In unfolded RNase A there is an interconversion between slow-folding and fast-folding forms (Us UF) that is known to show properties characteristic of proline isomerization in model peptides. Here, we accept the evidence that Us molecules contain nonnative proline isomers and we ask about the isomerization of these proline residues during folding. The Us UF reaction in unfolded RNase A is used both to provide data on the kinetics of proline isomerization in the unfolded protein and as the basis of an assay for measuring proline isomerization during folding.The tyrosine-detected folding kinetics at low temperatures have been compared to those of proline isomerization in unfolded RNase A. The comparison is based on the recent observation that the Us UF kinetics are independent of guanidinium chloride concentration, so that they can be extrapolated to low guanidinium chloride concentrations, at which folding takes place. At 00C the tyrosine-detected folding reaction is 100-fold faster than the conversion of Us to UF in unfolded RNase A. Consequently, the folding reaction is not rate-limited by proline isomerization as it occurs in unfolded RNase A.An assay is given for proline isomerization during folding. The principle is that native RNase A yields UF on unfolding, whereas protein molecules that still contain nonnative proline isomers yield Us. Unfolding takes place at 0C, at which proline isomerization is slow compared to unfolding. This assay yields two important results: (i) The kinetics of proline isomerization during folding are substantially faster than in unfolded RNase A-e.g., 40-fold at 0C. The mechanism of the rate enhancement is unknown. (ii) At low temperatures (0-100C), and also in the presence of (NH42S04, the tyrosine-detected folding reaction occurs before proline isomerization and yields a folded intermediate IN that is able to bind the specific inhibitor 2S-CMP. The results demonstrate that a folding intermediate is spectrally detectable when folding occurs at low temperatures. They suggest that low temperatures provide suitable conditions for determining the kinetic pathway of folding by characterizing folding intermediates.In unfolded RNase A there is an equilibrium between fastfolding (UF) and slow-folding (Us) forms (1-6). The kinetics of the UF Us reaction can be measured by "double-jump" experiments (4, 7). First, native RNase A (N) is unfolded in a reaction that yields UF. Under suitable conditionst this is the unfolding reaction measured by the change in tyrosine absorbance (7). When unfolding occurs at 00C, an almost quantitative conversion of N to UF can be obtained before there is a significant conversion of UF to Us, as we show here. Afterwards, the kinetics of the slow equilibration of UF