Decreasing the Level of Ethyl Acetate in Ethanolic Fermentation Broths of Escherichia coli KO11 by Expression of Pseudomonas putida estZ Esterase

Abstract

During the fermentation of sugars to ethanol relatively high levels of an undesirable coproduct, ethyl acetate, are also produced. With ethanologenic Escherichia coli strain KO11 as the biocatalyst, the level of ethyl acetate in beer containing 4.8% ethanol was 192 mg liter⁻¹. Although the E. coli genome encodes several proteins with esterase activity, neither wild-type strains nor KO11 contained significant ethyl acetate esterase activity. A simple method was developed to rapidly screen bacterial colonies for the presence of esterases which hydrolyze ethyl acetate based on pH change. This method allowed identification of Pseudomonas putida NRRL B-18435 as a source of this activity and the cloning of a new esterase gene, estZ. Recombinant EstZ esterase was purified to near homogeneity and characterized. It belongs to family IV of lipolytic enzymes and contains the conserved catalytic triad of serine, aspartic acid, and histidine. As expected, this serine esterase was inhibited by phenylmethylsulfonyl fluoride and the histidine reagent diethylpyrocarbonate. The native and subunit molecular weights of the recombinant protein were 36,000, indicating that the enzyme exists as a monomer. By using α-naphthyl acetate as a model substrate, optimal activity was observed at pH 7.5 and 40°C. The K_m and V_max for α-naphthyl acetate were 18 μM and 48.1 μmol · min⁻¹ · mg of protein⁻¹, respectively. Among the aliphatic esters tested, the highest activity was obtained with propyl acetate (96 μmol · min⁻¹ · mg of protein⁻¹), followed by ethyl acetate (66 μmol · min⁻¹ · mg of protein⁻¹). Expression of estZ in E. coli KO11 reduced the concentration of ethyl acetate in fermentation broth (4.8% ethanol) to less than 20 mg liter⁻¹.

In previous studies in our laboratory Ingram et al. replaced the native fermentation pathway in Escherichia coli and other enteric bacteria with the homo-ethanol pathway from Zymomonas mobilis (18). One of the altered organisms, E. coli KO11, has been investigated for commercial use and was shown to ferment the diverse array of sugars present in the polymers of agricultural residues. By using abundant agricultural residues as substrates together with yeast-based fermentation of grain, it may be possible to substantially reduce our dependence on imported petroleum as an automotive fuel (1).

Yeast-based ethanol fermentations result in minor products which copurify with ethanol (5, 6, 11, 26, 30, 38, 41). While many of these products are desirable as organoleptic agents and congeners in beverage alcohols, removal of the contaminating compounds to produce pure ethanol requires additional investment. Ethyl acetate is the most abundant ester produced by yeasts and is particularly difficult to separate from ethanol by distillation (12). This compound has also been found to be a minor product in mixed acid fermentations of many enteric bacteria (Klebsiella aerogenes, Enterobacter aerogenes, Citrobacter freundii, Enterobacter cloacae, and Hafnia alvei), but it was reported to be absent in E. coli fermentations (28). However, a preliminary investigation of distillates obtained from ethanologenic strain KO11 revealed a surprisingly high level of ethyl acetate, in excess of 2 g liter of ethanol⁻¹ (Greg Luli, B.C. International, personal communication). The necessity of postfermentation removal of this contaminant could add to the cost of producing pure ethanol with recombinant E. coli.

Ethyl acetate and other esters are produced by alcohol acetyltransferases and are hydrolyzed by esterases (11, 43). The concerted action of these two classes of enzymes determines the levels of ethyl acetate in fermentation broths. Synthesis is thought to result from the transfer of the acetyl moiety from acetyl coenzyme A to ethanol. Potentially, this reaction could be catalyzed by many different acetyltransferases as ethanol accumulates. In yeasts, ethyl acetate production has been ascribed to three acetyltransferases (26), although more than 10 additional candidates are also present on the annotatedgenome (http://genome-www.stanford.edu/Saccharomyces/).The E. coli genome contains at least 13 genes encoding acetyltransferase- or esterase-like proteins with various substrate specificities (4). In ethanologenic strain KO11, production of ethyl acetate during fermentation could result from high ethanol concentrations and a lack of strict substrate specificity. Although it should be possible to reduce ethyl acetate concentrations by eliminating enzymes responsible for ethyl acetate synthesis, these enzymes may also have essential cellular functions. A more efficient, if not more prudent, approach would be to increase the level of esterase with appropriate substrate specificity.

In this paper, we describe a simple method for direct identification of organisms and clones with recombinant DNA that hydrolyze volatile esters by using ethyl acetate as the substrate. This method was used to clone a gene encoding a short-chain aliphatic ester esterase (estZ) from Pseudomonas putida strain NRRL B-18435. The encoded protein was purified and characterized. Functional expression of estZ in E. coli KO11 substantially reduced the level of ethyl acetate in fermentation broth.

MATERIALS AND METHODS

Bacterial cultures.

Various derivatives of E. coli K-12, E. coli B, and other bacteria used in this study are listed in Table 1. Cultures were grown in L broth with appropriate supplements (24). For aerobic growth of nonethanologenic cultures, L broth was used without added sugar. For anaerobic growth, cultures of nonethanologenic strains were supplemented with 0.3% glucose. Ethanologenic strain KO11 (18, 34) was maintained on L agar with xylose (2%). Antibiotics were included in the media at the following concentrations: ampicillin, 100 μg ml⁻¹; tetracycline, 20 μg ml⁻¹; and chloramphenicol, 40 or 600 μg ml⁻¹ for KO11 and its derivatives.

TABLE 1.

Bacterial strains and plasmids used in this study

Organism or plasmid	Relevant genotype	Source or reference
Escherichia coli K-12 strains
AH222	SE2138 pcnB80 zad-2084::Tn10	This study
BL21(DE3)	ompT gal dcm lon hsdSB λDE3	Laboratory collection
DH5α	Δ(lacZYA-argF)U169 endA1 recA1 hsdR17 deoR thi-1 phoA supE44 gyrA96 relA1 φ80dlacZΔM15	Laboratory collection
ER1821	e14− (McrA−) endA1 supE44 thi-1 relA1? rfbD1? spoT1? Δ(mcrC-mrr)114::IS10	New England Biolabs
MRi93	pcnB80 zad-2084::Tn10	CGSC 7066
SE2138	Prototroph	Laboratory collection
Escherichia coli B	Prototroph	ATCC 11303
Escherichia coli KO11	B − Δfrd pfl::pdcZmadhBZmcat	Laboratory collection
Erwinia chrysanthemi P1	Prototroph	J. Preston
Klebsiella oxytoca M5A1	Prototroph	Laboratory collection
Pseudomonas aeruginosa PAO1	Prototroph	R. Jensen
Pseudomonas putida NRRL B-18435	Prototroph	J. Wolfram
Salmonella enterica serovar Typhimurium LT2	Prototroph	Laboratory collection
Enterococcus sp.	Prototroph	Laboratory collection
Plasmids
pAH181	pUC18 − P. putida ′pvdD estZ fpvA ′pvdE, Apr	This study
pAH185	pUC19 − P. putida ′pvdD estZ ′fpvA, Apr	This study
pAH188	pBR322 − P. putida estZ, Apr	This study
pAH191	pAH188 − estZ (− promoter), Apr	This study
pAH199	pAH191 − PZm (236 bp) − estZ, Apr	This study
pAH201	pAH191 − PZm (1.2 kbp) − estZ, Apr	This study
pAH208	pAH191 − PZm (147 bp) − estZ, Apr	This study
pAH213	pAH191 − PZm (785 bp) − estZ, Apr	This study
pAH219	pET15b − estZ, Apr	This study

Strain AH222, a pcnB derivative of wild-type strain SE2138, was constructed by transducing the pcnB mutation along with zad-2084::Tn10 from strain MRi93 with phage P1. Tetracycline-resistant transductants were selected, and the presence of the pcnB mutation was confirmed by the copy number of plasmid pBR322.

Fermentation of xylose by KO11.

Fermentations were carried out in L broth containing 10% xylose as previously described by using 500-ml vessels (29). The cultures were started with an initial cell concentration of 0.33 μg (dry weight) of cells ml⁻¹ and were incubated for 48 h. Temperature (35°C), pH (pH 6.5), and agitation (100 rpm) were controlled. Samples were removed at 12-h intervals to measure cell mass, ethanol, and ethyl acetate.

Whole-cell esterase assay (methyl red assay).

Esterase activity was determined in whole cells by using methyl red as a pH indicator of the acetate produced by hydrolysis of ethyl acetate. Whatman no. 1 filter paper disks (diameter, 12.5 cm) were soaked in a methyl red solution (1 mg ml⁻¹ in 95% ethanol) and allowed to dry. Colonies grown on L agar without added carbohydrate (to prevent acidification) were transferred to the methyl red paper by replica plating and incubated in a desiccator under ethyl acetate vapor. Positive colonies turned red due to the decrease in pH; negative clones appeared yellow on an amber background. With minor modifications, this assay can be used to rapidly screen for esterases which function under different conditions or with different substrates.

Construction of genomic DNA library.

Standard methods were used for construction of a genomic library of P. putida strain NRRL B-18435 DNA and other DNA manipulations (27). After partial hydrolysis with endonuclease Sau3AI, 4.0- to 6.0-kbp fragments of P. putida genomic DNA were purified by agarose gel electrophoresis and ligated into a dephosphorylated BamHI site of plasmid vector pUC18. The ligation mixture was transformed into E. coli strain DH5α. Plasmid DNA was isolated from the pooled ampicillin-resistant transformants and used as a P. putida gene library.

Construction of various esterase plasmids.

Additional plasmids were constructed by starting with pAH181 to allow for insertion of alternative promoters. The esterase gene (estZ) with flanking DNA was removed as a 2.4-kbp SalI fragment, cloned into vector plasmid pUC19 (plasmid pAH185), and sequenced (GenBank accession no. AY082397). In plasmid pAH185, the estZ gene is in the same orientation as the lac promoter. The estZ gene was removed from pAH185 as a 2.5-kbp HindIII-EcoRI fragment (the EcoRI site was filled in by Klenow polymerase) and cloned into the HindIII and AvaI sites (the AvaI site was blunt ended by using Klenow polymerase) of pBR322. The resulting plasmid, pAH188, was positive for ethyl acetate hydrolysis as determined by the methyl red assay. A promoter-probe version of pAH188, pAH191, was constructed, in which the region upstream from estZ was removed. For this construction, the 5′ 509-bp fragment encoding the N-terminal region of estZ was amplified by PCR by using pAH188 as the template, forward primer 5′-AAAAGTCGACGGATCCTAAGGAGTGTGACTTAATGTCCCTGAACCCTGACCTGGCGGCCTA-3′, andreverse primer 5′-CCAGGCTACCACCGACACTGT-3′. The 5′ end of the forward primer included a restriction site for SalI, translational stop codons in all three reading frames, a restriction site for BamHI for promoter DNA insertion, and a Shine-Dalgarno sequence for ribosomal binding. The PCR product was digested with SalI and DraIII, and the resulting 342-bp DNA fragment was used to replace the larger SalI and DraIII fragment in pAH188. Although E. coli strain SE2138(pAH191) was positive for ethyl acetate hydrolase activity, strain AH222(pAH191) carrying a plasmid copy number mutation (pcnB) was negative for this hydrolase activity.

Strain AH222(pAH191) was used to clone alternative promoters for estZ expression. In these constructs, promoter elements from Z. mobilis were used to minimize recombination between plasmid DNA and E. coli chromosomal DNA. Endonuclease Sau3AI fragments (0.5 to 2.0 kbp) of Z. mobilis genomic DNA were used as sources of surrogate promoters. After ligation into the dephosphorylated BamHI site of plasmid pAH191 and transformation of E. coli ER1821 (a restriction- and modification-deficient mutant), transformants were pooled for extraction of plasmid DNA. The resulting plasmid mixture was transformed into E. coli strain AH222 (pcnB), and ampicillin-resistant derivatives were screened for ethyl acetate hydrolase activity by using the methyl red assay. The following four clones with the highest activity (rate of color development) were selected for further study: pAH199, pAH201, pAH208, and pAH213.

Purification of EstZ.

An estZ expression plasmid was constructed to facilitate enzyme purification. The estZ coding region was amplified from plasmid pAH181 by PCR performed with forward primer 5′-AAAAAATAAACATATGTCCCTGAACCCTGACCTGGCGG-3′ and reverse primer 5′-CTAGTTATTGCTCAGCGCTTCTGATCGCCTGACGTTGA-3′. The forward primer included an NdeI site at the predicted ATG start codon. The reverse primer included a BlpI site downstream of the translation stop codon (TAA). The PCR-generated fragment was hydrolyzed with NdeI and BlpI, ligated into the corresponding sites in plasmid pET15b (Novagen), and transformed into E. coli strain BL21(DE3) to produce pAH219. With this plasmid, addition of isopropyl-β-d-thiogalactopyranoside (IPTG) induced the production of EstZ as a fusion protein containing an N-terminal histidine tag.

Fresh transformants of E. coli strain BL21(DE3) containing pAH219 were grown in 1 liter of L broth in a Fernbach flask with shaking (200 rpm) at 37°C for esterase isolation. When the optical density at 420 nm of the culture reached 0.7 (Spectronic 710; Bausch & Lomb), IPTG was added to a final concentration of 0.1 mM, and the culture was shifted to 23°C. After 4 h of incubation, cells were harvested, washed once with 50 ml of Tris buffer (50 mM Tris-HCl, pH 8.0), resuspended in 25 ml of Tris buffer, and broken by two passages through a French pressure cell at 20,000 lb/in². The cell lysate was centrifuged at 30,000 × g for 30 min at 4°C, and the supernatant was further clarified by centrifugation at 150,000 × g for 60 min at 4°C. The esterase-containing supernatant was loaded on a 5-ml HiTrap affinity chelating column (Pharmacia Biotech) precharged with NiCl₂. The column was washed with 50 ml of Tris buffer to remove unbound proteins. Proteins were eluted from the column by using a step gradient of imidazole in Tris buffer (pH 8.0). An imidazole concentration of 20 mM was used to remove nonspecifically bound proteins. EstZ was eluted with 70 mM imidazole in Tris buffer. Fractions containing EstZ were identified on the basis of apparent molecular weight and abundance by using sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE), pooled, and concentrated to 2.5 ml (Centriprep-10; Amicon). This preparation was loaded on a Sephacryl S-200 (Hi-prep 26/60) gel filtration column (Pharmacia Biotech) that had been equilibrated with 50 mM Tris buffer (pH 8.0) containing 150 mM NaCl. Proteins were eluted with the same buffer at a flow rate of 0.5 ml min⁻¹. Fractions containing EstZ were pooled and digested with thrombin (20 U; Pharmacia Biotech) at 16°C for 6 h in the presence of 1 mM CaCl₂ to remove the N-terminal His tag. After dialysis overnight against 50 mM Tris buffer at 4°C, EstZ was further purified by using a Q-Sepharose anion-exchange column (15 ml; Pharmacia Biotech) that had been equilibrated with Tris buffer. The esterase was eluted with 0.2 M NaCl, dialyzed overnight in 50 mM Tris buffer (pH 8.0), and stored on ice.

Molecular weight determination.

Native molecular weight was determined by gel filtration by using a Sephacryl S-200 gel filtration column (Pharmacia Biotech) in 50 mM Tris buffer (pH 8.0) with 0.15 M NaCl. The molecular weight standards (Sigma Chemical Co.) used were horse heart cytochrome c (molecular weight, 12,400), bovine erythrocyte carbonic anhydrase (29,000), bovine serum albumin (66,000), yeast alcohol dehydrogenase (150,000), and sweet potato β-amylase (200,000). The subunit molecular weight was determined by SDS-PAGE by using carbonic anhydrase (29,000), ovalbumin (45,000), and bovine serum albumin (66,000) as the standards.

Enzyme assays.

In all enzyme assays, 1 U of esterase activity was defined as the amount of enzyme required to hydrolyze 1 μmol of substrate per min. For routine assays, esterase activity was determined in 50 mM sodium phosphate buffer (pH 7.5) by using α-naphthyl acetate as the substrate. An increase in absorbance at 321 nm due to the release of α-naphthol was determined after a 5-min incubation at 37°C. An ɛ₃₂₁ of 2,740 M⁻¹ cm⁻¹ was used to calculate the specific activity. In kinetic experiments, the release of α-naphthol was monitored continuously at 37°C by using a water-jacketed cuvette and a Beckman model DU640 spectrophotometer. The cuvette temperature was maintained by using a circulating water bath. Reactions were initiated by adding substrate. Activity was calculated by using the initial rate of product formation. The effects of temperature and pH were also examined by using α-naphthyl acetate as the substrate. For temperature studies, reactants were preequilibrated for 10 min prior to the addition of substrate. For pH studies, Teorell and Stenhagen's citrate-phosphate-borate buffer (50 mM) was used instead of phosphate buffer to allow examination of a wide pH range (3).

Hydrolysis of ρ-nitrophenyl acetate was measured in 50 mM phosphate buffer (pH 7.5) by determining the increase in absorbance at 400 nm at 37°C due to release of ρ-nitrophenol (21). An ɛ₄₀₀ of 16,600 M⁻¹ cm⁻¹ was used to calculate the enzyme activity.

Esterase activity was also measured by using alcohol esters of acetate as substrates. In these assays, acetate was determined as NADH by using a coupled reaction (acetic acid kit; Boehringer Mannheim, Indianapolis, Ind.). The NADH concentration was determined at 340 nm. The 1-ml reaction mixture (50 mM sodium phosphate buffer [pH 7.5], enzyme, 1 mM substrate) was placed in a 2-ml glass vial and sealed with a serum stopper since many of the substrates are volatile. After 2 to 4 min of incubation at 37°C, the reactions were terminated by incubation at 85°C for 5 min. The acetate present in 0.1 ml of sample was determined by using the acetic acid kit. Under these conditions, the reaction was linear for up to 10 min.

Analytical methods.

Ethyl acetate was measured in fermentation broth by gas chromatography (model 3600 chromatograph equipped with a flame ionization detector; Varian, Palo Alto, Calif.) by using a Restek capillary column (30 m by 0.53 mm; Rtx-Volatiles, Bellefonte, Pa.). After injection, the column temperature was held at 50°C for 4 min, raised to 180°C at a rate of 25°C/min, and held at 180°C for 8 min. Dinitrogen was the carrier gas. Under these conditions, the lower limit for measurement of ethyl acetate in broth was approximately 20 mg liter⁻¹. Ethanol was measured as previously described (29).

Materials.

Biochemicals were purchased from Sigma Chemical Co. Other organic and inorganic chemicals were obtained from Fisher Scientific and were analytical grade. Restriction endonucleases and DNA-modifying enzymes were purchased from Promega and New England Biolabs. Preliminary sequence data for P. putida KT2440 were obtained from The Institute for Genomic Research website (http://www.tigr.org).

RESULTS AND DISCUSSION

The problem: ethyl acetate production by ethanologenic strain KO11.

In a typical fermentation containing 10% xylose, ethanologenic E. coli strain KO11 produced about 4.8% ethanol and 192 mg of ethyl acetate liter⁻¹ (Table 2). This level of ethyl acetate is 4- to 10-fold higher than the level typically observed with yeasts (5, 6, 11, 26, 30, 38, 41). In yeasts and presumably other organisms, the level of ethyl acetate is modulated by the activities of alcohol acetyltransferases and esterases (11). This compound has no known physiological function in yeasts. The higher level of this compound in fermentations with ethanologenic E. coli KO11 is presumed to result from either higher acetyltransferase activities which synthesize ethyl acetate or a lack of esterases which hydrolyze ethyl acetate. When tests were performed by using the methyl red assay (whole cells), E. coli B, KO11, and K-12 wild-type strains lacked the ability to hydrolyze ethyl acetate and failed to utilize ethyl acetate as a sole carbon source for growth. A variety of other bacteria (Klebsiella oxytoca strain M5A1, Salmonella enterica serovar Typhimurium strain LT2, Erwinia chrysanthemi strain P1, Pseudomonas aeruginosa strain PAO1, and Enterococcus spp.) were tested by using the methyl red assay and were found to lack the ability to hydrolyze ethyl acetate. Of the organisms tested, only P. putida NRRL B-18435 produced high levels of this activity.

TABLE 2.

Effect of EstZ on the fermentation of xylose (10%) by E. coli KO11^a

Organism	Cell mass (g liter−1)	Ethanol concn (g liter−1)	Ethyl acetate concn (mg liter−1)	Esterase activity (U mg of protein−1)
KO11	3.43	48	192	0.04
KO11(pAH181)	3.63	48	<20	0.16
KO11(pAH191)	3.52	49	77	0.10
KO11(pAH199)	3.83	51	51	NDb
KO11(pAH201)	3.33	51	51	ND
KO11(pAH208)	3.63	49	<20	0.20
KO11(pAH213)	3.52	48	48	ND

^aResults for completed fermentations (48 h). Esterase activities were measured in cells harvested in the stationary phase (24 h) by using α-naphthyl acetate as the substrate.

^bND, not determined.

Cloning the gene responsible for ethyl acetate hydrolysis.

E. coli strain SE2138 was transformed with a gene library from P. putida NRRL B-18435. Approximately 10,000 transformants were screened for ethyl acetate hydrolase activity by using the methyl red assay. Clones which turned bright red within 5 min of exposure to ethyl acetate vapor were selected from a replica plate and retested. Of the 20 positive clones, 1 clone expressing the highest activity (rate of color development) was selected for further study, and the plasmid was designated pAH181.

Plasmid pAH181 (Fig. 1) contained a 6.7-kbp insert with four open reading frames (ORFs). These ORFs were identified by comparison to other known genes by using Blast (www.ncbi.nlm.nih.gov). One of the ORFs contained the C-terminal 594 amino acids of a siderophore synthase (nonribosomal peptide synthesis) resembling PvdD in P. aeruginosa PAO1 (32) and P. putida WCS358 (GenBank accession no. CAC32046), with 61 and 62% identity, respectively. This C-terminal segment contained the thioesterase motif found in peptide synthetases associated with nonribosomal synthesis of peptide antibiotics (15, 25). The DNA region containing the pvdD-like gene was also localized on the incomplete, unannotated genome sequence of P. putida strain KT2440 (www.tigr.org). From this database, the full-length gene in P. putida NRRL B-18435 appears to encode a 3,089-amino-acid protein.

FIG. 1.

Comparison of the genetic organizations of the pvdD estZ fpvA pvdE gene cluster in P. putida NRRL B-18435 (chromosomal fragment in pAH181) and KT2440 (unannotated partial genome from The Institute for Genomic Research) to the genetic organization of the corresponding region in the chromosome of P. aeruginosa PAO1 (39).

The ORF immediately following pvdD encodes an esterase gene and was designated estZ. The translated estZ protein (318 amino acids) is similar to family IV esterases/lipases and to a hormone-sensitive lipase (Fig. 2) (2, 7, 23). The levels of amino acid sequence identity between EstZ and the other esterases/lipases listed in Fig. 2 varied between 30 and 41% (7, 8, 33, 42). The consensus motif for esterases, LAVVGDSVGG (10, 17), is located between positions 159 and 168 of the translated EstZ sequence. The amino acid serine at position 165 corresponds to the active site serine. EstZ also contains aspartate and histidine in the conserved region, completing a proposed catalytic triad (16, 35). Consistent with the lack of a signal sequence for translocation, EstZ activity was found in the cytoplasm of both P. putida NRRL B-18435 and recombinant E. coli (data not shown).

FIG. 2.

Comparison of the translated sequence of P. putida NRRL B-18435 EstZ with sequences of other family IV lipases/esterases (7, 8, 33, 42; www.tigr.org). The sequence of an esterase from an unidentified, petroleum-degrading bacterium, strain HD-1 (33), is also included. Amino acids that are conserved in all six protein sequences are indicated by asterisks below the sequences. Amino acids which are similar are indicated by colons, and functionally compatible amino acids are indicated by dots. Amino acids in the proposed catalytic triad are highlighted.

The estZ gene is located immediately downstream from the pvdD gene on the unannotated genome of P. putida KT2440 also (www.tigr.org). Both genes appear to be in the same operon, part of a pyoverdine synthesis gene cluster in P. putida. The predicted amino acid sequence of EstZ from strain KT2440 was 98% identical to the sequence of the cloned estZ gene product from strain NRRL-B18435. The five amino acids which differ reside in the nonconserved regions of family IV esterases.

The other two ORFs (pAH181) appear to be transcribed in the opposite direction from pvdD and estZ (Fig. 1). The third ORF (fpvA) was identified as a ferric pyoverdine receptor protein (39% identity with FpvA from P. aeruginosa PAO1) (32). Based on protein sequence similarity, the fourth ORF (pvdE) appears to encode the C-terminal portion of an ATP-dependent ABC membrane transporter associated with pyoverdine biosynthesis (66% identity with PvdE in P. aeruginosa PAO1) (31).

The ethyl acetate esterase activity encoded by plasmid pAH181 could be provided by either incomplete PvdD with the thioesterase motif or EstZ (Fig. 1). It is interesting that the genome of a related organism, P. aeruginosa PAO1, lacks the estZ gene between the homologous pvdD and fpvA genes (39). P. aeruginosa PAO1 failed to hydrolyze ethyl acetate, which is consistent with the hypothesis that estZ is the source of ethyl acetate esterase activity. To confirm this, a DNA fragment containing only the estZ gene was cloned into plasmid vector pBR322 to produce plasmid pAH188. Recombinant E. coli strains harboring pAH188 were ethyl acetate hydrolase positive. Although EstZ hydrolyzed ethyl acetate, the physiological role of this esterase in P. putida is apparently in pyoverdine biosynthesis. The specific role of EstZ in this process and its in vivo substrate have not been identified.

In E. coli, the aes gene encodes an enzyme related to family IV esterases (19, 36) which exhibits 23% identity with the translated EstZ from P. putida NRRL B-18435. The Aes enzyme has been shown to hydrolyze valerate esters preferentially over acetate as the acid moiety (19). The failure of native E. coli to hydrolyze ethyl acetate could be due to inadequate levels of Aes. In gene array studies, the mRNA levels for aes were found to be low (40).

Biochemical properties of EstZ.

Recombinant EstZ was purified 35-fold almost to homogeneity based on SDS-PAGE (data not shown). Both the native molecular weight (as determined by gel filtration) and the subunit molecular weight (as determined by SDS-PAGE) were approximately 36,000, indicating that EstZ functions as a monomer. The specific activities of purified EstZ were 42 and 66 U mg of protein⁻¹ for α-naphthyl acetate and ethyl acetate, respectively. Kinetic properties of the EstZ were determined by using two chromogenic substrates, α-naphthyl acetate and ρ-nitrophenyl acetate; the highly volatile nature of ethyl acetate hindered kinetic studies with this substrate (Table 3). Both chromogenic substrates were hydrolyzed with Michalean kinetics. The affinity for ρ-nitrophenyl acetate was threefold lower than that for α-naphthyl acetate. However, the V_max with ρ-nitrophenyl acetate was approximately twice that with α-nitrophenyl acetate. Thus, the higher turnover number obtained with ρ-nitrophenyl acetate was compensated for by the lower affinity for this substrate.

TABLE 3.

Kinetic properties of the P. putida esterase

Substrate	Km (μM)	Vmax (U mg of protein−1)	kcat (s−1)	kcat/Km (s−1 μM−1)
α-Naphthyl acetate	18	48	29	1.6
ρ-Nitrophenyl acetate	54	106	64	1.2

The effects of pH and temperature on EstZ activity were determined by using α-naphthyl acetate as the substrate. This enzyme was active between pH 6.0 and 10.5, with maximum activity at pH 7.5 (Fig. 3). Increasing the assay pH above this value led to a linear decrease in specific activity. EstZ retained 40% of the maximal activity at pH 10.0 but was inactive at pH 11.0. In this regard, EstZ appears to be similar to the lipase from Pseudomonas sp. strain B11-1 (7) and to an esterase from Streptomyces diastatochromogenes, (20) but differs from a Pseudomonas fluorescens esterase which retained 80% of the maximal (pH 7.5) activity at pH 10.0 (22). At pH 5.5, EstZ was only minimally active (about 10% of the maximal activity). The increase in activity between pH 5.5 and 7.5 with half-maximal activity near pH 6.5 suggests that protonation of a histidine residue is essential for optimal activity (histidine pK_a, 6.3). A histidine at position 291 is conserved in all family IV esterases and has been implicated in the catalytic triad of the hormone-sensitive lipase (2), an ortholog of the P. putida EstZ esterase.

FIG. 3.

Effect of pH on EstZ activity. α-Naphthyl acetate was used as the substrate.

EstZ esterase was optimally active at temperatures between 35 and 45°C, with maximal activity near 40°C (Fig. 4). The energy of activation was calculated to be 17.4 kcal mol⁻¹. This low activation energy is similar to the value of 11.2 kcal mol⁻¹ obtained with the lipase from Pseudomonas sp. strain B11-1 (7).

FIG. 4.

Effect of temperature on EstZ activity. α-Naphthyl acetate was used as the substrate. (A) Temperature plot; (B) Arrhenius plot.

Substrate specificity of EstZ.

The activity of EstZ decreased with increasing chain length of the acid moiety (Table 4). It is interesting that this esterase failed to hydrolyze α-naphthyl butyrate, although α-naphthyl caproate was hydrolyzed. Esters with chain lengths of eight carbons or more were not hydrolyzed. A similar decline in activity with increasing chain length of the aliphatic acid was observed for the lipase from Pseudomonas sp. strain B11-1 (7) and for an esterase from P. fluorescens (22). Although α-naphthyl acetate was hydrolyzed by EstZ with a maximum specific activity of 42 U mg of protein⁻¹, β-naphthyl acetate was a poor substrate (about 20% of the activity with α-naphthyl acetate). Other aromatic substrates, such as ρ-nitrophenyl acetate, were also hydrolyzed, but at a lower rate, by EstZ (Table 4).

TABLE 4.

Effect of chain length of the acid moiety on EstZ activity

Substrate	Esterase activity (U mg of protein−1)
α-Naphthyl acetate	42
α-Naphthyl propionate	33
α-Naphthyl butyrate	<1
α-Naphthyl caproate	15
α-Naphthyl caprylate	<1
α-Naphthyl laurate	<1
β-Naphthyl acetate	9
ρ-Nitrophenyl acetate	11

A series of aliphatic esters of acetate were also tested as substrates for EstZ (Fig. 5). Increasing the chain length of the alcohol moiety from one carbon to three carbons increased the hydrolysis rate of the ester (Fig. 5). The highest specific activity was observed with propyl acetate as the substrate (96 U mg of protein⁻¹). Increasing the chain length to more than three carbons led to a rapid decline in activity. With decyl acetate, the EstZ activity was less than 5% of that with propyl acetate. These results show that the preferred substrate for this enzyme is propyl acetate. Although EstZ hydrolyzed aromatic esters like naphthyl acetate and nitrophenyl acetate, higher activities were obtained with water-soluble aliphatic short-chain esters. This enzyme was sensitive to dimethyl sulfoxide (DMSO), which is consistent with a preference for aliphatic esters (Fig. 6). In the presence of 30% DMSO, the enzyme activity was less than 10% of the aqueous value. In this regard, EstZ is different from the related lipase from Pseudomonas sp. strain B11-1, whose activity was enhanced by DMSO (7).

FIG. 5.

Effect of aliphatic alcohol chain length on EstZ activity, determined by using acetate esters as substrates.

FIG. 6.

Effect of DMSO on EstZ activity. 4-Methylumbelliferyl acetate was used as the substrate. The specific activity without DMSO was 12.6 U mg of protein⁻¹.

Effect of inhibitors on EstZ activity.

All esterases and lipases, including the hormone-sensitive lipase, contain the α/β hydrolase fold and a catalytic triad with serine (nucleophile), a catalytic acid residue (asparate or glutamate), and histidine (35). EstZ from P. putida is also expected to conform to this structure based on the conservation of the amino acid sequence (Fig. 2). By using the crystal structure of a thermophilic esterase from Alicyclobacillus acidocaldarius as the model (9), a tertiary structure for EstZ was computed with SwissModel (13, 14, 37). The predicted structure exhibited an α/β hydrolase fold and close proximity of the conserved amino acids in the catalytic triad for EstZ. Requirements for serine and histidine for catalytic activity were confirmed by using specific inhibitors (Table 5). The serine protease inhibitor phenylmethylsulfonyl fluoride completely inhibited the EstZ activity, while a similar concentration of eserine resulted in 75% inhibition. Differences in the structure of the aromatic rings between the two inhibitors may influence the level of inhibition. Inhibition by diethylpyrocarbonate provided further evidence that a histidine residue is important for activity. EstZ has four cysteines, one of which (cysteine 107) is conserved in the six family IV esterases listed in Fig. 2. Two sulfhydryl reagents, mercuric chloride and naphthol AS-D chloroacetate, inhibited EstZ activity, suggesting that one or more of the cysteines have an important role. A lack of inhibition by EDTA indicates that a divalent cation is not essential for activity. These inhibitor studies and predicted models suggest that the general structure and reaction mechanism of EstZ from P. putida are similar to the general structure and reaction mechanism of other esterases.

TABLE 5.

Effects of inhibitors on EstZ activity

Compound	Esterase activity (U mg of protein−1)
None	40
Phenylmethylsulfonyl fluoride	<1
Eserine	11
Mercuric chloride	2
Naphthol AS-D chloroacetate	<1
Diethylpyrocarbonate	<1
EDTA	31

Utility of estZ: effect of E. coli KO11 on the level of ethyl acetate in ethanolic beers.

The EstZ esterase from P. putida exhibited a preference for short-chain aliphatic esters and may be useful for reduction of ethyl acetate in fermentation broths. To examine this possibility, KO11 was transformed with pAH181 containing the 6.7-kb fragment of DNA from P. putida NRRL B-18435 which contains estZ (Fig. 1). Expression of estZ in KO11 was confirmed by using α-naphthyl acetate as a substrate (Table 2). Native esterases in KO11 also hydrolyzed α-naphthyl acetate but had negligible activity with ethyl acetate. Based on studies performed with highly purified recombinant enzyme, the specific activity of EstZ with α-naphthyl acetate was found to be approximately 0.66 that with ethyl acetate (Table 4 and Fig. 5). Based on this ratio and the difference in esterase activity (α-naphthyl acetate hydrolysis) between KO11 and KO11(pAH181), the hydrolytic activity for ethyl acetate in KO11(pAH181) was estimated to be 0.18 U mg of cell protein⁻¹.

Fermentations with 10% xylose were completed within 48 h, producing more than 95% of the maximum theoretical yield for ethanol (0.51 g of ethanol per g of xylose) (Table 2). Although KO11(pAH181) and unmodified KO11 produced similar cell densities (3.63 and 3.43 g liter⁻¹, respectively) and similar levels of ethanol (4.8%), the concentration of ethyl acetate in the KO11(pAH181) beer (<20 mg liter⁻¹) was at least ninefold lower than that in the beer produced with unmodified KO11 (Table 2). These results clearly establish the utility of recombinant EstZ as an effective way to reduce ethyl acetate concentrations in ethanolic beers produced with recombinant E. coli.

Additional plasmid constructs were made with estZ to introduce promoter elements since in plasmid pAH181 estZ is apparently transcribed from the lacZ promoter in the vector. These constructs included a regulated T7-based expression vector (pAH219) which can be induced with IPTG to produce high levels of EstZ for enzyme purification or as a reagent for the treatment of beers, as well as pBR322-based vectors for expression in ethanologenic E. coli. The results obtained with five pBR322-based constructs are shown in Table 2. Plasmid pAH191 contains only the estZ coding region (with a Shine-Dalgarno sequence) preceded by a unique BamHI site. Expression of estZ in this plasmid is presumed to be from the upstream tet promoter or promoter-like DNA sequence in the vector, plasmid pBR322. Plasmids pAH199, pAH201, pAH208, and pAH213 are derivatives of pAH191 which contain different fragments of Z. mobilis DNA that serve as promoters in E. coli. When transformed into KO11, all of these pBR322-based plasmids expressed ethyl acetate-hydrolyzing activity (as determined by the methyl red assay) and reduced the level of ethyl acetate present in beers to less than one-third the level produced by unmodified KO11 (Table 2). Beer produced with the most effective pBR322-based construct, KO11(pAH208), contained less than 20 mg of ethyl acetate liter⁻¹. Cell growth and ethanol yield were not affected by inclusion of these plasmids.

Conclusions.

These studies showed that a genetic approach can be used to effectively reduce the level of ethyl acetate, an undesired coproduct, in ethanol beers produced with E. coli KO11. The genetic improvement, expression of plasmid-borne estZ from P. putida in KO11, was accomplished without decreasing the effectiveness of the biocatalyst. Cloning of estZ was facilitated by a simple screening method, minor modifications of which should prove to be generally useful for isolation of additional esterases with alternative reaction optima and substrates. Native or recombinant EstZ can be used to decrease ethyl acetate levels in fermentation broths as a postfermentation treatment or as an in vivo recombinant enzyme produced by the biocatalyst.