Two genes (MAT1A and MAT2A) encode for the essential enzyme methionine adenosyltransferase (MAT), which catalyzes the biosynthesis of S-adenosylmethionine (SAMe), the principal methyl donor and, in the liver, a precursor of glutathione. MAT1A is expressed mostly in the liver, whereas MAT2A is widely distributed. MAT2A is induced in the liver during periods of rapid growth and dedifferentiation. In human hepatocellular carcinoma (HCC) MAT1A is replaced by MAT2A. This is important pathogenetically because MAT2A expression is associated with lower SAMe levels and faster growth, whereas exogenous SAMe treatment inhibits growth. Rats fed ethanol intragastrically for 9 weeks also exhibit a relative switch in hepatic MAT expression, decreased SAMe levels, hypomethylation of c-myc, increased c-myc expression, and increased DNA strand break accumulation. Patients with alcoholic liver disease have decreased hepatic MAT activity owing to both decreased MAT1A expression and inactivation of the MAT1A-encoded isoenzymes, culminating in decreased SAMe biosynthesis. Consequences of chronic hepatic SAMe depletion have been examined in the MAT1A knockout mouse model. In this model, the liver is more susceptible to injury. In addition, spontaneous steatohepatitis develops by 8 months, and HCC develops by 18 months. Accumulating evidence shows that, in addition to being a methyl donor, SAMe controls hepatocyte growth response and death response. Whereas transient SAMe depletion is necessary for the liver to regenerate, chronic hepatic SAMe depletion may lead to malignant transformation. It is interesting that SAMe is antiapoptotic in normal hepatocytes, but proapoptotic in liver cancer cells. This should make SAMe an attractive agent for both chemoprevention and treatment of HCC.