Although many opinions have been offered regarding the penetrance of hemochromatosis in homozygotes for the 845G>A (C282Y) HFE mutation, a controlled study has shown that very few of these individuals develop clinical disease.1 Clearly, the homozygous state is a necessary but not sufficient condition for the development of the clinical disease. Finding the other factors that may play a significant role is of great importance, particularly if population screening for hemochromatosis is to be carried out. The studies of Fargion et al2suggesting that polymorphisms in the tumor necrosis factor alpha (TNF-α) promoter may be such modifiers of the hemochromatosis phenotype were, therefore, of special interest. In particular, it was suggested that the polymorphism at nucleotide 238 may protect against cirrhosis in homozygotes for hemochromatosis. The population studied by Fargion et al was drawn from patients who had been diagnosed clinically as having hemochromatosis. We have had the opportunity to genotype a large number of patients attending a health appraisal clinic, patients with a median age of 56 who represent the full spectrum of the phenotype associated with the homozygous state for the C282Y mutation. We found that among the 152 homozygotes detected there was a very broad range of putative disease manifestations and ferritin levels.1 If TNF-α genotypes had an effect on the hemochromatosis phenotype we might expect to see it in this population.
Collagen IV levels are an excellent surrogate for hepatic fibrosis in hemochromatosis,3 and, indeed, we have found that there is a slight excess of individuals with elevated collagen IV levels in homozygotes for the C282Y mutation, even among those without manifest liver disease.1 Figure 1shows the relationship of serum collagen IV levels to TNF-α promoter genotype in homozygotes for the C282Y mutation. Clearly, there is no major difference between these measurements of liver damage in the population that we have studied. Table 1compares the serum collagen, aspartate transaminase (AST), and ferritin levels in homozygotes for the C282Y mutation with different TNF-α promoter genotypes.
The distribution of plasma collagen IV levels in subjects with the C282Y/C282Y genotype.
Plasma collagen IV levels are a surrogate for hepatic fibrosis. The horizontal lines represent the geometric mean of collagen levels. Although the means are slightly lower in the heterozygous subjects, the difference does not approach statistical significance. Neither does a nonparametric analysis (Mann-Whitney) show a statistically significant difference.
The distribution of plasma collagen IV levels in subjects with the C282Y/C282Y genotype.
Plasma collagen IV levels are a surrogate for hepatic fibrosis. The horizontal lines represent the geometric mean of collagen levels. Although the means are slightly lower in the heterozygous subjects, the difference does not approach statistical significance. Neither does a nonparametric analysis (Mann-Whitney) show a statistically significant difference.
TNF-α promoter polymorphisms may be a risk factor for liver damage; if so, the effect is so small that it cannot be shown even with a group of more than 100 patients homozygous for the C282Y mutation. There must be other, more powerful, influences on the expression of theHFE mutations. So far our attempt to find such polymorphisms in the coding regions or promoters of HFE, calreticulin, β2-microglobulin, transferrin, transferrin receptor-2, DMT1 (nRamp2), ferroportin, transferrin receptor-1, IRP-1, IRP-2, hepcidin, the ferritin light and heavy chains, and ceruloplasmin have been unsuccessful. The search for genetic and environmental factors that influence the hemochromatosis phenotype must go on.
Supported by grant DK53505-02 from the National Institutes of Health and the Stein Endowment Fund.
