The underlying cause(s) of methionine-dependency and its relevance to cancer remains unclear. We aimed to determine whether (i) normal human lymphocytes exhibit methionine-dependency, (ii) baseline levels of genetic damage are related to methionine-dependency and (iii) methionine-dependence can be explained, in part, by common polymorphisms in methionine synthase and methylenetetrahydrofolate reductase (MTHFR). Genetic damage was measured in lymphocytes of 52 volunteers (29--65 years) using the cytokinesis-block micronucleus assay. Methionine-dependency was assessed by culturing cells in serum-free media containing 0.1 mM L-methionine and 0 mM D,L-homocysteine (met(+)hcy(-)) or 0 mM L-methionine and 0.2 or 0.4 mM D,L-homocysteine (0.2/0.4-hcy(+))(met(-)hcy(+)). Mitogenesis was stimulated with phytohaemagglutinin. Cytokinesis was inhibited by adding cytochalasin B at 44 h. Ninety-six hours after PHA, cells were transferred to microscope slides. Cell proliferation was measured by counting binucleated cell frequency and calculating nuclear division index. Volunteers were classified into tertiles of methionine-dependence according to the growth of their cells in met(-)hcy(+) media (relative to growth in met(+)hcy(-) media). Average cell division, as a percentage of division in met(+)hcy(-) media, was approximately 5, 26 and 70% in 0.2-hcy(+) media and 29, 70 and 142% in 0.4-hcy(+) media for the high, mid and low tertiles of methionine-dependence, respectively. Micronucleus frequency did not vary between these tertiles (P > 0.6). In both met(+)hcy(-) and met(-)hcy(+) media, cell division was not affected by polymorphisms in MTHFR (C677T, A1298C) or methionine synthase (A2756G). Cell division in met(-)hcy(+) media was negatively correlated with division in met(+)hcy(-) media (P = 0.05 and 0.007 for 0.2 and 0.4-hcy(+), respectively). Methionine-dependent lymphocytes had higher levels of cell proliferation in met(+)hcy(-) media than methionine-independent lymphocytes (P = 0.089 and 0.01 for 0.2 and 0.4-hcy(+), respectively). However, this difference was not apparent in previous experiments when cells were grown in media containing 10% fetal calf serum. These findings show that there is a wide inter-individual variation in the degree of methionine-dependency of normal human lymphocytes in vitro. Methionine-dependency does not appear to alter the risk for chromosomal mutation as measured by the micronucleus assay. We discuss the possible relevance to cancer of increased cell division in methionine-dependent cells under methionine-replete and serum-free media conditions.