Structural and biochemical characterization of a metagenome-derived esterase with a long N-terminal extension

Abstract

The genes encoding six novel esterolytic/lipolytic enzymes, termed LC-Est1∼6, were isolated from a fosmid library of a leaf-branch compost metagenome by functional screening using tributyrin agar plates. These enzymes greatly vary in size and amino acid sequence. The highest identity between the amino acid sequence of each enzyme and that available from the database varies from 44 to 73%. Of these metagenome-derived enzymes, LC-Est1 is characterized by the presence of a long N-terminal extension (LNTE, residues 26–283) between a putative signal peptide (residues 1–25) and a C-terminal esterase domain (residues 284–510). A putative esterase from Candidatus Solibacter usitatus (CSu-Est) is the only protein, which shows the significant amino acid sequence identity (46%) to the entire region of LC-Est1. To examine whether LC-Est1 exhibits activity and its LNTE is important for activity and stability of the esterase domain, LC-Est1 (residues 26–510), LC-Est1C (residues 284–510), and LC-Est1C* (residues 304–510) were overproduced in E. coli, purified, and characterized. LC-Est1C* was only used for structural analysis. The crystal structure of LC-Est1C* highly resembles that of the catalytic domain of Thermotoga maritima esterase, suggesting that LNTE is not required for folding of the esterase domain. The enzymatic activity of LC-Est1C was lower than that of LC-Est1 by 60%, although its substrate specificity was similar to that of LC-Est1. LC-Est1C was less stable than LC-Est1 by 3.3°C. These results suggest that LNTE of LC-Est1 rather exists as an independent domain but is required for maximal activity and stability of the esterase domain.

Introduction

Esterases [EC 3.1.1.1] catalyze the hydrolysis of an ester bond between a carboxylic acid and an alcohol group in water.¹ They hydrolyze this bond through the formation of an acyl-enzyme intermediate, in which the hydroxyl group of the catalytic serine residue is acylated by the substrate. Esterases assume an α/β hydrolase fold and contain a catalytic triad consisting of Ser, His, and Asp/Glu. This serine residue is usually located within a consensus sequence motif Gly-X-Ser-X-Gly (X: any amino acid). Esterases share a similar structure and the same catalytic mechanism with lipases.^1,2 However, esterases are different from lipases in substrate specificity.³ While lipases prefer water insoluble substrates with long acyl chains, esterases prefer water soluble substrates with short acyl chains. While lipases undergo interfacial activation because of the presence of a lid that covers the active site and opens upon contact with a water insoluble substrate, esterases show a Michaelis-Menten kinetic behavior because of the absence of a lid.⁴ Esterases have attracted much attention as industrial biocatalysts because of their catalytic versatility, robustness, stereoselectivity, and ability to promote synthetic reaction in organic solvent.^3,5,6

Metagenomic approach is an effective method to isolate genes encoding novel enzymes directly from environmental sources, which are useful not only for biotechnological application but also for growth of our knowledge on protein sequence space in nature.^7–11 Esterases/lipases are classified into eight families, I-VIII, in 1999, based on the differences in their amino acid sequences and biochemical properties.¹² However, since then, a variety of esterases and lipases, which do not belong to any one of these families have been identified mainly by a metagenomic approach.³

Compost is one of the promising sources of novel enzymes because a large variety of microorganisms live in compost.³ In EXPO Park, Osaka, Japan, leaves and branches cut from the trees are collected periodically, mixed with urea, and agitated for composting. We have isolated 11 novel ribonucleases H1 (LC1∼LC5- and LC7∼LC12-RNases H1)¹³ and one novel cutinase (LC-cutinase) with polyethylene terephthalate degrading activity¹⁴ from this compost by a metagenomic approach. Further structural and functional studies of these enzymes indicate that LC9-RNase H1 has an atypical DEDN active site motif,¹⁵ LC11-RNase H1 represents prokaryotic RNases H1 with a unique substrate recognition mechanism,^16,17 and LC-cutinase is a kinetically robust protein with a slow unfolding rate.¹⁸ The LC-cutinase gene has been isolated by function-based screening of a metagenomic DNA library.¹⁴ A library size and an average insert size of this library are 2.1 × 10⁴ and ∼35 kb, respectively. Clones containing cutinase genes are detected by their ability to form a halo on tributyrin agar plates (agar plates containing tributyrin) due to degradation of tributyrin. Of 6000 clones screened, 19 clones have been shown to give a halo on tributyrin agar plates.¹⁴ Three of them, which give the largest halo, contain the same gene encoding LC-cutinase. However, the genes responsible for formation of a halo of other 16 clones remain to be identified. Because not only cutinases but also esterases and lipases can degrade tributyrin, it is highly expected that these 16 clones contain novel esterase/lipase genes.

In this study, we identified six genes encoding novel metagenome-derived esterases (LC-Est1∼6) by determining the entire nucleotide sequences of the DNA inserts in the fosmids extracted from the clones that form a relatively large halo on tributyrin agar plates. Of them, LC-Est1 is a new type of esterase with a long N-terminal extension (LNTE). Biochemical characterization of LC-Est1 without a putative signal peptide (residues 26–510) and its C-terminal esterase domain (LC-Est1C, residues 284–510), and determination of the crystal structure of LC-Est1C* (residues 304–510) suggest that LNTE is not required for folding of LC-Est1 but is required for its maximal activity and stability.

Results and Discussion

Identification of novel metagenome-derived esterolytic/lipolytic enzymes

Functional screening of a fosmid library of a leaf-branch compost metagenome for genes encoding esterolytic/lipolytic enzymes using tributyrin agar plates has previously shown that 19 of 6000 clones form a halo around them due to degradation of tributyrin.¹⁴ Three of them, which form the largest halo at 50°C, contain the same gene encoding LC-cutinase. However, other 16 clones have not been analyzed for genes responsible for formation of a halo. Six of them form a relatively large halo, whereas the others form a small halo at 50°C. It is expected that activity, stability, and/or production (secretion) level of esterolytic/lipolytic enzymes from large halo-forming clones are higher than those from small halo-forming clones. To examine whether large halo-forming clones contain genes encoding novel esterolytic/lipolytic enzymes, fosmid vectors were extracted from these clones and the entire nucleotide sequences of the DNA inserts in these fosmid vectors were determined. An average size of these inserts is 35 kb. Searches of open reading frames (ORF), followed by blast searches of the amino acid sequences of ORFs deduced from their nucleotide sequences indicate that all six fosmid vectors contain a gene encoding novel esterolytic/lipolytic enzyme (one gene per one fosmid). These enzymes, which are different from one another, are termed LC-Est1∼6.

Amino acid sequences of LC-Est1∼6

Total amino acid residues of LC-Est1∼6 and the proteins, which show the highest amino acid sequence identities to these metagenome-derived enzymes, are summarized in Table I. Total amino acid residues of LC-Est1∼6 vary from 265 to 587. The highest identity between the amino acid sequence of either one of these enzymes and that available from the database varies from 44 to 73%. The nucleotide sequences of the genes encoding LC-Est1∼3 and LC-Est4∼6 are deposited in GenBank under accession numbers KM406409–KM406411 and KM406413–KM406415, respectively.

Table I.

List of Esterases/Lipases Isolated from Leaf-Branch Compost and Proteins with the Highest Amino Acid Sequence Identities

Cellulases	No. of residues	Protein with the highest sequence identity
		Protein	Source organism	Accession No.	Identity (%)
LC-Est1	510	Esterase	Candidatus Solibacter usitatus	ABJ82142	46
LC-Est2	383	Carboxylesterase typeB	Planctomyces brasiliensis	ADY57734	51
LC-Est3	265	Lipase	(bacterium enrichment culture)	ADR10200	73
LC-Est4	534	Putative carboxylesterase	Bradyrhizobium sp. STM 3843	WP_008970463	52
LC-Est5	303	Esterase/lipase	(uncultured bacterium)	AGF91880	53
LC-Est6	587	Putative esterase	Geobacillus thermoglucosidasius	GAJ43165	44

The amino acid sequences of LC-Est1∼6 are schematically shown in Figure 1. N-terminal sequence analysis using the SignalIP 3.0 Server (http://www.cbs.dtu.dk/services/SignalP-3.0/) and domain search using SMART (http://smart.embl.de) allow us to predict a signal peptide and a catalytic domain of esterase/lipase/peptidase (Pfam domain) respectively. LC-Est1, LC-Est2, and LC-Est4 have a putative signal peptide at their N-termini, suggesting that they are secretory proteins. Other proteins apparently do not have a signal peptide, suggesting that they are cytoplasmic proteins or are secreted via a Sec-independent pathway. LC-Est1, LC-Est2, and LC-Est4∼6 are probably esterases, because LC-Est1 has an “Esterase” domain, LC-Est2 and LC-Est4 have a “COesterase” (carboxylesterase) domain, and LC-Est5 and LC-Est6 have an “Abhydrolase” (α/β hydrolase) domain. The “Abhydrolase” family is represented by esterase (Est) from Pseudomonas putida. LC-Est6 has a “PepX C” (X-Pro dipeptidyl-peptidase C-terminal domain) domain in addition to an “Abhydrolase” domain, suggesting that it exhibit not only esterase activity but also peptidase activity. LC-Est3 may be a lipase, rather than an esterase, because it has a “Lipase 2” domain. Based on the difference in the amino acid sequences of the catalytic domains, LC-Est1∼6 are classified as family V (LC-Est1), family I-6 (LC-Est2), family III (LC-Est3), family VII (LC-Est4), family IV (LC-Est5), and family VI (LC-Est6) esterases/lipases, according to Arpigny and Jaeger.¹²

Figure 1.

Schematic representation of the primary structures of metagenome-derived esterolytic/lipolytic enzymes from leaf-branch compost. A putative signal peptide and a putative esterase/lipase/peptidase domain are shown by black and gray boxes respectively. “COesterase,” “Abhydrolase,” and “PepX C” represent “carboxylesterase,” “α/β-hydrolase,” and “X-Pro dipeptidyl-peptidase C-terminal domain,” respectively. The numbers above the sequence represent the positions of the N- and C-terminal residues of each protein. The numbers below the sequence represent the positions of the N- and C-terminal residues of each domain. The regions that are defined as LNTE and C-terminal esterase domain (LC-Est1C) of LC-Est1 in this study are also shown.

LC-Est1 consists of 510 amino acid residues with the calculated molecular mass of 55,620 Da and isoelectric point (pI) of 8.67, and is probably secreted in a form (residues 26–510) without a putative signal peptide. LC-Est1 is characterized by the presence of a LNTE with unknown function (Fig. 1). Residues 26–283 and residues 284–510 are arbitrarily defined as LNTE and a C-terminal esterase domain (LC-Est1C), respectively. Any domain is not detected in LNTE by domain search using SMART. Homology search indicated that only one protein shows a significant amino acid sequence similarity to LC-Est1 over the entire region. It is a putative esterase from Candidatus Solibacter usitatus Ellin6076 (CSu-Est, ABJ82142). C. S. usitatus Ellin6076 is a subdivision three member of Acidbacteria with a large genome size of 9.9 Mbp.¹⁹ Acidbacteria is one of the most widespread and abundant phyla with 26 major subdivisions found in soils and sediments. CSu-Est consists of 467 amino acid residues and shows the amino acid sequence identity of 46% to LC-Est1. It is noted, however, that many esterases/lipases show a significant amino acid sequence identity (at most 30%) to the C-terminal esterase domain of LC-Est1. Of them, thermostable esterase EstA from Thermotoga maritima MSB8 (Tm-EstA, NP_227849), for which the crystal structure is avaialble,²⁰ shows a relatively high amino acid sequence identity of 29% to the C-terminal esterase domain of LC-Est1. Because LC-Est1 and CSu-Est are a new type of esterase with LNTE and the role of LNTE remains to be analyzed, we decided to overproduce LC-Est1 and its derivative without LNTE (LC-Est1C), and purify and characterize them.

Comparison of amino acid sequences of LC-Est1, CSu-Est, and Tm-EstA

The amino acid sequence of LC-Est1 is compared with those of CSu-Est and Tm-EstA in Figure 2. CSu-Est and Tm-EstA have a putative 15-residue signal peptide at their N-termini, suggesting that they are secretory proteins as is LC-Est1. Three amino acid residues that form a catalytic triad of esterolytic/lipolytic enzymes are fully conserved as Ser399, Asp447, and His479 in LC-Est1. A pentapeptide GxSxG motif containing a catalytic serine residue is also conserved as GHSMG (residues 397–401) in LC-Est1. According to the crystal structure of Tm-EstA (PDB code 3DOH), Tm-EstA consists of an N-terminal immunoglobulin (Ig)-like domain (residues 16–157) and a C-terminal esterase domain (residues 158–395). The N-terminal Ig-like domain has been reported to be responsible for oligomerization of Tm-EstA and important for activity and stability of Tm-EstA.²⁰ An N-terminal half of LNTE of LC-Est1 and the corresponding region of CSu-Est show a weak amino acid sequence similarity to the Ig-like domain of Tm-EstA, suggesting that an N-terminal half of LNTEs of LC-Est1 and CSu-Est assumes a similar structure as that of the Ig-like domain of Tm-EstA.

Figure 2.

Alignment of amino acid sequences of LC-Est1, CSu-Est, and Tm-EstA. The amino acid residues, which are conserved in all three proteins, are denoted with white letters and highlighted in black. The amino acid residues, which are conserved in two different proteins, are highlighted in gray. The amino acid residues that form a catalytic triad (Ser399, Asp447 and His479 for LC-Est1) are denoted by asterisks. A signal peptide of each protein predicted by the SignalP 3.0 Server (http://www.cbs.dtu.dk/services/SignalP-3.0/) is underlined. Solid and open triangles above the sequences represent the positions, at which N-terminal regions of LC-Est1 are truncated to construct LC-Est1C and LC-Est1C*, respectively. The ranges of the secondary structures of LC-Est1C* and Tm-EstA are shown above and below the sequences, respectively. “α,” “β,” and “η” represent α helix, β strand, and 3₁₀ helix, respectively. Gaps are denoted by dashes. The numbers represent the positions of the amino acid residues starting from the N terminus of the protein. The accession numbers of these sequences are KM406409 for LC-Est1, ABJ82142 for CSu-Est, and NP_227849 for Tm-EstA.

Overproduction and purification of LC-Est1 and its derivatives

To examine whether LC-Est1 exhibits esterase activity and truncation of LNTE affects the activity and stability of LC-Est1, LC-Est1 without a signal peptide (Gln26-Lys510), LC-Est1C (Glu284-Lys510), and the LC-Est1C derivative without N-terminal 20 residues (LC-Est1C*, residues 304–510) were overproduced in E. coli in a His-tagged form. These proteins will be simply designated as LC-Est1, LC-Est1C, and LC-Est1C* hereafter. The amino acid residue of Tm-EstA (Thr145) corresponding to Glu284 of LC-Est1 is located in βH strand of the N-terminal Ig-like domain, which is connected to the C-terminal esterase domain through η3 3₁₀ helix (Fig. 2). Likewise, the amino acid residue of Tm-EstA (Pro173) corresponding to Pro304 of LC-Est1 is located in β2 strand of the esterase domain. Therefore, it is expected that LC-Est1C contains the entire esterase domain, whereas LC-Est1C* lacks β1 strand of the esterase domain.

Upon induction for overproduction, all proteins accumulated in E. coli cells in a soluble form. These proteins were purified to give a single band on SDS-PAGE by nickel affinity chromatography and gel filtration chromatography (data not shown). The amount of the protein purified from 1 L culture was typically 6 mg for LC-Est1, 2 mg for LC-Est1C, and 3 mg for LC-Est1C*. The molecular masses of LC-Est1, LC-Est1C, and LC-Est1C* were estimated to be 58, 28, and 26 kDa, respectively, by gel filtration chromatography. These values are comparable to the calculated ones (55,620 Da for LC-Est1, 27,382 Da for LC-Est1C, and 25,094 Da for LC-Est1C*), suggesting that LC-Est1, LC-Est1C, and LC-Est1C* exist as a monomer.

The far-UV circular dichroism (CD) spectra of LC-Est1 and LC-Est1C are shown in Figure 3. Both proteins give a broad trough with a minimum [θ] value around 220 nm. However, the trough of the spectrum of LC-Est1C is shallower than that of LC-Est1, suggesting that the helical content of LC-Est1C is lower than that of LC-Est1. The far-UV CD spectrum of LC-Est1C* was similar to that of LC-Est1C (data not shown).

Figure 3.

Far-UV CD spectra. The far-UV CD spectra of LC-Est1 (solid line) and LC-Est1C (broken line) were measured at 25°C in 10 mM Tris-HCl (pH 7.0) as described in Materials and Methods.

Crystal structure of LC-Est1C*

To examine whether truncation of LNTE affects the structure of the C-terminal esterase domain of LC-Est1, it is necessary to determine the crystal structures of LC-Est1 and its derivative without LNTE. Therefore, we tried to crystallize LC-Est1, LC-Est1C, and LC-Est1C*. However, the crystals suitable for X-ray diffraction were only obtained for LC-Est1C*. Attempts to crystallize LC-Est1 and LC-Est1C have so far been unsuccessful, probably because LNTE and an N-terminal region of LC-Est1C are highly flexible and prevent the formation of crystals suitable for X-ray diffraction.

The crystal structure of LC-Est1C* was determined by the multiwavelength anomalous dispersion (MAD) phasing method at 1.53 Å resolution. Data collection and refinement statistics are summarized in Table II. The asymmetric unit of the crystal structure consists of two molecules (molecular A and B). Molecular A contains 193 of 207 residues, with Arg361-Met366 and Ala503-Lys510 missing, probably due to structural disorder. Molecular B contains 191 of 207 residues, with Arg361-Met366 and Arg501-Lys510 missing. The structure of these two molecules are well superimposed with the root-mean-square deviation (RMSD) value of 0.24 Å for 191 Cα atoms. We used the structure of molecule A in this study.

Table II.

Data Collection and Refinement Statistics of LC-Est1C*

	Native	SeMet-peak	SeMet-inflection	SeMet-remote
Data collection statistics
Wavelength (Å)	0.900	0.978769	0.979101	0.994813
Space group	P212121	P212121	P212121	P212121
Cell parameters
a, b, c (Å)	45.86, 55.45, 156.27	46.13, 55.73, 156.24	46.13, 55.78, 156.32	46.13, 55.79, 156.35
α = β = γ (°)	90	90	90	90
Molecules/asymmetric unit	2	2	2	2
Resolution range (Å)	50.00–1.53 (1.56–1.53)a	50.00–2.30 (2.34–2.30)a	50.00–2.30 (2.34–2.30)a	50.00–2.30 (2.34–2.30)a
Reflections measured	392,694	123,978	122,975	121,933
Unique reflections	60,179	17,756	17,643	17,581
Redundancy	6.5 (7.2)a	7.0 (7.3)a	7.0 (7.3)a	6.9 (7.3)a
Completeness (%)	98.5 (100.0)a	94.6 (99.7)a	93.9 (99.5)a	93.6 (99.5)a
Rmerge (%)b	18.1 (28.3)a	11.5 (33.4)a	9.5 (36.5)a	10.3 (42.2)a
Average I/σ (I)	14.5 (3.6)a	23.3 (5.0)a	23.5 (4.5)a	22.2 (3.8)a
Refinement statistics
Resolution range (Å)	78.14–1.53
No. of atoms
Protein/Water	2992/451
Rwork (%)	19.1
Rfree (%)c	22.5
Rms deviations from ideal values
Bond lengths (Å)	0.014
Bond angles (°)	1.439
Average B factors (Å2)
Protein/Water	18.3/31.6
Ramachandran plot statistics
Favored region (%)	97.1
Allowed region (%)	2.9

^aValues in parentheses are for the highest-resolution shell.

^bR_merge = ∑|I_hkl — <I_hkl>|/∑ I_hkl, where I_hkl is an intensity measurement for reflection with indices hkl and <I_hkl> is the mean intensity for multiply recorded reflections.

^cFree R-value was calculated using 5% of the total reflections chosen randomly and omitted from refinement.

The structure of LC-Est1C* shows a typical α/β hydrolase fold (Fig. 4). This structure shows the highest similarity to the structure of the C-terminal esterase domain of Tm-EstA (PDB entry 3DOH) with the RMSD value of 1.25 Å for 191 Cα atoms. The steric configurations of the active-site residues forming a catalytic triad of LC-Est1C* (Ser399, Asp447, and His479) are nearly identical to those of Tm-EstA (Ser286, Asp334, and His374). The secondary structures of LC-Est1C∗ are similar to those of the esterase domain of Tm-EstA, except for loops between β4-strand and αB-helix and between β8-strand and αF-helix. The former loop of LC-Est1C* (Lys359-Gly369) is short and is not fully visible, whereas the corresponding loop of Tm-EstA (Pro234-Pro255) is long and contains Phe246 which participates in the formation of a tunnel at the active site and affects enzymatic activity.²⁰ Likewise, the latter loop of LC-Est1C* (Pro475-His479) is short, whereas the corresponding loop of Tm-EstA (Glu362-His374) is relatively long and contains η7-helix. The finding that the structure of LC-Est1C* highly resembles that of the esterase domain of Tm-EstA strongly suggests that truncation of LNTE does not significantly affect the structure of the esterase domain of LC-Est1. However, because LC-Est1C* may lack β1 strand of the esterase domain, we used LC-Est1C for biochemical characterization.

Figure 4.

Crystal structure of LC-Est1C*. The stereo view of the structure of LC-Est1C* (cyan) is superimposed on that of the C-terminal esterase domain of Tm-EstA (PDB entry 3DOH). Three active-site residues of LC-Est1C* (Ser399, Asp447, and His479) and the corresponding residues of Tm-EstA (Ser286, Asp334, and His374) are shown by stick models. N and C represent N- and C-termini.

Enzymatic activities of LC-Est1 and LC-Est1C

The enzymatic activities of LC-Est1 and LC-Est1C were determined at pH 7.5 and 30°C using various p-nitrophenyl (pNP) monoesters of fatty acids with acyl chain lengths of 2–14. The results are shown in Figure 5. LC-Est1 exhibited the highest activity toward the pNP-butyrate (C4) substrate. It exhibited comparable activity toward the pNP-hexanoate (C6) substrate. However, it exhibited 15–25% of the maximal activity toward the pNP-acetate (C2) and pNP-caprylate (C8) substrates and very weak activity toward the substrates with acyl chain lengths of ≥10. Thus, LC-Est1 shows a strong preference for the C4 and C6 substrates. LC-Est1C also showed a strong preference for these substrates, although it exhibited the highest activity toward the C6 substrate. Both enzymes did not show the activity to degrade olive oil. Thus, truncation of LNTE does not significantly affect the substrate specificity of LC-Est1. However, the specific activity of LC-Est1C determined toward the C4 substrate (101 units μmol⁻¹) was lower than that of LC-Est1 (255 units μmol⁻¹) by 60%, indicating that truncation of LNTE considerably decreases the activity of LC-Est1. The specific activity was defined as the enzymatic activity per μmol of protein, instead of mg of protein, because the molecular mass of LC-Est1C is nearly one-half of that of LC-Est1. If the specific activity is defined as the activity per mg of protein, it is 4.6 units mg⁻¹ for LC-Est1 and 3.7 units mg⁻¹ for LC-Est1C.

Figure 5.

Substrate selectivity of LC-Est1 and LC-Est1C. Specific activities of LC-Est1 (solid bar) and LC-Est1C (gray bar) toward triglyceride substrates are shown. C2∼C14 represent the acyl chain lengths of the substrates. The experiment was carried out in duplicate. Each value represents the average value and errors from the average values are shown.

The temperature dependencies of the activities of LC-Est1 and LC-Est1C were analyzed at various temperatures ranging from 20 to 80°C and pH 7.5 using pNP-butyrate (C4) as a substrate (Fig. 6). Both enzymes exhibited the highest activity at 40°C. The pH dependencies of LC-Est1 and LC-Est1C were analyzed at various pHs ranging from 4.0 to 9.0 and 30°C using pNP-butyrate (C4) as a substrate (Fig. 7). The pH dependencies and substrate specificities of LC-Est1 and LC-Est1C were analyzed at 30°C, instead of the optimum temperature for activity (40°C), because the stability of certain substrates, such as pNP-acetate decreases as the temperature increases beyond 40°C. Both enzymes exhibited the highest activity at pH 7.5, although LC-Est1C exhibited slightly higher activity at pH 8.0 than at pH 7.5 when the Tris-HCl buffers with various pHs ranging from 7.0 to 9.0 were used to determine the activity. These results indicate that truncation of LNTE does not significantly affect the optimum pH and optimum temperature for activity of LC-Est1, which are 40°C and pH 7.5, respectively. It is noted that the specific activity and optimum temperature for activity of LC-Est1C* were nearly identical to those of LC-Est1C, suggesting that truncation of N-terminal 20 residues does not significantly affect the activity and stability of LC-Est1C, although a possibility that this truncation slightly affects the stability of LC-Est1C cannot be ruled out.

Figure 6.

Optimum temperatures for activities of LC-Est1 and LC-Est1C. The temperature dependencies of the enzymatic activities of LC-Est1 (solid circle, solid line) and LC-Est1C (open circle, broken line) are shown. The activity was determined at pH 7.5 and the temperatures indicated using pNP-butyrate (C4) as a substrate, as described in Materials and Methods. The experiment was carried out at least twice, and errors from the average values are indicated by vertical lines.

Figure 7.

Optimum pH for activities of LC-Est1 and LC-Est1C. The pH dependencies of the enzymatic activities of LC-Est1 (solid line) and LC-Est1C (broken line) are shown. The activity was determined at 30°C and the pH values indicated using pNP-butyrate (C4) as a substrate, as described in Materials and Methods. The buffers used to analyze the pH dependence of activity were 20 mM sodium citrate (pH 4.0–6.0; open circle), 20 mM sodium phosphate (pH 6.0–8.0; cross) and 20 mM Tris-HCl (pH 7.0–9.0; solid square). The experiment was carried out at least twice, and errors from the average values are indicated by vertical lines.

To examine whether the enzymatic activity of LC-Est1C decreases as compared with that of LC-Est1 due to a decrease in substrate-binding affinity or turnover number, the kinetic parameters of LC-Est1 and LC-Est1C were determined at 30°C and pH 7.5 using pNP-butyrate (C4) as a substrate. Both enzymes follow a Michaelis-Menten kinetics. The kinetic parameters of these enzymes determined from the Lineweaver-Burk plots are summarized in Table III. The K_m value of LC-Est1C is slightly higher than but comparable to that of LC-Est1. Whereas, the k_cat value of LC-Est1C is lower than that of LC-Est1 by roughly 60%. These results suggest that truncation of LNTE reduces the turnover number of LC-Est1 without significantly affecting its substrate binding affinity. Truncation of LNTE probably slightly alters the conformation of the active site, in such a way that it is not optimum for activity.

Table III.

Specific Activities, Kinetic Parameters, and T_1/2 Values of LC-Est1 and LC-Est1C^a

Protein	Specific activity (unit/μmol)	Relative activity (%)	Km (mM)	kcat (s−1)	T1/2 (°C)	ΔT1/2b (°C)
LC-Est1	255 ± 31	100	0.30 ± 0.01	5.8 ± 0.3	67.2 ± 0.4	–
LC-Est1C	101 ± 19	40	0.48 ± 0.04	2.4 ± 0.1	63.9 ± 0.1	−3.3

^aThe enzymatic activity was determined at 30°C for 10 min, in 100 μL of 20 mM Tris-HCl (pH 7.5) containing 10% acetonitrile, by using pNP butyrate (C4) as a substrate. The substrate concentration was 1 mM. For determination of the kinetic parameters, the substrate concentration was varied from 0.25 to 6.0 mM. The relative activity was calculated by dividing the specific activity of LC-Est1C by that of LC-Est1. Experiments were carried out at least twice and the average values are shown together with the errors. The temperature of the midpoint of the thermal denaturation transition, T_1/2, was determined by monitoring the change in the CD value at 222 nm.

^bΔT_1/2 = T_1/2(LC-Est1C) – T_1/2(LC-Est1).

Thermal stability of LC-Est1 and LC-Est1C

To examine whether truncation of LNTE affects the stability of LC-Est1, thermal denaturation of LC-Est1 and LC-Est1C was analyzed at pH 7.0 by monitoring the change in CD values at 222 nm as the temperature was increased. Thermal denaturation of these proteins was irreversible at the condition examined. The thermal denaturation curves of these proteins were reproducible, unless the protein concentration, pH, and rate of temperature increase were seriously changed. The thermal denaturation curves of LC-Est1 and LC-Est1C measured in 10 mM Tris-HCl (pH 7.0) are shown in Figure 8. The midpoints of the transition of these thermal denaturation curves, T_1/2, are summarized in Table III. The T_1/2 value of LC-Est1C is lower than that of LC-Est1 by 3.3°C, suggesting that LNTE contributes to the stabilization of LC-Est1, but not so significantly.

Figure 8.

Thermal denaturation LC-Est1 and LC-Est1C. The thermal denaturation curves of LC-Est1 (solid line) and LC-Est1C (broken line) measured in 10 mM Tris-HCl (pH 7.0) are shown. The curves were recorded by monitoring the change in CD values at 222 nm as described in Materials and Methods.

Role of LNTE

Role of LNTE of LC-Est1 remains to be understood, because it shows little amino acid sequence similarity to any protein with known function and the crystal structure of LC-Est1 remains to be determined. It has been reported that the Tm-EstA derivative without the N-terminal Ig-like domain (Tm-EstAC, residues 158–395) exhibits much lower activity and stability than those of Tm-EstA, and loses an ability to oligomerize.²⁰ Tm-EstA exists as a hexamer in solution and is a highly thermostable enzyme with the optimum temperature for activity of ≥95°C and half-life at 100°C of 1.5 h. In contrast, Tm-EstAC exists as a monomer and exhibits 5% of the activity of Tm-EstA. The optimum temperature for activity of Tm-EstAC is 60°C and its half-life at 90°C is 15 min. Based on these results, it has been proposed that the Ig-like domain participates in the catalytic function by guiding the substrate to the active site of Tm-EstA. However, LNTE of LC-Est1 may not play a similar role as that of the Ig-like domain of Tm-EstA, because both LC-Est1 and LC-Est1C exist as a monomer in solution, and truncation of LNTE does not significantly affect the substrate binding affinity of LC-Est1 and only marginally destabilizes it. LNTE of LC-Est1 probably exists as an independent domain and weakly interacts with the esterase domain only at the region where it is connected to the esterase domain. Removal of these interactions probably destabilizes the protein and at the same time slightly alters the conformation of the active site.

Structural features of LC-Est1C∗ and Tm-EstAC

Tm-EstAC is greatly destabilized as compared to Tm-EstA, but is apparently still more stable than LC-Est1C*, because the optimum temperature for activity of Tm-EstAC (60°C) is higher than that of LC-Est1C* (40°C). Proteins are stabilized by the combination of a variety of the factors, such as increased number of ion pairs,²¹ increased number of hydrogen bonds,²² increased number of proline residues in loop regions,²³ increased number of disulfide bonds²⁴ increased interior hydrophobicity,²⁵ and anchoring of C-terminal tail.²⁶ Comparison of these factors between LC-Est1C* and Tm-EstAC indicates that the number of ion pairs, the number of hydrogen bonds, and the number of proline residues in loop regions of LC-Est1C* are lower than those of Tm-EstAC, although the difference is not so significant (Table IV). The number of ion pairs is 11 for LC-Est1C* and 14 for Tm-EstAC, the number of hydrogen bonds is 169 for LC-Est1C* and 206 for Tm-EstAC, and the number of proline residues in loop regions is 10 for LC-Est1C* and 12 for Tm-EstAC. However, both proteins do not contain a disulfide bond and their C-terminal tails are not anchored to the central region. The interior hydrophobicity of LC-Est1C* is rather higher than that of Tm-EstAC, because the ratio of interior apolar residues to total interior residues is 71.3% for LC-Est1C* and 65.4% for Tm-EstAC. These results suggest that the difference in the number of ion pairs, the number of hydrogen bonds, and interior hydrophobicity may at least partly account for the difference in stability between LC-Est1C* and Tm-EstAC.

Table IV.

Structural Features of LC-Est1C* and Tm-EstAC

	LC-Est1C*	Tm-EstAC
No. of amino acids	199	222
Opt. Temp. for activity (°C)a	40	60
Content of the residues (%)
Buried polar	15.6	20.2
Buried apolar	38.7	38.3
Buried apolar/buried total	71.3	65.4
No. of ion pairs (≤4.0 Å)	11	14
No. of hydrogen bonds	169	206
No. of disulfide bond	0	0
No. of Pro in loop	10	12

^aData from Ref.22 for Tm-EstAC.

Materials and Methods

Cells, plasmids, and enzymes

E. coli BL21-CodonPlus(DE3)-RP was from Stratagene (La Jolla, CA). Plasmid pET25b was from Novagen (Madison, WI). E. coli BL21-CodonPlus(DE3)-RP transformants were grown in Luria-Bertani broth medium (10 g Tryptone; 5 g Yeast extract; 10 g NaCl in 1 L H₂O) supplemented with 50 mg L⁻¹ ampicillin.

Plasmid construction

The pET25b derivatives for overproduction of LC-Est1 (residues 26–510), LC-Est1C (residues 284–510), and LC-Est1C* (residues 304–510) were constructed by polymerase chain reaction (PCR). The fosmid vector harboring the LC-Est1 gene ²⁰ was used as a template. The sequences of the PCR primers are 5′-TCAGCGCATATGCAGGATGCCGGGCAGGTG-3′ for primer 1, 5′-AAGAACCATATGGAGGCGCGCCGGGGGGAC-3′ for primer 2, 5′-AACACGCATATGCCTTATCGCCTCTACGTC-3 for primer 3, and 5′-GATGAATTCTCATTTATTCGTGGCCGC-3′ for primer 4, where the NdeI (primers 1–3), and EcoRI (primer 4) sites are underlined. Primers 1, 2, and 3 are designed such that the ATG codon is attached to the 5′-termini of the genes encoding LC-Est1, LC-Est1C, and LC-Est1C*, respectively. Primers 1 and 4, primers 2 and 4, and primers 3 and 4 were used to amplify the genes encoding LC-Est1, LC-Est1C, and LC-Est1C*, respectively. The resultant DNA fragments were digested with NdeI and EcoRI, and ligated into the NdeI-EcoRI sites of pET28a.

All DNA oligomers for PCR were synthesized by Hokkaido System Science (Sapporo, Japan). PCR was performed in 30 cycles using a thermal cycler (Gene Amp PCR System 2400; Applied Biosystems, Tokyo, Japan) and KOD DNA polymerase (Toyobo, Kyoto, Japan). The DNA sequences of the genes encoding all proteins mentioned above were confirmed by ABI Prism 310 DNA sequencer (Applied Biosystems).

Overproduction and purification

For overproduction of LC-Est1, LC-Est1C, and LC-Est1C*, E. coli BL21-CodonPlus(DE3)-RP transformants with the pET25b derivatives were grown at 37°C. When the absorbance of the culture at 600 nm reached around 0.5, isopropyl-β-d-thio galactopyranoside was added to the culture medium and cultivation was continued for an additional 4 h. The subsequent purification procedures were carried out at 4°C. Cells were harvested by centrifugation at 8000g for 10 min, suspended in 20 mM phosphate buffer (pH 7.0) containing 1 mM EDTA, disrupted by sonication lysis, and centrifuged at 30,000g for 30 min. The resultant supernatant was collected, dialyzed against 20 mM phosphate buffer (pH 6.0), and loaded onto a HiTrap SP HP column (GE Healthcare, Tokyo, Japan) equilibrated with the same buffer. The protein was eluted from the column by linearly increasing the NaCl concentration from 0 to 1M. The fractions containing the protein were collected and dialyzed against 20 mM phosphate buffer (pH 7.0) containing 10 mM imidazole and 0.3M NaCl, and applied to a Ni Sepharose 6 Fast Flow column (GE Healthcare, Tokyo, Japan) equilibrated with the same buffer. The protein was eluted from the column by linearly increasing the imidazole concentration from 10 to 300 mM. The fractions containing the protein were collected and dialyzed against 10 mM Tris-HCl (pH 7.5).

Selenomethionine-substituted LC-Est1C* was overproduced using methionine-auxotrophic E. coli strain B834 (DE3) pLysS in a defined medium.²⁷ The purification procedures were the same as those for the native enzyme.

The production level of the protein in E. coli cells and the purity of the protein were analyzed by SDS-polyacrylamide gel electrophoresis²⁸ using a 12% polyacrylamide gel, followed by staining with Coomassie brilliant blue (CBB). The amount of the protein was estimated from the intensity of the band visualized by CBB staining using the Scion Image program. The protein concentration was determined from the UV absorption on the basis that the absorbance of a 0.1% (1.0 mg mL⁻¹) solution at 280 nm is 0.75 for LC-Est1, 0.98 for LC-Est1C, and 1.01 for LC-Est1C*. These value was calculated by using ε = 1526 M⁻¹ cm⁻¹ for tyrosine and 5225 M⁻¹ cm⁻¹ for tryptophan at 280 nm.²⁹

Enzyme assay

The enzymatic activity was determined at the temperature indicated in 100 μL of 20 mM Tris-HCl (pH 7.5) containing 10% acetonitrile and 1 mM pNP monoesters of fatty acids with a chain length from C2 to C14. The amount of p-nitrophenol (pNP) released from the substrate was determined from the absorption at 412 nm (A₄₁₂) with an absorption coefficient of 14,200 M⁻¹ cm⁻¹. The initial velocity of the enzymatic reaction was determined by recording a time-dependent increase in A₄₁₂ values for 5 min by automatic spectrophotometer (Hitachi Spectrophotometer U-2810; Hitachi High-Technologies, Tokyo, Japan). One unit of enzymatic activity was defined as the amount of enzyme that produced 1 μmol of pNP per min. Specific activity was defined as the enzymatic activity per μmol of protein.

For analysis of the temperature dependence of activity, the activity was determined at 20 mM Tris-HCl (pH 7.5) and various temperatures ranging from 20 to 80°C. For the analysis of pH dependence of activity, the activity was determined at 30°C and various pHs ranging from pH 4.0 to 9.0. The buffers used for this analysis were 20 mM sodium citrate (pH 4.0–6.0), 20 mM sodium phosphate (pH 6.0–8.0), and 20 mM Tris-HCl (pH 7.0–9.0).

For determination of the kinetic parameters, pNP-butyrate (C4) was used as a substrate. The concentration of this substrate was varied from 0.25 to 6.0 mM. Hydrolysis of this substrate by the enzyme followed Michaeleis-Menten kinetics, and the kinetic parameters were determined from the Lineweaver-Burk plot.

CD spectra measurement

The far-UV (200–260 nm) CD spectrum of the protein was measured at 25°C on a J-725 spectropolarimeter (Japan Spectroscopic, Tokyo, Japan). The protein was dissolved in 10 mM Tris-HCl (pH 7.0). The protein concentration was 0.1 mg mL⁻¹ and a cell with an optical path length of 2 mm was used. The mean residual ellipticity (θ, deg cm² dmol⁻¹) was calculated using an average amino acid molecular mass of 110 Da.

Thermal denaturation

The thermal denaturation curve of the protein was obtained by monitoring the change in CD values at 222 nm as the temperature was increased. The protein was dissolved in 10 mM Tris-HCl (pH 7.0). The protein concentration and optical path length were 0.1 mg mL⁻¹ and 2 mm, respectively. The rate of temperature increase was 1.0°C min⁻¹. The thermal denaturation processes of the proteins examined were irreversible under this condition. The temperature of the midpoint of the transition, T_1/2, was calculated by curve fitting of the resultant CD values versus temperature data on the basis of a least-square analysis.

Crystallization

Prior to crystallization, native LC-Est1C* and SeMet-labeled LC-Est1C* were dialyzed against 10 mM Tris-HCl (pH 7.0) and concentrated to 19 mg mL⁻¹ using an ultrafiltration system Centricon (Millipore, Billerica, MA). The crystallization conditions were screened using crystallization kits from Hampton Research (Aliso Viejo, CA; Crystal Screens and Index) and Emerald BioStructures and Emerald BioSystems (Bainbridge Island, WA) (Wizard I-IV) by the sitting-drop vapor-diffusion method. Drop solutions were prepared by mixing 1 μL each of protein and reservoir solutions and equilibrated against a 100 μL reservoir solution. Crystals suitable for X-ray diffraction of native LC-Est1C* appeared after one week in Index No.71 [0.1M Bis-Tris (pH 6.5), 0.2M sodium chloride, and 25% (w/v) polyethylene glycol 3,350] at 20°C. Furthermore, the crystallization condition of SeMet-labeled LC-Est1C* was further optimized and the better quality crystals were obtained after two week with optimized reservoir solution [0.1M Bis-Tris (pH 6.5), 0.18M sodium chloride, and 21% (w/v) polyethylene glycol 3,350] at 4°C.

X-ray diffraction data collection and structure determination

X-ray diffraction data set of native LC-Est1C* and SeMet-labeled LC-Est1C* for MAD phasing was collected at −173°C under a nitrogen stream at beam line BL44XU in SPring-8 (Hyogo, Japan). Processing of the image data was performed using the HKL2000 program suite.³⁰ The structure of LC-Est1C* was solved by MAD phasing method with MAD data from the SeMet-labeled LC-Est1C* crystal and the native data set using the SHELX program suite³¹ in the HKL2MAP interface.³² Automated model building was conducted using ArpWarp.³³ Structural refinement was carried out by using REFMAC³⁴ of the CCP4 program and COOT.³⁵ The statistics for data collection and refinement are summarized in Table II. The figures were prepared using PyMol (http://www.pymol.org).

Protein data bank accession number: The coordinates and structure factors for LC-Est1C* have been deposited in the PDB under ID code 3WYD.