α‐Helix to random coil transitions of two‐chain coiled coils: Experiments on the thermal denaturation of ββ tropomyosin cross‐linked selectively at C36

Current ideas on unfolding equilibria in two-chain, coiled-coil proteins are examined by studies of a species of ββ tropomyosin that is sulfhydryl blocked at C190 and disulfide cross-linked at C36 ($ ^. \beta \_\beta ^. $). The desired species is produced by a seven-step process: (1) Rabbit skeletal muscle, comprising predominantly αα and ββ species, is oxidized with ferricyanide, cross-linking both species at C190. (2) The product is carbamylated at C36 of β chains, using cyanate in denaturing medium at pH 6. (3) All C190 cross-links are reduced with dithiothreitol (DTT). (4) All C190 sulfhydryls are permanently blocked by carboxyamidomethylation. (5) Chromatography on carboxymethylcellulose in denaturing medium is used to separate C190-blocked α chains from C190-blocked, C36-carbamylated β chains. (6) The latter are decarbamylated in denaturing medium by raising the pH to 8.0. (7) The C190-blocked β chains are renatured and cross-linked at C36 by ferricyanide. The procedure and the quality of the final product are judged by NaDodSO₄/polyacrylamide gel electrophoresis, titration of free sulfhydryls, and electrophoretic analysis of trypsin digestion products. Thermal unfolding curves are reported for the resulting pure $ ^. \beta \_\beta ^. $ species and for its DTT-reduction product. The latter ($ ^. \beta \beta ^. $) show equilibrium thermal unfolding curves that are very similar to those of the parent ββ noncross-linked species. The $ ^. \beta \_\beta ^. $ cross-linked species unfolds in a single-phase, cooperative transition with a melting temperature intermediate between the pretransition and posttransition shown by its cross-linked counterpart, the C190 cross-linked, C36-blocked species ($ .\beta^- \beta .$), which was studied earlier. These transitions are compared with one another and with that of the doubly cross-linked species, , in the light of two extant physical models for such transitions. The all-or-none segments model successfully rationalizes the data qualitatively for the $ .\beta^- \beta .$ and $ ^. \beta \_\beta ^. $ species if the usual postulates of greater inherent stability of the amino vs the carboxyl end of the molecule and of strain at each cross-link are accepted. However, the same model then requires that the species be the least stable of the three, whereas experiment shows the opposite, thus falsifying the all-or-none segments model. The continuum-of-states model is also qualitatively in accord with data on the $ .\beta^- \beta .$ and $ ^. \beta \_\beta ^. $ species. In fact, the general features of the transition in C36 cross-linked vs C190 cross-linked species were predicted by a statistical mechanical theory embodying the continuum-of-states model for singly cross-linked species. Moreover, since the same theory avers that loop entropy greatly stabilizes the large region between cross-links in the species, perhaps offsetting the effect of strain, qualitative considerations alone are insufficient to falsify this model in the face of the data on doubly cross-linked species. Thus, one model can be eliminated, but the second cannot. However, until quantitative simulations are done and found to agree with these data, the continuum-of-states model must still be considered questionable.