1. ¹Department of Psychiatry, Fudan University Affiliated Huashan Hospital, Shanghai, China
2. ²Shanghai Mental Health Center, Shanghai, China
3. ³Xinjiang Medical University Affiliated Traditional Chinese Medicine Hospital, Urumqi, China

Correspondence: Professor KD Jiang, Genetics, Shanghai Mental Health Center, 604 Lingling Road, Shanghai 200030, PR China. E-mail: jiangkaida@sh163.net

Received 7 September 2004; Revised 11 January 2005; Accepted 24 January 2005; Published online 15 March 2005.

Abstract

The etiology of schizophrenia is unclear, although family, twin, and linkage studies implicate genetic factors. Here, we identified adenomatous polyposis coli (APC), a tumor suppressor gene, as a risk factor for schizophrenia. We compared leukocytic gene expression patterns of six pairs of patients with schizophrenia and healthy controls by microarray. APC expression levels were significantly increased in all patients compared to healthy controls. To confirm the findings of microarray analysis, we measured expression levels of APC in the leukocytes from 30 relapse patients taking antipsychotic medication, 29 first-episode drug-naïve patients, and 30 healthy controls using real-time quantitative reverse transcription (RT)-polymerase chain reaction (PCR). APC expression levels were significantly increased in leukocytes of schizophrenics both taking and not taking antipsychotic medication and hence the increase of APC expression was not due to antipsychotic medication. APC is located at 5q21–22, which has been previously reported to be linked with schizophrenia. Further, we performed the transmission disequilibrium test (TDT) and TDT based on haplotypes to search for the association between schizophrenia and APC by examining 163 parent–offspring trios of Chinese descent. We analyzed three single-nucleotide polymorphisms (SNPs) (rs2229992, rs42427, rs465899) at the exon region of APC. TDT showed that the three SNPs are significantly associated with schizophrenia (TDT ²=4.23, P<0.05; 4.15, P<0.05; 8.49 P<0.01, respectively; HHRR²=5.54, P<0.05; 4.40, P<0.05; 9.79, P<0.01, respectively). We found a significant association between the APC haplotypes from rs2229992–rs42427–rs465899 and schizophrenia (Global ²=44.376,df=7, P<0.001). The C–A–T haplotype has a frequency of more than 57% and has a strong association with schizophrenia (²=15.04, P<0.001). These results indicate that the APC may be a candidate gene conferring susceptibility to schizophrenia and also may be associated with reduced vulnerability to cancer in schizophrenia.