Eur. J. Immunol. 1996.26: 2067-2074
`
`Rolien de Jong',
`Anja C. Bezeme8,
`Tim0 P. L. Zomerdijk',
`Tineke van de Pouw-Kraan3,
`Tom H. M. Ottenhoff' and
`Peter H. Nibbering
`' Department of
`Immunohematology and
`Bloodbank, University Hospital
`Leiden, Leiden, The Netherlands
`* Department of Infectious
`Diseases, University Hospital,
`Leiden, The Netherlands
`Department of Autoimmune
`Diseases, Central Laboratory of
`the Netherlands Red Cross Blood
`Transfusion Service and
`Laboratory for Experimental and
`Clinical Immunology, University
`of Amsterdam, The Netherlands
`
`Monomethylfumarate enhances Th2 cytokine secretion
`Selective stimulation of T helper 2 cytokine
`responses by the anti-psoriasis agent
`monometh ylfumarat e
`
`2067
`
`Type 2 cytokines are thought to have a protective role in psoriasis vulgaris by
`dampening the activity of T helper 1 (Thl) lymphocytes. The aim of the present
`study was to determine the effect of monomethylfumarate (MMF), the most
`active metabolite of the new anti-psoriatic drug Fumaderm@, on the production
`of cytokines and the development of Th subsets. MMF was found to enhance
`interleukin (1L)-4 and IL-5 production by CD2/CD8 monoclonal antibody-
`stimulated peripheral blood mononuclear cells (PBMC) in a dose-dependent
`manner. Maximal effects of MMF were found at a concentration of 200 pM and
`resulted in tenfold enhanced levels of IL-4 and IL-5 production. MMF did not
`affect the levels of IL-2 production, interferon (1FN)-y production or prolifera-
`tive T cell responses in these cultures. Similar effects of MMF were observed in
`cultures of purified peripheral blood T cells indicating that this compound can
`act directly on T cells. MMF did not influence cytokine production by purified
`CD4+CD45RAC (unprimed) T cells, but greatly enhanced IL-4 and IL-5 produc-
`tion without affecting IFN-y production by purified CD4+CD45RO+ (primed) T
`cells. Furthermore, MMF also augmented IL-4 and IL-5 production in estab-
`lished Thl/ThO clones that were stimulated with CD2/CD28 monoclonal anti-
`body. Finally, when PBMC were challenged with Mycobacterium tuberculosis
`that typically induces Thl recall responses with strong IFN-y secretion, MMF
`again appeared to induce high levels of IL-4 and IL-5 secretion while IFN-y pro-
`duction was unaffected. These results may be relevant for the development of
`therapeutic regimens designed to correct inappropriate Thl subset development
`in immunopathologic conditions.
`
`1 Introduction
`
`CD4' T helper (Th) cells can be classified in at least three
`subsets according to their functional program. Thl cells
`typically secrete high amounts of IL-2 and IFN-y whereas
`Th2 cells are characterized by their ability to produce high
`levels of IL-4 and IL-5 upon activation [l, 21. Th cells with
`shared features of Thl and Th2 cells are referred to as Tho
`cells [2]. Protective immunity towards intracellular patho-
`gens such as Listeria monocytogenes, Leishmania major
`and Mycobacterium species depends on a Thl response,
`whileTh2 cells promote disease [3,4]. In several models of
`organ-specific autoimmune diseases, on the other hand,
`Thl responses promote disease development whereas Th2
`responses are associated with disease resistance [5-71. The
`
`[I 154761
`
`Received February 29, 1996; in revised form June 5, 1996;
`accepted June 12, 1996.
`
`Correspondence: Rolien de Jong, Department of Immunohe-
`matology and Bloodbank, Building 1, E3-Q, University Hospital
`Leiden, P.O. Box 9600,2300 RC Leiden, The Netherlands
`Fax: +31-71-5216751
`
`Abbreviations: D M F Dimethylfumarate M E F Monoethylfu-
`marate MMF: Monomethylfumarate
`Key words: Immunomodulation I Cytokine I Thl and Th2 cell /
`Psoriasis
`
`0 VCH Verlagsgesellschaft mbH, D-6945 1 Weinheim, 1996
`
`most important factors that control Th cell differentiation
`are IL-12 and IL-4 that drive Thl and Th2 cell develop-
`ment, respectively [3, 8, 91. The products of Thl and Th2
`cells have a negative regulatory effect on the growth and
`function of the opposing subset, which may lead to a rein-
`forcement of polarizing differentiation events, particularly
`in chronic responses [lo, 111.
`Psoriasis vulgaris is characterized by epidermal hyperpla-
`sia with cellular infiltrates of lymphocytes and monocytes
`[12]. Psoriasis is presumed to be a Thl-associated autoim-
`mune disease based on a preferential isolation of Thl cells
`from psoriatic skin lesions [13, 141. Moreover, IFN-y could
`be localized in suction blister fluid from psoriatic lesions
`[15] and significantly higher levels of IFN-y were found in
`the serum of psoriasis patients than in the serum of healthy
`controls [16]. Clinical efficacy of cyclosporin A, CD4 mAb
`and IL-2 toxin in psoriasis treatment has been well estab-
`lished [12, 17, 181. In addition, clinical improvement in
`psoriasis vulgaris and psoriatic arthritis patients has also
`been observed upon therapy with Fumaderm@ [ 19-22].
`Fumaderm@ is composed of monoethylfumarate (MEF)
`and dimethylfumarate (DMF). The most effective fuma-
`rate metabolite of
`this drug is monomethylfumarate
`(MMF), which is formed in the circulation by hydrolysis of
`DMF [23]; (personal communication Dr. R. K. Joshi,
`ETH Dept. of Pharmacy, Zurich, Switzerland). Consider-
`ing the possibility that these compounds may act in psoria-
`sis by modulating ThliTh2 subset development, we ana-
`lyzed the ability of MMF to redirect T cell cytokine secre-
`tion profiles. Our data indicate that MMF is able select-
`ively to up-regulate the secretion of cytokines that are
`associated with type 2 immune responses.
`0014-298019610909-2067$10.00 + .25/0
`
`Sawai (IPR2019-00789), Ex. 1023, p. 001
`
`

`

`R. de Jong et al.
`2068
`2 Materials and methods
`
`2.1 Reagents
`
`mAb directed against CD2 (CLB-T11.1/1, CLB-T11.211 and
`CLB-CD2 clone HIK27, all IgGl), CD28 (CLB-CD2811,
`IgGl), CD3 (CLB-T3/4.E, IgE), CD8 (CLB-T8/4), CD19
`(CLB-CD19), CD14 (CLB-CD14) and CD16 (CLB-FcR-
`granl) were produced at the Central Laboratory of the
`Red Cross Blood Transfusion Service (CLB), Amsterdam,
`The Netherlands. CD45RA (2H4) mAb was purchased
`from Coulter Immunology (Hialeah, FL). CD45RO mAb
`was generously provided by Dr. P. Beverly (University
`College, London, GB). MMF (Fumapharm AG, Muri,
`Switzerland) was prepared as a stock solution of 10 mM in
`PBS; the pH was adjusted to 7.2 with 0.1 N NaOH. The
`stock was stored at 4" and diluted appropiately before use.
`Purified protein derivative (PPD) of Mycobacterium tuber-
`culosis was obtained from Statens Seruminstitute, Copen-
`hagen, Denmark.
`
`2.2 Tcells
`
`Peripheral blood was obtained from buffy coats of healthy
`blood bank donors. PBMC were isolated from heparinized
`blood by Ficoll-Amidotrizoate density centrifugation.
`Purified peripheral blood T cells were isolated after rosette
`aminoethylisothiouronium
`formation with
`bromide
`hydrobromide-treated sheep red blood cells followed by
`Ficoll-Amidotrizoate centrifugation. After lysis of sheep
`red blood cells, T cell fractions consisting of > 98 % CD2'
`cells were obtained. For the experiments on CD45R sub-
`sets, lymphocytes were isolated from PBMC by counter-
`flow centrifugation elutriation. CD4' T cell subsets were
`subsequently prepared by negative magnetic bead immu-
`noselection. Cells were incubated for 45 min at 4°C with
`saturating amounts of CD8, CD14, CD19, CD16 and
`CD45RA or CD45RO mAb. After two washes, sheep anti-
`mouse Ig-coated magnetic beads were added (Dynabeads
`M450, Dynal A.S., Oslo, Norway). After a 60-min incuba-
`tion at 4"C, rosetted cells were removed with a Dynal
`magnetic particle concentrator. The negatively selected
`fractions were >97% CD4+, < 2 % CD8+ and did not
`contain detectable levels of CD 14-expressing cells. The
`resulting purity for CD45RO and CD45RA antigen expres-
`sion was > 93 % and 95 % , respectively. The CD4+ T cell
`clones that were used in this study are specific for Myco-
`bacterium leprue and represent Thl or Tho type clones [24,
`251.
`
`2.3 Detection of binding sites for methylated fumarates
`
`Investigation of the presence of binding sites for methyl-
`ated fumarates was performed as described [26] with
`minor modifications. Briefly, approximately 1 x lo6 puri-
`fied T cells (> 98 '30 CD2+)/100 pl RPMI 1640 supple-
`mented with 20 mM Hepes (pH 7.4) and 10 mg BSA/ml
`were incubated with 49 pM ["C] DMF (specific activity 8.1
`Ci/mol, Hazleton, Harrogate, GB; a kind gift from Dr. R.
`K. Joshi) and various concentrations of unlabeled MMF or
`DME After 90 min at O"C, 1 ml cold RPMI 1640 supple-
`mented with 20 mM Hepes pH 7.4 and 10 mg/ml BSA/ml
`was added and the cells were centrifuged at 160 x g for
`
`Eur. J. Immunol. 1996.26: 2067-2074
`
`10 min. After washing twice, the cells were transferred for
`scintillation counting. From these data, the concentration
`of MMF causing 50 YO inhibition (IDSo) of the binding of
`["C] DMF to T cells was calculated. The specific activity of
`the radiolabeled methylated fumarates available is not
`high enough to perform calculations on the number of
`binding sites per T cell and the binding affinity of these
`sites for MMF.
`
`2.4 T cell stimulation
`
`PBMC were plated at 7.5 x lo5 cells/ml, purified periph-
`eral blood Tcells at 5 x lo" cells/ml and CD4+ T cell clones
`at lo6 cells/ml in 96-well plates (200 pYwell). Culture me-
`dium consisted of Iscove's Modified Dulbecco's medium
`(IMDM, Gibco BRL, Scotland) supplemented with 10 %
`FCS and antibiotics. Triplicate cultures of T cells were sti-
`mulated with a triplet of CD2 mAb (CLB-T11.1/1, CLB-
`T11.2/1 and CLB-CD2 clone HIK27) and CD28 mAb
`(CLB-CD28/1). In some experiments, a combination of
`CD28 mAb and CD3 mAb (CLB-T3/4.E) was used for
`stimulation. The mAb were used at a concentration of
`5 pg/ml and have been previously described to induce high
`levels of cytokine secretion [27, 281. Cells cultured in
`medium alone were included as negative controls. MMF
`was added at the initiation of the cultures in concentrations
`ranging from 50-400 pM. Culture supernatants were col-
`lected at 24 h to measure IL-2 secretion and at 72 h to
`quantitate IL-4, IL-5 and IFN-y production. These inter-
`vals were found to be optimal for the induction of the
`respective cytokines by T cells.
`
`For induction of recall antigen responses to M. tuberculo-
`sis, PBMC were stimulated with 5 pg/ml PPD either in the
`presence or absence of MMF. After 6 days, 20 U/ml rIL-2
`(Eurocetus, Amsterdam, The Netherlands) was added. At
`day 13, cells were washed and restimulated (7.5 x lo5
`cells/ml) with CD2 mAb and CD28 mAb. Supernatants
`were harvested for cytokine analysis 3 days later.
`
`2.5 T cell proliferation assay
`
`In parallel with the cultures that were set up for the mea-
`surement of cytokine secretion, T cells were plated for the
`analysis of proliferative responses. On day 4, 0.5 pCi
`[3H] thymidine ([3H] dThd) (6.7 Ci/mmol, Du Pont de
`Nemours Nederland B.C., Dordrecht, The Netherlands)
`was added to each well. After a 16-h pulse, incorporation
`was determined by liquid scintillation counting. Prolifera-
`tion is expressed as mean cpm of triplicate cultures. The
`SD between the triplicates was < 10 YO.
`
`2.6 Cytokme measurements
`
`IL-2 production was analyzed by bioassay using the IL-2-
`dependent CTLL-2 line [29]. Half-maximal proliferation
`in CTLL cells is induced with 1 U/ml rIL-2 (Eurocetus).
`IL-4, IL-5 and IFN-y production were assayed by ELISA.
`For detection of IL-4, flat-bottom microtiter plates (Maxi-
`sorb, Nunc, Roskilde, Denmark) were coated overnight
`using CLB-IL4I5 (CLB, Amsterdam, The Netherlands)
`[28]. Following incubation with serial dilutions of culture
`
`Sawai (IPR2019-00789), Ex. 1023, p. 002
`
`

`

`Eur. J. Immunol. 1996.26: 2067-2074
`
`supernatants and rIL-4 as standard, biotinylated CLB-IL4/
`1 (CLB) was added as a secondary mAb. Thereafter, plates
`were washed and incubated with horseradish peroxidase-
`conjugated streptavidin (Pierce, Rockford, IL) for 30 min
`washed
`and developed with 100 pg/ml 3,5,3',5'-
`tetramethylbenzidine
`(Merck, Darmstadt, Germany),
`0.003 % H202 in 0.11 M sodium acetate pH 5.5. The reac-
`tion was stopped by addition of 50 pl of 2 M H2S04/well.
`Plates were read at 450 nm in a Titertek microplate reader.
`ELISA for the other cytokines were performed in a similar
`manner. IL-5 production was determined using TRFK5
`mAb as a primary mAb and biotinylated JES1-5A10 mAb
`as capture antibody (both from Pharmingen, San Diego,
`CA). For detection of IFN-y, plates were coated with
`4SB3 mAb (European Collection of Animal Cell Cultures)
`and biotinylated MD-1 mAb was used as a secondary anti-
`IFN-y mAb [30]. One U/ml IFN-y was equal to 23 pg/ml.
`The detection level of the assays is 0.5 U/ml for IL-2,
`60 pg/ml for IL-4, 0.5 ng/ml for IL-5 and 100 U/ml for
`IFN-y.
`
`3 Results
`
`3.1 Effects of MMF on lymphocyte prolieration and
`cytokine production
`
`The presence of binding sites for MMF on purified T cells
`was inferred from the ability of both unlabeled MMF and
`DMF to compete equally well with ["C] DMF for binding
`to peripheral blood T cells in a dose-dependent manner
`(IDSo-6 mM MMF). To study the functional effects of
`MMF on T lymphocytes, a first series of experiments were
`performed in which PBMC were stimulated with a com-
`bination of CD2 mAb and CD28 mAb in the presence of
`titrated amounts of MMF. MMF in concentrations
`
`Monomethylfumarate enhances Th2 cytokine secretion
`
`2069
`
`5200 pM did not affect proliferative responses. Since
`400 pM MMF largely
`inhibited T cell proliferation
`(Fig. lA), in line with toxic concentrations of MMF for
`phagocytes and keratinocytes [23, 311, all further experi-
`ments were performed in concentrations of MMF ranging
`from 50-200 pM. MMF stimulated both IL-4 and IL-5
`production by activated T cells in a dose-dependent man-
`ner (Fig. 1D and E). Maximal effects were obtained using
`a concentration of 200 pM MMF and resulted in tenfold
`increased levels of both IL-4 and IL-5 production. Neither
`IL-2 nor IFN-y production was modulated by MMF
`(Fig. 1B and C). Similar findings were obtained when
`CD3 mAb instead of CD2 mAb were used for stimulation
`(data not shown). MMF by itself did not stimulate the
`release of these cytokines (Table 1 and 2).
`
`3.2 Kinetics of IL-4 and IL-5 release in the presence of
`MMF
`
`The levels of IL-4 and IL-5 secretion in the presence of
`MMF were analyzed at different time points after T cell
`stimulation. PBMC were stimulated with CD2/CD28 mAb
`in the presence of 100 pM and 200 pM MMF. MMF caused
`an increase in the levels of IL-4 and IL-5 production within
`2 days after T cell activation (Fig. 2). The enhancing
`effects of MMF became more evident at later time points
`and lasted throughout the 5- day culture period (Fig. 2).
`Peak levels in IL-4 were reached at day 3-4 and the
`highest levels in IL-5 at day 5.
`
`3.3 Regulatory effects of MMF on purified peripheral
`blood T cells
`
`We examined whether MMF can also affect T cell cytokine
`production in the absence of accessory cells. For this pur-
`
`A
`
`
`
`O-
`
`B
`
`0-
`
`C
`
`I loo
`
`
`
`;
`
`200
`
`0
`
`,
`E
`W
`2 w 4 0 0 6 w 8 o o l M K )
`11-2 U/ml
`
` ,
`
`z loo
`
`200
`
`0
`
`
`
`,
`,
`,
`20
`40
`60
`100
`80
`IFN-r U/ml (thousands)
`
`,
`
`120
`
`I200
`
`& I F ,
`
`400
`
`0
`
`20
`
`60 80
`1 w 120 140
`40
`cpm (thousands)
`
`200
`
`0
`
`5w
`
`1500
`
`MM)
`
`0
`
`50
`
`100
`200
`150
`1wo
`IL-5 ng/ml
`IL-4 p g h l
`Figure 1. MMF enhances IL-4 and IL-5 but not IL-2 and IFN-y production in PBMC. PBMC (7.5 X 10S/ml) were stimulated with CD2
`mAb (CLB-T11.1/1, CLB-T11.2/1 and CLB-CD2 clone HIK27) in combination with CD28 mAb (CLB-CD28/1). MMF was included at
`the indicated concentrations. Proliferative responses (['HI dThd incorporation; in cpm) were assayed at day 5. Supernatants were har-
`vested from parallel cultures 24 h after stimulation for detection of IL-2, or 72 h after stimulation for the measurement of IFN-y, IL-4
`and IL-5 secretion. Nonstimulated cultures contained undetectable levels of cytokines and incorporated < 200 cpm. The given data are
`representative for one out of three experiments.
`
`250 3w
`
`Sawai (IPR2019-00789), Ex. 1023, p. 003
`
`

`

`R. de Jong et al.
`
`2070
`pose, purified T cells were isolated from PBMC by E-
`rosette sedimentation. In line with the results on PBMC,
`about tenfold increased levels of IL-4 production were
`measured when purified T cells were stimulated in the
`presence of 200 pM MMF (Table 1). MMF also enhanced
`IL-5 production in CD2/CD28 mAb stimulated T cells,
`although less vigorously than in PBMC (stimulation indi-
`ces two- to threefold versus tenfold in PBMC, Table 1).
`Again, MMF did not induce cytokine secretion in unsti-
`mulated cultures. Comparable results were obtained when
`T cells were stimulated with either two or three distinct
`CD2 mAb in combination with CD28 mAb or when T cells
`were stimulated with CD3 mAb in combination with CD28
`mAb (Table 1). MMF did not influence the magnitude of T
`cell proliferative responses (Table 1).
`
`Eur. J. Imrnuool. 1996.26: 2067-2074
`
`3.4 MMF enhances IL-4 and IL-5 production in
`CD4+CD45RO+, but not in CD4*CD45RA+ T cells
`
`In peripheral blood, unprimed and primed CD4+ T cells
`can be discriminated on the basis of expression of
`CD45RA and CD45RO membrane molecules, respectively
`[32]. Functionally, these two subpopulations can be dis-
`tinguished in that CD4+CD45RO+ (primed) T cells are the
`main producers of IL-4, IL-5 and IFN-y upon T cell stimu-
`lation in vitro [32, 331. We studied whether MMF modula-
`tes cytokine secretion by CD4+CD45ROt T cells by
`CD4+CD45RA+ Tcells, or both. As shown in Fig. 3, addi-
`tion of MMF
`to CD2/CD28 mAb
`stimulated
`CD4+CD45RO+ T cells led to increased levels of IL-4
`(mean enhancement 2.5-fold, n = 3) and IL-5 (mean
`
`Table 1. Regulatory effects of MMF on purified peripheral blood T cells
`
`Readout"
`
`Stimulation")
`
`MMF~)
`
`Proliferation (cpm x
`
`IL-4 production (pg/ml)
`
`IL-5 production (ng/ml)
`
`100 pM
`200 pM
`0.3
`0.3
`Medium
`CD2 mAb (double) + CD28 mAb
`79
`80
`CD2 mAb (triple) + CD28 mAb
`91
`90
`CD3 mAb + CD28 mAb
`79
`77
`< 60
`<60
`Medium
`CD2 mAb (double) + CD28 mAb
`1912
`3546
`CD2 mAb (triple) + CD28 mAb
`1960
`3302
`CD3 mAb + CD28 mAb
`818
`2960
`< 1
`< 1
`Medium
`CD2 mAb (double) + CD28 mAb
`454
`563
`CD2 mAb (triple) + CD28 mAb
`280
`390
`CD3 mAb + CD28 mAb
`493
`586
`a) PurifiedTcells (5 X lO'/ml) were stimulated with either two CD2 mAb (CLB-T11.111 and CLB-T11.2/1) or three distinct CD2 mAb
`(CLB-T11.111, CLB-T11.2/1 and CLB-CD2 clone HIK27) in combination with CD28 mAb (CLB-CD28/1). Alternatively, cells were
`stimulated with CD3 mAb (CLB-T3/4.E) in combination with CD28 mAb (CLB-CD28/1). Unstimulated cells were included as a
`control.
`b) MMF was added at the initiation of the culture in concentrations as indicated.
`c) T cell proliferation was assayed by [3H] thymidine incorporation; IL-4 and IL-5 production was assayed by ELISA.
`
`-
`0.3
`69
`87
`71
`< 60
`458
`514
`184
`< 1
`194
`142
`313
`
`50 pM
`0.2
`77
`91
`80
`< 60
`940
`1026
`682
`< 1
`334
`249
`474
`
`Table 2. IL-4 and IL-5 production by CD4+ T cell clones in the presence of MMF
`
`T cell clonea)
`SCII-1G6
`
`SCVIII-2D3
`
`SCVIII-2D6
`
`R1F9
`
`SCIX-1F6
`
`SCI- 1 G2
`
`MMF pMh)
`0
`100
`200
`0
`100
`200
`0
`100
`200
`0
`100
`200
`0
`100
`200
`0
`100
`200
`
`Medium
`< 60"
`< 60
`< 60
`< 60
`< 60
`< 60
`< 60
`< 60
`<60
`< 60
`< 60
`< 60
`nt
`< 60
`< 60
`nt
`< 60
`< 60
`
`IL-4 pg/ml
`
`CD2/28
`120
`450
`600
`60
`160
`320
`60
`280
`520
`60
`120
`280
`160
`440
`840
`960
`1320
`1400
`
`Medium
`1"
`2
`1
`<1
`< 1
`< 1
`1
`< 1
`<1
`ntd)
`nt
`nt
`< 1
`< 1
`< 1
`< 1
`< 1
`< 1
`
`IL-5 ng/ml
`
`~~~
`
`CD2/28
`~ 76
`91
`90
`18
`28
`51
`32
`65
`80
`62
`99
`104
`40
`48
`44
`396
`368
`516
`
`a) CD4' T cell clones were cultured for 3 days with medium or CD2 mAb and CD28 mAb.
`b) MMF was added at the initiation of the culture at the indicated concentration.
`c) Data represent the mean cytokine contents of triplicate cultures measured by specific ELISA.
`d) nt: not tested.
`
`Sawai (IPR2019-00789), Ex. 1023, p. 004
`
`

`

`Eur. J. Immunol. 1996.26: 2067-2074
`
`2000 r
`
`1000
`
`600
`
`r
`
`-
`
`5 400
`
`In
`-
`I
`-I 200
`
`P
`I
`I
`I
`I
`I
`I
`I
`/
`d'
`I
`i
`
`0
`1 2 3 4 5
`days
`
`0
`1 2 3 4 5
`days
`
`Figure 2. Kinetics of IL-4 and IL-5 production in PBMC stimu-
`lated with CD2/CD28 mAb and MMF. PBMC (7.5 X l@/ml)
`were stimulated with CDUCD28 mAb in the absence (-0-) or
`presence of 100 pM (MMF (-V-)
`and 200 pM MMF (-0-).
`Supernatants were assayed for IL-4 and IL-5 production by
`ELISA at the indicated time points after stimulation.
`
`enhancement 3-fold, n = 3). MMF again did not affect the
`magnitude of the IFN-y response (Fig. 3), nor did it
`induce detectable levels of any cytokine in the absence of
`T cell stimuli. As expected, CD4+CD45RA+ T cells se-
`creted little or no IL-4 and IL-5 and only low levels of
`IFN-y following stimulation with CD2/CD28 mAb. This
`cytokine secretion pattern was not modified by MMF
`(Fig. 3). These data show that MMF exerts its regulatory
`effects (mainly) on the CD4+CD45RO+ T cell population.
`
`Monomethylfumarate enhances Th2 cytokine secretion
`
`2071
`
`tested the modulatory effects of MMF on established
`CD4' Thl clones that had frequently been stimulated with
`antigen over time. Thl (low IL-4/low IL-S/high IFNy pro-
`ducers) CD4+ T cell clones and a Tho (high IL4high IL-51
`high IFN-y producer) CD4+ T cell clone that we have
`described previously [24, 251 were activated with CD2/
`CD28 mAb in the absence or presence of MMF for 3 days.
`MMF enhanced IL-4 production in all CD4' T cell clones
`tested (mean enhancement about fivefold at 200 pM
`MMF, n = 6, Table 2). Enhanced levels of IL-5 secretion
`after addition of MMF were seen in five out of six clones
`(mean enhancement about twofold in these clones at
`200 pM MMF) (Table 2).
`
`3.6 MMF modulates T cell responses to recall antigen
`
`Finally, to analyze the modulatory effects of MMF on
`recall antigen responses, representing physiologically
`integrated T celYAPC interactions, PBMC were stimulated
`with PPD of Mycobacterium tuberculosis in the presence
`or absence of 100 or 200 pM MMF. T cell blasts were
`expanded by adding 20 U/ml rIL-2 at day 6. At day 13, T
`cells were harvested, washed and restimulated with CD2/
`CD28 mAb for 3 days in the absence of MME CD2/CD28
`mAb induced equally high levels of proliferation in all
`populations, showing that MMF had not affected the abil-
`ity of T cells to proliferate as expected (Fig. 4). PPD has
`been documented to mount typically Thl responses [24].
`Indeed, the generation of antigen-specific T cells in the
`absence of MMF resulted in a population of T cells that
`secreted high levels of IFN-y but little or no detectable
`IL-4 and IL-5 upon restimulation (Fig. 4). The antigen-
`specific T cells that were generated in the presence of
`MMF, in contrast, produced vast amounts of IL-4 and
`moderate quantities of IL-5 upon activation, while their
`ability to secrete IFN-y was unaffected by the MMF treat-
`ment (Fig. 4.) These findings are in agreement with our
`above observations and indicate that MMF resets the Th
`subset balance during antigen-specific memory T cell
`responses.
`
`3.5 MMF influences cytokine secretion by Thl CD4+ T
`cell clones
`
`4 Discussion
`
`The findings described above show the modulatory capac-
`ity of MMF to stimulate type 2 cytokines in polyclonal T
`cell populations of primed cells. Next, to investigate
`whether MMF is capable of modulating the T cell cytokine
`secretion profile of (virtually) fully differentiated cells
`away from their Thl phenotype at the clonal level, we
`
`The present results demonstrate that the anti-psoriatic
`agent MMF can selectively up-regulate IL-4 and IL-5
`secretion by human T cells. This modulatory effect of
`MMF was observed in cultures of PBMC, purified T cells,
`primed peripheral blood CD4+CD45RO+ T cells, recall
`responses to M. tuberculosis and, interestingly, even in
`
`IL-5 ng/ml
`IFN U/ml (thousands)
`11-4 pg/rnl
`Figure 3. MMF increases IL-4 and IL-5 production in CD4*CD45RO+ T cells and not in CD4+CD45RA+ T cells. Purified T cell subsets
`(5 X lo5 cells/ml) were stimulated with CDUCD28 mAb for 3 days. Supernatants were assayed for their cytokine content by ELISA.
`
`Sawai (IPR2019-00789), Ex. 1023, p. 005
`
`

`

`2072
`
`R. de Jong et al.
`
`established Mycobacterium-reactive Thl CD4' T cell clo-
`nes. These results indicate that the regulatory effects of
`MMF result from a direct effect on T cell function which is
`not influenced by the presence of APC, and provide a
`mechanistic explanation for the beneficial effects of MMF
`in psoriasis, a presumably Thl-associated autoimmune
`skin disorder [ 19-21].
`
`The MMF-mediated enhancement in IL-4 and IL-5 secre-
`tion is not established at the cost of IFN-y secretion. This
`suggests that the promoting effect of MMF on type 2 cyto-
`kine secretion is not mediated indirectly, i.e. via an inhibi-
`tory effect on type 1 cytokine secretion and a subsequent
`rescue of Th2 cells. In line with this view, it is demon-
`strated that MMF skews the typically Thl response that is
`induced by the recall antigen PPD towards a Tho pheno-
`type with a mixed cytokine secretion pattern. Interestingly,
`the immunomodulatory effect on T cell cytokine secretion
`profile is observed in established Thl T cell clones as well
`as in purified primed peripheral blood CD4TD45RO' T
`cells, but not in naive CD4+CD45RAC T cells. These
`results indicate that MMF does not initiate IL-4 and IL-5
`production by nonproducing naive T cells, but rather indu-
`ces or enhances the production of these cytokines in vivo
`by differentiated T cells. These results may be surprising at
`first sight, since modulation of established T cell responses
`by cytokines or soluble factors has been suggested to be
`more difficult than modulation of primary T cell responses
`[34]. However, Perez et al. [35] described that established
`mouse Thl cells can still be converted into IL-4 producers
`upon exposure to IL-4, whereas the Th2 phenotype could
`not be altered under the influence of cytokines. Our data
`further support the view that differentiated Thl cells can
`still be modulated and are not irreversibly committed.
`Importantly, this opens avenues for the design of immune
`intervention based theraDeutic strategies directed at alter-
`ing the cytokine balance:
`
`Eur. J. Immunol. 1996.26: 2067-2074
`
`So far, MMF seems more potent in stimulating IL-4 and
`IL-5 production than other factors, including glucocorti-
`coids, progesterone, retinoic acid and PGEz [36-401. It is
`important to understand the mechanism underlying the
`effect of MMF on cytokine production by T cells. Recent
`studies in granulocytes indicate that these cells express
`binding sites for methylated fumarates [23]. Interaction of
`MMF with these binding sites initiates a signal transduc-
`tion pathway comprising pertussis toxin-sensitive G prote-
`ins, protein tyrosine kinase activity, protein kinase H4
`activity as well as an increase in [Ca2+], and [CAMP],, but
`not phospholipase C and D activation ([23] and P.H. Nib-
`bering et al., submitted for publication). In addition,
`MMF increases [Ca2+Ii and [CAMP], in keratinocytes [31].
`Since specific membrane binding sites for methylated
`fumarate compounds are also expressed by T lymphocytes,
`it can be speculated that a similar pathway is employed by
`Tcells. In agreement with this notion, up-regulation of IL-
`4 and IL-5 production has been shown to depend on an
`increase in [CAMP], and the activation of protein kinase A
`and not on the activation of protein kinase C [41].
`
`The balance between Thl- and Th2-associated cytokines
`has been shown to affect the course of various immune dis-
`orders [l, 21. An immunopathologic role of polarized Thl
`responses
`has been proposed
`in
`organ-specific
`autoimmune diseases like experimental allergic encepha-
`lomyelitis (EAE) and type 1 insulin-dependent diabetes [3,
`6,421. Th2 cells may prevent the development of disease by
`down-regulating the activities of Thl cells [6, 71. In an
`adoptive model of mouse EAE, Khoruts et al. [43] ele-
`gantly demonstrated that even subtle changes in cytokine
`secretion patterns may have significant consequences on the
`development of T cell responses. In these studies, it was
`shown that whereas Thl type cells are highly pathogenic
`and Th2 cells do not cause EAE, incompletely skewed Tho
`cell populations that produce both type 1 and type 2 cytoki-
`
`A
`
`0
`
`cpm (thousands)
`
`IL-4 pglml
`
`e
`
`D
`
`PPO
`
`P P O m p M MMF
`
`P P O m v M MMF
`
`0
`
`2
`
`5
`
`
`
`0
`
`1
`
`0
`
`2
`
`5
`0
`3
`0
`0
`2
`5
`1
`0
`1
`IFN-v Ulml (thousands)
`11-5 nglml
`Figure 4. Modulation of recall antigen responses by MMF. PBMC were stimulated in primary cultures with PPD of M . tuberculosis or
`with PPD + 100 pM or 200 pM MMF. After 6 days, 20 Ulml rIL-2 was added. At day 13, cells were washed and restimulated (7.5 x Id
`cells/ml) with CD2 mAb and CD28 mAb. Supernatants were harvested for cytokine analysis 3 days later. The data are representative of
`three separate experiments. Closed bars: restimulation with CD2/CD28 mAb. Dashed bars: restimulation with medium.
`
`4
`
`0
`
`5
`
`0
`
`
`
`Sawai (IPR2019-00789), Ex. 1023, p. 006
`
`

`

`Eur. J. Immunol. 1996.26: 2067-2074
`
`nes are only weakly encephalitogenic and thus largely
`resemble Th2 type cells in their functional behavior [43].
`
`Thl Tcells and cytokines are also thought to be involved in
`the pathogenesis of psoriasis vulgaris. In support of this
`notion are the increased serum levels of IFN-y in psoriasis
`patients and the local production of IFN-y in psoriatic
`lesions [15, 161. A polarized Thl cytokine secretion profile
`in T cell clones established from skin lesions of psoriasis
`patients has been described [14], although such a clear se-
`gregation was not found in other studies [44,45]. The abil-
`ity of MMF to alter the balance between cytokines away
`from Thl responses as defined in the current study may at
`least in part explain the beneficial action of methylated
`fumarate derivatives on psoriatic skin lesions [19-211.
`Besides, the observation that eosinophil numbers are sig-
`nificantly enhanced in MMF-treated patients [20] corrobo-
`rates our observation of the ability of MMF to enhance the
`secretion of IL-5 in vitro.
`
`Animal studies revealed that fumaric acid derivatives in
`Fumaderm@’ tablets (120 mg DMF and 95 mg MEF) are
`completely resorbed within 2 h and at least DMF is quickly
`and completely hydrolyzed in the intestines to MMF
`(manufacturer’s product information). In human serum,
`DMF is also completely metabolized into MME The
`steady-state concentration of MMF in serum of patients
`receiving 1-3 tablets of Fumaderm@ per day amounts to
`36-108 pM, which falls within the range of concentrations
`that was applied in vitro in our present studies. Our further
`studies will focus on the analysis of T cells from patients
`before and after Fumaderm@ therapy to correlate our find-
`ings with the activities in vivo of MMF on Th subset deve-
`lopment. A better understanding of the molecular mecha-
`nisms of modalities, that activate type 2 cytokines may
`have important implications in diseases in which a redi-
`rection of Th subset activation is wanted to abrogate the
`pathogenic effects of Thl effector cells.
`
`We thank Dr. I: Mutis, Dr. B Koning and Prof. R. R f! De Vries
`for critically reading the manuscript. Dr. R. Van Lier is acknow-
`ledged for generously providing monoclonal antibodies. This work
`was supported by the Amsterdam Leiden Institute for Immunology
`(ALIFI), the MACROPA Foundation, the Commission of the
`European Communities (CEC) and the Netherlands Leprosy
`Relief Association (NSL).
`
`5 References
`
`Mosmann, T. R. and Coffman, R. L., Annu. Rev. Immunol.
`1989. 7: 145.
`Seder, R. A. and Paul, W. E., Annu. Rev. Immunol. 1994.12:
`635.
`Hsieh, C., Macatonia, S. E., Tripp, C. S., Wolf, S. E,
`O’Garra, A. and Murphy, K. M., Science 1993.260: 547.
`Scott, P., Natovitz, P., Coffman, R. L., Pearce, E. and Sher,
`A., J. Exp. Med. 1988. 168: 1675..
`Kuchroo, V. K., Prabhu Das, M., Brown, J. A., Ranger, A.
`M., Zamvil, S. S., Sobel, R. A., Weiner, H. L., Nabavi, N.
`and Glimcher, L. H., Cell 1995.80: 707.
`Liblau, R. S., Singer, S. M. and McDevitt, H. O., Immunol.
`Today 1995. 16: 34.
`Chen, Y., Kuchroo, K., Inobe, J., Hafler, D. A. and Weiner,
`H. L., Science 1994.265: 1237.
`Afonso, L. C. C., Scharton, T. M., Vierira, L. Q., Wysocka,
`M., Trinchieri, G. and Scott, P., Science 1994.263: 235.
`
`Monomethylfumarate enhances Th2 cytokine secretion
`
`2073
`
`9 Seder, R. A., Paul, W. E., Davis, M. M. and Fazekas de St.
`G