Download Multiple sclerosis: a two-stage disease - CCIS

© 2001 Nature Publishing Group http://immunol.nature.com
N EWS & V IEWS
Multiple sclerosis: a
two-stage disease
The pathogenesis of multiple sclerosis
consists of an inflammatory and
neurodegnerative phase. Better
understanding of these stages has aided the
development of specific therapeutic targets.
© 2001 Nature Publishing Group http://immunol.nature.com
LAWRENCE STEINMAN
Multiple sclerosis (MS) usually begins in early
adulthood with an autoimmune inflammatory
“strike” against components of the myelin
sheath. Paralysis, sensory disturbances, lack of
coordination and visual impairment are common features. The disease often starts with an
“attack” that lasts from a few days to weeks;
this is followed by remission that lasts from a
few months to years. This relapsing-remitting
phase often lasts five to ten years, but ∼30% of
individuals with this form of MS enter a secondary chronic-progressive state. This chronicprogressive state is often characterized by the
inability to walk, which leaves the MS patient
wheelchair-bound. In the chronic-progressive
phase, distinct attacks are rare and the disease
progresses insidiously. Occasionally, however,
clinical disability begins with this progressive
phase, in which case the disease is called “primary-progressive MS”. Evidence indicates that
the earlier phase of disease, characterized by
distinct attacks followed by remission, may be
mediated by an autoimmune reaction. The subsequent chronic phase of disease is due to
degeneration of both the myelin sheath, which
is synthesized by oligodendroglial cells, and
the underlying axon, which emanates from the
neuronal cell body some distance away. Indeed
it is axon loss in the spinal cord and spinal cord
atrophy that correlate most strongly with the
inability to walk and paralysis1,2.
Worldwide, approximately 1,000,000 individuals are afflicted with MS. Women with the
disease outnumber men two to one. This bias
towards females is seen in other autoimmune
diseases, for example, rheumatoid arthritis, systemic lupus erythematosus and thyroiditis.
Genome-wide studies have revealed that susceptibility to MS is linked to genes in the major histocompatibility complex (MHC) on chromosome 63–5. Alleles for certain class II genes,
HLA-DR and HLA-DQ, confer the strongest
risk of contracting MS. Other genes within the
HLA complex are involved in the pathogenesis
of MS, including expression of tumor necrosis
factor-α (TNF-α), various components of the
complement cascade and myelin oligodendroglial glycoprotein. More recently, transcriptional profiling with gene microarrays and largescale sequencing of transcripts from MS lesions
have identified a number of genes that are
involved in the pathogenesis of acute disease.
762
These include immunoglobulin and interleukin
6 (IL-6) as well as osteopontin, which plays a
role in the transition from relapsing-remitting to
chronic MS (Fig. 1a).
We have recently learned, from sequencing
the human genome and the genomes of various
microbes, that biological organisms share many
genes. Hence various proteins are used, in a
modular manner, to build structures whose
intrinsic components can resemble each other.
If a human and a microbe invading that human
share a common gene sequence that encodes
one of these conserved structural motifs, the
immune system, in recognizing a structure on
this foreign microbe, may mistakenly also
attack “self”. In the context of MS, many
microbial protein sequences share homologies
with structures found on the myelin sheath; this
leads to an attack on myelin via a process called
molecular mimicry. Relapses in MS are often
triggered by common viral infections. Viruses
such as herpesvirus 6, influenza, measles, papilloma virus and Epstein-Barr Virus all have
genes encoding sequences that mimic those
found in the major structural proteins of myelin.
Indeed, antibodies to components of the myelin
sheath cross-react and bind sequences from
these microbes. T cells also recognize
sequences from the myelin sheath that are
shared with these microbial sequences6. Once a
T cell, B cell or macrophage is activated by a
foreign microbe, self-protein or microbial
superantigen, it may penetrate the blood-brain
barrier.
Penetration of the blood-brain barrier by activated lymphocytes is a multistep process (Fig.
1). There are specialized capillary endothelial
cells in the central nervous system (CNS) that
are nonfenestrated and connected through tight
junctions. During the inflammatory response,
TNF-α and interferon-γ (IFN-γ) induce these
capillary endothelial cells to express vascular
cell adhesion molecule (VCAM) and MHC
class II molecules. Activated T cells express
integrins, such as very late antigen (VLA-4),
and members of the immunoglobulin superfamily, such as CD4, that can bind VCAM and
MHC class II molecules, respectively. Once
activated, any T cell expressing VLA-4, for
example, can bind to adhesion molecules on the
surface of inflamed endothelium and “walkthrough” the endothelium. In an animal model
nature immunology
•
volume 2 no 9
•
september 2001
•
of MS, acute experimental autoimmune
encephalomyelitis (EAE), blockade of VLA-4
reverses clinical paralysis and prevents further
relapses in the chronic model of this disease. In
acute MS lesions, VLA-4 is found on T cells
that collect in the “perivascular lymphocyte
cuff”, a region around veins and capillaries that
is limited by the extracellular matrix. Clinical
studies with a human antibody to VLA-4 are
now in Phase III following promising Phase II
trial results in which the incidence of MS
relapses was reduced.
Once the activated lymphocytes have
extravasated, they still must pass through a barrier of extracellular matrix, comprised of type
IV collagen, before they can enter the CNS.
Matrix metalloproteases (MMPs) are a family
of structurally and functionally related
enzymes that are involved in the degradation of
the extracellular matrix as well as the proteolysis of myelin components in MS. MMPs contain Zn2+at their active site, show TNF-α convertase activity and induce the cleavage of
TNF-α from a cell-bound to a soluble form.
Gelatinase A and B (also called MMP2 and
MMP9) play a key role in penetration of the
extracellular matrix. These MMPs are
detectable in the spinal fluid of MS patients,
and gelatinase B immunoreactivity is present in
endothelial cells, pericytes, macrophages and
astrocytes of MS lesions7. Myelin-specific T
cell clones derived from MS patients also produce gelatinase B upon activation with antigen.
The presence of gelatinase B in the perivascular infiltrate is associated with disruption of the
type IV collagen-positive basement membrane
and is critical in the opening of the blood-brain
barrier. Once the blood-brain barrier is
breached, inflammatory cells spread into the
white matter of the CNS. MMP inhibition by
tissue inhibitors of matrix metalloproteases
(TIMPs) can block TNF-α and thereby downregulate the induction of adhesion molecules
such as VCAM. TIMP-1 is present in the spinal
fluid of MS patients and is inducible by various
cytokines, including TNF-α. In terms of MS
therapy, IFN-β, a potent inhibitor of gelatinase
B activity, has been used relatively successfully in clinical trials. Inhibition of gelatinase is
thought to interfere with T cell migration into
the CNS as well as T cell secretion of TNF-α.
Other MMP inhibitors are currently under
http://immunol.nature.com
© 2001 Nature Publishing Group http://immunol.nature.com
N EWS & V IEWS
a
Inflammatory phase
Antigen-presenting cell
Neuron
VCAM
T cell
receptor
VLA-4
Myelin
fragment Complement
Myelin
sheath
T cell
reentry into
circulation
Axon
B cell
Venule
Collagen type IV
on the ECM
Adhesion and
penetration of
blood-brain barrier
Naked
axon
Antibody
Cytokines and
chemokines
eg TNF-α,
IFN-γ, IL-12,
Osteopontin
and IL-6
Macrophage
Demyelination
Release of
toxic mediator
b
Degenerative phase
Autoreactive
T cell
PLP
brain-specific
isoform
Excess
glutamate
Glutamate
receptor
AMPA or
kainate receptor
MHC class II
Antigenpresenting
cell
Myelin
sheath
Axon
Necrosis
Glutamate receptor
AMPA or kainate
receptor
Terminal
axon bulb
Bob Crimi
© 2001 Nature Publishing Group http://immunol.nature.com
MHC class II
Figure 1.The two stages of the progression of MS. (a) Autoimmune attack. (b) Neurodegeneration.
(ECM, extracellular matrix)
intense development for clinical MS trials.
Once immune cells have spread to the white
matter of the CNS, the immune response is targeted to the entire supramolecular myelin complex. Antibodies to various myelin proteins and
lipids of the myelin sheath, as well as to molecules expressed in the CNS, are secreted by B
cells that have migrated to the brain or from
serum that has extravasated across the bloodbrain barrier6,7. Activated complement proteins
appear in the spinal fluid along with membrane-attack complexes, which represent the
terminal components of this cascade. T cells
target certain proteins normally found in the
myelin sheath. These include myelin basic protein, myelin oligodendroglial glycoprotein and
http://immunol.nature.com
•
september 2001
•
volume 2 no 9
•
proteolipid protein, as well as stress proteins
such as αB crystallin, which is found in the
myelin sheath after activation via the inflammatory response. The T cells produce
cytokines, notably lymphotoxin-α (LT-α) and
TNF-α, which are members of the TNF family.
LT-α is secreted as a LT-α3 homotrimer and,
like TNF-α, can bind to the p55 TNF receptor
(p55-TNFRI) or the p75 TNFR (p75-TNFRII).
These cytokines induce macrophages,
microglial cells and astrocytes to produce NO
and osteopontin.
The free radical NO is a major mediator in
autoimmune diseases. NO is involved in the
killing of oligodendroglial cells by microglia.
Nitric oxide synthase (iNOS), which catalyzes
NO synthesis, has been found in demyelinating
lesions in MS. Both IFN-γ and TNF-α induce
iNOS transcription in astrocytes, microglia and
macrophages. The combined effect of antibody, complement, NO and TNF-α damages
myelin and induces the macrophage to phagocytose large chunks of the myelin sheath. In
addition, macrophages and T cells produce
osteopontin. This induces more T helper subset
1 (TH1) cytokines, including IFN-γ and IL-12,
and down-regulates TH2 cytokines such as IL10. TH1 cytokines may exacerbate MS, whereas TH2 cytokines may reduce the extent of MS
lesions8. This concerted attack by T cells, B
cells, complement and inflammatory mediators
such as cytokines, osteopontin and NO produces areas of demyelination, which impairs
electrical conduction along the axon and produces the pathophysiological defect.
TNF-α, LT and other members of the TNFR
family may also play key roles in the pathogenesis of oligodendroglial damage. TNF
expression is elevated in the spinal fluid in MS
relapses; TNF has also been found in MS
lesions. In addition, myelin basic protein–reactive T cells from HLA-DRB1*15-positive MS
patients express increased amounts of TNF-α.
Experimental trials with altered peptide
analogs of myelin basic protein that down-regulate the expression of TNF and up-regulate
TH2 cytokines can decrease the size of new
lesions in white matter8. Similarly the approved
drug, Copaxone, induces a shift towards TH2
cytokine production by myelin-reactive T cells.
This reduces the frequency of relapses in early
MS and decreases the degree of inflammatory
activity in white matter.
During chronic MS, when exacerbations
and remissions of MS are rare, there is evidence for axon loss and atrophy of the brain
and spinal cord1,2,9. In EAE and in MS, AMPA
(α-amino-3-hydroxy-5-methyl-4-isoxazoleppropionic acid), which mediates toxicity
induced by the excitatory neurotransmitter
nature immunology
763
© 2001 Nature Publishing Group http://immunol.nature.com
© 2001 Nature Publishing Group http://immunol.nature.com
N EWS & V IEWS
gluatamate, is present on oligodendroglial
cells and neurons. During inflammation in
both MS and EAE, lymphocytes, brain
microglia and macrophages release excessive
amounts of glutamate, which then activate
AMPA receptors (Fig. 1). Blockade of these
receptors with antagonists can ameliorate
EAE, including clinical relapses when treatment is begun after the onset of paralysis10.
The blockade of AMPA receptors does not
influence the immune response to myelin antigens but somehow protects oligodendroglial
cells and axons from immune-mediated damage. Damage may be due to increased fluxes
of calcium, which may cause necrotic damage
to oligodendroglial cells and axons. The use of
neuroprotective agents that block glutamate
receptor subtypes is the key focus in the development of new therapies for stroke and neurodegenerative conditions, and this approach
could prove useful for treatment of the chronic degenerative phase of MS as well.
The recognition of an inflammatory and a
neurodegenerative phase of MS has facilitated
the targeting of therapies that are specific for
various stages of MS. Thus, drugs such as
IFN-β interfere with lymphocyte migration to
brain; altered peptides and Copaxone affect
cytokine production by autoimmune T cells;
and glutamate receptor antagonists block the
Systemic lupus
erythematosus: an
autoimmune disease
of B cell hyperactivity
insidious atrophy and death of oligodendroglial cells and the underlying axon.
1.
2.
3.
4.
5.
6.
7.
Trapp, B. D. et al. New Engl. J. Med. 338, 278–285 (1998).
Loseff, N. A. et al. Brain 119, 701–708 (1996).
Haines, J. L. et al. Nature Genet. 13, 469–471 (1996).
Ebers, G. C. et al. Nature Genet. 13, 472–476 (1996).
Sawcer, S. et al. Nature Genet. 13, 464–468 (1996).
Wucherpfennig, K.W. et al. J. Clin. Invest. 100, 1114–1122 (1997).
Conlon, P., Oksenberg, J. R., Zhang, J. & Steinman, L. Neurobiol. Dis.
6, 149–166 (1999).
8. Kappos, L. et al. Nature Med. 6, 1176–1182 (2000).
9. Lucchinetti, C. F., Bruck,W., Rodriguez, M. & Lassmann, H. Brain
Pathol. 6, 259–274 (1996).
10. Pitt, D.,Werner, P. & Raine, C. Nature Med. 6, 67–70 (2000).
Department of Neurology and Neurological Sciences,
Stanford University School of Medicine, Stanford, CA
94305, USA. ([email protected])
B cells can regulate many aspects of immune
reactivity, as well as differentiate into
antibody-producing cells. In SLE, a systemic
autoimmune disease, recent research
suggests enhanced B cell function is the
defining pathogenic event.
PETER E. LIPSKY
Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease characterized
by the production of numerous autoantibodies
and involvement of skin, joints, kidneys,
brain, serosal surfaces, blood vessels, blood
cells, lungs and heart. As opposed to lupus in
animal models, SLE in humans is heterogeneous and affects different individuals with a
wide range of disease courses and manifestations. In addition, the progression of SLE in
an individual is difficult to predict at the
onset, although a number of factors are associated with more frequent or aggressive disease. These include female gender, AfricanAmerican and African-Caribbean origin and
restricted educational experience1.
SLE has been considered to be the prototypic systemic autoimmune disease because it
is associated with the production of a host of
autoantibodies, some of which appear to have
pathogenic consequences1,2. In general, systemic autoimmune diseases have been distinguished from organ-specific autoimmune diseases. The latter are thought to develop as a
result of immune responses to a limited set of
specific autoantigens or cross-reactive exogenous antigens, whereas the former are
believed to emerge from a more global abnor764
mality in immunoregulation. Indeed, for most
autoimmune conditions, it remains uncertain
whether such a clear distinction can be made
because systemic abnormalities, such as
defects in pathways of apoptosis, may lead to
limited expression of autoimmunity in
humans. Implicit in this delineation is the distinction between autoimmunity and autoimmune disease. Autoimmunity, the production
of autoantibodies or the activation and expansion of T cells that react with autologous
cells, proteins or tissues, is relatively common. Autoimmune disease, in which selfreactivity leads to tissue pathology, is much
less common. In most circumstances, the
determinants of tissue damage and their relationship to autoimmunity have not been fully
determined.
SLE is characterized by diffuse autoimmunity and characteristic tissue pathology.
Despite the presence of both autoantibodies
and tissue pathology in SLE, the relationship
remains controversial and exact explanations
for many of the clinical manifestations of this
disease remain unknown. Genetically determined characteristics of the immune
response—including its specificity, intensity
or resolution as well as the quality, magnitude
nature immunology
•
volume 2 no 9
•
september 2001
•
or diversity of the effector molecules produced or the nature of the end organs themselves—may contribute to tissue pathology in
SLE1,2.
SLE is associated with a myriad of
immunoregulatory
abnormalities.
Considerable interest has been focused on
abnormalities in T cell responses or production of T cell cytokines and/or defective control by regulatory T cells. One or other of
those abnormalities has been hypothesized to
play an essential role in the evolution of
autoimmunity and tissue pathology1. More
recent data suggest, however, that an intrinsic
tendency of B cells to respond excessively to
immune stimulation may be an essential feature of SLE2. One feature of the B cell hypothesis is the central relationship between SLE
and the production of characteristic patterns
of autoantibodies, some of which are clearly
involved in tissue damage. These include antiDNA in glomerulonephritis, anti-cardiolipin
in thrombosis and anti-Ro in congenital heart
block2. A challenge to the central role of
autoantibodies in SLE, however, has come
from results of genetically manipulated lupusprone MRL lpr/lpr mice that develop nephritis and vasculitis despite being unable to
http://immunol.nature.com