Link between Mutation in phoP and Secretion of ESAT-6 
As PhoP fulfills important regulatory functions in M. tuberculosis [21,22], it was of primary interest to identify and study potential effector molecules whose involvement in host pathogen interaction were influenced by the point mutation in phoP of H37Ra.
Since secreted proteins of the ESX-1 system of M. tuberculosis constitute a major interface between the bacterium and its host [23], we analyzed the different strains for their potential to induce T cell responses against ESAT-6 and its binding partner, the 10-kD culture filtrate protein CFP-10.
Groups of C57BL/6 mice were subcutaneously inoculated with H37Rv, H37Ra, or one of three recombinant H37Ra strains complemented with phoP, fadE5, or rpsL, respectively.
Two weeks after vaccination we assessed the interferon-gamma (IFN-gamma) production of splenocytes mounted against ESX-1 antigens or controls.
As anticipated, all tested strains induced IFN-gamma production in response to purified protein derivative (PPD) but not to a control peptide (Mal-E), indicating successful vaccination (Figure 3).
Most importantly, the various strains differed extensively in their potential to induce antigen specific T cell responses towards ESAT-6 and CFP-10.
As shown in Figure 3, splenocytes from mice that were inoculated with H37Rv produced high amounts of IFN-gamma upon stimulation with ESAT-6 or CFP-10, whereas the responses of H37Ra, H37Ra::fadE5, and H37Ra::rpsL infected mice were extremely weak.
In contrast, splenocytes from H37Ra::phoP inoculated mice showed very strong IFN-gamma production in response to incubation with ESAT-6 and CFP-10.
Exactly the same trend was observed when a highly sensitive T cell hybridoma assay was used to investigate ESAT-6 secretion.
This test is based on the presentation of the immunodominant epitope contained in the first 20 amino acids of ESAT-6 by H37Ra, recombinant H37Ra, or H37Rv infected bone marrow-derived dendritic cells (BM-DC) to an ESAT-6-specific T cell hybridoma (NB11), restricted by I-Ab.
Figure 4A shows that infection of BM-DC with H37Ra or H37Ra::rpsL induced no stimulation of the hybridoma as judged by IL-2 production, while H37Ra::phoP or H37Rv triggered high amounts of IL-2 production in this very sensitive and specific test.
All strains behaved comparably towards the Ag85A:241-260 control, emphasizing the specificity of the observed phenomenon for ESAT-6 (Figure 4B).
To further evaluate the involvement of phoP and the ESX-1 system in immunogenicity, C57BL/6 mice were vaccinated with additional strains.
Firstly, we constructed a partially diploid H37Ra knock-in strain carrying cosmid RD1-ppe68-ko that, in BCG yields the greatest amounts of ESAT-6 expression and secretion [24].
However, in H37Ra the presence of this cosmid could not increase the weak ESAT-6 and CFP-10 T cell responses (Figure 3B) in the splenocyte IFN-gamma assay.
Secondly, we also tested the previously described M. tuberculosis MT103 phoP knock-out (ko) strain SO2 [14,25] in this assay and found that this strain induced a potent PPD response as previously reported [25].
However, in comparison to the corresponding MT103 wild-type strain, the SO2 strain showed a strongly reduced ESAT-6 and CFP-10 specific T cell response, corresponding to a similar low level as the H37Ra strain (Figure 3B).
Altogether, our data obtained from multiple replicate experiments indicate that a direct link exists between a fully functional PhoP and the ability to generate T cell responses against ESAT-6 and CFP-10.
