Both CD4 + and CD8 + T Cells Are by kellena91


									Published August 1, 1996

                                   B o t h C D 4 + and C D 8 + T Cells Are Essential to I n d u c e
                                   E x p e r i m e n t a l A u t o i m m u n e M y a s t h e n i a Gravis
                                   By Guang-Xian Zhang,* Bao-Guo Xiao,* Moiz Bakhiet,*
                                   Peter van der Meide,* Hans Wigzell,g Hans Link,* and Tomas Olssonll

                                   From the *Division of Neurology, Huddinge Hospital, S-141 86 Huddinge, Stockholm; *Biomedical
                                   Primate Research Centre, Rijswijk, The Netherlands;$Division of Immunology, Karolinska Institute,
                                   Stockholm; and IIMotecularMedicine Unit, Karolinska Hospital, Stockholm, Sweden

                                   C D 4 + T cells have been shown to be crucial in the development o f experimental autoimmune
                                   myasthenia gravis (EAMG). The role o f C D 8 + T cells in E A M G is less well established. W e
                                   previously showed that antibody depletion o f C D 8 + T cells in rats effectively suppresses
                                   EAMG. T o further study the rote and relationship o f C D 4 + versus C D 8 + T cells in induction
                                   o f E A M G , C D 4 - / - , C D 8 - / - , and C D 4 - 8 - mutant C57BL/6 mice and the parent CD4+8 +
                                   wild-type mice were immunized with Torpedo acetylcholine receptor (ACht 0 plus complete

                                                                                                                                                      Downloaded from on January 5, 2010
                                   Freund's adjuvant. Clinical E A M G was nearly completely prevented in C D 4 - 8 - , C D 4 - / - ,
                                   and C D 8 - / - mice. This was associated with strongly reduced AChR-specific T and B celt re-
                                   sponses, and with reduced levels o f AChR-reactive interferon ~/(IFN-'y) and interleukin 4 (IL-4)
                                   mtLNA-expressing cells in lymphoid organs when compared with CD4+8 + wild-type mice.
                                   W e conclude that (a) both C D 4 + and C D 8 + T cells are essential for development o f EAMG,
                                   and a collaboration between these cell types may be necessary; (b) C D 4 + as well as C D 8 + T
                                   cells secrete I F N - y and IL-4, and both cytokines are involved in the development o f E A M G ;
                                   and (c), besides T cells, other immune cells might also be responsible for help o f anti-AChtL
                                   antibody production.

       xperimental autoimmune myasthenia gravis (EAMG) 1                      C D 4 + T cells and in mice lacking C D 4 + T cells, further
  E    can be induced in susceptible mouse strains by repeated
  injections o f acetylcholine receptor (AChR) emulsified in
                                                                              supporting a role for these T cells in the pathogenesis o f
                                                                              E A M G (9, 10),
  CFA, and serves as a useful animal model for exploring the                     W e have shown that depletion o f C D 8 + T cells in Lewis
  pathogenesis o f human myasthenia gravis (MG) (1). The                      rats also suppresses E A M G and the levels o f a n t i - A C h R an-
  muscle weakness and fatigue that are the hallmarks o f m y -                tibodies (11), indicating that C D 8 + T cells are also in-
  asthenia gravis and E A M G are due to an antibody-medi-                    volved as helper or effector cells in the pathogenesis o f this
  ated autoimmune attack directed against AChtL o f the                       disease. Another report, however, showed that [~2m - / -
  neuromuscular junctions (2). The production o f anti-                       mice, which are C D 8 - and M H C class I deficient, have a
  A C h R antibodies depends on cytokines produced by T                       higher incidence o f E A M G than control mice (12). To fur-
  cells (3). T cell responses to A C h R , its subunits, and the              ther analyze the relative contribution o f C D 4 and C D 8 T
  o~-subunit peptides have been described in myasthenia                       cells in the induction o f E A M G , we induced the disease in
  gravis (4, 5) and E A M G (6-8). AChtL subunit and peptide-                 C D 4 - / - , C D 8 - / - , and C D 4 - 8 - mutant C57BL/6J mice
  specific rat C D 4 + T cells secrete both T h l (IL-2, IFN-~/)              and the same strain o f CD4+8 + wild-type mice.
  and Th2 (IL-4) cytokines after stimulation with A C h R ,
  and provide help for a n t i - A C h R antibody production (3).
  E A M G is suppressed in mice treated with m A b to murine                  Materials and Methods
                                                                                 Animals and Antigen Preparation. Female C57BL/6J (H-2 b)
                                                                              CD8 -/-, CD4 -/-, and CD4-8- mutant mice were obtained
  1Abbreviations used in this paper: 0~-BGT, ~x-bungarotoxin; AChK, acetyl-   from Dr. Tak W. Mak using homologous recombination in
  choline receptor; EAMG, experimental autoimmune myasthenia gravis;          pluripotent embryonic stem cells, and were bred at the animal
  M-AChlK, muscle AChR extract; MBP, myelin basic protein; MNC,               house, Division of Immunology, Karolinska Institute. The same
  mononuclear cell; p.i., postprimary immunization; MG, myasthenia            strain of female CD4+8 + mice was obtained from the same ani-
  gravis; PILN, popliteal and inguinal lymph node; SI, stimulation index.     mal house. All mice used were 8-12 wk of age, weighing 20-30 g.

                                    349      j. Exp. Med. 9 The tLockefeller University Press - 0022-1007/96/08/349/08 $2.00
                                             Volume 184 August 1996 349-356
Published August 1, 1996

    Torpedo A C h R was purified from the electric organs of Torpedo        peroxidase staining, the red-brown immunospots that corre-
  Californica (Pacific Biomarine, Venice, CA) by affinity chroma-           sponded to cells having secreted anti-AChR IgG were counted and
  tography on an ot-cobrotoxin-agarose resin (Sigma Chemical Co.,           standardized to number per l0 s MNC.
  St. Louis, MO) as described previously (13). The isolated product            AChR-reactive IFN-y-secreting Cells. Nitrocellulose-bottomed
  was pure as judged by SDS-PAGE. Muscle A C h R extract from               microtiter plates were coated with 100/.zl rat IFN-~ capture anti-
  normal C57BL mice (M-AChR) was prepared as described (13).                body (DBI; 15) at 15 ~g/ml. Aliquots of 200-b~l suspensions
  Myelin basic protein (MBP) used as control antigen was purifed            containing 4 • 105 M N C were added to individual wells in trip-
 from normal mouse brains as described (14).                                licate, followed by antigen (AChR or MBP) or the mitogen Con
     Induction of EAMG. Mice were immunized with 20 p~g Tor-                A (Sigma Chemical Co.) in 10-~zl aliquots at a final concentration
 pedo A C h R emulsified in CFA in a total volume of 100 pJ, in-            of 10 p~g/ml A C h R or MBP, or 5 ~g/ml Con A. After 48 h of
 jected into five intradermal sites along the back, the hind foot-          culture, the wells were emptied. Secreted and bound IFN-y was
 pads, and the base of the tail (1), and were boosted on days 30 and        visualized by sequential application of polyclonal rabbit anti-rat
  60 after the primary sensitization. The mice were observed every          IFN-~ (15), biotinylated anti-rabbit IgG (Dakopatts), and ABC
  other day in a blinded fashion for signs of clinical muscle weak-         (Dakopatts). After peroxidase staining, the red-brown immuno-
 ness. The disease symptoms were graded as follows (9): 0, no def-          spots that corresponded to the cells that had secreted IFN-~/were
 inite muscle weakness; 1 +, normal strength at rest but weak with          enumerated in a dissection microscope. To calculate the numbers
  chin on the floor and inability to raise the head after exercise con-     of T cells responding to a particular antigen or mitogen, numbers
 sisting of 20 consecutive paw grips; 2 + , as grade 1+ weakness at         of spots in culture without antigen added were subtracted from
 rest; and 3 + , moribund, dehydrated, and paralyzed. Clinical              the values obtained after antigen or mitogen exposure. The data
 EAMG was confirmed by i.p. administration of neostigmine bro-              were expressed as numbers per 10s MNC.
 mide and atropine sulphate. Mice were killed 100 d postprimary                Lymphocyte Proliferation Responses. Triplicate aliquots (200 ~zl)
 immunization (p.i.).                                                       of M N C suspensions were applied to 96-well round-bottomed
     RIAforM-AChR Content. The concentration of A C h R of                  microtiter plates (Nunc, Copenhagen, Denmark) at a cell density

                                                                                                                                                   Downloaded from on January 5, 2010
  mouse muscle carcass was determined by ILIA (10). Briefly, tripli-        of 2 • 10('/ml. l()-bL1aliquots of either AChR, MBP, or Con A
 cate 2-pmol aliquots of lasI--ot-bungarotoxin (or             Amersham     were added to appropriate wells at a final concentration of 1() b~g/
  Corp., Arlington Heights, IL)-labeled Triton X-100 solubilized            ml (AChR or MBP) or 5 la,g/ml (Con A). After 60 h of incuba-
  mouse muscle extract were mixed with a standard pooled mouse              tion, the cells were pulsed for 12 h with 10-lxl aliquots containing
 anti-AChR antiserum. After overnight incubation, rabbit anti-              1 ~Ci of [~H]methylthymidine (specific activity, 42 Ci/mmol;
  mouse lg (Dakopatts, Copenhagen, Denmark) was added. The                  Amersham Corp.). Cells were harvested onto glass fiber filters and
  resulting precipitates were pelleted, washed, and pelleted again.         thymidine incorporation was measured. The results were expressed
  Radioactivity of the pellet was counted in a galmna counter (Pack-        as stimulation index (SI), which was calculated by dividing the
 ard Instrument Co., Inc., Meriden, CT). A C h R content is ex-             counts per minute from culture in the presence of antigen or mi-
 pressed as moles of 12sI--o~-BGT binding sites per milligram pro-          togen by the counts per minute from culture without antigen.
 tein of carcass, and the percentage loss of M - A C h R in test mouse         Detection of IFN-T, IL-4, and TGF-[3 mRNA Expression by In
 carcass was calculated (13).                                               Situ Hybridization. In situ hybridization was performed as de-
     RIA for Serum Anti-AChR Antibodies. Blood specimens were               scribed (16). Briefly, 200qxl aliquots of suspensions from PILN
 collected from the tail vein just before and then at weekly inter-         containing 4 X 10s M N C were plated into round-bottomed nil-
 vals, and serum antibody concentrations to M - A C h R were mea-           crotiter plates (Nunc, Roskilde, Denmark) in triplicate. 10-1xl ali-
 sured by RIA (13). Briefly, 1 nmol mouse A C h R was incubated             quots of either AChR, MBP, or Con A were added to appropri-
 with 2 nmo112sI--~-BGT. To 1 ml of labeled M - A C h R , 1 b~l se-         ate wells at a final concentration of 10 Ixg/nd (AChR or MBP) or
 rum was added, followed by rabbit anti-mouse Ig. The samples               5 Ixg/ml (Con A). After culture for 24 h, the cells were washed,
 were centrifuged, washed, and counted in a gamma counter. The              counted, and dried onto restricted areas of glass slides (ProbeOn
 A C h R precipitated minus the background value pemfits calcula-           slides; Fisher Scientific, Pittsburgh, PA). Synthetic oligonucle-
 tion of the titer in moles of toxin-binding sites bound per liter of       otide probes (Scandinavian Gene Synthesis AB, K6ping, Sweden)
  serum. Serum samples from CFA-injected CD4+8 + mice were                  were labeled using [3sS]deoxyadenosine-5'-c~-(thio)-triphosphate
  analyzed in parallel.                                                     with tenninal deoxynucleotidyl transferase (Amersham Corp.).
     Mononuclear Cell Suspensions. Suspensions of mononuclear cells         To increase the sensivity of the method, a mixture of four differ-
  (MNC) from the popliteal and inguinal lymph nodes (PILN),                 ent oligonucleotide probes was employed for each cytokine. The
  spleen, and thymus were prepared as described (11), and cells were        oligonucleotide sequences were obtained from EMBL/Gen-
  suspended in DME supplemented with 1% (vol/vol) MEM (both                 Bank/DDBJ using the MacVector software (IBI). The rat IFN-y
 from Gibco, Paisley, UK), 2 mM glutamine (Flow Laboratories,               probes (17) (EMBL/GenBank/DDBJ accession numbers M29315,
  lrvine, UK), 50 I U / m l penicillin, 50 I~g/ml streptomycin, and         M29316, and M29317) were complementary to bases 298-345
  10% (vol/vol) FCS (Gibco). The cells were washed three times              (exon 1), 80-125 (exon 2), 303-350 (exon 3), and 180-227 (exon
 and then rediluted to a cell concentration of 2 • 10a/ml.                  4). The rat IL-4 probes (18) (EMBL/GenBank/DDBJ accession
     Enumeration of A ChR-specific IgG Antibody-secreting Cells. A solid-   number X16058) corresponded to bases 83-130, 2(i)9-256, 270-
  phase enzyme-linked immunospot assay (ELISPOT) was used                   317, and 331-378. The TGF-IB probes (19) (EMBL/GenBank/
 with some modifications (11). Briefly, wells were coated with              DDBJ accession number X02812) were complenlentary to bases
  100 txI A C h R or MBP (I0 b~g/ml in PBS). AIiquots of 100-bd               766-1813 and 1953-2000. The oligonucleotide probes were ap-
  cell suspensions containing 2 • 10s M N C were added in tripli-           proximately 48 bases tong and checked for the absence of palin-
  cate to individual wells. After incubation for 24 h, the wells were       dromes or long sequence of homology within the species against
  emptied, followed by addition of rabbit anti-mouse IgG (Sigma             available EMBL/GenBank/DDBJ data. Cells were hybridized
  Chemical Co.), biotinylated swine anti-rabbit IgG (Dakopatts),            with 10(, cpm of labeled probe per 100 hal of hybridization nfix-
  and avidin-biotin peroxidase complex (ABC; Dakopatts). After              ture. After emulsion autoradiography, development, and fixation,

                                   350       CD4 + and CD8 + T Cells in Experimental Autoimmune Myasthenia Gravis
Published August 1, 1996

  the coded slides were examined by dark field microscopy for pos-                                                  These animals deteriorated progressively until two died and
  itive ceils containing >15 grains per cell in a starlike distribution.                                            two were humanely killed (Fig. 1). After the first boosting,
  The intracellular distribution of the grains was always checked by                                                two C D 8 - / - mice exhibited mild and moderate weakness,
  light microscopy, and labeled cells were expressed as numbers per                                                 respectively, lasting to the end. In the C D 4 - / - group, one
  lOs MNC.
                                                                                                                    mouse got mild E A M G after the first boosting and recov-
      Statistical Analysis. Overall differences between the four groups
  were evaluated by the Kruskal-Wallis one-way analysis of vari-                                                    ered about 1 m o later; another mouse developed moderate
  ance (ANOVA). Student's t test was used for the comparison of                                                     weakness after the second boosting and improved 20 d
  serum anti-AChR antibodies, and the Mann-Whitney U test was                                                       later. A m o n g the C D 4 - 8 - mutant mice, only one devel-
  used for the other variables analyzed. All significance tests were                                                oped mild and short-term muscle weakness 12 d after the
  two-sided.                                                                                                        first boosting (Fig. 1).
                                                                                                                       Loss o f M - A C h R . T h e mean values for losses o f
                                                                                                                    M - A C h R in the C D 4 - / - , C D 8 - / - , and C D 4 - 8 - mice af-
                                                                                                                    ter immunization with A C h R and C F A were 21 z 18%,
                                                                                                                    29 +- 15%, and 9 + 3%, respectively. In contrast, the mean
     Muscle Weakness in C D 4 - and~or CD8-knockout Mice.                                                           M - A C h R loss in the CD4+8 + wild mice was 54 • 11%,
  O n e out o f 15 wild-type (CD4+8 +) mice developed mus-                                                          being significantly higher than in the other three groups
  cle weakness 3 w k after primary sensitization without boost-                                                     (P <0.05, 0.05, and 0.002, respectively). N o differences
  ing. Seven m o r e mice developed paralysis after boosting.                                                       were found among C D 4 - / - , C D 8 - / - , and C D 4 - 8 - mice.

                        CD4+8+                                                                                                       3        CD8-/-

                                                                                                                                                                                                            Downloaded from on January 5, 2010
             2                                                                                                                       2


             0        ..~                        |        ,                                                                o

                  0         l0        20        30        40,        50     60        30     80        90     100                        0    I0    20   30   40       50    60    70    80    90     I00
                                                Days aft~ immm~timtion
                                                                                                                                                          Days ~ t ~   immuni~Uon

                                                                                                                                3        CD4-8-

                                                                          j L                                                  2

                                                                                 ,                                  r-,O


              0        10        20                                                                                         O'
                                           30        40         50        60         70    80     90        100                  u       iv    20   30   40     50      60    70    80
                                                                                                                                                                                          90    100
                                           Days fft~rimmunim~on                                                                                      Days~terimmutaiza~on

  Figure 1. Clinical course of EAMG in CD4+8 +, CD8 -/-, CD4 -/ , and CD4 8- C57BL/rJ mice (n = 15 in each group). Mice were immunized
  with AChR plus CFA and boosted twice monthly. (*) Died; (+) humanely killed.

                                                                      351                 Zhang et al.
Published August 1, 1996

                 -------o---   CD84-                                                         q)
            6                  CD4-g-                                            *#               10-                                              ~      CD8-/-
                               CD44-                                                                                                                      CD4-8-
            5         A                            *         *#
                               CFA-immunized                                                       8                                               ~      CD4+8+

            3                                                                                 m

  <         2


                          2       4         6        8        10            12         t4
                                                                                                             PILN                  Spleen               Thymus

                                      Weeks after immunization                               Figure 4. Numbers ofAChR-reactive IFN-"/~secretmg T cells per l0 s
                                                                                             MNC isolated from PILN, spleen, and thymus of the different groups.
  Figure 2.     Serum anti-M-AChR antibody levels in the different groups.                   Symbols refer to mean values, and bars to SD. P values refer to compari-
  The antibody concentrations were measured weekly by RIA and ex-                            sons between the CD4+8 + mice and the other three groups. (*) P <{).05;
  pressed as moles of {x-BGT-binding sites bound per liter of serum. Sym-                    (**) P <{}.01,
  bols refer to mean values, and bars to SD. The differences between
  CD4-8- and CD4+8 § mice were significant from week 2 p.i. up to end
  (significant P values were omitted). (*) P <0.05 between CD4+8 + and

                                                                                                                                                                                               Downloaded from on January 5, 2010
  CD8 -/ mice; (#) P <0.05 between CD4+8 + and CD4 - / - mice.                                    T h e highest n u m b e r s o f a n t i - A C h R I g G a n t i b o d y - s e c r e t -
                                                                                             i n g cells w e r e f o u n d in l y m p h n o d e s a n d t h e l o w e s t i n t h y -
                                                                                             m u s (Fig. 3). L y m p h n o d e s as well as spleens f r o m C D 4 + 8 +
    Anti-AChR Antibodies in Serum and AChR-specific IgG                                      m i c e c o n t a i n e d h i g h e r n u m b e r s o f t h e s e cells t h a n t h e o t h e r
  Antibody-secreting Cells. S e r u m a n t i - M - A C h R a n t i b o d i e s              groups. In t h y m u s , n u m b e r s o f a n t i - A C h R I g G - s e c r e t i n g
  w e r e d e t e c t e d in t h e C D 4 + 8 + w i l d - t y p e m i c e f r o m w e e k 2   cells w e r e l o w t h r o u g h o u t , b u t a significant difference was
  p.i. T h e levels i n c r e a s e d gradually o v e r t h e o b s e r v a t i o n p e -    noticed between CD4+8 + and CD4-8- mice.
  r i o d (Fig. 2). In contrast, t h e levels o f a n t i b o d i e s in t h e                    AChR-reactive IFN-y-secreting Cells. As s h o w n in Fig.
  C D 4 - 8 - m i c e w e r e significantly l o w e r f r o m w e e k 2 p.i.                 4, t h e l y m p h n o d e s f r o m C D 4 + 8 + E A M G m i c e c o n t a i n e d
  u p to t h e end. T h e levels in t h e C D 4 - / - a n d C D 8 - / - m i c e              h i g h e r n u m b e r s o f A C h R - r e a c t i v e I F N - ' v - s e c r e t i n g T cells
  w e r e also significantly l o w e r t h a n i n t h e C D 4 + 8 + w i l d -               c o m p a r e d w i t h C D 4 - 8 - , C D 4 - / - , o r C D 8 - / - m i c e . In
  type m i c e f r o m w e e k 5 p.i. o n w a r d s , e x c e p t at a f e w t i m e         spleen, C D 4 + 8 + m i c e also h a d h i g h e r n u m b e r s o f these
  points. T h e r e w e r e n o differences in a n t i b o d y levels b e -                  cells t h a n C D 4 - 8 - a n d C D 4 - / - m i c e . T h e r e w e r e n o dif-
  t w e e n CD4 -/-, CD8 -/-, a n d C D 4 - 8 -                     mice. N o a n t i -      ferences b e t w e e n t h e f o u r g r o u p s in t h y m u s . N o differ-
  A C h R antibodies w e r e f o u n d in sera f r o m t h e C F A - i m m u -               e n c e s w e r e f o u n d w h e n cells f r o m l y m p h o i d o r g a n s w e r e
  nized control mice.                                                                        e v a l u a t e d in c o n t r o l e x p e r i m e n t s after c u l t u r e w i t h M B P

       35                                              []     CD8-/-
                                                                                                  4-                                               9      CD8-/-
                                                       []     CD4-8-
       30                                                                                                                                          []     CD4-8-
                                                       []     CD4+8+
                                                                                                                                                   []     CD4+8+
                                                       []     CD4-/-
                                                                                                                                                   []     CD4-/-

   =~ 2 0 -                                                                                  o

  ,,~ lO

                  PILN                  Spleen              Thymus                                O,
                                                                                                            PILN                  Spleen                Thymus
  Figure 3.   Numbers ofanti-AChR IgG antibody-secreting cells per 105
  MNC isolated from PILN, spleen, and thymus of the different groups.                        Figure 5. Proliferative responses to AChR by MNC from PILN,
  Symbols refer to mean values, and bars to SD. P values refer to compari-                   spleen, and thynms of the different groups. Symbols refer to mean values,
  sons between the CD4+8 + mice and the other three groups. (*) P <0.05;                     and bars to SD. Pvalues refer to comparisons between the CD4+8 + mice
  (**) P <0.01; and (***) P <0.001.                                                          and the other groups. (*) P <0.05; (**) P <0.01.

                                             352        CD4 § and CD8 + T Cells in Experimental Autoimmune Myasthenia Gravis
Published August 1, 1996

  (0.4-0.8/10 s M N C ) or without antigen or mitogen (1.0-                between all four groups for TGF-13 (20.6 -+ 4.7-35.3 +
  2.1/105 M N C ) , respectively. After culture with C o n A for           13.2/105 MNC).
  M N C from lymph nodes, the lowest numbers o f IFN-~/-
  secreting cells were found in C D 4 - 8 - mice (6.5 + 2.4/10 s
  M N C ) and the highest in CD4+8 + mice (13.6 +- 6.1/105
  M N C ) (P <0.05). N o differences were found between
  other groups.                                                                This study provides evidence that both C D 4 and C D 8 T
     AChR-induced Lymphocyte Proliferation. Proliferative re-              cells are involved in the induction o f E A M G as helper or
  sponses to A C h R were higher in lymph node cells o f                   effector cells. The importance o f C D 4 + T cells in the
  C D 4 + 8 + mice compared with the other groups. Significant             pathogenesis o f E A M G is in line with previous data on
  differences were also found in spleens between CD4+8 +                   E A M G and human M G (6, 7, 10, 20). However, the role
  and C D 4 - 8 - or C D 8 - / - mice (Fig. 5). All groups showed          o f C D 8 + T cells in E A M G is unclear. Shenoy et al. (12) re-
  similar high SI levels after stimulation by C o n A (5.4-9.0),           ported that [32m - / - mice with deficient M H C class I ex-
  and similar low SI levels after culture with the control anti-           pression and low C D 8 cell counts showed more severe
  gen MBP (0.7-1.3).                                                       E A M G compared with controls, and postulated that the in-
     AChR-reactive IFN-% IL-4, and TGF-[3 mRNA-express-                    creased disease severity was a result o f a lack o f regulator/
  ing Cells. Lymph nodes from CD4+8 + mice contained                       suppressor C D 8 + T cells (12). By using C D 8 - / - and
  higher numbers ofAChlk-reactive IFN-'y and IL-4 m R N A -                C D 4 - 8 - mice, we n o w present evidence that C D 8 + T
  expressing cells when compared with C D 4 - / - , C D 8 - / - ,          cells also function as effector/helper cells in E A M G . There
  and C D 4 - 8 - mice. The numbers o f I F N - ~ / - m R N A -            are several reasons w h y results between C D 8 - / - and
  expressing cells in C D 4 - / - mice were also higher than in                                              m
                                                                           [32m - / - mice differ. (a) 132 - / - mice indeed process func-
                                                                           tional C D 8 + T cells (21, 22). After in vivo priming and in

                                                                                                                                                      Downloaded from on January 5, 2010
  C D 4 - 8 - mice. There were no differences among the three
  groups of mutant mice for IL-4 mRNA-expressing cells.                                                                             m
                                                                           vitro restimulation with H - 2 d spleen cells, 132 - / - (H-2 b)
  Numbers o f AChR-reactive TGF-[3 mRNA-expressing                         responder mice developed the same strong cytotoxic re-
  cells were lower in C D 4 - 8 - mice compared with CD4+8 +               sponses against the allospecific H-2a-expressing targets as
  wild-type mice, but not among the other groups (Fig. 6).                 [32m + / - control mice (21). The specific ability o f residual
  N o differences were detected between the four groups for                C D 8 + T cells in [32m-deficient mice to react with class I
  MBP-induced cytokine mRNA-expressing cells (2.0-3.5/                     H - 2 b molecules on normal H-2b-expressing cells indicates
  105 M N C for I F N - % 1.3-2.5 for IL-4, and 2.6-3.3 for                that these T cells have not undergone negative selection
  TGF-[3) or after culture o f M N C without antigen or mito-              (23), and CD8 + T cells specific for self-peptides conse-
  gen (2.2-3.8 for I F N - % 1.3-2.5 for IL-4, and 2.2-3 for               quently may escape elimination. (b) Besides their restriction
  TGF-[3). After being stimulated by Con A, the numbers of                 by M H C class I, CD8 + T cells are also restricted by M H C
  mRNA-expressiNg cells for IFN-3' were 43.6 -- 9.7 in                     class II (24). These class II-restricted C D 8 + T cell subsets,
  C D 4 - 8 - mice arid 62.3 + 10.2 in C D 4 + 8 + mice; for IL-4          which can secrete I F N - % IL-4, and IL-10 (23), wilt sur-
  they were 5.3 -+ 1.9 in C D 4 - 8 - mice and 10.2 -+ 4.0 in              vive in [32m-deficient mice. (c) Rats depleted o f C D 8 + cells
  CD4+8 + mice (P <0.05, respectively). There were no dif-                 with mAb had less severe E A M G accompanied by de-
  ferences between the other groups for I F N - y and IL-4, or             creased B and T cell responses to AChlL (11). (d) An unex-
                                                                           pected function o f M H C class I was observed in experi-
                                                                           mental SLE, an autoantibody-mediated disease, since SLE-
                                                                           susceptible mice lacking M H C class I did not develop any
  z;                                   9    CD8-/-                         clinical SLE manifestations, whereas mice depleted o f
                                       []   CD4-8-                         C D 8 + T cells remained susceptible (25). These data indi-
       20                              []   CD4+8+
                                       []   CD4-/-                         cate that experimental manipulation o f C D 8 + T cells will
                                                                           not parallel manipulation o f M H C class I.
  .~ 15                                                                        T cells have for a long time been functionally classified
                                                                           according to their C D phenotype, i.e., C D 4 + T cells are

                                                                           helper/inducer and C D 8 + T cells are suppressor/cytotoxic
  <                                                                        cells. R e c e n t studies suggest that the functions o f t cells are
  Z                                                                        more complicated (26). There are suppressor C D 4 + T cells
  E 5                                                                      that can produce TGF-[3 and are important in inducing
                                                                           tolerance (27). C D 8 + T helper cells can also be classified
        0                                                                  into T h l and Th2 subsets according to their phenotypes
                IFN~          IL-4           TGF-~3                        and production o f cytokines (28, 29). Based on the produc-
                                                                           tion o f l L - 4 and I F N - y , the C D w 6 0 + C D 8 + cells are simi-
  Figure 6. Numbers of AChK-reactive IFN-% IL-4, and TGF-[3                lar to the C D 4 + Th2 subset, whereas the C D w 6 0 - C D 8 +
  mRNA-expressing cells per 105 MNC isolated from PILN of the differ-
  ent groups. Symbols refer to mean values, and bars to SD. (*) P <0.05;   cells resemble the T h l subset (28). C D 8 + cells were also
  (**) P <0.01 between the CD4+8 + mice and the other three groups; and    found to help B cells in antibody production through the
  (#) P <0.05 between CD4 -/- and CD4-8- mice.                             expression o f C D 4 0 ligand (30). Elimination o f either CD4 +

                                     353    Zhang et al.
Published August 1, 1996

  or CD8 + T cells suppresses some T cell- or antibody-medi-                   secreting cells are reduced in association with decreased
  ated experimental autoimmune diseases (31-34). The present                   muscular weakness (8, 41). T h e present study suggests that
  study suggests that both C D 4 + and C D 8 + cells are essential             both C D 4 + and C D 8 + antigen-specific T cells produce
  for the development o f E A M G . Interestingly, in human                    I F N - y and, thereby, provide help to induce E A M G . In
  M G , there is a striking association with the M H C class I al-             EAE, both C D 4 + and C D 8 + MBP-specific T cells can se-
  lele H L A B8, which is stronger (higher relative risk), than                crete large amounts o f IFN-'y. Furthermore, oral tolerance
  with the M H C class II allele D1K3 (35). In view o f the fact               to EAE could be established after depletion o f either C D 4 +
  that M H C class I is the main restriction element for C D 8 +               or C D 8 + T cells (42). Taken together, I F N - y seems to be
  T cells, the present observation is in agreement with a                      an important p r o m o t o r in both B - and T cell-mediated
  M H C class I association in M G .                                           a u t o i m m u n e diseases.
     Mice lacking C D 4 + T cells have normal development                          The decrease o f a n t i - A C h l K antibodies in C D 4 and C D 8
  and function o f C D 8 + T cells and some TC1K-c~/]3+CD4-8 -                 nmtant mice might be a result o f reduced levels o f IL-4
  cells (36). Mice lacking C D 8 + T cells have normal devel-                  m R N A - e x p r e s s i n g cells. IL-4 is involved in the differentia-
  opment and function o f C D 4 + cells (37). In C D 4 - 8 - m u -             tion o f resting B cells to lg-secreting cells (43) and is the
  tant mice, B cells and TCR--ot/]3+CD4-8 - cells have                         major a n t i - A C h R a n t i b o d y - p r o m o t i n g factor (3). It has
  expanded to occupy the compartment that would other-                         been observed that human M G and E A M G are associated
  wise have been occupied by CD4+8 - and C D 4 - 8 + cells                     with elevated numbers o f IL-4 m R N A - e x p r e s s i n g cells (16,
  (38). These results indicate that a reduced c o m m u n i c a t i o n /      44). Besides C D 4 + T cells, type 2 C D 8 + T cells can p r o -
  adhesion between antigen-specific T cells and A P C , as a                   duce similar amounts or even two or three times more IL-4
  consequence o f the absence o f marker molecules (i.e.,                      m R N A (24). O u r results are in line with these observa-
  C D 4 / C D 8 surface antigens) in remaining T cells, might                  tions. In the present study, lack o f either C D 4 + or C D 8 + T

                                                                                                                                                                Downloaded from on January 5, 2010
  also play a role in the amelioration o f E A M G in our animal               cells does not substantially decrease the amounts o f TGF-]3,
  model. O n the other hand, C D 4 or C D 8 cells separately                   whereas lack o f both subsets does, indicating that both C D 4 +
  are not sufficient to induce E A M G , suggesting that a col-                and C D 8 + T cells are to some extent sources o f TGF-]3.
  laboration between C D 4 and C D 8 T cells may be neces-                     T h o u g h TGF-]3 functions as an immunosuppressive cyto-
  sary. In EAE, it was speculated that C D 4 + effector cells ini-             kine, the increased TGF-]3 anmunts in wild-type CD4+8 +
  tially h o m e to the central nervous system, and C D 8 + T                  mice are not high enough to counteract disease develop-
  cells are subsequently recruited and may subserve an effec-                  ment. It can be concluded that the suppression o f E A M G
  tor function that augments the damage o f myelin (31). In                    in C D 4 - a n d / o r C D 8 - k n o c k o u t mice is mainly caused by a
  nonobese diabetic mice, C D 4 + cells h o m e to the pancreas                decrease in the production o f effectors like IFN-'y and
  and p r o m o t e the influx o f C D 8 + cells, which finally leads          IL-4, but not mediated by T G F - ~ .
  to the full d e v e l o p m e n t o f insulitis (39). W h e t h e r or not       Notably, C D 4 - 8 - mice developed some amounts o f
  this is the case in antibody-mediated E A M G is currently                   a n t i - A C h R antibodies in serum after inmmnization with
  under investigation.                                                         A C h R plus CFA. This w o u l d suggest that B cells can be
     The Thl-associated proinflammatory I F N - y may play a                   activated to produce anti-AChlK antibodies to some degree
  role in both human M G and E A M G (5, 8). In transgenic                     without the help o f costimulatory cell surface antigens on
  mice expressing IFN-y in neuromuscular junctions, au-                        C D 4 + and C D 8 + cells. A n o t h e r possibility is that, besides
  toantibody deposits at m o t o r end plates imply that I F N - y             T cells, additional types o f cells such as macrophages (45)
  in the milieu o f the muscle tissue induces humoral auto-                    can provide B cell help. Further studies defining the roles
  immunity (40). After tolerance induction by oral or nasal ad-                o f i m m u n e cells b e y o n d T cells are thus crucial to under-
  ministration o f A C h l Z , numbm~ ofAChlK-reactive I F N - y -             stand the development o f M G and E A M G .

                                     We thank professor Tak W. Mak for providing the CD4 /-, C 1 ) 8 - , and CD4-8- mutant C57BL/6J
                                     mice, and Dr. Hans-GustafLjunggren for helpful discussion of the manuscript.

                                     This study was supported by grants from the Swedish Medical Research Council, the Swedish MS Society
                                     (NHR), Karolinska Institute, and the AFA Foundation.

                                     Address correspondence to Dr. Guang-Xian Zhang, Division of Neurology, Huddinge Hospital, S-14l 86
                                     Huddinge, Stockholm, Sweden.

                                     Received for publication 19 Februa~T 1996 and in revisedform 23 April I996.

   1. Berman, P.W., and J. Patrick. 1980. Experimental myasthe-                    1797-1810.
      nia gravis--a murine system.J. Exp. Med. 15l:204-223.                     3. Asthana, D., Y. Fujii, G. Huston, and J. Lindstrom. 1993.
   2. Drachman, D. 1994. Myasthenia gravis. N. Etl,gl.l. e~h'd. 330:               Regulation of antibody production by helper T cell clones in

                                     354       CD4 + and CD8 + T Cells in Experimental Autoimmune Myasthenia Gravis
Published August 1, 1996

      experimental autoimmune myasthenia gravis is mediated by                      R.K. Assoian, A.B. Roberts, M.B. Sporn, and D.V. Goeddel.
      IL-4 and antigen-specific T cell factors. Clin. Immunol. lmmu-                1985. Human transforming growth factor-J3 complementary
      nopathol. 67:240-248.                                                         DNA sequence and expression in normal and transformed
   4. Hohlfeld, R., K.V. Toyka, S.J. Tzartos, W. Carson, and                        cells. Nature (Lond.). 316:701-705.
      B.M. Conti-Tronconi. 1987. Human T helper lymphocytes                   20.   Waldor, M., S. Sriram, H. McDevitt, and L. Steinman. 1983.
      in myasthenia gravis recognize the nicotinic receptor subunit.                In vivo therapy with monoclonal anti-I-A antibody sup-
      Proc. Natl. Acad. Sci. USA. 84:5379-5383.                                     presses immune responses to acetylcholine receptor. Immunol-
   5. Link, H., O. Olsson, J.B. Sun, W.Z. Wang, G. Andersson,                       ogy. 80:2713-2717.
      H.P. Ekre, T. Brenner, O. Abramsky, and T. Olsson. 1991.                21.   Glas, R., C. Ohlen, P. H6glund, and K. K~irre. 1994. The
      Acetylcholine receptor reactive T and B cells in myasthenia                   CD8 + T cell repertoire in [32-microglobulin-deficient mice
      gravis and controls.J. Clin. Invest. 87:2191-2196.                            is biased towards reactivity against self-major histocompatibil-
   6. Fujii, Y., andJ. Lindstrom. 1988. Specificity of the T cell im-               ity class I.J. Exp. Med. 179:661-672.
      mune response to acetylcholine receptor in experimental au-             22.   Bix, M., and D. Raulet. 1992. Functionally conformed free
      toimmune myasthenia gravis. Response to subunits and syn-                     class I heavy chains exist on the surface of [32 microglobulin
      thetic peptides.J. Immunol. 140:1830-1837.                                    negative cells.J. Exp. Med. 176:829-834.
   7. Lennon, V.A., D.J. McCormick, E.H. Lambert, G.E. Gries-                 23.   Ljunggren, H.-G., L.V. Kaer, P.G. Ashton-Rickardt, S.
      mann, and M.Z. Atassi. 1985. Region ofpeptide 125-147 of                      Tonegawa, and H.L. Ploegh. 1995. Differential reactivity of
      acetylcholine receptor alpha subunit is exposed at neuromusclar               residual CD8 + T lymphocytes in TAP1 and [32-microglobu-
      junction and induces experimental autoimnmne myasthenia                       lin mutant mice. Eur.J. Immunol. 25:174-178.
      g-cavis. T cell immunity, and modulating autoantibodies. Proc.          24.   Kemeny, D., A. Noble, B. Holmes, and D. Diaz-Sanchez.
      Natl. Acad. Sci. USA. 82:8805-8809.                                           1994. Immune regulation: a new role for the CD8 + T cells.
   8. Wang, Z.Y., J. Qiao, A. Melms, and H. Link. 1993. T cell                      Immunol. Today. 15:107-110.
      reactivity to acetylcholine receptor in rats orally tolerized against   25.   Mozes, E., L.D. Kohn, F. Hakim, and D.S. Singer. 1993.

                                                                                                                                                       Downloaded from on January 5, 2010
       experimental autoimmune myasthenia gravis. Cell. Immunol.                    Resistance of MHC class I-deficient mice to experimental sys-
       152:394-404.                                                                 temic lupus erythematosus. Science (Wash. DC). 261:91-92.
   9. Christadoss, P., and M. Dauphinee. 1986. Immunotherapy                  26.   Kelso, A. 1995. Thl and Th2 subsets: paradigms lost? Immu-
      for myasthenia gravis: a murine model. J. Immunol. 136:                       nol. Today. 16:374-379.
      2437-2440.                                                              27.   Karpus, W.J., and R.H. Swanborg. 1991. CD4 + suppressor
  10. Kaul, R., M. Shenoy, E. Goluszko, and P. Christadoss. 1994.                   cells inhibit the function of effector cells of experimental au-
      Major histocompatibility class II gene disruption prevents ex-                toimmune encephalomyelitis though a mechanism involving
      perinaental autoimmune myasthenia gravis. J. Immunol. 152:                    transforming growth factor-]3. J. Immunol. 146:1163-1168.
       3152-3157.                                                             28.   Rieber, E.P., and G. Rank. 1994. CDw60: a marker for hu-
  11. Zhang, G.X., C.G. Ma, B.G. Xiao, M. Bakheit, H. Link, and                     man CD8 + T helper cells.J. Exp. Med. 179:1385-1390.
       T. Olsson. 1995. Depletion of CD8 + T cells suppresses ex-             29.   Salgame, P., J.S. Abrams, C. Clayberger, H. Goldstein, J. Con-
      perimental autoimmune myasthenia gravis in Lewis rats. Eur.                   vit, R.L. Modlin, and B.R. Bloom. 1991. Differing lympho-
      J. Immunol. 25:1191-1198.                                                     kine profiles of functional subsets of human CD4 and CD8 T
  12. Shenoy, M., R. Kaul, E. Goluszko, C. David, and P. Christa-                   cell clones. Science (Wash. DC). 254:279-282.
       doss. 1994. Effect of MHC class I and CD8 cell deficiency on           30.   Seder, R.A., and G.L. Gros. 1995. The functional role of
      experimental autoimmune myasthenia gravis pathogenesis. J.                    CD8 + T helper type 2 cells.J. Exp. Med. 181:5-7.
       lmmunol. 152:5330-5335.                                                31.   Koh, D.R., W.P. Fung-Leung, A. Ho, D. Gray, H. Acha-
  13. Lindstrom, J., B. Einarson, and S. Tzartos. 1981. Production                  Orbea, and T.W. Mak. 1992. Less mortality but more re-
       and assay of antibodies to acetylcholine receptors. Methods                  lapses in experimental allergic encephalomyelitis in CD8 - / -
       Enzymol. 74:432-460.                                                         mice. Science (Wash. DC). 256:1210-1213.
  14. Deibler, G.E., R.E. Martenson, and M.V. Kies. 1972. Large               32.   Fowell, D., and D. Mason. 1993. Evidence that the T cell
       scale preparation of myelin basic protein from central nervous               repertoire of normal rats contains cells with the potential to
       tissue of several mammalian species. Prep. Biochem. 2:139-                   cause diabetes. Characterization of the CD4 + T cell subset
       165.                                                                         that inhibits this autoimmune potential. J. Exp. Med. 177:
  15. Van der Meide, P.H., M. Dubbeld, K. Vijverberg, T. Kos,                       627-636.
      and H. Schellekens. 1986. The purification and characteriza-            33.   Kong, Y., H. Waldmann, S. Cobbold, A. Giraldo, B. Fuller,
       tion of rat gamma interferon by use of two monoclonal anti-                  and L. Simon. 1989. Pathogenic mechanisms in murine au-
      bodies.J. Gen. Virol. 67:1059-1071.                                           toimmune thyroiditis: short- and long-term effects of in vivo
  16. Wang, Z.Y., H. Link, .~.. Ljungdahl, B. H6jeberg, J. Link, J.                 depletion ofCD4 + and CD8 + cells. Clin. Exp. Immunol. 77:
       Qiao, A. Melms, and T. Olsson. 1994. Induction of IFN-%                      428-433.
       IL-4 and TGF-[3 in rats orally tolerized against experimental          34.   Pummerer, C., P. Berger, M. Fruhwirth, C. C)fner, and N.
      autoimmune myasthenia gravis. Ceil. Immunol. 157:353-368.                     Neu. 1991. Cellular infiltrate, major histocompatibility anti-
  17. Dijkema, R., P.H. van der Meide, M. Dubbeld, J. Wubben,                       gen expression and immunopathogenic mechanisms in car-
      and H. Schellekens. 1986. Cloning, expression and purifica-                   diac myosin-induced myocarditis. Lab. Invest. 65:538-547.
      tion of rat IFN-% Methods Enzymol. 119:453-463.                         35.   Dawkins, R.L., F.T. Christiansen, P.H. Kay, M. Garlepp, J.
  18. McKnight, A.J., A.N. Barclay, and D.W. Mason. 1991. Mo-                       McCluskey, P.N. Hollingsworth, and p.J. Zilko. 1983. Dis-
      lecular cloning of rat interleukin 4 cDNA and analysis of the                 ease associations with complotypes, subtypes and haplotypes.
       cytokine repertoire of subsets of CD4 + T cells. Eur. J. lmmu-               Immunol. Rev. 70:5-22.
       nol. 21:1187-1194.                                                     36.   Rahemtulla, A., W.P. Fung-Leung, M.W. Schilham, T.M.
  19. Derynck, R., J.A. Jarrett, E.Y. Chen, D.H. Eaton, J.R. Bell,                  Kundig, S.R. Sambhara, A. Narendran, A. Arabian, A.

                                    355       Zhang et al.
Published August 1, 1996

      Wakeham, C.J. Paige, P,.M. Zinkernagel, et al. 1991. Nor-         Exp. Med. 181:547-557.
      mal development and function of CD8 + cells but markedly      41. Ma, C.G., G.X. Zhang, B.G. Xiao, J. Link, T. Olsson, and
      decreased helper cell activity in mice lacking CD4. Nature        H. Link. 1995. Suppression of experimental autoimmune
       (Loud.). 353:180-184.                                            myasthenia gravis with nasal administration of acetylcholine
  37. Fung-Leung, W.P., M.W. Schilham, A. Rahemtulla, T.M.              receptor.J. Neuroimmunol. 58:51-60.
      Kundig, M. Vollenweider, J. Potter, W. van Ewijk, and         42. Chen, Y.,J. Inobe, and H.L. Weiner. 1995. Induction of oral
      T.W. Mak. 1991. CD8 is needed for development of cyto-            tolerance to myelin basic protein in CD8-depleted mice:
      toxic T cells but not helper T cells. Cell. 65:443-449.           both CD4 + and CD8 + cells mediate active suppression.J. Im-
  38. Schilham, M.W., W,P. Fung-Leung, A. Rahemtulla, T.                munol. 155:910-916.
      Kundig, L. Zhang, J. Potter, P,.G. Miller, H. Hengartner,     43. Lanzavecchia, A. 1985. Antigen-specific interaction between
      and T.W. Mak. 1993. Alloreactive cytotoxic T cells can de-        T and B cells. Nature (Lond.). 314:537-539.
      velop and function in mice lacking both CD4 and CD8. Eur.     44. Link, J., M. SSderstr6m, A. Ljungdahl, B. HSjeberg, T. Ols-
      J. Immunol. 23:1299-1304.                                         son, Z. Xu, S. Fredrikson, Z.Y. Wang, and H. Link. 1994.
  39. Thivolet, C.A., A. Bendelac, P. Bedossa, J-F. Bach, and C.        Organ-specific autoantigens induce IFN-y and IL-4 m R N A
       Carnaud. 1991. CD8 + T cell homing to the pancreas in the        expression in mononuclear cells in multiple sclerosis and my-
      nonobese diabetic mouse is CD4 + T cell-dependent. J. Irn-        asthenia gravis. Neurology. 44:728-734.
       munot. 145:85-93.                                            45. Kinoshita, I., T. Nakamura, A. Satoh, H. Matsuo, M. Seto, I.
  40. Gu, D., L. Wogensen, N.A. Calcutt, C. Xia, S. Zhu, J.P.           Tomita, M. Tsujihata, and S. Nagataki. 1988. Role of mac-
      Merlie, H.S. Fox, J. Lindatrom, H.C. Powell, and N. Sarvet-       rophage in the pathogenesis of experimental autoimmune my-
       nick. 1995. Myasthenia gravis-like syndrome induced by ex-       asthenia gravis.J. Neurol. Sci. 87:49-59.
      pression of interferon-"/ in the neuromuscular junction. J.

                                                                                                                                        Downloaded from on January 5, 2010

                                356     CD4 + and CD8 + T Cells in Experimental Autoimmune Myasthenia Gravis

To top