Arch Microbiol (2004) 181 : 155–162 DOI 10.1007/s00203-003-0642-4
O R I G I N A L PA P E R
Knut Verfürth · Antonio J. Pierik · Christina Leutwein · Susanne Zorn · Johann Heider
Substrate specificities and electron paramagnetic resonance properties of benzylsuccinate synthases in anaerobic toluene and m -xylene metabolism Received: 5 September 2003 / Revised: 27 November 2003 / Accepted: 28 November 2003 / Published online: 20 December 2003 © Springer-Verlag 2003
Abstract The anaerobic degradation pathways of toluene and m-xylene are initiated by addition of a fumarate cosubstrate to the methyl group of the hydrocarbon, yielding (R)-benzylsuccinate and (3-methylbenzyl)succinate, respectively, as first intermediates. These reactions are catalyzed by a novel glycyl-radical enzyme, (R)-benzylsuccinate synthase. Substrate specificities of benzylsuccinate synthases were analyzed in Azoarcus sp. strain T and Thauera aromatica strain K172. The enzyme of Azoarcus sp. strain T converts toluene, but also all xylene and cresol isomers, to the corresponding succinate adducts, whereas the enzyme of T. aromatica is active with toluene and all cresols, but not with any xylene isomer. This corresponds to the capabilities of Azoarcus sp. strain T to grow on either toluene or m-xylene, and of T. aromatica to grow on toluene as sole hydrocarbon substrate. Thus, differences in the substrate spectra of the respective benzylsuccinate synthases of the two strains contribute to utilization of different aromatic hydrocarbons, although growth on different substrates also depends on additional determinants. We also provide direct evidence by electron paramagnetic resonance (EPR) spectroscopy that glycyl radical enzymes corresponding to substrate-induced benzylsuccinate synthases are specifically detectable in anoxically prepared extracts of toluene- or m-xylene-grown cells. The presence of the EPR signals and the determined amount of the radical are consistent with the respective benzylsuccinate synthase activities. The properties of the EPR signals are highly similar to those of the prototype glycyl radical enzyme pyruvate formate lyase, but differ slightly from pre-
K. Verfürth · C. Leutwein · S. Zorn · J. Heider (✉) Mikrobiologie, Albert-Ludwigs-Universität, Schänzlestrasse 1, Institut für Biologie II, 79104 Freiburg, Germany Tel.: +49-761-2032774, Fax: +49-761-2032626, e-mail:
[email protected] A. J. Pierik Laboratorium für Mikrobiologie, Phillips-Universität, Fachbereich Biologie, 35032 Marburg, Germany
viously reported parameters for partially purified benzylsuccinate synthase. Keywords Anaerobic metabolism · Benzylsuccinate synthase · Glycyl radical enzyme · Electron paramagnetic spectroscopy · Substrate specificity · Toluene · Xylenes · Cresols · Thauera · Azoarcus
Introduction Various bacterial species utilizing aromatic hydrocarbons as sole carbon and energy sources under anoxic conditions have been isolated in the last decade. The pathways of hydrocarbon metabolism in these bacteria are fundamentally different from all known hydrocarbon-degradation pathways of aerobic organisms, which employ at least one oxygen-dependent reaction. A common initial pathway was recently detected in the anaerobic catabolism of toluene, m-xylene, 2-methylnaphthalene, and m- and p-cresol (Biegert et al. 1996; Beller and Spormann 1997a, b; Krieger et al. 1999; Annweiler et al. 2000; Müller et al 1999, 2001). A methyl group of all these compounds is anaerobically oxidized to the level of a carboxylic group, yielding, e.g., benzoyl-CoA from toluene or 3-methylbenzoyl-CoA from m-xylene (Biegert et al. 1996; Krieger et al. 1999). Remarkably, this oxidative pathway is initiated by a formal non-redox reaction, namely the addition of a fumarate cosubstrate to (one of) the methyl substituent(s), yielding (R)-benzylsuccinate from toluene (Biegert et al. 1996; Beller and Spormann 1997a, b, 1998; Leuthner et al. 1998; Leutwein and Heider 1999) (Fig. 1), and ringsubstituted benzylsuccinates from other compounds (Krieger et al. 1999; Müller et al. 1999; 2001; Annweiler et al. 2000) (Fig. 1). Similar initial reactions were even detected in the anaerobic metabolism of ethylbenzene and n-hexane, where a fumarate is added to the subterminal methylene group of the hydrocarbon (Rabus et al. 2001; Kniemeyer et al. 2003). The key enzyme of anaerobic toluene catabolism, (R)-benzylsuccinate synthase, has been detected in several
156
Fig. 1 Proposed pathways of anaerobic toluene and m-xylene catabolism. Enzymes: BSS (R)-benzylsuccinate synthase, BS-CT succinyl-CoA: (R)-benzylsuccinate CoA-transferase, MBS-CT succinyl-CoA: (3-methylbenzyl)succinate CoA-transferase. The dot at BSS represents the activated, glycyl-radical carrying form of the enzyme
denitrifying, iron(III)- and sulfate-reducing bacteria, in a phototrophic bacterium, and in toluene-degrading methanogenic consortia (Biegert et al. 1996; Beller and Spormann 1997a, b; Rabus and Heider 1998; Zengler et al. 1999; Meckenstock 1999; Beller and Edwards 2000; Kane et al. 2002). The enzyme has been purified from Thauera aromatica and Azoarcus sp. strain T (Leuthner et al. 1998; Beller and Spormann 1999), and toluene-induced operons containing the corresponding genes have been cloned from two T. aromatica strains (Leuthner et al. 1998; Coschigano et al. 1998), Azoarcus sp. strains T (Achong et al. 2002) and EbN1 (Kube et al. 2004), and Geobacter metallireducens (Kane et al. 2002). (R)-Benzylsuccinate synthase is an α2β2γ2 heterohexamer of three subunits of 98, 9, and 7 kDa. The derived amino acid sequence of the large subunit is similar to that of the other known glycylradical enzymes, which include pyruvate formate lyases and anaerobic ribonucleotide reductases (reviewed in Heider et al. 1999), as well as a recently discovered 4-hydroxyphenylacetate decarboxylase (Selmer and Andrei
2001) and a novel type of glycerol dehydratase (Raynaud et al. 2003). Glycyl radical enzymes are irreversibly inactivated by molecular oxygen, which causes oxygenolytic cleavage of the polypeptide chain at the radical site (Knappe and Sawers 1990; Wagner et al. 1992). Remarkably, only half of the α-subunits are cleaved in any characterized glycyl radical enzyme, which has been explained by assuming activation of only one active site-glycine per dimeric enzyme to a radical (Knappe and Sawers 1990). The occurrence of the expected oxygenolytic degradation product of the large (α) subunit of (R)-benzylsuccinate synthase, whose processing site coincides exactly with the predicted glycyl radical, has been shown (Leuthner et al. 1998). Furthermore, an electron paramagnetic resonance (EPR) spectrum of partially purified benzylsuccinate synthase from toluene-grown Azoarcus sp. strain T has been reported, which is similar to those of other glycyl radical enzymes (Krieger et al. 2001), and the very-high-field EPR spectroscopic properties of benzylsuccinate synthase have been compared to those of pyruvate formate lyase and anaerobic ribonucleotide reductase (Duboc-Toia et al. 2003). Further degradation of the first intermediate (R)-benzylsuccinate to benzoyl-CoA occurs via β-oxidation (Fig. 1). The enzymes involved in this part of the pathway are substrate-specific, and their synthesis is induced by toluene in T. aromatica (Leutwein and Heider 1999; Leuthner and Heider 2000). Analogous further metabolic pathways are assumed for other substrates activated by fumarate addition. In this work, we show spectroscopic evidence for the involvement of glycyl-radical enzymes in anaerobic toluene and m-xylene metabolism directly in cell extracts of denitrifying Azoarcus sp. strain T and T. aromatica. The EPR signals are highly characteristic for glycyl radicals and exclusively present in cells grown anaerobically on the respective hydrocarbons, which allows relatively facile screening for the presence of such enzymes. Moreover, we show that the benzylsuccinate synthases of Azoarcus sp. strain T and T. aromatica vary in their substrate specificities, which helps to explain the different capabilities of the strains in toluene and m-xylene degradation.
Materials and methods Growth of bacteria and preparation of cell extract T. aromatica strain K172 (DSMZ 6948; Anders et al. 1995) and Azoarcus sp. strain T (DSMZ 9506; Dolfing et al. 1990; Seyfried et al. 1994) were grown at 30 °C under denitrifying conditions in mineral salt medium. Cultures growing on aromatic hydrocarbons were additionally supplied with a paraffin carrier phase as described previously (Tschech and Fuchs 1987; Biegert et al. 1996). Both strains were grown on toluene or benzoate as sole substrates; Azoarcus sp. strain T was also grown on m-xylene or 3-methylbenzoate. Growth was monitored by measuring optical density at 578 nm. Cultures were grown in stoppered 1- or 2 l-bottles with discontinuous feeding of organic substrate and nitrate. Typical doubling times of T. aromatica on toluene were at 10 h, those of Azoarcus sp. strain T were 15 h for growth on toluene, and 35 h for growth on m-xylene. All following steps (except for centrifuga-
157 tions in air-tight beakers) were carried out at 25 °C under strictly anoxic conditions in a glove box with a N2/H2 (95:5; by volume) gas phase. Exponentially growing cells (at optical densities of 3.5–4.5) were harvested by centrifugation at 4 °C for 20 min (10,000×g), washed with anoxic growth medium without organic substrate, suspended in 1 volume (w/v) of 10 mM triethanolamine buffer (pH 7.5), and transferred to a French pressure cell (Aminco). Cells were lysed by one passage at 137 MPa and directly transferred into a degassed Hungate-tube, which had been connected to the French pressure cell in the glove box. The cell lysate was transferred into polycarbonate tubes and centrifuged for 60 min at 100,000×g. Protein concentrations of the supernatants (cell extracts) were determined according to the method of Bradford (1976) using bovine serum albumin as standard. Enzyme assays Unlabeled toluene or m-xylene (1 mM final concentration) was incubated with 1 mM of [2,3-14C]fumarate (9.5 MBq mmol–1) and cell extracts (final concentration 3–10 mg protein ml–1 ) in triethanolamine-HCl buffer (10 mM; pH 7.5) in an anaerobic glove box. Samples were incubated at 25 °C, and the reaction was stopped after different times (2–30 min) by adding H2SO4 to 0.5% (w/v) final concentration. Precipitated protein was removed by centrifugation, and the supernatants were analyzed by HPLC. Samples of 70 µl were applied on a LiChrospher 100 RPC-18 (5-µm) column (12.5×0.4 cm; Merck) and eluted at 1 ml min–1 with 20% (v/v) acetonitrile in 40 mM formic acid (pH 3) with simultaneous detection of UV absorbance (210 nm) and radioactivity (Ramona detector, Raytest). (R,S)-Benzylsuccinate eluted at 11.0 min retention time, as checked by cochromatography of authentic standards, and was quantitated by integration of the radiodetector peaks. Detector yields were normalized by integrating the signals of known amounts of [14C]benzoate, and concentrations were calculated from the known specific radioactivity of [14C]fumarate. Using cresol isomers as substrates, more polar labeled products (elution times 4.3, 3.5, and 2.7 min with o-, m- and p-cresol, respectively), and using xylene isomers, more hydrophobic products (elution times 20.7, 24.9, and 26.4 min with o-, m- and p-xylene, respectively) were obtained. These products were apparently correlated to the non-commercially available addition products of the cresol and xylene isomers and fumarate, based on their time-dependent accumulation, UV-Vis spectra, and migration positions on the reversedphase column. Mass spectrometric analysis Conversion assays with unlabeled substrates were adjusted to pH 2, the precipitated proteins were removed by centrifugation, and metabolites contained in the supernatants were separated by isocratic HPLC on a RP-C18 column (LiChroSpher 100, Merck) with an aqueous mobile phase containing 20% acetonitrile and 32 mM formic acid. HPLC fractions containing the putative benzylsuccinate derivatives were collected, lyophilized and dissolved in methanol. These samples were analyzed in a Thermoelectron LCQ-Advantage mass-spectrometer with X-Calibur software, using negative/positive ESI-MS and CI(NH3)-MS modes. Electron paramagnetic resonance (EPR) spectroscopy Sample preparation All samples for EPR analysis were prepared under strictly anoxic conditions inside a glove box. Freshly prepared extracts of T. aromatica or Azoarcus sp. strain T cells grown under different conditions were used. Extracts were either used directly or after solvent exchange by passage over a gel-filtration column inside the glove box. These experiments were carried out with PD10 gel-filtration columns (1.5×5 cm; Amersham-Pharmacia Biotech) that were first made anoxic by washing with 5 mM dithionite in water and then
equilibrated with anoxic water or D2O. Per run, 0.5 ml of cell extract were applied on the column. After elution with anoxic water or D2O, approximately the same volume of protein-containing eluate was obtained. The preparations were transferred to EPR tubes in 300-µl portions and immediately frozen and stored in liquid nitrogen. Recording and evaluation of EPR spectra EPR spectra were recorded with an EMX-6/1 X-band spectrometer (Bruker, Karlsruhe, Germany) with a standard TE102 rectangular cavity and an ESR-900 helium flow cryostat with variable temperature (Oxford instruments, Oxford, UK) or a liquid-nitrogen finger dewar. EPR spectra were recorded under non-saturating conditions, and averages of 5–20 scans were used; further conditions are given in the figure legends. Version 2.3.1 of the WINEPR program (Bruker) was used for data acquisition; version 2.11 for data manipulation (g-values and baseline determination, subtraction and integration). The standard used for spin integration was a solution of 10 mM CuSO4 in 2 M NaClO4 and 10 mM HCl. To calculate accurate g-values, the applied microwave frequency (ν) was measured by a built-in frequency counter (EMX 040–1161, Bruker), and the magnetic field (B) was measured by a Teslameter (E035, Bruker). Furthermore, g-values were calibrated by a commercial external standard (“strong pitch sample” Bruker), which was measured under identical conditions. Using the known g-value of the external standard (g=2.0028), a small, frequently occurring magnetic field deviation between sample and teslameter (0.26 mT) was corrected. The finally obtained g-value can be given with an error of ± 0.0002 at the resonating field corresponding to g=2. The halfsaturating microwave power (P1/2 value) was determined using established methods (Brudvig 1995). Materials Chemicals and biochemicals were obtained from Aldrich-Chemie, Gerbu, Fluka, Merck, Sigma or Roth. Gases were from Sauerstoffwerke Friedrichshafen. [2,3-14C]Fumarate (specific radioactivity 2.0 GBq mmol–1) was from American Radiolabeled Compounds/ Biotrend and was diluted to 9.5 MBq mmol–1 with unlabeled fumarate.
Results Benzylsuccinate synthase activities and substrate specificities Specific activities of benzylsuccinate synthase were determined in extracts of cells of T. aromatica strain K172 and Azoarcus sp. strain T grown on different substrates. The specific activity of benzylsuccinate synthase in extracts of toluene-grown Thauera aromatica cells was 23±5 nmol min–1 (mg protein)–1 with the enzyme test described here (Table 1), which is about 100-fold higher than observed previously in T. aromatica (Biegert et al. 1996). Specific activities of benzylsuccinate synthase in toluene- and m-xylene-grown cells of Azoarcus sp. strain T were 10.4±2 nmol min–1 (mg protein)–1 (Table 1), which is in the range of previously reported activities for this strain (Beller et al. 1998, 1999; Krieger et al. 1999). The specific activities of benzylsuccinate synthase, which were reproduced with several different cell batches, are sufficiently high to explain the observed doubling times of the cultures. By measuring addition of [2,3-14C]fumarate to unlabeled aromatic compounds, quantitative assessment of the sub-
158 Table 1 Benzylsuccinate synthase activities of extracts from Thauera aromatica and Azoarcus sp. stain T with different substrates. Specific activities are given in nmol min–1(mg protein)–1 Substrate
T. aromatica
Azoarcus sp. stain T
Toluene o-Xylene m-Xylene p-Xylene o-Cresol m-Cresol p-Cresol
23.0
