- Email: [email protected]
Firefly luciferase: the structure is known, but the mystery remains
Minireview
223
Firefly luciferase: the structure is known, but the mystery remains Thomas O Baldwin The structure of firefly luciferase reveals a new protein fold which may be representative of a growing family of homologous enzymes. Address: Department of Biochemistry and Biophysics and Department of Chemistry, Center for Macromolecular Design, Texas A&M University, College Station, TX 77843–2128, USA. Structure 15 March 1996, 4:223–228 © Current Biology Ltd ISSN 0969-2126
From the Atlantic Ocean to the eastern face of the Rocky Mountains, a summer sunset is usually accompanied by one of Nature’s most delightful light shows. Photinus pyralis, the North American firefly, has entertained countless observers, probably since the coming of humans to the continent. During the period between the 1950s and the 1980s, numerous young biologists earned their spending money as firefly collectors, first employed by Professor William D McElroy, then of Johns Hopkins University, and later as members of the Sigma Firefly Club. In 1985, when Marlene DeLuca and her colleagues cloned the cDNA encoding the luciferase of P. pyralis, an alternative source of the enzyme became available, and soon after, in laboratories around the world, numerous other organisms began to emit the characteristic yellow-green luminescence as a consequence of expression of the firefly luciferase in their cells. Historical perspective
Luciferase is a generic term describing any enzyme that catalyzes a reaction yielding visible light. Light emission is a consequence of formation of a product or intermediate in an electronically excited state; return to the ground state occurs via emission of a photon of light. Luciferases are highly diverse, catalyzing a great variety of reactions and using widely different substrates [1]. All have in common the involvement of oxygen. Luciferases are more different than, for example, proteases, which all carry out hydrolytic chemistry on peptide bonds. Luciferases all emit light, but they do so by very different means. Therefore, the luciferases from different organisms probably evolved independently, rather than from a common ancestral enzyme. Bacterial luciferase, the first luciferase to be cloned and also the first to be structurally characterized, is a flavin monooxygenase that utilizes flavin mononucleotide (FMN) to activate molecular oxygen, yielding a flavin C4a peroxide. Reaction of the peroxide with an aliphatic aldehyde substrate yields, ultimately, the carboxylic acid and the flavin C4a hydroxide in the first singlet excited state. Light emission, loss of the C4a hydroxide and dissociation of FMN returns the enzyme to
its starting condition. Firefly luciferase, on the other hand, catalyzes an oxidative reaction involving ATP, firefly luciferin and molecular oxygen, yielding an electronically excited oxyluciferin species [2]. This excited species emits visible light, which is employed by the firefly in its reproductive behavior [3]. Firefly luciferase was one of the first enzymes to be investigated in biochemical detail [4]. WD McElroy and his colleagues [5–9], as well as other investigators [10] working during the 1940s and 1950s with P. pyralis, determined the structures of the substrates and products of the enzyme. Evidence for the chemical mechanism proposed for the firefly luciferase reaction [11] has been discussed in a recent review [12]. The structures of luciferases
The structure of luciferase from the firefly P. pyralis has now been determined at high resolution and is reported by Conti et al. in this issue of Structure [13]. As beautifully described in their paper, the enzyme folds into two distinct domains, a large N-terminal domain comprising residues 4–436, and a C-terminal domain formed from residues 440–544. The structure is shown in Figure 1. The fold assumed by the luciferase polypeptide appears to be unique. The N-terminal domain consists of a b barrel and two b sheets flanked by a helices which form a five-layered ababa structure. The C-terminal domain, consisting of five b strands and three a helices, is folded into a compact structure that is connected to the N-terminal domain by a disordered loop (connecting residues 435 and 441). There are three other disordered loops not visible in the electron density, one in the C-terminal domain connecting residues 523 and 529, and two in the N-terminal domain (connecting residues 198–204 and residues 355–359). The structure presented is without bound substrates or other ligands. Conti et al. have taken advantage of the homology of firefly luciferase with other enzymes that catalyze similar reactions [14] to deduce the location of the active center (see Fig. 2). It is assumed that regions of greatest sequence conservation are most likely to be involved in the catalytic mechanism of the enzyme. Based on their extensive analysis, it is proposed that the active center is composed of residues on the surfaces of both domains, and that upon substrate binding, the domains move together to form the active center. The drawings presented and the comparisons made with the amino acid sequences of the large family of homologous proteins constitute a compelling argument. For a bioluminescence reaction to occur with a high quantum yield, as is the case for firefly luciferase, it is essential that water be excluded from the active site. It would appear that the two-domain structure presented by Conti et al. could serve such a purpose.
224
Structure 1996, Vol 4 No 3
Figure 1 Two orthogonal stereoviews of the surface of firefly luciferase, depicting an ‘anvil and hammer’ motif. Both views show the large N-terminal ‘anvil’ domain below the smaller C-terminal ‘hammer’ domain. The two views are the front view (top) and the right side view (bottom), obtained by rotating the front view through 90° to the left about the vertical axis. The color coding is the same as that used by Conti et al. [13], and indicates the domains and subdomains. The C-terminal domain (yellow) is the only domain composed exclusively of residues that are contiguous in the amino acid sequence, residues 440–544. The three subdomains of the N-terminal anvil consist of stretches that are not contiguous within the overall sequence. Subdomain A (blue) consists of residues 77–222 and 399–405. Subdomain B (purple) consists of residues 22–70 and 236–351. Subdomain C (green) consists of residues 4–10, 363–393 and 418–434. It appears that the active center comprises residues between the anvil and hammer, and it is suggested that the active center forms by movement of these two domains together following substrate binding [13]. (Figure courtesy of Peter Brick.)
The structure of bacterial luciferase has also been determined to high resolution without bound ligands [15]. Bacterial luciferase is a heterodimer composed of homologous subunits, a and b (see [16] for a review of the bacterial luciferase literature). Both subunits assume the wellknown (b/a)8, or TIM barrel structure, and are packed together by a parallel four-helix bundle. The active center is confined to the a subunit, and it has been proposed to reside within a large internal cavity which opens through a narrow crevice [17]. The opening to the cavity lies beneath a long disordered loop which appears to undergo a conformational rearrangement upon flavin binding and catalysis [16–20]. It has been proposed that this conformational change may be necessary to exclude water from the intermediates and the excited state of the flavin. Thus, it
appears reasonable to propose that both luciferases exclude water from their respective excited emitters by a conformational rearrangement that occurs upon or following substrate binding. Reaction catalyzed by firefly luciferase
Firefly luciferase catalyzes a multistep reaction [21]. In the first step, luciferin (compound I; Fig. 3) reacts with Mg2+-ATP to form luciferyl adenylate (compound II) and pyrophosphate. The luciferyl adenylate is oxidized by molecular oxygen, with the intermediate formation of the cyclic peroxide, a dioxetanone (compound III), and a molecule of AMP. The dioxetanone is decarboxylated as a result of intramolecular conversions (compound IV) to produce an electronically excited state of oxyluciferin in
Minireview Luciferases Baldwin
225
Figure 2
Alignment of the amino acid sequences, reported by Devine et al. [49], of the luciferases of Photinus pyralis (P.p), Luciola mingrelica (L.m), L. cruciata (L.c), L. lateralis (L.l), and the green-emitting strain of the click beetle (CbG). Also shown in this alignment is the 4-coumarate:CoA ligase (CoA). References to the sequences are given in [49]. The numbering at the top refers to the sequence of the
luciferase from P. pyralis. The extent of conservation at each position is indicated by the following color code: red=fully conserved in all six sequences; pink=2 different amino acids; green=3 different amino acids, and blue=4–6 different amino acids. At positions where a deletion has occurred in one or more of the proteins, the corresponding residues in the other proteins are shown in black.
the enol or keto form (compound V). Return to the ground state is accompanied by emission of a quantum of visible light with a wavelength of maximum light intensity (Imax) of 562–570 nm. Shimomura et al. [22] demonstrated that one oxygen atom of the product CO2 arises from the substrate oxygen. Non-enzymatic oxidation of luciferin yields oxyluciferin without luminescence [23]. (For a recent review, see [12].)
cruciata and Luciola lateralis [24], have been purified and characterized. The L. mingrelica and L. cruciata luciferases are similar to the P. pyralis luciferase, with Imax of 562–570 nm, whereas the reaction catalyzed by L. lateralis luciferase emits green light (Imax 552 nm). All Luciola luciferases inactivate rapidly at low ionic strength [21], whereas the P. pyralis luciferase crystallizes in active form under the same conditions [4].
Color of the bioluminescence of fireflies
The chemiluminescence emission spectra of a variety of luciferin analogs were studied by White et al. [9,23]. Based on the fluorescence emission spectra of these compounds in the presence and absence of proton acceptors, they suggested that the red emission arises from the keto anion form of the product molecule and the yellow-green
One of the most intriguing aspects of firefly bioluminescence pertains to the color of the light emitted. Recently, the luciferase from Luciola mingrelica fireflies, collected in the southern regions of Russia [21], as well as the luciferases of the fireflies indigenous to Japan, Luciola
226
Structure 1996, Vol 4 No 3
Figure 3 Proposed mechanism of the firefly bioluminescence reaction. The carboxylate group of firefly luciferin (I) is activated by reaction with ATP to form the adenylated luciferin (II). The a proton is lost, allowing reaction with molecular oxygen to yield (III). Cleavage of the dioxetanone ring (IV) yields the excited state of oxyluciferin (V).
emission from the enol anion. This hypothesis is consistent with the emission of red light in the bioluminescence reaction at lower pH, and has led to the suggestion that the various colors of light observed from different species of firefly result from a spectral mixing from these two emitters [5]. However, when luciferases were isolated from different organs of the click beetle Pyrophorus plagiophthalamus, each emitted a different color of bioluminescence, and each emission spectrum showed a single peak, not a superimposition of two or more different spectra [24,25]. The cDNAs of four luciferases from P. plagiophthalamus, chosen on the basis of differences in color of bioluminescence, have been cloned and sequenced [26]. The amino acid sequences of these luciferases are 95–99% identical, and fewer than two or three amino acid changes are needed for spectral shifts of up to 50 nm in Imax. The properties of these luciferases require some rethinking of the traditional explanations for spectral differences between insect luciferases, which historically had been attributed to shifts in the proposed equilibrium distribution between the keto and enol forms of the excited-state oxyluciferin product. If this were the sole explanation for the spectral variations, one would expect to see intermediate bimodal spectra, as have been reported for P. pyralis luciferase titrated with zinc ions [24]. However, the four recombinant click beetle luciferases emit light with sharp emission spectra with spectral maxima of 546 nm (green), 560 nm (yellow-green), 578 nm (yellow), and 593 nm (orange) [27]. This finding suggests that light emission from each enzyme occurs from a single molecular species within the environment of the enzyme and that differences in color result from differences between the microenvironments of the enzyme–oxyluciferin complexes. In a related study, L. cruciata luciferase cDNA was mutagenized and five mutants with different colors of bioluminescence ranging from green to red were isolated [28]. The mutations were found to be single amino acid changes. Of particular interest is
the fact that a large (50 nm) red shift (to an Imax of 612 nm) was caused by a single substitution of an invariant histidinyl residue (conserved in all insect luciferases) by a tyrosine (His433→Tyr). When the sequences of the mutant L. cruciata luciferases were compared with those of the click beetle enzymes, no common amino acid sequence affecting the color of light was apparent [26]. It may be concluded that the different colors of bioluminescence are caused by subtle changes in the tertiary structure of the luciferase molecule. The firefly luciferase superfamily of enzymes
The number of proteins related to firefly luciferase and apparently descended from a common ancestor is growing rapidly. These enzymes are involved in the biosynthesis of iron-binding siderophores [29,30], the antibiotics gramicidin S [31,32], tyrocidine [33], and penicillin [34–36], and pathogen defense agents in plants [37,38] and the biodegradation of halogenated aromatics [39]. Fatty acid:coenzyme A (CoA) ligase [40] and acetate:CoA ligase [41] use the same mechanism as 4-coumarate:CoA ligase [37] and the proteins show a high degree of sequence similarity. The amino acid sequence of AngR, a DNA-binding protein that modulates iron-regulated transcription, also shows homology to the firefly luciferases [42]. Given the diversity and importance of the related proteins already discovered, more are likely to be found as the number of sequenced proteins increases. Each of the enzymes catalyzes the adenylation of a carboxylic acid substrate using Mg2+-ATP (Fig. 4), followed by reaction of the activated carboxylate with an acceptor and release of AMP. Several of the related enzymes use CoA or an active-site thiol as an acceptor for displacing the AMP. There does not appear to be an essential thiol in the active site of firefly luciferase [43], although there have been several reports that the addition of CoA enhances the emission of light from firefly luciferase [43–45]. The role, if any, that CoA plays in the firefly luciferase reaction remains to be elucidated.
Minireview Luciferases Baldwin
Figure 4
227
non-fluorescent flavoprotein (see [16] and other references therein) — the function of which is unknown. Even as our knowledge of the structures and mechanisms of these delightful enzymes becomes more detailed, the beauty of the biological phenomenon of bioluminescence will not be diminished. The phenomenon of a flashing firefly will always be a source of joy to those who pause to observe it. The mysterious emission of light without heat will always intrigue the curious; and for the technically more sophisticated, the questions regarding the color of bioluminescence, and the details of the enzyme-catalyzed reaction remain to be resolved. Knowledge of the structure is a major step forward but there is still much to be learned. References
Comparisons of representative reactions catalyzed by members of the firefly luciferase family of enzymes. Reaction (a) is catalyzed by firefly luciferase [8]; reaction (b) by 4-coumarate CoA:ligase [37]; reaction (c) by gramicidin S synthetase I and tyrocidine synthetase [31,32]; and reaction (d) by 2,3-dihydroxybenzoate-AMP ligase [29].
Summary
In recent years, luciferases and the genes that encode them have become very useful for research purposes as well as for various commercial processes. The applications are too numerous to discuss here, but examples include the use of firefly luciferase to study the role of chaperones in protein folding [46] and of bacterial luciferase to study co-translational folding of polypeptides [47] and of the genes encoding luciferases to monitor transcriptional activities [45,48]. Numerous applications are described in the various publications of the International Symposia on Bioluminescence and Chemiluminescence published by Wiley. Clearly, the structures of firefly luciferase and bacterial luciferase will be of great value in further developing the applications and our basic understanding of these intriguing enzymes. Detailed mechanistic insights must await structural elucidation of the various complexes of the luciferases with their substrates and products, so there is much structural work yet to be done. It is interesting to note that the subunits of bacterial luciferase assume perhaps the most common of folding motifs, the (b/a)8 barrel, whereas firefly luciferase folds into a new structural form. While there are numerous homologs of firefly luciferase that catalyze similar reactions and have similar amino acid sequences, there is only one known non-luciferase homolog of bacterial luciferase — the
1. Hastings, J.W. (1983). Biological diversity, chemical mechanisms, and the evolutionary origins of bioluminescent systems. J. Mol. Evol. 19, 309–321. 2. DeLuca, M. (1976). Firefly luciferase. Advances in Enzymology and Related Areas of Molecular Biology 44, 37–68. 3. Case, J.F. & Strause, L.G. (1978). Neurally controlled luminescent systems. In Bioluminescence in Action. (Herring, P., ed), pp. 332–366, Academic Press, NY. 4. McElroy, W.D. (1960). Crystalline firefly luciferase. Methods Enzymol. 5, 445–448. 5. Seliger, M.M. &. McElroy, W.D. (1960). Spectral emission and quantum yield of firefly bioluminescence. Arch. Biochem. Biophys. 88, 136–141. 6. White, E.H., McCapra, F., Field, G.F. & McElroy, W.D. (1961). The structure and synthesis of firefly luciferin. J. Am. Chem. Soc. 83, 2402–2403. 7. Seliger, H.H. & McElroy, W.D. (1959). Quantum yield in the oxidation of firefly luciferin. Biochem. Biophys. Res. Commun. 1, 21–24. 8. DeLuca, M. & McElroy, W.D. (1974). Kinetics of the firefly luciferase catalyzed reactions. Biochemistry 13, 921–925. 9. White, E.H., Rapaport, E., Hopkins, T.A. & Seliger, H.H. (1969). Chemi- and bioluminescence of firefly luciferin. J. Am. Chem. Soc. 91, 2178–2180. 10. Suzuki, N. & Goto, T. (1972). Studies on firefly bioluminescence — II. Identification of oxyluciferin as a product in the bioluminescence of firefly lanterns and in the chemiluminescence of firefly luciferin. Tetrahedron 28, 4075–4082. 11. McElroy, W.D. & DeLuca, M. (1978). Chemistry of firefly luminescence. In Bioluminescence in Action. (Herring, P.J., ed), pp. 109–127, Academic Press, London. 12. Wood, K.V. (1995). The chemical mechanism and evolutionary development of beetle bioluminescence. Photochem. Photobiol. 62, 662–673. 13. Conti, E., Franks, N.P. & Brick, P. (1996). Crystal structure of firefly luciferase throws light on a superfamily of adenylate-forming enzymes. Structure 4, 287–298. 14. Babbitt, P.C., et al., & Dunaway-Mariano, D. (1992). Ancestry of the 4-chlorobenzoate dehalogenase: analysis of aminoacid sequence identities among families of acyl:adenyl ligases, enoyl-CoA hydratases/isomerases, and acyl-CoA thioesterases. Biochemistry 31, 5594–5604. 15. Fisher, A.J., Raushel, F.M., Baldwin, T.O. & Rayment, I. (1995). The three-dimensional structure of bacterial luciferase from Vibrio harveyi at 2.4 Å resolution. Biochemistry 34, 6581–6586. 16. Baldwin, T.O. & Ziegler, M.M. (1992). The biochemistry and molecular biology of bacterial bioluminescence. In Chemistry and Biochemistry of Flavoenzymes. (Müller, F, ed), vol. III, pp. 467–530, CRC Press, Boca Raton, FL. 17. Baldwin, T.O., et al., and Rayment, I. (1995). Structure of bacterial luciferase. Curr. Opin. Struct. Biol. 5, 798–809. 18. Baldwin, T.O. & Riley, P.L. (1980). Anion binding to bacterial luciferase: evidence for binding associated changes in enzyme structure. In Flavins and Flavoproteins. (Yagi, K. & Yamano, T, eds), pp. 139–147, Japan Scientific Societies Press, Tokyo and University Park Press, Baltimore.
228
Structure 1996, Vol 4 No 3
19. Holzman, T.F. & Baldwin, T.O. (1980). Proteolytic inactivation of luciferases from three species of luminous marine bacteria, Beneckea harveyi, Photobacterium fischeri, and Photobacterium phosphoreum: evidence of a conserved structural feature. Proc. Natl. Acad. Sci. USA 77, 6363–6367. 20. AbouKhair, N.K., Ziegler, M.M. & Baldwin, T.O. (1985). Bacterial luciferase: demonstration of a catalytically competent altered conformational state following a single turnover. Biochemistry 24, 3942–3947. 21. Ugarova, N.N. (1989). Luciferase of Luciola mingrelica fireflies. kinetics and regulation mechanism. J. Biolumin. Chemilumin. 4, 406–418. 22. Shimomura, O., Goto, T. & Johnson, F.H. (1977). Source of oxygen in the CO2 produced in the bioluminescent oxidation of firefly luciferin. Proc. Natl. Acad. Sci. USA 74, 2799–2802. 23. White, E.H., Rapaport, E., Seliger, H.H. & Hopkins, T.A. (1971). The chemi- and bioluminescence of firefly luciferin: an efficient chemical production of electronically excited states. Bioorg. Chem. 1, 92–122. 24. Seliger, H.H. & McElroy, W.D. (1964). The colors of firefly bioluminescence: enzyme configuration and species specificity. Proc. Natl. Acad. Sci. USA 52, 75–81. 25. Seliger, H.H., Buck, J.B., Fastie, W.G. & McElroy, W.D. (1964). The spectral distribution of firefly light. J. Gen. Physiol. 48, 95–104. 26. Wood, K.V., Lam, Y.A., Seliger, H.H. & McElroy, W.D. (1989). Complementary DNA coding click beetle luciferases can elicit bioluminescence of different colors. Science 244, 700–702. 27. Wood, K.V., Lam, Y.A. & McElroy, W.D. (1989). Introduction to beetle luciferases and their applications. J. Biolumin. Chemilumin. 4, 289–301. 28. Kajiyama, N. & Nakano, E. (1991). Isolation and characterization of mutants of firefly luciferase which produce different colors of light. Protein Eng. 4, 691–693. 29. Rusnak, F., Faraci, W.S. & Walsh, C.T. (1989). Subcloning, expression, and purification of the enterobactin biosynthetic enzyme 2,3dihydroxybenzoate-AMP ligase: demonstration of enzyme-bound (2,3dihydroxybenzoyl)adenylate product. Biochemistry 28, 6827–6835. 30. Rusnak, F., Sakaitani, M., Drueckhammer, D., Reichert, J. & Walsh, C.T. (1991). Biosynthesis of the Escherichia coli siderophore enterobactin: sequence of the entF gene, expression and purification of EntF, and analysis of covalent phosphopantetheine. Biochemistry 30, 2916–2927. 31. Kanda, M., Hori, K., Kurotsu, T., Miura, S. & Saito, Y. (1989). Reaction mechanism of gramicidin synthetase 1, phenylalanine racemase of Bacillus brevis. J. Biochem. (Tokyo) 105, 653–659. 32. Kratzschmar, J., Krause, M. & Maraheil, M.A. (1989). Gramacidin S biosynthesis operon containing the structural genes grsA and grsB has an open reading frame encoding a protein homologous to fatty acid thioesterases. J. Bacteriol. 171, 5422–5429. 33. Roskoski, R., Jr., Gevers, W., Kleinkauf, H. & Lipmann, F. (1970). Tyrocidine biosynthesis by three complementary fractions from Bacillus brevis (ATCC 8185). Biochemistry 9, 4839–4845. 34. Kleinkauf, H. & von Döhren, H. (1990) Nonribosomal biosynthesis of peptide antibiotics. Eur. J. Biochem. 192, 1–15. 35. Gutierrez, S., Diez, B., Montenegro, E. & Martin, J.F. (1991). Characterization of the Cephalosporium acremonium pcbAB gene encoding a-aminoadipyl-cysteinyl-valine synthetase, a large multidomain peptide synthetase: linkage to the pcbC gene as a cluster of early cephalosporin biosynthetic genes and evidence of multiple functional domains. J. Bacteriol. 173, 2354–2365. 36. MacCabe, et al., & Kinghorn, J.R. (1991). D-(L-a-aminoadipyl)-Lcysteinyl-D-valine synthetase from Aspergillus nidulans. J. Biol. Chem. 266, 12646–12654. 37. Knobloch, K.H. & Hahlbrock, K. (1977). 4-Coumarate:CoA ligase from cell suspension cultures of Petroselinum hortense Hoffm. Partial purification, substrate specificity, and further properties. Arch. Biochem. Biophys. 184, 237–248. 38. Becker-Andre, M., Schulze-Lefert, P. & Hahlbrock, K. (1991). Structural comparison, modes of expression, and putative cis-acting elements of the two 4-coumarate:CoA ligase genes in potato. J. Biol. Chem. 266, 8551–8559. 39. Scholten, J.D., Chang, K.H., Babbitt, P.C., Charest, H., Sylvestre, M. & Dunaway-Mariano, D. (1991). Novel enzymic hydrolytic dehalogenation of a chlorinated aromatic. Science 253, 182–185. 40. Suzuki, H., et al., & Yamomoto, T. (1990). Structure and regulation of rat long-chain acyl-CoA synthetase. J. Biol. Chem. 265, 8681–8685. 41. Connerton, I.F., Fincham. J.R.S., Sandeman, R.A. & Hynes, M.J. (1990). Comparison and cross species expression of the acetyl-CoA synthetase genes from the Ascomycete fungi Aspergillus nidulans and Neurospora crassa. Mol. Microbiol, 1, 451–459.
42. Farrel, D.H., Mikesell, P., Actis, L.A. & Crosa, J.H. (1990). A regulatory gene, angR, of the iron uptake system of Vibrio anguillarium: similarity with phage P22 cro and regulation by iron. Gene 86, 45–51. 43. Alter, S.C. & DeLuca, M. (1986). The sulfhydryls of firefly luciferase are not essential for activity. Biochemistry 25, 1599–1605. 44. Wood, K.V. (1992). Improved kinetics of light production by beetle luciferase using thiol reagents. Chem. Abstr. 116, 230750n. 45. Pazzagli, M., Devine, J.H., Peterson, D.O. & Baldwin, T.O. (1992). Use of bacterial and firefly luciferases as reporter genes in DEAE/dextranmediated transfection of mammalian cells. Anal. Biochem. 204, 315–323. 46. Szabo, A., Langer, T., Schröder, H., Flanagan, J., Bukau, B. & Hartl, F.U. (1994). The ATP hydrolysis-dependent reaction cycle of the Escherichia coli Hsp70 system — DnaK, DnaJ, and GrpE. Proc. Natl. Acad. Sci. USA 91, 10345–10349. 47. Fedorov, A.N. & Baldwin, T.O. (1995). Contribution of cotranslational folding to the rate of formation of native protein structure. Proc. Natl. Acad. Sci. USA 92, 1227–1231. 48. Wood, K.V. (1991). Recent advances and prospects for use of beetle luciferases as genetic reporters. In Bioluminescence and Chemiluminescence: Current Status. (Stanley, P.E. & Kricka, L.J., eds), pp. 543–546, John Wiley and Sons, Chichester. 49. Devine, J.H., Kutuzova, G., Green, V.W., Ugarova, N. & Baldwin, T.O. (1993). Luciferase from the European firefly Luciola mingrelica: cloning and nucleotide sequence of the cDNA, overexpression in E. coli and purification of the enzyme. Biochim. Biophys. Acta 1173, 121–132.
223
Firefly luciferase: the structure is known, but the mystery remains Thomas O Baldwin The structure of firefly luciferase reveals a new protein fold which may be representative of a growing family of homologous enzymes. Address: Department of Biochemistry and Biophysics and Department of Chemistry, Center for Macromolecular Design, Texas A&M University, College Station, TX 77843–2128, USA. Structure 15 March 1996, 4:223–228 © Current Biology Ltd ISSN 0969-2126
From the Atlantic Ocean to the eastern face of the Rocky Mountains, a summer sunset is usually accompanied by one of Nature’s most delightful light shows. Photinus pyralis, the North American firefly, has entertained countless observers, probably since the coming of humans to the continent. During the period between the 1950s and the 1980s, numerous young biologists earned their spending money as firefly collectors, first employed by Professor William D McElroy, then of Johns Hopkins University, and later as members of the Sigma Firefly Club. In 1985, when Marlene DeLuca and her colleagues cloned the cDNA encoding the luciferase of P. pyralis, an alternative source of the enzyme became available, and soon after, in laboratories around the world, numerous other organisms began to emit the characteristic yellow-green luminescence as a consequence of expression of the firefly luciferase in their cells. Historical perspective
Luciferase is a generic term describing any enzyme that catalyzes a reaction yielding visible light. Light emission is a consequence of formation of a product or intermediate in an electronically excited state; return to the ground state occurs via emission of a photon of light. Luciferases are highly diverse, catalyzing a great variety of reactions and using widely different substrates [1]. All have in common the involvement of oxygen. Luciferases are more different than, for example, proteases, which all carry out hydrolytic chemistry on peptide bonds. Luciferases all emit light, but they do so by very different means. Therefore, the luciferases from different organisms probably evolved independently, rather than from a common ancestral enzyme. Bacterial luciferase, the first luciferase to be cloned and also the first to be structurally characterized, is a flavin monooxygenase that utilizes flavin mononucleotide (FMN) to activate molecular oxygen, yielding a flavin C4a peroxide. Reaction of the peroxide with an aliphatic aldehyde substrate yields, ultimately, the carboxylic acid and the flavin C4a hydroxide in the first singlet excited state. Light emission, loss of the C4a hydroxide and dissociation of FMN returns the enzyme to
its starting condition. Firefly luciferase, on the other hand, catalyzes an oxidative reaction involving ATP, firefly luciferin and molecular oxygen, yielding an electronically excited oxyluciferin species [2]. This excited species emits visible light, which is employed by the firefly in its reproductive behavior [3]. Firefly luciferase was one of the first enzymes to be investigated in biochemical detail [4]. WD McElroy and his colleagues [5–9], as well as other investigators [10] working during the 1940s and 1950s with P. pyralis, determined the structures of the substrates and products of the enzyme. Evidence for the chemical mechanism proposed for the firefly luciferase reaction [11] has been discussed in a recent review [12]. The structures of luciferases
The structure of luciferase from the firefly P. pyralis has now been determined at high resolution and is reported by Conti et al. in this issue of Structure [13]. As beautifully described in their paper, the enzyme folds into two distinct domains, a large N-terminal domain comprising residues 4–436, and a C-terminal domain formed from residues 440–544. The structure is shown in Figure 1. The fold assumed by the luciferase polypeptide appears to be unique. The N-terminal domain consists of a b barrel and two b sheets flanked by a helices which form a five-layered ababa structure. The C-terminal domain, consisting of five b strands and three a helices, is folded into a compact structure that is connected to the N-terminal domain by a disordered loop (connecting residues 435 and 441). There are three other disordered loops not visible in the electron density, one in the C-terminal domain connecting residues 523 and 529, and two in the N-terminal domain (connecting residues 198–204 and residues 355–359). The structure presented is without bound substrates or other ligands. Conti et al. have taken advantage of the homology of firefly luciferase with other enzymes that catalyze similar reactions [14] to deduce the location of the active center (see Fig. 2). It is assumed that regions of greatest sequence conservation are most likely to be involved in the catalytic mechanism of the enzyme. Based on their extensive analysis, it is proposed that the active center is composed of residues on the surfaces of both domains, and that upon substrate binding, the domains move together to form the active center. The drawings presented and the comparisons made with the amino acid sequences of the large family of homologous proteins constitute a compelling argument. For a bioluminescence reaction to occur with a high quantum yield, as is the case for firefly luciferase, it is essential that water be excluded from the active site. It would appear that the two-domain structure presented by Conti et al. could serve such a purpose.
224
Structure 1996, Vol 4 No 3
Figure 1 Two orthogonal stereoviews of the surface of firefly luciferase, depicting an ‘anvil and hammer’ motif. Both views show the large N-terminal ‘anvil’ domain below the smaller C-terminal ‘hammer’ domain. The two views are the front view (top) and the right side view (bottom), obtained by rotating the front view through 90° to the left about the vertical axis. The color coding is the same as that used by Conti et al. [13], and indicates the domains and subdomains. The C-terminal domain (yellow) is the only domain composed exclusively of residues that are contiguous in the amino acid sequence, residues 440–544. The three subdomains of the N-terminal anvil consist of stretches that are not contiguous within the overall sequence. Subdomain A (blue) consists of residues 77–222 and 399–405. Subdomain B (purple) consists of residues 22–70 and 236–351. Subdomain C (green) consists of residues 4–10, 363–393 and 418–434. It appears that the active center comprises residues between the anvil and hammer, and it is suggested that the active center forms by movement of these two domains together following substrate binding [13]. (Figure courtesy of Peter Brick.)
The structure of bacterial luciferase has also been determined to high resolution without bound ligands [15]. Bacterial luciferase is a heterodimer composed of homologous subunits, a and b (see [16] for a review of the bacterial luciferase literature). Both subunits assume the wellknown (b/a)8, or TIM barrel structure, and are packed together by a parallel four-helix bundle. The active center is confined to the a subunit, and it has been proposed to reside within a large internal cavity which opens through a narrow crevice [17]. The opening to the cavity lies beneath a long disordered loop which appears to undergo a conformational rearrangement upon flavin binding and catalysis [16–20]. It has been proposed that this conformational change may be necessary to exclude water from the intermediates and the excited state of the flavin. Thus, it
appears reasonable to propose that both luciferases exclude water from their respective excited emitters by a conformational rearrangement that occurs upon or following substrate binding. Reaction catalyzed by firefly luciferase
Firefly luciferase catalyzes a multistep reaction [21]. In the first step, luciferin (compound I; Fig. 3) reacts with Mg2+-ATP to form luciferyl adenylate (compound II) and pyrophosphate. The luciferyl adenylate is oxidized by molecular oxygen, with the intermediate formation of the cyclic peroxide, a dioxetanone (compound III), and a molecule of AMP. The dioxetanone is decarboxylated as a result of intramolecular conversions (compound IV) to produce an electronically excited state of oxyluciferin in
Minireview Luciferases Baldwin
225
Figure 2
Alignment of the amino acid sequences, reported by Devine et al. [49], of the luciferases of Photinus pyralis (P.p), Luciola mingrelica (L.m), L. cruciata (L.c), L. lateralis (L.l), and the green-emitting strain of the click beetle (CbG). Also shown in this alignment is the 4-coumarate:CoA ligase (CoA). References to the sequences are given in [49]. The numbering at the top refers to the sequence of the
luciferase from P. pyralis. The extent of conservation at each position is indicated by the following color code: red=fully conserved in all six sequences; pink=2 different amino acids; green=3 different amino acids, and blue=4–6 different amino acids. At positions where a deletion has occurred in one or more of the proteins, the corresponding residues in the other proteins are shown in black.
the enol or keto form (compound V). Return to the ground state is accompanied by emission of a quantum of visible light with a wavelength of maximum light intensity (Imax) of 562–570 nm. Shimomura et al. [22] demonstrated that one oxygen atom of the product CO2 arises from the substrate oxygen. Non-enzymatic oxidation of luciferin yields oxyluciferin without luminescence [23]. (For a recent review, see [12].)
cruciata and Luciola lateralis [24], have been purified and characterized. The L. mingrelica and L. cruciata luciferases are similar to the P. pyralis luciferase, with Imax of 562–570 nm, whereas the reaction catalyzed by L. lateralis luciferase emits green light (Imax 552 nm). All Luciola luciferases inactivate rapidly at low ionic strength [21], whereas the P. pyralis luciferase crystallizes in active form under the same conditions [4].
Color of the bioluminescence of fireflies
The chemiluminescence emission spectra of a variety of luciferin analogs were studied by White et al. [9,23]. Based on the fluorescence emission spectra of these compounds in the presence and absence of proton acceptors, they suggested that the red emission arises from the keto anion form of the product molecule and the yellow-green
One of the most intriguing aspects of firefly bioluminescence pertains to the color of the light emitted. Recently, the luciferase from Luciola mingrelica fireflies, collected in the southern regions of Russia [21], as well as the luciferases of the fireflies indigenous to Japan, Luciola
226
Structure 1996, Vol 4 No 3
Figure 3 Proposed mechanism of the firefly bioluminescence reaction. The carboxylate group of firefly luciferin (I) is activated by reaction with ATP to form the adenylated luciferin (II). The a proton is lost, allowing reaction with molecular oxygen to yield (III). Cleavage of the dioxetanone ring (IV) yields the excited state of oxyluciferin (V).
emission from the enol anion. This hypothesis is consistent with the emission of red light in the bioluminescence reaction at lower pH, and has led to the suggestion that the various colors of light observed from different species of firefly result from a spectral mixing from these two emitters [5]. However, when luciferases were isolated from different organs of the click beetle Pyrophorus plagiophthalamus, each emitted a different color of bioluminescence, and each emission spectrum showed a single peak, not a superimposition of two or more different spectra [24,25]. The cDNAs of four luciferases from P. plagiophthalamus, chosen on the basis of differences in color of bioluminescence, have been cloned and sequenced [26]. The amino acid sequences of these luciferases are 95–99% identical, and fewer than two or three amino acid changes are needed for spectral shifts of up to 50 nm in Imax. The properties of these luciferases require some rethinking of the traditional explanations for spectral differences between insect luciferases, which historically had been attributed to shifts in the proposed equilibrium distribution between the keto and enol forms of the excited-state oxyluciferin product. If this were the sole explanation for the spectral variations, one would expect to see intermediate bimodal spectra, as have been reported for P. pyralis luciferase titrated with zinc ions [24]. However, the four recombinant click beetle luciferases emit light with sharp emission spectra with spectral maxima of 546 nm (green), 560 nm (yellow-green), 578 nm (yellow), and 593 nm (orange) [27]. This finding suggests that light emission from each enzyme occurs from a single molecular species within the environment of the enzyme and that differences in color result from differences between the microenvironments of the enzyme–oxyluciferin complexes. In a related study, L. cruciata luciferase cDNA was mutagenized and five mutants with different colors of bioluminescence ranging from green to red were isolated [28]. The mutations were found to be single amino acid changes. Of particular interest is
the fact that a large (50 nm) red shift (to an Imax of 612 nm) was caused by a single substitution of an invariant histidinyl residue (conserved in all insect luciferases) by a tyrosine (His433→Tyr). When the sequences of the mutant L. cruciata luciferases were compared with those of the click beetle enzymes, no common amino acid sequence affecting the color of light was apparent [26]. It may be concluded that the different colors of bioluminescence are caused by subtle changes in the tertiary structure of the luciferase molecule. The firefly luciferase superfamily of enzymes
The number of proteins related to firefly luciferase and apparently descended from a common ancestor is growing rapidly. These enzymes are involved in the biosynthesis of iron-binding siderophores [29,30], the antibiotics gramicidin S [31,32], tyrocidine [33], and penicillin [34–36], and pathogen defense agents in plants [37,38] and the biodegradation of halogenated aromatics [39]. Fatty acid:coenzyme A (CoA) ligase [40] and acetate:CoA ligase [41] use the same mechanism as 4-coumarate:CoA ligase [37] and the proteins show a high degree of sequence similarity. The amino acid sequence of AngR, a DNA-binding protein that modulates iron-regulated transcription, also shows homology to the firefly luciferases [42]. Given the diversity and importance of the related proteins already discovered, more are likely to be found as the number of sequenced proteins increases. Each of the enzymes catalyzes the adenylation of a carboxylic acid substrate using Mg2+-ATP (Fig. 4), followed by reaction of the activated carboxylate with an acceptor and release of AMP. Several of the related enzymes use CoA or an active-site thiol as an acceptor for displacing the AMP. There does not appear to be an essential thiol in the active site of firefly luciferase [43], although there have been several reports that the addition of CoA enhances the emission of light from firefly luciferase [43–45]. The role, if any, that CoA plays in the firefly luciferase reaction remains to be elucidated.
Minireview Luciferases Baldwin
Figure 4
227
non-fluorescent flavoprotein (see [16] and other references therein) — the function of which is unknown. Even as our knowledge of the structures and mechanisms of these delightful enzymes becomes more detailed, the beauty of the biological phenomenon of bioluminescence will not be diminished. The phenomenon of a flashing firefly will always be a source of joy to those who pause to observe it. The mysterious emission of light without heat will always intrigue the curious; and for the technically more sophisticated, the questions regarding the color of bioluminescence, and the details of the enzyme-catalyzed reaction remain to be resolved. Knowledge of the structure is a major step forward but there is still much to be learned. References
Comparisons of representative reactions catalyzed by members of the firefly luciferase family of enzymes. Reaction (a) is catalyzed by firefly luciferase [8]; reaction (b) by 4-coumarate CoA:ligase [37]; reaction (c) by gramicidin S synthetase I and tyrocidine synthetase [31,32]; and reaction (d) by 2,3-dihydroxybenzoate-AMP ligase [29].
Summary
In recent years, luciferases and the genes that encode them have become very useful for research purposes as well as for various commercial processes. The applications are too numerous to discuss here, but examples include the use of firefly luciferase to study the role of chaperones in protein folding [46] and of bacterial luciferase to study co-translational folding of polypeptides [47] and of the genes encoding luciferases to monitor transcriptional activities [45,48]. Numerous applications are described in the various publications of the International Symposia on Bioluminescence and Chemiluminescence published by Wiley. Clearly, the structures of firefly luciferase and bacterial luciferase will be of great value in further developing the applications and our basic understanding of these intriguing enzymes. Detailed mechanistic insights must await structural elucidation of the various complexes of the luciferases with their substrates and products, so there is much structural work yet to be done. It is interesting to note that the subunits of bacterial luciferase assume perhaps the most common of folding motifs, the (b/a)8 barrel, whereas firefly luciferase folds into a new structural form. While there are numerous homologs of firefly luciferase that catalyze similar reactions and have similar amino acid sequences, there is only one known non-luciferase homolog of bacterial luciferase — the
1. Hastings, J.W. (1983). Biological diversity, chemical mechanisms, and the evolutionary origins of bioluminescent systems. J. Mol. Evol. 19, 309–321. 2. DeLuca, M. (1976). Firefly luciferase. Advances in Enzymology and Related Areas of Molecular Biology 44, 37–68. 3. Case, J.F. & Strause, L.G. (1978). Neurally controlled luminescent systems. In Bioluminescence in Action. (Herring, P., ed), pp. 332–366, Academic Press, NY. 4. McElroy, W.D. (1960). Crystalline firefly luciferase. Methods Enzymol. 5, 445–448. 5. Seliger, M.M. &. McElroy, W.D. (1960). Spectral emission and quantum yield of firefly bioluminescence. Arch. Biochem. Biophys. 88, 136–141. 6. White, E.H., McCapra, F., Field, G.F. & McElroy, W.D. (1961). The structure and synthesis of firefly luciferin. J. Am. Chem. Soc. 83, 2402–2403. 7. Seliger, H.H. & McElroy, W.D. (1959). Quantum yield in the oxidation of firefly luciferin. Biochem. Biophys. Res. Commun. 1, 21–24. 8. DeLuca, M. & McElroy, W.D. (1974). Kinetics of the firefly luciferase catalyzed reactions. Biochemistry 13, 921–925. 9. White, E.H., Rapaport, E., Hopkins, T.A. & Seliger, H.H. (1969). Chemi- and bioluminescence of firefly luciferin. J. Am. Chem. Soc. 91, 2178–2180. 10. Suzuki, N. & Goto, T. (1972). Studies on firefly bioluminescence — II. Identification of oxyluciferin as a product in the bioluminescence of firefly lanterns and in the chemiluminescence of firefly luciferin. Tetrahedron 28, 4075–4082. 11. McElroy, W.D. & DeLuca, M. (1978). Chemistry of firefly luminescence. In Bioluminescence in Action. (Herring, P.J., ed), pp. 109–127, Academic Press, London. 12. Wood, K.V. (1995). The chemical mechanism and evolutionary development of beetle bioluminescence. Photochem. Photobiol. 62, 662–673. 13. Conti, E., Franks, N.P. & Brick, P. (1996). Crystal structure of firefly luciferase throws light on a superfamily of adenylate-forming enzymes. Structure 4, 287–298. 14. Babbitt, P.C., et al., & Dunaway-Mariano, D. (1992). Ancestry of the 4-chlorobenzoate dehalogenase: analysis of aminoacid sequence identities among families of acyl:adenyl ligases, enoyl-CoA hydratases/isomerases, and acyl-CoA thioesterases. Biochemistry 31, 5594–5604. 15. Fisher, A.J., Raushel, F.M., Baldwin, T.O. & Rayment, I. (1995). The three-dimensional structure of bacterial luciferase from Vibrio harveyi at 2.4 Å resolution. Biochemistry 34, 6581–6586. 16. Baldwin, T.O. & Ziegler, M.M. (1992). The biochemistry and molecular biology of bacterial bioluminescence. In Chemistry and Biochemistry of Flavoenzymes. (Müller, F, ed), vol. III, pp. 467–530, CRC Press, Boca Raton, FL. 17. Baldwin, T.O., et al., and Rayment, I. (1995). Structure of bacterial luciferase. Curr. Opin. Struct. Biol. 5, 798–809. 18. Baldwin, T.O. & Riley, P.L. (1980). Anion binding to bacterial luciferase: evidence for binding associated changes in enzyme structure. In Flavins and Flavoproteins. (Yagi, K. & Yamano, T, eds), pp. 139–147, Japan Scientific Societies Press, Tokyo and University Park Press, Baltimore.
228
Structure 1996, Vol 4 No 3
19. Holzman, T.F. & Baldwin, T.O. (1980). Proteolytic inactivation of luciferases from three species of luminous marine bacteria, Beneckea harveyi, Photobacterium fischeri, and Photobacterium phosphoreum: evidence of a conserved structural feature. Proc. Natl. Acad. Sci. USA 77, 6363–6367. 20. AbouKhair, N.K., Ziegler, M.M. & Baldwin, T.O. (1985). Bacterial luciferase: demonstration of a catalytically competent altered conformational state following a single turnover. Biochemistry 24, 3942–3947. 21. Ugarova, N.N. (1989). Luciferase of Luciola mingrelica fireflies. kinetics and regulation mechanism. J. Biolumin. Chemilumin. 4, 406–418. 22. Shimomura, O., Goto, T. & Johnson, F.H. (1977). Source of oxygen in the CO2 produced in the bioluminescent oxidation of firefly luciferin. Proc. Natl. Acad. Sci. USA 74, 2799–2802. 23. White, E.H., Rapaport, E., Seliger, H.H. & Hopkins, T.A. (1971). The chemi- and bioluminescence of firefly luciferin: an efficient chemical production of electronically excited states. Bioorg. Chem. 1, 92–122. 24. Seliger, H.H. & McElroy, W.D. (1964). The colors of firefly bioluminescence: enzyme configuration and species specificity. Proc. Natl. Acad. Sci. USA 52, 75–81. 25. Seliger, H.H., Buck, J.B., Fastie, W.G. & McElroy, W.D. (1964). The spectral distribution of firefly light. J. Gen. Physiol. 48, 95–104. 26. Wood, K.V., Lam, Y.A., Seliger, H.H. & McElroy, W.D. (1989). Complementary DNA coding click beetle luciferases can elicit bioluminescence of different colors. Science 244, 700–702. 27. Wood, K.V., Lam, Y.A. & McElroy, W.D. (1989). Introduction to beetle luciferases and their applications. J. Biolumin. Chemilumin. 4, 289–301. 28. Kajiyama, N. & Nakano, E. (1991). Isolation and characterization of mutants of firefly luciferase which produce different colors of light. Protein Eng. 4, 691–693. 29. Rusnak, F., Faraci, W.S. & Walsh, C.T. (1989). Subcloning, expression, and purification of the enterobactin biosynthetic enzyme 2,3dihydroxybenzoate-AMP ligase: demonstration of enzyme-bound (2,3dihydroxybenzoyl)adenylate product. Biochemistry 28, 6827–6835. 30. Rusnak, F., Sakaitani, M., Drueckhammer, D., Reichert, J. & Walsh, C.T. (1991). Biosynthesis of the Escherichia coli siderophore enterobactin: sequence of the entF gene, expression and purification of EntF, and analysis of covalent phosphopantetheine. Biochemistry 30, 2916–2927. 31. Kanda, M., Hori, K., Kurotsu, T., Miura, S. & Saito, Y. (1989). Reaction mechanism of gramicidin synthetase 1, phenylalanine racemase of Bacillus brevis. J. Biochem. (Tokyo) 105, 653–659. 32. Kratzschmar, J., Krause, M. & Maraheil, M.A. (1989). Gramacidin S biosynthesis operon containing the structural genes grsA and grsB has an open reading frame encoding a protein homologous to fatty acid thioesterases. J. Bacteriol. 171, 5422–5429. 33. Roskoski, R., Jr., Gevers, W., Kleinkauf, H. & Lipmann, F. (1970). Tyrocidine biosynthesis by three complementary fractions from Bacillus brevis (ATCC 8185). Biochemistry 9, 4839–4845. 34. Kleinkauf, H. & von Döhren, H. (1990) Nonribosomal biosynthesis of peptide antibiotics. Eur. J. Biochem. 192, 1–15. 35. Gutierrez, S., Diez, B., Montenegro, E. & Martin, J.F. (1991). Characterization of the Cephalosporium acremonium pcbAB gene encoding a-aminoadipyl-cysteinyl-valine synthetase, a large multidomain peptide synthetase: linkage to the pcbC gene as a cluster of early cephalosporin biosynthetic genes and evidence of multiple functional domains. J. Bacteriol. 173, 2354–2365. 36. MacCabe, et al., & Kinghorn, J.R. (1991). D-(L-a-aminoadipyl)-Lcysteinyl-D-valine synthetase from Aspergillus nidulans. J. Biol. Chem. 266, 12646–12654. 37. Knobloch, K.H. & Hahlbrock, K. (1977). 4-Coumarate:CoA ligase from cell suspension cultures of Petroselinum hortense Hoffm. Partial purification, substrate specificity, and further properties. Arch. Biochem. Biophys. 184, 237–248. 38. Becker-Andre, M., Schulze-Lefert, P. & Hahlbrock, K. (1991). Structural comparison, modes of expression, and putative cis-acting elements of the two 4-coumarate:CoA ligase genes in potato. J. Biol. Chem. 266, 8551–8559. 39. Scholten, J.D., Chang, K.H., Babbitt, P.C., Charest, H., Sylvestre, M. & Dunaway-Mariano, D. (1991). Novel enzymic hydrolytic dehalogenation of a chlorinated aromatic. Science 253, 182–185. 40. Suzuki, H., et al., & Yamomoto, T. (1990). Structure and regulation of rat long-chain acyl-CoA synthetase. J. Biol. Chem. 265, 8681–8685. 41. Connerton, I.F., Fincham. J.R.S., Sandeman, R.A. & Hynes, M.J. (1990). Comparison and cross species expression of the acetyl-CoA synthetase genes from the Ascomycete fungi Aspergillus nidulans and Neurospora crassa. Mol. Microbiol, 1, 451–459.
42. Farrel, D.H., Mikesell, P., Actis, L.A. & Crosa, J.H. (1990). A regulatory gene, angR, of the iron uptake system of Vibrio anguillarium: similarity with phage P22 cro and regulation by iron. Gene 86, 45–51. 43. Alter, S.C. & DeLuca, M. (1986). The sulfhydryls of firefly luciferase are not essential for activity. Biochemistry 25, 1599–1605. 44. Wood, K.V. (1992). Improved kinetics of light production by beetle luciferase using thiol reagents. Chem. Abstr. 116, 230750n. 45. Pazzagli, M., Devine, J.H., Peterson, D.O. & Baldwin, T.O. (1992). Use of bacterial and firefly luciferases as reporter genes in DEAE/dextranmediated transfection of mammalian cells. Anal. Biochem. 204, 315–323. 46. Szabo, A., Langer, T., Schröder, H., Flanagan, J., Bukau, B. & Hartl, F.U. (1994). The ATP hydrolysis-dependent reaction cycle of the Escherichia coli Hsp70 system — DnaK, DnaJ, and GrpE. Proc. Natl. Acad. Sci. USA 91, 10345–10349. 47. Fedorov, A.N. & Baldwin, T.O. (1995). Contribution of cotranslational folding to the rate of formation of native protein structure. Proc. Natl. Acad. Sci. USA 92, 1227–1231. 48. Wood, K.V. (1991). Recent advances and prospects for use of beetle luciferases as genetic reporters. In Bioluminescence and Chemiluminescence: Current Status. (Stanley, P.E. & Kricka, L.J., eds), pp. 543–546, John Wiley and Sons, Chichester. 49. Devine, J.H., Kutuzova, G., Green, V.W., Ugarova, N. & Baldwin, T.O. (1993). Luciferase from the European firefly Luciola mingrelica: cloning and nucleotide sequence of the cDNA, overexpression in E. coli and purification of the enzyme. Biochim. Biophys. Acta 1173, 121–132.