Stability of a preformed 125I-transferrin corona after exchange with excess of albumin. A, 125I-transferrin was first adsorbed to different PEGylated SPIOs. After removal of free unbound protein by filtration, a 3 fold excess of bovine albumin was added. FPLC analysis followed a 2 h incubation at room temperature. B, same experiment as in A but 125I-transferrin was covalently bound (EDC-coupling). A clear leakage of protein from the corona was observed when 125I-transferrin was initially adsorbed (upper lane) but not when covalently bound (bottom lane). 125I-activity in arbitrary units with a clearly less remaining activity with adsorbed transferrin than with covalently bound (see also ). Filled lines, UV 280 nm; dashed lines, 125I-activity in fractions. The peak at 33 min represents free albumin or transferrin. FPLC analyses showed that the primarily adsorbed transferrin is substantially exchanged by albumin, whereas the covalently bound co-elutes only with the SPIOs and not with the peak of free protein. Finally, an excess of whole plasma was added to the reaction mixture under the same experimental conditions ().