The Tuberculostatic Effect of Subtilin in Vitro and in Vivo.

The antituberculosis activity of subtilin, an antibiotic agent introduced by Jansen and Hirschmann (1944), has been the subject of conflicting reports in the literature. Salle and Jann (1945) reported that subtilin in a dilution of 1:50,000 prevented surface growth of Mycobacterium tuberculosis on Long's synthetic medium, and that transfers into fresh medium made from tubes containing 1:10,000 subtilin produced no growth. They concluded that "subtilin is bacteriostatic in high dilution and germicidal in greater concentration." Anderson and Wong (1946) and Wong, Hambly, and Anderson (1947) found that in the sorbitan monooleate albumin medium of Dubos and Middlebrook (1947), a 1:400,000 dilution of subtilin (2.5 jug per ml) regularly inhibited the growth of Mycobacterium tuberculosis for 12 days. To study the bactericidal effect, they inoculated 1 ml of each dilution not showing visible growth into each of two guinea pigs. The animals receiving material from tubes containing 1: 20,000 and 1: 10,000 dilutions of subtilin did not develop tuberculosis. These workers also found that subtilin inhibited tubercle formation in the allantoic membrane of the chick embryo, but they could demonstrate no effect on tuberculous infection of rabbit corneas by topical application of the drug, or on tuberculous infection in hamsters by subcutaneous injections of subtilin for 6 weeks. Knight and Tompsett (1948) observed that subtilin was effective in inhibiting the growth of tubercle bacilli only when the organisms were growing in a dispersed fashion, as in media containing wetting agents such as sorbitan monooleate. In media not containing wetting agents, where the bacilli were growing in aggregates, subtilin did not inhibit their growth. They reported that the course of tuberculous infection in mice was not affected by much larger daily doses of subtilin than those necessary for protection against pneumococcus or streptococcus infection. Farber et al. (1948) reported the sensitivity to subtilin of 6 strains of tubercle bacilli recently isolated from patients. Three were sensitive to 1:400,000, two were sensitive to 1:200,000, and one to 1:100,000 dilution of the drug. Presumably these tests were done in sorbitan monooleate albumin medium. This report describes experiments designed to clarify the status of subtilin as a tuberculostatic agent in the test tube and in guinea pigs. In vitro results. The effect of subtilin' on the growth of the virulent, human type tubercle bacillus, H37Rv, was tested in four different liquid media: Pros-