- Email: [email protected]
The quinohaemoprotein lupanine hydroxylase from Pseudomonas putida
Biochimica et Biophysica Acta 1647 (2003) 110 – 115 www.bba-direct.com
Review
The quinohaemoprotein lupanine hydroxylase from Pseudomonas putida David J. Hopper *, Mustak A. Kaderbhai Institute of Biological Sciences, University of Wales, Aberystwyth, Ceredigion SY23 3DD, UK Received 13 July 2002; received in revised form 31 October 2002; accepted 22 January 2003
Abstract Lupanine hydroxylase catalyses the first reaction in the catabolism of the alkaloid lupanine by Pseudomonas putida. It dehydrogenates the substrate, which can then be hydrated. It is a monomeric protein of Mr 72,000 and contains a covalently bound haem and a molecule of PQQ. The gene for this enzyme has been cloned and sequenced and the derived protein sequence has a 26 amino acid signal sequence at the Nterminal for translocation of the protein to the periplasm. Many of the features seen in the sequence of lupanine hydroxylase are common with other quinoproteins including the W-motifs that are characteristic of the eight-bladed propeller structure of methanol dehydrogenase. However, the unusual disulfide bridge between adjacent cysteines that is present in some PQQ-containing enzymes is absent in lupanine hydroxylase. The C-terminal domain contains characteristics of a cytochrome c and overall the sequence shows similarities with that of the quinohaemoprotein, alcohol dehydrogenase from Comamonas testosteroni. The gene coding for lupanine hydroxylase has been successfully expressed in Escherichia coli and a procedure has been developed to renature and reactivate the enzyme, which was found to be associated with the inclusion bodies. Reactivation required addition of PQQ and was dependent on calcium ions. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Quinohaemoprotein; Cytochrome; PQQ; Renaturation; Reactivation; Lupanine
1. Introduction Lupanine is one of a class of alkaloids known as the quinolizidine alkaloids and is found particularly in plants of the genus Lupinus [1,2]. These alkaloids give the plants some resistance to pathogens and also to herbivores as they are highly toxic to animals, inducing a lethal state of lupinosis. Besides this protective role, it is thought that they can also serve as nitrogen stores for developing seedlings [3]. In common with all plant products, degradation of these alkaloids by soil microorganisms is important in mobilization of the organic carbon in the carbon cycle. A number of strains of bacteria have been isolated that can grow on lupanine as their sole carbon and nitrogen source [4,5]. One of these was first isolated from soil by Moz˙ejko-Toczko in 1960 and identified as a Pseudomonas sp., while more recently, seven Gram-negative bacterial strains were isolated from soil in which lupins (Lupinus albus L.) had been grown [5]. Three of these were identified, two as Xanthomonas oryzae and one as Gluconobacter cerinus. * Corresponding author. Tel.: +44-1970-622-292; fax: +44-1970-622-307. E-mail address: [email protected] (D.J. Hopper).
Toczko et al. [6] reported that the first intermediate of lupanine breakdown by their Pseudomonas sp. is 17hydroxylupanine [6] produced by an enzyme referred to as lupanine hydroxylase.
2. Lupanine hydroxylase The enzyme, lupanine hydroxylase, was partially purified and shown to contain haem [7]. It required an electron acceptor for activity rather than a donor, as would be required for a monooxygenase type of hydroxylase. This suggested a mechanism by which the substrate is dehydrogenated and subsequently hydrated, with the oxygen of the hydroxyl group coming from water. It should be noted, however, that 17-hydroxylupanine is a carbinolamine with a pK of 10.5 and that its perchlorate and picrate salts analyse as the anhydronium form [8,9] (Fig. 1). The Pseudomonas enzyme is inducible by growth of the organism on (+) lupanine as a carbon source and is not active with the ( ) enantiomer [10]. Growth of the organism for enzymic studies depends, therefore, on a supply of lupanine, which is not commercially available. The need for the substrate to be isolated from lupin seeds
1570-9639/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1570-9639(03)00070-0
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
111
Fig. 1. The reaction catalysed by lupanine hydroxylase.
has dictated the pace of progress in the characterisation of this enzyme. Although another inducer of lupanine hydroxylase is 15-oxosparteine [11], this also is not readily available. Another related alkaloid is sparteine (Fig. 2); this is also a substrate for lupanine hydroxylase and is commercially available but, unfortunately, fails to act as an inducer of the enzyme [11]. Eventually, the enzyme was purified to homogeneity by various chromatographic techniques and some of its properties characterised [12]. The UV/VIS spectrum was that of a c-type cytochrome and a value of one haem per molecule of enzyme was calculated using the absorbance coefficient of cytochrome c. Moreover, the haem was covalently bound as expected for a c-type cytochrome. The haem will only accept one electron, and for the two-electron dehydrogenase reaction, a second electron acceptor must be present. Several pieces of evidence showed that this is PQQ; extracts from the enzyme were able to reactivate the PQQ-requiring glucose dehydrogenase apoenzyme from Acinetobacter calcoaceticus [13]; inactive apoenzyme produced by isoelectric focussing was reactivated by addition of PQQ, and the oxidised spectrum of this apoenzyme was only reduced by substrate if PQQ was also added. Titration of the apoenzyme with PQQ showed that full activity was recovered on the addition of one molecule of PQQ per molecule of enzyme [12]. Interaction of the enzyme with PQQ was also demonstrated by the effect of the cofactor on the midpoint redox potential of the haem group. This was + 193 mV for the
holoenzyme but shifted to + 98 mV when PQQ was removed. Addition of PQQ to this apoenzyme restored the original redox potential, clearly showing that binding of the cofactor affected the environment of the haem in the protein [14]. A similar shift in the midpoint redox potential has been described for another PQQ-containing quinohaemoprotein, ethanol dehydrogenase from Comamonas testosteroni, in which there was a shift from + 80 mV in the apoenzyme to + 140 mV in the holoenzyme [15]. Indeed, the lupanine hydroxylase has a number of other features in common with the ethanol dehydrogenase. It is found in the periplasmic compartment of the cell and is a monomeric protein Mr about 70,000. Gel filtration and ultracentrifuge studies gave values of 74,000 and 66,000, respectively, for the native lupanine hydroxylase while SDS-PAGE gave a value of 72,000 [12]. The assay developed for monitoring lupanine hydroxylase activity used horse heart cytochrome c as the electron acceptor and its reduction was followed spectrophotometrically by the increase in absorbance at 550 nm. Steady-state kinetics using this assay gave results suggestive of a pingpong mechanism with a Km value of 3.6 F 0.7 AM for lupanine and a kcat of 217 s 1 [12]. More recently, the same kinetic experiments have been performed with sparteine as substrate. Sparteine is another plant-produced quinolizidine alkaloid with a very similar structure to lupanine. The enzyme has a lower Km for sparteine with a value of 0.87 F 0.11 AM, but the kcat is also much lower at 11.8 s 1. Thus, the specificity constant for lupanine is higher at 60.3 AM 1 s 1 compared to 13.6 AM 1 s 1 for sparteine.
3. Cloning and sequence of the gene for lupanine hydroxylase
Fig. 2. The structure of sparteine.
The gene for lupanine hydroxylase (luh) has now been isolated using PCR with chromosomal DNA as template and
112
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
primers based on the sequence of the N- and C-terminals of the purified enzyme [16]. This was sequenced (accession number AJ318095) and found to code for a protein of Mr 72,256, which is in good agreement with the value of 72,000 found by SDS-PAGE of the purified enzyme. The derived sequence included the sequences of several internal peptides, isolated after digestion of the purified enzyme and then sequenced by Edman degradation, which suggests that the correct gene had been isolated. The ORF for the gene, however, starts at base 78 before the codon for the first amino acid of the mature protein. This was found by sequencing upstream of the coding region for the mature protein. Information on this part of the gene was obtained by amplifying a fragment of DNA encompassing the 5V end of the coding sequence for the mature protein and some of the upstream DNA by ligation-mediated PCR followed by
sequencing of the product. The protein sequence derived from this showed a 26 amino acid signal sequence, consistent with the need for translocation of the protein across the cell membrane to the periplasm. Analysis by the SignalP V2.0 computer program [17] gave a signal peptide probability of 1.000 and predicted the signal peptidase cleavage site to be between amino acids 26 and 27 with a probability of 0.986. The derived protein sequence showed similarities to those of many other PQQ-containing quinoproteins and particularly to the quinohaemoprotein, ethanol dehydrogenase from C. testosteroni, with which it had 36% identity in a BLAST search [18] (Fig. 3). The crystal structure of the ethanol dehydrogenase [19] shows the eight-bladed propeller structure that was described for the a-subunit PQQcontaining methanol dehydrogenase [20]. Alignment of the
Fig. 3. Alignment of the amino acid sequence of lupanine hydroxylase with that of the quinohaemoprotein ethanol dehydrogenase from C. testosteroni (edh). Residues that are common are shaded in black. The sequences above the asterisks are tryptophan docking motifs of the eight-bladed propeller structure found in the ethanol dehydrogenase and in methanol dehydrogenase. The haem-binding regions in the cytochrome c domains are italicised and doubly underlined. The signal sequence for each protein is shown in lower case.
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
113
sequence for lupanine hydroxylase with those for methanol dehydrogenase and ethanol dehydrogenase showed the presence of the eight W-motifs characteristic of this tertiary structure in the N-terminal domain suggesting a similar tertiary structure. The C-terminal domain incorporates the cytochrome c haem-binding consensus sequence of CXXCH as CGACH and the methionine that provides the sixth ligand to the iron in the ethanol dehydrogenase is conserved in lupanine hydroxylase (Fig. 3). Also conserved is the tryptophan residue that forms the floor of the active site chamber in both the methanol and ethanol dehydrogenases but a notable difference is the apparent absence in lupanine hydroxylase of the novel disulfide ring structure formed between adjacent cysteines that is a feature of the active site in the other two enzymes and in some other PQQ-enzymes.
4. Heterologous expression of the luh gene Conclusive evidence that the cloned luh gene is that of lupanine hydroxylase requires the heterologous expression of the gene to give active enzyme. Several attempts were made to express the gene in Escherichia coli but these were unsuccessful. Possible reasons for this are the incompatibility of the signal sequence with the secretion pathway in E. coli and the general difficulty in cytochrome c expression/maturation due to the requirement for haem insertion. To overcome these problems, we replaced the 26 amino acid signal sequence with the 21 residue alkaline phosphate
Fig. 4. Plasmid pEV-LH32 containing the gene for lupanine hydroxylase (LH) for expression in E. coli. SS is the signal sequence of alkaline phosphatase. S/D is the Shine/Delgarno sequence. Ppho is the phoA promotor. Ori is the origin of plasmid replication and Ampr codes for hlactamase.
Fig. 5. Reactivation of apoenzyme. Enzyme was incubated with 1 mM EGTA to chelate any calcium ions present in the preparation. It was then incubated at 20 jC with 0.05 mM PQQ (D), 0.05 mM PQQ and 2 mM CaCl2 (.), 2 mM CaCl2 (n), 0.05 mM PQQ and 0.1 mM lupanine (E) and 0.05 mM PQQ, 2 mM CaCl2, 0.1 mM lupanine (o). Samples were taken at timed intervals and assayed for lupanine hydroxylase activity.
signal sequence that is native to E. coli by ligation of the region of the luh gene encoding the mature protein into plasmid pLiQ [21] to give plasmid pEV-LH32 (Fig. 4). This also placed the luh gene under the control of the phoA
Fig. 6. Reactivation of apoenzyme with various cations. Enzyme was incubated at 20 jC with 1 mM EGTA to chelate any calcium ions in the preparation and then with 0.05 mM PQQ and 2 mM CaCl2 (.), MgCl2 (o), BaCl2 (E) or SrCl2 (D). Samples were taken at timed intervals and assayed for lupanine hydroxylase activity.
114
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
Fig. 7. SDS-PAGE analysis of recombinant lupanine hydroxylase in the inclusion bodies derived from E. coli harbouring either pEV-LH32 (lane 2) or pEV-LH32 and pEC86 (lane 3). The partially pure recombinant lupanine hydroxylase (arrow) is shown in lane 1.
promoter so that the tightly regulated gene expression can be switched on by shift to growth condition of low phosphate concentration. The recombinant plasmid was introduced into the E. coli host TB1 that already harboured the plasmid pEC86. This plasmid expresses constitutively the genes for cytochrome c maturation factors (ccm) contained on the ccm operon [22]. When the organism was grown (6 h aerobically followed by 12 h with low aeration) in phosphate-limiting medium (0.1 mM), a haem-containing protein of the same Mr as lupanine hydroxylase was produced, as shown by SDS-PAGE of total cell protein. However, this haemoprotein was present in amorphous inclusion bodies that were isolatable with the cell membranes following centrifugation of the lysed cells. Perhaps it was not surprising that the protein showed no enzyme activity as no PQQ had been included in the growth medium. Nevertheless, a procedure for the solubilisation and reactivation of the enzyme was developed. This involved the solubilisation of the isolated inclusion bodies with 8 M urea followed by rapid dilution (10-fold) with a buffered solution containing EDTA (1 mM) to yield a tancoloured soluble fraction. Incubation of this ‘renatured’ protein with PQQ gave some enzyme activity, which was further enhanced significantly by addition of Ca2 +. Chelation of any adventitious Ca2 + ions by pre-incubation with 1 mM EGTA prior to PQQ addition alone yielded no enzymic activity. Thus it could be demonstrated that both PQQ and an excess of Ca2 + ions were essential for activity (Fig. 5) as has been shown for other PQQ-containing enzymes. Other divalent cations such as Sr2 + and Ba2 +
were equally competent at reactivating the enzyme in conjunction with PQQ (Fig. 6). However, Mg2 + was much less effective. The importance of co-expression of the ccm factors for the production of enzyme with the haem correctly attached was demonstrated by comparing the activities of lupanine hydroxylase recovered in the renatured and reactivated inclusion bodies derived from cells that harboured either only the luh gene (pEV-LH32) or both the luh and ccm gene clusters (pEV-LH32 and pEC86) and grown under identical inducing conditions. A protein of the correct Mr was produced in the single plasmid-containing cells (Fig. 7) but it did not contain as much haem and could not be reactivated simply by addition of haem to the isolated protein. The specific activity of the enzyme in the reactivated extract of cells containing both plasmids was 11-fold higher than that from cells lacking the plasmid coding for the ccm operon. Thus we have been able to express the luh gene heterologously to obtain active enzyme and are, therefore, confident that this is the gene for lupanine hydroxylase. Furthermore, this allows us to produce sufficient enzyme, without the constraints of the limited availability of lupanine for growth, for further structural and other studies. References [1] M. Wink, C. Meissner, L. Witte, Patterns of quinolizidine alkaloids in 56 species of the genus Lupinus, Phytochemistry 38 (1995) 139 – 153. [2] M. Muzquiz, C. Cuadrado, G. Ayet, C. de la Cuadra, C. Burbano, A. Osagie, Variation of alkaloid components of lupin seeds in 49 genotypes of Lupinus albus L. from different countries and locations, J. Agric. Food Chem. 42 (1994) 1447 – 1450. [3] M. Wink, L. Witte, Quinolizidine alkaloids as nitrogen source for lupin seedlings and cell cultures, Z. Naturforsch. 40c (1985) 767 – 775. [4] M. Moz˙ejko-Toczko, Rozklad lupaniny przez Pseudomonas lupanine, Acta Microbiol. Pol. 9 (1960) 157 – 171. [5] F.M.C. Santana, A.M. Fialho, I. SaCorreia, J.M.A. Empis, Isolation of bacterial strains capable of using lupanine, the predominant quinolizidine alkaloid in white lupin, as sole carbon and energy source, J. Ind. Microbiol. 17 (1996) 110 – 115. [6] M. Toczko, W. Brzeski, K. Kaßkolewska-Baniuk, Microbial degradation of lupanine: V. Identification of 17-hydroxylupanine, Bull. Acad. Pol. Sci., Cl. II, 11 (1963) 161 – 164. [7] J. Rogozin´ski, Molecular properties of the inducible lupanine hydroxylase from growing cultures of Pseudomonas lupanine, Acta Biochim. Pol. 32 (1975) 57 – 66. [8] O.E. Edwards, F.H. Clarke, B. Douglas, 17-Hydroxylupanine and 17-oxylupanine, Can. J. Chem. 32 (1954) 235 – 241. [9] J. Thiel, W. Boczon, W. Wysocka, A stereostructural study of 17-hydroxylupanine and its perchlorate, Monatsh. Chem. 131 (2000) 1073 – 1081. [10] A. Niedzielska, J. Rogozin´ski, Inducible lupanine hydroxylase from resting cultures of Pseudomonas lupanini, Bull. Acad. Pol. Sci., Cl. II, XXI (1973) 1 – 5. [11] J. Droese, Inducers and substrates of inducible enzymes in the Pseudomonas sp. isolated from soil and degrading lupanine, Acta Microbiol. Pol., Ser. B 2 (1970) 95 – 101. [12] D.J. Hopper, J. Rogozin´ski, M. Toczko, Lupanine hydroxylase, a quinocytochrome c from an alkaloid-degrading Pseudomonas sp., Biochem. J. 279 (1991) 105 – 109.
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115 [13] R.A. van der Meer, B.W. Groen, M.A.G. van Kleef, J. Frank, J.A. Jongejan, J.A. Duine, Isolation, preparation, and assay of pyrroloquinoline quinone, Methods Enzymol. 188 (1990) 260 – 283. [14] D.J. Hopper, J. Rogozin´ski, Redox potential of the haem c group in the quinocytochrome, lupanine hydroxylase, an enzyme located in the periplasm of a Pseudomonas sp., Biochim. Biophys. Acta 1383 (1998) 160 – 164. [15] G.A.H. de Jong, J. Caldeira, J. Sun, J.A. Jongejan, S. de Vries, T.M. Lohr, I. Moura, J.A. Duine, Characterization of the interaction between PQQ and heme c in the quinohaemoprotein ethanol dehydrogenase from Comamonas testosteroni, Biochemistry 34 (1995) 9451 – 9458. [16] D.J. Hopper, M.A. Kaderbhai, S.A. Marriott, M. Young, J. Rogozin´ski, Cloning, sequencing and heterologous expression of the gene for lupanine hydroxylase, a quinocytochrome c from a Pseudomonas sp., Biochem. J. 369 (2002) 483 – 489. [17] H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Identification of
[18]
[19]
[20]
[21]
[22]
115
prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites, Protein Eng. 10 (1997) 1 – 6. S.F. Atschul, T.L. Madden, A.A. Schaffer, J.H. Zhang, Z. Zhang, W. Miller, D.J. Lipman, Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Nucleic Acids Res. 25 (1997) 3389 – 3402. A. Oubrie, H.J. Rozeboom, K.H. Kalk, E.G. Huizinga, B.W. Dijkstra, Crystal structure of quinohaemoprotein alcohol dehydrogenase from Comamonas testosteroni, J. Biol. Chem. 277 (2002) 3727 – 3732. M. Ghosh, C. Anthony, K. Harlos, M.G. Goodwin, C.C.F. Blake, The refined structure of the quinoprotein methanol dehydrogenase from ˙ , Structure 3 (1995) 177 – 187. Methylobacterium extorquens at 1.94 A M.A. Kaderbhai, C.C. Ugochukwu, S.L. Kelly, D.C. Lamb, Export of cytochrome P450 105D1 to the periplasmic space of Escherichia coli, Appl. Environ. Microbiol. 67 (2001) 2136 – 2138. L. Thony-Meyer, Biogenesis of respiratory cytochromes in bacteria, Microbiol. Mol. Biol. Rev. 61 (1997) 337 – 376.
Review
The quinohaemoprotein lupanine hydroxylase from Pseudomonas putida David J. Hopper *, Mustak A. Kaderbhai Institute of Biological Sciences, University of Wales, Aberystwyth, Ceredigion SY23 3DD, UK Received 13 July 2002; received in revised form 31 October 2002; accepted 22 January 2003
Abstract Lupanine hydroxylase catalyses the first reaction in the catabolism of the alkaloid lupanine by Pseudomonas putida. It dehydrogenates the substrate, which can then be hydrated. It is a monomeric protein of Mr 72,000 and contains a covalently bound haem and a molecule of PQQ. The gene for this enzyme has been cloned and sequenced and the derived protein sequence has a 26 amino acid signal sequence at the Nterminal for translocation of the protein to the periplasm. Many of the features seen in the sequence of lupanine hydroxylase are common with other quinoproteins including the W-motifs that are characteristic of the eight-bladed propeller structure of methanol dehydrogenase. However, the unusual disulfide bridge between adjacent cysteines that is present in some PQQ-containing enzymes is absent in lupanine hydroxylase. The C-terminal domain contains characteristics of a cytochrome c and overall the sequence shows similarities with that of the quinohaemoprotein, alcohol dehydrogenase from Comamonas testosteroni. The gene coding for lupanine hydroxylase has been successfully expressed in Escherichia coli and a procedure has been developed to renature and reactivate the enzyme, which was found to be associated with the inclusion bodies. Reactivation required addition of PQQ and was dependent on calcium ions. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Quinohaemoprotein; Cytochrome; PQQ; Renaturation; Reactivation; Lupanine
1. Introduction Lupanine is one of a class of alkaloids known as the quinolizidine alkaloids and is found particularly in plants of the genus Lupinus [1,2]. These alkaloids give the plants some resistance to pathogens and also to herbivores as they are highly toxic to animals, inducing a lethal state of lupinosis. Besides this protective role, it is thought that they can also serve as nitrogen stores for developing seedlings [3]. In common with all plant products, degradation of these alkaloids by soil microorganisms is important in mobilization of the organic carbon in the carbon cycle. A number of strains of bacteria have been isolated that can grow on lupanine as their sole carbon and nitrogen source [4,5]. One of these was first isolated from soil by Moz˙ejko-Toczko in 1960 and identified as a Pseudomonas sp., while more recently, seven Gram-negative bacterial strains were isolated from soil in which lupins (Lupinus albus L.) had been grown [5]. Three of these were identified, two as Xanthomonas oryzae and one as Gluconobacter cerinus. * Corresponding author. Tel.: +44-1970-622-292; fax: +44-1970-622-307. E-mail address: [email protected] (D.J. Hopper).
Toczko et al. [6] reported that the first intermediate of lupanine breakdown by their Pseudomonas sp. is 17hydroxylupanine [6] produced by an enzyme referred to as lupanine hydroxylase.
2. Lupanine hydroxylase The enzyme, lupanine hydroxylase, was partially purified and shown to contain haem [7]. It required an electron acceptor for activity rather than a donor, as would be required for a monooxygenase type of hydroxylase. This suggested a mechanism by which the substrate is dehydrogenated and subsequently hydrated, with the oxygen of the hydroxyl group coming from water. It should be noted, however, that 17-hydroxylupanine is a carbinolamine with a pK of 10.5 and that its perchlorate and picrate salts analyse as the anhydronium form [8,9] (Fig. 1). The Pseudomonas enzyme is inducible by growth of the organism on (+) lupanine as a carbon source and is not active with the ( ) enantiomer [10]. Growth of the organism for enzymic studies depends, therefore, on a supply of lupanine, which is not commercially available. The need for the substrate to be isolated from lupin seeds
1570-9639/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1570-9639(03)00070-0
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
111
Fig. 1. The reaction catalysed by lupanine hydroxylase.
has dictated the pace of progress in the characterisation of this enzyme. Although another inducer of lupanine hydroxylase is 15-oxosparteine [11], this also is not readily available. Another related alkaloid is sparteine (Fig. 2); this is also a substrate for lupanine hydroxylase and is commercially available but, unfortunately, fails to act as an inducer of the enzyme [11]. Eventually, the enzyme was purified to homogeneity by various chromatographic techniques and some of its properties characterised [12]. The UV/VIS spectrum was that of a c-type cytochrome and a value of one haem per molecule of enzyme was calculated using the absorbance coefficient of cytochrome c. Moreover, the haem was covalently bound as expected for a c-type cytochrome. The haem will only accept one electron, and for the two-electron dehydrogenase reaction, a second electron acceptor must be present. Several pieces of evidence showed that this is PQQ; extracts from the enzyme were able to reactivate the PQQ-requiring glucose dehydrogenase apoenzyme from Acinetobacter calcoaceticus [13]; inactive apoenzyme produced by isoelectric focussing was reactivated by addition of PQQ, and the oxidised spectrum of this apoenzyme was only reduced by substrate if PQQ was also added. Titration of the apoenzyme with PQQ showed that full activity was recovered on the addition of one molecule of PQQ per molecule of enzyme [12]. Interaction of the enzyme with PQQ was also demonstrated by the effect of the cofactor on the midpoint redox potential of the haem group. This was + 193 mV for the
holoenzyme but shifted to + 98 mV when PQQ was removed. Addition of PQQ to this apoenzyme restored the original redox potential, clearly showing that binding of the cofactor affected the environment of the haem in the protein [14]. A similar shift in the midpoint redox potential has been described for another PQQ-containing quinohaemoprotein, ethanol dehydrogenase from Comamonas testosteroni, in which there was a shift from + 80 mV in the apoenzyme to + 140 mV in the holoenzyme [15]. Indeed, the lupanine hydroxylase has a number of other features in common with the ethanol dehydrogenase. It is found in the periplasmic compartment of the cell and is a monomeric protein Mr about 70,000. Gel filtration and ultracentrifuge studies gave values of 74,000 and 66,000, respectively, for the native lupanine hydroxylase while SDS-PAGE gave a value of 72,000 [12]. The assay developed for monitoring lupanine hydroxylase activity used horse heart cytochrome c as the electron acceptor and its reduction was followed spectrophotometrically by the increase in absorbance at 550 nm. Steady-state kinetics using this assay gave results suggestive of a pingpong mechanism with a Km value of 3.6 F 0.7 AM for lupanine and a kcat of 217 s 1 [12]. More recently, the same kinetic experiments have been performed with sparteine as substrate. Sparteine is another plant-produced quinolizidine alkaloid with a very similar structure to lupanine. The enzyme has a lower Km for sparteine with a value of 0.87 F 0.11 AM, but the kcat is also much lower at 11.8 s 1. Thus, the specificity constant for lupanine is higher at 60.3 AM 1 s 1 compared to 13.6 AM 1 s 1 for sparteine.
3. Cloning and sequence of the gene for lupanine hydroxylase
Fig. 2. The structure of sparteine.
The gene for lupanine hydroxylase (luh) has now been isolated using PCR with chromosomal DNA as template and
112
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
primers based on the sequence of the N- and C-terminals of the purified enzyme [16]. This was sequenced (accession number AJ318095) and found to code for a protein of Mr 72,256, which is in good agreement with the value of 72,000 found by SDS-PAGE of the purified enzyme. The derived sequence included the sequences of several internal peptides, isolated after digestion of the purified enzyme and then sequenced by Edman degradation, which suggests that the correct gene had been isolated. The ORF for the gene, however, starts at base 78 before the codon for the first amino acid of the mature protein. This was found by sequencing upstream of the coding region for the mature protein. Information on this part of the gene was obtained by amplifying a fragment of DNA encompassing the 5V end of the coding sequence for the mature protein and some of the upstream DNA by ligation-mediated PCR followed by
sequencing of the product. The protein sequence derived from this showed a 26 amino acid signal sequence, consistent with the need for translocation of the protein across the cell membrane to the periplasm. Analysis by the SignalP V2.0 computer program [17] gave a signal peptide probability of 1.000 and predicted the signal peptidase cleavage site to be between amino acids 26 and 27 with a probability of 0.986. The derived protein sequence showed similarities to those of many other PQQ-containing quinoproteins and particularly to the quinohaemoprotein, ethanol dehydrogenase from C. testosteroni, with which it had 36% identity in a BLAST search [18] (Fig. 3). The crystal structure of the ethanol dehydrogenase [19] shows the eight-bladed propeller structure that was described for the a-subunit PQQcontaining methanol dehydrogenase [20]. Alignment of the
Fig. 3. Alignment of the amino acid sequence of lupanine hydroxylase with that of the quinohaemoprotein ethanol dehydrogenase from C. testosteroni (edh). Residues that are common are shaded in black. The sequences above the asterisks are tryptophan docking motifs of the eight-bladed propeller structure found in the ethanol dehydrogenase and in methanol dehydrogenase. The haem-binding regions in the cytochrome c domains are italicised and doubly underlined. The signal sequence for each protein is shown in lower case.
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
113
sequence for lupanine hydroxylase with those for methanol dehydrogenase and ethanol dehydrogenase showed the presence of the eight W-motifs characteristic of this tertiary structure in the N-terminal domain suggesting a similar tertiary structure. The C-terminal domain incorporates the cytochrome c haem-binding consensus sequence of CXXCH as CGACH and the methionine that provides the sixth ligand to the iron in the ethanol dehydrogenase is conserved in lupanine hydroxylase (Fig. 3). Also conserved is the tryptophan residue that forms the floor of the active site chamber in both the methanol and ethanol dehydrogenases but a notable difference is the apparent absence in lupanine hydroxylase of the novel disulfide ring structure formed between adjacent cysteines that is a feature of the active site in the other two enzymes and in some other PQQ-enzymes.
4. Heterologous expression of the luh gene Conclusive evidence that the cloned luh gene is that of lupanine hydroxylase requires the heterologous expression of the gene to give active enzyme. Several attempts were made to express the gene in Escherichia coli but these were unsuccessful. Possible reasons for this are the incompatibility of the signal sequence with the secretion pathway in E. coli and the general difficulty in cytochrome c expression/maturation due to the requirement for haem insertion. To overcome these problems, we replaced the 26 amino acid signal sequence with the 21 residue alkaline phosphate
Fig. 4. Plasmid pEV-LH32 containing the gene for lupanine hydroxylase (LH) for expression in E. coli. SS is the signal sequence of alkaline phosphatase. S/D is the Shine/Delgarno sequence. Ppho is the phoA promotor. Ori is the origin of plasmid replication and Ampr codes for hlactamase.
Fig. 5. Reactivation of apoenzyme. Enzyme was incubated with 1 mM EGTA to chelate any calcium ions present in the preparation. It was then incubated at 20 jC with 0.05 mM PQQ (D), 0.05 mM PQQ and 2 mM CaCl2 (.), 2 mM CaCl2 (n), 0.05 mM PQQ and 0.1 mM lupanine (E) and 0.05 mM PQQ, 2 mM CaCl2, 0.1 mM lupanine (o). Samples were taken at timed intervals and assayed for lupanine hydroxylase activity.
signal sequence that is native to E. coli by ligation of the region of the luh gene encoding the mature protein into plasmid pLiQ [21] to give plasmid pEV-LH32 (Fig. 4). This also placed the luh gene under the control of the phoA
Fig. 6. Reactivation of apoenzyme with various cations. Enzyme was incubated at 20 jC with 1 mM EGTA to chelate any calcium ions in the preparation and then with 0.05 mM PQQ and 2 mM CaCl2 (.), MgCl2 (o), BaCl2 (E) or SrCl2 (D). Samples were taken at timed intervals and assayed for lupanine hydroxylase activity.
114
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115
Fig. 7. SDS-PAGE analysis of recombinant lupanine hydroxylase in the inclusion bodies derived from E. coli harbouring either pEV-LH32 (lane 2) or pEV-LH32 and pEC86 (lane 3). The partially pure recombinant lupanine hydroxylase (arrow) is shown in lane 1.
promoter so that the tightly regulated gene expression can be switched on by shift to growth condition of low phosphate concentration. The recombinant plasmid was introduced into the E. coli host TB1 that already harboured the plasmid pEC86. This plasmid expresses constitutively the genes for cytochrome c maturation factors (ccm) contained on the ccm operon [22]. When the organism was grown (6 h aerobically followed by 12 h with low aeration) in phosphate-limiting medium (0.1 mM), a haem-containing protein of the same Mr as lupanine hydroxylase was produced, as shown by SDS-PAGE of total cell protein. However, this haemoprotein was present in amorphous inclusion bodies that were isolatable with the cell membranes following centrifugation of the lysed cells. Perhaps it was not surprising that the protein showed no enzyme activity as no PQQ had been included in the growth medium. Nevertheless, a procedure for the solubilisation and reactivation of the enzyme was developed. This involved the solubilisation of the isolated inclusion bodies with 8 M urea followed by rapid dilution (10-fold) with a buffered solution containing EDTA (1 mM) to yield a tancoloured soluble fraction. Incubation of this ‘renatured’ protein with PQQ gave some enzyme activity, which was further enhanced significantly by addition of Ca2 +. Chelation of any adventitious Ca2 + ions by pre-incubation with 1 mM EGTA prior to PQQ addition alone yielded no enzymic activity. Thus it could be demonstrated that both PQQ and an excess of Ca2 + ions were essential for activity (Fig. 5) as has been shown for other PQQ-containing enzymes. Other divalent cations such as Sr2 + and Ba2 +
were equally competent at reactivating the enzyme in conjunction with PQQ (Fig. 6). However, Mg2 + was much less effective. The importance of co-expression of the ccm factors for the production of enzyme with the haem correctly attached was demonstrated by comparing the activities of lupanine hydroxylase recovered in the renatured and reactivated inclusion bodies derived from cells that harboured either only the luh gene (pEV-LH32) or both the luh and ccm gene clusters (pEV-LH32 and pEC86) and grown under identical inducing conditions. A protein of the correct Mr was produced in the single plasmid-containing cells (Fig. 7) but it did not contain as much haem and could not be reactivated simply by addition of haem to the isolated protein. The specific activity of the enzyme in the reactivated extract of cells containing both plasmids was 11-fold higher than that from cells lacking the plasmid coding for the ccm operon. Thus we have been able to express the luh gene heterologously to obtain active enzyme and are, therefore, confident that this is the gene for lupanine hydroxylase. Furthermore, this allows us to produce sufficient enzyme, without the constraints of the limited availability of lupanine for growth, for further structural and other studies. References [1] M. Wink, C. Meissner, L. Witte, Patterns of quinolizidine alkaloids in 56 species of the genus Lupinus, Phytochemistry 38 (1995) 139 – 153. [2] M. Muzquiz, C. Cuadrado, G. Ayet, C. de la Cuadra, C. Burbano, A. Osagie, Variation of alkaloid components of lupin seeds in 49 genotypes of Lupinus albus L. from different countries and locations, J. Agric. Food Chem. 42 (1994) 1447 – 1450. [3] M. Wink, L. Witte, Quinolizidine alkaloids as nitrogen source for lupin seedlings and cell cultures, Z. Naturforsch. 40c (1985) 767 – 775. [4] M. Moz˙ejko-Toczko, Rozklad lupaniny przez Pseudomonas lupanine, Acta Microbiol. Pol. 9 (1960) 157 – 171. [5] F.M.C. Santana, A.M. Fialho, I. SaCorreia, J.M.A. Empis, Isolation of bacterial strains capable of using lupanine, the predominant quinolizidine alkaloid in white lupin, as sole carbon and energy source, J. Ind. Microbiol. 17 (1996) 110 – 115. [6] M. Toczko, W. Brzeski, K. Kaßkolewska-Baniuk, Microbial degradation of lupanine: V. Identification of 17-hydroxylupanine, Bull. Acad. Pol. Sci., Cl. II, 11 (1963) 161 – 164. [7] J. Rogozin´ski, Molecular properties of the inducible lupanine hydroxylase from growing cultures of Pseudomonas lupanine, Acta Biochim. Pol. 32 (1975) 57 – 66. [8] O.E. Edwards, F.H. Clarke, B. Douglas, 17-Hydroxylupanine and 17-oxylupanine, Can. J. Chem. 32 (1954) 235 – 241. [9] J. Thiel, W. Boczon, W. Wysocka, A stereostructural study of 17-hydroxylupanine and its perchlorate, Monatsh. Chem. 131 (2000) 1073 – 1081. [10] A. Niedzielska, J. Rogozin´ski, Inducible lupanine hydroxylase from resting cultures of Pseudomonas lupanini, Bull. Acad. Pol. Sci., Cl. II, XXI (1973) 1 – 5. [11] J. Droese, Inducers and substrates of inducible enzymes in the Pseudomonas sp. isolated from soil and degrading lupanine, Acta Microbiol. Pol., Ser. B 2 (1970) 95 – 101. [12] D.J. Hopper, J. Rogozin´ski, M. Toczko, Lupanine hydroxylase, a quinocytochrome c from an alkaloid-degrading Pseudomonas sp., Biochem. J. 279 (1991) 105 – 109.
D.J. Hopper, M.A. Kaderbhai / Biochimica et Biophysica Acta 1647 (2003) 110–115 [13] R.A. van der Meer, B.W. Groen, M.A.G. van Kleef, J. Frank, J.A. Jongejan, J.A. Duine, Isolation, preparation, and assay of pyrroloquinoline quinone, Methods Enzymol. 188 (1990) 260 – 283. [14] D.J. Hopper, J. Rogozin´ski, Redox potential of the haem c group in the quinocytochrome, lupanine hydroxylase, an enzyme located in the periplasm of a Pseudomonas sp., Biochim. Biophys. Acta 1383 (1998) 160 – 164. [15] G.A.H. de Jong, J. Caldeira, J. Sun, J.A. Jongejan, S. de Vries, T.M. Lohr, I. Moura, J.A. Duine, Characterization of the interaction between PQQ and heme c in the quinohaemoprotein ethanol dehydrogenase from Comamonas testosteroni, Biochemistry 34 (1995) 9451 – 9458. [16] D.J. Hopper, M.A. Kaderbhai, S.A. Marriott, M. Young, J. Rogozin´ski, Cloning, sequencing and heterologous expression of the gene for lupanine hydroxylase, a quinocytochrome c from a Pseudomonas sp., Biochem. J. 369 (2002) 483 – 489. [17] H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Identification of
[18]
[19]
[20]
[21]
[22]
115
prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites, Protein Eng. 10 (1997) 1 – 6. S.F. Atschul, T.L. Madden, A.A. Schaffer, J.H. Zhang, Z. Zhang, W. Miller, D.J. Lipman, Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Nucleic Acids Res. 25 (1997) 3389 – 3402. A. Oubrie, H.J. Rozeboom, K.H. Kalk, E.G. Huizinga, B.W. Dijkstra, Crystal structure of quinohaemoprotein alcohol dehydrogenase from Comamonas testosteroni, J. Biol. Chem. 277 (2002) 3727 – 3732. M. Ghosh, C. Anthony, K. Harlos, M.G. Goodwin, C.C.F. Blake, The refined structure of the quinoprotein methanol dehydrogenase from ˙ , Structure 3 (1995) 177 – 187. Methylobacterium extorquens at 1.94 A M.A. Kaderbhai, C.C. Ugochukwu, S.L. Kelly, D.C. Lamb, Export of cytochrome P450 105D1 to the periplasmic space of Escherichia coli, Appl. Environ. Microbiol. 67 (2001) 2136 – 2138. L. Thony-Meyer, Biogenesis of respiratory cytochromes in bacteria, Microbiol. Mol. Biol. Rev. 61 (1997) 337 – 376.