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

Article abstract of DOI:10.1002/pro.2591

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 s

Full text of DOI:10.1002/pro.2591

Structural and biochemical
characterization of a metagenome-derived
esterase with a long N-terminal extension
1
1
1
1,2
Hiroyuki Okano, Xun Hong, Eiko Kanaya, Clement Angkawidjaja,
1
and Shigenori Kanaya *
1
Department of Material and Life Science, Graduate School of Engineering, Osaka University, 565–0871, Japan
2
International College, Osaka University, Toyonaka, Osaka 560-0043, Japan
Received 25 September 2014; Revised 15 October 2014; Accepted 16 October 2014
DOI: 10.1002/pro.2591
Published online 28 October 2014 proteinscience.org
Abstract: The genes encoding six novel esterolytic/lipolytic enzymes, termed LC-Est1~6, were iso-
lated 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 (resi-
dues 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.
Keywords: metagenome; esterase; N-terminal extension; crystal structure; stability
Introduction
due is acylated by the substrate. Esterases assume
an a/b hydrolase fold and contain a catalytic triad
consisting of Ser, His, and Asp/Glu. This serine resi-
due is usually located within a consensus sequence
motif Gly-X-Ser-X-Gly (X: any amino acid). Ester-
ases share a similar structure and the same cata-
Esterases [EC 3.1.1.1] catalyze the hydrolysis of an
ester bond between a carboxylic acid and an alcohol
1
group in water. They hydrolyze this bond through
the formation of an acyl-enzyme intermediate, in
which the hydroxyl group of the catalytic serine resi-
1,2
lytic mechanism with lipases.
However, esterases
3
are different from lipases in substrate specificity.
While lipases prefer water insoluble substrates with
long acyl chains, esterases prefer water soluble sub-
strates 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
Grant sponsor: Ministry of Education, Culture, Sports, Science,
and Technology of Japan; Grant number: 24380055.
*Correspondence to: Shigenori Kanaya, Department of Material
and Life Science, Graduate School of Engineering, Osaka Uni-
versity, 2-1 Yamadaoka, Suita, Osaka 565–0871, Japan.
E-mail: kanaya@mls.eng.osaka-u.ac.jp
Published by Wiley-Blackwell. V
C
2014 The Protein Society
PROTEIN SCIENCE 2015 VOL 24:93—104 93

with a water insoluble substrate, esterases show a
Michaelis-Menten kinetic behavior because of the
chemical 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.
4
absence of a lid. Esterases have attracted much
attention as industrial biocatalysts because of their
catalytic versatility, robustness, stereoselectivity,
and ability to promote synthetic reaction in organic
3
,5,6
solvent.
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
Results and Discussion
Identification of novel metagenome-derived
esterolytic/lipolytic enzymes
7
–11
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 trib-
nature.
Esterases/lipases are classified into eight
families, I-VIII, in 1999, based on the differences in
their amino acid sequences and biochemical proper-
1
2
ties. 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 metage-
14
utyrin. Three of them, which form the largest halo
ꢀ
3
at 50 C, contain the same gene encoding LC-
nomic approach.
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,
Compost is one of the promising sources of novel
enzymes because a large variety of microorganisms
3
live in compost. In EXPO Park, Osaka, Japan,
ꢀ
whereas the others form a small halo at 50 C. It is
leaves and branches cut from the trees are collected
periodically, mixed with urea, and agitated for com-
posting. We have isolated 11 novel ribonucleases H1
expected that activity, stability, and/or production
(
secretion) level of esterolytic/lipolytic enzymes from
1
3
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 nucleo-
tide 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 con-
(
LC1ꢁLC5- and LC7ꢁLC12-RNases H1) and one
novel cutinase (LC-cutinase) with polyethylene ter-
1
4
ephthalate degrading activity 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
1
5
motif,
LC11-RNase H1 represents prokaryotic
RNases H1 with a unique substrate recognition
1
6,17
mechanism,
and LC-cutinase is a kinetically
1
8
robust protein with a slow unfolding rate. The LC-
cutinase gene has been isolated by function-based
1
4
tain a gene encoding novel esterolytic/lipolytic
screening of a metagenomic DNA library. A library
enzyme (one gene per one fosmid). These enzymes,
which are different from one another, are termed
LC-Est1ꢁ6.
size and an average insert size of this library are 2.1
4
3
10 and ꢁ35 kb, respectively. Clones containing
cutinase genes are detected by their ability to form a
halo on tributyrin agar plates (agar plates contain-
ing tributyrin) due to degradation of tributyrin. Of
Amino acid sequences of LC-Est1ꢁ6
6
000 clones screened, 19 clones have been shown to
Total amino acid residues of LC-Est1ꢁ6 and the pro-
teins, which show the highest amino acid sequence
identities to these metagenome-derived enzymes, are
summarized in Table I. Total amino acid residues of
LC-Est1ꢁ6 vary from 265 to 587. The highest iden-
tity between the amino acid sequence of either one
of these enzymes and that available from the data-
base varies from 44 to 73%. The nucleotide sequen-
ces of the genes encoding LC-Est1ꢁ3 and LC-
Est4ꢁ6 are deposited in GenBank under accession
numbers KM406409–KM406411 and KM406413–
KM406415, respectively.
1
4
give a halo on tributyrin agar plates. 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 tribu-
tyrin, it is highly expected that these 16 clones con-
tain novel esterase/lipase genes.
In this study, we identified six genes encoding
novel metagenome-derived esterases (LC-Est1ꢁ6) 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 ester-
ase with a long N-terminal extension (LNTE). Bio-
The amino acid sequences of LC-Est1ꢁ6 are sche-
matically 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
9
4
PROTEINSCIENCE.ORG
Characterization of Metagenome-Derived Esterase

Table I. List of Esterases/Lipases Isolated from Leaf-Branch Compost and Proteins with the Highest Amino Acid
Sequence Identities
Protein with the highest sequence identity
Cellulases No. of residues
Protein
Source organism
Accession No.
Identity (%)
LC-Est1
LC-Est2
LC-Est3
LC-Est4
LC-Est5
LC-Est6
510
383
265
534
303
587
Esterase
Carboxylesterase typeB
Lipase
Putative carboxylesterase Bradyrhizobium sp. STM 3843
Esterase/lipase
Candidatus Solibacter usitatus
Planctomyces brasiliensis
(bacterium enrichment culture)
ABJ82142
46
51
73
52
53
44
ADY57734
ADR10200
WP_008970463
AGF91880
(uncultured bacterium)
Geobacillus thermoglucosidasius GAJ43165
Putative esterase
using SMART (http://smart.embl.de) allow us to pre-
dict a signal peptide and a catalytic domain of ester-
ase/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 cyto-
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
2
0
plasmic proteins or are secreted via
a
Sec-
avaialble,
shows a relatively high amino acid
independent pathway. LC-Est1, LC-Est2, and LC-
Est4ꢁ6 are probably esterases, because LC-Est1 has
an “Esterase” domain, LC-Est2 and LC-Est4 have a
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 overpro-
duce LC-Est1 and its derivative without LNTE (LC-
Est1C), and purify and characterize them.
“COesterase” (carboxylesterase) domain, and LC-
Est5 and LC-Est6 have an “Abhydrolase” (a/b hydro-
lase) domain. The “Abhydrolase” family is repre-
sented by esterase (Est) from Pseudomonas putida.
LC-Est6 has a “PepX C” (X-Pro dipeptidyl-peptidase
C-terminal domain) domain in addition to an
Comparison of amino acid sequences of
LC-Est1, CSu-Est, and Tm-EstA
“Abhydrolase” domain, suggesting that it exhibit not
The amino acid sequence of LC-Est1 is compared
with those of CSu-Est and Tm-EstA in Figure 2.
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 cata-
lytic domains, LC-Est1ꢁ6 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 fam-
ily VI (LC-Est6) esterases/lipases, according to
1
2
Arpigny and Jaeger.
LC-Est1 consists of 510 amino acid residues with
the calculated molecular mass of 55,620 Da and iso-
electric 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 indi-
cated 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 Soli-
bacter usitatus Ellin6076 (CSu-Est, ABJ82142). C. S.
usitatus Ellin6076 is a subdivision three member of
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,” “a/b-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.
19
Acidbacteria with a large genome size of 9.9 Mbp.
Acidbacteria is one of the most widespread and abun-
dant phyla with 26 major subdivisions found in soils
and sediments. CSu-Est consists of 467 amino acid
Okano et al.
PROTEIN SCIENCE VOL 24:93—104 95

Figure 2. Alignment of amino acid sequences of LC-Est1, CSu-Est, and Tm-EstA. The amino acid residues, which are con-
served in all three proteins, are denoted with white letters and highlighted in black. The amino acid residues, which are con-
served 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 sec-
ondary structures of LC-Est1C* and Tm-EstA are shown above and below the sequences, respectively. “a,” “b,” and “h” repre-
sent a helix, b 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.
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 penta-
peptide GxSxG motif containing a catalytic serine
residue is also conserved as GHSMG (residues 397–
stability of LC-Est1, LC-Est1 without a signal pep-
tide (Gln26-Lys510), LC-Est1C (Glu284-Lys510), and
the LC-Est1C derivative without N-terminal 20 resi-
dues (LC-Est1C*, residues 304–510) were overpro-
duced 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 bH strand of the N-terminal Ig-
like domain, which is connected to the C-terminal
esterase domain through h3 310 helix (Fig. 2). Like-
wise, the amino acid residue of Tm-EstA (Pro173)
corresponding to Pro304 of LC-Est1 is located in b2
strand of the esterase domain. Therefore, it is
expected that LC-Est1C contains the entire esterase
domain, whereas LC-Est1C* lacks b1 strand of the
esterase domain.
4
01) 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 (res-
idues 16–157) and a C-terminal esterase domain
(
residues 158–395). The N-terminal Ig-like domain
has been reported to be responsible for oligomeriza-
tion of Tm-EstA and important for activity and sta-
2
0
bility of Tm-EstA. 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.
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 fil-
tration 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
Overproduction and purification of LC-Est1 and
its derivatives
To examine whether LC-Est1 exhibits esterase activ-
ity and truncation of LNTE affects the activity and
9
6
PROTEINSCIENCE.ORG
Characterization of Metagenome-Derived Esterase

proteins give a broad trough with a minimum [h]
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).
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 struc-
tures of LC-Est1 and its derivative without LNTE.
Therefore, we tried to crystallize LC-Est1, LC-
Est1C, and LC-Est1C*. However, the crystals suita-
ble 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.
Figure 3. Far-UV CD spectra. The far-UV CD spectra of LC-
Est1 (solid line) and LC-Est1C (broken line) were measured at
ꢀ
2
5 C in 10 mM Tris-HCl (pH 7.0) as described in Materials
and Methods.
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*), sug-
gesting that LC-Est1, LC-Est1C, and LC-Est1C*
exist as a monomer.
The crystal structure of LC-Est1C* was deter-
mined by the multiwavelength anomalous dispersion
(MAD) phasing method at 1.53 A˚ resolution. Data
The far-UV circular dichroism (CD) spectra of
LC-Est1 and LC-Est1C are shown in Figure 3. Both
collection and refinement statistics are summarized
in Table II. The asymmetric unit of the crystal
Table II. Data Collection and Refinement Statistics of LC-Est1C*
Native
SeMet-peak
SeMet-inflection
0.979101
SeMet-remote
0.994813
Data collection statistics
˚
Wavelength (A)
0.900
0.978769
Space group
1
P2 2
1
2
1
1
P2 2
1
2
1
P2
1
2
1
2
1
1 1 1
P2 2 2
Cell parameters
˚
a, b, c (A)
45.86, 55.45, 156.27
46.13, 55.73, 156.24
46.13, 55.78, 156.32
46.13, 55.79, 156.35
ꢀ
a 5 b 5 c ( )
Molecules/asymmetric unit
Resolution range (A)
90
2
90
2
90
2
90
2
˚
50.00–1.53
50.00–2.30
50.00–2.30
50.00–2.30
a
a
a
a
(
1.56–1.53)
392,694
60,179
(2.34–2.30)
(2.34–2.30)
(2.34–2.30)
Reflections measured
Unique reflections
Redundancy
123,978
17,756
122,975
17,643
121,933
17,581
a
a
a
a
6.5 (7.2)
7.0 (7.3)
7.0 (7.3)
6.9 (7.3)
a
a
a
a
Completeness (%)
98.5 (100.0)
94.6 (99.7)
11.5 (33.4)
23.3 (5.0)
93.9 (99.5)
9.5 (36.5)
23.5 (4.5)
93.6 (99.5)
b
a
a
a
a
R
merge (%)
18.1 (28.3)
10.3 (42.2)
a
a
a
a
Average I/r (I)
14.5 (3.6)
22.2 (3.8)
Refinement statistics
˚
Resolution range (A)
78.14–1.53
No. of atoms
Protein/Water
2992/451
19.1
22.5
R
work (%)
c
Rfree (%)
Rms deviations from ideal values
˚
Bond lengths (A)
0.014
1.439
ꢀ
Bond angles ( )
Average B factors (A )
Protein/Water
˚
2
18.3/31.6
Ramachandran plot statistics
Favored region (%)
Allowed region (%)
97.1
2.9
a
Values in parentheses are for the highest-resolution shell.
P
P
b
R
merge
5
|Ihkl — <Ihkl>|/
Ihkl, where Ihkl is an intensity measurement for reflection with indices hkl and <Ihkl> is
the mean intensity for multiply recorded reflections.
c
Free R-value was calculated using 5% of the total reflections chosen randomly and omitted from refinement.
Okano et al.
PROTEIN SCIENCE VOL 24:93—104 97

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 repre-
sent N- and C-termini.
structure consists of two molecules (molecular A and
B). Molecular A contains 193 of 207 residues, with
Arg361-Met366 and Ala503-Lys510 missing, prob-
ably due to structural disorder. Molecular B contains
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 second-
ary structures of LC-Est1C* are similar to those of
the esterase domain of Tm-EstA, except for loops
between b4-strand and aB-helix and between b8-
strand and aF-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 partici-
pates in the formation of a tunnel at the active site
1
91 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 A for
1
91 Ca atoms. We used the structure of molecule A
in this study.
The structure of LC-Est1C* shows a typical a/b
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)
2
0
and affects enzymatic activity. 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 h7-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 signifi-
cantly affect the structure of the esterase domain of
LC-Est1. However, because LC-Est1C* may lack b1
strand of the esterase domain, we used LC-Est1C
for biochemical characterization.
˚
with the RMSD value of 1.25 A for 191 Ca atoms.
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 maxi-
mal activity toward the pNP-acetate (C2) and pNP-
caprylate (C8) substrates and very weak activity
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. C2ꢁC14 repre-
sent the acyl chain lengths of the substrates. The experiment
was carried out in duplicate. Each value represents the aver-
age value and errors from the average values are shown.
9
8
PROTEINSCIENCE.ORG
Characterization of Metagenome-Derived Esterase

ꢀ
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 trun-
cation of N-terminal 20 residues does not signifi-
cantly 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.
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-
Figure 6. Optimum temperatures for activities of LC-Est1
and LC-Est1C. The temperature dependencies of the enzy-
matic 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.
ꢀ
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
toward the substrates with acyl chain lengths of
ꢂ10. 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 sub-
strate. 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
m
Table III. The K value of LC-Est1C is slightly
higher than but comparable to that of LC-Est1.
Whereas, the kcat value of LC-Est1C is lower than
that of LC-Est1 by roughly 60%. These results sug-
gest that truncation of LNTE reduces the turnover
number of LC-Est1 without significantly affecting its
substrate binding affinity. Truncation of LNTE
21
lmol ) was lower than that of LC-Est1 (255 units
21
lmol
) by 60%, indicating that truncation of
LNTE considerably decreases the activity of
LC-Est1. The specific activity was defined as the
enzymatic activity per lmol 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,
21
21
it is 4.6 units mg for LC-Est1 and 3.7 units mg
for LC-Est1C.
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
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.
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
ꢀ
The activity was determined at 30 C and the pH values indi-
were analyzed at various pHs ranging from 4.0 to
cated 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
ꢀ
9
.0 and 30 C using pNP-butyrate (C4) as a substrate
(
Fig. 7). The pH dependencies and substrate specific-
ities of LC-Est1 and LC-Est1C were analyzed at
4
.0–6.0; open circle), 20 mM sodium phosphate (pH 6.0–8.0;
ꢀ
3
0 C, instead of the optimum temperature for activ-
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.
ꢀ
ity (40 C), because the stability of certain sub-
strates, such as pNP-acetate decreases as the
Okano et al.
PROTEIN SCIENCE VOL 24:93—104 99

a
Table III. Specific Activities, Kinetic Parameters, and T1/2 Values of LC-Est1 and LC-Est1C
b
Specific activity
(unit/lmol)
Relative
activity (%)
DT1/2
21
ꢀ
ꢀ
Protein
K
m
(mM)
k
cat (s
)
T
1/2 ( C)
( C)
LC-Est1
LC-Est1C
255 6 31
101 6 19
100
40
0.30 6 0.01
0.48 6 0.04
5.8 6 0.3
2.4 6 0.1
67.2 6 0.4
63.9 6 0.1
–
23.3
a
ꢀ
The enzymatic activity was determined at 30 C for 10 min, in 100 lL of 20 mM Tris-HCl (pH 7.5) containing 10% acetoni-
trile, 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, T1/2, was
determined by monitoring the change in the CD value at 222 nm.
b
DT1/2 5 T1/2(LC-Est1C) – T1/2(LC-Est1).
probably slightly alters the conformation of the
active site, in such a way that it is not optimum for
activity.
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 independ-
ent 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.
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 moni-
toring 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 con-
centration, pH, and rate of temperature increase
were seriously changed. The thermal denaturation
curves of LC-Est1 and LC-Est1C measured in
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
1
0 mM Tris-HCl (pH 7.0) are shown in Figure 8.
ꢀ
activity of Tm-EstAC (60 C) is higher than that of
The midpoints of the transition of these thermal
denaturation curves, T1/2, are summarized in Table
III. The T1/2 value of LC-Est1C is lower than that of
ꢀ
LC-Est1C* (40 C). Proteins are stabilized by the
combination of a variety of the factors, such as
21
increased number of ion pairs, increased number
ꢀ
LC-Est1 by 3.3 C, suggesting that LNTE contributes
2
2
of hydrogen bonds, increased number of proline
to the stabilization of LC-Est1, but not so
significantly.
2
3
residues in loop regions,
increased number of
25
2
4
disulfide bonds increased interior hydrophobicity,
26
and anchoring of C-terminal tail. Comparison of
Role of LNTE
Role of LNTE of LC-Est1 remains to be understood,
because it shows little amino acid sequence similar-
ity to any protein with known function and the crys-
tal 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
2
0
to oligomerize. Tm-EstA exists as a hexamer in
solution and is a highly thermostable enzyme with
ꢀ
the optimum temperature for activity of ꢂ95 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
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.
1
00 PROTEINSCIENCE.ORG
Characterization of Metagenome-Derived Esterase

Table IV. Structural Features of LC-Est1C* and
CATATGCCTTATCGCCTCTACGTC-3 for primer 3,
Tm-EstAC
0
0
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
LC-Est1C*
Tm-EstAC
No. of amino acids
Opt. Temp. for activity ( C)
Content of the residues (%)
Buried polar
199
40
222
60
ꢀ
a
0
attached to the 5 -termini of the genes encoding LC-
15.6
38.7
71.3
11
169
0
20.2
38.3
65.4
14
206
0
Est1, LC-Est1C, and LC-Est1C*, respectively. Pri-
mers 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 result-
ant DNA fragments were digested with NdeI and
EcoRI, and ligated into the NdeI-EcoRI sites of
pET28a.
Buried apolar
Buried apolar/buried total
˚
No. of ion pairs (ꢃ4.0 A)
No. of hydrogen bonds
No. of disulfide bond
No. of Pro in loop
10
12
a
Data from Ref. [22 for Tm-EstAC.
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).
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 resi-
dues 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 hydro-
phobicity of LC-Est1C* is rather higher than that of
Tm-EstAC, because the ratio of interior apolar resi-
dues to total interior residues is 71.3% for LC-
Est1C* and 65.4% for Tm-EstAC. These results sug-
gest that the difference in the number of ion pairs,
the number of hydrogen bonds, and interior hydro-
phobicity may at least partly account for the differ-
ence in stability between LC-Est1C* and Tm-EstAC.
Overproduction and purification
For overproduction of LC-Est1, LC-Est1C, and LC-
Est1C*, E. coli BL21-CodonPlus(DE3)-RP transform-
ants with the pET25b derivatives were grown at
ꢀ
3
7 C. When the absorbance of the culture at 600 nm
reached around 0.5, isopropyl-b-D-thio galactopyra-
noside was added to the culture medium and culti-
vation 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, dia-
lyzed against 20 mM phosphate buffer (pH 6.0), and
loaded onto a HiTrap SP HP column (GE Health-
care, Tokyo, Japan) equilibrated with the same
buffer. The protein was eluted from the column by
linearly increasing the NaCl concentration from 0 to
Materials and Methods
Cells, plasmids, and enzymes
E. coli BL21-CodonPlus(DE3)-RP was from Strata-
gene (La Jolla, CA). Plasmid pET25b was from
Novagen (Madison, WI). E. coli BL21-Codon-
Plus(DE3)-RP transformants were grown in Luria-
Bertani broth medium (10 g Tryptone; 5 g Yeast
1
M. The fractions containing the protein were col-
lected 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 con-
centration from 10 to 300 mM. The fractions con-
taining the protein were collected and dialyzed
against 10 mM Tris-HCl (pH 7.5).
2
extract; 10 g NaCl in 1 L H O) supplemented with
21
5
0 mg L ampicillin.
Plasmid construction
The pET25b derivatives for overproduction of LC-
Est1 (residues 26–510), LC-Est1C (residues 284–
5
10), and LC-Est1C* (residues 304–510) were con-
Selenomethionine-substituted LC-Est1C* was
overproduced using methionine-auxotrophic E. coli
27
structed by polymerase chain reaction (PCR). The
20
fosmid vector harboring the LC-Est1 gene
was
strain B834 (DE3) pLysS in a defined medium.
used as a template. The sequences of the PCR pri-
The purification procedures were the same as those
for the native enzyme.
0
mers are 5 -TCAGCGCATATGCAGGATGCCGGGC
0
0
AGGTG-3 for primer 1, 5 -AAGAACCATATGGAGG
The production level of the protein in E. coli
cells and the purity of the protein were analyzed by
0
0
CGCGCCGGGGGGAC-3 for primer 2, 5 -AACACG
Okano et al.
PROTEIN SCIENCE VOL 24:93—104 101

2
8
SDS-polyacrylamide gel electrophoresis
using a
Thermal denaturation
1
2% polyacrylamide gel, followed by staining with
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 pro-
tein was dissolved in 10 mM Tris-HCl (pH 7.0). The
protein concentration and optical path length were
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 absorb-
21
0.1 mg mL
and 2 mm, respectively. The rate of
2
1
ꢀ
21
ance of a 0.1% (1.0 mg mL ) solution at 280 nm is
.75 for LC-Est1, 0.98 for LC-Est1C, and 1.01 for
LC-Est1C*. These value was calculated by using
temperature increase was 1.0 C min . The thermal
denaturation processes of the proteins examined
were irreversible under this condition. The tempera-
ture of the midpoint of the transition, T1/2, was cal-
culated by curve fitting of the resultant CD values
versus temperature data on the basis of a least-
square analysis.
0
2
1
21
21
21
e 5 1526 M cm for tyrosine and 5225 M cm
for tryptophan at 280 nm.
2
9
Enzyme assay
The enzymatic activity was determined at the tem-
perature indicated in 100 lL of 20 mM Tris-HCl (pH
Crystallization
7
.5) containing 10% acetonitrile and 1 mM pNP
Prior to crystallization, native LC-Est1C* and
SeMet-labeled LC-Est1C* were dialyzed against
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
1
1
0 mM Tris-HCl (pH 7.0) and concentrated to
2
1
9 mg mL
using an ultrafiltration system Cen-
the absorption at 412 nm (A 12) with an absorption
4
tricon (Millipore, Billerica, MA). The crystalliza-
tion conditions were screened using crystallization
kits from Hampton Research (Aliso Viejo, CA;
Crystal Screens and Index) and Emerald BioStruc-
tures and Emerald BioSystems (Bainbridge Island,
WA) (Wizard I-IV) by the sitting-drop vapor-diffu-
sion method. Drop solutions were prepared by mix-
ing 1 lL each of protein and reservoir solutions
and equilibrated against a 100 lL reservoir solu-
tion. 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]
21
21
cm . The initial velocity
coefficient of 14,200 M
of the enzymatic reaction was determined by record-
ing a time-dependent increase in A412 values for 5
min by automatic spectrophotometer (Hitachi Spec-
trophotometer U-2810; Hitachi High-Technologies,
Tokyo, Japan). One unit of enzymatic activity was
defined as the amount of enzyme that produced 1
lmol of pNP per min. Specific activity was defined
as the enzymatic activity per lmol 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 rang-
ꢀ
ing from 20 to 80 C. For the analysis of pH
ꢀ
at 20 C. Furthermore, the crystallization condition
dependence of activity, the activity was determined
of SeMet-labeled LC-Est1C* was further optimized
and the better quality crystals were obtained after
two week with optimized reservoir solution [0.1M
ꢀ
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 phos-
phate (pH 6.0–8.0), and 20 mM Tris-HCl (pH 7.0–
Bis-Tris (pH 6.5), 0.18M sodium chloride, and 21%
ꢀ
(
w/v) polyethylene glycol 3,350] at 4 C.
9
.0).
For determination of the kinetic parameters,
pNP-butyrate (C4) was used as a substrate. The con-
X-ray diffraction data collection and structure
determination
centration of this substrate was varied from 0.25 to
6
.0 mM. Hydrolysis of this substrate by the enzyme
X-ray diffraction data set of native LC-Est1C* and
followed Michaeleis-Menten kinetics, and the kinetic
parameters were determined from the Lineweaver-
Burk plot.
SeMet-labeled LC-Est1C* for MAD phasing was
ꢀ
collected at 2173 C under a nitrogen stream at
beam line BL44XU in SPring-8 (Hyogo, Japan).
Processing of the image data was performed using
3
0
CD spectra measurement
the HKL2000 program suite.
The structure of
The far-UV (200–260 nm) CD spectrum of the pro-
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
ꢀ
tein was measured at 25 C on a J-725 spectropo-
larimeter (Japan Spectroscopic, Tokyo, Japan). The
3
1
32
protein was dissolved in 10 mM Tris-HCl (pH 7.0).
program suite
in the HKL2MAP interface.
2
1
The protein concentration was 0.1 mg mL
cell with an optical path length of 2 mm was used.
and a
Automated model building was conducted using
3
3
ArpWarp. Structural refinement was carried out
34
2
21
The mean residual ellipticity (h, deg cm dmol
)
by using REFMAC
of the CCP4 program and
3
5
was calculated using an average amino acid molecu-
lar mass of 110 Da.
COOT.
The statistics for data collection and
refinement are summarized in Table II. The
1
02 PROTEINSCIENCE.ORG
Characterization of Metagenome-Derived Esterase

figures were prepared using PyMol (http://www.
pymol.org).
metagenomic approach. Appl Environ Microbiol 78:
556–1562.
5. Nguyen TN, You DJ, Kanaya E, Koga Y, Kanaya S,
2013) Crystal structure of metagenome-derived LC9-
1
1
Protein data bank accession number: The
coordinates and structure factors for LC-Est1C*
have been deposited in the PDB under ID code
(
RNase H1 with atypical DEDN active site motif. FEBS
Lett 587:1418–1423.
3
WYD.
16. Nguyen TN, Angkawidjaja C, Kanaya E, Koga Y,
Takano K, Kanaya S (2012) Activity, stability, and
structure of metagenome-derived LC11-RNase H1, a
homolog of Sulfolobus tokodaii RNase H1. Protein Sci
Acknowledgments
The synchrotron radiation experiments were per-
formed at Osaka University beam line BL44XU at
SPring-8 with the approval of the Japan Synchro-
tron Radiation Research Institute (JASRI; Proposal
No. 2013B6813, 2014A6915). The authors thank Dr.
Y. Koga for helpful discussions. The authors do not
have a conflict of interest to declare.
2
1:553–561.
1
1
1
7. Nguyen TN, You DJ, Matsumoto H, Kanaya E, Koga Y,
Kanaya S (2013) Crystal structure of metagenome-
derived LC11-RNase H1 in complex with RNA/DNA
hybrid. J Struct Biol 182:144–154.
8. Sulaiman S, You D-J, Kanaya E, Koga Y, Kanaya S
(
2014) Crystal structure and thermodynamic and
kinetic stability of metagenome-derived LC-cutinase.
Biochemistry 53:1858–1869.
9. Challacombe JF, Eichorst SA, Hauser L, Land M, Xie
G, Kuske CR (2011) Biological consequences of ancient
gene acquisition and duplication in the large genome of
Candidatus Solibacter usitatus Ellin6076. PLoS One 6:
e24882.
References
1
. Holmquist
M
(2000) Alpha/Beta-hydrolase fold
enzymes: structures, functions and mechanisms. Curr
Protein Pept Sci 1:209–235.
2
. Ollis DL, Cheah E, Cygler M, Dijkstra B, Frolow F,
Franken SM, Harel M, Remington SJ, Silman I,
Schrag J, Sussman JL, Verschueren KHG, Goldman
A (1992) The a/b hydrolase fold. Protein Eng 5:197–
20. Levisson M, Sun L, Hendriks S, Swinkels P, Akveld
T, Bultema JB, Barendregt A, van den Heuvel RH,
Dijkstra BW, van der Oost J, Kengen SW (2009)
Crystal structure and biochemical properties of a
2
11.
novel
thermostable
esterase
containing
an
3
. L ꢀo pez-L ꢀo pez O, Cerd aꢀ n ME, Gonzalez-Siso MI (2014)
New extremophilic lipases and esterases from metage-
nomics. Curr Protein Pept Sci 15:445–455.
immunoglobulin-like domain. J Mol Biol 385:949–
962.
21. Karshikoff A, Ladenstein R (2001) Ion pairs and the
thermotolerance of proteins from hyperthermophiles: a
“traffic rule” for hot roads. Trends Biochem Sci 26:550–
556.
4
5
. Verger R (1997) ‘Interfacial activation’ of lipases, facts
and artifacts. Tibtech 15:32–38.
. Bornscheuer UT (2002) Microbial carboxyl esterases:
classification, properties and application in biocataly-
sis. FEMS Microbiol Rev 26:73–81.
22. Shirley BA, Stanssens P, Hahn U, Pace CN (1992) Con-
tribution of hydrogen bonding to the conformational
stability of ribonuclease T1. Biochemistry 31:725–732.
23. Watanabe K, Chishiro K, Kitamura K, Suzuki Y (1991)
Proline residues responsible for thermostability occur
with high frequency in the loop regions of an extremely
thermostable oligo-1,6-glucosidase from Bacillus ther-
6
. Jochens H, Hesseler M, Stiba K, Padhi SK,
Kazlauskas RJ, Bornscheuer UT (2011) Protein engi-
neering of a/b-hydrolase fold enzymes. Chembiochem
1
2:1508–1517.
7
8
. Steele HL, Jaeger KE, Daniel R, Streit WR (2009)
Advances in recovery of novel biocatalysts from meta-
genomes. J Mol Microbiol Biotechnol 16:25–37.
moglucosidasius KP1006.
J Biol Chem 266:24287–
24294.
. Tuffin M, Anderson D, Heath C, Cowan DA (2009)
Metagenomic gene discovery: how far have we
moved into novel sequence space? Biotechnol J 4:
24. Boutz DR, Cascio D, Whitelegge J, Perry LJ, Yeates
TO (2007) Discovery of a thermophilic protein complex
stabilized by topologically interlinked chains. J Mol
Biol 368:1332–1344.
1
671–1683.
9
. Uchiyama T, Miyazaki K (2009) Functional metage-
nomics for enzyme discovery: challenges to efficient
screening. Curr Opin Biotechnol 20:616–622.
25. Pace CN, Fu H, Fryar KL, Landua J, Trevino SR,
Shirley BA, Hendricks MM, Iimura S, Gajiwala K,
Scholtz JM, Grimsley GR (2011) Contribution of hydro-
phobic interactions to protein stability. J Mol Biol 408:
514–528.
1
0. Simon C, Daniel R (2011) Metagenomic analyses: past
and future trends. Appl Environ Microbiol 77:1153–
1
161.
26. You DJ, Chon H, Koga Y, Takano K, Kanaya S (2007)
Crystal structure of type 1 ribonuclease H from hyper-
thermophilic archaeon Sulfolobus tokodaii: role of argi-
nine 118 and C-terminal anchoring. Biochemistry 46:
11494–11503.
1
1. Iqbal HA, Feng Z, Brady SF (2012) Biocatalysts and
small molecule products from metagenomic studies.
Curr Opin Chem Biol 16:109–116.
1
1
2. Arpigny JL, Jaeger KE (1999) Bacterial lipolytic
enzymes: classification and properties. Biochem J 343:
27. Hendrickson WA, Horton JR, LeMaster DM. (1990)
Selenomethionyl proteins produced for analysis by mul-
tiwavelength anomalous diffraction (MAD): a vehicle
for direct determination of three-dimensional structure.
EMBO J 9:1665–1672.
1
77–183.
3. Kanaya E, Sakabe T, Nguyen NT, Koikeda S, Koga Y,
Takano K, Kanaya S (2010) Cloning of the RNase H
genes from a metagenomic DNA library: identification
of a new type 1 RNase H without a typical active-site
motif. J Appl Microbiol 109:974–983.
28. Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227:680–685.
1
4. Sulaiman S, Yamato S, Kanaya E, Kim JJ, Koga Y,
Takano K, Kanaya S (2011) Isolation of a novel cuti-
nase homolog with polyethylene terephthalate degrad-
29. Goodwin TW, Morton RA (1946) The spectrophotomet-
ric determination of tyrosine and tryptophan in pro-
teins. Biochem J 40:628–632.
ing activity from leaf-branch compost using
a
Okano et al.
PROTEIN SCIENCE VOL 24:93—104 103

3
0. Otwinowski Z, Minor W (1997) Processing of X-ray dif-
fraction data collected in oscillation mode. Methods
Enzymol 276:307–326.
1. Sheldrick GM (2008) A short history of SHELX. Acta
Crystallogr A 64:112–122.
2. Pape T, Schneider TR (2004) HKL2MAP: a graphical
user interface for macromolecular phasing with
SHELX programs. J Appl Crystallogr 37:843–844.
3. Langer G, Cohen SX, Lamzin VS, Perrakis A (2008)
Automated macromolecular model building for X-ray
crystallography using ARP/wARP version 7. Nat Protoc
3:1171–1179.
3
4. Murshudov GN, Vagin AA, Dodson EJ (1997) Refine-
ment of macromolecular structures by the maximum-
likelihood method. Acta Crystallogr D Biol Crystallogr
3
3
5
3:240–255.
35. Emsley P, Cowtan K (2004) Coot: model-building tools
for molecular graphics. Acta Crystallogr D Biol Crys-
tallogr 60:2126–2132.
3
1
04 PROTEINSCIENCE.ORG
Characterization of Metagenome-Derived Esterase