Assessment of the charge selectivity of glomerular basement membrane using Ficoll sulfate.

The extent to which the glomerular basement membrane (GBM) contributes to the charge selectivity of the glomerular capillary wall has been controversial. To reexamine this issue, the size and charge selectivity of filters made from isolated rat GBM were assessed, using polydisperse Ficoll and Ficoll sulfate as test macromolecules. Ficoll sulfate, a novel tracer with spherical shape synthesized for this purpose, exhibited little or no binding to serum albumin, thereby avoiding a major difficulty that has been reported with dextran sulfate. The sieving coefficients of Ficoll sulfate were not different from those of Ficoll at physiological ionic strength, although the values for Ficoll sulfate were depressed at low ionic strength. These results confirm that the GBM possesses fixed negative charges but suggest that its charge density is insufficient to confer significant charge selectivity under physiological conditions, where electrostatic interactions are relatively well screened. The sieving coefficients of Ficoll sulfate and Ficoll were elevated significantly and by similar amounts when bovine serum albumin (BSA) was present in the retentate at 4 g/dl. This could be explained as the combined effect of two nonspecific physical factors, namely, the reduction in filtration velocity due to the osmotic pressure of BSA and the effect on macromolecular partitioning of repulsive solute-solute interactions. The view that BSA does not affect the intrinsic properties of the GBM is supported also by the absence of an effect on the hydraulic permeability of isolated GBM. The sieving coefficient of BSA was roughly half that of Ficoll or Ficoll sulfate of similar Stokes-Einstein radius. Given the finding of negligible charge selectivity, this difference may be attributed to the nonspherical shape of albumin. The results suggest that, to the extent that isolated GBM is similar to GBM in vivo, the charge selectivity of the glomerular capillary wall must be due to the endothelial and/or epithelial cell layers.

Random-coil model for glomerular sieving of dextran.

Dextran has been the most commonly employed test molecule for probing the selectivity of glomerular filtration to macromolecules of varying size. The usual theories for hindered transport of solid spheres through pores have limited utility in interpreting clearance data for dextran or other linear polymers because such polymers in solution more closely resemble random, solvent-filled coils than solid spheres. To provide a model for glomerular filtration of random-coil macromolecules, the equilibrium partitioning of random coils between cylindrical pores and bulk solution was simulated using Monte Carlo calculations, and those results were combined with a hydrodynamic theory for restricted motion of solvent-filled polymer coils in pores. The rates of transport predicted for either neutral random coils or for solid spores of the same Stokes-Einstein radius were significantly lower than observed transport rates of dextran through the glomerular capillary wall or across synthetic porous membranes. This facilitation of dextran transport was modeled by postulating weak, attractive interactions between dextran monomers and the pore wall. The random-coil model with attractive interactions, modeled using a short-range, square-well potential, was found to adequately represent dextran sieving data in normal rats. Various limitations of this approach are discussed.

Glomerular permeability barrier in the rat. Functional assessment by in vitro methods.

The formation of glomerular ultrafiltrate is dependent on the prevailing hemodynamic forces within the glomerular microcirculation and the intrinsic properties of the filtration barrier. However, direct assessment of the permeability barrier is difficult with most available techniques. We used confocal microscopy to image 1-micron thick optical cross-sections of isolated intact glomeruli and glomeruli denuded of cells and quantitated dextran (70,000 mol wt) diffusion from the capillary lumen. Dextran permeance was 11 times greater for the acellular filtration barrier than the intact peripheral capillary. Consideration of the basement membrane and cells as series resistors demonstrated that cells of the filtration barrier contribute 90% of the total resistance to macromolecular permeance. Using a different approach, dextran sieving coefficients for acellular glomeruli consolidated as a multilayer sheet in a filtration cell were similar to those for intact glomeruli in vivo at radii 30-36 A and approximately 50 times greater at a dextran radius of 60 A. The presence of cells significantly reduced hydraulic permeability determined on consolidated intact or acellular glomeruli in an ultrafiltration cell with 50 mmHg applied pressure. The glomerular basement membrane does restrict macromolecular permeability but cells are important determinants of the overall macromolecular and hydraulic permeability of the glomerulus.