SHORT COMMUNICATION

Inhibition of JNK in synovium by treatmentwith golimumab in rheumatoid arthritis

Katsuaki Kanbe • Junji Chiba • Atsushi Nakamura

Received: 14 June 2012 / Accepted: 9 December 2012

� Springer-Verlag Berlin Heidelberg 2013

Abstract The aim of this study was to investigate

immunohistological changes in mitogen-activated protein

kinases (MAPKs) in the synovium following treatment

with golimumab, compared with methotrexate (MTX). We

assessed synovial tissues for 13 different molecules to

detect cytokine levels histologically from 10 methotrexate

(MTX)-treated rheumatoid arthritis (RA) patients as con-

trols and 10 golimumab plus MTX-treated RA patients.

Synovium samples from both groups were assessed by

hematoxylin and eosin (HE) staining and analyzed for

expression of tumor necrosis factor-a (TNF-a), interleukin-

6 (IL-6), matrix metalloproteinase-3 (MMP-3), CD4 (T

cells), CD8 (T cells), CD20 (B cells), CD68 (macro-

phages), receptor activator of nuclear (kappa) B ligand

(RANKL), bromodeoxyuridine (BrdU), CD29 (b-1 inte-

grin), phospho-p38 MAPK (Tyr180/Tyr182), phospho-p44/

42 MAPK (ERK1/ERK2), and phospho-c-Jun N-terminal

kinase (JNK), by an immunohistological examination. HE

staining showed that there was a significant decrease in cell

proliferation in the synovium in RA patients who received

golimumab compared with the controls. TNF-a, IL-6,

MMP3, BrdU, p38, and ERK were not seen at significant

levels in either group. On the other hand, CD4, CD8,

CD20, CD29, CD68, RANKL, and JNK were significantly

decreased in the golimumab group compared with the

control. Based on a histological analysis of the synovium, it

appears that the efficacy of the treatment with golimumab

may involve the inhibition of cell proliferation, with

decreases in T cells, B cells, macrophages, b-1 integrin,

RANKL, and JNK in the synovium, compared with MTX

treatment, in RA.

Keywords Golimumab � Synovium � Histology �Rheumatoid arthritis � MAPK

Introduction

Rheumatoid arthritis (RA) is one of the most common

immune-mediated diseases and is characterized by syno-

vial inflammation and joint destruction [1]. Recently, sev-

eral anti-tumor necrosis factor therapies have been

developed for treating RA, including golimumab. Goli-

mumab is a human monoclonal antibody against tumor

necrosis factor-a (TNF-a).

Clinically, the effects of golimumab in RA have been

evaluated in three large, phase III clinical trials that

included patients with differing exposures to previous

therapies [2–4]. The ‘‘golimumab before employing

methotrexate as the first-line option in the treatment of

rheumatoid arthritis of early onset’’ (GO-BEFORE) study

was of patients with active RA who were naıve to meth-

otrexate (MTX), the ‘‘golimumab in active rheumatoid

arthritis despite methotrexate therapy’’ (GO-FORWARD)

study was of patients with active RA despite MTX therapy,

and the ‘‘golimumab in patients with active rheumatoid

arthritis after treatment with tumor necrosis factor-ainhibitors’’ (GO-AFTER) study was of patients with active

RA who had previously received anti-TNF-a therapy

[2–4]. These trials provided evidence of clinical efficacy

and are useful in starting therapy with golimumab. How-

ever, it still needs to be understood how golimumab affects

the synovium in joints to inhibit synovitis or bone erosion.

To date, there has been no reported investigation regarding

K. Kanbe (&) � J. Chiba � A. Nakamura

Department of Orthopaedic Surgery, Tokyo Women’s Medical

University, Medical Center East, 2-1-10 Nishiogu,

Arakawa, Tokyo 116-8567, Japan

e-mail: kanbeor@dnh.twmu.ac.jp

123

Rheumatol Int

DOI 10.1007/s00296-012-2626-7

the histological effects on the synovium treated with goli-

mumab, especially changes in the expression of mitogen-

activated protein kinase (MAPK). Based on the hypothesis

that anti-TNF-a therapy with golimumab indicates the

induction of different expression patterns of cytokines and

MAPK, we conducted a histological evaluation of 13

molecules (TNF-a, IL-6, MMP-3, CD4, CD8, CD20,

CD29, CD68, BrdU, RANKL, ERK, JNK, and p38 MAPK)

in the synovium of RA patients treated with golimumab.

This report presents findings relating to histological dif-

ferences in the synovium associated with treatment with

golimumab in RA.

Materials and methods

Ten patients (2 males, 8 females) who were treated with

golimumab had been treated for an average of 6 (range,

4–12) months and had a mean age of 65.7 (range, 51–75)

years and a mean duration of disease of 11 (range, 5–18)

years. All patients treated with golimumab were moderate

responders with mean disease activity (DAS28) [5], Clin-

ical Disease Activity Index (CDAI), and Simplified Dis-

ease Activity Index (SDAI) [6] scores of 3.4 ± 0.6,

12.8 ± 6.4, and 13.1 ± 7.1, respectively, at arthroscopic

synovectomy in the knee and shoulder joints. They

received 50 mg of golimumab every month plus an average

of 6.9 mg/week of MTX (4–12 mg) and 2.6 mg/day of

prednisolone (PSL; 2.5–10 mg). According to the Stein-

brocker criteria [7], four and six patients in the golimumab

group were classified with class II and class III disease,

respectively.

Control samples were obtained from ten RA patients

who did not receive golimumab, or any other biologic,

showing DAS28, CDAI, and SDAI scores of 3.3 ± 0.7,

11.7 ± 7.5, and 12.8 ± 8.3, respectively, at least 6 months

(average, 10 months) after receiving MTX. These 10

control patients (2 males, 8 females) had an average age of

61.3 (range, 52–75) years and, on average, received

6.7 mg/week of MTX (4–14 mg) and 2.9 mg/day of

prednisolone (2.5–10 mg) at the time of taking sample

specimens. Synovial tissue specimens were obtained from

all patients in the MTX group during arthroscopic syno-

vectomy of the knee and shoulder joints. According to the

Steinbrocker criteria [6], four and six patients in the control

group were classified as having class II and class III dis-

ease, respectively.

We selected patients with almost the same disease

activity for comparing the DAS28 (CRP), CDAI, and SDAI

scores in the two groups to determine the effect of

golimumab (Table 1). Inclusion criteria and indications

for synovectomy were MTX and PSL usage, moderate

disease activity with persistent pain of knee or shoulder

joints, and no previous history of RA surgery. Exclusion

criteria were low or high disease activity, other disease-

modifying anti-rheumatic drugs (DMARDs), non-steroidal

anti-inflammatory drugs (NSAIDs), hormonal therapy, and

other surgeries.

All patients who took part in this study provided written

informed consent. The ethics committee of our university

approved the study (1321).

All patients were diagnosed according to the criteria of

the American College of Rheumatology [8]. To evaluate

disease activity in the two groups, the erythrocyte sedi-

mentation rate (ESR), CRP, DAS28, CDAI, and SDAI

scores were measured to compare with the histological

findings in the synovium.

Serial paraffin sections of the synovium (5 lm) were

stained with hematoxylin and eosin (HE). For the immu-

nohistological examination, the tissue sections were

blocked for 10 min in phosphate-buffered saline containing

20 % rabbit serum and then incubated overnight at 4 �C

with the following antibodies: anti-TNF-a mouse mono-

clonal antibody (1:1,000; Biogenesis, Poole, United King-

dom), anti-human IL-6 rabbit polyclonal antibody

(Rockland Inc., Gilbertsville, PA, US), anti-human MMP-3

monoclonal antibody (1:100; Biogenesis), anti-human

CD20 monoclonal antibody (1:1,000; DAKO, Denmark),

anti-human CD68 monoclonal antibody, anti-human CD4,

and anti-human CD8 (1:1,000; DAKO), anti-BrdU mouse

monoclonal antibody (1:500; Chemicon, Temecula, CA,

US), anti-human CD29 (beta-1 integrin) mouse monoclo-

nal antibody (1:350; Novocastra #NCL-CD29, US), anti-

human receptor activator of nuclear (kappa) B ligand

(RANKL; FL-317) rabbit polyclonal antibody (1:200;

Santa Cruz Biotechnology, CA, US), anti-human phospho-

p38 MAPK (Tyr180/Tyr182) mouse monoclonal antibody

(1:500 dilution; Cell Signaling Technology #3216S), anti-

human phospho-p44/42 MAPK (Tyr202/Tyr204, ERK1/

ERK2) mouse monoclonal antibody (1:500 dilution; Cell

Table 1 Patient’s background for comparison of histology of synovium by treatment of golimumab

Groups Age (y) D.D. (y) MTX (mg/week) PSL (mg/day) CRP (mg/dl) DAS28

Control (n = 10) 61.3 ± 9 14 ± 7 6.7 ± 4 2.9 ± 3 0.7 ± 0.9 3.3 ± 0.7

Golimumab (n = 10) 65.7 ± 8 11 ± 6 6.9 ± 3 2.6 ± 2 0.8 ± 0.6 3.4 ± 0.6

D.D. disease duration, MTX methotrexate, PSL prednisolone

Rheumatol Int

123

Signaling Technology #9106S), and anti-human phospho-

JNK (1:500 dilution, G7, Santa Cruz Biotechnology

#sc-6254, US) mouse monoclonal antibody. After treat-

ment with a secondary antibody, we compared the patterns

of TNF-a, IL-6, MMP-3, CD4, CD8, CD20, CD29, CD68,

BrdU, RANKL, ERK, p38, and JNK between the goli-

mumab and control groups. After treating the samples with

a secondary antibody at room temperature for 10 min,

sections were incubated for 10 min with an appropriate

Vectastain ABC reagent (Vector), using 3,30-diam-

inobenzidine-4 HCl (DAB; Sigma) for the color reaction

for 5 min, resulting in clear brown staining of positive

cells. Microscopic evaluations were performed as descri-

bed previously [9]. The immunohistologically stained

samples were evaluated by estimating the number of pos-

itively stained cells in three different areas of the synovium

(superficial lining layer, perivascular area, and connective

tissue area). Positivity was noted when complete staining

of the cells was observed in all synovial samples at a

magnification of 2009 with HE staining. Three random

readings per high-power field were recorded, and the

results are expressed as the mean percentages with standard

deviations (SD) for each of the samples according to

Farahat et al. [9].

The Mann–Whitney U test was used to examine non-

parametric continuous variables between the two groups in

the histological examination. A p value \0.05 was con-

sidered to indicate statistical significance.

Results

HE staining showed that the synovium of the RA patients

who underwent golimumab treatment showed significantly

lower cell proliferation levels than those of the MTX

controls [20 (6.5) vs. 74 (7.8), respectively; p \ 0.05;

Table 2; Fig. 1a, b]. TNF-a was expressed at a low level,

and IL-6, MMP-3, and BrdU were also seen at low levels in

both groups; there was no significant difference between

the golimumab and control groups (Tables 2, 3; Fig. 1c–f).

The percentages of cells that were positively stained

for CD68 (p = 0.014) were significantly decreased in the

golimumab group compared with the controls (Table 3;

Fig. 2a, b). Further, there were significantly different levels

of CD4, CD8, CD20, BrdU, and RANKL between the

golimumab and control groups [mean (SD) 17 (6.8), 7.5

(4.2), 12 (5.6), 4.9 (5.3), and 18 (9.2) vs. 26 (6.7), 18 (6.5),

35 (7.4), 18 (9.7), and 32 (6.5), respectively; p = 0.021,

0.011, 0.013, 0.045, and 0.012, Table 3; Figs. 2c–h, 3a–f].

For MAPK immunostaining, neither P38 nor ERK in the

synovium was significantly different between the control

and golimumab groups [mean (SD) 3.6 (3.5) and 4.6

(3.7) vs. 3.1 (1.7) and 7.1 (4.5), respectively; p = 0.361,

p = 0.223; Fig. 4c–f]. However, there were significant

differences in CD29 and JNK expression between the

control and golimumab groups [mean (SD) 41 (7.5) and 46

(8.5) vs. 7.8 (6.7) and 3.5 (7.5), respectively; p \ 0.001,

p \ 0.001; Figs. 3e, f, 4a, b]. Thus, the expression patterns

Table 2 Comparison of histological findings of synovium by treat-

ment of golimumab

H.&E. TNF-a IL-6 MMP-3

HV CP

Control

(n = 10)

27 (5.5) 74 (7.8) 29 (7.8) 17 (4.7) 12 (7.2)

Golimumab

(n = 10)

22 (6.7) 20 (6.5)* 35 (8.9) 15 (6.7) 10 (5.5)

Results expressed as mean (SD) percentage of positive fields of

staining cell numbers by immunohistochemistry

HV hyper vascularity, CP cell proliferation

* Is significant difference (p \ 0.05)

A B

C D

E F

G H

Fig. 1 Histological comparison of HE staining and immunohistolog-

ical examination (magnification 9200) for the two groups. a, b HE;

c, d TNF-a; e, f interleukin-6 (IL-6): MMP-3; g, h; a, c, e, g control

group (MTX); b, d, f, h golimumab group. Arrows indicate positive

cells

Rheumatol Int

123

of CD29 (b-1 integrin), a mechanoreceptor, and JNK, of

the MAPKs, in the synovial tissues of patients treated with

golimumab were significantly different from those of the

MTX group.

Discussion

To our knowledge, this histological study is the first

report of evidence of the mechanism underlying the

clinical efficacy of golimumab as a therapy with TNF-ablockade. Despite golimumab not previously being rec-

ognized as a cytokine inhibitor, significant inhibition of

T cells, B cells, and macrophages was observed in the

synovium of RA, compared with MTX treatment. RANKL

expression was inhibited, leading to the possibility of

inhibiting bone and joint destruction via activation of the

osteoclast pathway. Synovial cell proliferation was also

inhibited, as seen in the decreased BrdU level. Thus, TNF

blockade with golimumab may have potential to inhibit

the development of RA in terms of synovial cell prolifera-

tion, concomitant with bone erosion, through inhibiting

RANKL.

It has been reported from an MRI study that 318 patients

who received golimumab plus MTX showed improvements

in inflammation, as evidenced by the rheumatoid arthritis

magnetic resonance imaging scoring (RAMRIS) [10] for

synovitis and bone edema/osteitis and bone erosion that

exceeded those observed in patients who received placebo

plus MTX [11]. In our data, cell proliferation, including

that of T cells, B cells, and macrophages, decreased sig-

nificantly, although TNF-a was expressed, in the goli-

mumab group. This suggests that the decreased synovitis,

recognized by MRI, indicated inhibition of the cytokines

that induce synovial proliferation. It is well known that

macrophages are one of the inflammatory cytokine-

producing cells in the RA synovium, and they also differ-

entiate into osteoclasts, which have been implicated in RA

inflammatory bone erosion and joint destruction [12, 13].

Thus, direct effects of inhibition on T cells, B cells,

and macrophages by golimumab may underlie its potent

efficacy.

To approach the signal transduction pathway involved in

the mechanism of action of golimumab, mitogen-activated

protein kinase (MAPK) cascade components, including

ERK, JNK, and p38, that mediate cell responses in the

Table 3 Comparison of immunohistology of synovium by treatment of golimumab

CD68 CD20 CD4 CD8 RANKL

Control (n = 10) 41 (4.5) 35 (7.4) 26 (6.7) 18 (6.5) 32 (6.5)

Golimumab (n = 10) 11 (5.9)* 12 (5.6)* 17 (6.8)* 7.5 (4.2)* 18 (9.2)*

BrdU CD29 JNK p38 ERK

Control (n = 10) 18 (9.7) 41 (7.5) 46 (8.5) 3.6 (3.5) 4.6 (3.7)

Golimumab (n = 10) 4.9 (5.3)* 7.8 (6.7)* 3.5 (7.5)* 3.1 (1.7) 7.1 (4.5)

Results expressed as mean (SD) percentage of positive fields of staining cell numbers by immunohistochemistry

* Significant difference (p \ 0.05)

A B

C D

E F

G H

Fig. 2 Immunohistological comparison of the expression of CD68,

CD20, CD4, and CD8 (magnification 9200). a, b CD68; c, d CD20;

e, f CD4: g, h CD8; a, c, e, g control group (MTX); b, d, f,h golimumab group. Arrows indicate positive cells

Rheumatol Int

123

synovium, were investigated. MAPK is highly activated in

rheumatoid synovium and may contribute to inflammatory

and destructive mechanisms [14]. The c-Jun N-terminal

kinase (JNK), which belongs to the MAPK family, plays

important roles in cytokine production and extracellular

matrix degradation by regulating matrix metalloproteinase

(MMP) in fibroblast-like synoviocytes and in animal models

of RA [15]. Of the three JNK isoforms, JNK1 has been

implicated as a pivotal regulator of synovial inflammation in

murine arthritis due to its role in macrophage migration [16].

JNK1 also contributes to osteoclast differentiation, because

JNK1-deficient osteoclast progenitors do not mature into

bone-resorbing osteoclasts [17]. These data suggest that JNK

participates in the synovial inflammation and joint destruc-

tion of RA and could potentially be targeted in RA treatment.

Also, important in these pathways are the MAPKs, which

phosphorylate amino acid residues on key intracellular pro-

teins. These kinases regulate cell survival, proliferation,

cytokine generation, and metalloproteinase production [18].

CD29 (b-1 integrin) and JNK were downregulated; thus, the

mechanoresponse through CD29, downstream of JNK, also

decreased. These finding indicated that TNF blocking is

possibly related to the expression levels of adhesion mole-

cules (CD29) and MAPK. Several kinds of CD molecules

decreased, including CD4, CD8, CD20, and CD68. The

ability of TNF binding with golimumab on cells may inhibit

signal transduction involving these molecules in regulating

synovitis. However, it is still unknown how golimumab

inhibits cell proliferation in the synovium, such as by

affecting apoptosis or cell-cycle changes. It has been sug-

gested that JNK of the MAPKs may be involved in cross talk

with the TNF cascade in synovial cells in RA.

A limitation of this study was that the mRNA levels of

these molecules, including MAPK, need to be confirmed

after treatment with golimumab. Furthermore, the effects

on signal transduction pathways downstream of the TNF

cascade, apart from MAPK, should be investigated to

further examine the mechanism(s) of action of golimumab

in the synovium in RA.

In conclusion, the effects of golimumab may involve the

inhibition of T cells, B cells, and macrophages, with

downregulation of signal transduction pathways with

effects on the levels of CD4, CD8, CD20, CD29, RANKL,

and JNK of MAPKs in the synovium of RA.

Acknowledgments This work was supported in part by a grant-in-

aid for Scientific Research (KAKENHI) (C) (24592284) from the

Ministry of Education, Culture, Sports, Science, and Technology

(MEXT), Japan Society for the Promotion of Science (JSPS).

References

1. Firestein GS (2003) Evolving concepts of rheumatoid arthritis.

Nature 423:356–361

A B

C D

E F

Fig. 3 Immunohistological comparison of the RANKL, BrdU, and

CD29 (magnification 9200). a, b RANKL; c, d BrdU; e, f CD29; a, c,

e control group (MTX); b, d, f golimumab group. Arrows indicate

positive cells

A B

C D

E F

Fig. 4 Immunohistological comparison of the expression of JNK,

p38, and ERK (magnification 9200). a, b JNK; c, d p38; e, f ERK, a,

c, e control group (MTX); b, d, f golimumab group. The arrowsindicate positive cells

Rheumatol Int

123

2. Emery P, Fleischmann RM, Moreland LW, Hsia EC, Strusberg I,

Durez P, Nash P, Amante EJ, Churchill M, Park W, Pons-Estel

BA, Doyle MK, Visvanathan S, Xu W, Rahman MU (2009)

Golimumab, a human anti-tumor necrosis factor a monoclonal

antibody, injected subcutaneously every four weeks in metho-

trexate-naıve patients with active rheumatoid arthritis: twenty-

four-week results of a phase III, multicenter, randomized, double-

blind, placebo-controlled study of golimumab before methotrex-

ate as first-line therapy for early-onset rheumatoid arthritis. Arthr

Rheum 60:2272–2283

3. Keystone EC, Genovese MC, Klareskog L, Hsia EC, Hall ST,

Miranda PC, Pazdur J, Bae SC, Palmer W, Zrubek J, Wiekowski

M, Visvanathan S, Wu Z, Rahman MU; GO-FORWARD Study

(2009) Golimumab, a human antibody to tumour necrosis factor agiven by monthly subcutaneous injections, inactive rheumatoid

arthritis despite methotrexate therapy: the GO-FORWARD study.

Ann Rheum Dis 68:789–796

4. Smolen JS, Kay J, Doyle MK, Landewe R, Matteson EL, Wol-

lenhaupt J, Gaylis N, Murphy FT, Neal JS, Zhou Y, Visvanathan

S, Hsia EC, Rahman MU; GO-AFTER study investigators (2009)

Golimumab in patients with active rheumatoid arthritis after

treatment with tumour necrosis factor a inhibitors (GO-AFTER

study): a multicentre, randomized, double-blind, placebo-con-

trolled, phase III trials. Lancet 374:210–221

5. van der Heijde DM, van ‘t Hof MA, van Riel PL, Theunisse LA,

Lubberts EW, van Leeuwen MA, van Rijswijk MH, van de Putte

LB (1990) Judging disease activity in clinical practice in rheu-

matoid arthritis: first step in the development of a disease activity

score. Ann Rheum Dis 49:916–920

6. Felson DT, Smolen JS, Wells G, Zhang B, van Tuyl LH, Funovits

J, Aletaha D, Allaart CF, Bathon J, Bombardieri S, Brooks P,

Brown A, Matucci-Cerinic M, Choi H, Combe B, de Wit M,

Dougados M, Emery P, Furst D, Gomez-Reino J, Hawker G,

Keystone E, Khanna D, Kirwan J, Kvien TK, Landewe R, Listing

J, Michaud K, Martin-Mola E, Montie P, Pincus T, Richards P,

Siegel JN, Simon LS, Sokka T, Strand V, Tugwell P, Tyndall A,

van der Heijde D, Verstappen S, White B, Wolfe F, Zink A,

Boers M, American College of Rheumatology; European League

Against Rheumatism (2011) American College of Rheumatology/

European League Against Rheumatism provisional definition of

remission in rheumatoid arthritis for clinical trials. Arthr Rheum

63:573–586

7. Steinbrocker O, Traeger CH, Battman RC (1949) Therapeutic

criteria in rheumatoid arthritis. JAMA 140:659–662

8. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF,

Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS

(1998) The American Rheumatism Association 1987 revised

criteria for the classification of rheumatoid arthritis. Arthr Rheum

31:315–324

9. Farahat MN, Yanni G, Poston R, Panayi GS (1993) Cytokine

expression in synovial membranes of patients with rheumatoid

arthritis and osteoarthritis. Ann Rheum Dis 52:870–875

10. Østergaard M, Peterfy C, Conaghan P et al (2003) OMERACT

rheumatoid arthritis magnetic resonance imaging studies: core set

MRI acquisitions, joint pathology definitions, and the OME-

RACT RA-MRI scoring system. J Rheumatol 30:1385–1386

11. Østergaard M, Peterfy C, Conaghan P, McQueen F, Bird P,

Ejbjerg B, Shnier R, O’Connor P, Klarlund M, Emery P, Genant

H, Lassere M, Edmonds J (2011) Significant improvement in

synovitis, osteitis, and bone erosion following golimumab and

methotrexate combination therapy as compared with methotrex-

ate alone. Arthr Rheum 63:3712–3722

12. Teitelbaum SL (2000) Bone resorption by osteoclasts. Science

289:1504–1508

13. Mclnnnes IB, Schett G (2007) Cytokines in rheumatoid arthritis.

Nat Rev Immunol 7:429–442

14. Thalhamer T, McGrath MA, Harnett MM (2008) MAPKs and

their relevance to arthritis and inflammation. Rheumatology

(Oxford) 47:409–414

15. Han Z, Boyle DL, Chang L, Bennett B, Karin M, Yang L,

Manning AM, Firestein GS (2001) C-Jun N-terminal kinase is

required for metalloproteinase expression and joint destruction in

inflammatory arthritis. J Clin Invest 108:73–81

16. Guma M, Ronacher LM, Firestein GS, Karin M, Corr M (2011)

JNK-1 deficiency limits macrophage-mediated antigen-induced

arthritis. Arthr Rheum 63:1603–1612

17. David JP, Sabapathy K, Hoffmann O, Idarraga MH, Wagner EF

(2002) JNK1 modulates osteoclastogenesis through both c-Jun

phosphorylation-dependent and -independent mechanisms. J Cell

Sci 115:4317–4325

18. Kumar S, Boehm J, Lee JC (2003) p38 MAP kinases: key sig-

naling molecules as therapeutic targets for inflammatory diseases.

Nat Rev Drug Discov 2:717–726

Rheumatol Int

123