THE PRODUCTION OF ANTIBODIES IN VITRO

By Linus Pauling

1942

By following the general procedure proposed in connection with a theory of serological phenomena, we have succeeded in making antibodies in the laboratory.

The procedure consists in subjecting normal globulin or other protein to the denaturing reagents or conditions in the presence of an antigen. the protein molecule unfolds, and then refolds in such a way as to assume a configuration complementary to that of the antigen, thus acquiring the properties of a specific homologous antibody.

Bovine γ-globulin was the protein used in most of our experiments. Some success was obtained also with other serum globulin fractions and with serum albumin. As antigens there have been used the triphenylmethane dye methyl blue (a mixture of the p-trisulfonated and disulfonated triphenylpararosanilines), the azo dye 1,3-dihydroxy-2,4,6-tri(p-azophenyl-arsonic acid) benzene and pneumococcus polysaccharide type III. Successful experiments were made by addition of alkali (to pH 11) and slow return to neutrality, by addition and slow removal of urea, and by heating to about 65° and slowly cooling; some success was also obtained by surface denaturation. The most satisfactory of the treatments tried was that of holding a solution of protein and antigen for several days at about 57°; this temperature seems to be high enough to cause the protein chains to unfold and to refold under the influence of the antigen into specific complementary configurations.

In one experiment a solution containing 0.01% of the azo dye mentioned above and 1% of bovine γ-globulin was held at 57° for about 14 days, then removed from the bath and dialyzed through Cellophane against 1% salt solution. Some precipitate formed during the heating and more during the dialysis. The mixture was then dialyzed against a 1% solution of the haptene arsanilic acid, which was changed several times. Most of the precipitate dissolved during the dialysis; that which remained was discarded. The solution, which was free of dye, want then dialyzed against 1% salt solution to remove the haptene.

The resulting protein solution was found to have many properties of an anti-serum specific to the phenylarsonic acid group. It gave precipitates with multihaptenic and azoproteins containing this group, and not with other dyes or azoproteins, and the specific precipitates were dissolved by excess of dyes or haptenes containing this group not by other dyes or haptenes, such as those containing the phenylsulfonic acid group.

A solution of 1% pneumococcus polysaccharide type III and 1% bovine γ-globulin was similarly held at 57° for 14 days. Some precipitate formed, which was removed. The resultant solution was found to precipitate type III polysaccharide but not types I or VIII (cross-reaction with type VIII was thus not shown) and to agglutinate pneumococci type III but not types I or II. Purified antibody solutions were made by adding 15% salt and bringing to pH 4 or adding 15% salt, calcium chloride, and calcium hydroxide to pH 8, thus precipitating polysaccharide and some protein. Each of the two solutions, after dialysis against 1% salt solution, was found to precipitate type III polysaccharide but not types I or VIII and to agglutinate type III pneumococci but not types of I or II. Mouse protection tests and swelling tests have not been carried out.

We acknowledge with thanks and support of a grant from the Rockefeller Foundation, the assistance of Dr. David Pressman, and the courtesy of Dr. W. Goebel in providing material.

Our experiments are being continued.