Biochemical studies on a versatile esterase that is most catalytically active with polyaromatic esters

Herein, we applied a community genomic approach using a naphthalene-enriched community (CN1) to isolate a versatile esterase (CN1E1) from the α/β-hydrolase family. The protein shares low-to-medium identity (≤ 57%) with known esterase/lipase-like proteins. The enzyme is most active at 25–30°C and pH 8.5; it retains approximately 55% of its activity at 4°C and less than 8% at ≥ 55°C, which indicates that it is a cold-adapted enzyme. CN1E1 has a distinct substrate preference compared with other α/β-hydrolases because it is catalytically most active for hydrolysing polyaromatic hydrocarbon (phenanthrene, anthracene, naphthalene, benzoyl, protocatechuate and phthalate) esters (7200–21 000 units g−1 protein at 40°C and pH 8.0). The enzyme also accepts 44 structurally different common esters with different levels of enantio-selectivity (1.0–55 000 units g−1 protein), including (±)-menthyl-acetate, (±)-neomenthyl acetate, (±)-pantolactone, (±)-methyl-mandelate, (±)-methyl-lactate and (±)-glycidyl 4-nitrobenzoate (in that order). The results provide the first biochemical evidence suggesting that such broad-spectrum esterases may be an ecological advantage for bacteria that mineralize recalcitrant pollutants (including oil refinery products, plasticizers and pesticides) as carbon sources under pollution pressure. They also offer a new tool for the stereo-assembly (i.e. through ester bonds) of multi-aromatic molecules with benzene rings that are useful for biology, chemistry and materials sciences for cases in which enzyme methods are not yet available.

because they partition and disorganize the cell membrane by removing lipids and proteins, which leads to cell death (Sikkema et al., 1995;von Wallbrunn et al., 2003). To cope with such activities, bacteria have developed multiple chemical tolerance mechanisms and an extensive enzyme arsenal (e.g. oxygenases, O-demetylases, CoA synthases and ligases, aldolases, alcohol dehydrogenases and α/β hydrolases) for mineralizing aromatic compounds by submitting them to the central metabolism (Pérez-Pantoja et al., 2008;Vilchez-Vargas et al., 2013).
Recent studies suggest that multifunctional esterase/ lipase-like proteins from the α/β hydrolase family that can hydrolyse both C-C and C-O bonds may exist in nature at much higher levels than previously thought . From an ecological perspective, such proteins may contribute to global carbon cycling processes for complex substrates, including recalcitrant organic pollutants. From a biotechnological perspective, such proteins may open unexpected research avenues for biotechnology applications. Esterases/lipases from the α/β hydrolase family have demonstrated such activities for catechol and biphenyl derivatives . However, the potential implication for the degradation of and biotechnology of complex organic molecules with two or more aromatic rings, specifically benzene rings (polyaromatic hydrocarbons -PAH), has not been described, even though the PAH degradation phenomenon is relatively well known (Pérez-Pantoja et al., 2008;Seo et al., 2009). Such activities are especially significant because the chemistry for PAH esters has been a subject of considerable interest due to their chemical and physical properties (e.g. polymers produced with versatile photo-reactivities) (Noh et al., 2001;Chae et al., 2002;Zhu et al., 2011). Heterogeneous catalysts have successfully been used to produce PAH-like (naphthalene, anthracene, phenanthrene, benzyl and phthalate) derivatives (Noh et al., 2001;Zhu et al., 2011;Maruyama et al., 2012). However, to the best of our knowledge, no study in the specialized literature has involved identification of potential applications for esterase/lipase-like proteins as catalysts for modifying PAHs.
Because contaminated environments and microbial communities derived therefrom are an excellent source for enzymes that act on aromatics (Pérez-Pantoja et al., 2008;Guazzaroni et al., 2013), we now ask whether such α/β hydrolases can be identified therein. We generated a subset of 5500 clones (in Escherichia coli EPI300-T1 R ; Epicentre Biotechnologies; Madison, WI, USA; Alcaide et al., 2013) from a naphthalene-enriched community (CN1) derived from PAH-contaminated soil (Guazzaroni et al., 2013), which included nearly 0.17 Gbp of community genomes, that were evaluated for their ability to hydrolyse α-naphthyl acetate (Reyes-Duarte et al., 2012). Four unique positive clones (hit rate 1:1375) were identified as active, and one (herein, CN1E1) was selected for in-depth analysis due to its high activity (halo/ colour formation) and capacity for hydrolysis in agarbased assays using the model PAH ester naphthalene carboxylic acid methyl ester (Fig. 1) (Altschul et al., 1997). Thirty-one predicted open reading frames were identified (Lukashin and Borodovsky, 1998); one encoded a putative esterase (CN1E1) with the α/β hydrolase fold. The protein [311 amino acids (AA); molecular weight (MW) 33 146 Da; isoelectric point (pI) 6.34] was produced in a soluble form (Fig. 2) upon expression in the pET-46 Ek/LIC vector (using the forward and reverse primers 5′-GACGACGACAAGATGGCGGTAGATCCG-3′ and 5′-GAGGAGAAGCCCGGTTATCTCGGTCCGGC-3′ respectively) and E. coli BL21 (DE3) (conditions described in Alcaide et al., 2013). The sequence was analysed, which Hydrolytic phenotype for E. coli BL21 (DE3) that either expressed or did not express CN1E1 using the pET-46 Ek/LIC vector. The cells were plated on fresh Luria Bertani (LB) plates with ampicillin (50 μg ml −1 ). The plates [with 1.0 mM isopropyl-β-Dgalactopyranoside (IPTG)] were incubated for 12 h at 37°C and then covered with a second layer that included the substrate [20 ml 5 mM N-(2-hydroxyethyl) piperazine-N′-(3-propanesulfonic acid) (EPPS) buffer, pH 8.0, 0.4% agarose and 320 μl of a naphthalene carboxylic acid methyl ester solution in acetonitrile (100 mg ml −1 )]. The active phenotype was apparent due to the colour change resulting from substrate hydrolysis and acetic acid formation, which changes the pH indicator from red to yellow.

Fig. 5.
Substrate profile for the wild-type CN1E1 α/β hydrolase using a set of structurally diverse esters. The specific activities were calculated in triplicate as described by Martínez-Martínez and colleagues (2013) at 40°C in 20 mM HEPES buffer pH 7.0 (for pNP esters) or 5 mM EPPS buffer pH 8.0 (for the remaining esters). The standard deviation (SD) for the triplicate assays is shown. The chemicals used for the enzymatic experiments were the purest grade available and were purchased from Fluka-Aldrich-Sigma Chemical Co. (St. Louis, MO, USA). The hydroxycinnamic-like esters were supplied by Apin Chemicals (Oxon, UK), the methyl phenanthrene-3-carboxylate and methyl phenanthrene-9-carboxylate were supplied by Wuhan Farthest Chemical (Mainland, China), and the anthracene-9-carboxylic acid methyl ester and anthracene-3-carboxylic acid methyl ester were obtained from Alfa Aesar (Karlsruhe, Germany). Insets (A) to (C) in the figure represent a zoom for substrates hydrolysed at low rates.
compounds were among the best substrates for CN1E1 (Fig. 5); this enzyme showed regio-selectivity for phenanthrene-3-carboxylate over methyl phenanthrene-9-carboxylate (2.7-fold) and, to a lesser extent, for anthracene-2-carboxylate over anthracene-9-carboxylate (1.7-fold). Although previous reports have shown that other esterases can degrade phthalate esters (Maruyama et al., 2005;Saito et al., 2010;Wu et al., 2013), to the best of our knowledge, the specialized literature has not reported an esterase with the capacity to degrade PAH esters with multiple aromatic (e.g. benzene) rings. The half-saturation (Michaelis) coefficient (K m), catalytic rate constant (kcat) and catalytic efficiency (kcat/Km) values were determined for the 14 best and structurally distinct substrates (Table 1). The parameters were determined as described previously  at pH 8.0 and 40°C (except for pNP-acetate were pH 7.0 was used to ensure substrate stability). The substrates analysed include pNP-acetate (the model p-nitrophenyl ester), triacetin (the model triacylglycerol), methyl bromoacetate (the model halogenated ester), tri-O-acetyl-glucal (the model polysaccharide-like ester), butyl acetate (the model alkyl ester), phenyl acetate (the model aryl ester), 2,5-dihydroxycinnamic acid methyl ester (the model hydroxycinnamic-like ester) and seven chemically distinct model PAH esters. As shown in Table 1, for catalytic efficiencies, pNP-acetate and triacetin were the most and least efficient substrates {[(k cat/Km)]pNPacetate/[(kcat/Km)]triacetin ∼36/1 factor}. The catalytic efficiency (kcat/Km) for PAH-like esters ranged from 4.5-(for methyl phenanthrene-9carboxylate) to 13.3-fold (for dimethyl phthalate) lower than for pNP-acetate, primarily due to a significant decrease in Km values (from 2.8-to 11.1-fold, depending on the substrate) for the pNP-acetate. The kcat/Km values were used to establish the following order: phenanthrene > naphthalene > benzoic > protocatechuic > anthracene > phthalate. In addition, the enzyme's marked regio-selectivity was demonstrated by a ∼ 3.0-fold higher efficiency for methyl phenanthrene-9-carboxylate compared with the 3-carboxylate ester.
Taken together, the results indicate that the esterase CN1E1 from a PAH-degrading microbial consortium includes a broad and unusual substrate profile. Threedimensional modelling (not shown) predicts a broad molecular environment in the CN1E1 active site, which is consistent with high-substrate accessibility and the specific activity and catalytic efficiency values for multiple multi-aromatic substrates ( Fig. 5 and Table 1). The observed functions may indicate metabolic and ecological capacities in vivo and provide research avenues for biotechnology applications. From an ecological perspective, the protein, which is likely an esterase/lipase, hydrolysed not only common esters but also (±)-pantolactone, γ-butyrolactone and γ-valerolactone; thus, CN1E1 may also be a lactone hydrolase. γ-Valerolactone hydrolases play a key role in degrading cyclopentanol (Iwaki et al., 2002), and the organism that produces CN1E1 may use cyclopentanol or cyclopentane as carbon sources. Notably, this protein also hydrolysed benzoic acid, 2-benzoyl-and methyl ester (Fig. 5), which has been proposed as an intermediate in the phenanthrene degradation pathway (Luan et al., 2006). In addition, this enzyme showed a high activity for both alkyl and benzyl groups on phthalate esters, which indicates that the esterase may also be involved in hydrolysing phthalate esters. Although ester formation is a common detoxification mechanism (e.g. Luan et al., 2006), little is known about whether PAH esters occur naturally. Because most of the aromatic esters investigated herein are known polymer components for plastic and pesticides, the results indicate that organism with the CN1E1 enzyme may initiate degradation of such recalcitrant molecules, which may be intro- Table 1. The half-saturation (Michaelis) coefficient (Km), catalytic rate constant (kcat) and catalytic efficiency (kcat/Km) values for the purified CN1E1 esterase for representative, structurally diverse esters. Three independent experiments at 40°C and pH 8.0 (5 mM EPPS buffer) were performed for each parameter, and the data are shown with the standard deviation. The parameters were determined using conditions and methods described previously . duced into ecosystems through anthropogenic activities or produced at intermediate stages in biodegradation pathways (Luan et al., 2006). A compositional similarity analysis using GOHTAM (Ménigaud et al., 2012) suggests that such organism is a α-proteobacterium in the order Rhizobiales (most likely Mesorizhobium); this was the only Rhizobiales member in the CN1 community detected by both full-length and partial 16S rRNA gene sequences analysis (Guazzaroni et al., 2013). Herein, we have enhanced our understanding of soil bacterial systems by demonstrating that aromatic molecule transformation mediated by esterases/lipase-like proteins may facilitate a deep breakdown of PAH components.
Our data show that CN1E1 is, to the best of our knowledge, the first efficient and catalytically active esterase from the α/β-hydrolase family for PAH ester hydrolysis; this suggests that this protein may be applied to generate a unique set of complex a la carte aromatic molecules with improved or unknown properties (Storms and Farrar, 1969;Noh et al., 2001;Chae et al., 2002;Yen et al., 2005;Jones and Sumner, 2006;Zhu et al., 2011;Kita et al., 2012;Maruyama et al., 2012). We believe that CN1E1 may expand the enzyme toolbox for new biotechnological opportunities involving heterocyclic aromatic compounds for future studies due to the inherent properties of enzymes compared with chemical heterogeneous-based processes and the unusual substrate range and preference and regio-and enantioselective properties of CN1E1 compared with previously reported esterases.