The article by Sukumar et al. (1) published in a recent issue of Diabetes concludes that the type 2 NADPH oxidase (Nox2) plays a critical role in insulin resistance–related endothelial dysfunction. However, in our opinion it is too premature to draw these conclusions when looking at the entire body of available literature. The authors show in two different genetic mouse models of insulin resistance a blunted endothelium-dependent aortic relaxation to acetylcholine that can be reversed to normal both in vitro and in vivo by applying the peptide gp91ds-tat, which interferes with protein–protein interactions during Nox2 activation. Similarly, knockout of Nox2 in one of the genetic models of insulin resistance improved aortic vascular function and reduced reactive oxygen species formation in pulmonary tissues. The effects of the other vascular Nox isoforms in the mouse, Nox1 and 4, were not studied.
Our group has recently studied all Nox isoforms in one model of diabetic vascular complications albeit in a state of insulin deficiency (2). Indeed, we found that Nox1, but neither Nox2 nor Nox4, is the relevant target in the mouse in diabetes-accelerated atherosclerosis. With respect to Nox2, expression was upregulated both in human aortic endothelial cells and in the aorta of diabetic mice, but unfortunately when knocking out Nox2 we found a dramatic increase in mortality due to overwhelming gram-negative sepsis. Indeed, mortality was 100% after 20 weeks of diabetes. The animals only survived when placed continuously on antibiotics for the entire duration of the study. Similar alarming findings with respect to Nox2 in diabetes have been reported by others (3).
Thus, we consider that currently it is an overstatement to conclude that Nox2 is “an attractive target to prevent early atherosclerosis in insulin resistance” (1). Nox2, an enzyme of white blood cells that mediates the oxidative burst of the innate immune response, plays a role in early inflammatory vascular events associated with insulin resistance and many other disease states. However, there is also substantial evidence that Nox2 is not a feasible target to treat diabetes-related vascular injury and advanced atherosclerosis. Indeed, in the setting of fully developed diabetes with frank hyperglycemia it appears more likely that this isoform should not be interfered with. The models used by Sukumar et al. (1) would be easier to interpret if other Nox inhibitors or knockout experiments of Nox1 and Nox4 were also studied. Furthermore, as suggested in the accompanying commentary (4), the experiments should have been complemented by studies in high-fat feeding and/or more advanced atherosclerosis models in ApoE−/− mice. Until then the verdict remains at least open as to whether and which isoform of Nox is a possible target in insulin resistance and diabetes-related vascular injury.
ACKNOWLEDGMENTS
No potential conflicts of interest relevant to this article were reported.
