High-affinity uptake of kynurenine and nitric oxide-mediated inhibition of
indoleamine 2,3-dioxygenase in bone marrow-derived myeloid dendritic cells
Toshiaki Hara a, Nanako Ogasawaraa, Hidetoshi Akimoto b, Osamu Takikawa b, Rie
Hiramatsu a, Tsutomu Kawabe a, Ken-ichi Isobe c, Fumihiko Nagase a*
a Department of Medical Technology, Nagoya University School of Health Sciences,
Nagoya, Aichi, Japan,b Institute of Longevity Science, National Center for Geriatrics
and Gerontology, Obu, Aichi, Japan, c Department of Immunology, Nagoya University
Graduate School of Medicine, Nagoya, Aichi, Japan
*Corresponding author: Fumihiko Nagase, Department of Medical Technology,
Nagoya University School of Health Sciences, 1-20 Daikominami-1-chome,
Higashi-ku, Nagoya, Aichi, 461-8673, Japan. Telephone and fax numbers:
+81-52-719-1189. E-mail address: email@example.com
Type of Article: Research Article
Indoleamine 2,3-dioxygenase (IDO)-initiated tryptophan metabolism along the
kynurenine (Kyn) pathway in some dendritic cells (DC) such as plasmacytoid DC
regulates T-cell responses.It is unclear whether bone marrow-derived myeloid DC
(BMDC) express functional IDO. The IDO expression was examined in
CD11c+CD11b+BMDC differentiated from mouse bone marrow cells using GM-CSF.
CpG oligodeoxynucleotides (CpG) induced the expression of IDO protein with the
production of nitric oxide (NO) in BMDC in cultures for 24 hr. In the enzyme assay
using cellular extracts of BMDC, the IDO activity of BMDC stimulated with CpG was
enhanced by the addition of a NO synthase (NOS) inhibitor, suggesting that IDO
activity was suppressed by NO production. On the other hand, the concentration of
Kyn in the culture supernatant of BMDC was not increased by stimulation with CpG.
Exogenously added Kyn was taken up by BMDC independently of CpG stimulation and
NO production, and the uptake of Kyn was inhibited by a transport system L-specific
inhibitor or high concentrations of tryptophan. The uptake of tryptophan by BMDC
was markedly lower than that of Kyn. In conclusion, IDO activity in BMDC is
down-regulated by NO production, whereas BMDC strongly take up exogenous Kyn.
Key words: Bone marrow-derived myeloid dendritic cells; Indoleamine
2,3-dioxygenase; Tryptophan; Kynurenine; Nitric oxide; Transport system L
Indoleamine 2,3-dioxygenase (IDO)-initiated tryptophan (Trp) metabolism along
the kynurenine (Kyn) pathway regulates T-cell responses in some dendritic cells (DC)
such as plasmacytoid DC or CD8+ DC in mouse spleen cells [1,2].Two mechanisms
of IDO to inhibit T-cell responses are proposed; the local depletion of Trp required for
cell proliferation and the induction of apoptosis or growth arrest by Trp metabolites .
Three functionally distinct subsets of DC are defined in mouse spleen cells and include
the plasmacytoid DC (pDC), CD8+ and CD8- conventional DC (cDC) . DC
generated in culture from mouse bone marrow precursors with GM-CSF  or
GM-CSF and IL-4 are mainly CD11c+CD11b+ myeloid DC, whereas cell culture from
mouse bone marrow cells with Fms-like tyrosine kinase 3 ligand (Flt3-L) allows the
generation of both cDC and pDC [5-7]. Human myeloid DC differentiated from blood
monocytes with GM-CSF and IL-4 or macrophages differentiated with M-CSF express
functionally active IDO [8,9]. It has recently been shown that thymosin 1 activates
IDO in GM-CSF/IL-4- or Flt3-L-developed DC from mouse bone marrow cells .
IDO is induced by inflammation or immune responses such as infectious or tumor
immunity. IDO expression is induced in DC by various stimuli such as IFN-, toll-like
receptor (TLR)-ligation by LPS or CpG oligodeoxynucleotides (CpG) or
CD80/CD86-ligation by CTLA-4 expressed on the regulatory T cells . CpG, which
is a strong immune stimulator, also possesses immune suppressive activity through the
induction of IDO [11-14]. TLR-9 signaling from CpG activates NF-B and p38
through MyD88 and IRF-8 in DC . CpG is the TLR ligand to induce IDO mRNA
in bone marrow-derived myeloid DC (BMDC) .
IDO protein can be expressed without functional enzymatic activity. Isolated
mouse splenic CD8+ DC were found to catabolize Trp when exposed to IFN-, whereas
other CD8- DC did not, even though both subsets expressed comparable amounts of
IDO protein as analyzed by Western blot . It is not clear yet why IDO protein
expressed in CD8- DC is not functional.
IDO activity of IFN--activated murine peritoneal macrophages was induced by
inhibition of nitric oxide synthase (NOS) . Incorporation of the heme prosthetic
group into active site is required for IDO activity, and inhibition of IDO activity by NO
generators was abrogated by co-addition of oxyhemoglobin, an antagonist of NO
function . Both the blocking of a heme site to O2 binding and conformational
changes induced by breaking the Fe-N bond have been proposed as important
mechanisms by which NO inhibits IDO . NO led to an accelerated degradation of
IDO protein in the proteasome . In addition, a peroxynitrite generator also
inhibited IDO activity through the nitration of the selective tyrosines of IDO . NO
production was induced in BMDC by stimulation with IFN- and LPS .
It has recently been published that BMDC expressing IDO upon IFN- stimulation
suppress OVA-specific CD8+ T cell proliferation [22,23]. However, these results are
not consistent with the published report that IFN- enhances antigen-presenting activity
in BMDC . In the present study, we examined the functional expression of IDO
in BMDC stimulated with CpG. BMDC expressed IDOprotein upon CpG stimulation
but its activity was inhibited by NO production. BMDC did not secrete Kyn upon
CpG stimulation but took up exogenous Kyn.
2. Materials and methods
The Phosphorothioate CpG1826 (5-TCC ATG ACG TTC CTG ACG TT-3),
NG-monomethyl-L-arginine acetate salt (NMA), L-Kyn, L-Trp,
2-amino-2-norbornanecarboxylic acid (BCH)and 1-methyl-DL-tryptophan (1-MT)
were purchased from Sigma-Aldrich (St. Louis, MO). PE-conjugated anti-mouse
CD11c and FITC-conjugated anti-mouse CD11b antibodies were purchased from
eBioscience (San Diego, CA). Anti-mouse IDO polyclonal antibody was prepared as
described previously . Anti-inducible NOS (iNOS) polyclonal antibody was
purchased from BD Bioscience (San Diego, CA).
2.2. Preparation of BMDC
C57BL/6 mice were purchased from Japan SLC (Shizuoka, Japan). BMDC were
generated as described previously . Briefly, bone marrow cells were cultured in
RPMI1640 medium (10% fetal calf serum, 300 g/ml glutamine, 100 U/ml penicillin,
100 g/ml streptomycin and 50 M 2-mercaptoethanol) containing 0.3% GM-CSF
supernatant (from murine GM-CSF producing Chinese hamster ovary cells, a gift from
T. Sudo, Toray Silicon, Tokyo, Japan). The DC culture medium was changed every 2
days to remove nonadherent cells. Loosely adherent clustering cells were collected on
day 6 and used as immature DC. BMDC were activated by stimulation with CpG (5
g/ml) alone or together with NMA (100 M) for 24 hr.
2.3. Flow cytometry
For detection of cell surface markers, cells were incubated with PE-conjugated
anti-CD11c and FITC-conjugated anti-CD11b antibodies at 4℃ for 30 min. These
cells were analyzed by an EPICS XL flow cytometer (Beckman Coulter).
2.4. Western blot
Western blot was carried out as described previously . BMDC were
stimulated with CpG (5 g/ml) with or without NMA (100 M) for 24 hr. Then, cells
were washed and lysed with 1×sample buffer and boiled for 3 min. The cell lysates
were passed though a syringe with a 26G needle before being applied on 10% sodium
dodecyl sulfate (SDS)-polyacrylamide gels. After electrophoresis, protein was
transferred to a nitrocellulose membrane and the membrane was blocked with PBS plus
0.05% Tween 20 (PBST) containing 0.3% skimmed milk for 1 hr at room temperature.
Then the membrane was incubated with anti-IDO, anti-iNOS or anti-actin antibody at
4℃ overnight. The membrane was then washed with PBST and incubated with
horseradish peroxidase-conjugated anti-rabbit IgG for 1 hr at room temperature.
Finally, the membrane was washed with PBST and developed with a Western lightning
chemiluminescence reagent (PerkinElmer Life Sciences, Boston, MA).
2.5. Assay of NO production
The amount of NO production in the medium was estimated by the assay of nitrite
using Griess reagent . Fifty microliter of each supernatant was mixed with an
equal volume of Griess reagent (1% sulfanilamide in 5% phosphoric acid and 0.1%
naphthylethylenediamine dihydrochloride in distilled water). The absorbance of the
mixture at 590 nm was determined by a plate reader, and the nitrite concentration was
determined using standard solutions of sodium nitrite.
2.6. Enzyme assay of IDO activity
IDO activity was determined by the assay previously described . After cells
were sonicated, the homogenate was centrifuged 10,000 rpm for 10 min. The
supernatant (100 l) was mixed with an equal volume of 2×reaction buffer (100 mM
potassium phosphate buffer pH 6.5, 40 mM sodium ascorbate, 20 M methylene blue,
200 g/ml catalase, 800 M Trp). The mixtures were incubated at 37℃ for 60 min to
permit IDO to convert Trp to N-formylkynurenine, and then 40 l of 30% (wt/vol) TCA
was added to stop the reaction. After heating at 50℃ for 30 min, the reaction
mixtures were centrifuged and Kyn concentration in the supernatant was measured by
high-pressure liquid chromatography (HPLC).
2.7. Assay of Kyn and Trp
Concentrations of Kyn and Trp were determined by HPLC as previously described
, with minor modification. Before HPLC assay, culture medium was deproteinized
by treatment with 86% methanol (1:6, vol/vol). Twenty-five microliter of sample was
injected into a 5 m endcapped Purospher RP-18 column (Merck, Darmstadt, Germany)
and analyses were carried out at a flow rate 1.0 ml/min. The mobile phase was 10 mM
acetic ammonium (pH 6.5) and 10% methanol. Kyn was detected by a UV-detector at
a wavelength of 360 nm and Trp by a fluorescence detector at an excitation wavelength
of 285 nm and an emission wavelength of 365 nm.
2.8. Assay of Kyn uptake
For the assay of Kyn uptake for a short time, BMDC were washed with
Tris-choline buffer (150 mM choline chloride, 10 mM Tris, pH 7.4) and suspended at 3
×105 cells/0.2 ml/well in Tris-choline buffer or Tris-Na buffer (150 mM sodium chloride,
10 mM Tris, pH 7.4). Kyn was added into cell suspensions with or without BCH, Trp,
or 1-MT in water bath at 37℃. Kyn uptake was stopped by cooling cell suspension on
ice. Kyn concentrations of the culture supernatant were assayed by HPLC, and the
decrease of Kyn content in the supernatant was estimated as Kyn uptake by BMDC.
3.1. Expression of IDO protein and NO production in CD11c+CD11b+ BMDC by
stimulation with CpG
BMDC were differentiated from bone marrow cells with GM-CSF for 6 days.
Most (73.2%) of the BMDC were CD11c+CD11b+ myeloid DC (Fig. 1A). The
expression of IDO and iNOS proteins was induced in BMDC but not original bone
marrow cells stimulated with CpG for 24 hr (Fig. 1B). CpG-mediated NO production
was also increased with the BMDC development (Fig. 1C). These results indicate that
the ability of BMDC to express IDO and iNOS proteins upon CpG stimulation is
induced in the fully developed BMDC.
3.2. Inhibition of IDO activity by NO production in BMDC
Effects of NO production on IDO activity in BMDC were tested. IDO protein was
expressed in BMDC by stimulation with CpG in the presence or absence of NMA, a
NOS inhibitor (Fig. 2A). CpG-mediated NO production was inhibited by NMA (Fig.
2B). In the enzyme assay using cellular extracts of BMDC, IDO activity of BMDC
stimulated with CpG was weak but enhanced by the addition of NMA (Fig. 2C).
These results show that expression of IDO protein is induced in BMDC stimulated with
CpG, although its activity is inhibited by NO production.
3.3. Non-secretion of Kyn by BMDC stimulated with CpG
We examined whether BMDC secreted Kyn upon stimulation with CpG. The Kyn
level in the culture supernatant of BMDC was not increased significantly by stimulation
with CpG even in the presence of exogenously added Trp (Fig. 3A). Correspondingly,
the concentration of Trp in the culture supernatant of BMDC stimulated with CpG was
hardly decreased unless NMA was added (Fig. 3B). These results indicate that BMDC
do not secrete Kyn upon CpG stimulation.
3.4. Uptake of exogenously added Kyn by BMDC without CpG stimulation
It has been recently shown that CD8- DC take up exogenously added Kyn upon
IFN- stimulation . Therefore, we examined whether exogenously added Kyn was
taken up by BMDC by CpG stimulation for 24 hr. The Kyn concentration in the
culture supernatant of BMDC was decreased by around 20% independently of CpG
stimulation and NMA when 50 M Kyn was exogenously added (Fig. 4A). The Kyn
concentration in the culture supernatant of BMDC was decreased dependently on the
concentration of exogenously added Kyn (10-100 M) (Fig. 4B). Concentrations of
Kyn in the culture supernatants of BMDC were decreased almost linearly with time for
24 hr (Fig. 4C). The ability of BMDC to take up Kyn was induced with the
development from bone marrow cells (Fig. 4D). These results suggest that BMDC take
up Kyn independently of IDO expression and NO production.
3.5. Inhibition of Kyn uptake by a transport system L-specific inhibitor and Trp
It has been shown that astrocytes take up Kyn through a Na+-independent transport
system L . Na+-dependency of Kyn uptake by BMDC was tested in cultures of
BMDC by using Na+ and Na+-free buffers for a short time. Kyn uptake by BMDC in
Na+-free buffer for 15 min was higher than that in Na+ buffer (Fig. 5A), indicating that
Kyn uptake by BMDC is mainly Na+-independent. Therefore, Kyn uptake by BMDC
was tested in Na+-free buffer. The uptake of exogenously added Kyn (10 M) by
BMDC (3×105 cells) increased rapidly to around 100 pmole (5% of Kyn exogenously
added) within 2 min and gradually thereafter (Fig. 5B). Effects of the addition of Trp,
BCH, a specific inhibitor of the transport system L, or 1-MT, an inhibitor of IDO and
the transport system L [31-33], on Kyn uptake by BMDC were tested (Fig. 5C). All
the BCH, Trp and 1-MT prevented the uptake of exogenously added Kyn by BMDC.
These results show that Kyn is taken up by BMDC mainly through the transport system
The present study shows that the expression of IDO activity in BMDC is regulated
at the post-transcriptional level by NO production. IDO activity of BMDC was
detected only by the enzyme assay using the cellular extracts of BMDC stimulated with
CpG in the presence of a NOS inhibitor. IFN--activated mouse peritoneal
macrophages secrete Kyn in the presence of a NOS inhibitor . However,
CpG-activated BMDC did not secrete Kyn even in the presence of a NOS inhibitor.
This may be caused by the weak activity of IDO in BMDC. Human myeloid DC
differentiated from blood monocytes with GM-CSF and IL-4 as well as macrophages
differentiated with M-CSF express IDO activity [8,9]. These differences are caused by
a clear species specificity regarding the induction of IDO versus iNOS in cultured cells
. In human monocytes/macrophages, IFN-or IFN-/LPS strongly induces IDO,
but not iNOS activity, while in mouse macrophages these stimuli strongly induce iNOS,
but not IDO activity. Thus, the inhibition of iNOS expression is at least required for
the induction of IDO activity.
We showed the expression of IDO protein, non-secretion of Kyn and NO
production in BMDC stimulated with CpG. IFN--activated CD8- DC in mouse spleen
express IDO protein but not IDO activity in contrast to CD8+ DC . CD8- DC
treated with IFN-produce significantly higher levels of NO than the CD8+ DC
counterpart . Thus, activated myeloid dendritic cells such as BMDC and CD8- DC
express IDO protein and produce NO but do not secrete Kyn. Unexpectedly, Park and
coworkers have recently published that IDO-expressing BMDC upon IFN- stimulation
suppress OVA-specific CD8+ T-cell proliferation [22,23]. However, IFN- induces
high levels of NO production , and enhances antigen-presenting activity in BMDC
. At present, it is difficult to reconcile these observations [22,23] with our finding
that the activity of IDO is suppressed by NO production in BMDC stimulated with CpG
and also with the enhancement of antigen presenting activity of BMDC stimulated with
Bone marrow cells neither expressed IDO and iNOS proteins nor produced NO
upon CpG stimulation without differentiation to BMDC with GM-CSF (Fig. 1B, C).
GM-CSF induces NO production in a skin dendritic cell line and enhances IDO
expression in eosinophils stimulated with IFN- [35, 36]. Thus, GM-CSF seems to be
an important factor for the induction of IDO and iNOS. However, GM-CSF
completely inhibits Flt3-L-induced pDC development from bone marrow cells [5, 6].
As far as we know, there is no report showing that high levels of NO production are
induced in pDC. Therefore, GM-CSF seems to be a much more critical factor in vitro
for the induction of iNOS than that of IDO with BMDC differentiation.
The development of CD8- DC from bone marrow cells in the presence of GM-CSF
depends on IRF-4, whereas the development of CD8+ DC and pDC in the presence of
Flt3-L mainly depends on IRF-8 [37,38]. Correspondingly, the negative regulation of
gene expression of IRF-8 inhibits the induction of IDO activity in CD8+ DC stimulated
with IFN-or human DC stimulated with LPS . Induction of IDO by LPS but not
IFN- in human monocytic THP-1 cells involves p38 and NF-κB pathways .
TLR-9 signaling from CpG activates p38 and NF-B through MyD88 and IRF-8 in DC
. BMDC developed from bone marrow cells with GM-CSF express significantly
IRF-8 [37,38]. Therefore, it may be possible that BMDC express IDO protein through
MyD88 and IRF-8 in response to CpG. However, it has recently been shown that LPS
but not IFN- induces the IDO expression in BMDC through the activation of PI3
kinase and JNK . Therefore, it is interesting to clarify which signal pathways are
required for the induction of IDO in BMDC by stimulation with CpG.
We showed that BMDC took up exogenous Kyn independently of CpG stimulation
and NO production. Na+-independent uptake of Kyn by BMDC was bloked by BCH, a
transport system L-specific inhibiter, Trp or 1-MT blocked. The transport of Kyn in
BMDC is similar to that in astrocytes, which is inhibited by BCH and Trp in Na+-free
solution . BCH, Trp, 1-MT and various other amino acids also inhibit Trp uptake
through the transport system L [31-33]. A BCH-sensitive and Na+-independent
transport is consistent with system L, a neutral amino acid transport mechanism, being
the major conduit of Trp [31-33]. Therefore, we conclude that BMDC take up Kyn
mainly through the transport system L.
We showed that BMDC took up Kyn much more preferentially than Trp, indicating
a higher affinity of Kyn than Trp to the transporter. A low affinity of Trp to the
transporter corresponds to the expression of a low IDO activity in BMDC. In fact, the
enzyme assay of IDO activity using the cellular extract of BMDC, which does not
require membrane transport of Trp, demonstrated IDO activity. A high-affinity,
Trp-selective amino acid transport system has been recently shown in human
macrophages, and speculated that this unique transport system allow macrophages to
take up Trp efficiently under low substrate concentration, such as may occur during
interaction between T cells and IDO-expressing antigen presenting cells . Taken
together with our findings, a low affinity of Trp to the transporter in BMDC causes the
expression of a low IDO activity, in addition to the suppression by NO production.
It has recently been shown that CD8- DC as well as CD8+ DC take up exogenously
added Kyn and secrete quinolinic acid upon IFN- stimulation . Therefore, DC
such as BMDC, CD8- DC and CD8+ DC take up exogenous Kyn. However, the uptake
of Kyn by BMDC is independent of CpG stimulation. Therefore, it is suggested that
CpG does not activate down stream enzymes of IDO along the Kyn pathway. On the
other hand, immunogenic CD8- DC became immunosuppressive DC through the
generation of Kyn metabolites such as quinolinic acid upon IFN- stimulation in the
presence of exogenous Kyn . However, we did not observe the
immunosuppressive activity of BMDC stimulated with IFN- in the presence of
exogenously added Kyn (unpublished data). Our results suggest a new possibility that
BMDC counteract Kyn-mediated induction of regulatory DC or T cells by scavenging
The utilization of Kyn by BMDC in the resting state might be physiologically
important for cell survival because IDO is not constitutively activated. Moffett et al.
have shown that intraperitoneal injections of Kyn did not result in any significant
increase in hepatocyte immunoreactivity with quinolinate-specific antibody, but rather
led to dramatic increase in immunoreactivity in tissue macrophages, splenic white pulp,
and thymic medulla [43,44]. Quinolinic acid formation was also induced most
strongly in spleen by systemic immune stimulation with pokeweed mitogen .
Therefore, it is suggested that extrahepatic Kyn is preferentially metabolized in immune
cells involving BMDC. It may be possible that Kyn is utilized for NAD synthesis for
the survival of BMDC as shown in RAW264.7 macrophages . Trp metabolism
along the Kyn pathway is also required for DC activation . Further study is
required in order to clarify the fate of Kyn taken up by BMDC.
GM-CSF induces in vivo as well as in vitro the development of myeloid DC,
whereas Flt3-L induces the development of both myeloid DC and pDC [5-7,48,49].
Thus, the activity of GM-CSF to induce in vitro the development of immunogenic
myeloid DC from bone marrow cells correlates with the physiological activity of
GM-CSF in vivo. In interaction between DC subsets, otherwise immunogenic CD8-
DC become tolerogenic in co-culture with CD8+ DC upon IFN- stimulation . The
present study suggests alternative possibility that myeloid DC differentiated with
GM-CSF up-regulates immune responses by counteracting tolerogenic activity of
IDO-expressing DC through the two independent mechanisms; the inhibition of IDO
activity by NO production and scavenging Kyn secreted from tolerogenic DC.
This study was supported by the Grant from Ministry of Health, Labour and
Welfare of Japan.
 Mellor AL, Munn DH. IDO expression by dendritic cells: tolerance and
tryptophan catabolism. Nat Rev Immunol 2004;4:762-74.
 Grohmann U, Fallarino F, Puccetti P. Tolerance, DCs and tryptophan: much ado
about IDO. Trends Immunol 2003;24:242-8.
 Shortman K, Liu Y-J. Mouse and human dendritic cell subtypes. Nat Rev
 Inaba K, Inaba M, Romani N, Aya H, Deguchi M, Ikehara S, et al. Generation of
large numbers of dendritic cells from mouse bone marrow cultures
supplemented with granulocyte/macrophage colony-stimulating factor. J Exp
 Gilliet M, Boonstra A, Paturel C, Antonenko S, Xu X-L, Trinchieri G, et al. The
development of murine plasmacytoid dendritic cell precursors is differentially
regulated by FLT3-ligand and granulocyte/macrophage colony-stimulating factor.
J Exp Med 2002;195:953-8.
 Brawand P, Fitzpatrick DR, Greenfield BW, Brasel K, Maliszewski CR, De
15 Download full-text
Smedt TD. Murine plasmacytoid pre-dendritic cells generated from Flt3
ligand-supplemented bone marrow cultures are immature APCs. J Immunol
 Naik SH, Proietto AI, Wilson NS, Dakic A, Schnorrer P, Fuchsberger M, et al.
Cutting edge: generation of splenic CD8+ and CD8- dendritic cell equivalents in
Fms-like tyrosine kinase 3 ligand bone marrow cultures. J Immunol
 Hwu P, Du MX, Lapointe R, Do M, Taylor MW, Young HA. Indoleamine
2,3-dioxygenase production by human dendritic cells results in the inhibition of
T cell proliferation. J Immunol 2000;164:3596-9.
 Munn DH, Shafizadeh E, Attwood JT, Bondarev I, Pashine A, Mellor AL.
Inhibition of T cell proliferation by macrophage tryptophan catabolism. J Exp
 Romani L, Bistoni F, Perruccio K, Montagnoli C, Gaziano R, Bozza S, et al.
Thymosin 1 activates dendritic cell tryptophan catabolism and establishes a
regulatory environment for balance of inflammation and tolerance. Blood
 Fallarino F, Puccetti P. Toll-like receptor 9-mediated induction of the
immunosuppressive pathway of tryptophan catabolism. Eur J Immunol
 Hayashi T, Beck L, Rossetto C, Gong X, Takikawa O, Takabayashi K, et al.
Inhibition of experimental asthma by indoleamine 2,3-dioxygenase. J Clin Invest
 Mellor AL, Baban B, Chandler PR, Manlapat A, Kahler DJ, Munn DH. Cutting
edge: CpG oligonucleotides induce splenic CD19+ dendritic cells to acquire
potent indoleamine 2,3-dioxygenase-dependent T cell regulatory functions via
IFN Type 1 signaling. J Immunol 2005;175:5601-5.