Molecular Characterization of Novel Family IV and VIII Esterases from a Compost Metagenomic Library

Two novel esterase genes, est8L and est13L, were isolated and identified from a compost metagenomic library. The encoded Est8L and Est13L had molecular masses of 33,181 and 44,913 Da consisting of 314 and 411 amino acids, respectively, without signal peptides. Est8L showed the highest identity (32.9%) to a hyper-thermophilic carboxylesterase AFEST from Archaeoglobus fulgidus compared to other esterases reported and was classified to be a novel member of family IV esterases with conserved regions such as HGGG, DY, GXSXG, DPL, and GXIH. Est13L showed the highest identity (98.5%) to the family VIII esterase Est7K from the metagenome library. Est8L and Est13L had the highest activities for p-nitrophenyl butyrate (C4) and p-nitrophenyl caproate (C6), respectively, and Est13L showed a broad substrate specificity for p-nitrophenyl substrates. Est8L and Est13L effectively hydrolyzed glyceryl tributyrate. The optimum temperatures for activities of Est8L and Est13L were identical (40 °C), and the optimum pH values were 9.0 and 10.0, respectively. Est13L showed higher thermostability than Est8L. Sephacryl S-200 HR chromatography showed that the native form of Est8L was a dimer. Interestingly, Est13L was found to be a tetramer, contrary to other family VIII esterases reported. Est8L was inhibited by 30% isopropanol, methanol, and acetonitrile; however, Est13L was activated to 182.9% and 356.1%, respectively, by 30% isopropanol and methanol. Est8L showed enantioselectivity for the S-form, but Est13L showed no enantioselectivity. These results show that intracellular Est8L and/or Est13L are oligomeric in terms of native forms and can be used for pharmaceutical and industrial applications with organic solvents under alkaline conditions.


Introduction
Lipolytic enzymes, such as esterase (EC 3.1.1.1) and lipase (EC 3.1.1.3), hydrolyze the ester bond to carboxylic acid and alcohol. Esterases hydrolyze ester bonds in short chains with <10 carbons, whereas lipase hydrolyzes ester bonds in long chains with >10 carbons [1]. The application of esterases and lipases has been widely researched for chemical reactions and for application in the pulp and paper, cosmetics, and pharmaceutical industries [2,3]. Especially, esterases are applied for the improvement of the lipophilicity for ester prodrugs to increase cell wall penetration [4]. Bacterial esterases were first classified into eight families based on their primary amino acid sequences [1], and recently other families have been reported up to the 19th family [5].
The family IV esterase, which is called hormone-sensitive lipase (HSL), has a catalytic triad consisting of serine (S), aspartic acid (D), and histidine (H). The serine residue in the catalytic triad is placed in the conserved GXSXG motif [6]. The bacterial HSL (bHSL) is a member of the α/β hydrolase family, which have similar structures, i.e., β-sheets covered by α-helixes [7]. The bHSL is divided into two domains: cap and catalytic domains. The catalytic domain forms α-helix and β-sheets with catalytic residues, and the cap domain were analyzed and performed using the Clustal W method in DNA/MAN (Lynnon Biosoft, version 4.11, Quebec, Canada). Phylogenetic trees were constructed by MEGA version X [23] using the maximum likelihood method.

Preparation of Crude Enzymes
Esterase-positive clones were inoculated in 200 mL LB broth and incubated for 15 h at 200 rpm and 37 • C. After incubation, the medium was centrifuged for 15 min at 6000× g and 4 • C. The pellet was washed two times with 20 mL of 20 mM Tris-HCl (pH 8.0) and centrifuged for 5 min at 6000× g and 4 • C. The pellet was resuspended in 5 mL of the same buffer and sonicated three times under an amplitude of 38% and pulse on for 1 sec and pulse off for 1 sec for 1 min using a microtip equipped on a sonicator (VCX500, Sonics & Materials, Newtown, CT, USA). After centrifugation for 15 min at 6000× g and 4 • C, the supernatant was collected as a crude enzyme.

Purification of the Enzymes and Activity Staining
The crude enzymes were separated further using a Biologic LP System (Bio-Rad). HiTrap Q anion exchange chromatography was performed using 20 mM Tris-HCl (pH 8.0) as a low buffer and 20 mM Tris-HCl (pH 8.0) containing 1 M NaCl as a high buffer at a flow rate of 1.0 mL/min for 90 min. The HiTrap t-butyl HIC was employed as a second column using 50 mM sodium phosphate (pH 7.0) buffer containing 1.5 M (NH 4 ) 2 SO 4 as a high buffer and 50 mM sodium phosphate (pH 7.0) buffer as a low buffer at a flow rate of 1.0 mL/min for 60 min. The HiTrap capto MMC was performed for further purification using 25 mM sodium acetate (pH 4.5) buffer as a low buffer and 50 mM sodium phosphate (pH 7.0) buffer including 1 M NaCl as a high buffer at a flow rate of 0.5 mL/min for 2 h. Size exclusion chromatography was performed using Sephacryl S-200 with 50 mM sodium phosphate (pH 7.0) containing 0.15 M NaCl at a flow rate of 0.5 mL/min for 240 min. Molecular mass standards were used with β-amylase, bovine serum albumin (BSA), and trypsinogen (200, 66.4, and 24.0 kDa, respectively).
To confirm the activity and molecular mass of each cloned enzyme at the same time, HiTrap Q pools were separated by native polyacrylamide gel electrophoresis (PAGE) without sodium dodecyl sulfate (SDS) and heat treatment of the sample, and the gel was analyzed by activity staining. The native gel was soaked with 50 mM Tris-HCl (pH 8.0) for 30 min, tightly overlapped to a 3% agar strip containing 1.5% glyceryl tributyrate, wrapped, and incubated for 30 min at 40 • C until a clear band was visible. The region of the gel corresponding to the clear band of the strip was sliced, chopped, and eluted in 20 mM Tris-HCl buffer overnight at 4 • C. After centrifugation for 5 min at 10,000× g and 4 • C, the supernatant was collected and used as a purified enzyme. To verify the purity, SDS-PAGE was performed by loading the samples in 11.5% acrylamide gel [24]; then, proteins in the gel were stained with silver. Protein concentration was determined by the Bradford assay [25].

Enzyme Assays
The esterase activity was calculated as the amount of p-nitrophenol, which is the product of the reaction between esterase and p-nitrophenyl ester. The enzyme was added into 50 mM Tris-HCl buffer (pH 8.0) containing 1 mM pNPB. The absorbances of the reaction mixtures were measured continuously for 2 min at 400 nm and 25 • C as a standard assay using the kinetic mode of a spectrophotometer (Optizen, K-Lab, Daejon, Korea). One unit of the enzyme activity was the enzyme amount producing 1 µmol of p-nitrophenol per minute using a molar extinction coefficient of 16,400/M/cm at pH 8.0. Acetylcholinesterase and butyrylcholinesterase activities were measured using ATCI and BTCI, respectively, as described previously [26]. Enzymes reacted with 0.5 mM substrate and 0.5 mM DTNB in 100 mM sodium phosphate (pH 7.5), and the reaction was monitored continuously for 10 min at 412 nm and 25 • C in the kinetic mode.

Characterization of Enzymes
The optimum temperature was identified using the standard assay method at 20, 30, 40, 50, 60, and 70 • C. The optimum pH was determined using 50 mM universal buffer (boric acid/citric acid/trisodium orthophosphate) in the range of pH 6.0 and 11.0, and molecular extinction of each pH was used, as described previously [18]. The heat stability of the purified enzyme was analyzed by preincubating at 30, 40, 50, and 60 • C for 1 h at designated times.
Substrate specificity was investigated using 1 mM p-NP esters (C2~C16). Five different concentrations of substrate, i.e., pNPB (C4), were used for the kinetic study, and Lineweaver-Burk plots were constructed to calculate the K m and V max .
The lipid hydrolysis activity of the enzyme was confirmed using a pH shift assay [27] using 1% oils (fish oil and olive oil) or glyceryl triesters (glyceryl tributyrate, glyceryl trioctanoate, and glyceryl trioleate) and 0.1% phenol red in 20 mM Tris-HCl (pH 8.0). Absorbance was measured using a spectrophotometer in the kinetic mode at every 5 min for 60 min at 560 nm and 25 • C, and the amount of remaining substrate was determined. To confirm the enantioselectivity of the enzyme, 1% (R)-or (S)-methyl-3-hydroxy-2-methylpropionate was used as a substrate [18].

Accession Numbers of the Est8L and Est13L
The sequences of est8L and est13Lwere deposited under accession numbers MZ484407 and MZ484408 at GenBank, respectively.

Sequence Analyses and Multiple Alignments of Est8L and Est13L
The positive clones YH-E8 and YH-E13 were selected in this study based on their halo sizes. The insert DNA of YH-E8 was 3202 bp in length, and an open reading frame (ORF) of 945 bp coding an esterase was identified and named est8L. The insert DNA of YH-E13 was 3871 bp long, and an ORF of an esterase consisting of 1236 bp was identified and named est13L. The encoded proteins Est8L and Est13L had molecular weights of 33,181 and 44,913 Da, respectively, were consisted of 314 and 411 amino acid residues, respectively; their theoretical pI values were 4.66 and 6.34, respectively. Est8L and Est13L had no signal peptides, as observed by SignalP-5.0 analysis. Analysis of the phylogenetic tree predicted that Est8L was a novel member of family IV esterases, and Est13L was a novel member of family VIII esterases ( Figure 1). A BLASTp search showed that Est8L has the highest identity (100%) for an alpha/beta hydrolase from Sphingorhabdus sp. (GenBank MBF6602187) in annotated sequences. However, Est8L showed the highest identity (32.9%) to a hyper-thermophilic carboxylesterase AFEST from Archaeoglobus fulgidus compared to other esterases enzymatically reported to date. Est13L had the highest identity (98.5%) for the family VIII esterase Est7K from an uncultured bacterium (AJN91095).

Purification of Est8L and Est13L
Each enzyme of Est8L and Est13L was purified using HiTrap Q as the first column. Both enzymes were bound to the resin and eluted through a gradient step (Supplementary Figures S1a and S2a). Next, a t-butyl HIC column was employed as a second column; however, recovery of Est8L activity was too low to be detectable, whereas Est13L was successfully purified. Therefore, HiTrap capto MMC and HIC columns were used as the second columns for Est8L and Est13L, respectively (Supplementary Figures S1b and S2b). In SDS-PAGE, purified Est8L was observed with smear bands at the predicted position of the gel, i.e., corresponding to~33 kDa (Figure 3a), probably because of the internal proteolytic cleavage of foreign proteins as in other cases [28][29][30][31][32].
Est13L appeared as a major band at the predicted position, corresponding to~45 kDa ( Figure 3b).
Figures S1a and S2a). Next, a t-butyl HIC column was employed as a second colum however, recovery of Est8L activity was too low to be detectable, whereas Est13L w successfully purified. Therefore, HiTrap capto MMC and HIC columns were used as t second columns for Est8L and Est13L, respectively (Supplementary Figures S1b and S2 In SDS-PAGE, purified Est8L was observed with smear bands at the predicted position the gel, i.e., corresponding to ~33 kDa (Figure 3a), probably because of the internal pro olytic cleavage of foreign proteins as in other cases [28][29][30][31][32]. Est13L appeared as a ma band at the predicted position, corresponding to ~45 kDa (Figure 3b). Through this procedure, the purification fold and yield of Est8L were 11.9% a 12.1%, respectively, and those of Est13L were 40.3% and 43.2%, respectively (Table 1). To confirm the size of each cloned enzyme, activity staining was performed usi HiTrap Q pools after native-PAGE. In the overlapped-agar strips, the respective acti bands of Est8L and Est13L were observed (Figure 4a,b). The region of the gel correspon ing to the clear band of the strip was sliced, and then proteins were recovered and an lyzed by SDS-PAGE to determine their molecular masses through silver staining. A result, Est8L and Est13L were observed at the bands corresponding to 33 and 45 kD respectively, similar to their expected values (Figure 4c,d). Through this procedure, the purification fold and yield of Est8L were 11.9% and 12.1%, respectively, and those of Est13L were 40.3% and 43.2%, respectively (Table 1). To confirm the size of each cloned enzyme, activity staining was performed using HiTrap Q pools after native-PAGE. In the overlapped-agar strips, the respective active bands of Est8L and Est13L were observed (Figure 4a,b). The region of the gel corresponding to the clear band of the strip was sliced, and then proteins were recovered and analyzed by SDS-PAGE to determine their molecular masses through silver staining. As a result, Est8L and Est13L were observed at the bands corresponding to 33 and 45 kDa, respectively, similar to their expected values (Figure 4c,d). . Activity staining of Est8L (a) and Est13L (b) onto the agar strips after native-PAGE and analysis of recovered Est8L (c) and Est13L (d) eluted from the activity-stained agar strips using SDS-PAGE and staining with silver nitrate. Native-PAGE was performed using the HiTrap Q fraction with the highest activity, and activity staining was performed according to the procedure described in the Materials and Methods. Red arrows in (a,b) represent the active bands for the enzymes. Red arrows in (c,d) indicate the predicted positions of Est8L and Est13L, respectively. S, size markers; C, crude extracts; Q, fraction pools from HiTrap Q column chromatography; A, pools from activity staining.

Determination of Molecular Masses of Native Est8L and Est13L
To determine the molecular masses of native forms of Est8L and Est13L, size exclusion chromatography using a Sephacryl S-200 HR column was performed. The elution volumes of Est8L and Est13L were 61.5 and 47.5 mL, respectively, and their molecular masses were predicted to be 67.2 and 160.0 kDa, respectively. These results indicated that their native forms are a dimer and tetramer, respectively ( Table 2).  . Activity staining of Est8L (a) and Est13L (b) onto the agar strips after native-PAGE and analysis of recovered Est8L (c) and Est13L (d) eluted from the activity-stained agar strips using SDS-PAGE and staining with silver nitrate. Native-PAGE was performed using the HiTrap Q fraction with the highest activity, and activity staining was performed according to the procedure described in the Materials and Methods. Red arrows in (a,b) represent the active bands for the enzymes. Red arrows in (c,d) indicate the predicted positions of Est8L and Est13L, respectively. S, size markers; C, crude extracts; Q, fraction pools from HiTrap Q column chromatography; A, pools from activity staining.

Determination of Molecular Masses of Native Est8L and Est13L
To determine the molecular masses of native forms of Est8L and Est13L, size exclusion chromatography using a Sephacryl S-200 HR column was performed. The elution volumes of Est8L and Est13L were 61.5 and 47.5 mL, respectively, and their molecular masses were predicted to be 67.2 and 160.0 kDa, respectively. These results indicated that their native forms are a dimer and tetramer, respectively ( Table 2).

Properties of Est8L and Est13L
Est8L and Est13L were optimally active at 40 • C, and the optimum pH values of Est8L and Est13L were 9.0 and 10.0, respectively, indicating that both are alkaline esterases ( Figure 5). Est13L had better thermal stability with a half-life of 3.2 min at 60 • C, whereas Est8L showed a half-life of 3.2 min at 50 • C ( Figure 6).

Properties of Est8L and Est13L
Est8L and Est13L were optimally active at 40 °C, and the optimum pH values of Est8L and Est13L were 9.0 and 10.0, respectively, indicating that both are alkaline esterases (Figure 5). Est13L had better thermal stability with a half-life of 3.2 min at 60 °C, whereas Est8L showed a half-life of 3.2 min at 50 °C ( Figure 6). (a) (b) Figure 6. Thermal stabilities of Est8L (a) and Est13L (b). The enzyme was preincubated at the designated temperatures for the indicated times prior to activity measurements.

Properties of Est8L and Est13L
Est8L and Est13L were optimally active at 40 °C, and the optimum pH values of Est8L and Est13L were 9.0 and 10.0, respectively, indicating that both are alkaline esterases (Figure 5). Est13L had better thermal stability with a half-life of 3.2 min at 60 °C, whereas Est8L showed a half-life of 3.2 min at 50 °C (Figure 6). (a) (b) Figure 6. Thermal stabilities of Est8L (a) and Est13L (b). The enzyme was preincubated at the designated temperatures for the indicated times prior to activity measurements.
Est8L was significantly inhibited to 0%, 13.2%, and 29.0% at 5 mM Cu 2+ , Zn 2+ , and Cu 2+ , respectively, whereas Est13L was not significantly inhibited by most ions tested, except moderate inhibitions to 39.1% and 58.8% at 5 mM Cu 2+ and Zn 2+ , respectively ( Figure  10). In the hydrolysis of glyceryl esters, Est8L and Est13L preferred glyceryl tributyrate (C4). After hydrolysis for 60 min under the condition described in Materials and Methods, Est8L and Est13L showed similar patterns for hydrolysis of glyceryl triesters and oils; for glyceryl tributyrate, they showed hydrolysis of 16.1% and 15.6%, respectively, and for fish oil, they showed hydrolysis of 86.7% and 82.6%, respectively. Both enzymes showed no significant activities with glyceryl trioleate (C18) and olive oil ( Figure 11).  In the hydrolysis of glyceryl esters, Est8L and Est13L preferred glyceryl tributyrate (C4). After hydrolysis for 60 min under the condition described in Materials and Methods, Est8L and Est13L showed similar patterns for hydrolysis of glyceryl triesters and oils; for glyceryl tributyrate, they showed hydrolysis of 16.1% and 15.6%, respectively, and for fish oil, they showed hydrolysis of 86.7% and 82.6%, respectively. Both enzymes showed no significant activities with glyceryl trioleate (C18) and olive oil ( Figure 11). In enantioselectivity, Est8L hydrolyzed the S-form with 10.1% higher activity than the R-form, suggesting preferences for the S-form, whereas Est13L showed no selectivity for the enantiomers (Figure 12). In the t-test, it was observed that the enantioselectivity for S-form was significant compared to R-form with a p-value < 0.05.
In comparison with family IV esterases from similar compost metagenomic sources, Est8L had poor organic solvents stabilities, whereas EstCs1 from a compost metagenomic library was stable at 30% isopropanol, ethanol, acetonitrile, acetone, methanol, dimethyl formamide, and dimethyl sulfoxide [10]. In addition, Est8L was optimally active at pH 10.0, whereas EstCs1 was optimally active at pH 8.0 [10]. In enantioselectivity, Est8L hydrolyzed the S-form with 10.1% higher activity than the R-form, suggesting preferences for the S-form, whereas Est13L showed no selectivity for the enantiomers (Figure 12). In the t-test, it was observed that the enantioselectivity for S-form was significant compared to R-form with a p-value < 0.05. In enantioselectivity, Est8L hydrolyzed the S-form with 10.1% higher activity than the R-form, suggesting preferences for the S-form, whereas Est13L showed no selectivity for the enantiomers (Figure 12). In the t-test, it was observed that the enantioselectivity for S-form was significant compared to R-form with a p-value < 0.05.
In the case of family VIII esterases from similar compost metagenome, Est13L showed similar specific activity EstCs3 from a compost metagenomic library (42.3 vs. 50.2 U/mg). On the other hand, Est13L was stable or activated under organic solvents, whereas EstCs3 was abolished or significantly inactivated by organic solvents such as 30% isopropanol, acetonitrile, ethanol, and methanol [11]. Est13L and EstCs3 showed similar molecular weights; however, their native forms were different (tetramer vs. monomer) [11].
To compare the properties of the Est8L and Est13L with each family, we searched the characterized enzymes at Pubmed (https://pubmed.ncbi.nlm.nih.gov/, accessed on 25 May 2021). The summary of characterization of each enzyme-containing Est8L and Est13L were described (Tables 3 and 4). Family IV and VIII esterases have been practically interested in the usage of ester synthesis and transesterification reactions, and especially hydrolysis of antibiotics by family VIII [10,11].
The native form of Est8L was observed as a dimer, similar to other HSLs such as PestE [34,35] and Est22 [45] ( Table 3). The native form of Est13L was a tetramer, which was unique and unlikely the other family VIII esterases; most family VIII esterases are monomers in their native forms, such as EstC [49], EstBL [51], EstCE1 [52], EstIII [53], and EstCS3 [11]. Est22 was a trimer in its native form [54], and EstA3 required an oligomeric form containing less than six subunits for activation [52] (Table 4).
Est8L and Est13L are alkaline esterases. The optimum pH range of bHSLs is broad, from 5.0 to 9.0. In contrast, the optimum pH range of most family VIII esterases ranges from 7.2 to 10.0, indicating that most family VIII esterases are alkaline esterases. Only a few family VIII esterases have been reported as neutral esterases, the optimum pH of which is 7.0, such as LipBL [55], EstB [56], Lpc53E1 [50], and Lip8 [57] (Tables 3 and 4).
Collectively, Est8L showed a low identity and a higher specific activity than most other family IV esterases reported and different properties in enantioselectivity and sensitivity to the internal proteolytic activity. In contrast, Est13L showed different properties in terms of substrate specificity, specific activity, enantioselectivity, and metal ion inhibition compared to Est7K, although both differed in only six amino acids, similar to the reports that one or several amino acids could cause changes in enzyme properties such as specific activities, substrates specificities, thermostabilities, solvent tolerances, and stereoselectivities in esterases and lipases [67][68][69][70][71].

Conclusions
In this study, two novel esterase genes est8L and est13L encoding intracellular Est8L and Est13L, respectively, were isolated from the compost metagenomic library. Est8L and 13 L were revealed to be novel members of family IV esterase (bHSL) and family VIII esterase, respectively. Est8L had similar characteristics to bHSL, such as substrate specificity for C4 and optimum temperature and pH values of 40 • C and 9.0, respectively. Est8L exhibited enantioselectivity for the S-form (10.16% higher than the R-form). Est13L also had similar characteristics such as molecular weight and optimum temperature and pH. However, Est13L was a tetramer in its native form; unlikely, other family VIII esterases were reported. Est8L was sensitive to isopropanol and acetonitrile, whereas Est13L was activated by 30% isopropanol or methanol. In contrast, Est8L and Est13L were inhibited by Cu 2+ and Zn 2+ with similar patterns, but Est13L was more stable than Est8L. In conclusion, Est8L had higher specific activity than most of other HSLs and enantioselectivity for the S-form; however, it was sensitive to organic solvents; Est13L had higher stability against ions, organic solvents such as alcoholic solvents (isopropanol and methanol), unusual tetrameric form in family VIII esterases, and thermal stress. These results suggested that Est8L and Est13L can be used for chemical reactions with enantioselectivity and for detergent/chemical reactions under alkaline conditions, respectively.