Developmental plasticity of the cardiovascular system in oviparous vertebrates: effects of chronic hypoxia and interactive stressors in the context of climate change

Oviparous (egg-laying) vertebrates typically develop in fluctuating environments with little or no parental care. This reproductive strategy has some advantages over viviparity (Shine, 2015), but it exposes the embryo to environmental stress at a critical stage of life when defence mechanisms may not be fully developed. The consequences can be severe, because environmental fluctuations during development can permanently alter organismal structure, function and behaviour, and these traits can even be inherited by subsequent generations (Sultan, 2017). Therefore, the developmental plasticity of oviparous vertebrates plays a critical role in determining their future fitness and survival. This is especially true in an era of climate change, where rising concentrations of CO₂ in the atmosphere are driving global warming and increasing the frequency and intensity of environmental hypoxia and hypercapnia (Pörtner et al., 2014). Such rapid changes in the severity, frequency and spatial scale of these stressors will significantly challenge the developmental plasticity of oviparous species. Thus, it is important to gain an understanding of both the short- and long-term consequences of environmental stress on the embryonic physiology of these vulnerable animals.

Oviparous vertebrates commonly experience hypoxia during embryonic development (Box 1). Importantly, studies across a wide range of species have shown that chronic developmental hypoxia (CDH; see Glossary) has persistent effects on the cardiovascular system of oviparous vertebrates (Fig. 1). It appears that some cardiovascular responses to CDH are well-conserved among mammals, birds, reptiles and fish (Galli et al., 2023; Tables S1, S2 and S3). However, there are many interspecific differences, and the outcome of CDH appears to be dependent on multiple factors, including the magnitude and duration of hypoxia, as well as developmental stage. Furthermore, the hypoxic response can be altered by the interactive effects of other environmental stressors, such as temperature and hypercapnia (Box 2). These interactions are becoming increasingly important in the context of climate change.

The overall aim of this Review is to discuss and compare the phenotypic consequences of CDH on the cardiovascular system of oviparous birds, fish and reptiles. We define CDH here as periods of hypoxia that last for days, weeks or months. When we refer to oxygen levels, we present values as % O₂ saturation (21% O₂=100% air saturation). We start the Review with an overview of the effects of CDH on the embryonic cardiovascular system of each vertebrate class, as well as the persistent effects on juvenile and adult life stages. Unless specified, the data we present on juveniles and adults are taken from studies that exposed embryos to CDH for a defined period during development, and then returned them to normoxia and investigated the cardiovascular phenotype in later life. These kinds of studies reveal traits that arise from persistent developmental plasticity, rather than plasticity due to acclimation (Earhart et al., 2022). Where possible, we attempt to identify species-specific responses, the threshold for response and critical windows (see Glossary). Although data are extremely limited, we also review the interactive effects of CDH, hypercapnia and temperature on cardiovascular outcomes. Surprisingly, we were unable to find sufficient literature to warrant a review of the effects of CDH on the cardiovascular system of amphibians (only one relevant paper: Fritsche and Burggren, 1996).

Much of what we know about the effects of CDH on the avian cardiovascular system comes from studies on domestic chickens (Table S1). These investigations were largely devised to improve farming practices or to study the clinical implications of CDH without the confounding influence of maternal responses (Itani et al., 2018). However, the levels of hypoxia used in these studies (13–17% O₂ saturation) are within the natural range of some avian nests (Box 1), which makes them ecologically relevant.

The most common consequence of CDH is embryonic growth restriction. In chickens, isobaric or hypobaric hypoxia (see Glossary) at 13–15% O₂ (≈2500–3500 m) consistently leads to a reduction in embryonic body mass (Table S1A), and the critical window occurs at 30–60% of incubation (Dzialowski et al., 2002; Ruijtenbeek et al., 2000). In addition, embryos from various chicken strains (broilers, red junglefowl, white Leghorn) exposed to isobaric or hypobaric hypoxia have an increased brain-to-body weight ratio (Giussani et al., 2007; Salinas et al., 2010; Skeffington et al., 2020) and/or an increased heart-to-body weight ratio (Table S1C). Asymmetric growth restriction is usually a consequence of the ‘brain-sparing’ response (Giussani, 2016), which involves a systemic vasoconstriction that shunts blood to hypoxia-sensitive organs, such as the brain and heart. Although protective in the short term, it can ultimately lead to systemic hypertension and cardiac remodelling (Giussani, 2016). Indeed, some studies have shown that isobaric or hypobaric hypoxia leads to an increase in chick embryonic heart mass, aortic wall thickness and ventricular wall thickness (Table S1B; Salinas et al., 2010; Villamor et al., 2004). However, other studies have found a decrease in heart mass, or no effect (Table S1B), and there appears to be no clear correlation between the outcome and the length or duration of the hypoxic exposure.

CDH triggers a range of responses to improve O₂-carrying capacity and delivery in vertebrates (Galli et al., 2023). Embryonic capillary density and chorioallantoic membrane (CAM) vascularity is increased in the CDH-exposed Canada goose and domestic chicken, respectively (Snyder et al., 1984; Table S1E). Haematocrit is also increased in chicken embryos exposed to CDH (13–15%, Table S1E), and the critical window lies between day 6 and 12 of development (Dzialowski et al., 2002). However, at least in the case of chickens at sea-level, the increase in embryonic O₂-carrying capacity is not enough to offset the negative effects of hypoxia, and cardiac function is compromised. In one study, hypoxia reduced chicken embryonic ventricular peak systolic pressure, dorsal aortic peak systolic pressure (see Glossary), stroke volume and cardiac output (see Glossary), while diastolic function (see Glossary) was preserved (Sharma et al., 2006). In other studies, hypoxic chick embryos had signs of cardiomyopathy, including left ventricular dilatation, reduced ventricular wall thickness, increased apoptosis (Tintu et al., 2009), a reduced left ventricular ejection fraction, aortic thickening, reduced contractility, reduced cardiac output and diastolic dysfunction (Itani et al., 2016, 2020; Jonker et al., 2015; Rouwet et al., 2002). These problems were associated with a significant increase in cardiac oxidative stress and a reduction in cardiac antioxidant capacity (Itani et al., 2016, 2020). Heart rate is generally reduced by acute hypoxia in chickens (Akiyama et al., 1999; Altimiras and Phu, 2000; Crossley et al., 2003; Mortola et al., 2010; Sharma et al., 2006; Tazawa, 1981), but it eventually returns to control values with longer hypoxic periods; Table S1D). This is despite a significant increase in adrenal concentrations of adrenaline and noradrenaline, which is associated with a greater sensitivity of cardiac β-receptors and enhanced sympathetic innervation in the peripheral vasculature (Table S1H).

The effects of CDH on chicken embryonic cardiac structure and function are associated with multiple cellular abnormalities. Ventricular protein content and protein/DNA ratios are reduced in hypoxic embryonic chickens, which is associated with a reduction in heart mass (Asson-Batres et al., 1989). In another study, CDH initially caused cardiac myocyte hyperplasia in chicken embryos, but this eventually led to hypertrophy (see Glossary) with more myofibrils, larger Golgi complexes, less glycogen and fewer, larger secretory granules (see Glossary; Maksimov and Korostyshevskaia, 2012). This response was also accompanied by an increase in cardiac collagen (Table S1G), and a decrease in myosin heavy chain and titin proteins (Tintu et al., 2009). There is also reduced expression of genes involved in cardiac calcium handling, as well as a shift from compliant to stiff isoforms of titin and increased vascular endothelial growth factor (VEGF) expression (Jonker et al., 2015; Tintu et al., 2009). CDH also increases mitochondrial-derived oxidative stress in the hearts of chicken embryos and reduces mitochondrial efficiency and capacity (Table S1F).

Numerous studies have shown that developmental hypoxia has long-term effects on the avian cardiovascular system. Firstly, the growth restriction and increased heart-to-body weight ratio associated with CDH often continues into adulthood in chickens (Table S1A; Lindgren and Altimiras, 2013). Changes in absolute heart mass are often absent after hatching or later in life, indicative of a degree of cardiac plasticity (Table S1B). However, the cardiomyopathy in hypoxic embryonic chickens observed by Tintu et al. (2009) persists into adulthood, with severe left ventricular dilatation, decreased left ventricular ejection fraction, fibrosis and diastolic dysfunction. Lindgren and Altimiras (2013) showed that adult chickens exposed to CDH have signs of systolic, but not diastolic, dysfunction and increased expression of B1 adrenoreceptors without any change in collagen content. Lastly, Skeffington et al. (2020) found a range of cardiovascular abnormalities in adult chickens exposed to CDH, including hypertension, increased cardiac work, enhanced baroreflex gain (see Glossary), left ventricular wall thickening and increased contractility. Overall, adult chickens exposed to CDH share many of the pathological cardiovascular signatures observed in adult mammals from hypoxic pregnancies (Table S3; Itani et al., 2018), and the phenotype is reminiscent of both compensated and decompensated heart failure. This makes chickens an excellent model for studying the programming of cardiovascular disease by CDH in the absence of confounding maternal factors.

Many embryonic reptiles experience CDH as part of their natural development, but climate change is increasing the frequency and intensity of these events (Boxes 1 and 2). It is particularly interesting to study cardiovascular programming in reptiles, because cardiac design differs substantially between the reptilian classes (Burggren et al., 2020). Most turtles (testudines), snakes and lizards (squamates) have a single undivided ventricle with no pressure separation between the pulmonary and systemic circulations. However, monitor lizards and pythons have a functionally divided ventricle, and crocodilians (alligators, crocodiles, caimans and gharial) have a fully divided ventricle, allowing for high systemic arterial pressures and an elevated metabolic rate. These differences place variable metabolic demands on the reptilian cardiovascular system, which could be expected to lead to species-specific responses to CDH.

As in chickens, there is no effect of CDH on reptilian embryonic body mass if the O₂ concentration is at or above 17% (Table S1A). However, isobaric and hypobaric hypoxia at O₂ concentrations of 10–15% consistently reduces embryonic body mass and/or body length in American alligators, snapping turtles, Florida red-bellied turtles, common wall lizards, viperine snakes and leopard geckos; but total incubation time is unchanged (Table S1A). The critical window for growth restriction is between 70 and 90% incubation in American alligators (Tate et al., 2016), whereas embryonic mass in common snapping turtles is dependent on the total hypoxic exposure time (Tate et al., 2015). Hypoxia also causes an increase in the total amount of yolk present at the end of development in American alligators, Florida red-bellied turtles, common wall lizards and viperine snakes; indicating a reduced conversion of yolk to tissue (Crossley et al., 2017; Crossley and Altimiras, 2005; Kam, 1993; Owerkowicz et al., 2009).

CDH is also associated with an increase in heart-to-body weight ratio in American alligators, snapping turtles and Florida red-bellied turtles (Table S1C). The critical window for the response in American alligators is at 20–40% of development (Tate et al., 2016), and 50–70% in common snapping turtles (Tate et al., 2015). The asymmetric growth restriction suggests that reptiles exhibit the brain-sparing response, which is supported by a recent study that found a modest increase in brain blood flow in embryonic turtles exposed to CDH (Sartori et al., 2019). An increase in absolute heart mass is also evident in hypoxic embryonic snapping turtles, lizard geckos and common wall lizards (Table S1B). However, in most studies, absolute heart mass does not change with hypoxia, suggesting that heart growth is preserved at the expense of somatic growth. Nevertheless, Crossley's laboratory thoroughly investigated the critical windows for this response and showed that cardiac enlargement occurs before somatic growth restriction (Tate et al., 2015, 2016). This finding suggests that cardiac enlargement in reptiles is a direct response to CDH, rather than a consequence of reduced somatic growth.

In contrast to birds, CDH leads to long-term changes in reptilian heart rate, but the responses are species-specific. CDH causes bradycardia in American alligator embryos (70–90% development, 10% O₂) and common wall lizards chronically exposed to high-altitude hypoxia [15–17% O₂ sea-level equivalent (SLE, see Glossary); Table S1D], but it causes a significant tachycardia (see Glossary) in embryonic snapping turtles (10% O₂) and scincid lizards (Table S1D). The underlying reason for these species-specific differences is unknown, and it is also unclear why reptiles modulate heart rate during CDH, whereas mammals and birds do not (Table S1D; Table S2C).

As in birds, chronic levels of hypoxia in embryonic reptiles trigger adaptive cardiovascular responses that improve O₂-carrying capacity and delivery. American alligators and Florida red-bellied turtles increase haematocrit during chronic hypoxia exposure (Kam, 1993; Warburton et al., 1995), but haemoglobin isoform expression and affinity is unchanged (Bautista et al., 2021; Grigg et al., 1993). CDH also increases angiogenesis in the CAM in American alligators (Corona and Warburton, 2000), which lowers the resistance of the chorioallantoic circulation by adding parallel vascular beds. This response ultimately reduces systemic blood pressure (Crossley and Altimiras, 2005; Eme et al., 2011b, 2013). The critical window for hypotension is at 20–70% of development in snapping turtles and 50–70% in American alligators (Tate et al., 2015, 2016). However, despite arterial hypotension, blood flow to the American alligator CAM increases during hypoxia, which presumably serves to enhance gas exchange (Eme et al., 2011a; Sartori et al., 2019). Given that total blood flow remains constant, the increase in CAM blood flow may be driven by increased intraembryonic vascular resistance, which could also explain the observed cardiac enlargement in snapping turtles and lizard geckos (Eme et al., 2021; Parker and Dimkovikj, 2019).

In addition to baseline changes in cardiovascular function, CDH alters the embryonic cardiovascular response to acute hypoxia in reptiles. American alligator and snapping turtle embryos exposed to CDH have an attenuated response to an acute hypoxic challenge, with blunted heart rate and blood pressure responses (Crossley and Altimiras, 2005; Eme et al., 2011b). In agreement with these findings, critical O₂ tension (P_Crit; see Glossary) is lower in snapping turtles and American alligator embryos exposed to CDH, compared with their normoxic counterparts (Crossley et al., 2017; Kam, 1993). However, the enhanced hypoxia tolerance does not appear to be associated with mitochondrial remodelling (Galli et al., 2016). Collectively, these results suggest that embryos exposed to CDH are less responsive to acute hypoxic stress and may tolerate lower levels of hypoxia.

Most of our understanding of the long-term effects of CDH have come from studies on American alligators and common snapping turtles. It is interesting to compare and contrast these two reptiles because crocodilians are archosaurs and more closely related to birds than testudines and squamates (Brusatte et al., 2010). Given that crocodilians also have a fully divided heart and higher metabolic rates, one may expect American alligators to respond to CDH more similarly to birds than to snapping turtles.

Juvenile American alligators and snapping turtles exposed to CDH most commonly experience catch-up growth, but some studies have reported persistent growth restriction (Table S1A), as well as an increased heart-to-body weight ratio (Crossley et al., 2022; Galli et al., 2016, 2021; Joyce et al., 2018; Ruhr et al., 2021; Smith et al., 2023, 2019). Despite cardiac enlargement, most resting cardiovascular parameters in juvenile American alligators and snapping turtles are similar between individuals from normoxic or hypoxic incubations. In particular, the systemic and pulmonary hypertension, as well as systolic and diastolic ventricular dysfunction that is often present in mammals and birds exposed to CDH appears to be absent in American alligators and turtles (Table S3). However, there are some reptilian cardiovascular parameters that are permanently affected by CDH. Left ventricular stroke volume is increased and pulmonary blood flow is decreased in juvenile American alligators exposed to CDH (Joyce et al., 2018; Smith et al., 2019). Likewise, heart rate is reduced, and total cardiac output is increased in juvenile common snapping turtles exposed to CDH (Wearing et al., 2017, 2016).

More differences in the long-term cardiovascular phenotype are revealed when reptiles are placed under physiological stress. Compared to normoxic controls, juvenile American alligators from hypoxic incubations that are swimming or stimulated with β-adrenergic agonists (see Glossary) have a faster rate of ventricular relaxation, greater left ventricle stroke volume, increased carotid blood flow and lower pulmonary blood flow (Joyce et al., 2018; Smith et al., 2019). Furthermore, the blunted cardiovascular response to acute hypoxia that is observed at the embryonic level is also present in juvenile alligators, suggesting a long-term improvement in hypoxia tolerance (Crossley et al., 2022, 2023; Smith et al., 2019). This is also the case for juvenile turtles exposed to CDH, as they are able to maintain cardiac output two-fold higher than controls during 2 h of anoxia (Ruhr et al., 2021). The improved anoxia tolerance is also apparent at the cellular level, and is associated with increased myofilament calcium sensitivity, a superior ability to suppress cardiac myocyte reactive oxygen species (ROS) production during anoxia and lower basal cardiac ROS production (Galli et al., 2021; Ruhr et al., 2019). These adaptations could be useful for turtles in juvenile and adult life stages, as they often experience long bouts of anoxia and reoxygenation following breath-hold dives and overwintering under ice-covered lakes (Jackson, 2002). Exposure to CDH also affects the response to digestion in snapping turtles. Compared with controls, peak postprandial metabolic rates are higher in juvenile turtles exposed to CDH (suggesting an increased metabolic cost of digestion) and this is supported by higher systemic blood flows (Wearing et al., 2017).

The cellular and molecular mechanisms driving cardiovascular programming in reptiles may involve mitochondrial remodelling, as CDH appears to improve mitochondrial efficiency in American alligators and snapping turtles, and this is driven by a lower proton leak (Galli et al., 2016, 2021). Furthermore, CDH induces substantial changes in the cardiac proteome of American alligators prior to hatching, and these changes are largely maintained into juvenile life, with animals from hypoxic incubations showing a shift in protein synthesis (transcription and translation), cellular organization, metabolic adjustments and protein degradation (Alderman et al., 2019). Proteins involved in metabolism are particularly enriched in juvenile alligator hearts from hypoxic incubations, including those with roles in fatty acid oxidation, the citric acid cycle and oxidative phosphorylation (see Glossary). Also worth noting is an increased protein expression of the antioxidant superoxide dismutase, which – in addition to the improved ability to recycle proteins – may help to manage ROS production (Alderman et al., 2019). Finally, we have recently shown that cardiac programming by CDH in snapping turtles is supported by differential expression and DNA methylation of genes associated with sarcomere function, ion-channels, cardiomyocyte survival and heart rate (Ruhr et al., 2021).

In summary, it is clear that CDH programmes the cardiovascular physiology of American alligators and snapping turtles, but in contrast to birds and mammals, the phenotype is not overtly dysfunctional; in fact, in many cases, it appears to be beneficial. The fact that these two species lack many of the pathological signatures associated with CDH (Table S3) suggests the long-term outcome of CDH may be more dependent on body temperature and metabolic rate, rather than phylogeny. It is possible that the higher metabolic costs associated with endothermy place an additional metabolic burden on juvenile and adult birds and mammals exposed to CDH, leading to pathological outcomes.

Among the vertebrate classes, fish are prone to experiencing the most severe levels of hypoxia during development, particularly in climate change scenarios (Box 1). Previous work has shown that CDH alters a wide range of phenotypic traits in teleosts, including metabolic rate (Del Rio et al., 2021), swimming performance (Johnston et al., 2013), sex ratios (Robertson et al., 2014), the balance of sex hormones (Shang and Wu, 2004) and brain development (Mikloska et al., 2022). Nevertheless, surprisingly little is known about the effects of CDH on the teleost cardiovascular system. Comparisons with the other oviparous classes is also difficult because the levels of hypoxia used in fish studies are considerably more severe than those used in studies of reptiles and birds.

Similarly to the other vertebrate classes, fish embryos or larvae exposed to CDH have reduced body mass (Table S1A), which renders the individuals less competitive and more vulnerable to predation (Mason, 1969). The growth restriction is driven by the activation of hypoxia inducible factor (HIF), which ultimately supresses the insulin-like growth factor (IGF) pathway (Kajimura et al., 2004; Sun et al., 2011). Fish embryos exposed to CDH also have reduced developmental rates, delayed hatching and delayed heart morphogenesis (Bagatto, 2005; Ciuhandu et al., 2005; Del Rio et al., 2021; Kajimura et al., 2005; Miller et al., 2011, 2008). These effects are particularly prevalent when fish are exposed to hypoxia in the later embryonic stages, which is presumably due to the increasing O₂ demands of the developing organism and the O₂-diffusion limitations across the egg membrane (Rombough, 1988).

Although acute hypoxia exposure slows growth and delays development during embryogenesis, upon reoxygenation, hypoxia-exposed embryos often (but not always) return to the same size as control animals (Table S1A). Zebrafish embryos exposed to ∼1–2% O₂ from 24 to 36 hours post-fertilisation (hpf) are shorter than control animals, but the embryos catch up if they are returned to normoxia (Kamei et al., 2018). The catch-up growth in zebrafish embryos is mediated in part by the IGF pathway (Kamei et al., 2011). Specifically, IGF pathway activity, stimulated by insulin receptor substrate 1 (IRS1)-mediated IGF signalling, helps maintain neural crest cell populations during hypoxia (Kamei et al., 2018). Reductions in neural crest cell numbers – either through ablation or by a combination of hypoxia and reduced IRS1-stimulated IGF signalling – prevents catch-up growth upon reoxygenation in zebrafish (Kamei et al., 2018).

To our knowledge, the effects of CDH on cardiac mass in embryonic/larval fishes have not been directly studied, but there have been measurements of ventricular volume. In zebrafish, hypoxia (3% O₂) leads to a reduction in ventricular end diastolic and systolic volume at 96 hpf, but an increase at 5 days (Table S1B). This suggests that hypoxia initially causes a reduction in heart size in embryonic zebrafish, but cardiac enlargement occurs once they reach the larval stages. Interestingly, in vivo imaging of zebrafish larvae has shown that brain blood flow is unchanged by hypoxia (Schwerte et al., 2003), despite an overall redistribution of blood to the red layer of muscle to enhance O₂ uptake at 7 days post fertilisation (dpf). This suggests that although blood flow distribution is changed, the brain-sparing effect is absent (El–Fiky and Wieser, 1988). Although these studies have only been performed on one species, it is possible that the brain-sparing effect is unnecessary in fish. Instead, blood is redistributed towards the muscle to enhance O₂ uptake to the body.

Like other vertebrates (Galli et al., 2023), embryonic and larval fishes exposed to hypoxia trigger mechanisms to enhance O₂ extraction. Stage-matched comparisons reveal a greater expression of the higher O₂ affinity form of embryonic haemoglobin in fish incubated in hypoxia (6% O₂) compared with those in normoxia (Bianchini and Wright, 2013). Similarly, erythropoiesis is stimulated from 7 dpf in hypoxic zebrafish larvae (Schwerte et al., 2003), and intersegmental blood vessel vascularisation is increased from 6 dpf (Yaqoob and Schwerte, 2010). O₂ extraction may also be enhanced through the activation of O₂-sensitive transcription factors, such as HIF. Lake whitefish and zebrafish embryos and larvae show hypoxia-induced, stage-specific changes in the expression of HIF1a and its associated downstream targets, which are known to stimulate haematopoiesis (Wang and Semenza, 1996) and angiogenesis (Iyer et al., 1998), and have been shown to enhance hypoxia tolerance in early life in some studies (Mandic et al., 2020; Robertson et al., 2014; Whitehouse and Manzon, 2019), but not others (Levesque et al., 2019). Finally, behavioural adaptations may also lead to increased O₂ extraction. For example, hypoxia (3% O₂) has been shown to induce pectoral fin motions in zebrafish (from 2 dpf) to aid O₂ uptake (Jonz and Nurse, 2005), and acute hypoxia exposure causes suppression of O₂ uptake while simultaneously increasing tail beat frequency – potentially in an attempt to reoxygenate the egg case – in little skate embryos (Di Santo et al., 2016).

In addition to increasing O₂ extraction, embryonic and larval fishes can also increase O₂ transport to the tissues through alterations in cardiovascular dynamics. During early embryogenesis under normal conditions, fishes rely on diffusion for the supply of O₂ to their respiring tissues (Burggren, 2004; Grillitsch et al., 2005). This has been demonstrated in developing zebrafish where, prior to ∼14 dpf, reducing the blood's O₂-carrying capacity elicits no changes in either cardiac output or anaerobic metabolism, implying that under standard conditions, there is no essential role for convective O₂ (Jacob et al., 2002). However, this is not the case under hypoxic conditions. Zebrafish incubated in hypoxia (∼10% O₂) display greater heart rate and cardiac output than those in normoxia from 4 dpf onwards, which is likely to increase convective O₂ transport and act to complement the O₂ obtained through diffusion to meet the organism's total O₂ demand (Grillitsch et al., 2005; Jacob et al., 2002). Interestingly, this implies that the afferent nervous system can sense and respond to hypoxia by increasing heart rate from 4 dpf, approximately 10 days before convective O₂ transport is required under normoxic conditions. These studies suggest that CDH hastens the shift from diffusion to convection-based O₂ provision in zebrafish embryos (Jacob et al., 2002), but further work is required on this topic. Similarly to embryonic reptiles, there is evidence that these cardiovascular adjustments may improve hypoxia tolerance in the short-term, as P_Crit is lower in hypoxic zebrafish (Robertson et al., 2014) and Atlantic salmon (Wood et al., 2019b) compared with that of normoxic counterparts.

CDH also causes long-term changes in heart rate in embryonic zebrafish (Table S1D), but the magnitude and direction are variable. In general, tachycardia is the dominant response for embryonic zebrafish exposed to relatively mild or moderate levels of hypoxia (8–10% O₂) at temperatures of 28–31°C. However, severe hypoxia (2–4% O₂) causes bradycardia (Table S1D), which is mediated by a release of vagal tone or increase in catecholamines (Steele et al., 2011, 2009). Nevertheless, cardiac output remains constant in chronically hypoxic larval or embryonic zebrafish owing to an elevated stroke volume, and in some cases it is even increased (Cypher et al., 2018; Jacob et al., 2002; Moore et al., 2006; Yaqoob and Schwerte, 2010). Larval zebrafish subjected to hypoxia (4% O₂) also have significantly increased gene expression of β1, β2a and β2b adrenergic receptors (ARs) at 4 dpf relative to normoxic fish (Steele et al., 2009), and CDH increases cardiac responsiveness to agonists of adrenergic signalling and delays the onset of cholinergic control (see Glossary) in the rainbow trout (Miller et al., 2011). However, sympathetic sensitisation in zebrafish is likely to be dependent on the duration of hypoxia exposure and developmental stage, as the expression of β1AR does not change in whole zebrafish embryos (2 dpf) exposed to only 12 h or 24 h of hypoxia (5% O₂; Ton et al., 2002, 2003).

Despite the ecological importance, the long-term effects of CDH are poorly studied in fishes, and the results are highly variable. Trout larvae exposed to CDH exhibit catch-up growth with a significantly greater increase in weight (278% for CDH versus 188% for control) and length (64% for CDH versus 27% for control), eventually leading to significantly larger fry body weights and lengths compared with controls (Johnston et al., 2013). In contrast, juvenile Chinook salmon and European seabass exposed to CDH during embryogenesis are significantly smaller than controls (Del Rio et al., 2019), and growth restriction in hypoxic zebrafish embryos also persists into adulthood (Table S1A). However, no effect of CDH has been found on body weight in adult Atlantic salmon (Wood et al., 2017). Collectively, these studies show that the long-term effect of CDH on body mass is extremely variable in teleosts, and it depends on multiple factors, including species and body temperature.

To our knowledge, nothing is known about the long-term effects of CDH on juvenile and adult teleost cardiac structure or function. However, there is evidence of differential cardiac gene expression in rainbow trout exposed to CDH, including that of the common housekeeping genes 18S ribosomal RNA and acidic ribosomal phosphoprotein, and protein expression of cardiac troponin I (Johnston et al., 2013). Furthermore, previous work has shown that zebrafish cardiac morphology can be altered by other environmental stressors during development, including temperature and CO₂ (see below), as well as crude oil and polycyclic aromatic hydrocarbons (for a review, see Takeshita et al., 2021). Therefore, there is ample evidence that the fish heart is capable of developmental plasticity, but there is a distinct lack of studies on CDH.

Although few studies have explicitly investigated hypoxic programming in the fish heart, several studies address aspects of whole-organism performance and fitness that potentially link to cardiac performance. Hypoxic-incubated (10% O₂) rainbow trout show a consistently lower maximum relative swimming speed than normoxic controls across three developmental stages, which is thought to be caused by a delay in cardiac maturation (Johnston et al., 2013). Zebrafish and Atlantic salmon larvae exposed to CDH also show an improved whole-animal hypoxia tolerance. However, this phenotype does not persist into the juvenile and adult life stages (Del Rio et al., 2021; Robertson et al., 2014; Vanderplancke et al., 2015; Wood et al., 2019a; Wood et al., 2017). In fact, European seabass larvae raised in hypoxia (8% O₂) show a reduced hypoxia tolerance as juveniles, which is associated with an increased prevalence of opercular abnormalities (Cadiz et al., 2017). Similarly, 15-month-old Atlantic salmon exposed to CDH (10% O₂) are marginally less hypoxia tolerant than normoxia-incubated animals, although their aerobic scope is similar (Wood et al., 2017), and there is no effect of CDH (10% O₂) on hypoxia tolerance in juvenile Chinook salmon (Del Rio et al., 2021). Nevertheless, the physiological response to hypoxia can be affected by CDH in some fish. For example, when seabass are exposed to hypoxia as juveniles, fish that experienced hypoxia during embryogenesis show changes in haemoglobin subtype expression, but no differences in overall haemoglobin concentration (Cadiz et al., 2017).

Oviparous vertebrates rarely experience CDH in isolation because other developmental stressors often occur simultaneously (Box 2). Indeed, under natural conditions, CDH often occurs alongside fluctuations in CO₂, temperature, pH and salinity. Given that climate change is increasing the magnitude and frequency of these events, it is becoming increasingly important to study these interactive effects.

Although maternal nest choice and behaviour may partly shield terrestrial embryos from thermal stress, recent models suggest that global warming will increase the incubation temperatures of avian and reptilian eggs (Du et al., 2023; DuRant et al., 2019). Extensive research has shown that thermal stress can dramatically alter the morphology and physiology of reptilian and avian embryos, including changes in growth, body mass, cardiac mass, heart rate, mitochondrial density and respiration (Ben-Ezra and Burness, 2017; Du et al., 2023; Du and Shine, 2015; Du et al., 2010; Singh et al., 2020). Although data are scarce, some studies have investigated the interactive effects of temperature and hypoxia in avian and reptilian embryos. For example, Lourens et al. (2007) undertook a study in chickens where incubation temperature was increased from 37.8°C to 38.9°C at either 17% or 21% O₂. Temperature and hypoxia had independent effects on hatch time, body weight, yolk-free body weight and relative heart weight; however, there were no interactions between O₂ and temperature (Lourens et al., 2007). Another study in chickens found that mild levels of hypoxia (17% O₂) did not produce any effects on embryonic body mass or heart mass, even when temperature was increased from 37.8 or 38.9°C (Table S1A). By contrast, the negative effects of hypobaric hypoxia (2877 m, 15% O₂ SLE) during embryonic development on body mass, swimming speed and heart rate in adult viperine snakes at 28°C disappear when temperature is reduced to 24°C (Souchet et al., 2020a,b). Interestingly, increasing the temperature to 32°C produces a completely different phenotype, with a reduced heart rate, smaller body mass and faster swimming speed. The surprising improvement in swimming performance in adult snakes at high altitude at the warmest temperature persisted after relocation to low elevation (Souchet et al., 2020a). The authors suggest that constraints on development may be offset by the preservation of performance traits (perhaps through cardiorespiratory plasticity). Collectively, these studies suggest that the vertical colonisation potential of reptiles and birds (see below) will be affected by the interaction between temperature and O₂ availability.

Interestingly, a recent meta-analysis found that aquatic embryonic ectotherms are more than three times as plastic as terrestrial ectotherms when exposed to thermal stress during development (Pottier et al., 2022). Indeed, a large body of literature has shown that an increase in developmental temperature affects embryonic and larval fish growth rate, sex ratio, body size, metabolism, heart rate, cardiac morphology, hypoxia tolerance and swimming performance (Dimitriadi et al., 2018; Eme et al., 2015; Melendez and Mueller, 2021; Mueller et al., 2011; Pelster, 1999; Vagner et al., 2019; Zambonino-Infante et al., 2013). Some of these studies found effects that lasted into adulthood, including increased ventricular roundness in juvenile and adult male zebrafish exposed to elevated temperatures during embryogenesis (Dimitriadi et al., 2018, 2021). However, the short- and long-term effects of developmental temperature are highly variable in fish, and interestingly, the same meta-analysis found that persistent effects on thermal tolerance limits in adulthood were surprisingly weak (Pottier et al., 2022). Whether the same is true when elevated temperature occurs in combination with hypoxia is largely unknown, because surprisingly little is known about this interaction. One study on Chinook salmon investigated developmental outcomes in fish that were reared from fertilization to the fry stage at two temperatures (10°C and 14°C) and two O₂ levels (100% or 50% air saturation). Although temperature and O₂ saturation had independent effects on growth and acute hypoxia tolerance, there was no interaction between the two stressors (Del Rio et al., 2019). This was also the case in European sea bass exposed to different temperature and hypoxia combinations (40% or 100% air saturation×15°C and 20°C) from the flexion stage until the end of larval development (Cadiz et al., 2018). However, there were significant interactions on hatching success and thermal tolerance in Chinook salmon, with higher temperature generally potentiating the effects of hypoxia (Del Rio et al., 2019). Lastly, CDH causes an increase in cardiac output and heart rate in zebrafish embryos at 25–31°C, but the magnitude of the response is lowest at 31°C, presumably because the fish had neared their maximal cardiovascular capacity (Jacob et al., 2002). Clearly, more studies are warranted and necessary to understand the physiological implications of temperature and hypoxia interactions during development.

Global warming is driving some reptilian and avian species to shift their geographical distributions towards higher-elevation habitats with lower O₂ availability (Neate-Clegg and Tingley, 2023; Rubenstein et al., 2023). Developmental plasticity will therefore play a pivotal role in successful colonization of high-altitude environments. One approach to predicting the effects of climate-driven elevational range shifts is the so-called ‘transplant’ experiment, whereby gravid females or embryos from one elevation are transported and maintained at another. In this regard, recent work on the viperine snake has been particularly insightful, because this species has repeatedly migrated across elevational gradients to colonise high-altitude environments, in association with historical warming and cooling cycles (Gómez and Lunt, 2007). Transplanting viperine snake embryos at 28°C from 436 m (20% O₂ SLE) to 2877 m (15% O₂ SLE) increases heart rate, reduces body mass and decreases swimming ability (Souchet et al., 2020b). Importantly, post-hatching reciprocal transplant of snakes back to 436 m does not fully recover swimming performance, and the response is significantly temperature sensitive (see ‘Interactive effects of CDH and temperature’, above). Similar results were found in common wall lizards, where transplantation of embryos from sea level to 2877 m (15–16% O₂ SLE) leads to suppressed embryonic metabolism, cardiac hypertrophy and larger eggs that produce hatchlings with relatively low mass (Cordero et al., 2017). In contrast, transplantation of lowland Mongolia racerunner lizards to 2036 m (16–17 O₂ SLE) had no effect on embryonic development (hatching time and success) or hatchling phenotypes (body size and locomotor performance), which suggests this species can buffer the impact of hypobaric hypoxia (Li et al., 2020).

Another approach to predicting the effects of climate-driven elevational range shifts is to compare embryonic outcomes in native highland versus native lowland individuals from the same species. These types of studies reveal genetic adaptations that arise over successive generations. Perhaps unsurprisingly, numerous studies have clearly shown that embryonic highland oviparous species are less sensitive to hypoxia than their lowland counterparts. For example, there is no effect of 12% O₂ exposure on embryonic body weight in geese raised at high altitude (1600 m, Table S1A), and hatchling masses of high-altitude coots (4100 m) are similar or slightly greater than those at sea-level (Carey et al., 1989). Native high-altitude ptarmigan and coot embryos (4200 m; León-Velarde and Monge-C, 2004) and bar-headed goose embryos (Snyder et al., 1984) have a greater O₂-carrying capacity than their sea-level counterparts, with increased haematocrit, haemoglobin, capillary density and blood O₂ affinity. Adult fishes from high-altitude habitats in China also possess adaptations related to haemoglobin, as well as expansions of gene families associated with energy metabolism, ion transport and the response to hypoxia (Kang et al., 2017; Lei et al., 2021; Tong et al., 2017). Lastly, cardiac citrate synthase activity in white-tailed ptarmigan (4200 m) is higher than that of its sea level counterparts, suggesting increased mitochondrial density and oxidative capacity (Carey and Martin, 1997). These studies demonstrate that prolonged high-altitude residence in oviparous vertebrates confers some protection against hypobaric hypoxia (similarly to humans; Giussani et al., 2001), and this is associated with adaptations in both O₂-carrying capacity and utilisation. Nevertheless, living at high altitude for six successive generations does not completely protect chicken embryos from the effects of hypobaric hypoxia. Growth restriction in chickens is improved by high-altitude residence, but there is still a significant reduction in embryonic body mass with hypobaric hypoxia, as well as cardiac hypertrophy, ventricular wall thickening, aortic medial thickening and an increase in adrenal catecholamines (Giussani et al., 2007; Salinas et al., 2010, 2011). The effects can be prevented if high-altitude hens are given O₂ supplementation, which confirms that hypoxia rather than hypobaria is driving the cardiovascular response. Furthermore, the effects persist into adulthood when chickens are maintained at high altitude for a further 6 months, and there is also evidence of pulmonary hypertension, right-sided heart dysfunction and hypotension (Herrera et al., 2013; Salinas et al., 2014). Interestingly, American alligators exposed to CDH and maintained in hypoxia into juvenile life also have signs of pulmonary hypertension, including a decreased ratio of the right ventricle to left ventricle (Owerkowicz et al., 2009). Collectively, these studies suggest some of the problems associated with CDH in chickens cannot be prevented by residence at high altitude (at least across six generations) and post-hatch exposure to hypoxia may cause further damage, including pulmonary hypertension. Whether later generations would eventually evolve better protection awaits investigation.

Despite the fact that oviparous vertebrates often experience hypoxia and hypercapnia simultaneously (both naturally and in climate change scenarios; Box 2), we are unaware of any studies that have investigated the combined effects of CDH and chronic hypercapnia. There are however, several studies that have shown interactive effects of acute hypoxia and hypercapnia (<1 day) on chick embryonic O₂-carrying capacity and acid–base balance (Andrewartha et al., 2011, 2014; Burggren et al., 2023, 2012; Mueller et al., 2017). Furthermore, there is ample evidence that embryonic growth and cardiovascular outcomes can be affected by chronic hypercapnia alone, even at physiological levels. For example, exposure of embryonic chickens and ducks to CO₂ concentrations that they would normally encounter in the nest (1%) or higher (4%) increases body mass, compared with atmospheric levels (0.004%) and this effect persists into adulthood (De Smit et al., 2006; El-Hanoun et al., 2019; Everaert et al., 2007; Fares et al., 2012; Verhoelst et al., 2011). A similar observation has been made in common snapping turtles exposed to 3.5% CO₂ (Wearing et al., 2014), and American alligator embryos exposed to 3.5% and 7% CO₂ have increased relative heart mass and reduced arterial blood pressure (Eme and Crossley, 2015). Lastly, embryonic chickens and ducks exposed to 1% CO₂ have increased embryonic haemoglobin, packed cell volume (i.e the proportion of blood made up of cells) and red blood cell count (El-Hanoun et al., 2019; Fares et al., 2012). Collectively, these studies suggest that hypercapnia during development could offset hypoxic growth restriction in birds and reptiles, and it could potentiate some of the cardiovascular responses to hypoxia.

A large body of evidence suggests that juvenile and adult fish possess sufficient acid–base and osmoregulatory capabilities to tolerate very high CO₂ levels (>2000 µatm; Murray et al., 2016). However, a recent metanalysis confirmed that fish embryos and larvae are significantly more sensitive to hypercapnia than their adult counterparts (Cattano et al., 2018). Indeed, embryonic or larval fish have significantly higher levels of mortality and reduced growth at P_CO2 levels consistent with climate change projections (∼1000 atm). The increased sensitivity is likely due to ontogenic differences in respiration modes (dermal versus gills) and insufficient acid–base regulation prior to gill formation (Ishimatsu et al., 2008). There is also evidence that chronic hypercapnia affects cardiac function in some larval fish species. Chronic exposure to P_CO2 at ∼1100–1300 μatm causes tachycardia in Pacific herring, garfish and zebrafish larvae (Alter and Peck, 2021; Miller, 2013; Villalobos et al., 2020). However, numerous other studies have found no effect of hypercapnia on growth, heart rate, haemoglobin and mitochondrial function, and some have even found increased growth (Esbaugh, 2018; Leo et al., 2018; Mu et al., 2015; Scheuffele, 2017; Sun et al., 2019). Therefore, although there is certainly a case to study the interaction between hypercapnia and hypoxia in fish embryos and larvae, the effects may be relatively modest compared with those of temperature.

Oviparous ectotherms produce viable young when eggs are exposed to CDH, but there are numerous effects on the cardiovascular system at multiple levels of biological organisation, both during development and in postnatal life (Fig. 1). Despite vastly different cardiac designs and body temperatures, the embryonic cardiovascular responses are generally well conserved among vertebrates, and include asymmetric growth restriction, relative cardiac enlargement, alterations in heart rate, enhanced sympathetic activity and an increase in O₂-carrying capacity. In the long term, these phenotypic changes programme cardiovascular abnormalities in chickens that are very similar to those of mammals, leading to reduced cardiac performance and pathological cardiovascular signatures. The impact of CDH in American alligators and snapping turtles is less severe in juvenile life and may even be beneficial under circumstances of increased physiological stress. This suggests that the increased metabolic demand associated with endothermy places an additional burden on the avian and mammalian heart.

Unsurprisingly, the embryonic and postnatal response to CDH depends on the severity of hypoxia. In birds and reptiles, most responses are only evident at O₂ concentrations at or below 15% saturation. These levels of O₂ are commonly experienced by many embryonic reptilian species, which suggests that CDH is a significant driver of individual variation. In contrast, most lowland embryonic avians are unlikely to experience O₂ concentrations below 20% O₂, which makes CDH less ecologically relevant. However, megapode species develop at O₂ concentrations below 15%, so it would be interesting to see whether these species are uniquely adapted to hypoxia. The situation in fishes is far more complex, and there does not seem to be any obvious O₂ threshold for a cardiovascular response, even within the same species. This is probably because the levels of CDH are much more severe in the fish studies (45–95% reduction in O₂) versus the avian and reptilian studies (20–50% reduction in O₂), probably leading to higher levels of variation, and making comparisons between these groups complicated.

More work needs to be done to characterise the phenotypic responses and thresholds for CDH in the presence of other stressors, such as hypercapnia and temperature in all oviparous vertebrate groups. Interestingly, hypercapnia alone appears to have both synergistic and antagonistic responses to hypoxia in oviparous vertebrates, which means that the combination of these two stressors is expected to produce entirely different phenotypes. This is relevant to normal development because reptiles and birds experience hypoxia and hypercapnia simultaneously, and most studies use non-physiological levels of CO₂ when investigating hypoxia. It is also important in the context of climate change because the prevalence and intensity of hypercapnia is increasing, particularly in aquatic environments. Unsurprisingly, warming temperatures exacerbate the effects of developmental hypoxia in some oviparous species, which is concerning considering global warming and the increased prevalence and intensity of heat waves. The timing of extreme weather events is also crucial, because most species possess critical windows in development where the cardiovascular system is especially sensitive to stress. Furthermore, we expect species with faster developmental rates and shorter gestations to be disproportionately affected by heat waves and extreme weather events, compared with longer-gestation species, because a greater proportion of their development will be affected. Obviously, the challenge is to study the integrative effects of CDH, hypercapnia and warming on embryonic and adult phenotypic outcomes. In this regard, it is also critically important to gather accurate data about the effects of climate change on nest gas tensions and temperatures.

Future work should also focus on transplantation studies to determine the effects of high-altitude acclimation on reptilian and avian developmental outcomes. Studies like these are important because the phenotypic response to high-altitude hypoxia in lowland species will ultimately determine the colonization potential of these animals as the planet continues to warm. From the limited data available, it is clear that reptiles and birds respond to hypobaric hypoxia in a similar fashion to isobaric hypoxia, and some of the traits cannot be reversed by returning the animals to sea level. Long-term residence at high altitude affords protection in most avian and reptilian species, but domestic chickens raised at high altitude for six generations still undergo some level of growth restriction and cardiac remodelling in response to CDH. Importantly, the phenotype worsens with continued exposure to hypoxia post-hatch. Clearly, more multigenerational studies are necessary to understand the impact of cardiovascular plasticity on the vertical colonisation potential of oviparous birds and reptiles.

Lastly, there are some questions in this field that are almost completely unstudied. For example, our understanding of the effects of CDH and other stressors on the amphibian cardiovascular system is severely lacking. This is surprising, as this class of vertebrates is one of the most likely to experience fluctuations in developmental O₂, CO₂ and temperature (Box 2). There is also very little known about sex-dependent differences in the response to CDH among oviparous vertebrates. It is well established in the mammalian literature that cardiometabolic responses to developmental stressors are strongly sex-dependent, with females often being protected against detrimental long-term health outcomes compared to males (Giussani, 2021; Sandovici et al., 2022). Sex-dependent differences have been observed in some avian studies, but these effects are largely unstudied in ectothermic vertebrates. Similarly, the transgenerational effects of CDH and the underlying epigenetic mechanisms are very poorly studied in oviparous vertebrates. In this regard, several studies have shown that parental exposure to hypoxia can improve hypoxia tolerance in zebrafish offspring (Burggren, 2014; Ragsdale et al., 2022). These kinds of phenomena are particularly important to study, because transgenerational plasticity will play a crucial role in determining a species' ability to cope with a rapidly changing environment (Donelson et al., 2018).