Site-directed mutation at residues near the catalytic site of histamine dehydrogenase from Nocardioides simplex and its effects on substrate inhibition

Article abstract of DOI:10.1093/jb/mvp162

Histamine dehydrogenase from Nocardioides simplex (nHmDH) is a homodimer containing one 6-S-cysteinyl FMN (CFMN) and one [4Fe-4S] cluster per monomer. nHmDH catalyses the oxidative deamination of histamine to ammonia and imidazole acetaldehyde, but histamine inhibits its catalytic activity at high concentrations. We mutated gene-encoded residues (Tyr180, Gly268 and Asp269) near CFMN to understand the biophysical meaning of the substrate inhibition. Three mutants Y180F, G268D/D269C and Y180F/G268D/D269C were expressed by considering the DNA sequence alignment of histamine dehydrogenase from Rhizobium sp. 4-9 (rHmDH), which does not suffer from the substrate inhibition. The Y180F/G268D/D269C mutation to mimic rHmDH successfully suppressed the inhibition, although the catalytic activity decreased. The substrate inhibition was weakened by the Y180F mutation, but G268D/D269C was still susceptible to the inhibition. It was found that it also causes changes in the UV-vis absorption spectra of the substrate-reduced form and the redox potential of the enzymes. The characterization suggests that the thermodynamic preference of the semiquinone form of CFMN in the two-electron-reduced subunit of the enzyme is responsible for the substrate inhibition. However, destabilization of the semiquinone form leads to kinetic hindrance due to the uphill single electron transfer from the fully reduced CFMN to the [4Fe-4S] cluster.

Full text of DOI:10.1093/jb/mvp162

J. Biochem. 2010;147(2):257–264 doi:10.1093/jb/mvp162
Site-directed mutation at residues near the catalytic site of
histamine dehydrogenase from Nocardioides simplex and
its effects on substrate inhibition
Received September 16, 2009; accepted October 4, 2009; published online October 20, 2009
Maiko Tsutsumi, Noriaki Tsuse,
Nobutaka Fujieda* and Kenji Kano
simplex; rHmDH, histamine dehydrogenase from
Rhizobium sp. 4—9; TMADH, trimethylamine
dehydrogenase.
y
Division of Applied Life Sciences, Graduate School of Agriculture,
Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
*
Present address: Department of Material and Life Science,
Histamine dehydrogenase from Nocardioides simplex
Division of Advanced Science and Biotechnology, Graduate School
of Engineering, Osaka University, Suita, Osaka 565-0871, Japan.
(nHmDH) is an iron—sulfur-containing flavoprotein.
y
nHmDH catalyses the oxidative deamination of hista-
mine to ammonia and imidazole acetaldehyde (1). The
enzyme is a homodimeric protein composed of sub-
units with a molecular mass of 76 kDa. Each subunit
contains a covalently linked 6-S-cysteinyl flavin
mononucleotide cofactor (CFMN) and a [4Fe—4S]
iron—sulfur cluster. Each subunit also possesses one
adenosine diphosphate of unknown function (2—4).
The DNA sequence alignment shows that nHmDH is
closely related to histamine dehydrogenase from
Rhizobium sp. 4—9 (rHmDH; 64% identical) and
trimethylamine dehydrogenase (TMADH; 40% identi-
cal), both of which also contain one CFMN and one
Kenji Kano, Division of Applied Life Sciences, Graduate School of
Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan.
Tel: þ81-75-753-6392, Fax: þ81-75-753-6456,
E-mail: kkano@kais.kyoto-u.ac.jp
Histamine dehydrogenase from Nocardioides simplex
(nHmDH) is a homodimer containing one 6-S-cysteinyl
FMN (CFMN) and one [4Fe—4S] cluster per monomer.
nHmDH catalyses the oxidative deamination of
histamine to ammonia and imidazole acetaldehyde,
but histamine inhibits its catalytic activity at high
concentrations. We mutated gene-encoded residues
(
Tyr180, Gly268 and Asp269) near CFMN to under-
stand the biophysical meaning of the substrate inhibi-
tion. Three mutants Y180F, G268D/D269C and
Y180F/G268D/D269C were expressed by considering
the DNA sequence alignment of histamine dehydroge-
nase from Rhizobium sp. 4-9 (rHmDH), which does
not suffer from the substrate inhibition. The Y180F/
G268D/D269C mutation to mimic rHmDH success-
fully suppressed the inhibition, although the catalytic
activity decreased. The substrate inhibition was weak-
ened by the Y180F mutation, but G268D/D269C
was still susceptible to the inhibition. It was found
that it also causes changes in the UV-vis absorption
spectra of the substrate-reduced form and the redox
potential of the enzymes. The characterization suggests
that the thermodynamic preference of the semiquinone
form of CFMN in the two-electron-reduced subunit of
the enzyme is responsible for the substrate inhibition.
However, destabilization of the semiquinone form
leads to kinetic hindrance due to the uphill single elec-
tron transfer from the fully reduced CFMN to the
[
4Fe—4S] iron—sulfur cluster per subunit (3,4).
All these iron—sulfur-containing flavoproteins
require three electrons per subunit for full reduction;
two for reduction of CFMN and one for the reduction
of the [4Fe—4S] cluster. It has been considered that in
these enzymes two electrons in the substrate are passed
to the oxidized CFMN (CFMN ) and two sequential
O
single electron transfers occur from the fully reduced
CFMN (CFMN ) to the oxidized [4Fe—4S] cluster
R
(
FeS ), and finally another two sequential single elec-
O
tron transfers are followed from the reduced [4Fe—4S]
cluster (FeS ) to electron acceptor. The physiological
electron acceptor is electron transferring flavoprotein
R
(
ETF) for TMADH (5), but not identified for nHmDH
and rHmDH. Several artificial electron acceptors
such as phenazine methosulfate (with dichloropheno-
lindophenol) (6), phenazine ethosulfate (PES) (with
DCIP) (7) and ferricenium hexafluorophosphate
(Fc ) (8) may be utilized in place of the physiological
one. Even artificial electron acceptor cannot receive
the electron from the reduced flavin site at least in
nHmDH and TMADH (8). The enzyme kinetics is
strongly influenced by the electron acceptor used. In
this article, the argument on the effect of the electron
acceptor will be minimized for simplicity.
þ
[
4Fe—4S] cluster.
Keywords: histamine dehydrogenase/redox potential/
substrate inhibition/6-S-cysteinyl flavin mononucleo-
tide/[4Fe—4S] iron-sulfur cluster.
Abbreviations: CFMN, 6-S-cysteinyl flavin mononu-
cleotide; CFMN , the oxidized form of CFMN;
The unique redox properties of nHmDH are
observed during the reductive titration with histamine.
In the reductive titration with histamine, two-electron
reduction occurs per subunit of the enzyme at pH59,
while single-electron reduction occurs at pH49 (9).
In the following, we will define the number of elec-
tron and the amount of histamine added as those
O
CFMN , the semiquinone form of CFMN; CFMN ,
R
S
the fully reduced form of CFMN; FeS , the oxidized
O
form of [4Fe—4S] iron—sulfur cluster; FeS , the
R
reduced form of [4Fe—4S] iron—sulfur cluster;
nHmDH, histamine dehydrogenase from Nocardioides
ß The Authors 2009. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
257

M. Tsutsumi et al.
per subunit. Therefore, the enzyme reduced with one
equivalent amount of histamine per subunit corre-
sponds to the enzyme composed of the two-electron-
reduced subunits. The two-electron-reduced form
of nHmDH subunit is in equilibrium between two ulti-
mate states: one involves the semiquinone form of
(closed circles, Fig. 1). The amino acid in TMADH
and nHmDH of 180 position is conserved, but Phe
in rHmDH. In addition, the position of Asp265
in rHmDH, which perhaps forms a catalytic triad,
is different from that of the corresponding one in
TMADH (Asp267) and nHmDH (Asp269). In order
to study whether the differences are responsible for
the substrate inhibition characteristics, three mutant
proteins of nHmDH involving a single mutation
(Y180F), double mutation (G268D/D269C) and
triple mutation (Y180F/G268D/D269C) as well as
recombinant wild-type (WT) were expressed and char-
acterized in view of enzymatic kinetics, spectroscopy
and electrochemistry.
CFMN (CFMN ) and FeSR and the other involves
S
CFMNR and FeSO (9). The distribution is controlled
by individual redox potentials which are the function
of pH (9).
Both nHmDH and rHmDH show high substrate
specificity towards histamine. Secondary and tertiary
amines including biogenic amines such as tyramine
do not react with the enzymes (1, 10). Therefore, the
enzymes can be utilized in histamine analysis in food
and clinical analysis (11—13). nHmDH is much higher
in the catalytic activity than rHmDH, but is strongly
susceptible to substrate inhibition from histamine
at high concentrations. Similar substrate inhibition is
also observed for TMADH, which is inhibited by
its natural substrate trimethylamine (7, 8, 14). In con-
trast, rHmDH does not suffer from such substrate
inhibition (10).
Experimental procedures
Mutagenesis and expression of nHmDH mutants
Techniques for restriction enzyme digestions, ligation transforma-
tion and other standard molecular biology manipulations were
based on the methods previously described (19) and in part in accor-
dance with the manufacturer’s instructions.
The DNA fragment coding for Y180F mutant enzyme (Y180F)
was generated by the Quikchange method using plasmid pGEM-
hmd (3, 19) as a template for KOD-plus polymerase with the primers
X-ray crystallographic analysis and sequence align-
ment have revealed that the amino acid residues near
the catalytic site responsible for the substrate recogni-
tion and the catalytic reaction are highly conserved,
but some residues are different from one another
0
0
(
and (reverse) 5 -GACGCCGCTGAAGCCGTGTGCGCCGTA
forward) 5 -GTGTACGGCGCACACGGCTTCAGCGGCGTC-3
0
0
CAC-3 . The DNA fragments coding for G268D/D269C mutant
enzyme (G268D/D269C) and for Y180F/G268D/D269C mutant
enzyme (Y180F/G268D/D269C) were generated by the Quik-
change method using plasmid pGEM-hmd and pGEM-hmdY180F
as templates, respectively, for KOD-plus DNA polymerase with the
(
Fig. 1). In a previous paper, we suggested the impor-
tance of three amino acid residues: Trp266 (aromatic
bowl which accommodates the substrate by cation-p
bonding (15), Ser270 and Phe567, for the substrate
recognition of nHmDH from the homology modelling
of the 3D structure (3). The imidazole ring of hista-
mine might be stabilized via hydrogen-bond formation
with the hydroxyl group of Ser270 and p interaction
with the aromatic groups of Trp266 and Phe567.
rHmDH has the corresponding Trp and Ser residues
0
primers (forward) 5 -GGCAGCTGGGAGGACTGCTCCGTCAC
0
0
GTCC-3 and (reverse) 5 -GGACGTGACGGAGCAGTCCTCCC
0
AGCTGCC-3 . The final PCR products were gel purified and
digested with Dpn I. Each of the purified PCR products was digested
with Nde I and Hind III and inserted into pET-26b(þ) (Novagen) to
yield the expression construct pET-hmdY180F, pET-hmd/G268D/
D269C or pET-hmdY180F/G268D/D269C, respectively.
Expression and purification of nHmDH mutants
The nHmDH mutants were expressed in Escherichia coli strain
Rosetta (DE3) and purified as described previously (19). Each pro-
tein concentration was determined using a modified Lowry method
with DC Protein Assay Kit (Bio-Rad, USA) with bovine serum albu-
min as the standard protein. All chemicals used in this study were
of analytical reagent grade and all solutions were prepared with
distilled water.
(Fig. 1, Trp263 and Ser267), but only the correspond-
ing Phe residue is replaced with Leu567. It has been
proposed that Tyr169, His172 and Asp267 form a
catalytic triad in TMADH (16—18), and the three
amino acids were conserved in nHmDH (Fig. 1,
boxed). It has been shown that the catalytic center of
nHmDH considerably resembles that of rHmDH,
but there are some differences especially at the 180,
UV-vis spectroscopy
Measurements were performed with a quartz cuvette of a 1-cm light
ꢀ
path at a final volume of 2.5 ml under anaerobic conditions at 30 C
2
68 and 269 positions in the vicinity of CFMN
using a water-jacket cell holder and a thermostat. UV-vis spectra
were obtained on a Shimadzu UV-2500PC UV-vis recording spectro-
photometer (Japan). Histamine dihydrochloride, agmatine sulfate
and putrescine were purchased from Sigma, Wako and Nacalai,
respectively, and used as substrates. The nHmDH solutions were
prepared at concentrations of 2.6ꢁ3.5 mM in a 100 mM potassium
phosphate buffer, pH 7.5. The nHmDH solutions were kept anaer-
174
170
168
265
262
263
nHmDH
rHmDH
TMADH
VYGAHGYSGV
lYGAHGFGIF
VYGAHSYL-P
SWEGDSVTSRF
TWEDCSGPSRF
EWGEDAGPSRF
2
obic by repeated evacuation and flushing with O -free moisturized
*
* **
*
***
argon. The spectra of the reduced forms were taken 3, 15 and 60 min
after the addition of histamine, agmatine and putrescine.
Fig. 1 Amino acid sequence alignment around CFMN of nHmDH,
rHmDH and TMADH. nHmDH, Nocardioides simplex histamine
degydrogenase; rHmDH, Rhizobium sp. 4-9 histamine dehydrogen-
Steady-state kinetic measurement
Measurements were performed with a quartz cuvette of a 1-cm light
ase; TMADH, bacterium W
3
A
1
trimethylamine dehydrogenase.
ꢀ
Asterisks indicate amino acid residues which are conserved among
the aligned sequences. Closed triangles indicate the amino acid
residues concerning substrate specificity and the catalytic reaction,
which were reported in references (3,10 and 18). Putative catalytic
triad is boxed. Closed circles indicate the amino acid residues
mutated in nHmDH and its homologous ones in rHmDH and
TMADH.
path at a final volume of 3 ml under anaerobic conditions at 30 C
using a water-jacket cell holder and a thermostat. Fc was obtained
þ
from Aldrich and used as an artificial electron acceptor, because its
redox potential is sufficiently positive than that of the enzyme. Each
of stock solutions of substrate and Fc was added to the assay mix-
þ
ture to reach desired concentrations. Assays for the determination of
kinetic parameters were performed in a 0.1 M potassium phosphate
2
58

Substrate inhibition of histamine dehydrogenase
buffer of pH 7.5. The reactions were initiated by the addition of the
enzymes and the decrease in the absorbance at 617 nm due to the
reduction of Fc was measured using a Shimadzu UV-2500PC UV-
200
A
þ
vis recording spectrophotometer (Japan) (20). The enzyme kinetics
was analysed by a non-linear least-squares method in accordance
with the following equation (21):
1
1
50
v
kcat
00
¼
ð1Þ
½
Eꢂ
KS ½Sꢂ
1
þ
þ
½
Sꢂ Ki
50
where kcat, KS and Ki are the catalytic constant, the Michaelis
constant of the substrate and the inhibition constant of the corre-
sponding substrate, respectively, and were optimized as adjustable
parameters to fit Eq. (1) to the data. [E] and [S] denote the concen-
tration of the enzyme and substrate, respectively.
0
0
250
500
750
1000
[histamine] / µM
Spectroelectrochemistry
Mediated spectroelectrochemical titrations for the redox potential
measurement of the enzymes were performed by controlling the
solution potential in a quartz cuvette using a Hokuto Denko
HSV-100 potentiostat (Japan). Absorption spectra were recorded
simultaneously on a Shimadzu UV-2500PC UV-vis spectropho-
tometer as described previously (9, 22). Two osmium complexes,
B
10
5
[
acid) (bipyridylamine)
Os(bipyridylamine)
2
Cl
2
](PF and [Os(4-imidazole carboxylic
2
6 2
](PF ) , were synthesized according to the
6
)
2
2
literature (23) and used as mediators of the redox titration. The
electrolysis solution was a 0.1 M Britton and Robinson buffer con-
taining 5—10 mM nHmDH, 35 mM [Os(bipyridylamine)
and 43 mM [Os(4-imidazole carboxylic
bipyridylamine)
2
Cl
2
](PF
6
)
acid)
2
2
(
2
](PF
6
)
2
, pH 7.5, at a total volume of 1.7 ml. An
antifoaming reagent (Antifoam PE-L, Wako, Japan) was added to
the electrolyte solution to a final concentration of 0.3% (w/w). When
the spectra became independent of the electrolysis time (i.e. when the
solution potential became identical with the electrode potential),
the UV-vis spectra were measured. The electrode potential was
changed stepwise at several given values. All these experiments
on spectroscopy and electrochemistry were performed in a COY
0
0
500
1000
1500
2000
[
agmatine] / µM
3
C
Laboratory Products Model A glove box (USA). The concentration
ꢀ
of O
2
in the glove box was maintained less than 10 ppm at 30 C.
2
1
0
Cyclic voltammetry
Cyclic voltammetry was carried out on a three-electrode system
using a BAS CV-50 W electrochemical analyser (Japan) under anaer-
obic conditions. A platinum wire and Ag|AgCl|saturated KCl were
used as reference and counter electrodes, respectively. The working
electrodes used were indium tin oxide (ITO) electrodes (5 ꢃ5 mm,
BAS) (9). The electrolysis solution was a 0.1 M Britton and
Robinson buffer containing 50—70 mM nHmDH and 0.2 M NaCl,
pH 7.5, at a total volume of 1 ml. All potentials are referred to the
standard hydrogen electrode in this article.
0
5
10
15
20
[
putrescine] / mM
Fig. 2 Steady-state kinetic analysis of the reaction of WT, Y180F,
G268D/D269C and Y180F/G268D/D269C nHmDH. (A) Histamine,
Results
(
B) agmatine and (C) putrescine. Closed circles, data for recombi-
Enzyme kinetics
nant wild-type nHmDH; open circles, data for Y180F; closed
triangles, data for G268D/D269C; open triangles, data for Y180F/
G268D/D269C. Curves were fitted using Eq. (1) representing
uncompetitive inhibition.
nHmDH is highly selective to primary amine sub-
strates. Histamine is the best substrate but very
small reactivity is observed for agmatine and putres-
cine (1, 2, 4). Figure 2A shows the histamine concen-
tration dependence of the steady-state enzyme kinetics
(
v) for WT and the three mutants (Y180F, G268D/
D269C and Y180F/G268D/D269C) in the presence
susceptible to substrate inhibition giving a waved pat-
tern dependence on the concentration similar to that of
WT. Such substrate inhibition was partially weakened
for the Y180F mutant (open circle) and almost
completely suppressed in the Y180F/G268D/D269C
mutant (open triangle), although the enzymatic activ-
ity itself was decreased. Agmatine and putrescine are
poor substrates, and they exhibited standard satura-
tion dependence of the kinetics on the substrate
concentration for WT and the three mutants (Fig. 2B
and C). The dotted lines in Fig. 2 represent the regres-
sion curves, and the refined enzyme kinetic parameters
þ
of Fc as an artificial electron acceptor at pH 7.5
and 30 C. The enzyme kinetics exhibited a standard
Michaelis—Menten-like dependence on the Fc con-
centration (data not shown). For simplicity (that is,
in order to minimize the effect of the electron
acceptor), the Fc concentration was kept as excess
ꢀ
þ
þ
(0.33 mM) in the following. On the other hand, hista-
mine showed strong substrate inhibition against WT,
as shown by the closed circle. The G268D/D269C
mutant (indicated by the closed triangle) was also
2
59

M. Tsutsumi et al.
a
Table 1. Steady-state enzyme kinetic parameters for wild-type and mutant nHmDHs with various amine substrates .
WT
Y180F
G268D/D269C
Y180F/G268D/D269C
Histamine
kcat s
KS mmol
kcat/KS s mmol
Ki mmol
ꢁ1
2
2
2
2
2
2.5 (ꢄ0.1) ꢃ10
33 ꢄ2
2.0 (ꢄ0.1) ꢃ10
1.2 (ꢄ0.1) ꢃ10
1.5 (ꢄ0.1) ꢃ10
50 ꢄ1
8.9 (ꢄ0.1) ꢃ10
76 ꢄ1
ꢁ
1
ꢁ1
ꢁ1
ꢁ1
b
b
b
b
7.2 ꢄ0.1 (100)
5.8 (ꢄ0.1) ꢃ10
1.6 ꢄ0.2 (100)
8.3 (ꢄ0.4) ꢃ10
2.9 ꢄ0.1 (100)
5.9 (ꢄ0.1) ꢃ10
1.2 ꢄ0.1 (100)
2
2
1
Agmatine
kcat s
ꢁ1
5.3 ꢄ0.2
12 ꢄ1
3.7 ꢄ0.8
3.5 ꢄ0.6
2
2
2
2
K
S
mmol
2.4 (ꢄ0.1) ꢃ10
2.8 (ꢄ0.1) ꢃ10
1.2 (ꢄ0.1) ꢃ10
1.9 (ꢄ0.1) ꢃ10
ꢁ1
ꢁ2
b
ꢁ2
b
ꢁ2
b
ꢁ2
b
kcat/KS s mmol
Ki mmol
2.2 (ꢄ0.1) ꢃ10 (0.31)
4.3 (ꢄ0.3) ꢃ10 (2.7)
3.1 (ꢄ0.8) ꢃ10 (1.1)
1.8 (ꢄ0.2) ꢃ10 (1.5)
1
1
1
1
Putrescine
kcat s
ꢁ1
ꢁ1
2.8 ꢄ0.1
1.9 ꢄ0.1
2.8 ꢄ0.6
7.8 (ꢄ0.3) ꢃ10
3
3
3
3
KS mmol
kcat/KSS mmol
Ki mmol
4.7 (ꢄ0.5) ꢃ10
2.5 (ꢄ0.1) ꢃ10
3.1 ꢃ10
6.1(ꢄ0.2) ꢃ10
ꢁ
1
ꢁ4
b
ꢁ4
b
ꢁ4
b
ꢁ4
b
6.0 (ꢄ0.4) ꢃ10 (0.0083)
7.6 (ꢄ0.4) ꢃ10 (0.048)
9.0 (ꢄ2.0) ꢃ10 (0.031)
1.3 (ꢄ0.1) ꢃ10 (0.011)
1
1
1
1
a
ꢀ
þ
Reactions were performed in a 0.1 M potassium phosphate buffer, pH 7.5 at 30 C, and were initiated by the addition of Fc .
þ
b
Fc concentration was 0.33 mM. Numbers in parentheses represent the % (kcatKS ) versus histamine.
ꢁ1
are summarized in Table 1. In the case of TMADH
also, a pure substrate, diethylmethylamine, does
not inhibit the enzyme activity even at high
concentrations (8, 24).
the maximum at around 360 nm (9). In the reductive
half-reaction of the WT enzyme with histamine, two-
electron reduction occurs per subunit at pH59 (9),
and the two-electron-reduced WT gave the absorption
bands at 365 and 445 nm at pH 7.5 (Fig. 3A, solid line),
The inhibitory characteristics are well expressed
by K . Both WT and G268D/D269C species have
i
indicating the presence of CFMN . However, the
S
small K . The site-directed mutation of Tyr180 to Phe
i
absorption band (shoulder) at 445 nm was weak
compared with that observed for one ultimate state
containing CFMNS predominantly over CFMNR.
This means that part of the two-electron-reduced
form of WT is in the other ultimate state containing
CFMNR.
seems to be effective to reduce the substrate inhibi-
tion as observed as an increase in the Ki value in the
Y180F mutation and more drastically in the Y180F/
G268D/D269C mutation. However, all these mutation
cause a decrease in the bimolecular rate constant
between the free enzyme and the substrate, as judged
On the addition of excess histamine (20-equivalent),
the absorption bands at 365 and 445 nm were slightly
sharpened, and the small broad absorption band
appeared at around 500 nm (Fig. 3A, dashed line).
All these results indicate an equilibrium shift from
the CFMN -FeS state to the CFMN -FeS state
ꢁ1
from the kcat KS values for histamine (Table 1). The
kcat value is also an important parameter which repre-
sents the rate constant of the forward decomposition
of enzyme—substrate complex as well as enzyme—
electron acceptor complex. The triple mutation as
Y180F/G268D/D269C decreased kcat significantly.
R
O
S
R
under the conditions.
ꢁ
1
The comparison of the kcat KS values for three sub-
strates reveals that the mutation also induces the
change in the substrate specificity. WT showed very
high substrate specificity towards histamine as judged
In the case of agmatine and putrescine, the absorp-
tion spectra of the two-electron-reduced form of WT
did not change by the addition of excess substrate
(20-equivalent) (Fig. 3B and C). The two-electron-
ꢁ
1
from the small kcat KS values for agmatine and putres-
cine compared with that for histamine (Table 1). The
three kinds of the mutation as Y180F, G268D/D269C
and Y180F/G268D/D269C decreased the substrate
specificity to some extents, as judged from the fact
reduced form of WT with putrescine gave a smaller
absorption peak assigned to CFMNS than that
observed with histamine and agmatine. The equilib-
rium in the two-electron-reduced form of WT is influ-
enced by the substrate used. In this paper, the
argument on the effect of the substrate will be mini-
mized for simplicity.
Figure 4 illustrates the UV-vis spectra of the two-
electron-reduced form as well as the oxidized form for
the three mutants: Y180F, G268D/D269C and Y180F/
G268D/D269C. The absorption spectra of the fully
oxidized form are almost identical with that of WT,
ꢁ
1
that the relative value of kcat K of agmatine against
S
ꢁ
K of histamine increased drastically by the
S
1
kcat
mutation (Table 1).
UV-vis spectra of substrate-reduced nHmDH
Figure 3A shows the UV-vis absorption spectra of the
oxidized and the histamine-reduced WT enzyme under
anaerobic conditions at pH 7.5. The broad and strong
absorption band with a maximum at 440 nm is
assigned to CFMNO (Fig. 3A, dotted line). CFMNS
gives two sharp absorption bands at 365 nm and
indicating that the electronic state of CFMN remains
O
practically unchanged on the mutation. In addition,
the absorption spectrum of the two-electron-reduced
form of G268D/D269C (Fig. 4B) is also almost iden-
tical to that of WT (Fig. 3), indicating the favoured
equilibrium of the CFMN -FeS state rather than the
445 nm and one small broad band at around 500 nm,
while CFMNR gives a broad absorption band with
S
R
2
60

Substrate inhibition of histamine dehydrogenase
0
.15
A
B
C
0.15
A
0.1
0
.1
0
.05
0
0.05
2
00
400
600
800
0
2
00
400
600
800
Wavelength / nm
Wavelength / nm
B
0
0
.15
0
0
.15
0.1
0
.1
.05
0
.05
0
2
00
400
600
800
Wavelength / nm
200
400
600
800
Wavelength /nm
0
0
.15
C
0.15
0
.1
0.1
.05
0
0.05
2
00
400
600
800
0
Wavelength / nm
200
400
600
800
Fig. 4 UV-vis spectra of (A) Y180F, (B) G268D/D269C and (C)
Y180F/G268D/D269C nHmDH. Spectra of oxidized (dotted line)
and substrate-reduced (solid line) in a 0.1 M potassium phosphate
buffer, pH 7.5, respectively.
Wavelength /nm
Fig. 3 UV-vis spectra of wild-type nHmDH with (A) histamine,
(
(
B) agmatine and (C) putrescine as substrates. Spectra of the oxidized
dotted line), substrate (2-equivalent amount)-reduced (solid line)
and substrate (20-equivalent)-reduced wild-type nHmDH
dashed line) in a 0.1 M potassium phosphate buffer, pH 7.5.
state in the two-electron-reduced form of the Y180F
and Y180F/G268D/D269C mutants.
(
Evaluation of the redox potentials
CFMN -FeS state in the two-electron-reduced form
O
at pH 7.5.
On the other hand, interestingly, the two-electron-
reduced form of Y180F and Y180F/G268D/D269C
did not give any absorption peak assigned to
In order to verify the equilibrium shift mentioned
above, we attempted to evaluate the redox potential
of the CFMN and [4Fe—4S] cluster in the enzymes.
The CFMN cofactor is silent in direct electrochemistry
but shows a characteristic absorption spectral change
during the redox reaction (9, 25). Therefore, a medi-
ated spectroelectrochemical titration method is useful
to evaluate the redox potential of CFMN (Fig. S1) (9).
R
CFMN , but a board absorption band at around
S
360 nm assigned to CFMNR (Fig. 4A and C). This
means that the equilibrium lies to the CFMN -FeS
R
O
2
61

M. Tsutsumi et al.
Table 2. Redox potentials of the flavin and the [4Fe—4S] iron—sulfur cluster in nHmDH at pH 7.5.
ꢀ
0
O=S
ꢀ0
S=R
ꢀ0
O=R
ꢀ0
FeS
a
a
a
b
c
S
d
E
=mV
E
=mV
E
=mV
E
=mV
K
K
2
e
WT
Y180F
G268D/D269C
Y180F/G268D/D269C
16
—
0
—
8
27
23
ꢁ26
1.8
ꢆ1
1.9
ꢆ1
2.4
ꢆ1
1.8
ꢆ1
f
f
ꢁ138
ꢁ155
ꢁ146
ꢁ140
f
—
f
—
20
37
a
b
Evaluated potentials by mediated spectroelectrochemical titrations. Midpoint potential determined by cyclic voltammetry using an ITO
c
d
e
f
electrode. Calculated values using Eq. (3). Calculated values using Eq. (4). Data taken from Tsutsumi et al. (2008). Not determined.
Considering the existence of an intermediate CFMNS
during the two-electron reduction of CFMN:
Discussion
The present work has shown that the site-direct muta-
tions of Tyr180, Gly268 and Asp269 located in the
vicinity of CFMN strongly affect the enzyme kinetics
including substrate inhibition and substrate specificity,
the spectroscopic property of two-electron-reduced
form and the redox potential of the redox sites. The
substrate inhibition from a natural substrate histamine
has been successfully suppressed by the Y180F/
G268D/D269C mutation, although the bimolecular
rate constant between the enzyme and the substrate
ꢀ
0
O=S
ꢁ
ꢁ
CFMN þe
CFMNS
E
E
E
ð2aÞ
ð2bÞ
O
ꢁ!
ꢀ
0
S=R
ꢁ
ꢁ
CFMN þe
CFMNR
S
ꢁ!
ꢀ
0
O=R
ꢀ0
O=S
ꢁ
ꢁ
CFMNO þ2e
CFMNR
½¼ðE
ꢁ
!
ꢀ
0
þE Þ=2ꢂ,
ð2cÞ
S=R
a two-step single-electron transfer model was adopted
for the Nernstian analysis of the absorbance change
ꢀ
0
O=S
ꢀ0
S=R
at 365 nm and 440 nm (9). E
and E
as the
ꢁ1
(kcat KS ) decreases to 17% of WT (Table 1). This
ꢀ
O=R
redox potentials of Eqs. (2a) and (2b) as well as E
seems to be reasonable in part, since the Y180F/
G268D/D269C mutation was designed to imitate the
amino acid sequence in the vicinity of the catalytic site
of rHmDH, which does not suffer from the histamine
inhibition. Interestingly, the kcat KS value of rHmDH
is only 23% of the WT nHmDH (unpublished data)
and is close to that of the Y180F/G268D/D269C
as the redox potential of Eq. (2c) were thus evaluated
and are summarized in Table 2.
The G268D/D269C mutation caused a drastic
ꢀ
0
ꢀ0
S=R
ꢁ
1
shift in E
potential. In the case of the Y180F and Y180F/
and E
to the direction of the negative
O=S
ꢀ
0
O=S
G268D/D269C mutant, we could not evaluate E
ꢀ
0
S=R
ꢁ1
and E
separately, because no clear absorption
mutant. However, the decrease in the kcat KS value is
not simply responsible for the substrate inhibition-free
characteristics, as evidenced by the enzyme kinetics
of the G268D/D269C mutant, which shows the k
band assigned to CFMN was observed during the
S
titration and the titration curve was simply analysed
on a conventional one-step two-electron transfer
model [Eq. (2c)]. The common property is ascribed
cat
ꢁ
1
K
value of only 22% of WT but is susceptible to
S
to the thermodynamic instability of CFMN in the
S
the histamine inhibition as in the case of WT.
The spectroscopic data suggest that the mutations
examined here scarcely affect the electronic state of
overall redox reaction, that is, an extremely low
value of the semiquinone formation constant (or com-
proportionation constant, KS) defined by:
CFMN . However, the electronic state of the two-
O
electron-reduced form of the enzyme is very sensitive
to the mutation. The G268D/D269C mutant as well as
WT contains obviously CFMN in the two-electron-
2
½
CFMN ꢂ
S
¼expðFðEꢀ0
ꢁEꢀS0=RÞ=RTÞ,
ð3Þ
KS ¼
O=S
½
CFMN ꢂ½CFMN ꢂ
O
R
S
reduced form, but CFMN is fully reduced to CFMNR
in Y180F and G268D/D269C mutants. The situation
can be reasonably explained in view of thermo-
dynamics as follows.
where F, R and T are the Faraday constant, the gas
constant and the absolute temperature, respectively.
The data suggest that the Y180F mutation is respon-
nHmDH is three-electron equivalent per subunit
for full reduction, but the substrate can provide only
two electrons per subunit at pH59 (9). Therefore, the
substrate-derived two-electron-reduced form has two
ultimate states: CFMN -FeS and CFMN -FeS .
sible for strong destabilization of CFMN , but details
S
remain unknown.
On the other hand, the spectroscopic information of
the [4Fe—4S] iron—sulfur cluster is not easy to extract.
However, the [4Fe—4S] cluster in the enzyme can com-
municate directly with ITO electrodes and then cyclic
voltammetry is available for evaluation of the redox
S
R
R
O
ꢀ
0
The distribution equilibrium is a function of E
S=R
ꢀ
0
FeS
and E
:
!
ꢀ
FeS
ꢀ0
S=R
ꢀ
0
FðE ꢁE
Þ
potential of the [4Fe—4S] cluster, E
(Fig. S2)
9, 25). The data evaluated for the enzymes are sum-
½CFMNS ꢁFeSRꢂ
FeS
K ꢅ
¼exp
: ð4Þ
2
(
marized in Table 2. The large negative shift in E was
½
CFMNR ꢁFeSOꢂ
RT
ꢀ
0
FeS
observed for the Y180F and G268D/D269C mutants,
The K value is about 2 for WT and G268D/D269C
2
ꢀ
0
FeS
but E
of Y180F/G268D/D269C was close to that
mutant as calculated from the value in Table 2, but
that is close to 0 for the Y180F and G268D/D269C
ꢀ
0
FeS
of WT at pH 7.5. The data indicate that E
is also
ꢀ
0
S=R
sensitive to the mutation even in the vicinity of
CFMN, but the reason is not clear at this stage.
mutants. Note here that the E
values of Y180F and
G268D/D269C could not be evaluated here but should
2
62

Substrate inhibition of histamine dehydrogenase
ꢀ
0
O=R
be much positive than E , since the semiquinone
In conclusion, the amino acid residues in the vicinity
of CFMN play very important role in governing the
redox potential of CFMN and [4Fe—4S] cluster, and
then the semiquinone formation constant KS and
2
the electron distribution constant K between CFMN
and the [4Fe—4S] cluster are easily influenced by the
formation constant KS is close to 0. Considering the
enzyme kinetics and the thermodynamic argument as
well as spectroscopic properties, it is reasonably con-
cluded that large K2 value (probably more than about
unity) is responsible for the substrate inhibition.
Substrate inhibition is considered to be a special case
of uncompetitive inhibition, and thus some interaction
must exist between the enzyme in the substrate-reduced
state and the substrate. The idea is supported by the
spectral data in Fig. 3: the absorption peaks assigned
amino acid residues. The reason is not clarified in this
work, but it has been found that the enzyme kinetics
is very sensitive to the thermodynamic property. The
substrate inhibition from histamine is strongly related
to the stability of CFMN in the two-electron-reduced
S
to CFMN of the two-electron-reduced form of WT
S
form, as represented by large K . In contrast, however,
2
becomes more evident in the presence of excess amount
of histamine. There seems to be the plausible second
binding site for the substrate. The site is not responsi-
ble for the enzyme catalysis, but the binding seems to
small K value leads to uphill single electron transfer
2
from CFMNR to FeS , which results in the decrease
O
in the kinetics. Because of the conflicting properties
ascribed to the K value, it seems to be very difficult
2
ꢁ
1
perturb the redox potential to increase the K value.
2
to generate mutant enzymes with large kcat
K
value
S
Interestingly, such spectral perturbation in the two-
electron-reduced form was observed in the G268D/
D269C mutants on the addition of excess amount of
histamine (data not shown) and similar spectroscopic
phenomena are reported for TMADH (8). At high
trimethylamine concentrations, the spectrum also exhi-
bits a high absorption at 365 nm that is characteristic
of the semiquinone form of the flavin (8).
without substrate inhibition. When we consider some
application of the enzyme as catalysts of biosensors or
bioassay of histamine, nHmDH with a large K value
2
is convenient for low concentration sample analysis
ꢁ
1
because of large kcat
small K value is convenient for high concentration
K
value, while rHmDH with
S
2
sample analysis because the enzyme does not suffer
from the substrate inhibition. The negative shift of
ꢀ
0
O=R
ꢀ0
FeS
The property of large K2 value is strongly related to
E
occurred in the G268D/D269C or that of E
ꢀ
0
O=S
ꢀ0
S=R
that of large K value, that is, E 4 E
(Table 2).
S
occurred in the Y180F also leads to the kinetic hin-
drance derived from the thermodynamic situation.
Under such situations, the second step of the two
sequential single electron transfers from the reduced
flavin to the [4Fe—4S] cluster (that is, the single elec-
tron transfer from CFMN to FeS ) is not favoured
Supplementary data
S
O
compared with that of the first step electron transfer
the electron transfer from CFMNR to FeS ) in view
Supplementary Data are available at JB Online.
(
O
of the thermodynamics and then becomes one of the
key steps to govern the kcat value. The binding of the
second substrate enhances the situation, which should
reduce the enzyme kinetics. We believe that this is the
situation explaining the substrate inhibition.
Acknowledgement
The authors thank Kikkoman Corporation for their kind gift of
rHmDH.
On the other hand, Y180F and G268D/D269C
mutants did not show any spectral change on the addi-
tion of excess amount of histamine. The small pertur-
bation of the redox potential induced by the second
binding of the substrate, if any, might not be enough
Funding
This work was supported in part by Grants-in-Aids for Scientific
Research from the Ministry of Education, Science, Sports and
Culture of Japan (to M.T. 20-3618 and to K.K. 15380082 and
1
9310070).
to bring CFMN into existence for the Y180F and
S
G268D/D269C mutants, because the K2 value is too
small before the second binding.
In contrast, the K and K values of Y180F/G268D/
Conflict of interest
None declared.
S
2
D269C are much smaller than those of WT. Small
K2 value means uphill single electron transfer from
References
the CFMN to FeS . This situation is not convenient
R
O
1. Siddiqui, J.A., Shoeb, S.M., Takayama, S., Shimizu, E.,
and Yorifuji, T. (2000) Histamine dehydrogenase of
Nocardioides simplex: a second bacterial enzyme for his-
tamine degradiation. J. Biochem. Mol. Bio. Biophys. 5,
37—43
for fast enzyme turnover and kcat should be small. The
G268D/D269C mutation causes a decrease in the
enzyme activity (decrease in kcat and increase in KS)
(Table 1). This might also be explained in terms of
the redox potential. The negative shift in E
ꢀ
0
O=R
2
. Suddiqui, J.A., Shoeb, S.M., Takayama, S., Shimizu, E.,
and Yorifuji, T. (2000) Purification and characterization
of histamine dehydrogenase from Nocardioides simplex
IFO 12069. FEMS Microbiol. Lett. 189, 183—187
. Fujieda, N., Satoh, A., Tsuse, N., Kano, K., and
Ikeda, T. (2004) 6-S-cysteinyl flavin mononucleotide-
containing histamine dehydrogenase from Nocardioides
simplex: molecular cloning, sequencing, overexpression,
and characterization of redox centers of enzyme.
Biochemistry 43, 10800—10808
and
ꢀ
0
FeS
E
caused by the G268D/D269C mutation results
in the decrease in the driving force of the reaction
between the free enzyme and the substrate. In a similar
way, the decrease in the enzyme activity by the Y180F
3
ꢀ
0
might be ascribed to the positive shift of E
the negative shift of E
and
(Table 2). The situation
O=R
ꢀ
0
FeS
leads to the uphill electron transfer from CFMN to
4Fe—4S] cluster.
[
2
63

M. Tsutsumi et al.
4
5
. Limburg, J., Mure, M., and Klinman, J.P. (2005)
Cloning and characterization of histamine dehydroge-
nase from Nocardioides simplex. Arch. Biochem.
Biophys. 436, 8—22
. Huang, L., Rohlfs, R.J., and Hille, R. (1995) The reac-
tion of trimethylamine dehydrogenase with electron
transferring flavoprotein. J. Biol. Chem. 270,
15. Deakyne, C.A. and Meot-Ner, M. (1985) Unconven-
tional ionic hydrogen bonds. 2. NH ꢇꢇꢇp. Complexes
þ
of onium ions with olefins and benzene derivatives.
J. Am. Chem. Soc. 107, 474—479
16. Bellamy, H.D., Lim, L.W., Mathews, F.S., and
Dunham, W.R. (1989) Studies of crystalline trimethyla-
mine dehydrogenase in three oxidation state and in the
presence of substrate and inhibitor. J. Biol. Chem. 264,
11887—11892
2
3958—23965
. Jang, M.-H., Basran, J., Scrutton, N.S., and Hille, R.
1999) The reaction of trimethylamine dehydrogenase
with trimethylamine. J. Biol. Chem. 274, 13147—13154
. Falzon, L. and Davidson, V.L. (1996) Kinetic model for
regulation by substrate of intramolecular electron trans-
fer in trimethylamine dehydrogenase. Biochemistry 35,
6
7
1
7. Barber, M.J., Neame, P.J., Lim, L.W., White, S., and
Mathews, F.S. (1992) Correlation of X-ray deduced
and experimental amino acid sequences of trimethyl-
amine dehydrogenase. J. Biol. Chem. 267, 6611—6619
8. Basran, J., Mewies, M., Mathews, F.S., and
Scrutton, N.S. (1997) Selective modification of alkyl-
ammonium ion specificity in trimethylamine dehydroge-
nase by the rational engineering of cation-bonding.
Biochemistry 36, 1989—1998
(
1
2445—2452
8
. Roberts, P., Basran, J., Wilson, E.K., Hille, R., and
Scrutton, N.S. (1999) Redox cycles in trimethylamine
dehydrogenase and mechanism of substrate inhibition.
Biochemistry 38, 14927—14940
1
2
9. Fujieda, N., Tsuse, N., Satoh, A., Ikeda, T., and
Kano, K. (2005) Production of completely flavinylated
histamine dehydrogenase, unique covalently bound
flavin, and iron-sulfur cluster-containing enzyme of
Nocardioides simplex in Escherichia coli, and its proper-
ties. Biosci. Biotechnol. Biochem. 69, 2459—2462
0. Wilson, E.K., Mathews, F.S., Packman, L.C., and
Scrutton, N.S. (1995) Electron tunneling in substrate-
reduced trimethylamine dehydrogenase: kinetics of elec-
tron transfer and analysis of the tunneling pathway.
Biochemistry 34, 2584—2591
9
. Tsutsumi, M., Fujieda, N., Tsujimura, S., Shirai, S., and
Kano, K. (2008) Thermodynamic redox properties
governing half-reduction characteristics of histamine
dehydrogenase from Nocardioides simplex. Biosci.
Biotechnol. Biochem. 72, 786—796
1
0. Bakke, M., Sato, T., Ichikawa, K., and Nishimura, I.
(
2005) Histamine dehysrogenase from Rhizobium sp.:
gene cloning, expression in Escherichia coli, characteriza-
tion and application to histamine determination.
J. Biotechnology 119, 260—271
2
2
1. Segal, I.H. (1993) Uncompetitive inhibition in Enzyme
Kinetics, pp. 136—143, Wiley-Interscience, New York
2. Tsujimura, S., Kuriyama, A., Fujieda, N., Kano, K., and
Ikeda, T. (2005) Mediated spectroelectrochemical titra-
tion of proteins for redox potential measurements by a
separator-less one-component bulk electrolysis methods.
Anal. Biochem. 337, 325—331
1
1. Takagi, K. and Shikata, S. (2004) Flow injection deter-
mination of histamine with a histamine dehydrogenase-
based electrode. Anal. Chim. Acta 505, 189—193
2. Sato, T., Horiuchi, T., and Nishimura, I. (2005) Simple
and rapid determination of histamine in food using a new
histamine dehydrogenase from Rhizobium sp. Anal.
Biochem. 346, 320—326
3. Yamada, R., Fujieda, N., Tsutsumi, M., Tsujimura, S.,
Shirai, O., and Kano, K. (2008) Bioelectrochemical
determination at histamine dehydrogenase-based elec-
trodes. Electrochemistry 76, 600—602
4. Steenkamp, D.J. and Beinert, H. (1982) Mechanistic
studies on the dehydrogenases of methylotrophic bacte-
ria. 1. The influence of substrate binding to reduced
trimethylamine dehydrogenase on the intramolecular
electron transfer between its prosthetic groups.
Biochem. J. 207, 233—239
1
2
3. Nagabayashi, Y., Omayu, A., Yagi, S., Nakamura, K.,
and Motonaka, J. (2001) Evaluation of Osmium (II)
complexes as electron transfer mediators accessible for
amperometric glucose sensors. Anal. Sci. 17, 945—950
4. Rohlfs, R.J. and Hille, R. (1994) The reaction of
trimethylamine dehydrogenase with diethylmethylamine.
J. Biol. Chem. 269, 30869—30879
1
2
2
1
5. Tsutsumi, M., Tsujimura, S., Shirai, O., and Kano, K.
(
2009) Direct electrochemistry of histamine dehydroge-
nase from Nocardioides simplex. J. Electroanal. Chem.
625, 144—148
2
64