Gene cloning and characterization of α-amino acid ester acyl transferase in empedobacter brevis ATCC14234 and Sphingobacterium siyangensis AJ2458

Article abstract of DOI:10.1271/bbb.110181

The gene encoding α-amino acid ester acyl transferase (AET), the enzyme that catalyzes the peptideforming reaction from amino acid methyl esters and amino acids, was cloned from Empedobacter brevis ATCC14234 and Sphingobacterium siyangensis AJ2458 and expressed in Escherichia coli. This is the first report on the aet gene. It encodes a polypeptide composed of 616 (ATCC14234) and 619 (AJ2458) amino acids residues. The V_max values of these recombinant enzymes during the catalysis of L-alanyl-L-glutamine formation from L-alanine methylester and L-glutamine were 1,010 U/mg (ATCC14234) and 1,154 U/mg (AJ2458). An amino acid sequence similarity search revealed 35% (ATCC14234) and 36% (AJ2458) identity with an α-amino acid ester hydrolase from Acetobacter pasteurianus, which contains an active-site serine in the consensus serine enzyme motif, GxSYxG. In the deduced amino acid sequences of AET from both bacteria, the GxSYxG motif was conserved, suggesting that AET is a serine enzyme.

Full text of DOI:10.1271/bbb.110181

Biosci. Biotechnol. Biochem., 75 (11), 2087–2092, 2011
Gene Cloning and Characterization of ꢀ-Amino Acid Ester Acyl Transferase
in Empedobacter brevis ATCC14234 and Sphingobacterium siyangensis AJ2458
Isao ABE,y Seiichi HARA, and Kenzo YOKOZEKI
Research Institute for Bioscience Products and Fine Chemicals, Ajinomoto Co., Inc.,
1-1 Suzuki-cho, Kawasaki-ku, Kawasaki 210-8681, Japan
Received March 8, 2011; Accepted July 31, 2011; Online Publication, November 7, 2011
[doi:10.1271/bbb.110181]
The gene encoding ꢀ-amino acid ester acyl trans-
ferase (AET), the enzyme that catalyzes the peptide-
forming reaction from amino acid methyl esters and
amino acids, was cloned from Empedobacter brevis
ATCC14234 and Sphingobacterium siyangensis AJ2458
and expressed in Escherichia coli. This is the first report
on the aet gene. It encodes a polypeptide composed of
616 (ATCC14234) and 619 (AJ2458) amino acids
residues. The Vmax values of these recombinant enzymes
during the catalysis of L-alanyl-L-glutamine formation
from L-alanine methylester and L-glutamine were
1,010 U/mg (ATCC14234) and 1,154 U/mg (AJ2458).
An amino acid sequence similarity search revealed
35% (ATCC14234) and 36% (AJ2458) identity with an
ꢀ-amino acid ester hydrolase from Acetobacter pasteur-
ianus, which contains an active-site serine in the
consensus serine enzyme motif, GxSYxG. In the
deduced amino acid sequences of AET from both
bacteria, the GxSYxG motif was conserved, suggesting
that AET is a serine enzyme.
We have developed a novel enzymatic method to
generate Ala-Gln that overcomes the disadvantages of
conventional methods, using L-alanine methyl ester
(Ala-OMe) as an acyl donor and Gln as a nucleophile.6)
We found that ꢀ-amino acid ester acyl transferase
(AET), purified from Empedobacter brevis ATCC
14234 (EAET), generates large amounts of Ala-Gln
from Ala-OMe and Gln in aqueous solution, with yields
exceeding 80% relative to Ala-OMe. In addition, the
enzyme has wide substrate specificity for both acyl
donors and nucleophiles, and can catalyze peptide-
forming reactions not only to produce various dipeptides
from the corresponding amino acid esters and amino
acids but also to produce various oligopeptides from the
corresponding amino acid esters and peptides.6)
Here we report the gene cloning of AET from two
strains, and the characterization of the resulting re-
combinant enzymes. First we cloned the EAET-
encoding gene and expressed it in E. coli. Next, we
hypothesized that forms of AET other than EAET are
more highly expressed in E. coli and screened for AET
homologs. Sphingobacterium siyangensis AJ2458 was
selected as a representative strain, and the gene encoding
AET from S. siyangensis AJ2458 (SAET) was cloned.
The Ala-Gln producing activity in a culture broth of
recombinant E. coli expressing SAET was higher than
that of recombinant E. coli expressing EAET. These
recombinant enzymes were then purified and character-
ized, and analysis of the deduced amino acid sequence
indicated that AET is a serine enzyme.
Key words: L-alanyl-L-glutamine; ꢀ-amino acid ester
acyl transferase; Empedobacter brevis;
Sphingobacterium siyangensis; L-glutamine
L-Glutamine (Gln) is a conditionally essential amino
acid due to its central function in nitrogen metabolism.1)
Despite its nutritional importance, it is rarely used as a
component of parenteral nutrition due to its low
solubility and instability in solution. One approach to
overcome its physicochemical limitations is to supply it
as a dipeptide by conjugating it with other amino acids.
Consequently, L-alanyl-L-glutamine (Ala-Gln) is one of
the most suitable Gln-containing dipeptides.2)
Several methods of producing Ala-Gln have been
reported, including Ala-Gln chemical synthesis via D-2-
chloropropionyl-glutamine, as reported by Sano et al.,3)
but these methods require many complicated steps, due
to the need to introduce protecting groups into the amino
acids and to remove them from the resulting peptides.
Recently, Tabata et al. reported a simple method of Ala-
Gln synthesis through fermentative production from
glucose and ammonia,4,5) but this system cannot recog-
nize the fixation of the desired amino acid at a specified
position. Hence this method has the potential disadvant-
age that undesired peptide byproducts may occur in
accordance with the substrate specificity of the enzyme.
Materials and Methods
Materials. The amino acid methyl esters used were hydrochloride
salts obtained from Bachem (Bubendorf, Switzerland), and the amino
acids were from Ajinomoto (Tokyo). A HiLoad 26/10 Q Sepharose
column, a Resource PHE column, and a HiLoad 16/60 Superdex 200
pg column were purchased from GE Healthcare (Buckinghamshire,
UK). A Bio-Scale CHT5-I column was from Bio-Rad Laboratories
(Hercules, CA).
Microorganisms and culture conditions. E. brevis ATCC14234 and
S. siyangensis AJ2458 were grown in Terrific Broth (TB) medium,
containing glycerol 4 g/L, peptone 12 g/L, yeast extract 24 g/L,
KH2PO4 2.3 g/L, and K2HPO4 12.5 g/L at 20 ꢀC. E. coli JM109,
harboring expression plasmids, was grown in TB medium containing
100 mg/L of ampicillin sodium salt (amp) at 20 ꢀC.
Cloning of genes encoding ꢀ-amino acid ester acyl transferase. We
cloned the gene encoding AET from E. brevis ATCC 14234 (eaet), as
y
To whom correspondence should be addressed. Fax: +81-44-244-6581; E-mail: isao abe@ajinomoto.com
Abbreviations: AET, ꢀ-amino acid ester acyl transferase; amp, ampicillin sodium salt; ORF, open reading frame; TB, Terrific Broth

2088
I. ABE et al.
follows: Consensus primers (50-TTYACNGCNATHTAYCARCC-30
and 50-TCNGGCATNGTRTANGTNACRTT-30; H: A, T, or C; N: A,
C, G, T; R: A or G; Y: C or T) were designed based on the internal
amino acid sequence of EAET. A 1.5-kb DNA fragment was amplified
from E. brevis ATCC 14234 genomic DNA, and a 4.5-kb DNA
fragment was hybridized to the 1.5-kb DNA fragment by Southern
hybridization. The 4.5-kb fragment was inserted into the HindIII site of
the pUC118 vector. Then pUC118 Eaet was obtained by colony
hybridization using the amplified DNA fragment as probe.
volume of 100 mM potassium phosphate buffer containing 4 M
ammonium sulfate (pH 6.5).
The mixed solution (60 mL) was applied to a Resource PHE column
(0:64 ꢁ 3 cm) equilibrated with 100 mM potassium phosphate buffer
containing 2 M ammonium sulfate (pH 6.5). The column was washed
with 100 mM potassium phosphate buffer containing 2 M ammonium
sulfate (pH 6.5), and then the enzyme was eluted with a linear gradient
of 2.0–0 M ammonium sulfate. The purified enzyme (0.6 mL) was
dialyzed against 20 mM potassium phosphate buffer (pH 6.5) and
stored at 4 ꢀC.
Expression plasmids for eaet were constructed as follows: Three
initiation codons were identified as candidates, with which we
designed corresponding primers, GTG2 50-GGGAATTCCATATG-
AAAAAATTAACATTAAAAGTAACT-30 and 50-GGGGGCTGCA-
GTACTTGTACGGTTTCGCCCGATAAA-30, GTG1 50-GGGAATT-
CCATATGCGCGACAATTACGAAAAAATAGAA-30 and 50-GGG-
GGCTGCAGTACTTGTACGGTTTCGCCCGATAAA-30, and ATG
50-GGGAATTCCATATGCGCGATGGTACAAAGTTATTTACA-30
and 50-GGGGGCTGCAGTACTTGTACGGTTTCGCCCGATAAA-30.
The eaet genes were amplified by PCR using the GTG2, GTG1, and
ATG primers and E. brevis ATCC14234 genomic DNA as template.
During PCR, the GTG initiation codon was intentionally converted to
ATG, because the former is used infrequently in E. coli.
E. coli Saet was cultured under the same conditions as E. coli Gtg2,
harvested by centrifugation (8;000 ꢁ g, 10 min), and washed with
100 mM potassium phosphate buffer (pH 6.5). All procedures were
carried out at 4 ꢀC or on ice. Wet cells (5 g) were resuspended in 50 mL
of 100 mM potassium phosphate buffer (pH 6.5), and the cell
suspension was sonicated with an Insonator 201 at 195 W for 45 min.
To remove the insoluble fraction, the supernatant was ultracentrifuged
at 100;000 ꢁ g for 20 min, and the resulting supernatant was used as
the cell-free extract.
The cell-free extract (47 mL) was applied to a Bio-Scale CHT5-I
column (1 ꢁ 6:4 cm) equilibrated with 100 mM potassium phosphate
buffer (pH 6.5). The column was washed with 100 mM potassium
phosphate buffer (pH 6.5), and then the enzyme was eluted with a
linear gradient of 0.1–0.5 M potassium phosphate. The active fractions
were collected and dialyzed against 100 mM potassium phosphate
(pH 6.5). The dialyzed solution was then mixed with an equal volume
of 100 mM potassium phosphate buffer containing 4 M ammonium
sulfate (pH 6.5).
The amplified DNA fragments were digested with NdeI and PstI
and inserted into the NdeI and PstI sites of pTrpT, an expression
plasmid that contains the tryptophan promoter and the rrnB terminator
of E. coli in the pUC19 vector.7) The resulting plasmids were named
pTrpT Gtg2, pTrpT Gtg1, and pTrpT Atg. E. coli JM109 was
transformed with pTrpT Gtg2, pTrpT Gtg1, and pTrpT Atg, and the
transformants were named E. coli Gtg2, E. coli Gtg1, and E. coli Atg.
We cloned the gene encoding S. siyangensis AJ2458 AET (saet)
as follows: The probe for hybridization was generated by PCR using
pTrpT Gtg2 as template. Primers (50-GGGAATTCCATATG-
AAAAAATTAACATTAAAAGTAACT-30 and 50-GGGGGCTGCA-
GTACTTGTACGGTTTCGCCCGATAAA-30) were designed based
on the open reading frame (ORF) of eaet. A 3.5-kb DNA fragment was
hybridized to the probe by Southern hybridization using S. siyangensis
AJ2458 genomic DNA. This fragment was inserted into the HindIII
site of pUC19 vector, and pUC19 Saet was obtained by colony
hybridization using the amplified DNA as probe.
The mixed solution (44 mL) was applied to a Resource PHE column
(0:64 ꢁ 3 cm) equilibrated with 100 mM potassium phosphate buffer
containing 2 M ammonium sulfate (pH 6.5). The column was washed
with 100 mM potassium phosphate buffer containing 2 M ammonium
sulfate (pH 6.5), ant then the enzyme was eluted with a linear gradient
of 2.0–0 M ammonium sulfate. Five mL of the active fraction was
collected and concentrated to 1 mL by ultrafiltration with an Ultracel-
3K filter (Millipore, Billerica, MA) applied to a HiLoad 16/60
Superdex 200 pg column (1:6 ꢁ 60 cm) equilibrated with 20 mM
potassium phosphate buffer containing 1 M sodium chloride (pH 6.5),
and eluted with 20 mM potassium phosphate buffer containing 1 M
sodium chloride (pH 6.5). The purified enzyme (3 mL) was then
dialyzed against 20 mM potassium phosphate buffer (pH 6.5) and
stored at 4 ꢀC.
The expression plasmid for saet was constructed as follows: Saet
was amplified by PCR using primers 50-GGGAATTCCATAT-
GAAAAATACAATTTCGT-30 and 50-GCTCTAGACTAATCTTT-
GAGGACAGAAAA-30 with pUC19 Saet as template. The amplified
DNA fragment was digested with NdeI and PstI and inserted into the
NdeI and PstI sites of pTrpT. The resulting plasmid was named pTrpT
Saet. E. coli JM109 was transformed with pTrpT Saet, and the
transformant was named E. coli Saet.
Protein analysis. Protein concentrations were determined by the
Bradford method8) with bovine serum albumin as standard. Sodium
dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) was
performed on a 10–20% polyacrylamide gel (Daiichi Pure Chemicals,
Tokyo) with Precision Protein Marker (Bio-Rad Laboratories,
Hercules, CA) as marker proteins. The native relative molecular mass
was determined with a HiLoad 16/60 Superdex 200 pg column using
Gel Filtration LMW and Gel Filtration HMW calibration kits (GE
Healthcare, Buckinghamshire, UK).
All DNA samples were sequenced with a 3100 Genetic Analyzer
(Applied Biosystems, Foster City, CA).
Purification of ꢀ-amino acid ester acyl transferase. E. coli Gtg2
was subcultured at 20 ꢀC for 24 h in Luria-Bertani medium containing
100 mg/L of amp and 15 g/L of agar. The subcultured cells were then
used to inoculate 100 mL of TB medium containing 100 mg/L of amp
in a 500-mL flask. The cells were grown at 20 ꢀC for 40 h, harvested by
centrifugation (8;000 ꢁ g, 10 min), and washed with 50 mM Tris–HCl
buffer (pH 8.0). All procedures were carried out at 4 ꢀC or on ice. Wet
cells (15 g) were resuspended in 100 mL of 50 mM Tris–HCl buffer
(pH 8.0), and the cell suspension was sonicated using Insonator 201
(Kubota, Tokyo) at 195 W for 45 min. To remove the insoluble
fraction, the supernatant was ultracentrifuged at 100;000 ꢁ g for
20 min, and the resulting supernatant was used as the cell-free extract.
The cell-free extract (95 mL) was then applied to a HiLoad 26/10 Q
Sepharose column (2:6 ꢁ 10 cm) equilibrated with 50 mM Tris–HCl
(pH 8.0). The active fractions, collected from the non-adsorbed
fractions, were dialyzed against 100 mM potassium phosphate buffer
(pH 6.5). The dialyzed solution (111 mL) was then applied to a Bio-
Scale CHT5-I column (1 ꢁ 6:4 cm) equilibrated with 100 mM potas-
sium phosphate buffer (pH 6.5). The column was washed with 100 mM
potassium phosphate buffer (pH 6.5), and then the enzyme was eluted
with a linear gradient of 0.1–0.5 M potassium phosphate. The active
fractions were collected and dialyzed against 100 mM potassium
phosphate (pH 6.5). The dialyzed solution was mixed with an equal
Enzyme assay. To measure Ala-Gln producing activity, we used a
mixture of 100 mM borate buffer (pH 9.0), 100 mM Ala-OMe, and
200 mM Gln at a final volume of 0.1 mL. Because Ala-OMe hydro-
chloride was used as substrate, the substrate solution was adjusted to
pH 9.0 with 6 N NaOH before it was added to the enzyme. The enzyme
reaction was carried out at 25 ꢀC for 10 min. To stop the reaction, 1 mL
of 1% w/v H3PO4 solution was added. One unit of Ala-Gln producing
activity was defined as the amount of enzyme that produces 1 mmol
Ala-Gln per min.
Optimal reaction pH and temperature. The optimal pH values for
the reactions were determined at 25 ꢀC using various buffers at
100 mM: sodium acetate (pH 4.0, 4.5, 5.0, and 5.5), MES (pH 5.5, 6.0,
and 6.5), potassium phosphate (pH 6.5, 7.0, and 7.5), Tris–HCl
(pH 7.5 and 8.0), and borate (pH 8.0, 8.5, 9.0, 9.5, and 10.0). The
optimal reaction temperature was selected from among 10 ꢀC, 15 ꢀC,
20 ꢀC, 25 ꢀC, 30 ꢀC, 35 ꢀC, 40 ꢀC, 45 ꢀC, 50 ꢀC, and 55 ꢀC at pH 9.0.
Kinetic assay. The kinetic parameters with regard to the Ala-Gln
forming reaction were determined by measuring the Ala-Gln produced

Gene Cloning of ꢀ-Amino Acid Ester Acyl Transferase
2089
Fig. 1. Nucleotide and Deduced Amino Acid Sequences of eaet.
The deduced amino acid sequence lies below the nucleotide sequence. Possible initiation codons are boxed (GTG2, GTG1, ATG). Internal
amino acid sequences are underlined.
by varying the concentration of Ala-OMe (4.6, 9.1, 18.2, 36.4, and
cleavage with lysyl endopeptidase and determined the
internal amino acid sequences of EAET to be
72.7 mM) against several fixed concentrations of Gln (4.6, 9.1, 18.2,
36.4, and 72.7 mM) at 25 ꢀC in 100 mM borate buffer (pH 9.0). The Km
LFTAIYQPK and ETNVTYTMPD. The gene was
values toward Ala-OMe and Gln and the Vmax values were determined
obtained as described above in ‘‘Materials and
from secondary plots of Lineweaver-Burk plots.
Methods.’’ The nucleotide sequence contained a 1,851-
bp ORF, encoding a polypeptide of 616 amino acids
(Fig. 1). The estimated molecular mass was approx-
imately 71 kDa.
High-performance liquid chromatography analysis. The amounts of
dipeptides produced were measured by high-performance liquid
chromatography under the following conditions: column, Inertsil
ODS-2 (4:6 mm ꢁ 250 mm; GL Science, Tokyo); mobile phase,
H3PO4 solution containing a 10:1.5–5 mixture of 5 mM sodium
1-octanesulfonate (pH 2.1) and methanol; flow rate, 1.0 mL/min;
temperature 40 ꢀC; detection, 210 nm.
To confirm that the DNA fragment we cloned was
indeed eaet, expression plasmids for eaet were con-
structed. Because some initiation codons were observed
(Fig. 1), E. coli JM109 was transformed with the
various expression plasmids and grown at 20 ꢀC in TB
medium containing 100 mg/L of amp for 40 h. After
culture, the Ala-Gln producing activities of E. coli Gtg2,
E. coli Gtg1, and E. coli Atg culture broths were
examined. Only that of E. coli Gtg2 demonstrated
similar activity (0.4 U/mL of culture broth), approx-
imately 11 times higher than that of the wild-type strain.
Based on these results, we concluded that the DNA
Results
Cloning and expression of the gene encoding ꢀ-amino
acid ester acyl transferase from E. brevis ATCC14234
To analyze the N-terminal amino acid sequence of
EAET, EAET was purified as described previously,6) but
we could not determine whether the N terminal was
protected. Next, we analyzed fragments of EAET after

2090
I. ABE et al.
Fig. 2. Alignment of EAET and SAET.
The putative signal peptide is underlined. The serine enzyme consensus sequence (GxSYxG) is boxed (x is a nonconserved amino acid).
Putative Ser-Asp-His catalytic residues are indicated by an asterisk. EAET, ꢀ-amino acid ester acyl transferase from E. brevis ATCC14234
(DDBJ accession no. AB610977); SAET, ꢀ-amino acid ester acyl transferase from S. siyangensis AJ2458 (no. AB610978).
Table 1. Purification of EAET and SAET
fragment contained a 1.8-kb ORF corresponding to
eaet and that GTG2 was the initiation codon, as shown
in Fig. 1.
Total
activity
(U)
Total
protein
(mg)
Specific
activitya
(U/mg)
Yield
(%)
Fraction
Screening of AET from Cytophaga, Flavobacterium,
and Sphingobacterium
EAET
Cell-free extract
Q-Sepharose
CHT5-I
127
100
60
729
62
1.7
0.00036
0.17
1.6
100
79
We hypothesized that forms of AET other than EAET
are more highly expressed in E. coli. Hence we screened
strains that demonstrated Ala-Gln producing activity.
Because E. brevis belongs to the phylum CFB, which
includes the genera Cytophaga, Flavobacterium, and
Sphingobacterium,9) we examined 57 such strains for
their ability to produce Ala-Gln from Ala-OMe and Gln,
as described previously.6) S. siyangensis AJ2458 pro-
duced amounts of Ala-Gln equal to E. brevis
ATCC14234, prompting us to select it as a representa-
tive producer of Ala-Gln.
36
998
48
Resource-PHE
0.36
0.29
SAET
Cell free extract
CHT5-I
3935
1704
743
93
206
19
385
100
43
4.4
0.85
0.092
Resource-PHE
Superdex 200 pg
872
1015
19
2.4
aAla-Gln producing activity was assayed at 25 ꢀC and at pH 9.0.
Cloning and expression of the gene encoding AET
from S. siyangensis AJ2458
Ala-Gln producing activities of E. coli Saet and E. coli
Gtg2 were 39.4 and 0.4 U/mL of culture broth respec-
tively, and the activity of E. coli Saet was 98.5 times
higher than that of E. coli Gtg2.
The gene encoding S. siyangensis AJ2458 ꢀ-amino
acid ester acyl transferase (SAET) was cloned as
described in ‘‘Materials and Methods.’’ We noted a
1,860-bp ORF in the nucleotide sequence, encoding a
polypeptide of 619 amino acid residues whose estimated
molecular mass was 71 kDa, the same as EAET.
To confirm that the DNA fragment was saet, a saet
expression plasmid was constructed, and E. coli harbor-
ing this plasmid (E. coli Saet) was grown at 20 ꢀC in TB
medium containing 100 mg/L of amp for 40 h. E. coli
Saet demonstrated Ala-Gln producing activity, leading
us to conclude that the DNA fragment was saet. The
Purification and characterization of recombinant
enzymes
To determine whether the difference in Ala-Gln
producing activity of culture broth as between E. coli
Gtg2 and E. coli Saet depended on the specific activity
of the enzyme, recombinant enzymes were purified
(Table 1). The enzymes were purified in 3-column
chromatography runs, and the purified enzymes ap-
peared to be homogeneous, as determined by SDS–

Gene Cloning of ꢀ-Amino Acid Ester Acyl Transferase
2091
PAGE (data not shown). The apparent native molecular
Peptides produced (mM)
mass of SAET was approximately 115 kDa as deter-
mined by gel-filtration. The molecular mass of the
denatured SAET based on SDS–PAGE was approxi-
mately 68 kDa, suggesting that SAET consists of two
identical subunits, like as EAET.6) The specific activities
of EAET and SAET were 998 U/mg and 1,015 U/mg
respectively. Because the specific activities were nearly
equal, we concluded that SAET was more highly
expressed than EAET in E. coli. In addition, the specific
activity of EAET (998 U/mg) was almost identical to
that of native EAET (1,073 U/mg) isolated from
E. brevis ATCC14234.6)
The Km values for Ala-OMe during the Ala-Gln
forming reaction from Ala-OMe and Gln were 21.3 mM
(EAET) and 23.3 mM (SAET), and thus the Km values
for Ala-OMe were nearly equal. In contrast, the Km
values for Gln were 9.9 mM (EAET) and 15.9 mM
(SAET), and thus the Km value for Gln of SAET was
about 1.6 times higher than that of EAET. The Vmax
values were 1,010 U/mg (EAET) and 1,154 U/mg
(SAET), and thus the Vmax values were nearly equal.
The optimal pH for the Ala-Gln producing activity
of EAET and SAET was 8.5, and the optimal temper-
atures for EAET and SAET were 30 ꢀC and 20 ꢀC
respectively.
The substrate specificity of SAET was examined as
described previously.6) Hence the production yield of
L-alanyl-L-alanine (Ala-Ala) was high, and the produc-
tion yields of L-alanyl-L-lysine (Ala-Lys), L-alanyl-L-
arginine (Ala-Arg), L-alanyl-L-histidine (Ala-His), and
L-alanyl-L-tryptophan (Ala-Trp) were low as compared
with those of EAET reported in a previous study,6)
and thus basic amino acids (L-lysine, L-arginine, and
L-histidine) and L-tryptophan were unsuitable nucleo-
philes for SAET (Fig. 3A). The production yields of
L-glycyl-L-glutamine (Gly-Gln), L-seryl-L-glutamine
(Ser-Gln), and L-threonyl-L-glutamine (Thr-Gln) were
also low as compared with those of EAET,6) and L-
glycine methyl ester, L-serine methyl ester, and L-
threonine methyl ester were unsuitable acyl donors for
SAET (Fig. 3B).
0
20
40
60
80
Ala-Gly
Ala-Ala
Ala-Val
Ala-Leu
Ala-Ile
A
15 min
180 min
Ala-Met
Ala-Phe
Ala-Trp
Ala-Ser
Ala-Thr
Ala-Asn
Ala-Gln
Ala-Tyr
Ala-Lys
Ala-Arg
Ala-His
Ala-Asp
Ala-Glu
Gly-Gln
Ala-Gln
Val-Gln
Ile-Gln
B
Met-Gln
Pro-Gln
Phe-Gln
Ser-Gln
Thr-Gln
Gln-Gln
Tyr-Gln
Arg-Gln
Fig. 3. Substrate Specificity of SAET.
A, Nucleophile specificity of SAET; B, acyl donor specificity of
SAET. The various reaction solutions (100 mL), containing 100 mM
amino acid methyl ester, 150 mM amino acid, and 0.37 U purified
enzyme in 100 mM borate buffer (pH 9.0), were incubated at 25 ꢀC.
Discussion
Hence we hypothesize that these signal peptides are
necessary to produce active EAET and SAET and to
mediate their transport to the periplasm.10) While the
reason for these differences in the Ala-Gln producing
activity of culture broth as between E. coli Gtg2 and
E. coli Saet is not clear at present, we believe that a
difference in the manners of processing of EAET and
SAET in E. coli causes the disparity.
Through database searches using BLAST (blastp
2.1.2), EAET and SAET were found to be homologous
with several proteins (Table 2). Though the homology
was 35%, the most homologous protein was the ꢀ-amino
acid ester hydrolase (AEH) from Acetobacter pasteur-
ianus.11) AEH catalyzes the transfer of the acyl group
from an ꢀ-amino acid ester to amine nucleophiles, such
as 7-aminocephem and 6-penam compounds (synthesis),
or to water (hydrolysis). AEH has not been reported to
catalyze peptide-forming reactions with amino acids as a
nucleophile.
In this study, we isolated two novel genes encoding
AET from E. brevis ATCC14234 and S. siyangensis
AJ2458. The Ala-Gln producing activity of an E. coli
Saet culture broth was 98.5 times higher than that of an
E. coli Gtg2, although the specific activities of EAET
and SAET were nearly equal. In addition, the codon
usages of saet and eaet in E. coli were very similar (data
not shown). Based on these findings, the difference in
the Ala-Gln producing activity of culture broth as
between E. coli Gtg2 and E. coli Saet did not appear to
depend on the degree of translation.
The deduced amino acid sequence of EAET was 64%
identical to that of SAET and three amino acids shorter
than SAET (Fig. 2). By analysis with Genetyx software
(ver. 6.0.4), the first 22 amino acids of EAET and the
first 20 amino acids of SAET were identified as signal
peptides, and AET in E. brevis and S. siyangensis was
expected to localize to the periplasm. In addition, the
signal peptide is believed to facilitate proper folding.
The second most homologous protein was Bacillus
laterosporus 7ꢁ-(4-carboxybutanamido) cephalosporanic

2092
I. ABE et al.
Table 2. Proteins Homologous to EAET and SAET
Homology (%)
EAET SAET
Protein IDa
Organism
Function
Motifb
AAL60195
I40217
AAF42807
Acetobacter pasteurianus
Brevibacillus laterosporus
Rhodococcus sp.
AEH11)
Glutaryl 7-ACA acylase12)
Cocaine esterase16)
35
26
26
36
25
25
GSSYEG
GLSYMA
GVSYLG
aAccession number in the protein database of the National Center for Biotechnology Information.
bBold, consensus sequence of serine enzyme.
acid (glutaryl 7-ACA) acylase (GA), which hydrolyzes
glutaryl 7-ACA to 7-amino cephalosporanic acid, a
starting material for semisynthetic cephalosporin anti-
biotics.12) According to the Structural Classification of
Proteins (SCOP) classification, GA belongs to the N-
terminal nucleophile amidohydrolase superfamily. In
addition, it belongs to the penicillin acylase catalytic
domain family, which includes penicillin G acylase
(PGA). GA and PGA catalyze not only the hydrolysis of
amide bonds but also, like AET, the synthesis of amide
bonds.13,14) PGA synthesizes ampicillin from phenyl-
glycine methyl ester and 6-amino penicillanic acid
(6-APA) in a suitable solvent. Conversely, 6-APA is a
competitive inhibitor; enzymes that are bound to 6-APA
do not hydrolyze ampicillin.15)
adjacent to the catalytic His. Site-directed mutagenesis
of EAET F330 and SAET Y328 may help to elucidate
the function of the residue next to the catalytic His in
AET.
This is the first report on the aet gene. Directed
evolution, such as site-directed mutagenesis, might
increase the Ala-Gln producing activity of the enzyme
to improve the production yield of Ala-Gln.
Acknowledgments
We thank Dr. T. Ohnuki, Dr. K. Izawa, Dr. Y. Mihara,
and Dr. S. Suzuki for encouragement during the course
of this work, Mrs. S. Takeuchi for technical assistance,
and Dr. I. Kira for useful comments on the manuscript.
Ala-Gln was not hydrolyzed in the presence of Gln
during Ala-Gln production by AET.6) Hence the reaction
mechanism of AET has been suggested to resemble
those of GA and PGA. However, AET can synthesize
Ala-Gln at high yields in aqueous solution, unlike GA
and PGA. A comparative analysis of the reaction
mechanisms of AET, GA, and PGA might lead to
further consideration of these enzymes.
The third most homologous protein was cocaine
esterase from Rhodococcus sp. strain MB1. This enzyme
hydrolyzes the ester bond in cocaine to benzoate and
ecgonine methyl ester.16) Cocaine esterase and AEH
both contain an active-site serine in the consensus serine
enzyme motif (GxSYxG, where x is a nonconserved
amino acid). In GA, the GxSYxG motif is not
completely conserved, and the last glycine is replaced
by alanine. EAET and SAET contain the GxSYxG
motif. Asp and His residues, which constitute the
catalytic residues of serine enzymes, are also conserved
(Fig. 2). In addition, AET activity was inhibited by
4-nitrophenyl 4-guanidinobenzoate (p-NPGB), a serine
enzyme inhibitor (data not shown). Based on these
results, we propose that AET is a serine enzyme.
References
1) Wernerman J, Clin. Nutr. Suppl., 1, 37–42 (2004).
2) Furst P, Albers S, and Stehle P, Metab. Clin. Exp., 38, 67–72
¨
(1989).
3) Sano T, Sugaya T, Inoue K, Mizutaki S, Ono Y, and Kasai M,
Org. Process Res. Dev., 4, 147–152 (2000).
4) Tabata K, Ikeda H, and Hashimoto S, J. Bacteriol., 187, 5195–
5202 (2005).
5) Tabata K and Hashimoto S, Appl. Environ. Microbiol., 73,
6378–6385 (2007).
6) Yokozeki K and Hara S, J. Biotechnol., 115, 211–220 (2005).
7) Nozaki H, Kira I, Watanabe K, and Yokozeki K, J. Mol. Cat. B:
Enzymatic, 32, 205–211 (2005).
8) Bradford M, Anal. Biochem., 72, 248–254 (1976).
9) Kimura M, J. Mol. Evol., 16, 111–120 (1980).
10) Arkowitz RA and Bassilana M, Biochim. Biophys. Acta, 1197,
311–343 (1994).
11) Polderman-Tijmes JJ, Jekel PA, de Vries EJ, van Merode AE,
Floris R, van der Laan JM, Sonke T, and Janssen DB, Appl.
Environ. Microbiol., 68, 211–218 (2002).
12) Aramori I, Fukagawa M, Tsumura M, Iwami M, Ono H, Kojo
H, Kohsaka M, Ueda Y, and Imanaka H, J. Bacteriol., 173,
7848–7855 (1991).
13) Biffi S, De Martin L, Ebert C, Gardossi L, and Linda P, J. Mol.
Cat. B: Enzymatic, 19–20, 135–141 (2002).
The amino acid adjacent to the catalytic His differs
between EAET and SAET (EAET, F330; SAET, Y328).
Nakagawa et al. reported that a mutation at the residue
next to the catalytic His caused a selective increase in
the amidase activity of lipase, and the effect of this
mutation on the function of catalytic His was essential to
the increased activity.17) We propose, then, that the
difference in substrate specificity between EAET and
SAET is caused by the difference in the amino acid
14) Arroyo M, de la Mata I, Acebal C, and Castillon MP, Appl.
Microbiol. Biotechnol., 60, 507–514 (2003).
15) Goncalves LRB, Fernandez-Lafuente R, Guisan JM, and
Giordano RLC, Enzyme Microb. Technol., 31, 464–471 (2002).
16) Bresler MM, Rosser SJ, Basran A, and Bruce NC, Appl.
Environ. Microbiol., 66, 904–908 (2000).
17) Nakagawa Y, Hasegawa A, Hiratake J, and Sakata K, Protein
Eng. Des. Sel., 20, 339–346 (2007).