Mycobacterium tuberculosis FBA Binds Human Plasminogen
The first report of an interaction between M. tuberculosis and plasminogen came from Monroy and colleagues who used flow cytometry to demonstrate binding (Monroy et al. 2000). Subsequent far-Western blotting of separated soluble protein extracts and cell-wall proteins indicated the presence of four plasminogen-binding proteins of molecular weight 66, 60, 55, and 30 kDa. Plasminogen binding by these proteins was abolished in the presence of the lysine analog e-aminocaproic acid (EACA), suggesting the involvement of lysine residues in the interactions (Monroy et al. 2000). Further work by the same group involved separating the proteins present in soluble extracts and culture filtrates from M. tuberculosis by 2D gel electrophoresis. Following transfer to PVDF, membranes were probed with human plasminogen and mass spectrometry and amino- terminal amino acid sequencing identified 15 putative plasminogen-binding proteins, one of which was M. tuberculosis FBA (FBA-tb) (Xolalpa et al. 2007).
More recent studies have shown the presence of FBA-tb in the cytosol, culture filtrate, cell wall, and cell-membrane fractions (de la Paz Santangelo et al. 2011). Flow cytometry and enzymatic activity measurements revealed that FBA-tb was exported to the cell surface and produced under various axenic growth conditions, including oxygen depletion, and hence by non-replicating bacilli (de la Paz Santangelo et al. 2011). Importantly, FBA-tb expression was demonstrated in vivo in the lungs of experimentally infected guinea pigs and mice (de la Paz Santangelo et al. 2011). ELISA experiments confirmed that FBA-tb binds human plasminogen in a dose-dependent manner (KD = 6.7 ± 3 nM); an interaction which could be abolished by EACA (de la Paz Santangelo et al. 2011). In contrast, the presence of TD3, a competitive inhibitor of aldolase activity, had no effect on plasminogen binding, demonstrating that plasminogen binding was independent of aldolase activity (de la Paz Santangelo et al. 2011). Tissue plasminogen activator (tPA), but not FBA-tb itself, was able to activate FBA-bound plasminogen to plasmin, while a2-antiplasmin reduced plasmin activity in the absence of FBA-tb. In contrast, the presence of FBA-tb reduced this inhibition, perhaps suggesting competition between FBA-tb and a2-antiplasmin for the same binding site on plasmin. Interestingly, FBA from M. leprae (which shares 87% amino acid identity with FBA-tb) was also detected in cell wall and membrane fractions of the leprosy bacillus purified from chronically infected armadillo tissues, suggesting that the moonlighting properties of FBA in Mycobacterium species may not be limited to M. tuberculosis (de la Paz Santangelo et al. 2011).
Attempts at inactivating the gene encoding FBA in M. tuberculosis followed by growth in standard in vitro culture media have been unsuccessful, suggesting that FBA is essential for viability in this organism (Griffin et al. 2011). However, the use of conditional mutants has shown that strains lacking FBA-tb expression are viable if supplied with an appropriate combination of carbon substrates entering metabolism above and below the FBA-catalyzed reaction (de la Paz Santangelo et al. 2011; Puckett et al. 2014). Mouse in vivo experiments confirmed that M. tuberculosis requires FBA activity for growth during acute infections and for persistence during chronic infections, presumably due either to the lack of a growth permissive ratio of carbon sources that can compensate for the lack of FBA-tb and/or the loss of moonlighting functions (Puckett et al. 2014). Given the emergence of multidrug-resistant M. tuberculosis, and since human (i.e., class I) and bacterial (class II) FBAs are structurally and mechanistically distinct, inhibitors of FBA-tb are currently being developed as potential anti-tuberculosis therapies (Daher et al. 2010; Labbe et al. 2012).