Background: Proteomic analysis is a promising tool for biomarker discovery.
Objectives: To explore the urinary proteome of cats with chronic kidney disease (CKD). Animals: Seventeen client-owned cats with naturally occurring CKD (NatCKD); 14 purpose-bred cats with surgically induced CKD (remnant kidney model, RemCKD); and 17 or 14 community-owned healthy cats that were age- and sex-matched to group NatCKD or RemCKD, respectively.
Methods: Retrospective study using banked urine from previous prospective studies. A tandem mass spectrometry proteomic protocol was established for feline urine samples. Data were processed using software MaxQuant and R, analyzing the presence and relative quantities of known feline proteins. Each protein’s quantity was compared between CKD and the corresponding control group; for proteins showing diagnostic potential, receiver operating characteristic (ROC) curves were drawn.
Results: Six hundred nineteen and 592 proteins were identified in the naturally occurring or surgically induced CKD dataset, respectively. Among these, 194 and 241 proteins met the qualitative control criteria and were included in further analysis. Thirty-four proteins differed significantly (Benjamini-Hochberg adjusted p < .05) between NatCKD and its corresponding control by > 2 fold; no protein differed significantly between RemCKD and its corresponding control. Retinol binding protein 4 (RBP4) was the top contributor of protein composition variation in both datasets. ROC curves for RBP4 showed areas under the curve of 0.945 (95% confidence interval, 0.865-1.000) and 0.781 (95% confidence interval, 0.603-0.958) for differentiating NatCKD or RemCKD from their respective control groups.
Conclusions: Urinary proteomic analysis revealed urinary RBP4 as promising CKD marker in cats.
