Pressure versus Heat-Induced Unfolding of Ribonuclease A: The Case of Hydrophobic Interactions within a Chain-Folding Initiation Site ² Joan Torrent, ‡ James Patrick Connelly, § Maria Gra`cia Coll, ‡ Marc Ribo´, ‡ Reinhard Lange, § and Maria Vilanova* ,‡ Laboratori d’Enginyeria de Proteı ¨nes, Departament de Biologia, Facultat de Cie` ncies, UniVersitat de Girona, Campus de MontiliVi. E-17071 Girona, Spain, INSERM U128, IFR 24, 1919 Route de Mende (CNRS), F-34293 Montpellier, Cedex 5, France ReceiVed June 25, 1999; ReVised Manuscript ReceiVed September 21, 1999 ABSTRACT: To investigate the characteristics of the postulated carboxy terminal chain-folding initiation site in bovine pancreatic ribonuclease A (RNase A) (residues 106-118), important in the early stages of the folding pathway, we have engineered by site-directed mutagenesis a set of 14 predominantly conservative hydrophobic variants of the protein. The stability of each variant has been compared by pressure and temperature-induced unfolding, monitored by fourth derivative UV absorbance spectroscopy. Apparently simple two-state, reversible unfolding transitions are observed, suggesting that the disruption of tertiary structure of each protein at high pressure or temperature is strongly cooperative. Within the limits of the technique, we are unable to detect significant differences between the two processes of denaturation. Both steady-state kinetic parameters for the enzyme reaction and UV CD spectra of each RNase A variant indicate that truncation of hydrophobic side chains in this region has, in general, little or no effect on the native structure and function of the enzyme. Furthermore, the decreases in free energy of unfolding upon pressure and thermal denaturation of all the variants, particularly those modified at residues 106 and 108, suggest that the hydrophobic residues and side chain packing interactions of this region play an important role in maintaining the conformational stability of RNase A. We also demonstrate the potential of Tyr115 replacement by Trp as a non-destabilizing fluorescence probe of conformational changes local to the region. Protein unfolding has gained a lot of interest in the past few years. This is partly driven by the biotechnological potential of enzymes withstanding extreme conditions and the pathological role of partly denatured proteins in a variety of diseases. However, despite numerous studies, no generally accepted mechanistic model is available. Protein unfolding has been variously described as a two-state or a multi-state process comprising several intermediates (1, 2). Other authors have dismissed the idea of defined intermediates and explained experimental as well as simulation data in terms of more or less rugged conformational landscapes allowing multiple pathways from the folded to the unfolded form (3). These often contradicting views are easily explained by three factors: (a) due to the complexity of the macromolecular nature and the diversity of proteins, the experimental interpretation is necessarily model dependent; (b) different detection methods, such as NMR, 1 CD, fluorescence, etc., are likely to focus on particular unfolding events; (c) protein unfolding is induced by changes in different parameters, mainly by chemical denaturants, by heat or cold, and by pressure. While the use of chemical denaturants has the disadvantage that thermodynamic parameters have to be extrapolated from zero denaturant concentration, and it is often difficult to specify the precise interaction of the chemical agent with the protein, the use of heat is hampered by the fact that many proteins precipitate at high temperature or undergo irreversible structural changes. Pressure is a much less used effect that provides an elegant alternative, as it perturbs (within a limited range) chemical equilibrium reversibly. As most of the protein unfolding studies to date have been undertaken by the use of chemical denaturants or heat, it is of interest how the pressure data can be connected to them. This issue has been addressed most extensively in the case of staphylococcal nuclease wild-type and several variants by Royer and co-workers (4-8). ² This work was supported by Grants PB96-1172-CO2-02 from the DGES of the Ministerio de Educacio´n y Cultura, Spain, and ACI-96- 49 and ACI-97-20 from the CIRIT of the Generalitat de Catalunya, Spain. J.T. thanks the CIRIT of the Generalitat de Catalunya, Spain, for a short-term fellowship. J.P.C. was a Poste Vert (INSERM) postdoctoral fellow (1996-1998). Support was also received from the “Fundacio´ M. F. de Roviralta” of Barcelona for equipment purchasing grants. * To whom correspondence should be addressed. Fax: +34-972- 418-150. E-mail: dbmvb@fc.udg.es. ‡ Universitat de Girona. § CNRS. 1 Abbreviations: NMR, nuclear magnetic resonance; CD, circular dichroism; RNase A, bovine pancreatic ribonuclease A; CFIS, chain- folding initiation site; Us II and Uvf are unfolded species of RNase A corresponding to the major slow and the very fast refolding phases, respectively; C-terminal, carboxy-terminal; PCR, polymerase chain reaction; Tris, tris(hydroxymethyl)aminomethane; AcOH, acetic acid; FPLC, fast protein liquid chromatography; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; HPLC, high- performance liquid chromatography; MES, 4-morpholineethanesulfonic acid; CDA, cumulative difference amplitude; UV, ultraviolet; poly- (C), poly(cytidylic acid); C>p, cytidine 2′:3′-cyclic monophosphate; ATEE, N-acetyl-L-tyrosine ethyl ester; FTIR, Fourier transform infrared. 15952 Biochemistry 1999, 38, 15952-15961 10.1021/bi991460b CCC: $18.00 © 1999 American Chemical Society Published on Web 11/09/1999