The bacterial DNA content of mouse organs in the Cornell model of dormant tuberculosis

Setting: The Cornell model of murine tuberculosis has proved useful for demonstrating and studying dormancy. However, it has not previously been used to investigate the molecular aspects of dormancy. Objective: To obtain a profile of the amount of Mycobacterium tuberculosis DNA at various stages of the Cornell model. Design: BALB/C mice were infected intravenously with 2.7 x 10(6) cfu M. tuberculosis strain H37Rv; they were left for two weeks, treated for 14 weeks,vith isoniazid and pyrazinamide and left untreated for a further 14 weeks. Spleens and lungs at start of treatment and at 8, 12, 14, 18, 24 and 28 weeks thereafter were examined by culture, and DNA in tissue homogenates was quantitated by polymerase chain reaction (PCR) and dot blot hybridisation. Results: Culture and quantitative PCR estimated initial bacillary content at about 10(7) per organ. Thereafter, organ cultures rapidly declined and were usually negative between 14 and 28 weeks. However, during this period quantitative PCR consistently estimated about 5.5 log(10) bacilli equivalents in spleens and lungs, Dot blot hybridisation sensitive to about 10 ng bacillary DNA was usually positive pretreatment and occasionally during and after treatment, hence confirming the PCR results. Conclusions: There is persistence of significant quantities of M. tuberculosis DNA throughout the various stages of the model. This may represent dead bacilli, free DNA or dormant forms.