The Levels Of TXNIP In Different Glucose Tolerance Groups And The Relationship Between TXNIP And IL-1β

Objective: Diabetes mellitus has emerged as an important public healthand economic problem in the worldwide. The prevalence of diabetes is9.7%in Chinese adults. Among them, more than95%is Type2diabetesmellitus(T2DM).Insulin resistance (IR) and islet β cell dysfunction wasinvolved as the main pathogenesis of T2DM. Some studies have indicatedthat oxidative stress and inflammation were also involved in thepathogenesis of T2DM in past few years. But the exact etiology of thisdisease remained unclear. Recently, some animal studies have indicatedthat thioredoxin interacting protein(TXNIP) plays an important role in IRand insulin secretion in T2DM models. However, data on plasma TXNIPlevels in patients with diabetes were scarce. Our present study is aimed toinvestigate the plasma TXNIP levels in different glucose tolerance groupsand discuss the relationship between TXNIP and β-cell dysfunction/IR indiabetes and prediabetes. Moreover, we intend to further investigate thepotential relationship between TXNIP and proinflammatory Interleukin-1β (IL-1β).Methods: According to oral glucose tolerance test (OGTT) results,93participants were divided into3groups: diabetes mellitus(DM) group,prediabetes (PD) group and normal glucose tolerance (NGT) group. PlasmaTXNIP, IL-1β, malondialdehyde (MAD), superoxide dismutase(SOD),fasting blood glucose(FBG), glycosylated hemoglobin (HbA1c) and otherbiochemical index were measured in all the patients. Insulin sensitivity wasexpressed by calculated Homa-insulin resistance index(HOMA-IR). And βcell function was expressed by calculated HOMA-β cell function(HOMA-β).The relationship between TXNIP and glucose, IL-1β, SOD,MDA, HOMA-IR, HOMA-β were analysed by using multiple linearregression techniques and Pearson’s linear correlation analysis.Results: Age, BMI, SBP, DBP, TC, TG, LDL and HDL did not showsignificant differences in three groups(P>0.05)；Plasma TXNIP level washigher in PD group compared with NGT group(355.35±31.88vs.274.36±33.86ng/ml, P<0.05), but lower in PD group compared with DM group(355.35±31.88vs.426.16±63.15ng/ml, P<0.05). Plasma IL-1β and MDAwere higher in PD group compared with NGT group(22.06±2.60vs.18.27±2.31ng/l,4.85±0.48vs.3.80±0.61nmmol/mL, P<0.05), but lower in PDgroup compared with DM group(22.06±2.60vs.29.36±2.96ng/l,4.85±0.48vs.6.83±0.66nmmol/ml, P<0.05). Pearson’s linear correlation analysisindicated that TXNIP was positively correlated with IL-1β(r=0.799， P<0.05), MDA(r=0.763，P<0.05) and HOMA-IR(r=0.524，P<0.05), andnegatively correlated with SOD(r=-0.705, P<0.05) and HOMA-β(r=-0.518,P<0.05). Multiple linear regression analysis indicated that IL-1β hadsignificant influence on TXNIP(P＜0.05).Conclusion: In our study, Plasma TXNIP level is influenced by bloodglucose concentration. In high glucose condition, plasma TXNIP level issignificantly increased, and the level of TXNIP is positively correlated withIR, but negatively correlated with β cell function. Moreover, the correlationcoefficient of TXNIP on IR and β cell function are similar. In prediabetespatient, the TXNIP levels have already increased, which indicates thatTXNIP may be a potential indicator in the early stage of diabetes. In ourstudy, IL-1β is positively associated with TXNIP levels, and SOD isnegatively associated with TXNIP levels. Moreover, Multiple linearregression analysis indicates that L-1β is a main influence factor onTXNIP.