Rhodobacter sphaeroides periplasmic nitrate reductase (Rs NapAB) is one of the enzymes whose assays give odd results: in spectrophotometric assays with methyl viologen as the electron donor, the activity increases as the reaction progresses, whereas the driving force provided by the soluble redox partner decreases; in protein film voltammetry (PFV), whereby the enzyme directly exchanges electrons with an electrode, the activity of NapAB decreases at large overpotential, whereas a monotonic increase is expected [Elliott, S. J., et al. (2002) Biochim. Biophys. Acta 1555, 54-59]. The relations between these phenomena and the catalytic mechanism are still debated. By studying NapAB mutants, we found that the peculiar dependences of electrochemical and solution activities on driving force are greatly affected by substituting certain amino acids that are located in the vicinity of the active site (M153, Q384, R392); this led us to establish and discuss the relation between the experimental parameters of the electrochemical and spectrophotometric assays: we show that the rate of reduction of the enzyme (which depends on the electrode potential or on the concentration of reduced MV) modulates the activity of the enzyme, but the "solution potential" does not. Our results also support the view that the complex profiles of activity versus potential are fingerprints of the active site chemistry, rather than direct consequences of changes in the redox states of relays that are remote from the active site.