Review
`
`TRENDS in Molecular Medicine Vol.11 No.1 January 2005
`
`Dimethylfumarate for psoriasis: more
`than a dietary curiosity*
`
`Ulrich Mrowietz1 and Khusru Asadullah2
`
`1Department of Dermatology, University Clinic of Schleswig-Holstein, Campus Kiel, 24105 Kiel, Germany
`2Schering AG, 13342 Berlin, Germany
`
`Swiss company Fumapharm (www.fumapharm.ch),
`became available [3]. Coating of the tablets enabled the
`liberation of FAEs in the small intestine and decreased the
`rate of adverse GI effects. A schedule of dosing was
`established on an empirical base, starting with the low-
`strength formulation for three weeks before the introduc-
`tion of high-strength tablets by increasing the number of
`tablets weekly. The maximum dose was defined as a total
`amount of 1.2 g FAE/day (two high-strength tablets three
`times a day), representing 720 mg dimethylfumarate
`(DMF). In 1994, this mixture of FAE was registered by
`the German drug administration (BfArM) as Fumadermw
`initial (low-strength tablets) and Fumadermw (high-
`strength tablets) for the systemic treatment of severe
`psoriasis.
`Since its official registration, Fumadermw has become
`the number one drug for the systemic therapy of psoriasis
`in Germany (approximately 66% of all prescriptions for
`systemic psoriasis therapy;
`‘EUROPSO patient survey
`
`(a)
`
`–
`
`–
`
`COO
`CH2
`CH2
`COO
`Succinate
`
`Succinatdehydrogenase
`
`FAD
`
`FADH 2
`
`–
`
`COO
`CH
`CH
`COO–
`Fumarate
`
`Fumarase
`
`OH
`
`H2O
`
`–
`
`COO
`CH
`CH2
`COO
`Malate
`
`–
`
`O
`
`R1
`
`O
`C7
`
`H
`
`C1
`
`C2
`
`H
`
`6
`C
`O
`
`R2
`
`O
`
`(b)
`
`CH3
`
`CH
`
`2
`
`O
`
`EHF
`
`O
`H
`C43
`C38
`
`H
`
`C3
`
`9
`
`OH
`
`42
`C
`O
`
`O
`
`CH3
`
`MHF
`O
`H
`C31
`C26
`
`H
`
`C2
`
`7
`
`OH
`
`30
`C
`O
`
`DMF
`O
`H
`C19
`C14
`
`O
`
`CH3
`
`H
`
`C1
`
`5
`
`18
`C
`O
`
`CH3
`
`O
`
`TRENDS in Molecular Medicine
`
`Figure 1. Fumarate and its esters. (a) Fumarate metabolism in the citric acid cycle.
`(b) The chemical structure of free fumaric acid (R1ZR2ZH) and the fumaric acid
`esters (R1ZR2ZAlkyl) dimethylfumarate (DMF), methylhydrogenfumarate (MHF)
`and ethylhydrogenfumarate (EHF).
`
`Fumaric acid esters (FAEs) have been used for the oral
`treatment of psoriasis since 1959 and have been
`registered for this indication in Germany since 1994.
`Dimethylfumarate (DMF) and its metabolite methyl-
`hydrogenfumarate (MHF) are the pharmacologically
`active compounds, with DMF being the main com-
`ponent of the marketed FAE-mixture. However, the
`mechanism of action of FAE is yet to be fully understood.
`It has been shown that DMF inhibits NFkB translocation,
`which leads to (i) the inhibition of pro-inflammatory
`cytokine production and adhesion molecule expression,
`(ii) the inhibition of dendritic cell differentiation and, at
`higher concentrations, (iii) the induction of apoptosis.
`Recent evidence also shows that these effects are
`mediated through the interference of the intracellular
`redox system by DMF. Here, the mode of action of FAE
`and its clinical use for psoriasis will be discussed.
`
`Introduction
`Fumaric acid is a simple-structured dicarbonic acid, which
`has an important role in the citric acid cycle in humans
`(Figure 1). Fumaric acid deficiencies, or other disorders
`related to a disturbed metabolism of this compound, are
`unknown as a cause of disease. Nevertheless, fumaric acid
`esters are the number one drug for the oral treatment of
`severe psoriasis in Germany.
`Fumaric acid is used as a nutritional additive in various
`forms in the food and farming industries, without
`untoward effects [1]. The clinical use of fumaric acid
`derivatives started in 1959 when the German chemist
`Schweckendiek, suffering from psoriasis, raised the
`hypothesis that the disease might be caused, at least in
`part, by disturbances in the citric acid cycle. He postulated
`that the exogenous supplementation of fumaric acid might
`reverse the pathological process. Schweckendiek tried to
`prove his hypothesis by treating himself with fumaric acid
`derivatives. Because of a potential irritant effect of free
`fumaric acid when taken orally, he used esters of fumaric
`acid. Gastrointestinal (GI) irritant effects, however, could
`not fully be circumvented. Taking a self-designed mixture
`of different fumaric acid esters (FAEs; see Glossary), his
`psoriasis cleared [2].
`To improve the pharmaceutical quality, a new product
`consisting of a defined mixture of FAEs, produced by the
`
`*Schering AG has no commercial interest in Fumadermw.
`Corresponding author: Mrowietz, U. (umrowietz@dermatology.uni-kiel.de).
`Available online 2 December 2004
`
`www.sciencedirect.com 1471-4914/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.molmed.2004.11.003
`
`Sawai (IPR2019-00789), Ex. 1019, p. 001
`
`

`

`44
`
`Review
`
`TRENDS in Molecular Medicine Vol.11 No.1 January 2005
`
`Glossary
`
`DMF: dimethylfumarate
`EHF: ethylhydrogenfumarate
`FAEs: fumaric acid esters
`MHF: methylhydrogenfumarate
`NFkB: nuclear factor kB
`
`2002’, www.europso.org). Clinical studies are limited in
`number; however, the reported efficacy in psoriasis is
`high, together with a favourable long-term safety profile
`(see below). The first data concerning combination treat-
`ment with FAEs were recently published which broadens
`the clinical usage in patients with severe psoriasis [4,5].
`Within the last few years, there has been increasing
`evidence that FAEs have potent immunomodulatory
`effects on various types of cells with importance not only
`for the psoriatic tissue reaction.
`
`Pathogenetic concepts of psoriasis
`Psoriasis is regarded as an immune disorder in which a
`putative antigen presented to T cells by antigen-presenting
`dendritic cells leads to the generation of specifically
`activated T cells [6]. According to the T-cell cytokine
`expression profile, psoriasis is classified as a Th1-type
`immune response.
`Because several other inflammatory diseases, such as
`rheumatoid arthritis, Crohn’s disease and multiple scler-
`osis, follow similar immunological pathways of T-cell
`activation, psoriasis can be regarded as a visible disease
`model. The crucial involvement of T-cells in the patho-
`genesis of psoriasis was highlighted by transplantation
`experiments of psoriasis skin onto severe combined
`immunodeficiency mice and investigations to analyse the
`T-cell receptor repertoire [7,8]. In addition to the inflam-
`matory component, the hyperproliferation of epidermal
`keratinocytes, together with a disturbed cellular differen-
`tiation (leading to the disease characteristic hyper-
`parakeratosis), seems to be dependent on a genetic
`susceptibility that is, among others, associated with the
`epidermal differentiation complex (psoriasis susceptibility
`locus; PSORS4) and corneodesmosin (PSORS1) [9].
`The mechanism of action of FAEs in psoriasis might,
`therefore, be of interest for future use in the treatment of
`diseases with a pathogenetic background similar to this
`chronic skin disorder. Here, a summary of the current
`knowledge about the clinical use of FAEs and its mode of
`action is given.
`
`Clinical use of FAE for the systemic treatment of
`psoriasis
`The consensus guidelines [3] recommend treating psoriasis
`patients with Fumadermw until there is an improvement of
`lesions or the patient is satisfied with the treatment
`outcome. A maintenance treatment to stabilize clinical
`improvement can be performed for up to two years. Clinical
`trials have shown a decrease of the psoriasis area and
`severity index (PASI) – a measure for psoriasis severity –
`between 50 and 80% after 12–16 weeks of therapy [10,11].
`Recently, Hoefnagel and colleagues [12] presented data
`about the long-term safety of Fumadermw therapy in
`
`www.sciencedirect.com
`
`patients treated continuously for up to 14 years. Notably,
`no increased risk for infections or malignancies could be
`found within this study population.
`In earlier clinical trials, the question of which of the
`different esters of fumaric acid is the active compound was
`addressed. In 1989, Nieboer et al. [13] published the
`results of an extensive study evaluating the clinical effects
`of different FAEs in various concentrations as mono-
`therapies and in combination in psoriasis patients. The
`data showed that ethylhydrogenfumarate (EHF) mono-
`therapy was not superior to placebo. However, by using
`DMF (240 mg/d), a 40% improvement was seen in the
`DMF-treated group after six weeks of therapy, whereas
`in the placebo-group, worsening of psoriasis was reported.
`It was concluded that the main active ingredient of
`Fumadermw is DMF.
`
`Adverse events related to FAE treatment
`The most common adverse event reported in all clinical
`trials with FAE is GI complaints. Clinical symptoms
`include diarrhoea (most common), stomach ache and
`cramps, increased frequency of stools, nausea and vomit-
`ing. GI-complaints are clearly related to the tablet
`formulation of FAE. With a newly developed microtablet
`formulation, the rate of GI complaints is decreased to
`below 6% (unpublished data*).
`Flush is another characteristic adverse event related to
`FAE-treatment. The symptoms are highly variable and
`are reported mainly as a transient feeling of warmth or
`heat normally lasting for minutes. Remarkably, these side
`effects frequently occur only during the initial phase of
`therapy. Other symptoms occurring as adverse events
`during FAE therapy are rare. There is no report about
`an increased frequency of infections, neither of bacterial
`(e.g. folliculitis) nor viral origin (e.g. herpes simplex,
`zoster or human papillomavirus-related infections, such
`as common warts, common cold or flu). There are no
`reports in the literature about the development of any
`form of malignancies (e.g. lymphoma, myeloproliferative
`disorders and solid tumours) during long-term treatment
`with FAE.
`FAE-induced leukocytopenia is frequently observed
`[10,14]. Severe lymphocytopenia might be a reason to
`terminate FAE-therapy. Although there are no published
`follow-up data in lymphopenic patients after FAE-with-
`drawal, there is a complete recovery up to baseline values
`within 6–12 weeks after termination of FAE-therapy
`(U. Mrowietz, unpublished data).
`A characteristic of FAE treatment is an increase in
`peripheral blood eosinophils [10,11]. Maximum eosinophil
`numbers are usually seen between week four and eight of
`therapy with a peak at week six. In individual patients, the
`percentage of eosinophils in the differential count can exceed
`25%. However, clinical symptoms or disorders related to
`eosinophilia have not been observed. When FAE-therapy is
`continued, eosinophil numbers normally decline and, in
`most patients, return to baseline levels [11].
`
`* Langner, A. et al. (2004) Results of a phase II study of a novel oral fumarate,
`BG-12, in the treatment of severe psoriasis. J. Eur. Acad. Dermatol. Venerol. 18, 798.
`Presented at the European Congress on Psoriasis, October 21–24 2004, Paris, France.
`
`Sawai (IPR2019-00789), Ex. 1019, p. 002
`
`

`

`Review
`
`TRENDS in Molecular Medicine Vol.11 No.1 January 2005
`
`45
`
`Pharmacokinetics of fumaric esters in humans
`There is little knowledge about the pharmacokinetics of
`FAEs in humans. FAEs are almost completely absorbed in
`the small intestine. The postulated main active ingredient of
`Fumadermw, DMF, is rapidly hydrolysed by ubiquitous
`esterases to methylhydrogenfumarate (MHF), its main
`metabolite. MHF is further metabolized into fumaric acid
`and finally into carbon dioxide and water. There is no evi-
`dence for an organ-specific toxicity due to drug metabolism.
`Metabolism is independent of cytochrome P-450-
`dependent pathways. This is of particular importance
`because psoriasis is frequently associated with other
`diseases, such as diabetes mellitus or coronary heart
`disease, requiring concomitant medications [15]. Drug
`interactions have not been reported.
`In a recent investigation in healthy subjects given
`one tablet containing 240 mg DMF and 95 mg calcium-
`monoethylfumarate, only MHF, the metabolite of DMF,
`could be detected in serum after a mean time of 120 min,
`when the tablet was taken under fasting conditions. DMF
`and free fumaric acid were below detection limit of the
`HPLC-system used [16].
`By using an intestinal tissue homogenate the half-life
`of DMF was less than 2 min. The enzymes responsible for
`DMF degradation are carboxylesterases rather than
`cholinesterases [17], but the significance of this finding
`is not yet clear.
`
`Mode of action
`FAEs can modulate the functions, including proliferation
`and mediator production and secretion, of several cell
`types. Can the various known effects of FAE be linked to a
`common mechanism of action?
`
`FAEs affect T-cell numbers
`Early studies clearly demonstrated that FAEs have
`profound effects on the T-cell system. In an open study
`using Fumadermw in 16 patients with severe psoriasis
`(PASIO20), flow-cytometric determination of leukocytes
`was performed weekly for 12 weeks [18]. Leukocytopenia
`was observed in 94% of the patients, with a mean
`reduction in leukocyte numbers of 26.6% (maximum
`C
`T cells decreased by 45.4%
`60%). The number of CD4
`C
`C
`T cells by 44.5%. In one patient, CD8
`cells
`and CD8
`decreased by 87%; however, clinical signs of immuno-
`suppression-like infections were not observed.
`In seven of 16 patients, absolute numbers of CD4
`T cells were w200/ml. There was no change in the
`CD4:CD8 ratio. HLA-DR-expression on T cells remained
`unaltered and there was no significant alteration in CD25
`expression (IL-2 receptor a-chain). The decrease in the
`number of peripheral T cells is associated with a reduction
`of the lesional T-cell infiltrate. An immunohistochemical
`study of 33 patients, from whom biopsies were taken
`before and after two, six and eight weeks of Fumadermw
`therapy, revealed a decrease in the epidermal inflamma-
`tory infiltrate of 74.4% after eight weeks of treatment,
`C
`T cells of 51.2% [19]. The
`with a decrease of CD4
`subepidermal
`infiltrate decreased by 52.1%, with a
`C
`cells of 84.4% after eight weeks.
`reduction in CD4
`C
`memory T cells slightly increased
`Interestingly, CD45RO
`
`C
`
`www.sciencedirect.com
`
`within the epidermis from 26.8% to 33%, in the sub-
`epidermal layer from 28.9% to 33.6% and decreased in the
`deeper dermis from 20.3% to 11.9% after eight weeks of
`Fumadermw therapy. Within the eight-week treatment
`period there was a complete clearing of neutrophil
`granulocytes from the skin in pre-treatment biopsies.
`Although treatment with fumarates lead to a profound
`decrease in the number of T cells in some patients, which
`might be a reason to stop therapy, adverse events related
`to a decreased immunosurveillance have not been
`observed in short term or long-term studies [12].
`In an in vitro study, the effect of MHF on T-cell cytokine
`production was analysed in greater detail [20]. The
`C
`authors demonstrated, in purified T cells (98% CD2
`)
`stimulated with antibodies against CD2, CD28 or a
`mixture of CD28 and CD3, that !200 mM MHF did not
`influence cell proliferation. MHF induced the production
`of the Th2-type cytokines IL-4 and IL-5 (EC50 100 mM), but
`was without effect on IL-2 or IFNg production, which are
`prominent Th1-type cytokines. The stimulating effect of
`MHF on IL-4 and IL-5 production was also shown for
`several different T-cell clones. By using radioactively
`labelled MHF and DMF, evidence for specific binding
`sites for MHF on T-cells was provided. In a separate set of
`C
`C
`CD45RO
`memory T cells were
`experiments, CD4
`stimulated with antibodies against CD2 and CD28 and
`the effect of MHF was evaluated. MHF upregulated the
`secretion of IL-4 2.5-fold and IL-5 threefold in stimulated
`memory T cells, without any effect on IFNg secretion. The
`induction of IL-4 and IL-5, but not of IFNg, by MHF was
`further seen in peripheral blood mononuclear cells
`(PBMCs) stimulated with purified protein derivative
`(PPD) as a pathogenic stimulus, mimicking an immune
`response against Mycobacterium tuberculosis. MHF
`enhanced the release of TNFa and IL-10 in lipopolysac-
`charide-stimulated monocytes without having an effect on
`IL-12 and IL-1RA secretion [21]. Newer data showed a
`reduced secretion of IFNg after the treatment of dendritic
`cells with MHF and subsequent T-cell stimulation [22].
`Goreschi et al. (unpublished data†) analysed the ex vivo
`Th1- and Th2-cytokine repertoire of T cells from patients
`with psoriasis treated with FAEs. They found a significant
`C
`suppression of the intracellular IFNg:IL-4 ratio in CD4
`T cells, which began after three weeks of therapy and
`paralleled clinical improvement.
`From these data,
`it was concluded that the main
`mechanism by which FAE induces the remission of
`psoriatic lesions is a shift of the immunological balance
`from a Th1- towards a Th2-like response. However, the
`prominent decrease in the number of T cells that is
`associated with FAE therapy prompted the investigation
`of the effect of these compounds on programmed cell
`death.
`
`FAE-induced apoptosis
`In U937-cells, Sebo¨ck et al.
`[23] were the first to
`demonstrate the potent apoptosis-inducing activity of
`
`† Goreschi, K. et al. (2002) Fumaric acid ester an anti-psoriatic drug abolishes the
`capacity of T cells to induce Th1-mediated autoimmune disease. Arch. Dermatol. Res.
`294, 28. Presented at the XXIX Annual meeting of the Arbeitgemeinschaft
`Dermatologische Forschung ADF, Ferbruary 28–March 2 2002, Berlin, Germany.
`
`Sawai (IPR2019-00789), Ex. 1019, p. 003
`
`

`

`46
`
`Review
`
`TRENDS in Molecular Medicine Vol.11 No.1 January 2005
`
`DMF. These data were confirmed in human monocyte-
`derived dendritic cells as an important regulator of specific
`T-cell activation. Zhu and Mrowietz [24] showed that DMF
`and MHF could completely prevent the GM-CSF- and
`IL-4-induced differentiation of monocytes into dendritic
`cells. DMF also potently induced apoptosis in these cells.
`Recently, the effect of FAEs on T-cell apoptosis was
`investigated in greater detail. In purified human T cells,
`DMF, but not MHF, could induce apoptosis. This was
`substantiated by measuring two independent markers
`[Apo2.7 expression and DNA fragmentation, using
`TUNEL (Tdt-mediated dUTP nick end labeling)]. T cells
`stimulated with IL-2, antibodies against OKT3 or both
`were more susceptible to DMF-mediated apoptosis com-
`pared with unstimulated cells. In parallel to an increase in
`Apo2.7 expression, DMF induced a dose-dependent down-
`regulation of anti-apoptotic Bcl2-expression [25].
`
`FAEs modulate cytokine production
`In a series of experiments, the effect of FAEs, notably
`DMF, on inflammatory mediator transcription and pro-
`duction was investigated. Using human PBMCs, DMF
`inhibited lipopolysaccharide-induced protein secretion of
`the chemokines IL-8, Mig and IP-10 in a dose-dependent
`manner without altering cell viability [27].
`Because keratinocytes themselves are actively produc-
`ing inflammatory mediators, such as chemokines, Stoof
`et al. [26] investigated the effect of DMF on phorbol-ester-
`or
`IFNg-stimulated normal human keratinocytes
`obtained from healthy human skin. They showed that
`DMF caused a dose-dependent inhibition of groa, IL-8,
`Mig, IP-10 and IP-9/I-TAC mRNA expression. Further-
`more, protein secretion of IL-8, Mig and IP-10 was
`decreased by the DMF treatment of keratinocytes.
`Ockenfels et al. [27] co-cultured normal human kera-
`tinocytes obtained from patients with psoriasis together
`with HUT78-cells (a T-cell-like cell line). DMF inhibited
`IFNg, IL-6 and TGFa release and increased IL-10 release
`in all co-culture experiments.
`
`FAE modulates adhesion molecule expression
`Because the inflammatory response in psoriasis is associ-
`ated with an increased expression of adhesion molecules
`on endothelial cells and keratinocytes [28], the effect of FAEs
`was investigated. Vandermeeren and co-workers [29]
`showed that DMF inhibited TNFa- or IL-1a-stimulated
`expression of the adhesion molecule ICAM-1 in human
`fibroblasts.
`In human umbilical vein endothelial cells (HUVECs),
`DMF blocked the TNFa-induced expression of the
`adhesion molecules ICAM-1, VCAM-1 and E-selectin
`[29]. Furthermore, the adherence of monocytic U937
`cells to TNFa- and IL-4-treated HUVECs was inhibited
`by DMF.
`
`FAEs inhibit NFkB activity
`Because the transcription of chemokines and adhesion
`factors, including IL-8 and E-selectin, is dependent upon
`NFkB, it is possible that DMF modulates this regulatory
`pathway. Additional evidence for a possible role of NFkB
`was provided by the observation that apoptosis is, at least
`
`www.sciencedirect.com
`
`in part, dependent on the modulation of NFkB activity as a
`regulator of cell survival. Bureau et al. [30] demonstrated
`the importance of constitutively active NFkB for the
`survival of quiescent mature immune cells, such as T and
`B cells, monocytes and macrophages and neutrophil
`granulocytes. The blockade of NFkB lead to apoptotic
`cell death in human macrophages, which depended on the
`activation of caspase 9, but not caspases 3 or 8 [31]. In
`human T cells, the inhibition of NFkB translocation was
`associated with apoptotic cell death independent of
`caspases 1 and 3 [32]. Ward et al. [33] found TNFa to be
`a survival
`factor for neutrophils linked to NFkB
`activation.
`When the effect of DMF on the NFkB-pathway was
`analysed more closely, a significant inhibition of kB1/p50
`in the nucleus was found, whereas there was little influ-
`ence on IkBa, b and 3 or RelA/p65 or c-Rel. AP-1-mediated
`gene transcription remained unaltered [34].
`The effect of DMF on the NFkB cascade was further
`investigated using the TNFa- or VEGF-induced mRNA
`expression of tissue factor in human endothelial cells [35].
`DMF inhibited TNFa-, but not VEGF-, induced tissue factor
`expression. Further experiments revealed an inhibition of
`TNFa-mediated nuclear entry of p65. This was not due to
`the inhibition of TNFa-induced signalling to IkB, because
`IkBa phosphorylation and degradation were not altered
`[36]. By interfering with NFkB translocation, DMF acts not
`only as an inhibitor of TNFa-induced cellular functions but
`also as an inhibitor of TNFa-production, because its
`transcription is, at least in part, dependent on NFkB.
`Litjens et al. [22] provided evidence that, in addition to
`DMF, MHF inhibited LPS-induced NFkB activation in
`human dendritic cells.
`
`FAEs interfere with cellular redox-systems
`Barchowsky et al. [37], using porcine aortic endothelial
`cells, found an association between DMF-mediated inhi-
`bition of NFkB translocation and intracellular thiol levels.
`The phosphorylation and ubiquitination of IkB is depen-
`dent on the balance of reduced glutathione (GSH) and
`oxidized glutathione-disulfide (GSSG), which is influenced
`by intracellular reactive oxygen species (ROS) through
`redox-sensitive kinases [38].
`Data from the toxicology literature indicate a direct
`link between cell death by necrosis or apoptosis and
`intracellular levels of thiols. In a model of acetaminophen-
`induced liver toxicity, hepatocyte necrosis was observed in
`relation to decreased glutathione levels [39]. The addition
`of antioxidant glutathione-monoethylester shifted necrotic
`cell death to apoptosis. An increase of oxidized and a
`reduction of reduced glutathione lead to increased
`apoptosis in M14 melanoma cells, which was associated
`with a downregulation of the c-myc protooncogene [40].
`The effect of DMF on intracellular thiols is time
`dependent. The addition of DMF to Chinese hamster
`ovary cells for 5 min was followed by a decrease of
`intracellular glutathione to 10% of basal
`levels [41].
`When the activity of the glutathione-synthesis-limiting
`enzyme g-glutamine–cysteine synthase in guinea pig
`hippocampus cells was inhibited by the addition of the
`irreversible inhibitor buthioninsulfoximine, intracellular
`
`Sawai (IPR2019-00789), Ex. 1019, p. 004
`
`

`

`Review
`
`TRENDS in Molecular Medicine Vol.11 No.1 January 2005
`
`47
`
`levels of reduced glutathione were decreased by 30% [42].
`The subsequent addition of DMF further reduced gluta-
`thione to 4% of baseline levels. However, Duffy et al. [43]
`demonstrated an increase of glutathione in a retinal-
`neuroblastoma hybrid cell-line by DMF. This discrepancy
`was further investigated by using human retinal pigment
`epithelial cells. Nelson and co-workers [44] found a time-
`dependent effect of DMF on intracellular thiols: an initial
`decrease of glutathione followed by a sustained increase of
`more than twofold.
`These data were confirmed in animal experiments in
`which DMF was added to the diet of mice and rats for two
`weeks, leading to a significant increase of glutathione and
`quinone-reductase, which is another detoxifying enzyme
`that is found in various organs, such as the liver and small
`intestine [45].
`From these experiments, it can be concluded that DMF
`can significantly modulate intracellular levels of detoxify-
`ing enzyme systems, such as the GSH–GSSG system,
`upregulating reduced-glutathione levels after prolonged
`exposure or dietary supplementation. There are no data
`about the activity of MHF in these systems.
`
`FAEs: hypothesized molecular mode of action
`Taken together, the following hypothesis about the
`mechanism of action of DMF can be proposed: (i) DMF
`interferes with intracellular thiols resulting in increased
`levels of reduced glutathione after prolonged exposure,
`through an unknown mechanism; (ii) increased gluta-
`thione levels inhibit redox-sensitive kinases, resulting in
`(iii) an inhibition of the phosphorylation and ubiquitina-
`tion of IkB,
`leading to (iv) an inhibition of NFkB
`translocation (Figure 2). This pathway leads to the
`modulation of the NFkB-dependent cascades of inflam-
`matory cytokine production and adhesion molecule
`expression. Because DMF seems not to interfere with
`the basal cytokine expression that is necessary for
`immune defence reactions, the documented long-term
`safety regarding the lack of an increased risk of infections
`or tumour development might be explained.
`
`Concluding remarks
`The empirically observed effect of FAE on the immuno-
`logically mediated skin disease psoriasis has led to the
`discovery of the potent immunomodulatory activity of the
`active compound DMF. FAEs, particularly DMF, might be
`regarded as potent immunomodulators with clinically
`proven effectiveness in psoriasis. From the present data,
`the inhibition of NFkB translocation and downstream
`NFkB-dependent pro-inflammatory pathways are the
`hallmarks of the mode of action of DMF. The interference
`with intracellular redox-systems might lead directly to the
`effects observed.
`Unfortunately, the product approved only in Germany
`(Fumadermw), being a mixture of different FAEs com-
`posed by empirical means, could not be licensed else-
`where. However, a new drug, termed BG-12, is currently
`in clinical development and consists of DMF as a
`monosubstance in a microtablet formulation.
`Future research should be directed towards determin-
`ing how FAEs, namely DMF and its metabolites, including
`
`www.sciencedirect.com
`
`DMF
`
`+
`
`–
`
`GSH:GSSG
`
`ROI
`
`+
`–
`Redox-sensitive kinases
`
`Phosphorylation of IκB
`
`Ubiquitination of IκB
`
`Cytosol
`
`Poly-ubiquitinated IκB
`is degraded by
`26S proteasome
`
`NF-κB activation
`
`
`
`TranslocationTranslocation
`
`Nucleus
`GSSG
`+ –
`NFκB
`Binding
`
`DMF
`
`+
`
`–
`
`GSH
`
`–
`DMF
`
`NFκB-site
`
`TRENDS in Molecular Medicine
`
`Figure 2. Proposed mode of action of DMF on NFkB-regulated gene transcription.
`The balance between oxidized glutathione (GSSG) and reduced glutathione (GSH)
`as well as the presence of reactive oxygen intermediates (ROI) regulates
`degradation of the inhibitor of kB (IkB) and subsequent activation of NFkB. DMF
`increases GSH, leading to an inhibition of NFkB-activation. Furthermore, DMF
`inhibits NFkB p65 subunit translocation from the cytosol into the nucleus, and by
`increasing GSH levels decreases the binding of NFkB to DNA.
`
`free fumaric acid, interact with intracellular structures
`and components at the molecular level. The establishment
`of more sophisticated methods for analysing FAE metab-
`olism in vivo and ex vivo is an essential prerequisite. The
`identification of the mode of action of FAEs in more detail
`might lead to the identification of promising new drug
`targets. More clinical studies are also needed; these
`should include larger patient populations over longer
`periods of time, to verify the suspected favourable long-
`term safety.
`
`References
`1 Blank, R. et al. (1999) Effect of fumaric acid and dietary buffering
`capacity on ileal and fecal amino acid digestibilities in early-weaned
`pigs. J. Anim. Sci. 77, 2974–2984
`2 Schweckendiek, W. (1959) Heilung von Psoriasis. Med. Monatschr. 13,
`103–104
`3 Mrowietz, U. et al. (1999) Treatment of severe psoriasis with fumaric
`acid esters: scientific background and guidelines for therapeutic use.
`Br. J. Dermatol. 141, 424–429
`4 Balasubramaniam, P. et al. (2004) Fumaric acid esters in severe
`psoriasis, including experience of use in combination with other
`systemic modalities. Br. J. Dermatol. 150, 741–746
`5 Ormerod, A.D. and Mrowietz, U. (2004) Fumaric acid esters, their
`place in the treatment of psoriasis. Br. J. Dermatol. 150, 630–632
`6 Prinz, J.C. (2001) Psoriasis vulgaris – a sterile antibacterial skin
`reaction mediated by cross-reactive T cells? An immunological view of
`the pathophysiology of psoriasis. Clin. Exp. Dermatol. 26, 326–332
`7 Boehncke, W.H. et al. (1996) Pulling the trigger on psoriasis. Nature
`379, 777
`8 Vollmer, S. et al. (2001) Dominant lesional T cell receptor rearrange-
`ments persist in relapsing psoriasis but are absent from nonlesional
`skin: evidence for a stable antigen-specific pathogenic T cell response
`in psoriasis vulgaris. J. Invest. Dermatol. 117, 1296–1301
`9 Elder, J.T. et al. (2001) The genetics of psoriasis 2001: the odyssey
`continues. Arch. Dermatol. 137, 1447–1454
`
`Sawai (IPR2019-00789), Ex. 1019, p. 005
`
`

`

`48
`
`Review
`
`TRENDS in Molecular Medicine Vol.11 No.1 January 2005
`
`(1994) Antipsoriatic effect of fumaric acid
`10 Altmeyer, P. et al.
`derivatives. Results of a multicenter double-blind study in 100
`patients. J. Am. Acad. Dermatol. 30, 977–981
`11 Mrowietz, U. et al. (1998) Treatment of psoriasis with fumaric acid
`esters. Results of a prospective multicenter study. Br. J. Dermatol.
`138, 456–460
`12 Hoefnagel, J.J. et al. (2003) Long-term safety aspects of systemic
`therapy with fumaric acid esters in severe psoriasis. Br. J. Dermatol.
`149, 363–369
`13 Nieboer, C. et al. (1989) Systemic therapy with fumaric acid derivates:
`new possibilities in the treatment of psoriasis. J. Am. Acad. Dermatol.
`20, 601–608
`14 Ho¨xtermann, S. et al. (1998) Fumaric acid esters suppress peripheral
`CD4- and CD8-positive lymphocytes in psoriasis. Dermatology 196,
`223–230
`15 Henseler, T. and Christophers, E. (1995) Disease concomitance in
`psoriasis. J. Am. Acad. Dermatol. 32, 982–986
`16 Litjens, N.H. et al. (2004) Pharmacokinetics of oral fumarates in
`healthy subjects. Br. J. Clin. Pharmacol. 58, 429–432
`17 Werdenberg, D. et al. (2003) Presystemic metabolism and intestinal
`absorption of antipsoriatic fumaric acid esters. Biopharm. Drug
`Dispos. 24, 259–273
`18 Altmeyer, P. et al. (1996) Verlaufsbeobachtungen der Lymphozyten-
`subpopulationen bei Psoriasis-Patienten unter oraler Therapie mit
`Fumaraten. Aktuel. Dermatol. 22, 272–277
`19 Bacharach-Buhles, M. et al. (1994) Fumaric acid esters (FAEs)
`suppress CD15- and ODP 4-positive cells in psoriasis. Acta Derm.
`Venereol. Suppl. (Stockh.) 186, 79–82
`20 De Jong, P. et al. (1996) Selective stimulation of T helper 2 cytokine
`responses by the anti-psoriasis agent monomethylfumarate. Eur.
`J. Immunol. 26, 2067–2074
`21 Asadullah, K. et al. (1997) Influence of monomethylfumarate on
`monocytic cytokine formation-explanation for adverse and thera-
`peutic effects in psoriasis? Arch. Dermatol. Res. 289, 623–630
`22 Litjens, N.H. et al. (2004) Monomethylfumarate affects polarization of
`monocyte-derived dendritic cells resulting in down-regulated Th1
`lymphocyte responses. Eur. J. Immunol. 34, 565–575
`23 Sebo¨k, B. et al. (2000) Dimethylfumarat ist unter den Fumadermw-
`Inhaltsstoffen der sta¨ rkste Induktor von Apoptose-Pha¨ nomenen in
`lympho-histiozyta¨ ren U-937-Zellen. HGZ. Hautkr. 75, 1–5
`24 Zhu, K. and Mrowietz, U.
`(2001) Inhibition of dendritic cell
`differentiation by fumaric acid esters. J. Invest. Dermatol. 116,
`203–208
`25 Treumer, F. et al. (2003) Dimethylfumarate is a potent inducer of
`apoptotic cell death in human T cells. J. Invest. Dermatol. 121,
`1383–1388
`26 Stoof, T.J. et al. (2001) The antipsoriatic drug dimethylfumarate
`strongly suppresses chemokine production in human keratinocytes
`and peripheral blood mononuclear cells. Br. J. Dermatol. 144,
`1114–1120
`27 Ockenfels, H.M. et al. (1998) The antipsoriatic agent dimethylfuma-
`rate immunomodulates T-cell cytokine secretion and inhibits cyto-
`kines of the psoriatic cytokine network. Br. J. Dermatol. 139, 390–395
`
`28 Veale, D. et al. (1995) Immunolocalization of adhesion molecules in
`psoriatic arthritis, psoriatic and normal skin. Br. J. Dermatol.
`132, 32–38
`29 Vandermeeren, M. et al. (1997) Dimethylfumarate is an inhibitor of
`cytokine-induced E-selectin, VCAM-1, and ICAM-1 expression in
`human endothelial cells. Biochem. Biophys. Res. Commun. 234, 19–23
`30 Bureau, F. et al. (2002) Constitut