Type 2 diabetes (T2D) has a replicated linkage on chromosome12q24.2, the non-insulin-dependent-diabetes 2 (NIDDM2) locus. PSMD9 (which rarely causes T2D in Italians) lies in the NIDDM2 region and is implicated in beta cell insulin transcription and diabetes onset in mice. Thus, PSMD9 is a candidate T2D gene for the NIDDM2 locus. We aimed at identifying any linkage of the PSMD9 A/T/G haplotype, or of any of its single variants, to Italian T2D siblings/families. We screened 201 T2D siblings/families for PSMD9 variants and performed a parametric and non-parametric linkage study, including linkage disequilibrium (LD) modeling and simulation analyses. Our results show a consistent significant LOD score in linkage with T2D for each single PSMD9 SNP variant (IVS3 + nt460A, P = 3.546E - 007, IVS3 + nt437T, P = 7.723E - 008, and 197G, P = 4.921E - 007) and for the haplotype (A/T/G for the above-cited variants, P = 3.078E - 015) using the non-parametric analysis, as well as the LD modeling test (P = 4.178E - 009) and the parametric linkage analysis. The strongest signal is present under the recessive model (P = 4.905E - 011). Our statistical power in the present study relies on the presence of T2D affected siblings, which represent an ideal dataset to identify linkage with a recessive disease model. Our 1,000 simulation analyses, performed for each single test, confirm that the results are not due to random chance. In summary, the A/T/G haplotype in PSMD9 is linked via the recessive allelic model to T2D in Italians. By our observation and testing, the linkage strategy can identify a gene contributing to T2D in a homogeneous subject dataset.
(c) 2009 Wiley-Liss, Inc.