Crystal structure of a complex of Escherichia coli glycerol kinase and an allosteric effector fructose 1,6-bisphosphate.

The three-dimensional structures of Escherichia coli glycerol kinase (GK) with bound glycerol in the presence and absence of one of the allosteric regulators of its activity, fructose 1,6-bisphosphate (FBP), at 3.2 and 3.0 A, are presented. The molecule crystallized in space group P41212, and the structure was solved by molecular replacement. The models were refined with good stereochemistry to final R-factors of 21.1 and 21.9%, respectively. A tetrameric arrangement of monomers was observed which was essentially identical to the proposed inactive tetramer II previously described [Feese, M. D., Faber, H. R., Bystrom, C. E., Pettigrew, D. W., and Remington, S. J. (1998) Structure (in press)]. However, the crystal packing in this form was especially open, permitting the FBP binding site to be occupied and identified. The crystallographic data revealed a most unusual type of FBP binding site formed between two glycine-arginine loops (residues 234-236) where one-half of the binding site is donated by each monomer at the regulatory interface. The molecule of FBP binds in two mutually exclusive modes on a noncrystallographic 2-fold axis at 50% occupancy each; thus, a tetramer of GK binds two molecules of FBP. Ionic interactions between the 1- and 6-phosphates of FBP and Arg 236 were observed in addition to hydrogen bonding interactions between the backbone amide of Gly 234 and the 6-phosphate. No contacts between the protein and the furanose ring were observed. Mutagenesis of Arg 236 to alanine drastically reduced the extent of inhibition of GK by FBP and lowered, but did not eliminate, the ability of FBP to promote tetramer association. These observations are consistent with the previously characterized mechanism of FBP inhibition of GK, in which FBP acts both to promote dimer-tetramer assembly and to inactivate the tetramers.

Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein.

The 2.1-A resolution crystal structure of wild-type green fluorescent protein and comparison of it with the recently determined structure of the Ser-65 --> Thr (S65T) mutant explains the dual wavelength absorption and photoisomerization properties of the wild-type protein. The two absorption maxima are caused by a change in the ionization state of the chromophore. The equilibrium between these states appears to be governed by a hydrogen bond network that permits proton transfer between the chromophore and neighboring side chains. The predominant neutral form of the fluorophore maximally absorbs at 395 nm. It is maintained by the carboxylate of Glu-222 through electrostatic repulsion and hydrogen bonding via a bound water molecule and Ser-205. The ionized form of the fluorophore, absorbing at 475 nm, is present in a minor fraction of the native protein. Glu-222 donates its charge to the fluorophore by proton abstraction through a hydrogen bond network, involving Ser-205 and bound water. Further stabilization of the ionized state of the fluorophore occurs through a rearrangement of the side chains of Thr-203 and His-148. UV irradiation shifts the ratio of the two absorption maxima by pumping a proton relay from the neutral chromophore's excited state to Glu-222. Loss of the Ser-205-Glu-222 hydrogen bond and isomerization of neutral Glu-222 explains the slow return to the equilibrium dark-adapted state of the chromophore. In the S65T structure, steric hindrance by the extra methyl group stabilizes a hydrogen bonding network, which prevents ionization of Glu-222. Therefore the fluorophore is permanently ionized, causing only a 489-nm excitation peak. This new understanding of proton redistribution in green fluorescent protein should enable engineering of environmentally sensitive fluorescent indicators and UV-triggered fluorescent markers of protein diffusion and trafficking in living cells.

The Cys292-->Ala substitution in protein R1 of class I ribonucleotide reductase from Escherichia coli has a global effect on nucleotide binding at the specificity-determining allosteric site.

Ribonucleotide reductase from aerobically grown Escherichia coli is allosterically regulated, both with respect to general activity and substrate specificity. Protein R1, the homodimeric enzyme component which harbours binding sites for allosteric effectors (nucleoside triphosphates) as well as substrates (ribonucleoside diphosphates), has been engineered at Cys292 close to the dimer interaction area. This residue was earlier shown to be specifically photoaffinity labelled with the allosteric nucleotide dTTP. In this study the effect of the Cys292-->Ala substitution is shown to be an overall diminished nucleotide binding at the specificity site reflected in Kd values for dTTP, dGTP and dATP higher by more than one order of magnitude with respect to wild type. The mutant protein's interaction with other protein components of the ribonucleotide reductase system was unaffected by the mutation. These results show that Cys292 in protein R1 of class I ribonucleotide reductase from E. coli is located in the allosteric specificity site.

Crystal structure of the Aequorea victoria green fluorescent protein.

The green fluorescent protein (GFP) from the Pacific Northwest jellyfish Aequorea victoria has generated intense interest as a marker for gene expression and localization of gene products. The chromophore, resulting from the spontaneous cyclization and oxidation of the sequence -Ser65 (or Thr65)-Tyr66-Gly67-, requires the native protein fold for both formation and fluorescence emission. The structure of Thr65 GFP has been determined at 1.9 angstrom resolution. The protein fold consists of an 11-stranded beta barrel with a coaxial helix, with the chromophore forming from the central helix. Directed mutagenesis of one residue adjacent to the chromophore, Thr203, to Tyr or His results in significantly red-shifted excitation and emission maxima.

Residues important for radical stability in ribonucleotide reductase from Escherichia coli.

The R2 protein of ribonucleotide reductase contains at the side chain of tyrosine 122 a stable free radical, which is essential for enzyme catalysis. The tyrosyl radical is buried in the protein matrix close to a dinuclear iron center and a cluster of three hydrophobic residues (Phe-208, Phe-212, and Ile-234) conserved throughout the R2 family. A key step in the generation of the tyrosyl radical is the activation of molecular oxygen at the iron center. It has been suggested that the hydrophobic cluster provides an inert binding pocket for molecular oxygen bound to the iron center and that it may play a role in directing the oxidative power of a highly reactive intermediate toward tyrosine 122. We have tested these hypotheses by constructing the following mutant R2 proteins:F208Y, F212Y, F212W, and I234N. The resulting mutant proteins all have the ability to form a tyrosine radical, which indicates that binding of molecular oxygen can occur. In the case of F208Y, the yield of tyrosyl radical is substantially lower than in the wild-type case. A competing reaction resulting in hydroxylation of Tyr-208 implies that the phenylalanine at position 208 may influence the choice of target for electron abstraction. The most prominent result is that all mutant proteins show impaired radical half-life; in three of the four mutants, the half-lives are several orders of magnitude shorter than that of the wild-type radical. This suggests that the major role of the hydrophobic pocket is to stabilize the tyrosyl radical. This hypothesis is corroborated by comparative studies of the environment of other naturally occurring tyrosyl radicals.

The conserved serine 211 is essential for reduction of the dinuclear iron center in protein R2 of Escherichia coli ribonucleotide reductase.

The R2 protein family of class I ribonucleotide reductases contains a highly conserved serine residue close to the essential tyrosyl radical and the dinuclear iron center. In order to test its physiological importance, we have engineered the Ser-211 of Escherichia coli R2 to an alanine and a cysteine residue. The three-dimensional structure of R2 S211A solved to 2.4-A resolution is virtually identical to the wild-type structure apart from the substituted residue. Both mutant proteins contain oxidized dinuclear iron and tyrosyl radical, and their specific enzyme activity per radical are comparable to that of the wild-type protein. In R2 S211A the stability of the tyrosyl radical is substantially decreased, probably caused by movement of Gln-80 into hydrogen bonding distance of Tyr-122. The major defect in R2 S211A, however, is the inability of its iron center to be reduced by enzymic or chemical means, a characteristic not found in R2 S211C. We propose that Ser-211 is needed as a proton donor/transporter during reduction of the iron center of R2, a reaction which in vivo precedes reconstitution of the tyrosyl radical. This offers a physiological explanation for the high conservation of a serine residue at this position in the R2 family.

Autocatalytic generation of dopa in the engineered protein R2 F208Y from Escherichia coli ribonucleotide reductase and crystal structure of the dopa-208 protein.

The mutant form Phe-208-->Tyr of the R2 protein of Escherichia coli ribonucleotide reductase contains an intrinsic ferric-Dopa cofactor with characteristic absorption bands at 460 and ca. 700 nm [Ormö, M., de Maré, F., Regnström, K., Aberg, A., Sahlin, M., Ling, J., Loehr, T. M., Sanders-Loehr, J., & Sjöberg, B. M. (1992) J. Biol. Chem. 267, 8711-8714]. The three-dimensional structure of the mutant protein, solved to 2.5-A resolution, shows that the Dopa is localized to residue 208 and that it is a bidentate ligand of Fe1 of the binuclear iron center of protein R2. Nascent apoR2 F208Y, lacking metal ions, can be purified from overproducing cells grown in iron-depleted medium. ApoR2 F208Y is rapidly and quantitatively converted to the Dopa-208 form in vitro by addition of ferrous iron in the presence of oxygen. Other metal ions (Cu2+, Mn2+, Co2+) known to bind to the metal site of wild-type apoR2 do not generate a Dopa in apoR2 F208Y. The autocatalytic generation of Dopa does not require the presence of a tyrosine residue at position 122, the tyrosine which in a wild-type R2 protein acquires the catalytically essential tyrosyl radical. It is proposed that generation of Dopa initially follows the suggested reaction mechanism for tyrosyl radical generation in the wild-type protein and involves a ferryl intermediate, which in the case of the mutant R2 protein oxygenates Tyr 208. This autocatalytic metal-mediated reaction in the engineered R2 F208Y protein may serve as a model for formation of covalently bound quinones in other proteins.

Correlation between active form and dimeric structure of mitochondrial nicotinamide nucleotide transhydrogenase from beef heart.

The active form of purified mitochondrial nicotinamide nucleotide transhydrogenase from beef heart was investigated by crosslinking with dimethylsuberimidate and SDS-PAGE, with or without pretreatment with the inactivating detergent Triton X-100. In the absence of detergent, crosslinked isomers of the dimeric form of 208-235 kDa were obtained. Addition of detergent led to the simultaneous loss of the dimers and the bulk of the activity. Removal of the detergent led to a partial restoration of both activity and the dimeric forms. The results suggest that the active form is a dimer, and that the detergent-dependent conversion to the largely inactive monomer is reversible. It is proposed that the mechanism of inactivation of transhydrogenase by Triton X-100 involves a disruption of essential hydrophobic interactions between the membrane-spanning regions of the monomers.

Engineering of the iron site in ribonucleotide reductase to a self-hydroxylating monooxygenase.

Protein R2 of ribonucleotide reductase contains a dinuclear ferric iron center adjacent to a tyrosyl radical in the interior of the protein matrix. A patch of hydrophobic residues surrounds the iron-radical cofactor. Its importance during the oxidative generation of the iron-radical cofactor was investigated by site-directed mutagenesis of Phe-208 to tyrosine. The mutant protein R2 F208Y has prominent absorption bands at 460 and 720 nm reminiscent of those in ferric-catecholate complexes. Resonance Raman spectroscopy shows that the iron center of R2 F208Y contains a bidentate catechol ligand. The mechanism for generation of this protein-derived dihydroxyphenylalanine may be similar to the catalytic cycle of methane monooxygenase.

An ultrafiltration assay for nucleotide binding to ribonucleotide reductase.

Direct partition through ultrafiltration was applied to develop a method for the study of nucleotide binding to ribonucleotide reductase from Escherichia coli. The assay involved a 0.5- to 1-min centrifugation step where bound and unbound nucleotides are separated over an ultrafiltration membrane. No effects were seen due to hyperconcentration of protein at the membrane surface. The method was verified by measuring binding of dATP, ATP, dTTP, dGTP, and GDP at 25 and 4 degrees C with dissociation constants ranging from 0.1 to 80 microM. The results were in good agreement with earlier data obtained by other techniques and extend our knowledge in the case of ATP and dGTP binding at 25 degrees C.

Evidence for two different classes of redox-active cysteines in ribonucleotide reductase of Escherichia coli.

The large subunit of ribonucleotide reductase from Escherichia coli contains redox-active cysteine residues. In separate experiments, five conserved and 2 nonconserved cysteine residues were substituted with alanines by oligonucleotide-directed mutagenesis. The activities of the mutant proteins were determined in the presence of three different reductants: thioredoxin, glutaredoxin, or dithiothreitol. The results indicate two different classes of redox-active cysteines in ribonucleotide reductase: 1) C-terminal Cys-754 and Cys-759 responsible for the interaction with thioredoxin and glutaredoxin; and 2) Cys-225 and Cys-439 located at the nucleotide-binding site. Our classification of redox-active cysteines differs from the location of the active site cysteines in E. coli ribonucleotide reductase suggested previously (Lin, A.-N. I., Ashley, G. W., and Stubbe, J. (1987) Biochemistry 26, 6905-6909).