Non-specific binding of protein-stabilized gold sols as a source of error in immunocytochemistry.

The observation that protein-A conjugated gold sols bound to fibronectin-collagen (FNC) fibres in human fibroblast cultures prompted a series of studies on the binding of gold particles stabilized in various ways (Staphylococcal protein A, bovine serum albumin, avidin, streptavidin, gelatin, hemoglobin, polyethylene glycol (MW 20 000), methylcellulose and the nonionic detergent Tween 20) to cell and tissue components, to protein dot blots and SDS-PAGE blots on nitrocellulose paper. We found that binding of gold particles to certain cell and tissue components and to various immobilized proteins did occur irrespective of the stabilizing agent. We argue that, albeit gold sols are stabilized against salt coagulation by adsorption of proteins and other stabilizing agents, "naked areas" are (constantly or intermittently) present on particle surfaces, available for interaction with cell and tissue components that have a high electrostatic affinity for the charged gold surface under prevailing experimental conditions. Non-specific binding may be reduced or abolished by competing proteins (i.e. proteins with a higher affinity for gold than any component in the object studied) provided the proteins and the gold conjugate are present concomitantly during incubation. We found gelatin (Bloom number 60-100) to be an effective competitive protein probably due to its high affinity for gold over a wide pH range. Further, gelatin did not appreciably inhibit the specific interaction in dot blots between SpA and IgG except at very low IgG concentrations. A protocol for the use of gold-protein conjugates to circumvent the hazards of unspecific gold binding is suggested.