One of the best examples of a gene-environment interaction in autoimmunity involves instances in which certain drugs produce a lupus-like syndrome in genetically susceptible individuals. It turns out that people who carry a variant of the N-acetyltransferase-2 gene, so-called slow acetylators, do not metabolize the drugs procainamide and hydralazine very well. Thus, they are predisposed to lupus induced by these drugs because these substances remain active for extended periods in the body. Removing the drugs, however, eliminates the autoimmunity.

Although the mechanisms that cause drug-induced lupus are believed to be somewhat different from those of systemic lupus erythematosus (SLE), the similarities encouraged some investigators to look for a connection. According to Bruce Richardson, a professor of medicine at the University of Michigan Medical School, one link may involve DNA methylation, a reaction that plays a key role in suppressing gene expression (E. Ballestar et al., "The epigenetic face of systemic lupus erythematosus," J Immunol, 176:7143-7, 2006).

Richardson and his colleagues discovered that procainamide and hydralazine inhibit the methylation of DNA in T cells, so genes that would otherwise be silenced by methylation are now expressed. "If you look at women with idiopathic lupus [SLE], they have decreased levels of DNA-methyltransferase in their T cells," Richardson says. Autoreactive T cells in these women may be over-expressing genes that are causing the SLE, according to Richardson.

Richardson and his colleagues identified six genes that are over-expressed in lupus patients, including the TNFSF7 gene, which codes for CD70, a T-cell surface molecule that stimulates B cells. "The autoreactive T cells hyperstimulate the B cells to make immunoglobulin [antibodies], and also promiscuously kill every macrophage they contact," says Richardson. Without macrophages, which normally clean up cellular debris from dead cells, chromatin from apoptic cells drives the anti-DNA response seen in lupus, according to Richardson.

What environmental factors might contribute to the lupus-associated DNA hypomethylation? Richardson doesn't know for sure, but it could come from diet. For example, crucial components in DNA methylation come from an essential amino acid, methionine, as well as various B vitamins, zinc, and choline. A diet deficient in these materials might contribute to the onset of lupus.

Richardson adds another intriguing observation: If lupus is a disease that involves a genetic predisposition, why doesn't it typically appear until people are in their 30s and 40s? According to Richardson, DNA-methyltransferase levels decrease as one ages. "It may be that you can tolerate a poor diet when you're younger, when DNA-methyltransferase levels are high," he says, "but as your levels drop you become susceptible to agents that may demethylate DNA until you get lupus, assuming you've got the predisposing genes." Unfortunately, there are no data to support the hypothesis that a poor diet is linked to lupus.