“…However, prolonged digestion with neuraminidase, an enzyme which specifically removes sialoglycoproteins [27,40,43], did not affect label ling with PHM 5, although such treatment is reported to have removed all three otherwise similar antigens recently purified from the glomerular epithelial sialoprotein coat of rats [34]. Similarly, the use of nitrous acid oxidation to re move heparin and heparan sulphate had no effect on label ling, despite the presence of large amounts of heparan sulphate throughout the GBM [10,25] and possibly on glomerular epithelial cells [44], The two other related GAG, chondroitin sulphate and hyaluronic acid, also appear un likely to be involved; absorption with chondroitin sulphate B (dermatan sulphate) had no effect, nor did PHM 5 bind to interstitial cells of kidney papillae or the luminal sur face of Henle's loop and distal tubules which are all sites suggested to contain chondroitin sulphate and hyaluronic acid [27,28,33]. In addition, fibronectin, a glycoprotein present in mesangial areas and within the GBM [11,37], did not absorb PHM 5, nor does PHM 5 appear to be directed against the amyloid P component, since absorp tion with GBM and serum had no effect [12], Finally, the failure of absorption by human milk, and its different renal distribution, namely glomerular but not tubular epithelial cells, indicates PHM 5 is not directed against the recently described epithelial membrane antigen [21].…”