Apo-dihydrofolate reductase from Escherichia coli samples two distinct environments slowly on the NMR time scale at room temperature. Several assigned resonances belong to residues in, or proximal to, a loop (loop I) which is comprised of residues 9-24. This exchange process was altered (either removed or made fast on the NMR time scale) by deleting three hairpin turn forming residues from the loop and filling the gap with a single glycine [Li, L., Falzone, C. J., Wright, P. E., & Benkovic, S. J. (1992) Biochemistry 31, 7826-7833]. An approximate value of 35 s-1 for the exchange rate associated with loop I in apo-DHFR was obtained in two-dimensional nuclear Overhauser spectra by analyzing the time dependence of the cross-peak volume for N epsilon H of Trp-22, a residue which is located in this loop and which has resolved cross-peaks. Owing to the critical role that this loop plays in catalysis, the correspondence between this rate of conformational exchange and off-rates for tetrahydrofolate and the reduced nicotinamide cofactor from product and substrate complexes suggests that loop movement may be a limiting factor in substrate turnover.