Forrest E. Kendall

Reaction Between Proteins and Diazotized Aromatic Amines in Neutral Solution.

In a series of 3 challenging papers entitled “Chemospecific Antigens” Klopstock and Selter 1 have raised a number of important questions dealing with the chemical basis of certain immune reactions. While several of their minor points will be dealt with in connection with other work now in progress, the present note concerns the underlying thought and principal conclusion of their work, namely, that no chemical interaction takes place in neutral solutions of diazotized aromatic amines and proteins or lipoids. It is held that such solutions are merely simple mixtures, and that “eine chemisch zu verstehende Substitution des Chemikals im Eiweissmolekül bei der chemospezifischen Komplexantigenbildung keine Rolle spielt.” Klopstock and Selter have failed entirely to report chemical control experiments in testing this conclusion, and the following are therefore submitted: I. Conditions the same as in Paper II, p. 455, showing appearance of chemospecific antigen only after mixtures had stood in the cold 1 to 24 hours. 1% diazotized arsanilic acid, “neutralized to litmus.” 2 vols, rabbit serum (K. and S. used beef serum in this experiment, rabbit serum in others). Since color formation could conceivably have been due to some other serum component than protein, a solution of 3-times recrystallized, dialyzed egg albumen (containing 0.018 g. per cc.) was next used. II. No difference to the eye immediately after mixing. Since the neutral diazo arsanilic acid solution darkened rapidly, as noted by Klopstock and Selter, making readings after several hours unsatisfactory, diazotized sulfanilic acid was substituted. This darkens comparatively little on standing in the cold, and, in addition, antisera to sulfanilic-azo-proteins were available. Moreover, since “neutrality to litmus” is a vague term an experiment was run at pH 6.8, but since it might be objected that this is toward the alkaline side of “neutral to litmus” and that actual coupling took place for this reason,† there are given below only the results at pH 6.3-6.4, at which blue litmus is faintly reddened. III. Dialyzed egg albumen solution, 0.022 g. per cc, pH 6.3. 0.75% solution of diazotized sulfanilic acid, pH 6.4.

Protein-lipid relationships in experimental canine atherosclerosis.

The applicability of Cohns method number 10 to the study of dog plasma has made possible this investigation of experimental canine atherosclerosis. Administration of thiouracil which causes hypercholesterolemia but no atherosclerosis in dogs produces no considerable alteration in the distribution of plasma lipoproteins. High cholesterol feeding and administration of thiouracil causes both atherosclerosis and significant alterations in the distribution of lipoproteins.

Preparation of the Specific Polysaccharide of Type I Pneumococcus

It was shown 1 that preparation of the specific polysaccharides of Types II and III pneumococcus in the cold and without the use of strong acid or alkali gave products forming much more viscous solutions than did the polysaccharides isolated by the older methods, 2 but apparently differing otherwise from the older preparations only in their ability to precipitate more antibody from homologous rabbit antisera. These differences were attributed to the greater chain length of the molecules of the new preparations. The acetyl polysaccharide from Type I pneumococcus has now been similarly prepared. This product had the chemical and physical properties of the acetyl polysaccharide reported by Avery and Goebel 3 but gave solutions of much higher viscosity than did samples prepared according to Reference 3, and precipitated twice as much antibody nitrogen from a Type I antipneumococcus rabbit serum. Thus a solution of the new preparation, S 120, containing 1 mg. per ml. in 0.9% saline had a relative viscosity, ηr, of 1.69. After 8 hours in a sealed tube at 100° this solution showed ηr = 1.05. After treating with N/2 NaOH for 40 hours at 37° C. ηr was 1.20. A preparation, S91a, isolated from culture filtrate concentrated on the steam bath gave ηr = 1.10. Table I shows the amount of antibody nitrogen precipitated from rabbit and horse Type I antipneumococcus sera by an excess of the different preparations of Type I pneumococcus specific polysaccharide (SI). As was found with S II and S III the use of heat as in the initial concentration of S I lowers its power to precipitate rabbit antisera without greatly affecting its reaction with horse antisera.

Preparative Changes Necessitated by a Quantitative Study of Precipitating Power of Pneumococcus Polysaccharides.

In an extension of the quantitative study of the precipitin reaction between the specific polysaccharide of Pneumococcus III (S III) and homologous antibody 1 it was found that several S III preparations which precipitated the same amount of antibody from horse serum threw down widely varying quantities of antibody from a rabbit serum. The present preliminary report deals with experiments on S II and S III undertaken to find the cause of variations in the precipitating power of these substances toward homologous rabbit antisera. 1. Specific Polysaccharide of Type III pneumococcus (S III). A Chamberland filtrate of 24-hour cultures in phosphate meat infusion glucose broth was concentrated about 12-fold in vacuo and precipitated several times with an equal volume of alcohol. The caseous S III was redissolved and separated from protein over the barium salt with barium chloride. Barium was removed by repeated reprecipitation of the S III with glacial acetic acid after addition of sodium acetate. The properties of several lots obtained in this way are given in the table. Preparations in which heat and strong acid were avoided showed far greater viscosity than did samples prepared by the original method. 2 Comparison of viscosities was made in 0.9% salt solution. 3 As will be seen from the table a new preparation precipitates approximately the same amount of antibody from serum as does the original culture filtrate. The new preparations precipitate more antibody from Type III antipneumococcus rabbit serum than those prepared by the original method, although all precipitate about the same amount of antibody from homologous horse antiserum. The sera used were first absorbed with “C”-polysaccharide and pneumococcus protein so that the remaining antibody would be mainly anti-S.†