How a small molecule flips a vital tuberculosis enzyme

Tuberculosis bacteria survive long-term inside human hosts by switching how they make and use energy. A key player in that switch is Mycobacterium tuberculosis isocitrate lyase 2 (ICL2), an enzyme that lets the bacterium grow on non-glycolytic substrates such as fatty acids during infection. Previous work had shown that small cellular metabolites like acetyl-CoA and propionyl-CoA can turn ICL2 on, but exactly how those molecules did so was unknown. In new work led by Ivanhoe K. H. Leung, researchers set out to uncover the molecular mechanism behind this allosteric regulation—how a molecule binding at one site changes activity at another. They focused on two structural features that set ICL2 apart: its C-terminal domain and a unique helical substructure on its N-terminal catalytic domain. By examining how these pieces move and interact, the team discovered that acetyl-CoA binding does not serve as a substrate reaction but instead promotes changes in the enzyme’s quaternary structure. Those changes disrupt internal contacts and trigger conformational shifts that switch ICL2 into an active state, explaining how a central metabolite can remotely control this survival enzyme.

To reveal this mechanism the team combined several complementary techniques: protein NMR, crystallography, molecular dynamics and mutagenesis. Crystallography provided static snapshots of ICL2’s architecture, highlighting the distinct C-terminal domain and the unique helical substructure on the N-terminal catalytic domain. protein NMR reported on how different regions move in solution and respond to ligand binding, while molecular dynamics simulations traced the sequence of movements triggered when acetyl-CoA docks. Mutagenesis experiments—changing specific amino acids—tested which contacts were essential for the observed changes. Together these approaches showed that acetyl-CoA binding promotes the dimerisation of the C-terminal domain and disrupts its interactions with the helical substructure on the N-terminal catalytic domain. Those disrupted contacts lead to larger conformational changes across ICL2 and produce enzyme activation. The study also places acetyl-CoA alongside propionyl-CoA as central metabolites able to allosterically activate ICL2, and it contrasts these regulators with true enzyme substrates by noting that acetyl-CoA is not an ICL2 substrate.

These findings matter for two linked reasons. Scientifically, the work reveals, for the first time, how the binding of acetyl-CoA—which does not undergo reaction by ICL2—induces activation through a dynamic allosteric mechanism involving domain dimerisation and disruption of specific interdomain contacts. That advances our understanding of protein allostery and offers a clear example of how metabolite signals can rewire enzyme activity through structural rearrangement. Medically, ICL2, together with its isoform ICL1, is essential for Mycobacterium tuberculosis survival and pathogenesis, and the bacterium is responsible for the most deaths worldwide due to a single bacterial agent. By identifying distinct structural elements and a discrete allosteric pathway that control ICL2 activity, the study points to new opportunities to design drugs that block this regulatory switch. Finally, the combination of protein NMR, crystallography, molecular dynamics and mutagenesis used here provides a transferable toolkit for dissecting allostery in other proteins and for guiding therapeutic manipulation of metabolic control points in pathogens.