Neurotensin and its high affinity receptor (NTSR1) localise within dopaminergic neurones in the mesocortical, mesolimbic and nigrostriatal systems and it is now clear that neurotensin can selectively modulate dopaminergic neurotransmission. This has led to the hypothesis that altered neurotensin function contributes to the pathogenesis of schizophrenia and other psychoses. This hypothesis has been supported circumstantially by a number of lines of evidence. (1) Central administration of neurotensin produces effects similar to those produced by the peripheral administration of atypical antipsychotics. (2) Observations of low levels of neurotensin in the CSF of schizophrenics. (3) Reduced numbers of neurotensin receptors in the brains of schizophrenics. Given the above link between neurotensin and dopamine, and the evidence implicating altered neurotensin function in psychosis, we have postulated that DNA sequence variation in neurotensin or its receptors might be associated with schizophrenia. In keeping with this hypothesis, an association has recently been reported between schizophrenia and the gene encoding the neurotensin high affinity receptor (NTSR1). However, caution is required because the associated marker, a tetranucleotide repeat, is located 3 kb away from the 3' end of the gene and there is no evidence that it is functional. Therefore, as a follow-up to our earlier work on neurotensin, we have now sought to test the hypothesis that DNA sequence variants that alter the structure or expression of the NTSR1 gene (VAPSEs) are associated with schizophrenia. However, while we found 14 novel sequence variants in 28 probands with psychosis, none resulted in an amino acid change, and neither direct nor indirect association studies suggested these are involved in susceptibility to schizophrenia.