Background: Urinary proteomics identifies the totality of urinary proteins and can therefore help in getting an early and precise diagnosis of various pathological processes in the kidneys. In infants, non-invasive urine collection is most commonly accomplished with a urine bag or clean catch. The influence of those two collection methods on urinary proteomics was assessed in this study.
Methods: Thirty-two urine samples were collected in infants using urine bag and clean catch within 24 h. Nine boys and seven girls with a mean age of 4.3 ± 2.9 months were included (5 × post-pyelonephritis, 10 × non-kidney disease, 1 × chronic kidney disease (CKD)). Liquid chromatography-mass spectrometry (LC-MS/MS) was performed in data-independent acquisition (DIA) mode. Protein identification and quantification were achieved using Spectronaut.
Results: A total of 1454 urinary proteins were detected. Albumin and α-1-microglobulin were detected the most. The 18 top-abundant proteins accounted for 50% of total abundance. The number of proteins was slightly, but insignificantly higher in clean catch (957 ± 245) than in bag urine (876 ± 255). The median intensity was 1.2 × higher in the clean catch. Overall, differential detection of proteins was 29% between the collection methods; however, it diminished to 3% in the 96 top-abundant proteins. Pearson's correlation coefficient was 0.81 ± 0.11, demonstrating a high intraindividual correlation. A principal component analysis and a heat map showed clustering according to diagnoses and patients rather than to the collection method.
Conclusion: Urinary proteomics shows a high correlation with minor variation in low-abundant proteins between the two urine collection methods. The biological characteristics overrule this variation. Graphical abstract A higher resolution version of the Graphical abstract is available as Supplementary information.
Keywords: Chronic kidney disease; Infants; Liquid chromatography-mass spectrometry; Urinary proteomics; Urine collection.
© 2023. The Author(s).