Type 2 diabetes is both a phenotypically and geneotypically heterogeneous disease. It is caused by defective insulin secretion and action. Protein tyrosine phosphatases (PTPases) play important roles in insulin cascade signal transduction and have been suggested to be related to insulin resistance (1,2). PTP-1B, a member of the PTP family, is expressed widely in many tissues, acting as a negative regulator in the insulin receptor signal transduction pathway (3-5). The PTP-1B gene is located on the long arm of human chromosome 20, in the region of q13.1-q13.2, which has been linked to quantitative trait loci of obesity and insulin (6,7). A recent study by Echwald S.M. et al. (8) demonstrated that a Pro387Leu variation of the PTP-1B gene, which resulted in the impairment of the serine phosphorylation of the PTP-1B peptide (in vitro experiment), was associated with type 2 diabetes in a Danish Caucasian population with a genotype relative risk of 3.7 (CI 1.26～10.93, P=0.02). Since studies involving the association between the genetic variations and type 2 diabetes are often controversial and inconsistent in different ethnic populations, we tested the association between the Pro387Leu variation of PTP-1B gene with type 2 diabetes in a Chinese Han population for the first time. The Pro387Leu variation of PTP-1B gene was detected usingPCRand restriction ragment–length polymorphism in 589 subjects chosen from the Han population living in southern China, including 329 type 2 diabetic patients (men/women 143/186, age 59.4±9.9 years, BMI 23.9±3.5kg/m2) and 238 control subjects (men/women 100/138, age 57.5±8.3 years, BMI 23.8±3.1kg/m2). The control subjects underwent a 75-g oral glucose tolerance test and were diagnosed with normal glucose tolerance (NGT) in accordance with the 1997 American Diabetes Association criteria. The study was approved by the ethnics committee of our institution. All the subjects gave informed consent. In our study, only two subjects heterozygous for the mutation were found in the NGT control group, with genotype and allele frequencies of 0.008 and 0.004, respectively. We found another two heterozygotes in the diabetic patient group; the genotype and allele frequencies were 0.006 and 0.003, respectively. The differences did not reach statistical significance between groups (P= 0.05 for both). The distribution was consistent with Hardy-Weinberg equilibrium. We then examined the impacts of the mutation on metabolic and anthropometric parameters in both groups. Among NGT control subjects, there were no significant differences in age, fasting plasma glucose (FPG), or lipid profile between the two subgroups with or without the Leu387 mutation (P=0.05), while BMI was significantly higher in subjects with the Leu387 allele (23.74±3.05 vs. 28.55±2.19 kg/m2, P=0.027). In the diabetic patient group, no differences were observed in age, BMI, FPG, HbA1c, Cpeptide, or lipid profile (P=0.05). Since the mutation rate was quite low in the examined Chinese Han population and at the same time there were 31 subjects with a BMI＞27kg/m2 in the subgroup without the Leu387 mutation, the difference found in BMI between the mutation carriers and noncarriers in the control group was likely attributed to individual variance rather than the true difference caused by the presence of the mutation. In conclusion, our data indicated that the mutation of Pro387Leu in PTP-1B gene was present in the Chinese Han population examined, but this variation wa