Enzymes of the Phenylpropionate Catabolic Pathway of Escherichia coli
The persistence in the environment of man-made chemicals, particularly chlorinated aromatics such as DDT and polychlorinated biphenyls (PCB's), in the form of pesticides and industrial chemicals is an issue of increasing public concern. By studying bacterial pathways involved in biodegradation of aromatic compounds, we seek to understand the enzymatic processes involved in their breakdown. We hope that a better understanding of the molecular basis for aromatic biodegradation will eventually lead to the design of improved biological catalysts for bio-remediation of chemical pollutants.
We are currently studying a catabolic pathway found in Escherichia coli responsible for the degradation of phenylpropionic acid, on which E. coli can grow as a sole carbon source. Oxidative ring cleavage is catalysed by 2,3-dihydroxyphenylpropionate 1,2-dioxygenase (MhpB), to give a yellow ring fission product which is cleaved by an unusual C-C hydrolase enzyme MhpC to give succinic acid and 2-hydroxypentadienoic acid. The latter unstable dienol is hydrated by a Mn2+-dependent hydratase enzyme MhpD to give 4-hydroxy-2-keto-pentanoic acid, which is the substrate for retro-aldol cleavage by aldolase MhpE.
Most of the research in the group has focused on extradiol dioxygenase MhpB and C-C hydrolase MhpC, however we have also examined the final two enzymes on the pathway, hydratase MhpD and alsolase MhpE.
1. 2,3-Dihydroxyphenylpropionate 1,2-dioxygenase (MhpB)
2,3-Dihydroxyphenylpropionate 1,2-dioxygenase (MhpB) is a non-heme iron (II) dependent extradiol dioxygenase, which has been overexpressed, purified and characterised . Amino acid sequence alignments have revealed that this enzyme is a member of a distinct class of extradiol dioxygenases, containing a catechol 2,3-dioxygenase from Alcaligenes eutrophus, which we have also purified . We have investigated the catalytic mechanism of MhpB using 18O2labelling experiments and synthesis of reaction intermediates. From these experiments we have obtained evidence in favour of a lactone intermediate in the mechanism, formed by a type of Criegee rearrangement .
We have also synthesised cyclopropyl-containing substrate analogues to act as radical traps for this enzyme, and we have observed a novel cis-trans isomerisation of the cyclopropyl ring during enzymatic processing, due to a reversible opening of the cyclopropyl ring. This provides evidence for a semiquinone radical intermediate in the enzyme mechanism .
We have synthesised carba- analogues of the two possible hydroperoxide intermediates, in order to identify which is the true intermediate in the MhpB reaction mechanism. The C-1 carba analogue (R=Me) showed no inhibition of MhpB, nor did the R=Me C-2 analogue. However, the R=tBu C-2 analogue, in which the hydroxymethyl substituent is held in an axial orientation, acts as a reversible competitive inhibitor for MhpB, consistent with the intermediacy of the C-2 hydroperoxide intermediate in the catalytic mechanism .
We have discovered a model reaction for extradiol catechol cleavage involving 1,4,7-triazacyclononane (TACN), FeCl2or FeCl3, pyridine (3 equiv) in methanol, which gives a good yield of muconic semialdehyde methyl ester. The reaction is highly selective for the macrocyclic ligand, which mimics the tridentate co-ordination chemistry observed in the extradiol dioxygenase active site. Interestingly, the FeCl2reaction shows higher extradiol/intradiol selectivity (7:1) than the FeCl3reaction (2:1), indicating that Fe(II) has an inherent chemical preference for extradiol cleavage, as found in the enzyme active site .
We have explored the mechanism of this reaction, which is the only iron (II) based model reaction to give a derivative of the true extradiol cleavage product. The reaction proceeds using the mono-sodium salt of catechol in the absence of pyridine, but not using the di-sodium salt, implying that catechol binds as its mono-anion. However, in this case the product distribution is very different, favouring intradiol cleavage by 15:1, implying that pyridine has an additional role, perhaps as a proton donor. Repeating the reaction with catechol mono-sodium salt in the presence of 1 equivalent of pyridine hydrochloride results in a switch back to extradiol cleavage, in a 4:1 ratio. These results suggest that the role of pyridine is two-fold: one equivalent is required as a base, but subsequently the protonated pyridinium salt acts as a proton donor, probably to catalyse the Creigee rearrangement step. These results provide some insight into the acid/base catalysis of the extradiol dioxygenases, and suggests roles for acid/base groups in the enzyme active site .
Two active site histidine residues, His-115 and His-179, were identified in MhpB that might act as acid-base catalytic residues. Replacement of either His residue by Gln by site-directed mutagenesis gave inactive mutant enzymes. Replacement of adjacent Asp-114 by Ala also gave an inactive mutant, however the D114N mutant retained a low level of activity. Analysis of the pH/rate profile of this latter mutant showed a loss of activity at high pH, indicating an effect on acid catalysis. Further studies indicate that His-115 is the catalytic base, while His-179 is an acidic group required for the final lactone hydrolysis step .
Recent studies have focussed on how the extradiol and intradiol catechol dioxygenase active sites catalyse two different Criegee rearrangements on a common hydroperoxide intermediates, as first discussed in 2001. Using error-prone PCR, we have selected 6 MhpB mutants of modified substrate specificity: two of these mutants were found to catalyse the formation of 5-15% intradiol product, consistent with the partitioning of a common intermediate . We are currently using mechanistic probes to study the alkenyl and acyl migration mechanisms for Criegee rearrangement in the extradiol and intradiol catechol dioxygenases.
2. 2-Hydroxy-6-Keto-Nona-2,4-Diene 1,9-Dioic Acid 5,6-Hydrolase (MhpC)
The dienol ring fission product has been characterised and used to purify the hydrolase enzyme MhpC to near homogeneity. The identity of the MhpC products has been established, and by performing the enzymatic reaction in 2H2O the stereochemistry of the reaction has been established as proceeding with overall retention of regiochemistry . Evidence for an MhpC-catalysed ketonisation of the dienol substrate has been obtained by 2H exchange experiments .
Pre-steady state kinetic analysis of the MhpC reaction has confirmed the existence of a discrete intermediate, and has revealed that approximately 50% of the keto-intermediate is released from the active site per catalytic cycle. A high Kdfor the substrate was measured (30 µM), although Km= 7 µM, suggesting that the substrate is bound in astrained non-planar conformation, thus promoting ketonisation .
Amino acid sequence alignments reveal that this family of C-C hydrolase enzymes is in the alpha/beta-hydrolase superfamily, containing a serine-histidine-aspartate catalytic triad, as found in the serine proteases. However, rapid quench studies using a synthetic radiolabelled substrate have yielded a very low yield of labelled protein, suggesting that there is no covalent acyl enzyme intermediate. Evidence in favour of a general base mechanism (involving base-catalysed attack of water on the ketonised substrate) has been obtained by incorporation of a small amount of two atoms of 18O from H218O. A non-hydrolysable substrate analogue, 4-keto-nona-1,9-dioic acid (KNDA) was found to act as a time-dependent inhibitor of MhpC, and was found to undergo
time-dependent incorporation of 18O from H218O into the ketone carbonyl, consistent with the formation of a gem-diol intermediate .
The crystal structure of E. coli MhpC has been determined, in collaboration with Prof. S. Wood (Dept. of Biochemistry, University of Southampton), in the presence of a non-cleavable analogue . The catalytic Ser-110 is positioned very close to the carbonyl group of the substrate analogue, however the Ser-110/His-263 pair straddle the substrate in an arrangement different to that found in the serine proteases. Pre-steady state kinetic analysis of S110A and H263A mutants have shown that Ser-110 is not required for the initial keto-enol tautomerisation step, but is required for C-C cleavage; while His-263 is required for both tautomerisation and C-C cleavage . Arg-188, positioned near the C-1 carboxylate group, is responsible for substrate binding, but also has a catalytic role in keto-enol tautomerisation, probably in substrate destabilisation .
We have developed a general synthetic route to the meta-ring cleavage products, using a palladium-mediated Heck coupling to synthesise the central diene. This route has been used to synthesise a series of substituted aryl-containing substrates for C-C hydrolase BphD, found on the biphenyl degradation pathway of Pseudomonas sp. LB400. A Hammett plot of log(kcat) versus sigma gave a reaction coefficient of –0.71 for C-C cleavage, consistent with a departing carbanion in the transition state of the enzyme catalysed reaction. Furthermore, a reduced substrate analogue containing a C-6 hydroxyl group was found to be a substrate for BphD-catalysed aldol-type cleavage, consistent with a general base mechanism . Using this synthetic route, we have synthesised substrate containing a 13C label at C-6. Treatment of BphD with this labelled substrate gives a new NMR signal at 128 ppm, consistent with a gem-diol intermediate, with no appearance of an acyl enzyme intermediate at 170 ppm .
MhpC shares sequence similarity with a/b-hydrolases of diverse catalytic function: esterases, haloperoxidases, even cofactor-independent dioxygenases! In order to rationalise this diverse range of catalytic activities, we have suggested that these enzymes are able to activate a range of nucleophiles and diatomic substrates, including water, hydrogen peroxide, and dioxygen . We have shown that MhpC and BphD both possess esterase activity, which has allowed a physical organic chemical study of esterase activity, by means of a Hammett plot . We have recently shown that MhpC is also able to activate hydroxylamine as a nucleophile, and is able to catalyse C-C bond formation reactions in organic solvents .
3. 2-Hydroxypentadienoic Acid Hydratase (MhpD)
The ketonisation of 2-hydroxypentadienoic acid in aqueous solution has been studied and compared with its homologue enolpyruvic acid . Hydratase MhpD is dependent on a Mn2+cofactor, which appears to be involved in substrate activation. The enzyme has been purified to homogeneity, and mechanistic studies undertaken . The mechanism is thought to involve ketonisation of the substrate, followed by attack of a Mn2+-activated water.
4. 4-Hydroxy-2-Keto-Pentanoic Acid Aldolase (MhpE)
Aldolase MhpE catalyses the aldol condensation of acetaldehyde and pyruvic acid to give 4-hydroxy-2-keto-pentanoic acid. We have purified the enzyme 30-fold from Escherichia coli, and we have determined that the enzyme is a class 1 aldolase utilising an imine linkage with an active site lysine residue to activate the ketone of pyruvic acid for aldol reaction. The enzyme has been shown to catalyse the reverse reaction (i.e. C-C bond formation), but the equilibrium constant lies heavily in favour of C-C cleavage. The enzyme is selective for acetaldehyde as carbonyl acceptor, but is able to utilise a range of -keto acids as carbonyl donor . This enzyme could therefore be amenable for use in selective C-C bond forming biotransformations.
 T.D.H. Bugg, Biochem. Biophys. Acta 1202, 258-264 (1993)
 E.L. Spence, M. Kawamukai, J. Sanvoisin, H. Braven and T.D.H. Bugg, J. Bacteriol., 178, 5249-5256 (1996)
 J. Sanvoisin, G.J. Langley & T.D.H. Bugg, J. Am. Chem. Soc. 117, 7836-7837 (1995)
 E.L. Spence, G.J. Langley, and T.D.H. Bugg, J. Am. Chem. Soc., 118, 8336-8343 (1996)
 C.J. Winfield, Z. Al-Mahrizy, M. Gravestock and T.D.H. Bugg, J. Chem. Soc. Perkin Trans. 1, 3277-3289 (2000)
 G. Lin, G. Reid, and T.D.H. Bugg, J. Chem. Soc. Chem. Commun., 1119-1120 (2000)
 G. Lin, G. Reid, and T.D.H. Bugg, J. Am. Chem. Soc., 123, 5030-5039 (2001)
 S. Mendel, A. Arndt, and T.D.H. Bugg, Biochemistry, 43, 13390-13396 (2004)
 T.D.H. Bugg and G. Lin, J. Chem. Soc. Chem. Commun., 941-952 (2001)
 W.W.Y. Lam & T.D.H. Bugg, Chem. Commun., 1163-4 (1994)
 W.W.Y. Lam & T.D.H. Bugg, Biochemistry, 36, 12242-12251 (1997)
 I.M.J. Henderson & T.D.H. Bugg, Biochemistry, 36, 12252-12258 (1997)
 S.M. Fleming, T.A. Robertson, G.J. Langley, and T.D.H. Bugg, Biochemistry, 39, 1522-1531 (2000)
 G. Dunn, M.G. Montgomery, F. Mohammed, A.Coker, J.B.Cooper, T. Robertson, J-L. Garcia, T.D.H. Bugg & S.P. Wood, J. Mol. Biol., 346, 253-265 (2005)
 C. Li, M.G. Montgomery, F. Mohammed, J-J. Li, S.P. Wood, and T.D.H. Bugg, J. Mol. Biol.,346, 241-251 (2005)
 C. Li, J-J.Li, M.G. Montgomery, S.P. Wood, and T.D.H. Bugg, Biochemistry, 45, 12470-12479 (2006)
 D.M. Speare, S.M. Fleming, M.N. Beckett, J-J. Li and T.D.H. Bugg, Org. Biomol. Chem., 2, 2942-2950 (2004)
 J-J. Li, C. Li, C.A. Blindauer, and T.D.H. Bugg,Biochemistry, 45, 12461-12469 (2006)
 T.D.H. Bugg, Bio-Organic Chemistry, 32, 367-375 (2004)
 J-J. Li and T.D.H. Bugg, Org. Biomol. Chem., 5, 507-513 (2007)
 C. Li, M. Hassler, and T.D.H. Bugg,ChemBioChem 9, 71-76 (2008)
 J.R. Pollard, I.M.J. Henderson, and T.D.H. Bugg, J. Chem. Soc. Chem. Commun., 1885-1886 (1997)
 J.R. Pollard & T.D.H. Bugg, Eur. J. Biochem., 251, 98-106 (1998)
 J.R. Pollard, D. Rialland, and T.D.H. Bugg, Appl. Environ. Microbiol.., 64, 4093-4094 (1998)