To the Editor:

Two recent studies1,2 using lipopolysaccharide (LPS)-induced proteinuria in mice as a model of human proteinuric kidney disease have suggested that the proteinuria is caused by dysfunction in the kidney barrier, specifically relating to podocyte injury. However, there are serious concerns about the model used and the conclusions made in these studies. First, the fold changes in protein excretion are small in LPS-mediated proteinuria—threefold in proteinuria1 and five- to sevenfold in albuminuria2. This model does not compare to the 100–10,000-fold changes in albumin excretion seen in human proteinuric disease, including focal segment glomerulosclerosis3, diabetic nephropathy4 and minimal change disease5 (reviewed in ref. 6). Even in nephrotic disease induced by puromycin aminonucleoside in rats, there is a more than 50-fold increase in albumin excretion7. Second, it is notable that such pronounced changes in podocyte components, which the authors attribute to governing the glomerular barrier1,2, result in such a mild form of proteinuria.

Further, claims that kidney barrier function is altered in LPS-mediated proteinuria1,2 were not supported experimentally, as no studies directly measuring glomerular permeability function were performed. There is a need to examine the effects of LPS on glomerular permeability to albumin directly, as changes in albumin excretion in human proteinuric disease cannot be accounted for by changes in the glomerular permeability of inert probes such as dextran and Ficoll, particularly for probes the same size as albumin3,4,5,6. Similar findings have been also been made in rat proteinuric disease.6