Recently, Koike et al. 1 identified a 25-bp deletion in a coding exon of the Disrupted-In-Schizophrenia (DISC1) gene in the 129S6/SvEv strain, which was also confirmed at the genomic level for all extant 129 mouse inbred substrains. 2 This mutation could interfere with the production of the full-length DISC1 protein. When the 129S6/SvEv-derived DISC1 gene was transferred to C57BL/6J genetic background, the resultant mice displayed a subtle behavioral abnormality in working memory, without any other major deficits in behavior and brain anatomy. 1 Several lines of evidence from animal models with RNA interference to DISC1, as well as studies in patient-derived lymphoblasts, suggest that loss of DISC1 function may be involved in the abnormalities underlying the pathophysiology of major mental conditions. 3 The minor behavioral abnormality resulting from the 25-bp deletion is in contrast to the results of three independent groups who have generated partial lossof-function models by expressing dominant-negative DISC1 constructs and obtained substantial changes in behavior, including deficits in prepulse inhibition, latent inhibition and working memory. 4 Moreover, the absence of obvious anatomical changes in the mutant mice1 conflicts with cellular models based on the knockdown of DISC1 or expression of dominantnegative DISC1, which disrupts developmental processes critical for normal cortical architecture. 5 Thus, if the 25-bp deletion completely abolishes the fulllength DISC1 protein that is crucial for proper neurodevelopment, why are there such small phenotypic changes in mice with this mutation? To address this question, our collaborative group examined expression of DISC1 protein in the 129S6/SvEv strain, and systematically compared expression with that in the C57BL/6J strain by using antibodies against more than 10 independent epitopes that were generated from eight independent groups, including commercially available antibodies (Figure 1a). Although DISC1 has several isoforms, we focused our analyses of immunoreactivity on the one that corresponded to full-length DISC1. Interestingly, all the antibodies except the one that was a generous gift from Dr Joseph A Gogos1 showed almost identical immunoreactivity to DISC1 at this molecular weight band, or at least immunoreactivity that was indistinguishable at the detection levels, between these two strains (Figure 1b and Supplementary Figure 1). The specificity of these antibodies against DISC1 was tested by western blotting with extracts from monkey COS7 cells expressing full-length C57BL/6J-derived mouse DISC1 or with mock transfection (Supplementary Figure 2). The specificity of the mExon3 antibody was also confirmed by preabsorption tests by both western blotting and immunohistochemistry (Supplementary Figure 3). D27 is the best-characterized antibody against mouse DISC1 in both western blotting and immunohistochemistry. D27 recognizes two distinct bands at 100–110kDa, corresponding in size to full-length DISC1. 6 We conducted immunoprecipitation (IP) with mouse brain extracts by using the mExon3 antibody, which was followed by western blotting with the D27 antibody, and found that the molecule corresponding to the lower band detected by D27 was specifically precipitated by mExon3. Consistent with the results of western blotting, we failed to observe any difference in IP between the 129S6/SvEv and the C57BL/6J strains.