The interaction of proteins with silicic acid.

Although silicon is a constituent of normal animal tissues, silicic acid formed by the dissolution of inhaled quartz particles can induce the formation of pathological lesions in which the normal tissue is replaced by fibrotic tissue arranged concentrically around the silica particles. The mechanism of the change which constitutes the disease, silicosis, is unknown. Schulman & Rideal (1937) demonstratedthe interaction ofaproteinwith silicic acidby injecting the acid beneath a monolayer of gliadin at pH 6.2. The silicic acid was adsorbed to the protein layer and after 12 hr. produced a film which could not be penetrated or dispersed by sodium oleate. This communication describes a more extensive study designed to investigate the conditions under which proteins react with silicic acid and to determine which of their functional groups are responsible for the interaction. Two methods of study were used. Protein monolayers were spread on silicic acid solutions and the characteristics of the films were compared with monolayers on subsolutions which contained no silicic acid. Silicic acid and protein solutions were also mixed in bulk and their interaction was followed turbidimetrically. The two proteins examined, bovine serum albumin and methaemoglobin, were chosen because of the considerable difference between their isoelectric points. The properties of gliadin films on silicic acid substrates indicate that the interaction involves two processes: (a) the adsorption of silicic acid to the protein, and (b) the polymerization of the silicic acid. When silicic acid, in the form of the monomer or low polymers, is adsorbed beneath a protein film, polymerization between molecules adsorbed on adjacent protein sites follows and results in the formation of a complete network underneath the protein film, which is thereby solidified. If, however, the protein is spread on a solution containing silicic acid already in the form of high polymers and adsorption occurs, the protein is anchored by the silicic acid and is prevented from spreading completely to a monolayer. If interaction takes place, the area to which the protein will spread may be considerably decreased. The rate of polymerization of silicic acid depends markedly onpH and, except in the region pH 5 5-6, it is slow. To simplify the interpretation of our measurements and to allow a study of proteinsilicic acid interaction without the complicating polymerization reaction, the silicic acid sols were pre-polymerized at pH 6 before use as subsolutions. A parallel study of the interaction of monolayers of a carboxylic acid, an alcohol and an amine allows some deductions to be made as to which groups of the protein are likely to interact with silicic acid. The influence of pH on the formation of silicic acid-protein complexes is shown most clearly by turbidimetric measurements on bulk solutions.