Chronic intermittent hypoxia‐induced cardiovascular and renal dysfunction: from adaptation to maladaptation

Chronic intermittent hypoxia (CIH) is the dominant pathological feature of human obstructive sleep apnoea (OSA), which is highly prevalent and associated with cardiovascular and renal diseases. CIH causes hypertension, centred on sympathetic nervous overactivity, which persists following removal of the CIH stimulus. Molecular mechanisms contributing to CIH‐induced hypertension have been carefully delineated. However, there is a dearth of knowledge on the efficacy of interventions to ameliorate high blood pressure in established disease. CIH causes endothelial dysfunction, aberrant structural remodelling of vessels and accelerates atherosclerotic processes. Pro‐inflammatory and pro‐oxidant pathways converge on disrupted nitric oxide signalling driving vascular dysfunction. In addition, CIH has adverse effects on the myocardium, manifesting atrial fibrillation, and cardiac remodelling progressing to contractile dysfunction. Sympatho‐vagal imbalance, oxidative stress, inflammation, dysregulated HIF‐1α transcriptional responses and resultant pro‐apoptotic ER stress, calcium dysregulation, and mitochondrial dysfunction conspire to drive myocardial injury and failure. CIH elaborates direct and indirect effects in the kidney that initially contribute to the development of hypertension and later to chronic kidney disease. CIH‐induced morphological damage of the kidney is dependent on TLR4/NF‐κB/NLRP3/caspase‐1 inflammasome activation and associated pyroptosis. Emerging potential therapies related to the gut–kidney axis and blockade of aryl hydrocarbon receptors (AhR) are promising. Cardiorenal outcomes in response to intermittent hypoxia present along a continuum from adaptation to maladaptation and are dependent on the intensity and duration of exposure to intermittent hypoxia. This heterogeneity of OSA is relevant to therapeutic treatment options and we argue the need for better stratification of OSA phenotypes.

with male predominance, affecting around 14% of men and 5% of women aged 30-69 years (Benjafield et al., 2019).OSA represents a significant risk factor for cardiovascular and kidney diseases (Garvey et al., 2009) and is a secondary cause of arterial hypertension (HTN) (Brown et al., 2022;Peppard et al., 2000).
Chronic intermittent hypoxia (CIH) is considered the dominant feature of OSA.CIH refers to a pathological pattern of oxygen dysregulation, characterised by brief (seconds-to-minutes), recurrent cycling between normal physiological and low levels of O 2 (multiple events per hour) that is prolonged (days-to-years).CIH occurs not only during the collapse of the upper airway in OSA patients but also when there is a defective respiratory rhythm generated by the central nervous system (central apnoeas).
Most of the anti-hypertensive drugs used in OSA patients target the sympathetic system and/or the RAAS and have beneficial anti-hypertensive effects (Ou et al., 2023;Svedmyr et al., 2021).Although, RAAS inhibitors are proven to reduce HTN in OSA patients (Heitmann et al., 2010;Kraiczi et al., 2000;Morgan et al., 2018;Thunström et al., 2016) treatment with beta blockers alone or in combination with a diuretic was associated with the lowest systolic pressure in a cohort of 5818 OSA patients (Svedmyr et al., 2021).It is likely that these patients had mild HTN since they received monotherapy.In small cohorts, the number of OSA patients treated with beta blockers is low, which does not readily facilitate comparisons with other therapies (Diogo, Pinto et al., 2015).Curiously, the beta blocker drug carvedilol failed to reverse established HTN in a CIH rat model (Diogo, Pinto et al., 2015).Despite the beneficial effect of anti-hypertensive drugs, there is a high prevalence (60%) of poorly controlled BP in OSA patients (Ou et al., 2023;Svedmyr et al., 2021).The association between OSA and drug refractory HTN is well established.There is a high prevalence (∼80%) of sleep apnoea patients with drug-resistant or refractory HTN, and OSA has been considered a secondary cause of drug-resistant HTN (Chedier et al., 2022;Logan et al., 2001;Pedrosa et al., 2011).
In short, the standard of care in OSA patients has limitations in the control and treatment of HTN and cardiovascular risk and there is a need to improve therapeutic strategies.Hypertension and general cardiovascular and renal damage induced by CIH ultimately results from carotid body-induced autonomic imbalance with sympathetic hyperactivity together with oxidative stress and inflammation.There is an enormous literature on the effects of CIH on the carotid body (recently reviewed by Iturriaga, 2023) as well as excellent reviews detailing the effects of CIH/OSA on sympathetic nerve activity in humans (e.g.Puri et al., 2021).The present review is focused on the effect of CIH on the molecular phenotypes of the heart, vessels and kidney.The latter tissues are downstream of the confluence J Physiol 601.24 of complex functional (e.g.carotid body, sympathetic, and RAAS activity) and biochemical mechanisms (e.g.HIF stabilization, oxidative stress and inflammation).This review emphasizes the transition from CIH-induced adaptation to maladaptation and consequently the need for personalised therapeutic approaches in humans based on molecular biomarkers of chronicity and severity of CIH, other than nocturnal oxygen desaturations, and the apnoea/hypopnoea index.

The phenotype of arterial hypertension induced by chronic intermittent hypoxia
HTN emerges in OSA patients experiencing five apnoeic events per hour and therefore even in mild disease (Bouloukaki et al., 2020).A dose-response relationship between OSA and HTN is well documented (Hou et al., 2018;Peppard et al., 2000) with a direct association between the severity of apnoea/hypopnoea index (Peppard et al., 2000;Yildiz et al., 2022) and nocturnal oxygen desaturation (Wang et al., 2023) with increased blood pressure (BP).Indeed, HTN was the first maladaptive response directly attributed to CIH (Fletcher, Lesske, Qian et al., 1992).Since then, the evidence produced in animal models and clinical studies in OSA patients and/or in healthy volunteers has allowed the identification of several phenotypic characteristics of this form of HTN, summarized as follows.
CIH increases both systolic and diastolic BP.Increases in both systolic (SBP) and diastolic (DBP) blood pressure caused by CIH are reported in healthy humans and rodents.In healthy human volunteers, very mild CIH increases SBP by 4.2-7.1% and DBP by 6.3% (Gilmartin et al., 2010;Tamisier et al., 2011).In rodents, the cardiovascular outcome of exposure to CIH, first described by Fletcher et al. (1992), was a rise of 14 mmHg (intra-arterially) or 21 mmHg (tail cuff) in mean BP after 35 days of exposure, but SBP and DBP values were not reported.Since then, increases in SBP and DBP have been described in rodent models between 7% and 24%, and 9% and 27% for SBP and DBP, respectively (for a review see AlMarabeh et al., 2019).These changes correspond to increases in mean BP of 8-37 mmHg, mainly depending on the method used to record BP and duration of CIH exposure (for a review see Diogo & Monteiro, 2014).
CIH stimulus is during sleep-time but high BP is maintained during wake-time.OSA patients present high blood pressure and high levels of sympathetic activation that persist even in the absence of the apnoea/CIH stimulus (Somers et al., 1995).The prevailing effects of CIH exposure during sleep-time on wake-time HTN has been also shown in rodent models since 1992 (Fletcher, Lesske, Qian et al., 1992) to date (Pimpão et al., 2023) and in healthy human volunteers (Tamisier et al., 2011).Unlike CIH, arousals from sleep elevate blood pressure acutely, but have no persistent effect during wake-time (Brooks et al., 1997;Fletcher, 2001).
CIH is associated with a non-dipping profile of BP during sleep.Exposure to CIH impairs the physiological fall in sympathetic activity that occurs during non-rapid eye movement (NREM) sleep, which determines a physiological drop in BP >10% (Rosenzweig et al., 2023;Silvani & Dampney, 2013).The loss of nocturnal dipping is frequently observed in OSA patients (Seif et al., 2014) and mimicked in CIH-exposed rodents (Fletcher et al., 1999;Marciante et al., 2020;Pimpão et al., 2023).Loss of the dipping pattern of BP poses a considerable risk to patients (Salles et al., 2016).
Increase in BP varies with the intensity and duration of CIH.In rodents, the increase in BP is gradual and seen from the third day of exposure to CIH (Marcus et al., 2009(Marcus et al., , 2012)), but the development of stable nocturnal and diurnal HTN was found between week 2 and 5, depending on the CIH paradigm and type of BP measurements (AlMarabeh et al., 2019;Coelho et al., 2020;Fletcher, Lesske, Qian et al., 1992;Takahashi et al., 2018).In humans, it is difficult to establish the total duration of OSA disease (Farré et al., 2022).In healthy volunteers, reversible increases in BP have been seen after 2-4 weeks of exposure to CIH (Gilmartin et al., 2010;Tamisier et al., 2009Tamisier et al., , 2011)).In male rats, increases in mean arterial BP ranged from 6.7% (10% O 2 , 15 cycles/h, 12 h/day, for 14 days) to 18.9% in animals exposed to 2-5% O 2 , 48 cycles/h, 6 h/day, for 30 days.Healthy volunteers exposed to 30 cycles/h of CIH (13% O 2 , 8 h/day, for 14 days) also showed higher BP increases than those exposed to 17 cycles/h (13% O 2 , 9 h/day, for 28 days) (AlMarabeh et al., 2019).

CIH-induced hypertension occurs in males and females.
Most of the in vivo studies exploring the relationship between CIH and HTN were performed in male animals, a sex bias which is not uncommon in biomedical science, but may also be partly related to the fact that men are up to 4 times more likely to have OSA, compared to women (Peppard et al., 2013).Of note, no sex differences in the relationship between HTN and OSA are reported in OSA patients (Bixler et al., 2000).Whether sex differences exist in animal models has not been systematically analysed but using a paradigm of CIH (6% O 2 , 6 cycles/h, 8 h/day) for 35 days, the hypertensive response (mean arterial pressure) in female rats was 7-9 mmHg (Souza et al., 2015) and 9-12 mmHg in male rats (Souza et al., 2016;Zoccal et al., 2009).In addition, upon exposure to a 7-day paradigm of intermittent hypoxia (3 min of normoxia (21% oxygen) alternating with 3 min of 10% oxygen between 08.00 and 16.00 h, with remainder of the day at normoxia), mean blood pressure increased to a lesser extent in intact females (1.6 ± 0.6 mmHg, n = 11) than in females studied after ovariectomy (5.1 ± 1.1 mmHg, n = 6) or males (5.4 ± 1.0 mmHg, n = 10) (Hinojosa-Laborde & Mifflin, 2005).
CIH relationship with HTN is strongest in younger participants.Age is an important factor in the development of CIH-induced hypertension.Young animals aged 3 months exposed to 14 days of CIH showed increases in mean arterial pressure from 101.9 ± 7.2 to 150.9 ± 25.4 mmHg, which were not observed in animals aged 24 months exposed to the same protocol (128.1 ± 26.9 to 132.9 ± 16.0 mmHg) (Quintero et al., 2016).In humans, the prevalence of OSA increases with age, but the cardiovascular significance decreases with age and the relationship between HTN and OSA is strongest in young participants, especially those of normal weight (Bixler et al., 2000;Haas et al., 2005).

Hypertension may persist even after cessation of daily
CIH exposure.There is consensus that exposure to CIH only during sleep increases blood pressure during the waking phase, but what remains unclear in rodents is whether hypertension is reversed when nocturnal CIH is discontinued.Coelho et al (2020) demonstrated that after 6 weeks of exposure of rats to a very mild paradigm of CIH, cessation did not restore basal levels of BP.Probably, reversibility of CIH-induced HTN is dependent on the chronicity and severity of previous exposure but this has been poorly studied in animal models.In humans, the efficacy of CPAP in reducing BP is variable from one individual to another and is currently unpredictable (Muxfeldt et al., 2015;Sánchez-de-la-Torre et al., 2015).
Evidence of new anti-hypertensive drugs studied in CIH models is scarce.In rodent models of CIH, information about the efficacy of drugs in reversing established hypertension is scarce.Most of the drugs (ibuprofen, ascorbic acid, bosentan, allopurinol, apocyanin, l-propargylglycine) tested in animal models were given concurrently with CIH exposure (Del Rio et al., 2010, 2012;Marcus et al., 2010;Morgan et al., 2016;Yuan et al., 2016).Thus, they were used for prevention and to delineate pathophysiological mechanisms rather than for treatment of already established cardiovascular and renal maladaptations.Nevertheless, examples of drugs that reversed already established HTN (>21 days of exposure) in rodent models include losartan (Fletcher et al., 1999), tempol (Míčová et al., 2017) and the aryl hydrocarbon receptor (AhR) antagonist (CH-223191) (Coelho et al., 2020).These are preliminary/proof of concept studies, limited to one dose and short-term administration (days) and remain far from benefit-risk approaches to drug screening for potential therapeutic application.
In summary, early diagnosis of OSA is of the utmost importance but unfortunately, even when the diagnosis is made, it remains unclear for how long OSA has been established and the degree of cardiovascular maladaptation.The evidence available on drug targetable mechanisms and biomarkers of maladaptation induced by CIH in vessels, heart and kidney is discussed in the next sections.

Mechanisms of CIH-induced vascular alterations: focus
on inflammation and oxidative stress.CIH-dependent mechanisms underpinning vascular dysfunction are multiple and intertwined, converging on disruption of protective NO signalling (Fig. 1).CIH-induced inflammation and oxidative stress have been identified as key mechanisms contributing to vascular dysfunction in OSA patients (Stiefel et al., 2013).
From cardiac remodelling to heart failure.Pathological cardiac remodelling is an early mechanism that occurs before the onset of heart failure and is characterized by left ventricular (LV) hypertrophy, increased fibrosis and apoptosis, leading to ventricular stiffness, risk of arrhythmias and impaired cardiac function (Martin et al., 2023).Systematic reviews and meta-analysis highlight J Physiol 601.24 LV hypertrophy and dilatation in OSA patients, which appears related to hypoxia severity (Cuspidi et al., 2020(Cuspidi et al., , 2021;;Ogilvie et al., 2020).This structural remodelling is accompanied by myocardial fibrosis and increased plasma levels of galectin-3 that also correlate with the severity of the disease (Cicco et al., 2021;Shah et al., 2020;Singh et al., 2019;Wang et al., 2021).CIH-related cardiac remodelling (Fig. 2) could represent the first step triggering subsequent contractile dysfunction in OSA patients (Ogilvie et al., 2020;Yu et al., 2020).In the context of acute coronary syndrome, the prevalence of sleep disturbances is estimated to be up to 60% (Ludka et al., 2014), and patients with sleep disorders exhibit less myo-cardial salvage, structural heart remodelling and increase in infarct size (Buchner et al., 2014(Buchner et al., , 2015)).On the other hand, some clinical data suggest that patients with OSA could be preconditioned by CIH, and thus could respond favourably after an acute coronary event (Ludka et al., 2017;Shah et al., 2013).Recent meta-analysis of studies performed in rodents reported that, as observed in OSA patients, CIH induces cardiac remodelling (i.e.LV hypertrophy, LV dilatation, interstitial fibrosis and apoptosis), associated or not with contractile dysfunction, as well as a dichotomous response to myocardial ischaemia and reperfusion, with both an increase and a decrease in infarct size (Belaidi et al., 2022;Hu et al., 2020).This

Figure 2. Schematic representation of the deleterious impact of CIH on myocardium
Intermittent hypoxia induces a dichotomous response on myocardium, depending on both the intensity and the duration of the hypoxic stimulus.Acute and short-term CIH induces sympathetic activation, HIF-1 activation and oxidative stress that are associated with early adaptive effects on myocardium (i.e.increased contractile function, myocardial perfusion and decreased infarct size).On the other hand, CIH and long-term activation of these mechanisms drive maladaptive responses in myocardium, inducing pro-apoptotic ER stress, alterations of calcium handling and mitochondrial disruption resulting in myocardial damage (i.e.cardiac arrhythmia, sudden death, left ventricular pathological remodelling, contractile dysfunction and increased infarct size).Casp-3, caspase 3; CIH, chronic intermittent hypoxia; CHOP, C/EBP homologous protein; EPO, erythropoietin; ER, endoplasmic reticulum; ET-1, endothelin-1; HIF-1, hypoxia-inducible factor-1; iNOS, inducible nitric oxide synthase; LV, left ventricular; MAM, mitochondria-associated ER membrane; NOX-2, NADPH oxidase 2; Nrf2, nuclear factor erythroid-2-related factor 2; VEGF, vascular endothelial growth factor; SOD, superoxide dismutase.
divergent effect of CIH on myocardium could rely on several factors, such as species (Yin et al., 2012), age (Castro-Grattoni et al., 2020, 2021;Wei et al., 2022) or the presence of comorbidities (Détrait et al., 2021;Rodriguez et al., 2014).In addition, the pattern of CIH (intensity, number of cycles and duration) likely represents a major determinant of the cardiac response to CIH.Whereas short-term intermittent hypoxia exposure (from 1 day to 4 weeks) exerts either no effect on cardiac function or a sympathetic-dependent increase in cardiac contractility in mice (Détrait et al., 2021;Guo et al., 2015;Naghshin et al., 2009Naghshin et al., , 2012;;Rodriguez et al., 2014;Wei et al., 2022), cardiac dysfunction seems to be closely dependent on the duration of the exposure to intermittent hypoxia (more than 4 weeks) (Chen et al., 2005(Chen et al., , 2008(Chen et al., , 2010;;Ding et al., 2014;Guo et al., 2015;Wei et al., 2022;Williams et al., 2010).Similarly, in rats, it was demonstrated that a 2-week exposure to mild CIH increased cardiac output (Lucking et al., 2014), with evidence of a β 1 -adrenoceptor increase in cardiac contractility evident after 3 days of exposure to intermittent hypoxia (Marullo et al., 2023).In the context of ischaemic disease, intermittent hypoxia also induces a dichotomous response, with acute, moderate (>7% F iO 2 ) and short-term (<4 h per day) exposure resulting in a decrease in infarct size, whereas severe (<7% F iO 2 ) and chronic exposure (several days) results in an increase in infarct size (Belaidi et al., 2022).These results from rodent studies confirm the biphasic character of the cardiac response to intermittent hypoxia, with a transition from an early cardiac adaptation to a later maladaptation (Belaidi et al., 2022).This raises a crucial question in the era of personalized medicine and prediction of the trajectories of OSA complications: could characterisation of the hypoxic pattern help predict those patients that will develop detrimental consequences?This discussion started some years ago (for a review see Randerath et al., 2019), but remains unresolved to date.

Mechanisms involved in the cardiac response to CIH
Oxidative stress and inflammation.As described for blood vessels, CIH induces the production of ROS in rodent myocardium (Bai et al., 2021;Ramond et al., 2013;Totoson et al., 2013;Yeung et al., 2015;Yuan et al., 2015;Zhou, Wang, et al., 2014, 2017) and in H9c2 cardiomyoblasts (Bai et al., 2021).In addition, therapeutic strategies (i.e.methallothionein, atorvastatin, melatonin, telmisartan and others) targeting CIH-induced oxidative stress and associated cardiac inflammation prevents myocardial apoptosis and fibrosis (Han et al., 2014;Yeung et al., 2015;Yuan et al., 2014Yuan et al., , 2015;;Zhou, Wang, et al., 2014).CIH-induced oxidative stress could contribute to both the early adaptive and late maladaptive cardiac response to intermittent hypoxia (Lavie, 2015).Indeed, very short-term intermittent hypoxia exposure (3 days) induces an increased cardiac expression of nuclear factor erythroid-2-related factor 2 (Nrf2) and SOD, as compensatory mechanisms preventing early oxidative damages and dysfunctions.In contrast, long-term exposure to CIH (more than 4 weeks) decreases the expression of Nrf2 and SOD, increases oxidative damage and mitochondrial pro-apoptotic pathways, which result in mitochondrial fragmentation and apoptosis, but also contribute to cardiac inflammation, fibrosis and finally contractile dysfunction (Han et al., 2018;Pai et al., 2016Pai et al., , 2022;;Zhou et al., 2017).Similar mechanisms have been proposed to explain the differential response to intermittent hypoxia observed in young and aged animals (Wei et al., 2022).Finally, in the context of myocardial ischaemia, oxidative stress contributes to the increase in infarct size following severe CIH protocols, and antioxidant treatments administered during the CIH exposure prevent the CIH-induced increase in infarct size (Ramond et al., 2013;Totoson et al., 2013).

Sympathetic
activation.As previously stated, sympathetic hyperactivity (Fig. 2) has been extensively studied in CIH models and humans (Puri et al., 2021).This autonomic dysfunction is driven by oxidative stress and HIF-1 activation (Kumar et al., 2006;Peng et al., 2006), and contributes to the early intermittent hypoxia-induced increase in cardiac contractility (Naghshin et al., 2009).However, CIH-induced persistent sympathetic over-activity triggers a switch from an initial adaptive to a late maladaptive response.This results in alterations of ventricular repolarization and occurrence of lethal ventricular arrhythmias (Morand et al., 2018), increased infarct size (Bourdier et al., 2016), and finally exacerbation of ischaemic-dependent cardiac remodelling and contractile dysfunction (Bourdier et al., 2020).

Kidney phenotypes associated with CIH
Exposure to CIH potentially elaborates direct and indirect effects in the kidney that initially contribute to the development of HTN and later to chronic kidney disease (CKD).Direct and lasting effects may be caused by local hypoxia and/or re-oxygenation cycles associated with oxidative stress and inflammation of renal tissues (AlMarabeh et al., 2019).Indirect effects refer mainly to carotid body-mediated efferent renal sympathetic nerve activation (Fig. 3).Activation of β1-adrenoceptors in the juxtaglomerular cells triggers the release of renin and the subsequent activation of the angiotensin and aldosterone cascade.Ultimately, catecholamines and angiotensin II perpetuate the vicious cycle of oxidative stress and inflammation in vessels, carotid body, heart, kidney and brain.
Like cardiovascular tissues, the pathophysiological effects of CIH in the kidney appear to depend on the paradigm of intermittent hypoxia.In a model of moderate CIH exposure (6 % O 2 , 12 cycles/h, for 8 h/day, for 14 days), it was determined that CIH can: (1) elicit increased BP in the absence of overt kidney injury; (2) does not impair the low-pressure baroreflex control of renal sympathetic nerve activity; (3) has modest effects on the high-pressure baroreflex control of renal sympathetic nerve activity, most likely indirectly due to HTN; but (4) impairs diuretic and natriuretic responses to fluid overload (AlMarabeh et al., 2021).The importance of local kidney inflammation to aberrant stimulation of renal afferent nerves signalling to the nucleus of the solitary tract has been discussed in the context of the pathophysiology of CIH-induced HTN (AlMarabeh et al., 2019, 2022).Local kidney inflammation may be driven by altered haemodynamics and decreased tissue oxygen tension (P O 2 ).The effect of CIH on glomerular filtration and perfusion pressure is variable with no changes reported in one study (O'Neill et al., 2019), whereas hyperfiltration and reduced renal blood flow (RBF) is reported in a more recent study (Kious et al., 2023), perhaps related to differences in the pattern of CIH, state (anaesthetized vs. conscious) and/or the use of different anaesthetic agents in the two studies.However, in both studies, there is consensus that exposure to a moderate paradigm of CIH (6.5-10.0%O 2 , 10-20 cycles/h, 8 h/day for 2 weeks) causes a reduction in renal cortical and medullary P O 2 (Kious et al., 2023;O'Neill et al., 2019) Interest in the direct molecular effects of CIH on the kidneys of adult rodents emerged in 2012 and has grown since 2016 (Coelho et al., 2020;Correia et al., 2021;Ding et al., 2016;Guan et al., 2019;O'Neill et al., 2019;Wu et al., 2018Wu et al., , 2020;;Zhang, Cai et al., 2019;Zhao et al., 2021).However, owing to significant differences in the CIH paradigms used in these studies (frequency of bouts, intensity of hypoxia and total duration of exposure to intermittent hypoxia), it is difficult to arrive at a clear consensus.There is convincing evidence in rodents that long-term and/or severe CIH paradigms (>30 cycles/h and/or >8 h/day and/or <10% O 2 and/or >4 weeks) evoke detrimental processes in kidney morphology and function (Ding et al., 2016;Sun et al., 2012;Wu et al., 2020;Zhang, Su et al., 2019;Zhao et al., 2021).In rats, inefficient O 2 utilization in the kidney and sustained decreases in cortical P O 2 occurs following long-term (2 weeks) but not acute intermittent hypoxia (4 h) (O'Neill et al., 2019).This finding was observed in response to exposure to a relatively mild CIH paradigm (6.5% O 2 , 10 cycles/h, 8 h/day) (O'Neill et al., 2019), and is consistent with HIF-1α stabilization (protein level) in kidney described after 8 weeks of exposure to CIH in mice (Sun et al., 2012).An increase in Hif3a mRNA, much higher than Hif1a in kidney cortex, increased VEGFa and HIF2a in kidney medulla, with no renal changes of the HIF binding partner, HIF-1β, was found in rats after 3 weeks of very mild CIH (Coelho et al., 2020).The effects of CIH on the HIF-dependent downstream renal hormone erythropoietin are unknown in animal models and contradictory findings are reported in sleep apnoea patients.In adults, erythropoietin levels are significantly elevated in some subgroups of sleep apnoea patients, with <30 kg/m 2 associated with cardiovascular complications (Zhang, Zeng et al., 2017).Sun et al. (2012) were the first to show that short-term exposure to a severe paradigm of CIH exposure in mice (8% O 2 , 30 cycles/h, 12 h/day, for 3-7 days) induced an acute renal inflammatory response with significant upregulation of Nrf2-regulated antioxidants (haem oxygenase-1 and metallothionein) and inhibition of oxidative damage.In contrast, long-term CIH (8% O 2 , 30 cycles/h, 12 h/day, for 8 weeks) induced significant renal inflammation and downregulation of antioxidants along with significant renal oxidative damage, cell death and fibrosis (Sun et al., 2012).Intermediate exposure (10% O 2 , 20 cycles/h, 8 h/day, for 2 weeks) is associated with activation of pro-oxidative and pro-fibrotic gene programs (Kious et al., 2023).Consistent with the notion of a temporal response to CIH, an increase in antioxidant defence was shown using a mild short-term CIH paradigm (5% O 2 , 5.6 cycles/h, 10.5 h/day) by Correia et al. (2021).In this model, short-term CIH (up to 7 days), wherein no or mild changes in BP were observed (Diogo, Pereira et al., 2015), increased total glutathione availability (Correia et al., 2021) and protein S-thiolation, a post-translational modification that might protect proteins from irreversible oxidation (Dalle-Donne et al.,

Figure 3. Schematic representation of the deleterious effects of chronic intermittent hypoxia in the kidney
Molecular and pathophysiological signatures in kidney are dependent on the frequency, intensity and duration of intermittent hypoxia.Short-term and very mild paradigms have been reported to induce adaptive pathways related to oxidative stress and increased markers of fibrosis without remodelling.On the other hand, long-term and moderate-to-severe paradigms are characterized by a phase (14-21 days) of severe metabolic and molecular changes that appear to be reversible.These changes include reduced antioxidant defences, increased production of reactive oxygen species (ROS), activation of the AhR-CYP1A1 axis, renin-angiotensin-aldosterone system activation, pro-inflammatory cascades, inflammasome activation, increased production of cell growth factors and iron overload.Moderate paradigms (6.5-10.0%O 2 , 10-20 cycles/h, 8 h/day, 14 days) are systematically associated with decreases in cortical and medullary P O 2 (CPO 2 and MPO 2 ) but with variable effects in renal blood flow (RBF) or glomerular filtrate rates (GFR).When the hypoxia paradigm is maintained for a prolonged period (35-60 days), it promotes irreversible damage, including cell death, renal remodelling and reduced renal function.ACR, albumin-to-creatinine ratio; BUN, blood urea nitrogen; CYP1A1, cytochrome P4501A1; EMT, extracelular matrix remodelling; ER, endoplasmic reticulum; HIF, hypoxia inducible factor (1, 2, 3); JNK, c-Jun N-terminal kinase; MAPK-p38, mitogen-activated protein kinase p38; NA, noradrenaline; NF-κB, nuclear factor κB; NQO1, NAD(P)H quinone oxidoreductase; Nrf2, nuclear factor erythroid-2-related factor 2; VEGF, vascular endothelial growth factor.
Very mild paradigms (5% O 2 , 5.6 cycles/h, 10.5 h/day) do not induce histological changes in the kidney, even after 8 weeks of exposure, but the duration of CIH is associated with an increase in the albumin-to-creatinine ratio in urine, in addition to lower food intake and water consumption, and lower body weight gain (Correia et al., 2021).In this model, with established HTN (3 weeks of CIH), increased mRNA levels of mesothelial markers (vimentin and fibronectin-1) were observed, without changes in epithelial (E-cadherin) and fibrosis (col1a1 and a-sma) markers.Therefore, mild and early stages of exposure to CIH are preferentially associated with adaptive responses to redox stress without significant morphological changes in the kidney (Bai et al., 2022;Correia et al., 2021;Zhang, Su et al., 2019).
Pharmacological and transgenic strategies in rodents have been used to elucidate the molecular mechanisms contributing to CIH-induced renal damage.Biomarkers of renal injury in response to CIH are decreased in response to blockade of angiotensin AT1 receptors (Zhang, Cai et al., 2019), administration of angiotensin 1−7 (Lu Kang, Hu, Tang, Zhou, Yu et al., 2017), blockade of cannabinoid CB1 receptors (Zhao et al., 2021), administration of recombinant sRAGE (soluble form of receptor for advanced glycation end products) protein (Wu et al., 2016), administration of adiponectin (Ding et al., 2016), inhalation of hydrogen (Guan et al., 2019) and administration of NLRP3 inhibitor to reduce apoptosis and pyroptosis (Bai et al., 2022).TLR-4 deficiency also alleviates renal injury, inflammation and fibrosis associated with exposure to CIH (Zhang, Su et al., 2019).Activated angiotensin II AT1 (Wang et al., 2018) and cannabinoid CB1 receptors (Zhao et al., 2021) are well documented stimulators of ROS production in the kidney of animals exposed to CIH.Adiponectin protects against CIH-induced renal cell apoptosis by inhibiting ROS-related ER stress, but adiponectin is known as an activator of the AMPK pathway regulating urine protein and inhibiting glycogen synthesis in the glomerulus and distal renal tubule, respectively (Ding et al., 2016).CIH promotes renal transferrin receptor and divalent metal transporter-1 expression, and hydrogen attenuates CIH-induced renal injury at least partially by inhibiting renal iron overload (Guan et al., 2019).
AhR is a ligand-activated transcription factor initially known to be activated by some environmental pollutants, which uses HIF-1β/ARNT as the binding partner of HIF (for a review see Coelho et al., 2021).Emerging putative inflammasome activators in the kidney are AhR agonists.In rats, a mild CIH paradigm activates, after 21 days, the AhR pathway in the kidney, which is associated with higher expression of markers of inflammation and epithelial-to-mesenchymal transition (Coelho et al., 2020;Correia et al., 2021).Endogenous substances now established as AhR agonists in the kidney include several tryptophan metabolites that are uraemic toxins such as indoxylsulfate (for a review see Coelho et al., 2020Coelho et al., , 2022)).The gut is the source of indole, which is processed in the liver to the AhR agonist indoxylsulfate, highlighting the AhR ligand crosstalk between gut microbiota and the kidney (Lowenstein & Nigam, 2021).Lu et al. (2020) demonstrated that Sprague-Dawley rats exposed to 6 weeks of CIH exhibited an altered gut microbiota and renal fibrosis, which were reversed by renal denervation, indicating a relationship with the sympathetic nervous system.Acute intermittent hypoxia inhibited the AhR-CYP1A1 axis in kidney cortex, which increased progressively with the duration of exposure to CIH.A link between AhR and oxidative stress in CIH has also been recently established, demonstrating that cystine, the product of cysteine oxidation, acts as an AhR activator (Correia et al., 2021).AhR interferes with the expression of clock genes (Jaeger & Tischkau, 2016), which are expressed in the kidney (Gumz, 2016).The resetting of clock genes in the kidney is implicated in the nocturnal dipping pattern of BP and is dependent on a small decrease in renal P O 2 during the inactive period (Emans et al., 2017), in a HIF-1α dependent manner (Adamovich et al., 2017).In a single dose exploratory study, AhR blockade reversed CIH-induced hypertension during the day but did not prevent the loss of BP dipping associated with CIH (Pimpão et al., 2023).
Final remarks and future directions.Cardiovascular and renal maladaptation during exposure to CIH is a significant contributor to the global burden of cardiovascular diseases.
In general, in rodent models of CIH, a common pattern of molecular mechanisms contributing to dysfunction presents in heart, blood vessels and kidney.Short term (<2 weeks) and mild paradigms induce compensatory antioxidant upregulation in response to increased generation of ROS.In contrast, long-term exposures of mild, moderate and severe intensities result in oxidative damage and also contribute to tissue inflammation, fibrosis and finally to contractile (heart and vessels) dysfunction, cardiac dysrhythmias, acceleration of atherosclerotic processes in vessels and renal dysfunction.
The factors driving transition from adaptation to maladaptation remain unresolved.Also remaining to be fully determined is the extent of the capacity to mitigate or fully reverse cardiorenal damage.It is evident that duration of exposure to CIH is a major contributor to disease progression wherein CIH elaborates 'dose-dependent' disruption to regulatory transcription factors (e.g.HIF) and non-transcriptionally regulated stimuli such as ROS and DNA methyl transferases.The intermittent pattern of exposure to hypoxia (and reoxygenation) appears to be the main driver of maladaptive signalling mechanisms, in contrast to the adaptive mechanisms that can occur in chronic sustained hypoxia (e.g.altitude).We argue that there is an urgent need for identification of molecular biomarkers of chronicity and severity of CIH in humans, other than nocturnal oxygen desaturations, and the apnoea/hypopnoea index.These markers should aid in phenotypic categorisation of disease and help to quantify the extent of maladaptation as well as contribute to informed predictions of the trajectories of downstream complications of OSA.
Collectively, these studies highlight the need for personalised approaches in the treatment of OSA patients due to heterogeneity in the exposure to CIH, relating to intensity and duration.Moreover, greater consideration of the effects of CIH in other tissues (adipose, liver, microbiota) is required given their proclivity to affect cardiovascular risk.
In closing, we draw focus to the urgent need to reconsider experimental study design in pre-clinical studies employing animal models of CIH, modelling human OSA, to not only elucidate pathophysiological mechanisms contributing to OSA-related morbidities such as cardiorenal disease, but also to advance our understanding of the efficacy of established and emerging pharmacotherapies to ameliorate or reverse CIH-induced dysfunction through designs that allow for the assessment of efficacy in the face of established disease and benefit/risk with the aim and ambition to improve cardiovascular outcome in people with sleep apnoea.Is Aberrant reno-renal reflex control of blood pressure a contributor to chronic intermittent hypoxia-induced hypertension?Frontiers in Physiology, 10, 465.Almarabeh, S., Lucking, E. F., O'halloran, K. D., & Abdulla, M. H. (2022).Intrarenal pelvic bradykinin-induced sympathoexcitatory reno-renal reflex is attenuated in rats exposed to chronic intermittent hypoxia.Journal of Hypertension, 40(1), 46-64.the final version of this manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

References
Introduction: sleep apnoea, chronic intermittent hypoxia and cardiovascular risk -relevance and unmet needs Obstructive sleep apnoea (OSA) is a highly prevalent condition that affects 936 million individuals worldwide, 0 Claire Arnaud is Investigator of the Unit of Hypoxia and Cardiovascular and Respiratory Pathophysiology at Université Grenoble Alpes (France) mainly focused on sleep disorders.Sofia Pereira (Associate Professor of Pharmacology) is based in NOVA Medical School in Lisbon (Portugal) and have complementary expertise (Biochemistry and MD) in the study of anti-hypertensive drugs in chronic intermittent hypoxia conditions.Ken O'Halloran is Professor of Physiology at University College Cork (Ireland), mainly interested in the control of breathing in health and disease.Emilia Monteiro (Professor of Pharmacology) is based in NOVA Medical School in Lisbon (Portugal) and have complementary expertise (Biochemistry and MD) in the study of anti-hypertensive drugs in chronic intermittent hypoxia conditions.

Figure 1 .
Figure 1.Schematic representation of the deleterious impact of CIH on vessel structure and function Several devices have been developed to reproduce the intermittent hypoxic component of sleep apnoea in humans, rodents and cells through induction of cyclic variations in the inspired oxygen fraction (F iO 2 ).Underlying mechanisms of CIH-induced vascular dysfunction have been identified, such as endothelial dysfunction, inflammation, oxidative stress and sympathetic hyperactivity, and they converge in the vessel on disruption of protective nitric oxide (NO) signalling, contributing to vascular remodelling, vascular tone dysfunction and atherosclerosis.ET-1, endothelin-1; ETC, electron transport chain; LF/HF, low frequency/high frequency ratio based on heart rate variability analysis; NF-κB, Nuclear factor-κB; NO, nitric oxide; NOX, NADPH oxidase; peNOS, phosphorylated endothelial nitric oxide synthase; ROS, reactive oxygen species; SOD, superoxide dismutase; XO, xanthine oxidase.