ESMOND R. LONG 1023 



solutions of tuberculin protein with ammonium sulphate and adding acetic acid until 

 a reaction of pH 4.8 was reached, she was able to induce a portion of the protein pres- 

 ent to crystallize in needle form. Usually the needles crystallized out in company 

 with amorphous protein. In a few instances they were pure. The crystals obtained 

 took a blue stain with methylene blue, gave the biuret and Millort tests, and were heat 

 coagulable. One of the purest preparations obtained caused a marked skin reaction in 

 tuberculous guinea pigs. The total weight of product, five times recrystallized, from 

 20 gm. of water-soluble protein was only 70 mg., but this beginning appears the most 

 hopeful of all the attempts made at purification of the active principle. 



An extensive investigation has been carried out by Johnson and his colleagues 

 on the proteins in the body of the tubercle bacillus. Coghill,' following the analysis 

 outline of Johnson,' prepared an albumin-like protein from tubercle bacilli grown on 

 a synthetic medium, which proved to have marked tuberculin potency. Distinctive 

 chemical features were a high content in basic amino acids and failure to reduce 

 Benedict's solution even after hydrolysis. Inasmuch as no protein precipitation oc- 

 curred on dialysis of a 5 per cent sodium chloride extract of tubercle bacilli, it was con- 

 cluded that globulins are practically absent from the structure of the bacillus. 



CoghilP found that alkaline extraction of the bacillary mass after removal of the 

 water-soluble protein yielded a protein containing 14.2 per cent nitrogen and consti- 

 tuting 20 per cent of the dry weight of the bacilli. This proved practically devoid of 

 tuberculin activity. 



A preliminary study carried out jointly by Seibert and Coghill^ brought out the 

 fact that a larger amount of potent water-soluble tuberculin protein is obtainable 

 from the filtered culture medium on which the tubercle bacilli have grown for six 

 weeks than from the bodies of the corresponding bacilli. With shorter periods of 

 growth larger amounts remained in the bacilli, while with incubation of three months 

 most of the activity appeared to have passed into the medium. The proteins obtained 

 from the two sources were qualitatively similar. 



These studies re-emphasize a fact appreciated since Koch's time, viz., a funda- 

 mental difference between the active principle of tuberculin and the active principles 

 of the true exotoxins. The latter, as exemplified typically in diphtheria toxin, appear 

 to represent true secretions or excretions from the bacteria elaborating them. Ex- 

 tracts of diphtheria bacilli do not contain appreciable amounts of diphtheria toxin. 

 Extracts of recently grown tubercle bacilli, on the other hand, are rich in the active 

 principle of tuberculin, and the active substance found in the culture medium appears 

 to arise by extraction, possibly associated with autolysis, of the growing bacilli. 



THE NATURE OF THE TUBERCULIN REACTION 

 ANATOMICAL CHANGES AND EXPLANATIONS OF THEIR CAUSE 



When dilute solutions of tuberculin are injected into the skin of a normal animal, 

 no grossly visible change occurs. Microscopically, a transient influx of a few leuko- 

 ' Coghill, R. D.: /. Biol. Chem., 70, 439. 1926. - 

 ^Johnson, T. B.: Am. Rev. Tuberc, 14, 164. 1926. 

 3 Coghill, R. D.: /. Biol. Chem., 70, 449. 1926. 

 -i Seibert, F. B., and Coghill, R. D.: unpublished. 



