Differential impact of glucose levels and advanced glycation end-products on tubular cell viability and pro-inflammatory/profibrotic functions.

High glucose (HG) or synthetic advanced glycation end-products (AGE) conditions are generally used to mimic diabetes in cellular models. Both models have shown an increase of apoptosis, oxidative stress and pro-inflammatory cytokine production in tubular cells. However, the impact of the two conditions combined has rarely been studied. In addition, the impact of glucose level variation due to cellular consumption is not clearly characterized in such experiments. Therefore, the aim of this study was to compare the effect of HG and AGE separately and of both on tubular cell phenotype changes in the HK2 cell line. Moreover, glucose consumption was monitored every hour to maintain the glucose level by supplementation throughout the experiments. We thus observed a significant decrease of apoptosis and H2O2 production in the HK2 cell. HG or AGE treatment induced an increase of total and mitochondrial apoptosis as well as TGF-ß release compared to control conditions; however, AGE or HG led to apoptosis preferentially involving the mitochondria pathway. No cumulative effect of HG and AGE treatment was observed on apoptosis. However, a pretreatment with RAGE antibodies partially abolished the apoptotic effect of HG and completely abolished the apoptotic effect of AGE. In conclusion, tubular cells are sensitive to the lack of glucose as well as to the HG and AGE treatments, the AGE effect being more deleterious than the HG effect. Absence of a potential synergistic effect of HG and AGE could indicate that they act through a common pathway, possibly via the activation of the RAGE receptors.

RAGE and CYBA polymorphisms are associated with microalbuminuria and end-stage renal disease onset in a cohort of type 1 diabetes mellitus patients over a 20-year follow-up.

AIMS: We investigated the association of polymorphisms of three genes implicated in oxidative stress: CYBA C242T, RAGE -374T/A and -429T/C, and ALOX12 Arg261Gln, with the delay of microalbuminuria onset in patients with type 1 diabetes mellitus (DT1). METHODS: A total of 162 T1D patients presenting with diabetes for 32.9 ± 9 years were included in the study; 53 had persistent microalbuminuria (>30 mg/l) and 109 did not. Onset of diabetes, microalbuminuria and end-stage renal disease (ESRD) were recorded as bio-clinical data. We determined polymorphism association of microalbuminuria with a Cox regression model. RESULTS: All polymorphisms respected the Hardy-Weinberg equilibrium. The Cox regression model validated four significant variables associated with microalbuminuria: RAGE 374AA (HR 4.19 [1.84-9.58] (p = 0.001)), CYBA TT+TC (HR 2.1 [1.16-3.80], p = 0.015), male sex (HR 1.92 [1.07-3.43], p = 0.028) and diabetes diagnosis at the pediatric stage (HR 1.85 [1.03-3.32], p = 0.039). The same association was found with ESRD (p = 0.028 and p = 0.033 for CYBA TC+TT and RAGE 374AA, respectively). CYBA C242T and RAGE 374T/A were not significantly associated with diabetic retinopathy. CONCLUSIONS: CYBA C242T and RAGE -374T/A correlate with microalbuminuria onset in the French DT1 cohort. The same correlation with ESRD onset supports the argument for the involvement of a genetic predisposition involving kidney-specific oxidative stress for diabetic nephropathy.