ARTHRITIS & RHEUMATISM
Vol. 58, No. 3, March 2008, pp 754–763
© 2008, American College of Rheumatology
Crucial Role of the Interleukin-6/Interleukin-17 Cytokine Axis
in the Induction of Arthritis by Glucose-6-Phosphate Isomerase
Keiichi Iwanami,1Isao Matsumoto,2Yoko Tanaka-Watanabe,1Asuka Inoue,1
Masahiko Mihara,3Yoshiyuki Ohsugi,4Mizuko Mamura,1Daisuke Goto,1Satoshi Ito,1
Akito Tsutsumi,1Tadamitsu Kishimoto,5and Takayuki Sumida1
Objective. To clarify the glucose-6-phosphate
isomerase (GPI)–specific CD4? T cell lineage involved
in GPI-induced arthritis and to investigate their patho-
logic and regulatory roles in the induction of the
Methods. DBA/1 mice were immunized with GPI
to induce arthritis. CD4? T cells and antigen-presenting
cells were cocultured with GPI, and cytokines in the
supernatant were analyzed by enzyme-linked immunosor-
bent assay. Anti–interferon-? (anti-IFN?) monoclonal
antibody (mAb), anti–interleukin-17 (anti–IL-17) mAb,
or the murine IL-6 receptor (IL-6R) mAb MR16-1 was
injected at different time points, and arthritis develop-
ment was monitored visually. After MR16-1 was in-
jected, percentages of Th1, Th2, Th17, and Treg cells
were analyzed by flow cytometry, and CD4? T cell
proliferation was analyzed using carboxyfluorescein di-
acetate succinimidyl ester.
Results. GPI-specific CD4? T cells were found to
be differentiated to Th1 and Th17 cells, but not Th2
cells. Administration of anti–IL-17 mAb on day 7 sig-
nificantly ameliorated arthritis (P < 0.01), whereas
administration of anti-IFN? mAb exacerbated arthritis.
Neither anti–IL-17 mAb nor anti-IFN? mAb adminis-
tration on day 14 ameliorated arthritis. Administration
of MR16-1 on day 0 or day 3 protected against arthritis
induction, and MR16-1 administration on day 8 signif-
icantly ameliorated existing arthritis (P < 0.05). After
administration of MR16-1, there was marked suppres-
sion of Th17 differentiation, without an increase in Th1,
Th2, or Treg cells, and CD4? T cell proliferation was
Conclusion. IL-6 and Th17 play an essential role
in GPI-induced arthritis. Since it has previously been
shown that treatment with a humanized anti–IL-6R
mAb has excellent effects in patients with rheumatoid
arthritis (RA), we propose that the IL-6/IL-17 axis
might also be involved in the generation of RA, espe-
cially in the early effector phase.
Rheumatoid arthritis (RA) is characterized by
symmetric polyarthritis and joint destruction. Although
the etiology of RA is considered to be an autoimmune
reactivity to antigens that are specifically expressed in
joints, this remains a controversial hypothesis. It has
been reported that autoimmune reactivity to a ubiqui-
tous cytoplasmic enzyme, glucose-6-phosphate isomer-
ase (GPI), provokes joint-specific inflammation in
K/BxN mice (1,2). This finding highlights the potential
role of systemic autoreactivity to certain ubiquitous
autoantigens in the pathogenesis of RA.
More recently, it was reported that arthritis can
also be induced in DBA/1 mice by immunization with
GPI (3). GPI-induced arthritis is different from
collagen-induced arthritis (CIA) with regard to the
priority of T cells and B cells. In CIA, treatment with
anti-CD4 monoclonal antibodies (mAb) is ineffective
after the mice have produced antibodies to type II
collagen (4,5), and CD4-deficient mice can develop CIA
at the same incidence and severity as untreated mice (6).
Adoptive transfer of IgG antibodies purified from mice
Drs. Matsumoto and Sumida’s work was supported by a grant
from the Japanese Ministry of Science and Culture.
1Keiichi Iwanami, MD, Yoko Tanaka-Watanabe, MSc, Asuka
Inoue, BSc, Mizuko Mamura, MD, PhD, Daisuke Goto, MD, PhD,
Satoshi Ito, MD, PhD, Akito Tsutsumi, MD, PhD, Takayuki Sumida,
MD, PhD: University of Tsukuba, Tsukuba, Japan;2Isao Matsumoto,
MD, PhD: University of Tsukuba, Tsukuba, and PRESTO, Japan
Science and Technology Agency, Saitama, Japan;3Masahiko Mihara,
PhD: Chugai Pharmaceutical Company, Ltd., Shizuoka, Japan;4Yo-
shiyuki Ohsugi, PhD: Chugai Pharmaceutical Company, Ltd., Tokyo,
Japan;5Tadamitsu Kishimoto, MD, PhD: Osaka University, Osaka,
Address correspondence and reprint requests to Isao Matsu-
moto, MD, PhD, Division of Clinical Immunology, Major of Advanced
Biochemical Applications, Graduate School of Comprehensive Hu-
man Science, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba 305-
8575, Japan. E-mail: email@example.com.
Submitted for publication May 9, 2007; accepted in revised
form November 14, 2007.
with CIA can induce arthritis even in strains that are not
susceptible to CIA induction by conventional immuni-
zation. In GPI-induced arthritis, administration of anti-
CD4 mAb after arthritis onset rapidly ameliorates the
arthritis, despite the absence of changes in the anti-GPI
antibody titers. Fc? receptor–deficient mice are resistant
to GPI-induced arthritis, and adoptive transfer of puri-
fied IgG antibodies alone is not able to induce arthritis
in these mice (3). These findings indicate that although
autoantibodies are necessary for GPI-induced arthritis,
CD4? T cells are indispensable even after antibody
The present study was designed to further char-
acterize the importance of CD4? T cells in GPI-induced
arthritis. Specifically, we investigated the CD4? T cell
lineage involved in GPI-induced arthritis and the regu-
latory mechanisms of pathogenic T cells. The results
demonstrated that GPI-specific CD4? T cells shifted to
Th1 and Th17 cells and that Th17 played a crucial role
in the development of GPI-induced arthritis. We also
found that blockade of interleukin-6 receptor (IL-6R)
significantly suppressed the arthritis and inhibited Th17
differentiation. The main message of this study is that
the IL-6/IL-17 axis may be essential for the development
of T cell–dependent autoimmune arthritis.
MATERIALS AND METHODS
Mice. Male DBA/1 mice were purchased from Charles
River Laboratories (Yokohama, Japan). All mice were main-
tained under specific pathogen–free conditions, and all exper-
iments were conducted in accordance with the institutional
GPI-induced arthritis. Recombinant human GPI was
prepared as described previously (7). Briefly, human GPI
complementary DNA was inserted into plasmid pGEX-4T3
(Pharmacia, Uppsala, Sweden) for expression of glutathione
S-transferase–tagged proteins. The Escherichia coli–harboring
pGEX-hGPI plasmid was allowed to proliferate overnight at
37°C before the addition of 0.1 mM IPTG to the medium,
which was followed by a further culture overnight at 30°C. The
bacteria were lysed with a sonicator, and the supernatant was
purified with a glutathione–Sepharose column (Pharmacia).
The purity was estimated by sodium dodecyl sulfate–
polyacrylamide gel electrophoresis.
Mice were immunized intradermally with 300 ?g of
recombinant human GPI in Freund’s complete adjuvant
(Difco, Detroit, MI). Recombinant human GPI and Freund’s
complete adjuvant were emulsified at a 1:1 ratio (volume/
volume). For induction of arthritis, 150 ?l of the emulsion was
injected intradermally into the base of the tail. For intracell-
ular staining and cell proliferation assay, 50 ?l was injected
into each footpad of the hind paw. Arthritis was evaluated
visually, and changes in each paw were scored on a scale of
0–3, where 0 ? no evidence of inflammation, 1 ? subtle
inflammation or localized edema, 2 ? easily identified swelling
that was localized to either the dorsal or ventral surface of the
paw, and 3 ? swelling of all aspects of the paw.
Analysis of cytokine profiles. Mice were killed on day
7 or day 14. Spleens were harvested and hemolyzed with a
solution of 0.83% NH4Cl, 0.12% NaHCO3, and 0.004% diso-
dium EDTA in phosphate buffered saline (PBS). Single-cell
suspensions were prepared in RPMI 1640 medium (Sigma-
Aldrich, St. Louis, MO) containing 10% fetal bovine serum
(FBS), 100 units/ml of penicillin, 100 ?g/ml of streptomycin,
Figure 1. Induction of severe polyarthritis by immunization with
recombinant human glucose-6-phosphate isomerase (GPI). DBA/1
mice were immunized with 300 ?g of recombinant human GPI, and the
development of arthritis was monitored visually and scored on a scale
of 0–3 (see Materials and Methods for details). Arthritis was clinically
apparent beginning on days 7–8, peaked in severity on day 14, and then
gradually subsided. A, Severe swelling of the wrist and ankle joints
(arrowheads) in mice immunized with GPI as compared with control
mice. B, Mean ? SEM arthritis scores on days 0–28 in 10 mice from a
ROLE OF THE IL-6/IL-17 AXIS IN GPI-INDUCED ARTHRITIS 755
and 50 ?M 2-mercaptoethanol. CD4? T cells were isolated by
magnetic-activated cell sorting (Miltenyi Biotec, Bergisch
Gladbach, Germany). The purity (?97%) was confirmed by
flow cytometry. Splenic feeder cells treated with 50 ?g/ml of
mitomycin C were used as antigen-presenting cells (APCs).
Purified CD4? T cells and APCs were cocultured with 5 ?g/ml
of GPI at a ratio of 5:1 for 24 hours at 37°C in an atmosphere
containing 5% CO2. The supernatants were assayed for
interferon-? (IFN?), IL-4, and IL-17 by enzyme-linked immu-
nosorbent assay (ELISA) using a Quantikine ELISA kit (R&D
Systems, Minneapolis, MN).
Treatment of arthritis with antibodies. To neutralize
IL-17 and IFN?, mice were injected intraperitoneally with 100
?g of neutralizing antibody or isotype control on day 7 or day
14. Anti–IL-17 mAb MAB421 (IgG2a) and anti-IFN? mAb
MAB485 (IgG2a) were purchased from R&D Systems. IgG2a
isotype control was purchased from eBioscience (San Diego,
CA). For IL-6 neutralization, mice were injected intraperito-
neally with 2 mg or 4 mg of MR16-1 (an IgG1-specific mAb
against murine IL-6R) or control IgG (purified from the serum
of nonimmunized rats) on day 0, 3, 8, or 14. MR16-1 was a gift
from Chugai Pharmaceutical (Tokyo, Japan), and control IgG
was purchased from Jackson ImmunoResearch (West Grove,
Anti-GPI antibody analysis. Sera were obtained on day
28 or day 35 and diluted 1:500 in blocking solution (25%
Block-Ace [Dainippon Sumitomo Pharma, Osaka, Japan] in
PBS) for analysis of antibody. Then, 96-well plates (Sumitomo
Bakelite, Tokyo, Japan) were coated with 5 ?g/ml of recom-
binant human GPI for 12 hours at 4°C. After washing twice
with washing buffer (0.05% Tween 20 in PBS), the blocking
solution was applied for 2 hours at room temperature to block
nonspecific binding. After 2 washes, 150 ?l of diluted sera was
added, and the plates were incubated for 2 hours at room
temperature. After 3 washes, alkaline phosphatase (AP)–
conjugated anti-mouse IgG was added at a final dilution of
1:5,000 for 1 hour at room temperature. After 3 washes, color
was developed with substrate solution, consisting of 1 tablet of
AP tablet (Sigma-Aldrich) per 5 ml of AP reaction solution
(9.6% diethanolamine and 0.25 mM MgCl2, pH 9.8). Plates
were incubated for 20 minutes at room temperature, and the
optical density was read at 405 nm using a microplate reader.
Intracellular cytokine staining and flow cytometric
analysis. Mice were killed on day 7. Popliteal lymph nodes
were harvested, and single-cell suspensions were prepared as
described above. Cells (1 ? 106/ml) were stimulated with 100
?g/ml of recombinant human GPI in 96-well round-bottomed
plates (Nunc, Roskilde, Denmark) for 24 hours. GolgiStop
(BD PharMingen, San Diego, CA) was added during the last 2
hours of each culture. Cells were stained extracellularly, fixed,
and permeabilized with Cytofix/Cytoperm solution (BD
PharMingen), then the cells were stained intracellularly. A
mouse Treg cell staining kit with forkhead box P3 (FoxP3)
(eBioscience) was used to stain Treg cells according to the
protocol supplied by the manufacturer. Samples were analyzed
with a FACSCalibur flow cytometer (Becton Dickinson,
Mountain View, CA), and data were analyzed with FlowJo
software (Tree Star, Ashland, OR).
Cell proliferation assay. Mice were killed on day 10.
Popliteal lymph nodes were harvested, and single-cell suspen-
sions were prepared as described above. Cells (2 ? 107/ml) in
PBS were stained with 1.25 ?M carboxyfluorescein diacetate
succinimidyl ester (CFSE-DA; Molecular Probes, Eugene,
OR) for 8 minutes. Stained cells were cultured with 25 ?g/ml
of recombinant human GPI at 1 ? 106/ml in 96-well round-
bottomed plates (Nunc) for 60 hours and then analyzed by flow
Statistical analysis. Data are expressed as the mean ?
SEM or mean ? SD. Differences between groups were exam-
ined for statistical significance using the Mann-Whitney U test.
P values less than 0.05 were considered significant.
Induction of severe symmetric polyarthritis by
immunization with GPI. For the induction of arthritis,
we immunized DBA/1 mice with 300 ?g of recombinant
human GPI emulsified with Freund’s complete adjuvant.
Of the 177 mice immunized with recombinant human
GPI, 167 (94.4%) developed severe swelling of the wrist
and ankle joints (Figure 1A). The arthritis appeared on
Figure 2. Differentiation of glucose-6-phosphate isomerase (GPI)–specific CD4? T cells
into Th1 and Th17 cells. CD4? T cells and mitomycin C–treated antigen-presenting cells
were stimulated for 24 hours with GPI on either day 7 (induction phase) or day 14 (effector
phase) and then assessed for the production of interferon-? (IFN?), interleukin-17 (IL-17),
and IL-4 by enzyme-linked immunosorbent assay. Values are the mean and SD of 3
independent experiments (n ? 3 mice per experiment). ? ? P ? 0.05 versus cells stimulated
on day 14, by Mann-Whitney U test. UD ? undetectable (?2 pg/ml).
756IWANAMI ET AL
days 7–8, showed peak severity on day 14, then gradually
subsided (Figure 1B).
Differentiation of GPI-specific CD4? effector T
cells to Th1 and Th17 cells, but not Th2 cells. CD4? T
cells are indispensable for both the induction phase and
the effector phase of GPI-induced arthritis (3); however,
the lineage to which GPI-specific CD4? effector T cells
are differentiated remains to be elucidated. To deter-
mine the lineage, we stimulated CD4? T cells with
recombinant human GPI on day 7 (induction phase) or
day 14 (effector phase) in vitro and then assessed
cytokine production by ELISA. GPI-specific CD4? T
cells produced IFN? and IL-17, but not IL-4, on days 7
and 14 (Figure 2). Interestingly, IFN? production was
lower on day 7 than on day 14 (P ? 0.05), whereas IL-17
production was higher on day 7 than on day 14 (P ?
0.05). These data demonstrated that GPI-specific CD4?
effector T cells are differentiated to Th1 and Th17 and
are regulated differently during the development of
Crucial role of Th17 cells in the induction phase.
If GPI-specific CD4? T cells produce both IFN? and
IL-17, then which of these two cytokines affects the
development of arthritis? To answer this question, we
injected 100 ?g of anti-IFN? mAb or anti–IL-17 mAb
intraperitoneally on day 7 or day 14 after immunization
with recombinant human GPI. Injection of anti–IL-17
mAb on day 7 resulted in significant improvement in the
arthritis scores as compared with injection of isotype
control (P ? 0.01), but injection of anti–IL-17 mAb on
day 14 did not affect the course of the disease (Figure
3A). In contrast, injection of anti-IFN? mAb on day 7
and day 14 did not ameliorate arthritis, but rather,
tended to exacerbate it (Figure 3A).
Figure 3. Suppression of the development of glucose-6-phosphate isomerase (GPI)–induced arthritis by treatment with
anti–interleukin-17 (anti–IL-17) monoclonal antibody (mAb). A, Arthritis scores following intraperitoneal injection of 100 ?g
of anti–IL-17 mAb or anti–interferon-? (anti-IFN?) mAb on day 7 or day 14 after GPI immunization (arrow). Values are the
mean and SEM of 5 mice per group. Results are representative of 2 independent experiments. ? ? P ? 0.05; ?? ? P ? 0.01
versus isotype control at the same time point, by Mann-Whitney U test. B, Titers of anti-GPI antibody in sera obtained on day
35 following intraperitoneal injection of 100 ?g of anti–IL-17 mAb on day 7 or day 14 after GPI immunization, as determined
by enzyme-linked immunosorbent assay. Each symbol represents a single mouse. Bars show the mean ? SD optical density
(OD) at 405 nm. NS ? not significant (by Mann-Whitney U test).
ROLE OF THE IL-6/IL-17 AXIS IN GPI-INDUCED ARTHRITIS 757
Next, we explored whether anti–IL-17 mAb af-
fects the production of anti-GPI antibodies. Treatment
of mice with anti–IL-17 mAb on day 7 or on day 14 did
not appreciably affect the titers of anti-GPI antibody
(Figure 3B). These results indicate that Th17 cells are
involved in the development of GPI-induced arthritis
independently of anti-GPI antibody titers.
Inhibition of arthritis by anti–IL-6R mAb. It has
been reported that IL-6 plays an important role in the
differentiation of Th17 cells from naive T cells (8,9). We
speculated that blockade of IL-6 might inhibit the
development of arthritis, and we examined the effects of
anti–IL-6R mAb MR16-1 on the development of arthri-
tis. We injected 2 mg of MR16-1 intraperitoneally on day
0, 3, or 8 after immunization with recombinant human
GPI, or we injected 4 mg on day 14 after immunization.
As we anticipated, injection of MR16-1 on day 0 com-
pletely blocked the development of arthritis (Figure 4A),
and injection on day 3 showed an almost complete
inhibition (Figure 4B). Even after the development of
arthritis, injection of MR16-1 on day 8 significantly
suppressed the progression of arthritis (Figure 4C);
however, injection of 4 mg of MR16-1 on day 14, at the
peak of arthritis, did not ameliorate arthritis (Figure
4D). These results suggest that blockade of IL-6R has
protective effects and some therapeutic effects on GPI-
Inhibition of the development of Th17 cells,
without an increase in Th1, Th2, or Treg cell popula-
tions, by anti–IL-6R mAb. To examine whether MR16-1
affects Th1, Th2, and Treg cells, we cultured cells from
draining lymph nodes obtained on day 7 in the presence
of recombinant human GPI for 24 hours. Since the
majority of cells that produce IL-17 are of the CD4high
population, we analyzed IFN? and IL-4 production
gating on the CD4highpopulation. We found that the
majority of cells that produced cytokines such as IL-17
expressed CD4highcells (data not shown).
Figure 4. Inhibition of the development of arthritis by treatment with anti–interleukin-6 receptor
(anti–IL-6R) monoclonal antibody (mAb). Mice were immunized with glucose-6-phosphate isomerase
(GPI) and injected intraperitoneally with 2 mg of the anti–IL-6R mAb MR16-1 or control Ig on day 0 (A),
day 3 (B), or day 8 (C) or with 4 mg of MR16-1 or control Ig on day 14 (D) after GPI immunization. The
development of arthritis was monitored visually and scored on a scale of 0–3 (see Materials and Methods
for details). Arrow indicates the date of mAb injection. Values are the mean and SEM of 5 mice per group.
Results are representative of 2 independent experiments. ? ? P ? 0.05 versus controls, by Mann-Whitney
758IWANAMI ET AL
We performed intracellular cytokine staining for
IL-17, IFN?, and IL-4 without nonspecific stimulants, such
as phorbol myristate acetate or ionomycin, to assess phys-
iologic cytokine production. Injection of MR16-1 on day 0
resulted in a significant decrease in IL-17 production by
CD4highT cells (1.39%) as compared with injection of
control Ig (15.1%) (P ? 0.05), and there was a similar
tendency with injection on day 3 (7.59% versus 14.3%; P ?
0.05) (Figure 5A). IFN? production was not significantly
increased by MR16-1 injection on day 0 (1.35% versus
1.56%) or on day 3 (0.73% versus 1.4%) (Figure 5A).
There was no difference in IL-4 production (Figure 5A).
We also used intracellular staining methods to
examine FoxP3 expression after treatment with MR16-1.
FoxP3-positive CD4? T cells were essentially unaffected
by MR16-1 treatment on day 0 or day 3 (Figure 5B).
These data indicate that MR16-1 prevents the differen-
tiation of naive T cells to Th17 cells, but does not affect
other cell lineages.
Inhibition of the production of antigen-specific
antibodies and antigen-specific proliferation of CD4? T
cells by anti–IL-6R mAb. We next explored whether
MR16-1 affects the production of anti-GPI antibodies.
Treatment of mice with MR16-1 resulted in significant
Figure 5. Inhibition of the differentiation of draining lymph node cells into Th17 cells by treatment with
anti–interleukin-6 receptor (anti–IL-6R) monoclonal antibody (mAb). Mice were immunized with glucose-6-
phosphate isomerase (GPI) and injected intraperitoneally with 2 mg of the anti–IL-6R mAb MR16-1 or with rat
IgG (control) on day 0 or day 3 after GPI immunization. A, Cells from draining lymph nodes obtained on day 7
were cultured in the presence of 100 ?g of recombinant human GPI. GolgiStop was added during the last 2 hours
of each culture, and flow cytometric analysis of IL-17 and either interferon-? (IFN?) or IL-4 was performed,
gating on CD4highcells. Results are representative of 3 independent experiments (n ? 2 mice per experiment).
B, Cells from draining lymph nodes (DLN) and spleen obtained on day 7 were stained with forkhead box P3
(FoxP3), and flow cytometric analysis of FoxP3 and CD4 was performed. Results are representative of 3
independent experiments (n ? 2 mice per experiment). Values shown in the dot plots are the percentages of
positive cells in the compartment.
ROLE OF THE IL-6/IL-17 AXIS IN GPI-INDUCED ARTHRITIS759
reductions of anti-GPI antibody titers on days 3, 8, and
14 (P ? 0.0283, P ? 0.0090, P ? 0.0283, respectively) as
compared with mice injected with control Ig (Figure
6A). These results emphasize the inhibitory effects of
MR16-1 on the production of anti-GPI antibodies irre-
spective of the phase of arthritis when treatment is
In addition to antibody production, IL-6 is in-
volved in T cell proliferation (10). Therefore, we ex-
plored whether MR16-1 affects antigen-specific prolif-
eration of CD4? T cells. Mice were injected
intraperitoneally with 2 mg of MR16-1 on day 0, 3, or 8
after immunization of recombinant human GPI. Popli-
teal lymph nodes were harvested on day 10, cells stained
with CFSE-DA were cultured with recombinant human
GPI for 60 hours, and cell proliferation was analyzed by
flow cytometry. As expected, CD4? T cells treated with
MR16-1 in vivo proliferated significantly less than those
treated with control IgG (21.7% versus 2.22% on day 0,
30.3% versus 20.2% on day 3, 36.2% versus 27.7% on
day 8) (P ? 0.05) (Figure 6B). These data suggest that
MR16-1 inhibits antigen-specific proliferation of CD4?
T cells, leading to a reduced population of antigen-
specific CD4? T cells in draining lymph nodes.
GPI, a ubiquitous glycolytic enzyme, is a new
candidate autoantigen in the initiation of autoimmune
arthritis (2). The arthritogenicity of GPI was first de-
scribed in T cell receptor–transgenic K/BxN mice (2). In
K/BxN mice, CD4? T cells (especially KRN T cells)
were required for the development of arthritis, although
they appeared to be dispensable after the mice produced
arthritogenic autoantibodies to GPI (11). While the
K/BxN mouse is a striking model of spontaneous arthri-
tis, the effectiveness of biologic agents used to treat the
arthritis is limited. Tumor necrosis factor ? (TNF?)
blockade had no effect on the development and progres-
sion of arthritis in K/BxN mice (12), and serum transfer
from arthritic K/BxN mice into IL-6–deficient mice did
not affect the course of arthritis as compared with that in
wild-type mice (13).
GPI-induced arthritis is produced by immuniza-
tion of genetically unaltered DBA/1 mice with GPI. In
GPI-induced arthritis, administration of either anti-
TNF? mAb or CTLA-4Ig after the onset of arthritis
shows a significant amelioration of the arthritis (Matsu-
moto I, et al: unpublished observations). This model is
different from the CIA model in a T cell–dependent
manner. In GPI-induced arthritis, administration of
anti-CD4 mAb around the time of immunization was
shown to completely prevent arthritis, and more note-
worthy, administration of anti-CD4 mAb on day 11 and
Figure 6. Inhibition of the production of anti–glucose-6-phosphate
isomerase (anti-GPI) antibodies and the proliferation of CD4? T cells
by treatment with anti–interleukin-6 receptor (anti–IL-6R) monoclo-
nal antibody (mAb). A, Mice were immunized with glucose-6-
phosphate isomerase (GPI) and injected intraperitoneally with 2 mg of
the anti–IL-6R mAb MR16-1 or rat IgG (control) on day 0, 3, or 8, or
with 4 mg of mAb MR16-1 or control Ig on day 14 after GPI
immunization. Sera were obtained on day 28, and the titers of anti-GPI
antibodies were analyzed by enzyme-linked immunosorbent assay.
Each symbol represents a single mouse. Bars show the mean ? SD
optical density (OD) at 405 nm. P values were determined by
Mann-Whitney U test. B, Mice were injected intraperitoneally with 2 mg
of mAb MR16-1 or rat IgG (control) on day 0, 3, or 8 after immunization.
Cells from draining lymph nodes (DLN) obtained on day 10 were stained
with carboxyfluorescein diacetate succinimidyl ester (CFSE-DA), cul-
tured with 25 ?g of recombinant human GPI for 60 hours, and cell
proliferation was analyzed by flow cytometry. Values are the percentage
of proliferating cells. Results are representative of 3 independent exper-
iments (n ? 2 mice per experiment). ? ? P ? 0.05 versus controls, by
Mann-Whitney U test.
760IWANAMI ET AL
on day 14 was shown to induce rapid remission of the
arthritis (3). These findings highlight the importance of
CD4? T cells in the induction phase and the effector
phase of GPI-induced arthritis. In contrast, in CIA,
CD4? T cells are indispensable only until the B cells
produce autoantibodies, since anti-CD4 mAb treatment
is ineffective when administered after anti-GPI antibod-
ies have appeared (4,5). Judging from these findings,
GPI-induced arthritis is considered a useful murine
model for analyzing the role of CD4? T cells in the
effector phase of the arthritis.
Several studies have examined the roles of Th17
cells, a distinct lineage of CD4? effector T cells, in
various arthritis models (14–17). CIA was shown to be
partially suppressed in IL-17–deficient mice (16),
whereas it was exacerbated in IFN?-deficient mice or
IFN? receptor–deficient mice (18–20). Despite the sim-
ilarity of Th1 and Th17, the efficacy of anti–IL-17 mAb
treatment in GPI-induced arthritis was more marked
than in CIA. In the CIA model, administration of
anti–IL-17 antibodies during the induction phase of
arthritis was shown to only partially inhibit the develop-
ment of arthritis (21). This difference between GPI-
induced arthritis and CIA may reflect a more substantial
contribution from cells of the Th17 lineage. In our
experiments, the production of IL-17 on day 7 was
higher than that on day 14, and for IFN?, the inverse was
true, with lower production of IFN? on day 7 than on
day 14. It has been reported that IFN? suppresses the
production of IL-17 by inhibiting IL-23R (22,23); there-
fore, a cytokine milieu in which little IFN? is present
during the induction phase of arthritis might boost the
production of a large amount of IL-17, and conversely, a
milieu in which large amounts of IFN? are present
during the effector phase of arthritis might inhibit the
production of IL-17. This might also account for the fact
that spontaneous remission began on day 14 in mice with
Recent in vitro studies indicated that IL-6 is an
essential inducer of the differentiation of Th17 cells
(8,9). In our experiments, blockade of IL-6R on days 0
and 3 markedly suppressed the production of IL-17 and
the proliferation of GPI-specific CD4? T cells in vivo.
In contrast, GPI-induced arthritis was suppressed by
MR16-1 administration on days 0 and 3, and when
MR16-1 was administered on day 8, the arthritis was
ameliorated, which likely occurred through inhibition of
T cell proliferation and autoantibody production, rather
than blockade of Th17 differentiation. MR16-1 also
suppressed autoantibody production most significantly
when administered on day 8. This effect was probably
mediated through a direct action on B cells (24,25)
because the production of anti-GPI antibodies was
highest around day 8 (Matsumoto I, et al: unpublished
In the present experiments, the dose of MR16-1
we administered was 20–40 times higher than the dose
of the anti–IL-17 mAb. MR16-1 is a mAb against murine
IL-6R, and for there to be sufficient inhibition of the
biologic activity of IL-6 in vivo, soluble IL-6 receptors,
which are consistently present in the blood, would have
to be blocked. Therefore, a relatively high dose would be
needed compared with the titer of antibodies to the
cytokine itself. This idea is supported by our unpub-
lished data (Matsumoto I, et al: unpublished observa-
tions) showing that MR16-1 inhibited the biologic activ-
ity of IL-6 in vitro when administered at the same
concentration as other antibodies to the cytokine itself.
Are these scenarios applicable to RA in humans?
The therapeutic effects of a humanized anti–IL-6R?
antibody (tocilizumab) on RA have recently been re-
ported (26,27). Patients with severe forms of RA re-
tained high titers of anti-GPI antibodies (7,28,29), al-
though a few control subjects also had these antibodies.
In anti-GPI antibody–positive individuals, GPI-reactive
CD4? T cells, especially Th1-type cells, were specifically
detected in peripheral blood mononuclear cells from
RA patients who shared either the HLA–DR*0405 or
*0901 haplotype (30). What about mice with GPI-
induced arthritis? High titers of anti-GPI antibodies
have been found to be produced by arthritis-resistant
C57BL/6 mice as well, although their T cells exhibited
weak GPI responses (ref. 3 and Matsumoto I, et al:
unpublished observations) as compared with the re-
sponses of T cells from arthritis-susceptible DBA/1 mice.
These findings indicate that anti-GPI antibodies
are not sufficient for the induction of arthritis; it is
probable that the support of antigen-specific T cell
activation is indispensable. In this regard, GPI-induced
arthritis seems to be a useful model for analyzing the
pathology of RA in humans. In addition, it has been
shown that TNF antagonists clearly inhibit the progres-
sion of GPI-induced arthritis (3), even after clinical
onset of disease (Matsumoto I, et al: unpublished obser-
vations). In our present study, administration of anti–
IL-17 mAb or MR16-1 on day 14 (late effector phase)
was not able to ameliorate GPI-induced arthritis. How-
ever, both the IL-6/IL-17 axis and TNF? might play a
crucial role in established RA, since both tocilizumab
and TNF antagonists have shown marked therapeutic
efficacy in humans with established RA (26,27,31–34),
although administration of MR16-1 or anti-TNF mAb
has shown no effect or only a weak effect on fully
established CIA in mouse models (35,36). Further ana-
ROLE OF THE IL-6/IL-17 AXIS IN GPI-INDUCED ARTHRITIS 761
lysis is necessary to determine whether GPI-reactive
Th17 cells exist in the peripheral blood or joints of
patients with RA who have anti-GPI antibodies.
In conclusion, the findings of our study highlight
the importance of the IL-6/IL-17 axis in GPI-induced
arthritis, a murine model of RA. Blockade of IL-6R
might be a useful therapeutic strategy in Th17-mediated
arthritis. Since a humanized anti–IL-6R mAb has been
shown to have an excellent therapeutic effect on RA,
further studies are needed to confirm that the IL-6/IL-17
axis is also crucial in RA.
Dr. Matsumoto had full access to all of the data in the study
and takes responsibility for the integrity of the data and the accuracy
of the data analysis.
Study design. Iwanami, Matsumoto, Sumida.
Acquisition of data. Iwanami, Matsumoto, Tanaka-Watanabe, Inoue,
Mihara, Ohsugi, Mamura, Goto, Ito, Tsutsumi, Kishimoto, Sumida.
Analysis and interpretation of data. Iwanami, Matsumoto, Sumida.
Manuscript preparation. Iwanami, Matsumoto, Sumida.
Statistical analysis. Iwanami, Matsumoto.
1. Kouskoff V, Korganow AS, Duchatelle V, Degott C, Benoist C,
Mathis D. Organ-specific disease provoked by systemic autoreac-
tivity. Cell 1996;87:811–22.
2. Matsumoto I, Staub A, Benoist C, Mathis D. Arthritis provoked by
linked T and B recognition of a glycolytic enzyme. Science
3. Schubert D, Maier B, Morawietz L, Krenn V, Kamradt T. Immu-
nization with glucose-6-phosphate isomerase induces T cell-de-
pendent peripheral polyarthritis in generally unaltered mice. J Im-
4. Ranges GE, Sriram S, Cooper SM. Prevention of type II collagen-
induced arthritis by in vivo treatment with anti-L3T4. J Exp Med
5. Williams RO, Whyte A. Anti-CD4 monoclonal antibodies sup-
press murine collagen-induced arthritis only at the time of primary
immunization. Cell Immunol 1996;170:291–5.
6. Tada Y, Ho A, Koh DR, Mak TW. Collagen-induced arthritis in
CD4- or CD8-deficient mice: CD8 T cells play a role in initiation
and regulate recovery phase of collagen-induced arthritis. J Im-
7. Matsumoto I, Lee DM, Goldbach-Mansky R, Sumida T, Hitchon
CA, Schur PH, et al. Low prevalence of antibodies to glucose-6-
phosphate isomerase in patients with rheumatoid arthritis and a
spectrum of other chronic autoimmune disorders. Arthritis
8. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, et al.
Reciprocal development pathways for the generation of patho-
genic effector TH17 and regulatory T cells. Nature 2006;441:235–8.
9. Mangan PR, Harrington LE, O’Quinn DB, Helms WS, Bullard
DC, Elson CO, et al. Transforming growth factor-? induces
development of the TH17 lineage. Nature 2006;441:231–4.
10. Takeda K, Kaisho T, Yoshida N, Takeda J, Kishimoto T, Akira S.
Stat 3 activation is responsible for IL-6 dependent T cell prolifer-
ation through preventing apoptosis: generation and characteriza-
tion of T cell-specific Stat3-deficient mice. J Immunol 1998;161:
11. Korganow AS, Ji H, Mangialaio S, Duchatelle V, Pelanda R,
Martin T, et al. From systemic T cell self-reactivity to organ-
specific autoimmune disease via immunoglobulins. Immunity
12. Kyburz D, Carson DA, Corr M. The role of CD40 ligand and
tumor necrosis factor ? signaling in the transgenic K/BxN mouse
model of rheumatoid arthritis. Arthritis Rheum 2000;43:2571–7.
13. Ji H, Pettit A, Ohmura K, Ortiz-Lopez A, Duchatelle V, Degott C,
et al. Critical roles for interleukin 1 and tumor necrosis factor ? in
antibody-induced arthritis. J Exp Med 2002;196:77–85.
14. Nakae S, Nambu A, Sudo K, Iwakura Y. Suppression of immune
induction of collagen-induced arthritis in IL-17-deficient mice.
J Immunol 2003;171:6173–7.
15. Nakae S, Saijo S, Horai R, Sudo K, Mori S, Iwakura Y. IL-17
production from activated T cells is required for the spontaneous
development of destructive arthritis in mice deficient in IL-1
receptor antagonist. Proc Natl Acad Sci U S A 2003;100:5986–90.
16. Koenders MI, Lubberts E, Oppers-Walgreen B, van den Bersse-
laar L, Helsen MM, Di Padova FE, et al. Blocking of interleu-
kin-17 during reactivation of experimental arthritis prevents joint
inflammation and bone erosion by decreasing RANKL and inter-
leukin-1. Am J Pathol 2005;167:141–9.
17. Koenders MI, Kolls JK, Oppers-Walgreen B, van den Bersselaar
L, Joosten LA, Schurr JR, et al. Interleukin-17 receptor deficiency
results in impaired synovial expression of interleukin-1 and matrix
metalloproteinase 3, 9, and 13 and prevents cartilage destruction
during chronic reactivated streptococcal cell wall–induced arthri-
tis. Arthritis Rheum 2005;52:3239–47.
18. Chu CQ, Song Z, Mayton L, Wu B, Wooley PH. IFN? deficient
C57BL/6 (H-2b) mice develop collagen induced arthritis with
predominant usage of T cell receptor V?6 and V?8 in arthritic
joints. Ann Rheum Dis 2003;62:983–90.
19. Manoury-Schwartz B, Chiocchia G, Bessis N, Abehsira-Amar O,
Batteux F, Muller S, et al. High susceptibility to collagen-induced
arthritis in mice lacking IFN-? receptors. J Immunol 1997;158:
20. Vermeire K, Heremans H, Vandeputte M, Huang S, Billiau A,
Matthys P. Accelerated collagen-induced arthritis in IFN-? recep-
tor-deficient mice. J Immunol 1997;158:5507–13.
21. Lubberts E, Koenders MI, Oppers-Walgreen B, van den Bersse-
laar L, Coenen-de Roo CJ, Joosten LA, et al. Treatment with a
neutralizing anti-murine interleukin-17 antibody after the onset of
collagen-induced arthritis reduces joint inflammation, cartilage
destruction, and bone erosion. Arthritis Rheum 2004;50:650–9.
22. Harrington JE, Hatton RD, Mangan PR, Turner H, Murphy TL,
Murphy KM, et al. Interleukin-17-producing CD4?effector T cells
develop via a lineage distinct from the T helper type 1 and 2
lineages. Nat Immunol 2005;6:1123–32.
23. Park H, Li Z, Yang XO, Chang SH, Nurieva R, Wang Y, et al. A
distinct lineage of CD4 T cells regulates tissue inflammation by
producing interleukin 17. Nat Immunol 2005;6:1133–41.
24. Muraguchi A, Kishimoto T, Miki Y, Kuritani T, Kaieda T,
Yoshizaki K, et al. T cell-replacing factor (TRF)-induced IgG
secretion in human B blastoid cell line and demonstration of
acceptors for TRF. J Immunol 1981;127:412–6.
25. Yoshizaki K, Nakagawa T, Kaieda T, Muraguchi A, Yamamura Y,
Kishimoto T. Induction of proliferation and Igs-production in
human B leukemic cells by anti-immunoglobulins and T cell
factors. J Immunol 1982;128:1296–301.
26. Choy EH, Isenberg DA, Garrood T, Farrow S, Ioannou Y, Bird H,
et al. Therapeutic benefit of blocking interleukin-6 activity with an
anti–interleukin-6 receptor monoclonal antibody in rheumatoid
arthritis: a randomized, double-blind, placebo-controlled, dose-
escalation trial. Arthritis Rheum 2002;46:3143–50.
27. Nishimoto N, Yoshizaki K, Miyasaka N, Yamamoto K, Kawai S,
Takeuchi T, et al. Treatment of rheumatoid arthritis with human-
762 IWANAMI ET AL
ized anti–interleukin-6 receptor antibody: a multicenter, double-
blind, placebo-controlled trial. Arthritis Rheum 2004;50:1761–9.
28. Schaller M, Burton DR, Ditzel HJ. Autoantibodies to GPI and
creatine kinase in RA, and few human autoimmune sera detect
GPI. Nat Immunol 2002;3:412–3.
29. Van Gaalen FA, Toes RE, Ditzel HJ, Schaller M, Breedveld FC,
Verweij CL, et al. Association of autoantibodies to glucose-6-
phosphate isomerase with extraarticular complications in rheuma-
toid arthritis. Arthritis Rheum 2004;50:395–9.
30. Kori Y, Matsumoto I, Zhang H, Yasukochi T, Hayashi T, Iwanami
K, et al. Characterisation of Th1/Th2 type, glucose-6-phosphate
isomerase reactive T cells in the generation of rheumatoid arthri-
tis. Ann Rheum Dis 2006;65:968–9.
31. Maini R, St.Clair EW, Breedveld F, Furst D, Kalden J, Weisman
M, et al, for the ATTRACT Study Group. Infliximab (chimeric
anti-tumour necrosis factor ? monoclonal antibody) versus pla-
cebo in rheumatoid arthritis patients receiving concomitant meth-
otrexate: a randomized phase III trial. Lancet 1999;354:1932–9.
32. Lipsky PE, van der Heijde DM, St.Clair EW, Furst DE, Breedveld
FC, Kalden JR, et al, for the Anti–Tumor Necrosis Factor Trial in
Rheumatoid Arthritis with Concomitant Therapy Study Group.
Infliximab and methotrexate in the treatment of rheumatoid
arthritis. N Engl J Med 2000;343:1594–602.
33. Moreland LW, Baumgartner SW, Schiff MH, Tindall EA, Fleisch-
mann RM, Weaver AL, et al. Treatment of rheumatoid arthritis
with a recombinant human necrosis factor receptor (p75)-Fc
fusion protein. N Engl J Med 1997;337:141–7.
34. Weinblatt ME, Kremer JM, Bankhurst AD, Bulpitt KJ, Fleisch-
mann RM, Fox RI, et al. A trial of etanercept, a recombination
tumor necrosis factor receptor:Fc fusion protein, in patients with
rheumatoid arthritis receiving methotrexate. N Engl J Med 1999;
35. Takagi N, Mihara M, Moriya Y, Nishimoto N, Yoshizaki K,
Kishimoto T, et al. Blockage of interleukin-6 receptor ameliorates
joint disease in murine collagen-induced arthritis. Arthritis Rheum
36. Joosten LA, Helsen MM, van de Loo FA, van den Berg WB.
Anticytokine treatment of established type II collagen–induced
arthritis in DBA/1 mice: a comparative study using anti-TNF?,
anti–IL-1?/?, and IL-1Ra. Arthritis Rheum 1996;39:797–809.
ROLE OF THE IL-6/IL-17 AXIS IN GPI-INDUCED ARTHRITIS763