DOI: 10.1093/brain/awg269
Advanced Access publication August 22, 2003
Brain (2003), 126, 2638±2647
Blocking effects of serum reactive antibodies
induced by glatiramer acetate treatment in multiple
sclerosis
Hassan H. Salama,1,3 Jian Hong,1 Ying C. Q. Zang,1 Azza El-Mongui3 and Jingwu Zhang1,2
1Multiple
Sclerosis Research Unit, Department of
Neurology, Baylor-Methodist Multiple Sclerosis Center,
2Department of Immunology, Baylor College of Medicine,
Houston, Texas, USA, and 3Department of Neurology,
Faculty of Medicine, Mansoura University, Mansoura,
Egypt
Summary
Glatiramer acetate (GA) is a treatment option for multiple sclerosis. Although its mechanism of action
remains unclear, evidence has emerged supporting the
role of GA as an immunomodulatory drug that regulates T-cell function. It has been demonstrated that
long-term GA treatment induces a serum antibody
response; however, the functional properties of these
`reactive antibodies' are unknown. It has been speculated that GA-induced antibodies may have a blocking
effect that can inhibit the immunologic activity of GA.
This study was conducted to determine whether serum
antibodies induced by GA treatment can block the
in vitro immunoregulatory effects of GA on T-cell proliferation and cytokine production. Forty-two patients
with relapsing-remitting multiple sclerosis who were
Correspondence to: Hassan H. Salama MD, Department of
Neurology, Mansoura University, Mansoura, Egypt
E-mail: [email protected]
treated with GA for 1±5 years were examined for GA
antibody titres using enzyme-linked immunoabsorbent
assay (ELISA). Thirty-three percent of patients developed high antibody titres [antibody binding index
(ABI) = 16±64] and 14% had low antibody titres
(ABI = 4) after 1 year on treatment. Results showed
that puri®ed GA antibodies blocked the stimulatory
effects of GA on GA-speci®c T-cell lines of Th0 cytokine
pro®le. The increase in interleukin-10 (IL-10) and IL-4
levels and the decrease in IL-12 and tumour necrosis
factor-a levels, normally seen with GA stimulation,
were reversed in the presence of GA antibodies. The
study has important implications in our understanding
of the potential role of high-titre GA antibodies in the
treatment of multiple sclerosis.
Keywords: glatiramer acetate; Copaxone; cytokines; serum reactive antibodies; multiple sclerosis
Abbreviations: ABI = antibody binding index; CPM = counts per minute; EDSS = Expanded Disability Status Scale;
ELISA = enzyme-linked immunoabsorbent assay; GA = glatiramer acetate; HEL = hen egg lysozyme; Ig = immunoglobulin;
IL = interleukin; OD = optical density; PBMCs = peripheral blood mononuclear cells; TNF-a = tumour necrosis factor-a
Introduction
The development of immunomodulatory agents, including
glatiramer acetate (GA) and interferon b, represents a major
advance in the treatment of multiple sclerosis. As with
interferon b, it has been shown that GA favourably alters the
natural history of the disease by reducing relapse rate and
decreasing brain in¯ammation as measured by MRI (Johnson
et al., 1995; Comi et al., 2001a, b). Although the mechanism
of action of GA remains unclear, there is evidence that the
clinical effects of the agent are related to its immunomodulatory properties. In particular, GA has been found to increase
the ratio of anti-in¯ammatory (Th2) to pro-in¯ammatory
(Th1) cytokines (Aharoni et al., 1997, 1998, 2000, 2002;
Chen et al., 2001, 2002; Farina et al., 2001; Hussien et al.,
2001; Neuhaus et al., 2001; Maron et al., 2002). Other
mechanisms that may account for the treatment effects of GA
include: (i) interaction/competition with major histocompatibility complex class II molecules (speci®cally DR2 molecules), thereby interfering with presentation of self-myelin
antigens to autoreactive T-cells (Racke et al., 1992;
Teitelbaum et al., 1992, 1996; Fridkis-Hareli et al., 1994;
Ben-Nun et al., 1996; Fridkis-Hareli and Stominger, 1998);
(ii) induction of anergy that renders T-cells unresponsive to
myelin antigens (Gran et al., 2000); (iii) upregulation of
CD8+ regulatory T-cells (Karandikar et al., 2002); (iv)
Brain Vol. 126 No. 12 ã Guarantors of Brain 2003; all rights reserved
Blocking effects of GA antibodies in multiple sclerosis
modi®cation of dendritic cell costimulation processes
(Hussien et al., 2001) or activity as a weak/partial T-cell
receptor agonist to activate naõÈve T-cells (Wiesemann et al.,
2001); and (v) impairment of activated T-cells to interact with
microglia that produce cytokines in the CNS (Chabot et al.,
2002).
Standard treatment with daily subcutaneous injections of
GA is known to induce the development of `reactive
antibodies' by the host immune system (Brenner et al.,
2001; Farina et al., 2002). It has been speculated that GA
antibodies may interfere with the regulatory properties of GA
by blocking the active groups on the molecule or by forming
complexes that are rapidly cleared by the immune system.
The issue is particularly relevant in the clinical management
of multiple sclerosis because there is a large body of evidence
that neutralizing antibodies develop to varying degrees to
interferon b formulations; the incidence of neutralizing
antibodies has been reported in 2±6% of patients receiving
intramuscular injections of interferon b-1a (Herndon et al.,
1999; Jacobs et al., 2000; Clanet et al., 2002), 12±25% of
patients receiving subcutaneous injections of interferon b-1a
(Antonelli et al., 1998; PRISMS Study Group, 1998; Panitch
et al., 2002), and 28±45% of patients receiving interferon b1b (IFNB Multiple Sclerosis Study Group, 1993, 1996;
European Study Group on Interferon Beta-1b in Secondary
Progressive Multiple Sclerosis, 1998). Data from two
Phase III trials of interferon b in multiple sclerosis patients
have shown that neutralizing antibodies are associated with
diminished ef®cacy starting after 18±24 months of treatment
(IFNB Multiple Sclerosis Study Group, 1996; PRISMS Study
Group, 2001).
To date, no study has been reported that addresses the
biological effects or therapeutic implications of serum
antibodies induced by GA treatment in patients with multiple
sclerosis. This paper reports the results of a study conducted
to determine whether GA antibodies in¯uence the stimulatory
and regulatory effects of GA on T-cells in functional assays.
Because both the activation of Th2 cells and the regulation of
cytokine production by T-cells are considered to be important
regulatory mechanisms of GA, the functional assays were
designed to analyse the effects of GA antibodies on these
parameters. In addition, although the number of patients in
the study is limited and insuf®cient for meaningful statistical
analysis, preliminary attempts were made to evaluate whether
the occurrence and titres of GA antibodies are associated with
a decrease in the therapeutic effects of GA.
Methods
Study design and patients
Multiple sclerosis patients who had started GA treatment
(20 mg subcutaneously once daily) at the Baylor Methodist
Multiple Sclerosis Center between 1996 and 2000, and who
had been tested for serum reactive antibodies to GA before
the treatment (as pre-treatment specimens) and periodically
2639
after the treatment (as post-treatment specimens), were
identi®ed for inclusion in the study by investigators via
retrospective chart review. Serum specimens from all patients
were stored at ±80°C within 2 h of blood draw at various time
points. Serum specimens were included in the study if
patients met the following criteria: (i) a diagnosis of clinically
de®nite multiple sclerosis and evidence of gadoliniumenhanced T1-weighted lesions of the brain and the spinal
cord on MRI scans; (ii) a relapsing-remitting course of
multiple sclerosis at the start of GA treatment; and (iii)
complete clinical records documenting concurrent use of
medications, Expand Disability Status Scale (EDSS) scores,
and annual relapse rate 1 year before GA treatment and
during the course of uninterrupted treatment over a period of
1±5 years. Serum specimens were excluded from the study if
patients received treatment with an interferon b product or
prolonged immunosuppressive therapy during the study.
Patients were treated for acute exacerbations with methylprednisolone (1 g administered intravenously for 3 days with
an oral prednisone taper) when episodes occurred.
Table 1 shows the demographic and clinical characteristics
of 42 patients who met inclusion/exclusion criteria and were
included in the study. Sixty-four percent of patients were
female, the mean age was 44 6 11.4 years, and patients had
multiple sclerosis for a mean of 11.3 6 7.1 years. Patients
were treated with GA for a mean of 3.3 6 1.6 years, ranging
from 1 to 5 years.
Reagents
Culture medium used in the study was RPMI 1640 supplemented with 10% heat-inactivated foetal calf serum and
L-glutamine, sodium pyruvate, non-essential amino acids and
10 mM HEPES buffer (Hyclone, Logan, UT, USA). GA
(Copaxone) used in this study was obtained from Teva
Pharmaceuticals, Inc. (St Louis, MO, USA). Control antigens,
hen egg lysozyme (HEL) and phytohemagglutinin, were
purchased from Sigma (St Louis, MO, USA).
Antibody reactivity by enzyme-linked
immunoabsorbent assay (ELISA)
Stored serum specimens (at ±80°C) were examined for
antibody reactivity to GA. Brie¯y, microtitre plates were
coated overnight at 4°C with either GA or HEL at a
concentration of 1 mg/well. Wells were then blocked at
37°C for 2 h with phosphate-buffered saline (PBS) containing
2% bovine serum albumin (BSA) (Sigma) and were subsequently washed four times with 0.02% Tween 20 in a 0.9%
NaCl solution. Each diluted serum sample and its control
were added to the adjacent wells and incubated for 2 h. Plates
were washed four times and incubated for 30 min with a goat
anti-human immunoglobulin (Ig) (IgG) IgG/IgM antibody
conjugated with horseradish peroxidase at a dilution of
1:1500 (Sigma). A solution of 0.0125% tetramethylbenzi-
2640
H. H. Salama et al.
dine/0.008% H2O2 in citrate buffer (pH 5.0) was used as a
substrate, and colour development was stopped using 2N
H2SO4. Optical densities (ODs) were measured using an
ELISA reader (Bio-Rad, Hercules, CA, USA). Antibody titres
were expressed in the form of an antibody binding index
(ABI) as follows: mean OD of GA-bound wells/mean OD of
control wells coated with HEL. In all experiments, the mean
OD obtained from HEL-coated control wells was equivalent
to the background OD in GA-coated wells (control sera
obtained from healthy controls that had no antibody reactivity
to GA). A speci®c response is de®ned as an ABI of >4.
Inhibition assay
Peripheral blood mononuclear cells (PBMCs) derived from
healthy individuals were cultured at 100 000 cells/well with
GA at a predetermined concentration of 100 mg/ml or
phytohemagglutinin (1 mg/ml) as a control. Cultures were
set up in duplicate in the presence and absence of sera (1:100
dilution) pre-dialyzed against culture media. Cultures were
retained for 3 days and were pulsed with [3H]-thymidine
(Nycomed Amersham, Arlington Heights, IL, USA) at 1 mCi
per well during the ®nal 16 h of culture. Cells were then
harvested using an automated cell harvester and [3H]thymidine incorporation was measured in a b-plate counter.
Percent inhibition is calculated as: {1 ± [mean counts per
minute (CPM) of PBMC culture in the presence of GA serum/
mean CPM of PBMC culture in the absence of GA
serum] 3 100%}.
Immunoblot analysis
GA was electrophoresed at the indicated amount using 10%
SDS±PAGE. After blotting, nitrocellulose membranes were
cut into strips and then blocked with 5% low-fat milk powder
in Tris-buffered saline containing 0.1% Tween 20 (milkTable 1 Demographic and clinical characteristics of
patients at enrolment
Number of patients at enrolment
Sex (% female/male)
Age:
Mean in years 6 SD
Range
Disease duration:
Mean in years 6 SD
Range
EDSS:
Mean score 6 SD
Range
Relapse rate:
Mean 6 SD
Range
Treatment duration:
Mean in years 6 SD
Range
42
64/36
44.0 6 11.4
18±67
11.3 6 7.1
2±38
3.7 6 1.3
2.0±7.0
TBST). The strips were then incubated with pre-treatment and
post-treatment serum preparations at a dilution of 1:100 in miniincubation trays for 1 h at room temperature. A goat anti-human
IgG and IgM (H+L chains) coupled to horseradish peroxidase
was used as a secondary antibody (100 ng/ml in 2% milk-TBST)
and incubated with pre-washed strips for 45 min, followed by
enhanced chemiluminescent visualization using a reagent kit
purchased from Amersham (Piscataway, NJ, USA).
Ammonium sulphate precipitation of serum
immunoglobulin
Sera were brought to 100% saturation with ammonia sulphate
and were stirred continuously overnight at 4°C. The resulting
solution was centrifuged at 120 000 g for 30 min, and the
precipitates were dissolved in 1 ml of 100 mM sodium
phosphate buffer (pH 7.3) and dialyzed subsequently against
RPMI 1640 medium.
Generation of GA-speci®c T-cell lines
PBMCs were plated at 200 000 cells/well (for a total of
96 wells) in the presence of GA (100 mg/ml). Seven days
later, all cultures were restimulated with GA in the presence
of irradiated autologous PBMCs. After another week, each
well was split into four aliquots (~104 cells per aliquot) and
cultured in duplicate with 105 irradiated autologous PBMCs
in the presence and absence of GA. Cultures were retained for
3 days and were pulsed with [3H]-thymidine at 1 mCi per well
during the ®nal 16 h of culture. Cells were then harvested
using an automated cell harvester, and [3H]-thymidine
incorporation was measured. A well/culture was de®ned as
speci®c for GA when the CPM was >1000 and exceeded the
reference CPM (in the absence of GA) by at least three times.
Speci®c T-cell lines were restimulated and expanded by
culturing 20 000 cells/well with 100 000 cells/well of
irradiated autologous PBMCs in the presence of GA
(100 mg/ml).
T-cell proliferation assay
GA-speci®c T-cell lines were derived from healthy individuals who were not exposed to GA and were cultured at
20 000 cells/well in the presence of irradiated autologous
PBMC at 100 000 cells/well and GA at a predetermined
concentration of 100 mg/ml. Cultures were set up in triplicate
in the presence and absence of different dilutions of puri®ed
GA antibody. Cultures were retained for 3 days and [3H]thymidine incorporation was measured as described above.
The results were expressed in CPM.
1.7 6 0.7
0±4
3.3 6 1.6
1±5
Cytokine measurements by ELISA
Culture supernatants were collected from the GA T-cell line
culture 48 h after initiation of the culture and were
Blocking effects of GA antibodies in multiple sclerosis
2641
Table 2 GA antibody titres and clinical parameters
Treatment
duration
(years)
ABI
n
1
<4
16±64
<4
16±64
<4
16±64
<4
16±64
<4
16±64
9
0
4
2
4
1
5
3
6
8
2
3
4
5
EDSS
Pre-treatment
Post-treatment
Relapse rate
Pre-treatment
Treatment
Post-treatment
2.9
2.1
1.9
1
0.6
4.25
4.3
2.8
7
3.5
4.3
3.8
3.9
4
4.8
2
7.5
3.1
4.8
2.8
4.1
1.5
2
1.8
1
1.4
2
1.7
1.6
0.75
1
0.8
2
0.2
2.7
0.3
1.8
0.5
1.5
0.3
2
0.2
1.7
0.17
0.9
Low antibody titre is de®ned as ABI = 4. High antibody titre is de®ned as ABI = 16±64. Patients with an
ABI = 1 (considered as insigni®cant GA antibody titre) and with low GA antibody titre (ABI = 4) are
combined as one group (ABI < 4).
measured for concentrations of cytokines using ELISA kits
according to the manufacturer's instructions (PharMingen,
San Diego, CA, USA). Microtitre plates were coated
overnight at 4°C with mouse monoclonal antibodies (capturing antibody). Wells were then blocked at 37°C for 2 h with
PBS containing 2% BSA (Sigma) and were subsequently
washed four times with 0.02% Tween 20 in a 0.9% NaCl
solution. Samples were added and incubated for 2 h with a
biotinylated detecting antibody (0.25 mg/ml of each monoclonal antibody, respectively) in 2% BSA/PBS/Tween 20.
The remainder of the procedure for colour development is as
described previously. The detection limits for all four
cytokines were <15 pg/ml in all assays.
Statistical analysis
The percentage of patients who developed antibodies during
treatment was calculated to determine the incidence of
antibody-positive patients. Using the ABI, low titres of GA
antibodies were de®ned as 4, whereas high titres were de®ned
as 16±64. Student's t-tests were used to analyse normally
distributed variables. All reported P values are based on twotailed statistical tests, with a signi®cance level of 0.05.
Results
Incidence and binding properties of GA
antibodies
All pre-treatment serum specimens derived from 42 patients
included in this study and an additional panel of 12 serum
specimens obtained from untreated patients with multiple
sclerosis and healthy individuals tested negative for GAspeci®c antibodies (ABI = 1). Post-treatment, 48% of patients
tested positive for GA-speci®c antibodies post-treatment,
with 33% developing high antibody titres (ABI = 16±64) and
14% developing low titres (ABI = 4); 52% of patients tested
negative for antibodies post-treatment (ABI = 1) (Table 2). A
trend was noted that patients with higher ABIs (16±64)
showed more deterioration in EDSS scores and more relapses
from pre-treatment to post-treatment compared with patients
with an ABI <4 and the same length of treatment (Table 2).
However, in this retrospective study, the number of patients
was not suf®cient to detect any statistically signi®cant effects.
Fourteen serum specimens of high GA antibody titres were
pre-screened for an inhibitory effect on the proliferation of
PBMCs in response to GA. Six specimens exhibited a
signi®cant blocking effect (>60%), at a serum dilution of
1:100, on the ability of GA to induce the T-cell proliferation
in PBMCs, while others had either no effect or an effect that
was not speci®c for GA.
Six serum specimens that had an inhibitory effect on GAinduced T-cell proliferation were selected for detailed
characterization. These paired specimens consisted of pretreatment sera without detectable GA antibodies and posttreatment sera with high titres of GA antibodies (ABI = 64)
from the same patients. They are designated as MS-1 through
MS-6. As shown in Fig. 1, the selected post-treatment serum
antibodies reacted speci®cally with GA but not with the
control antigens, interferon b or HEL. In contrast, paired
serum specimens derived from the same patients before
treatment had no detectable reactivity to GA. The speci®c
reactivity of serum antibodies to GA was con®rmed by
immunoblot analysis (Fig. 2).
Effects of GA antibodies on GA-speci®c T-cell
lines
We then addressed whether serum GA antibodies would
interfere with the effect of GA on GA-speci®c T-cell lines
and on their cytokine production. We therefore generated a
small panel of well-characterized T-cell lines that reacted
speci®cally to GA and analysed the effect of GA antibodies
on the selected T-cell lines. These T-cell lines were generated
2642
H. H. Salama et al.
Fig. 1 Six paired pre-treatment (A) and post-treatment (B) serum preparations selected for high GA antibody titres (ABI = 64) were tested
at serial dilutions for reactivity to GA using ELISA. HEL and interferon-b were used as control antigens. Background OD (wells coated
with no antigens) was 0.27 6 0.01.
from a healthy individual and had the following characteristics: (i) speci®cally recognized GA but not HEL (used as a
control antigen); (ii) exhibited high stimulation indices (i.e.
speci®c proliferation to GA/proliferation to a control antigen)
of >20; and (iii) had a cytokine pro®le consistent with the Th0
phenotype. In addition, to avoid potential interference caused
by unknown serum factors in the cell functional assays,
immunoglobulin fractions were puri®ed by ammonia sulphate
precipitation and dialyzed thoroughly against culture media
to remove other serum factors. The precipitated antibody
fractions were tested to con®rm the speci®c reactivity to GA
by ELISA prior to use in the functional assays. As shown in
Fig. 3, antibodies puri®ed from post-treatment sera signi®cantly inhibited the proliferation of three representative
T-cell lines speci®c for GA at antibody dilutions of 1:5 to
1:10 (roughly equivalent to 1:50 to 1:100 serum dilutions). In
contrast, paired antibodies puri®ed from pre-treatment sera of
the same patients did not demonstrate a similar effect.
Furthermore, GA antibodies also were found to revert the
in vitro effect of GA-induced cytokine production of GAspeci®c T-cell lines. As shown in Fig. 4, GA stimulation
of the T-cell lines resulted in up-regulation of interleukin-10
(IL-10) and IL-4, and down-regulation of tumour necrosis
factor-a (TNF-a) and IL-12, which are the two prominent
Blocking effects of GA antibodies in multiple sclerosis
2643
Fig. 2 Immunoblot analysis of pre-treatment and post-treatment serum specimens derived from six
selected multiple sclerosis patients. GA was loaded at the indicated concentrations and analysed by
SDS±PAGE electrophoresis (left panel). Paired serum specimens derived from six multiple sclerosis
patients before (±) and after (+) GA treatment were examined by immunoblot analysis (right panel).
pro-in¯ammatory cytokines associated with multiple
sclerosis. These regulatory effects of GA were substantially
reverted by the addition of GA antibodies.
Correlation of high titres of GA antibody with
serum cytokine pro®le
To further delineate the in vivo effect of GA antibodies on
serum cytokine production, we examined the change in serum
cytokine pro®le before and after GA treatment in patients
with high GA antibody titres as compared with that in patients
with low titres. To this end, serum cytokine pro®les of two
groups were compared: 12 multiple sclerosis patients who
were on GA treatment for 3±5 years and exhibited low titres
of GA antibodies (ABI = 4) and 13 multiple sclerosis patients
who were treated for the same length of time and who had
high GA antibody titres (ABI = 16±64). As shown in Fig. 5, a
signi®cant increase in the serum concentrations of IL-10
and a reduction in TNF-a and IL-12 were observed in posttreatment specimens in the low GA antibody titre group. This
effect also has been reported by other investigators (Li et al.,
1998; Hussien et al., 2001). In contrast, a reversed cytokine
pro®le was seen in the high antibody titre group as evident by
high concentrations of TNF-a and IL-12 and a low concentration of IL-10 in post-treatment specimens versus pretreatment specimens (Fig. 5). These results suggest that the
in vivo effect of GA is reverted by GA antibodies in patients
who developed high titres of GA antibodies.
Discussion
GA has been available in the USA as a therapeutic option for
the treatment of relapsing-remitting multiple sclerosis since
1995. Administration of GA may induce the development of
substantial `reactive antibody' responses in multiple sclerosis
patients within 1 year of treatment initiation (Brenner et al.,
2001; Farina et al., 2002). However, there is no clear
evidence as to whether the development of GA-reactive
antibodies during prolonged treatment is associated with
reduced therapeutic ef®cacy. The ®ndings described in the
present study provide, for the ®rst time, important experimental evidence for considering the potential blocking effect
of serum antibodies induced by treatment with GA. GAspeci®c antibodies were shown to reverse GA-stimulated
proliferation of GA-speci®c T-cells, up-regulation of the antiin¯ammatory cytokines IL-10 and IL-4, and down-regulation
of the pro-in¯ammatory cytokines TNF-a and IL-12 in
T-cells.
The results of this study raise a number of important issues.
First, what is the potential mechanism of action that underlies
the blocking/neutralizing effect of GA antibodies on the
regulatory properties of the drug? Although the present study
involved only a T-cell functional analysis and was not
designed to investigate molecular interactions or to address
mechanistic issues, several possibilities exist that may
explain the ®ndings. For example, GA antibodies may bind
to and block certain amino acid groups on the polymer, which
act as functional epitopes capable of activating GA-speci®c
T-cells (e.g. Th2 or Th0 subsets). This possible action may
explain the loss of the stimulatory effect of GA on T-cells and
the diminution of its intrinsic regulatory properties on T-cellmediated cytokine production when GA antibodies were
added into the system. Alternatively, antibody reactivity to
GA may interfere with the binding of GA to the molecules of
the major histocompatibility complex, a proposed mechanism
by which GA may competitively impair the presentation of
self-myelin antigens/peptides to T-cells (Racke et al., 1992;
Teitelbaum et al., 1992; Fridkis-Hareli et al., 1994; FridkisHareli and Strominger, 1998). It is also possible that GA
antibodies may facilitate rapid internalization and intracellular degradation of GA through Fc and Fc receptors
2644
H. H. Salama et al.
Fig. 3 Serum preparations were puri®ed by ammonia sulphate precipitation and tested at the indicated dilutions for inhibition of
proliferation of a representative GA-speci®c T-cell line. Cells of three independent GA-speci®c T-cell lines (GA-1, GA-2 and GA-3) were
cultured at 20 000 cells/well together with irradiated autologous PBMC (100 000 cells/well) in the presence (solid horizontal lines) and
absence (dotted horizontal lines) of GA at a concentration of 100 mg/ml. The same antibodies were tested with a representative T-cell line
speci®c for HEL-1 using the same setting as described for the GA-speci®c T-cell lines.
commonly expressed on B-cells, macrophages and other
blood cells, resulting in substantially reduced bioavailability
of GA in the system. Further investigations are needed to
delineate the mechanism whereby GA antibodies block the
regulatory effects of GA on T-cells.
The ®ndings also raise an important clinical issue as to
whether the occurrence and high titres of GA antibodies may
impair the treatment effect of GA in multiple sclerosis
patients. A trend demonstrating a potential correlation
between high GA antibody titres and clinical deterioration
in EDSS scores and relapse rate was noted in the present
study. However, this retrospective study was not originally
designed to address the clinical relevance of elevated titres of
GA antibodies and, hence, the number of patients examined
was not suf®cient to detect a statistically signi®cant effect. It
should be cautioned that development of antibody responses
to GA is dynamic in multiple sclerosis patients during
treatment, and that GA antibodies have been reported to have
Blocking effects of GA antibodies in multiple sclerosis
2645
Fig. 4 Supernatants were collected from cultures described for Fig. 3 at 48 h after the onset of the culture
and were analysed for the concentrations of the indicated cytokines using ELISA kits. `0' represents
background cytokine production in supernatants collected from cultures of a representative GA-speci®c
T-cell line (GA-3) in the absence of GA. `GA' indicates the cytokine production in the absence of
puri®ed GA antibodies.
intrinsic properties that potentially promote myelin repair in a
murine model of demyelinating disease (Ure and Rodriguez,
2002). Therefore, the issue regarding clinical consequences
of GA antibodies induced by GA treatment is complex, and
the outcome may depend mainly on interplay between the
blocking effects on immune regulation induced by GA and
the direct bene®cial effects of the antibodies. In conclusion,
the clinical relevance of GA antibodies must be evaluated in
clinical trials.
Acknowledgements
We wish to thank Mrs De La Rosa for patient and sample
coordination. This work was supported by a grant from the
2646
H. H. Salama et al.
Fig. 5 Serum cytokine concentrations of IL-4, IL-10, IL-12 and TNF-a were determined by ELISA in
12 patients who had been on GA treatment for 3±5 years and who had low antibody titres (ABI = 4) and
13 patients who had undergone treatment for the same length of time and who developed high titres of
GA antibodies (ABI = 16±64). Serum specimens obtained from a group of 14 untreated
relapsing remitting multiple sclerosis patients were tested as a reference.
Richardson Foundation and a fellowship grant from the
Egyptian government to H.H.S. The study was not related to
any pharmaceutical companies.
References
Aharoni R, Teitelbaum D, Sela M, Arnon R. Copolymer 1 induces
T-cells of the T helper type 2 that crossreact with myelin basic
protein and suppress experimental autoimmune encephalomyelitis.
Proc Natl Acad Sci USA 1997; 94: 10821±6.
Aharoni R, Teitelbaum D, Sela M, Arnon R. Bystander suppression
of experimental autoimmune encephalomyelitis by T cell lines and
clones of the Th2 type induced by copolymer 1. J Neuroimmunol
1998; 91: 135±46.
Aharoni R, Teitelbaum D, Leitner O, Meshorer A, Sela M,
Arnon R. Speci®c Th2 cells accumulate in the central nervous
system of mice protected against experimental autoimmune
encephalomyelitis by copolymer I. Proc Natl Acad Sci USA
2000; 97: 11472±7.
Aharoni R, Meshorer A, Sela M, Arnon R. Oral treatment of mice
with copolymer 1 (glatiramer acetate) results in the accumulation of
speci®c Th2 cells in the central nervous system. J Neuroimmunol
2002; 126: 58±68.
Antonelli G, Bagnato F, Pozzilli C, Simeoni E, Bastianelli S,
Currenti M, et al. Development of neutralizing antibodies in
patients with relapsing-remitting multiple sclerosis treated with
IFN-b1a. J Interferon Cytokine Res 1998; 18: 345±50.
Ben-Nun A, Mendel I, Bakimer R, Fridkis-Hareli M, Teitelbaum D,
Arnon R, et al. The autoimmune reactivity to myelin
oligodendrocyte glycoprotein (MOG) in multiple sclerosis is
potentially pathogenic: effect of copolymer 1 on MOG-induced
disease. J Neurol 1996; 243 (4 Suppl 1): S14±22.
Brenner T, Arnon R, Sela M, Abramsky O, Meiner Z, RivenKreitman R, et al. Humoral and cellular immune responses to
Copolymer 1 in multiple sclerosis patients treated with Copaxone.
J Neuroimmunol 2001; 115: 152±60.
Chabot S, Yong FP, Le DM, Metz LM, Myles T, Yong VW.
Cytokine production in T lymphocyte-microglia interaction is
attenuated by glatiramer acetate: a mechanism for therapeutic
ef®cacy in multiple sclerosis. Mult Scler 2002; 8: 299±306.
Chen M, Gran B, Costello K, Johnson K, Martin R, Dhib-Jalbut S.
Glatiramer acetate induces a Th2-biased response and
crossreactivity with myelin basic protein in patients with multiple
sclerosis. Mult Scler 2001; 7: 209±19.
Chen M, Conway K, Johnson K, Martin R, Dhib-Jalbut S. Sustained
immunological effects of glatiramer acetate in patients with
multiple sclerosis treated for over 6 years. J Neurol Sci 2002;
201: 71±7.
Clanet M, Radue EW, Kappos L, Hartung HP, Hohlfeld R,
Sandberg-Wollheim M, et al. A randomized, double-blind, dosecomparison study of weekly interferon b-1a in relapsing multiple
sclerosis. Neurology 2002; 59: 1507±17.
Comi G, Filippi M, Wolinsky JS. European/Canadian multicenter,
double-blind, randomized, placebo-controlled study of the effects of
glatiramer acetate on magnetic resonance imagingÐmeasured
disease activity and burden in patients with relapsing multiple
sclerosis. European/Canadian Glatiramer Acetate Study Group. Ann
Neurol 2001a; 49: 290±7.
Comi G, Filippi M, Wolinsky JS. The extension phase of the
European-Canadian MRI study demonstrates a sustained effect of
glatiramer acetate in patients with relapsing-remitting multiple
sclerosis [abstract]. European/Canadian Glatiramer Acetate Study
Group. Neurology 2001b; 56 (8 Suppl 3): A255.
European Study Group on Interferon b-1b in Secondary Progressive
Multiple Sclerosis. Placebo controlled multicentre randomised trial
of interferon b-1b in treatment of secondary progressive multiple
sclerosis. Lancet 1998; 352: 1491±7.
Farina C, Then Bergh F, Albrecht H, Meinl E, Yassouridis A,
Neuhaus O, et al. Treatment of multiple sclerosis with Copaxone
(COP). Elispot assay detects COP-induced interleukin-4 and
interferon-g response in blood cells. Brain 2001; 124: 705±19.
Farina C, Vargas V, Heydari N, Kumpfel T, Meinl E, Hohlfeld
R. Treatment with glatiramer acetate induces speci®c IgG4
Blocking effects of GA antibodies in multiple sclerosis
antibodies in multiple sclerosis patients. J Neuroimmunol 2002;
123: 188±92.
Fridkis-Hareli M, Strominger JL. Promiscuous binding of synthetic
copolymer 1 to puri®ed HLA-DR molecules. J Immunol 1998; 160:
4386±97.
Fridkis-Hareli M, Teitelbaum D, Gurevich E, Pecht I, Brautbar C,
Kwon OJ, et al. Direct binding of myelin basic protein and synthetic
copolymer 1 to class II major histocompatibility complex molecules
on living antigen-presenting cellsÐspeci®city and promiscuity.
Proc Natl Acad Sci USA 1994; 91: 4872±6.
Gran B, Tranquill LR, Chen M, Bielekova B, Zhou W, Dhib-Jalbut
S, et al. Mechanisms of immunomodulation by glatiramer acetate.
Neurology 2000; 55: 1704±14.
Herndon RM, Jacobs LD, Coats ME, Goodkin DE, Mass MK,
Richert JR, et al. Results of an ongoing, open-label, safetyextension study of interferon b-1a (Avonex) treatment in multiple
sclerosis. Int J MS Care [serial online]. Dec 1999; 2: 1±6. Available
from http://mscare.com.
Hussien Y, Sanna A, SoÈderstroÈm M, Link H, Huang Y-M.
Glatiramer acetate and IFN-b act on dendritic cells in multiple
sclerosis. J Neuroimmunol 2001; 121: 102±10.
IFNB Multiple Sclerosis Study Group. Interferon b-1b is effective
in relapsing-remitting multiple sclerosis. 1. Clinical results of a
multicenter, randomized, double-blind, placebo-controlled trial.
Neurology 1993; 43: 655±61.
IFNB Multiple Sclerosis Study Group and the University of British
Columbia Multiple Sclerosis/MRI Analysis Group. Neutralizing
antibodies during treatment of multiple sclerosis with interferon b1b: experience during the ®rst three years. Neurology 1996; 47:
889±94.
Jacobs LD, Beck RW, Simon JH, Kinkel RP, Brownscheidle CM,
Murray TJ, et al. Intramuscular interferon b-1a therapy initiated
during a ®rst demyelinating event in multiple sclerosis. New Engl J
Med 2000; 343: 898±904.
Johnson KP, Brooks BR, Cohen JA, Ford CC, Goldstein J, Lisak
RP, et al. Copolymer 1 reduces relapse rate and improves disability
in relapsing-remitting multiple sclerosis: results of a phase III
multicenter, double-blind, placebo-controlled trial. Neurology
1995; 45: 1268±76.
Karandikar NJ, Crawford MP, Yan X, Ratts RB, Brenchley JM,
Ambrozak DR, et al. Glatiramer acetate (Copaxone) therapy
induces CD8+ T cell responses in patients with multiple sclerosis.
J Clin Invest 2002; 109: 641±9.
2647
Li Q, Milo R, Panitch H, Swoveland P, Bever CT Jr. Glatiramer
acetate blocks the activation of THP-1 cells by interferon-gamma.
Eur J Pharmacol 1998; 342: 303±10.
Maron R, Slavin AJ, Hoffmann E, Komagata Y, Weiner HL. Oral
tolerance to copolymer 1 in myelin basic protein (MBP) TCR
transgenic mice: cross-reactivity with MBP-speci®c TCR and
differential induction of anti-in¯ammatory cytokines. Int Immunol
2002; 14: 131±8.
Neuhaus O, Farina C, Wekerle H, Hohlfeld R. Mechanisms of
action of glatiramer acetate in multiple sclerosis. Neurology 2001;
56: 702±86.
Panitch H, Goodin DS, Francis G, Chang P, Coyle PK, O'Connor P,
et al. Randomized, comparative study of interferon b-1a treatment
regimens in multiple sclerosis. The EVIDENCE Trial. Neurology
2002; 59: 1496±1506.
PRISMS (Prevention of Relapses and Disability by Interferon-b-1a
Subcutaneously in Multiple Sclerosis) Study Group. Randomized
double-blind placebo-controlled study of interferon b-1a in
relapsing/remitting multiple sclerosis. Lancet 1998; 352: 1498±504.
PRISMS Study Group and the University of British Columbia MS/
MRI Analysis Group. PRISMS-4: Long-term ef®cacy of interferonb-1a in relapsing multiple sclerosis. Neurology 2001; 56: 1628±36.
Racke MK, Martin R, McFarland H, Fritz RB. Copolymer-1induced inhibition of antigen-speci®c T cell activation: interference
with antigen presentation. J Neuroimmunol 1992; 37: 75±84.
Teitelbaum D, Milo R, Arnon R, Sela M. Synthetic copolymer 1
inhibits human T-cell lines speci®c for myelin basic protein. Proc
Natl Acad Sci USA 1992; 89: 137±41.
Teitelbaum D, Fridkis-Hareli M, Arnon R, Sela M. Copolymer 1
inhibits chronic relapsing experimental allergic encephalomyelitis
induced by proteolipid protein (PLP) peptides in mice and interferes
with PLP-speci®c T cell responses. J Neuroimmunol 1996; 64: 209±
17.
Ure DR, Rodriguez M. Polyreactive antibodies to glatiramer acetate
promote myelin repair in murine model of demyelinating disease.
FASEB J 2002; 16: 1260±2.
Wiesemann E, Klatt J, SoÈnmez D, Blasczyk R, Heidenreich F,
Windhagen A. Glatiramer acetate (GA) induces IL-13/IL-5
secretion in naõÈve T-cells. J Neuroimmunol 2001; 119: 137±44.
Received December 2, 2002. Revised May 28, 2003.
Second revision June 20, 2003. Accepted June 23, 2003