Good soldier falls: the liver in sleep apnea.

SEVERE OBSTRUCTIVE SLEEP APNEA fires a complex array of physiological stressors rapidly and relentlessly across each period of sleep. With each occlusion of the upper airway, particularly in obese individuals, arterial oxygen content and delivery fall precipitously, carbon dioxide rises, acidosis ensues, hemodynamics are perturbed, cortisol levels rise, and sympathetic activation surges (for review, see Refs. 10 and 14). Although many of these rapidly fluctuating physiological stressors wreak havoc with homeostasis, the repeated attacks from fluctuating oxygen levels are increasingly recognized as a major contributor to morbidities in sleep apnea (11, 14). In this issue of the Journal of Applied Physiology, Dr. Li and colleagues have examined the liver’s response to longer term exposure to hypoxia-reoxygenation events modeling the oxygenation patterns of obstructive sleep apnea (6). Having previously observed hyperlipidemia in response to brief (5 days) exposure to severe hypoxia-reoxygenation (8), the present study was designed to determine whether exposure duration and/or severity of arterial oxygen nadir influences intermittent hypoxia-induced hyperlipidemia. The present study finds hyperlipidemia of similar magnitude when the exposure duration is extended to 4 wk. In contrast, severity is an important factor in intermittent hypoxia hyperlipidemia and is only present with severe hypoxia (6). Therefore, severe intermittent hypoxia has a persistent effect on lipid levels, and yet this effect does not appear to be cumulative. While raising serum lipids seems unhealthy, this pattern of lipid response is consistent with an adaptive response to severe hypoxia. In support, a similar response to hypoxia is present in simpler life forms, including fission yeast. Lipid homeostasis in mammalian cells is controlled primarily by sterol regulatory element binding proteins (SREBPs). SREBPs activate the transcription of over 30 genes regulating lipid synthesis (for review, see Ref. 9). There is an important feedback in this system so that low lipid levels activate SREBP, and high sterol levels will inactivate SREBP, thereby maintaining stable serum lipid levels. However, recent investigation of SREBP homologs in fission yeast has unveiled a second controller for the yeast homolog of SREBP (SRE-1), and that controller is hypoxia (5). Sterol synthesis requires oxygen and will quickly cease under conditions of low oxygen. In the fission yeast, Saccharomyces pombe, severe hypoxia activates SRE-1, and this in turn upregulates genes for low oxygen sterol synthesis and genes promoting the switch from aerobic to anaerobic survival and growth, genes essential for adapting to hypoxia and allowing sterol synthesis even during hypoxia. This system, at least in fission yeast, can override the sterol negative feedback, such that it is possible under conditions of hypoxia to have persistently higher levels of specific lipids, necessary for maintaining membrane stability and fluidity (7). There are several lines of evidence to support a parallel system in mammals, whereby hypoxia may be permissive to, or even promote, hyperlipidemia. Newborn humans who suffer perinatal severe hypoxemia have high triglyceride concentrations in cord blood and serum, where the level of triglycerides varies with the severity of hypoxic insult (1, 3). The present work suggests that intermittent hypoxia-induced hypertriglyceridemia, at least under some circumstances, may be an SREBP-mediated response. Li et al. (6) have shown that intermittent hypoxia raises hepatic stearoyl coenzyme A desaturase-1 (SCD-1), an enzyme that, in a hypoxic environment, is expected to increase triglyceride synthesis more than cholesterol and is expected to enhance hepatic release of lipids. Through this mechanism, hypoxia in mammals could promote persistent hypertriglyceridemia. In support, this same group of researchers previously found that mice with transgenic partial deficiency in hypoxia-inducible factor-1 (HIF-1 / ) showed smaller increases in SREBP and SCD-1 and proportionately smaller increases in serum lipids in response to intermittent hypoxia (7). Future studies are needed to determine whether the hypertriglyceridemia is of benefit in mammals and whether this benefit outweighs the adverse health consequences of 15–20% increase in serum triglycerides. While it appears that the liver may be responding to defend the body from the effect of hypoxia on sterol synthesis, the present study by Li et al. (6) shows that the liver is wounded by intermittent hypoxia. The wound includes lipid peroxidation, suggesting significant oxidative stress from intermittent hypoxia. Oxidative stress and lipid peroxidation are implicated in the pathogenesis of many hepatic disorders including nonalcoholic steatohepatitis and fibrosis (4). In turn, severe obstructive sleep apnea is a significant risk factor for NASH independent of obesity (12). Just how sleep apnea and intermittent hypoxia induce steatohepatitis is not known, but recently identified mechanisms of intermittent hypoxia brain injury may shed light on the pathogenesis of sleep apnea-induced liver injury. Intermittent hypoxia increases lipid peroxidation in selective brain regions (13, 15), and this process is dependent upon NADPH oxidase activation (16). Transgenic absence of functional NADPH oxidase (Nox-2) and Nox-2 inhibition with apocynin throughout exposure completely confer resistance to lipid peroxidation in susceptible brain regions (16). Support for the concept that a similar pathway underlies hepatic injury comes from the recent observation that NADPH oxidase activation in stellate cells in the liver is essential for steatohepatitis and fibrosis (2). Thus, in light of the findings of Li et al. it will be important to determine whether the hepatic lipid peroxidation and liver injury from intermittent hypoxia and sleep apnea are downstream from NADPH oxidase activation. We are far from securing effective pharmacological therapeutics for sleep apnea, and many patients have difficulties tolerating the various mechanical therapies for this disorder. Thus while we continue to search for pharmacotherapeutics to treat sleep apnea, reasonable interim objectives should include preventing or minimizing end-organ damage from this disease, including this hepatic injury and dyslipidemia identified by Li et al. (6).