Practical limitations of estimation of protein adsorption to polymer surfaces

The estimation of protein adsorption to polymeric surfaces is a prerequisite for analysing the biological activity of a coated protein both in the evaluation of polymeric biomaterials and in the standardisation of ELISA assays. Direct quantitative methods utilise either measurements of the physical characteristics of the coated surface (equipment necessary for this is not always available), or measurements of radiolabelled protein. We demonstrate that proteins labelled with 125I using chloramine-T contain a significant proportion of 125I label which is unable to adsorb to polyvinyl microtitre wells. This non-bindable label is not separable from protein by fractionation on Sephadex G-25 but does dissociate upon SDS gel electrophoresis. The proportion of non-adsorbable label increases with storage time at -70 degrees C, the effect being accelerated at higher temperatures. A method is described for estimating protein adsorption which is both independent of the non adsorbable fraction but can also be used to estimate it. The use of biotin as a protein label does not apparently result in deterioration of adsorption with short storage times. The problem of subsequent desorption in ELISA buffers, tissue culture media, or body fluids is discussed.