© 2004 American Thoracic Society
Slip of the TongueDavid Geffen School of Medicine at University of California, Los Angeles Los Angeles, California The tongue is essential for speech/phonation, tasting, chewing, swallowing, suckling, and licking, but to do all of this it evolved to lie partially in the pharynx, where in humans it is at best a nuisance for breathing. At its worst, physics, sleep, and gravity conspire to make inspiration difficult (and noisy) or even impossible, because while one is lying supine the tongue can collapse on the back of the pharynx, obstructing the airway and causing apnea. Obstructive sleep apneas (OSAs) compromise health, and the cumulative effect of OSA over months and years can be devastating to brain function, causing the loss of gray and white matter (1), and have long-term effects on cognitive function. Veasey and coworkers in the current issue of the Journal (pp. 665–672) suggest that this damage can also affect regulatory actions of the brain, in particular by compromising the proper functioning of motoneurons that control tongue muscles, which leads to a vicious cycle of progressively worse sleep-disordered breathing. The etiology of an OSA is not known and is a matter of substantial investigation. At the top of likely suspects are age, Y chromosome–associated genes, morbid obesity, and structural abnormalities of the head, face, and soft tissue. None of these factors, alone or in combination, produce obstructions during wakefulness. Yet during sleep, especially when motoneuronal activity is suppressed during rapid eye movement (REM) sleep, upper airway muscles lose tone and obstruction can occur. Hypoglossal (XII) motoneurons, whose activity determines the position, shape, and stiffness of the tongue, are the principle focus for basic research aimed at understanding control of upper airway function. The reduction in release of the neuromodulator serotonin (5-HT) during REM sleep is likely a major contributor to decreases in XII motoneuronal activity, as 5-HT acting on 5-HT2A/2C receptors elevates neuronal excitability in rats and cats (2). This observation led to a concept that serotonergic drugs may help increase airway muscle tone and reduce OSA in humans, but clinical trials have been disappointing. Here, Veasey and coworkers provide explanations as to why this might be the case. They found that in rats exposed to long-term (3 weeks) intermittent hypoxia (LTIH), injections of 5-HT (or the glutamate receptor agonist NMDA) into the XII nucleus were less effective in inducing XII nerve activity compared with controls. After excluding several obvious possible causes such as downregulation of the number or binding affinity of 5-HT2A/2C receptors or even the numbers of XII motoneurons, they hypothesized that oxidative stress induced by LTIH reduced XII motoneuronal excitability. Oxidative stress appears to play a role in the pathology of neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease and amyotropic lateral sclerosis (3), and it is the cornerstone of the "free radical theory of brain aging" that posits that a decline in neural function is due to the accumulation of deleterious effects on critical cellular components by the normal metabolic production of free radicals (4). LTIH produces neurocognitive deficits in rodents, and these deficits are attenuated after administration of an antioxidant (5). In mice, LTIH induces neuronal apoptosis in the cerebral cortex and neurocognitive dysfunction (6). Hypoxia-induced oxidative stress leads to lipid peroxidation and protein and nucleic acid oxidation. The extent of tissue damage can be evaluated by measuring phospolipid composition, reactive aldehydes, and isoprostanes, with F2-isoprostane commonly used to index lipid peroxidation; Veasey and coworkers found that LTIH elevated this isoprastane isomer in the region of the medulla including the XII nucleus. Antioxidant compounds such as vitamins C and E, lipoic acid, ubiquinones, superoxide dismutase, catalase, and glucathione peroxidase can limit oxidative damage. Tempol, a superoxide dismutase mimetic, reduces both nitration and oxidation of key cellular proteins. Here, treatment with tempol prevented the LTIH-induced increase in isoprastane level and the decrease in the responsiveness of the XII nerve to 5-HT and NMDA injections. This observation supports the hypothesis that oxidative stress contributes to pathologic responses caused by LTIH. Multiple groups of respiratory neurons may be affected by oxidative stress including those involved in rhythm generation or chemosensation (7). However, as Veasey and coworkers rightly point out, motoneurons are at greater risk of oxidative injury because of their large somas, high metabolic activity, high mitochondrial function and high content of neurofilament protein susceptible to oxidative and nitrative changes. Motoneurons also have very little cytosolic Ca2+ buffers, which may also make them particular vulnerable to oxidative stress. Responses to LTIH are clearly different from those induced by intermittent hypoxia (IH). IH in neonatal or adult rats induces a 5-HT2A/2C–dependent increase in respiratory (including XII) motoneuronal activity, called long-term facilitation (LTF), that is postulated to be a regulatory mechanism to reduce the frequency and severity of airway obstructions (7, 8). Interestingly, in humans, significant responses to IH are seen during sleep in habitual snorers (9). In brainstem slices from neonatal rats that generate respiratory rhythm, three consecutive applications of a 5-HT2 receptor agonist to the XII nucleus augments motoneuronal output for several hours by potentiating the function of AMPA receptors (the principle glutamate receptors transmitting excitatory respiratory drive) (10). In the present study, LTIH did not affect the response to AMPA 12 hours after the last of approximately 20,000 hypoxic episodes. Clearly the severity and number of hypoxic episodes changes the efficiency of 5-HT–dependent compensatory mechanisms. Moreover, LTIH may compromise the role that LTF is postulated to play in preventing or reducing the incidence of OSAs. Veasey and coworkers performed their experimental study in rats. Rodents, cats, and dogs (with the notable exception of the English bulldog) do not suffer from OSA. When rodents are used to study the effects of hypoxia, the protocols are markedly different from the experience of humans with OSA. In humans, hypoxic episodes occur only during sleep and are associated with hypercapnea and alveolar hypoventilation, cessation of respiratory flow with consequent large intrathoracic and upper airway intraluminal pressure swings during inspiratory efforts against an occluded airway, a marked elevation of blood pressure (often greater than 200 mm Hg systolic), substantially increased sympathetic outflow, and ultimately frequent arousals. In contrast, rats are kept in chambers where they are exposed during both wakefulness and sleep to hypoxic gas mixtures (FIO2 = 5–10%); as a result, they hyperventilate with consequent hypoxic hypocapnea, and their upper airways are invariably patent. Nonetheless, the rodent model mimics hypoxia-reoxygenation swings of the severe forms of OSA, suggesting that oxidative injury also occurs in humans. An important consequence of the mechanism proposed by Veasey and coworkers is that OSA initiates a self-perpetuating process, whereby airway obstruction and resulting episodes of hypoxia-reoxygenation impair airway protective mechanisms and lead to more obstructive episodes. Breaking this cycle or even slowing it down has considerable therapeutic potential. FOOTNOTES Conflict of Interest Statement: J.L.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.A.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
