We prospectively assessed the impact of a sterol regulatory element-binding factor-2 (SREBF-2) polymorphism on the risk of developing nonalcoholic fatty liver disease (NAFLD) and on liver histology and lipoprotein and glucose metabolism in biopsy-proven NAFLD. In a population-based study, we followed 175 nonobese, nondiabetic participants without NAFLD or metabolic syndrome at baseline, characterized for the SREBF-2 rs133291 C/T polymorphism, dietary habits, physical activity, adipokines, C-reactive protein (CRP), and endothelial adhesion molecules. A comparable cohort of NAFLD patients underwent liver biopsy, an oral glucose tolerance test with minimal model analysis to yield glucose homeostasis parameters, and an oral fat tolerance test with measurement of plasma lipoproteins, adipokines, and cytokeratin-18 fragments. After 7 years, 27% of subjects developed NAFLD and 5% developed diabetes. SREBF-2 predicted incident NAFLD and diabetes and CRP and endothelial adhesion molecule changes. In biopsy-proven NAFLD patients, SREBF-2 predicted nonalcoholic steatohepatitis (odds ratio 2.92 [95% CI 2.08–4.18], P = 0.002) and the severity of tissue insulin resistance, β-cell dysfunction, and oral fat intolerance (characterized by higher postprandial lipemia, cholesterol enrichment of triglyceride-rich lipoproteins and oxidized LDLs, HDL cholesterol fall, adipokine imbalance, and postprandial apoptosis activation). An SREBF-2 polymorphism predisposes individuals to NAFLD and associated cardiometabolic abnormalities and affects liver histology and glucose and lipid metabolism in biopsy-proven NAFLD.