Characterization of a dihydrolipoyl dehydrogenase having diaphorase activity of Clostridium kluyveri

Article abstract of DOI:10.1271/bbb.70724

The Clostridium kluyveri bfmBC gene encoding a putative dihydrolipoyl dehydrogenase (DLD; EC 1.8.1.4) was expressed in Escherichia coli, and the recombinant enzyme rBfmBC was characterized. UV-visible absorption spectrum and thin layer chromatography analysis of rBfmBC indicated that the enzyme contained a noncovalently but tightly attached FAD molecule. rBfmBC catalyzed the oxidation of dihydrolipoamide (DLA) with NAD⁺ as a specific electron acceptor, and the apparent K_m values for DLA and NAD⁺ were 0.3 and 0.5 mM respectively. In the reverse reaction, the apparent K _m values for lipoamide and NADH were 0.42 and 0.038 mM respectively. Like other DLDs, this enzyme showed NADH dehydrogenase (diaphorase) activity with some synthetic dyes, such as 2,6-dichlorophenolindophenol and nitro blue tetrazolium. rBfmBC was optimally active at 40°C at pH 7.0, and the enzyme maintained some activity after a 30-min incubation at 60°C.

Full text of DOI:10.1271/bbb.70724

Biosci. Biotechnol. Biochem., 72 (4), 982–988, 2008
Characterization of a Dihydrolipoyl Dehydrogenase
Having Diaphorase Activity of Clostridium kluyveri
y
Saikat CHAKRABORTY, Makiko SAKKA, Tetsuya KIMURA, and Kazuo SAKKA
Graduate School of Bioresources, Mie University, Tsu 514-8507, Japan
Received November 6, 2007; Accepted January 8, 2008; Online Publication, April 7, 2008
[doi:10.1271/bbb.70724]
The Clostridium kluyveri bfmBC gene encoding a
putative dihydrolipoyl dehydrogenase (DLD; EC
1.8.1.4) was expressed in Escherichia coli, and the
recombinant enzyme rBfmBC was characterized. UV-
visible absorption spectrum and thin layer chromatog-
raphy analysis of rBfmBC indicated that the enzyme
contained a noncovalently but tightly attached FAD
molecule. rBfmBC catalyzed the oxidation of dihydro-
lipoamide (DLA) with NADþ as a specific electron
acceptor, and the apparent Km values for DLA and
NADþ were 0.3 and 0.5 mM respectively. In the reverse
reaction, the apparent Km values for lipoamide and
NADH were 0.42 and 0.038 mM respectively. Like other
DLDs, this enzyme showed NADH dehydrogenase
(diaphorase) activity with some synthetic dyes, such as
2,6-dichlorophenolindophenol and nitro blue tetra-
zolium. rBfmBC was optimally active at 40 ꢀC at
pH 7.0, and the enzyme maintained some activity after
a 30-min incubation at 60 ꢀC.
dophenol (DCPIP).1) DLD has been extensively charac-
terized from a variety of organisms belonging to
bacteria, archaea, and eucarya. Enzyme properties are
similar among the members of these groups.2,3) DLDs
normally function as integral components of the pyr-
uvate, 2-oxoglutarate, and branched chain 2-oxoacid
dehydrogenase multienzyme complexes4,5) and of the
glycine cleavage system.6) Therefore, DLDs play a
significant role in metabolism and energy production.
DLDs might be homodimers with a subunit size of about
50 kDa. Each monomer contains a tightly but non-
covalently attached flavin nucleotide molecule, a redox
disulfide, and one NADþ-binding site. Upon reoxidation
of the dihydrolipoyl moiety, electrons are transferred
from the dihydrolipoyl moiety to redox disulfide, then to
the FAD cofactor, and finally to the NADþ cofactor.2)
We are interested in enzymes with diaphorase
activity, since they are useful in the colorimetric
determination of NAD(P)H, and hence various dehy-
drogenase activities can be assayed when coupled with
various dyes that act as hydrogen acceptors from
NAD(P)H. NADH diaphorase also has been used
extensively in detecting nitric oxide synthases activity
in various cell types, including neuronal cell bodies,
vascular endothelium, cells of the immune system,
and epithelial cells, and thus have gained considerable
importance as a diagnostic chemical.7) In fact, several
kinds of diaphorases are commercially available, and
their origins are bacteria such as Clostridium kluyveri.8)
In an earlier study, an enzyme from C. kluyveri was
purified and characterized by Kaplan et al.,8) who used
starting material equivalent to the currently available
commercial product (Worthington Biochemical, Lake-
wood, NJ). The molecular size of the purified enzyme
was determined to be about 24,000, and the enzyme was
expected to be a proteolysis product of a parental DLD
judging by its small size. However, we recently cloned
the gene encoding the small diaphorase from C. kluy-
veri, and found that the encoded diaphorase consisted of
Key words: Clostridium kluyveri; dihydrolipoyl dehy-
drogenase; diaphorase
Dihydrolipoyl dehydrogenase (DLD; EC 1.8.1.4) is a
component of the multi-enzyme 2-oxo-acid dehydrogen-
ase complexes, and it catalyzes the oxidation of a
dihydrolipoyl residue covalently bound to a component
protein and the reverse reaction as follows (homepage of
Enzyme Nomenclature of IUBMB: http://www.chem.
qmul.ac.uk/iubmb/enzyme/):
Protein N6-(dihydrolipoyl)lysine þ NADþ
$ protein N6-(lipoyl)lysine þ NADH
DLD can also use free dihydrolipoate, dihydrolipoamide
(DLA), or dihydrolipoyllysine as substrate. It is often
termed ‘‘diaphorase’’ for its characteristic flavin moiety
and NADH oxidase activity, where the electron acceptor
is generically an oxidized dye such as methylene blue,
nitro blue tetrazolium (NBT), or 2,6-dichlorophenolin-
y
To whom correspondence should be addressed. Tel: +81-59-231-9621; Fax: +81-59-231-9684; E-mail: sakka@bio.mie-u.ac.jp
Abbreviations: DCPIP, 2,6-dichlorophenolindophenol; DLA, dihydrolipoamide; DLD, dihydrolipoyl dehydrogenase; LA, lipoamide; NBT, nitro
blue tetrazolium; PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; TLC, thin layer chromatography; Tris,
Tris(hydroxymethyl)aminomethane

Dihydrolipoyl Dehydrogenase of Clostridium kluyveri
983
an N-terminal flavin reductase-like domain and a C-
terminal Fe-containing rubredoxin-like domain. This
domain organization was completely different from
those of well-known DLDs, which consist of a
Pyr redox 2 domain and a Pyr redox dim domain.9)
Furthermore, we found that a gene highly homologous
with authentic DLD genes existed in the unpublicized
genomic DNA sequence database of C. kluyveri, con-
structed by the Research Institute of Innovative Tech-
nology for the Earth (RITE).10) We were interested in
comparing the enzymatic properties of this DLD enzyme
with the small diaphorase.
on a thermal cycler (Bio-Rad Laboratories, Hercules,
CA). The PCR fragments were gel-purified by with GFX
PCR DNA and a gel band purification kit (Amersham
Biosciences, Buckinghamshire, UK), ligated into
pCR2.1, and introduced into E. coli DH5ꢀ by a standard
transformation protocol. After the integrity of the PCR-
amplified fragment was confirmed by DNA sequencing,
the insert DNA was recovered from the recombinant
plasmid by digestion with BamHI and SalI and ligated
into pET-28a(+), yielding pET-bfmBC. The produced
recombinant enzyme (termed rBfmBC) contained a
6xHis-tag at the N-terminus.
In this paper, we report the expression of a putative
DLD gene encoding 455 amino acid residues from
C. kluyveri on the basis of the genome database of
RITE. Recently, we found that the DLD gene described
in this paper is identical to bfmBC (DDBJ accession
no. CP000673; protein id EDK33811.1) reported by a
German group. Hence we use bfmBC and BfmBC as
names of the DLD gene and the gene product respec-
tively. We also deal with the purification and character-
ization of BfmBC from recombinant Escherichia coli.
Purification of rBfmBC. E. coli BL21 (DE3) harbor-
ing pET-bfmBC was grown overnight at 37 ꢀC in 500 ml
of LB medium containing 40 mg/ml of kanamycin.
IPTG was added to a final concentration of 1 mM and
incubated for a further 5 h at 37 ꢀC. The cells were
harvested by centrifugation at 4;000 ꢁ g for 20 min, and
resuspended in 30 ml of lysis buffer (pH 7.4) containing
50 mM NaH2PO4, 300 mM NaCl, and 10 mM imidazole.
The cells were disrupted by sonication, and the soluble
fraction of the recombinant protein was obtained by
centrifugation at 4;500 ꢁ g for 25 min. The supernatant
was used as the crude enzyme solution and was purified
by HiTrap Chelating HP (Amersham Biosciences)
column chromatography according to the supplier’s
protocol. Active fractions were collected and desalted by
dialysis using 50 mM sodium phosphate buffer (pH 7.4).
The presence of about 50-kDa recombinant protein
rBfmBC was confirmed by SDS–PAGE (12.5%) analy-
sis. Protein was determined by the Bradford method12) or
by UV absorption.
Materials and Methods
Bacterial strains, plasmids, and cultivation. C. kluy-
veri ATCC 8527 was cultivated in an ATCC medium
(broth 1120) at 37 ꢀC for 5 d. E. coli strains DH5ꢀ and
BL21(DE3) were used in the cloning and expression
respectively of the C. kluyveri bfmBC gene encoding
a DLD. The plasmid pCR2.1-TOPO with single 30-
thymidine overhangs was used in TA cloning. The
plasmid pET-28a (+) was obtained from Novagen
(Madison, WI). Recombinant E. coli stains were rou-
tinely grown in Lenox Broth (LB) medium supplement-
ed with ampicillin (50 mg/ml) or kanamycin sulfate
(40 mg/ml). For high expression of the recombinant
protein, LB Broth EZMix (Sigma-Aldrich, St. Louis,
MO.) was also used.
Molecular weight determination by gel filtration. The
molecular weight of native rBfmBC was determined by
gel filtration on a Superdex 200 HR 10/30 column
(Amersham Biosciences). The column was eluted with
50 mM morpholineethanesulfonic acid buffer (pH 7.5)
containing 0.15 M NaCl at a flow rate of 30 ml/h. Horse
spleen apoferritin (440 kDa), aldolase (158 kDa), oval-
bumin (43 kDa), and chymotrypsinogen A (25 kDa)
were used as molecular weight standards in this experi-
ment.
Chemicals. All chemicals were obtained from Nacalai
Tesque (Kyoto, Japan). DCPIP was purchased from
merck (Darmstadt, Germany). Restriction endonucleases
BamHI and SalI were from Takara Bio (Shiga, Japan).
DL-DLA was prepared by reduction of DL-lipoamide
(LA) (Sigma-Aldrich) with NaBH4.11)
Enzyme assays. DLD activity was assayed at 30 ꢀC
in 50 mM potassium phosphate-2 mM EDTA (pH 7.0)
containing 0.4 mM DLA and 1 mM NADþ. The reaction
mixture (final volume, 1 ml) was started with the
Cloning of the bfmBC gene. C. kluyveri chromosomal
DNA was isolated using a QIAamp mini spin column
(Qiagen, Valencia, CA), and used as a template for PCR
amplification. Oligonucleotide primers 50-CCCGGATC-
CATGGCTTATAAATATGATCTGATTG-30 and 50-
CCCGTCGACTTACTCCCTCATCAATTCCC-30, in-
cluding BamHI and SalI sites (underlined), were
designed to amplify a 1,368-bp fragment of the bfmBC
gene (DDBJ accession no. CP000673;protein id
EDK33811.1). PCR amplification was carried out using
KOD DASH DNA polymerase (Toyobo, Osaka, Japan)
addition of the enzyme, and its progress was monitored
13)
by the increase in A340
.
The extinction coefficient for
NADH (6:22 ꢁ 103 Mꢂ1 cmꢂ1) was used in the calcu-
lation of enzyme activity. One unit of enzyme activity
was defined as the amount of enzyme required for the
production of 1 mmol of NADH in 1 min at 25 ꢀC.
Diaphorase activity against DCPIP (extinction coef-
ficient, 2:1 ꢁ 104 Mꢂ1 cmꢂ1) was determined by meas-
uring the decrease in absorbance at 600 nm.8) Diaphor-

984
S. CHAKRABORTY et al.
ase activity against 0.5 mM NBT was assayed by
measuring the reduction in absorbance at 585 nm with
9 mM NADH in 0.3 ml of 50 mM Tris–HCl (pH 8.0) at
37 ꢀC.14) Unless otherwise stated, enzyme activity was
measured with a Shimadzu spectrophotometer BioSpec-
1600 (Kyoto, Japan) at 30 ꢀC. One unit of enzyme was
defined as the amount of enzyme that reduces 1 mmol of
2,6-DCPIP per min at 25 ꢀC at pH 8.5.
excitation, the emission pattern was recorded from 300
to 600 nm and 300 to 560 nm respectively at room
temperature. The influence of NADþ on the fluorescence
spectrum was monitored by the addition of NADþ to a
final concentration of 0.05 mM. The effect of DLA on
the fluorescence spectrum was monitored after the
addition of the substrate to give a final concentration
of 0.25 mM at 0 min and 5 min.
The effect of pH on DLD and diaphorase activities
was investigated using Britton and Robinson’s universal
buffer (50 mM phosphoric acid, 50 mM boric acid, 50 mM
acetic acid; the pH was adjusted to 3–11 using 1 M
NaOH). To determine pH stability, the enzymes
were pre-incubated overnight at 4 ꢀC in Britton and
Robinson’s universal buffer solution at different pHs.
The residual activities were then measured under
standard assay conditions.
The effect of temperature on DLD and diaphorase
activities was measured using a Shimadzu spectropho-
tometer UV-1200 with a temperature control system.
Thermostability was measured by pre-incubating the
enzyme at various temperatures (30 to 70 ꢀC) for 30 min,
followed by measurement of the residual enzyme
activities under standard assay conditions.
The apparent Km values of rBfmBC for LA, DLA,
NADH, and NADþ were determined in 50 mM potas-
sium phosphate-2 mM EDTA (pH 7.0). The concentra-
tions of LA and DLA were varied between 0.2 and
1.0 mM, with fixed concentrations of NADH (0.1 mM)
and NADþ (1 mM) respectively. The apparent Km values
for NADH and NADþ were measured with varying
concentrations of NADH (0.02–0.1 mM) and NADþ
(0.1–1.0 mM) at constant concentrations of LA (1 mM)
and DLA (1 mM) respectively.
Results and Discussion
Expression and purification of rBfmBC
rBfmBC was highly expressed by E. coli BL21(DE3)
harboring pET-rbfmBC, and purified using a HiTrap HP
Chelating column in one step, with a recovery of about
70%. The expression level of rBfmBC was calculated to
be about 10 mg per liter of the culture. The purified
fractions showed a strong yellow color, suggesting that
rBfmBC is a flavoprotein. Expression was found to be
better in LB Broth EZMix (Sigma-Aldrich) than in
normal LB or Super broth (3.5% Bacto Tryptone, 2%
yeast extract, and 0.5% NaCl) media. As shown in
Fig. 1, the purified protein showed a single band on
SDS–PAGE. The molecular size of the purified protein
(about 50 kDa) was in good agreement with the
calculated molecular weight of rBfmBC, including a
6xHis-tag (52,983).
The apparent molecular weight of native rBfmBC was
determined by gel filtration on a Superdex 200 HR 10/
30 column as described in ‘‘Materials and Methods.’’
After repeated experiments, we concluded that the
molecular size of rBfmBC in the native form was about
110 kDa. A comparison of this value with the theoretical
molecular size (53 kDa) suggested that rBfmBC might
exist as a dimer in the native state. This was consistent
with previous observations in which native forms of
many DLDs were reported to exist as dimers consisting
of two identical subunits (Table 1), which contained
1 mol of FAD and 2 redox-active Cys residues per
subunit.16,17) Although several dihydrolipoyl dehydro-
genases with a typical molecular mass or nicotinamide-
Cofactor determination. Identification of flavin co-
factor present in recombinant enzyme was carried out
as described previously.15) rBfmBC was heated at
100 ꢀC for 20 min and then centrifuged at 14;000 ꢁ g
to remove white precipitated protein. An appropriate
amount of the supernatant was applied to a Kieselgel 60
TLC plate (merck) along with FAD and FMN as
standards, and developed with a solvent system of
butanol:acetic acid:water (4:1:1). The molar content
of cofactor FAD present in the recombinant enzyme
was determined using extinction coefficient 10:3 ꢁ
M
1
2
3
103 Mꢂ1 cmꢂ1
.
66.0
45.0
UV-Visible and fluorescence spectral analyses of
rBfmBC. UV-Visible absorption spectra of rBfmBC
were obtained with Shimadzu Spectrophotometer Bio-
Spec-1600. Measurement was carried out between 250
and 800 nm at room temperature in 50 mM phosphate
buffer (pH 7.4).
Fluorescent spectra were recorded using an RF-
spectrophotometer (Shimadzu). The protein solution
(1 mM) in 50 mM phosphate buffer (pH 7.4) was excited
at 260 nm and 296 nm. Following 260-nm and 296-nm
30.0
Fig. 1. SDS–PAGE of rBfmBC.
Protein samples were analyzed on a 12.5% polyacrylamide gel.
Lane 1, cell-free extracts from uninduced cells; lane 2, cell-free
extracts from cells induced with 1 mM IPTG; lane 3, purified
rBfmBC; M, molecular size marker.

Dihydrolipoyl Dehydrogenase of Clostridium kluyveri
985
Table 1. Comparison of Structural and Functional Properties of Dihydrolipoamide Dehydrogenase from Different Sources
Molecular mass
Coenzyme
(kDa)
Source
Metabolic role
References
Native
Subunita
NADH
NADPH
Mammalian, yeast
Trypanosoma cruzi
Aerobic eubacteria
100–115
100
100–110
50–55
55
50–55
+
+
+
ꢂ
ꢂ
ꢂ
Mitochondrial 2-oxo acid oxidation
Probably 2-oxo acid oxidation
2-oxo acid oxidation, but cf. (29) for a
third DLD of P. putida without known
function
2, 3, 26, 27
28
2, 3, 29
Archaebacteria
Anabaena sp.
115–122
104
58–61
53
+
+
ꢂ
ꢂ
Function not known
30–32
16
Probably photorespiratory glycine
oxidation
Uric acid and glycine utilization
Clostridium
105
52
+
+
17
cylindrosporum
Clostridium
sporogenes
17
100
68
52
34.5
ꢂ
+
+
Function not known
Glycine utilization
Eubacterium
acidaminophilum
Clostridium kluyveri
(+)b
17
110c
50
+
ꢂ
Function not known
This study
aValues obtained by SDS–PAGE.
bNADPH was the preferred cofactor; this enzyme is actually thioredoxin reductase, like flavoprotein, exhibiting some DLD activity.
cThis value was determined for the recombinant protein that contained a 6xHis-tag and some additional amino acid residues.
nucleotide specificity are known in anaerobic and
glycine-utilizing bacteria,17) they were reported to form
dimers in the native form.
a FAD-binding site conserved in many DLDs. The
molar ratio of FAD to rBfmBC monomer was calculated
to be 0.97.
BfmBC is expected to be a component of the glycine
cleavage system, since some genes downstream of
bfmBC have high similarity to those responsible for
glycine cleavage, but the molecular size and nicotina-
mide-nucleotide specificity of rBfmBC were similar to
the usual DLDs, but not to clostridial enzymes involved
in glycine cleavage (Table 1). A comparison of the
BfmBC sequence with DLD sequences registered in the
protein sequence database on Pfam website (http://
pfam.sanger.ac.uk/) revealed that BfmBC consisted of a
Pyr redox 2 domain and a Pyr redox dim domain, and
that the former domain included a flavin-binding domain
(the 6th to the 27th amino acids), and highly conserved
redox-active sites (the 39th to the 53rd amino acids),
including two conserved Cys residues.
Enzyme properties
When the DLD and diaphorase activities of rBfmBC
were assayed at various temperatures, rBfmBC was
optimally active at 40 ꢀC. rBfmBC activities gradually
decreased at temperatures higher than 40 ꢀC, and
rBfmBC showed about 20% of the highest activity at
70 ꢀC. The enzyme was stable to incubation at 40 ꢀC for
30 min in the absence of the substrates, and it kept its
activity after 3 months of storage at 4 ꢀC. It showed
slight activity even after incubation in the absence of the
substrate at 60 ꢀC for 30 min. The small diaphorase,
which is a major component of the commercial
diaphorase of C. kluyveri, showed negligible activity at
60 ꢀC.9) C. cylindrosporum and C. sporogenes DLDs
lost about 60% and 10% respectively of their specific
activities upon storage at ꢂ20 ꢀC for 6 months.18) Based
on these observations, rBfmBC appears to be compara-
tively thermostable, although some microorganisms
have been found to produce thermostable DLDs, e.g.,
DLD of thermophilic bacterium Bacillus stearothermoc-
philus was stable at 65 ꢀC,19) and incubation of DLDs
from halophilic archaebacteria at 95 ꢀC for 15 min in
the presence of 4 M NaCl resulted in no detectable loss
of their enzyme activities.20) Although high stability
of enzymes is generally an important factor for usage,
the requirement of extreme reaction conditions is not
desirable.
TLC analysis clearly revealed the presence of FAD
but not that of FMN in the recombinant enzyme as
a cofactor (Fig. 2), as expected from the presence of
The optimum pH of rBfmBC was found to be about
7.0, and it showed high activity in a pH range of 6.0 to
8.0 in both DLD and diaphorase assays. Overnight
incubation of the enzyme at 4 ꢀC in Britton and
Robinson’s buffer solution at various pH, the levels
FMN
S
FAD
Fig. 2. Identification of Flavin Cofactor Present in rBfmBC.
The flavin cofactor of rBfmBC was released by boiling and
analyzed by TLC with FMN and FAD as standards. S, sample.

986
S. CHAKRABORTY et al.
Table 2. Kinetic Constants of C. kluyveri rBfmBC
3
2
A
Specific Activity
(U/mg)
Type of activity
Substrate
Km (mM)
DLA
NAD
0.3
0.5
139
131
DLD
1
0
LA
NADH
0.42
0.038
112
88
DCPIP
NADH
0.12
0.25
675
1200
Diaphorase
300 400 500 600 700 800
Wavelength (nm)
1000
B
indicated that the recombinant enzymes were stable at
pH 6.0, and retained about half of the initial activity at
pH 6.0 and 9.0.
1
2
The influence of metal ions on DLD and diaphorase
activities were examined by assaying enzyme activity in
the presence of each metal ion at 1 mM of metal ions
such as Naþ, Ca2þ, Mg2þ, Mn2þ, Fe3þ, Zn2þ, Pb2þ, and
Hg2þ, no ion enhanced the enzyme activity of rBfmBC.
Pb2þ and Hg2þ ions showed strong inhibitory effects
on rBfmBC activity (more than 80% inhibition). About
30% inhibition of rBfmBC activity was observed with
5 mM Zn2þ ions.
3
4
0
300
600
Wavelength (nm)
1000
C
1
4
3
Kinetic parameters
2
From graphs showing reaction rates plotted against
respective substrate concentrations, the apparent Km
values for DLA and NADþ were estimated to be 0.3 mM
and 0.5 mM (Table 2) respectively under the assay
conditions described above under ‘‘Materials and Meth-
ods.’’ In the reverse reaction, the apparent Km values for
LA and NADH were determined to be 0.42 mM and
0.038 mM respectively. Although the Km values for DLA
and LA were similar to each other, the Km value for
NADH was smaller than that for NADþ. This enzyme
did not use NADPH as the electron donor. The similar
trend was observed in many DLDs, as shown in Table 1,
but DLDs from some anaerobic Clostridium spp. and
glycine-utilizing bacteria differ from other DLDs in their
coenzyme specificity, i.e., the latter is highly specific to
NADH.21) It is interesting that C. kluyveri, being strictly
anaerobic bacteria, has a preference for NADH as a
coenzyme, in contrast to some other DLDs from
anaerobic bacteria (Table 1).
0
300
560
Wavelength (nm)
Fig. 3. UV-Visible Absorption Spectrum (A) and Fluorescent Spec-
tra (B and C) of rBfmBC.
A, Measurement was carried out between 250 and 800 nm at room
temperature in 50 mM phosphate buffer (pH 7.4). B, Fluorescent
spectra were recorded between 300 and 600 nm by excitation at
260 nm in 50 mM phosphate buffer (pH 7.4). Treatments were as
follows: trace 1, as-isolated rBfmBC; trace 2, rBfmBC plus 50 mM
NADþ after 5 min; trace 3, rBfmBC plus 50 mM NADþ and 250 mM
DLA; trace 4, rBfmBC plus 50 mM NADþ and 250 mM DLA after
5 min. C, Fluorescent spectra were recorded between 300 and
560 nm by excitation at 296 nm in 50 mM phosphate buffer (pH 7.4).
Treatments were as follows: trace 1, as-isolated rBfmBC; trace 2,
rBfmBC plus 50 mM NADþ after 5 min; trace 3, rBfmBC plus 50 mM
NADþ and 250 mM DLA; trace 4, rBfmBC plus 50 mM NADþ and
250 mM DLA after 5 min.
One of the characteristics of DLDs is that they can
function with benzyl viologen, DCPIP, ferricyanide,
menadione, methylene blue, and NBT in addition to LA
and lipoic acid in a wide range of artificial electron
acceptors, and with NADH or NADPH as electron
donor.22) In diaphorase activity, the apparent Km values
of rBfmBC for DCPIP and NADH were 0.12 mM and
0.25 mM respectively. No diaphorase activity was
detected when NADH was replaced by NADPH, as
observed in the DLD of halophilic archaebacteria20) and
the hyperthermophilic archaeon Thermococcus kodakar-
aensis.23) At high concentrations of DCPIP, the reaction
mixture was too dark in color to detect a change in
absorbance, leading to the limited assay conditions used
in this study, as described under ‘‘Materials and
Methods.’’ rBfmBC showed strong diaphorase activity
with NBT as an electron acceptor, although quantitative
analysis was not carried out because of insoluble
precipitate (formazan) formation due to the enzyme
reaction.
UV-Visible and fluorescence spectral analyses of
rBfmBC
UV-Visible spectral analysis of the recombinant
protein showed two absorption peaks, at 270 nm due to
protein and at 453 nm due to the FAD cofactor

Dihydrolipoyl Dehydrogenase of Clostridium kluyveri
987
(Fig. 3A). The ratio of absorption at 280 nm to that of
References
450 nm was 4.98. This factor was conveniently used as a
criterion to determine the purity of DLD. The calculated
value was comparable to those of enzymes previously
reported,18,24) confirming that a subunit molecule con-
tained an FAD molecule.
1) Argyrou, A., Sun, G., Palfey, B. A., and Blanchard, J. S.,
Catalysis of diaphorase reactions by Mycobacterium
tuberculosis lipoamide dehydrogenase occurs at the EH4
level. Biochemistry, 42, 2218–2228 (2003).
2) Williams, C. H., Jr., ‘‘Chemistry and Biochemistry of
Flavoenzymes’’ Vol. III, ed. Muller, F., CRC Press, Boca
Raton, Florida, pp. 121–211 (1992).
3) Carothers, D. J., Pons, G., and Patel, M. S., Dihydroli-
poamide dehydrogenase: functional similarities and
divergent evolution of the pyridine nucleotide-disulfide
oxidoreductases. Arch. Biochem. Biophys., 268, 409–425
(1989).
4) Mattevi, A., de Kok, A., and Perham, R. N., The
pyruvate dehydrogenase multienzyme complex. Curr.
Opin. Struct. Biol., 2, 877–887 (1992).
Since rBfmBC contains no Trp residue but does
contain 14 Tyr and 10 Phe residues, we attempted to
measure the intrinsic fluorescence spectrum due to Tyr
or Phe residues at an excitation wavelength of 260 nm.23)
When the intrinsic fluorescence spectrum of as-isolated
rBfmBC was recorded, emission peaks were observed
around 330 and 510 nm (Fig. 3B). Excitation at 260 nm
excited amino acid residues, probably tyrosine residues
in rBfmBC, which emitted the absorbed energy at
330 nm. This emitted energy was transferred to the FAD
cofactor, which in turn emitted energy at 510 nm. When
NADþ was added to the enzyme mixture, the peaks at
330 and 510 nm decreased greatly. The decrease in these
peaks upon the addition of NADþ suggests a change in
the local environment of FAD and the aromatic amino
acid residues due to binding of NADþ to the protein.
When the enzyme was reduced with DLA in the
presence of NADþ, both emission peaks drastically
decreased at 330 nm and 510 nm, since the FAD
molecule was reduced. Concomitantly, a 460 nm emis-
sion appeared. It represented the formation of NADH as
result of the oxidation of DLA by rBfmBC. This peak
increased after 5 min of incubation, indicating the
ongoing reaction and formation of NADH, but this
460 nm peak disappeared over time (data not shown).
This might have been due to dissociation of NADH from
rBfmBC. These observations were similar to those for
recombinant human DLD.25)
In many intrinsic fluorescence spectral studies, a
wave length of about 295 nm has been used to excite
Trp residues and to avoid excitation of Tyr and Phe
residue(s). Although rBfmBC contains no Trp residue
and does contain 14 Tyr and 10 Phe residues in the
molecule, emission peaks were obtained at 330 and
510 nm by excitation at 296 nm. The fluorescence
spectrum due to 296-nm excitation was similar to that
due to 260 nm excitation (Fig. 3C).
DLDs have been studied in eukaryotes and a wide
range of aerobic bacteria, but less studied in strictly
anaerobic clostridia such as C. kluyveri. In this study,
we expressed and purified a C. kluyveri DLD (BfmBC)
that was different from the major diaphorase of this
bacterium, and found that the recombinant protein
showed enzyme properties similar to those of DLDs
isolated from other origins (Table 1). However, the
high stability of rBfmBC may make it useful for
practical use.
5) Perham, R. N., Domains, motifs and linkers in 2-oxo
acid dehydrogenase multienzyme complexes: a para-
digm in the design of a multifunctional protein.
Biochemistry, 30, 8501–8512 (1991).
6) Kikuchi, G., and Hiraga, K., The mitochondrial glycine
cleavage system. Mol. Cell. Biochem., 45, 137–149
(1982).
7) Hon, W. M., Chhatwal, V. J. S., Khoo, H. E., and
Moochhala, S. M., Histochemical method for detecting
nitric oxide synthase activity in cell cultures. Biotech.
Histochem., 78, 29–32 (1997).
8) Kaplan, F., Setlow, P., and Kaplan, N. O., Purification
and properties of a DPHN-TPNH diaphorase from
Clostridium kluyveri. Arch. Biochem. Biophys., 132,
91–98 (1969).
9) Chakraborty, S., Sakka, M., Kimura, T., and Sakka, K.,
Cloning and expression of a Clostridium kluyveri gene
responsible for diaphorase activity. Biosci. Biotechnol.
Biochem., in press.
10) Annual RITE report FY2003, The international co-
operation program for global environment promotion,
research and development of a highly efficient biopro-
cess for production of valuable chemicals from protein-
containing biomass and organic waste. 46–122 (2003).
11) Reed, L. K., Koike, M., Levitch, M. E., and Leach, F. R.,
Studies on the nature and reactions of protein-bound
lipoic acid. J. Biol. Chem., 232, 143–158 (1958).
12) Bradford, M. M., A rapid and sensitive method for
quantitation of microgram quantities of protein utilizing
the principle of protein-dye binding. Anal. Biochem., 72,
248–254 (1976).
13) Layne, E., Spectrophotometric and turbidimetric meth-
ods for measuring proteins. Methods Enzymol., 3, 447–
454 (1957).
14) Hope, B. T., Michael, G. J., Knigge, K. M., and Vincent,
S. R., Neuronal NADPH diaphorase is a nitric oxide
synthase. Proc. Natl. Acad. Sci. USA, 88, 2811–2814
(1991).
15) Gonzalez, C. F., Ackerley, D. F., Park, C. H., and
Martin, A., Soluble flavoprotein contributes to chromate
reduction and tolerance by Pseudomonas putida. Acta
Biotechnol., 23, 233–239 (2003).
Acknowledgments
16) Serrano, A., Purification, characterization and function
of dihydrolipoamide dehydrogenase from cyanobacteri-
um Anabaena sp. strain P.C.C.7119. Biochem. J., 288,
823–830 (1992).
We thank Dr. Hideaki Yukawa of RITE for providing
unpublished sequence data for C. kluyveri.

988
S. CHAKRABORTY et al.
17) Dietrichs, D., Meyer, M., Schmidt, B., and Andreesen,
J. R., Purification of NADPH-dependent electron trans-
ferring flavoproteins and N-terminal protein sequence
data of the dihydrolipoamide dehydrogenase from
anaerobic, glycine-utilizing bacteria. J. Bacteriol., 172,
2088–2095 (1990).
18) Dietrichs, D., and Andreesen, J. R., Purification and
comparative studies of dihydrolipoamide dehydrogenase
from anaerobic, glycine-utilizing bacteria Peptostrepto-
coccus glycinophilus, Clostridium cylindrosporum, and
Clostridium sporogenes. J. Bacteriol., 172, 243–251
(1990).
19) Nishimoto, E., Aso, Y., Koga, T., and Yamashita, S.,
Thermal unfolding process of dihydrolipoamide dehy-
drogenase studies by fluorescence spectroscopy. J.
Biochem., 140, 349–357 (2006).
20) Danson, M. J., Eisenthal, R., Hall, S., Kessel, S. R., and
Williams, D. L., Dihydrolipoamide dehydrogenase from
halophilic archaebacteria. Biochem. J., 218, 811–818
(1984).
21) Fazekas, A. G., and Kokai, K., Extraction, purification,
and separation of tissue flavins for spectrophotometric
determination. Methods Enzymol., 18B, 385–398 (1971).
22) Oppermann, B. F., Schmidt, B., and Steinbuchel, A.,
Purification and characterization of acetoin:2,6-dichloro-
phenolindophenol oxidoreductase, dihydrolipoamide
dehydrogenase, and dihydrolipoamide acetyltransferase
of the Pelobacter carbinolicus acetoindehydrogenase
enzyme system. J. Bacteriol., 173, 757–767 (1991).
23) Fukushima, E., Shinka, Y., Fukui, T., Atomi, H., and
Imanaka, T., Methionine sulfoxide reductase from the
hyperthermophilic archaeon Thermococcus kodakaraen-
sis, an enzyme designed to function at suboptimal
growth temperatures. J. Bacteriol., 189, 7134–7144
(2007).
24) Schmincke-Ott, E., and Bisswanger, H., Dihydrolipoa-
mide dehydrogenase component of the pyruvate dehy-
drogenase complex from Escherichia coli K12. Eur. J.
Biochem., 114, 413–420 (1981).
25) Liu, T., Lioubov, G., Susannah, L. H., and Mulchand, P.,
Spectroscopic studies of the characterization of recombi-
nant human dihydrolipoamide dehydrogenase and its
site-directed mutants. J. Biol. Chem., 270, 15545–15550
(1995).
26) Kenney, W. C., Zakim, D., Hogue, P. K., and Singer,
T. P., Multiplicity and origin of isoenzymes of lipoyl
dehydrogenase. Eur. J. Biochem., 28, 253–260 (1972).
27) Wilson, J. E., A comparative study of the multiple forms
of pig heart lipoyl dehydrogenase. Arch. Biochem.
Biophys., 144, 216–223 (1971).
28) Lohrer, H., and Krauth-Siegel, R. L., Purification and
characterization of lipoamide dehydrogenase from Try-
panosoma cruzi. Eur. J. Biochem., 194, 863–869 (1990).
29) Burns, G., Sykes, P. J., Hatter, K., and Sokatch, J. R.,
Isolation of a third lipoamide dehydrogenase from
Pseudomonas putida. J. Bacteriol., 171, 665–668 (1989).
30) Danson, M. J., Archaebacteria: the comparative enzy-
mology of their central metabolic pathways. Adv.
Microb. Physiol., 29, 165–231 (1988).
31) Danson, M. J., McQuattie, A., and Stevenson, K. J.,
Dihydrolipoamide dehydrogenase from halophilic
archaebacteria: purification and properties of the enzyme
from Halobacterium halobium. Biochemistry, 25, 3880–
3884 (1986).
32) Sundquist, A. R., and Fahey, R. C., The novel disulfide
reductases bis-ꢁ-glutamylcystine reductase and dihydro-
lipoamide dehydrogenase from Halobacterium halobi-
um: purification by immobilized-metal-ion affinity
chromatography and properties of the enzymes. J.
Bacteriol., 170, 3459–3467 (1988).