Structural and mechanistic insights into Mycobacterium abscessus as-partate decarboxylase PanD and a pyrazinoic acid-derived inhibitor

Abstract

Mycobacterium tuberculosis (Mtb) aspartate decarboxylase PanD is required for biosynthesis of the essential cofactor coenzyme A and targeted by the first line drug pyrazinamide (PZA). PZA is a prodrug that is converted by a bacterial amidase into its bioactive form pyrazinoic acid (POA). Employing structure-function analyses we previously identified POA-based inhibitors of Mtb PanD showing much improved inhibitory activity against the enzyme. Here, we performed the first structure-function studies on PanD encoded by the non-tuberculous mycobacterial lung pathogen Mycobacterium abscessus (Mab), shedding light on the differences and similarities of Mab and Mtb PanD. Solution X-ray scattering data provided the solution structure of the entire tetrameric Mab PanD, which in comparison to the structure of the derived C-terminal trun-cated Mab PanD1-114 mutant, revealed the orientation of the four flexible C-termini relative to the catalytic core. Enzymatic studies of Mab PanD1-114 explored the essentiality of the C-terminus for catalysis. A library of recombinant Mab PanD mutants based on structural information and PZA/POA resistant PanD mutations in Mtb, illuminated critical residues involved in the substrate tunnel and enzymatic activity. Using our library of POA analogs, we identified (3-(1-naphthamido)pyrazine-2-car-boxylic acid) (analog 2) as the first potent inhibitor of Mab PanD. The inhibitor shows mainly electrostatic- and hydrogen bonding interaction with the target enzyme as explored by isothermal titration calorimetry and confirmed by docking studies. The observed unfavorable entropy indicates that significant conformational changes are involved in the binding process of analog 2 to Mab PanD. In contrast to PZA and POA, which are whole-cell inactive, analog 2 exerts appreciable antibacterial activity against the three subspecies of Mab.