Chemical detoxification of Flexner dysentery antigen; a mouse test to measure toxicity relative to antigenicity.

The problem of developing a detoxified Flexner dysentery antigen for use in the armed forces during the recent war (Barnes, Dewey, Henry, and Lupfer, 1947; Barnes and Dewey, 1947) required a test by which the results of a given chemical or physical manipulation of the antigen could be evaluated with respect to both toxicity and protection. It was necessary to do this with such precision and definition of standards that progress in detoxification could be followed quantitatively. This meant that altered antigens had to be compared with crude vaccines in terms of a smaller or a larger number of toxic units that were capable of yielding an identical degree of protection against infection. To accomplish these ends, it was necessary to adapt several methods to one test and to make this applicable to the dysentery organism. No one test or combination of tests fulfilling these needs was found available or easily adaptable. In the first place, a representative measure of toxicity of successive dilutions of antigen had to be worked out for the dysentery organism. Although weight loss has been used for many tests of physiological effect, it has not been shown whether it is a generally reliable index of toxicity and it had not been clearly applied to Flexner dysentery toxicity in the mouse as a well-defined test. This convenient property of weight loss was chosen because it was found to be proportional to the logarithm of the dose of antigen given (see below), and in all tests with either unaltered or detoxified material, no discrepancy between weight loss and observed symptoms in mice could be noted. Secondly, it was important to develop corresponding precision in a protection test rather than by immune body titration because high protectivity against infection was the chief and practical aim; a nuimber of steps had to be taken to achieve adequate precision in this part of the test. For purposes of exploration, the mouse was employed rather than humans, although final results have been checked in the latter (Barnes, Dewey, Henry, and Lupfer, 1947). The protective power of an antigen is titrated a week after its administration by injecting each of the same mice with a carefully standardized dose of live virulent organisms. Thirdly, it was felt necessary to use the same mice for assaying both toxicity and protection of a given sample; by choice of the tests for weight loss and protection and by adjustment of dosages it was found possible to accomplish this,