T-cell vaccination in Experimental Autoimmune Encephalomyelitis

1. Journal of Neuroimmunology, 44 (1993) 85-94 85
© 1993 Elsevier Science Publishers B.V. All rights reserved 0165-5728/93/$06.00
JNI 02350
Enhanced response to antigen within lymph nodes of SJL/J mice
that were protected against experimental allergic encephalomyelitis
by T cell vaccination
R a n a Zeine "~, Diane H e a t h ¢ and Trevor Owens a,b
a Department of Mech'cine and b Department of Neurology and Neurosurgery, and c Montreal Neurological Institute, McGill University,
Montreal, Quebec, Canada
(Received 19 August 1992)
(Revision received 11 November 1992)
(Accepted 11 November 1992)
Key words: Experimental allergic encephalomyelitis; T cell vaccination; Anti-ergotypic response
Summary
The effects of T cell vaccination on peripheral immune responsiveness are not yet fully understood. We have
induced resistance to rat spinal cord homogenate (RSCH)-induced experimental allergic encephalomyelitis (EAE)
in S J L / J mice by vaccination with four T cell lines (RZ8, RZ15, RZ16, and A51) which were reactive to myelin
basic protein (MBP) but not to proteolipid protein (PLP). The effect was relatively neuroantigen-specific since
vaccination with ovalbumin (OVA)-reactive and alloantigen-specific cells did not prevent EAE induction. Alloanti-
gen-reactive cells reduced the rate of relapse. The number of central nervous system (CNS) infiltrates and mean
clinical EAE scores were significantly reduced. This is the first report demonstrating T cell vaccination in the SJL/J
mouse, a strain in which PLP is the predominant encephalitogen in RSCH. The vaccinating cells were of the
memory/effector (CD44 high, CD45RB ~°~) surface phenotype. We examined the effect of T cell vaccination on
lymph node T cell proliferative responses to MBP, encephalitogenic peptides of PLP and MBP, OVA and anti-CD3.
With the exception of polyclonal cytokirre responses to anti-CD3, which remained unchanged, vaccination led to a
5-10-fold augmentation in all, including background, responses. By comparison with lymph node cell (LNC)
responses from naive mice and mice primed with OVA, it appeared that T cell vaccination restored cellular
activation levels which had been depleted in peripheral lymphoid tissues of unvaccinated animals with EAE.
Introduction injection of encephalitogenic T cells that have been
attenuated by irradiation, hydrostatic pressure, glu-
Experimental allergic encephalomyelitis (EAE) is an taraldehyde fixation or ganglioside treatment (Ben-Nun
autoimmune demyelinating disease of the central ner- et al., 1981; Lider et al., 1986; Offner et al., 1989). The
vous system (CNS) which is induced by CD4 ÷ T cells TcR repertoire among encephalitogenic T cells is re-
specific for neuroantigens (Raine, 1985). A fundamen- stricted in many EAE models (Acha-Orbea et al., 1988;
tal requirement for the encephalitogenicity of T cells is Sakai et al., 1988; Padula et al., 1991) and it is possible
their expression of T cell receptors (TcR) that recog- to prevent the disease by immunization with TcR V-re-
nize relevant epitopes present on CNS proteins such as gion peptides (Vandenbark et al., 1989; Howell et al.,
MBP (Schluesener and Wekerle, 1985; Vandenbark, 1989).
1985) or proteolipid protein (PLP) (Satch et al., 1987; The cellular mechanisms involved in vaccination with
Kennedy et al., 1990). EAE can be prevented by prior T cells or TcR peptides are not yet fully understood.
Anti-T cell immune responses have been characterized
in protected animals including cytolytic T - T cell inter-
Correspondence to: T. Owens, Montreal Neurological Institute, 3801 actions involving cytotoxic CD8 ÷ T cells (Sun et al.,
University Street, Montreal, Quebec, Canada H3A 2B4. 1988), and suppressive responses mediated by anti-idi-

2. 86
otypic CD8 ÷ cells (Vandenbark et al., 1989; Lider et cells/ml in 1-ml flat-bottomed tissue culture wells
al., 1988; Janeway, 1989). Non-specific effector mecha- (Falcon/Fisher, Montreal, Quebec) in RPMI 1640
nisms such as TGF~ secretion have been described for (Gibco/BRL, Burlington, Ontario), supplemented with
antigen specific suppressive T cells (Miller et al., 1992). 10% fetal calf serum (FCS) (ICN, Mississuaga, On-
In addition to specific regulatory mechanisms directed tario), 2 mM L-glutamine (Calbiochem, SanDiego, CA),
at T cells of a defined antigen specificity, more general 5 x 10 -5 M 2-mercaptoethanol (2-ME) (Sigma, St.
responses against activated T cells ('anti-ergotypic') Louis, MO), and 100 U / m l penicillin-100 /zg/ml
have been detected in vaccinated animals. These have streptomycin (Gibco/BRL Burlington, Ontario). Cul-
been demonstrated by positive delayed type hypersen- tures were stimulated by the addition of 50 /xg/ml
sitivity (DTH) reactions to T cell lines and clones MBP or OVA. Responsiveness to MBP or OVA was
(Offner et al., 1989; Lohse et al., 1989) and by the assessed in parallel microcultures by [3H]thymidine
prevention of passive transfer of EAE by anti-ergotypic incorporation at 4 days following an overnight pulse
T cells that recognize not the idiotype of the autoim- (0.5 p, Ci/culture) (ICN Biomedicals Inc., Mississauga,
mune T cells, but as yet unidentified markers induced Ontario). MBP-reactive cells were collected after 4
by activation (Lohse et al., 1989). days in culture, centrifuged on Ficoll-Hypaque (Phar-
To examine the effects of T cell vaccination on macia, Montreal, Quebec), washed in HBSS and 1 × 107
peripheral immune responses, we developed a series of blasts per mouse were injected intravenously for as-
T cell lines that were encephalitogenic in SJL/J mice. sessment of encephalitogenicity. The MBP-specific cell
SJL/J mice vaccinated with these MBP-reactive T cells lines (RZ8, RZ15, RZ16 and A51) and OVA-specific
were protected against RSCH-induced EAE. The pro- cell lines were derived by sequential restimulation of
liferative response of lymph node (LN) T cell popula- such cultures. Cells were collected, centrifuged on Fi-
tions from vaccinated mice was compared to that of coil and recultured at 10-day intervals at 1 x 106
untreated controls. Both background responses and cells/ml with MBP (50 /xg/mi), 5 x 10 6 irradiated
responses to the encephalitogenic PLP peptide p139- (3000 R) SJL/J spleen ceils (SC) and appropriate
151 were enhanced by vaccination. The antigen cross- concentrations of IL-2-containing supernatant from
reactivity of vaccination suggested the operation of ConA-stimulated rat spleen cells (CAS, an ammonium
non-specific effector mechanisms in mediating protec- sulphate-purified preparation from ConA-stimulated
tion. The enhancement of LN responses could be at- rat spleen cells) (Owens and Miller, 1987). The anti-H-
tributed to non-specific (e.g. anti-ergotypic) responses, 2 k cell line Sk4d2 was derived from an in vitro mixed
elicited by the vaccinating T cells. The component of lymphocyte reaction (MLR) (B10.S anti-B10.BR) and
augmented LNC responses in vaccinated mice that was was maintained in a similar manner.
antigen-specific may be attributable to the persistence
of effector functions, normally depleted from the pe- Assessment of proliferation and antigen reactivity
riphery during EAE. LN were isolated from mice and 8 X 105 LNC were
assayed in 200/.tl cultures with or without antigen. For
T cell lines, ceils were collected from culture, cen-
Materials and Methods trifuged on Ficoll, and 2 x 105 ceils were assayed in
200 /xi cultures with 5 x 105 irradiated (3000 R) syn-
Mice geneic spleen ceils as antigen-presenting cells (APC).
SJL/J female mice of between 4 and 8 weeks old Protein antigens (purified protein derivative (PPD)
were purchased from Harlan Sprague Dawley (Indi- (Cedarlane, Hornby, Ontario), OVA (Calbiochem, La
anapolis, IN). Jolla, CA) and MBP) were added at 50/zg/mi. Peptide
antigens (guinea pig MBP p90-99 and p88-99 (HF-
Derication of T cell lines and assessment of encephalito- FKNIVTPRTP (Hashim et al., 1986), generously pro-
genicity vided by Dr. G. Hashim, St. Luke's-Roosevelt Hospital
Mice were immunized by subcutaneous injection Center, New York); mouse MBP p90-102 (FFKNIVT-
with 400 p.g of either bovine MBP (Sigma, St. Louis, PRTPPP (Whitham et al., 1991), Multiple Peptide
MO) or ovalbumin (OVA) (Calbiochem, La Jolla, CA) Systems, San Diego, CA); PLP p139-151
in complete Freund's adjuvant (CFA) (Difco, Detroit, (HSLGKWLGHPDKF (Whitham et al., 1991), Sheldon
MI), containing 50 ~.g H37RA per mouse, and boosted Biotechnology Centre, Montreal, Quebec); and an ir-
after 7 days. Brachial, axillary, inguinal, and para-aortic relevant control peptide (CKEQFLDGWTDRWIES,
lymph nodes (LN) were collected 14 days after initial obtained from Dr. M. Newkirk, Montreal General
immunization and a single-cell suspension prepared by Hospital)) were added at 15/~g/ml. After 48 or 60 h
disruption through stainless steel mesh into Hanks' [3H]thymidine (0.5 p.Ci/culturc)was added. Cultures
balanced salt solution (HBSS) (Gibco/BRL Burling- were harvested either 6 (for cell lines) or 18 (for LNC)
ton, Ontario). LNC were washed and plated at 4 x 10 6 h later onto glass fiber filters (Skatron, ICN Biomedi-

3. 87
cals Inc., Mississauga, Ontario) and incorporated ra- mononuclear cells was determined for each brain from
dioactivity measured by liquid scintillation counting. three sections 0.25 cm apart. The mean percentage of
perivascular infiltrates was calculated for each group
Antibodies and surface phenotype analysis (n = 4).
Monoclonal antibodies included phycoerythrin-cou-
pied anti-CD4 (PE-CD4) (Becton-Dickinson, Mountain Assessment of LNC lymphokine responses
View, CA), anti-CD8 (53-6.7) (Ledbetter and Herzen- Brachial, axillary, inguinal and para-aortic LN were
berg, 1979), anti-CD45RB (23G2) (Birkeland et al., collected on day 14 from vaccinated and control mice
1989), anti-CD44/Pgp-1 (IM7.8.1)(Trowbridge et al., that had been primed and boosted with either RSCH
1982) and anti-CD3, (145-2Cll) (Leo et al., 1987). or OVA in CFA. LNC (4 × 106) were stimulated by
Fluorescinated and biotinylated antibodies were puri- overnight culture in 1 mi of RPMI 1640, 5% FCS, 50
fied from culture medium by Protein G-sepharose tzM 2-ME and 2 mM t.-glutamine with either 15/zg/ml
affinity chromatography (Pharmacia, Montreal, Que- MBP p90-102, 50 tzg/ml OVA, 15/zg/ml PLP p139-
bec) and either coupled with biotin (53-6.7) using 151), or in 1-ml culture wells (Falcon/Fisher, Mon-
biotinamidocaproate N-hydroxy succinimide ester treal, Quebec) to which anti-CD3 (145-2Cll) (10
(Sigma), or fluorescinated (145-2Cll) using fluores- p.g/ml) had been adsorbed. Culture Supernatants were
cein isothiocyanate (FITC) (Sigma). For surface stain- collected after 24 h for measurement of lymphokine
ing, T cells (1 x 106) were collected, centrifuged on content.
Ficoll and incubated with either PE-CD4, biotin-53-
6.7, FITC-145-2Cll, 23G2 or IM7.8.1 culture medium Cytokine bioassays
at 4°C for 20 min and then washed. Where required, IL-2 and IL-4 titers in supernatants from T cell
the cells were then incubated with either FITC-cou- activation cultures were measured by bioassay on the
pled streptavidin (Bio-Can Scientific, Toronto, On- CTLL-6 (Firestein et al., 1989) and CT-4S (Hu-Li et
tario), or FITC-coupled goat anti-rat Ig (Southern al., 1989) cell lines, respectively. IFN-7 titers were
Biotechnology, Birmingham, AL). Surface staining was determined by inhibition of the proliferation of the
analyzed using a FACScan (Becton Dickinson, Moun- WEHI-279 cell line (Kelso, 1990). For all assays, 104
tain View, CA). cells were added to 100 /zl cultures containing titra-
tions of supernatants or standards. Cell survival a n d / o r
Vaccination with T cells proliferation were read out colorimetrically after 2
T cells were collected from culture 2-3 days follow- (IL-2 and IL-4) or 3 days (IFN-7), using MTF (Sigma)
ing stimulation with either MBP or OVA (50 p.g/ml) (Mosmann, 1983). Optical densities at 550 nm with a
or H2 k spleen ceils (106/well), and centrifuged on reference at 690 nm were read using an SLT EAT 400
Ficoil. The T cells were irradiated (2000 R), washed in muitiwell photometer (Fisher). Lymphokine titers in
HBSS and between 5 × 106 and 1 × 107 were injected units per ml were calculated by reference to titrations
intravenously per mouse in PBS. of rIL-2 (Cetus, San Francisco, CA; a gift from Dr. A.
Kelso), rlL-4 (DNAX, Palo Alto, CA; a gift from Dr.
Induction and assessment of EAE T. Mosmann), and rIFN-7 (Genzyme, Markham, On-
Three weeks post vaccination, vaccinated and con- tario).
trol mice were immunized with 0.5 RSCH in CFA for
the induction of EAE and boosted 7 days later. Mice
were monitored daily for symptoms and assigned clini- Results
cal scores as follows: 0 (no symptoms), 1 (flaccid tail,
clumsiness), 2 (moderate paresis), 3 (severe paresis or Characterization of cells used for vaccination
unilateral hind limb paralysis), 4 (bilateral hindlimb Four MBP-reactive cell lines were derived from
paralysis), 5 (moribund). LNC isolated from S J L / J mice that had been primed
and boosted with MBP in CFA. The RZ8, RZ16 and
Histology A51 cell lines were encephalitogenic when tested after
Mice were anaesthetized with chloral hydrate (3.5 4 days in culture (not shown). Encephalitogenic cell
g/kg) (Fisher), then perfused through the left ventricle lines, in culture, lost their pathogenicity within a few
with 100 ml paraformaldehyde (PFA) (Fisher) (4%) in weeks. For instance, the RZ15 cell line was not en-
PBS. Brains were removed, post-fixed in PFA and cephalitogenic when tested after 3 weeks in culture.
quick-frozen in O.C.T. compound (Miles) in isopen- The cells used for vaccination had been propagated in
tane on dry ice. Coronal sections were cut on a cryo- vitro for 7-9 weeks and were not encephalitogenic at
stat, dried overnight and stained with hematoxylin- the time of vaccination. These cell lines were MBP-
eosin for assessment of mononuclear cell infiltration. specific and did not respond to OVA or PPD (Fig. 1).
The percent of vessels surrounded by more than 10 In vitro culture with MBP selected for cells with reac-

4. 88
Reactivity of RZ15 MBP-specific cell lines did not respond to the PLP
control p.
peptide p139-151 which is the predominant encephali-
I J
togen in RSCH-induced E A E in S J L / J mice (Whitham
88-99
I' et al., 1991) (SI between 0.4 and 1.3 were obtained).
90-99
14
All lines expressed high levels of CD3, CD4 and a/fl
.~ 90-102
,-i
TcR on the day they were used for vaccination (Fig. 2).
E MBP The majority of cells had acquired the CD44 high CD45
OVA 1 RB ~°* phenotype which is associated with activation by
PPO l antigen recognition and with effector function in vivo
SJL (Zeine and Owens, 1992; Jensen et al., 1992; Weinberg
1
medium et al., 1992) (Fig. 2).
I I I
10000 20000 30000
Thymidlne incorporation (cpm) Vaccination of SJL / J mice against EAE
EAE was induced by two injections of RSCH in
CFA. Clinical signs of EAE included flaccid tails,
Reactivity of RZ16 clumsiness, and moderate paresis which progressed in
some mice to severe paresis. The severity of disease
88-99
was reduced following T cell vaccination. Figure 3
90-99 shows that the daily mean clinical scores of mice vacci-
nated with either RZ15, RZ16, or A51 were reduced as
e~ MBP
compared to unvaccinated controls. In the RZ15 and
=1
E OVA RZ16 vaccination experiments, the proportion of mice
m
PPD
that showed signs of EAE was similar in vaccinated
and unvaccinated groups (100% and 50%, respectively).
SJL
The reduction in mean clinical scores reflected de-
medium creased disease severity per mouse, as well as a more
! | !
20000 40000 60000
rapid recovery in vaccinated groups (Fig. 3). Whereas
ThymIdlneincor~mtion (¢pm) the maximum scores recorded in control mice were 3 in
the RZ16 experiment and 2 in the RZ15 experiment,
clinical scores never exceeded 1 in vaccinated mice. In
Reactivity of A51 the third experiment, vaccination with A51 reduced the
proportion of affected mice from 50% to 30%.
controlp. •
In addition, histological assessment of brains from
88-99
I animals that had been vaccinated with either RZ8 or
90-99
I IJ
RZ15 showed markedly reduced numbers of CNS
90-102 mononuclear infiltrates as compared to untreated con-
"~ MBP trois (Table 1). In one experiment, mice were also
E
OVA • preimmunized with syngeneic nylon wool purified LN
• T cells (NWT), prepared from unimmunized mice. This
• also reduced CNS infiltration, but not as profoundly as
medium with RZ8 (Table 1).
! i i
Control cell lines reactive to OVA and to allo-anti-
10000 20000 30000
Thymidlne Inco~oraflon(cpm) gen (H2 k, SK4D2) did not affect the severity of the
Fig. 1. Antigen-reactivity of T cell lines. SJL/J mice were primed
first episode of disease indicating specificity of the
with 0.4 mg MBP in CFA and boosted after 7 days. LN were isolated effect of MBP-specific cell lines. A maximum clinical
at 14 days and LNC were cultured with MBP and IL-2 as described score of 2 was recorded following vaccination with
in Materials and Methods. Responsiveness of 2× 105 cells to pro- SK4D2, similar to unvaccinated groups mentioned
teins (50 ;zg/ml MBP, PPD, or OVA) and peptides (15 p.g/ml MBP above. However, both MBP-specific and control lines
p88-99, p90-99, p90-102, or irrelevant peptide) was assessed in
influenced the rate of relapse. Table 2 shows the
microcultures by [3H]thymidine incorporation at 2 days. Results
show counts per minute (cpm):l: standard error of the mean (SEM). results of those experiments where relapse occurred.
The MBP-reactive lines RZ16 and A51, as well as the
allo-antigen-specific line SK4D2, reduced the inci-
tivities to MBP peptides such as p90-99 and p90-102, dence of relapse, whereas the OVA-reactive line did
which are encephalitogenic in S J L / J mice (Kono et al., not (Table 2). There were no relapses in either the
1988) (Fig. 1). Whereas stimulation indices (SI) of 3 to mice vaccinated with RZ15 or the corresponding con-
10 were obtained in response to whole MBP (Fig. 1), trol group.

5. 89
TABLE 1
i i
Reduction in the extent of CNS infiltration following T cell vaccina-
tion. Assessment of brain mononuclear infiltrates
CD3
Vaccination Mean of percent
perivascular infiltrates
Exp. i
None 16.8_+ 4.9
NWT 7.5+ 3
RZ8 0 + 0
TCR~p
Exp. 2
None 20.7 + I 1
RZ15 11 _+ 4
Mice were primed with RSCH + CFA 3 weeks after vaccination with
I X 10 7 irradiated T cells, and boosted after 7 days. Three weeks
later the mice were perfused with fixative and the brains collected
C04 for H&E staining. Percent perivascular infiltrates for each mouse
was calculated as: (Number of vessels surrounded by > 10 mononu-
clear cells/ Total number of vessels observed in three sections 0.25
cm apart)x 100. The figures represent means from four animals+
standard deviation. Percent perivascular infiltrates for normal un-
F.
I "
primed SJL/J mice in our animal facility is nil.
[ ,
--- CD44
~" " r " Furthermore, in our experience only 5-10-fold differ-
CD45RB ences or greater correlate with biological significance
in this assay. IL-2 and IL-4 titers were not affected by
1ol Io 2 lo 3 vaccination (Table 3).
Fluorescence i n t e n s i t y Fig. 4A compares proliferation in response to neu-
Fig. 2. Surface phenotype of cells used for vaccination. RZ15 cells roantigens and O V A of LNC from vaccinated and
were collected from culture, centrifuged on Ficoll, and stained with control mice that had been immunized with either
either FITC-145-2Cll, PE-CD4, biotinylated H57, 1M7.8.1 super- R S C H or OVA. Immunization of mice with R S C H in
natant, or 23G2 supernatant. H57 was visualized with FITC-strep-
the absence of vaccination induced weak responses to
tavidin. IM7.8.1 and 23G2 were visualized with FITC-goat-anti-rat
Ig. The cells were analyzed by FACS as described in Materials and the encephalitogenic PLP peptide 139-151 and to MBP
Methods. Dead cells were excluded by side scatter gating. 5000 (Fig. 4A). Responses to whole MBP were stronger than
events are shown for each histogram, stippled line represents control those to p91-102, reflecting the multi-determinant na-
staining in the absence of primary antibody. ture of the whole protein. Vaccination with the RZ15
or A51 cell lines enhanced thymidine incorporation in
response to all stimuli, including background prolifera-
Responsiveness of LNC from vaccinated and control tion (Fig. 4A). The stimulation indices did not change.
mice In the case where animals were immunized with R S C H
T o c o m p a r e p e r i p h e r a l i m m u n e r e s p o n s i v e n e s s in following vaccination, animals responded to neuroanti-
p r o t e c t e d a n d u n p r o t e c t e d m i c e , w e m e a s u r e d re- gens, and not to the control antigen O V A (Fig. 4A).
s p o n s e s o f i s o l a t e d L N C to a u t o a n t i g e n s ( M B P , M B P However, responses to MBP p91-102 wcre not en-
p90-102 and PLP p139-151), a control antigen (OVA),
a n d t h e p a n - T cell r e a g e n t a n t i - C D 3 . In e a c h e x p e r i - TABLE 2
ment, LN from three either vaccinated or untreated
Effect of T cell vaccination on relapse rate
m i c e w e r e c o l l e c t e d o n t h e d a y o f o n s e t o f t h e first
clinical e p i s o d e a n d p o o l e d for in v i t r o analysis. Experiment Vaccination Number of mice that relapsed
L N C p o l y c l o n a l r e s p o n s e s (to a n t i - C D 3 ) w e r e m e a s -
Control Vaccinated
u r e d by c y t o k i n e p r o d u c t i o n , this b e i n g a m o r e r e l i a b l e
1 RZ16 2/4 0/4
assay for r e s p o n s e s to i m m o b i l i z e d a n t i - C D 3 t h a n
2 A51 4/4 2/4
thymidine incorporation (Owens, unpublished observa- 3 SK4D2 4/4 0/4
tions). R e s p o n s e s f r o m v a c c i n a t e d a n d c o n t r o l m i c e 4 anti-OVA 2/4 2/4
w e r e e q u i v a l e n t ( T a b l e 3). I F N - y t i t e r s w e r e h i g h e r in
Mice were vaccinated with either anti-MBP (RZI6 and A51), anti-
L N C f r o m v a c c i n a t e d m i c e as c o m p a r e d to u n t r e a t e d
H2 k (SK4D2), or anti-OVA cell lines. Three weeks later mice were
c o n t r o l s ( T a b l e 3), b u t t h e s e d i f f e r e n c e s w e r e n o t sig- primed with RSCH in CFA and boosted after 7 days. Clinical signs
n i f i c a n t in an u n p a i r e d , t w o - t a i l e d t-test, P = 0.2074). of EAE were evaluated as described in Materials and Methods.

6. 90
hanced over background by vaccination, whereas re- Immunization of mice with OVA led to increased
sponses to PLP p139-151 and MBP were evident. background thymidine incorporation, compared to that
Vaccination with MBP-specific T cells did not specifi- seen in naive mice (Fig. 4B), and this was not affected
cally affect responsiveness to OVA that was induced by by vaccination with MBP-specific cell lines (Fig. 4A).
immunization with OVA and CFA. The SI of 2-3-fold By contrast, background thymidine incorporation by
in response to OVA in Fig. 4A was similar to the SI in LNC from unvaccinated mice that were injected with
the control experiment shown in Fig. 4B. RSCH and CFA was low, similar to that of naive mice
3.0
P 2.5
0
o
m 2.0
._o control
1.5
1¢
,.
m
Vaccinated
o
f- 1.0 _ ~ ,
el
e
E 0.5
0.0 -
12 14 16 18 20 22 24 26
Days after immunization with RSCWCFA
2.5-
B
.= 2.0
O
o
t
1.5 O~ control
o
v
A
vaccinated
o
r-
el
O
E
. . . . . .
• - • i ! I
00 1 20 24 28 32
Daya after immunization with RSCWCFA
1.5
C
O
o
g 1.0
A control
4, vaccinated
0.5
g
E
= . =
0.0 ; i i
8 12 6 20 24 28 32
Days after immunization with RSCH/CFA
Fig. 3. Protection of S J L / J mice against E A E by vaccination with RZI5, R Z I 6 and ASI cells. S J L / J female mice were vaccinated with
0.5-1 ),( 107 irradiated (2000 R) cells. Vaccinated and control mice were immunized for the induction of E A E by repeated injections of RSCH in
C F A and monitored for clinical signs as detailed in Materials and Methods. The graphs show the mean (n = 4) clinical scores plotted against
time for each group. Closed symbols represent mice vaccinated with (A) RZI5, (B) RZ16, (C) A51. Open symbols represent unvaccinated
controls. Bars represent SEM.

7. 91
TABLE 3 (Fig. 4). These mice had active EAE. The level of
Polyclonal lymphokine responses in LNC from vaccinated and con- background thymidine incorporation was therefore in-
trol mice versely proportional to the disease state of the mice,
and was increased by vaccination that reduced disease
Titer ( U / m l )
severity. The overall effect of vaccination, therefore,
Vaccination: RZ15, RZ16, A51 Untreated appears to have been able to restore a background
Immunization: RSCH OVA RSCH level of proliferation in peripheral lymphoid tissue.
Cytokine
IL-2 5.60± 1.8 5.80± 0.09 6.80± 1.9
IL-4 13.9 ± 8.9 12.2 + 2.9 10.5 -+ 9.2
IFN-7 530 ±240 600 + 100 185 _+45 Discussion
Polyclonal i m m u n e responses of L N C from vaccinated and control
mice were similar. Mice were vaccinated with the anti-MBP cell lines We have shown that SJL/J mice can be protected
RZ15, RZ16, or A51. T h r e e weeks later mice were primed with against the induction of EAE by vaccination with acti-
either RSCH or O V A in C F A and boosted after 7 days. LN were vated autoreactive T cells. Loss of encephalitogenicity
isolated at 14 days and L N C were cultured at 4 × l06 cells per ml in
did not influence the potential for vaccination. This is
1 ml wells to which anti-CD3 had been adsorbed. Supernatants were
collected after 24 h of culture and lymphokine contents were mea-
the first report of protection against RSCH-induced
sured by bioassay. The titers represent m e a n s of three vaccination EAE by T cell vaccination in SJL/J mice (Fig. 3 and
and four control e x p e r i m e n t s ± standard error of the m e a n (SEM). Table 1). We have studied LNC responses in mice that
A
Vaccination Immunization
None RSCH
..1111.111.'i .... )11111111111_ f
RZ15 RSCH ,, ,,,,,,,,,,,,,,,, ,,,,,' ...... ,,,,,,,,,~
RZ 15 OVA (.-.;.;.;.~.~.-.~.;.~.~...:././././----.--. ,
[] MBPp91-102
None RSCH [] MBP
• OVA
A51 RSCH [] PLP p139-15t
A51 OVA
0 6000 12000 18000 24000 30000
Thymidineincorporation(cpm)
B
Stimulus
OVA
• mmunizedwith OVA
Medium • Naive mice
Alone
0 gO~00 "18000 27000" 36000
Thymidine incorporation (cpm)
Fig. 4. Responses of LNC from vaccinated, control and naive mice. Vaccinated and control mice were primed and boosted with either RSCH or
O V A in CFA as described in Materials and Methods. LN were isolated on day 14 from vaccinated, control, or naive mice. For assessment of
proliferation, 8 x 105 I.NC were cultured in triplicate wells containing 200/~1 of complete medium with or without antigen. O V A and MBP were
added at 50 ,~g/ml; peptides were added at 15/~g/ml. Proliferation was assessed by [3H]thymidine incorporation after 3 days. Results are shown
as cpm ± SEM. (A) Proliferation of LNC from either control mice or mice vaccinated with RZ15, and ,~St. (B) Proliferation of ENC from naive
and OVA-primed mice.

8. 92
had been vaccinated 3 weeks prior to the induction of minants. In addition, anti-ergotypic activity may have
EAE. In mice with E A E neuroantigen-specific T cells contributed to the augmented responses seen in LNC
and non-specific recruits migrate to the CNS (Zeine from vaccinated mice. Vaccination, therefore, ap-
and Owens, 1992). It is probable that effector T cell peared to rescue peripheral T cell effector functions
migration to the CNS results in the hyporesponsive LN that are depleted in mice with EAE (Fig. 4). The
that we have now described in mice immunized with stimulatory effects of T cell vaccination on peripheral
RSCH, since mice primed with OVA in CFA showed immune responsiveness may reflect the rescue of nor-
increased background levels of LNC activity compared mal cellular activity.
to naive mice. We interpret this background response Protective mechanisms other than anti-ergotypic re-
to reflect ongoing T cell proliferation in vitro that was sponses may also play a role in vaccination. For in-
initiated in LN in vivo. We have demonstrated a clear stance, LNC from mice vaccinated with the RZ16 cell
difference between vaccinated and untreated mice in line did not show increased background levels of prolif-
this LNC proliferation. eration (not shown), and disease resistance in these
The T cells used for vaccination were MBP-specific mice may be due to anti-idiotypic mechanisms (see
(Fig. 1). MBP is unlikely to be the predominant en- Lohse and Cohen (1991) for review). Alternatively,
cephalitogen in this strain (Kennedy et al., 1990; anti-ergotypic recognition in our RZ16 vaccination may
Whitham et al, 1991). It is noteworthy that MBP-reac- have been incomplete, so that effective disease inhibi-
tive T cells protected against RSCH-induced EAE. T tion occurred in the absence of any changes in levels of
cell lines of irrelevant specificity (OVA, allo-MHC) did LNC activity.
not prevent disease induction. One explanation for Finally, there are reports of cross-induction of PLP
protection by MBP-specific vaccination against PLP- and MBP-reactive T cells in EAE (Perry et al., 1991;
mediated disease is that the effector mechanism of Cross et al., 1991), and our data may reflect interplay
protection is non-specific, but neuroantigen restricted. in E A E of these two reactivities. Recent evidence
Miller et al. (1992) have described antigen-specific suggests that, even within immunodominant determi-
induction of suppressor ceils whose effector mecha- nants, there is a hierarchy in the T cell contact residues
nism (TGF~ release) was non-specific. In our experi- composed of a single primary residue and a few sec-
ments, protective or suppressor ceils with specificity ondary residues (Evavold et al., 1991). Theoretically,
either for MBP or anti-MBP idiotypes might have any cross-reactivity between MBP and PLP epitopes
inhibited the activation or migration from LN of PLP- might allow protective immune responses that are in-
reactive T ceils by such a non-specific mechanism and duced by MBP-reactive T cells to become directed
this is currently being studied in our laboratory. The against PLP-reactive T cells.
relative lack of effect of immunization with cell lines of In summary, we have examined peripheral immune
irrelevant specificity probably reflects the absence of responses in mice that were protected from EAE by T
those antigens from our EAE experiments, and, there- cell vaccination. In contrast to the CNS where the
fore, no induction of suppression. However, the opera- extent of perivascular infiltration was reduced, the
tion of such a non-specific effector mechanism would level of cellular activation within the LN was either
not explain the LN hyperresponsiveness that we have unchanged or increased. This implies that the mecha-
observed. nism of protection induced by preimmunization with T
An alternative interpretation is that protection was cells is an active process involving multiple cellular
due to cellular mechanisms, that involved the induction immune responses, and that responses that are associ-
of responses against activated T ceils. Such responses ated with effector function (T cell proliferation) may
have been described in rats tested 6 days after vaccina- also reflect the operation of regulatory processes.
tion and have been termed anti-ergotypic (Lohse et al.,
1989). Anti-ergotypic T ceils could either inhibit the
emigration of effector T cells from LN, or might induce
a general T cell hyperresponsiveness through T : T in- Acknowledgements
teractions. The former would lead to reduced numbers
of CNS infiltrates, and both could produce increased We thank Dr. Jack Antel and Dr. Michael Ratcliffe
LNC responses to encephalitogens. Our evidence is for comments on the manuscript and for helpful dis-
consistent with both of these possibilities. Vaccination, cussions, and Gary Spector for his contribution to the
even with unprimed nylon wool purified T cells, re- early stages of this research. We thank Drs. George
duced the degree of CNS perivascular infiltration (Ta- Hashim and Marianna Newkirk for provision of pep-
ble 1). This may well represent anti-ergotypic reactivity tides, and Drs. Timothy Mosmann and Anne Kelso for
even though the T cells were not deliberately activated. provision of recombinant lymphokines. This work was
These populations may have included non-resting T funded by The Multiple Sclerosis Society of Canada.
cells with the potential of expressing 'ergotypic' deter- T.O. is an MRC Canada Scholar. R.Z. was supported

9. 93
by FCAR (Quebec) and The Multiple Sclerosis Society Lider, O., Shinitzky, M. and Cohen, I.R. (1986) Vaccination against
of Canada. experimental autoimmune diseases using T lymphocytes treated
with hydrostatic pressure. Proc. N.Y. Acad. Sci. 275, 267-273.
Lohse, A. and Cohen, I.R. (1991) Mechanisms of resistance to
autoimmune disease induced by T-cell vaccination. Autoimmu-
References nity 9, 119-121.
Lohse, A.W., Mor, F., Karin, N. and Cohen, I.R. (1989) Control of
Acha-Orbea, H., Mitchell, D.J., Timmermann, L., Wraith, D.C., experimental autoimmune encephalomyelitis by T cells respond-
Tausch, G.S., Waldor, M.K., Zamvil, S.S., McDevitt, H.O. and ing to activated T cells. Science 244, 820-822.
Steinman L. (1988) Limited heterogeneity of T cell receptors Miller, A., Lider, O., Roberts, A.B., Sporn, M.B. and Weiner, H.L.
from lymphocytes mediating autoimmune encephalomyelitis al- (1992) Suppressor T cells generated by oral tolerization to myelin
lows specific immune intervention. Cell 54, 263-273. basic protein suppresses both in vitro and in vivo immune re-
Ben-Nun, A., Wekerle, H. and Cohen, I.R. (1981) Vaccination sponses by the release of transforming growth factor beta after
against autoimmune encephalomyelitis with T lymphocyte line antigen-specific triggering. Proc. Natl. Acad. Sci. USA 89, 421-
cells reactive against myelin basic protein. Nature 292, 60-61. 425.
Birkland, M.L., Johnson, P., Trowbridge, I.S. and Pure, E. (1989) Mosmann, T.R. (1983) Rapid colorimetric assay for cellular growth
Changes in CD45 isoform expression accompany antigen-induced and survival: application to proliferation and cytotoxicity assays.
murine T-cell activation. Proc. Natl. Acad. Sci. USA 86, 6734- J. Immunol. Methods 65, 55-63.
6738. Offner, O., Jones, R., Celnik, B. and Vandenbark, A.A. (1989)
Cross, A.H., Tuohy, V.K. and Raine, C.S. (1991) Switching of anti- Lymphocyte vaccination against experimental autoimmune en-
gen responsiveness during relapsing autoimmune demyelination cephalomyelitis: evaluation of vaccination protocols. J. Neuroim-
(Abstract). Ann. Neurol. 30, 269. munol. 21, 13-22.
Evavold, B.D., Williams, S.G., Chen, J.S. and Allen, P.M. (1991) T Owens, T. and Miller, J.F.A.P. (1987) Interactions in vivo between
cell inducing determinants contain a hierarchy of residues con- hapten-specific suppressor T cells and an in vitro cultured helper
tacting the T cell receptor. Sere. Immunol. 3, 225-229. T cell line. J. Immunol. 138, 1687-1692.
Firestein, G.S., Roeder, W.D., Laxer, J.A., Townsend, K.S., Weaver, Padula, S.J., Lingnheld, E.G., Stabach, P.R., Chou, C.J., Kono, D.H.
C.T., Horn, J.T., Linton, J., Torbett, B.E. and Glasebrook, A.L. and Clark, R.B. (1991) Identification of encephalitogenic V/34-
(1989) A new murine CD4 ÷ T cell subset with an unrestricted bearing T cells in S J L / J mice: Further evidence for the V region
cytokine profile. J. lmmunol. 143, 518-525. disease hypothesis? J. Immunol. 146, 879-883.
Hashim, G.A., Day, E.D., Fredane, L., Intintola, P., Carvalho, E. Perry, L.L., Barzaga-Gilbert, E. and Trotter, J.L. (1991) T cell
Biological activity of region 65-102 of the myelin basic protein. J. sensitization to proteolipid protein in myelin basic protein-in-
Neurosci. Res. 1986; 16, 467-478. duced relapsing experimental allergic encephalomyelitis. J. Neu-
Howell, M.D., Winters, S.T., Olee, T., Powell, H.C., Carlo, D.J. and roimmunol. 33, 7-15.
Brostoff, S.W. (1989) Vaccination against experimental allergic Raine, C.S. (1985) Experimental allergic encephalomyelitis. In: J.C.
encephalomyelitis with T cell receptor peptides. Science 246, Koetsier (Ed.), Handbook of Clinical Neurology Vol. 3(47). Else-
668-670. vier, Amsterdam, pp. 429-466.
Hu-Li, J., Ohara, J., Watson, C., Tsang, W. and Paul, W. (1989) Sakai, K., Sinha, A.A., Mitchell, D.J., Zamvil, S.S., Rothbard, J.B.,
Derivation of a T-cell line that is highly responsive to IL-4 and McDevitt, H.O., Steinman, L. and Acha-Orbea, H. (1988) In-
IL-2 (CT.4 R) and an IL-2 hyporesponsive mutant of that line volvement of distinct murine T-cell receptors in autoimmune
(CT.4S). J. Immunol. 142, 800-807. encephalitogenic response to nested epitopes of myelin basic
Janeway, C.A. (1989) Immunotherapy by peptides. 1989; Nature 341, protein. Proc. Natl. Acad. Sci. USA 85, 8608-8612.
482-483. Satch, J., Sakai, K., Endoh, M., Koike, F., Kunishita, T., Namikawa,
Jensen, M.A., Arnason, B.G.W., Toscas, A. and Noronha, A. (1992) T., Yamamura, I. and Tabira, T. (1987) Experimental allergic
Preferential increase of IL-2R ~- CD4 + T cells and CD45 RB- encephalomyelitis mediated by murine encephalitogenic T cell
CD4* T cells in the central nervous system in experimental lines specific for myelin proteolipid apoprotein. J. Immunol. 138,
allergic encephalomyelitis. J. Neuroimmunol. 38, 255-262. 179-184.
Kelso, A. (1990) Frequency analysis of lymphokine-secreting CD4 ÷ Schluesener, H.J. and Wekerle, H. (1985) Autoaggressive T lympho-
and CD8 + T cells activated in a graft-versus-host reaction. J. cyte lines recognizing the encephalitogenic region of myelin basic
lmmunol. 145, 2167-2176. protein: in vitro selection from unprimed rat T lymphocyte popu-
Kennedy, M.K., Tan, L.-J., Dal Canto, M.C. and Miller, S.D. (1990) lations. J. Immunol. 135, 3128-3133.
Regulation of effector stages of experimental autoimmune en- Sun, D., Quin, Y., Chluba, J., Epplen, J.T. and Wekerle, H. (1988)
cephalomyelitis via neuroantigen-specific tolerance induction. J. Suppression of experimentally induced autoimmune en-
lmmunol. 145, 117-126. cephalomyelitis by cytolytic T - T cell interactions. Nature 332,
Kono, D.H., Urban, J.L., Hovrath, S.J., Ando, D.G., Saavedra, R.A. 842-844.
and Hood, L. (1988) Two minor determinants of myelin basic Trowbridge, I.S., Lesley, J., Schulte, R., Hyman, R. and Trotter, J.
protein induce experimental allergic encephalomyelitis in S J L / J (1982) Biochemical characterization and cellular distribution of a
mice. J. Exp. Med. 168, 213-227. polymorphic, murine cell-surface glycoprotein expressed on lym-
Ledbetter, J.A. and Herzenberg, L.A. (1979) Xenogeneic monoclonal phoid cells. Immunogenetics 15, 229-312.
antibodies to mouse lymphoid differentiation antigens. Immunol. Vandenbark, A., Hashim, G. and Offner, H. (1989) Immunization
Rev.. 47, 63-90. with a synthetic T-cell receptor V-region peptide protects against
Leo, O., Fop, M., Sachs, D.H., Sanelson, L.E. and Bluestone, J.A. experimental autoimmune encephalomyelitis. Nature 1989; 341,
(1987) Identification of a monoclonal antibody specific for a 541-544.
murine T3 polypeptide. Proc. Natl. Acad. Sci. USA 84, 1374-1378. Vandenbark, A.A., Offner, H., Reshef, T., Fritz, R., Chou, C.-H.J.
Lider, O., Reshef, T., Beraud, E., Ben-Nun, A. and Cohen, I.R. and Cohen, I.R. (1985) Specificity of T lymphocyte lines for
(1988) Anti-idiotypic network induced by T cell vaccination against peptides of myelin basic protein. J. Immunol. 135, 229-234.
experimental autoimmune encephalomyelitis. Science 239, 181- Weinberg, A.D., Whitham, R., Swain, S.L., Morrison, W.J., Wyrick,
183. G., Hoy, C., Vandenbark, A.A. and Offner, H. (1992)Transform-

10. 94
ing growth factor-/3 enhances the in vivo effector function and peptide of proteolipid protein and transfer relapsing demyelinat-
memory phenotype of antigen-specific T helper cells in experi- ing experimental autoimmune encephalomyelitis. J. lmmunol.
mental autoimmune encephalomyelitis. J. Immunol. 148, 2109- 146, 101-107.
2117. Zeine, R. and Owens, T. (1992) Direct demonstration of the infiltra-
Whitham, R.H., Bourdette, D.N., Hashim, G.A., Herndon, R.M., Ilg, tion of murine CNS by Pgp-l/CD44 hi~ CD45RB I°~ CD4 + T
R.C., Vandenbark, A.A. and Offner, H. (1991) Lymphocytes from cells that induce experimental allergic encephalomyelitis. J. Neu-
S J L / J mice immunized with spinal cord respond selectively to a roimmunol. 40, 57-70.