We have studied CD4؉ T cells that mediate immunological memory to an intravenous infection with Mycobacterium tuberculosis. The studies were conducted with a mouse model of memory immunity in which mice are rendered immune by a primary infection followed by antibiotic treatment and rest. Shortly after reinfection, tuberculosis-specific memory cells were recruited from the recirculating pool, leading to rapidly increasing precursor frequencies in the liver and a simultaneous decrease in the blood. A small subset of the infiltrating T cells was rapidly activated (<20 h) and expressed high levels of intracellular gamma interferon and the T-cell activation markers CD69 and CD25.

It is estimated that one-third of the world's population is infected with Mycobacterium tuberculosis, but that only 10% of infected people break down with the disease. In the remaining 90% the infection remains clinically latent. In the present study, the immune mechanisms controlling the latent phase of tuberculosis infection were evaluated in a mouse model of latency and reactivation. Mice aerosol-infected with M. tuberculosis were treated with anti-mycobacterial drugs resulting in very low, stable bacterial numbers (<500 CFU in the spleen and lung) for 10-12 weeks followed by reactivation of the disease with increasing bacterial numbers. During latency, pathological changes in the lung had almost completely resolved and lymphocyte number and turnover were at the pre-infection level. The CD4 subset was highly active during the acute phase of infection and could be detected by intracellular staining for IFN-gamma as well as after antigen-specific stimulation with mycobacterial antigens. The CD8 subset was not involved in the acute stage of infection, but this subset was active and produced IFN-gamma during the latent phase of infection. In vivo depletion of T cell subsets supported these findings with a 6-7-fold increase in bacterial numbers in the lung following anti-CD4 treatment during the acute phase, while anti-CD8 treatment did not have an effect. The opposite was found during the latent phase where anti-CD8 treatment as well as anti-IFN-gamma treatment both resulted in a 10-fold increase in bacterial numbers in the lung, while anti-CD4 treatment induced only a modest change.

The delayed-type hypersensitivity (DTH) to purified protein derivative (PPD) test has been used to infer about protective immunity to Mycobacterium tuberculosis and to diagnose tuberculosis. We showed that in memory tuberculosis-immune mice both DTH to PPD and resistance to M. tuberculosis could be effectively elicited in the footpad and both reactions led to the accumulation of reactive T cells in the regional lymph nodes with a CD4+ phenotype and characterized by the secretion of high levels of interleukin-2 (IL-2) and interferon-gamma (IFN-gamma) and no IL-4. By adoptive transfer into nude mice of highly purified CD4+ T cells harvested during the recall of protective immunity it was confirmed that this population mediated both manifestations. However, the specificity of the T cells recruited during these processes were found to differ markedly; T cells involved in protection to a challenge with live tuberculosis bacilli recognized predominantly low-mass culture filtrate antigens below 15 000 MW, while cells recruited during DTH to PPD were directed to molecular mass fractions between 15 000 and 31 000. Using single purified antigens we showed that the latter cells recognized the secreted mycobacterial protein Ag85B and the heat-shock proteins, DnaK and GroEL. Protective T cells, in contrast, were characterized by a very high frequency of T cells directed to the ESAT-6 peptide 1-20.

It is estimated that one‐third of the world's population is infected with Mycobacterium tuberculosis, but that only 10% of infected people break down with the disease. In the remaining 90% the infection remains clinically latent. In the present study, the immune mechanisms controlling the latent phase of tuberculosis infection were evaluated in a mouse model of latency and reactivation. Mice aerosol‐infected with M. tuberculosis were treated with anti‐mycobacterial drugs resulting in very low, stable bacterial numbers (<500 CFU in the spleen and lung) for 10–12 weeks followed by reactivation of the disease with increasing bacterial numbers. During latency, pathological changes in the lung had almost completely resolved and lymphocyte number and turnover were at the pre‐infection level. The CD4 subset was highly active during the acute phase of infection and could be detected by intracellular staining for IFN‐γ as well as after antigen‐specific stimulation with mycobacterial antigens. The CD8 subset was not involved in the acute stage of infection, but this subset was active and produced IFN‐γ during the latent phase of infection. In vivo depletion of T cell subsets supported these findings with a 6–7‐fold increase in bacterial numbers in the lung following anti‐CD4 treatment during the acute phase, while anti‐CD8 treatment did not have an effect. The opposite was found during the latent phase where anti‐CD8 treatment as well as anti‐IFN‐γ treatment both resulted in a 10‐fold increase in bacterial numbers in the lung, while anti‐CD4 treatment induced only a modest change.