nucleotides. J Physiol 583: 521–536,2007.

801. Vardhan A, Kachroo A, Sapru HN. Excitatory amino acid recep-tors in commissural nucleus of the NTS mediate carotid chemore-ceptor responses. Am J Physiol Regul Integr Comp Physiol 264:R41–R50, 1993.

802. Vicario I, Obeso A, Rocher A, Lopez-Lopez JR, Gonzalez C.

Intracellular Ca2� stores in chemoreceptor cells of the rabbit ca-rotid body: significance for chemoreception. Am J Physiol Cell

Physiol 279: C51–C61, 2000.803. Viemari JC. Noradrenergic modulation of the respiratory neural

network. Respir Physiol Neurobiol 164: 123–130, 2008.804. Villafuerte FC, Cardenas-Alayza R, Macarlupu JL, Monge C,

Leon-Velarde F. Ventilatory response to acute hypoxia in trans-genic mice over-expressing erythropoietin: effect of acclimation to3-week hypobaric hypoxia. Respir Physiol Neurobiol 158: 243–250,2007.

805. Vincent SG, Waddell AE, Caron MG, Walker JK, Fisher JT. Amurine model of hyperdopaminergic state displays altered respira-tory control. FASEB J 21: 1463–1471, 2007.

806. Vizek M, Bonora M. Diaphragmatic activity during biphasic ven-tilatory response to hypoxia in rats. Respir Physiol 111: 153–162,1998.

807. Vizek M, Pickett CK, Weil JV. Biphasic ventilatory response ofadult cats to sustained hypoxia has central origin. J Appl Physiol

63: 1658–1664, 1987.808. Vizek M, Pickett CK, Weil JV. Increased carotid body hypoxic

sensitivity during acclimatization to hypobaric hypoxia. J Appl

Physiol 63: 2403–2410, 1987.809. Vizek M, Pickett CK, Weil JV. Interindividual variation in

hypoxic ventilatory response: potential role of carotid body. J Appl

Physiol 63: 1884–1889, 1987.810. Vlasic V, Simakajornboon N, Gozal E, Gozal D. PDGF-beta

receptor expression in the dorsocaudal brainstem parallelshypoxic ventilatory depression in the developing rat. Pediatr Res

50: 236–241, 2001.811. Wach RA, Bee D, Barer GR. Dopamine and ventilatory effects of

hypoxia and almitrine in chronically hypoxic rats. J Appl Physiol

67: 186–192, 1989.812. Wade JG, Larson CP Jr, Hickey RF, Ehrenfeld WK, Severing-

haus JW. Effect of carotid endarterectomy on carotid chemore-ceptor and baroreceptor function in man. N Engl J Med 282:823–829, 1970.

813. Waldrop TG, Bauer RM, Iwamoto GA. Microinjection of GABAantagonists into the posterior hypothalamus elicits locomotor ac-tivity and a cardiorespiratory activation. Brain Res 444: 84–94,1988.

814. Walker DW. Hypoxic inhibition of breathing and motor activity inthe foetus and newborn. Clin Exp Pharmacol Physiol 22: 533–536,1995.

815. Walker DW, Lee B, Nitsos I. Effect of hypoxia on respiratoryactivity in the foetus. Clin Exp Pharmacol Physiol 27: 110–113,2000.

816. Walsh BK, Brooks TM, Grenier BM. Oxygen therapy in theneonatal care environment. Respir Care 54: 1193–1202, 2009.

817. Wang JS, Chen LY, Fu LL, Chen ML, Wong MK. Effects ofmoderate and severe intermittent hypoxia on vascular endothelialfunction and haemodynamic control in sedentary men. Eur J Appl

Physiol 100: 127–135, 2007.818. Wang ZY, Bisgard GE. Chronic hypoxia-induced morphological

and neurochemical changes in the carotid body. Microsc Res Tech

59: 168–177, 2002.819. Wang ZY, Olson EB Jr, Bjorling DE, Mitchell GS, Bisgard GE.

Sustained hypoxia-induced proliferation of carotid body type I cellsin rats. J Appl Physiol 104: 803–808, 2008.

820. Wang ZZ, Dinger B, Fidone SJ, Stensaas LJ. Changes in ty-rosine hydroxylase and substance P immunoreactivity in the catcarotid body following chronic hypoxia and denervation. Neuro-

science 83: 1273–1281, 1998.821. Ward DS. The effects of dopamine on the ventilatory response to

sustained hypoxia in humans. In: Control of Breathing and Its

Modeling Perspective, edited by Honda Y, Miyamoto Y, Konno Kand Widdicombe JG. New York: Plenum, 1992, p. 291–298.

822. Ward DS. Stimulation of hypoxic ventilatory drive by droperidol.Anesth Analg 63: 106–110, 1984.

823. Ward DS, Berkenbosch A, DeGoede J, Olievier CN. Dynamicsof the ventilatory response to central hypoxia in cats. J Appl

Physiol 68: 1107–1113, 1990.



824. Ward DS, Dahan A, Mann CB. Modelling the dynamic ventilatoryresponse to hypoxia in humans. Ann Biomed Eng 20: 181–194,1992.

825. Ward DS, Voter WA, Karan S. The effects of hypo- and hyper-glycaemia on the hypoxic ventilatory response in humans.J Physiol 582: 859–869, 2007.

826. Wasicko MJ, Sterni LM, Bamford OS, Montrose MH, Carroll

JL. Resetting and postnatal maturation of oxygen chemosensitivityin rat carotid chemoreceptor cells. J Physiol 514: 493–503, 1999.

827. Waters KA, Gozal D. Responses to hypoxia during early develop-ment. Respir Physiol Neurobiol 136: 115–129, 2003.

828. Waters KA, Laferriere A, Paquette J, Goodyer C, Moss IR.

Curtailed respiration by repeated vs. isolated hypoxia in maturingpiglets is unrelated to NTS ME or SP levels. J Appl Physiol 83:522–529, 1997.

829. Waters KA, Tinworth KD. Depression of ventilatory responsesafter daily, cyclic hypercapnic hypoxia in piglets. J Appl Physiol 90:1065–1073, 2001.

830. Watson CS, Schaefer R, White SE, Homan JH, Fraher L,

Harding R, Bocking AD. Effect of intermittent umbilical cordocclusion on fetal respiratory activity and brain adenosine in late-gestation sheep. Reprod Fertil Dev 14: 35–42, 2002.

831. Watson CS, White SE, Homan J, Abdollah S, Brien JF, Challis

JR, Bocking AD. Fetal breathing is not inhibited by ethanol expo-sure during prolonged reduced uterine blood flow. Can J Physiol

Pharmacol 74: 1016–1024, 1996.832. Weil JV. Variation in human ventilatory control-genetic influence

on the hypoxic ventilatory response. Respir Physiol Neurobiol 135:239–246, 2003.

833. Weil JV, Byrne-Quinn E, Sodal IE, Friesen WO, Underhill B,

Filley GF, Grover RF. Hypoxic ventilatory drive in normal man.J Clin Invest 49: 1061–1072, 1970.

834. Weil JV, Stevens T, Pickett CK, Tatsumi K, Dickinson MG,

Jacoby CR, Rodman DM. Strain-associated differences in hypoxicchemosensitivity of the carotid body in rats. Am J Physiol Lung

Cell Mol Physiol 274: L767–L774, 1998.835. Weir EK, Lopez-Barneo J, Buckler KJ, Archer SL. Acute oxy-

gen-sensing mechanisms. N Engl J Med 353: 2042–2055, 2005.836. Weizhen N, Engwall MJ, Daristotle L, Pizarro J, Bisgard GE.

Ventilatory effects of prolonged systemic CNS hypoxia in awakegoats. Respir Physiol 87: 37–48, 1992.

837. Wenger RH. Mitochondria: oxygen sinks rather than sensors? Med

Hypotheses 66: 380–383, 2006.838. Wenninger JM, Olson EB, Wang Z, Keith IM, Mitchell GS,

Bisgard GE. Carotid sinus nerve responses and ventilatory accli-matization to hypoxia in adult rats following 2 weeks of postnatalhyperoxia. Respir Physiol Neurobiol 150: 155–164, 2006.

839. West JB. Rate of ventilatory acclimatization to extreme altitude.Respir Physiol 74: 323–333, 1988.

840. White DP, Douglas NJ, Pickett CK, Weil JV, Zwillich CW.

Sexual influence on the control of breathing. J Appl Physiol 54:874–879, 1983.

841. White DP, Schneider BK, Santen RJ, McDermott M, Pickett

CK, Zwillich CW, Weil JV. Influence of testosterone on ventila-tion and chemosensitivity in male subjects. J Appl Physiol 59:1452–1457, 1985.

842. Williams BA, Buckler KJ. Biophysical properties and metabolicregulation of a TASK-like potassium channel in rat carotid bodytype 1 cells. Am J Physiol Lung Cell Mol Physiol 286: L221–L230,2004.

843. Williams BA, Smyth J, Boon AW, Hanson MA, Kumar P,

Blanco CE. Development of respiratory chemoreflexes in re-sponse to alternations of fractional inspired oxygen in the newborninfant. J Physiol 442: 81–90, 1991.

844. Williams SE, Wootton P, Mason HS, Bould J, Iles DE, Riccardi

D, Peers C, Kemp PJ. Hemoxygenase-2 is an oxygen sensor for acalcium-sensitive potassium channel. Science 306: 2093–2097, 2004.

845. Wolsink JG, Berkenbosch A, DeGoede J, Olievier CN. Venti-latory interaction between hypoxia and hypercapnia in pigletsshortly after birth. Respir Physiol 96: 25–35, 1994.

846. Wong-Riley MT, Liu Q. Neurochemical development of brainstem nuclei involved in the control of respiration. Respir Physiol

Neurobiol 149: 83–98, 2005.

847. Wong-Riley MT, Liu Q. Neurochemical and physiological corre-lates of a critical period of respiratory development in the rat.Respir Physiol Neurobiol 164: 28–37, 2008.

848. Wood HE, Fatemian M, Robbins PA. Prior sustained hypoxiaattenuates interaction between hypoxia and exercise as ventilatorystimuli in humans. Exp Physiol 92: 273–286, 2007.

849. Wrobel LJ, Ogier M, Chatonnet F, Autran S, Mezieres V,

Thoby-Brisson M, McLean H, Taeron C, Champagnat J. Ab-normal inspiratory depth in Phox2a haploinsufficient mice. Neuro-

science 145: 384–392, 2007.850. Wu W, Dave NB, Yu G, Strollo PJ, Kovkarova-Naumovski E,

Ryter SW, Reeves SR, Dayyat E, Wang Y, Choi AM, Gozal D,

Kaminski N. Network analysis of temporal effects of intermittentand sustained hypoxia on rat lungs. Physiol Genomics 36: 24–34,2008.

851. Wyatt CN, Buckler KJ. The effect of mitochondrial inhibitors onmembrane currents in isolated neonatal rat carotid body type Icells. J Physiol 556: 175–191, 2004.

852. Wyatt CN, Evans AM. AMP-activated protein kinase and chemo-transduction in the carotid body. Respir Physiol Neurobiol 157:22–29, 2007.

853. Wyatt CN, Peers C. Ca2�-activated K� channels in isolated type Icells of the neonatal rat carotid body. J Physiol 483: 559–565, 1995.

854. Wyatt CN, Wright C, Bee D, Peers C. O2-sensitive K� currents incarotid body chemoreceptor cells from normoxic and chronicallyhypoxic rats and their roles in hypoxic chemotransduction. Proc

Natl Acad Sci USA 92: 295–299, 1995.855. Wyatt CN, Mustard KJ, Pearson SA, Dallas ML, Atkinson L,

Kumar P, Peers C, Hardie DG, Evans AM. AMP-activated pro-tein kinase mediates carotid body excitation by hypoxia. J Biol

Chem 282: 8092–8098, 2007.856. Xu F, Xu J, Tse FW, Tse A. Adenosine stimulates depolarization

and rise in cytoplasmic [Ca2�] in type I cells of rat carotid bodies.Am J Physiol Cell Physiol 290: C1592–C1598, 2006.

857. Xu W, Chi L, Row BW, Xu R, Ke Y, Xu B, Luo C, Kheirandish

L, Gozal D, Liu R. Increased oxidative stress is associated withchronic intermittent hypoxia-mediated brain cortical neuronal cellapoptosis in a mouse model of sleep apnea. Neuroscience 126:313–323, 2004.

858. Yamaguchi S, Balbir A, Schofield B, Coram J, Tankersley CG,

Fitzgerald RS, O’Donnell CP, Shirahata M. Structural and func-tional differences of the carotid body between DBA/2J and A/Jstrains of mice. J Appl Physiol 94: 1536–1542, 2003.

859. Yamamoto M, Nishimura M, Kobayashi S, Akiyama Y, Miy-

amoto K, Kawakami Y. Role of endogenous adenosine in hypoxicventilatory response in humans: a study with dipyridamole. J Appl

Physiol 76: 196–203, 1994.860. Yamamoto Y, Konig P, Henrich M, Dedio J, Kummer W. Hyp-

oxia induces production of nitric oxide and reactive oxygen spe-cies in glomus cells of rat carotid body. Cell Tissue Res 325: 3–11,2006.

861. Yamamoto Y, Kummer W, Atoji Y, Suzuki Y. TASK-1, TASK-2,TASK-3 and TRAAK immunoreactivities in the rat carotid body.Brain Res 950: 304–307, 2002.

862. Yan X, Koos BJ, Kruger L, Linden J, Murray TF. Characteriza-tion of [125I]ZM 241385 binding to adenosine A2A receptors in thepineal of sheep brain. Brain Res 1096: 30–39, 2006.

863. Yang F, Khoo MC. Ventilatory response to randomly modulatedhypercapnia and hypoxia in humans. J Appl Physiol 76: 2216–2223,1994.

864. Yang JJ, Chou YC, Lin MT, Chiu TH. Hypoxia-induced differen-tial electrophysiological changes in rat locus coeruleus neurons.Life Sci 61: 1763–1773, 1997.

865. Ye JS, Tipoe GL, Fung PC, Fung ML. Augmentation ofhypoxia-induced nitric oxide generation in the rat carotid bodyadapted to chronic hypoxia: an involvement of constitutive and induc-ible nitric oxide synthases. Pflugers Arch 444: 178–185, 2002.

866. Yelmen NK, Turgut G, Sahin G, Oruc T. Effects of intravenousand intracerebroventricular theophylline on hypoxic ventilatorydepression in anesthetized cats. Med Princ Pract 13: 277–281, 2004.

867. Yuan G, Adhikary G, McCormick AA, Holcroft JJ, Kumar GK,

Prabhakar NR. Role of oxidative stress in intermittent hypoxia-



induced immediate early gene activation in rat PC12 cells.J Physiol 557: 773–783, 2004.

868. Zapata P. Is ATP a suitable co-transmitter in carotid body arterialchemoreceptors? Respir Physiol Neurobiol 157: 106–115, 2007.

869. Zapata P, Torrealba F. Blockade of dopamine-induced chemosen-sory inhibition by domperidone. Neurosci Lett 51: 359–364, 1984.

870. Zhang M, Buttigieg J, Nurse CA. Neurotransmitter mechanismsmediating low-glucose signalling in cocultures and fresh tissueslices of rat carotid body. J Physiol 578: 735–750, 2007.

871. Zhang S, Robbins PA. Methodological and physiological variabil-ity within the ventilatory response to hypoxia in humans. J Appl

Physiol 88: 1924–1932, 2000.

872. Zhang W, Carreno FR, Cunningham JT, Mifflin SW. Chronicsustained and intermittent hypoxia reduce the function of ATP-sensitive potassium channels in the nucleus of the solitary tract.Am J Physiol Regul Integr Comp Physiol 295: R1555–R1562,2008.

873. Zhang W, Carreno FR, Cunningham JT, Mifflin SW. Chronicsustained hypoxia enhances both evoked EPSCs and norepineph-rine inhibition of glutamatergic afferent inputs in the nucleus of thesolitary tract. J Neurosci 29: 3093–3102, 2009.

874. Zwemer CF, Song MY, Carello KA, D’Alecy LG. Strain differ-ences in response to acute hypoxia: CD-1 versus C57BL/6J mice.J Appl Physiol 102: 286–293, 2007.



90:675-754, 2010. doi:10.1152/physrev.00012.2009 Physiol RevLuc J. Teppema and Albert Dahan

You might find this additional information useful...

858 articles, 463 of which you can access free at: This article cites http://physrev.physiology.org/cgi/content/full/90/2/675#BIBL

on the following topics: http://highwire.stanford.edu/lists/artbytopic.dtlcan be found at Medline items on this article's topics

Medicine .. Anoxia Medicine .. Respiration Medicine .. Environment and Public Health Physiology .. Carotid Chemoreceptors Oncology .. Hypoxia Oncology .. Transcription Factors

including high-resolution figures, can be found at: Updated information and services http://physrev.physiology.org/cgi/content/full/90/2/675

can be found at: Physiological Reviewsabout Additional material and information http://www.the-aps.org/publications/prv

This information is current as of September 2, 2010 .

http://www.the-aps.org/.20814-3991. Copyright © 2010 by the American Physiological Society. ISSN: 0031-9333, ESSN: 1522-1210. Visit our website at

MDpublished quarterly in January, April, July, and October by the American Physiological Society, 9650 Rockville Pike, Bethesda provides state of the art coverage of timely issues in the physiological and biomedical sciences. It isPhysiological Reviews

http://physrev.physiology.org/cgi/content/full/90/2/675#BIBL

http://highwire.stanford.edu/lists/artbytopic.dtl

http://physrev.physiology.org/cgi/content/full/90/2/675

http://www.the-aps.org/publications/prv

http://www.the-aps.org/

the ventilatory response to hypoxia in mammals

Documents