`Vol. 94, pp. 10821–10826, September 1997
`Immunology
`
`Copolymer 1 induces T cells of the T helper type 2 that crossreact
`with myelin basic protein and suppress experimental
`autoimmune encephalomyelitis
`(immunoregulation兾multiple sclerosis兾cytokines)
`
`RINA AHARONI, DVORA TEITELBAUM, MICHAEL SELA*, AND RUTH ARNON
`Department of Immunology, The Weizmann Institute of Science, Rehovot 76100, Israel
`
`Contributed by Michael Sela, August 6, 1997
`
`The synthetic amino acid copolymer copoly-
`ABSTRACT
`mer 1 (Cop 1) suppresses experimental autoimmune enceph-
`alomyelitis (EAE) and is beneficial in multiple sclerosis. To
`further understand Cop 1 suppressive activity, we studied the
`cytokine secretion profile of various Cop 1-induced T cell lines
`and clones. Unlike T cell lines induced by myelin basic protein
`(MBP), which secreted either T cell helper type 1 (Th1) or both
`Th1 and Th2 cytokines, the T cell lines兾clones induced by Cop
`1 showed a progressively polarized development toward the
`Th2 pathway, until they completely lost the ability to secrete
`Th1 cytokines. Our findings indicate that the polarization of
`the Cop 1-induced lines did not result from the immunization
`vehicle or the in vitro growing conditions, but rather from the
`tendency of Cop 1 to preferentially induce a Th2 response. The
`response of all of the Cop 1 specific lines兾clones, which were
`originated in the (SJL兾JⴛBALB兾c)F1 hybrids, was restricted
`to the BALB兾c parental haplotype. Even though the Cop
`1-induced T cells had not been exposed to the autoantigen
`MBP, they crossreacted with MBP by secretion of interleukin
`(IL)-4, IL-6, and IL-10. Administration of these T cells in vivo
`resulted in suppression of EAE induced by whole mouse spinal
`cord homogenate,
`in which several autoantigens may be
`involved. Secretion of anti-inflammatory cytokines by Cop
`1-induced suppressor cells, in response to either Cop 1 or
`MBP, may explain the therapeutic effect of Cop 1 in EAE and
`in multiple sclerosis.
`
`The existence of T lymphocyte subsets that produce distinct
`sets of cytokines and, as a result, perform distinct effector
`functions, has been well established in both mice and humans
`(1–3). Naive CD4⫹ lymphocytes (Th0) triggered by antigen
`differentiate either into T helper type 1 (Th1) or into T helper
`type 2 (Th2) cells that can crossregulate one another; their
`respective cytokines act antagonistically. Th1 cells produce
`interleukin (IL)-2, interferon (IFN)-␥, and tumor necrosis
`factor, activate cell-mediated immunity, and induce delayed-
`type hypersensitivity. Th2 cells produce IL-4, IL-5, IL-6, and
`IL-10 and down-regulate cell-mediated immunity. During the
`last few years many studies have implicated the preferential
`development and activation of Th1兾Th2 subpopulations in a
`variety of autoimmune diseases (3–5). Experimental autoim-
`mune encephalomyelitis (EAE) is an inflammatory autoim-
`mune disease of the central nervous system (CNS) that serves
`as an animal model for multiple sclerosis (MS) (6). EAE can
`be induced in several animal species by immunization with
`myelin basic protein (MBP) (6), proteolipid protein (7), or
`myelin oligodendrocyte glycoprotein (8). During the progres-
`sion of EAE, Th1, and not Th2, cytokines are present in the
`
`The publication costs of this article were defrayed in part by page charge
`payment. This article must therefore be hereby marked ‘‘advertisement’’ in
`accordance with 18 U.S.C. §1734 solely to indicate this fact.
`© 1997 by The National Academy of Sciences 0027-8424兾97兾9410821-6$2.00兾0
`PNAS is available online at http:兾兾www.pnas.org.
`
`inflammatory lesions in the CNS (4, 9). The autoreactive T
`cells that induce the disease generally display a Th1 phenotype,
`and the adoptive transfer of these Th1 cells is sufficient to
`induce EAE (4–6, 10, 11). On the other hand, recovery from
`EAE is associated with an elevation of Th2 cells and cytokines
`in the CNS (4, 12). Regulatory T cells that suppress the
`development of EAE produce cytokines that correspond to the
`Th2 profile and mediate their activity by secretion of these
`suppressive cytokines (4, 13, 14). These findings, along with the
`observation that Th2 cytokines can inhibit the action of Th1
`cytokines (3, 4, 11), suggest that the induction and activation
`of Th2 cells may potentially prevent EAE and other autoim-
`mune diseases that are mediated by Th1 cells.
`Copolymer 1 (Cop 1) is a synthetic random copolymer of
`amino acids composed of L-Ala, L-Glu, L-Lys, and L-Tyr (15,
`16). Cop 1 exerts a marked suppressive and protective effect on
`EAE in various animal species, including primates, and on
`chronic relapsing EAE. Cop 1 also was shown to slow the
`progression of disability and to reduce the relapse rate in MS
`patients (17), and it recently was approved as a drug for MS
`under the trade name of Copaxone. The mechanism of Cop 1
`activity in EAE and MS involves high-affinity promiscuous
`binding to various class II major histocompatibility complex
`(MHC) molecules (18). This efficient MHC binding results in
`both competition with myelin antigens for T cell activation and
`induction of specific regulatory T cells. We previously have
`demonstrated that the unresponsiveness to EAE induced by
`Cop 1 is regulated by T suppressor (Ts) cells, because it can be
`adoptively transferred to normal recipients and abrogated by
`pretreatment with cyclophosphamide (19). Cop 1-specific Ts
`cell lines and Ts hybridomas were established from spleens of
`mice that had been rendered unresponsive to EAE by Cop 1
`(20). These Cop 1-induced suppressor cells, or their superna-
`tants, inhibited the in vitro response of an encephalitogenic line
`to MBP when cocultured. Furthermore, they prevented the
`development of EAE induced by whole mouse spinal cord
`homogenate (MSCH) in vivo. To further understand the
`mechanism of Cop 1 activity, we studied the cytokine secretion
`profile of various Cop 1-induced T cell lines and clones in
`response to Cop 1, as well as to the autoantigen MBP. In the
`following we demonstrate that Cop 1 induces CD4⫹ T cell
`lines兾clones of
`the Th2 subtype, which secrete anti-
`inflammatory Th2 cytokines in response to either Cop 1 or
`MBP, and suppress EAE induced by whole spinal cord ho-
`mogenate.
`
`Abbreviations: EAE, experimental autoimmune encephalomyelitis;
`MS, multiple sclerosis; MSCH, mouse spinal cord homogenate; MBP,
`myelin basic protein; Cop 1, copolymer 1; Ts, T suppressor; Th1 and
`Th2, T helper types 1 and 2; IL, interleukin; IFN-␥, interferon-␥; LN,
`lymph nodes; ICFA, incomplete Freund’s adjuvant; CFA, complete
`Freund’s adjuvant; CNS, central nervous system.
`*To whom reprint requests should be addressed. e-mail:
`weizmann.weizmann.ac.il.
`
`lisela@
`
`10821
`
`Page 1 of 6
`
`YEDA EXHIBIT NO. 2063
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`10822
`
`Immunology: Aharoni et al.
`
`Proc. Natl. Acad. Sci. USA 94 (1997)
`
`MATERIALS AND METHODS
`Mice. SJL兾J, BALB兾c and (SJL兾J⫻BALB兾c)F1 mice were
`purchased from The Jackson Laboratory. Female mice, 7–12
`weeks old, were used for all experiments.
`Antigens and Antibodies. Cop 1 is a synthetic random basic
`polymer, prepared by polymerization of the N-carboxyanhy-
`drides of L-alanine, ␥-benzyl-L-glutamate, ,N-trifluoroacetyl
`L-lysine, and L-tyrosine (15) followed by removal of blocking
`groups. Two Cop 1 batches (batches 02095 and 55495) ob-
`tained from Teva Pharmaceutical Industries (Petach Tikva,
`Israel) were used through the study, with average molecular
`mass of 6,000 Da and 5,800 Da, respectively. MBP was isolated
`from spinal cords of mice or rats, as previously described (21).
`MSCH was prepared as previously described (19). Lysozyme
`from egg white was obtained from Sigma. Cell lines producing
`the mAbs, mouse anti-mouse I-Ad (MK-D6), were obtained
`from American Type Culture Collection. Monoclonal anti-
`mouse I-Ek antibodies crossreactive with I-Ed were purchased
`from Serotec.
`T Cell Lines and Clones. T cell lines were established either
`from spleens of mice that had been rendered unresponsive to
`EAE by subcutaneous injection of Cop 1 or MBP (5–10
`mg兾mouse), emulsified in incomplete Freund’s adjuvant
`(ICFA, Difco) 15 to 35 days earlier, or from lymph nodes (LN)
`of mice that had been immunized with Cop 1 or MBP (200
`g兾mouse) emulsified in complete Freund’s adjuvant (CFA,
`Difco) supplemented with 4 mg兾ml of Mycobacterium tuber-
`culosis H37Ra (Difco), 10 days earlier. Cells were cultured and
`selected in vitro using the immunizing antigen (1–0.5 mg兾
`plate), as described (20). Every 14–21 days, cells were stimu-
`lated by 3-day exposure to Cop 1 or MBP presented on
`syngeneic-irradiated (3,000 rad) spleen cells (50 ⫻ 106兾plate),
`followed by propagation in 10% supernatant of Con A-
`activated normal mouse spleen cells as T cell growth factor.
`Control lines specific to lysozyme were similarly obtained by
`subcutaneous immunization with lysozyme (5 mg兾mouse)
`emulsified in ICFA and stimulated by exposures to lysozyme.
`Cloning of T cell lines was performed by limiting dilution at 0.3
`cells兾well.
`Proliferation Assay. T-Cells (1.5 ⫻ 104) were cultured with
`irradiated spleen cells (5 ⫻ 105) and with the indicated
`antigens. At the end of 48-hr incubation, cultures were pulsed
`with 1Ci[3H]thymidine and harvested 6–12 hr later. Results
`are expressed as mean cpm thymidine incorporation for trip-
`licate cultures. SDs were under 20% of the mean cpm.
`Cytokine Assays. Spleen cells (5 ⫻ 106兾ml) or T cell from
`lines and clones (1 ⫻ 106兾ml), were incubated with the
`indicated antigens, presented on irradiated spleen cells (5 ⫻
`106兾ml), in a final volume of 1 ml. Supernatants were collected
`24 hr later and assayed for cytokine levels using either
`indicator cells or mAbs in ELISA.
`Cytokine assay by indicator cells. The presence of IL-2 or IL-4
`in culture supernatants was evaluated by their ability to
`support the proliferation of the IL-2-dependent CTLD line
`and the IL-4-dependent CT4-S line, respectively. The tested
`supernatants were incubated with the indicator cells (1 ⫻
`104兾well) at a 1:1 dilution for 48 hr and then labeled with 1
`Ci-thymidine. Results are expressed as mean cpm thymidine
`incorporation for triplicate cultures, and the SDs were less then
`20%.
`Cytokine assay by ELISA. IL-2, IL-4, IL-6, IL-10, IFN-␥, and
`tumor necrosis factor-␣ were measured using a quantative
`sandwich ELISA using pairs of mAbs obtained from Phar-
`Mingen, according to the manufacturer’s instructions. The
`threshold detection for all cytokines was 10–50 pg兾ml. Results
`are expressed in ng as mean concentration of duplicate culture
`supernatants (SDs under 20%), measured in duplicate wells by
`ELISA (SDs under 10%).
`
`Induction of EAE. (SJL兾J⫻BALB兾c)F1 2–3-month-old fe-
`male mice were injected in all four footpads with 3.5 mg兾
`mouse spinal cord homogenate emulsified in a 1:1 ratio in CFA
`supplemented with 4 mg兾ml H37Ra. Pertussis toxin (0.25 ml,
`250 ng, Sigma) was injected intravenously, immediately after,
`and 48 hr later. Mice were examined daily for signs of EAE and
`assessed for clinical severity using a 1–5 score as described
`(20).
`
`RESULTS
`Cop 1-specific T cell lines were established from spleens of
`(SJL兾J⫻BALB兾c)F1 mice that had been rendered unrespon-
`sive to EAE by injection of Cop 1 in ICFA 15–35 days earlier
`or from lymph nodes of mice that had been immunized with
`Cop 1 in enriched CFA 10 days earlier. The cells were cultured
`and selected in vitro by repeated exposures to Cop 1. Twelve
`different Cop 1-specific T cell lines and clones were generated
`and characterized. The lines were tested for their CD4兾CD8
`phenotype, and in all cases 96% of the cells were found to bear
`the CD4 phenotype.
`IL-2 and IL-4 Secretion by Cop 1-Specific Lines. The IL-2
`and IL-4 secretion of eight different Cop 1 Ts lines and clones
`first was measured using specific indicator cell lines. CTLD and
`CT4-S lines were used for measuring IL-2 and IL-4 secretion,
`respectively. T cell lines兾clones were tested with various an-
`tigen concentrations, and the results obtained with the optimal
`antigen dose is presented. All of the Cop 1-specific lines and
`clones secreted IL-4, and not IL-2, when incubated with Cop
`1 (Fig. 1). This restriction toward IL-4 secretion was observed
`in the Cop 1 lines and clones originating from spleens of mice
`that had been rendered unresponsive to EAE with Cop 1 in
`ICFA (S-1–4, S-22–1, S-22–5, S-2, and S-3), as well as in the
`Cop 1 lines originating from LN of mice that had been
`immunized with Cop 1 in CFA (LN-1, LN-3, and LN-7). The
`whole spleen cell population from the Cop 1兾ICFA immu-
`nized mice (Cop1兾ICFA spleen cells) secreted both IL-2 and
`IL-4 in response to Cop 1 (see Figs. 1 and 3A) However, after
`one or two cycles of stimulation with Cop 1, when Cop
`1-specific lines have been established, only IL-4 could be
`detected. The clones originating from LN of mice immunized
`with Cop 1兾CFA exhibited a mixed IL-2兾IL-4 secretion for a
`longer period (IL-2 was measured for 10 stimulations), but
`
`lines兾clones. Eight
`IL-2 and IL-4 secretion by T cell
`FIG. 1.
`different Cop 1-specific T cell lines and clones, spleen cells of mice
`injected with Cop 1 in ICFA, and a rat MBP-specific line (EF-S-RBP)
`were cultured with Cop 1 (10 g兾culture) or rat MBP (20 g兾culture).
`The presence of IL-2 and IL-4 in the supernatants was determined by
`their ability to support IL-2-dependent CTLD and IL-4-dependent
`CT4S cell lines. Results are expressed as mean cpm of thymidine
`incorporation for triplicate cultures. SDs were under 20% of the mean
`cpm. Results are from one representative experiment of more than 10
`performed.
`
`Page 2 of 6
`
`YEDA EXHIBIT NO. 2063
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Immunology: Aharoni et al.
`
`Proc. Natl. Acad. Sci. USA 94 (1997)
`
`10823
`
`Table 1. Restriction of Cop 1-specific S-2 line to the
`parental haplotypes
`
`eventually they lost their ability to secrete IL-2 as well. In
`contrast to the Cop 1 lines兾clones, the EF-S-RBP effector line
`originating in mice that had been immunized with the enceph-
`alitogenic antigen rat MBP in CFA and selected using MBP,
`secreted IL-2 from the initial stimulation through the whole
`study period, when cocultured with the immunizing antigen.
`IL-4 could not be detected by the MBP-specific line in
`response to MBP (Fig. 1).
`Comparison Between Cop 1- and MBP-Induced Suppressor
`Lines. Two different lines originating in mice that had been
`rendered unresponsive to EAE by injection of either Cop 1 or
`mouse MBP emulsified in ICFA were selected in vitro using the
`homologous antigen. Each line exhibited a high proliferation
`response to its specific antigen. The cytokine profile of these
`two lines, after three in vitro stimulations, was tested by using
`mAbs in ELISA (Fig. 2). For each line the response to the
`optimal antigen concentration is presented. Whereas both
`lines responded to the nonspecific mitogen Con A with pro-
`duction of high amounts of all the cytokines tested, they
`differed in the secretion profile obtained when each line was
`exposed to its specific antigen. The MBP line reacted to its
`specific antigen MBP by secretion of Th1 cytokines (IL-2,
`IFN-␥ and tumor necrosis factor-␣), as well as Th2 cytokines
`(IL-4, IL-6, and IL-10). In contrast, the Cop 1 line secreted
`only Th2 cytokines, but not Th1 cytokines, in response to Cop
`1, although it still had the potential of producing Th1 cytokines
`as indicated by the response to Con A. Thus, whereas the MBP
`line exhibited a mixed Th1兾Th2 cytokine secretion, the Cop 1
`line was confined to the Th2 pathway.
`Restriction of Cop 1 Specific F1-Ts Lines. The F1-Ts-Cop 1
`lines兾clones were originated from (SJL兾J⫻BALB兾c)F1 mice.
`The restriction of S-2 line to the two parental haplotypes SJL兾J
`(H-2s) and BALB兾c (H-2d) is shown in Table 1. Whereas the
`response to the mitogen Con A was not restricted to one of the
`parental haplotypes, a restricted pattern was observed in
`response to Cop 1. Maximal proliferation as well as secretion
`of IL-4, IL-6, and IL-10 were obtained when Cop 1 was
`presented on BALB兾c or F1 antigen-presenting cells (APC).
`On the other hand, the proliferation and Th2 secretion in
`response to Cop 1 on SJL兾J APC were much lower, and IL-4
`was not secreted when Cop 1 was presented by SJL兾J. Neither
`IL-2 nor IFN-␥ were measured in response to Cop 1 with all
`strains of APC. Similar results were obtained with the other
`Cop 1-specific lines兾clones. Thus, the activity of F1-Ts-Cop1
`lines兾clones toward Cop 1, manifested both by proliferation
`
`FIG. 2. Comparison between cytokine secretion by Cop 1- and
`MBP-induced Ts lines. Quantative ELISA of supernatants of two T
`cell lines established from spleens of mice, which had been rendered
`unresponsive to EAE by injection of Cop 1 or MBP in ICFA,
`stimulated by no antigen, Cop 1 (50 g兾ml), MBP (100 g兾ml), and
`ConA (5 g兾ml). Results are expressed as mean cytokine concentra-
`tion of duplicate wells. SDs for both assays were under 20% of the
`mean.
`
`IL-4, ng兾ml
`
`IL-6, ng兾ml
`
`IL-10, ng兾ml
`
`Proliferation, ⫻10⫺3
`
`IL-2, ng兾ml
`
`INF-␥, ng兾ml
`
`BALB兾c
`SJL兾J
`F1
`Antigen
`0.16
`0.26
`0.21
`—
`21.61
`4.80
`19.50
`Cop 1
`n.d.
`n.d.
`n.d.
`Con A
`⬍0.05
`⬍0.05
`⬍0.05
`—
`⬍0.05
`⬍0.05
`⬍0.05
`Cop 1
`18.7
`14.7
`16.7
`Con A
`⬍0.05
`⬍0.05
`⬍0.05
`—
`⬍0.05
`⬍0.05
`⬍0.05
`Cop 1
`9.1
`9.2
`9.8
`Con A
`⬍0.3
`⬍0.3
`⬍0.3
`—
`⬍0.3
`2.1
`2.0
`Cop 1
`11.7
`10.3
`10.8
`Con A
`⬍0.1
`⬍0.1
`⬍0.1
`—
`15.1
`6.5
`14.1
`Cop 1
`35.5
`36.8
`36.6
`Con A
`⬍0.2
`⬍0.2
`⬍0.2
`—
`8.6
`2.1
`9.8
`Cop 1
`11.4
`10.9
`12.5
`Con A
`The Cop 1-specific line S-2 was tested for its restriction to the
`parental haplotype, by comparing its proliferation and cytokine se-
`cretion in response to Cop 1 (50 g兾ml) or Con A (5 g兾ml), on
`irradiated spleen cells (5 ⫻ 106 cells兾ml) from (SJL兾J⫻BALB兾c)F1,
`SJL兾J, or BALB兾c origin. SDs of all the indicated values were under
`20%. Results are from one representative experiment of two per-
`formed. n.d., not determined.
`and by Th2 cytokine secretion, was restricted to the BALB兾c
`H-2d haplotype.
`A further analysis of the restriction of the Cop 1-specific
`lines兾clones within the H-2d region was performed using
`anti-I-Ad and anti-I-Ed mAbs (Table 2). Whereas no inhibition
`was obtained by antibodies to the I-E subregion of the BALB兾c
`haplotype, the response to Cop 1 manifested by both prolif-
`eration and IL-4 secretion was strongly inhibited in the pres-
`ence of I-Ad antibodies. Similarly the other Cop 1-specific
`lines兾clones were inhibited only by antibodies to the I-Ad. Thus
`a restriction of the response toward Cop 1 to the I-A subregion
`of the H-2d was demonstrated.
`Crossreactivity of Cop 1-Specific F1-Ts Lines兾Clones with
`MBP. The pattern of cytokine secretion by several F1-Ts-Cop
`1 lines兾clones in response to Cop 1 and to the autoantigen
`MBP was studied. A detailed proliferation and cytokine profile
`of S-2 line during its development is demonstrated in Fig. 3.
`Initially, when whole spleen cell population from mice that had
`been rendered unresponsive to EAE by Cop 1 was tested (Fig.
`3A) a low response to Cop 1 was measured in comparison to
`the response obtained with the mitogen Con A. The spleen
`cells reacted to Cop 1 by proliferation and by secretion of both
`Th1 (IL-2 and IFN-␥) and Th2 (IL-4, IL-6, and IL-10) cyto-
`kines. On the other hand, neither proliferation nor cytokine
`
`Table 2. Restriction of Cop 1-specific S-2 line within the
`H-2d region
`
`IL-4, cpm
`
`Proliferation, cpm
`
`Inhibitor
`Anti-I-Ed
`Anti-I-Ad
`—
`Antigen
`390
`748
`243
`—
`8,573
`796
`7,405
`Cop 1
`64
`43
`28
`—
`8,563
`61
`9,685
`Cop 1
`The Cop 1-specific line S-2 was tested for its restriction within the
`H-2d region, by inhibition of the proliferation and IL-4 secretion in
`response to Cop 1 (10 g兾well) presented on BALB兾c irradiated
`spleen cells (5 ⫻ 106兾well), with mAbs specific either to the I-Ad or
`I-Ed region. SDs of all the indicated values were under 20%. Results
`are from one representative experiment of two performed.
`
`Page 3 of 6
`
`YEDA EXHIBIT NO. 2063
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`10824
`
`Immunology: Aharoni et al.
`
`Proc. Natl. Acad. Sci. USA 94 (1997)
`
`line secreted even higher amounts of Th2 cytokines in response
`to MBP than to Cop 1.
`It should be noted that although all of the Cop 1-specific
`lines兾clones secreted significant amounts of Th2 cytokines in
`response to Cop 1, they differed in the extent of crossreactivity
`with MBP. Other lines兾clones obtained demonstrated a
`smaller degree of crossreactivity with MBP, usually on the
`level of IL-4 secretion. The crossreactivity with MBP was
`exhibited only by Cop 1-induced lines兾clones. Thus control
`lysozyme specific lines, which secreted Th2 cytokines in re-
`sponse to lysozyme, did not respond at all to MBP or to Cop
`1 (data not shown).
`Suppressive Activity of Cop 1-Specific F1-Ts Lines in Vivo.
`The ability of the Cop 1-specific lines兾clones to prevent EAE
`induced by whole spinal cord homogenate in vivo was inves-
`tigated. Activated F1-Ts-Cop1 cells were injected to (SJL兾
`J⫻BALB兾c)F1 mice followed by EAE induction. The S-2 line
`originating from spleens of mice that had been rendered
`unresponsive to EAE with Cop 1 in ICFA was tested both after
`2 months (Fig. 4A) and after 6 months (Fig. 4B) ofin vitro
`growth. This line was significantly effective in suppressing
`EAE, because the mean clinical score of mice injected with the
`line was always lower than that of the control group. However
`the S-2 cells grown in vitro for 6 months were more effective
`in disease prevention than those grown for 2 months. Thus,
`after 2 months 20 ⫻ 106 cells inhibited only 40% and 58% of
`disease incidence and severity respectively, P ⬍ 0.04, whereas
`10 ⫻ 106 cells of the line grown for 6 months inhibited 80%
`incidence and 86% severity of the disease, P ⬍ 0.01.
`The Cop 1-specific clone S-22–1, originating similarly to S-2,
`completely prevented the disease, and none of the animals
`showed any signs of EAE (P ⬍ 0.002). This is in contrast to the
`lysozyme-specific line, which did not show any inhibitory effect
`on the disease (P ⬍ 0.7), even though the same number of cells
`were injected (Fig. 4C).
`
`Inhibition of MSCH-induced EAE by T cell lines and
`FIG. 4.
`clones. (A) Cop 1-specific line S-2 after 2 months in culture (20 ⫻
`106兾mouse). (B) S-2 line after 6 months in culture (10 ⫻ 106兾mouse).
`(C) Cop 1-specific clone S-22–1 and lysozyme-specific line Lys-1 after
`6 months in culture (15 ⫻ 106兾mouse). (D) Cop 1-specific clone LN-3
`after 6 months in culture (20 ⫻ 106兾mouse). Cells were injected
`intravenously 3 days after stimulation with Cop 1 to (SJL兾J⫻BALB兾
`c)F1 mice 5–10 in a group, followed by EAE induction by MSCH.
`Control mice were induced with EAE alone. Results are expressed as
`mean daily clinical score of 5–10 mice in a group. Statistical analysis
`by Student’s t test: (A) S-2 after 2 months in culture, P ⬍ 0.04. (B) S-2
`after 6 months in culture, P ⬍ 0.01. (C) L-22–1, P ⬍ 0.002, Lys-1, P ⬍
`0.7. (D) LN-3 clone, P ⬍ 0.09.
`
`FIG. 3. Proliferation and cytokine secretion profile of S-2 line
`during its development. (A) Initial response of spleen cells. (B) After
`6 weeks in culture. (C) After 6 months in culture. Cells were cultured
`with no antigen, Cop 1 (50 g/ml), MBP (100 g兾ml), and ConA (5
`g兾ml). Proliferation was measured by thymidine incorporation for
`triplicate cultures. Cytokine concentration was measured by quanti-
`tative ELISA in duplicate wells for each one of duplicate culture
`supernatants. SDs were under 20% of the mean. Results represent one
`of three independent experiments.
`
`secretion could be measured when the whole spleen cell
`population was incubated with MBP.
`Six weeks later, after the cells had been exposed to Cop 1 for
`three cycles of stimulation, the response to Cop 1 was more
`confined (Fig. 3B). The S-2 line responded to Cop 1 by
`proliferation, but not by secretion of IL-2 and IFN-␥. These
`Th1 cytokines were secreted only in response to the mitogen
`Con A. On the other hand, higher amounts of IL-4, IL-6, and
`IL-10 were secreted by these cells in response to Cop 1,
`approaching the amounts induced by Con A. Moreover, S-2
`cells secreted IL-4 not only in response to Cop 1 but also in
`response to MBP (7-fold more than the secretion with no
`antigen).
`After 6 months of in vitro growth in which the S-2 line had
`been repeatedly exposed to Cop 1, IL-2 and INF-␥ were not
`detected in response to Cop 1, nor after stimulation with Con
`A (Fig. 3C). Although the cells did not secrete Th1 cytokines,
`they secreted high amounts of Th2 cytokines in response to
`Cop 1 and to Con A. The crossreactive response to MBP, which
`had been observed after 2 months only on the level of IL-4
`secretion, now was demonstrated on the level of all Th2
`cytokines measured (IL-4, IL-6, and IL-10). Furthermore, this
`
`Page 4 of 6
`
`YEDA EXHIBIT NO. 2063
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Immunology: Aharoni et al.
`
`Proc. Natl. Acad. Sci. USA 94 (1997)
`
`10825
`
`The suppressive activity of another Cop 1-specific clone,
`LN-3, originating from LN of mice that had been immunized
`with Cop 1 in CFA and grown in vitro for 6 months, also was
`tested. As demonstrated (Fig. 4D), this clone also induced
`EAE suppression, but it was less effective than S-2 and S-22–1,
`because even after 6 months, 20 ⫻ 106 cells induced only 60%
`and 50% inhibition of the disease incidence and severity,
`respectively (P ⬍ 0.09).
`
`DISCUSSION
`The cumulative evidence on the protective role of Th2 cells in
`a variety of cellular immune responses, including autoimmune
`diseases, led to the design of immuno-intervention approaches
`aimed at inducing an antigen-specific shift from the Th1
`pathogenic response to the protective Th2 response (4, 5, 13,
`14, 22). In this study we demonstrated that Cop 1, which had
`been shown to suppress EAE as well as MS (16, 17), induces
`a predominant Th2 response. All 12 T cell lines and clones
`generated from Cop 1-immunized mice and selected for Cop
`1 exhibited a Th2 cytokine profile and secreted high amounts
`of IL-4, IL-6, and IL-10 in response to Cop 1 (Figs. 1 and 3A,
`Tables 1 and 2). Secretion of tumor necrosis factor-␣never was
`detected by Cop 1-induced cells (data not shown). IL-2 and
`IFN-␥ were secreted when the whole spleen cell population
`was tested (Figs. 1 and 3). However, after the cells were
`repeatedly exposed to Cop 1 and T cell lines兾clones were
`established, these Th1 cytokines could not be detected,
`whereas CD4⫹ Th2 cells prevailed.
`The bias toward the Th2 phenotype was found in the Cop 1
`lines兾clones originating from spleens of mice that had been
`rendered unresponsive to EAE by injection of Cop 1 in ICFA
`15–35 days earlier. This regimen has been previously shown to
`induce antigen-specific supressor cells (19, 20), and the Cop 1
`lines兾clones originating in this way showed strong disposition
`toward the Th2 pathway, because after 1–3 cycles of stimula-
`tion with Cop 1 they already were confined to the Th2 pathway
`(Figs. 2 and 3B). Furthermore, even clones originating from
`lymph nodes of mice that had been immunized with Cop 1 in
`enriched CFA 10 days earlier, a procedure usually used to
`obtain effector T cell lines (6), exhibited this tendency to Th2
`diversion. In this case a mixed Th1兾Th2 secretion was detected
`for a longer period but eventually, those clones, too, lost the
`ability to secrete IL-2, and a complete shift to IL-4 secretion
`was observed. T cell lines obtained from LN of mice that
`produced Th2 cytokines also were demonstrated in the case of
`mice that had been immunized with altered peptide ligand of
`proteolipid protein and CFA (23).
`It should be noted that our T cell lines were not grown in the
`presence of either IL-4 or neutralizing antibodies to Th1
`interleukins, which frequently are used to selectively promote
`Th2 polarized cells (3). Rather, T cell growth factor medium
`that contained the whole spectrum of cytokines secreted by
`normal spleen cells in response to Con A was used as a growth
`factor source. In this way the role of the specific antigen in
`determining the cytokine profile could be revealed. Indeed,
`the effector line EF-S-RBP, secreted IL-2, and not IL-4, when
`cocultured with its specific antigen MBP (Fig. 1). Further-
`more, even when MBP was injected according to the suppres-
`sion regime (in ICFA, 1 month earlier), the MBP-specific line
`obtained exhibited a mixed Th1兾Th2 cytokine profile (Fig. 2).
`In contrast, the Cop 1-induced lines were confined to the Th2
`pathway already after three stimulations. These findings indi-
`cate that the polarization of Cop 1-induced lines was not
`obtained due to the immunization vehicle or the in vitro culture
`conditions, but rather due to the uniqueness of Cop 1, which
`preferentially induces Th2 protective response. This trait could
`stem from the polymeric nature of Cop 1, which contains
`multiple major histocompatibility complex class II binding
`epitopes, and does not require processing for presentation
`
`(18). This results in high density of the antigen on the
`antigen-presenting cells, thus favoring Th2 development (3).
`Of interest is the genetic restriction of the Cop 1-specific
`lines that were originated from (SJL兾J⫻BALB兾c)F1 mice to
`the BALB兾c (H-2d) parental haplotype (Table 1), and within
`the H-2d region, to the I-A and not the I-E subregion (Table
`2). This restriction was manifested by proliferation and by Th2
`cytokine secretion. We demonstrated before that the unre-
`sponsiveness to EAE of the BALB兾c strain results from the
`high level of regulatory cells because it could be abrogated by
`pretreatment with cyclophosphamide (19). The restriction of
`the Cop 1 Th2 lines to the BALB兾c haplotype is also consistent
`with the dominant BALB兾c genetic predisposition toward Th2
`differentiation (24), which was reported in the literature for
`Leishmania major (4, 25) and autoimmune diabetes (4, 26).
`The crossreactivity of Cop 1 with the natural autoantigen
`MBP, on the level of B as well as T cell response, previously
`was demonstrated (16, 27, 28). This crossreactivity was shown
`to correlate with Cop 1 suppressive activity in vivo (29). It was
`therefore interesting to find that the crossreactivity between
`Cop 1 and MBP also is expressed on the level of cytokine
`secretion (Fig. 3). This crossreactivity, manifested only on the
`level of the Th2 cytokine secretion, became more evident with
`more stimulations with Cop 1, probably due to selection of
`crossreactive clone from the cell line population. Thus, initially
`it was confined to the signature Th2 cytokine, IL-4 (Fig. 3B),
`but with additional stimulations with Cop 1, the crossreactivity
`with MBP became much more pronounced and expanded to
`the three Th2 cytokines tested (Fig. 3C). The S-2 line secreted
`even higher amounts of Th2 cytokines in response to MBP to
`which it had never been exposed before, than to Cop 1, which
`had been used for immunization and stimulation. Interestingly,
`whereas the response of encephalitogenic lines to MBP is Th1-
`and H-2s-restricted (4, 6), the crossreactive response to MBP
`of our Cop 1-specific lines is Th2- and H-2d-restricted. The way
`by which Cop 1 induces this crossreactive Th2-confined re-
`sponse has not been revealed in this study. Whether the Cop
`1-specific cells recognize a common ‘‘suppressive’’ determi-
`nant shared by Cop 1 and MBP, but different from the
`encephalogenic regions of MBP that induce Th1 inflammatory
`processes, or bear a mutated T cell receptor that transduces
`different signals as a consequence of antigen binding, is not
`clear at present.
`The most meaningful criterion for the biological relevance
`of the Cop 1-specific Th2 lines is their ability to suppress the
`disease in vivo. Indeed a significant suppression was demon-
`strated by these cells on the development of EAE, as reflected
`both in the incidence and in the clinical score of the disease
`(Fig. 4). The relevance of Cop 1 Th2-specific lines to the in vivo
`effect induced by Cop 1 is not obvious, because whole spleen
`population demonstrated a mixed Th1兾Th2 response (Figs. 1
`and 3A). However it should be noted that adoptive transfer of
`protection was demonstrated using whole spleen cell popula-
`tion from Cop 1-treated mice as well (19). Because activated
`T cells cross the blood brain barrier irrespective of their
`antigenic specificity (30), it is possible that Cop 1-specific cells
`induced or stimulated by the injection of Cop 1 in the
`periphery, penetrate through the blood brain barrier to the
`CNS. The Cop 1-induced cells, which crossreact with MBP on
`the level of Th2 cytokines, are further stimulated by exposure
`to MBP, which is abundantly present in the CNS, and can be
`presented by Ia-inducible glia cells (30). Thus the secretion of
`the anti-inflammatory cytokines is carried on or even ampli-
`fied in the case of the crossreactive response to MBP (Fig. 3C).
`Interestingly, even those Cop 1 lines that showed only small or
`moderate crossreactivity with MBP induced an in vivo protec-
`tive effect on the disease (Fig. 4 A andD). This effect can be
`attributed