The coincidence of AIDS and tuberculosis is such a major public health threat that it has been called the """"""""cursed duet"""""""". We seek to learn how HIV-1 affects the pathogenicity of M. tuberculosis and vice versa. In regions where tuberculosis is endemic, co-infection mainly reflects re-activation of latent tuberculosis with the onset of AIDS. However, in some populations, infection by HIV-1 precedes acute exposure to M. tuberculosis. Moreover, active tuberculosis also might affect the course of initial HIV-1 infection. In vitro studies of co-infection have not compared order of infection. We will test the hypothesis that the order of infection will influence the outcome of co-infection. This can be done most effectively with a functional genomics approach that is not constrained by pre-judgement of what differences are likely. Functional genomic tests of global hypotheses also provide abundant data to generate new hypotheses that may lead in unexpected directions. For example, our functional genomic data from human macrophages infected by M. tuberculosis, revealed increased expression of the HIV-1 co-receptor CXCR4. That single observation suggested a novel and significant aspect of HIV-macrophage interaction during tuberculosis that we then tested in co-infection experiments. To date, neither infection of macrophages by HIV-1 nor co-infection has been studied with functional genomics.
Our aim i s to characterize differences in human macrophage gene expression profiles in cells coinfected by HIV-1 and M. tuberculosis compared to cells that are uninfected or infected by only one pathogen, for two distinct co-infection scenarios modeled in vitro. We will infect first by HIV-1 and then by M. tuberculosis or vice versa. The gene expression profiles will be characterized using a spotted micro-array of highly specific 60-base oligonucleotide probes for approximately 19,000 human genes. Replicate experiments will be performed so that statistically significant changes in gene expression can be determined, and these will be analyzed using algorithms for Boolean logic, clustering, and profiling. Based on bioinformatic analyses and on informed appreciation of individual changes in gene expression, we will develop new hypotheses. Moreover, our data will clearly show whether co-infected cells exhibit gene expression profiles that differ based on whether they were infected first by HIV-1 and then by Mr. tuberculosis or vice versa. Such differences are expected to provide a basis for addressing the course of disease in different groups of patients represented by the models.
