RESUMEN
In an effort to gain greater insight into the molecular mechanism of the electron-transfer reactions of cytochrome b(5), the bovine cytochrome b(5)-horse cytochrome c complex has been investigated by high-resolution multidimensional NMR spectroscopy using (13)C, (15)N-labeled cytochrome b(5) expressed from a synthetic gene. Chemical shifts of the backbone (15)N, (1)H, and (13)C resonances for 81 of the 82 residues of [U-90% (13)C,U-90% (15)N]-ferrous cytochrome b(5) in a 1:1 complex with ferrous cytochrome c were compared with those of ferrous cytochrome b(5) in the absence of cytochrome c. A total of 51% of these residues showed small, but significant, changes in chemical shifts (the largest shifts were 0.1 ppm for the amide (1)H, 1.15 for (13)C(alpha), 1.03 ppm for the amide (15)N, and 0.15 ppm for the (1)H(alpha) resonances). Some of the residues exhibiting chemical shift changes are located in a region that has been implicated as the binding surface to cyt c [Salemme, F. R. (1976) J. Mol. Biol. 10, 563-568]. Surprisingly, many of the residues with changes are not located on this surface. Instead, they are located within and around a cleft observed to form in a molecular dynamics study of cytochrome b(5) [Storch, E. M., and Daggett, V. (1995) Biochemistry 34, 9682-9693](.) The rim of this cleft can readily accommodate cytochrome c. Molecular dynamics simulations of the Salemme and cleft complexes were performed for 2 ns and both complexes were stable.
Asunto(s)
Grupo Citocromo c/química , Citocromos b5/química , Resonancia Magnética Nuclear Biomolecular , Secuencia de Aminoácidos , Animales , Secuencia de Bases , Sitios de Unión , Isótopos de Carbono , Bovinos , Compuestos Ferrosos/química , Caballos , Datos de Secuencia Molecular , Resonancia Magnética Nuclear Biomolecular/métodos , Conformación Proteica , Protones , TermodinámicaRESUMEN
In the accompanying paper [Storch et al. (1999) Biochemistry 38, 5054-5064] equilibrium denaturation studies and molecular dynamics (MD) simulations were used to investigate localized dynamics on the surface of cytochrome b5 (cyt b5) that result in the formation of a cleft. In those studies, an S18C:R47C disulfide mutant was engineered to inhibit cleft mobility. Temperature- and urea-induced denaturation studies revealed significant differences in Trp 22 fluorescence between the wild-type and mutant proteins. On the basis of the results, it was proposed that wild type populates a conformational ensemble that is unavailable to the disulfide mutant and is mediated by cleft mobility. As a result, the solvent accessibility of Trp 22 is decreased in S18C:R47C, suggesting that the local environment of this residue is less mobile due to the constraining effects of the disulfide on cleft dynamics. To further probe the structural effects on the local environment of Trp 22 caused by inhibition of cleft formation, we report here the results of steady-state and time-resolved fluorescence quenching, differential phase/modulation fluorescence anisotropy, and 1H NMR studies. In Trp fluorescence experiments, the Stern-Volmer quenching constant increases in wild type versus the oxidized disulfide mutant with increasing temperature. At 50 degrees C, KSV is nearly 1.5-fold greater in wild type compared to the oxidized disulfide mutant. In the reduced disulfide mutant, KSV was the same as wild type. The bimolecular collisional quenching constant, kq, for acrylamide quenching of Trp 22 increases 2.7-fold for wild type and only 1.8-fold for S18C:R47C, upon increasing the temperature from 25 to 50 degrees C. The time-resolved anisotropy decay at 25 degrees C was fit to a double-exponential decay for both the wild type and S18C:R47C. Both proteins exhibited a minor contribution from a low-amplitude fast decay, consistent with local motion of Trp 22. This component was more prevalent in the wild type, and the fractional contribution increased significantly upon raising the temperature. The fast rotational component of the S18C:R47C mutant was less sensitive to increasing temperature. A comparison of the 1H NMR monitored temperature titration of the delta-methyl protons of Ile 76 for wild type and oxidized disulfide mutant, S18C:R47C, showed a significantly smaller downfield shift for the mutant protein, suggesting that Trp 22 in the mutant protein experiences comparatively decreased cleft dynamics in core 2 at higher temperatures. Furthermore, comparison of the delta'-methyl protons of Leu 25 in the two proteins revealed a difference in the ratio of the equilibrium heme conformers of 1.2:1 for S18C:R47C versus 1.5:1 for wild type at 40 degrees C. The difference in equilibrium heme orientations between wild type and S18C:R47C suggests that the disulfide bond affects heme binding within core 1, possibly through damped cleft fluctuations. Taken together, the NMR and fluorescence studies support the proposal that an engineered disulfide bond inhibits the formation of a dynamic cleft on the surface of cyt b5.
Asunto(s)
Citocromos b5/química , Disulfuros/química , Triptófano/química , Sustitución de Aminoácidos/genética , Animales , Arginina/química , Arginina/genética , Citocromos b5/genética , Polarización de Fluorescencia , Isoleucina/química , Isoleucina/genética , Modelos Moleculares , Resonancia Magnética Nuclear Biomolecular , Ingeniería de Proteínas , Ratas , Serina/química , Serina/genética , Espectrometría de Fluorescencia , Temperatura , TermodinámicaRESUMEN
A previous molecular dynamics (MD) simulation of cytochrome b5 (cyt b5) at 25 degrees C displayed localized dynamics on the surface of the protein giving rise to the periodic formation of a cleft that provides access to the heme through a protected hydrophobic channel [Storch and Daggett (1995) Biochemistry 34, 9682]. Here we describe the production and testing of mutants designed to prevent the cleft from opening using a combination of experimental and theoretical techniques. Two mutants have been designed to close the surface cleft: S18D to introduce a salt bridge and S18C:R47C to incorporate a disulfide bond. The putative cleft forms between two separate cores of the protein: one is structural in nature and can be monitored through the fluorescence of Trp 22, and the other binds the heme prosthetic group and can be tracked via heme absorbance. An increase in motion localized to the cleft region was observed for each protein, except for the disulfide-containing variant, in MD simulations at 50 degrees C compared to simulations at 25 degrees C. For the disulfide-containing variant, the cleft remained closed. Both urea and temperature denaturation curves were nearly identical for wild-type and mutant proteins when heme absorbance was monitored. In contrast, fluorescence studies revealed oxidized S18C:R47C to be considerably more stable based on the midpoints of the denaturation transitions, Tm and U1/2. Moreover, the fluorescence changes for each protein were complete at approximately 50 degrees C and a urea concentration of approximately 3.9 M, significantly below the temperature and urea concentration (62 degrees C, 5 M urea) required to observe heme release. In addition, solvent accessibility based on acrylamide quenching of Trp 22 was lower in the S18C:R47C mutant, particularly at 50 degrees C, before heme release [presented in the accompanying paper (58)]. The results suggest that a constraining disulfide bond can be designed to inhibit dynamic cleft formation on the surface of cyt b5. Located near the heme, the native dynamics of the cleft may be functionally important for protein-protein recognition and/or complex stabilization.
Asunto(s)
Citocromos b5/química , Disulfuros/química , Sales (Química)/química , Animales , Citocromos b5/genética , Citocromos b5/metabolismo , Hemo/química , Hemo/metabolismo , Calor , Modelos Moleculares , Mutagénesis Sitio-Dirigida , Oxidación-Reducción , Unión Proteica/genética , Desnaturalización Proteica , Ingeniería de Proteínas , Ratas , Espectrometría de Fluorescencia , Termodinámica , Triptófano/química , Urea/químicaRESUMEN
Molecular dynamics simulations of rat and bovine apocytochrome b5 were performed to investigate the structural and dynamical consequences of heme removal. A crystal structure is available for the bovine holoprotein, while experimental studies of apocytochrome b5 have focused on the rat protein. The rat and bovine proteins are 93% homologous by sequence, and the sequence differences (six residues) appear to have no effect on the structure of the native holoprotein, as seen by the correlation of a bovine simulation with rat holocytochrome b5 experimental data (Storch & Daggett, 1995). There was a marked effect, however, on the structure and dynamics of the apo form. The bovine apocytochrome b5 simulation displayed subtle inconsistencies when compared to the experimental results on the rat apoprotein. Therefore, the rat protein was constructed from the bovine crystal structure coordinates. The MD simulation of the rat apoprotein displayed greater deviations from the crystal structure, yet it was in much closer agreement to the experimental data for the apoprotein. Additionally, the six variant residues fall in the regions where the bovine protein deviated from experiment. The two hydrophobic cores of the rat protein behaved very differently. Core 2 was well maintained, retained native-like structure, and is in good agreement with NMR data (Moore & Lecomte, 1990). Conversely, core 1, which is normally constrained by the prosthetic heme group, exhibited conformational heterogeneity, increased mobility, and some loss of secondary structure. Thus, the model of rat apocytochrome b5 complements past studies by providing structural information about core 1 that has proved difficult to obtain by experiment. The bovine simulation serves as a prediction, since little to no experimental data exist for this form of the apoprotein.
Asunto(s)
Apoproteínas/química , Grupo Citocromo b/química , Hemo/química , Conformación Proteica , Animales , Bovinos , Fenómenos Químicos , Química Física , Simulación por Computador , Citocromos b , Bases de Datos Factuales , Espectroscopía de Resonancia Magnética , Modelos Moleculares , Pliegue de Proteína , Estructura Secundaria de Proteína , Estructura Terciaria de Proteína , RatasRESUMEN
Cytochrome b5 participates in electron-transfer reactions with a variety of different proteins. To explore how this protein might discern between structurally varied proteins, we have performed a molecular dynamics simulation focusing on its structural stability and dynamic behavior in solution. The protein was simulated in water at 298 K and pH 6.9 for 2.5 ns. The protein deviated significantly from the crystal structure midway through the simulation, but ultimately the crystalline conformation was regained. The simulation was at all times well behaved as judged by comparison to structural NMR data obtained in solution. One region of the protein backbone that deviated from the crystal conformation contains acidic residues implicated in electrostatic-based protein-protein recognition. The mobility in this region caused the protein to display different patterns of residues at the surface with time, as well as the formation of a large cleft partially exposing the hydrophobic core lining the heme pocket. Furthermore, the position and cyclical formation of this cleft suggest that hydrophobic interactions may be important in protein-protein recognition events and possibly even electron transfer, as the cleft allows for easy access to the heme group. These results indicate that thermal motion could provide a low-energy mechanism for controlling recognition events. Thus, the dynamical behavior observed through the varying solution conformations sampled may be important in influencing the diverse range of protein-protein interactions in which cytochrome b5 participates.