Electron transfer chain reaction of the extracellular
flavocytochrome cellobiose dehydrogenase from the
basidiomycete Phanerochaete chrysosporium
Kiyohiko Igarashi
1
, Makoto Yoshida
1
, Hirotoshi Matsumura
2
, Nobuhumi Nakamura
2
,
Hiroyuki Ohno
2
, Masahiro Samejima
1
and Takeshi Nishino
3
1 Department of Biomaterials Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Japan
2 Department of Biotechnology, Tokyo University of Agricultural and Technology, Japan
3 Department of Biochemistry and Molecular Biology, Nippon Medical School, Tokyo, Japan
Cellulose is the most abundant natural polymer on
earth, and its degradation is thus an important compo-
nent of the carbon cycle. Although cellulose is often
referred to as a b-linked glucose polymer, cellobiose, a
b-1,4-linked glucose dimer, should strictly be regarded
as the repeating unit of cellulose, because adjacent
glucoses show opposing faces to each other in the cel-
lulose chain [1,2]. Many microorganisms recognize this
repeating unit and hydrolyze cellulose to cellobiose as
an initial step in the metabolism [3]. In filamentous

steady-state reduction of flavin was not affected by the mutation, the rate
of subsequent electron transfer from flavin to heme was halved in F166Y.
When WT or F166Y was reduced with cellobiose and then mixed with
cytochrome c, heme re-oxidation and cytochrome c reduction occurred syn-
chronously, suggesting that the initial electron is transferred from reduced
heme to cytochrome c. Moreover, in both enzymes the observed rate of
the initial phase of cytochrome c reduction was concentration dependent,
whereas the second phase of cytochrome c reduction was dependent on the
rate of electron transfer from flavin to heme, but not on the cytochrome c
concentration. In addition, the electron transfer rate from flavin to heme
was identical to the steady-state reduction rate of cytochrome c in both
WT and F166Y. These results clearly indicate that the first and second
electrons of two-electron-reduced CDH are both transferred via heme, and
that the redox reaction of CDH involves an electron-transfer chain mech-
anism in cytochrome c reduction.
Abbreviations
CDH, cellobiose dehydrogenase; F166Y, Phe166Tyr mutant CDH; NHE, normal hydrogen electrode; WT, wild-type CDH.
FEBS Journal 272 (2005) 2869–2877 ª 2005 FEBS 2869
concerning the contribution of cellobiose dehydroge-
nase (CDH; EC 1.1.99.18) to the extracellular cellulose
metabolism of the white-rot fungus Phanerochaete
chrysosporium, and the importance of a combination
of hydrolytic and oxidative reactions in cellulose-
degrading fungi as discussed previously [7–13].
CDH is the only extracellular flavocytochrome
known to be secreted by filamentous fungi during cel-
lulose degradation [11–13]. This enzyme carries flavin
and a b-type heme in different domains, and the flavin
domain catalyzes the dehydrogenation of cellobiose
and cello-oligosaccharides to the corresponding d-lac-

Pichia pastoris. Presteady-state reduction of cyto-
chrome c and re-oxidation of heme in the recombinant
wild-type (WT) and mutant enzymes were observed by
sequential mixing in a three-syringe stopped-flow spec-
trophotometer to clarify the redox mechanism of
CDH.
Results
Redox properties of WT and F166Y CDH
The redox potentials of WT and F166Y were com-
pared at various pH values, as shown in Fig. 2A. The
potential of F166Y was lower than that of WT at all
pH values, but the difference was larger at lower pH
than at neutral pH. The potentials of WT and F166Y
were estimated to be 176 and 151 mV vs. normal
hydrogen electrode (NHE) at pH 4.0. Although the
potential of F166Y is 25 mV lower than that of WT, it
is still high enough to receive the electron from
reduced flavin, because addition of cellobiose causes
spectral changes in F166Y from the oxidized to the
reduced form (Fig. 2B). Reduced F166Y has typical
absorption maxima of CDH at 562, 533 and 428 nm
due to a-, b- and c-(Soret) bands, respectively, and
the absorption decreased at around 480 nm mainly
because of flavin reduction. These results are essen-
tially the same as those for the WT enzyme [24] and
suggest that both flavin and heme are active in the
mutant enzyme.
Presteady-state and steady-state kinetics
of F166Y
The presteady-state reduction of flavin and the subse-

values
for heme reduction of WT and F166Y were
30.2 ± 1.2 and 12.5 ± 0.9 s
)1
, respectively. These
results suggest that only the electron transfer step from
flavin to heme was halved in F166Y, whereas the ini-
tial flavin reduction was not affected by the muta-
tion. Presteady-state kinetic experiments for flavin and
heme reduction were carried out at various substrate
concentrations, as shown in Fig. 4. The k
obs
values for
flavin reduction were identical for WT and F166Y
(Fig. 4A), whereas those of heme reduction in F166Y
was almost half of that of WT at all substrate concen-
trations tested (Fig. 4B). The dissociation constants of
WT (K
WT
d
) and F166Y (K
F166Y
d
) obtained from the
plots were 107 ± 6.6 and 111 ± 6.5 lm, respectively.
As shown in a previous presteady-state kinetic study
of WT CDH, heme reduction was inhibited at high
substrate concentrations. In this study, this phenom-
A
B

1230 ± 180 lm), reflecting the similar K
d
values of the
two enzymes. The limiting rates of heme reduction for
WT (k
WT
lim
) and F166Y (k
F166Y
lim
) were 46.9 ± 6.8 and
18.8 ± 1.0 s
)1
, respectively.
The steady-state kinetic parameters are summarized
in Table 1. As expected from the presteady-state
experiment, there is no difference in these parameters
between WT and F166Y using ubiquinone as an
electron acceptor, whereas the mutation affected the
kinetic parameters when the redox reaction was mon-
itored in terms of cytochrome c reduction. The k
cat
values of cellobiose oxidation monitored in terms of
cytochrome c reduction were quite similar to the k
lim
values of heme reduction in both WT and F166Y.
Interestingly, an increase in cytochrome c concentra-
tion inhibited its reduction only in the case of
F166Y.
Sequential mixing experiment of WT and F166Y

(k
obs
) for flavin (A) and heme (B) reduction in WT (s) and F166Y
(j). k
obs
values for both prosthetic groups were obtained under the
same conditions as in Fig. 3, using 25–500 l
M cellobiose as a sub-
strate. The fitting of the data was performed as described in
Experimental procedures.
Table 1. Steady-state kinetic constants for WT and F166Y. All measurements were carried out at 30 °Cin50mM sodium acetate buffer,
pH 4.0. Cellobiose oxidation was monitored by following the reduction of 1 m
M ubiquinone or 50 lM cytochrome c as described in Experi-
mental procedures. ND, no significant substrate inhibition was observed.
Cellobiose oxidation with electron acceptors
Ubiquinone Cytochrome c
Ubiquinone
reduction Cytochrome c reduction
K
m
(lM)
k
cat
(s
)1
)
K
m
(lM)
k

heme reduction in both enzymes at the same cellobiose
concentration. As shown in Fig. 6B,D, heme in WT
and F166Y remained oxidized during this phase, but
was re-reduced with reduction of cytochrome c. The
k
obs
of cytochrome c reduction depended on the con-
centration of cytochrome c in the region tested (5–
20 lm, data not shown), and was almost identical for
WT and F166Y with limiting values of 1460 ± 140
and 1380 ± 80 s
)1
, respectively.
Discussion
Two mechanisms, electron transfer chain and electron
sink, have been proposed for the redox reaction of
CDH, and several previous kinetic studies have attemp-
ted to clarify the overall reaction of this enzyme [21–23].
However, uncertainty remains, possibly because of the
special features of this enzyme. The optimum pH values
of flavin reduction by cellobiose (pH 4.5–5.0) and of
electron transfer from flavin to heme (pH 3.5–4.0) differ
from each other, and the rate-limiting step of the reac-
tion thus depends on the pH of the reaction mixture
[20,25]. Moreover, a higher concentration of substrate
(cellobiose) inhibits presteady-state heme reduction, but
not flavin reduction [20,26], suggesting that binding of
substrate to the active site of the flavin domain inhibits
electron transfer from flavin to heme. This phenomenon
makes it difficult to solve the redox mechanism of this

transfer step from flavin to heme, but not initial flavin
reduction or cytochrome c reduction.
WT or F166Y was first mixed with cellobiose, and
then the cellobiose–CDH mixture was mixed with cyto-
chrome c after 0.1 s (WT) or 0.2 s (F166Y). At the
second mixing time,  90% of heme is in the ferrous
state and the flavin forms a semiquinone, as confirmed
A
B
Fig. 5. Absorption spectra of heme in CDH (A) and cytochrome c
(B) for comparison of the isosbestic points. The spectra of the oxi-
dized (solid line) and reduced (dashed line) forms are compared in
the range of 450–650 nm. Dotted lines at 549.0 (left) and 556.7
(right) show the isosbestic points of heme in CDH and cyto-
chrome c, respectively.
K. Igarashi et al. Electron transfer chain reaction of CDH
FEBS Journal 272 (2005) 2869–2877 ª 2005 FEBS 2873
previously by EPR [20], suggesting that most of the
enzyme is in the two-electron reduced form. In previous
studies, prereduced CDH was often used to observe the
electron transfer from heme to cytochrome c with cello-
biose or ascorbate as an electron donor [22,23,28]. With-
out a sequential mixing technique, however, it is difficult
to monitor cytochrome c reduction, because premixing
with the electron donors produces one- (ascorbate) or
three- (cellobiose) electron-reduced CDH, but not the
two-electron-reduced form with flavin radical and
reduced heme. Soon after the two-electron-reduced
CDH was mixed with cytochrome c, synchronous cyto-
chrome c reduction and re-oxidation of heme were

M) was mixed with 100 lM cellobiose, and 20 lM cytochrome c was then added and mixed after 0.1 s (WT) or 0.2 s (F166Y)
using a sequential mixing stopped-flow apparatus. The absorption changes were monitored after mixing with cytochrome c, and the initial
(0–0.020 s) and secondary (0–0.2 s) phase are seen in left (A, C) and right (B, D) panels, respectively. Conditions: 50 m
M sodium acetate buf-
fer (pH 4.0) at 30 °C.
Electron transfer chain reaction of CDH K. Igarashi et al.
2874 FEBS Journal 272 (2005) 2869–2877 ª 2005 FEBS
In the same report, moreover, they proposed that the
heme in CDH acts as an electron sink, because Rogers
and co-workers reported that heme is oxidized during
steady-state cytochrome c reduction [28]. Although this
phenomenon was also observed in the second phase of
presteady-state cytochrome c reduction, it is because of
the significant difference between the electron transfer
rate from flavin to heme (30 s
)1
) and that from heme to
cytochrome c ( 1500 s
)1
). After the electron is loaded
from flavin to heme, it is transferred to cytochrome c
without any significant time lag. Consequently, all the
results obtained in this study apparently indicate that
the overall redox reaction of CDH occurs through
the electron transfer chain mechanism, as shown in
Scheme 1.
Although this study clearly demonstrates an electron
transfer chain mechanism of CDH when cytochrome c
is used as an electron acceptor, it is too early to con-
clude that the mechanism is also used during cellulose

cytochrome c were also measured at various concentrations
(0–1 mm for ubiquinone and 0–50 lm for cytochrome c)
using 500 lm cellobiose as a substrate. The reductions of
ubiquinone and cytochrome c were monitored photometri-
cally at 406 nm (De
406
¼ 0.745 mm
)1
cm
)1
) and 550 nm
(De
550
¼ 17.5 mm
)1
cm
)1
), respectively. Because apparent
substrate inhibition was observed when cytochrome c was
used as an electron acceptor, the obtained substrate
dependence plots were fitted to the Michaelis–Menten equa-
tion with a substrate inhibition constant (K
i
). Unless other-
wise noted, steady-state kinetic parameters (K
m
and k
cat
)
were estimated by nonlinear fitting of the data to the

facturer’s instructions, and the vector pCR4
Ò
Blunt-
TOPO ⁄ f166y was digested with EcoRI and XbaI (TaKaRa
Bio, Shiga, Japan) and ligated into the pPICZa-A vector
(Invitrogen) at the same restriction sites. The vector
pPICZa-A ⁄ f166y was then linearized with Bpu1102I
(TaKaRa Bio) and transformed into Pichia pastoris KM-
71H using a MicroPulser electroporation device (Bio-Rad
Laboratories, Hercules, CA, USA). The Zeocin-resistant
transformant was cultivated in a growth medium (1% yeast
extract, 2% polypeptone, 1% glycerol; w ⁄ v) for 24 h at
30 °C, followed by the induction medium (1% yeast
extract, 2% polypeptone, 1% methanol; w ⁄ v) for 48 h at
26.5 °C, and F166Y was purified from the culture filtrate
with the same protocol as described previously [24]. The
purity of wild-type CDH and F166Y was confirmed by
SDS ⁄ PAGE and by the absorption spectrum.
Measurement of midpoint potential of heme
in CDH
A direct electrochemical technique was used to measure the
mid-point potential according to our previous report [30].
Glassy carbon, platinum, and Ag ⁄ AgCl were used as the
working, counter, and reference (+205 mV vs. NHE at
25 °C) electrodes, respectively. Cyclic voltammetry was
performed in the presence of 50 mm MgCl
2
using an ALS
Electrochemical Analyzer 624A, and the potential was deter-
mined by averaging the anodic and cathodic peak potentials.

16.8 mm
)1
Æcm
)1
) and 556.7 nm for heme in CDH
(De
556.7
¼ 9.29 mm
)1
Æcm
)1
), was monitored. All presteady-
state measurements were carried out at least three times in
50 mm sodium acetate buffer pH 4.0 at 30 °C, and the data
were analyzed as described previously [20].
Acknowledgements
This research was supported by Grants-in-Aid for Sci-
entific Research to KI (No. 15780206), MS (No.
14360094), and TN (No. 16205021), and by Grant-in-
Aid for Scientific Research on Priority Area to TN
(No. 12147208) from the Ministry of Education, Cul-
ture, Sports, Science and Technology, and a Research
Fellowship to MY (No. 08446) from the Japan Society
for the Promotion of Science.
References
1 Nishiyama Y, Sugiyama J, Chanzy H & Langan P
(2003) Crystal structure and hydrogen bonding system
in cellulose 1a, from synchrotron X-ray and neutron
fiber diffraction. J Am Chem Soc 125, 14300–14306.
2 Nishiyama Y, Langan P & Chanzy H (2002) Crystal

novel extracellular hemoflavoenzyme from the white-rot
fungus Phanerochaete chrysosporium. Arch Biochem
Biophys 300, 705–713.
10 Bao W & Renganathan V (1992) Cellobiose oxidase of
Phanerochaete chrysosporium enhances crystalline cellu-
lose degradation by cellulases. FEBS Lett 302, 77–80.
11 Eriksson K-EL, Habu N & Samejima M (1993) Recent
advances in fungal cellobiose oxidoreductases. Enzyme
Microb Technol 15, 1002–1008.
12 Henriksson G, Johansson G & Pettersson G (2000) A
critical review of cellobiose dehydrogenases. J Biotech-
nol 78, 93–113.
13 Cameron MD & Aust SD (2001) Cellobiose dehydro-
genase. An extracellular fungal flavocytochrome.
Enzyme Microb Technol 28, 129–138.
14 Morpeth FF (1985) Some properties of cellobiose oxi-
dase from the white-rot fungus Sporotrichum pulverulen-
tum. Biochem J 228, 557–564.
15 Henriksson G, Pettersson G, Johansson G, Ruiz A &
Uzcategui E (1991) Cellobiose oxidase from Phanero-
chaete chrysosporium can be cleaved by papain into two
domains. Eur J Biochem 196, 101–106.
16 Wilson MT, Hogg N & Jones GD (1990) Reactions
of reduced cellobiose oxidase with oxygen. Is cello-
biose oxidase primarily an oxidase? Biochem J 270,
265–267.
17 Kremer SM & Wood PM (1992) Evidence that cello-
biose oxidase from Phanerochaete chrysosporium is pri-
marily an Fe(III) reductase. Kinetic comparison with
neutrophil NADPH oxidase and yeast flavocytochrome

Phanerochaete chrysosporium cellobiose dehydrogenase
in the methylotrophic yeast Pichia pastoris. Biosci Bio-
technol Biochem 65, 2050–2057.
25 Samejima M, Phillips RS & Eriksson K-EL (1992)
Cellobiose oxidase from Phanerochaete chrysosporium:
stopped-flow spectrophotometric analysis of pH-depen-
dent reduction. FEBS Lett 306, 165–168.
26 Jones GD & Wilson MT (1988) Rapid kinetic studies of
the reduction of cellobiose oxidase from the white-rot
fungus Sporotrichum pulverulentum by cellobiose. Bio-
chem J 256, 713–718.
27 Page CC, Moser CC & Dutton PL (2003) Mechanism
for electron transfer within and between proteins. Curr
Opin Chem Biol 7, 551–556.
28 Rogers MS, Jones GD, Antonini G, Wilson MT &
Brunori M (1994) Electron transfer from Phanerochaete
chrysosporium cellobiose oxidase to equine cytochrome c
and Pseudomonas aeruginosa cytochrome c-551. Biochem
J 298, 329–334.
29 Martinez D, Larrondo LF, Putnam N, Gelpke MD,
Huang K, Chapman J, Helfenbein KG, Ramaiya P,
Detter JC, Larimer F et al. (2004) Genome sequence of
the lignocellulose degrading fungus Phanerochaete chrys-
osporium strain RP78. Nat Biotechnol 22, 695–700.
30 Igarashi K, Verhagen MFJM, Samejima M, Schu
¨
lein
M, Eriksson K-EL & Nishino T (1999) Cellobiose dehy-
drogenase from the fungi Phanerochaete chrysosporium
and Humicola insolens: a flavohemoprotein from Humi-