Overexpression and characterization of a carboxypeptidase from the hyperthermophilic archaeon Thermococcus sp. NA1

Article abstract of DOI:10.1271/bbb.70.1140

Genomic analysis of a hyperthermophilic archaeon, Thermococcus sp. NA1, revealed the presence of an 1,497 bp open reading frame, encoding a protein of 499 amino acids. The deduced amino acid sequence was similar to thermostable carboxypeptidase 1 from Pyrococcus furiosus, a member of peptidase family M32. Five motifs, including the HEXXH motif with two histidines coordinated with the active site metal, were conserved. The carboxypeptidase gene was cloned and overexpressed in Escherichia coli. Molecular masses assessed by SDS-PAGE and gel filtration were 61 kDa and 125 kDa respectively, which points to a dimeric structure for the recombinant enzyme, designated TNA1_CP. The enzyme showed optimum activity toward Z-Ala-Arg at pH 6.5 and 70-80°C (k _cat=K_m = 8.3 mM^-1 s^-1). In comparison with that of P. furiosus CP (k_cat/K_m = 667 mM^-1 s^-1), TNA1_CP exhibited 80-fold lower catalytic efficiency. The enzyme showed broad substrate specificity with a preference for basic, aliphatic, and aromatic C-terminal amino acids. This broad specificity was confirmed by C-terminal ladder sequencing of porcine N-acetyl-renin substrate by TNA1_CP.

Full text of DOI:10.1271/bbb.70.1140

Biosci. Biotechnol. Biochem., 70 (5), 1140–1147, 2006
Overexpression and Characterization of a Carboxypeptidase
from the Hyperthermophilic Archaeon Thermococcus sp. NA1
Hyun Sook LEE, Yun Jae KIM, Seung Seob BAE, Jeong Ho JEON,
y
Jae Kyu LIM, Sung Gyun KANG, and Jung-Hyun LEE
Korea Ocean Research & Development Institute, Ansan P.O. Box 29, Seoul 425-600, Korea
Received September 14, 2005; Accepted January 11, 2006
Genomic analysis of a hyperthermophilic archaeon,
Thermococcus sp. NA1, revealed the presence of an
1,497 bp open reading frame, encoding a protein of 499
amino acids. The deduced amino acid sequence was
similar to thermostable carboxypeptidase 1 from Pyro-
coccus furiosus, a member of peptidase family M32. Five
motifs, including the HEXXH motif with two histidines
coordinated with the active site metal, were conserved.
The carboxypeptidase gene was cloned and overex-
pressed in Escherichia coli. Molecular masses assessed
by SDS–PAGE and gel filtration were 61 kDa and
125 kDa respectively, which points to a dimeric struc-
ture for the recombinant enzyme, designated TNA1 CP.
The enzyme showed optimum activity toward Z-Ala-
Arg at pH 6.5 and 70–80 ꢀC (kcat=Km ¼ 8:3 mMꢁ1 sꢁ1).
peptidase Taq),2) T. thermophilus,3) and the hyperther-
mophilic archaeon Pyrococcus furiosus4) have been
purified and characterized so far. Thermostable carboxy-
peptidase 1 is known to be very thermostable, with a
temperature optimum of 80–100 ꢀC.2,4) High optimum
temperatures for activity have been reported for other
carboxypeptidases purified from the bacterium Ther-
moactinomyces vulgaris5) and the archaeon Sulfolobus
solfataricus,6) with temperature optima of 60 and 85 ꢀC
respectively.
Thermostable carboxypeptidase 1 exhibits broad sub-
strate specificities. For instance, P. furiosus CP releases
basic, aromatic, neutral, and polar amino acids from the
C-terminus while it is unable to digest peptides with C-
terminal Gly, Pro, or acidic residues (Asp and Glu).4) In
the case of carboxypeptidase Taq, it cleaves C-terminal
neutral, basic, and acidic amino acids most readily, with
the exception of Pro and hydrolyses amino acids with
long side chains.2)
The biological role of microbial metallocarboxypep-
tidases has not been established, while mammalian
metallocarboxypeptidases play key roles in major bio-
logical processes, ranging from digestive-protein degra-
dation, as effected by carboxypeptidase A and B,7,8) to
specific proteolytic processing, as in the maturation of
biologically active peptides, as effected by carboxypep-
tidase N and enkephalin convertase.9,10) It has been
suggested that the broad specificities of P. furiosus CP
against synthetic and natural peptide substrates might
play a role in digestion and protein turnover as opposed
to specific post-translational modification as seen in
higher organisms.4)
In comparison with that of P. furiosus CP (kcat=Km
¼
667 mMꢁ1 sꢁ1), TNA1 CP exhibited 80-fold lower cata-
lytic efficiency. The enzyme showed broad substrate
specificity with a preference for basic, aliphatic, and
aromatic C-terminal amino acids. This broad specificity
was confirmed by C-terminal ladder sequencing of
porcine N-acetyl-renin substrate by TNA1 CP.
Key words: genomic analysis; hyperthermophile; clon-
ing; expression; carboxypeptidase
Carboxypeptidases (CPs) act at C-terminals of the
polypeptide chain and liberate a single amino acid. They
can be divided into three major groups, serine carboxy-
peptidases, metallocarboxypeptidases, and cysteine car-
boxypeptidases, based on the nature of the amino acid
residues at the active site of the enzymes.
Metallocarboxyproteases are characterized by the
requirement of a divalent metal ion for their activity.
Thermostable carboxypeptidase 1 (EC 3.4.17.19), be-
longing to peptidase family M32, is distinct from other
metallocarboxypeptidases, since it contains a HEXXH
metal-binding motif.1) Several thermostable carboxy-
peptidase 1 genes that span all three kingdoms of life,
namely archaea, bacteria, and eucarya, have been
identified. Among them, the only three thermostable
carboxypeptidase 1 from Thermus aquaticus (carboxy-
There is growing interest in thermostable enzymes
because of their broad range of biotechnological
applications and as useful models to understand better
the structural basis of thermostability.11,12) The broad
substrate specificities of thermostable carboxypeptidase
1 allow their use in C-terminal sequencing of pro-
teins.2,4) Using N-acetyl-renin substrate, P. furiosus CP
was able to sequence up to seven residues from the C-
terminus releasing Leu, Val, Phe, Tyr, Ser, and His
residues.
y
To whom correspondence should be addressed. Tel: +82-31-400-6243; Fax: +82-31-406-2495; E-mail: jlee@kordi.re.kr

Characterization of an Archaeal Carboxypeptidase
1141
Recently, we isolated a hyperthermophilic archaeon,
Thermococcus sp. NA1, and the whole genome se-
quence was determined in order to search for valuable,
extremely thermostable enzymes. Analysis of the ge-
nome information of Thermococcus sp. NA1 showed a
carboxypeptidase gene belonging to thermostable car-
boxypeptidase 1. In this study, we describe the cloning
of the gene and characterization of the recombinant
enzyme with regard to its enzymatic properties.
1 h at 4 ꢀC). To purify TNA1 CP with the His6-Tag,
the resulting supernatant was applied to a column of
TALONꢀ metal affinity resin (BD Biosciences Clon-
tech, Palo Alto, CA) and washed with 10 mM imidazole
(Sigma, St. Louis, MO) in 50 mM Tris–HCl buffer (pH
8.0) containing 0.1 M KCl and 10% glycerol, TNA1 CP
was eluted with 300 mM imidazole in the buffer. To
purify TNA1 CP (no tag), the supernatant was applied to
a column of HiTrap Q (5 ml, Amersham Biosciences)
equilibrated with 50 mM Tris–HCl (pH 7.5). A gradient
from 0 to 1 M KCl in the same buffer was applied to the
column. The active fractions were pooled and concen-
trated using Centricon YM-10 (Millipore, Bedford,
MA). The sample was applied to a Superdex 200
10/300 GL column (Amersham Biosciences) equili-
brated with 50 mM Tris–HCl (pH 8.0) and 0.15 M NaCl.
The active fractions of TNA1 CP and TNA1 CP (no tag)
were then buffer exchanged to 50 mM Tris–HCl buffer
(pH 8.0) containing 10% glycerol using Centricon
YM-10.
Materials and Methods
Strains and growth conditions. E. coli DH5ꢀ was
used for plasmid propagation and nucleotide sequenc-
ing. E. coli BL21-CodonPlus(DE3)-RIL cells (Strata-
gene, LaJolla, CA) and plasmid pET-24a(þ) (Novagen,
Madison, WI) were used for gene expression. E. coli
strains were cultivated in Luria-Bertani medium at
37 ꢀC, and kanamycin was added to the medium at a
final concentration of 50 mg/ml.
The protein concentration was estimated from the
absorbance at 280 nm using an extinction coefficient of
90,750 Mꢂ1 cmꢂ1. The protein purity was examined
by sodium dodecyl sulfate–polyacrylamide gel electro-
phoresis (SDS–PAGE) analysis performed according to
the standard procedure. Gel filtration of the pure enzyme
was performed on a Superdex 200 10/300 GL column
DNA manipulation and sequencing. DNA manipula-
tions were performed by standard procedures, as
described by Sambrook and Russell.13) Restriction
enzymes and other modifying enzymes were purchased
from Promega (Madison, WI). Small-scale preparation
of plasmid DNA from E. coli cells was performed with a
plasmid mini kit (Qiagen, Hilden, Germany). DNA
sequencing was performed with the automated sequenc-
er (ABI3100) using a BigDye terminator kit (PE Applied
Biosystems, Foster City, CA).
¨
using fast protein liquid chromatography (Akta FPLC
System, Amersham Biosciences), equilibrated with 50
mM Tris–HCl (pH 8.0) and 0.2 M KCl. Thirty mg protein
was loaded and eluted at a flow rate of 0.4 ml/min.
Cloning and expression of the carboxypeptidase-
encoding gene. Based on the whole genome sequence
of Thermococcus sp. NA1, the full length of a
carboxypeptidase gene (TNA1 CP) flanked by the NdeI
and HindIII sites was amplified by PCR with genomic
DNA and three primers (sense [50-CGACCCGGCAT-
ATGGAGGAAGTTTTCCAGAACGAAACC-30], anti-
sense 1 [50-CTCCACATAAGCTTGAGGTACCTCTCC-
TTCACCCAGCG-30], and antisense 2 [50-CTCCACA-
TAAGCTTTCAGAGGTACCTCTCCTTCACCCAGCG-
30]; the italicized sequences indicate the NdeI site in
the sense primer and the HindIII site in the antisense
primers). The amplified DNA fragments were digested
with NdeI and HindIII. The fragments were then ligated
with NdeI/HindIII digested pET-24a(þ), and the result-
ing recombinant plasmids were used to transform E. coli
DH5ꢀ. The recombinant plasmids were introduced
into BL21-CodonPlus(DE3)-RIL for expression after
sequence confirmation. Overexpression was induced
by the addition of isopropyl-ꢁ-D-thiogalactopyranoside
(IPTG) at the mid-exponential growth phase and
incubation for 3 h at 37 ꢀC. The cells were harvested
by centrifugation (6;000 ꢁ g for 20 min at 4 ꢀC) and
resuspended in 50 mM Tris–HCl buffer (pH 8.0) con-
taining 0.1 M KCl and 10% glycerol. The cells were
disrupted by sonication and centrifuged (20;000 ꢁ g for
Enzyme assays. TNA1 CP activity was assayed by
measuring the hydrolysis of N-Cbz-Alanyl-Arginine (Z-
Ala-Arg) (Bachem AG, Bubendorf, Switzerland). The
assay mixture (250 ml) contained 50 mM KMES (potas-
sium 2-[N-morpholino]ethanesulfonic acid) buffer (pH
6.5), 0.4 mM CoCl2, and 8 mM substrate. The reaction
was initiated by addition of the enzyme and was
incubated at 80 ꢀC for 10 min, followed by quenching
on ice. Subsequently, 600 ml of cadmium-ninhydrin
reagent was added, and the samples were incubated for
an additional 5 min at 80 ꢀC for color development,
followed by quenching on ice.14) The absorbance was
read at 500 nm and the enzyme activity was calculated
using the arginine standard curve. One unit of activity
was defined as the amount of enzyme that liberates
1 mmol of arginine from Z-Ala-Arg per min at 80 ꢀC.
Activity assays for other Z-Ala-X amino acid substrates
(X = Ala, Asp, His, Leu, Met, Asn, Pro, Val, and Tyr)
were performed under otherwise identical assay con-
ditions. Appropriate amino acids were used to prepare
standard curves.
C-Terminal ladder sequencing of N-acetyl rennin
substrate with TNA1 CP. The sequencing reaction was
carried out with the following reaction mixture in a total
volume of 20 ml: 5 mM KMES buffer (pH 6.5), 50 mM

1142
H. S. LEE et al.
Fig. 1. Sequence Comparison of Peptidase Family M32.
Dashes indicate gaps and numbers on the right represent the position of the last residue in the original sequence. Identical residues among the
four enzymes are marked with *, and residues with conserved substitutions and semi-conserved substitutions are marked with : and, .
respectively. Conserved motifs appear at 238–240 (HPF), DXRXT (248–252), HESQ (298–301), HEXXH (269–273), IRXXAD (350–355), and
GXXQDXHW (405–412). Accession numbers with percent identity to carboxypeptidase from Thermococcus sp. NA1 (T. NA1) are as follows:
P. furiosus DSM 3638 (NP 578185), 84%, T. aquaticus YT-1 (P42663), 35% and T. thermophilus HB27 (YP 005684), 35%.
porcine N-acetyl-renin substrate (Sigma, St. Louis, MO),
0.2 mM CoCl2, and 150 nM TNA1 CP. The sample was
incubated at 80 ꢀC for 3 min, followed by quenching on
ice and the addition of 0.1% TFA. It was analyzed by
matrix-assisted laser desorption/ionization time-of-
flight (MALDI-TOF) mass spectrometry (Ultraflex,
Bruker Daltonik, Bremen, Germany), and the peaks
were assigned to within 0.3 Da of their monoisotopic
masses. The amino acid sequence of porcine N-acetyl-
renin substrate (MW 1801.3) was Ac-Asp-Arg-Val-Tyr-
Ile-His-Pro-Phe-His-Leu-Leu-Val-Tyr-Ser. The matrix
used was ꢀ-cyanohydroxycinnamic acid and the accel-
eration voltage was 25 kV.

Characterization of an Archaeal Carboxypeptidase
1143
Nucleotide sequence accession number. The reported
nucleotide sequence has been submitted to the Gen-
Bank/EMBL Data Bank under accession no. DQ144135.
Results and Discussion
Primary structure of the TNA1 CP gene and expres-
sion of the recombinant enzyme
Recently, we isolated a hyperthermophilic archaeon,
Thermococcus sp. NA1, growing at high temperatures
(70–90 ꢀC), and, by analyzing the genome sequence,
we found that the open reading frame, composed of
1,497 bp, encodes a protein homologous to thermostable
carboxypeptidase 1, which were purified and character-
ized from P. furiosus DSM 3638 (84% identity),15,16)
T. aquaticus YT-1 (35% identity),17) and T. thermophi-
lus HB27 (35% identity)3) (Fig. 1). The encoded protein
consists of 499 amino acids with a predicted molecular
mass of 59,198 Da and a pI of 5.61. It appeared that it is
an intracellular protein due to a lack of signal peptide,
which is consistent with other peptidase family M32.
Similarity analysis has shown the presence of CP-
homologous genes in genome sequences of other
archaea whose deduced amino acid sequences showed
fairly high identities to TNA1 CP (33 to 92%). In
particular, those from Thermococcales, including Ther-
mococcus kodakaraensis,18) Pyrococcus horikoshii,19)
and Pyrococcus abyssi,20) were quite similar to
TNA1 CP (83 to 92%). On the other hand, similarities
were lower between TNA1 CP and bacterial orthologs
(29 to 37% identity). The sequence alignment of
TNA1 CP and its homologs revealed that five motifs
in addition to the HEXXH motif, based on the data
obtained from P. furiosus CP,16) are well conserved,
except for one amino acid, A354, from the IRXXAD
exchanged with S354. The protein ligands (H269, H273,
and E299) and the proton shuttle (E270) forming the
active site of TNA1 CP are also conserved, hence the
protein might adopt a conformation similar to those
found in other peptidases containing the HEXXH motif.
The TNA1 CP gene was amplified with PCR and the
expressed enzyme was purified from a soluble cell
extract. Analysis by SDS/PAGE showed that the 61-
kDa protein (Fig. 2), which was the expected size of the
fusion product comprising the 59.2-kDa CP protein and
a 1.5-kDa peptide corresponding to the -KLAAALEH6-
(His6-Tag) at the C-terminal region of protein, was the
major component of the purified sample. Gel filtration
under non-denaturing conditions showed a protein peak
corresponding to a molecular mass of 125 kDa. These
data point to a dimeric structure for the enzyme (data not
shown).
Fig. 2. SDS–PAGE (12%) of the Purified Enzyme.
The molecular mass standards (lane M) were phosphorylase b
(103 kDa), bovine serum albumin (77 kDa), ovalbumin (50 kDa),
carbonic anhydrase (34.3 kDa), soybean trypsin inhibitor (28.8 kDa),
and lysozyme (20.7 kDa). The band corresponding to the enzyme is
indicated by the arrow.
temperature optimum at 70–80 ꢀC, but the activity
decreased rapidly above 80 ꢀC (Fig. 3A). The influence
of pH on TNA1 CP activity was evaluated by using
different buffers, sodium acetate (pH 4.0–5.0), KMES
(pH 5.5–6.5), HEPES (N-[2-hydroxyethyl]piperazine-
N0-[2-ethanesulfonic acid]) (pH 7–7.5), Tris–HCl (pH
8–10), and the optimum values were pH 6.5 (Fig. 3B). It
is worth noting that carboxypeptidase Taq from eubac-
teria T. aquaticus has an optimum pH of 8, but archaeal
P. furiosus CP and TNA1 CP have optimum pHs of 6.5.
As shown in Fig. 1, protein ligands for metal binding
were conserved in the HEXXH motif and the HESQ
sequence, implying that the activity of TNA1 CP is
influenced by the addition of metal ions. The addition of
Co2þ ions to the enzyme solution stimulated activity
3.7-fold, but Co2þ ions could not be replaced with other
divalent cations (Ba2þ, Ca2þ, Cu2þ, Fe2þ, Mg2þ, Mn2þ
,
Ni2þ, and Zn2þ) at a concentration of 0.2 mM (Fig. 4).
The control exhibited residual activity prior to the
addition of Co2þ, probably caused by the remaining
metal ions after buffer exchange or tight binding of
Co2þ. No CP activity was observed in the presence of a
metal chelating agent, EDTA (1 mM). Carboxypeptidase
Taq was activated by metal ions in the order Co2þ
>
2)
Ca2þ > Mg2þ > Cu2þ > Zn2þ
,
and T. thermophilus
CP required a divalent metal ion such as Zn2þ or
Co2þ for its activity.3) On the other hand, P. furiosus CP
were activated by Co2þ but not by Zn2þ,4) and the metal
requirement of TNA1 CP (data not shown) was similar
to that of P. furiosus CP.4) It has been reported that
P. furiosus CP and T. thermophilus CP are associated
with a variety of metals, such as Pb2þ, Yb2þ, Mg2þ and
Zn2þ in the crystal structures,3,16) and further investiga-
tion is needed to determine what is the biologically
relevant metal for those CPs, including TNA1 CP.
Based on the result that Cu2þ, Fe2þ, Ni2þ and Zn2þ
lowered the residual activity below 50% (Fig. 4),
TNA1 CP was treated with metal ions, and then Co2þ
Biochemical characterization of purified TNA1 CP
The functionality of TNA1 CP was confirmed by its
ability to hydrolyze Z-Ala-Arg, and this substrate was
used in all routine assays. Carboxypeptidase activity was
strongly stimulated at high temperatures and showed a

1144
H. S. LEE et al.
A
B
100
80
60
40
20
0
4
5
6
7
8
9
10
30
40
50
60
70
80
90
100
pH
o
Temperature ( C)
Fig. 3. The Effects of Temperature (A) and pH (B) on the Activities of TNA1 CP.
A, Activity assays were performed under standard conditions as the sample temperature was increased from 30 to 100 ꢀC. B, Activity assays
were performed under standard conditions with the following buffers (each at 50 mM): sodium acetate, pH 4.0–5.0; KMES, pH 5.5–6.5; HEPES,
pH 7.0–7.5; Tris–HCl, pH 8.0–10.0.
350
300
250
200
150
100
50
0
(-) Ba Ca Co Cu Fe Mg Mn Ni Zn Ba/ Ca/ Co/ Cu/ Fe/ Mg/ Mn/ Ni/ Zn/EDTA
Co Co Co Co Co Co Co Co Co
Metal ions
Fig. 4. Effects of Metal Ions on the Activities of TNA1 CP.
Activity assays were performed under standard conditions, while various divalent metal ions (0.2 mM) and EDTA (1 mM) alone or in
combination with Co2þ (0.2 mM) were added to the diafiltered enzyme solution. (ꢂ) means control without any metal ions.
was added to assess their inhibitory effect. As shown in
Fig. 4, the inhibitory activity of Cu2þ, Fe2þ, Ni2þ, and
better thermostability than P. furiosus CP (40 min).4)
The inactivation of TNA1 CP at 90 ꢀC was quite fast,
50% of the activity remaining after only 8 min. The
thermostability of TNA1 CP was highly affected by the
presence of Co2þ. In the absence of Co2þ in the reaction
during heating, the half-life values of TNA1 CP de-
creased to 7.4 min and 5.0 min at 80 ꢀC and 90 ꢀC
respectively, which means that the conformation of
TNA1 CP-Co2þ might be more stable (data not shown).
Without Co2þ, no activation of TNA1 CP by heat
treatment (at 80 ꢀC up to 20 min, data not shown) was
observed, which might explain the rapid inactivation of
the enzyme at 80 ꢀC. Cheng et al.4) suggested that
Zn2þ was not overcome by the presence of Co2þ
,
suggesting that the inhibitory metal ions bind TNA1 CP
tightly in place of Co2þ and perhaps accompany a
conformational change. On the other hand, Ba2þ and
Mn2þ had little effect, and Ca2þ and Mg2þ exhibited no
effect on TNA1 CP activity.
The thermostability of TNA1 CP was evaluated by
incubating the enzyme in 50 mM KMES buffer (pH 6.5)
containing 0.2 mM CoCl2 at 80 ꢀC and 90 ꢀC for up to
100 min. TNA1 CP lost enzymatic activity with a half-
life (t1=2) of 84 min at 80 ꢀC (Fig. 5), and thus exhibits

Characterization of an Archaeal Carboxypeptidase
1145
Table 1. Substrate Specificities of TNA1 CP and P. furiosus CP as
Determined with Z-Ala-X Substrates
100
Rate of hydrolysis (%)ꢃ with
Substrate
50
40
TNA1 CP
P. furiosus CP#
Z-Ala-Arg
Z-Ala-Leu
Z-Ala-Met
Z-Ala-Tyr
Z-Ala-Ala
Z-Ala-Val
100
62
49
24
16
12
100
27
61
12
14
6
30
20
10
ꢃAll values are percentages expressed relative to the rate of hydrolysis of Z-
Ala-Arg (100%) for each protein.
#Data from reference 4.
Table 2. Kinetic Parameters of TNA1 CPs and P. furiosus CP as
Determined with Z-Ala-Arg Substrateꢃ
1
0
20
40
60
80
100
Km
kcat
(sꢂ1
kcat=Km
(mMꢂ1 sꢂ1
)
Time (min)
Enzyme
TNA1 CP
TNA1 CP (no tag)
(mM)
)
Fig. 5. Effect of Temperature on TNA1 CP Stability.
TNA1 CP (0.8 mg/ml) was incubated at 80 ꢀC ( ) and 90 ꢀC (
in 50 mM KMES buffer, pH 6.5, containing 0.2 mM CoCl2. At the
times shown, aliquots were taken out, and the activities were
measured in the same buffer at 80 ꢀC using Z-Ala-Arg as substrate.
1.4
1.6
0.9
11.6
12.8
8.3
8.0
)
P. furiosus CP#
600
667
ꢃThe typical error in Km and kcat determinations did not exceed 13%.
#Data from reference 4.
P. furiosus CP might be stabilized in vivo by solutes
such as cyclic-2,3-diphosphoglycerate, di-myo-inositol-
1,10-phosphate, and mannosyl glycerate. However, in
vitro supplementation of 12.5% glycerol did not protect
TNA1 CP from inactivation at either temperature (data
not shown).
kinetic parameters such as Km and kcat values were
calculated from the measured activity for TNA1 CP
(Table 2). In comparison with those of P. furiosus CP
(Km ¼ 0:9 mM and kcat ¼ 600 sꢂ1), TNA1 CP exhibited
80-fold lower catalytic efficiency toward Z-Ala-Arg. To
investigate the possibility of the His6-Tag at the C-
terminus of TNA1 CP affecting affinity or activity
toward the substrate, TNA1 CP protein having no
His6-Tag, TNA1 CP (no tag), was prepared, and kinetic
constants were obtained (Table 2). TNA1 CP (no tag)
gave a virtually identical result, indicating that the
decrease in catalytic efficiency of TNA1 CP was not
caused by the presence of a His6-Tag at the C-terminus.
It is not yet understood why the two enzymes showing
high identity (84%) were different in catalytic efficiency.
The oligomeric status of TNA1 CP was not different
from that of P. furiosus CP. Amino acids for dimer
formation in P. furiosus CP16) were also present in
TNA1 CP, and the dimeric structure was confirmed by
gel filtration column chromatography (125 kDa on a
Superdex 200, data not shown). In addition, as discussed
above, most of the critical residues for carboxypeptidase
activity were found in TNA1 CP: an HPF sequence, a
DXRXT sequence, an HESQ sequence, and a GXXQ-
DXHW sequence in addition to the characteristic metal-
loprotease HEXXH motif. One possible candidate to
explain the decrease in catalytic efficiency is S354, since
an A354 in the IRXXAD sequence was changed to S354
in TNA1 CP. Arndt et al. (2002)16) proposed that the
IRXXAD and GXXQDXHW signature sequences might
work in concert to bind the substrate C-terminal carbox-
ylate and promote release of the cleaved amino acid
TNA1 CP exhibited relatively broad specificity with
a preference toward basic, aliphatic, and aromatic C-
terminal amino acids (Table 1). No activity was detected
under standard conditions against Z-Ala-Asp, Z-Ala-
Pro, Z-Ala-His, or Z-Ala-Asn among the substrates
tested. In comparison with P. furiosus CP,4) the prefer-
ence was in a similar pattern, but not identical, viz.,
Arg > Met > Leu > Ala > Tyr > Val for P. furiosus
CP, and Arg > Leu > Met > Tyr > Ala > Val for
TNA1 CP1. According to Umetsu et al.,21) the hydrol-
ysis rates were strikingly affected by amino acids in the
penultimate position from the C-termini of the sub-
strates. Therefore, other amino acids than Ala in the Z-
Ala-X (X = Asp, Pro, His, and Asn) substrate might
enhance the cleavage of substrates. Many carboxypepti-
dases exhibiting broad substrate specificities have been
reported.22) On the basis of preferences for amino acids
at the C-terminus, Penicillium janthinellum CP,23)
a
member of the serine carboxypeptidases, and carboxy-
peptidase T from Thermoactinomyces sp.24) cleaved the
peptide bonds formed by basic and hydrophobic C-
terminal amino acid residues, whereas carboxypeptidase
A and carboxypeptidase C, and carboxypeptidase B and
carboxypeptidase D exhibited strong preferences for
hydrophobic and basic residues respectively.1,25)
Kinetic analysis was conducted using Z-Ala-Arg, and

1146
H. S. LEE et al.
Fig. 6. MALDI-TOF Spectrum of C-Terminal Ladder Sequencing of N-Acetyl-renin Substrate by TNA1 CP.
Fifty mM N-acetyl-renin substrate was digested in 5 mM KMES buffer (pH 6.5) containing 0.2 mM CoCl2 and 150 nM TNA1 CP at 80 ꢀC for
3 min, and the sample was analyzed by MALDI-TOF mass spectrometry.
product. Nevertheless, it is not certain how the change of
A354 to S354 can effect the decrease in kinetic value,
and this should be investigated further. In addition,
further experiments are required to determine whether
the enzyme was folded properly during purification from
E. coli or exhibited non-ideal behavior.
Maritime Affairs and Fisheries, of the Republic of
Korea.
References
1) Rawlings, N. D., and Barrett, A. J., Evolutionary
families of metallopeptidases. Methods Enzymol., 248,
183–228 (1995).
2) Lee, S. H., Minagawa, E., Taguchi, H., Matsuzawa, H.,
Ohta, T., Kaminogawa, S., and Yamauchi, K., Purifica-
tion and characterization of a thermostable carboxypep-
tidase (carboxypeptidase Taq) from Thermus aquaticus
YT-1. Biosci. Biotechnol. Biochem., 56, 1839–1844
(1992).
3) Nagata, K., Tsutsui, S., Lee, W. C., Ito, K., Kamo, M.,
Inoue, Y., and Tanokura, M., Crystallization and
preliminary X-ray analysis of carboxypeptidase 1 from
Thermus thermophilus. Acta Crystallogr. D Biol. Crys-
tallogr., 60, 1445–1446 (2004).
C-Terminal ladder sequencing of the peptide
Sequential C-terminal hydrolysis of the peptide N-
acetyl-renin substrate by TNA1 CP, followed by MAL-
DI-TOF mass detection, offered unambiguous proof of
carboxypeptidase activity (Fig. 6). Under the reaction
conditions, TNA1 CP sequenced up to seven residues
from the C-terminus of substrate, releasing Ser, Tyr,
Val, Leu, His, and Phe residues. These sequencing
experiments confirmed the broad specificity of
TNA1 CP demonstrated in the experiments with ZAX
dipeptides.
4) Cheng, T. C., Ramakrishnan, V., and Chan, S. I.,
Purification and characterization of a cobalt-activated
carboxypeptidase from the hyperthermophilic archaeon
Pyrococcus furiosus. Protein Sci., 8, 2474–2486 (1999).
5) Stepanov, V. M., Carboxypeptidase T. Methods Enzy-
mol., 248, 675–683 (1995).
In this study, we characterized a thermostable
carboxypeptidase from a hyperthermophilic archaeon,
Thermococcus sp. NA1. The TNA1 CP can be applied to
C-terminal sequencing of peptides, and practical appli-
cation is in progress.
6) Colombo, S., D’Auria, S., Fusi, P., Zecca, L., Raia, C.
A., and Tortora, P., Purification and characterization of
a thermostable carboxypeptidase from the extreme
thermophilic archaebacterium Sulfolobus solfataricus.
Eur. J. Biochem., 206, 349–357 (1992).
7) Hartsuck, J. A., and Lipscomb, W. N., Carboxypeptidase
A. In ‘‘The Enzymes’’ 3rd ed., Vol. 3, ed. Boyer, P. D.,
Academic Press, New York, pp. 1–56 (1971).
Acknowledgments
We would like to thank Dr. J. H. Hong and Mr. S. C.
Shin of Macrogen, Inc., for valuable information. This
work was been supported by the KORDI in-house
program (PE97201) and the Marine and Extreme
Genome Research Center Program of the Ministry of
8) Folk, J. E., Carboxypeptidase B. In ‘‘The Enzymes’’ 3rd

Characterization of an Archaeal Carboxypeptidase
1147
ed., Vol. 3, ed. Boyer, P. D., Academic Press, New York,
pp. 57–59 (1971).
9) Levin, Y., Skidgel, R. A., and Erdos, E. G., Isolation and
characterization of the subunits of human plasma
carboxypeptidase N (kininase i). Proc. Natl. Acad. Sci.
USA, 79, 4618–4622 (1982).
10) Fricker, L. D., and Snyder, S. H., Enkephalin convertase:
purification and characterization of a specific enkepha-
lin-synthesizing carboxypeptidase localized to adrenal
chromaffin granules. Proc. Natl. Acad. Sci. USA, 79,
3886–3890 (1982).
11) Sonnleitner, B., and Fiechter, A., Advantages of using
thermophiles in biotechnological processes: expectations
and reality. Trends Biotechnol., 1, 74–80 (1983).
12) Cowan, D., Daniel, R., and Morgan, H., Thermophilic
proteases: properties and potential applications. Trends
Biotechnol., 3, 68–72 (1985).
S., and Imanaka, T., Complete genome sequence of the
hyperthermophilic archaeon Thermococcus kodakaraen-
sis KOD1 and comparison with Pyrococcus genomes.
Genome Res., 15, 352–363 (2005).
19) Kawarabayasi, Y., Sawada, M., Horikawa, H., Haikawa,
Y., Hino, Y., Yamamoto, S., Sekine, M., Baba, S.,
Kosugi, H., Hosoyama, A., Nagai, Y., Sakai, M., Ogura,
K., Otsuka, R., Nakazawa, H., Takamiya, M., Ohfuku,
Y., Funahashi, T., Tanaka, T., Kudoh, Y., Yamazaki, J.,
Kushida, N., Oguchi, A., Aoki, K., and Kikuchi, H.,
Complete sequence and gene organization of the genome
of a hyper-thermophilic archaebacterium, Pyrococcus
horikoshii OT3. DNA Res., 30, 55–76 (1998).
20) Cohen, G. N., Barbe, V., Flament, D., Galperin, M.,
Heilig, R., Lecompte, O., Poch, O., Prieur, D.,
Querellou, J., Ripp, R., Thierry, J. C., Van der Oost, J.,
Weissenbach, J., Zivanovic, Y., and Forterre, P., An
integrated analysis of the genome of the hyperthermo-
philic archaeon Pyrococcus abyssi. Mol. Microbiol., 47,
1495–1512 (2003).
13) Sambrook, J., and Russell, D. W., Molecular cloning: a
laboratory manual, 3rd ed., Cold Spring Harbor Labo-
ratory, Cold Spring Harbor (2001).
14) Doi, E., Shibata, D., and Matoba, T., Modified colori-
metric ninhydrin methods for peptidase assay. Anal.
Biochem., 118, 173–184 (1981).
15) Robb, F. T., Maeder, D. L., Brown, J. R., DiRuggiero, J.,
Stump, M. D., Yeh, R. K., Weiss, R. B., and Dunn, D.
M., Genomic sequence of hyperthermophile, Pyrococcus
furiosus: implications for physiology and enzymology.
Meth. Enzymol., 330, 134–157 (2001).
16) Arndt, J. W., Hao, B., Ramakrishnan, V., Cheng, T.,
Chan, S. I., and Chan, M. K., Crystal structure of a novel
carboxypeptidase from the hyperthermophilic archaeon
Pyrococcus furiosus. Structure (Cambridge), 10, 215–
224 (2002).
21) Umetsu, H., Hishinuma, K., Wake, H., and Ichishima,
E., Production, purification, and properties of serine
carboxypeptidase from Paecilomyces carneus. Curr.
Microbiol., 33, 44–48 (1996).
22) Colombo, S., D’Auria, S., Fusi, P., Zecca, L., Raia, C.
A., and Tortora, P., Purification and characterization of
a thermostable carboxypeptidase from the extreme
thermophilic archaebacterium Sulfolobus solfataricus.
Eur. J. Biochem., 206, 349–357 (1992).
23) Breddam, K., Carboxypeptidase S-1 from Penicillium
janthinellum: enzymatic properties in hydrolysis and
aminolysis reactions. Carlsberg Res. Commun., 53, 309–
320 (1988).
17) Lee, S. H., Taguchi, H., Yoshimura, E., Minagawa, E.,
Kaminogawa, S., Ohta, T., and Matsuzawa, H., Carboxy-
peptidase Taq, a thermostable zinc enzyme, from
Thermus aquaticus YT-1: molecular cloning, sequenc-
ing, and expression of the encoding gene in Escherichia
coli. Biosci. Biotechnol. Biochem., 58, 1490–1495
(1994).
24) Osterman, A. L., Stepanov, V. M., Rudenskaia, G. N.,
Khodova, O. M., and Tsaplina, I. A., Carboxypeptidase
T-intracellular carboxypeptidase of Thermoactinomy-
cetes: a distant analog of animal carboxypeptidase.
Biokhimiia, 49, 292–301 (1984).
25) Remington, S. J., and Breddam, K., Carboxypeptidases
C and D. Methods Enzymol., 244, 231–248 (1994).
18) Fukui, T., Atomi, H., Kanai, T., Matsumi, R., Fujiwara,