TRENDSin Molecular Medicine Vol. 7 No. 3 March 2001
`
`115
`
`Heterogeneity of multiple sclerosis
`pathogenesis: implications for
`diagnosis and therapy
`
`Hans Lassmann, Wolfgang Brick and Claudia Lucchinetti
`
`Multiple sclerosisis a chronic inflammatory disease of the nervous system in which
`a T-cell-mediated inflammatory process is associated with destruction of myelin
`sheaths. Although demyelinationis the primary event, axonsare also destroyedin
`the lesions, and the lossof axonscorrelates with permanentfunctionaldeficit.
`Here, we discuss evidence that demyelination and axonaldestruction follow
`different pathogenetic pathwaysin subgroupsofpatients. This might,at least in
`part, explain the heterogeneity in genetic susceptibility, clinical presentation
`and responseto treatment observed betweenindividuals.
`
`Multiple sclerosis (MS) is the most common
`neurological disorder in young adults in the developed.
`world.It is a chronic inflammatory disease ofthe
`central nervous system (CNS), whichleads to large
`focal lesions ofPRIMARY DEMYELINATION(see Glossary)
`with relative axonal preservation. It is considered to be
`an autoimmunedisease that is induced when T HELPER 1
`CELLS (Th1) recognize components ofthe myelin sheath.
`Activated, autoreactiveT cells within the lesions are
`believed to drive the chronic inflammatory process and
`activate local or hematogenous macrophages that
`destroy myelin. However, we suggestthat this
`pathogenetic schemeis oversimplified and cannot
`explain lesion formation. It is known that T-cell
`populationsother than classical Th1 cells contribute to
`inflammation in MS and that amplification of
`demyelination in a chronic inflammatory reaction in the
`brain requires additional factors. Furthermore, the
`patterns ofdemyelination are different between
`different subgroups ofMS patients, which suggests that
`the disease is heterogeneous. Depending upon the
`patient, factors that amplify demyelination can be
`either antibodies directed against surface components
`ofmyelinor factors that impair the metabolism of
`myelin-supporting oligodendrocytes. Although these
`insights are based mainly on pathological studies of
`biopsy or autopsy tissue, new clinical and paraclinical
`markers to distinguish these different patterns of
`demyelination look promising. The heterogeneity in the
`pathology ofMS could form thebasis ofboth the
`polygenic nature and the profound heterogeneity ofthe
`disease with respectto clinical presentation and
`responseto therapy.
`
`Basic features of MSlesions
`
`The pathology of MSis distinguished from thatof
`other inflammatory diseases ofthe nervous system by
`the presenceoflarge, multifocal, demyelinated
`
`Hans Lassmann*
`Division of
`Neuroimmunology,Brain
`Research Institute,
`University of Vienna,
`Spitalgasse 4, A-1090
`Wien,Austria.
`*e-mail:
`hans.lassmann@
`univie.ac.at
`
`Wolfgang Brick
`Institute of
`Neuropathology, Charité
`AugustenburgerPlatz 1,
`D-13353 Berlin, Germany.
`Claudia Lucchinetti
`Dept Neurology, Mayo
`Clinic, 200 First Street SW,
`Rochester, MN 55905,
`USA.
`
`plaqueswith reactive glial scar formation!?. This
`demyelinating process is accompanied by an
`inflammatory reaction with infiltrates composed
`mainly ofT cells and macrophages. Although myelin
`sheathsare the primary target oftissue destruction,
`axons,nervecells and astrocytes are also affected,
`althoughto a lesser degree. Activelesions, defined by
`the ongoing destruction ofmyelin, are heavily
`infiltrated by macrophages andactivated microglial
`cells. These cells are closely associated with the
`disintegrating myelin sheaths and are responsible for
`the uptake and removal of myelin debris.
`Although MSis a primary demyelinating disease
`with relative sparing ofaxons, the emphasisis on the
`term ‘relative’. Acute axonalinjury is frequent in
`actively demyelinating MSlesions*, and this leads to
`a 50-70% reduction in neurite density in chronic
`plaques, compared with normal tissue. Although
`demyelination can be repaired,at least in part, by
`remyelination?, axonal destruction is irreversible.
`Thus, in MS patients, the relapsing—remitting
`functional impairmentis caused mainly by
`inflammation and demyelination, whereas the
`accumulation of an irreversible neurological deficit is
`caused mainly by axonal destruction andloss.
`
`Is Th1-mediated autoimmunity against myelin the
`causeofinflammation in MS?
`
`Thepathology ofinflammation in MSlesionsis
`consistent with a T-cell-mediated immunereaction,
`leading to the recruitment ofhematogenous
`macrophages andactivation ofmicroglia‘. This is
`similar to the pathology ofexperimental autoimmune
`encephalomyelitis (EAE), a disease induced by
`immunization ofanimals with CNStissue, myelin or
`myelin proteins. Several features ofMS lesions suggest
`that, as in EAE,the inflammatory process in MSis
`driven by a Th1-mediated autoimmuneresponse. For
`example, the numberofautoreactive T cells with the
`cytokine spectrum ofTh1 cells is increasedin blood of
`MSpatients compared with controls. Furthermore,
`within actively demyelinating lesions, Th1-related
`cytokines, such as interferon y, tumour necrosis factor
`o (TNF-o) or interleukin 2, are expressed in invading
`leukocytes andlocal glial cells®. The spectrum of
`chemokines and chemokinereceptorsis also consistent
`with a Th1-driven inflammatory response®, and MSis
`
`http://tmm.trends.com 1471-4914/01/$ - see front matter © 2001 Elsevier ScienceLtd. All rights reserved. Pll: S1471-4914(00)01909-2
`
`1
`
`Hopewell EX1036
`
`Hopewell EX1036
`
`1
`
`

`

`116
`
`TRENDSin Molecular Medicine Vol. 7 No. 3 March 2001
`
`
`
`Th1
`
`Th2
`
`Tc1
`
`Inflammation ©
`
`Demyelination
`Pattern |
`
`Pattern II
`
`PatternIll
`
`Pattern IV
`
`Glossary
`Primary demyelination:destruction of myelin sheaths with
`relative sparing of axons.
`T helper1 cells (Th1): MHCclassIl-restricted T cells that secrete
`a spectrum of cytokinesincludinginterleukin 2, interferon y and
`lymphotoxin o, andelicit a delayed-type hypersensitivity reaction.
`T helper2 cells (Th2): MHCclassIl-restricted T cells that secrete
`mainly interleukins 4, 5, and 6. They stimulate antibody
`production andareinvolvedin allergic reactions.
`ClassI-restricted cytotoxic T cells (Tc1): cells that secrete a
`similar spectrum of cytokinesto Th1 cells.
`
`+t Re
`
`Primary
`Distal
`Antibody-
`Macrophage-
`_oligodendro- oligodendroglia
`mediated
`mediated
`demyelination demyelination gliopathy & degeneration
`TNF-a
`anti-MOG
`apoptosis metabolic defect
`ROI
`anti-GC
`ischemia/toxic
`proteinases
`Others?
`virus induced
`
`
`
`
`
`A
`
`Acute axonalinjury during phaseof active demyelination
`Macrophage toxins: Proteases; NO-radicals, TNFa
`Cytotoxic T cells
`
`Onepossible reasonfor this might be that the
`pathogenesis of MSlesions is more complex than a
`pure Th1-mediated CNS autoimmunedisease. There
`is evidence thatcells other than classical Th1 cells
`contribute to the inflammatory response in MSlesions
`(Fig. 1). Numerically, CD8,class I-restricted T cells
`outnumber CD4cells®. Furthermore,classI-restricted
`T cells are predominantat the site oftissue
`destruction in actively demyelinating lesions, whereas
`CD4 cells are retained mainly in perivascular
`inflammatory infiltrates!°. Recent studies using PCR
`to analyze single cells show that clonal expansionis
`much more prominentin the CD8 population than in
`the CD4 T-cell population!®. Additionally, the extent of
`axonal injury and tissue destruction correlates better
`with the numberof macrophages and CD8cells in the
`Fig. 1. Summary of pathogenetic mechanismsinvolvedin the formation of
`lesions than with CD4cells. Taken together, these
`multiple sclerosis lesions. Inflammation: evidence indicates that
`data suggestthat class I-restricted T cells could play
`Thelper1 (Th1) cells have a role in inducing inflammatory reactionsin the
`an importantrole in the pathogenesis of MS.
`central nervoussystem. Proinflammatory cytokines released from
`Anotherfeature of inflammation in MSthatis
`Th1cells activate macrophages,which are responsiblefor the majority of
`demyelination and axonalinjury. In addition, however, T helper 2 (Th2)
`incompatible with the conceptofa purely Th-1-mediated
`cells and cytotoxic, CLASS RESTRICTED CYTOTOXICTCELLS (T¢1) might modify the
`disease is the abundance ofgranulocytes and eosinophils
`outcomeofthe lesions. Demyelination: myelin sheaths and
`in active lesions of acute and fulminate variants,in
`oligodendrocytes (OG) can be destroyed,possibly bydifferent
`mechanismsindifferent individuals.This results in distinctly different
`particular in Marburg’s type ofacute MS and Devic's
`patterns of demyelination in active lesions. Demyelination may be induced
`type ofneuromyelitis optica. An inflammatory reaction
`by macrophages(M)and/ortheir toxic products (resulting in pattem |), by
`similar to that found in Devic's disease can be induced by
`specific demyelinating antibodies and complement (C,resulting in pattern
`transfer ofTh2-polarized autoreactiveT cells to
`Il), by degenerative changesin distal processes,in particular those of
`periaxonal oligodendrocytes(distal oligodendrogliopathy), followed by
`immunocompromised animals". These data suggest
`apoptosis (resulting in pattern III) or by a primary degeneration of
`that T HELPER 2 CELL(TH2)-mediated mechanisms might
`oligodendrocytes followed by myelin destruction (resulting in pattern IV).
`contribute to inflammation in subsets ofpatients with
`Possible mediators of myelin and oligodendrocyte destruction include
`MS(Ref. 12).
`tumornecrosis factor o (TNF- 0), reactive oxygenintermediates (RON),
`antibodies against myelin oligodendrocyte glycoprotein (anti-MOG)or
`galactocerebroside(anti-GC). Axonalinjury: axonalinjury follows acute
`destruction of myelin sheaths.In the active phase of demyelination, axonal
`injury is likely to be induced by macrophagetoxinsorby thedirect effects
`ofcytotoxic T cells. The chronic axonalinjury observedinactive plaques
`maybe caused bya lack oftrophic support byglial cells, such as
`oligodendrocytes, but could also involve inflammatory mediators,
`produced by macrophages,that persist in most active chronic lesions.
`
`Chronic axonalinjury in inactive demyelinated plaques
`Lack of trophic support by oligodendrocytes
`TRENDSin Molecular Medicine
`
`Additional demyelinating amplification factors are
`required to form demyelinated plaques
`In mostvertebrates, including rats, guinea pigs and
`primates,pure T-cell-mediated inflammationof the
`brain does not lead to demyelination. An exception to
`this is the mouse, in which extensive activation of
`macrophages and microglia in EAE,or in certain
`transgenic models,is associated with primary
`destruction of myelin. In these mouse models,
`signaling through TNFreceptor 1 is required for the
`destruction of myelin and oligodendrocytes".
`Pure inflammatory T-cell-mediated
`encephalomyelitis can, however, become an MS-like
`demyelinating disease in the presenceofspecific
`demyelinating amplification factors. One such factor,
`which has been extensively characterized recently, is
`
`associated with certain major histocompatibility
`complexclass IT haplotypes’. Thus, itis generally
`believed that the immunological mechanisms
`responsible for inflammation in MSaresimilar to those
`in EAE,an experimental paradigm ofTh1-mediated
`autoimmunedisease, and that immunomodulatory
`therapeutic strategies that work in EAE should also be
`beneficial in MS. Unfortunately, in many instances this
`approachhas been disappointing®.
`
`http://tmm.trends.com
`
`2
`
`

`

`TRENDSin Molecular Medicine Vol. 7 No. 3 March 2001
`
`117
`
`Box 1. Multiple sclerosis cases includedin a study of the heterogeneity of multiple sclerosis lesions?
`
`however, are samplesfrom 18 patients with
`different formsof chronic MS, which have a
`
`demyelination is present and appropriate
`careis taken in staging ofthe lesions.
`
`References
`
`As the mechanisms of demyelination can
`only be studied in MSlesions in which myelin
`is at the stage of disintegration, observations
`on the heterogeneity of demyelination are
`from studiesofactive lesions(Table1).
`Becauseactive lesions are more commonin
`
`patients with fulminate acute disease, such
`as that in acute MSorin biopsies from early
`MS,these conditions are over-representedin
`our material. Thus, the study contains an
`inherent bias towardspatients with
`exceptionally severe disease. Also included,
`
`similar spectrum oflesional patterns.
`Undoubtedly,the full spectrum of
`pathologywithin active MS lesionsis best
`recognized following autopsy, where large
`areasof the lesions plus surroundingbrain
`tissue can be analyzed. However, the
`different patterns of demyelination,
`described here (Table Il), can be identified
`unequivocally even in small stereotactic
`biopsies, provided an area ofactive
`
`a Lucchinetti, C. et a/. (2000) Heterogeneity of
`multiple sclerosis lesions: Implications for the
`pathogenesis of demyelination. Ann. Neurol.
`47, 707-717
`b Griot, C. et al. (1990). Selective degeneration of
`oligodendrocytes mediated by reactive oxygen
`species. Free Radic. Res. Commun. 11, 181-193
`c Brick, W.et al. (1995) Monocyte/macrophage
`differentiation in early multiple sclerosis. Ann.
`Neurol. 38, 788-796
`
`
`
` Lesionswithinfiltration by macrophages, which contain intracytoplasmic granules, immunoreactivefor all myelin
`
`
`Table I. Multiple sclerosis cases includedin this study
`Numberof cases
`Definition of clinical MS courses
`
`Acute multiple sclerosis (Marburg's Type):
`Fulminate inflammatory demyelinating disease with typical multiple sclerosis pathology leading to patient's
`death within one yearafter onset.
`Early multiple sclerosis (biopsies):
`Patients with severe acute neurological disease with atypical clinical and MRI presentation (e.g. large monofocal
`lesions; no intrathecal immunoglobulin synthesis);
`Biopsy taken generally during the first months after disease onset with pathologyof inflammatory demyelinating
`disease resembling MS;
`77% of these patients developclinically definite MS in follow up (averagefollow up time: 37 months).
`Chronic active multiple sclerosis:
`Patients with clinically definite chronic multiple sclerosis (relapsing-remitting; primary progressive or secondary progressive);
`At least one actively demyelinating lesion present within the CNS.
`Chronic inactive multiple sclerosis:
`Patients withclinically definite multiple sclerosis; (relapsing-remitting; primary progressive or secondary progressive);
`
`No actively demyelinating lesions in whole CNS.
`
`14 autopsies
`
`51 biopsies
`
`18 autopsies
`
`not included
`
`TableIl. Defining demyelinating activity®
`Numberof lesions
`Definition of demyelinating activity
`
`Active lesions:
`
`proteins, including minor myelin components such as myelin oligodendrocyte glycoprotein
`Inactive lesions:
`Demyelinated or remyelinated lesions, with or without macrophageinfiltration;
`Macrophagesdonotcontain degradation products, immunoreactive for minor myelin proteins;
`
`173 lesions in autopsies
`62 lesionsin biopsies
`
`152 lesions in autopsies
`9 lesions in biopsies
`
`direct T-cell-mediated cytotoxicity!5 and metabolic
`the presence ofdemyelinating antibodies directed
`against epitopes expressed on the surface ofmyelin
`impairmentofoligodendrocytes through ischemia’®,
`sheaths and oligodendrocytes". In this experimental
`toxins’’or virus infection!*®. Thus, a major challenge for
`future research,is to define which ofthese mechanisms
`paradigm,brain inflammation, mediatedbyTcells,
`induceslocal activation ofhematogenous macrophages
`operate in actively demyelinating MSplaques.
`or microglia, and impairs the blood—brain barrier. The
`latter allows circulating demyelinating antibodies and
`complement components to enter the CNS and destroy
`myelin, either by complementactivation or an
`antibody-dependentcellular cytotoxicity reaction. The
`pathological hallmark ofsuchlesions is the local
`precipitation ofthe lytic terminal complement complex
`on the surface ofmyelin sheaths and oligodendrocytes.
`However, other mechanisms that have been
`identified could induce myelin damage during the course
`ofan inflammatory process in the CNS.These include
`
`Heterogenous pattemsof demyelination in MS
`Amajorrestriction ofpathogenetic studies ofMS is
`the limited material available from actively
`demyelinating lesions, and it has required a large
`international effort to collect sufficient specimensto
`perform such a study!92° (Box 1). A detailed
`immunopathological investigation ofthis material
`has revealed a profound heterogeneity in the patterns
`of demyelination betweendifferent patients, although
`active plaques from the samepatient were very
`
`http://tmm.trends.com
`
`3
`
`

`

`
`
`(Il) Macrophage mediated
`
`(I) Antibody mediated
`
`(Ill) Distal oligodendrogliopathy
`
`(IV) Primary oligodendrocyte damage
`with secondary demyelination
`
`Perivenousdistribution oflesions;
`Radial expansionofthe lesions;
`Inflammatory infiltrates composed of
`T-cells and macrophages;
`Activated macrophagesand microglia
`associated with degenerating myelin.
`Similar lesionsasin | but additional
`
`deposition of immunoglobulin and
`activated complementatsites of active
`myelin destruction
`Inflammationby T-cells and macrophages;
`Small vessel vasculitis with endothelial cell
`damageand microvessel thrombosis;
`Degenerationof distal oligodendrocyte
`processes, followed by oligodendrocyte
`apoptosis and demyelination
`Similar lesionasin (I), but prominent
`oligodendrocyte degeneration in a small
`rim of periplaque white matter
`
`F-cell-mediated inflammation with
`
`macrophage/microglia activation;
`Demyelination induced by
`macrophagetoxins.
`
`T-cell-mediated inflammation with
`
`macrophage/microglia activation;
`Complementmediatedlysis of
`antibody-targeted myelin
`T-cell-mediated small vessel vasculitis
`
`with secondary ischemic damage of
`the white matter
`
`F-cell-mediated inflammation with
`
`
`
`TRENDSin Molecular Medicine Vol. 7 No. 3 March 2001
`
`
`Table 1. Essential characteristics of different patterns of demyelination in multiple sclerosis
`Putative mechanisms
`Pattemsof demyelination
`Pathology
`
`macrophage/microglia activation;
`Demyelination induced by
`macrophagetoxins on the background
`of metabolically impaired
`oligodendrocytes;
`Genetic defect of oligodendrocytes?
`
`similar. All actively demyelinating lesions were
`associated with an inflammatory process, with the
`inflammatory infiltrates composed mainlyofT cells
`and macrophages. Despite the similarities in the
`inflammatory reaction the lesions segregated into
`four patterns of myelin destruction (Table 1, Figs 1,2).
`Pattern I (macrophage-associated demyelination)
`closely resembles myelin destruction in mouse
`models of autoimmune encephalomyelitis. In these
`models, toxic products of activated macrophages,
`such as TNF-a(Ref. 13) or reactive oxygen species,
`are mainly responsible for the destruction of myelin
`sheaths?!. Lesions similar to pattern II
`(antibody-mediated demyelination) are found in
`models of EAE that are inducedby sensitization
`with myelin oligodendrocyte glycoprotein (MOG). In
`this model, demyelination is induced by cooperation
`between encephalitogenic T cells, which are responsible
`for inflammation, and demyelinating anti-MOG
`antibodies‘. So far, patterns III and IV have not been
`identified in experimental models of demyelinating
`disease. Distal oligodendrogliopathy-associated
`demyelination (pattern ITI), however, is commonly
`found in virus-induced human white-matter
`diseases”? and is also seen in the penumbraregion
`of white-matter strokes (Rauschkaet al,
`unpublished). Preliminary evidence from our
`laboratory suggests that white-matter ischemiais a
`major pathogenetic factor for demyelination and
`tissue damagein such lesions. The mechanisms
`responsible for pattern IV lesions (primary
`oligodendrocyte degeneration), which are the most
`infrequent in the MSpopulation and are restricted
`to a subset of patients with primary progressive
`disease”, are notclear.It is tempting to speculate
`
`http://tmm.trends.com
`
`that, in these patients, a genuine metabolic
`disturbanceof oligodendrocytes could render these
`cells particularly vulnerable to the toxic action of
`inflammatory mediators.
`The heterogenous patternsof pathology so far
`could only be related to a specific clinical disease
`variant in patients with most acute and fulminate
`diseasecourse.All patients with Devic’s type of
`neuromyelitis optica had antibody-mediated tissue
`damage?(patternII), whereas all patients with
`Balo’s type of concentric lesions hadlesionsof distal
`oligodendrogliopathy (pattern III). Furthermore,
`primary oligodendrocyte degeneration (pattern IV)
`has only been found in a small subsetofpatients with
`primary progressive disease. Exceptfor this
`segregation in the mostsevere disease variants, no
`specific association between the pattern of
`demyelination and clinical disease was found.
`
`Axonalinjury and loss: a consequenceof inflammatory
`demyelination
`Early descriptions ofthe pathology ofMS, published in
`the early 1900s, emphasized the functional importance
`of axonal destructionin the lesions, whichled to
`secondary (Wallerian) tract degeneration and brain
`atrophy°. This aspect ofMS pathology has received
`increased attention recently, as serial magnetic
`resonance imaging (MRI) investigations indicate that
`axonal loss within the lesions and brain atrophy
`correlate with permanent, progressive functional
`deficits. Axonal injury in MS plaques occurs in two
`stages”. A high incidence of acute axonal injury is
`found in lesions during the active stage ofmyelin
`destruction”®6, Thus, even during the earliest stages
`ofthe disease, every newly formedplaqueis associated
`
`4
`
`

`

`
`PPWM
` -
`j
`
`
`“>
`
`(e)Mog, NI)A it
`es VE wha J.
`
`119
`
`damage. Althoughtheinitial event that triggers axonal
`damage can differ, the final pathway ofaxonal
`destruction appearsto be similar in all conditions of
`cerebral damage;alterations in ion-channel
`permeability disturb calcium homeostasis within the
`axons, which leadsto activationofcalcium-dependent
`proteases,local degradation of cytoskeletal elements,
`blockade of axonal transport and,finally, axonal
`disruption”®.
`The extent of axonal damage in MSlesionsis
`variable and depends upontheseverity ofthe
`inflammatory process duringtheactive stage of
`demyelination. Other factors that influence the
`degreeof axonal injury are the pathogenetic
`mechanism of demyelination and,possibly,
`heterogeneity in the susceptibility ofindividual
`patients. Within the MS population, axonalloss
`appearsto be most importantin patients with either
`primary or secondary progressive disease courses.
`These data provide support for the concept that the
`mechanism of demyelination and tissue destruction is
`heterogenous betweendifferent subgroups of
`patients, and could have profound consequencesfor
`our understandingof disease pathogenesis andfor
`the future design of novel therapeutic strategies.
`
`Is inflammationin MSlesionsalways deleterious?
`Aspreviously discussed, T-cell-mediated
`inflammation is the apparent driving force behind the
`pathological process in MS lesions. Morerecent data,
`however, indicate that inflammation might also be
`neuroprotective, or have a role in the repair of
`damaged tissue within the CNS. For example,
`autoimmuneT cells protect neurons in the optic nerve
`from secondary degeneration after injury by partial
`crushing®°, The presence ofmacrophages stimulates
`remyelination in organ culture*!, an observation that
`could be particularly relevant for MSlesions.
`Furthermore, inflammatory cells within MSlesions
`synthesize neurotrophic factors, such as brain-derived
`neurotrophic factor*?. These data suggest that the
`inflammatory response within demyelinated plaques
`of MS patients mightalso have a role in the repair
`process and, thus, complete blockageof all
`inflammatory processes within the lesions could be
`counterproductive.
`
`
`
`TRENDSin Molecular Medicine Vol. 7 No. 3 March 2001
`
`
`
`Fig. 2. Histopathology ofdifferent patterns of demyelination in multiple sclerosis.(a) Actively
`demyelinating lesion following patterns| andII. The active plaque (PL)isfilled with activated
`macrophagesand microglia. Thereis a sharp demarcation betweenthe actively demyelinating lesions
`and theperiplaque white matter (PPWM). Immunocytochemistry for CD68 (to identify activated
`macrophages/microglia). Magnification x 200. (b) Actively demyelinating plaqueofpatternIl that shows
`massive deposition of complement C9neo-antigen (brownstaining) on degenerating myelin sheaths
`and in myelin degradation products taken up by macrophagesin the zoneof active demyelination
`(ADM).Thereis faint CSneoreactivity on myelin sheaths in the PPWM. Immunocytochemistry for
`C9neo-antigen. Magnification x 500.(c) Actively demyelinating lesion following patternIll. Myelin
`staining using Luxolfast blue showsanill-demarcated demyelinated plaque(PL). In the centre ofthe
`lesion is an inflamed blood vessel surrounded by a small rim of preserved myelin (arrow). Magnification
`x30. (d) The samelesion as shownin (c) stained with the leukocyte marker CD45. Myelin around the
`central vessel has a lowerdensity of inflammatory cells compared to therestofthe lesion (arrow). In
`addition,this lesion hasan indistinct boundary compared with thelesion in panel(a).
`Immunocytochemistry for CD45. Magnification x 30. (e) Higher magnification ofthe area indicated by
`the arrowin panels(c) and (d) stained for myelin-oligodendrocyte glycoprotein (MOG,brownstaining).
`There are numerous MOG-reactivefibers preservedin the lesion. Magnification x 300.(f) Higher
`magnificationofthe area indicated by the arrow in panels (c) and (d) stained for myelin associated
`glycoprotein (MAG). Thereis very little MAG immunoreactivity. Magnification x 300. (g) Actively
`demyelinating lesion following pattern IV. The plaque contains numerous macrophagescontaining
`myelin degradation products (stained blue with the Luxol fast blue myelin stain) and has a sharply
`demarcated edge. Magnification x 300.(h) The periplaque white matterofthelesion in (g). The myelin
`appears vacuolated and contains numerousoligodendrocytes with fragmented DNA(black nuclei)
`identified using anin situtailing reaction for DNA fragmentation. Magnification x 400.
`
`with a significant loss ofaxons. In this phase, the
`Clinical identification of subgroups of MS patients
`The heterogenouspatterns ofdemyelination in
`extent ofaxonal injury correlates with the numberof
`individuals with MSpatients are defined by the
`macrophages andclassI-restricted T cells in the
`lesions, and is apparently mediated by toxic products of
`pathology of active lesions. However, we expect that
`
`macrophages andTcells. A few mediators have been clinical diagnosis and therapy would be aided by
`developing clinical and paraclinical markers to identify
`identified, such asnitric oxide and proteases, that can
`induce axonal injury in vitroor in vive?’8. In addition
`patient subgroups accordingto their pathogenetic
`to axonal degenerationin active plaques,there is also a
`pathways oflesion formation.As the association
`low level of continuous axonal destruction and loss in
`betweenclinical forms ofthe disease and pathological
`chronic inactive demyelinated plaques*+. This is not
`subtypesis limited, this approach will largely depend
`seen in inactive, remyelinated lesions, so the lack of
`upon the developmentofsuitable paraclinical markers.
`trophic support by oligodendrocytes in demyelinated
`It is possible that subsets of MS patients will be
`lesions might render axons vulnerable to progressive
`identified using MRI and magnetic resonance
`
`http://tmm.trends.com
`
`5
`
`

`

`120
`
`TRENDSin Molecular Medicine Vol. 7 No. 3 March 2001
`
`Outstanding questions
`
`MOGisonly oneofthe possible target antigens for
`antibody-mediated demyelination. More precise
`analysis ofdemyelinating anti-MOGantibodies, and the
`identification ofother potential targets ofpathogenic
`autoantibodies, should leadto tools that allow a more
`detailed categorization ofMS patient subsets.
`
`+ What immunological and neurobiological mechanisms underlie the
`heterogenouspatterns of MS pathology?
`+ Which are the mostsuitable clinical and paraclinical markers to define
`patient subgroupswith different pathogenetic pathwaysin the
`formation of demyelinated plaques?
`Do patients with different pathogenetic pathways of demyelination
`require
`subtype-specific therapy?
`+ Arerecent therapies developed against specific mechanisms of
`demyelination or axonal injury effective in MS patients?
`
`+
`
`Consequencesof MS therapy
`Demyelination in MS develops by a T-cell-driven
`inflammatory process. Thus, the primary nature of
`inflammation is undisputed and will remain central
`for treatments that modulate the immune system®.
`There are, however, several aspects that limit the
`therapeuticefficacy ofstrategies directed exclusively
`against the inflammatory componentofthe disease.
`spectroscopy. These techniques focus on two major
`Currently, immune suppression is unable to stop the
`issues: determining the imagingcorrelate of basic
`inflammatory reaction in the CNS and immune
`histopathological MS features, such as
`
`inflammation, demyelination—remyelination,gliosis modulatory regimesusing interferonfor co-
`and axonalloss; and identifying different
`polymerI decrease,but do not abolish, inflammation.
`demyelination pathways. From their pathological
`It is not possible to intervene morespecifically in the
`appearance”,it is predicted that pattern I and II
`inflammatory process becauseneitherthetrigger of
`inflammation (virus induced versus autoimmunity)
`lesions, which reflect classical autoimmunity, will be
`characterized by their sharp borders, and early and
`northe specific target antigen in the CNSof affected
`pronounced contrast enhancementasa reflection of
`patients is known.It should also be rememberedthat,
`damageto the blood-brain barrier. By contrast,
`in MS,evidence that inflammation is driven by a Th1
`lesions that involve oligodendrocytes (patterns III
`responseis circumstantial, and a role for otherT cells,
`and IV) expand morediffusely into the white matter.
`for example a response mediatedbyclass-I-restricted
`Thus,these lesions have ill-defined borders and less
`cells,is likely.
`pronouncedanddelayed blood-brain barrier
`These uncertainties indicate the importance of
`damage. Lesions with these features have been
`identifying supplementary therapeutic strategies to
`described in subgroupsof MSpatients*3 and
`prevent demyelination and tissue destruction in MS.
`preliminary evidence indicates that they correlate
`Possible approachesinclude blocking macrophage
`with the respective pathological patterns.
`responses orspecific macrophagetoxins, the elimination
`Cases ofMSin which antibody-mediated
`ofspecific demyelinating antibodies, neuroprotective
`demyelination occurs could, additionally, be identified
`therapies to prevent axonal injury or the consequences
`by the presence ofserum-demyelinating antibodies.
`ofbrain ischemia, and stimulation of remyelination.
`Such antibodies have been found in the serum ofa
`Although attractive conceptually, these strategies have
`subsetofMSpatients**. Oneofthe major targets of
`so far failed. It is possible thatthis is because ofthe
`demyelinating antibodies is MOG.Although anti-MOG
`heterogeneity in the pathogenetic mechanismsthat
`leads to the formation ofdemyelinated plaques in this
`antibodies occur in serum andcerebrospinal fluid ofMS
`disease, which we havedescribed earlier.
`patients,a classification ofMS patient subsets on the
`basis oftheir presence or absenceis notyet possible**36,
`Thus, a major challenge for MS researchis to
`An additional complication is that these antibodies are
`develop paraclinical markers that identify the
`also foundin control patients without MS andin
`heterogenous pathogenetic components involved in
`patients with other neurological diseases. Furthermore,
`the formation of MSplaquesin individuals at
`not all anti-MOG antibodies induce demyelination
`different stagesof their disease. This could lead to
`in vivoandin vitro. most anti-MOGantibodies
`thestratification of MS patients into smaller
`recognize regions ofMOGthat are not exposed on the
`subgroups with common,defined mechanismsof
`inflammation, demyelination and tissue damage,
`extracellular surface ofoligodendrocytes and are
`therefore not targets for antibodies in vivo. Lastly,
`and to subtype-specific therapy.
`
`Acknowledgements
`Theseprojects were
`funded bygrants of the
`Bundesministerium far
`Bildung, Wissenschaft
`und Kunst (GZ 70.056/
`2-Pr/4/99), the
`Gemeinnitzige Hertie
`Stiftung (GHS 2/540/99)
`and the US Multiple
`Sclerosis Society
`(RG 3051-A-1).
`
`References
`1 Lassmann,H.(1998) Multiple sclerosis pathology.
`In McAlpine’s Multiple Sclerosis (3rd edn)
`(Compston,A., ed.), Churchill Livingstone
`2 Prineas, J.W. (1985) The neuropathology of
`multiple sclerosis. In Handbook ofClinical
`Neurology (Vol 47. Demyelinating diseases)
`(Koetsier, J.C., ed.), pp. 337-395, Elsevier Science
`3 Kornek, B. and Lassmann,H.(1999) Axonal
`pathology in multiple sclerosis: a historical note.
`Brain Pathol. 9,651-656
`
`http://tmm.trends.com
`
`4