American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 127B:5–10 (2004) Evidence for Linkage Between Regulatory Enzymes in Glycolysis and Schizophrenia in a Multiplex Sample William S. Stone, 1,2 * Stephen V. Faraone, 2,4 Jessica Su, 1,2 Sarah I. Tarbox, 1 Paul Van Eerdewegh, 1,3 and Ming T. Tsuang 1,2,3,4 1 Harvard Medical School Department of Psychiatry at Massachusetts Mental Health Center, Boston, Massachusetts 2 Harvard Institute of Psychiatric Epidemiology and Genetics, Massachusetts 3 Department of Epidemiology, Harvard School of Public Health, Massachusetts 4 Psychiatry Service and Harvard Department of Psychiatry, Massachusetts General Hospital, Massachusetts Observations of impaired glucose regula- tion in schizophrenia are long-standing, although their pathological and etiological significance is uncertain. One approach to the issue that minimizes environmental variables (e.g., medication and diet) is to determine whether genes related to glucose regulation show genetic linkage to schizo- phrenia. We examined the potential role of glucose metabolism in schizophrenia through a genome scan of affection status in schizophrenia and an empirical method for deriving P-values. Data were utilized from the NIMH Genetics Initiative for Schi- zophrenia dataset, which comprises a total sample consisting of 71 pedigrees containing 218 nuclear families and 987 individuals. A genome scan with 459 markers spaced at an average of 10 cM intervals was con- ducted using the linkage analysis program Genehunter separately for European- and African-American groups. Enzymes that regulate glycolysis were identified and the genes regulating these enzymes were locat- ed through the Online Mendelian Inheri- tance in Man (OMIM) website. The focus in this study was on genes located near pre- viously reported schizophrenia susceptibil- ity regions. The genome-wide significance of these genes to schizophrenia was as- sessed using permutation testing. When re- sults were adjusted for multiple testing within and across ethnic groups, 6-phospho- fructo-2-kinase/fructose-2,6-bisphosphatase 2 (PFKFB2; chromosome 1q32.2) achieved genome-wide significance (P ¼ 0.04). In addi- tion, hexokinase 3 (HK3; chromosome 5q35.3) was also suggestive of linkage (P ¼ 0.09). For the European-American sample, PFKFB2 (1q32.2), hexokinase 3 (HK3; 5q35.3), and pyruvate kinase 3 (PK3; chromosome 15q23) achieved significance at the 0.05 level. None of the genes showed significance in the African-American sample. Our results pro- vide further support for the view that genes that regulate glucose metabolism may also influence susceptibility to schizophrenia. More generally, they support the view that relationships between glucose dysregula- tion and schizophrenia are inherent to the disorder, and are not merely epiphenomena related to medication or other treatment factors. ß 2003 Wiley-Liss, Inc. KEY WORDS: genetics; schizophrenia; linkage; glycolysis; phospho- fructokinase; hexokinase INTRODUCTION Abnormal glucose regulation in schizophrenia has been reported over much of the last century [e.g., Kooy, 1919; Braceland et al., 1945; Franzen and Nilsson, 1968]. Between 1930s and 1950s (when neuroleptics Grant sponsor: The National Institute of Mental Health (to Dr. Ming T. Tsuang); Grant numbers: T32 MH 17119, U01- MH46318, 2R01-MH43518, R01-MH 50647, R25 MH 60485; Grant sponsor: The Veterans Administration’s Medical Research, Health Services Research, and Development and Cooperative Studies Programs; Grant sponsor: NARSAD Distinguished Inves- tigator Award (to Dr. Tsuang); Grant sponsor: NARSAD Young Investigator Award (to Dr. Stone). Sarah I. Tarbox’s present address is Department of Psychology, University of Pittsburgh, Pittsburgh, PA. *Correspondence to: William S. Stone, Ph.D., Massachusetts Mental Health Center, 74 Fenwood Road, Boston, MA 02115. E-mail: william_stone@hms.harvard.edu Received 21 October 2002; Accepted 8 August 2003 DOI 10.1002/ajmg.b.20132 ß 2003 Wiley-Liss, Inc.