Reactive oxygen species in mechanotransduction.

A RANDOMIZED CLINICAL TRIAL by the Acute Respiratory Distress Syndrome Network (1) demonstrated a 22% reduction in mortality in patients by reducing the tidal volume for mechanical ventilation from the conventional setting of 12 ml/kg to a lower setting of 6 ml/kg. This dramatic decrease in mortality has stimulated significant interest in the mechanisms of ventilatorinduced lung injury and the development of lung protective strategies. One theory is that higher tidal volumes induce excessive stretch of lung tissue that results in stimulation of inflammatory responses and exacerbation of the underlying lung injury (reviewed in Refs. 12, 13, 23, 25, 26, 28). The central premise is that cells in the lungs sense changes in mechanical forces and transduce the mechanical signal into a biological response (mechanotransduction) (12, 21, 31). However, the mechanisms of mechanotransduction remain to be elucidated. A study by Ali et al. (Ref. 2a, page L486 in this issue), a current article in focus, examines one of the potential early mechanisms involved in mechanotransduction, the production of reactive oxygen species (ROS) in endothelial cells in response to mechanical stretch, and relates this signal to activation of inflammatory responses. The foundation for the theory that changes in mechanical forces in the lungs stimulate inflammation can be found in studies of the development of atherosclerosis. Atherosclerotic plaques typically develop in regions of the circulation near branch points or areas of curvature that promote flow reversal or flow stagnation. Thus the hypothesis is that oscillatory shear stress or the loss of shear stress promotes proinflammatory pathways in endothelial cells in these regions (reviewed in Refs. 2, 4, 20, 27). Furthermore, it has been postulated that changes in shear stress regulate proinflammatory mechanisms in endothelial cells by stimulating the production of ROS (9–11, 15, 17, 24). Stimulation of ROS production has been demonstrated in response to changes in shear stress (8, 9, 19, 24, 30, 33) and cyclic mechanical stretch (6, 7, 18, 22, 29, 32) in cultured endothelial cells. However, the source of ROS remains controversial. As described by Ali et al. (2a) ROS generation in response to mechanical forces may originate from an NAD(P)H oxidase system, from xanthine oxidase, from mitochondrial production, or from other oxidase systems such as nitric oxide synthase. De Keulenaer et al. (9) demonstrated that oscillatory shear stress of human umbilical vein endothelial cells (HUVEC) over 24 h stimulated a progressive increase in NADH oxidase activity as measured by lucigenin chemiluminescence, whereas steady laminar shear stress induced a transient increase in NADH oxidase activity that returned to baseline levels by 24 h. Oscillatory shear stress stimulated higher levels of intracellular superoxide compared with static controls as measured by dihydroethidium fluorescence. Cyclic stretch also increased NAD(P)H oxidase activity and ROS production in endothelial cells, and this increase correlated with an increase in message for the p22 subunit of NAD(P)H oxidase (22). Interestingly, mechanical stretch-induced stimulation of ROS production [measured by dichlorofluorescin (DCF) diacetate fluorescence] by vascular smooth muscle cells was not observed in cells isolated from mice deficient in the p47 subunit of NAD(P)H oxidase (16). A recent study by McNally et al. (24) demonstrated that xanthinedependent superoxide production was increased in endothelial cells exposed to oscillatory shear stress and that cells isolated from p47 / mice had low superoxide production and decreased xanthine oxidase expression and activity. The study by Ali et al. (2a) shows that mechanical stretch stimulates ROS production by HUVEC through mitochondrial mechanisms. ROS production, assessed by DCF fluorescence, was significantly increased after 6 h of stretch, and the increase was abolished by the mitochondrial complex I inhibitor rotenone. Inhibitors of NAD(P)H oxidase (apocynin), xanthine oxidase (allopurinol), or nitric oxide synthase (N-nitro-L-arginine) did not block the stretch-induced increase in DCF fluorescence, suggesting that these pathways were not important in the response to stretch. To further examine the role of mitochondrial pathways in the stretch-induced stimulation of ROS, mutant HUVEC cells were generated that lacked mitochondrial DNA ( 0 cells). When ROS production was measured in 0 cells, the baseline levels were substantially reduced, and there was a reduced increase in DCF fluorescence in response to stretch that was insensitive to rotenone. To determine whether the increased ROS production was functionally related to inflammatory pathways, the authors demonstrated that control HUVEC responded to stretch with increased activation of NFB and VCAM-1 expression. Cyclic stretch did not induce either of these proinflammatory events in 0 cells, but treatment with lipopolysaccharide did induce NFB activation and VCAM-1 expression, indicating that the respiration-deficient cells were capable of activating proinflammatory pathways. The study by Ali and colleagues (2a) raises many interesting questions. Studies from different groups have now demonstrated that mechanosensitive stimulation of ROS is generated by NAD(P)H oxidase, xanthine oxidase, or mitochondrial pathways. Which of these pathways is predominant in pulmonary endothelial cells? The majority of work in the area of mechanosensitive ROS production has focused on HUVEC or other large vessel endothelial cells, but few studies have examined the sensitivity of pulmonary endothelial cells (3, 14, 30) or other lung cells such as epithelial cells, macrophages, eosinophils, or fibroblasts. It is also not clear what levels of mechanical forces are necessary to induce ROS production or proinflammatory pathways. The lungs typically undergo cyclic deformation during normal breathing, and this may be increased during mechanical ventilation, but the levels of stretch have not been well characterized and are likely to vary substantially throughout the lungs. Endothelial cells are exposed to a wide range of shear stresses in the pulmonary microcirculation since red blood cells must transit through capillaries that Address for reprint requests and other correspondence: C. M. Waters, Dept. of Physiology, Univ. of Tennessee Health Science Center, 894 Union Ave., 426 Nash, Memphis, TN 38163 (E-mail: [email protected]). Am J Physiol Lung Cell Mol Physiol 287: L484–L485, 2004; 10.1152/ajplung.00161.2004.