Effects of Monomethylfumarate on Human
`Granulocytes
`
`Peter H. Nibbering,* Bing Thio,t Timo P.L. Zomerdijk,* Anja C. Bezemer,* Roderick L. Beijersbergen,* and
`Ralph van Furth*
`Departments of "Infectious Diseases and i"Dcrmatology, Univcrsiry Hospital, Leiden, The Netherlands
`
`Monomethylfumarate (MMF) is the most active metabolite
`of the new antipsoriasis drug Fumaderm. Because granulo(cid:173)
`cytes play an important role in the pathophysiology of psoria(cid:173)
`sis, the effects of this drug on the functional activities of these
`cells were investigated. MMF stimulated polarization and
`elastase release, and enhanced the intracellular killing ofbac(cid:173)
`teria by granulocytes. This compound suppressed the for(cid:173)
`myl-Met-Nle-Phe (FMLP) - stimulated respiratory burst in
`these cells. MMF and dimethyl fumarate but not its stereo(cid:173)
`isomer dimethyl maleate, fumaric acid, or dimethylmalate
`stimulated polarization of and elastase release by granulo(cid:173)
`cytes, indicating that methylated fumarate derivatives inter(cid:173)
`act with granulocytes in a specific fashion. MMF did not
`
`affect the binding of formyl-Nle-Leu-Phe-Nle-Tyr-Lys(cid:173)
`fluorescein isothiocyanate to the FMLP receptor on granulo(cid:173)
`cytes. This compound induced an increase in the intracellular
`Ca++ ([Ca++L) and cyclic adenosine monophosphate con(cid:173)
`centration.
`The agonistic effects of MMF on granulocytes are thought
`to be mediated by the rise in the [Ca++] i and the antagonistic
`effects by the increase in the cyclic adenosine monophos(cid:173)
`phate concentration. These effects of MMF on granulocytes
`may in part explain the beneficial action of methylated fuma(cid:173)
`rate derivatives on psoriatic skin lesions. ] In.vest Dermatol
`101:37-42,1993
`
`Psoriasis, a chronic inflammatory skin disease affecting up
`
`to 3% of the population, is characterized by epidermal
`hyperplasia and infiltration of lymphocytes and granu(cid:173)
`locytes into the skin [l,2J. Treatment is based on inhibi(cid:173)
`tion of the proliferation of epidermal cells and interfer(cid:173)
`ence in the inflammatory process. In addition to the present
`therapeutic regimens a new systemic antipsoriasis drug, which con(cid:173)
`sists of dimethyl fumarate (DMF) and monoethylfumarate (MEF) ,
`has been introduced with successful results [3,4]'* Monomethylfu(cid:173)
`marate (MMF), which is formed in the circulation by hydrolysis of
`DMF, is believed to be th e most potent metabolite of this drug.
`Granulocytes are retained at the epidermis by means of a number of
`factors [5 - 7]. The exact role of granulocytes in the pathophysiol(cid:173)
`ogy of psoriasis is not known. IL-8 in combination with TNF-a,
`which is also present in psoriatic skin lesions [2,5], stimulates granu(cid:173)
`locytes to produce reactive oxygen intermediates and proteolytic
`enzymes [8,9], which can damage epidermal cells and probably
`influence the growth and differentiation of keratinocytes. Because
`
`Manuscript received September 17, 1992; accepted for publication Febru(cid:173)
`ary 12, 1993.
`Reprint requests to: R. van Furth, Department of Infectious Diseases,
`Universiry Hospital, Building 1, C5-P, P.O. Box 9600, 2300 RC Leiden,
`The Netherlands.
`Abbreviations: db-cAMP, diburyryl-cyclic adenosine monophosphate;
`DMF, dimethylfumarate; DMM, dimethylmaleate; FA, fumaric acid; FITC,
`fluorescein isothiocyanate; FLPEP FITC, formyl-Nle-Leu-Phe-Nle-Tyr(cid:173)
`Lys-FITC; FMLP, formyl-Met-Leu-Phe; FURA2/ AM, pentakis (acetoxy(cid:173)
`methyl) ester of [1-[2-(5-carboxyoxazol-2-yl)-6-aminobenzofuran-5-oxy]-
`2-(2' -amino-S' -methylphenoxy)-ethane-N ,N ,N',N' -tetraacetic acid; hpf,
`high-power field; MEF, monoethylfumarate; MMF, monomethylfumarate;
`Tw, Tween 8.
`:j: Thio HB, van der SchroeR" JG, Nugteren-Huying WH, Vermeer BJ:
`Long-term therapy with dimethylfumarate and monoethylfumarate (Fuma(cid:173)
`derm) in psoriasis (in preparation).
`
`granulocytes play an important role in the pathophysiology of psori(cid:173)
`asis the beneficial effect of treatment with fumaric acid derivatives
`may be caused by its effects on granulocytes. The present study
`concerns the effects of MMF on granulocytes from healthy donors.
`
`MATERIALS AND METHODS
`Compounds Used to Stimulate Granulocytes Cells were in(cid:173)
`cubated with the following compounds: MMF (Sigma Chemical
`Co., St. Louis, MO), dimethyl fumarate (DMF; Merck, Darmstadt,
`Germany), dimethylmaleate (DMM; Merck), fumaric acid (FA;
`Merck), and S- or R-dimethylmalate (S-DMM and R-DMM; Jans(cid:173)
`sen Chimica, Geel, Belgium). In all experiments the pH of the stock
`solutions of these compounds was adjusted to about 7.2 with 0.1 N
`NaOH. At the concentrations and exposure time used in this study
`the stimuli did not affect cell viability, as determined by trypan blue
`exclusion and the lactate dehydrogenase release assay.
`For comparison, cells were stimulated with 10 to 100 nM for(cid:173)
`myl-Met-Leu-Phe (FMLP) (Sigma) or 1 nM rlL-8 (a gift from Dr. I.
`Lindley, Department of Immunostimulation, Sandoz Forschung(cid:173)
`sinstitut Gesellschaft H.B.H., Vienna, Austria). Because both
`FMLP and rlL-8 were dissolved in dimethylsulfoxide (Serva, Hei(cid:173)
`delberg, Germany) and diluted in PBS, cells were also exposed to
`PBS with an equivalent concentration of dimethylsulfoxide. Phor(cid:173)
`bol myristate acetate (PMA) and dibutyryl(db) -cyclic adenosine
`monophosphate (cAMP) were obtained from Sigma.
`
`Preparation of Human Leukocytes Buffy coats prepared from
`blood from healthy donors were diluted in PBS (pH 7.4) and then
`subjected to Ficoll-amidotrizoate (p = 1.077 gjml; Pharmacia Inc.,
`UPf-sala, Sweden) density centrifugation at 440 X g for 20 min
`[10 . Granulocytes were purified from the Ficoll-Amidotrizoate
`pellets by dextran 1 X g sedimentation (Plasmasteril, Fresenius
`A.G., Bad Homberg, Germany) at 37°C for 10 min followed by
`removal of the remaining erythrocytes by a single hypotonic lysis.
`
`0022-202X/93/S06.00 Copyright © 1993 by The Sociery for Investigative Dermatology, Inc.
`
`37
`
`Sawai (IPR2019-00789), Ex. 1024, p. 001
`
`

`

`38 NIBBERING ET AL
`
`THE JOURNAL OF INV ESTIGATIVE DERMATOLOGY
`
`The final cell suspensions consisted of 97 ± 2% granulocytes. The
`viability of the cell suspensions exceeded 95%, as determined by
`trypan blue dye exclusion.
`
`Measurement of Polarization of Granulocytes The assay to
`determine polarization of granulocytes was performed as described
`[11] with minor modifications. Briefly, 400-fLl aliquots of a suspe.n(cid:173)
`sion of 1 X 106 granulocytes/ml RPMI 1640 (Flow LaboratorIes
`Ltd., Irvine, UK) were incubated with 100 fLl of the stimulus di(cid:173)
`luted in RPMI 1640 at 3r C for 15 min. Incubation was stopped by
`the addition of 500 fLl 9% v /v formaldehyde in PBS; the cells were
`maintained in this fixative until light microscopical determination
`of the percentage of cells that were not spherical. Two different
`investigators each analyzed 200 cells from each preparation.
`
`Chemotaxis Assay Chemotaxis of granulocytes was measured
`with the 48-well microchemotaxis chamber (Neuroprobe, Cabin
`John, MD) assay as described (12). In short, chemotaxins 0: PBS
`were placed in the lower compartments. Two filters were adjusted
`between the lower and upper compartments. The lower filter had a
`pore width of 0.45 fLm and the upper filter of 8 fLm. Granulocytes
`were placed in the upper compartment and, after incubation for
`1.5 h instead of 2.5 h at 37 °C, the filters were fixed, stained, and
`cleared with xylene. The number of granulocytes that had passed
`the upperfilter was determined by light microscopy (magnification
`X 400). The results are expressed as the number of cells per 10
`high-power fields (hpf).
`
`Measurement of Elastase Release by Granulocytes Granulo(cid:173)
`in 137 mM NaCI, 2.7 mM KCI, 8.1 mM
`cytes suspended
`Na2HP04, 0.9 mM CaCI2 , 0.49 mM MgClz, 1.5 mM KHzP0 4 ,
`and 2.5 mg bovine serum albumin (BSA)/ml at a concentration of
`5 X 106 cells/ml were incubated with the indicated stimuli at 37 °C
`for 30 min. The reaction was stopped by transferring the tubes to
`crushed ice. After centrifugation at 500 X g for 10 min, samples of
`the cell-free supernatants were assayed for elastase activity utili~ing
`substrate S2484 (Kabivitrum, Stockholm, Sweden) as descnbed
`[13J. The absorbance of the reaction product after 60 min was read at
`405 nm using a Titertek multiscan (EFLAB, Helsinki, Finland).
`Granulocytes were incubated with 5 fLg cytochalasin E {Serva)/ml
`PBS under slow rotation (4 rpm) at 37°C for 5 min, washed, and
`suspended to a concentration of 5 X 106 cells/ml.
`
`Measurement ofIntracellular Killing of Mycobacterium fortui(cid:173)
`tum by Granulocytes Preparation ~f M. JortuJtul1l (ATCC
`12790, American Type Culture CollectIOn, RockVille, MD) and
`the assay to determine the intracellular killing of this mycobacter(cid:173)
`ium by granulocytes has been described [14]. Briefly,S X 106 gran(cid:173)
`ulocytes/ml Hanks' balanced salt solution (HBSS) (Oxoid Ltd.,
`Basingstoke, UK) containing 0.01 % gelatin and 0.01 % Tween 80,
`further called HBSS-gel-Tw, and 5 X 106 M. Jortuitum/ml HBSS(cid:173)
`gel-Tw were mixed and incubated in the presence of 10% v /v
`human serum at 37°C for 10 min. The supernatant contammg
`extracellular bacteria was discarded, the granulocytes washed, and
`then 5 X 106 granulocytes containing bacteria were stimulated with
`10% serum with or without 200 fLM MMF at 37°C. One and two
`hours after stimulation samples were taken and, after lysis of the
`granulocytes, the number of viable cell-associated bacteria was ?e(cid:173)
`termined microbiologically. As control, granulocytes contammg
`bacteria were incubated with MMF or no stimulus.
`
`Measurement of the Oxygen Consumption by Granulo(cid:173)
`cytes Oxygen consumption by granulocytes at rest and after stim(cid:173)
`ulation was determined polarographically [15]. In short, 300 fLl of a
`suspension of 2.5 X 1 06leukocytes/ml RPMI 16~0 ~ere added .to a
`thermostated vessel (3rC), and the oxygen tenSIOn In the medIUm
`was measured with a Clark electrode. The rate of oxygen consump(cid:173)
`tion was calculated from the slope of the oxygen tension trace over a
`period of 5 to 10 min. ~o discriminate between.oxygen c~nsum~­
`tion by the NADPH OXidase complex and the mltochondnal respI(cid:173)
`ration, cells were also stimulated in the presence of 1 mM NaN 3' an
`inhibitor of mitochondrial respiration.
`
`Measurement of Superoxide Anion Production by Granu,
`locytes Superoxide anion (02-) production by cytochalasin E ~
`incubated granulocytes at rest and after stimulation was assessed by
`the superoxide dismutase-inhibitable reduction of ferricytochrom~
`c (type IV; Sigma) as described [1 6] with minor modifications.
`Briefly, the stimulus was added to 1 X 106 cells/ml PBS supple(cid:173)
`mented with 1.2 mM CaCI 2 , 1 mM MgClz, 5 mM glucose.
`20 mM Hepes, and 0.1 mM ferricytochrome c at 3rC and th~
`absorbance at 550 nm was measured every 30 seconds during a
`1 O-min period. The production of superoxide anion by the cells was
`calculated from a standard curve obtained with reduced cy(cid:173)
`tochrome.
`
`Measurement of the NADPH Oxidase Activity in a Cell-Free
`System NADPH oxidase activity was measured by means of a
`C lark oxygen electrode at 2rC as the rate of oxygen consumption
`by the cytosolic and membranous components of NADPH oxidase
`that were isolated from granulocytes as described [17]. Briefly,
`360 plIO mM Hepes (pH 7.0) containing 0.17 M sucrose, 75 roM
`NaCI, 0.5 mM ethyleneglycol-bis-(-amino-ethylether)N,N' -tetra(cid:173)
`acetic acid (EGTA), 1 mM MgCI2 , 2 mM NaN3 • and 10 ILM
`GTPyS (Calbiochem Corp, La Jolla, CA) were mixed with 10111
`membranous components and 10 fLl cytosolic components, each of
`which is the equivalent of 1 X 106 granulocytes. Assembly of the
`NADPH oxidase complex was initiated by adding sodium dodecyl(cid:173)
`sulfate at a final concentration of 100 fLM and closing the reaction
`vessel. After 3 min, NADPH atafinal concentration of250 pM was
`added and the oxygen consumption measured.
`
`Determination of the Binding of FLPEP-FITC to Granulo(cid:173)
`cytes by FACS Analysis The fluorescent probe formyl-Nle(cid:173)
`Leu-Phe-Nle-Tyr-Lys -fluorescein isothiocyanate (FLPEP-FITC;
`Molecular Probes Inc., Eugene, OR) was used to determine binding
`to the receptor as described [19]. Briefly, granulocytes at a concen(cid:173)
`tration of 2 X 1.06 cells/ml buffer (pH 7.4) were incubated with
`various concentrations of FLPEP-FITC in the dark at 4 0 C for 60
`min and then the fluorescence of the cells was measured on a
`FACScan (Becton and Dickinson, Mountainville, CA) equipped
`with a 15 m W argon-ion laser. For assessment of the non-specific
`binding of this probe, cells were incubated with 1 fLM FMLP at 4 ° C
`for 15 min prior to addition ofFLPEP-FITC. The results expressed
`as mean fluorescence intensity values are in arbitrary units (AU).
`
`Competition Binding Assay for cAMP At selected intervals
`after stimulation, 50 fLl of a suspension of 1 X 106 granulocytes/ml
`were stimulated at 3rc for various intervals and then 450 fLl ice(cid:173)
`cold I-propanol (Merck) were added to disrupt the cells. After soni(cid:173)
`cation for 5 min this material was centrifuged at 7000 X g for 1 min
`and then the supernatant was lyophilized. The residue was dissolved
`in 100 fLl distilled water and the cAMP content of these samples was
`determined as described previously [19]. In short, cAMP in the
`samples was allowed to compete with 5',8-f3H]-cAMP (specific
`activity 40-60 Ci/mmol; Amersham International, Amersham,
`Bucks, UK) for binding to sites in the bovine microsomal prepara(cid:173)
`tion. The cAMP content of these samples was calculated from a
`standard curve constructed in parallel with unlabeled cAMP (Amer.
`sham). The cAMP concentration was calculated from this value and
`the mean cell volume of the granulocytes.
`
`Measurement of the [Ca++]; in Granulocytes For measure(cid:173)
`ment of the [Ca-] i, 2 X 107 leukocytes/ml 20 mM Hepes supple(cid:173)
`mented with 138 mM NaCI, 6 mM ICCI, 1 mM MgS04' 1.1 m.M
`CaCI2 , 1 mM NaH2P04, 5.5 mM D-glucose, 0.1 mMEGTA, and
`0.1 % BSA (Ca- buffer, pH 7.4) were incubated with 2 JlM of the
`acetoxymethyl ester of FURA-2 (FURA-2/ AM; Sigma) in the dark
`at 3rc for 30 min. Next. the cells were washed with Ca++ buffer
`and then suspended in either Ca- buffer or Ca++ buffer without
`CaCl 2 but supplemented with 2 mM EGTA, further called Ca++·
`free buffer. Two millimeter aliquots of 5 X 106 cells/ml Ca++
`buffer or Ca++-free buffer were transferred to a quartz cuvette with
`magnetic stirring. The cell suspensions were excited alternately at
`340, 360, and 380 nm (slit width 2.5 11m) and emission was
`
`Sawai (IPR2019-00789), Ex. 1024, p. 002
`
`

`

`VOL. 101, NO.1 JULY 1993
`
`EFFECTS OF MMF ON GRANULOCYTES 39
`
`100
`
`80
`
`~ 60
`!!!
`Q;
`u
`U
`
`.~ 40
`'" "0
`
`20
`
`0
`
`a. 00
`
`0
`
`0.2
`
`20
`2
`MMF (pM)
`
`200
`
`500
`
`Figure 1. Effect of various concentrations of MMF on the polarization of
`granulocytes. Granulocytes were incubated for 15 min with concentrations
`ofMMF ranging from 0.2.uM to 500 .uM and the percentage cells that were
`not spherical was assessed. Results are means ± SEM; 11 = 3.
`
`recorded at 500 nm (slit width 5 nm) with a Perkin-Elmer 3000
`spectrofluorometer (Perkin-Elmer, Ueberlingen, Germany). The
`[Ca++] i was calculated from the ratio of the fluorescence excited at
`340 nm and at 380 nm as described [20]. Calibration of the FURA-
`2 fluorescence was performed by lysis of the cells with 0.1 % wtjv
`Triton X-I00 (Sigma) in the presence of 1.1 mM CaClz and the
`subsequent addition of 10 mM EGT A.
`Statistical Analysis The significance of the differences between
`the values found for cells exposed to stimuli and cells incubated with
`the vehicule was determined with the Mann-Whitney U test. p <
`0.05 was considered significant.
`
`RESULTS
`Induction of the Polarization and Chemotaxis and Elastase
`Release by Granulocytes by Various Stimuli MMF induced
`polarization of granulocytes in a dose-dependent fashion; the maxi(cid:173)
`mum percentage polarized cells was obtained with 200 .aM MMF
`(Fig 1). DMF but not DMM, FA, S-DMM, or R-DMM stimulated
`the polarization of granulocytes (Table I). About 90% of the granu(cid:173)
`locytes had a polarized morphology after stimulation with FMLP
`(Table I).
`The number of cells in the upper filter was slightly but signifi(cid:173)
`cantly higher when MMF and DMF was present in the lower com-
`
`partment of the chambers than when no stimulus was applied
`(Table I). The effect on the number of cells in the filter did not
`depend on the concentration of MMF (range 2 nM to 2 mM; results
`not shown). These results indicate that MMF and DMF are hardly
`chemotactic for granulocytes; FMLP was a potent chemotaxin for
`these cells (Table I). Addition of MMF to the upper compartment of
`the n1.icrochemotaxis chamber did not affect this response to FMLP.
`Granulocytes incubated with MMF for 5 min responded to FMLP as
`well as cells incubated with PBS (results not shown). MMF but not
`DMF, DMM, FA, S-DMM, or R-DMM stimulated granulocytes to
`release elastase (Table I). Stimulation by MMF was considerably less
`than that by rIL-8 or FMLP (Table I). MMF-incubated granulocytes
`released significantly less elastase after rIL-8 or FMLP stimulation
`than PBS-incubated cells.
`
`Intracellular Killing of M. fortuitum by Granulocytes The
`phagocytosis of M. fortuitul1I by granulocytes and the growth of this
`bacterium in HBSS - gel-Tw were not affected by MMF (results not
`shown). The intracellular killing of M. Jortllitum by granulocytes
`stimulated with serum was limited but addition of MMF enhanced
`the killing process significantly; the percentage of M. Jortuitum
`killed 2 h after stimulation with and without MMF was 60 ± 4.5%
`and 29 ± 3.5% (means ± SEM; n = 3), respectively. In the absence
`of serum, intracellular killing of M. Jortuitul1I by granulocytes did
`not occur and MMF had no effect. The cell-free medium, obtained
`after incubation for 2 h at 37"C of 5 X 106 granulocytes and 5 X
`106 M. Jortuitul1I in 1 111.1 HBSS-gel-Tw containing 10% serum
`with MMF, did not affect the growth of M. Jortuitum, indicating
`that no extracellular killing of bacteria occurred (results not shown).
`Stimulation of Oxygen Consumption and O 2 Production by
`Granulocytes with MMF and FMLP Oxygen consumption
`and O 2 production by resting granulocytes and by granulocytes
`after stimulation with various concentrations of MMF, FMLP, or
`PMA are shown in Table II. Both MMF and FMLP stimulated
`oxygen consumption by these cells. Optimal concentrations for
`MMF and FMLP were 200.aM and 100 11.M, respectively. No addi(cid:173)
`tive effect of MMF on the FMLP-stimulated oxygen consumption
`by granulocytes was found (results not shown). In the presence of
`NaN3 , an inhibitor of mitochondrial respiration, the stimulatory
`effect of MMF on oxygen consumption was almost completely
`abolished whereas that of FMLP was hardly affected (Table II),
`indicating that MMF does not activate the NADPH oxidase com(cid:173)
`plex. In agreement with this notion, MMF did not stimulate granu(cid:173)
`locytes to produce 0 2". This compound did not scavenge 02".
`To find out whether incubation with MMF suppresses the
`FMLP-stimulated oxygen consumption, granulocytes were ex(cid:173)
`posed to MMF in the presence ofNaN 3 , or as control to PBS, prior
`
`Table I. Effects of Various Stimuli on the Polarization and Chemotaxis of, Elastase Release by, and Increase in the [Ca++t in Granulocytes
`
`Stimulus'
`
`Polarized Cellsb (%)
`
`Chemotaxis'
`(cell/tO hpf)
`
`Elastase Released
`(00405 nm)
`
`[Ca++]i'
`(nM)
`
`18 ± 3f
`0.30 ± 0.01
`60±6
`None
`58 ± 5g
`91 ± 109
`0.43 ± O.Ols
`MMF
`40 ± 5:
`113 ± 9:
`0.30 ± 0.01
`OMF
`0.27 ± 0.01
`NO
`18 ± 3
`OMM
`0.30 ± 0.02
`FA
`12 ± 2
`NO
`0.34 ± 0.01
`NO
`19 ± 4
`S-OMM
`0.31 ± 0.01
`NO
`12 ± 4
`R-OMM
`NOb
`1.61 ± O.llg
`NO
`rIL-8
`88 ± 3g
`1129 ± 308g
`2.10 ± 0.27:
`FMLP
`• Final concentration ofMMF, OMF, and OMM was 200 tIM; of FA, S-OMM, and R-OMM 1 mM; of rlL-8 1 nM. The final concentration ofFMLP in the polarization assay, the
`chemotaxis assay and the elastase release assay was 10 nM and for the measurement of the [Ca++]; it was 100 nM.
`b Various stimuli were added to granulocytes and after 15 min the percentage of cells that were not spherical was assessed. n = 3.
`, Various stimuli were included in the lower compartment of the microchemotaxis chamber and after an incubation for 1.5 h at 37 "C the number of cells in the upper filter was
`determined light microscopically. The results expressed as the number of cells per 10 hpf. n = 3.
`, Various stimuli were incubated with granulocytes for 30 min; after the reaction was terminated, the cell-frec supcrnatant of this suspension was assayed for elastase activity. n = 3.
`l Various stimuli were added to FURA 2/AM-incubatcd granulocytes and the [Ca++]j was monitored. Data represent max iinuln [Ca++]j vaJucs. n = 4- 16.
`fResults are means ± SEM.
`t Indicates p < 0.05 for the difference between the values for stimu lated and control cells.
`b NO, not done.
`
`110 ± 6
`258 ± 14g
`235 ± 8g
`129 ± 9
`119±1
`74 ± 10
`85 ± 4
`450 ± 26·~
`618 ±38g
`
`Sawai (IPR2019-00789), Ex. 1024, p. 003
`
`

`

`40 NIBBERING ET AL
`
`THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
`
`Table ll. Oxygen Consumption and Superoxide Anion Production by Granulocytes
`at Rest and After Stimulation with MMF, FMLP, and PMAa
`
`Incubation
`
`Stimulusb
`
`None
`PMA
`MMF
`FMLP
`MMF
`FMLP
`FMLP
`
`Oxygen Consumption
`(nmol/[5 X 106 cells X min)
`3 ± 2' (6)
`25 ± 5J (6)
`18 ± 6J (6)
`16 ± Y (8)
`2 ± 1 (6)
`15±4J (3)
`5±1'(3)
`
`O2 Production
`(I1mol/[l X 106 cells X 10 mjn))
`6 ± 0.4 (7)
`51±4J(5)
`6 ± 0.4 (7)
`34 ± 1J (7)
`N[)f
`ND
`24 ± 0.4' (7)
`
`• Oxygen consumption due to mitochondrial respiration was inhibited with 1 mM NaN,. Where indicated, granulocytes were incubated with MMF together with NaN , for 15
`min before measurement of rhe FMLP-stimulated oxygen consumption and O. production.
`• Final concentrations of200)JM MMF, 100 nM FMLP, 100 ng/m1 PMA, and 1 mM NaN, .
`, Results are means ± SEM, with the number of experiments in parentheses.
`, Indicates P < 0.05 for the difference between the values for stimulated and control cells.
`, Indicates p < 0.05 for the difference between the values for MMF-incubated and control cells.
`I ND, not done.
`
`to stimulation with FMLP. The results showed that the respiratory
`burst in MMF-incubated cells was significantly lower than in con(cid:173)
`trol cells (Table II). Because it has been suggested that an increase in
`the cAMP concentration inhibits the oxidative metabolism of gran(cid:173)
`incubated granulocytes with 0.1 mM
`ulocytes [21,22], we
`db-cAMP for 5 min at 37°C before measurement of the FMLP(cid:173)
`stimulated 02" production. The results revealed that the FMLP(cid:173)
`stimulated production of 02" by db-cAMP-incubated granulocytes
`was 40 ± 9% (mean ± SEM) less (n = 6) than that by control gran(cid:173)
`ulocytes.
`Effect ofMMF on the NADPH Oxidase Activity in the Cell(cid:173)
`Free System To elucidate the mechanisms underlying the inhibi(cid:173)
`tion of the respiratory burst in MMF-incubated cells, we measured
`the effect of MMF on the NADPH oxidase activity in the cell-free
`system. For this purpose MMF, or as control PBS, was added to the
`mixture of cytosolic and membranous components of the NADPH
`oxidase compl.ex before the addition ofSDS. The results revealed no
`effect of MMF on the NADPH oxidase activity; the results of the
`addition of MMF and PBS were 8.4 ± 0.2 J.LM oxygen/min and
`8.2 ± 0.6 JLM oxygen/min (means ± SEM; n = 5), respectively.
`Effect ofMMF on the Binding ofFLPEP-FITC to Granulo(cid:173)
`cytes To find out whether the MMF-mediated inhibition of the
`respiratory burst in granulocytes was due to down-regulation of the
`FMLP-receptor, the effect of MMF on binding of FLPEP-FITC to
`these cells was determined. The results revealed no effect of expo(cid:173)
`sure of the cells to MMF for 15 min prior to the incubation with the
`fluorescent probe (Fig 2). Addition of MMF during the incubation
`with FLPEP-FITC was without effect as well (results not shown).
`
`Changes in the cAMP Concentration in Granulocytes After
`Stimulation with MMF and FMLP Because an increase in the
`cAMP concentration in cells is involved in the negative feedback
`processes in cells [24], we investigated whether MMF induces a rise
`in the cAMP concentration in granulocytes. The cAMP concentra(cid:173)
`tion in resting granulocytes was 0.72 ± 0.19 J.LM (mean ± SEM;
`n = 4). Stimulation of granulocytes with MMF resulted in a tran(cid:173)
`sient increase in the cAMP concentration to a maximum value of
`1.33 ± 0.14 J.LM (mean ± SEM; n = 4) 20 seconds after stimula(cid:173)
`tion, which returned to resting values within 1 min after addition of
`the stimulus. FMLP induced a threefold increase in the cAMP con(cid:173)
`centration in granulocytes with a maximum value of 2.67 ±
`0.60 J.LM (mean ± SEM; n = 4) 30 seconds after stimulation.
`Changes in the [Ca++}; in Granulocytes Induced by Various
`Stimuli The [Ca++] i in resting granulocytes amounted to 110 ±
`6 nM (mean ± SEM; Table I). Addition of2 f1.M to 20 mM MMF to
`granulocytes induced a dose-dependent increase in the [Ca++Ji;
`maximum values were found after stimulation with 200 JLM MMF
`(Fig. 3). The rise in the [Ca++L was already apparent within 5
`seconds of the addition of MMF and the [Ca++] i gradually returned
`
`to resting values within the next 2 - 3 min. The increase in the
`[Ca++] i after stimulation with MMF is due mainly to the release of
`Ca+t from intracellular stores. Similar kinetics for changes in the
`[Ca++]; were found for granulocytes stimulated with FMLP and
`rlL-8 (results not shown). Furthermore, the FMLP-induced in(cid:173)
`crease in the [Ca++]; in MMF-incubated and PBS-incubated granu(cid:173)
`locytes did not differ (results not shown), indicating that MMF did
`not affect signaling in granulocytes.
`To find out whether the effects of methylated fumarate deriva(cid:173)
`tives on granulocytes depend on the presence of methyl groups, the
`changes in the [Ca++] i in granulocytes upon stimulation with DMF,
`DMM, FA, S-DMM, or R-DMM were assessed. The results reo
`vealed that of these compounds only DMF induced an increase in
`the [Ca++]; (Table I).
`
`DISCUSSION
`The main conclusion that can be drawn from the present results is
`that the methylated fumarate derivatives MMF and DMF modulate
`several functional activities of granulocytes. These fumarate deriva(cid:173)
`tives are agonists of granulocytes because they stimulate the polar(cid:173)
`ization and elastase release by granulocytes and enhance the intra·
`cell ular killing of bacteria by these cells. Some of these effects might
`be mediated by the rise in the [Ca+t] i in granulocytes stimulated by
`these compounds, since Ca++ -specific ionophores also stimulate the
`polarization and elastase release by granulocytes [23 - 25]. MMF and
`DMF are not chemotactic for granulocytes or stimulate the respira-
`
`300
`
`200
`
`100
`
`::
`'iii
`c:
`J!!
`.!:
`
`'" " c: '" " '" ~
`
`0
`~
`
`25
`FLPEP (nM)
`
`SO
`
`Figure 2. Effect of MMF on the binding of FLPEP-FITC to granulocytes,
`Granulocytes were exposed to 200 liM MMF ~, PBS (0). or 1 liM FMLP
`(0) for 15 min before jncubation with variolls concentrations of FLPEP,
`FITe.' After a 60-min incubation with this probe, the fluorescence of th~
`cells was measured with a FACScan f1owcytometer. The results are given fot
`one representative experiment out of three experiments.
`
`Sawai (IPR2019-00789), Ex. 1024, p. 004
`
`

`

`VOL. 101, NO.1 JULY 1993
`
`300
`
`~ 200
`
`.s -. .
`
`'" ~ u_~
`
`100
`
`0
`
`0 0.002 0.02 0.2
`
`20
`
`MMF (mM)
`
`Figure 3. Effect of various concentrations of MMF on the [Ca++] i in granu(cid:173)
`locytes. For investigation of the effect of MMF on the [Ca++]., FURA
`2/ AM-incubated granulocytes were stimulated with 2,uM to 20 mM MMF
`and the changes in the [Ca++]. monitored. Data, i.e., means ± SEM, repre(cid:173)
`sent the peak values (at 10 seconds) after stimulation. n = 3.
`
`tory burst in these cells. On the contrary, MMF suppressed the
`FMLP-stimulated respiratory burst in these cells; in this respect
`methylated fumarate derivatives are antagonists of granulocytes.
`What are the mechanisms underlying the inhibitory effect of
`MMF on the FMLP-stimulated respiratory burst? It could be that
`MMF interferes with the formation of the NADPH oxidase com(cid:173)
`plex. This possibility can be excluded because we found no effect of
`MMF on the NADPH oxidase activity in the cell-free system. An(cid:173)
`other possibility is that MMF down-regulates the FMLP receptor.
`Our observation that MMF did not affect the binding of FLPEP(cid:173)
`FITC to granulocytes or the FMLP-induced increase in the [Ca++].
`excludes the possibility that it inhibits expression of the receptor for
`FMLP or the signaling through this receptor. It may be that cAPK
`interferes with the signaling, which arises from the interaction be(cid:173)
`rween FMLP and its receptor [21,22], and thus inhibits the respira(cid:173)
`tory burst in granulocytes after stimulation with this ligand. This
`possibility could be correct because MMF induced an increase in the
`cAMP concentration, which can inhibit 02-production by granu(cid:173)
`locytes [22].
`Our observation that MMF induces polarization of granulocytes
`without being chemotactic for these cells is not in agreement with
`the generally accepted view that polarization of phagocytes is
`closely correlated to chemotactic responsiveness. This discrepancy
`cannot yet be explained. An increased adherence of MMF-stimu(cid:173)
`lated granulocytes to glass or plastic surfaces, as described for IgG
`stimulation [26], has not been observed. It has been reported that
`activation of protein kinase C in lymphocytes by phorbol esters
`leads to a decreased reaction to chemotactic stimuli [27]. Because
`MMF does not activate this enzyme in granulocytes (unpublished
`observations), this cannot explain the observed discrepancy. It thus
`appears that MMF induces a shape change rather than being chemo(cid:173)
`tactic for granulocytes.
`Previous studies showed that human granulocytes efficiently
`phagocytose M.JortuitulII whereas subsequent killing of the ingested
`bacteria is limited [14] .§ Because intracellular killing ofM.JortuitulII
`by granulocytes activated by interferon-y [14] or stimulated by IL-
`8§ is enhanced compared to normal granulocytes, we used this
`bacterium to investigate whether MMF also augments the intracel(cid:173)
`lular killing of this bacterium by granulocytes.
`Interactions between methylated fumarate derivatives and granu(cid:173)
`locytes are specific because compounds with an ethyl group (mon(cid:173)
`oethylfumarate; unpublished observations) instead of a methyl
`group or without a methyl group, i.e., fumaric acid, did not stimu-
`
`§ Nibbering PH, Pos 0, Stevenhagen A, van Furth R: Interleukin-8
`enhances the non-oxidative intracellular killing of Mycobacterium fortui(cid:173)
`tum by human granulocytes (submitted for publication).
`
`EFFECTS OF MMF ON GRANULOCYTES 41
`
`late polarization, degranulation, or changes in the [Ca++] i in granu(cid:173)
`locytes. The most convincing evidence for the specificity of the
`binding site for methylated fumarate derivatives is our observation
`that DMF, but not its stereoisomer DMM, stimulates granulocytes.
`Furthermore, we did not observe a change in the [Ca++] i or oxygen
`consumption by purified monocytes after stimulation with MMF
`(unpublished observations), indicating that methylated fumarate
`derivatives react with cells in a selective fashion .
`The present results indicate that MMF suppresses the production
`and release of important mediators of the inflammatory process,
`such as reactive oxygen intermediates and elastase. In addition to
`these effects on granulocytes, MMF suppresses the proliferation of
`human cultured keratinocytes and SVK 14 keratinocyte cell-line
`cells (unpublished observations). The therapeutic effect of methyl(cid:173)
`ated fumarate derivatives in psoriasis can most probably be ex(cid:173)
`plained by its anti-inflammatory effects on granulocytes and its
`anti proliferative effects on keratinocytes.
`
`Tllis 1V0rk was 5Ilpported by a gratlt from the Kasswaars F,md, University Hospital
`Leidw, Tire Netlrerlands. We thank M. TIL IIa" dell Barselaar, A . Steve"hagen, atld
`C. Widijanto for tlreir ted'tlical assiSlarlCe a"d Dr. L. KOelldenllm, (Department of
`PlIlmo"ary Diseases, U"iversity Hospit