Identification of secreted proteins of Mycobacterium tuberculosis by a bioinformatic approach.

Proteins secreted by Mycobacterium tuberculosis are usually targets of immune responses in the infected host. Here we describe a search for secreted proteins that combined the use of bioinformatics and phoA' fusion technology. The 3,924 proteins deduced from the M. tuberculosis genome were analyzed with several computer programs. We identified 52 proteins carrying an NH(2)-terminal secretory signal peptide but lacking additional membrane-anchoring moieties. Of these 52 proteins-the TM1 subgroup-only 7 had been previously reported to be secreted proteins. Our predictions were confirmed in 9 of 10 TM1 genes that were fused to Escherichia coli phoA', a marker of subcellular localization. These findings demonstrate that the systematic computer search described in this work identified secreted proteins of M. tuberculosis with high efficiency and 90% accuracy.