Modeling of the bacterial luciferase-flavin mononucleotide complex combining flexible docking with structure-activity data

Although the crystal structure of Vibrio harveyi luciferase has been elucidated, the binding sites for the flavin mononucleotide and fatty aldehyde substrates are still unknown. The determined location of the phosphate-binding site close to Arg 107 on the alpha subunit of luciferase is supported here by point mutagenesis. This information, together with previous structure-activity data for the length of the linker connecting the phosphate group to the isoalloxazine ring represent important characteristics of the luciferase-bound conformation of the flavin mononucleotide. A model of the luciferase-flavin complex is developed here using flexible docking supplemented by these structural constraints. The location of the phosphate moiety was used as the anchor in a flexible docking procedure performed by conformation search by using the Monte Carlo minimization approach. The resulting databases of energy-ranked feasible conformations of the luciferase complexes with flavin mononucleotide, omega-phosphopentylflavin, omega-phosphobutylflavin, and omega-phosphopropylflavin were filtered according to the structure-activity profile of these analogs. A unique model was sought not only on energetic criteria but also on the geometric requirement that the isoalloxazine ring of the active flavin analogs must assume a common orientation in the luciferase-binding site, an orientation that is also inaccessible to the inactive flavin analog. The resulting model of the bacterial luciferase-flavin mononucleotide complex is consistent with the experimental data available in the literature. Specifically, the isoalloxazine ring of the flavin mononucleotide interacts with the Ala 74-Ala 75 cis-peptide bond as well as with the Cys 106 side chain in the alpha subunit of luciferase. The model of the binary complex reveals a distinct cavity suitable for aldehyde binding adjacent to the isoalloxazine ring and flanked by other key residues (His 44 and Trp 250) implicated in the active site.     

Implications of the reactive thiol and the proximal non-proline cis-peptide bond in the Structure and function of Vibrio harveyi luciferase

The role of a highly reactive cysteine residue, Cys106, in Vibrio harveyi luciferase in modulating the substrate-enzyme interactions and in turn affecting the enzyme activity has been extensively investigated over the past three decades. Replacing Cys106 with valine dramatically hinders the ability of luciferase to stabilize the C4a-hydroperoxyflavin intermediate [Abu-Soud, H. M., Clark, A. C., Francisco, W. A., Baldwin, T. O., and Raushel, F. M. (1993) J. Biol. Chem. 268, 7699-7706] and consume aldehyde substrate [Xi, L., Cho, K.-W., Herndon, M. E., and Tu, S.-C. (1990) J. Biol. Chem. 265, 4200-4203], therefore markedly decreasing enzyme activity. On the basis of the structure-activity relationship of flavin analogues and the location of the phosphate binding site of flavin mononucleotide (FMN) coupled with molecular modeling, the functional part of the isoalloxazine ring of FMN, the thiol side chain of Cys106, the methyl group of Ala75, and the unique non-prolyl cis-peptide bond between Ala74 and Ala75 were found to be closely packed [Lin, L. Y., Sulea, T., Szittner, R., Vassilyev, V., Purisima, E. O., and Meighen, E. A. (2001) Protein Sci. 10, 1563-1571]. Here, by mutating Ala75 to Gly, we restored key wild-type properties to the C106V mutant, in particular, high enzyme activity and a stable C4a-hydroperoxyflavin intermediate, demonstrating that the primary reason for the dark phenotype of the C106V mutant was the unfavorable steric interaction between Val106 and Ala75 side chains, which could in turn disturb the cis-oriented amide linkage of Ala74 and Ala75. Moreover, significant red shifts in light emission of 3-10 nm were measured for luciferases carrying Val106 with the spectrum of the double mutant C106V/A75G now red shifted to that of Photobacterium phosphoreum luciferase, which also has Val and Gly at positions 106 and 75, respectively. These results strengthen the validity of the binding geometry of the modeled flavin with the re-face of the pyrimidine end of the isoalloxazine ring next to Cys106 and implicate the Ala74-Ala75 cis-peptide as a key component in the bioluminescence reaction.     

Structure of the flavin adduct formed in the suicide reaction of alpha-hydroxybutynoate with D-lactate dehydrogenase.

The Zn-dependent flavoenzyme D-lactate dehydrogenase from Megasphaera elsdenii is irreversibly inactivated by the D form of the suicide substrate 2-hydroxy-3-butynoic acid. The process of inactivation involves formation of a new pink chromophore, which can be released in intact form from the protein and which was purified to homogeneity by affinity chromatography. Inactivation involves covalent addition of the suicide substrate to the flavin coenzyme. The optical spectra indicate an elongation of the flavin chromophore, and the chemical reactivity suggests a derivative of reduced flavin. The structure of this adduct was deduced from Fourier transform NMR, from the chemical properties, and from comparison with appropriate models, which were synthesized chemically. This structure involves the covalent linkage of the acetylenic inhibitor to positions N(5) and C(6) of the flavin coenzyme via carbon atoms 2 and 4 of the inhibitor to form an additional fused aromatic ring. The pink adduct can be reconverted to an isoalloxazine chromophore by reduction with borohydride and subsequent reoxidation with oxygen. This new isoalloxazine has the spectral properties of an isoflavin, and it is proposed to carry the moiety of the inactivator molecule as substituent at position C(6). The structure of the pink chromophore representing a cyclic adduct to the flavin positions N(5) and C(6) is compared to that of the adduct obtained from L-lactate oxidase from Mycobacterium smegmatis and the L form of the same inhibitor [C(4a)--N(5) cyclic adduct; Schonbrunn, A., Abeles, R. H., Walsh, C. T., Ghisla, S., Ogata, H., and Massey, V. (1976) Biochemistry 15, 1978]. This comparison allows deductions about the relative orientation of substrate, coenzyme, and active center functional groups in the two enzymes.     

Effects of mutations of the alpha His45 residue of Vibrio harveyi luciferase on the yield and reactivity of the flavin peroxide intermediate

This work was undertaken to investigate the functional consequences of mutations of the essential alpha His45 residue of Vibrio harveyi luciferase, especially with respect to the yield and reactivity of the flavin 4a-hydroperoxide intermediate II. A total of 14 luciferase variants, each with a different single-residue replacement for the alpha His45, were examined. These variants showed changes, mostly slight, in their light decay rates of the nonturnover luminescence reaction and in their Km values for decanal and reduced riboflavin 5'-phosphate (FMNH2). All alpha His45 mutants, however, showed markedly reduced bioluminescence activities, the magnitude of the reduction ranging from about 300-fold to 6 orders of magnitude. Remarkably, a good correlation was obtained for the wild-type luciferase, 12 alpha His45-mutated luciferases, and six additional variants with mutations of other alpha-subunit histidine residues between the degrees of luminescence activity reduction and the dark decay rates of intermediate II. Such a correlation further indicates that the activation of the O-O bond fission is an important function of the flavin 4a-hydroperoxide intermediate II. Both alpha H45G and alpha H45W were found to bind near-stoichiometric amounts of FMNH2. Moreover, each variant catalyzed the oxidation of bound FMNH2 by two mechanisms, with a minor pathway leading to the formation of a luminescence-active intermediate II and a major dark pathway not involving any detectable flavin 4a-hydroperoxide species. This latter pathway mimics that in the normal catalysis by flavooxidases, and its elicitation in luciferase was demonstrated for the first time by single-residue mutations.     

Functional roles of conserved residues in the unstructured loop of Vibrio harveyi bacterial luciferase

Residues 257-291 of the Vibrio harveyi bacterial luciferase alpha subunit comprise a highly conserved, protease-labile, disordered loop region, most of which is unresolved in the previously determined X-ray structures of the native enzyme. This loop region has been shown to display a time- dependent proteolysis resistance upon single catalytic turnover and was postulated to undergo conformational changes during catalysis ([AbouKhair, N. K., Ziegler, M. M., and Baldwin, T. O. (1985) Biochemistry 24, 3942-3947]. To investigate the role of this region in catalysis, we have performed site-specific mutations of different conserved loop residues. In comparison with V(max) and V(max)/K(m,flavin) of the native luciferase, the bioluminescence activities of alphaG284P were decreased to 1-2% whereas those of alphaG275P and alphaF261D were reduced by 4-6 orders of magnitude. Stopped-flow results indicate that both alphaG275P and alphaF261D were able to form the 4a-hydroperoxy-FMN intermediate II but at lower yields. Both mutants also had enhanced rates for the intermediate II nonproductive dark decay and significantly compromised abilities to oxidize the decanal substrate. Additional mutations were introduced into the alphaG275 and alphaF261 positions, and the activities of the resulting mutants were characterized. Results indicate that the torsional flexibility of the alphaG275 residue and the bulky and hydrophobic nature of the alphaF261 residue were critical to the luciferase activity. Our results also support a functional role for the alpha subunit unstructured loop itself, possibly by serving as a mobile gating mechanism in shielding critical intermediates (including the excited flavin emitter) from exposure to medium.     

Active site hydrophobicity is critical to the bioluminescence activity of Vibrio harveyi luciferase

Vibrio harveyi luciferase is an alphabeta heterodimer containing a single active site, proposed earlier to be at a cleft in the alpha subunit. In this work, six conserved phenylalanine residues at this proposed active site were subjected to site-directed mutations to investigate their possible functional roles and to delineate the makeup of luciferase active site. After initial screening of Phe --> Ala mutants, alphaF46, alphaF49, alphaF114, and alphaF117 were chosen for additional mutations to Asp, Ser, and Tyr. Comparisons of the general kinetic properties of wild-type and mutated luciferases indicated that the hydrophobic nature of alphaF46, alphaF49, alphaF114, and alphaF117 was important to luciferase V(max) and V(max)/K(m), which were reduced by 3-5 orders of magnitude for the Phe --> Asp mutants. Both alphaF46 and alphaF117 also appeared to be involved in the binding of reduced flavin substrate. Additional studies on the stability and yield of the 4a-hydroperoxyflavin intermediate II and measurements of decanal substrate oxidation by alphaF46D, alphaF49D, alphaF114D, and alphaF117D revealed that their marked reductions in the overall quantum yield (phi( degrees )) were a consequence of diminished yields of luciferase intermediates and, with the exception of alphaF114D, emission quantum yield of the excited emitter due to the replacement of the hydrophobic Phe by the anionic Asp. The locations of these four critical Phe residues in relation to other essential and/or hydrophobic residues are depicted in a refined map of the active site. Functional implications of these residues are discussed.     

Structural backgrounds for the formation of a catalytically competent complex with NADP(H) during hydride transfer in ferredoxin-NADP(+) reductases

The role of the highly conserved C266 and L268 of pea ferredoxin-NADP(+) reductase (FNR) in formation of the catalytically competent complex of the enzyme with NADP(H) was investigated. Previous studies suggest that the volume of these side-chains, situated facing the side of the C-terminal Y308 catalytic residue not stacking the flavin isoalloxazine ring, may be directly involved in the fine-tuning of the catalytic efficiency of the enzyme. Wild-type pea FNR as well as single and double mutants of C266 and L268 residues were analysed by fast transient-kinetic techniques and their midpoint reduction potentials were determined. For the C266A, C266M and C266A/L268A mutants a significant reduction in the overall hydride transfer (HT) rates was observed along with the absence of charge-transfer complex formation. The HT rate constants for NADPH oxidation were lower than those for NADP(+) reduction, reaching a 30-fold decrease in the double mutant. In agreement, these variants exhibited more negative midpoint potentials with respect to the wild-type enzyme. The three-dimensional structures of C266M and L268V variants were solved. The C266M mutant shows a displacement of E306 away from the relevant residue S90 to accommodate the bulky methionine introduced. The overall findings indicate that in FNR the volume of the residue at position 266 is essential to attain the catalytic architecture between the nicotinamide and isoalloxazine rings at the active site and, therefore, for an efficient HT process. In addition, flexibility of the 268-270 loop appears to be critical for FNR to achieve catalytically competent complexes with NADP(H).     

Mechanism of flavin mononucleotide cofactor binding to the Desulfovibrio vulgaris flavodoxin. 2. Evidence for cooperative conformational changes involving tryptophan 60 in the interaction between the phosphate- and ring-binding subsites

A mechanism has been proposed for the binding of flavin mononucleotide (FMN) and riboflavin to the apoflavodoxin from Desulfovibrio vulgaris [Murray, T. A., and Swenson, R. P. (2003) Biochemistry 42, 2307-2316]. In this model, the binding of the flavin isoalloxazine ring is dependent on the presence of a phosphate moiety in the phosphate-binding subsite, suggesting a cooperative interaction between that region and the ring-binding subsite. In the absence of inorganic phosphate, FMN can bind through the initial association of its 5'-phosphate group in the phosphate-binding subsite followed by insertion of the flavin ring. Because riboflavin lacks the 5'-phosphate group, it is unable to bind to this apoprotein in the absence of inorganic phosphate in solution. However, inorganic phosphate can enhance the rate of ring binding by occupying the phosphate-binding subsite. In this paper, NMR, near-UV circular dichroism (CD), and fluorescence spectroscopy provide evidence for a phosphate-induced conformational change within the isoalloxazine ring-binding subsite. Phosphate-dependent changes in the chemical shifts of 22 amide groups were observed in (1)H-(15)N HSQC NMR spectra. The majority of these groups are proximal to the phosphate-binding subsite or the loops that constitute the isoalloxazine ring-binding site. Also, a phosphate-dependent change in the environment or position of the Trp60 side chain was apparent in the NMR data and was confirmed by associated changes in the near-UV CD and tryptophan fluorescence spectra when compared to the spectra of the W60A mutant. These data suggest that phosphate, either the 5'-phosphate of the FMN or inorganic phosphate from solution, facilitates the movement of the side chain of Trp60 out of the isoalloxazine ring-binding site and other associated conformational changes, creating a population of apoflavodoxin that is capable of binding the isoalloxazine ring. This conformational switch may explain why some apoflavodoxins cannot bind riboflavin and also supports the "aromatic gate" model proposed from the crystal structure of the Anabaena apoflavodoxin [Genzor, C. G., Perales-Alcon, A., Sancho, J., and Romero, A. (1996) Nat. Struct. Biol. 3, 329-332].     

On the structure of flavin-oxygen intermediates involved in enzymatic reactions.

During the catalytic reactions of flavoprotein hydroxylases and bacterial luciferase, flavin peroxides are formed as intermediates [see Massey, V. and Hemmerich, P. (1976) in The Enzymes, 3rd edn (P. Boyer, ed.) pp. 421--505, Academic Press, New York]. These intermediates have been postulated to be C(4a) derivatives of the flavin coenzyme. To test this hypothesis, modified flavin coenzymes carrying an oxygen substituent at position C(4a) of the isoalloxazine ring were synthesized. They are tightly bound by the apoenzymes of D-amino acid oxidase, p-hydroxybenzoate hydroxylase and lactate oxidase; the resulting complexes show spectral properties closely similar to those of the transient oxygen adducts of the hydroxylases. The optical spectra of the lumiflavin model compounds were found to be highly dependent on the solvent environment and nature of the subsituents. Under appropriate conditions they simulate satisfactorily the spectra of the transient enzymatic oxygen adducts. The results support the proposal that the primary oxygen adducts formed with these flavoproteins on reaction of the reduced enzymes with oxygen are flavin C(4a) peroxides.     

Two Lysine Residues in the Bacterial Luciferase Mobile Loop Stabilize Reaction Intermediates

Bacterial luciferase catalyzes the reaction of FMNH₂, O₂, and a long chain aliphatic aldehyde, yielding FMN, carboxylic acid, and blue-green light. The most conserved contiguous region of the primary sequence corresponds to a crystallographically disordered loop adjacent to the active center (Fisher, A. J., Raushel, F. M., Baldwin, T. O., and Rayment, I. (1995) Biochemistry 34, 6581–6586; Fisher, A. J., Thompson, T. B., Thoden, J. B., Baldwin, T. O., and Rayment, I. (1996) J. Biol. Chem. 271, 21956–21968). Deletion of the mobile loop does not alter the chemistry of the reaction but decreases the total quantum yield of bioluminescence by 2 orders of magnitude (Sparks, J. M., and Baldwin, T. O. (2001) Biochemistry 40, 15436–15443). In this study, we attempt to localize the loss of activity observed in the loop deletion mutant to individual residues in the mobile loop. Using alanine mutagenesis, the effects of substitution at 15 of the 29 mobile loop residues were examined. Nine of the point mutants had reduced activity in vivo. Two mutations, K283A and K286A, resulted in a loss in quantum yield comparable with that of the loop deletion mutant. The bioluminescence emission spectrum of both mutants was normal, and both yielded the carboxylic acid chemical product at the same efficiency as the wild-type enzyme. Substitution of Lys²⁸³ with alanine resulted in destabilization of intermediate II, whereas mutation of Lys²⁸⁶ had an increase in exposure of reaction intermediates to a dynamic quencher. Based on a model of the enzyme-reduced flavin complex, the two critical lysine residues are adjacent to the quininoidal edge of the isoalloxazine.     

Structural backgrounds for the formation of a catalytically competent complex with NADP(H) during hydride transfer in ferredoxin–NADP+ reductases

The role of the highly conserved C266 and L268 of pea ferredoxin–NADP⁺ reductase (FNR) in formation of the catalytically competent complex of the enzyme with NADP(H) was investigated. Previous studies suggest that the volume of these side-chains, situated facing the side of the C-terminal Y308 catalytic residue not stacking the flavin isoalloxazine ring, may be directly involved in the fine-tuning of the catalytic efficiency of the enzyme. Wild-type pea FNR as well as single and double mutants of C266 and L268 residues were analysed by fast transient-kinetic techniques and their midpoint reduction potentials were determined. For the C266A, C266M and C266A/L268A mutants a significant reduction in the overall hydride transfer (HT) rates was observed along with the absence of charge-transfer complex formation. The HT rate constants for NADPH oxidation were lower than those for NADP⁺ reduction, reaching a 30-fold decrease in the double mutant. In agreement, these variants exhibited more negative midpoint potentials with respect to the wild-type enzyme. The three-dimensional structures of C266M and L268V variants were solved. The C266M mutant shows a displacement of E306 away from the relevant residue S90 to accommodate the bulky methionine introduced. The overall findings indicate that in FNR the volume of the residue at position 266 is essential to attain the catalytic architecture between the nicotinamide and isoalloxazine rings at the active site and, therefore, for an efficient HT process. In addition, flexibility of the 268–270 loop appears to be critical for FNR to achieve catalytically competent complexes with NADP(H).     

Cytochrome b5 reductase: role of the si-face residues, proline 92 and tyrosine 93, in structure and catalysis

The conserved sequence motif "RxY(T)(S)xx(S)(N)" coordinates flavin binding in NADH:cytochrome b(5) reductase (cb(5)r) and other members of the flavin transhydrogenase superfamily of oxidoreductases. To investigate the roles of Y93, the third and only aromatic residue of the "RxY(T)(S)xx(S)(N)" motif, that stacks against the si-face of the flavin isoalloxazine ring, and P92, the second residue in the motif that is also in close proximity to the FAD moiety, a series of rat cb(5)r variants were produced with substitutions at either P92 or Y93, respectively. The proline mutants P92A, G, and S together with the tyrosine mutants Y93A, D, F, H, S, and W were recombinantly expressed in E. coli and purified to homogeneity. Each mutant protein was found to bind FAD in a 1:1 cofactor:protein stoichiometry while UV CD spectra suggested similar secondary structure organization among all nine variants. The tyrosine variants Y93A, D, F, H, and S exhibited varying degrees of blue-shift in the flavin visible absorption maxima while visible CD spectra of the Y93A, D, H, S, and W mutants exhibited similar blue-shifted maxima together with changes in absorption intensity. Intrinsic flavin fluorescence was quenched in the wild type, P92S and A, and Y93H and W mutants while Y93A, D, F, and S mutants exhibited increased fluorescence when compared to free FAD. The tyrosine variants Y93A, D, F, and S also exhibited greater thermolability of FAD binding. The specificity constant (k(cat)/K(m)(NADH)) for NADH:FR activity decreased in the order wild type > P92S > P92A > P92G > Y93F > Y93S > Y93A > Y93D > Y93H > Y93W with the Y93W variant retaining only 0.5% of wild-type efficiency. Both K(s)(H4NAD) and K(s)(NAD+) values suggested that Y93A, F, and W mutants had compromised NADH and NAD(+) binding. Thermodynamic measurements of the midpoint potential (E degrees ', n = 2) of the FAD/FADH(2) redox couple revealed that the potentials of the Y93A and S variants were approximately 30 mV more positive than that of wild-type cb(5)r (E degrees ' = -268 mV) while that of Y93H was approximately 30 mV more negative. These results indicate that neither P92 nor Y93 are critical for flavin incorporation in cb(5)r and that an aromatic side chain is not essential at position 93, but they demonstrate that Y93 forms contacts with the FAD that effectively modulate the spectroscopic, catalytic, and thermodynamic properties of the bound cofactor.     

Probing the agonist domain of the nicotinic acetylcholine receptor by cysteine scanning mutagenesis reveals residues in proximity to the alpha-bungarotoxin binding site

We have constructed a series of cysteine-substitution mutants in order to identify residues in the mouse muscle nicotinic acetylcholine receptor (AChR) that are involved in alpha-bungarotoxin (alpha-Bgtx) binding. Following transient expression in HEK 293-derived TSA-201 cells, covalent modification of the introduced cysteines with thiol-specific reagents reveals that alpha subunit residues W187, V188, F189, Y190, and P194 are solvent accessible and are in a position to contribute to the alpha-Bgtx binding site in native receptors. These results with the intact receptor are consistent with NMR studies of an alpha-Bgtx/receptor-dodecapeptide complex [Basus, V., Song., G., and Hawrot, E. (1993) Biochemistry 32, 12290-12298]. We pursued a more detailed analysis of the F189C mutant as this site varies substantially between AChRs that bind Bgtx and certain neuronal AChRs that do not. Treatment of intact cells expressing F189C with either bromoacetylcholine (BrACh) or [2-(trimethylammonium)ethyl] methane-thiosulfonate (MTSET), both methylammonium-containing thiol-modifying reagents with agonist properties, results in a marked decrease ( approximately 55-70%) in the number of alpha-Bgtx binding sites, as measured under saturating conditions. The decrease in sites appears to affect both alpha/gamma and alpha/delta sites to the same extent, as shown for alphaW187C and alphaF189C which were the two mutants examined on this issue. In contrast to the results obtained with MTSET and BrACh, modification with reagents that lack the alkylammonium entity, such as methylmethanethiosulfonate (MMTS), the negatively charged 2-sulfonatoethyl methane-thiosulfonate (MTSES), or the positively charged aminoethyl methylthiosulfonate (MTSEA), has little or no effect on the maximal binding of alpha-Bgtx to the alphaW187C, alphaV188C, or alphaF189C mutant receptors. The striking alkylammonium dependency suggests that an interaction of the tethered modifying group with the negative subsite within the agonist binding domain is primarily responsible for the observed blockade of toxin binding.     

Structural and spectral response of Aequorea victoria green fluorescent proteins to chromophore fluorination

Global replacements of tyrosine by 2- and 3-fluorotyrosine in "enhanced green" and "enhanced yellow" mutants of Aequorea victoria green fluorescent proteins (avGFPs) provided protein variants with novel biophysical properties. While crystallographic and modeled structures of these proteins are indistinguishable from those of their native counterparts (i.e., they are perfectly isomorphous), there are considerable differences in their spectroscopic properties. The fluorine being an integral part of the avGFP chromophore induces changes in the titration curves, variations in the intensity of the absorbance and fluorescence, and spectral shifts in the emission maxima. Furthermore, targeted fluorination in close proximity to the fluorinated chromophore yielded additional variants with considerably enhanced spectral changes. These unique spectral properties are intrinsic features of the fluorinated avGFPs, in the context of the rigid chromophore-microenvironment interactions. The availability of the isomorpohous crystal structures of fluorinated avGFPs allowed mapping of novel, unusual interaction distances created by the presence of fluorine atoms. In addition, fluorine atoms in the ortho position of the chromophore tyrosyl moiety exhibit a single conformation, while in the meta position two conformer states were observed in the crystalline state. Such global replacements in chromophores of avGFPs and similar proteins result in "atomic mutations" (i.e., H --> F replacements) in the structures, offering unprecedented opportunities to understand and manipulate the relationships between protein structure and spectroscopic properties.     

Mechanism of flavin mononucleotide cofactor binding to the Desulfovibrio vulgaris flavodoxin. 1. Kinetic evidence for cooperative effects associated with the binding of inorganic phosphate and the 5'-phosphate moiety of the cofactor

The pathway(s) by which the flavin cofactor binds to the apoflavoprotein is the subject of some debate. The crystal and NMR structures of several different flavodoxins have provided some insight, although there is disagreement about the location of the initial interaction between the flavin mononucleotide (FMN) and the apoflavodoxin and the degree of protein conformational change associated with cofactor binding [Genzor, C. G., Perales-Alcon, A., Sancho, J., and Romero, A. (1996) Nat. Struct. Biol. 3, 329-332; Steensma, E., and van Mierlo, C. P. M. (1998) J. Mol. Biol. 282, 653-666]. Binding kinetics using stopped-flow spectrofluorimetry and phosphate competition studies were used to develop a model for flavin binding to the flavodoxin from Desulfovibrio vulgaris. In the presence of phosphate, the time course of fluorescence quenching associated with FMN binding to apoflavodoxin was biphasic, whereas riboflavin, which lacks the 5'-phosphate group of FMN, displayed monophasic binding kinetics. When the concentration of phosphate in solution was increased, the FMN binding rates of the two phases behaved differently; the rate of one phase decreased, while the rate of the other increased. A similar increase in the single phase associated with riboflavin binding was also observed. This has led to the following model. The binding of the flavin isoalloxazine ring to its subsite is dependent on the presence of a phosphate group in the phosphate-binding subsite. When phosphate is in the buffer solution, FMN can bind in either of two ways: by the initial insertion of the 5'-phosphate group followed by ring binding or, when inorganic phosphate from solution is bound, the insertion of the isoalloxazine ring first. Riboflavin, which lacks the phosphate moiety of FMN, binds only in the presence of inorganic phosphate, presumably due to the binding of this group in the phosphate-binding subsite. These results suggest that cooperative interactions exist between the phosphate subsite and the ring-binding region in the D. vulgaris flavodoxin that are necessary for isoalloxazine ring binding.     

Site-directed mutagenesis of firefly luciferase active site amino acids: a proposed model for bioluminescence color.

Under physiological conditions firefly luciferase catalyzes the highly efficient emission of yellow-green light from the substrates luciferin, Mg-ATP, and oxygen. In nature, bioluminescence emission by beetle luciferases is observed in colors ranging from green (approximately 530 nm) to red (approximately 635 nm), yet all known luciferases use the same luciferin substrate. In an earlier report [Branchini, B. R., Magyar, R. M., Murtiashaw, M. H., Anderson, S. M., and Zimmer, M. (1998) Biochemistry 37, 15311-15319], we described the effects of mutations at His245 on luciferase activity. In the context of molecular modeling results, we proposed that His245 is located at the luciferase active site. We noted too that the H245 mutants displayed red-shifted bioluminescent emission spectra. We report here the construction and purification of additional His245 mutants, as well as mutants at residues Lys529 and Thr343, all of which are stringently conserved in the beetle luciferase sequences. Analysis of specific activity and steady-state kinetic constants suggested that these residues are involved in luciferase catalysis and the productive binding of substrates. Bioluminescence emission spectroscopy studies indicated that point mutations at His245 and Thr343 produced luciferases that emitted light over the color range from green to red. The results of mutational and biochemical studies with luciferase reported here have enabled us to propose speculative mechanisms for color determination in firefly bioluminescence. An essential role for Thr343, the participation of His245 and Arg218, and the involvement of bound AMP are indicated.     

Using Raman spectroscopy to monitor the solvent-exposed and "buried" forms of flavin in p-hydroxybenzoate hydroxylase

X-ray crystallographic studies of several complexes involving FAD bound to p-hydroxybenzoate hydroxylase (PHBH) have revealed that the isoalloxazine ring system of FAD is capable of adopting in two positions on the protein. In one, the "in" form, the ring is surrounded by protein groups and has little contact with solvent; in the second, "out" form, the ring is largely solvent exposed. Using Raman difference spectroscopy, it has been possible to obtain Raman spectra for the flavin ring in both conformational states for different complexes in solution. The spectra consist of a rich assortment of isoalloxazine ring modes whose normal mode origin can be assigned by using density functional theory and ab initio calculations. Further insight into the sensitivity of these modes to changes in environment is provided by the Raman spectra of lumiflavin in the solid state, in DMSO and in aqueous solution. For the protein complexes, the Raman difference spectra of flavin bound to wt PHBH and wt PHBH plus substrate, p-hydroxybenzoate, provided examples of the "in" conformation. These data are compared to those for flavin bound to wt PHBH plus 2,4-dihydroxybenzoate, where X-ray analysis show that the flavin is "out". There are several spectral regions where characteristic differences exist for flavin in the "in" or "out" conformation, these occur near 1700, 1500, 1410, 1350, 1235, and 1145 cm(-)(1). These spectral features can be used as empirical marker bands to determine the populations of "in" and "out" for any complex of PHBH and to monitor changes in those populations with perturbations to the system, e.g., by changing temperature or pH. Thus, it will now be possible to determine the conformational state of the flavin in PHBH for those complexes that have resisted X-ray crystallographic analysis. Raman difference data are also presented for the Tyr222Phe mutant. The Raman data show that the isoalloxazine ring is predominantly "out" for Tyr222Phe. However, in the presence of the substrate p-hydroxybenzoate there is clear evidence from the Raman marker bands that a mixed population of "in" and "out" exists with the majority being in the "out" state. This is consistent with the conclusions drawn from crystallographic studies on this complex (Gatti, D. L., Palfey, B. A., Lah, M. S., Entsch, B., Massey, V., Ballou, D. P., and Ludwig, M. L. (1994) Science, 266, 110-114).     

Activity and stability of the luciferase--flavin intermediate.

A luciferase intermediate in the bacterial bioluminescence system, which is formed by reaction of enzyme with reduced flavin mononucleotide (FMNH2) and oxygen, is shown to emit light with added aldehyde under anaerobic conditions. The reaction with oxygen is thus effectively irreversible under the conditions used. The flavin chromophore has an absorption maximum at about 370 nm and the potential activity (bioluminescence yield) in the further reaction of the isolated intermediate with aldehyde is strictly proportional to the amount of this flavin chromophore.     

Factors That Control the Chemistry of the LOV Domain Photocycle

Algae, plants, bacteria and fungi contain Light-Oxygen-Voltage (LOV) domains that function as blue light sensors to control cellular responses to light. All LOV domains contain a bound flavin chromophore that is reduced upon photon absorption and forms a reversible, metastable covalent bond with a nearby cysteine residue. In Avena sativa LOV2 (AsLOV2), the photocycle is accompanied by an allosteric conformational change that activates the attached phototropin kinase in the full-length protein. Both the conformational change and formation of the cysteinyl-flavin adduct are stabilized by the reduction of the N5 atom in the flavin’s isoalloxazine ring. In this study, we perform a mutational analysis to investigate the requirements for LOV2 to photocycle. We mutated all the residues that interact with the chromophore isoalloxazine ring to inert functional groups but none could fully inhibit the photocycle except those to the active-site cysteine. However, electronegative side chains in the vicinity of the chromophore accelerate the N5 deprotonation and the return to the dark state. Mutations to the N414 and Q513 residues identify a potential water gate and H₂O coordination sites. These residues affect the electronic nature of the chromophore and photocycle time by helping catalyze the N5 reduction leading to the completion of the photocycle. In addition, we demonstrate that dehydration leads to drastically slower photocycle times. Finally, to investigate the requirements of an active-site cysteine for photocycling, we moved the nearby cysteine to alternative locations and found that some variants can still photocycle. We propose a new model of the LOV domain photocycle that involves all of these components.