The effect of salicylates on insulin sensitivity

Kim et al. Reply

We would like to thank Netea et al. for their thoughtful comments regarding our article (1). First of all, we would like to emphasize that we are not advocating the use of high-dose aspirin for the treatment of type 2 diabetes. In fact, we strongly recommend against it in view of aspirin’s potential long-term toxicities at high doses. Our studies used high-dose aspirin as a pharmacological tool to identify IKK-βas a potentially novel therapeutic target for type 2 diabetes. This hypothesis, developed by Yuan et al. (2), led to our finding that Ikk-β^–/+ mice were protected from obesity- and diet-induced insulin resistance (2) as well as following 5-hour lipid infusion (1). Because aspirin is a weak inhibitor of IKK-β, high doses of the drug restore insulin sensitivity in each of these models of insulin resistance.

We are aware of the previous studies that have shown that aspirin therapy actually causes insulin resistance (3–5). We believe the major reason for the discrepancy is not due to species differences as suggested by Netea et al. but rather the lower dose and duration of salicylate therapy used in these human studies. Even older clinical trials clearly showed hypoglycemic effects of high doses (4–10 g/day) of aspirin and salicylate that occur progressively over 1 to 2 or 3 weeks (6). Using corresponding therapeutic regimens, we saw beneficial effects in rodents (1, 2) and, in more recent studies by our group, in humans (7). High-dose salicylate treatment (about 7 g/day for 2 weeks) was very successful in lowering fasting and postprandial hyperglycemia in patients with type 2 diabetes, which could partly be attributed to increased peripheral insulin sensitivity as assessed by a hyperinsulinemic-euglycemic clamp (7).

In summary, we believe that our data (1) support the hypothesis that fat induces insulin resistance through activation of a serine/threonine kinase cascade leading to decreased IRS-1/IRS-2–associated phosphatidylinositol 3-kinase activity in muscle and liver (2, 8), and that IKK-β, and the steps leading to its activation, are potential novel therapeutic targets for type 2 diabetes mellitus.