Hypofunction of the N-methyl-D-aspartate (NMDA) receptor has been hypothesized to underlie the pathophysiology of schizophrenia, based on the observation that non-competitive antagonists of the NMDA receptor, such as phencyclidine, induce schizophrenia-like symptoms. Mice lacking the NR2A subunit of the NMDA receptor complex are known to display abnormal behaviour, similar to schizophrenic symptoms. The expression of NR2A starts at puberty, a period corresponding to the clinical onset of schizophrenia. This evidence suggests that the NR2A (GRIN2A) gene may play a role in the development of schizophrenia and disease phenotypes. In this study, we performed a genetic analysis of this gene in schizophrenia. Analysis of the GRIN2A gene detected four single nucleotide polymorphisms, and a variable (GT)(n) repeat in the promoter region of the gene. A case-control study (375 schizophrenics and 378 controls) demonstrated evidence of an association between the repeat polymorphism and the disease (P = 0.05, Mann-Whitney test), with longer alleles overly represented in patients. An in-vitro promoter assay revealed a length dependent inhibition of transcriptional activity by the (GT)(n) repeat, which was consistent with a receptor binding assay in postmortem brains. Significantly, the score of symptom severity in chronic patients correlated with repeat size (P = 0.01, Spearman's Rank test). These results illustrate a genotype-phenotype correlation in schizophrenia and suggest that the longer (GT)(n) stretch may act as a risk-conferring factor that worsens chronic outcome by reducing GRIN2A levels in the brain.