BoxCar increases the depth and reproducibility of diabetic urinary proteome analysis

Abstract

Purpose: Mass spectrometry-based proteomics performs well in high throughput detection of urinary proteins. Nonetheless, protein identification depth and reproducibility remain the challenges in diabetic urinary proteome with high complexity and broad dynamic range, especially for low-abundant proteins. As a new data acquisition strategy, the BoxCar method was reported to benefit for low-abundant protein identification. Whether it is propitious to diabetic samples with high dynamic range proteomes has not been discussed yet. We aimed to apply BoxCar method to diabetic urine sample analysis, and to compare it with standard data dependent acquisition (DDA) method on protein identification in detail. Experimental Design:We performed seven technical replicates analysis on two urine samples from healthy individuals and diabetic patients to evaluate protein detection of BoxCar and standard DDA methods on single sample. Further comparison of two methods was made on multiple diabetic urine samples.Results: BoxCar could increase over 20% of identified proteins and performed better quantitative reproducibility than standard DDA method either in single or multiple diabetic urinary samples. BoxCar also improved the detection of low-abundant proteins.Functional enrichment analysis of normal albuminuria or microalbuminuria samples indicated that BoxCar acquired more diabetes-related biological information. Conclusions and Clinical Relevance:The study demonstrates that BoxCar could enhance the depth and reproducibility in diabetic urinary proteome analysis, which provides reference for mass spectrometry approach selection in clinical urinary proteomic research.