We have analyzed glomerular sieving data from humans, rats in vivo, and from isolated perfused rat kidneys (IPK) and present a unifying hypothesis that seems to resolve most of the conflicting results that exist in the literature. Particularly important are the data obtained in the cooled IPK, because they allow a variety of experimental conditions for careful analysis of the glomerular barrier; conditions that never can be obtained in vivo. The data strongly support the classic concept of a negative charge barrier, but separate components seem to be responsible for charge and size selectivity. The new model is composed of a dynamic gel and a more static membrane layer. First, the charged gel structure close to the blood compartment has a charge density of 35-45 meq/l, reducing the concentration of albumin to 5-10% of that in plasma, due to ion-ion interactions. Second, the size-selective structure has numerous functional small pores (radius 45-50 A) and far less frequent large pores (radius 75-115 A), the latter accounting for 1% of the total hydraulic conductance. Both structures are required for the maintenance of an intact glomerular barrier.

The fractional clearances (theta) for FITC-Ficoll and albumin were estimated in isolated perfused rat kidneys in which the tubular activity was inhibited by low temperature (8 degrees C) and/or 10 mM NH(4)Cl. The Ficoll data were analyzed according to a two-pore model giving small and large pore radii of 46 A and 80-87 A, respectively. The estimated negative charge density was 35-45 meq/l at 8 degrees C. Perfusion with erythrocyte-free solutions of kidneys at 37 degrees C reduced glomerular size and charge permselectivity. Thus the large pore fraction of the glomerular filtrate (f(L)) was 1.64% at 37 degrees C compared with 0.94% at 8 degrees C. The theta for albumin was four times higher at 37 degrees C than at 8 degrees C (0.86% vs. 0.19%, respectively). NH(4)Cl caused further irreversible damage to the glomerular barrier. We conclude that there are no deleterious effects on the glomerular barrier of a reduction in temperature from 37 degrees C to 8 degrees C. Therefore our data seem to disprove the hypothesis of low glomerular permselectivity and transtubular uptake of intact albumin and support the classic concept of a highly selective glomerular barrier.

The effect of shape on the transglomerular passage of solutes has not been hitherto systematically studied. We perfused isolated rat kidneys to determine the fractional clearances (theta) at various filtration rates for four molecules of different shapes but with similar Stokes-Einstein radii (aSE = 34-36 A). The theta for hyaluronan, bikunin, and Ficoll36 A were 66, 16, and 11%, respectively, at a glomerular filtration rate (GFR) of 0.07 ml x min(-1) x g wet wt(-1) and decreased to 46, 14, and 7%, respectively, on a fivefold increase in GFR. Under the same conditions, theta for albumin increased from 0.15 to 0.74%, and similar behavior was observed for larger Ficolls (aSE >45 A). Pore analysis showed that the "apparent neutral" solute radii of Ficoll, albumin, bikunin, and hyaluronan were 35, 64, 33, and 24 A, respectively, despite similar aSE. In addition, the properties of the glomerular filter changed with increasing GFR and hydrostatic pressure. We conclude that elongated shape, irrespective of size and charge, drastically increases the transglomerular passage of a solute, an effect that is related to its frictional ratio.

The serum protein orosomucoid (α1-acid glycoprotein) is needed to maintain the high capillary permselectivity required for normal homeostasis. It is not known how the protein executes its action, but it seems to contribute to the charge barrier. Moreover, recent studies suggest that the endothelial glycocalyx is essential for the charge barrier. The main site of orosomucoid synthesis is the liver, but we wanted to explore the possibility that orosomucoid was synthesized in endothelial cells. Primary cultures of human microvascular endothelial cells (HMVEC) from dermal tissue were established. Human liver cells were used as positive controls, and total RNA was prepared from both cell types. Reverse transcription-polymerase chain reaction (RT-PCR) was performed and demonstrated orosomucoid expression. After RT-PCR, the identities of the PCR products were confirmed by sequencing. RNase protection assay performed on total RNA from the HMVEC confirmed the results from the RT-PCR, i.e., orosomucoid mRNA is expressed by endothelial cells. Synthesis of orosomucoid in both liver and endothelial cells was demonstrated by immunoprecipitation. In conclusion, endothelial cells normally produce orosomucoid, which is essential for capillary charge selectivity. We suggest that orosomucoid exerts its effect by interacting with other components of the endothelial glycocalyx.

The classical concept of a negative glomerular charge barrier has recently been questioned, mainly based on the somewhat high clearance for anionic horseradish peroxidase (HRP). The validity of using anionic HRP can be tested by changing the properties of the charge barrier. A rather unequivocal approach is to reduce the ionic composition of the perfusate and hence increase the Debye length. We determined the glomerular clearance for horseradish peroxidase and serum albumin, using isolated rat kidneys perfused at 8 degrees C to reduce the tubular modification of the primary urine. The perfusate contained trace amounts of the neutral 125I-nHRP and the anionic 131I-aHRP and were otherwise identical except for different ionic strengths, 152 mM and 34 mM, respectively. During control, the fractional clearance (theta) was 0.11 +/- 0.015 for nHRP and 0.045 +/- 0.010 for aHRP, with an average clearance ratio (n/a) of 2.8 +/- 0.24. Low ionic strength reduced theta for aHRP to 0.027 +/- 0.006, giving an increased clearance ratio for HRP of 4.2 +/- 0.44. The existence of a negative charge barrier is supported by the experiments. The result obtained during normal perfusion is compatible with a charge density (omega) of 34 mEq L-1, using a model of homogeneously charged membrane. Low ionic strength perfusion reversibly reduced the concentration of fixed charges to 12 mEq L-1, suggesting an almost threefold increase of the glomerular membrane volume. Thus, the glomerular charge barrier should be regarded to have a dynamic gel structure rather than being a rigid membrane.