Characterization of a Novel Alkaline Family VIII Esterase with S-Enantiomer Preference from a Compost Metagenomic Library

Abstract

A novel esterase gene, est7K, was isolated from a compost metagenomic library. The gene encoded a protein of 411 amino acids and the molecular mass of the Est7K was estimated to be 44,969 Da with no signal peptide. Est7K showed the highest identity of 57% to EstA3, which is an esterase from a drinking water metagenome, when compared with the enzymes with reported properties. Est7K had three motifs, SMTK, YSV, and WGG, which correspond to the typical motifs of family VIII esterases, SxxK, Yxx, and WGG, respectively. Est7K did not have the GxSxG motif in most lipolytic enzymes. Three additional motifs, LxxxPGxxW, PLGMxDTxF, and GGxG, were found to be conserved in family VIII enzymes. The results of the phylogenetic analysis and the alignment study suggest that family VIII enzymes could be classified into two subfamilies, VIII.1 and VIII.2. The purified Est7K was optimally active at 40ºC and pH 10.0. It was activated to exhibit a 2.1-fold higher activity by the presence of 30% methanol. It preferred short-length p-nitrophenyl esters, particularly p-nitrophenyl butyrate, and efficiently hydrolyzed glyceryl tributyrate. It did not hydrolyze β-lactamase substrates, tertiary alcohol esters, glyceryl trioleate, fish oil, and olive oil. Est7K preferred an S-enantiomer, such as (S)-methyl-3-hydroxy-2-methylpropionate, as the substrate. The tolerance to methanol and the substrate specificity may provide potential advantage in the use of the enzyme in pharmaceutical and other biotechnological processes.

Introduction

Lipolytic enzymes, esterases and lipases, are carboxylic ester hydrolases. Esterases (E.C. 3.1.1.1) hydrolyze ester bonds of short fatty acids with less than 10 carbons, and lipases (E.C. 3.1.1.3) hydrolyze ester bonds of long fatty acids with more than 10 carbons. Esterases and lipases, particularly with their enantioselectivity, have a wide range of biological applications in the synthesis of biopolymers, pharmaceuticals, agrochemicals, and flavor compounds [12]. Bacterial lipolytic enzymes were originally grouped into eight families [1] and the classification has been expanded to 15 families with the recent discovery of many new lipolytic enzymes [6]. Originally, three enzymes that were approximately 380 residues long and showed striking similarity to several class C β-lactamases were classified as family VIII enzymes [1]. Although lipolytic family VIII enzymes share common motifs such as GxxK with the β-lactamase family, they are thought to be evolutionally rather loosely related to other esterases [1,30]. Very recently, several family VIII esterases from diverse metagenomic sources have been experimentally studied [3,7,21,29].

The microbial community in compost includes myriad microorganisms, and the indigenous microbes of compost produce various enzymes. The microbial community of compost is expected to be more diverse than that of any other environment, since it varies greatly depending on the nature of raw materials and the progress of composting [8,45]. Many compost microorganisms would not be cultivated in laboratories, and metagenomic approaches could be employed to exploit various valuable genes from compost microorganisms that are not easily culturable or totally unculturable. Biocatalysts have been isolated from uncultured microorganisms using a metagenomic approach [37,39].

We have previously reported a novel esterase, Est2K, from a compost metagenomic library [15]. In this study, we isolated another novel esterase gene, est7K, from the library and examined some properties of Est7K. The classification of the family VIII enzymes into subfamilies, VIII.1 and VIII.2, was also discussed with the analysis of additional motifs in the family VIII enzymes.

 

Materials and Methods

Selection of Esterase-Positive Clones

A compost metagenomic library was constructed using a fosmid vector, and esterase-positive clones were identified with their clear zones on the LB agar plates containing 1% glyceryl tributyrate (Sigma) as the substrate [15]. YH-E7, one of the 18 esterase-positive clones obtained, was used in this study.

Sequence Analysis and Construction of Phylogenetic Tree

The nucleotide sequence of the plasmid in an esterase-positive clone was determined by SolGent (Daejeon, Korea). The conserved region of the esterase was analyzed by BLASTp of NCBI (http://www.ncbi.nlm.nih.gov), and the signal peptide was predicted by SignalP 4.1 in CBS (http://www.cbs.dtu.dk/services/SignalP/) [33]. The molecular mass and pI of the encoded protein were analyzed via the ExPASy site (http://www.expasy.ch/tools/protparam.html). To construct a phylogenetic tree, the evolutionary history was inferred using the neighbor-joining method [38]. Multiple alignments of the amino acid sequences were performed using Clustal W [44] and analyzed with GeneDoc 2.7 [25]. Evolutionary analyses were conducted using the MEGA6 program [43].

Site-Directed Mutagenesis

Site-directed mutagenesis of the gene was conducted using a QuikChange II kit (Stratagene, Santa Clara, CA, USA) [22]. The primer pair for S73A mutation was 5’-CCGGATTGCTGCCATGACCAA-3’ (forward) and 5’-TTGGTCATGGCAGCAATCCGG-3’ (reverse). The primer pair for Y191A mutation was 5’-CGCAGTGGAATGCCTCTGTTTCAA-3’ (forward) and 5’-TTGAAACAGAGGCATTCCACTGCG-3’ (reverse). The primer pair for S356A mutation was 5’-TCATGCCCGCGGCCAAGGGCGAAT-3’ (forward) and 5’- ATTCGCCCTTGGCCGCGGGCATGA-3’ (reverse). After 15 cycles, the PCR products were treated with DpnI, and the resulting products were transformed to E. coli XL-1 blue supercompetent cells. The substitutions were confirmed by nucleotide sequencing.

Determination of Esterase Activity

Esterase activity was determined by measuring the amount of p-nitrophenol generated from p-nitrophenyl butyrate (Sigma), as described previously [15]. Unless otherwise stated, the reaction was performed for 1 min at 25℃ with 1 mM p-nitrophenyl butyrate in 50 mM Tris-HCl (pH 8.0) and the absorbance at 400 nm of the reaction mixture was continuously measured. One unit of esterase activity was defined as the amount of enzyme that generated 1 μmol of p-nitrophenol in 1 min under the conditions.

Purification of the Enzyme

Crude enzyme preparation and purification of the enzyme were performed as previously described [15,42]. Briefly, the transformants were grown in LB broth containing ampicillin for 12 h at 37℃, harvested, and dispersed in 50 mM sodium citrate (pH 5.5), and dialyzed against 50 mM Tris-HCl (pH 8.0) buffer. The enzyme was purified by High-Q (5 ml; Bio-Rad, CA, USA), CHT-II (5 ml; Bio-Rad), and t-Butyl HIC (5 ml; Bio-Rad) column chromatographies. The progress of the purification was monitored by determining the amount of protein [19] and SDS-PAGE on an 11.5% polyacrylamide gel [18].

Characterization of the Enzyme

The optimum temperature, thermostability, and optimum pH of the enzyme activity were determined as described previously [15]. The influences of cations and phenylmethylsulfonyl fluoride (PMSF), substrate specificity of the enzyme, and hydrolysis of the β-lactam antibiotic ampicillin were determined as described previously [15].

Enantioselectivity was analyzed using a pH shift assay; Est7K was reacted with 300 mM (R)- or (S)-methyl-3-hydroxy-2-methylpropionate in 20 mM Tris-HCl (pH 8.0) containing phenol red (2 g/l) [14]. The absorbance spectra of the solutions were recorded at 350–600 nm. For the hydrolysis of tertiary alcohol esters, Est7K was reacted with 25 mM t-butyl acetate, linalyl acetate, or α-terpinyl acetate in 20 mM Tris-HCl, pH 8.0, containing phenol red [24]. Hydrolysis of glyceryl esters (glyceryl butyrate and glyceryl trioleate) and oils (fish oil and olive oil) were analyzed with 1% substrates.

Nucleotide Sequence Accession Number

The nucleotide sequence of the esterase gene est7K has been deposited at GenBank under the accession number KP756684.

 

Results

Characterization of Esterase Gene est7K and Est7K

In the previous study, 18 esterase-positive subclones were obtained from the mixed DNA of the 19 active fosmid clones of a compost metagenome, and a novel esterase gene, est2K, was isolated and characterized from one of the subclones [15]. In this study, the other 17 subclones were tested for their lipolytic activity and 13 were found to be positive and four were very weakly positive. The sequences of the 13 putative lipolytic genes from the positive clones were determined and eight different lipolytic genes were identified (data not shown). The lipolytic gene of subclone YH-E7 showed the lowest similarity to the reported lipolytic genes and was selected for further study.

The recombinant plasmid of YH-E7 had an inserted DNA fragment of about 2.4 kb. Sequence analysis of the inserted DNA fragment revealed an ORF of 1,236 bp, and the ORF was identified to be an esterase gene with a domain that showed high similarity to the β-lactamase superfamily and was named est7K. The esterase gene est7K was expected to encode a protein of 411 amino acid residues with no signal peptide. Est7K was calculated to be an acidic protein of 44,969 Da with a theoretical pI of 5.89.

The amino acid sequence of Est7K showed the highest identity of 74% to that of β-lactamase of Citromicrobium sp. JLT1363 (WP_033193084), followed by 74% identity to the β-lactamase of C. bathyomarinum (WP_010239180), 71% identity to the β-lactamase of uncultured Sphingomonadales bacterium HF0500_24B12 (ADI19464), and 66% identity to the β-lactamase of Erythrobacter litoralis (WP_011413033). The above sequences have been deposited very recently and their enzymatic properties have not yet been reported. Among the family VIII lipolytic enzymes with reported enzymatic properties, EstA3 from drinking water metagenome [9] and EstF4K from soil/water metagenome [29] showed the highest identity of 57% to Est7K.

Analysis of the amino acid sequences of Est7K and 23 lipolytic family VIII enzymes revealed that Est7K contained conserved regions such as SMTK (73rd to 76th), YSV (191st to 193rd), and WGG (362nd to 364th) motifs (Fig. 1).

Fig. 1.Alignment of the amino acid sequences of Est7K (Accession No. KP756684) and related lipolytic enzymes. Sequences: esterases EstA3 and EstCE1 from drinking water and soil metagenomes (DQ022078 and DQ022079, respectively), putative β-lactamase class C EstF4K from soil and water metagenomes (JN001202), hypothetical protein Est22 (shown as Est22-S in this figure) of a soil metagenome (HQ156921), lipolytic enzyme LipBL from Marinobacter lipolyticus SM19 (FR719924), esterases EstM-N1 and EstM-N2 from an arctic soil metagenome (HQ154132 and HQ154133, respectively), EstC of a soil metagenome (FJ025785), Est2K from a compost metagenome (GQ426329), EstU1 from a soil metagenome (JF791800), Est22 from a leachate metagenome (KF052088), SBLip1 from a forest soil metagenome (JQ780827), Est01 from a biogas slurry metagenome (HQ444406), EstB from Burkholderia gladioli (AF123455), Lpc53E1 from a marine sponge metagenome (JQ659262), PBS-2 from Paenibacillus sp. PBS-2 (KF972440), EstBL from Burkholderia multivorans UWC10 (AAX78516), Lip8 from Pseudomonas aeruginosa LST-03 (AB126049), esterase III (Est III) from Pseudomonas fluorescens SIK WI (AAC60471), EstA (EstA-Sc) from Streptomyces chrysomallus X2 (CAA78842), EstA (EstA-An) from Arthrobacter nitroguajacolicus Rü61a (CAD61039), EstA (EstA-P) from Pseudomonas sp. LS107d2 (M68491), and Est from Arthrobacter globiformis SC-6-98-28 (AAA99492). Consensus sequences are shown with uppercase or lowercase letters. Number 6 represents the branched hydrophobic amino acid residues Val/Leu/Ile. The conserved motifs are boxed in yellow and newly identified motifs are boxed in red. The region with the xxSxG sequence is shown in a blue box.

The alignment suggests that, in addition to the three motifs mentioned above, three more motifs are conserved in most of the family VIII enzymes: LxxxPGxxW (181st to 189th) at the N-terminus of the YSV motif, PLGMxDTxF (221st to 229th), and GGxG (271st to 274th) (Fig. 1). The amino acid residues Gly54, Pro103, and Gly140, which can form β-turns in the 3D structure of the enzymes, are observed in all the family VIII enzymes (Fig. 1). Asp281 and Gly358 are conserved in the 19 enzymes of family VIII, but are replaced with Gly and Arg, respectively, in the five family VIII enzymes (Fig. 1). The GxSxG motif was found to be not conserved in all the family VIII enzymes. Only two enzymes, EstC and Est2K, have the GxSxG sequence; nine enzymes, including Est7K, have xxSxG (354th to 358th), and the other enzymes do not have the sequence at all (Fig. 1).

In the phylogenetic tree of Est7K and other related proteins, Est7K was clustered with family VIII esterases and the location of Est7K in the tree suggested Est7K to be a novel member of the family (Fig. 2). It was also suggested from the tree that the family VIII enzymes could be divided into two subfamilies (Fig. 2).

Fig. 2.Phylogenetic tree showing the evolutionary relationships and levels of homology of the lipolytic enzymes. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The analysis involved 24 amino acid sequences. Evolutionary analyses were conducted in MEGA6 [43].

Site-Directed Mutagenesis

Site-directed mutagenesis of Ser73 to Ala73 resulted in complete loss of enzyme activity, and that of Tyr191 to Ala191 caused almost complete loss (93.4%) of the activity. The mutagenesis of Ser356 to Ala356 did not cause any significant change in the enzyme activity (data not shown).

Purification of Est7K

Est7K produced by the clone YH-E7 was purified by High-Q, CHT-II, and t-butyl HIC chromatography. The recovery rate, purification fold, and specific activity of the enzyme were 15.2%, 67.5-fold, and 790.2 U/mg protein, respectively (data not shown). The specific activity of Est7K was 46.2 times higher than that of Est2K (17.1 U/mg protein) under the same assay conditions [15]. Purified Est7K appeared as a single band on an SDS-PAGE gel and had a molecular mass of 42.4 kDa (Fig. 3).

Fig. 3.SDS-PAGE of the purified Est7K. M, molecular weight markers; lane 1, crude extract of the clone; lane 2, pooled from High-Q chromatography; lane 3, pooled from CHT-II chromatography; lane 4, pooled from HIC chromatography.

Properties of Est7K

Est7K exerted its maximal activity at 40℃, and showed 80% of the maximal activity at 50℃ (Fig. 4A). Est7K retained 65% of the original activity after 30 min of heat treatment at 30℃, but was completely inactivated after 15 min at 40℃ (Fig. 4B). Est7K showed its maximal activity at pH 10 (Fig. 4C). Mono- and divalent cations showed no significant influence on the enzyme activity, except that 5 mM Cu2+ inhibited 53.9% of the activity (Table 1). PMSF at a concentration of 1 mM inhibited 62.4% of the activity (Table 1), thereby indicating that serine residue is responsible for the catalytic activity. Est7K was stable in the presence of polar organic solvents methanol and isopropanol, but was sensitive to acetonitrile, losing more than 65% of the activity in the presence of 5% and 30% acetonitrile (Table 1). The increase in the concentration of methanol from 5% to 30% resulted in a significant increase in the enzyme activity, from 107.5% to 208.5% (Table 1). No such effect was observed with isopropanol.

Fig. 4.Effects of temperature and pH on the enzyme activity of Est7K. (A) Optimum temperature; (B) thermostability; and (C) optimum pH. Enzyme activities were measured by a continuous method at each designated pH. For more details, refer to the Materials and Methods section.

Table 1.aRelative activity was expressed as specific activities relative to the activity 790.2 U/mg protein. The values represent the average of the results from independent triplicate experiments.

When pNP esters were used as substrates, purified Est7K efficiently hydrolyzed ester bonds of short-chain fatty acids, and the relative activities toward pNP-butyrate (C4), pNP-acetate (C2), pNP-octanoate (C8), and pNP-caprate (C10) were 100%, 59.3%, 11.4%, and 3.5%, respectively (Table 2). The ratio of Est7K activity for pNP-C10 to pNP-C4 was very low (0.035). The value was similar to those of some family VIII esterases, EstCE1 (0.03) and EstA3 (0) [9], but was lower than those of other family VIII esterases, Est2K (0.4) [15] and Est22 (0.3) [21].

Table 2.ND, not determined.

The Km values for pNP-C4, pNP-C2, pNP-C8, and pNP-C10 were 61.7, 123.9, 86.6, and 25.0 μM, respectively (Table 2). Est7K showed no or negligible β-lactamase activity toward ampicillin and nitrocefin under the experimental conditions, even though Est7K was grouped as the same family VIII with class C β-lactamases (data not shown). Est7K hydrolyzed an S-enantiomer, (S)-methyl-3-hydroxy-2-methylpropionate, more efficiently than an R-enantiomer (Fig. 5A). The conversion ratio of R:S was calculated to be 1:2.53 by comparing A560 after 10 min reaction (Fig. 5B). Est7K could not hydrolyze the tertiary alcohols such as t-butyl acetate, linalyl acetate, or α-terpinyl acetate (data not shown). Furthermore, Est7K efficiently hydrolyzed glyceryl tributyrate, but did not hydrolyze glyceryl trioleate, fish oil, and olive oil (Fig. 5C).

Fig. 5.Analysis of enantioselectivity and lipid hydrolysis of Est7K using pH shift assay. (A) Enantioselectivity analysis with (R)- or (S)-methyl-3-hydroxy-2-methylpropionate containing phenol red: 1, (R) substrate; 2, (R) substrate + enzyme; 3, (S) substrate; 4, (S) substrate + enzyme. (B) Absorbance spectra of the reaction mixtures in (A). (C) Hydrolysis of glyceryl esters and oils.

 

Discussion

Est7K showed high identities to the recently reported β-lactamases of unreported enzymatic properties. Of approximately 100 listed sequences at NCBI BLAST that shows over 49% identity with Query coverage of 97%, only a few sequences have been published with information on their enzymatic properties; EstA3 from drinking water metagenome, EstCE1 from soil metagenome [9], EstF4K from soil/water metagenome [29], and Est22 from soil metagenome [23]. Recently, enzymatic properties of several lipolytic family VIII enzymes have been reported: Est22 from leachate metagenome [21], SBLip1 from forest soil metagenome [3], Est01 from biogas slurry metagenome [7], and PBS-2 from Paenibacillus sp. PBS-2 [16].

The SMTK motif in Est7K corresponds to the SxxK motif found in most of the β-lactamase superfamily [1,46]. The YSV and WGG motifs of Est7K correspond to the YSL (or YXN) and WGG (or KTG box) motifs, respectively, that are conserved in family VIII enzymes [46]. It has been suggested that the Ser in the SxxK motif functions as a nucleophile, and the Tyr in the YSL motif activates the Ser as a general base for attacking the ester-carbonyl of the substrate molecule and is stabilized by the proximity of the side chains of Lys in the SxxK motif and Trp in the WGG otif [36,46]. The loss of enzyme activity due to the site-directed mutagenesis of Ser73 to Ala73 and of Tyr191 to Ala191 indicates that the Ser in the SMTK motif and Tyr in the YSV motif are involved in catalytic function. The conserved residue Gly140 is positioned at the end of the motif LLxHxxG, which was described as a family VIII motif [32,34]. However, the motif is not highly conserved, except that the Gly is absolutely conserved and the first residue is moderately conserved as branched hydrophobic amino acids Leu/Val/Ile.

The typical pentamotif GxSxG has been reported to contain a serine residue that has a catalytic function in most esterase families [4]. The GxSxG sequence was found not to be conserved in Est7K as in most of the family VIII enzymes aligned. The pentamotif sequence was reported to be only partially conserved when a family VIII esterase, EstA from Arthrobacter nitroguajacolicus, two other family VIII esterases, and AmpC were aligned [40]. No loss of the enzyme activity by the site-directed mutagenesis of Ser356 in Est7K to Ala356 indicates that the Ser residue in the xxSxG sequence has no important catalytic function, although Est7K has the xxSxG (354th to 358th) sequence. The motifs GMS372RG of Est2K, GMS373EG of EstC, GIS149DG of EstB, and GLS321VG of LipBL were reported to be the GxSxG motif, but the Ser in the motifs was found not to be essential for the catalytic activity [15,30-32,36]. The role of the nucleophilic Ser residue in the SxxK motif in the family VIII enzymes is thought to be the same as that of the Ser residue in the GxSxG motif in most of other lipolytic enzymes and in the GDSL motif in family II (GDSL family). Based on the above results, it could be concluded that the catalytic residues in the active sites of the family VIII esterases are Ser, Lys, and Tyr, whereas those of the superfamily of α/β-hydrolases, including lipolytic enzymes, are Ser, Asp, and His, which form a catalytic triad [1,28].

In the phylogenetic tree of family VIII enzymes, five enzymes that showed identities lower than 25% to Est7K were clustered together and were clearly separated from other enzymes (Fig. 2). The five enzymes were Est from A. globiformis [26], EstA from A. nitroguajacolicus [40], EstA from Streptomyces chrysomallus [2], esterase III from Pseudomonas fluorescens [17], and EstA from Pseudomonas sp. LS107d2 [20]. Interestingly, the alignment analysis showed that the conserved residues in the six motifs of the five enzymes were different from those of the other 19 enzymes: (i) the second residue in SxxK of the five enzymes is C (M/Vin the 19 enzymes); ii) the second and third residues in YSV is H/E and A, respectively; iii) the first residue in WGG is H; iv) the first and the last residues in LxxxPGxxW are P and H/F, respectively; v) the last residue in PLGMxDTxF is V/L; and vi) the first and the last residues in GGxG are W/S and A/M, respectively (Table 1). With these results, it is suggested that family VIII be classified into two subfamilies, VIII.1 and VIII.2, and that subfamily VIII.1 include the reference enzyme Est from A. globiformis (AAA99492), as in the first classification of family VIII by Argipny and Jaeger [1]. Family I lipolytic enzymes have already been divided into six subfamilies [1].

Enzymes belonging to lipolytic family VIII are listed with some of their characteristics in order, based on their identities to Est7K, in Table 3. Est7K had no signal peptide like most of the lipolytic family VIII enzymes reported. To date, only five lipolytic family VIII enzymes have been reported to have a signal peptide: EstC [36], Est2K [15], EstU1 [13], Est22 [21], and SBLip1 [3]. The five enzymes were d erived from m etag enomic s ources a nd s howed identities of 34–40% to Est7K (Table 3).

Table 3.aEnzymes are listed in order based on their identity to Est7K and availability of experimental data. bp-Nitrophenyl esters; crelative activities (%) at the concentration of solvents. dNumerals in parentheses represent query coverage; eTriacetin; fEster bond, not β-lactam ring, of 7-amino cephalosporinic acid was cleaved; gstereoselectivity for production of (+)-trans-chrysanthemic acid. -, Negative activity; +, positive activity; ++, high activity; L, low activity. Blank, not available.

Est7K was a mid-sized enzyme with 411 amino acid residues among the family VIII enzymes that had 378 to 445 amino acid residues (Table 3). The optimum temperature (40℃) of Est7K was in the middle range compared with those of other family VIII enzymes (20–80℃) and the optimum pH of Est7K was in the alkaline region, pH 10.0 (Table 3).

Est7K was tolerant to methanol and a more than 2-fold increase in the activity was observed in the presence of 30% methanol. A similar effect of methanol was reported with EstF4K (175.8% at 30% methanol) and Lpc53E1 (220% at 20% methanol), and a more striking effect with EstC (600% at 20% methanol) (Table 3).

Est7K, like most of the family VIII enzymes, preferred a short-chain fatty acid (C4) as the substrate (Table 3). Only Lpc53E1 preferred a long-chain fatty acid (C16) as the substrate [41]. The substrate preference indicates that Est7K is a typical carboxylesterase rather than a lipase [1]. The substrate specificity of Est7K might have been changed during the evolutionary process, as it has been suggested for the family VIII esterases that did not show β-lactamase activity due to steric reasons [46] or the length of the Ω-loop [5]. Est7K showed moderate to low levels of enantioselectivity for (S)-enantiomer of methyl-3-hydroxy-2-methylpropionate. Although EstCE1, which has low identity to Est7K, was highly enantioselective for (+)-menthyl acetate [9], the enzymes with high identity to Est7K, EstA3, EstF4K, and LipBL showed relatively low levels of enantioselectivity [9,29,31] (Table 3).

Est7K did not hydrolyze tertiary alcohol esters. Only EstC and EstB have been reported to hydrolyze the esters of tertiary alcohol [36,46] (Table 3). No other enzymes have been studied for their ability to hydrolyze tertiary alcohols (Table 3). The GGGX motif, which is located in the active site adjacent to the oxyanion, is linked with the hydrolysis of esters of tertiary alcohols [10,36]. Although the GGGL sequence in EstC, corresponding to the GGGX motif, was found to be located toward the C-terminus of the enzyme, the ability of EstC to hydrolyze linalyl acetate was suggested to support the importance of this motif [36]. Many of the family VIII enzymes, including Est7K, have the GGGL sequence at the corresponding position of EstC (Fig. 1). However, Est7K could not hydrolyze the tertiary alcohol, and hydrolysis by other enzymes has not yet been reported (Table 3). EstB hydrolyzing the tertiary alcohol has a GAGM sequence, but not GGGX. It is likely that the GGGL sequence in Est7K and EstC is not responsible for the hydrolysis of the tertiary alcohol.

In this study, we characterized a new family VIII esterase, Est7K, without the GxSxG motif found in most of the β-lactamase superfamily. The family VIII enzymes were found to have three additional motifs conserved and suggested to be classified into two subfamilies, VIII.1 and VIII.2. The tolerance to methanol and the enantioselectivity of Est7K may provide potential advantage in the use of the enzyme in pharmaceutical and other biotechnological processes.