Our initial studies of hydrogen-deuterium (H-D) exchange of tryptophan 109 in Escherichia coli alkaline phosphatase (AP) suggested that significant local unfolding of the protein might occur to allow for the exchange reaction, which is very slow at room temperature (Fischer et al., Biochemistry 39 (2000) 1455-1461). In order to investigate whether the partial unfolding and/or 'breathing' motions leading to H-D exchange were part of the unfolding pathway of the protein we prepared a series of mutants, designed to produce cavities around the exchanging residue, and compared their rates of H-D exchange to their lability (rate of inactivation) in guanidine hydrochloride (Gd:HCl). The complex unfolding kinetics of the mutants in the presence of Gd:HCl showed several components with rates that differed substantially among these proteins, but none of the rates of denaturation induced with Gd:HCl was consistently correlated with the H-D exchange rates. We conclude that the partial opening of the AP structure during the H-D exchange of tryptophan 109, although very slow, is not a rate determining step in the unfolding of this protein.