Alterations to the supply of oxygen during early life pose a profound stress to physiological processes during critical periods of development, with potentially deleterious outcomes that can be long-lasting. Sleep-related breathing disorders are characterized by intermittent pauses in breathing during sleep with associated blood oxygen desaturations. Obstructive sleep apnea (OSA) is recognized as a common condition in adults, but more recently there is increased focus on the incidence of OSA in childhood and during pregnancy that has long been overlooked. Apnea of prematurity is a recognized phenomenon characteristic of preterm babies. Chronic intermittent hypoxia (CIH) is a central feature of disordered breathing and is increasingly recognized as a major driving force in the development of a wide range of morbidities, including maladaptive remodeling in cardiorespiratory control (O'Halloran, 2016). The early life environment should prepare offspring for long-term success and it is therefore important that developmental conditions do not produce a mismatch with the adult environment. Physiological adaptations to perturbations during development may sustain life before birth or in early life, but may come at a cost in terms of biological trade-offs, with deleterious consequences for homeostatic regulation in later life.

The physiological changes that occur during pregnancy predispose pregnant women to the development of OSA (Loube et al., 1996; Izci et al., 2006). Gestational OSA has been identified in many recent studies as a major clinical problem. In the absence of large scale longitudinal studies, the prevalence of gestational OSA remains unclear, but evidence of gestational OSA has been reported in several studies (Edwards et al., 2005; Chen et al., 2012; Pien et al., 2014) and may be more common in women with higher body mass index (Ko et al., 2012; Louis et al., 2012; Sarberg et al., 2016). Gestational OSA is likely to result in pre-placental oxygen desaturations whereby both the mother and her fetus become hypoxic. Maternal O₂ desaturations have been shown to coincide with decelerations in fetal heart rate (Sahin et al., 2008), despite the greater oxygen-carrying capacity of fetal hemoglobin, which left-shifts the hemoglobin-oxygen dissociation curve, which means oxygen loading occurs at lower oxygen concentrations in the fetus. Recent studies report that mothers with sleep-disordered breathing had increased evidence of uteroplacental hypoxia and fetal distress (Bourjeily et al., 2015a,b; Ravishankar et al., 2015). Studies to date show that gestational CIH may be responsible for some of the poor fetal outcomes following gestational OSA. Gestational CIH is reported to cause reductions in fetal growth (reversible), increased normoxic ventilation, attenuated peak hypoxic ventilatory response (short-term), and accentuated hypoxic ventilatory depression in postnatal pups, as well as glucose dysregulation (Gozal et al., 2003; Iqbal and Ciriello, 2013; Iqbal et al., 2014).

Preterm birth is associated with aberrant control of breathing owing to amongst other things, immaturity in the central respiratory networks governing respiratory control. Periodic breathing with recurrent central apneas are common and are associated with repeated exposure to (IH Razi et al., 2002; Martin et al., 2011; Di Fiore et al., 2013; Decima et al., 2015). Several studies have reported long-lasting deleterious effects of exposure to hypoxia during critical periods of the postnatal period in rodents (Okubo and Mortola, 1988; Gozal et al., 2002; Reeves et al., 2003, 2006). Notwithstanding the critical role of respiratory muscles as the final effector in the respiratory control system there is a general paucity of information concerning the short- and long-term effects of exposure to hypoxia during development on respiratory muscle function and respiratory mechanics. There are no studies to date exploring the effects of early life exposure to CIH on diaphragm function. This represents an important gap in our knowledge, of potential clinical relevance. The aim of this study was to investigate the effects of exposure to gestational CIH and postnatal CIH on diaphragm muscle function in male and female rats. Both male and female animals were studied given the evidence of male susceptibility to early life stressors (Zeitlin et al., 2002; Mage and Donner, 2006; Johnston and Hagberg, 2007), notwithstanding our previous observation that there are no sex differences in the effects of pCIH on sternohyoid muscle force (McDonald et al., 2015b, 2016).

The principal finding of this study is that early life exposure to CIH has no effect on diaphragm muscle force-generating capacity. Neither gestational CIH nor postnatal CIH affected diaphragm force or endurance properties in either male or female animals. Similarly, hypoxic tolerance in diaphragm muscle of both sexes was unaffected by antecedent hypoxic exposure. The study extends and complements a recent report by our group, which observed that diaphragm function in adulthood is unaffected by postnatal exposure to CIH (McDonald et al., 2016). The findings in diaphragm muscle contrast with our observations in sternohyoid muscle from the same animals (McDonald et al., 2015b). Thus, differential effects of early life exposure to hypoxia are expressed in sternohyoid and diaphragm muscle, with evidence of weakness in the airway dilator muscle that persists for several weeks upon return to normoxia (McDonald et al., 2015b) and increased susceptibility to subsequent hypoxic insult (McDonald et al., 2016), whereas no functional impairment is evident in diaphragm muscle at any stage, revealing an apparent relative resilience in rat diaphragm muscle to hypoxic stress in early life (McDonald et al., 2015b, 2016; this study). The latter is consistent with observations of sternohyoid, but not diaphragm, dysfunction following chronic sustained hypoxia during development in rats (Carberry et al., 2014), and some reports in adult rats documenting no functional impairment in diaphragm following chronic sustained hypoxia (El-Khoury et al., 2003; McMorrow et al., 2011; Gamboa and Andrade, 2012).

Our study is the first report of the effect of early life exposure to CIH on diaphragm function. Though some subtle effects of CIH exposure on diaphragm muscle performance were evident, on balance we conclude that early life exposure to CIH has no major effect on diaphragm force in either sex, at least under the conditions established in our study. We do not discount the possibility that CIH has adverse consequences for diaphragm muscle performance per-se. Indeed, diaphragm weakness and fatigue was reported in adult male rats exposed to CIH, an outcome dependent on the duration and intensity of the hypoxic stimulus (Shortt et al., 2014). Moreover, chronic sustained hypoxia is reported to cause diaphragm weakness, an effect dependent on the timing of presentation of the stimulus during the developmental period (Kass and Bazzy, 2001), consistent with observations of sternohyoid muscle dysfunction following chronic sustained hypoxia (Carberry et al., 2014). Whilst, the entire gestational and neonatal period is generally regarded as a relatively vulnerable period wherein the respiratory control system is quite malleable, we acknowledge that within the timeframe of our study there may well have been periods of maladaptive remodeling leading to dysfunction that in turn were countered by adaptive or compensatory responses. There are likely complex temporal responses to hypoxia in developing respiratory muscle and our study design was focused on just two distinct end-points (P22 and 42). In particular, the first week of life in the rat is known to be especially sensitive to perturbation with consequences for cardiorespiratory control (Bavis, 2005; Bavis et al., 2007; Pozo et al., 2012; Fournier et al., 2015). It would be interesting to explore the effects of CIH on respiratory muscle performance during this period i.e., functional assessment immediately following exposure to CIH during the first week of life. The latter may have greater relevance to preterm infants who experience CIH resulting from apnea of prematurity. Similarly, earlier assessment of diaphragm muscle performance following exposure to gCIH (e.g., during the first week of life) might have revealed evidence of maladaptive plasticity with functional consequences. Our initial study point at P22 was 3 weeks after termination of the gCIH exposure, potentially providing a window for recovery during the rapidly changing developmental timeline (O'Connell et al., 2013). Thus, further work is required to comprehensively determine the consequences of exposure to early life CIH on diaphragm muscle form and function, but the evidence to hand at this juncture suggests no major lasting effect of early life CIH on diaphragm performance. Of note in the present study, exposure to gCIH increased diaphragm fatigue during severe hypoxia at P22 in female but not male animals. The effect is small but might point to increased susceptibility to respiratory impairment under such conditions should they present in vivo. The observation suggests that further evaluation of hypoxic tolerance and potential differences between the sexes following exposure to gCIH are warranted.

It is important to consider the findings of this study in the context of our previous observations revealing sternohyoid muscle weakness in pCIH-exposed rats with lasting effects leading to increased susceptibility to hypoxic insult in adulthood. The co-ordinated activation of pharyngeal upper airway muscles (e.g., sternohyoid) and thoracic muscles (e.g., diaphragm) optimizes ventilation and is critical for airway stability. The sternohyoid appears more vulnerable to hypoxia-induced force decline (McDonald et al., 2015b, 2016), perhaps due to its composition of fast muscle fiber types compared to predominantly slow fiber types of the diaphragm (O'Connell et al., 2013). Differential functional responses to hypoxia in fast and slow limb muscles has been observed (Howlett and Hogan, 2007). A study of the structure and function of rat sternohyoid and diaphragm muscle during the first month of life revealed a faster maturation rate in the sternohyoid compared with the diaphragm (O'Connell et al., 2013). Fiber remodeling may partly explain pCIH-induced sternohyoid weakness (McDonald et al., 2015b, 2016), and as such intrinsic differences in fiber complement between sternohyoid and diaphragm, with related differences in metabolic strategies (glycolytic vs. oxidative), combined with the different functional requirements of the two muscles might provide a substrate for maladaptive remodeling in one muscle (sternohyoid), but not the other (diaphragm). It is also plausible to consider that the diaphragm may be intrinsically protected and more robust in response to adversity because of its essential role in respiratory homeostasis. Redox remodeling appears pivotal to CIH-induced respiratory muscle dysfunction (Williams et al., 2015), and antioxidant supplementation has been shown to ameliorate or prevent CIH-induced weakness and fatigue (Dunleavy et al., 2008; Skelly et al., 2012; Shortt et al., 2014). Differential redox responses in diaphragm and sternohyoid could plausibly underpin the apparent difference in muscle responses to CIH-related stress.

Whatever the mechanism, a mismatch in the performance of diaphragm and upper airway dilator muscles potentially places the compliant pharyngeal airway at risk. As such, weakened airway dilator muscle activity in the face of normal pump muscle function threatens airway caliber, at least in species such as human with collapsible upper airways. Thus, our findings may have relevance to the control of airway patency in humans, where maintenance of diaphragm muscle force with coincident depressed upper airway muscle force would increase the risk of airway collapse. Weak upper airway muscles may not adequately counteract the negative pressure generated by the diaphragm during inspiration, and/or may not be able to recover a collapsed pharyngeal airway with persistent diaphragm contraction.

In summary, the present study demonstrates that exposure to CIH during early life has little effect on diaphragm force or fatigue. Differential effects of early life CIH on airway dilator and diaphragm muscle performance may result in a mismatch with deleterious consequences for the control of airway caliber. As such, extrapolating (with caution) the findings to humans, we speculate that early life hypoxic stress may have long-lasting implications for the control of airway patency and respiratory homeostasis. In this way the findings suggest that early life adversity characterized by CIH may have implications for the development or worsening of obstructive sleep apnea in later life.

FM and KO designed the study. FM conducted experiments and analyzed data. FM and KO interpreted data sets and performed statistical analyses. All authors contributed to the writing of the manuscript.

Funded by Health Research Board (Ireland). FM was funded by the School of Medicine and Medical Science, University College Dublin, Ireland.