Effect of heat treatment on bovine beta-lactoglobulin A, B, and C explored using thiol availability and fluorescence

Dilute solutions of beta-lactoglobulin (beta-Lg) A, B, and C were heated at temperatures between about 40 and 94 degrees C for 10 min, cooled, and analyzed using Trp fluorescence and extrinsic fluorescence spectra of the probe 1,8-anilinonaphthalene sulfonate (ANS). Thiol availabilities using 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB) were determined using a separate set of samples. The normalized ANS fluorescence emission intensity and the thiol availability results showed a 1:1 relationship with the loss of nativelike but not SDS-monomeric protein, as determined by PAGE analysis. The normalized Trp emission intensity results did not show a comparable 1:1 relationship with the loss of nativelike protein, indicating that the Trp intensity arose from consequential disulfide bond reorganization and not the initial unfolding reaction. The results were also analyzed in terms of two-state models, and the midpoint temperatures (T(mid)) for the proteins were generally beta-Lg C > beta-Lg A > beta-Lg B, and the slopes at the midpoint temperatures for the A variant were generally less than those for the B and C variants indicating that beta-Lg A may denature by a different mechanism from that of beta-Lg B or beta-Lg C. The T(mid) parameters derived from the ANS fluorescence intensity results were similar to those for thiol availability and both were lower than the T(mid) values for Trp emission intensity showing that creation of an ANS binding site on a beta-Lg molecule was linked to the irreversible exposure of a thiol group and the loss of native beta-Lg but preceded the decrease in Trp(61) fluorescence quenching. These results for the differences between the behavior of the A and B or the C variants involved the creation of a destabilizing cavity by the Val(118)Ala (A --> B) substitution and the changed charge distribution within the CD loop caused by the Asp(64)Gly (A --> B) substitution.