The pathogenesis of hepatic fibrosis and cirrhosis in genetic hemochromatosis may involve a direct effect of excess iron on collagen synthesis in the liver. To investigate this theory, we measured procollagen messenger RNA levels (types I, III and IV) in the livers of rats in which we produced chronic parenchymal iron overload by feeding them dietary carbonyl iron (2.5%, wt/wt) for up to 18 mo. This feeding resulted in predominantly parenchymal iron deposition in a periportal distribution similar to that seen in genetic hemochromatosis. Increased amounts of collagen fibrils were observed in iron-loaded livers on electron microscopy; all iron-loaded livers showed some periportal fibrosis. Although very high hepatic iron concentrations (range = 340 to 1,100 mumol/gm dry wt) were achieved in the carbonyl iron-loaded rats, we saw no consistent difference between steady-state messenger RNA levels for procollagens types I, III and IV in control and iron-loaded livers examined at five different time points up to 18 mo. Messenger RNA levels of the cytokine transforming growth factor-beta1, which has been implicated as having a role in the production of extracellular matrix proteins, were also measured. No significant differences were observed between iron-loaded and control livers. These results suggest that excess parenchymal iron does not have a direct effect on the expression of the procollagens or transforming growth factor-beta1 genes in iron-loaded livers and that factors other than, or in addition to, iron are necessary for fibrosis to occur.