Predicting the protein composition of human urine in normal and pathological states: Quantitative description based on Dent1 disease (CLCN5 mutation).

ABSTRACT

KEY POINTS The presence of plasma proteins in urine is difficult to interpret quantitatively. It may be a result of impaired glomerular filtration or impaired proximal tubule (PT) reabsorption, or both. Dent1 disease (CLCN5 mutation) abolishes PT protein reabsorption leaving glomerular function intact. Using urine protein measurements from patients with Dent1 disease and normal individuals, we devised a mathematical model that incorporates two PT transport processes with distinct kinetics. This model predicts albumin, α1 -microglobulin (α1 -m), ß2 -microglobulin (ß2 -m) and retinol-binding protein 4 (RBP4) urine concentrations. Our results indicate that the urinary excretion of ß2 -m and RBP4 differs from that of albumin and α1 -m in their sensitivity to changes in the glomerular filtration rate, glomerular protein leak, tubular protein uptake via endocytosis and PT water reabsorption. The model predicts quantitatively how hyperfiltration and glomerular leak interact to promote albuminuria. Our model should contribute to improved understanding and interpretation of urine protein measurements in renal disease. ABSTRACT To clarify the relative contributions of glomerular filtration and tubular uptake to urinary protein excretion, we developed a mathematical model of protein reabsorption in the human proximal tubule (PT) using Michaelis-Menten kinetics and molar urinary protein measurements taken from human Dent1 disease (CLCN5 loss-of-function mutation). ß2 -Microglobulin (ß2 -m) and retinol-binding protein 4 (RBP4) are normally reabsorbed with 'very high' efficiency uptake kinetics and fractional urinary excretion of 0.025%, whereas albumin and α1 -microglobulin (α1 -m) are reabsorbed by 'high' efficiency uptake kinetics and 50-fold higher fractional urinary excretion of 1.15%. Our model correctly predicts the urinary ß2 -m, RBP4 and α1 -m content in aristolochic acid nephropathy, and elevated ß2 -m excretion with increased single nephron glomerular filtration rate (SNGFR) following unilateral-nephrectomy. We explored how altered endocytic uptake, water reabsorption, SNGFR and glomerular protein filtration affect excretion. Our results help to explain why ß2 -m and RBP4 are more sensitive markers of PT dysfunction than albumin or α1 -m, and suggest that reduced PT sodium and water reabsorption in Fanconi syndrome may contribute to proteinuria. Transition of albumin excretion from normal to microalbuminuria, a 5-fold increase, corresponds to a 3.5-fold elevation in albumin glomerular filtration, supporting the use of microalbuminuria screening to detect glomerular leak in diabetes. In macroalbuminuria, small albumin permeability changes produce large changes in excretion. However, changes in SNGFR can alter protein excretion, and hyperfiltration with glomerular leak can combine to increase albuminuria. Our model provides a validated quantitative description of the transport processes underlying the protein composition of human urine in normal and pathophysiological states.