The reaction of PQQ (2,7,9-tricarboxypyrroloquinoline quinone)-dependent MDH (methanol dehydrogenase) from Methylophilus methylotrophus has been studied under steady-state conditions in the presence of an alternative activator [GEE (glycine ethyl ester)] and compared with similar reactions performed with ammonium (used more generally as an activator in steady-state analysis of MDH). Studies of initial velocity with methanol (protiated methanol, C1H3O1H) and [2H]methanol (deuteriated methanol, C2H3O2H) as substrate, performed with different concentrations of GEE and PES (phenazine ethosulphate), indicate competitive binding effects for substrate and PES on the stimulation and inhibition of enzyme activity by GEE. GEE is more effective at stimulating activity than ammonium at low concentrations, suggesting tighter binding of GEE to the active site. Inhibition of activity at high GEE concentration is less pronounced than at high ammonium concentration. This suggests a close spatial relationship between the stimulatory (KS) and inhibitory (KI) binding sites in that binding of GEE to the KS site sterically impairs the binding of GEE to the KI site. The binding of GEE is also competitive with the binding of PES, and GEE is more effective than ammonium in competing with PES. This competitive binding of GEE and PES lowers the effective concentration of PES at the site competent for electron transfer. Accordingly, the oxidative half-reaction, which is second-order with respect to PES concentration, is more rate-limiting in steady-state turnover with GEE than with ammonium. The smaller methanol C-1H/C-2H kinetic isotope effects observed with GEE are consistent with a larger contribution made by the oxidative half-reaction to rate limitation. Cyanide is much less effective at suppressing 'endogenous' activity in the presence of GEE than with ammonium, which is attributed to impaired binding of cyanide to the catalytic site through steric interaction with GEE bound at the KS site. The kinetic model developed previously for reactions of MDH with ammonium [Hothi, Basran, Sutcliffe and Scrutton (2003) Biochemistry 42, 3966-3978] is consistent with data obtained with GEE, although a more detailed structural interpretation is given here. Molecular-modelling studies rationalize the kinetic observations in terms of a complex binding scenario at the molecular level involving two spatially distinct inhibitory sites (KI and KI'). The KI' site caps the entrance to the active site and is interpreted as the PES binding site. The KI site is adjacent to, and, for GEE, overlaps with, the KS site, and is located in the active-site cavity close to the PQQ cofactor and the catalytic site for methanol oxidation.