In human nephrectomy specimens, the kidney level of tubular transport proteins does not correlate with their abundance in urinary extracellular vesicles

Abstract

Human urinary extracellular vesicles (uEVs) contain proteins from all nephron segments. An assumption for years has been that uEVs might provide a noninvasive liquid biopsy that reflect physiological regulation of transporter protein expression in humans. We hypothesized that protein abundance in human kidney tissue and uEVs are directly related and tested this in paired collections of nephrectomy tissue and urine sample from 12 patients. Kidney tissue was fractioned into total kidney protein, crude membrane (plasma membrane and large intracellular vesicles)-enriched, and intracellular vesicle-enriched fractions as well as sections for immunolabeling. uEVs were isolated from spot urine samples. Antibodies were used to quantify six segment-specific proteins [proximal tubule-expressed Na+-phosphate cotransporters (NaPi-2a), thick ascending limb-expressed Tamm-Horsfall protein and renal outer medullary K+ channels, distal convoluted tubule-expressed NaCl cotransporters, intercalated cell-expressed V-type H+-ATPase subunit G3 (ATP6V1G3), and principal cell-expressed aquaporin 2] and three uEV markers (exosomal CD63, microvesicle marker vesicle‐associated membrane protein 3, and β-actin) in each fraction. By Western blot analysis and immunofluorescence labeling, we found significant positive correlations between the abundance of CD63, NaCl cotransporters, aquaporin 2, and ATP6V1G3, respectively, within the different kidney-derived fractions. We detected all nine proteins in uEVs, but their level did not correlate with kidney tissue protein abundance. uEV protein levels showed higher interpatient variability than kidney-derived fractions, indicating that factors, besides kidney protein abundance, contribute to the uEV protein level. Our data suggest that, in a random sample of nephrectomy patients, uEV protein level is not a predictor of kidney protein abundance.