Mc Boissier

Intra-articular electrotransfer of plasmid encoding soluble TNF receptor variants in normal and arthritic mice

Anti‐inflammatory gene therapy is promising in inflammatory diseases such as rheumatoid arthritis (RA). We have previously demonstrated that intra‐muscular (i.m.) electrotransfer (ET) of plasmids encoding three different human tumor necrosis factor‐α‐soluble receptor I variants (hTNFR‐Is) exert protective effects in an experimental RA model. However, such a systemic approach could be responsible for side effects. The present study aimed at performing an intra‐articular (i.a.) gene therapy by electrotransfer using the hTNFR‐Is plasmids.

Anti-IL6R treatment acts on CD4+ FoxP3+ regulatory T cells in a model of rheumatoid arthritis

Backgroundand objectives Studies have demonstrated the clinical efficacy of tocilizumab, a humanised anti-IL-6 receptor (R) antibody (Ab), in patients with rheumatoid arthritis (RA). The rational for blocking IL-6 in this disease mainly lays on the pro-inflammatory role of this cytokine in the disease. Moreover, IL-6 acts as a ‘pivot’, determining whether an immune response is dominated by regulatory CD4+Foxp3+ T cells or Th17 effector T cells. However, only few works have studied the consequences of anti-IL-6R treatment on Tregs cells and only focuses on their frequency. In this study, the authors hypothesised that IL-6 inhibition leads to Tregs modifications in RA models given that (i) anti-IL6R is therapeutic in RA and its models by inhibiting Th17 cells, and (ii) IL-6 plays a determining role in inducing Th17 when present, or Tregs when it is absent. Our objective was thus to elucidate anti-IL-6R mode of action in RA models by studying the consequences of this treatment on Tregs phenotype and biological activity. Methods Mice with collagen-induced arthritis were treated by MR16-1 (a rat antimouse IL-6 receptor monoclonal Ab provided by Chugai Pharmaceutical Co LTD, Japan) and the evolution of Tregs (defined as CD4+ FoxP3+) during arthritis course was assessed at key time points (day 9–17–28 and 43 after CIA induction) by studying their number, frequency and phenotype (expression of GITR, ICOS, Helios, CD62L, CTLA-4 and CD39) in lymph nodes (LN) and spleen by flow cytometry. Finally, the immunosuppressive activity of the Treg cells on CFSE-labeled CD4+ CD25- T effector (Teff) cells proliferation was also studied. Numerical analysis of Th17 and Th1 cells was also performed at the same times by flow cytometry. Results Clinical evaluation of arthritides in mice treated with anti-mouse IL-6R mAb showed, as expected, a less severe disease as compared to control Ig treated mice. A decreased Teff/Tregs ratio overtime were observed in the LN of MR16-1 treated mice. Tregs phenotype was modified in treated mice mostly at day 17, with a decreased frequency of Tregs expressing CD62L or Helios (spleen) but an enhanced intensity of CD39 expression on Tregs (spleen and LN) and increased frequency of CD39+ Tregs (LN). No modification in ICOS, GITR and CTLA-4 were observed on Tregs at any time in any compartments. Lastly, the immunosuppressive activity of the Treg cells on Teff cells proliferation was not modified, suggesting that Tregs immunosuppression activity was probably due to another mechanism.. Conclusion Tregs are modified by anti-IL-6R treatment in CIA, that could result in an enhanced peripheral generation of Tregs (Helios down regulation), an increased Tregs activation (CD39+ upregulation) and a better capacity to leave the secondary lymphoid organs and to migrate within the inflammatory joint (decreased CD62L down regulation). Altogether, this study shows that inhibition of IL-6 in CIA also acts by modifying Tregs state.

IL-35 gene therapy in a model of rheumatoid arthritis

Backgroundand objectives Interleukin (IL)-35 is a novel interleukin described as an immunosuppressive cytokine specifically produced by CD4+FoxP3+ regulatory T (Tregs) cells but not by conventional T cells. Since Tregs play a major role in autoimmunity maintenance and protects from inflammation, the authors aimed at evaluating the role of IL-35 in a model of rheumatoid arthritis in mice using a non viral gene transfer strategy. Methods Two plasmids were used and compared: pIGneo-mouse (m)IL-35-GFP (gift from Dario Vignali) and pORF-mIL-35, where IL-35 is expressed under CMV or HTLV promoter, respectively. pIGneo-mouse (m)IL-35-GFP contains a ‘native’ form of IL-35 made of Ebi3 and p35 linked subunits, while pORF-mIL-35contains a ‘single chain’ construct. In vitro, IL-35 expression was detected in lysate of transfected HEK cells by western blot (p35 or Ebi3) and immunoprecipitation (IP). The in vivo trangene expression was detected by fluorescence analysis of GFP on muscle slides after intra-muscular electrotransfer (ET) of 60 µg pIGneo-mIL-35-GFP plasmid. The clinical effect of IL-35 was assessed in collagen-induced arthritis in mice receiving two im ET of 60 µg pIGneo-mIL-35-GFP or pORF-IL-35, 3 and 18 days after CIA induction. Control groups received the corresponding empty plasmids or PBS. Tregs and Th17 frequencies and phenotypes were performed by flow cytometry in the spleen and lymph nodes (LN) of pORF-IL-35 treated mice at day 34. Results Both subunits p35 and Ebi3 were expressed in lysates from either pIGneo-mIL-35-GFP or pORF-IL-35 HEK transfected cells, and formed the IL-35 heterodimer as shown by IP. At least 3 days after intramuscular ET of pIGneo-mIL-35-GFP plasmid, the authors were able to detect GFP expression, revealing in vivo IL-35 expression, on muscle histological slides. Surprisingly, whatever the plasmid used, IL-35 gene transfer resulted in a statistically significant increase in clinical scores of collagen-induced arthritis as compared to empty plasmid. The underlying cellular mechanisms of this effect were shown to be related to an increased in number and frequency of Th17 cells in the spleen of pORF-mIL-35 treated mice, leading to an increased Th17/Tregs ratio. Interestingly, CD39 expression on Tregs increased (frequency and intensity) in the LN of pORF-mIL-35 treated mice as compared to control groups, but CD62L expression also increased (spleen and LN) on these cells. Conclusion The authors show an unexpected but clear exacerbating effect of IL-35 gene transfer in an autoimmune and inflammatory model of RA. The authors can hypothesise that IL-35, as already known, induced an activation of Tregs by up-regulating CD39 expression on these cells, but that the concomitant up-regulation of CD62L prevent their migration into the inflammatory joint by retaining them in secondary lymphoid organs, thus impeding them from exerting their immunosuppressive role.

Differentiation and activation of regulatory T lymphocytes in a TNF transgenic model of arthritis, and the impact of passive or active anti-TNF therapies

Background and objectives Tumour necrosis factor (TNF) is a critical cytokine in rheumatoid arthritis. The authors recently demonstrated the efficacy of active immunotherapy against TNF (TNF-K) in a human TNF transgenic (TTg) mouse model of arthritis. Studies with anti-TNF treatments in autoimmune diseases suggest that TNF and TNF antagonists might act through the involvement of a T cells subset, namely CD4+ CD25+ FoxP3+ regulatory T cells (Treg). In the present study, the authors aimed to better characterise the impact on the Treg population of (1) a TNF driven disease, specifically arthritis in TTg mice and (2) active or passive anti-TNF immunotherapy in this same model. Methods First, untreated TTg mice were euthanised sequentially at the age of 7 weeks, 12 weeks, 17 weeks and 24 weeks. Subsequently, three groups of mice were used: TTg mice given three doses of TNF-K at 15, 16 and 19 weeks of age, TTg mice treated with infliximab and untreated TTg mice. Then in each group, mice were euthanised at 31 weeks. After each euthanasia, the frequency of Treg and the percentage of Treg cells expressing CTLA-4, TNFR2 and CD62L from the spleen and lymph nodes (LN) were determined. Finally, the immunosuppressive activity of the Treg cells was studied. Results hTNF overexpression induced an initial decline in Treg. Then, once chronic inflammation was established, the frequency of Treg cells increased with time. The authors also observed a progressive and dramatic increase in TNFR2 expression (65.8–88.6% at weeks 7 and 24, respectively, p<0,0001) and MFI on Treg from LN during the course of arthritis. Compared to untreated mice, hTNF blockade with either infliximab or TNF-K resulted in an increased Treg frequency. The study also showed that hTNF blockade induces an upregulation of CTLA-4 expression by Treg cells in LN, accompanied by an increased Treg suppressive activity on CD4+ CD25− effector T cell proliferation. Finally, in LN of infliximab and TNF-K treated TTg mice, the authors observed an expansion of induced Treg cells defined by the CD4+ CD25+ FoxP3+ CD62L− phenotype. Conclusion In this study, in a strictly hTNF dependent inflammation model, the authors show for the first time that TNF can have different effects on Treg, depending on the duration of exposure and on disease phase. The work also shows that TNF blockade either by TNF-K or infliximab could depend not only on TNF neutralisation but also on Treg upregulation.

A7.03 Il-33 inhibits experimental arthritis through CD25+ and CD25- regulatory T cells activation

Background and objectives Interleukin (IL)-33 is a new member of the IL-1 family that exerts pleiotropic activities in innate and adaptive immunity. Some evidence also suggest that the IL-33/ST2 (IL-33 receptor) axis is strongly involved in the pathophysiology of rheumatoid arthritis (RA). We recently showed that IL-33 was able to inhibit collagen-induced arthritis (CIA) by promoting a regulatory T cell (Tregs) population overexpressing CD39 and ST2L. This work aimed at further enlightening the involvement of regulatory T cells in the IL-33-mediated effect in arthritis. Results To further study the implication of Treg in IL-33 therapeutic effect, we evaluated IL-33 mediated effect in the absence of CD25+ Treg in CIA by co-administering IL-33 together with a depleting anti-CD25 antibody. As expected, this antibody obviously suppressed CD25 expressing CD4+FoxP3+ Treg population. In contrast, while CD4+FoxP3+CD25+ Treg were depleted, IL-33 dramatically amplified a population of CD25 negative CD4+FoxP3+ Treg in mice with CIA. We confirmed in an in vitro system this strong impact of IL-33 on CD25- Treg. On the contrary, IL-2 had no impact in vitro on this non-expressing CD25 cells. Conclusion In absence of the alpha chain of the IL-2 receptor (CD25), IL-33 is still able to induce Treg expansion probably independently of IL-2 by a direct effect on ST2L-expressing Treg. Finally, this set of experiments reinforces the strong impact of IL-33 on both CD25+ and CD25- Treg subpopulations, therefore strengthening the involvement of these cells in the therapeutic effect of IL-33 in CIA.

AB0125 Differential Expression of Vascular Inflammation Markers in Collagen-Induced Arthritis and Experimental Atherosclerosis

Background Cardiovascular disease (CVD) is a major cause of morbidity and mortality in rheumatoid arthritis (RA). Increased cardiovascular risk in RA is attributed to both traditional risk factors and systemic inflammation over a long period. Limited experimental data exist on vascular involvement in arthritis models1,2,3. Thus, we developed a model that would mimic both vascular dysfunction and articular inflammation in RA, using collagen-induced arthritis (CIA) in in C57Bl/6 (B6) mice. Objectives To study the expression of vascular inflammatory markers and atheroma formation in CIA with and without concomitant hyperlipidic diet (HD). Methods CIA was induced in B6 mice with chicken type-II collagen (cCII) followed by a boost 21 days later. These mice were fed with a standard (chow) diet or with HD given for 12 weeks starting from the day of the boost. The control mice were not immunized (NI) but were fed either the standard diet or HD. Arthritis severity was evaluated using a validated clinical score. Aorta and synovial membranes were removed 15 weeks after the first cCII immunization. We analysed vascular cell adhesion molecule-1 (VCAM-1), inducible NO-synthase (iNOS) and interleukin-17 mRNA level in aorta with real-time quantitative PCR. VCAM-1 localisation in the aortic sinus was determined by immunohistochemistry. In addition, atherosclerotic plaques were assessed in the aorta. Results CIA was associated with high expression of VCAM-1, independently of the fed diet. VCAM-1 overexpression was detectable 4 weeks after cCII immunization and persisted at 15 weeks. HD induced atheroma plaque formation and aortic iNOS expression which were both independent from CIA. Concomitant CIA and HD had no additive effect on atheroma, VCAM-1 or iNOS expression. Conclusions In B6 mice, HD and CIA induce distinct and independent expression of large vessel inflammation markers. This model, sensitive to both CIA and experimental atherosclerosis may be relevant for the study of CVD in RA. References Asquith, D. L., Miller A. M., Hueber A. J., et al. Apolipoprotein E-deficient mice are resistant to the development of collagen-induced arthritis. Arthritis Rheum 2010.62:472-7. Postigo, J., Genre F., Iglesias M., et al. Exacerbation of type II collagen-induced arthritis in apolipoprotein E-deficient mice in association with the expansion of Th1 and Th17 cells. Arthritis Rheum 2011.63:971-80. Rose, S., Eren M., Murphy S., et al. A novel mouse model that develops spontaneous arthritis and is predisposed towards atherosclerosis. Ann Rheum Dis 2013.72:89-95. Disclosure of Interest None declared

IL-23 specific targeting by active immunisation improves inflammation and destruction in a murine model of rheumatoid arthritis

Interleukin 23 (IL-23) is a pro-inflammatory cytokine known to be essential for the differentiation of the Th17 lymphocytes, a subtype of T lymphocyte implied in auto-immunity. Its subunit, IL-23 p19, is specific of this cytokine. We had previously demonstrated for IL-1β and tumour necrosis factor α, that active immunisation against these cytokines could be protective in animal models of arthritis. The …

In vivo induction of foxp3 in collagen-induced arthritis treatment with modified dendritic cells

We studied the prevention of a mice model of rheumatoid arthritis, the collagen-induced arthritis (CIA) model, with DNA-matured dendritic cell injection and then studied the induction of Treg through this innate immunity stimulation. CIA was induced as described in DBA/1 mice with bovine type II collagen intradermal injections. Intermediate DNA-matured dendritic cells were injected at D14 intraperitoneally. The clinical course of arthritis was followed and B-cell and T-cell responses were assayed. The induction of CD4+CD25+ was tested by flow cytometry and the induction of Treg markers was quantified by quantitative RT-PCR. Treatment of CIA with intermediate DNA-matured dendritic cells could prevent arthritis as well as lipopolysaccharide-matured dendritic cells. Neither B-cell and T-cell responses were not modified nor was a TH2 response observed. The induction of Treg (CD4+CD25+) cells was observed in blood, and lymph nodes. The induction of foxp3 could be quantified and increased with DNA-matured dendritic cells in peripheral nodes. In conclusion we observed a prevention of CIA with the injection of DNA-matured dendritic cells that did not modify the specific response against bovine type II collagen. Because of the absence of T-cell and B-cell response modification as well as TH2 modification, we believed that the induction of CD4+CD25+ cells that expressed foxp3 are involved in the prevention of CIA we observed.

Electrotransfer of low doses of plasmid encoding interleukin-10 in gene therapy of collagen-induced arthritis.

Gene therapy is extremely promising in rheumatoid arthritis (RA). Electrotransfer (ET) is a recent method reported to enhance the in vivo effects of intra-muscular DNA injection. Interleukin-10 (IL-10) has anti-inflammatory effects in RA and in collagen-induced arthritis (CIA), a murine model of RA. We used ET to enhance penetration into skeletal muscle of plasmids encoding IL-10. CIA was induced in DBA/1 mice by immunization with bovine type II collagen. Injection into the tibial cranial muscle of low-dose (200 ng) pCOR plasmid encoding murine IL-10 (pCOR-CMV-mIL-10) was immediately followed by application of square-wave electric pulses (8 pulses of 200 V/cm, 20 ms duration at 2 Hz). Control groups received empty plasmid or saline before ET. When ET was performed twice on days 10 and 25 post-immunization, CIA was significantly delayed (P < 0.05) and attenuated (P < 0.001) in pCOR-CMV-mIL-10 ET groups vs. control groups. When pCOR-CMV-mIL-10 ET was started at disease onset (days 25 and 40), the clinical severity of CIA was reduced (P < 0.05). All groups treated early or late by pCOR-CMV-mIL-10 ET showed dramatic suppression of histological signs of arthritis compared to control groups. Taken together, these data indicate that administration of an anti-inflammatory gene by ET of naked DNA is effective in vivo in an arthritis model in preventive and curative protocols, even when low doses were given.