| Resumo: |
Owing to its availability, ease of collection and correlation with (patho-) physiology, urine is an attractive source for clinical proteomics. However, the lack of comparable datasets from large cohorts has greatly hindered development in this field. Here we report the establishment of a high resolution proteome database of naturally occurring human urinary peptides and proteins - ranging from 800-17,000 Da - from over 3,600 individual samples using capillary electrophoresis coupled to mass spectrometry, yielding an average of 1,500 peptides per sample. All processed data were deposited in an SQL database, currently containing 5,010 relevant unique urinary peptides that serve as classifiers for diagnosis and monitoring of diseases, including kidney and vascular diseases. Of these, 352 have been sequenced to date. To demonstrate the applicability of this database, two examples of disease diagnosis were provided: For renal damage diagnosis, patients with a specific renal disease were identified with high specificity and sensitivity in a blinded cohort of 131 individuals. We further show definition of biomarkers specific for immunosuppression and complications after transplantation (Kaposi's sarcoma). Due to its high information content, this database will be a powerful tool for the validation of biomarkers for both renal and non-renal diseases.
|
