DISSECTING THE DISULFIDE-COUPLED FOLDING PATHWAY OF BOVINE PANCREATIC TRYPSIN-INHIBITOR - FORMING THE 1ST DISULFIDE BONDS IN ANALOGS OF THE REDUCED PROTEIN

The kinetics of disulphide bond formation and breakage have been measured in five analogues of the single-disulphide intermediates that occur in folding of bovine pancreatic trypsin inhibitor (BPTI), in which the cysteine residues not involved in disulphide bonds have been replaced by serine residues. Only a single disulphide bond can be made in each analogue, allowing the rates and equilibrium constants of the corresponding microscopic steps to be determined. These steps cannot be characterized in normal BPTT with six cysteine residues, as only the total overall rate of forming all 15 possible disulphide bonds can be measured. The intramolecular rate of forming each disulphide bond was found in the dithiol forms of reduced BPTI to be approximately proportional inversely to the size of the disulphide loop formed. On this basis, the 30-51 disulphide bond is not formed preferentially in the fully reduced protein, even though it predominates amongst the one-disulphide intermediates as a result of rapid intramolecular disulphide rearrangements. It is found to be more stable than the other one-disulphide intermediates because its disulphide is more stable to reduction than the others. This is undoubtedly due to the stable partially folded structure of (30-51), as the other intermediates are much less structured under folding conditions. Urea (8 M) uniformly decreased by a factor of 5 the effective concentrations of cysteine thiol groups in the analogues and in reduced BPTI, suggesting that reduced BPTI under normal refolding conditions is not an ideal random coil; any non-random conformations do not, however, favour formation of any specific disulphide bonds. Findings from these studies have led to further refinement and development of various approaches to measuring and interpreting the intramolecular rates of disulphide bond formation and breakage in proteins.