Catalase-peroxidase (KatG) is a key enzyme in M. tuberculosis because it is the sole catalase in this pathogen, and because it is responsible for activation of the oldest, most potent anti-TB agent, isoniazid (INH). Peroxidative activation of INH by KatG leads to an intermediate that acylates nicotinamide adenine dinucleotide cofactor, forming a potent inhibitor of a mycolic acid biosynthetic enzyme known as InhA and causing cell wall damage. Activation is deficient in the Ser315Thr KatG mutant associated with INH resistance worldwide. The early phase of catalytic turnover in KatG and three Ser315 mutants was probed using peroxide and the peroxide surrogate, HCN. Equilibrium and kinetics methods along with EPR measurements were used to characterize ligand binding. kON and kOFF rates were evaluated and ligand binding to five or six coordinate heme iron species in the resting enzymes could be separately probed using cyanide. While peroxidases generally contain five coordinate heme iron in their resting state, in KatG, the six-coordinate heme sub-population, which contains a distal water molecule associated with heme iron, was found to be the preferred species in the ligand binding process. According to its three-dimensional crystal structure, the Ser315Thr KatG mutant lacks structured water in the active site and in a substrate access channel, and manifests slow ligand binding. The results suggest that the reaction mechanism in catalase-peroxidases should include a specific role for water molecules in the active site required for substrate (peroxide) access and deprotonation allowing coordination of HOO- to heme iron for initiation of turnover.