• Potential conflict of interest: Nothing to report.


We would like to thank Valenti and colleagues for their interest in our article and for pointing out the higher-than-expected prevalence of the C282Y mutation among Caucasians with nonalcoholic steatohepatitis (NASH) in our study1 compared to the prevalence reported among Caucasians in the general population in two recent large studies, similar to their previous findings.2–4 We agree that larger case controlled studies are needed to determine whether hemochromatosis (HFE) mutations confer increased susceptibility to NASH.

Valenti et al., found that subjects with nonalcoholic fatty liver disease (NAFLD) who had C282Y mutations had a lower body mass index (26.3 versus 27.9; P = 0.035) and serum triglyceride level (mmol/L; 1.49 versus 2.07; P = 0.05) and a trend toward a lower β-cell index (173.1 versus 224.4; P = 0.161) compared to subjects with NAFLD without the C282Y mutation. Among the Caucasian NASH subjects in our study, we did not find significant differences in body mass index or serum triglyceride levels between subjects with and without the C282Y mutation. This may reflect dietary differences in the two countries and the higher prevalence of obesity in the United States compared to Italy. Alternatively, because all our patients had biopsy-proven NASH compared to only 31% of the subjects in the study by Valenti et al., it is conceivable that the effect of iron on lipid metabolism is more pronounced in the early stages of NAFLD, which is later diminished as the disease progresses toward NASH.

There is a growing body of work suggesting that iron and glucose metabolism is interdependent and that increased iron stores may contribute to “first hit” insulin resistance (IR). For example, hyperinsulinism and IR and/or diabetes are common in patients with iron overload.5 Increased iron stores inhibit insulin extraction and metabolism in the liver, leading to hyperinsulinemia.6 Serum iron and transferrin may contribute to IR via increased adipocyte lipolysis7 and impairment of glucose transport.8 In vitro, iron can reduce binding of insulin to its receptor and reduces insulin receptor gene expression.9

A recent proteomic analysis on livers of iron-loaded mice that showed down-regulation of several enzymes involved in fatty acid oxidation lends support for the hypothesis that body iron stores appear to have an effect on lipid metabolism.10 Petrak et al., compared the protein profiles of HFE-deficient mice to wild-type mice with equivalent hepatic iron stores obtained through dietary iron loading.11 Among the proteins differentially expressed in the two groups of mice were proteins that function in tumor necrosis factor-α signaling (glutathione S-transferase P1), and cholesterol and fatty acid metabolism (liver carboxylesterase and a protein of unknown function recently implicated in liver fibrogenesis [selenium binding protein 2]).

In summary, there is increasing evidence that iron, and potentially HFE mutations independent of body iron stores, play a role in the natural history of NAFLD in a subset of the population and contribute to both the first and second “hits” in the development of this disease through a variety of mechanisms such as altered lipid and glucose metabolism, insulin signaling, and the generation of oxidative stress.

James E. Nelson Ph.D.*, Kris V. Kowdley M.D.*, * Benaroya Research Institute, Virginia Mason Medical Center, Seattle, WA.