Analysis of Glomerular Permselectivity in the Rat Using

Changes in glomerular permselectivity are the physiological basis for proteinuria, which typically accompanies chronic kidney disease. Traditionally changes in the sizerestrictive properties of the glomerular barrier have been quantified by measuring the fractional clearance ( ) of exogenous infused dextrans and interpreting the results using the hindered transport theory for uncharged, solid spheres in cylindrical pores. The observed values for for dextran in normal rats and healthy humans are unexpectedly large, however, given the normal absence of proteinuria. Recent in vitro diffusion studies show that dextran’s hindered transport behavior is better represented by a random coil with attractive pore/solute iteractions, while Ficoll follows the predictions for a neutral solid sphere. Thus Ficoll sieving data should provide a better indication of the sizeselective properties of the glomerular capillary wall. In one of the studies for this thesis, healthy Munich-Wistar rats were infused with either H-dextran or H-Ficoll. Plasma and urine samples were fractionated on gel chromatography columns which had been calibrated with narrowly-sized dextran and Ficoll standards. While values of for Ficoll were similar to those measured for nearly-neutral proteins, dextran values were significantly higher at all molecular sizes. On this basis, the glomerular capillary wall is shown to be considerably more size-restrictive than had been previously determined. The random coil model was applied to the dextran data, and the attractive energy required to explain the dextran/Ficoll discrepancies in vivo was nearly the same as that required in vitro. Thus is appears that dextran’s enhanced transport is nearly independent of the medium through which it travels. The implications of the more size-restrictive barrier were examined in a second experimental protocol, in which H-Ficoll was infused into four groups of fawn-hooded (FH) rats: a two-kidney (2K) control group, an uninephrectomized (UNX), a UNX group treated with the angiotensin converting enzyme inhibitor enalapril (ENA), and a UNX group treated with the nitrous oxide inhibitor NAME. The UNX and NAME groups had significantly higher glomerular filtration pressures ( P ) and more proteinuria than the 2K and ENA groups, but the membrane pore size parameters of the four groups were essentially the same. The extent of albumininuria correlated strongly with P , while the total rate of excretion of non-albumin proteins did not. Because albumin is negatively charged, these findings imply that albuminuria in FH rats results from a specific defect in glomerular charge-, rather than size-selectivity induced by chronic glomerular hypertension. Finally to examine the potential for using fractional clearances to estimate filtration pressures in the clinical setting, the mathematical model was modified to make P an adjustable parameter, and fitted values were compared with those actually measured in the FH rats. The fitted and measured P did not show significant correlation, suggesting that may not be sufficiently sensitive to P to allow such estimations. Thesis Supervisor: William M. Deen Title: Professor of Chemical Engineering