SIR – Recent studies have suggested the possibility that expansion of a polyglutamine tract in a yet unknown protein may be involved in a small subset of cases of schizophrenia.1, 2 Since expanded polyglutamine tracts causing several inherited neurodegenerative diseases always express the mutated protein in lymphoblasts, studies of schizophrenia were based on Western blots of protein extracts of lymphoblasts probed with an antibody (termed 1C2) which preferentially detects expanded polyglutamine tracts.3 A band in the 50–60 kDa range was found to be strongly reactive with the 1C2 antibody in two out of 32 unrelated childhood onset schizophrenia patients collected by Judith Rapoport and colleagues at the National Institute of Mental Health, and this band was identified as an acidic protein termed GRAP (Glutamine Repeat Acidic Protein).2 A band of approximately 50 kDa reactive with 1C2 was also detected in two out of 57 patients affected by adult onset schizophrenia.1 While the identification of a gene implicated in even a small subset of cases of schizophrenia would constitute a significant gain of know- ledge, there is a clear need for additional studies before proving that expanded polyglutamine tracts are involved.4 In this respect, one may pursue several research directions in parallel, the value of all of them increasing with the number of individuals tested. One important aspect is to investigate the possibility that the detection of polyglutamine expansion signals in schizophrenia patients might arise from artefacts at the level of protein expression. For example, regarding childhood onset schizophrenia, it is unlikely that there are age differences in the expression of GRAP.2 Variations in protein expression might also arise from differences in DNA methylation in promoter regions. A loss of methylation in the vicinity of the myotonic dystrophy (DM) untranslated CTG repeat subsequent to EBV transformation of lymphocytes has been recently reported.5 The authors raise the possibility that the detection of proteins with an expanded polyglutamine tract in schizophrenia patients might be an artefact caused by a loss of methylation near a long CAG repeat, and subsequent translation of this repeat into polyglutamines after lymphocyte transformation with EBV.5 However, it is interesting that a strong band of approximately 50 kDa was also found to react with the antibody 1C2 in protein extracts from post-mortem human brains of a limited number of schizophrenia patients (A Sharp and CA Ross, personal communication). Whether this 50-kDa band correlates with the polyglutamine proteins previously detected in some schizophrenia cases1, 2 will be investigated. In addition, while DM involves a very unstable CTG repeat which lies in a very large unmethylated CpG island,6 the CAG repeat which codes for an expanded polyglutamine tract found in schizophrenia is likely to be moderately enlarged. The vast majority of long and unstable CAG repeats previously detected in schizophrenia patients by using the Repeat Expansion Detection technique indeed corresponds to CAG repeat loci (Dir 1 and CTG18.1) which are also expanded in the normal population, and might not be involved in the disease.7 Consistently, the detection of an expanded polygluta- mine tract in GRAP does not correlate with Dir 1 and CTG18.1 sizes.2 A critical objective to further study polyglutamine tracts in schizophrenia is to test larger and/or more diverse cohorts of patients and controls for the presence of an expanded polyglutamine tract. While polyglutamine tracts do not appear to be involved in bipolar disorder,8 expansion of a polyglutamine tract in a protein of approximately 50 kDa has been detected with the antibody 1C2 in a small number of adult onset schizophrenia patients who belong to a large cohort comprising over 200 individuals (tested by the group of GA Rouleau, unpublished data). Since large cohorts would be best screened at the DNA level, an equally important objective being pursued is to identify the gene which codes for the protein with an expanded polyglutamine tract found in some schizophrenia patients. In this respect, it is useful to compare the proteins detected with the antibody 1C2 in the 50–60 kDa range in different studies of schizophrenia1, 2 for size and pI, ideally on the same electrophoretic gel. Preliminary results based on standard one-dimensional gel electrophoresis suggest that GRAP2 and the protein detected in adult onset schizophrenia1 migrate to the same position in the gel (unpublished data), and bi-dimensional gel electrophoresis will be performed to determine whether the two proteins are similar. Considering the diversity of CAG/CTG repeat features and dynamics in the human genome, it is expected that conducting detailed protein analysis in a series of patients and controls will efficiently determine whether expanded polyglutamine tracts are associated with schizophrenia.
This is a preview of subscription content, access via your institution