Probing the Vibrio harveyi luciferase beta subunit functionality and the intersubunit domain by site-directed mutagenesis

Protonation status and control mechanism of flavin-oxygen intermediates in the reaction of bacterial luciferase

Bacterial luciferase catalyzes a bioluminescent reaction by oxidizing long-chain aldehydes to acids using reduced FMN and oxygen as co-substrates. Although a flavin C4a-peroxide anion is postulated to be the intermediate reacting with aldehyde prior to light liberation, no clear identification of the protonation status of this intermediate has been reported. Here, transient kinetics, pH variation, and site-directed mutagenesis were employed to probe the protonation state of the flavin C4a-hydroperoxide in bacterial luciferase. The first observed intermediate, with a λ_max of 385 nm, transformed to an intermediate with a λ_max of 375 nm. Spectra of the first observed intermediate were pH-dependent, with a λ_max of 385 nm at pH < 8.5 and 375 at pH > 9, correlating with a pK_a of 7.7-8.1. These data are consistent with the first observed flavin C4a intermediate at pH < 8.5 being the protonated flavin C4a-hydroperoxide, which loses a proton to become an active flavin C4a-peroxide. Stopped-flow studies of His44Ala, His44Asp, and His44Asn variants showed only a single intermediate with a λ_max of 385 nm at all pH values, and none of these variants generate light. These data indicate that His44 variants only form a flavin C4a-hydroperoxide, but not an active flavin C4a-peroxide, indicating an essential role for His44 in deprotonating the flavin C4a-hydroperoxide and initiating chemical catalysis. We also investigated the function of the adjacent His45; stopped-flow data and molecular dynamics simulations identify the role of this residue in binding reduced FMN.

Bacterial luciferase: Molecular mechanisms and applications

Bacterial luciferase is a flavin-dependent monooxygenase which is remarkable for its distinctive feature in transforming chemical energy to photons of visible light. The bacterial luciferase catalyzes bioluminescent reaction using reduced flavin mononucleotide, long-chain aldehyde and oxygen to yield oxidized flavin, corresponding acid, water and light at λ_max around 490nm. The enzyme comprises of two non-identical α and β subunits, where α subunit is a catalytic center and β subunit is crucially required for maintaining catalytic function of the α subunit. The crystal structure with FMN bound and mutagenesis studies have assigned a number of amino acid residues that are important in coordinating critical reactions and stabilizing intermediates to attain optimum reaction efficiency. The enzyme achieves monooxygenation by generating C4a-hydroperoxyflavin intermediate that later changes its protonation status to become C4a-peroxyflavin, which is necessary for the nucleophilic attacking with aldehyde substrate. The decomposing of C4a-peroxyhemiacetal produces excited C4a-hydroxyflavin and acid product. The chemical basis regrading bioluminophore generation in Lux reaction remains an inconclusive issue. However, current data can, at least, demonstrate the involvement of electron transfer to create radical molecules which is the key step in this mechanism. Lux is a self-sufficient bioluminescent system in which all substrates can be recycled and produced by a group of enzymes from the lux operon. This makes Lux distinctively advantageous over other luciferases for reporter enzyme application. The progression of understanding of Lux catalysis is beneficial to improve light emitting efficiency in order to expand the robustness of Lux application.

Crystal structure of the bacterial luciferase/flavin complex provides insight into the function of the beta subunit

Bacterial luciferase from Vibrio harveyi is a heterodimer composed of a catalytic alpha subunit and a homologous but noncatalytic beta subunit. Despite decades of enzymological investigation, structural evidence defining the active center has been elusive. We report here the crystal structure of V. harveyi luciferase bound to flavin mononucleotide (FMN) at 2.3 A. The isoalloxazine ring is coordinated by an unusual cis-Ala-Ala peptide bond. The reactive sulfhydryl group of Cys106 projects toward position C-4a, the site of flavin oxygenation. This structure also provides the first data specifying the conformations of a mobile loop that is crystallographically disordered in both prior crystal structures [(1995) Biochemistry 34, 6581-6586; (1996) J. Biol. Chem. 271, 21956 21968]. This loop appears to be a boundary between solvent and the active center. Within this portion of the protein, a single contact was observed between Phe272 of the alpha subunit, not seen in the previous structures, and Tyr151 of the beta subunit. Substitutions at position 151 on the beta subunit caused reductions in activity and total quantum yield. Several of these mutants were found to have decreased affinity for reduced flavin mononucleotide (FMNH(2)). These findings partially address the long-standing question of how the beta subunit stabilizes the active conformation of the alpha subunit, thereby participating in the catalytic mechanism.

Mutational analysis of the subunit interface of Vibrio harveyi bacterial luciferase

Bacterial luciferase is a heterodimeric (alphabeta) enzyme which catalyzes a light-producing reaction in Vibrio harveyi. In addition to the alphabeta enzyme, the beta subunit can self-associate to form a stable but inactive homodimer [Sinclair, J. F., Ziegler, M. M., and Baldwin, T. O. (1994) Nat. Struct. Biol. 1, 320-326]. The studies reported here were undertaken to explore the role of the subunit interface in the conformational stability of the enzyme. To this end, we constructed four mutant heterodimers in which residues at the subunit interface were changed in an effort to alter the volume of an apparent solvent accessible channel at the interface or to alter H-bonding groups. Equilibrium unfolding data for the heterodimer have been interpreted in terms of a three-state mechanism [Clark, C. A., Sinclair, J. F., and Baldwin, T. O. (1993) J. Biol. Chem. 268, 10773-10779]. However, we found that unfolding for the wild-type and mutant luciferases is better described by a four-state model. This change in the proposed mechanism of unfolding is based on observation of residual structure in the subunits following dissociation of the heterodimeric intermediate. All of the mutants display modest reductions in activity but, surprisingly, no change in the DeltaG2H2O value for subunit dissociation and no measurable change in the equilibrium dissociation constant relative to that of the wild-type heterodimer. However, the DeltaG1H2O value for the formation of the dimeric intermediate that precedes subunit dissociation is reduced for three of the mutants, indicating that mutations at the interface can alter the stability of a region of the alpha subunit that is distant from the interface. We conclude that the interface region communicates with the distal domains of this subunit, probably through the active center region of the enzyme.

Characterization of the binding of Photobacterium phosphoreum P-flavin by Vibrio harveyi Luciferase

The isolated Photobacterium phosphoreum luciferase is associated with a bound flavin designated P-flavin and tentatively identified as 6-(3"-myristic acid)-FMN. Since FMN and myristic acid are products of the normal luciferase reaction, we explored the possibility that P-flavin can also be bound by luciferase from other luminous bacteria and serve as an active site probe. P-flavin has never been detected in Vibrio harveyi cells. We found that the V. harveyi luciferase binds P. phosphoreum P-flavin, at a ratio of 1 P-flavin per luciferase alphabeta dimer, and with concomitant absorption spectral perturbation of P-flavin, fluorescence quenching of P-flavin and luciferase, and activity inhibition of luciferase. Isolated P-flavin can be fully reduced photochemically. V. harveyi luciferase bound the oxidized P-flavin with a K(d) (or K(i) competitively against decanal) of 0.1-0.16 microM, which is three orders of magnitude lower than the K(d) for FMN binding but similar to that of reduced FMN binding. The reduced P-flavin exhibited a K(i) (competitively against the reduced FMN substrate) of 0.16 microM, also similar to the K(d) for reduced FMN. Hence, the covalent attachment of myristic acid to FMN greatly and preferentially enhanced the binding of oxidized P-flavin. The dissociation of P-flavin was slow in comparison with the binding of reduced FMN and decanal substrates. Modification of the alphaCys106 near the active site by N-ethylmaleimide can be retarded by P-flavin. These findings indicate that P-flavin is potentially a superb active site probe for luciferase. We hypothesize that P-flavin is a by-product of luciferase generated by a side reaction which is trivial with the V. harveyi luciferase but significant in the P. phosphoreum luciferase-catalyzed reaction.

Catalytically active forms of the individual subunits of Vibrio harveyi luciferase and their kinetic and binding properties

Contradictory findings have recently been reported regarding the (in)abilities of individual subunits of the Vibrio harveyi alpha beta dimeric luciferase to catalyze bioluminescence. We have produced individual alpha and beta subunits separately in Escherichia coli JM109 cells by recombinant DNA techniques. Both subunits were purified to more than 90% homogeneity and found to be catalytically active, with their general catalytic properties and the specific activities similar to those reported earlier (Sinclair, J. F., Waddle, J. J., Waddill, E. F., and Baldwin, T. O. (1993) Biochemistry 32, 5036-5044). Individual subunits were significantly distinct from the native luciferase with respect to inactivations by trypsin and N-ethylmaleimide, and the stability of the flavin 4a-hydroperoxide intermediate. The active species in isolated alpha and beta samples were each the predominant protein species, corresponding to a 42,000 M(r) alpha monomer and a 67,000 M(r) beta dimer, respectively. These findings clearly indicate that the activities of the individual subunits are not due to trace contaminations of the respective counter subunits. The much reduced specific activities of the individual subunits are, in part, a consequence of diminished abilities to oxidize the aldehyde substrate. Kinetic and equilibrium measurements indicate that alpha and beta 2 each contained a reduced flavin site, an aldehyde substrate site, and an aldehyde inhibitor site. The on and off rates of the decanal inhibitor binding were substantially slower than the bindings of decanal and reduced riboflavin 5'-phosphate substrates. These findings are consistent with a scheme that the aldehyde inhibitor blocks the binding of the reduced flavin substrate.