In vitro modulation of cytochrome p450 reductase supported indoleamine 2,3-dioxygenase activity by allosteric effectors cytochrome b5 and methylene blue

Article abstract of DOI:10.1021/bi100022c

Indoleamine 2,3-dioxygenase (IDO) is a heme-containing dioxygenase involved in the degradation of several indoleamine derivatives and has been indicated as an immunosuppressive. IDO is an attractive target for therapeutic intervention in diseases which are known to capitalize on immune suppression, including cancer, HIV, and inflammatory diseases. Conventionally, IDO activity is measured through chemical reduction by the addition of ascorbate and methylene blue. Identification of potential coenzymes involved in the reduction of IDO in vivo should improve in vitro reconstitution systems used to identify potential IDO inhibitors. In this study we show that NADPH-cytochrome P450 reductase (CPR) is capable of supporting IDO activity in vitro and that oxidation of L-Trp follows substrate inhibition kinetics (k_cat = 0.89 ± 0.04 s ^-1 , K_m = 0.72 ±0.15 μM, and K_i = 9.4 ± 2.0 μM). Addition of cytochrome b5 to CPR-supported L-Trp incubations results in modulation from substrate inhibition to sigmoidal kinetics (k_cat = 1.7 ± 0.3 s^-1, K_m = 1.5 ± 0.9 μM, and K_i = 1.9 ± 0.3). CPR-supported D-Trp oxidations (±cytochrome b₅) exhibit Michaelis-Menten kinetics. Addition of methylene blue (minus ascorbate) to CPR-supported reactions resulted in inhibition of DTrp turnover and modulation of L-Trp kinetics from allosteric to Michaelis-Menten with a concurrent decrease in substrate affinity for IDO. Our data indicate that CPR is capable of supporting IDO activity in vitro and oxidation of tryptophan by IDO displays substrate stereochemistry dependent atypical kinetics which can be modulated by the addition of cytochrome b₅.

Full text of DOI:10.1021/bi100022c

Biochemistry 2010, 49, 2647–2656 2647
DOI: 10.1021/bi100022c
In Vitro Modulation of Cytochrome P450 Reductase Supported Indoleamine
2,3-Dioxygenase Activity by Allosteric Effectors Cytochrome b5 and Methylene Blue
Josh T. Pearson,*,‡ Sophia Siu,§ David P. Meininger,§ Larry C. Wienkers,‡ and Dan A. Rock‡
§
‡Biochemistry and Biophysics Group, Department of Pharmacokinetics and Drug Metabolism and Department of Protein Sciences,
Amgen Inc., 1201 Amgen Court West, Seattle, Washington 98119-3105
Received November 20, 2009; Revised Manuscript Received February 23, 2010
ABSTRACT: Indoleamine 2,3-dioxygenase (IDO) is a heme-containing dioxygenase involved in the degradation
of several indoleamine derivatives and has been indicated as an immunosuppressive. IDO is an attractive target
for therapeutic intervention in diseases which are known to capitalize on immune suppression, including
cancer, HIV, and inflammatory diseases. Conventionally, IDO activity is measured through chemical
reduction by the addition of ascorbate and methylene blue. Identification of potential coenzymes involved
in the reduction of IDO in vivo should improve in vitro reconstitution systems used to identify potential IDO
inhibitors. In this study we show that NADPH-cytochrome P450 reductase (CPR) is capable of supporting
IDO activity in vitro and that oxidation of
m = 0.72 ( 0.15 μM, and Ki = 9.4 ( 2.0 μM). Addition of cytochrome b5 to CPR-supported
tions results in modulation from substrate inhibition to sigmoidal kinetics (kcat = 1.7 ( 0.3 s-1, Km = 1.5 ( 0.9
μM, and Ki = 1.9 ( 0.3). CPR-supported -Trp oxidations ((cytochrome b5) exhibit Michaelis-Menten
kinetics. Addition of methylene blue (minus ascorbate) to CPR-supported reactions resulted in inhibition of
Trp turnover and modulation of -Trp kinetics from allosteric to Michaelis-Menten with a concurrent
L
-Trp follows substrate inhibition kinetics (kcat = 0.89 ( 0.04 s-1
-Trp incuba-
,
K
L
D
D-
L
decrease in substrate affinity for IDO. Our data indicate that CPR is capable of supporting IDO activity in vitro
and oxidation of tryptophan by IDO displays substrate stereochemistry dependent atypical kinetics which can
be modulated by the addition of cytochrome b5.
Indoleamine 2,3-dioxygenase (IDO)1 is a heme-containing
dioxygenase responsible for the oxidative cleavage of several
indoleamine derivatives (1-3), including the initial and rate-
limiting step of tryptophan degradation along the kynurenine
pathway (4). IDO is widely expressed in a variety of cell types
including macrophages, eosinophils, dendritic cells, B cells,
endothelial cells, and various types of tumor cells (5-10), with
IDO gene expression regulated by various cytokines, most
importantly by interferon-γ (IFNγ) (11, 12). IDO has been
implicated as an immunosupporessive, contributing to maternal
tolerance toward the allogeneic fetus (13), suppression of trans-
plant rejection (14, 15), and regulation of autoimmune
disorders (16-18), with its expression in cancer cells correlating
with poor prognosis (19, 20).
the subsequent protein folding environment used to bind the
heme. In combination, the two structural features create the
active protein (holoenzyme), whereby the binding of various
cofactors, electron transfer partners, and/or ligands can impact
electron transfer and modulate catalytic activity in order to
control the extent of biochemical response (24). Therefore, prior
to targeting a heme protein for disease modification it is
imperative to develop a basic understanding of a heme protein’s
sensitivity toward different reducing systems. Such efforts can
provide assurance that the reconstituted in vitro enzyme activity
represents the activity under physiological conditions.
For IDO, the endogenous cofactor has yet to be identified.
Recently, Maghzal et al. reported cytochrome b5 is capable of
reducing Fe3þ-IDO to support catalytic activity (25). In this
study, the authors showed direct reduction of IDO by cyto-
chrome b5 in a system containing NADPH-cytochrome P450
reductase (CPR) and a NADPH regenerating system. Activity
assays with increasing concentrations of cytochrome b5 resulted
in an increase in kynurenine formation, and knock down of
cytochrome b5 protein expression in HEK 293 cells by siRNA
Heme-containing proteins represent a vital class of enzymes
often involved in physiological homeostasis, whereupon their
activity can be modulated by their electron transfer
partners (21-23). Moreover, heme protein activity is impacted
by the ring structure used to sequester the iron (heme form) and
*Corresponding author. Telephone: (206) 265-7823. Fax: (206) 217-
0492. E-mail: joshuap@amgen.com.
resulted in decreased IDO activity. At 400 μM L-Trp, CPR alone
1Abbreviations: IDO, indoleamine 2,3-dioxygenase;
L
-Trp,
L-trypto-
showed no activity but was a requirement for cytochrome b5
reduction of Fe3þ-IDO and enzymatic activity. Although indir-
ectly, Vottero and co-workers were the first to observe a potential
role for cytochrome b5 as an electron transfer coenzyme to IDO
by the use of a tryptophan auxotroph yeast strain transduced
with human IDO (26). Their data showed that decreased levels of
cytochrome b5 led to an increase in cell growth, which was
interpreted to be due to lower activation of IDO and therefore
inhibition of Trp catabolism.
phan; D-Trp, D-tryptophan; CPR, NADPH-cytochrome P450 reduc-
tase (alternatively named P450 oxidoreductase); b5, cytochrome b5;
MethBlue, methylene blue; PBS, phosphate-buffered saline; P450,
cytochrome P450; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-
1-propanesulfonate; TCA, trichloroacetic acid; NADPH, β-nicotina-
mide adenine dinucleotide 20-phosphate reduced tetrasodium salt
hydrate; 1-
L
-MT, 1-L-methyl-Trp; DOPC, 1,2-dioleoyl-sn-glycero-3-
phosphocholine; DLPS, 1,2-dilauroyl-sn-glycero-3-phospho-
L-serine;
DLPC, 1,2-dilauroyl-sn-glycero-3-phosphocholine; kan, kanamycin;
TB, terrific broth.
r
2010 American Chemical Society
Published on Web 02/23/2010
pubs.acs.org/Biochemistry

2648 Biochemistry, Vol. 49, No. 12, 2010
Pearson et al.
The discovery of a potential in vivo electron transfer partner
capable of reducing IDO and supporting activity is fundamental
to furthering our understanding of IDO biochemistry and
elucidating inhibitors to modulate its activity in vivo. To date,
kinetic characterization of purified human IDO has been con-
ducted through the use of a chemical reducing system that
consists of ascorbate and methylene blue (1). The use of chemical
reducing systems can provide insight into the potential catalytic
rate and efficiency of IDO, but the data acquired may misrepre-
sent or conceal key ligand-enzyme interactions masking the
ability to understand ligand structure function relationships. For
example, interactions with electron transfer partners may lead to
conformational changes that alter kinetics or affinities of the
enzyme for its substrate(s), inhibitor(s), cofactors, or even
additional electron transfer partners.
CPR is the endogenous electron transfer partner for the
cytochrome P450 family of enzymes, transferring the obligate
two electrons from NADPH individually to the P450 at two
separate steps in the catalytic cycle (27). Cytochrome b5 has been
shown to stimulate P450 activity for various isoforms, either
through transfer of the second electron (from either CPR or
NADH cytochrome b5) (21, 22, 28) to P450 or through an
allosteric effect resulting in more efficient catalysis (23, 29-31).
Multiple ligand binding, homotropic or heterotropic, can also
induce an allosteric effect, resulting in atypical kinetics for P450
enzymes (32-36). Sono showed similar results for IDO, where-
upon binding of an effector increased the affinities for the
supplemented with 500 μM δ-aminolevulinic acid (ALA). Solu-
ble His(6)-IDO expression was induced via addition of 0.5 ng/mL
N-(β-ketocaproyl)-DL-homoserine lactone (Sigma) at cell densi-
ties of ∼0.5 OD at 600 nm. Overnight induction at 18 °C
produced purified protein lots of higher specific activity than
those induced for shorter periods of time or at higher tempera-
tures. Final cell densities of 12-13 OD at 600 nm were typical for
successful productions.
Cells were harvested by centrifugation at 5000g for 30 min at
4 °C. Cell pellets were homogenized in lysis buffer and passed
through a Microfluids model 110 L microfluidizer operating at
16000 psi nitrogen gas with the lysis cell and lysate receptacle
chilled on ice. Lysates were clarified via centrifugation at 4 °C for
45 min at 28000g prior to purification.
IDO lysis buffer consisted of 25 mM Tris, 20 mM imidazole,
pH 7.4, and 150 mM NaCl supplemented with EDTA-free
Complete protease inhibitor (Roche) and benzonase nuclease
HC (Novagen). Clarified cell lysate from 1 L of culture volume
was sterile filtered and loaded onto a 5 mL HisTrap FF (GE
Healthcare) column, preequilibrated in 25 mM Tris, pH 7.4,
20 mM imidazole, and 300 mM NaCl. The column was washed
with 25 mM Tris, 40 mM imidazole, pH 7.4, and 300 mM NaCl
and protein eluted with a 5 column volume linear gradient from
40 to 250 mM imidazole in 25 mM Tris, pH 7.4, and 300 mM
NaCl. Elution fractions containing human IDO were further
purified by preparative size exclusion chromatography (SEC) on
a HiLoad 26/60 Superdex 200 column (GE Healthcare) with a
mobile phase consisting of 50 mM Tris, pH 7.4, and 1 mM
EDTA. All purification steps were carried out at 4 °C on an Akta
Express chromatography system (GE Healthcare). SEC aliquots
were pooled based on relative His(6)-IDO content as assessed
by SDS-PAGE, flash-frozen in liquid nitrogen, and stored at
-80 °C. Purified protein was >95% pure as assessed by
analytical SEC with an A406/A280 ratio of 1.65.
LC-MS/MS Quantitation of Analytes. All quantitative
measurements were conducted on a 4000 Q-Trap from Applied
Biosystems (San Jose, CA) connected to a Shimadzu HPLC
including a degasser. A sample volume of 20 μL was injected onto
a Waters YMC ODS-AQ column (2.0 ꢀ 50 mm) at a flow rate of
250 μL/min. Initial HPLC conditions were 100% solvent A
(2.0% formic acid in water). The elution gradient increased
from 0% to 35% B (0.1% formic acid in acetonitrile) in 2 min,
35-90% B in 0.2 min, and was held at 90% B for 0.5 min before
equilibration back to initial conditions. Quantitation of kynu-
renine was conducted by using the MRM function monitored by
a Q1 set at m/z 209.07 and Q3 at m/z 146.40, with the declustering
potential at 41 and a collision energy of 25 (arbitrary units). Trp
levels were monitored by a Q1 m/z of 205.05 and Q3 188.10, with
a declustering potential of 55 and a collision energy of 17. For
quantitation, an internal standard, 3-NO-tyrosine, was moni-
tored by a Q1 set at m/z 227.09 and a Q3 m/z of 181.0, with the
declustering potential at 36 and a collision energy of 21.
Mass spectral settings common to all analytes were the dwell
time, 250 ms; curtain gas, 10; IonSpray gas 1 and 2 set at 40;
IonSpray voltage, 4500; and the source temperature at 450 °C.
Reconstitution of Purified Human Cytochrome CPR and
Human Cytochrome b5. Reconstitutions of purified CPR and
purified cytochrome b5 were conducted as originally described by
Shaw et al. (40). Briefly, purified CPR or purified CPR mixed
with purified cytochrome b5 was placed on ice for 5 min. A 1:1:1
mixture of the lipids, DOPC, DLPC, and DLPS, at a total lipid
concentration of 1 mg/mL was added to yield a final 0.1 mg/mL
substrates
changes in reaction kinetics (37). Substrate inhibition kinetics for
-Trp has also been observed for rabbit IDO (1, 38). Lu et al.
recently expanded upon these observations showing substrate
inhibition kinetics for -Trp with recombinant human IDO and
L- and D-Trp and led to substrate/effector-dependent
L
L
modulation of these kinetics by the effector 3-indole ethanol (39).
Even though evolutionarily distinct, the similarity to cyto-
chrome P450 enzymes in regard to effector binding, atypical
kinetics, and the potential interactions with b5 (25, 26) led us to
further investigate the potential roles CPR and cytochrome b5
play in modulating IDO enzymology. The present study reveals
that CPR alone is capable of supporting IDO enzymatic activity
and that IDO can be modulated differentially by the addition of
the coenzyme cytochrome b5 or the ligand methylene blue.
MATERIALS AND METHODS
Materials. L-kynurenine, L- and D-Trp, 1-L-methyl-Trp, for-
mic acid, 3-NO-tyrosine, methylene blue, trichloroacetic acid,
acetonitrile, and ascorbic acid were of the highest grade available
from Sigma-Aldrich (St. Louis, MO). Reduced β-NADPH was
purchased from EMD Chemicals, Inc. (Gibbstown, NJ), and
DOPC, DLPC, and DLPS were purchased from Avanti Polar
Lipids, Inc. (Alabaster, AL). Purified recombinant human cyto-
chrome b5 was purchased from Invitrogen, and purified recom-
binant human cytochrome P450 reductase was purchased from
BD Gentest.
Protein Expression and Purification. Residues 2-403 of
human IDO were fused to an N-terminal hexahistidine (His(6)
tag via PCR. Escherichia coli strain BL21 Star (Stratagene)
was transformed with pAMG21 His(6)-IDO for expression of
soluble enzyme. Transformed colonies selected on agarose plates
containing 50 μg/mL kanamycin (kan) were used to generate
small starter cultures in TB media with 50 μg/mL kan at 37 °C.
Starter cultures were scaled up to 1 L in TB/kan media

Article
Biochemistry, Vol. 49, No. 12, 2010 2649
concentration after dilution and allowed to sit on ice for an
additional 5 min. Following 5 min equilibration, CHAPS
(5 mg/mL) was then added to yield a final concentration of
0.5 mg/mL after dilution and allowed to sit on ice for 5 min.
Samples were then diluted with PBS and used as cofactor
working stocks.
kcat values, measured product concentrations were corrected for
time and protein concentration prior to determination of Vmax
values. IC50 data were fit using a sigmoidal dose-response model
(eq 3).
% remaining activity ¼ min þ ðmax - minÞ=
ð1 þ 10∧ðlog ½Iꢁ - log IC50ÞÞ
ð3Þ
IDO Enzymatic Activity Assays. IDO activity was assessed
by measuring the formation of kynurenine from either
L- or
After visual inspection of the Lineweaver-Burk plots, Ki data
were fit globally to the appropriate model and evaluated against
the other two potential models. Data were fit globally to the
mixed-inhibition model (eq 4), competitive (eq 5), or noncompe-
titive (eq 6).
D
-Trp versus time. Purified human IDO was diluted into PBS
(pH 7.4) containing either 20 mM ascorbate (brought to pH 7.4
by addition of NaOH) or reconstituted CPR ((cytochrome b5)
working stock and divided into 800 μL aliquots. Substrate
at varying concentrations, dissolved in 1:2 DMSO to water,
was diluted 100-fold into each respective aliquot, mixed by
pipet, and plated into eight 90 μL aliquots. Plates were pre-
incubated at 37 °C for 5 min before initiation of reaction with
10 μL of either 100 μM methylene blue (ascorbate incubations),
10 mM NADPH (CPR cofactor supported incubations), or PBS
for negative control (four of the eight replicates). Individual
incubations (100 μL) contained a final 0.2 pmol of IDO.
Incubations were allowed to proceed for 2 min at 37 °C with
shaking before being quenched by the addition of 200 μL of 10%
TCA in water containing 2 μM 3-NO-tryosine as the internal
standard. Plates were heat sealed and placed at 60 °C for 1 h (41),
centrifuged, and then analyzed by LC-MS/MS. Standard curves
ν ¼ ðVmax=ð1 þ ½Iꢁ=RKiÞÞ½Sꢁ=ððKmð1 þ ½Iꢁ=KiÞ=
ð1 þ ½Iꢁ=RKiÞÞ þ ½SꢁÞ
ð4Þ
ð5Þ
ð6Þ
ν ¼ Vmax½Sꢁ=ððKmð1 þ ½Iꢁ=KiÞÞ þ ½SꢁÞ
ν ¼ ðVmax=ð1 þ ½Iꢁ=KiÞÞ½Sꢁ=ðKm þ ½SꢁÞ
Definition of terms for the above equations are [I] is the
concentration of inhibitor in the system, [S] is the concentration
of substrate, Km is equal to the substrate concentration at half the
maximal reaction velocity, Vmax is equal to the maximal velocity,
Ki is the dissociation constant for the enzyme-inhibitor complex,
R describes the mechanism of inhibition, and n is the Hill
coefficient.
of L-kynurenine were prepared fresh each day and treated under
the same conditions as incubation samples. Negative control
samples were analyzed for kynurenine impurities at each sub-
strate concentration studied, averaged, and subtracted from the
measured activities.
Active Site Modeling. The crystallography coordinates of
IDO were used from 2D0T.pdb (43). The resolution of this
˚
ligand-bound structure was at 2.3 A. The pdb coordinates were
input into Maestro (Schrodinger, NY). The protein structure was
prepared using protein preparation script which steps through
structural optimization of protein structure followed by mini-
mization. Briefly, the initial structure was preprocessed and
analyzed for correct bond order, overlapping residues, added
any missing hydrogens, and removed water molecules beyond
IC50 Determination. The incubation time and protein con-
centrations used were within the linear range for each respective
cofactor system. In addition, no more than 15-20% turnover of
L-Trp occurred over the 2 min incubation time. L-Trp was run at a
concentration of 2 μM for the three cofactor systems, and all
incubations contained less than 1% organic solvent (v/v; acet-
onitrile). Incubation conditions (run in triplicate) were identical
to those listed above for enzymatic activity assays.
˚
5 A. Heteroatom ionization states were for the heme iron
followed by hydrogen bond optimization. The resulting structure
was minimized using an OLPS2005 force field to an rmsd of
Ki Determination. A matrix of six probe substrate concen-
˚
0.3 A. SiteMap (Schrodinger, NY) was used to identify R sites
trations (5, 10, 20, 25, 50, and 150 μM D-Trp) and six inhibitor
(active binding sites) on IDO. The SiteMap defined active site
concentrations (0, 0.01, 0.1, 1, 10, and 100 μM) was used to
determine a Ki value. Reaction conditions and sample prepara-
tion procedures were identical to those described above for the
IC50 determinations. All Ki determinations were run in triplicate.
Statistical Analysis. Standard curves and mass spectrometry
data were fit using Analyst (version 1.4; Applied Biosystems,
Foster City, CA). Standard curves were weighted using 1/x unless
stated otherwise. Analysis of kinetic and inhibition constant data
was performed using GraphPad Prism (version 5.00; GraphPad
Software Inc., San Diego, CA). Enzyme kinetic data were fit to
each of three models, either Michaelis-Menten, substrate in-
hibition (eq 1)
was used to generate a docking grid for Glide docking of
The final structure was minimized. SiteMap was reexamined
using the new L-Trp-bound structure as described above.
L-Trp.
RESULTS
CPR-Supported IDO Activity. The activity of purified
human IDO was examined with the addition of varying concen-
trations of reconstituted human CPR (0.7-75 nM). Incubations
utilizing -Trp as the substrate (0-20 μM) display substrate
L
inhibition kinetics (Figure 1A; Supporting Information Table 1)
with increasing levels of CPR increasing kcat but having no effect
on Km or Ki. The fitted parameters from eq 1 for Km and Ki are
0.72 ( 0.15 and 9.4 ( 2.0 μM, respectively. The kcat values range
from 0.29 ( 0.02 to 0.67 ( 0.06 s-1 for incubations with 0.7 to
ν ¼ Vmax½Sꢁ=ðKm þ ½Sꢁð1 þ ½Sꢁ=KiÞÞ
or allosteric sigmoidal (eq 2)
ð1Þ
75 nM CPR, respectively. Incubations with -Trp (0-300 μM)
D
n
n
ν ¼ Vmax½Sꢁ =ðK0 þ ½Sꢁ Þ
ð2Þ
display Michaelis-Menten kinetics (Figure 1B; Supporting
Information Table 1) with a Km value of 25.4 ( 3.2 μM and kcat
values that range from 0.36 ( 0.01 to 0.68 ( 0.03 s-1 for incuba-
tions with 0.7 and 75 nM CPR, respectively. IDO activity as a
function of CPR concentration (0-500 nM) in the presence of
and compared by statistical analysis and visual inspection
through the use of reciprocal plots. For the Hill equation (e.g.,
eq 2), K0 is a constant comprising the interaction factors and no
longer equals Km except when n is equal to 1 (42). For reported
saturating -Trp (300 μM) yields an “effective” Km for CPR of
D

2650 Biochemistry, Vol. 49, No. 12, 2010
Pearson et al.
F
IGURE 2: Cytochrome b5 dependent modulation of IDO activity.
(A) Modulation of CPR- (200 nM) supported L-Trp oxidation
kinetics with increasing concentrations of cytochrome b5. Concen-
trations of cytochrome b5 utilized were 0.0 (b), 6.6 (9), 13 (2), 33 (1),
and 66 ([). (B) D-Trp oxidation kinetics with increasing amounts of
cytochrome b5. (C) Plot of enzyme activity for L-Trp (20 μM) and
D-Trp (300 μM) versus cytochrome b5 concentration.
F
IGURE 1: CPR-supported IDO activity. (A) Kinetic traces display-
ing substrate inhibition for L-Trp oxidation by CPR-supported IDO
activity. CPR concentrations were 0.7 (b), 2.2 (9), 4.5 (2), 20 (1),
37.5 (]), and 75 nM (O). (B) Michaelis-Menten kinetics for CPR-
supported IDO utilizing D-Trp as substrate. CPR concentrations
were matched to (A). (C) Plot of enzyme activity for D-Trp (300 μM)
versus CPR concentration.
45.9 ( 1.3 μM in the presence of cytochrome b5 concentrations
ranging from 33 to 400 nM (Figure 2B; Supporting Information
Table 2). IDO activity as a function of cytochrome b5 concentra-
tion (5-1000 nM) in the presence of saturating concentrations
6.0 ( 0.6 nM (Figure 1C) and maximal kcat value of 0.87 (
0.01 s-1. Subsequent experiments utilized saturating concentra-
tions of CPR (200 nM) with varying amounts of cytochrome b5.
At 200 nM CPR,
L
-Trp catabolism by IDO has a maximal kcat
of
for cytochrome b5 of 33.4 ( 2.4 μM (
-Trp) with maximal kcat values for -Trp and
and 3.5 ( 0.1 s-1, respectively (Figure 2C).
Methylene Blue-Supported Tryptophan Oxidation. As-
corbate/methylene blue-supported oxidation of -Trp (0-200 μM)
L-Trp (20 μM) and
D
-Trp (300 μM) yield “effective” Km values
-Trp) and 28.0 ( 2.7 μM
-Trp of 1.7 ( 0.3
value of 0.89 ( 0.04 s-1 (Supporting Information Table 1).
Cytochrome b5 Modulation of Substrate Inhibition
L
(
D
L
D
Kinetics. At 6.6 nM cytochrome b5, -Trp (0-20 μM) oxidation
L
continues to display substrate inhibition kinetics but with an
increase in the Ki to 15.6 ( 1.9 μM (Figure 2A; Supporting
Information Table 2). No significant changes in Km or kcat are
observed compared to activities measured in the absence of
cytochrome b5. However, as cytochrome b5 concentrations in-
L
displays substrate inhibition kinetics with a Km of 3.1 ( 0.3 μM,
a Ki of 225 ( 36 μM, and a kcat of 0.99 ( 0.4 s-1 (Supporting
Information Figure 1A).
D-Trp (0-10 mM) with ascorbate/
crease from 13 to 400 nM, the kinetic behavior of
L-Trp changes
methylene blue-supported activity follows Michaelis-Menten
kinetics with a Km of 1.03 ( 0.23 mM and kcat of 1.7 ( 0.1 s-1
(Supporting Information Figure 1B). Experiments conducted
significantly. At high concentrations of cytochrome b5 the
kinetics display sigmoidal behavior, with Km and n (Hill co-
efficient) values of 1.5 ( 0.9 μM and 1.9 ( 0.3, respectively. The
which employ an extended concentration range of
L-Trp
oxidation of
D-Trp with increasing concentrations of cytochrome
(0-200 μM) in CPR-supported incubations display partial
substrate inhibition, i.e., never reaching complete inhibition of
activity (Figure 3). Addition of 250 nM cytochrome b5 to this
b5 continued to display Michaelis-Menten kinetics but results
in an increase in Km and kcat, with Km increasing ∼1.7-fold to

Article
Biochemistry, Vol. 49, No. 12, 2010 2651
F
IGURE 3: Methylene blue-supported tryptophan oxidation. Plot of
IDO activity versus log L-Trp concentration showing partial sub-
strate inhibition of CPR-supported incubations. Addition of 250 nM
cytochrome b5 abolishes the observed substrate inhibition. Methy-
lene blue-supported reactions also observe substrate inhibition but
with a lower affinity Ki. Added lines represent the Ki values for CPR-
(solid line) and methylene blue- (dashed line) supported L-Trp
oxidation reactions.
F
IGURE 5: Methylene blue-dependent modulation of CPR-sup-
ported L-Trp oxidation. (A) Modulation of CPR- (200 nM) sup-
ported L-Trp oxidation kinetics with increasing concentrations of
methylene blue: 0.0 (b), 0.1 (9), and 0.5 μM (2). (B) Addition of
methylene blue to CPR- (200 nm) plus cytochrome b5- (250 nM)
supported reactions continues to display allosteric sigmoidal kinetics
with increasing amounts of methylene blue up to 1 μM, but with a
concomitant decrease in the Hill coefficient. At 10 μM L-Trp reac-
tions display Michaelis-Menten kinetics with, by definition, a Hill
coefficient of 1 (see Table 1). (C) Plot of percent catalytic efficiency
(kcat/Km) from kinetic constants normalized to 0 μM methylene
blue (control) for CPR- (200 nM) and CPR (200 nM) with cyto-
chrome b5- (250 nM) supported incubations of L- and D-Trp. Addi-
tion of methylene blue (minus ascorbate) leads to a concentration-
dependent loss in catalytic efficiency.
F
IGURE 4: Inhibition of CPR-supported D-Trp oxidation by methy-
lene blue. Lineweaver-Burk plots for visual inspection of methylene
blue-dependent inhibition of D-Trp oxidation in CPR- (A) and CPR
plus cytochrome b5- (B) supported reactions. Concentrations of CPR
and cytochrome b5 were 200 and 250 nM, respectively.
system abolishes the observed partial substrate inhibition for
L-Trp but continues to display nonlinear, or atypical, kinetics.
Methylene blue-supported -Trp oxidation also displays sub-
L
strate inhibition kinetics, although with a higher Ki, which
renders the distinction between complete or partial substrate
inhibition indiscernible.
Methylene Blue Modulation of CPR-Supported L-Trp
Oxidation. Addition of methylene blue (0-30 μM) to CPR-
supported L-Trp oxidation reactions results in a switch from
Methylene Blue Inhibition of CPR-Supported D-Trp
Oxidation. Addition of methylene blue (0-100 μM) to CPR-
substrate inhibition kinetics to Michaelis-Menten with increas-
ing methylene blue concentration decreasing the affinity of IDO
supported
D
-Trp incubations ((cytochrome b5) leads to concen-
for
L-Trp (Figure 5A and Table 1). CPR plus cytochrome b5-
tration-dependent inhibition of IDO activity (Figure 4). For
incubations minus cytochrome b5, i.e., CPR alone, the data are
best described by a competitive inhibition model, with a Ki value
of 57.2 ( 13.6 μM. For CPR plus cytochrome b5, methylene blue
supported
L-Trp oxidation reactions continue to display sigmoi-
dal kinetics upon the addition of methylene blue up to a
concentration of 1 μM (Figure 5B) but result in a methylene
blue concentration-dependent decrease in affinity for the sub-
inhibition of
D-Trp oxidation followed a mixed mode of inhibi-
strate
L-Trp and the Hill coefficient (n value). At 10 and 30 μM
tion (44), with a Ki of 9.0 ( 1.9 μM and an R value of 3.6 ( 2.2.
methylene blue, a decrease in the affinity for
L-Trp continues to

2652 Biochemistry, Vol. 49, No. 12, 2010
Pearson et al.
Table 1: Kinetic Constants Determined for CPR-Supported L
-Trp Oxidation in the Presence of Methylene Bluea
MethBlue (μM)
[b5] (nM)
kinetic model
Kmb (μM)
kcat (1/s)
Kcat/Km (s-1μM-1
)
Ki (μM)
n (Hill coeff)
0.0
0.1
0.5
1.0
10
0
0
SIc
M-M
M-M
M-M
M-M
M-M
AS
0.72 ( 0.15
1.7 ( 0.1
3.9 ( 0.4
4.9 ( 0.4
12.9 ( 1.0
27.9 ( 8.0
1.5 ( 0.9
1.9 ( 0.2
4.0 ( 0.4
5.4 ( 0.7
12.4 ( 1.1
22.4 ( 3.9
0.87 ( 0.01
1.3 ( 0.1
2.5 ( 0.1
3.0 ( 0.1
6.4 ( 0.3
8.1 ( 1.6
1.7 ( 0.3
2.2 ( 0.1
3.4 ( 0.1
4.2 ( 0.2
10.0 ( 0.5
12.0 ( 1.5
1.2 ( 0.4
9.4 ( 2.0
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/Ad
0.76 ( 0.05
0.64 ( 0.06
0.61 ( 0.05
0.49 ( 0.04
0.29 ( 0.08
1.1 ( 0.71
1.0 ( 0.21
0.85 ( 0.17
0.74 ( 0.16
0.80 ( 0.16
0.54 ( 0.13
N/A
0
N/A
0
N/A
0
N/A
30
0
N/A
0.0
0.1
0.5
1.0
10
250
250
250
250
250
250
1.9 ( 0.3
1.3 ( 0.1
1.3 ( 0.1
1.2 ( 0.1
1.1 ( 0.1e
0.9 ( 0.1e
AS
AS
AS
M-M
M-M
b
30
aCPR concentration equal to 200 nM. Kinetic traces fit to the allosteric sigmoidal model (Hill equation) report K0 values where K0 comprises the interaction
factors for the n sites of substrate binding. cKinetic model abbreviations: SI, substrate inhibition; M-M, Michaelis-Menten; AS, allosteric sigmoidal. dN/A,
not applicable. eDetermined from fit to AS model; by statistical criteria the data were better represented by the Michaelis-Menten equation which is
equivalent to a Hill coefficient of 1.
IDO (Figure 7C) accounts for the volume displaced by
L
-Trp as
3
˚
evidenced by the similar R site volumes of 332 A compared to
3
˚
308 A for
L-Trp versus the apo structure, respectively. This
newly formed R site presents a novel site for cofactor interaction.
DISCUSSION
IDO is an attractive target for therapeutic intervention in
diseases which are known to capitalize on immune suppression
including cancer, HIV, and inflammatory diseases. Despite the
established role of IDO in immune suppression, only one clinical
candidate has been evaluated to date, 1-
-1MT), a weak inhibitor of IDO (Figure 8). The natural
product annulin B and menadione represent a structurally
distinct class of inhibitors from -1MT, and both compounds
D-methyltryptophan
(
D
F
IGURE 6: Reduction system dependent differences in menadione
IC50 values. Plot of the percent remaining activity for the oxidation of
L-Trp by IDO versus the log concentration of menadione when
utilizing our three different reducing systems, CPR (200 nM), CPR
(200 nM) þ b5 (250 nM), and ascorbate/methylene blue.
D
have been shown to inhibit IDO in vitro (41, 45). The inhibitors
shown in Figure 8 demonstrate the flexibility of IDO to bind
different chemical motifs, e.g., the indole of Trp to the naphtho-
quinoline moiety of annulin B, which possess unique physio-
chemical properties and spatially distinct pharmacophores.
Moreover, the redox activities of menadione may contribute to
the mechanism by which it inhibits IDO. The presence or absence
of different enzyme cofactors and/or electron transfer partners
may impact the dynamics of the active site. The sensitivity of IDO
toward different effectors, in combination with the limited
knowledge regarding its endogenous electron transfer partners
or cofactors essential for oxidative activity, requires further
attention to understanding the biochemistry of IDO.
be observed in parallel with increasing concentrations of methy-
lene blue, with the Hill coefficient reaching a value of 1 and
resulting in Michaelis-Menten kinetics (Table 1). Figure 5C
shows the concentration-dependent changes in catalytic effi-
ciency (kcat/Km) for CPR-supported
increasing methylene blue concentrations.
Cofactor-Dependent Inhibition of L-Trp Oxidation by
Menadione. Incubations of -Trp (at the corresponding Km
L- and D-Trp oxidation with
L
values) and IDO with varying concentrations of menadione
result in differential inhibition. With reactions supported by
CPR, menadione failed to inhibit
or absence of cytochrome b5 in the incubation (Figure 6).
Conversely, menadione was shown to effectively inhibit -Trp
oxidation from reactions supported with ascorbate/methylene
blue, yielding an IC50 value of 0.64 ( 0.08 μM, similar to the
previously reported value of 1.0 μM (41).
Model of Potential Effector Binding Site Adjacent to
Active Site. The principal active site was identified above the
heme using SiteMap as described in the Materials and Methods
L
-Trp regardless of the presence
Conventionally, IDO is reconstituted with methylene blue and
ascorbic acid to achieve oxidative activity toward Trp. The
artificial nature of this reconstitution system has the potential
to mask or alter the biochemical characteristics of the enzyme.
Here we show that reconstitution of recombinant purified CPR
into lipids according to the protocol published by Shaw (40),
followed by dilution into IDO reaction mixture (PBS, IDO
protein, and substrate), results in NADPH-dependent formation
of N-formylkynurenine, which is subsequently converted to
L
section (Figure 7A). Subsequent docking of
L-Trp was performed
kynurenine through acidification. With D-Trp as the substrate,
to generate a new ligand bound structure. The new ligand-bound
model was subjected to SiteMap to identify a potential allosteric
binding site from the newly formed complex (Figure 7B). The pri-
mary R site identified from the complex overlaps with the original
binding site but also occupies additional space adjacent to the
IDO follows Michaelis-Menten kinetics (Figure 1B, Supporting
Information Table 1) but observes substrate inhibition kinetics
with its endogenous substrate L-Trp (Figure 1A, Supporting
Information Table 1). Varying the concentration of CPR per
incubation leads to an increase in kcat with no change in the
heme and near bulk solvent. The new pocket from
L-Trp-bound
kinetic behavior observed for either substrate, L- or D-Trp

Article
Biochemistry, Vol. 49, No. 12, 2010 2653
FIGURE 8: Relevant chemical structures. Structures of 1-methyltryp-
tophan (1-MT), annulin B, menadione, and methylene blue.
concentrations of CPR of 0.87 ( 0.01 s-1 (Figure 1C). Due to the
potential differences in the conformation of IDO when
is bound versus -Trp (substrate inhibition kinetics versus
Michaelis-Menten), the “effective” Km value for oxidation of
-Trp may not accurately describe the interaction of CPR with
IDO for -Trp oxidation.
L-Trp
D
D
L
The addition of cytochrome b5 to reconstitutions of recombi-
nant CPR in lipids leads to an increase in catalytic rate for both
L
- and
D
-Trp, with
D
-Trp continuing to display Michaelis-
Menten kinetics but
L-Trp kinetics undergoing a cytochrome b5
concentration-dependent change from substrate inhibition to
sigmoidal. A plot of cytochrome b5 concentration versus reaction
velocity at saturating concentrations for both L- and D-Trp yields
equivalent “effective” Km values, arguing that in the presence of
CPR the interaction of IDO with cytochrome b5 is not altered by
the stereochemistry of the tryptophan substrate. We are currently
taking more direct approaches to measure the interactions
between CPR/IDO and CPR/IDO/cytochrome b5 in the presence
of various ligands and substrates.
The results herein demonstrate that CPR with NADPH
supports the biotransformation of tryptophan to N-formylky-
nurenine. These findings contradict previously reported data
from Maghzal et al., where CPR alone did not support NADPH-
dependent oxidation of tryptophan. This discrepancy could be
explained by two differences in our assays. Foremost, in our
hands the dilution of CPR without prior reconstitution into lipids
resulted in undetectable formation of kynurenine over back-
ground. However, CPR premixed with cytochrome b5 (without
lipids) before dilution into IDO reaction mixes did show slight
activity over background (data not shown). Second, the pre-
F
IGURE 7: Active site map of Apo IDO and L-Trp-bound IDO. (A)
Mapped surface area of the primary R site identified from Apo
IDO structure from 2D0T.pdb. The surface area was calculated to
3
be 1038 A . The hydrophobic surface contact points are highlighted
˚
vious study assessed activity at a single concentration of L-Trp
in purple. The relative hydrophobic/hydrophilic contact area of the R
site was 0.67. (B) Mapped surface area of the primary R site identified
from IDO with L-Trp docked within the active site. The hydrophobic
surface contact points are highlighted in purple. The relative hydro-
phobic/hydrophilic contact area of the R site was 0.57. (C) Compa-
rison ofthe R sites fromthe Apo vs L-Trp-docked modelsshowa large
degree of overlap. The R site from the bound structure displaces
accessible volume distal to the heme where L-Trp docks but adds
additional volume adjacent to heme with additional area for hydro-
phobic contact.
(400 μM). This concentration is approximately 45-fold above our
Ki for the observed substrate inhibition kinetics with CPR as the
cofactor. Therefore, under such conditions the activity may be
markedly suppressed in the absence of cytochrome b5.
The kinetic constants for ascorbate/methylene blue-supported
oxidation reactions of
reported in the literature with our Km values being slightly higher
affinity (for -Trp a value of 3 vs 7.1-20 μM and for -Trp 1 vs
-Trp a value
-Trp 1.7 vs 2.6-5.9 s-1) (39, 46, 47).
L- and D-Trp are comparable to values
L
D
2.6-5 mM) and our kcat values slightly slower (for
L
(Figure 1A,B). A plot of CPR concentration versus reaction
of ∼1 vs 2-5 s-1 and for
D
velocity at a saturating concentration of
D-Trp (300 μM) follows
These differences may be attributed to differences in assay design
(pH, buffering system, detection method, etc.). As reported
previously (37, 39), the ascorbate/methylene blue-supported
Michaelis-Menten kinetics with a Km of electron transfer from
CPR to IDO equal to 6.0 ( 0.6 nM and a maximal reaction
velocity for the IDO dependent oxidation of
D-Trp at saturating
oxidation of L-Trp displays substrate inhibition kinetics as

2654 Biochemistry, Vol. 49, No. 12, 2010
Pearson et al.
observed in our CPR system, but with an ∼24-fold increase in Ki;
surface area to that of the
L
-Trp binding pocket, suggesting the
i.e., in the presence of methylene blue the inhibitory substrate
new R site may also be suitable for binding an additional small
molecule. Moreover, the proximity of the secondary site is close
enough to directly impact interactions at the active site, either
through direct contact with the primary ligand or through
indirect interaction with the heme or adjacent protein structure
used to sequester the heme.
Whether the observed homotropic effects are due to substrate-
induced conformational changes or ligand/ligand interactions
within the active site requires further investigation. IDO is
susceptible to allosteric modulation by substrates, electron
transfer partners (cytochrome b5), and small molecule effec-
tors (37, 39). As summarized by Lu, the development of more
efficient IDO inhibitors will require further understanding of
IDO enzymology and further characterization of the effector
binding site. As shown here for menadione, incubation condi-
tions can lead to differential inhibition.
binding site has a lower affinity for the substrate
L-Trp (Figure 3,
Supporting Information Figure 1A). In parallel,
D-Trp sees a
similar decrease in affinity (∼12-20-fold change in Km) with the
ascorbate/methylene blue reduction system compared to the
CPR ((cytochrome b5) system. These observations warranted
further investigation into the potential role methylene blue plays
in disrupting substrate binding. Addition of methylene blue
(minus ascorbate) to CPR- ((cytochrome b5) supported IDO
incubations leads to altered kinetics for both isomers of Trp
(Figures 4 and 5). For D-Trp, addition of methylene blue results
in competitive inhibition for CPR-supported reactions which
changes to mixed-mode inhibition in the presence of cytochrome
b5 (250 nM). Addition of methylene blue to CPR- ((cytochrome
b5) supported L-Trp oxidation reactions results in loss of the
observed atypical kinetics (substrate inhibition for CPR and
sigmoidal for CPR plus cytochrome b5), yielding Michaelis-
Menadione effectively inhibits the oxidation of L-Trp using the
Menten kinetics. For both
L- and
D
-Trp, addition of methylene
chemical reductant system of ascorbate and methylene blue
whereas menadione does not exhibit inhibition using the CPR
((cytochrome b5) reduction system described herein. The me-
chanism for loss of inhibition is unclear; one proposal involves
the possibility that menadione disrupts methylene blue binding
and subsequent electron transfer to IDO but does not interfere
blue to CPR-supported incubations (( cytochrome b5) results in
concentration-dependent decrease in catalytic efficiency
a
(Figure 5C) driven primarily by a decrease in affinity for the
substrate.
The observed substrate inhibition kinetics for ascorbate/
methylene blue-supported IDO oxidation of
lished for concentrations of -Trp above 50 μM (37, 39). Lu et al.
proposed that the observed substrate inhibition kinetics were due
to binding of a second molecule of -Trp to an inhibitory site, Si,
which becomes available upon binding of the first molecule of
L
-Trp is well estab-
with L-Trp binding such that total inhibitory activity by mena-
L
dione is lost with the CPR reducing system. Other mechanisms
are possible. Incubations with CPR ( b5 require reconstitution of
the coenzymes into lipids, which could potentially sequester the
inhibitor menadione. One would expect, however, sequestration
of menadione to result in a shift of the IC50 curve to the right due
to a lower free fraction of menadione, not a complete loss of
inhibition. Another possible explanation for the lack of inhibition
by menadione in the CPR reducing system is the potential for
menadione to be reduced by methylene blue. Reduction of
menadione in the methylene blue/ascorbate system would disrupt
electron transfer to IDO. We are currently investigating further
the mechanism of menadione inhibition as menadione has been
reported to inhibit IDO in a clonal T-REx-derived cell line stably
transfected with doxycyclin-inducible IDO. Menadione, a known
cytotoxic agent (54, 55), exhibited an ∼4-fold difference between
inhibitory potency toward IDO (IC50 = 28.9 μM) and the
measured LD50 for cellular cytotoxicity (128 μM) (41), but these
data are difficult to interpret as cellular kynurenine levels were
measured utilizing the nonspecific aldehyde and ketone reactive
Ehrlich reagent (p-dimethylaminobenzaldehyde in strong acid
medium) which has been shown to lack selectivity for kynu-
renine (56).
The potential for alternate electron transfer partners for IDO
is a distinct possibility. Herein we present enzymatic activity from
IDO when reconstituted with CPR. However, other active
reductases, including NADH CoQ reductase (NOQ) and indu-
cible nitric oxide synthase (iNOS), share similar properties and
need to be investigated further as potential in vivo coenzymes of
IDO. iNOS in particular is of interest, as it has been characterized
as having a CPR-like domain (57), and iNOS and nitric oxide
formation can impact IDO activity (58, 59). Regardless of which
reductase(s) is the actual in vivo cofactor to IDO, the idea of IDO
allostery is an attractive one knowing its immunosuppressive
role, where enzymatic activity can be tightly regulated by
cofactors, effectors, or electron transfer partners and not by
changes in endogenous tryptophan levels (50-100 μM in
humans (60-62)). For example, the observed partial substrate
L
L-
Trp (39). Nickel et al. in collaboration with Lu have recently
shown by CO flash photolysis that CO rebinding is hindered by
elevated concentrations of
on this work and their ascorbate/methylene blue-supported
kinetics, the authors propose that -Trp binding at the Si site
L-Trp (4.8 and 20 mM) (48). Based
L
leads to substrate inhibition through the inhibition of O2
association (39, 48).
In principle, partial substrate inhibition kinetics arise from
allosteric interactions (36, 49). Allosteric interactions can arise
through ligand-induced conformational changes resulting from
substrate or effector binding or through substrate/effector or
substrate/substrate interactions within the active site (35, 50, 51).
CPR-supported L-Trp oxidation steady-state kinetics display
partial substrate inhibition which can be altered by the addition
of cytochrome b5, resulting in sigmoidal kinetics. The switch from
substrate inhibition to sigmoidal kinetics leads to an increase in
the rate of catalysis without loss of allosteric kinetics. This
increase in catalytic rate upon the addition of cytochrome b5,
therefore, could result from cytochrome b5/IDO protein interac-
tions leading to rearrangement of the IDO active site making O2
more accessible to the heme iron (as proposed by Nickel) or
rearrangement of bound substrate into a more productive
orientation above the catalytic iron.
In silico modeling was used to identify a potential effector
molecule binding site. SiteMap has been successfully utilized to
identify novel R sites (active sites) in proteins (52, 53). The
combination of SiteMap with the docking algorithm Glide
was employed in an attempt to identify the potential allosteric
site from an L-Trp-docked IDO structure (PDB entry 2D0T;
4-phenylimidazole-bound structure (43)). A site directly adjacent
to heme was identified as the primary R site (active site) from this
L-Trp-docked structure (Figure 7). Interestingly, the new site
created from the docked model possessed similar hydrophobic

Article
Biochemistry, Vol. 49, No. 12, 2010 2655
inhibition may function to maintain a basal level of IDO activity,
where formation of kynurenine pathway metabolites is mini-
mized. Subsequent changes in the gene expression of allosteric
effectors, i.e., cytochrome b5, could then lead to activation of
IDO generating elevated levels of kynurenine and downstream
metabolites that participate in immunosuppression (63-66).
8. Uyttenhove, C. (2003) Evidence for a tumoral immune resistance
mechanism based on tryptophan degradation by indoleamine 2,3-
dioxygenase. Nat. Med. 9, 1269–1274.
9. Orabona, C., Puccetti, P., Vacca, C., Bicciato, S., Luchini, A.,
Fallarino, F., Bianchi, R., Velardi, E., Perruccio, K., Velardi, A.,
Bronte, V., Fioretti, M. C., and Grohmann, U. (2006) Toward the
identification of a tolerogenic signature in IDO-competent dendritic
cells. Blood 107, 2846–2854.
10. Odemuyiwa, S. O., Ghahary, A., Li, Y., Puttagunta, L., Lee, J. E.,
Musat-Marcu, S., Ghahary, A., and Moqbel, R. (2004) Cutting edge:
human eosinophils regulate T cell subset selection through indolea-
mine 2,3-dioxygenase. J. Immunol. 173, 5909–5913.
11. Yoshida, R., Imanishi, J., Oku, T., Kishida, T., and Hayaishi, O.
(1981) Induction of pulmonary indoleamine 2,3-dioxygenase by
interferon. Proc. Natl. Acad. Sci. U.S.A. 78, 129–132.
12. Pfefferkorn, E. R. (1984) Interferon gamma blocks the growth of
Toxoplasma gondii in human fibroblasts by inducing the host cells to
degrade tryptophan. Proc. Natl. Acad. Sci. U.S.A. 81, 908–912.
13. Munn, D. H. (1998) Prevention of allogeneic fetal rejection by
tryptophan catabolism. Science 281, 1191–1193.
14. Grohmann, U. (2002) CTLA-4-Ig regulates tryptophan catabolism
in vivo. Nat. Immunol. 3, 1097–1101.
15. Swanson, K. A., Zheng, Y., Heidler, K. M., Mizobuchi, T., and
Wilkes, D. S. (2004) CDllcþ cells modulate pulmonary immune
responses by production of indoleamine 2,3-dioxygenase. Am. J.
Resp. Cell Mol. Biol. 30, 311–318.
16. Gurtner, G. J., Newberry, R. D., Schloemann, S. R., McDonald,
K. G., and Stenson, W. F. (2003) Inhibition of indoleamine 2,3-
dioxygenase augments trinitrobenzene sulfonic acid colitis in mice.
Gastroenterology 125, 1762–1773.
CONCLUSION
The current study investigated whether CPR alone can support
IDO enzymatic activity and whether the addition of additional
electron transfer partners or cofactors used to support IDO
activity such as cytochrome b5 and methylene blue differentially
impact IDO activity. The data presented demonstrate that CPR
is capable of supporting IDO enzymatic activity and display
differential kinetics for the two isomers of Trp, with the observed
L
-Trp partial substrate inhibition arguing for the binding of two
substrate molecules. Addition of cytochrome b5 to CPR-sup-
ported -Trp incubations resulted in a switch from negative to
L
positive homotropic cooperativity. Addition of methylene blue
(minus ascorbate) to CPR-supported incubations also resulted in
modulation of IDO enzymatic activity but differed from that of
cytochrome b5 in that it resulted in a decrease in catalytic
efficiency for both L- and D-Trp oxidations driven by a decrease
in affinity of IDO for both substrates. In conclusion, our data
indicate that CPR is capable of supporting IDO activity in vitro
and oxidation of tryptophan by IDO displays substrate stereo-
chemistry dependent atypical kinetics which can be modulated by
the addition of cytochrome b5.
17. Hayashi, T., Beck, L., Rossetto, C., Gong, X., Takikawa, O.,
Takabayashi, K., Broide, D. H., Carson, D. A., and Raz, E. (2004)
Inhibition of experimental asthma by indoleamine 2,3-dioxygenase.
J. Clin. Invest. 114, 270–279.
18. Kwidzinski, E., Bunse, J., Aktas, O., Richter, D., Mutlu, L., Zipp, F.,
Nitsch, R., and Bechmann, I. (2005) Indolamine 2,3-dioxygenase is
expressed in the CNS and down-regulates autoimmune inflammation.
FASEB J. 19, 1347–1349.
ACKNOWLEDGMENT
19. Brandacher, G. (2006) Prognostic value of indoleamine 2,3-dioxygen-
ase expression in colorectal cancer: effect on tumor-infiltrating T cells.
Clin. Cancer Res. 12, 1144–1151.
We are grateful to Dr. John J. Hill at Amgen Department of
Protein Sciences for helpful discussions and critical review of the
manuscript.
20. Okamoto, A., Nikaido, T., Ochiai, K., Takakura, S., Saito, M., Aoki,
Y., Ishii, N., Yanaihara, N., Yamada, K., Takikawa, O., Kawaguchi,
R., Isonishi, S., Tanaka, T., and Urashima, M. (2005) Indoleamine 2,3-
dioxygenase serves as a marker of poor prognosis in gene expression
profiles of serous ovarian cancer cells. Clin. Cancer Res. 11, 6030–6039.
21. Yamazaki, H., Nakano, M., Imai, Y., Ueng, Y. F., Guengerich, F. P.,
and Shimada, T. (1996) Roles of cytochrome b5 in the oxidation of
testosterone and nifedipine by recombinant cytochrome P450 3A4 and
by human liver microsomes. Arch. Biochem. Biophys. 325, 174–182.
22. Bell, L. C., and Guengerich, F. P. (1997) Oxidation kinetics of ethanol
by human cytochrome P450 2E1. Rate-limiting product release
accounts for effects of isotopic hydrogen substitution and cytochrome
b5 on steady-state kinetics. J. Biol. Chem. 272, 29643–29651.
23. Perret, A., and Pompon, D. (1998) Electron shuttle between mem-
brane-bound cytochrome P450 3A4 and b5 rules uncoupling mechan-
isms. Biochemistry 37, 11412–11424.
SUPPORTING INFORMATION AVAILABLE
Two tables presenting data on the changes in kinetic constants
for L- and D-Trp oxidation by recombinant human IDO in the
presence of CPR (Table 1) and CPR plus cytochrome b5 (Table 2)
with two figures presenting the UV-vis spectrum of our recom-
binant IDO preparation and a figure of the observed steady-state
turnover rates for L- and D-Trp with ascorbate and methylene
blue as the reducing system. This material is available free of
charge via the Internet at http://pubs.acs.org.
24. Reedy, C. J., Elvekrog, M. M., and Gibney, B. R. (2008) Development
of a heme protein structure-electrochemical function database. Nu-
cleic Acids Res. 36, D307–D313.
REFERENCES
1. Yamamoto, S., and Hayaishi, O. (1967) Tryptophan pyrrolase of
rabbit intestine. D- and L-tryptophan-cleaving enzyme or enzymes.
J. Biol. Chem. 242, 5260–5266.
2. Higuchi, K., and Hayaishi, O. (1967) Enzymic formation of D-
kynurenine from D-tryptophan. Arch. Biochem. Biophys. 120, 397–
403.
3. Hirata, F., and Hayaishi, O. (1972) New degradative routes of
5-hydroxytryptophan and serotonin by intestinal tryptophan 2,3-
dioxygenase. Biochem. Biophys. Res. Commun. 47, 1112–1119.
4. Moffett, J. R., and Namboodiri, M. A. (2003) Tryptophan and the
immune response. Immunol. Cell Biol. 81, 247–265.
5. Mellor, A. L., and Munn, D. H. (2004) IDO expression by dendritic
cells: tolerance and tryptophan catabolism. Nat. Rev. 4, 762–774.
6. Beutelspacher, S. C. (2006) Expression of indoleamine 2,3-dioxygen-
ase (IDO) by endothelial cells: implications for the control of allor-
esponses. Am. J. Transplant. 6, 1320–1330.
25. Maghzal, G. J., Thomas, S. R., Hunt, N. H., and Stocker, R. (2008)
Cytochrome b5, not superoxide anion radical, is a major reductant of
indoleamine 2,3-dioxygenase in human cells. J. Biol. Chem. 283,
12014–12025.
26. Vottero, E., Mitchell, D. A., Page, M. J., MacGillivray, R. T.,
Sadowski, I. J., Roberge, M., and Mauk, A. G. (2006) Cytochrome
b(5) is a major reductant in vivo of human indoleamine 2,3-dioxy-
genase expressed in yeast. FEBS Lett. 580, 2265–2268.
27. Paine, M. J. I. (2005) Electron Transfer Partners of Cytochrome P450,
in Cytochrome P450: Structure, Mechanism, and Biochemistry (Ortiz
de Montellano, P. R., Ed.) 3rd ed., pp 115-148, Kluwer Academic
Publishers, New York.
28. Schenkman, J. B., Voznesensky, A. I., and Jansson, I. (1994) Influence
of ionic strength on the P450 monooxygenase reaction and role of
cytochrome b5 in the process. Arch. Biochem. Biophys. 314, 234–241.
29. Yamazaki, H., Johnson, W. W., Ueng, Y. F., Shimada, T., and
Guengerich, F. P. (1996) Lack of electron transfer from cytochrome b5
in stimulation of catalytic activities of cytochrome P450 3A4. Characteri-
zation of a reconstituted cytochrome P450 3A4/NADPH-cytochrome
7. Munn, D. H. (2004) Expression of indoleamine 2,3-dioxygenase by
plasmacytoid dendritic cells in tumor-draining lymph nodes. J. Clin.
Invest. 114, 280–290.

2656 Biochemistry, Vol. 49, No. 12, 2010
Pearson et al.
P450 reductase system and studies with apo-cytochrome b5. J. Biol.
Chem. 271, 27438–27444.
function of human indoleamine 2,3-dioxygenase. J. Biol. Chem. 278,
29525–29531.
30. Yamazaki, H., Shimada, T., Martin, M. V., and Guengerich, F. P.
(2001) Stimulation of cytochrome P450 reactions by apo-cytochrome
b5: evidence against transfer of heme from cytochrome P450 3A4 to
apo-cytochrome b5 or heme oxygenase. J. Biol. Chem. 276, 30885–
30891.
47. Papadopoulou, N. D., Mewies, M., McLean, K. J., Seward, H. E.,
Svistunenko, D. A., Munro, A. W., and Raven, E. L. (2005) Redox
and spectroscopic properties of human indoleamine 2,3-dioxygenase
and a His303Ala variant: implications for catalysis. Biochemistry 44,
14318–14328.
31. Gruenke, L. D., Konopka, K., Cadieu, M., and Waskell, L. (1995)
The stoichiometry of the cytochrome P-450-catalyzed metabolism of
methoxyflurane and benzphetamine in the presence and absence of
cytochrome b5. J. Biol. Chem. 270, 24707–24718.
32. Korzekwa, K. R., Krishnamachary, N., Shou, M., Ogai, A., Parise,
R. A., Rettie, A. E., Gonzalez, F. J., and Tracy, T. S. (1998)
Evaluation of atypical cytochrome P450 kinetics with two-substrate
models: evidence that multiple substrates can simultaneously bind to
cytochrome P450 active sites. Biochemistry 37, 4137–4147.
33. Shou, M., Dai, R., Cui, D., Korzekwa, K. R., Baillie, T. A., and
Rushmore, T. H. (2001) A kinetic model for the metabolic interaction
of two substrates at the active site of cytochrome P450 3A4. J. Biol.
Chem. 276, 2256–2262.
48. Nickel, E., Nienhaus, K., Lu, C., Yeh, S. R., Nienhaus, G. U. (2009)
Ligand and substrate migration in human indoleamine 2,3-dioxygen-
ase, J. Biol. Chem. (in press).
49. Atkins, W. M. (2005) Non-Michaelis-Menten kinetics in cytochrome
P450-catalyzed reactions. Annu. Rev. Pharmacol. Toxicol. 45, 291–
310.
50. Davydov, D. R., Davydova, N. Y., Tsalkova, T. N., and Halpert, J. R.
(2008) Effect of glutathione on homo- and heterotropic cooperativity
in cytochrome P450 3A4. Arch. Biochem. Biophys. 471, 134–145.
51. Davydov, D. R., and Halpert, J. R. (2008) Allosteric P450 mechan-
isms: multiple binding sites, multiple conformers or both? Expert
Opin. Drug Metab. Toxicol. 4, 1523–1535.
52. Halgren, T. (2007) New method for fast and accurate binding-site
identification and analysis. Chem. Biol. Drug Des. 69, 146–148.
53. Halgren, T. A. (2009) Identifying and characterizing binding sites and
assessing druggability. J. Chem. Inf. Modeling 49, 377–389.
54. McAmis, W. C., Schaeffer, R. C., Jr., Baynes, J. W., and Wolf, M. B.
(2003) Menadione causes endothelial barrier failure by a direct effect
on intracellular thiols, independent of reactive oxidant production.
Biochim. Biophys. Acta 1641, 43–53.
34. Wang, R. W., Newton, D. J., Liu, N., Atkins, W. M., and Lu, A. Y.
(2000) Human cytochrome P-450 3A4: in vitro drug-drug interaction
patterns are substrate-dependent. Drug Metab. Dispos. 28, 360–366.
35. Dabrowski, M. J., Schrag, M. L., Wienkers, L. C., and Atkins, W. M.
(2002) Pyrene pyrene complexes at the active site of cytochrome P450
3A4: evidence for a multiple substrate binding site. J. Am. Chem. Soc.
124, 11866–11867.
3
36. Atkins, W. M., Wang, R. W., and Lu, A. Y. (2001) Allosteric behavior
in cytochrome p450-dependent in vitro drug-drug interactions: a
prospective based on conformational dynamics. Chem. Res. Toxicol.
14, 338–347.
37. Sono, M. (1989) Enzyme kinetic and spectroscopic studies of inhibitor
and effector interactions with indoleamine 2,3-dioxygenase. 2. Evi-
dence for the existence of another binding site in the enzyme for indole
derivative effectors. Biochemistry 28, 5400–5407.
38. Sono, M., Taniguchi, T., Watanabe, Y., and Hayaishi, O. (1980)
Indoleamine 2,3-dioxygenase. Equilibrium studies of the tryptophan
binding to the ferric, ferrous, and CO-bound enzymes. J. Biol. Chem.
255, 1339–1345.
39. Lu, C., Lin, Y., and Yeh, S. R. (2009) Inhibitory substrate binding site
of human indoleamine 2,3-dioxygenase. J. Am. Chem. Soc. 131,
12866–12867.
40. Shaw, P. M., Hosea, N. A., Thompson, D. V., Lenius, J. M., and
Guengerich, F. P. (1997) Reconstitution premixes for assays using
purified recombinant human cytochrome P450, NADPH-cyto-
chrome P450 reductase, and cytochrome b5. Arch. Biochem. Biophys.
348, 107–115.
41. Kumar, S. (2008) Structure based development of phenylimidazole-
derived inhibitors of indoleamine 2,3-dioxygenase. J. Med. Chem. 51,
4968–4977.
42. Shou, M. (2002) Kinetic analysis for multiple substrate interaction at
the active site of cytochrome P450. Methods Enzymol. 357, 261–276.
43. Sugimoto, H. (2006) Crystal structure of human indoleamine 2,3-
dioxygenase: catalytic mechanism of O2 incorporation by a heme-
containing dioxygenase. Proc. Natl. Acad. Sci. U.S.A. 103, 2611–
2616.
55. Ross, D., Thor, H., Orrenius, S., and Moldeus, P. (1985) Interaction
of menadione (2-methyl-1,4-naphthoquinone) with glutathione.
Chem.-Biol. Interact. 55, 177–184.
56. Alegre, E., Lopez, A. S., and Gonzalez, A. (2005) Tryptophan
metabolites interfere with the Ehrlich reaction used for the measure-
ment of kynurenine. Anal. Biochem. 339, 188–189.
57. Yamamoto, K., Kimura, S., Shiro, Y., and Iyanagi, T. (2005) Inter-
flavin one-electron transfer in the inducible nitric oxide synthase
reductase domain and NADPH-cytochrome P450 reductase. Arch.
Biochem. Biophys. 440, 65–78.
58. Samelson-Jones, B. J., and Yeh, S. R. (2006) Interactions between
nitric oxide and indoleamine 2,3-dioxygenase. Biochemistry 45, 8527–
8538.
59. Thomas, S. R., Mohr, D., and Stocker, R. (1994) Nitric oxide inhibits
indoleamine 2,3-dioxygenase activity in interferon-gamma primed
mononuclear phagocytes. J. Biol. Chem. 269, 14457–14464.
60. Amirkhani, A., Heldin, E., Markides, K. E., and Bergquist, J. (2002)
Quantitation of tryptophan, kynurenine and kynurenic acid in human
plasma by capillary liquid chromatography-electrospray ionization
tandem mass spectrometry. J. Chromatogr. 780, 381–387.
61. Huengsberg, M., Winer, J. B., Gompels, M., Round, R., Ross, J., and
Shahmanesh, M. (1998) Serum kynurenine-to-tryptophan ratio in-
creases with progressive disease in HIV-infected patients. Clin. Chem.
44, 858–862.
62. Laich, A., Neurauter, G., Widner, B., and Fuchs, D. (2002) More
rapid method for simultaneous measurement of tryptophan and
kynurenine by HPLC. Clin. Chem. 48, 579–581.
63. Fallarino, F. (2003) T cell apoptosis by kynurenines. Adv. Exp. Med.
Biol. 527, 183–190.
44. Houston, J. B. (2003) Typical and Atypical Enzyme Kinetics, in Drug
Metabolizing Enzymes: Cytochrome P450 and Other Enzymes in
Drug Discovery and Development (Jae, S., Lee, R. S. O., and Fischer,
M. B., Eds.) pp 211-251, CRC PressBoca Raton, FL.
45. Pereira, A., Vottero, E., Roberge, M., Mauk, A. G., and Andersen,
R. J. (2006) Indoleamine 2,3-dioxygenase inhibitors from the North-
eastern Pacific marine hydroid Garveia annulata. J. Nat. Prod. 69,
1496–1499.
64. Frumento, G. (2002) Tryptophan-derived catabolites are responsible
for inhibition of T and natural killer cell proliferation induced by
indoleamine 2,3-dioxygenase. J. Exp. Med. 196, 459–468.
65. Terness, P. (2002) Inhibition of allogeneic T cell proliferation by
indoleamine 2,3-dioxygenase-expressing dendritic cells: mediation of
suppression by tryptophan metabolites. J. Exp. Med. 196, 447–457.
66. Chen, W., Liang, X., Peterson, A. J., Munn, D. H., and Blazar, B. R.
(2008) The indoleamine 2,3-dioxygenase pathway is essential for
human plasmacytoid dendritic cell-induced adaptive T regulatory cell
generation. J. Immunol. 181, 5396–5404.
46. Littlejohn, T. K., Takikawa, O., Truscott, R. J., and Walker, M. J.
(2003) Asp274 and his346 are essential for heme binding and catalytic