Crystal structure of d -stereospecific amidohydrolase from Streptomyces sp. 82F2 - Insight into the structural factors for substrate specificity

Article abstract of DOI:10.1111/febs.13579

d-Stereospecific amidohydrolase (DAH) from Streptomyces sp. 82F2, which catalyzes amide bond formation from d-aminoacyl esters and l-amino acids (aminolysis), can be used to synthesize short peptides with a dl-configuration. We found that DAH can use 1,8-diaminooctane and other amino compounds as acyl acceptors in the aminolysis reaction. Low concentrations of 1,8-diaminooctane inhibited acyl-DAH intermediate formation. By contrast, excess 1,8-diaminooctane promoted aminolysis by DAH, producing d-Phe-1,8-diaminooctane via nucleophilic attack of the diamine on enzyme-bound d-Phe. To clarify the mechanism of substrate specificity and amide bond formation by DAH, the crystal structure of the enzyme that binds 1,8-diaminooctane was determined at a resolution of 1.49 ?. Comparison of the DAH crystal structure with those of other members of the S12 peptidase family indicated that the substrate specificity of DAH arises from its active site structure. The 1,8-diaminooctane molecule binds at the entrance of the active site pocket. The electrkon density map showed that another potential 1,8-diaminooctane binding site, probably with lower affinity, is present close to the active site. The enzyme kinetics and structural comparisons suggest that the location of enzyme-bound diamine can explain the inhibition of the acyl-enzyme intermediate formation, although the bound diamine is too far from the active site for aminolysis. Despite difficulty in locating the diamine binding site for aminolysis definitively, we propose that the excess diamine also binds at or near the second binding site to attack the acyl-enzyme intermediate during aminolysis. Database The coordinates and structure factors for d-stereospecific amidohydrolase have been deposited in the Protein Data Bank at the Research Collaboratory for Structural Bioinformatics under code: 3WWX. D-Stereospecific amidohydrolase (DAH) can use diamine as acyl acceptors in the aminolysis. Low concentrations of 1,8-diaminooctane inhibited acyl-DAH intermediate formation, but excess 1,8-diaminooctane promoted aminolysis. The structural analysis suggests that the location of enzyme-bound 1,8-diaminooctane can explain the inhibition of the acyl-enzyme intermediate formation. In addition, the excess 1,8-diaminooctane is considered to bind to attack the acyl-enzyme intermediate during aminolysis.

Full text of DOI:10.1111/febs.13579

Crystal structure of D-stereospecific amidohydrolase from
Streptomyces sp. 82F2 – insight into the structural factors
for substrate specificity
Jiro Arima1, Kana Shimone1, Kazusa Miyatani1, Yuka Tsunehara1, Yoshitaka Isoda2, Tomoya Hino3
and Shingo Nagano3
1 Department of Agricultural, Biological, and Environmental Sciences, Faculty of Agriculture, Tottori University, Japan
2 United Graduate School of Agricultural Sciences, Tottori University, Japan
3 Department of Chemistry and Biotechnology, Graduate School of Engineering, Tottori University, Japan
Keywords
D-Stereospecific amidohydrolase (DAH) from Streptomyces sp. 82F2, which
catalyzes amide bond formation from D-aminoacyl esters and L-amino
acyl acceptor; amide bond formation;
aminolysis; crystal structure;
acids (aminolysis), can be used to synthesize short peptides with a DL-con-
figuration. We found that DAH can use 1,8-diaminooctane and other
amino compounds as acyl acceptors in the aminolysis reaction. Low con-
D-stereospecific amidohydrolase
Correspondence
centrations of 1,8-diaminooctane inhibited acyl-DAH intermediate forma-
tion. By contrast, excess 1,8-diaminooctane promoted aminolysis by DAH,
producing D-Phe-1,8-diaminooctane via nucleophilic attack of the diamine
on enzyme-bound D-Phe. To clarify the mechanism of substrate specificity
and amide bond formation by DAH, the crystal structure of the enzyme
J. Arima, Department of Agricultural,
Biological, and Environmental Sciences,
Faculty of Agriculture, Tottori University,
Tottori 680-8553, Japan
Fax: +81 857 31 5347
Tel: +81 857 31 5363
ꢀ
that binds 1,8-diaminooctane was determined at a resolution of 1.49 A.
E-mail: arima@muses.tottori-u.ac.jp
and
Comparison of the DAH crystal structure with those of other members of
the S12 peptidase family indicated that the substrate specificity of DAH
arises from its active site structure. The 1,8-diaminooctane molecule binds
at the entrance of the active site pocket. The electrkon density map showed
that another potential 1,8-diaminooctane binding site, probably with lower
affinity, is present close to the active site. The enzyme kinetics and struc-
tural comparisons suggest that the location of enzyme-bound diamine can
explain the inhibition of the acyl-enzyme intermediate formation, although
the bound diamine is too far from the active site for aminolysis. Despite
difficulty in locating the diamine binding site for aminolysis definitively, we
propose that the excess diamine also binds at or near the second binding
site to attack the acyl-enzyme intermediate during aminolysis.
S. Nagano, Department of Chemistry and
Biotechnology, Graduate School of
Engineering, Tottori University, Tottori 680-
8552, Japan
Fax/Tel: +81 857 31 5273
E-mail: snagano@bio.tottori-u.ac.jp
(Received 13 August 2015, revised 17
October 2015, accepted 26 October 2015)
doi:10.1111/febs.13579
Database
The coordinates and structure factors for D-stereospecific amidohydrolase have been deposited in the
Protein Data Bank at the Research Collaboratory for Structural Bioinformatics under code: 3WWX.
Introduction
Peptidases play vital physiological roles through the
biosynthesis, degradation and regulation of peptides or
proteins. Rawlings et al. [1] classified 400 000 pepti-
dases into 55 clans and 244 families based on their
primary structures in the MEROPS database (http://
merops.sanger.ac.uk). Peptidases are also classified by
Abbreviations
DAA, D-amino acid amidase; DAH, D-stereospecific amidohydrolase; OBzl, benzyl ester; pNA, p-nitroanilide; UPLC, ultra-performance liquid
chromatography.
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
337

Structure of D-stereospecific amidohydrolase
J. Arima et al.
the nucleophile in the active site (e.g. as serine, thre-
onine, cysteine, aspartate, glutamate or metal pepti-
dases). Serine peptidases, which are classified into 53
families (S1–S81 in MEROPS), account for over 30%
of the sequences in MEROPS. Although many pepti-
dases cleave L-amino acid peptide bonds, some pepti-
dases prefer peptides consisting of D-amino acids as a
substrate. The database also contains enzymes that
can cleave aminoacyl amide and ester bonds.
their substrate specificities and the mechanism for
recognizing acyl acceptors for the aminolysis reaction
are not fully understood.
Recently, we found a new serine peptidase, D-stereo-
specific amidohydrolase (DAH), in Streptomyces sp.
82F2. The enzyme belongs to the S12 peptidase family
and has high aminolysis activity in addition to
hydrolyzing various D-aminoacyl and D,D-configuration
dipeptide derivatives [20]. In the DAH aminolysis reac-
tion, the enzyme preferentially uses D-aminoacyl
derivatives as acyl donors and L-amino acids and their
derivatives as acyl acceptors, producing dipeptides
with a DL-configuration [21]. Therefore, the enzyme is
of particular interest for synthesizing various peptides
incorporating D-amino acids. DAH has been used to
synthesize a cyclic dipeptide that inhibits family 18
chitinase, cyclo (D-Pro-L-Arg) in a one-step/one-pot
reaction.
The most studied active site in serine peptidase is
the classic Ser/His/Asp catalytic triad [2]. Variations in
the active site architectures, including the Ser/Ser/Lys
triad and the Ser/Lys dyad, have also been reported
[3–5]. In the catalytic mechanism of all known serine
peptidases, a covalent acyl-enzyme intermediate is gen-
erated and the substrate is an acyl donor. The interme-
diate is then hydrolyzed by the nucleophilic attack of
an activated water molecule. In the presence of rela-
tively high concentrations of primary amine, which are
not achieved under physiological conditions, nucle-
ophilic attack by an amino group instead of a water
molecule forms an amide bond (Fig. 1) [6–11]. In this
reaction, termed aminolysis, the primary amine acts as
an acyl acceptor. Because amino acids can act as acyl
acceptors, the unnatural aminolysis reaction of serine
peptidases has attracted attention as a method for syn-
thesizing various peptides that cannot be produced
under natural conditions.
In the present study, we showed that DAH can use
1,8-diaminooctane and other amino compounds as
acyl acceptors in the aminolysis reaction. We crystal-
lized this enzyme in the presence of 1,8-diaminooctane
and the crystal structure of the DAH complex with
1,8-diaminooctane was obtained at a resolution of
ꢀ
1.49 A. Comparative analysis with homologous
enzymes provided insights into structural factors for
the substrate specificity and aminolysis of DAH.
Several enzymes in the MEROPS peptidase database
belonging to the S12 family, which have a catalytic Ser/
Lys dyad, exhibit high aminolysis activity for various
types of substrates such as amides esters and peptides
[12–14]. Four crystal structures of S12 family peptidases
have been reported: D-Ala-D-Ala carboxypeptidase B
Results and Discussion
Diamino-alkane acts as an acyl acceptor
substrate
First, we assessed the effect of 1,8-diaminooctane on
the catalytic function of DAH. The p-nitroaniline
release activity of DAH for D-Phe-p-nitroanilide
(pNA) is accompanied by the acyl-enzyme intermedi-
ate formation (Fig. 2A). The intermediate is then dea-
cylated by the nucleophilic attack of activated water to
form D-Phe (Fig. 3A). If the amino group of 1,8-dia-
minooctane attacks the acyl-DAH intermediate instead
(D,D-peptidase), class
amidase (DAA) and D-stereospecific aminopeptidase
[15–18]. D,D-Peptidase and D-stereospecific
C b-lactamase, D-amino acid
aminopeptidase catalyze peptide bond formation by
their aminolysis activity [12–14]. Based on these crystal
structures, the structural factors responsible for the
substrate specificities have been identified [19]. However,
Fig. 1. Scheme of the hydrolysis and aminolysis reactions catalyzed by serine peptidases.
338
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS

J. Arima et al.
Structure of D-stereospecific amidohydrolase
Fig. 2. Effect of 1,8-diaminooctane on acyl-enzyme intermediate
formation of DAH. (A) Reaction scheme of the formation of the
Fig. 3. Effect of 1,8-diaminooctane on the hydrolysis and
aminolysis reactions catalyzed by DAH. (A) Reaction scheme after
the formation of the acyl-enzyme intermediate when D-Phe-pNA
was used a substrate. D-Phe is the product of the nucleophilic
attack of water on the carbonyl carbon of the acyl-enzyme
intermediate. By contrast, 1,8-diaminooctane attacks the acyl-
enzyme intermediate to produce D-Phe-1,8-diaminooctane. (B, C)
Extracted ion chromatograms of the products of the DAH reaction,
D-Phe and D-Phe-1,8-diaminooctane, using 0.8 mM D-Phe-pNA for
120 min. Mass spectra data: D-Phe, calculated for C9H11NO2
[M + H]+: 166.0869, found: 166.0851–166.0882. D-Phe-1,8-
diaminooctane, calculated for C17H29N3O [M + H]+: 292.2391,
found: 292.2385–292.2413.
acyl-enzyme intermediate. p-Nitroaniline is
a product of the
formation of the acyl-enzyme intermediate when the D-Phe-pNA
substrate is used. (B) Effect of 1,8-diaminooctane on the acyl-
enzyme intermediate formation. The release of p-nitroaniline from
D-Phe-pNA associated with the acyl-enzyme intermediate formation
was measured every minute. Each value represents the
mean Æ SD of values from four independent experiments.
of a water molecule, (R)-2-amino-N-(8-aminooctyl)-3-
phenylpropanamide (D-Phe-1,8-diaminooctane) should
be formed (Fig. 3A).
As shown in Fig. 2B, the release of p-nitroaniline by
DAH (acyl-enzyme intermediate formation) decreased
as the 1,8-diaminooctane concentration increased to
~ 5 mM. However, in the presence of more than
10 mM 1,8-diaminooctane, p-nitroaniline release by
DAH gradually increased with the 1,8-diaminooctane
concentration (Fig. 2B). These results indicate that
1,8-diaminooctane up to a concentration of ~ 5 mM
inhibits acyl-enzyme intermediate formation, whereas
higher 1,8-diaminooctane concentrations rescue the
acyl-enzyme intermediate formation.
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
339

Structure of D-stereospecific amidohydrolase
J. Arima et al.
Next, we examined the reaction after the acyl-enzyme
intermediate formation. Figure 3B shows that the
production of D-Phe, which is a hydrolysis product of
the acyl-enzyme intermediate, was retarded in the
presence of 5 mM 1,8-diaminooctane, which is consistent
with the observation that up to ~ 5 mM 1,8-
diaminooctane suppressed the acyl-enzyme intermediate
formation (Fig. 2B). The addition of 160 mM 1,8-
diaminooctane increased the formation of D-Phe, which
is also consistent with the excess of 1,8-diaminooctane
rescuing p-nitroaniline formation. Although acyl-enzyme
intermediate formation is retarded by up to 5 mM 1,8-
diaminooctane, D-Phe-1,8-diaminooctane, which is
produced by aminolysis of the intermediate by 1,8-
diaminooctane, was detected under the same conditions.
Further addition of 1,8-diaminooctane markedly
increased the aminolysis reaction (Fig. 3C). Because
amount of 1,8-diaminooctane. The increased amount of
acyl-enzyme intermediate was competitively consumed
by deacylation with a water molecule and aminolysis
with 1,8-diaminooctane.
The addition of 1,8-diaminooctane, other alkylami-
nes, amino alcohols and L-amino acids to the enzyme
reaction mixture also results in an aminolysis reaction
(Table 1), indicating that DAH recognizes these amine
compounds as acyl acceptor substrates. By contrast,
DAH did not recognize D-amino acids, secondary and
tertiary amine compounds and some amines, including
Tris and amino sugars, as acyl acceptor substrates
(data not shown).
Next, we investigated the enzyme kinetics of the
aminolysis reaction with several acyl donor and accep-
tor substrates. The Km value for D-Phe-pNA could not
be determined because of its low solubility (< 0.9 mM
at pH 8.5). Therefore, D-Phe-benzyl ester (OBzl) and
1,8-diaminooctane were used as acyl donor and
acceptor substrates, respectively. The Km value for D-
Phe-OBzl hydrolysis (0.036 mM) was determined in a
previous study [21]. As shown in Table 2, the apparent
Km values for D-Phe-OBzl and 1,8-diaminooctane for
160 mM
1,8-diaminooctane
rescued
acyl-enzyme
intermediate formation, D-Phe formation was also
slightly rescued. These observations of the acyl-
intermediate formation and deacylation of the
intermediate suggest that acyl-enzyme intermediate
formation was facilitated by the presence of a large
Table 1. Recognition of amine compounds as acyl acceptor substrates for aminolysis. Recognition of the acyl acceptor was evaluated by
UPLC ESI-TOF-MS after the enzyme reaction with 20 mM D-Phe-OBzl, 200 mM additive and 0.2 M Tris-HCl (pH 8.5). Mass spectra data of
the D-phenylalanyl derivatives are listed.
Recognition
Chemicals
as acyl acceptor
Aminolysis product (formula)
Calculated [M + H]+
Observed
a
1,3-Diaminopropane
1,4-Diaminobutane
1,5-Diaminopentane
1,6-Diaminohexane
1,7-Diaminoheptane
1,8-Diaminooctane
1,10-Diaminodecane
Ethylamine
Æ
D-phenylalanyl 1,3-diaminopropane (C12H19N3O)
D-phenylalanyl 1,4-diaminobutane, (C13H21N3O)
D-phenylalanyl 1,5-diaminopentane (C14H23N3O)
D-phenylalanyl 1,6-diaminohexane (C15H25N3O)
D-phenylalanyl 1,7-diaminoheptane (C16H27N3O)
D-phenylalanyl 1,8-diaminooctane (C17H29N3O)
D-phenylalanyl 1,10-diaminodecane (C19H33N3O)
D-phenylalanyl ethylamine (C11H16N2O)
222.1608
236.1765
250.1921
264.2078
278.2234
292.2391
320.2704
193.1342
207.1499
222.1648
236.1794
250.1930
264.2054
278.2271
292.2397
320.2705
193.1342
207.1500
+
+
+
+
+
+
a
Æ
Propylamine
+
D-phenylalanyl propylamine (C12H18N2O)
b
Isopropylamine
Butylamine
–
+
+
+
D-phenylalanyl butylamine (C13H20N2O)
D-phenylalanyl hexylamine (C15H24N2O)
D-phenylalanyl 1-aminodecane (C19H32N2O)
D-phenylalanyl 2-aminoethanol (C11H16N2O2)
D-phenylalanyl 4-amino-1-butanol (C13H20N2O2)
D-phenylalanyl 6-amino-1-hexanol (C15H24N2O2)
D-phenylalanyl 10-amino-1-decanol (C15H24N2O2)
D-Phe-Gly
221.1655
249.1969
305.2595
209.1291
237.1604
265.1918
321.2544
221.1687
249.1959
305.2599
209.1286
237.1635
265.1917
321.2547
Hexylamine
1-Aminodecane
2-Aminoethanol
4-Amino-1-butanol
6-Amino-1-hexanol
10-Amino-1-decanol
Gly
a
Æ
+
+
+
c
+
c
L-Ala
+
D-Phe-L-Ala
c
L-Lys
+
D-Phe-L-Lys
c
L-Leu
+
D-Phe-L-Leu
c
L-Trp
+
D-Phe-L-Trp
a
Æ, Low amount of product.
–, Not detectable.
b
c
+, Refer to Arima et al. [21].
340
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS

J. Arima et al.
Structure of D-stereospecific amidohydrolase
Table 2. Km values of DAH for hydrolysis and amide bond
formation activity. Km values were calculated from a nonlinear
regression fit to the Michaelis–Menten equation using initial
estimates from double-reciprocal plots.
In the structure of the catalytic center of DAH,
Ser86 Oc forms a hydrogen bond with the backbone
amide of Gly337, which makes the hydroxyl group
point further away from Lys89. Lys89 Nf forms
hydrogen bonds with Tyr 191 Og, Asn193 Od and the
backbone carbonyl oxygen of Ala270 (Fig. 4C). The
interaction decreases the pKa of general base Tyr to
form a phenolate anion that activates the nucleophilic
Ser [23,24].
Reaction
Substrate
Km (mM)
D-Phe-OBzl hydrolysis
D-Phe-OBzl
0.036a
14.3
D-Phe-1,8-diaminooctane synthesis D-Phe-OBzl
1,8-Diaminooctane 626
a
Value given in Kato et al. [13].
Substantially elongated electron densities in the
Fo À Fc and 2Fo À Fc simulated annealing omit maps
were observed at the entrance of the active site pocket
(Fig. 4D). The length of this electron density is com-
parable to 1,8-diaminooctane, which is contained in
the crystallization solution, and no other compounds
are consistent with this electron density shape. There-
fore, we assigned this electron density as 1,8-diami-
nooctane. The distance between the active site Ser
residue and the bound 1,8-diaminooctane molecule
D-Phe-1,8-diaminooctane synthesis were 14.3 and
626 mM, respectively. The substantially larger D-Phe-
OBzl Km value for aminolysis than for hydrolysis
implies that 1,8-diaminooctane inhibits D-Phe-OBzl
binding to the enzyme.
Structure of DAH
ꢀ
was 7.3 A (Fig. 4E). The binding model shows that
Because 1,8-diaminooctane and other compounds
containing an amino group accept an acyl group
and undergo aminolysis by DAH, we crystallized
DAH in the presence of amino compounds. The
addition of 3% 1,8-diaminooctane to DAH produced
pyramidal crystals. Although several types of crystals
were obtained in the presence of alkylamines (data
not shown), the quality of these crystals was mostly
the amino group of 1,8-diaminooctane forms hydrogen
bonds with the backbone carbonyl of Phe155 and a
water molecule that is hydrogen bonded to another
water molecule and Asp362 Od. The other amino
group of 1,8-diaminooctane forms a hydrogen bond
with Ser342 Oc (Fig. 4D). In addition, Phe155 is
involved in a hydrophobic interaction with the alkyl
chain of 1,8-diaminooctane. We also found another
elongated Fo À Fc simulated annealing omit map close
to the active site. This is another potential 1,8-diami-
nooctane binding site because the shape of the Fo À Fc
electron density is similar to that for the other binding
site. In addition, one end of the density is close to the
Ser367 and Ser343 side chains, which enable hydro-
gen-bonding interactions with the diamine. However,
because the 2Fo À Fc simulated annealing omit map is
poor, probably as a result of low affinity and low
occupancy, the diamine cannot be assigned unambigu-
ously, and thus only the diamine bound to the former
binding site was included in the final stage of the
refinement.
poor. We determined the crystal structure of DAH
ꢀ
at
crystals.
The overall structure of the enzyme is presented in
a
resolution of 1.49 A using the pyramidal
Fig. 4A. The enzyme consists of a-rich and b-rich
regions. The b-rich region is formed by residues 38–86
and 224–379, and consists of six main antiparallel b-
strands flanked by four a-helices. The a-helix rich
region (residues 87–223) includes another 11 a-helices
(Fig. 4A).
There are two motifs, Ser-x-x-Lys and Tyr-x-Asn,
which are necessary for the expression of the enzyme
activity [15,22] in the N-terminal (Ser86–Lys89) and
the internal region (Tyr191–Asn193) of DAH, respec-
tively (Fig. 5). Crystallographic analysis of the enzyme
revealed that DAH possesses a large cavity that leads
to the catalytic center. The side chains of Ser86 and
Lys89 that create a catalytic Ser/Lys dyad in S12 fam-
ily peptidases [15] were positioned at the center of the
large cavity (Fig. 4B). Tyr191, which is close to the
Ser/Lys dyad, acts as the general base to activate Ser
as a nucleophile for the acyl donor during the acyla-
tion step and the water molecule during hydrolysis
[22–24]. In addition, there is a small pocket close to
the catalytic center of Ser86, Ly89 and Tyr191
(Fig. 4B).
Selectivity of substrate length and bulkiness
DAH prefers D-aminoacyl derivatives and D,D-config-
uration dipeptide derivatives with large, hydrophobic
side chains as acyl donors [20,21]. By contrast to
DAH, D,D-peptidase and DAA recognize the C-term-
inal D-Ala of oligopeptides and D-aminoacyl amides
with bulky hydrophobic side chains, respectively, as
acyl donors. Structural comparison of these enzymes
with DAH revealed that the overall structure of DAH
is analogous to D,D-peptidase and DAA (rmsd values:
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
341

Structure of D-stereospecific amidohydrolase
J. Arima et al.
Fig. 4. Overall structure and local
structure around the active site Ser of
DAH. (A) Overall structure of DAH that
interacts with 1,8-diaminooctane. a-Helices
and b sheets are shown in red and yellow,
respectively. The bound 1,8-diaminooctane
and active site residues (Ser86, Lys89 and
Tyr191) are shown as a purple and green
ball and stick models, respectively. (B)
Surface model of DAH. The bound 1,8-
diaminooctane and active site residues
(Ser86, Lys89 and Tyr191) are shown as a
purple and green ball and stick model,
respectively. (C) Residues surrounding the
active site Ser86 and Lys89. Ser86 and
Lys89 are shown as a ball and stick model
according to the atom type. Other
residues are shown as sticks colored
according to the atom type. Hydrogen
bonds are represented as broken lines. (D)
Fo À Fc and 2Fo À Fc simulated annealing
omit electron density maps (countered at
3.0 and 1.0 r, respectively) are shown as
cyan and orange meshes, respectively.
The length of this electron density is
comparable to 1,8-diaminooctane.
Hydrogen bonds are represented as
broken lines. (E) Residues surrounding 1,8-
diaminooctane. Carbon and nitrogen atoms
of the bound 1,8-diaminooctane are purple
and blue, respectively. Residues are
shown as sticks colored according to the
atom type.
ꢀ
1.5 and 2.3 A for D,D-peptidase and DAA, respec-
(Fig. 6A), the enzyme has a large cavity and active site
pocket. Therefore, the large cavity in DAH allows the
peptide substrate to enter, and the large space in the
active site pocket accommodates the large side chain
of the acyl donor substrate.
tively) (Fig. 6, left). Of these enzymes, DAH possesses
the largest cavity leading to the catalytic center (Fig. 6,
middle). DAH and DAA have a larger pocket at the
bottom of the cavity than D,D-peptidase (Fig. 6, right).
Previous studies of D,D-peptidase and DAA have
shown that the large D,D-peptidase cavity accommo-
dates the peptide substrate, and the narrow space in
the DAA cavity limits the length of the substrate
[17,25,26]. By contrast, the DAA active site pocket is
larger than that of D,D-peptidase (Fig. 6B,C, right).
The larger active site pocket of DAA can recognize
and accommodate D-aminoacyl amides with bulky side
chains. The crystal structure of DAA complexed with
D-Phe reveals that the ligand fits snugly in the DAA
active site pocket, as described below. In DAH
Structural factors for recognizing the substrate
N-terminal amino acid
Among DAH, D,D-peptidase, and DAA, only D,D-pep-
tidase exhibits carboxypeptidase activity. The structure
of D,D-peptidase complexed with
a peptideglycan
mimic, b-sultam, suggests that Arg285 recognizes the
terminal carboxylate of the peptide substrate and
expresses the carboxypeptidase activity [27]. The termi-
nal carboxylate recognition by the Arg residue is also
342
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS

J. Arima et al.
Structure of D-stereospecific amidohydrolase
Fig. 5. Alignment of primary structures of DAH and S12 family peptidases. Multiple sequence alignment was performed with CLUSTAL
algorithm in CLUSTALW (http://www.genome.jp/tools/clustalw). Asterisks (*) denote conserved residues; colons (:) denote conserved
substitutions in all sequences and periods (.) denote semi-conserved substitutions in all sequences. The two motifs necessary for
expressing the enzyme activity, Ser-x-x-Lys and Tyr-x-Asn, are shown in bold. Proteins: bac_Dpep, alkaline D-peptidase from Bacillus cereus
(MEROPS ID: MER002489); DDpep, D,D-peptidase from Streptomyces sp. R61 (MEROPS ID: MER000459); DAA, D-amino acid amidase
from Ochrobactrum anthropi (MEROPS ID: MER005964); DAP, D-aminopeptidase from Ochrobactrum anthropi (MEROPS ID: 000457); MlrB,
microsystinase from Sphingomonas sp. ACM-3962 (MEROPS ID: MER017069); blaC, class C b-lactamase from Enterobacter cloacae
(MEROPS ID: MER000463).
supported by the fact that introducing the Arg residue
into a mutant D-aminopeptidase at the position equiv-
alent to Arg285 in D,D-peptidase increases D,D-carbox-
ypeptidase activity by 28-fold [19]. Figure 7A shows
superimposed active site structures of D,D-peptidase
complexed with b-sultam and DAH. The Arg residue
in D,D-peptidase is replaced with Met in DAH. There-
fore, DAH exhibits no carboxypeptidase activity.
DAA also does not have the Arg residing at the corre-
sponding position and exhibits no carboxypeptidase
activity.
structure of DAH shows that the pocket of this
enzyme is sufficiently large to accommodate D-Phe.
The active site Ser is found at similar positions for
DAH and DAA, and many hydrophobic side chains,
Phe150, Val154, Phe155, Ile266, Ile338 and Tyr144,
line the pocket of DAH (Fig. 7C,D). Furthermore, the
backbone NH and carbonyl of Gly337 can assist D-
Phe binding in the pocket of DAH. Therefore, the
binding mode of D-Phe in DAH would be very similar
to that found in DAA. Consequently, the interactions
of the bound D-amino acid with backbone NH and
carbonyl would be important for the acyl-enzyme
intermediate formation for peptidases using D-amino
acid as an acyl donor.
D,D-Peptidase, DAA and DAH recognize an amide
bond involving D-amino acid residues [28–33]. The
crystal structure of the DAA-D-Phe acyl-enzyme inter-
mediate (Fig. 7B) revealed that D-Phe (shown in
magenta) fits into the active site pocket. The amino
group of the bound D-Phe is hydrogen bonded with
the backbone carbonyl of Gln310, the Glu114 side
chain and a water molecule, which forms a hydrogen
bond with Gln310. The carbonyl group of the ligand
also forms a hydrogen bond with the backbone NH of
Gln310. In addition, the phenyl ring of the ligand is
involved in hydrophobic interactions with Phe113,
Met117, Trp215, Phe234 and Ile311 (Fig. 7B).
Effects of 1,8-diaminooctane on the DAH reaction
The release of p-nitroaniline, which is equivalent to the
acyl-enzyme intermediate formation, is retarded by
1,8-diaminooctane concentrations up to 5 mM (Fig. 3).
However, in the presence of high concentrations of
1,8-diaminooctane, p-nitroaniline release is reactivated.
The formation of the aminolysis and hydrolysis prod-
ucts also increases in the presence of high concentra-
tions of 1,8-diaminooctane. The D-Phe-OBzl Km value
for the hydrolysis is substantially smaller than that for
DAH can use D-amino acids as an acyl donor to
produce an acyl-enzyme intermediate. The crystal
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
343

Structure of D-stereospecific amidohydrolase
J. Arima et al.
Fig. 6. Comparison of the folds and cavity
shapes of (A) DAH, (B) D,D-peptidase
(Protein Data Bank code: 3PTE) [15] and
(C) DAA (Protein Data Bank code: 2EFX)
[17]. Corresponding ribbon models are
shown on the left with a-helices and b
sheets in red and yellow, respectively.
Active site Ser residues of the enzymes
are shown as ball and stick models
colored according to the atom type. The
hydrophobic surfaces are also shown
(middle). Active site Ser residues of the
enzymes are shown as ball and stick
models colored according to the atom
type. Corresponding cross-sectional views
are shown on the right. The structures of
respective enzymes are superimposed on
the model of DAH, and then the cross-
sectional view of respective enzymes with
the same angle is created. The enzyme
molecules are shown as hydrophobic
surfaces.
aminolysis in the presence of 1,8-diaminooctane. The
crystal structures of 1,8-diaminooctane-bound DAH
and an inhibitor-bound D,D-peptidase provide a sce-
nario that explains these observations. The crystal
structure of DAH revealed that the active site pocket
of this enzyme is narrowed by the diamine, and
thereby the entry of the acyl donor is inhibited by the
presence of 1,8-diaminooctane. The Fo À Fc simulated
annealing omit electron density map implies that the
diamine could bind to the second site, which is close
to the active site Ser. The diffuse 2Fo À Fc electron
density of the second site suggests that the affinity for
the diamine is relatively low. It has been reported that
D,D-peptidase is completely inhibited by a peptidogly-
can mimic [34]. The peptidoglycan mimic covers the
active site pocket, and thus the ligand completely
blocks the access of the acyl donor to the active site
Ser. Comparing 1,8-diaminooctane-bound DAH with
D,D-peptidase inhibited by the peptidoglycan mimic
indicated that 1,8-diaminooctane overlaps with the
Gly-L-diaminopimelic acid moiety of the peptideglycan
mimic. However, the active site pocket and the Ser86
side chain in DAH are accessible for an acyl donor,
even when the potential second diamine binding site is
occupied by 1,8-diaminooctane, unlike the peptidogly-
can mimic-bound D,D-peptidase. Therefore, the crystal-
lographic data can explain the partial inhibition of
acylation of the enzyme. Aminolysis, which is deacyla-
tion of the acyl-enzyme intermediate by the diamine, is
hindered by very high concentrations of water
(> 5500 mM), which consumes the acyl-enzyme inter-
mediate by hydrolysis. Addition of a large amount of
1,8-diaminooctane leads to the deacylation of the acyl-
enzyme intermediate by the diamine, which is probably
located close to the enzyme-bound acyl group and
Ser86. The increased deacylation rescues acyl-enzyme
intermediate formation and the intermediate is com-
petitively attacked by water molecules and the dia-
mine. The 1,8-diaminooctane in the crystal structure is
unlikely to act as an acyl acceptor because the amino
groups of 1,8-diaminooctane are too far (12.8 and
c
ꢀ
8.0 A) from the active site Ser O . Based on the crystal
structure, it is difficult to identify unambiguously the
diamine binding site where 1,8-diaminooctane binds
and attacks the acyl-enzyme intermediate during
aminolysis. However, there is a possible binding site
that probably has lower affinity for the diamine close
to the second diamine binding site (Fig. 7C,D). If 1,8-
diaminooctane bound to the second site, one of the
diamine amino groups would form a hydrogen bond
with Ser367 and/or Ser343, pointing the other amino
group toward the acyl group of the intermediate,
344
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS

J. Arima et al.
Structure of D-stereospecific amidohydrolase
Fig. 7. Comparison of the active sites of DAH, D,D-peptidase and DAA. (A) Superimposed ligand binding modes of DAH and D,D-peptidase.
The main chains of D,D-peptidase binding a peptideglycan mimic (Protein data bank code: 1PW1) [34] and 1,8-diaminooctane-bound DAH are
shown as magenta and green loop models, respectively. The DAH ligand, 1,8-diaminooctane, and the D,D-peptidase peptide glycan mimic
ligand are shown as stick models. The molecular surface of D,D-peptidase is shown as magenta transparent surface. (B) Active site structure
of D-Phe-bound DAA. Main chain and side chain carbon atoms of DAA binding D-Phe (Protein data bank code: 2DNS) is shown in yellow
[17]. The DAA ligand, D-Phe, is shown in magenta. The molecular surface of DAA is shown as a yellow transparent surface. (C) Overlaid
active site structures of DAH and D-Phe-bound DAA. Main chain and side chain carbon atoms of DAH are shown are shown in green.
Fo À Fc and 2Fo À Fc simulated annealing omit electron density maps (countered at 3.0 and 1.0 r, respectively) are shown as cyan and
orange meshes, respectively. The molecular surface of DAH is shown as a light green transparent surface. (D) Superimposed active site
structures of DAH and DAA as shown in (C) viewed from different angle.
allowing the diamine to attack. Although DAH can
use L-amino acids as acyl acceptors, free D-amino acids
are very poor acyl acceptors for DAH. Therefore, both
the acyl group at the active site and the enzyme sur-
face around the acyl group would be required for
proper binding of 1,8-diaminooctane to accept the acyl
group. To clarify the recognition mechanisms for the
acyl acceptor in DAH aminolysis, further crystallo-
graphic studies are required, such as crystallization of
DAH bound to a stable acyl group analogue in the
presence of 1,8-diaminooctane.
D,D-peptidase, which is expected to be necessary for
the carboxypeptidase activity as shown by a previous
mutagenesis study, is replaced by Met321 in DAH.
This structural feature is probably responsible for the
substrate specificity of DAH. Because 1,8-diaminooc-
tane inhibits acyl-enzyme intermediate formation, the
1,8-diaminooctane-bound DAH structure can explain
the inhibition mode of intermediate formation. The
excess 1,8-diaminooctane acts as an acyl acceptor sub-
strate in DAH aminolysis. Although we identified a
potential binding mode of the diamine that attacks the
acyl-enzyme intermediate, the exact location of the
diamine binding site for aminolysis should be deter-
mined by X-ray crystallography to shed further light
on the aminolysis mechanism.
Conclusions
We solved the crystal structure of DAH complexed
with 1,8-diaminooctane. The overall enzyme structure
is analogous to D,D-peptidase and DAA. Comparing
the structures indicated that DAH possesses the largest
cavity leading to the catalytic center. By contrast, the
arrangement of the residues in the active site pocket is
similar to that of DAA complexed with D-Phe. In
addition, the residue corresponding to Arg285 of
Materials and methods
Expression and purification
We cultivated Escherichia coli Rosetta (DE3) harboring the
expression vector for DAH from Streptomyces sp. 82F2,
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
345

Structure of D-stereospecific amidohydrolase
J. Arima et al.
Table 3. Data collection and refinement statistics for the crystal
structure of DAH. Rmerge = Σ(|Ii À <I>|)/Σ|Ii|, where Ii is the intensity
of an observation, and <I> is the mean value for that reflection,
and summations are over all reflections. Rfactor = Σ||Fo|À|Fc||/Σ|Fo|,
where Fo and Fc are the observed and calculated structure factors,
respectively. Rfree factor was calculated with 5% of the data.
pET–82F2DAP [21] at 25 °C for 48 h in Overnight Expres-
sion Instant TB medium (50 mL; Novagen Inc., Madison,
WI, USA) in 10 test tubes (5 mL of medium in each tube).
The culture was centrifuged to remove the cells, and the
recombinant enzyme was purified. The culture supernatant
was dialyzed against 20 mM sodium acetate (pH 5.5). The
dialysate was loaded onto a spin column (Vivapure-S;
Sartorius, Gottingen, Germany) equilibrated with 20 mM
sodium acetate (pH 5.5). The bound protein was washed
with sodium acetate buffer containing 0.1 M NaCl, and
eluted with sodium acetate buffer containing 0.4 M NaCl.
The eluate was pooled and dialyzed against 20 mM Tris-
HCl (pH 8.0) containing 0.5 M NaCl. Then, it was concen-
trated to more than 10 mgÀ1ÁmLÀ1 with a centrifugal filter
(Amicon Ultra; Millipore Corp., Billerica, MA, USA). The
homogeneity of the purified proteins was confirmed by
12% SDS/PAGE under denaturing conditions [35].
Crystal cell parameter and data collection
Unit cell parameter
ꢀ
a, b, c (A)
59.29, 74.68, 76.45
Space group
P212121
37 394
2.26
Relative molecular mass
ꢀ
Matthews coefficient (A /Da)
3
Solvent content (%)
45.7
Collection and reduction
ꢀ
Wavelength (A)
1.000
ꢀ
Resolution limit (A)
1.49
Number of total reflections
321 232
Number of unique reflections
54 051
Completeness (%)
96.3 (75.2)a
20.0 (3.68)a
5.3 (24.4)a
0.990 (0.928)a
14.9
I/r
Crystallization
Rmerge (%)
CC1/2
DAH crystals were grown at 20 °C through sitting drop
vapor diffusion by mixing a protein solution (17 mgÁmLÀ1
protein in 20 mM Tris-HCl, pH 8.0, with 0.5 M NaCl) and
a reservoir solution (1.5 M citrate, pH 6.0). Although sev-
eral types of crystals were obtained in the presence of alky-
lamines (3% 1,6-diaminohexane, 1,7-diaminoheptane, and
6-amino-1-hexanol) during additive screening, the quality
of these crystals was mostly poor. Only the pyramidal crys-
tals obtained in the presence of 1,8-diaminooctane were of
sufficient quality for diffraction experiments.
2
ꢀ
Wilson B-factor (A )
Refinement
ꢀ
Resolution range (A)
29.5–1.49
14.1
Rwork (%)
Rfree (%)
ꢀ
rmsd (bond length) (A)
16.2
0.009
rmsd (bond angle) (°)
1.32
Number of protein residues
340 of 349
1,8-diaminooctane
Chemical component
2
ꢀ
Average Β-factor (A )
All atoms
17.4
15.4
31.4
Protein atoms
1,8-diaminooctane
Ramachandran plotb
Favored (%)
Data collection and structure determination
The crystals were harvested in solution containing 1.5 M
citrate (pH 6.0), 3% diamino-alkane and 20% glycerol, and
were flash-cooled in a nitrogen stream at 100 K. DAH
crystallized in the P212121 space group, with unit cell
97.3
2.1
Allowed (%)
Outliers (%)
0.6
a
ꢀ
ꢀ
ꢀ
dimensions: a = 59.29 A, b = 74.68 A and c = 76.45 A.
Diffraction data was collected at SPring-8 BL26B1
(Hyogo, Japan). The data were indexed, scaled, merged
and reduced with HKL2000 [36] and IMOSFLM packages [37].
The initial phases were obtained by molecular replacement
with PHASER [38] using the structure of D,D-peptidase (Pro-
tein data bank code: 1CEF; 37% sequence identity) as a
search model.
Values in parenthesis are for the highest resolution shell.
Ramachandran plot was calculated using MOLPROBITY [41].
b
also observed another elongated 2Fo À Fc simulated
annealing omit map close to the active site. Because the
shape of this density is similar to that of the other binding
site, this is another potential 1,8-diaminooctane binding
site. However, the 2Fo À Fc simulated annealing omit maps
of this site is weak, probably as a result of the low affinity
and low occupancy of the ligand. Thus, we did not add the
second 1,8-diaminooctane to the structural model. A final
round of refinement was performed using a model with
only one 1,8-diaminooctane molecule. Coordinates and
structure factors for DAH have been deposited in the Pro-
tein Data Bank at the Research Collaboratory for Struc-
tural Bioinformatics under code 3WWX. Figures 4–6 were
ꢀ
The structure was refined by PHENIX [39] against 1.5 A
diffraction data. The final atomic model contained 341
amino acid residues. The data collection and refinement
statistics for the crystal structure of DAH are presented in
Table 3. The asymmetric unit contained one protomer. The
Fo À Fc and 2Fo À Fc simulated annealing omit maps sug-
gested that a long ligand bound at the entrance of the
active site pocket. Thus, 1,8-diaminooctane was fitted into
the elongated electron density in the structural model. We
346
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS

J. Arima et al.
Structure of D-stereospecific amidohydrolase
prepared using UCSF CHIMERA [40] and PYMOL (The
PyMOL Molecular Graphics System, Version 1.7.4;
Premier XE; Waters Corp., Milford, MA, USA). Each
sample was eluted with solvent A–solvent B 95 : 5 for
2 min, solvent A–solvent B 80 : 20 for 1 min, solvent A–
solvent B gradient of 80 : 20 to 50 : 50 for 2 min, and sol-
vent A–solvent B of 20 : 80 for 2 min. Solvent A was
Milli-Q water containing 0.1% formic acid and solvent B
was acetonitrile containing 0.1% formic acid. Data were
processed using MASSLYNX (Waters Corp.).
€
Schrodinger, LLC, New York, NY, USA).
Effect of 1,8-diaminooctane on the acyl-enzyme
intermediate formation
The effect of the 1,8-diaminooctane concentration on the
acyl-enzyme intermediate formation of DAH was investi-
gated by
a continuous spectrophotometric assay with
Acknowledgements
0.8 mM D-Phe-pNA as the substrate. In the assay, DAH
(5 lL, 100 lgÁmLÀ1) was incubated with a mixture contain-
ing 0.3 M Tris-HCl (75 lL, pH 8.5) and various concentra-
tions of 1,8-diaminooctane (5–160 mM) for 120 min at
25 °C, followed by the addition of 4 mM D-Phe-pNA
(20 lL). The increased in A405 caused by the release of p-
nitroaniline associated with the acyl-enzyme intermediate
formation was monitored continuously and measured every
minute using a Microtiter plate reader (680; Bio-Rad Labo-
ratories Inc., Hercules, CA, USA). The initial activity rate
was determined from a linear part of the optical density
profile of p-nitroaniline measured with the same instru-
ment.
This work was supported in part by a Grant-in-Aid
for Scientific Research (C) from the Japan Society for
the Promotion of Science to J.A. (No. 26450124). The
evaluation of the acyl acceptor substrate was sup-
ported by the Naito Foundation through funding to
J.A.
Author contributions
J.A. and S.N. planned the project. J.A. performed
experiments depicted in Figs 1, 2, and 3. K.S. and
K.M. performed the purification and crystallization of
DAH. Y.T contributed experiments depicted in
Table 1. Y.I. contributed experiments depicted in
Table 2. T.H. and S.N. determined the crystal struc-
ture of DAH. J.A. and S.N. wrote the paper. All
authors discussed the results and commented on the
manuscript.
Aminolysis reaction using a diamino-alkane as an
acyl acceptor
An aminolysis reaction using a diamino-alkane as an acyl
acceptor was performed. 0.5 M D-Phe-OBzl (2 lL) dissolved
in dimethyl sulfoxide and 0.5 M diamino-alkane (20 lL)
were added to 0.5 M Tris-HCl (26 lL, pH 8.5). The reac-
tion was initiated by adding 0.2 mgÁmLÀ1 DAH solution
(2 lL) to the mixture. The reaction was continued at 4 °C
for 4 h, and was terminated by adding 0.5 M HCl (50 lL)
to the mixture. The reaction mixture was analyzed by MS.
To determine the apparent Km value for D-Phe-OBzl for
D-Phe-1,8-diaminooctane synthesis, the enzyme reaction
was conducted at different D-Phe-OBzl concentrations (5–
15 mM) in the presence of 0.2 M 1,8-diaminooctane at 4 °C
and pH 8.5 for 5 min. To determine the apparent Km value
for 1,8-diaminooctane for D-Phe-1,8-diaminooctane synthe-
sis, the enzyme reaction was performed at different 1,8-dia-
minooctane concentrations (50–150 mM) in the presence of
20 mM D-Phe-OBz at 4 °C and pH 8.5 for 5 min.
References
1 Rawlings ND, Waller M, Barrett AJ & Bateman A
(2014) MEROPS: the database of proteolytic enzymes,
their substrates and inhibitors. Nucleic Acids Res 42,
D503–D509.
2 Dodson G & Wlodawer A (1998) Catalytic triads and
their relatives. Trends Biochem Sci 23, 347–352.
3 Shin S, Yun YS, Koo HM, Kim YS, Choi KY & Oh
BH (2003) Characterization of a novel Ser-cisSer-Lys
catalytic triad in comparison with the classical Ser-His-
Asp triad. J Biol Chem 278, 24937–24943.
4 Paetzel M & Strynadka NC (1999) Common protein
architecture and binding sites in proteases utilizing a
Ser/Lys dyad mechanism. Protein Sci 8, 2533–2536.
5 Paetzel M & Dalbey RE (1997) Catalytic hydroxyl/
amine dyads with serine proteases. Trends Biochem Sci
22, 28–31.
6 Bratovanova EK & Petkov DD (1987) Glycine flanked
by hydrophobic bulky amino acid residues as minimal
sequence for effective subtilisin catalysis. Biochem J 248,
957–960.
MS analysis
The molecular mass of the aminolysis reaction products
was determined by ultra-performance liquid chromatogra-
phy (UPLC)-ESI-TOF MS equipped with a C18 reverse-
phase system. For UPLC-ESI-TOF MS analysis, the reac-
tion mixture was diluted by 1000-fold v/v with 0.1% formic
acid. After the solution was filtered, aliquots of each sam-
ple (5 lL) were analyzed by UPLC-ESI-TOF MS (LCT
7 Yokozeki K & Hara S (2005) A novel and efficient
enzymatic method for the production of peptides from
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
347

Structure of D-stereospecific amidohydrolase
J. Arima et al.
unprotected starting materials. J Biotechnol 115,
211–220.
8 Arima J, Morimoto M, Usuki H, Mori N & Hatanaka
T (2010) b-Alanyl-peptide synthesis by Streptomyces S9
aminopeptidase. J Biotechnol 147, 52–58.
D-carboxypeptidase with new penicillin binding activity
by directed mutagenesis. Protein Sci 14, 2296–2303.
20 Arima J, Ito H, Hatanaka T & Mori N (2011)
Aminolytic reaction catalyzed by D-stereospecific
amidohydrolases from Streptomyces spp. Biochimie 93,
1460–1469.
9 Arima J, Morimoto M, Usuki H, Mori N & Hatanaka
T (2010) The aminolysis reaction of Streptomyces S9
aminopeptidase promotes the synthesis of diverse prolyl
dipeptides. Appl Environ Microbiol 76, 4109–4112.
10 Arima J, Chiba M, Ichiyanagi T, Yabuta Y, Mori N &
Aimi T (2010) Eryngase: a Pleurotus eryngii
aminopeptidase exhibiting peptide bond formation
activity. Appl Microbiol Biotechnol 87, 1791–1801.
11 Arima J, Tanaka A, Morimoto M & Mori N (2013)
Mutation of active site serine residue with cysteine
displays change in acyl-acceptor preference of b-
peptidyl aminopeptidase from Pseudomonas aeruginosa
PAO1. Appl Microbiol Biotechnol 98, 1631–1640.
12 Fre‘re JM (2004) Streptomyces R61 D-Ala-D-Ala
carboxypeptidase. In Handbook of Proteolytic
Enzymes, 2 edn (Barrett AJ, Rawlings ND & Woessner
JF, eds), pp. 1959–1962. Elsevier, London.
13 Kato Y, Asano Y, Nakazawa A & Kondo K (1989)
First stereoselective synthesis of D-amino acid alkyl
amide catalyzed by D-aminopeptidase. Tetrahedron 45,
5743–5754.
14 Kato Y, Asano Y, Nakazawa A & Kondo K (1990)
Synthesis of D-alanine oligopeptides catalyzed by D-
aminopeptidase in non-aqueous media. Biocatalysis 3,
207–215.
21 Arima J, Hatanaka T & Mori N (2011) One-pot
synthesis of diverse DL-configuration dipeptides by a
Streptomyces D-stereospecific amidohydrolase. Appl
Environ Microbiol 77, 8209–8218.
ꢂ
22 Dubus A, Ledent P, Lamotte-Brasseur J & Frere JM
(1996) The roles of residues Tyr150, Glu272, and
His314 in class C b-lactamases. Proteins 25, 473–485.
23 Lamotte-Brasseur J, Dubus A & Wade RC (2000) pKa
calculation for class C b-lactamases: the role of Tyr150.
Proteins 40, 23–28.
24 Silvaggi NR, Anderson JW, Brinsmade SR, Pratt RF &
Kelly JA (2003) The crystal structure of phosphonate-
inhibited D-Ala-D-Ala peptidase reveals an analogue of
a tetrahedral transition state. Biochemistry 42,
1199–1208.
25 Lee W, McDonough MA, Kotra L, Li ZH, Silvaggi
NR, Takeda Y, Kelly JA & Mobashery S (2001) A 1.2-
A snapshot of the final step of bacterial cell wall
biosynthesis. Proc Natl Acad Sci USA 98, 1427–1431.
26 Kuzin AP, Liu H, Kelly AJ & Knox JR (1995) Binding
of cephalothin and cefotaxime to D-Ala-D-Ala-
peptidase reveals a functional basis of a natural
mutation in a low-affinity penicillin-binding protein and
in extended-spectrum b-lactamases. Biochemistry 34,
9532–9540.
15 Kelly JA & Kuzin AP (1995) The refined
ꢁ
ꢂ
crystallographic structure of a DD-peptidase penicillin-
ꢀ
target enzyme at 1.6 A resolution. J Mol Biol 254,
27 Llinas A, Ahmed N, Cordaro M, Laws AP, Frere JM,
Delmarcelle M, Silvaggi NR, Kelly JA & Page MI
(2005) Inactivation of bacterial DD-peptidase by beta-
sultams. Biochemistry 44, 7738–7746.
223–236.
16 Bompard-Gilles C, Remaut H, Villeret V, Prange T,
ꢁ
ꢂ
Fanuel L, Delmarcelle M, Joris B, Frere J & Van
28 Asano Y, Mori T, Hanamoto S, Kato Y & Nakazawa
A (1989) A new D-stereospecific amino acid amidase
from Ochrobactrum anthropi. Biochem Biophys Res
Commun 162, 470–474.
29 Asano Y, Nakazawa A, Kato Y & Kondo K (1989)
Properties of a novel D-stereospecific aminopeptidase
from Ochrobactrum anthropi. J Biol Chem 264, 14233–
14239.
Beeumen J (2000) Crystal structure of a D-
aminopeptidase from Ochrobactrum anthropi, a new
member of the ‘penicillin-recognizing enzyme’ family.
Structure 8, 971–980.
17 Okazaki S, Suzuki A, Komeda H, Yamaguchi S, Asano
Y & Yamane T (2007) Crystal structure and functional
characterization of a D-stereospecific amino acid
amidase from Ochrobactrum anthropi SV3, a new
member of the penicillin-recognizing proteins. J Mol
Biol 368, 79–91.
18 Lobkovsky E, Billings EM, Moews PC, Rahil J, Pratt
RF & Knox JR (1994) Crystallographic structure of a
phosphonate derivative of the Enterobacter cloacae P99
cephalosporinase: mechanistic interpretation of a beta-
lactamase transition-state analog. Biochemistry 33,
6762–6772.
30 Asano Y, Ito H, Dairi T & Kato Y (1996) An alkaline
D-stereo-specific endopeptidase with b-lactamase
activity from Bacillus cereus. J Biol Chem 271, 30256–
30262.
ꢁ
31 Arthur M, Depardieu F, Cabanie L, Reynolds P &
Courvalin P (1998) Requirement of the VanY and
VanX D, D-peptidases for glycopeptide resistance in
enterococci. Mol Microbiol 30, 819–830.
32 Goffin C & Ghuysen JM (1998) Multimodular
penicillin-binding proteins: an enigmatic family of
orthologs and paralogs. Microbiol Mol Biol Rev 62,
1079–1093.
ꢁ
19 Delmarcelle M, Boursoit MC, Filee P, Baurin SL, Frere
ꢂ
JM & Joris B (2005) Specificity inversion of
Ochrobactrum anthropi D-aminopeptidase to a D,
348
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS

J. Arima et al.
Structure of D-stereospecific amidohydrolase
33 Reynolds PE, Depardieu F, Dutka-Malen S, Arthur M
& Courvalin P (1994) Glycopeptide resistance mediated
by enterococcal transposon Tn1546 requires production
of VanX for hydrolysis of D-alanyl-D-alanine. Mol
Microbiol 13, 1065–1070.
38 McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn
MD, Storoni LC & Read RJ (2007) Phaser
crystallographic software. J Appl Crystallogr 40,
658–674.
ꢁ
39 Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis
34 Silvaggi NR, Josephine HR, Kuzin AP, Nagarajan R,
Pratt RF & Kelly JA (2005) Crystal structures of
complexes between the R61 DD-peptidase and
peptidoglycan-mimetic beta-lactams: a non-covalent
complex with a ‘perfect penicillin’. J Mol Biol 345,
521–533.
35 Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227, 680–685.
36 Otwinowski Z & Minor W (1997) Processing of X-ray
diffraction data collected in oscillation mode. Methods
in Enzymology, 276, Macromolecular Crystallography,
part (Carter ACW Jr & Sweet RM, eds), pp. 307–326.
Academic Press, New York.
IW, Echols N, Headd JJ, Hung L-W, Kapral GJ,
Grosse-Kunstleve RW et al. (2010) PHENIX: a
comprehensive Python-based system for
macromolecular structure solution. Acta Crystallogr
D66, 213–221.
40 Pettersen EF, Goddard TD, Huang CC, Couch GS,
Greenblatt DM, Meng EC & Ferrin TE (2004) UCSF
Chimera – a visualization system for exploratory
research and analysis. J Comput Chem 25, 1605–1612.
41 Chen VB, Arendall WB 3rd, Headd JJ, Keedy DA,
Immormino RM, Kapral GJ, Murray LW, Richardson
JS & Richardson DC (2010) MolProbity: all-atom
structure validation for macromolecular
crystallography. Acta Crystallogr D Biol Crystallogr 66,
12–21.
37 Leslie AGW & Powell HR (2007) Evolving Methods
for Macromolecular Crystallography. Springer,
Dordrecht. 245, pp. 41–51. ISBN 978-1-4020-6314-5.
FEBS Journal 283 (2016) 337–349 ª 2015 FEBS
349