Competition between Albumin and Beta-2-Microglobulin for Renal Tubular Uptake: Brush Border and/or Lysosomes?

J.P. Revillard, Hôpital E.-Herriot-INSERM U.80, F-69437 Lyon Cédex 3 (France) Dear Sir, We were most interested in the recent paper by Bernard et al. [1] showing that lowand highmolecular-weight protein can compete for renal uptake. These authors challenged the current opinion that tubular protein reabsorption is achieved through distinct pathways for small and large molecules. Their data showed that injection of 1 g/kg of bovine serum ablumin in rats, which resulted in an estimated concentration of 6–13 μg/ml in glomerular ultrafiltrate (assuming a sieving coefficient of 10 3 to 5 × 10 4) induced an increased excretion of human ßimicroglobulin (ßim) (336-fold), rat ßam (10-fold) and retinol-binding protein (4-fold). Human serum albumin, at comparable concentrations in the ultrafiltrate, merely induced a 4-fold increase in human ß∑m excretion. Human ß∑m injected at 16 mg/kg, a dose which increases ß2irι levels in glomerular filtrate by a factor of 10–15, resulted in a 4-fold increase of albumin excretion. Our own studies [2] using injections of 0.2–150 mg/kg of human ß2m in rats demonstrated a possible increase in albumin excretion but mainly the occurrence of several low-molecularweight proteins in the urine. Intravenous infusion of arginine in humans was reported to induce a rapid increase of urinary excretion of albumin and a much greater increase of ß∑m [3]. Considering the relatively high concentration of low-molecular-weight proteins in primitive urine, as compared to that of albumin, the data from these studies do not exclude the hypothesis of a common reabsorption pathway. However, as pointed out by Bernard et al. [1], the postulated higher affinity of bovine versus human serum albumins for a tubular binding site is surprising in view of the great similarities in the amino acid composition, net charge and size of the two molecules. Furthermore, in selective glomerular proteinuria there is usually no substantial increase of low-molecular-weight proteins despite a major elevation of the albumin filtered load [4]. Tubular reabsorption of proteins is a complex process in which several subsequent interrelated steps can be distinguished: binding to a putative brush border acceptor site, endocytosis, cellular transport of endocytic vesicles, fusion with lysosomes and proteolysis [5, 6]. Kinetic studies of ßmi reabsorption in the rat demonstrated tubular uptake within 5 min, and increase in tubular reabsorption between 10 and 20 min after ßmi injection [2]. Degradation reached a maximum between 30 and 60 min after injection [2, 7]. After infusion of 150 mg/kg of ßmi, urinary excretion of some rat low molecular-weight-proteins and albumin did not reach its maximum