`
`www.multiplesclerosisjournal.com
`
`Cytokine production in T lymphocyte–microglia interaction is attenuated by
`glatiramer acetate: a mechanism for therapeutic efficacy in multiple
`sclerosis
`
`S Chabot1, FP Yong1, DM Le1, LM Metz1, T Myles1and VW Yong*,1,2
`1Department of Clinical Neurosciences, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada; 2Department of
`Oncology, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada
`
`The efficacy of glatiramer acetate in multiple sclerosis (MS) is thought to involve the production of Th2 regulatory lymphocytes that secrete
`anti-inflammatory cytokines; however, other mechanisms cannot be excluded. Given that activated T lymphocytes infiltrate into the CNS and
`become in close proximity to microglia, we evaluated whether glatiramer acetate affects the potential
`interaction between T cells and
`microglia. We report that the co-culture of activated T lymphocytes with microglia led to the induction of several cytokines, and that these
`were reduced by glatiramer acetate treatment. Morphological transformation of bipolar/ramified microglia into an activated ameboid form was
`attenuated by glatiramer acetate. These results reveal a novel mechanism for glatiramer acetate: the impairment of activated T cells to
`effectively interact with microglia to produce cytokines. The net result of a non-inflammatory milieu within the CNS,
`in spite of T cell
`infiltration, may help account for the amelioration of disease activity in MS patients on glatiramer acetate therapy.
`Multiple Sclerosis (2002) 8, 299–306
`
`Key words: cytokine; lymphocyte; microglia; neuroinflammation; therapeutics
`
`Introduction
`
`Glatiramer acetate (CopaxoneR) is a polymer of four amino
`acids (L-alanine, L-glutamate, L-lysine and L-tyrosine) with
`efficacy in the treatment of patients with relapsing-
`remitting multiple sclerosis (MS).1 ± 4 In an animal model
`of MS, experimental autoimmune encephalomyelitis
`(EAE), glatiramer acetate suppresses both the acute and
`chronic disease induced by several myelin proteins.5 ± 7
`The principal mechanism of action of glatiramer acetate in
`MS is thought to involve antigen presentation (reviewed
`in Refs. [8,9]). In this regard, glatiramer acetate has been
`shown to compete with, or displace, myelin peptides from
`binding to major histocompatibility complex (MHC) mole-
`cules; to act as an altered peptide ligand of immunogenic
`epitopes of myelin basic protein; or as an antigen itself 10 ± 15
`Glatiramer acetate has also been described as an antago-
`nist at the T-cell receptor to the immunodominant epitope
`of MBP.16 Consequences of affecting antigen presentation
`include the apoptosis or functional inactivation (anergy)
`of encephalitogenic T helper 1 (Th1) lymphocytes that
`mediate the disease, and the formation of glatiramer
`acetate-specific Th2 suppressor/regulatory cells that pro-
`duce anti-inflammatory cytokines. The latter are thought
`to traffic into the CNS and participate in the attenuation
`of disease through bystander suppression.14,15,17,18 In sup-
`
`*Correspondence: VW Yong, PhD, Neuroscience Research
`Group, 3330 Hospital Drive NW, Calgary, Alberta, Canada
`T2N 4N1.
`E-mail: vyong@ucalgary.ca
`Received 29 October 2001; accepted 11 December 2001
`
`© Arnold 2002
`
`port of these concepts, mononuclear cells from MS sub-
`jects treated with glatiramer acetate show evidence of
`a shift from a pro-inflammatory Th1 cytokine profile to a
`Th2 bias.15,19 ± 23 Furthermore, copaxone reactive Th2 cells
`are found to accumulate in the CNS of experimental
`animals.14
`It is possible that other mechanisms may also account
`for the efficacy of glatiramer acetate in MS. When
`T lymphocytes enter the parenchyma of the CNS, they
`become in close proximity to microglia. Although it
`is thought
`that T cells become reactivated when they
`re-encounter antigen presented by microglia, direct
`interaction between activated T cells and microglia, inde-
`pendent of antigen presentation, may also occur. In recent
`studies, we demonstrated that activated human T cells
`and microglia can interact in a contact-dependent manner,
`whether these were syngeneic or allogeneic, in an appa-
`rently non-antigen-dependent
`fashion. This interaction
`tumour necrosis factor-a
`yields substantial amounts of
`(TNF-a),24 IL-10,25 IL-Ib,
`IL-4, IL-6, IL-12 and IL-13.26
`Given that TNF-a can be toxic to oligodendrocytes,27 ± 29
`the encounter of activated T lymphocytes with microglia
`can lead potentially to the oligodendrocyte/myelin pathol-
`ogy that is characteristic of MS. The upregulation of a
`variety of cytokines would further propagate an inflam-
`matory milieu within the CNS.
`In this manuscript, we have addressed whether the
`production of cytokines in T cell ± microglia interaction
`would be affected by glatiramer acetate. If so, this would
`represent a novel mechanism of glatiramer acetate in
`MS.
`
`10.1191/1352458502ms810oa
`
`
`
`msj.sagepub.comDownloaded from
`
` at Infotrieve on November 3, 2015
`
`Page 1 of 8
`
`YEDA EXHIBIT NO. 2048
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`300
`
`Materials and methods
`
`Mechanisms of glatiramer acetate
`S Chabot et al
`
`Isolation of mononuclear cells and treatment with glatiramer
`acetate
`Heparinized blood was collected from normal volunteers,
`and subjected to Ficoll-Hypaque (Pharmacia Biotech,
`Mississauga, Ontario, Canada) centrifugation to obtain peri-
`pheral blood mononuclear cells as described previously.24
`After two washes, cells were suspended at a density of 1 ± 2
`million/ml in horizontal T-25 flasks (Nunc, Becton Dickin-
`son, Mississauga, Ontario, Canada) in the serum-free
`medium, AIM V (Gibco BRL, Burlington, Ontario, Canada),
`to which 1 mg/ml of an anti-CD3 antibody (OKT3) was
`added. Three hours after the anti-CD3 addition, the T-25
`flasks were stood upright from their horizontal position in
`order to kill monocytes that had adhered. Floating cells,
`which were mostly lymphocytes, were left for a period of 72 h
`at 37°C. Flow cytometry analysis of the mononuclear cell
`population at this 72-h period indicated that CD3+ cells
`constituted about 90% of the total cell population, with
`approximately 60% CD4+ and 30% CD8+ ratio. B lympho-
`cytes (CD19+) and NK cells (CD56+) consisted of 5 ± 6% of
`the total mononuclear cell population, and no monocytes
`(CD14+) were detected. Henceforth, given that the majority
`of cells in the mononuclear population are T cells, these will
`be referred to as T lymphocytes.
`In experiments where glatiramer acetate (from TEVA
`Pharmaceutical Industries, Petach-Tikva, Israel; batches
`242993097 and 242994498) was used, this drug (5 ± 50
`mg/ml), diluted in phosphate-buffered saline, was added
`to cultures 3 h after the initiation of CD3 ligation, at the
`time that the T-25 flasks were altered from the horizontal to
`upright positions. Cells were left for 69 h at 37°C, then
`collected, counted and resuspended in fresh AIM-V at a
`density of 500,000 cells/ml. Following a second treatment
`with glatiramer acetate, cells were left at 37°C for an
`additional 3 h. Thereafter, 100 ml (thus, 50,000 cells) of
`cell suspension was added to individual wells of a 96-well
`plate already containing microglia or U937 monocytoid
`cells (see below). Thus, unless otherwise stated, most
`experiments involved 72-h treatment of T cells with glatir-
`amer acetate, administered at two time points. The purity
`of T cells after 72 h of glatiramer acetate treatment was not
`different from that of non-treated controls (unpublished
`results).
`We have pretreated T cells with glatiramer acetate as
`this simulates the exposure of leukocytes to this drug at
`sites of subcutaneous injection and at draining lymph
`nodes. The literature does not provide evidence that glatir-
`amer acetate itself can enter the CNS. Note also that the
`concentrations of glatiramer acetate that are used here are
`comparable to those used by other laboratories in tissue
`culture studies.15,20 ± 23
`
`MicrogliaÐT cell interactions
`Adult human microglia of over 95% purity was isolated
`from the resected brain tissue of patients undergoing surgi-
`£
`cal resection to treat intractable epilepsy, as in previous
`reports.24,25 A total of 2.5
`104 microglia were plated per
`well of a 96-well plate. Microglia culture medium was
`
`minimum essential medium supplemented with 5% FCS,
`0.1% dextrose and 20 mg/ml gentamicin (all constituents
`from Gibco BRL). In microglia ± T cell interaction assays,
`100 ml containing 50,000 T cells in AIM-V (as described
`above) was added to individual wells of a 96-well plate
`already containing 25,000 microglia (or U937 monocytoid
`cells Ð see below) in 100 ml microglia culture medium.
`Twenty-four hours after, conditioned medium was collected
`for cytokine quantifications by ELISA.
`Fetal human microglia was isolated from brains obtained
`legal and therapeutic abortions using a protoc ol
`at
`described by Lee et al.30 Specimens ranged in gestational
`age from 14 to 20 weeks. The use of these and all other
`human materials has been approved by our local institu-
`tional ethics committee. Cells were used for interactions
`with T cells in a manner identical to that described for their
`adult counterparts.
`To complement the studies of primary human microglia
`cultures, we have used a human pro-monocytoid cell line.
`Cells in this U937 line become microglia-like, as assessed
`by morphology and expression of cell surface molecules,
`when treated sequentially with 50 ng/ml phorbol-12-
`myristate-13-acetate (PMA) (time 0 ± 48 h) and 100 U/ml
`interferon-g (IFNg ) (from 48 to 72 h).31 Cells were used
`1 ± 3 days after the IFNg treatment. As with microglia,
`50,000 T cells in AIM-V was added to individual wells of a
`96-well plate already containing 25,000 PMA/IFNg -treated
`U937 cells, and conditioned medium was collected after
`24 h of co-cultures.
`
`Cytokine protein quantification
`Cytokine protein levels in the conditioned medium of
`microglia ± T cell co-cultures were measured using enzyme-
`linked immunoabsorbent assay (ELISA) kits bought from
`BioSource International (Montreal, Quebec, Canada). Assays
`were performed following manufacturer instructions.
`
`Figure 1 Glatiramer acetate pretreatment of T cells suppresses TNF
`a and IL 10 production that is generated in T cell adult human
`microglia interactions. T cells or adult microglia in isolation produce
`undetectable TNF a or IL 10. In co culture, both IL 10 and TNF a
`levels are significantly elevated and this is reduced dose dependently
`by glatiramer acetate (GA) pretreatment of T cells. Values are
`mean‹SEM of triplicate analyses; results have been reproduced in
`three series of experiments involving different microglia and blood
`donors. **p<0.01, ***p<0.001, one way ANOVA with Bonferroni
`post hoc comparisons
`
`Multiple Sclerosis
`
`Downloaded from
`
`msj.sagepub.com
`
` at Infotrieve on November 3, 2015
`
`Page 2 of 8
`
`YEDA EXHIBIT NO. 2048
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Mechanisms of glatiramer acetate
`S Chabot et al
`
`301
`
`with 0.2% gelatin, which evaluates b2-integrin adhesive
`function.32 Coating was performed with 100 ml per well at
`37°C overnight. The fluid was then removed, the wells were
`washed once with saline, and 50,000 T cells in 100 ml AIM-V
`medium were then added. These T cells had been pretreated
`for 3 days with glatiramer acetate after the anti-CD3 ligation
`as described above, or were untreated controls. After 1 h at
`37°C, the fluid from each well was removed and processed
`through a Coulter Z1 counter to enumerate the number of
`unadhered cells. Adherent cells were floated by 0.25%
`trypsin and similarly counted. Thus, the total number of
`T cells was elucidated from each well and the proportion
`that remained floating was tabulated.
`
`Figure 2 Glatiramer acetate treatment of T cells also lead to the
`suppression of IL 1b, IL 6 and IL 12 in T cell adult human microglia
`interactions. Both activated T cells and adult human microglia do not
`secrete detectable amounts of IL 1b and IL 12 into their culture
`medium. With co culture, levels of IL 1b and IL 12 are increase, to 17
`and 114 pg/ml, respectively. These inducible cytokines are dose
`dependently reduced by glatiramer acetate pretreatment of T cells. IL
`6 is a constitutively expressed cytokine that can be readily detected in
`culture medium of T cells or microglia in isolation. Glatiramer acetate
`treatment of T cells also attenuated the IL 6 levels found in T cell
`microglia co cultures (level of IL 6 in control T cell microglia co
`culture in this example: 368 pg/ml). *p<0.05, **p<0.01,
`***p<0.001, one way ANOVA with Bonferroni post hoc comparisons
`
`Adhesion assays
`Wells in 96-well plates were coated with 20 mg/ml fibronec-
`tin (Sigma, St. Louis, MO, USA), a b1-integrin substrate, or
`
`Results
`
`Glatiramer acetate decreases cytokine levels in co-cultures of
`T cells and adult human microglia
`When cultured in isolation,
`the medium of anti-CD3
`activated human T cells or adult human microglia con-
`tained undetectable amounts of TNF-a or IL-10 as assayed
`by ELISAs. In co-culture of T cells and adult human
`microglia, however, substantial levels of TNF-a and IL-10
`were elicited (Figure 1). Significantly, the pretreatment of
`T cells with glatiramer acetate prior to their encounter
`with adult microglia reduced, in a concentration-dependent
`manner, the levels of both cytokines (Figure 1). The effect of
`glatiramer acetate was principally on T cells since the
`pretreatment of microglia with glatiramer acetate (25 mg/ml,
`
`Figure 3 Morphology of microglia in T cell adult microglia co cultures. Adult human microglia are mostly bipolar in morphology in culture
`(A). T cells are present as single cells or clumps when plated into wells of a 96 well plate (B). When T cells are co cultured with microglia in the
`absence of glatiramer acetate, bipolar microglia become rounded/ameboid in morphology (C, some microglia are shown by arrows). This
`£
`morphological transformation is prevented by glatiramer acetate pretreatment of T cells (D, with some bipolar microglia indicated by arrows).
`All frames are of the same original magnification,
`400
`
`Downloaded from
`
`msj.sagepub.com
`
` at Infotrieve on November 3, 2015
`
`Multiple Sclerosis
`
`Page 3 of 8
`
`YEDA EXHIBIT NO. 2048
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`302
`
`Mechanisms of glatiramer acetate
`S Chabot et al
`
`co-culture of anti-CD3 activated T cells with adult micro-
`glia, IL-1b and IL-12 became detectable while levels of IL-6
`remained unaltered from their high basal amounts. Glatir-
`amer acetate pretreatment of T cells dose dependently
`attenuated the induction of IL-1b and IL-12 in T cell ±
`microglia co-cultures, and also suppressed the amounts of
`the constitutively expressed cytokine, IL-6 (Figure 2).
`We noted that when microglia encounter activated
`T cells, the morphology of microglia transforms from a
`ramified/bipolar morphology to an amoeboid rounded form
`(Figure 3). This is reminiscent of an activated microglia
`in vivo, which transforms progressively from a ramified
`resting morphology to an amoeboid form.36 However, when
`T cells were pretreated with glatiramer acetate (25 mg/ml),
`the morphological transformation of microglia in T cell ±
`microglia co-culture was attenuated. Overall, the lack of a
`morphological transformation of microglia is another indi-
`cation that glatiramer acetate pretreatment of T cells results
`in their decreased ability to interact with microglia to
`produce cytokines.
`
`Adhesion of T cells is not altered by glatiramer acetate
`A possible explanation of the reduction of cytokine produc-
`tion by glatiramer acetate in T cell ±microglia co-culture is
`that glatiramer-acetate-treated T cells were less able to
`adhere onto microglia compared to control T cells. Although
`this did not appear qualitatively to be the case as observed
`by examination of live cultures (Figure 3), we utilized an
`adhesion assay to test this possibility. Figure 4 demonstrates
`that glatiramer-acetate-pretreated T cells did not have a
`reduced tendency to adhere onto fibronectin or gelatin,
`which evaluated b1- and b2-integrin interaction, respec-
`tively.32 Similarly, the adhesion of T cells onto a monolayer
`of PMA/IFNg -treated U937 cells, a model of microglia (see
`next section), was not affected by glatiramer acetate (unpub-
`lished observations).
`
`Figure 5 Cytokine production in co cultures of T cells with fetal
`human microglia is reduced by glatiramer acetate. As with their
`adult counterparts,
`fetal human microglia in isolation produce
`negligible quantities of IL 10 and TNF a. In co cultures of T cells with
`microglia, increases of IL 10 and TNF a are elicited and these are
`reduced by the pretreatment of T cells with glatiramer acetate. Values
`are mean‹SEM of three analyses. *p<0.05, **p<0.01, ***p<0.001,
`1 way ANOVA with Bonferroni post hoc comparisons
`
`Figure 4 The adhesive property of activated T cells is not altered by
`glatiramer acetate. Following 3 days of treatment of activated T cells
`with glatiramer acetate, 50,000 T cells were plated onto fibronectin
`(to evaluate b1 integrin function) or gelatin (b2 integrin) for 1 h. The
`proportion of cells that remained floating was enumerated and
`expressed as a percentage of the total adherent and floating cell
`population. None of the concentrations of glatiramer acetate affected
`adhesion when compared to controls (mean‹SEM of triplicates).
`This result was reproduced in three different sets of experiments
`
`for 1 ±3 d) did not influence cytokine production in sub-
`sequent T cell ± microglia interactions (results not shown).
`The effect of glatiramer acetate in reducing cytokine
`production in T cell ± adult microglia co-cultures is not
`the result of a decrease in the proliferation or survival of
`T cells. When the total number of T cells was counted after
`£
`72 h of glatiramer acetate treatment, cell numbers were
`103) versus glatiramer ace-
`comparable in control (24‹2
`tate (5, 25 and 50 mg/ml) groups (26‹1, 26‹2 and 22‹1,
`£
`103). It should be noted that while glatiramer
`respectively,
`acetate does affect the proliferation of lymphocytes, this
`generally requires a longer term of exposure (over 5 days) to
`drug in vitro.15,21,33 Furthermore, dye exclusion assays
`revealed no overt toxicity to T cells after 3 days of glatira-
`mer acetate treatment (unpublished observations). Finally,
`equal number of T cells was added to microglia in all test
`situations.
`We have addressed other features that could be impor-
`tant determinants for understanding the mechanisms by
`which glatiramer acetate affects T cell ±microglia interac-
`tions. First, T cells had to be activated with anti-CD3
`antibody since co-cultures of unactivated T cells (even in
`the presence of 50 U/ml IL-2) with microglia did not result
`in increased production of TNF-a. Moreover, it was neces-
`sary for T cells to be pretreated with glatiramer acetate
`since its reducing effect on cytokine production did not
`occur when added at the time of co-culture. Indeed, our
`current results indicate that the pretreatment period of T
`cells should be at least 24 h before these cells could alter
`cytokine production in T cell ± microglia interaction
`(unpublished observations).
`We evaluated further the expression of other cytokines
`implicated in MS, specifically IL-1b, IL-6 and IL-12.28,34,35
`Levels of IL-1b and IL-12 were undetectable in the condi-
`tioned media of T cells and adult microglia in isolation,
`while IL-6 was present in substantial amounts (between 200
`to 1000 pg/ml in six different sets of cultures). Following the
`
`Multiple Sclerosis
`
`Downloaded from
`
`msj.sagepub.com
`
` at Infotrieve on November 3, 2015
`
`Page 4 of 8
`
`YEDA EXHIBIT NO. 2048
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Mechanisms of glatiramer acetate
`S Chabot et al
`
`303
`
`mechanism of action of glatiramer acetate in MS. By an
`effect on antigen presentation, glatiramer acetate is thought
`to cause the production of Th2 suppressor/regulatory cells,
`which then migrate into the CNS. Within the CNS, and
`upon antigen restimulation, the glatiramer acetate-specific
`cells produce anti-inflammatory Th2 cytokines, including
`IL-5, IL-10 and IL-13, which are thought to inhibit the
`expansion of autoreactive Th1 cells, through a process
`termed bystander suppression (Figure 7).
`The experiments of this manuscript were aimed at
`uncovering other mechanisms by which glatiramer acetate
`may work in MS.
`Activated T cells traffic into the CNS as part of the
`immunological surveillance mechanism of the CNS. It is
`thought that those that recognize antigens are expanded and
`retained within the CNS while those that do not either
`undergo apoptosis or leave the CNS. In MS, increased
`numbers of activated T cells transit into the CNS paren-
`chyma.34,37 EAE experiments have revealed that while the
`initial wave of T cells that enters the brain is auto-antigen
`specific, the majority (over 98%) of the later arriving acti-
`vated T cells have many different specificities; 39,40 similar
`kinetics are thought to occur in MS. In view of the large
`number of activated T cells of many specificities that enters
`the CNS, one might consider whether these can interact
`with CNS constituents in a promiscuous non-antigen-
`dependent manner to produce cytokines. In support, it
`was noted that in a facial nerve resection model in mouse,
`T cells infiltrated into the CNS and aggregated around
`microglia, and this was correspondent with an increase in
`IL-1b and TNF-a.41 In graft-versus-host disease, activated
`microglia cell clusters are invariably associated with infil-
`trating T cell blasts.42 Futhermore, microglia isolated from
`the brains of healthy adult mice stimulated the differentia-
`tion of naõÈve T cells into Th1 effector cells without affecting
`T-cell proliferation. 43
`The current model that polyclonally activated human T
`cells can interact with human microglia to upregulate a
`variety of inflammatory cytokines further sheds a new
`perspective on immune propagation within the CNS. Cyto-
`kine production in this T cell ± microglia interaction occurs
`in the absence of any identifiable antigen, whether or not
`the microglia and T cells are MHC matched or mis-
`matched.24 TNF-a is produced predominantly by microglia,
`whereas IL-10 is produced by both cell types in the co-
`culture.25 Ligand pairs that modulate IL-10 and TNF-a
`production in T cell ± microglia interaction were found to
`include B7/CTLA-4:CD28, VLA-4:VCAM-1, CD40:CD40L
`and CD23:CD21/CD11.25
`Thus, when activated T cells enter the CNS, they have
`the capacity to engage microglia to produce a variety of
`cytokines to promote CNS inflammation. In this manu-
`script, we have expanded the disease relevance of the T
`cell ± microglia interaction by demonstrating that T cells
`pretreated with glatiramer acetate have a decreased capacity
`to evoke cytokine production in T cell ± microglia co-culture.
`Indeed, all cytokines tested were reduced in a dose-depend-
`ent manner by glatiramer acetate.
`At present,
`the mechanisms by which glatiramer
`acetate affect cytokine production in T cell ± microglia
`interactions are unknown. Some clues may be found in
`
`Figure 6 Glatiramer acetate on cytokine production in co cultures
`of T lymphocytes with PMA/IFNg treated U937 cells. In co culture of
`PMA/IFNg treated U937 cells with T lymphocytes,
`the levels
`of cytokines induced after 24 h were as follows: 969 pg/ml for
`TNF a, 13 pg/ml for IL 4, 667 pg/ml for IL 10, 20 pg/ml for IL 12, and
`51 pg/ml for IL 13. Values are mean‹SEM of triplicate cultures,
`and all results have been expressed as a percentage of control T cell
`microglia cultures (i.e., without glatiramer acetate)
`
`Glatiramer acetate in other models of T cell ±microglia/
`macrophage interactions
`As with their adult counterparts, fetal human microglia in
`isolation do not secrete detectable amounts of IL-10 or TNF-
`a into the culture medium. When fetal human microglia
`were co-cultured with activated T cells, however, significant
`amounts of IL-10 and TNF-a were induced. With glatiramer
`acetate pretreatment of T cells, the resultant IL-10 and TNF-
`a produced in T cell ± microglia co-cultures was reduced in
`a concentration-dependent manner (Figure 5).
`U937 is a human pro-monocytoid cell line that assumes a
`bipolar microglia/macrophage-like morphology when
`sequentially treated with PMA and IFNg . Indeed the PMA/
`IFNg -treated U937 assumes several characteristics of micro-
`glia particularly with respect to cytokine production.31 In
`this manuscript, we demonstrate that the PMA/IFNg -treated
`U937 cells have low to negligible amounts of IL-10, IL-12
`and TNF-a, but that their co-culture with anti-CD3 activated
`T cells resulted in the substantial increase in levels of these
`cytokines. Similarly, two Th2 cytokines, IL-4 and IL-13, were
`also induced in co-cultures of T cells with PMA/IFNg -treated
`U937 cells. Significantly, the pretreatment of T cells with
`glatiramer acetate prior to their encounter with PMA/IFNg -
`treated U937 cells reduced, in a concentration-dependent
`manner, the levels of all these cytokines (Figure 6).
`
`Discussion
`
`The treatment of patients with MS has improved signifi-
`cantly in the past few years, with drugs such as glatiramer
`acetate and IFNb having a favourable impact on the clinical
`course of the disease. Despite this, the mechanisms of how
`these drugs work in MS are not fully elucidated.29,37 Under-
`standing the modes of action of current MS drugs and
`uncovering their critical targets will lead to a more rational
`approach to design better therapeutics.
`A spate of recent papers15,16,20,22,23,38 has provided sup-
`port to an older literature6,10,11,13,17,18 on the proposed
`
`Downloaded from
`
`msj.sagepub.com
`
` at Infotrieve on November 3, 2015
`
`Multiple Sclerosis
`
`Page 5 of 8
`
`YEDA EXHIBIT NO. 2048
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`304
`
`Mechanisms of glatiramer acetate
`S Chabot et al
`
`Figure 7 Proposed mechanisms of glatiramer acetate (CopaxoneR) within the CNS. We suggest two mechanisms of action of glatiramer acetate
`within the CNS. The first model of bystander suppression has been suggested by the results of several groups (see text) whereby copaxone
`specific Th2 cells, with degenerate T cell receptor (TCR), are generated in the periphery following interference with antigen presentation. These
`cells migrate into the CNS and are reactivated upon encounter with antigens, including cross reactive myelin basic protein (MBP) peptides,
`presented by antigen presenting cells such as microglia. The resultant production of Th2 anti inflammatory cytokines then negatively regulates
`the expansion of autoreactive Th1 cells through bystander suppression. The result of this model is increased Th2 cytokines within the CNS, and
`decreased Th1 cytokines. The second model of copaxone activity stems from the results of this manuscript, where copaxone modulates the
`properties of Th1 and Th2 cells. When these traffic into the CNS, they are less able to engage microglia to produce cytokines. The result is
`the decrease of both Th1 and Th2 cytokines within the CNS. Both models likely coexist to produce the net result of decreasing inflammation
`within the CNS
`
`the observations that T cells, rather than microglia, have
`to be pretreated with glatiramer acetate, and that a period
`of at least 24 h of pretreatment is necessary. While this
`indicates that some properties of the T cells are being
`altered by glatiramer acetate, the ligand pairs cited earlier
`to modulate cytokine production25 were not altered by
`glatiramer acetate treatment, at
`least as assessed on
`expression levels by flow cytometry (unpublished ob-
`servations). Also,
`the lack of cytokine production did
`
`not appear to be an effect of glatiramer acetate on the
`adhesion of T cells (Figure 4). Others have noted that
`glatiramer acetate did not affect the adhesiveness of brain
`endothelial cells in vitro.44 Because the mechanism of the
`glatiramer acetate effect herein is still unknown, our
`preference is to refer to glatiramer-acetate-treated T lym-
`phocytes as `modulated’ T cells (Figure 7). Since both Th1
`(IL-1b, IL-12, TNF-a) and Th2 (IL-4, IL-6, IL-10, IL-13)
`cytokines are downregulated, a presumption is that both
`
`Multiple Sclerosis
`
`Downloaded from
`
`msj.sagepub.com
`
` at Infotrieve on November 3, 2015
`
`Page 6 of 8
`
`YEDA EXHIBIT NO. 2048
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Mechanisms of glatiramer acetate
`S Chabot et al
`
`305
`
`the Th1 and Th2 cells are modulated by glatiramer acetate
`to have a reduced capacity to interact with microglia.
`It should be noted that there has been very spare
`information on the pharmacokinetic properties of glatir-
`amer acetate. Because radiolabelled glatiramer acetate is
`rapidly degraded into smaller molecular weight fragments
`following subcutaneous injections in animals, the serum
`concentration of this drug is presumed to be low or non-
`detectable following clinical doses in man.45 Thus, one
`should consider whether the exposure of T cells with
`glatiramer acetate in this study is physiologically relevant.
`We propose that it is, since T cells in proximity to the
`subcutaneous injection sites, including those that are at
`draining lymph nodes, would be exposed to a reasonable
`concentration of glatiramer acetate. Also, the concentrations
`of glatiramer acetate in this study are similar to those
`employed in other laboratories in order to polarize glatir-
`amer-acetate-specific T cells.15,19,21 ± 23
`The modulation of T cell ± microglia engagement may be
`an important target for MS drugs. We have noted that IFNb-
`pretreated T cells are also altered in their ability to engage
`microglia to produce cytokines. Differences do exist, how-
`ever, since, unlike the diminution of all cytokines noted in
`this study, the interaction of IFNb-treated T cells with
`microglia led to the upregulation of IL-10 while diminishing
`IL-1b, IL-4, IL-12 and IL-13. In contrast, IL-6 remained
`unchanged.26
`In summary, our findings have revealed a novel mecha-
`nism of glatiramer acetate in the CNS. We suggest that when
`T cells infiltrate the CNS of MS patients on glatiramer
`acetate therapy, the `modulated’ T cells are unable to engage
`microglia to promote inflammation within the CNS; a con-
`sequence is the reduction of both Th1 and Th2 cytokines in
`the CNS (Figure 7). This mechanism does not exclude the
`more established model of the generation of copaxone-
`reactive Th2 lines, which traffic into the CNS to produce
`bystander suppression of autoreactive Th1 cells (Figure 7).
`With both mechanisms ongoing, the net result of glatiramer
`acetate treatment would be the dampening of inflammation
`within the CNS.
`
`Acknowledgements
`This project was supported by a research grant provided
`by TEVA Pharmaceutical Industries, Israel. We are grate-
`ful
`for the strong support of Dr Nora Tarcic, Project
`Manager, Autoimmune Section, TEVA Pharmaceutical
`Industries, Israel. We thank all the nursing staff in the
`operating theatre who coordinated the transfer of excised
`adult human brain samples to the Yong laboratory.
`
`References
`
`1 Bornstein MB, Miller A, Slagle S, et al. A pilot trial of Cop 1 in
`exacerbating remitting multiple sclerosis. N Engl J Med 1987;
`317: 408 14.
`2 Johnson KP, Brooks BR, Cohen JA, et al. Copolymer 1 reduces
`relapse rate and improves disability in relapsing remitting mul
`tiple sclerosis: results of a phase III multicenter double blind
`placebo controlled trial. The Copolymer 1 Multiple Sclerosis
`Study Group. Neurology 1995; 45: 1268 76.
`
`3 Johnson KP, Brooks BR, Ford CC, et al. Sustained clinical bene
`fits of glatiramer acetate in relapsing multiple sclerosis patients
`observed for 6 years. Mult Scler 2000; 6: 255 66.
`4 Wolinsky JS, Narayana PA, Johnson KP and the Copolymer 1
`Multiple Sclerosis Study Group and the MRI Analysis Center.
`United States open label glatiramer acetate extension trial for
`relapsing multiple sclerosis: MRI and clinical correlates. Mult
`Scler 2001; 7: 33 41.
`5 Ben Nun A, Mendel I, Bakimer R, et al. The autoimmune reac
`tivity to myelin oligodendrocyte glycoprotein (MOG) in multiple
`sclerosis is potentially pathogenic: effect of copolymer 1 on
`MOG induced disease. J Neurol 1996; 243: S14 22.
`6 Teitelbaum D, Fridkis Hareli M, Arnon R, Sela M. Copolymer 1
`inhibits chronic relapsing experimental allergic encephalomye
`litis induced by proteolipid protein (PLP) peptides in mice and
`interferes with PLP specific T cell responses. J Neuroimmunol
`1996; 64: 209 17.
`7 Teitelbaum D, Arnon R, Sela M. Immunomod ulation of
`experimental autoimmune encephalomyelitis by oral admin
`istration of copolymer 1. Proc Natl Acad Sci USA 1999; 96:
`3842 47.
`8 Neuhaus O, Farina C, Yassouridis A, et al. Multiple sclerosis:
`comparison of copolymer 1 reactive T cell lines from treated
`and untreated subjects reveals cytokine shift from T helper
`1 to T helper 2 cells. Proc Natl Acad Sci USA 2000; 97:
`7452 57.
`9 Sela M, Teitelbaum D. Glatiramer acetate in the treatment