NMR characterization of a single-cysteine mutant of Escherichia coli thioredoxin and a covalent thioredoxin-peptide complex.

The mechanism of disulfide reduction by thioredoxin in the cell is thought to occur through the formation and subsequent destruction of a mixed-disulfide intermediate between thioredoxin and the substrate. In order to model the interaction, we have prepared a mutant of Escherichia coli thioredoxin where the second cysteine residue of the active site has been replaced by an alanine residue. A specific covalent complex has been prepared between the remaining cysteine residue and a short cysteine-containing peptide. This paper describes the preparation and characterization of the mutant protein both free and in the peptide complex.

Effects of buried charged groups on cysteine thiol ionization and reactivity in Escherichia coli thioredoxin: structural and functional characterization of mutants of Asp 26 and Lys 57.

To investigate the role of Asp 26 and Lys 57, two conserved, buried residues, in the redox mechanism of Escherichia coli thioredoxin (Trx), three mutant proteins, Asp 26 --> Ala (D26A), Lys 57 --> Met (K57M), and the double mutant D26A/K57M, were prepared, replacing the charged amino acids with hydrophobic residues with similar sizes. Both the oxidized (Trx-S2) and reduced [Trx-(SH)2] forms of the mutant thioredoxins are fully folded and similar in overall structure to the wild-type protein (wt). The structure of the active site hydrophobic surface is unchanged by the mutation of Asp 26 and Lys 57, since DNA polymerase activity in the 1:1 complex of the T7 gene 5 protein and mutant Trx-(SH)2 shows similar Kd values (approximately 5 nM) for both mutants and wt. In contrast, redox reactions involving thioredoxin as a catalyst of the reduction of disulfides or oxidation of dithiols are strongly affected by the mutations. In the reaction of Trx-S2 with thioredoxin reductase at pH 8.0, the kcat/Km value for the D26A mutant is decreased by a factor of 10 from that of wt, while the value for the D26A/K57M mutant is reduced 40-fold. The activity of Trx-(SH)2 as a protein disulfide reductase was measured with insulin, using fluorescence to detect oxidation of thioredoxin. At 15 degrees C and pH 8.0, both the D26A and K57M mutants showed 5--10-fold decreases in rates of reaction compared to those of the wild type, and the pH-rate profiles for the mutants were shifted 1 (K57M) and 2 (D26A) units to higher pH compared with the wt curve. NMR measurements for the three mutant proteins indicate that the proteins have the same global fold as that of the wild type, although changes in the chemical shifts of a number of resonances indicate local structural changes in the active site region. The resonances of oxidized D26A and D26A/K57M are pH-independent between pH 6.0 and 10.0, confirming the identification of the active site group titrating with a pKa of 7.5 in wt Trx-S2 as Asp 26. A profound change in the pKa of Asp 26, from 7.5 in the wild type to 9.4 in the mutant, is observed for K57M Trx-S2. The pH-dependent behavior of the resonances is affected in all mutant Trx-(SH)2 proteins. A single pKa shifted to higher values is observed on both the Cys 32 and Cys 35 Cbeta resonances. Ultraviolet absorbance measurements (A240) as a function of pH for wt Trx-(SH)2 demonstrate that the cysteine thiols titrate with apparent pK(a)s of about 7.1 and 9.9. The mutant proteins each show a single transition in the A240 measurements, with a midpoint at pH 7.8-8.0, consistent with the NMR results. The change in absorbance at 240 nm with increasing pH indicates that the number of thiols titrating in each mutant is greater than one but less than two. It is clear that both thiol pK(a)s have been significantly shifted by the mutations. The Cys 32 pKa is moved from 7.1 in wt to 7.8-8.0 in the mutants. The value of the Cys 35 pKa either is indistinguishable from that of Cys 32, thus accounting for more than one thiol titrating in the UV absorbance measurements or else is shifted to much higher pHs (> 10) where its transition is masked in both UV and NMR measurements by the effects of ionization of the tyrosine residues and unfolding of the protein. Our results strongly suggest that the buried Asp 26 carboxyl and Lys 57 epsilon-amino groups significantly affect the pK(a)s of the active site thiols, particularly that of the exposed low-pKa thiol Cys 32, thereby enhancing the rates of thiol-disulfide reactions at physiological pH.

Replacement of Trp28 in Escherichia coli thioredoxin by site-directed mutagenesis affects thermodynamic stability but not function.

Escherichia coli thioredoxin contains two tryptophan residues (Trp28 and Trp31) situated close to the active site disulfide/dithiol. In order to probe the structural and functional roles of tryptophan in the mechanism of E. coli thioredoxin (Trx), we have replaced Trp28 with alanine using site-directed mutagenesis and expressed the mutant protein W28A in E. coli. Changes in the behavior of the mutant protein compared with the wild-type protein have been monitored by a number of physical and spectroscopic techniques and enzyme assays. As expected, removal of a tryptophan residue causes profound changes in the fluorescence spectrum of thioredoxin, particularly for the reduced protein (Trx-(SH)2), and to a lesser extent for the oxidized protein (Trx-S2). These results show that the major contribution to the strongly quenched fluorescence of Trx-S2 in both wild-type and mutant proteins is from Trp31, whereas the higher fluorescence quantum yield of Trx-(SH)2 in the wild-type protein is dominated by the emission from Trp28. The fluorescence, CD, and 1H NMR spectra are all indicative that the mutant protein is fully folded at pH 7 and room temperature, and, despite the significance of the change, from a tryptophan in close proximity to the active site to an alanine, the functions of the protein appear to be largely intact. W28A Trx-S2 is a good substrate for thioredoxin reductase, and W28A Trx-(SH)2 is as efficient as wild-type protein in reduction of insulin disulfides. DNA polymerase activity exhibited by the complex of phage T7 gene 5 protein and Trx-(SH)2 is affected only marginally by the W28A substitution, consistent with the buried position of Trp28 in the protein. However, the thermodynamic stability of the molecule appears to have been greatly reduced by the mutation: guanidine hydrochloride unfolds the protein at a significantly lower concentration for the mutant than for wild type, and the thermal stability is reduced by about 10 degrees C in each case. The stability of each form of the protein appears to be reduced by the same amount, an indication that the effect of the mutation is identical in both forms of the protein. Thus, despite its close proximity to the active site, the Trp28 residue of thioredoxin is not apparently essential to the electron transfer mechanism, but rather contributes to the stability of the protein fold in the active site region.

Direct measurement of the aspartic acid 26 pKa for reduced Escherichia coli thioredoxin by 13C NMR.

Because of interference from the pH-dependent behavior of nearby groups in the active site of Escherichia coli thioredoxin, the pKa of the buried carboxyl group of the aspartic acid at position 26 has been difficult to quantitate. We report a direct measurement of this pKa using an NMR method utilizing the correlation between the C beta H proton resonances and the 13CO of the titrating carboxyl group. The experiments show unequivocally that the pKa is 7.3-7.5, rather than the value of 9 or greater recently proposed by Wilson, N. A., et al. [(1995) Biochemistry 34, 8931-8939]. The assignment of the titrating resonances to Asp 26 is unambiguous: the values of the C beta H chemical shifts correspond exactly to those of Asp 26, and their titration in the pH range 5.7-10.0 is the same as that observed previously for the proton resonances alone. In addition, the chemical shift of the carboxyl 13C resonance at pH 5.7 is upfield of those of the other carboxyl and carboxamide resonances, diagnostic for a protonated carboxyl group. The resonances assigned to Asp 26 are the only ones that titrate in the pH range 5.7-10.5. None of the other aspartate and glutamate residues in the molecule are titrated in this pH range, consistent with their positions on the surface of the molecule. The pKa measured for Asp 26 in reduced thioredoxin is identical within experimental error to that measured in the oxidized form of the protein. This is significant for the reductive mechanism of thioredoxin: the buried salt bridged/hydrogen-bonded side chains of Asp 26 and Lys 57 are likely to contribute to the facility of the reaction by providing a convenient source and sink for protons in the hydrophobic environment of the complex between thioredoxin and its substrates.

Proton sharing between cysteine thiols in Escherichia coli thioredoxin: implications for the mechanism of protein disulfide reduction.

Proton sharing between acidic groups has been observed in the active sites of several enzymes, including bacteriorhodopsin, aspartic proteases, and ribonuclease HI. We here report NMR observations suggestive of proton sharing between cysteine thiols in the active site of the oxidation-reduction enzyme thioredoxin. The pKas of the two cysteine thiols in the Escherichia coli protein are removed from the expected value of 8.4 by approximately 1 pH unit in either direction, upward and downward. Further, the C beta resonances of both residues show clearly the effects of both of these pKas, indicating that the titrations of the two thiol groups are intimately linked. This behavior strongly suggests that the low pKa ascribed to the deprotonation of the Cys 32 thiol most likely arises through the interaction and close approach of the thiol of Cys 35, with the thiolate anion of Cys 32 stabilized through the sharing of the remaining thiol proton, nominally attached to Cys 35. These observations provide a rationale for the mediation of active site pH control, an important aspect of the mechanism of thioredoxin and other proteins with catalytic thioredoxin domains, such as protein disulfide isomerases.

Comparison of the hydrogen-exchange behavior of reduced and oxidized Escherichia coli thioredoxin.

Hydrogen-deuterium exchange rates for the amide protons in oxidized (disulfide) and reduced (dithiol) thioredoxin have been measured using a series of 15N-1H HSQC spectra at various times after buffer exchange into 99% 2H2O. Information on exchange rates and protection factors was obtained for both forms of thioredoxin for 68 amide protons using this method; in general, the rates obtained by this method were for amide protons of residues in the hydrogen-bonded beta-sheet and alpha-helix secondary structure of thioredoxin. Estimates of the exchange rate for those amide protons that exchanged with rates too fast to measure by hydrogen--deuterium exchange were made by saturation-transfer measurements, which were particularly useful in defining the hydrogen exchange behavior of the active site Cys-Gly-Pro-Cys sequence and of the loops adjacent to it (residues 73-75 and 91-98). Amide proton exchange rates provide a qualitative estimate of the backbone mobility, and the differences in hydrogen exchange behavior between the two forms of thioredoxin are consistent with those observed in calculations of polypeptide chain dynamics obtained from 15N relaxation measurements [Stone, M. J., et al. (1993) Biochemistry 32, 426-435]. For most of the protein, the exchange rates are close to identical in the two forms, consistent with their very close similarity in structure and backbone dynamics. Significant differences in behavior are observed in the active site sequence and in the regions of the protein that are close to this sequence in the three-dimensional structure, including portions of the beta-strand and alpha-helical sequences immediately adjacent to the active site.(ABSTRACT TRUNCATED AT 250 WORDS)

High-resolution solution structures of oxidized and reduced Escherichia coli thioredoxin.

BACKGROUND: Thioredoxin participates in thiol-disulfide exchange reactions and both oxidized thioredoxin (disulfide form) and reduced thioredoxin (dithiol form) are found under physiological conditions. Previous structural studies suggested that the two forms were extremely similar, although significant functional and spectroscopic differences exist. We therefore undertook high-resolution solution structural studies of the two forms of Escherichia coli thioredoxin in order to detect subtle conformational differences. RESULTS: The solution structures of reduced and oxidized thioredoxin are extremely similar. Backbone structure is largely identical in the two forms, with slight differences in the region of the active site, which includes Cys32 and Cys35. The side chain sulfur atom of Cys32 is tilted away from that of Cys35 in the reduced form of the protein to accommodate the increase in S-S distance that occurs upon reduction of the disulfide, but the chi 1 angles of the two cysteines remain the same in the two forms. CONCLUSIONS: Only subtle conformational changes occur upon changing the oxidation state of the active site cysteines, including the positions of some side chains and in hydrogen bonding patterns in the active site region. Functional differences between the two forms are probably therefore related to differences in local conformational flexibility in and near the active site loop.

Effect of disulfide bridge formation on the NMR spectrum of a protein: studies on oxidized and reduced Escherichia coli thioredoxin.

As a prelude to complete structure calculations of both the oxidized and reduced forms of Escherichia coli thioredoxin (M(r) 11,700), we have analyzed the NMR data obtained for the two proteins under identical conditions. The complete aliphatic 13C assignments for both oxidized and reduced thioredoxin are reported. Correlations previously noted between 13C chemical shifts and secondary structure are confirmed in this work, and significant differences are observed in the C beta and C gamma shifts between cis- and trans-proline, consistent with previous work that identifies this as a simple and unambiguous method of identifying cis-proline residues in proteins. Reduction of the disulfide bond in the active-site Cys32-Gly-Pro-Cys35 sequence causes changes in the 1H, 15N and 13C chemical shifts of residues close to the active site, some of them quite far distant in the amino acid sequence. Coupling constants, both backbone and side chain, show some differences between the two proteins, and the NOE connectivities and chemical shifts are consistent with small changes in the positions of several side chains, including the two tryptophan rings (Trp28 and Trp31). These results show that, consistent with the biochemical behavior of thioredoxin, there are minimal differences in backbone configuration between the oxidized and reduced forms of the protein.

Structural dynamics in an electron-transfer complex.

The dynamic behaviour of the complex of horse cytochrome c with cytochrome c peroxidase, an electron-transfer complex, was studied in solution by a hydrogen exchange labelling method together with two-dimensional NMR analysis. Although cytochrome c hydrogens in the expected binding region exhibit slowed exchange, the measured slowing factors are very small, indicating that hydrogen-exchange occurs with little hindrance from within the binding interface. The complex in solution must therefore be highly mobile rather than rigidly defined, as implied by the crystalline complex. This result is in conflict with the concept that biological electron transfer occurs by way of predetermined covalent pathways.

Characterization by 1H NMR of a C32S,C35S double mutant of Escherichia coli thioredoxin confirms its resemblance to the reduced wild-type protein.

A mutant of Escherichia coli thioredoxin containing serine residues in place of the two active-site cysteines, termed C32S,C35S, previously shown to be partially able to substitute for reduced thioredoxin in certain phage systems, has been characterized by 1H NMR spectroscopy at pH values between 5.5 and 10. The 1H NMR spectrum of the mutant at pH 5.5 is very similar to that of the wild-type protein in either the reduced or oxidized state. Chemical shift changes in the vicinity of the active site serines indicate that the nearby hydrophobic pocket is somewhat changed, probably as a result of the replacement of the cysteine thiols with the smaller, more hydrophilic hydroxyl side chains and a change in the preferred chi 1 angles of the side chains. Although the pattern of amide protons persistent in 2H2O differs only slightly between the two forms of the wild-type protein, the pattern observed for the C32S,C35S mutant shows characteristic features that correspond closely with those of the reduced wild-type protein rather than with the oxidized form. The pH dependence of the mutant protein shows a single group titrating close to the active site with a pKa of 8.3, which we assign to the buried carboxyl group of Asp 26 by analogy with the behavior of wild-type thioredoxin. The pKa is significantly higher for the mutant protein, consistent with an increase in the hydrophobicity of the pocket where the carboxyl is buried, probably due to repacking caused by the removal of the cysteine thiols and the placement of the serine hydroxyls in positions where they interact better with solvent. The results demonstrate that the solution behavior of the mutant protein is similar in many ways to that of reduced wild-type thioredoxin, explaining its partial activity in the two essential roles of reduced thioredoxin as a subunit of phage T7 DNA polymerase and in the assembly of filamentous phage.

Isotope effects in peptide group hydrogen exchange.

Kinetic and equilibrium isotope effects in peptide group hydrogen exchange reactions were evaluated. Unlike many other reactions, kinetic isotope effects in amide hydrogen exchange are small because exchange pathways are not limited by bond-breaking steps. Rate constants for the acid-catalyzed exchange of peptide group NH, ND, and NT in H2O are essentially identical, but a solvent isotope effect doubles the rate in D2O. Rate constants for base-catalyzed exchange in H2O decrease slowly in the order NH > ND > NT. The alkaline rate constant in D2O is very close to that in H2O when account is taken of the glass electrode pH artifact and the difference in solvent ionization constant. Small equilibrium isotope effects lead to an excess equilibrium accumulation of the heavier isotopes by the peptide group. Results obtained are expressed in terms of rate constants for the random coil polypeptide, poly-DL-alanine, to provide reference rates for protein hydrogen exchange studies as described in Bai et al. [preceding paper in this issued].

Stable submolecular folding units in a non-compact form of cytochrome c.

Studies of structure, dynamics, and stability of cytochrome c (cyt c) at low pH in a non-compact pre-molten globule state indicate that the protein contains submolecular folding units that are independently stable. In high salt, acid cyt c (pD 2.2; where D is deuterium) is nearly as compact as the native form. Nuclear magnetic resonance (n.m.r.) line broadening typical of the molten globule form is seen, indicating loosened packing and increased mobility not only for side-chains but also for the main chain. As NaCl concentration is decreased below 0.05 M, cyt c expands due to the deshielding of electrostatic repulsions, attaining a linear extent perhaps double that of the native protein (viscosity, fluorescence). In the extended form, tertiary structural hydrogen bonds are largely broken (hydrogen exchange rate), some normally buried parts of the protein are exposed to water (fluorescence), and many of the native side-chain contacts must be lost. Nevertheless, almost all of the helical content is retained (circular dichroism). The helices involve the same amino acid residues that are helical in the native state (hydrogen exchange labeling monitored by 2-dimensional n.m.r.). The equilibrium constant for helix formation at 20 degrees C (0.02 M-NaCl, pD 2.2) is about 10 (hydrogen exchange rate), even though the individual helical segments when isolated have little or no structure. Additional experiments were done to check assumptions and calibrate parameters that underlie the hydrogen exchange analysis of protein folding. These results indicate that the native-like helical segments in the expanded non-globular form of cyt c exist as part of somewhat larger submolecular folding units that possess significant equilibrium stability. Results from equilibrium and kinetic studies of protein folding support the generality of this conclusion. This view is contrary to the two-state paradigm for equilibrium folding and inconsistent with the idea that side-chain packing constraints determine folding motifs. The result suggests an extension of the thermodynamic hypothesis for protein structure to kinetic folding processes, so that the amino acid code for equilibrium and kinetic folding may be the same, and also seems pertinent to the biological evolution of contemporary protein structures.

Structural description of acid-denatured cytochrome c by hydrogen exchange and 2D NMR.

Hydrogen exchange and two-dimensional nuclear magnetic resonance (2D NMR) techniques were used to characterize the structure of oxidized horse cytochrome c at acid pH and high ionic strength. Under these conditions, cytochrome c is known to assume a globular conformation (A state) with properties resembling those of the molten globule state described for other proteins. In order to measure the rate of hydrogen-deuterium exchange for individual backbone amide protons in the A state, samples of oxidized cytochrome c were incubated at 20 degrees C in D2O buffer (pD 2.2, 1.5 M NaCl) for time periods ranging from 2 min to 500 h. The exchange reaction was then quenched by transferring the protein to native conditions (pD 5.3). The extent of exchange for 44 amide protons trapped in the refolded protein was measured by 2D NMR spectroscopy. The results show that this approach can provide detailed information on H-bonded secondary and tertiary structure in partially folded equilibrium forms of a protein. All of the slowly exchanging amide protons in the three major helices of native cytochrome c are strongly protected from exchange at acid pH, indicating that the A state contains native-like elements of helical secondary structure. By contrast, a number of amide protons involved in irregular tertiary H-bonds of the native structure (Gly37, Arg38, Gln42, Ile57, Lys79, and Met80) are only marginally protected in the A state, indicating that these H-bonds are unstable or absent. The H-exchange results suggest that the major helices of cytochrome c and their common hydrophobic domain are largely preserved in the globular acidic form while the loop region of the native structure is flexible and partly disordered.