Novel Metagenome-Derived Carboxylesterase That Hydrolyzes β-Lactam Antibiotics^▿,†

Strains, library construction, and screening.

E. coli strains DH5α and BL21(DE3) were used for all cloning and expression studies. A soil sample was collected from the Upo wetland in South Korea. A metagenomic library was constructed in the vector pSuperCosI, as previously described (18).

Subcloning and sequence analysis.

A cosmid clone (pCosU1) showing lipolytic activity on the tributyrin (TBN) agar plate was inoculated into 50 ml of LB broth containing 50 μg/ml of ampicillin. After an overnight incubation at 30°C, the cells were harvested by centrifugation at 5,000 × g for 15 min and washed twice with distilled water. The cosmid DNA was purified using the alkaline lysis method (4) with minor modifications and was randomly sheared by nebulization according to the manufacturer's instructions (Invitrogen, Carlsbad, CA). After nebulization, DNA fragments of 2 to 4 kb were isolated from a 0.6% low-melting-point agarose (FMC Bioproducts, Rockland, ME) gel and end repaired to generate blunt ends. The blunt-ended DNA was ligated into the plasmid pUC118/HincII/BAP purchased from Takara (Kyoto, Japan), and the ligation products were introduced into E. coli DH5α cells (Takara, Kyoto, Japan). The E. coli transformants were plated onto LB agar plates containing 100 μg/ml of ampicillin and 1% tributyrin. After incubation at 37°C for 24 h, a colony surrounded by a clear halo was selected. Nucleotide sequencing was performed with an ABI 3100 automated sequencer using a BigDye Terminator kit (PE Applied Biosystems, Foster City, CA). The DNA sequence was determined by primer walking in both directions and assembled using the ContigExpress program of the Vector NTI Suite, version 7, software package (InforMax, North Bethesda, MD). The open reading frame (ORF) was detected using the ORF search tool provided by the National Center for Biotechnology Information (NCBI). Sequence homology searches were performed with the BLAST program (1). A signal sequence search was performed with the SignalP, version 3.0, program (13). Multiple alignments between protein sequences were performed with the ClustalW program (30). A phylogenetic tree was constructed by the neighbor-joining method (27) using Molecular Evolutionary Genetics Analysis (MEGA; version 4.1, software (29).

Expression and purification of recombinant EstU1.

The estU1 gene was amplified without its signal sequence using pUCU1 as a template and the following primers: (5′-GACCTCCCATATGGAAGGGCCGGTTACG-3′ and 5′-CTCTCTCGAGTCGATCAAACGCTCCATAGACAATATTTC-3′ (NdeI and XhoI restriction enzyme sites are underlined). The estU1 gene was cloned into the expression vector pET-24a(+), and the recombinant plasmid was transformed into E. coli BL21(DE3) cells. As cell density reached a turbidity of about 0.6 at 600 nm, 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) was added to the culture to induce protein expression. After 10 h of cultivation at 25°C, the cells were harvested by centrifugation (5,000 × g for 20 min at 4°C) and resuspended in 50 mM Tris-HCl buffer (pH 8.0) containing 100 mM KCl and 10% glycerol. The cells were disrupted by sonication and centrifuged (15,000 × g for 1 h at 4°C). To purify EstU1 with a His₆ tag, the resulting supernatant was applied to a column of Talon metal affinity resin (BD Biosciences Clontech, Palo Alto, CA) and washed with resuspension buffer containing 10 mM imidazole in 50 mM Tris-HCl buffer (pH 8.0) containing 100 mM KCl and 10% glycerol. EstU1 was eluted with buffer containing 300 mM imidazole, followed by size exclusion chromatography with a Superdex-75 (16/60) column, equilibrated with 20 mM Tris-HCl buffer, pH 7.8, and 150 mM NaCl at a 1 ml/min flow rate. Protein concentration was measured using a Bio-Rad protein assay kit (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard (6). The purity of the protein was examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under denaturing conditions, as described by Laemmli (21).

Esterase assay.

Enzyme activity was measured by a spectrophotometric method using p-nitrophenyl esters (Sigma, St. Louis, MO) as the substrate. After incubation at each optimum temperature for 5 min, the absorbance at 405 nm was measured to detect the released p-nitrophenol. One unit of esterase activity was defined as the amount of enzyme required to release 1 μmol of p-nitrophenol from p-nitrophenyl esters per min.

Biochemical properties of esterase.

Substrate specificity was determined by using p-nitrophenyl esters with different aliphatic side chains: C₂ (acetate), C₄ (butyrate), C₆ (hexanoate), C₈ (octanoate), C₁₀ (decanoate), C₁₂ (laurate), C₁₄ (myristate), C₁₆ (palmitate), and C₁₈ (stearate). The kinetic parameters (k_cat and K_m) of enzymatic conversion were determined by analysis of the dependence of the initial reaction rates on the p-nitrophenyl butyrate concentration (10 to 200 μM). Each protein was used at a concentration of around 34 nM. The molar extinction coefficient measured under the assay conditions was 13,500 M⁻¹ cm⁻¹. The kinetic parameters k_cat and K_m were determined by fitting the data to the Michaelis-Menten equation. The optimum temperature of the enzyme reaction was determined in the same substrate solution described above at various temperatures ranging from 5 to 70°C. The optimum pH was determined over a pH range of 4.0 to 10.0, using the following buffer systems: 50 mM sodium acetate (pH 4.0 to 6.0), 50 mM sodium phosphate (pH 6.0 to 7.5), 50 mM Tris-HCl (pH 7.5 to 8.5), and 50 mM CHES (N-cyclohexyl-2-aminoethanesulfonic acid; pH 8.5 to 10.0). Various metal ions (MnCl₂, MgCl₂, CaCl₂, CuCl₂, ZnSO₄, FeSO₄, CoSO₄, and NiSO₄) and enzyme inhibitors (phenylmethylsulfonyl fluoride [PMSF] and EDTA) at final concentrations of 1 mM were incubated with the enzyme in 50 mM Tris-HCl buffer (pH 7.5) at 35°C for 1 h, and then the enzyme activity was assayed.

β-Lactamase assay.

The β-lactam hydrolytic activity of EstU1 was determined spectrophotometrically using a chromogenic β-lactam substrate such as nitrocefin [3-(2, 4 dinitrostyrl)-(6R,7R-7-(2-thienylacetamido)-ceph-3-em-4-carboxylic acid, E-isomer] as a substrate (Unipath, Basingstoke, United Kingdom). The enzyme was incubated with a 1 mM nitrocefin solution (in 0.1 M phosphate, 1 mM EDTA, pH 7.0) at 35°C, and the rate change at 486 nm was recorded. The molar extinction coefficient of nitrocefin at 486 nm is 20,500 M⁻¹ cm⁻¹. The hydrolyzing activity toward the β-lactam substrates was determined by a paper disc method as described previously with slight modification (11). Antibiotics (ampicillin, penicillin G, cephaloridine, cephalothin, cefazolin, cefuroxime, and cefotaxime) and class C β-lactamase from Enterobacter cloacae were obtained from Sigma-Aldrich (St. Louis, MO). A negative control with antibiotics only and a positive control incubated with the class C β-lactamase and antibiotics were included for comparison. The hydrolytic activity of the enzyme toward nonchromogenic β-lactam antibiotics was measured by incubating 200 μM purified EstU1 with antibiotic substrates, such as 1 mM ampicillin, 4 mM penicillin G, 3 mM cephaloridine, 3 mM cephalothin, 3 mM cefazolin, 1 mM cefuroxime, and 1 mM cefotaxime, in 50 mM Tri-HCl (pH 8.0) for 2 h at 35°C. Then, the resulting reaction mixtures were put onto small paper discs placed on the E. coli BL21(DE3) lawn. To prepare the bacterial lawn of E. coli BL21(DE3) in advance, a suspension of E. coli BL21(DE3) grown in LB medium at 37°C to an optical density at 600 nm (OD₆₀₀) of 0.6 was added to 80 ml of LB agar. The inoculated medium was then poured into a flat-bottomed square dish (inner dimensions, 125 by 125 mm; SPL), and a thin filter paper disc 0.8 cm in diameter (Advantec, Japan) was carefully put onto the gelled plate. The diameters of the inhibition zones around the discs were recorded after an overnight incubation at 37°C.

HPLC analysis.

EstU1 (100 μM or 200 μM) was incubated with 2 mM cephaloridine, 1 mM cephalothin, or 2 mM cefazolin in 50 mM Tris-HCl (pH 8.0) for 1 h at 35°C, and the resulting mixtures were analyzed by high-performance liquid chromatography (HPLC). The standard samples containing cephalosporins (cephaloridine, cephalothin, and cefazolin) were used as a reference, and a positive control incubated with class C β-lactamase and cephalosporins was included. HPLC analysis of the reaction mixtures was conducted on an Atlantic dC₁₈ column (particle size, 5 μm; inner dimensions, 4.6 by 150 mm; Waters, Ireland) using water containing 0.1% trifluoroacetic acid as mobile phase A and methanol containing 0.1% trifluoroacetic acid as mobile phase B. The gradient started at 0% B and increased to 100% B over 30 min at a flow rate of 1 ml/min. The retention times of cephaloridine, cephalothin, and cefazolin were 12.191, 12.331, and 13.488 min, respectively. The retention times of the reaction products of cephaloridine and cefazolin were 3.482 and 11.073 min, respectively.

Determination of kinetic parameters of β-lactamase activity.

All the kinetic measurements of antibiotic substrates were performed at 35°C in 100 mM sodium phosphate (pH 7.0). The initial rates of hydrolysis were determined by following the absorbance variation, using a UV-2401PC spectrophotometer (Shimadzu, Japan). For cefazolin (extinction coefficient at 273 nm [ϵ₂₇₃] = 6,600 M⁻¹ cm⁻¹), the K_m and k_cat values were determined by fitting the data according to the methods of Hanes-Woolf (7). The K_m values of cephaloridine and cephalothin were determined as competitive inhibition constants, K_is, in the presence of p-nitrophenyl butyrate as a reporter substrate.

Site-directed mutagenesis of EstU1.

To understand whether a single catalytic triad is important for both enzyme activities, site-directed changes to alanine were made at S100A, K103A, and Y218A using a Stratagene (La Jolla, CA) Quik Change kit according to the manufacturer's instructions. The primers used to introduce the S100A, K103A, and Y218A mutations were as follows: S100A_forward (5′-CGATCTTCCGCATCTACGCGATGTCGAAGCCAATCACG-3′), S100A_reverse (5′-CGTGATTGGCTTCGACATCGCGTAGATGCGGAAGATCG-3′), K103A_forward (5′-CATCTACTCGATGTCGGCGCCAATCACGACGGTGG-3′), K103A_reverse (5′-CCACCGTCGTGATTGGCGCCGACATCGAGTAGATG-3′), Y218A_forward (5′-CACGACCTGGGATGCCGGCCACAGCACTGAC-3′), and Y218A_reverse (5′-GTCAGTGCTGTGGCCGGCATCCCAGGTCGTG-3′). The positions of the mutated codons are underlined. The catalytic activities of the three variants were tested and compared to the activity of the wild-type enzyme.

Nucleotide sequence accession number.

The nucleotide sequence of EstU1 was deposited in the GenBank database under accession number JF791800.

Screening and sequence analysis of a clone with lipolytic activity.

A metagenomic library from a soil sample from Upo, South Korea, consisting of 6,912 cosmid clones had previously been constructed (18). To screen for an esterase-producing clone, the cosmid library clones were plated on LB agar containing 1% tributyrin (TBN). Sequence analysis of the pUCU1 insert DNA showed the presence of a 1,281-bp ORF (estU1) encoding a polypeptide of 426 amino acids. BlastP analysis of the amino acid sequence of EstU1 indicated that it was similar to a β-lactamase (YP_004154831) from Variovorax paradoxus EPS (58% identity), a hypothetical protein (NP_772348) from Bradyrhizobium japonicum USDA 110 (54% identity), a β-lactamase (YP_531482) of Rhodopseudomonas palustris BisB18 (51% identity), and a β-lactamase (YP_674851) from Mesorhizobium sp. strain BNC1 (48% identity). A multiple sequence alignment of EstU1 and its homologs showed that the S-X-X-K motif is well conserved in class C β-lactamases (20), penicillin-binding proteins (PBPs) (17), and family VIII carboxylesterases (12, 19, 26) (Fig. 1).

Purification and characterization of EstU1.

To investigate its functionality in hydrolyzing esters and β-lactam antibiotics, the estU1 gene was overexpressed in E. coli. A putative signal peptide of 25 amino acids in the EstU1 amino acid sequence was found by the SignalP, version 3.0, program, and the gene was amplified with primer pairs designed to remove the signal peptide. SDS-PAGE analysis of purified EstU1 showed a single band corresponding to approximately 44 kDa, which correlates well with the size of the mature protein (Fig. 2).

Purified EstU1 could hydrolyze a wide range of substrates (C₂ to C₁₀), with the highest activity toward p-nitrophenyl butyrate (C₄), while no enzyme activity was detected toward longer p-nitrophenyl esters (C₁₂ to C₁₈), indicating that this protein is a bona fide esterase (Table 1). The enzyme appeared to have maximum hydrolytic activity at 45°C in the range of 5 to 75°C (Fig. 3A) and was active in the range of pH 7.5 to 9.5, with maximal activity at pH 8.5 (Fig. 3B). To determine the resistance to various chemical agents that might affect its activity, the enzyme was incubated under various conditions with compounds that might inhibit its activity. EstU1 activity was not affected by the presence of Ca²⁺(91%), Co²⁺ (89%), Cu²⁺ (88%), Fe²⁺ (84%), Mg²⁺ (91%), Mn²⁺ (74%), Ni²⁺ (89%), Zn²⁺ (78%), or EDTA (94%), but it lost approximately 70% of its activity in the presence of 1 mM PMSF, which is known to bind specifically to the active-site serine residue in serine proteases and inhibit their activity.

Determination of β-lactamase activity.

Because the amino acid sequence of EstU1 is also similar to the sequences of class C β-lactamases, we examined whether EstU1 has β-lactam hydrolytic activity toward nitrocefin and antibiotics such as ampicillin, penicillin G, and several first-generation cephalosporins (cephaloridine, cephalothin, and cefazolin). EstU1 showed β-lactam hydrolytic activity against nitrocefin, a chromogenic β-lactamase substrate (data not shown). Based on this observation, we tested the β-lactam hydrolytic activity of EstU1 toward various β-lactam antibiotic substrates. As shown in Fig. 4, the diameters of the inhibition zones around the discs containing penicillin G or cephaloridine together with EstU1 were slightly decreased compared to the disc diameter of the negative control, implying that the efficacy of the antibiotic was affected by EstU1 activity. The decrease in the size of the clear zone around the disc was even more obvious when cephalothin or cefazolin was incubated with EstU1 (Fig. 4). No inhibition zone was observed around the disc of cefazolin incubated with EstU1, indicating that cefazolin became completely ineffective against the bacteria. On the other hand, EstU1 addition did not seem to alter the antibiotic efficacy of ampicillin (Fig. 4), cefuroxime (second-generation cephalosporin), or cefotaxime (third-generation cephalosporin) (see Fig. S1 in the supplemental material) because the sizes of the clear zones did not change at all compared to the clear zone of the negative control. EstU1 seems to have β-lactam hydrolytic activity toward first-generation cephalosporins and penicillin G.

To verify the β-lactam-hydrolyzing activity of EstU1 toward first-generation cephalosporins (cephaloridine, cephalothin, and cefazolin), the changes in cephalosporins or reaction products made by EstU1 were analyzed by reverse-phase HLPC. The HPLC spectra of cephalosporins or reaction products made by class C β-lactamase were used as references (Fig. 5). When cephaloridine (or cefazolin) was incubated with EstU1, the peak pattern was consistent with peaks of reaction mixtures resulting from the incubation of class C β-lactamase and cephaloridine (or cefazolin). The peaks corresponding to antibiotics were decreased with the appearance of a product peak (Fig. 5A and C). Hydrolysis of cephaloridine (or cefazolin) by a β-lactamase opens the β-lactam ring and thus generates an acidic functional group in the molecule. Consequently, the reaction product is eluted earlier than the substrate. In the case of cephalothin hydrolysis, the cephalothin peak was significantly decreased by EstU1 as the incubation time elapsed. The peak profile was identical to that made by class C β-lactamase although the reaction product could not be detected in either EstU1 or class C β-lactamase (Fig. 5B). Taken together, we concluded that EstU1 hydrolyzes the cephalosporins in the same way class C β-lactamase does.

Determination of kinetic parameters.

The kinetic parameters of EstU1 toward p-nitrophenyl butyrate and the first-generation cephalosporins (cephaloridine, cephalothin, and cefazolin) were investigated (Table 2). The K_m values for p-nitrophenyl butyrate and cefazolin were determined to be within a similar range; however, the k_cat values toward cefazolin were approximately 5 orders of magnitude lower than those toward p-nitrophenyl butyrate. Because the enzyme activities of EstU1 toward cephaloridine and cephalothin were too low to allow an accurate determination of kinetic parameters, we estimated the K_i values (∼215 and ∼140 μM) for these two compounds by using them as competitive inhibitors of p-nitrophenyl butyrate.

Site-directed mutagenesis.

The inhibition by PMSF and the presence of the conserved S-X-X-K motif implicate the involvement of serine in both esterase and β-lactamase activities. Additionally, EstU1 does not seem to harbor a separate active site for each activity, given the relatively small size of the enzyme. To confirm whether a single nucleophilic serine or catalytic triad can play a crucial role in both activities, serine 100, lysine 103, and tyrosine 218 of EstU1 were each separately replaced with an alanine residue, and the activities of mutant proteins were investigated. The three variants constructed were expressed as soluble and properly folded proteins, as analyzed by SDS-PAGE and column purification, with the same retention times as the wild type (data not shown). Three mutants of EstU1 were inactive toward p-nitrophenyl butyrate and nitrocefin, confirming that the three residues are essential for both activities (data not shown). A multiple sequence alignment revealed that these three residues are highly conserved in family VIII carboxylesterases, class C β-lactamases, and penicillin-binding proteins (Fig. 1).