Application Of Protein Fingerprinting Model In The Diagnosis Of Type 2 Diabetic Nephropathy

Objective:There is great interest in identifying molecules whose levels are significantly altered in type 2 diabetic nephropathy, because this knowledge may provide clues regarding the pathogenesis of this disorder. This study aims to screen biomarkers in the serum from patients with type 2 diabetic nephropathy using surface-enhanced laser desorption-ionization time of flight mass spectrometry (SELDI-TOF-MS), and coupled with artificial neural network (ANN) to develop a diagnostic model and applied to diagnose type 2 diabetic nephropathy.Methods:SELDI-TOF-MS and CM 10 (cationic chip array) protein chips were used to detect the serum proteomic patterns of 51 serum samples from patients with type 2 diabetic nephropathy and 66 matched controls. Patients and controls were randomly assigned into a training group (n=70, 30 patients with type 2 diabetic nephropathy, 40 controls) and a testing group (n=47, 21 patients with type 2 diabetic nephropathy, 26 controls). An ANN diagnostic model was trained on data from the training set, and the testing set were analyzed with blind prediction using established model to validate its diagnostic efficiency.Results:Total of 175 peaks were obtained from the serum samples. From these peaks, about 17 differentially expressed mass peaks (P<0.01) were chosen to establishing and training an ANN model. Six specific mass peaks (M/Z at 5420,5758,6435,6634,10283 and 11744 ) were chosen to develop an ANN diagnostic model and showed a high efficiency. This model was blind tested and yielded a sensitivity of 81.0%, a specificity of 96.2%.Conclusion:Our results suggested these specific biomarkers can provide a novel mode for the laboratory diagnosis of type 2 diabetic nephropathy, even the potency in early diagnosis and differentiation of type 2 diabetic nephropathy as a whole. We may draw the conclusion that protein profiling using serum from individuals with type 2 diabetic nephropathy would be especially informative, since such studies may lead to a better understanding of the molecular mechanisms underlying type 2 diabetic nephropathy, as well as the development of diagnostic and prognostic tests.