Characterization of an extradiol dioxygenase involved in the catabolism of lignin-derived biphenyl

Abstract

In the catabolism of lignin-derived biphenyl by Sphingobium sp. SYK-6, LigZ catalyzes the cleavage of 2,2′,3-trihydroxy-3′-methoxy-5,5′-dicarboxybiphenyl (OH-DDVA) to a meta-cleavage product (MCP) identified here as 4,11-dicarboxy-8-hydroxy-9-methoxy-2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoate (DCHM-HOPDA). DCHM-HOPDA is transformed nonenzymatically, likely to a lactone (k = 0.13 ± 0.01 min⁻¹, pH 7.5). This is hydrolyzed to the dienolate at alkaline pH (apparent pK_a ~ 11.3). Only the dienolate is a substrate for LigY, the putative MCP hydrolase. LigZ has higher specificity for OH-DDVA (k_cat/K_m = 2.20 ± 0.02 × 10⁷ s⁻¹·m⁻¹) than for protocatechuate (PCA; 6 ± 1 × 10² s⁻¹·m⁻¹). PCA also inactivates LigZ (partition ratio of 50), but at rates too low to be physiologically relevant. This study provides insight into the bacterial catabolism of lignin and facilitates the study of downstream catabolic enzymes.

Abbreviations

DCHM-HOPDA, 4,11-dicarboxy-8-hydroxy-9-methoxy-2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoate

IMAC, immobilized metal ion affinity chromatography

MCP, meta-cleavage product

OH-DDVA, 2,2′,3-trihydroxy-3′-methoxy-5,5′-dicarboxybiphenyl

PCA, protocatechuate

PCD, protocatechuate 4,5-dioxygenase

Lignin is a heterogeneous aromatic polymer that contributes to the recalcitrance of plant cell walls. Deconstructing lignin is of great interest due to efforts to transform the various components of lignocellulose to biofuels, commodity chemicals, and biomaterials [1]. Recent studies indicating that bacteria are able to at least partially degrade lignin have accelerated the study of enzymes and pathways involved in lignin depolymerization and the catabolism of the depolymerization products [2]. Sphingobium sp. SYK-6 (SYK-6 hereafter) has emerged as one of the best characterized lignin-transforming bacterial strains [3], with pathways having been identified for the catabolism of vanillate [4], syringate [4], lignin-derived biaryl ethers [5], phenylcoumaran [6], and 2,2′-dihydroxy-3,3′-dimethoxy-5,5′-dicarboxybiphenyl (DDVA) [4].

The catabolism of DDVA by SYK-6 has been genetically elucidated and several of the enzymes have been characterized (Fig. 1). Briefly, the demethylation of DDVA by LigX yields 2,2′,3-trihydroxy-3′-methoxy-5,5′-dicarboxybiphenyl (OH-DDVA) [7]. This catechol is subject to meta-ring cleavage catalyzed by LigZ, an extradiol dioxygenase [8]. The resulting meta-cleavage product (MCP) is then transformed by LigY, a proposed MCP hydrolase [9]. Several aspects of this metabolism remain unclear, including the identity of the MCP, due to its instability, and its transformation products [8], and the nature of LigY, which does not appear to have the conserved Ser-His-Asp catalytic triad of MCP hydrolases [9].

Figure 1

Catabolic pathway of 2,2′-dihydroxy-3,3′-dimethoxy-5,5′-dicarboxybiphenyl (DDVA) in Sphingobium sp. SYK-6. DDVA is demethylated by a ligX-encoded oxygenase to generate OH-DDVA. LigZ catalyzes the extradiol fission of OH-DDVA to yield DCHM-HOPDA, a meta-cleavage product (MCP), which is hydrolyzed by LigY to produce 5CVA and CHPD. MCPs form stable pyridine derivatives in the presence of ammonia. Metabolites in gray have not been experimentally validated.

SYK-6 possesses at least four extradiol dioxygenases involved in catabolizing lignin depolymerization products: DesB, a gallate dioxygenase [10, 11]; DesZ, a 3-methylgallate dioxygenase [12]; LigAB, a protocatechuate dioxygenase [13]; and LigZ, OH-DDVA dioxygenase [8]. Extradiol dioxygenases utilize Fe²⁺ or sometimes Mn²⁺ coordinated by a 2-His-1-carboxylate facial triad to catalyze the oxygenolytic cleavage of catechols meta to the hydroxyl groups [14]. Three families of extradiol dioxygenases have been identified based on their structural folds, of which the best characterized are the Type I enzymes. These are exemplified by BphC of Burkholderia xenovorans LB400 [15, 16] and homoprotocatechuate 2,3-dioxygenase of Brevibacterium fuscum [17], which belong to the vicinal oxygen-chelate superfamily. Despite the different structural folds, all extradiol dioxygenases utilize similar catalytic machinery and mechanisms [18, 19]. The SYK-6 enzymes are Fe²⁺-dependent, Type II extradiol dioxygenases, which belong to the LigB-like superfamily, but nevertheless show interesting structural diversity and features. Thus, LigAB is a canonical Type II enzyme in which LigB, the larger subunit, harbors the catalytic machinery and LigA forms a cap above the active site. DesB, DesZ, and LigZ are all homologous to LigB. However, DesB also possesses a cap domain that is homologous to LigA. In contrast, DesZ and LigZ are ‘cap-less’ enzymes. Studies of DesB have further revealed that gallate-binding induces displacement of the active site metal ion [11]. Finally, vanillate has been reported to activate LigAB in a rare example of feed-forward activation [20].

Herein, we report the characterization of LigZ. The enzyme's reaction product and its nonenzymatic transformation were characterized using mass spectrometry and NMR. An oxygraph assay was used to evaluate LigZ's steady-state kinetic parameters as well as the enzyme's susceptibility to inactivation by various substrates. The results are discussed with respect to extradiol dioxygenases and bacterial lignin catabolism.

Materials and methods

Chemicals and reagents

All reagents were of at least analytical grade unless otherwise noted. OH-DDVA was synthesized as described previously [8]. Restriction enzymes and the Phusion PCR system were purchased from New England Biolabs. Water for buffers was purified using a Barnstead Nanopure Diamond™ system to a resistance of at least 18 megaohms. Protocatechuate-4,5-dioxygenase (PCD) was a kind gift from Prof. John Lipscomb. LigY was produced as described elsewhere (E. Kuatsjah, M. J. Kobylarz, A. C. K. Chan, H.-M. Chen, M. E. P. Murphy & L. D. Eltis, in preparation).

Cloning of ligZ

DNA was purified, manipulated, and propagated using standard procedures [21]. The ligZ gene was amplified from genomic DNA prepared from SYK-6 (NBRC 103272) using the following primers (restriction sites underlined): 5′-CCTCATATGGCCGAAATCGTGCTG-3′ and 5′-GGTAAGCTTGCCCTGAAAATACAGGTTTTCTTGCCAGGCGACGAAGC-3′. The resulting amplicon was cloned into pET41b (Novagen, Madison, WI, USA) and its nucleotide sequence was confirmed to match the locus tag SLG_07720. The resulting construct, pET41LigZ_His, produces LigZ with a C-terminal His-tag that can be removed with the tobacco etch virus (TEV) protease.

Protein production and purification

LigZ was heterologously produced using Escherichia coli BL21 containing pET41LigZ_His. Freshly transformed cells were used to inoculate 4 L LB supplemented with 30 μg·mL⁻¹ kanamycin at 37 °C and grown to an optical density of ~ 0.7. Gene expression was induced with 1 mm isopropyl β-d-thiogalactopyranoside, and the cells were further incubated overnight at 20 °C. Cells were harvested by centrifugation and stored at −80 °C until further use. Cells were suspended in 40 mL ice-cold 20 mm HEPES, pH 7.5 and lysed at 4 °C using an EmulsiFlex-C5 homogenizer (Avestin, Ottawa, ON, Canada). Cellular debris was removed by centrifugation. LigZ was purified using Ni Sepharose 6 Fast Flow resin (GE Healthcare, Mississauga, ON, Canada) according to the manufacturer's instructions. The His-tag was removed by digestion with TEV protease at room temperature in 20 mm HEPES, pH 7.5 supplemented with 1 mm DTT and 10 mm EDTA. Uncleaved LigZ and TEV were removed by passing the dialyzed digestion mix through the Ni Sepharose 6 Fast Flow resin. LigZ was further purified using a MonoQ 10/100 GL and an ÄKTA Purifier (GE Healthcare). The protein was eluted with a linear gradient from 0 to 0.5 m NaCl in 20 mm HEPES, pH 7.5. Fractions containing LigZ were pooled and dialyzed into 20 mm HEPES, pH 7.5, concentrated to ~ 30 mg·mL⁻¹, and brought inside a Labmaster Model 100 glovebox (Mbraun, Peabody, MA, USA) to equilibrate overnight. The deoxygenated LigZ was incubated with 10 × molar excess of (NH₄)₂Fe(SO₄)₂·6H₂O, applied to Sephadex G-25 resin (GE Healthcare) to remove excess metal, flash frozen as beads in liquid N₂, and stored at −80 °C.

Protein analytical methods

Protein purity was evaluated using SDS-polyacrylamide gel stained with Coomassie Blue according to established procedures [21]. Protein concentrations were determined using micro BCA™ Protein Assay Kit (Thermo Fisher Scientific, Bothell, WA, USA) using bovine serum albumin as a standard. Iron concentrations were determined colorimetrically using the Ferene S assay and ferric chloride solution as a standard [22]. The metal species contained in the LigZ preparation were determined using a NexION 300d inductively coupled plasma mass spectrometry (ICP-MS) instrument (Perkin Elmer) calibrated using IV-Stock-4 synthetic standard (Inorganic Ventures). The protein samples were treated with concentrated HNO₃ and H₂O₂ as previously described [23].

Steady-state kinetics

Kinetic assays were performed by monitoring the consumption of O₂ using a Clark-type polarographic O₂ electrode OXYG1 (Hansatech, Pentney, UK) connected to a circulating water bath. Assays were performed in 1 mL of air-saturated 40 mm HEPES (I = 0.1 m, pH 7.5) at 25 °C, 0.2% DMSO and initiated by the addition of 5 nm of LigZ. Reaction velocities were corrected for the background reading prior to LigZ addition. The electrode was calibrated daily by a two-point calibration using air-saturated water and O₂-depleted water via addition of sodium hydrosulfite, according to the manufacturer's instructions. Stock solutions were prepared fresh daily. LigZ stock solution was prepared anaerobically, stored in a sealed vial on ice, and aliquoted using a gas-tight syringe for use. Steady-state kinetic parameters were evaluated by fitting the Michaelis–Menten equation to the data using the least-squares fitting of LEONORA [24].

To determine the reaction rate over a pH range of 6–9, a composite buffer containing 50 mm acetate, 50 mm MES, and 100 mm Tris was used. The assay contained 100 μm OH-DDVA and was initiated with the addition of 50 nm LigZ. To determine the apparent steady-state kinetic parameters for the organic substrate, the concentration of OH-DDVA was varied from 3.1 to 400 μm. To determine the apparent steady-state kinetic parameters for O₂, 400 μm OH-DDVA was used and the initial concentration of O₂ was varied by equilibrating the reaction mixture with humidified mixtures of O₂ or N₂ gasses. Final O₂ levels were normalized to the resting ambient O₂ level prior to the adjustment. The electrode was equilibrated with air-saturated buffer between runs.

Inactivation kinetics

The susceptibility of LigZ to inactivation was evaluated according to the different modes summarized in Scheme 1. The stability of LigZ in air-saturated buffer was evaluated by incubating the enzyme in the oxygraph cuvette under ambient conditions and measuring the residual activity, A_t, at different incubation times, t. Activity was determined by adding 400 μm OH-DDVA. The apparent first-order rate constant of inactivation, , was evaluated by fitting the following equation to the data where A_max is the activity observed without preincubation:

(1)

Scheme 1

Potential modes of inactivation of extradiol dioxygenases during steady-state turnover [15]. E* represents inactivated enzyme.

The apparent rate constant of inactivation during catalytic turnover in air-saturated buffer, , was calculated from the partition ratio and :

(2)

The partition ratio was obtained from the slope of a line fitted to a plot of total O₂ consumed versus the amount of LigZ added to the reaction. Total O₂ consumed was taken to represent the number of times LigZ turned over, and the amount of LigZ added to the reaction was determined from the reaction's initial velocity.

Cleavage product characterization

DCHM-HOPDA was prepared by reacting 200 μm OH-DDVA with excess LigZ (~ 1 μm) in air-saturated buffer. Reactions were monitored to completion using the oxygraph. LigZ was removed by ultrafiltration using a membrane with a 30 kDa cut-off. For NMR, DCHM-HOPDA was prepared using 5 mm potassium phosphate, pH 7.5, and the pooled filtrate was derivatized with 2 mm ammonium hydroxide and freeze dried. The dried sample was suspended in D₂O (Sigma-Aldrich, St. Louis, MO, USA) to a final concentration of ~ 4 mm. NMR spectra were collected using cryoprobe-equipped Bruker Avance III 600 MHz spectrometer and processed using ACD/NMR Processor Academic Edition version 12.01 (ACD/Labs). For high accuracy mass determination, DCHM-HOPDA prepared using 20 mm ammonium acetate, pH 8.0 was run on a LC-MS system equipped with 80 × 0.25 mm Luna 3 μm PFP2 analytical column (Phenomenex, Torrance, CA, USA). The pyridine derivative of DCHM-HOPDA was eluted using a mobile phase of 100 mm ammonium acetate, pH 4.5: methanol. Mass spectra were acquired using an Agilent 6460 Triple Quadrupole mass spectrometer equipped with an electrospray ion source operated in positive ion mode. Data were processed using MassHunter Qualitative Analysis ver. B.06.00 (Agilent Technologies, Santa Clara, CA, USA).

To identify the LigZ cleavage product of protocatechuate (PCA), the cleavage reaction was performed in 20 mm ammonium bicarbonate, pH 8.0 to form the pyridine derivative. Upon completion, the reaction was quenched with formic acid to a final concentration of ~ 1%. The pyridine derivative was resolved on a Waters 2695 Separation HPLC module (Milford, MA, USA) equipped with a Waters 2996 photodiode array detector and an Aqua 5 μm C18 250 × 4.6 mm column (Phenomenex). The derivative was eluted using 0.1% formic acid and monitored at 270 nm.

For stability and apparent pK_a determination, DCHM-HOPDA was prepared using 100 mm phosphate, pH 7.5 as described above. Aliquots of enzyme-free filtrate were titrated with sodium hydroxide from pH 7.6 to 13.2, as measured with a pH electrode, while keeping the concentration of DCHM-HOPDA constant at ~ 20 μm. Absorbance spectra were recorded using a Cary 60 UV-vis spectrophotometer (Agilent). The nonenzymatic transformation of DCHM-HOPDA was monitored spectrophotometrically at 480 nm. The half-life was evaluated by fitting a first-order exponential curve to the data. The pK_a values and relative chemical distributions were estimated using MarvinSketch 6.0.0 (ChemAxon, Cambridge, MA, USA).

Results and Discussion

Purification of LigZ

Heterologously produced LigZ was purified using immobilized metal ion affinity chromatography and anion exchange chromatographies to > 98% apparent homogeneity as judged from SDS/PAGE at yields of 5–10 mg·L⁻¹ of cell culture. However, this procedure yielded apoenzyme with no detectable iron or OH-DDVA-cleavage activity, likely due to the aerobic treatment and EDTA in the TEV protease buffer. Incubation of the LigZ with excess Fe²⁺ yielded preparations containing ~ 1.2 m ratio of iron and possessing a specific activity of 210 μmol·min⁻¹·mg⁻¹. This specific activity is approximately four orders of magnitude higher than what was previously reported [8].

Substrate specificity

In an oxygraph assay using a composite buffer, LigZ was most active at pH 7.5 (Fig. S1). Accordingly, the enzyme was subsequently characterized using 40 mm HEPES (I = 0.1 m), pH 7.5 at 25 °C. Under these conditions, the initial rate of O₂ consumption displayed Michaelis–Menten kinetics at up to 400 μm OH-DDVA (Fig. 2A), with apparent k_cat and K_m values for OH-DDVA of 129 ± 2 s⁻¹ and 6.0 ± 0.6 μm, respectively (Table 1). The apparent k_cat/K_m value of LigZ for OH-DDVA, 2.20 ± 0.02 × 10⁷ s⁻¹·m⁻¹, is comparable to what has been reported for BphC and protocatechuate 2,3-dioxygenase for their respective physiological substrates [16, 25, 26]. LigZ also displayed Michaelis–Menten behavior with respect to O₂ concentration (Fig. 2B) with a value of 67 ± 3 μm.

Figure 2

Steady-state kinetic analyses of the LigZ-catalyzed reaction. (A) Dependence of initial velocity on OH-DDVA concentration in air-saturated buffer. (B) Dependence of initial velocity on the oxygen concentration in the presence of saturating amount of OH-DDVA. Red lines represent fits of the Michaelis–Menten equation to the data using LEONORA.

Table 1. Apparent steady-state kinetic and inactivation parameters of LigZ

Substrate	(μm)	(s^−1)	/ (× 10^4 m^−1 s^−1)	Partition ratio	(× 10^−2 s^−1)	/ (m^−1 s^−1)
OH-DDVA	6.0 (0.6)	129 (2)	2200 (200)	18 000 (600)	0.72 (0.03)	1000 (100)
PCA	8000 (1000)	5.2 (0.3)	0.07 (0.01)	50 (5)	10 (1)	13 (2)

• Experiments were performed using 40 mm HEPES (I = 0.1 m), pH 7.5 at 25 ^°C. The values in parentheses represent standard errors.

The high specific activity of LigZ preparations afforded a reexamination of the enzyme's substrate specificity. As previously reported, LigZ did not detectably cleave catechol, 2,3-dihydroxybiphenyl (DHB), gallate, or ortho-methylgallate [8]. However, LigZ cleaved PCA with an apparent k_cat/K_m value of 7 ± 1 × 10² s⁻¹·m⁻¹ (Table 1). The addition of vanillate to the reaction mixture did not alter this value. Overall, the apparent preference of LigZ for OH-DDVA over PCA is ~ 30 000-fold. By comparison, the preference of BphC for DHB over catechol is ~ 350-fold [16].

Inactivation of LigZ

In the presence of higher concentrations of OH-DDVA, the activity of LigZ decreased to zero prior to substrate depletion. Moreover, O₂ consumption resumed upon addition of fresh enzyme. Because extradiol dioxygenases are O₂ labile, we investigated the stability of LigZ in air-saturated buffer at 25 °C (40 mm HEPES (I = 0.1 m) pH 7.5). Under these conditions, the enzyme's activity decayed with an apparent pseudo first-order rate constant of inactivation, , of 1.8 ± 0.2 × 10⁻² s⁻¹ (Fig. 3A), which corresponds to a half-life of 39 ± 4 s. Importantly, the activity of LigZ did not significantly decrease upon anaerobic incubation for up to 1 h, either with or without OH-DDVA, suggesting that j₂ and j₅ are negligible. Thus, LigZ seems to be quite O₂ labile. By comparison, the half-life of BphC, a Type I extradiol dioxygenase, is 16 ± 2 min [27] and that of LigAB, another Type II enzyme, is ~ 3 h [13]. This inactivation of LigZ appears to be due to the oxidation of the active site iron because the activity could be at least partially restored upon anaerobic incubation with Fe²⁺.

Figure 3

Inactivation of LigZ. (A) The decay of LigZ activity upon incubation of the enzyme in air-saturated buffer. The red line represents a fit of Eqn 1 to the data. (B) Total O₂ consumed in steady-state reactions as a function of LigZ concentration. The partition ratio was calculated from a linear fit (red line) of the data. That value for j₃^app was calculated from Eqn 2.

To determine the apparent rate constant of inactivation during catalytic turnover, , we measured the partition ratio. Values of 18 000 ± 900 and 50 ± 5 were measured for OH-DDVA (Fig. 3B) and PCA, respectively, which correspond to values of 7.2 ± 0.3 ms⁻¹ and 100 ± 10 ms⁻¹ (Table 1). These values are within a factor of two of those reported for the inactivation of BphC by DHB and catechol, respectively [16]. Nevertheless, PCA is unlikely to inactivate LigZ in vivo given the relatively low specificity of the enzyme for this substrate. Thus, the estimated / value, ~ 130 s⁻¹·m⁻¹, is about three orders of magnitude lower than that of BphC for either 3-Cl catechol [16] or 2′,6′-diCl DHB [28]. Finally, although the partition ratio of LigZ for OH-DDVA is significantly lower than that of BphC (85 000 ± 1000 for DHB) [16] and XylE (1 400 000 for catechol) [29] for their physiological substrates, it is comparable to that of LigAB (18 000 for PCA) [13]. The similar susceptibilities of LigZ and LigAB to inactivation during catalytic turnover do not have a ready structural explanation given that LigAB has a cap domain and LigZ does not.

Product identification

Cleavage of OH-DDVA by LigZ yielded a yellow-colored compound (λ_max = 480 nm), consistent with a dienolate anion. This species decayed rapidly in solution [t_½ = 5.5 ± 0.4 min, potassium phosphate (I = 0.1 m) pH 7.5 (Fig. 4)], complicating characterization. To facilitate identification of the MCPs derived from OH-DDVA and PCA, they were derivatized with ammonia to form stable pyridine derivatives (Fig. 1) [30]. High-resolution LC-MS of the OH-DDVA-derived compound revealed a parent ion with an m/z value of 334.0562, which agrees well with predicted mass of the singly protonated pyridine derivative of DCHM-HOPDA (334.0563). The identification of the pyridine derivative of DCHM-HOPDA was further validated by 1D proton NMR and HSQC analysis (¹H NMR (D₂O, 600 MHz): δ 8.52 (d, 1 H, Ar-H), 8.28 (d, 1 H, Ar-H), 8.21 (d, 1 H, Ar-H), 7.62 (d, 1 H, Ar-H), 3.97 (s, 3 H, OCH₃)). Moreover, while lacking in resolution, signal counts from 1D carbon NMR were consistent with 15 carbon nuclei found in the pyridine derivative of DCHM-HOPDA (¹³C NMR (D₂O, 600 MHz): 177.50, 175.17, 173.49, 158.99, 155.09, 152.37, 150.97, 150.65, 128.95, 123.95, 123.66, 123.13, 123.10, 120.12, 116.46, 58.58).

Figure 4

Decay of DCHM-HOPDA. DCHM-HOPDA was generated by cleaving OH-DDVA with an excess of LigZ (potassium phosphate (I = 0.1 m), pH 7.5). The dashed line depicts the nonenzymatic decay of the dienolate absorbance band (t_½ = 5.5 ± 0.4 min). The solid line depicts the decay of the dienolate absorbance band upon the addition of 200 nm LigY (arrow).

Similarly, the LigZ-catalyzed cleavage of PCA, performed in an ammonium bicarbonate buffer to trap the MCP as a pyridine derivative, yielded a product that comigrated with that formed by the PCD-catalyzed cleavage of PCA as well as 2,4-pyridinedicarboxylic acid (t_R = 9.2 min, λ_max = 274 nm) (Fig. 5). In contrast, 2,5-pyridinedicarboxylic acid eluted later (t_R = 11.5 min, λ_max = 270 nm). This result indicates that LigZ catalyzes the 4,5-cleavage of PCA to 4-carboxy-2-hydroxymuconate-6-semialdehyde.

Figure 5

HPLC chromatograms of the pyridine derivatives of PCA (green) cleavage products. LigZ- and PCD-catalyzed cleavage reactions were performed in ammonium bicarbonate buffer. 2,4-Pyridinedicarboxylic (pink) and 2,5-pyridinedicarboxylic (blue) acids were run as controls.

Nonenzymatic transformation of DCHM-HOPDA

As noted above, yellow-colored DCHM-HOPDA decayed rapidly in solution. The decay rate increased with pH and ionic strength, and gave rise to a stable species characterized by a λ_max of 340 nm (Fig. 6). This species is likely to be a γ-lactone similar to those formed by muconates and HOPDA [31, 32]. Consistent with this interpretation, the dienolate species was regenerated under alkaline condition (λ_max = 452 nm, pH > 13), and decayed again when the pH was neutralized. Moreover, equilibration between the dienolate and the lactone took several minutes after perturbation of the pH. Nevertheless, titration of DCHM-HOPDA with base yielded an apparent pK_a value of 11.3 (Fig. 6, inset). This value is commensurate with the predicted pK_a for the C2 hydroxyl of DCHM-HOPDA. In HOPDA and its chlorinated congeners, the pK_a of the C2 hydroxyl correlated with the stability of the dienolate in solution [33]. That is, a lower pK_a of the C2 hydroxyl favors the dienolate form.

Figure 6

UV-visible absorption spectra of DCHM-HOPDA. Spectra were recorded at pH 7.5 (solid line) and 13.0 (dashed line). The inset depicts the apparent pK_a determined by titrating the absorbance band at 450 nm.

The decay of the MCP was accelerated in the presence of LigY (Fig. 4), suggesting that the dienolate species is the substrate for LigY. The half-life of DCHM-HOPDA could be confirmed indirectly by quenching the nonenzymatic transformation of the dienolate with excess LigY after various nonenzymatic incubation times and quantifying the nonenzymatically transformed MCP by adding NaOH to generate the dienolate (Fig. S2). The half-life calculated in this manner was similar to that calculated by spectrophotometrically monitoring the decay of the dienolate (~ 5 min). This result also indicated that only the dienolate is a substrate for LigY. The nonenzymatic transformation of DCHM-HOPDA to the γ-lactone and inability of LigY to transform the latter explains the lack of hydrolysis product upon delayed addition of LigY [9]. At the same time, characterization of DCHM-HOPDA and its stability facilitates further characterization of LigY.

Acknowledgements

Dr. Mark Okon and Prof. Lawrence McIntosh assisted with the NMR spectroscopy. Mariko Ikehata and Prof. Michael E P Murphy assisted with the ICP-MS. This research was supported by Large Scale Applied Research Project 2108 from Genome Canada, Discovery Grant 171359 from the Natural Sciences and Engineering Research Council of Canada (NSERC) to LDE, and NSERC-CREATE through Genome Science and Technology program studentship to EK.

Author contributions

EK and LDE conceived and designed the research, analyzed and interpreted the data, and wrote the manuscript. EK conducted most of the experiments. HMC synthesized the HO-DDVA with supervision from SWG, and both contributed to product characterization. All authors read and approved the manuscript.