Dimethylfumarate Induces Immunosuppression via Glutathione Depletion and Subsequent Induction of Heme Oxygenase 1  Joachim C.U. Lehmann, Joanna J. Listopad, Christine U. Rentzsch, Frederik H. Igney, Arne von Bonin, Hartwig H. Hennekes, Khusru Asadullah, Wolf-Dietrich F. Docke  Journal of Investigative Dermatology  Volume 127, Issue 4, Pages 835-845 (April 2007) DOI: 10.1038/sj.jid.5700686 Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 1 DMF and DEF suppress inflammatory cytokine secretion. The influence of FAE on cytokine secretion of primary human peripheral blood immune cells was determined after 24hours culture with different stimuli and FAEs. Cytokine concentrations are shown as percentage (mean±SD) of levels in control cultures. (a) LPS/IFNγ (1μg/ml, 10ng/ml) induced cytokine secretion in human PBMC (n=6). Concentrations in control culture: TNFα 34.1±9.2ng/ml, IL-12p70 28.6±3.7pg/ml. (b) LPS/IFNγ (1μg/ml, 10ng/ml) induced cytokine secretion in separated human monocytes (n=6). Concentrations in control culture: TNFα 28.9±3.1ng/ml, IL-10 0.6±0.4ng/ml, and IL-12p40 4.7±1.4ng/ml. (c) PHA (100μg/ml) induced cytokine secretion in human PBMC (n=6). Concentrations in control culture: IFNγ 3.7±0.3ng/ml, IL-1β 9.7±1.2ng/ml, IL-2 2.8±0.9ng/ml, IL-4 126±23pg/ml, IL-6 53.4±12.3ng/ml, and IL-10 2.8±0.8ng/ml. (d) Anti-CD3/ant-CD28 induced cytokine secretion in purified human T cells (n=5–6). IFNγ 1.5±0.8ng/ml, IL-2 26.8±8.9ng/ml, IL-4 0.3±0.4ng/ml, IL-5 0.4±0.5ng/ml, and IL-10 2.2±1.2ng/ml. **P<0.01 and *P<0.05 versus control culture (100%). Journal of Investigative Dermatology 2007 127, 835-845DOI: (10.1038/sj.jid.5700686) Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 2 DMF and DEF inhibit lymphocyte proliferation in MLR. The influence of FAE on alloreactive lymphocyte proliferation was determined in triplicated cultures in MLR. Responder and allogeneic stimulator PBMCs were cultured for 5 days in the presence of the different FAE at the indicated concentrations. (a) Lymphocyte proliferation was analyzed by 3H-thymidine incorporation for additional 6hours. (b) IL-2 concentrations in the supernatants were determined before the addition of 3H-thymidine. Values from five independent experiments are shown as percentage (mean±SD) of those in control cultures (100%): 3H-thymidine incorporation 6,329±1,230c.p.m., IL-2 concentration 111.7±27.4pg/ml. **P<0.01 and *P<0.05 versus control culture (100%). Journal of Investigative Dermatology 2007 127, 835-845DOI: (10.1038/sj.jid.5700686) Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 3 DMF induces HO-1 expression. Expression of the anti-inflammatory stress protein HO-1 was analyzed by quantitative real-time RT-PCR in peripheral blood immune cells treated with FAE. (a) HO-1 mRNA expression was determined in PBMC cultured with the different FAE for 4hours. (b) Kinetics of HO-1 mRNA expression was determined in PBMC treated with 25μM DMF for the indicated times. (c) To clarify the responsible cell type for HO-1 induction, HO-1 mRNA expression was determined after 4hours of culture of negatively separated monocytes and T cells with different concentrations of DMF. (d) To establish GSH depletion as mechanism of DMF-induced HO-1 induction, HO-1 mRNA expression was determined in PBMC treated with 25μM DMF with or without addition of 1mM GSH-OEt. 1,3-Bis-2-chloroethyl-1-nitrosourea (GSH depletor) and CoPP (cobaltic protoporphyrin IX, GSH-independent HO-1 inducer) at the indicated concentrations served as controls. Values are shown as mean±SD from three independent experiments. Journal of Investigative Dermatology 2007 127, 835-845DOI: (10.1038/sj.jid.5700686) Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 4 Substitution of GSH abolishes the inhibitory effects of DMF on inflammatory cytokine secretion and alloreactive lymphocyte proliferation. Exogenous GSH-OEt substitution was used to verify whether GSH depletion is responsible for the immunosuppressive in vitro effects of DMF. (a, b) PBMCs were stimulated with LPS/IFNγ (1μg/ml, 10ng/ml) or PHA (100μg/ml) in the presence of (a) DMF and (b) DEF with or without addition of 1mM GSH-OEt. Cytokine concentrations in the supernatants were determined after 24hours). Values from two independent experiments (one experiment for IL-6) are given as percentage (mean±SD) of the levels in control cultures (100%) for LPS stimulation (TNFα 51.0±14.8ng/ml, IL-10 3.0±1.4ng/ml, and IL-12p70 473.5±175.7pg/ml) and PHA stimulation (IFNγ 4.1±0.4ng/ml, IL-1β 7.9±0.7ng/ml, and IL-6 49.1ng/ml). (c, d) Five-day MLR with responder and allogeneic stimulator PBMC was performed in the presence of (c) DMF and (d) CoPP, respectively, with or without 0.5mM GSH-OEt substitution. Lymphocyte proliferation was assessed by 3H-thymidine incorporation during the last 6hours. Values from five independent experiments are shown as percentage (mean±SD) from those in control cultures (100%): 3H-thymidine incorporation 30,780±10,714c.p.m. **P<0.01 and *P<0.05 versus control culture (100%). Journal of Investigative Dermatology 2007 127, 835-845DOI: (10.1038/sj.jid.5700686) Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 5 HO-1 antagonization counteracts inhibition of IL-12 and IFNγ secretion by DMF and DEF. (a) Kinetics of HO-1 knockdown by siRNA in unstimulated and LPS (100ng/ml) stimulated THP-1 cells (one representative out of two independent experiments). (b) Effect of HO-1 knockdown on inhibition of IL-12p40 secretion by DMF and DEF in LPS (100ng/ml) stimulated THP-1 cells after 24hours culture. IL-12p40 concentrations are given as percentage (mean±SD of two independent experiments) of the levels in control culture (100%): IL-12p40 605±669pg/ml. (c) Effect of HO-1 inhibition by SnPPIX (50μM) on inhibition of IL-12p40 and TNFα secretion by DMF and DEF in LPS (100ng/ml) stimulated PBMC after 6hours culture. Cytokine concentrations are given as mean±SD of six independent experiments. **P<0.01 versus similarly treated cultures without SnPPIX. (d) Effect of HO-1 inhibition by SnPPIX (50μM) on inhibition of IFNγ secretion by DMF and DEF in PHA (100μM) stimulated PBMC after 6hours culture. IFNγ concentrations are given as mean±SD of six independent experiments. **P<0.01 versus cultures without SnPPIX. Journal of Investigative Dermatology 2007 127, 835-845DOI: (10.1038/sj.jid.5700686) Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 6 Substitution of GSH abolishes the casual stimulatory effect of low-dose DMF on TNFα secretion. (a) In some experiments monocytic endotoxin-induced TNFα secretion was increased by treatment with low concentrations of DMF. PBMCs were stimulated with high doses of LPS (1μg/ml) and IFNγ (10ng/ml) in the presence of DMF at the indicated concentrations. Cytokine concentrations were determined in pooled supernatants of duplicated cultures after 24hours. Data from one representative out of two independent experiments are shown. This effect may be an equivalent to the transient increase in TNFα plasma levels, which have been observed in psoriasis patients during the clinical treatment with FAE. (b) Data from the same experiments are shown for treatment with 10μM DMF or DEF with or without substitution of 1mM GSH-OEt. Data from two independent experiments are shown as percentage (mean±SD) of values in the control cultures (100%): TNFα 33.0±4.7ng/ml. Journal of Investigative Dermatology 2007 127, 835-845DOI: (10.1038/sj.jid.5700686) Copyright © 2007 The Society for Investigative Dermatology, Inc Terms and Conditions