Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`131
`
`Immunomodulation in multiple
`sclerosis: from immunosuppression
`to neuroprotection
`Oliver Neuhaus1, Juan J. Archelos2 and Hans-Peter Hartung1
`
`1Department of Neurology, Heinrich-Heine-Universita¨ t, Moorenstrasse 5, 40225 Du¨ sseldorf, Germany
`2Multiple Sclerosis Research Group, Department of Neurology, Karl-Franzens-Universita¨ t, Auenbruggerplatz 22, 8036 Graz, Austria
`
`Multiple sclerosis (MS) is the most common disabling
`neurological disease of young adulthood. Following
`advances in the understanding of the immunological
`mechanisms that underlie the pathogenesis of MS, a
`growing arsenal of immunomodulatory agents is avail-
`able. Two classes of immunomodulators are approved
`for long-term treatment of MS, the efficacy of several
`promising new concepts is being tested in clinical trials
`and classical immunosuppressive agents used in MS
`treatment have been shown to exert specific, immuno-
`modulatory effects. Furthermore, two recent obser-
`vations have changed our basic understanding of the
`pathogenesis of MS. First, immune cells in MS lesions
`have neuroprotective activity, which indicates a bene-
`ficial role of neuroinflammation. Second, there is evi-
`dence that axonal
`loss, rather than demyelination,
`underlies the progression of MS and, hence, constitutes
`a therapeutic target.
`
`‘The surest way to lose a reputation in neurology is to
`advocate a treatment for multiple sclerosis.’ (H. Houston
`Merritt)
`
`Multiple sclerosis (MS) is the most common inflamma-
`tory disorder of the CNS and the leading cause of
`neurological disability in young adults [1]. Many immune
`abnormalities have been described in MS, which indicates
`that the immune system plays a central role in its
`pathogenesis [2– 4]. Although immune responses contrib-
`ute to the formation and maintenance of MS lesions [5],
`neuroinflammation might have neuroprotective effects
`[6,7]. This crucial role of the immune system in disease
`pathogenesis has important therapeutic implications. For
`a long while corticosteroids were the only proven therapy
`for MS. However, these only shorten an acute attack and
`effective, long-term drug treatment was not available.
`Although several immunosuppressive agents (i.e. inhibi-
`tors of crucial components of the immune system that
`cause generalized immune dysfunction) were used off-
`label, the adverse systemic effects, such as increased risk
`of cancer and infection, limited the potential benefits in
`MS. More recently, two classes of immunomodulatory
`agents, interferon b (IFN-b) and glatiramer acetate (GA),
`
`Corresponding author: Oliver Neuhaus (oliver.neuhaus@uni-duesseldorf.de).
`
`have been approved for the treatment of MS [8 – 10].
`Immunomodulators, which do not cause general suppres-
`sion of the immune system, shift immune responses from
`pro-inflammatory autoimmune conditions [mediated by
`T helper 1 (Th1) cytokines that are released by auto-
`reactive T cells] towards more beneficial anti-inflamma-
`tory circumstances (mediated through Th2 cytokines that
`are secreted by regulatory T cells). Both IFN-b and GA
`have been proven to be partially effective in clinical trials
`[1,11 – 14]. In the search for more efficacious agents, many
`new drugs are under investigation in preclinical and
`clinical trials, but several promising approaches have
`failed [15]. In parallel there have been advances in
`understanding the underlying pathogenesis of the disease
`as well as modes of action of the different agents.
`Here, we summarize current concepts about the mech-
`anisms of action of therapies already approved for MS and
`the most promising future candidates.
`
`Disease-relevant immune processes
`Since the first description of MS in 1835 by J. Cruveilhier
`as ‘scle´rose en taches, en ıˆles par masses disse´mine´es’ [16],
`the concepts of
`its pathogenesis have been adapted
`continuously [1,2,4,17]. Although unproven, the current
`consensus is that MS pathogenesis comprises an initial
`inflammatory phase, which fulfils the criteria for an
`autoimmune disease [18], followed by a phase of selective
`demyelination and last, a neurodegenerative phase [4,17].
`Subjects with genetically determined susceptibility to MS
`[19] harbor T cells that react with CNS autoantigens.
`Although these can remain dormant for decades, at some
`point they are activated in the periphery, probably by
`molecular mimicry (i.e. recognition of epitopes that are com-
`mon to autoantigens and microbial antigens as exogenous
`triggers [20,21]). This enables them to migrate through the
`blood– brain barrier to the brain and spinal cord. Reacti-
`þ
`helper or
`vated in the CNS, these T cells of either CD4
`þ
`CD8
`cytotoxic phenotype [22] release pro-inflammatory
`Th1 cytokines and orchestrate the destruction of the
`myelin sheath by various types of immune cells. Destruc-
`tion follows the first two of four pathological patterns [5]:
`(1) T-cell- and macrophage-mediated demyelination; (2)
`antibody-mediated demyelination that involves comple-
`ment activation [3,23]; (3) distal oligodendrogliopathy and
`
`http://tips.trends.com 0165-6147/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0165-6147(03)00028-2
`1
`
`Hopewell EX1007
`
`

`

`132
`
`Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`Fig. 1. Pro-inflammatory T cells in the periphery are activated when foreign antigens (Ags) and self-antigens that are presented on major histocompatibility complex class II
`(MHC-II) by Ag-presenting cells, such as macrophages, bind to T-cell receptors (TCRs). Activated T cells migrate to, adhere at and penetrate through the blood– brain bar-
`rier, steps that are mediated by adhesion molecules, proteases and chemokines. In the CNS, the T cells are reactivated by CNS Ags presented on MHC-II by other Ag-pre-
`senting cells, predominantly microglial cells. The reactivated T cells secrete pro-inflammatory cytokines, such as interferon g (IFN-g) or interleukin 2 (IL-2), which induce
`CNS inflammation by subsequent activation of macrophages, other T cells and B cells. Macrophages and T cells attack the myelin sheath of oligodendrocytes (OGs) by
`cytotoxic mediators, mainly tumor necrosis factor a (TNF-a), O2 radicals and nitric oxide (NO). B cells differentiate into plasma cells. These secrete demyelinating antibodies
`that can guide and activate macrophages, and ignite the complement cascade, which causes assembly of the membrane attack complex and causes pore formation in
`myelin membranes. Demyelination occurs by four different pathological patterns (1–4), as described in the main text.
`
`oligodendrocyte apoptosis; and (4) primary oligodendrocyte
`degeneration. The mechanisms of the latter two patterns
`remain elusive.
`In addition to this autoaggressive inflammatory phase,
`axonal loss, which causes irreversible disability, occurs
`early in the course of the disease [24,25]. It is unclear
`whether axonal damage is the consequence of a primary
`active destructive process executed by,
`for example,
`macrophages and cytotoxic molecules derived from CD8
`cells [26], or a (patho)physiological response that occurs
`secondarily to demyelination and is based on increased
`vulnerability [24,27]. Axonal damage appears to be
`initiated by increased membrane permeability followed
`by enhanced Ca2þ
`influx. Disruption of axonal transport
`alters the cytoskeleton and leads to axonal swelling,
`lobulation and, finally, disconnection [28].
`The different molecules involved in each phase of MS
`are summarized in Fig. 1 [1,17,29].
`
`MS therapeutics: immunomodulatory profile in vitro and
`in vivo
`The structural features of the therapeutic agents are
`shown in Fig. 2. Interferon b1a (IFN-b1a), IFN-b1b and
`glatiramer acetate (GA) are approved for the treatment of
`relapsing-remitting (RR) MS, IFN-b1b for secondary-
`progressive (SP) MS, IFN-b1a for SP MS with super-
`imposed relapses and mitoxantrone for worsening forms of
`RR and SP MS.
`
`Glucocorticosteroids
`Methylprednisolone and prednisolone are the mainstays of
`treatment for acute attacks in MS [30]. Because of side-
`effects in long-term treatment regimens and superior
`efficacy compared with oral application, steroids are
`mostly delivered in intravenous pulses. Most of the
`immunological effects of glucocorticosteroids are mediated
`by specific, ubiquitously distributed intracellular recep-
`tors. These form a multiprotein complex with two
`
`http://tips.trends.com
`
`2
`
`molecules of heat shock protein 90 (HSP90), one molecule
`of HSP70, one molecule of HSP56, one so-called immuno-
`philin, and other, less well characterized, proteins. After
`binding lipophilic glucocorticosteroid, the receptor dis-
`sociates irreversibly from the rest of the complex, and the
`steroid – receptor complex ligates specific glucocorticoid-
`responsive elements in nuclear DNA. This either upregu-
`lates or downregulates the transcription of target genes.
`The nuclear factor kB (NF-kB) transcription factor induces
`the expression of multiple inflammatory and immune
`genes [31]. Glucocorticosteroids inhibit NF-kB, both by
`direct binding of the activated receptor to NF-kB and by
`inducing expression of the specific inhibitory protein IkB
`in lymphocytes.
`The immunological effects of glucocorticosteroids
`include: (1) inhibition of T-cell activation and production
`of pro-inflammatory cytokines, such as interleukin 2 (IL-2)
`and IFN-g; (2) increased production of anti-inflammatory
`Th2 cytokines by T cells; (3) inhibition of IFN-g-induced
`major histocompatibility complex (MHC) class II expres-
`sion on macrophages; (4) decreased production of pro-
`inflammatory cytokines, prostaglandins and leukotrienes
`by macrophages; (5) diminished adhesion of neutrophils to
`endothelial cells; and (6) inhibition of endothelial-cell
`activation, and expression of MHC class II and adhesion
`molecules. Taken together, these actions underlie the
`sealing effect of steroids on the blood– brain barrier, which
`prevents further access of immune cells and molecules to
`the brain.
`In addition to the genomic effects mediated by steroid
`receptor activation, there is recent evidence that high
`doses of glucocorticosteroids induce apoptosis of target
`cells by a direct, non-genomic effect [32]. In this case,
`apoptosis is thought to be mediated by a direct effect on
`cellular membranes, which influences transmembranous
`ion transport and, subsequently, reduces the availability of
`ATP. Steroid-induced apoptosis of autoreactive T cells is
`important for the termination of an acute MS attack [32].
`
`

`

`Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`133
`
`IFN-b
`IFN-b is highly species specific. In humans, the IFN-b
`polypeptide is produced and secreted by fibroblasts, but
`virtually all mammalian cells can produce IFN-b on stimu-
`lation. Two recombinant IFN-b preparations, IFN-b1a and
`IFN-b1b, are approved for treatment of MS. The optimal
`dosage and route of administration (subcutaneously
`versus intramuscularly), the resulting pharmacodynamic
`properties of IFN-b and the role of neutralizing antibodies
`are controversial [33 – 37].
`IFN-b has immunomodulatory properties, antiviral and
`anti-proliferative effects, and promotes cell differentiation
`[38]. Although the mechanisms of action of IFN-b are not
`fully understood, there is agreement that the major effects
`are mediated by activation of a transmembrane IFN
`receptor, which leads to either the upregulation or down-
`regulation of target genes [39,40]. Unlike IFN-g, the ‘type
`II interferon’, IFN-b and IFN-a (type I interferons) share
`a receptor, which consists of two chains, IFN-aR1 and
`IFN-aR2, or several subvariants. Binding of IFN-b (and
`other type I interferons) to the extracellular domain of the
`receptor induces an intracellular signal transduction
`cascade that involves: (1) recruitment and activation of
`the cytoplasmic tyrosine kinase 2 by IFN-aR1, and Janus
`kinase 1 by IFN-aR2; (2) subsequent phosphorylation and
`recruitment of signal transducers and activators of tran-
`scription (STAT1 and STAT2) to form a STAT1– STAT2
`heterodimer; (3) migration of the STAT1– STAT2 hetero-
`dimer to the nucleus; (4) association of STAT1 –STAT2
`with the p48 protein, to form the active ‘IFN-stimulated
`gene factor 3’; (5) binding of IFN-stimulated gene factor 3
`to promoter elements and initiation of the transcription
`of target genes. The variations in this rough scheme
`of signal transduction that lead to different effects of IFN-b
`in different target cells are based on variations in each
`of the steps outlined above, which are not yet fully
`understood [39].
`The range of immune effects attributed to IFN-b is wide
`and ever broadening [38,41]. It suppresses T-cell pro-
`liferation, diminishes IFN-g-induced upregulation of MHC
`class II expression, induces the production of Th2 cyto-
`kines and reduces synthesis of Th1 cytokines, and inhibits
`monocyte activation. In addition, IFN-b downregulates
`matrix metalloproteinases (MMPs), decreases surface-
`expressed adhesion molecules and increases the release
`of soluble adhesion molecules, which combine to reduce the
`migratory potential of T cells.
`
`GA
`GA is the acetate salt of a standardized, randomized
`mixture of synthetic polypeptides. After subcutaneous
`administration, GA is quickly degraded to free amino acids
`and small oligopeptides and, thus, most probably initiates
`its major immunological effects in the periphery. Unlike
`the multiple immunological effects of IFN-b, which are
`antigen nonspecific, the immunomodulatory potential of
`GA is based on immune cells that are specific for myelin
`basic protein (MBP) and, probably, other myelin antigens
`[38,42]. Four major mechanisms of GA activity have been
`identified: (1) competition between GA and MBP for
`binding to MHC molecules;
`(2) competition between
`
`http://tips.trends.com
`
`3
`
`GA – MHC complexes and MBP – MHC complexes for bind-
`ing to the T-cell receptor (TCR); (3) activation and toler-
`ance induction in MBP-specific T cells through an altered
`peptide ligand; and (4) induction of GA-reactive, Th2-like
`regulatory cells that mediate local bystander suppression.
`Whereas the first two effects are thought to take
`place only in vitro, the latter two are also likely to occur
`in vivo and could contribute to the anti-inflammatory
`effects of GA [42].
`
`Intravenous immunoglobulins
`Intravenous immunoglobulins (IVIgs) are pooled, purified,
`human Igs with virtually unlimited specificities. Consist-
`ent with their composition, several mechanisms of action
`of IVIgs have been suggested [43,44]. These include:
`(1) anti-idiotype antibodies (binding and inactivation of
`pathogenic antibodies by IVIgs);
`(2) blockade of Fc
`receptors on mononuclear phagocytes; (3) downregulation
`of the endogenous production of Igs; (4) attenuation and
`abrogation of complement-mediated effects (partially by
`‘consumption’ of complement components); (5) neutraliz-
`ation of molecules (TCR, MHC, costimulatory molecules
`and cytokines)
`that are involved in inflammation;
`(6) induction of anti-inflammatory cytokines; and (7) induc-
`tion of apoptosis. Experimental evidence indicates that
`IVIgs might also be involved in myelin repair, but clinical
`proof of this is lacking [44]. Currently, the role of IVIgs in
`treating MS is undecided.
`
`Immunosuppressive agents
`Mitoxantrone is effective in the treatment of severe active
`forms of MS [45,46]. Although mitoxantrone suppresses
`both T cells and B cells, in vitro experiments indicate
`that major sites of action in MS are antigen-presenting
`cells, which are induced to undergo apoptosis, and macro-
`phages, the major effector cells of demyelination, which
`are deactivated [47].
`Azathioprine is widely used in organ transplantation
`and autoimmune disorders [48] and is considered a second-
`line drug in MS. It mainly targets the activation, pro-
`liferation and differentiation of both T cells and B cells by
`competition between its metabolites and DNA nucleotides.
`Specific immunomodulatory properties have not been
`reported to date.
`Cyclophosphamide is used to treat severe and rapidly
`progressive forms of MS, although evidence of its efficacy
`in clinical trials is conflicting. In addition to strong
`immunosuppression, cyclophosphamide exerts immuno-
`modulatory effects that shift immune responses from Th1
`towards Th2 by an unknown mechanism [49].
`One study has shown methotrexate to convey some
`therapeutic benefit in progressive forms of MS [10], but its
`mechanisms of action in autoimmune diseases are largely
`unknown.
`
`Potential new agents
`Anti-adhesion molecules
`Adhesion molecules comprise several families of mol-
`ecules that are essential in virtually all cellular inter-
`actions of immune cells [50,51]. The selectins (e.g. L-selectin,
`E-selectin and P-selectin), which have a lectin-binding
`
`

`

`134
`
`Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`(g) Cyclophosphamide
`
`Cl
`
`H2C
`
`O
`
`O
`
`CH2
`N
`
`P
`
`NH
`
`CH2
`H2C Cl
`
`279.1 Da
`
`CH2NCH3
`
`NN
`
`(h) Methotrexate
`
`H2N
`
`N
`
`N
`
`NH2
`
`HOOC
`
`HOOC
`CH2 CH2
`CH NH CO
`454.4 Da
`
`(i) Anti-adhesion molecule antibodies
`150 kDa
`
`Murine
`Human
`
`(j) Statins
`
`150 kDa
`
`O
`
`O
`
`O
`
`O
`
`(a) Glucocorticosteroids
`
`HO
`CH3
`
`H
`
`CH3
`
`H
`
`H
`
`O
`
`CH2OH
`C
`O
`OH
`
`CH3
`Methylprednisolone (374.5 Da)
`(b) Interferon β (IFN-β)
`Cys31
`
`Cys141
`
`Cys17
`
`A
`
`B
`
`C
`
`D
`
`E
`
`NH2
`Met1
`
`Asn80
`
`COOH
`Asn166
`Sugar
`residues
`Disulfide bridge:
`Cys31–141
`
`IFN-β1a
`(22.5 kDa)
`
`Cys31
`
`Cys141
`
`Ser17
`
`A
`
`B
`
`C
`
`D
`
`E
`
`NH2
`Ser2
`
`IFN-β1b (18.5 kDa)
`(c) Glatiramer acetate (GA)
`
`COOH
`Asn166
`
`O
`
`H
`
`Simvastatin (418.6 Da)
`
`(k) Neurotrophic factors
`
`NH2
`
`COOH
`
`BDNF
`(13.5 kDa)
`
`Disulfide bridges:
`Cys13–80; Cys58–109; Cys68–111
`
`(l) Neuroprotective agents
`
`NH2
`
`NS
`
`O
`
`CF3
`
`Riluzole (234.2 Da)
`
`TRENDS in Pharmacological Sciences
`
`Lys Ala
`Ala
`Lys
`Ala
`Lys
`Glu
`Ala
`Lys
`Tyr
`
`Ala
`Ala
`Ala Tyr
`Glu
`Ala
`Ala
`Lys
`Lys
`Glu
`Ala
`Ala Ala
`Lys
`Tyr
`Ala Ala
`Lys
`Lys Lys
`Glu
`Ala
`Ala Ala Ala
`
`Lys Lys
`Ala Ala Ala
`Lys
`Lys Lys
`Tyr
`Glu
`Ala Ala
`Ala
`Lys
`Lys
`Glu
`Ala
`Ala
`Lys
`Tyr
`Glu
`Ala
`Lys
`Lys
`Lys
`
`Glu
`Ala
`Tyr
`Glu
`Ala
`
`Lys
`Ala
`Lys
`Glu
`Ala
`Lys
`Lys Lys
`Ala
`Lys
`
`4.7–11.0 kDa
`
`(d) Intravenous immunoglobulins (IVIgs)
`
`Fab fragment
`
`Hinge region
`
`Fc fragment
`
`CL
`
`VL
`
`CH1
`
`VH
`
`CH2
`CH3
`
`150 kDa
`
`Light chain
`Heavy chain
`
`Variable region
`Constant region
`
`Disulfide bridge
`
`(e) Mitoxantrone
`
`OH
`
`O
`
`H
`
`N
`
`CH2
`
`CH2 NH CH2 CH2 OH
`
`2 HCI
`
`OH O
`
`H
`
`N CH2
`
`CH2 NH CH2 CH2 OH
`517.4 Da
`
`N
`
`HN
`
`O
`
`N
`
`H3C
`
`O−
`
`N+
`
`S
`
`N
`
`N
`
`N
`277.3 Da
`
`(f) Azathioprine
`
`http://tips.trends.com
`
`4
`
`

`

`Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`135
`
`domain that binds to glycosylated and sialylated ligands,
`are involved in the ‘rolling’ of
`leukocytes. Integrins
`comprise a4b1, a5b1, a6b1 integrins [also known as
`very late antigen 4 (VLA-4) VLA-5 and VLA-6, respec-
`tively] and leukocyte function antigen 1 (LFA-1). VLA-4
`and LFA-1 are present on circulating leukocytes and
`mediate their migration across the vascular endothelium;
`in addition, LFA-1 is a costimulatory molecule in T-cell
`activation. Cadherins (including classical cadherins,
`desmosomal cadherins and protocadherins) form molecu-
`lar links between adjacent cells by zipper-like structures.
`Members of the Ig superfamily, such as intercellular
`adhesion molecule 1 (ICAM-1), ICAM-2 and ICAM-3,
`vascular cell adhesion molecule (VCAM), and LFA-2 and
`LFA-3 have Ig-like domains. ICAM-1 and VCAM-1 are the
`counter-receptors of LFA-1 and VLA-4, respectively, and
`are thus involved in leukocyte migration.
`Antibodies against single adhesion molecules can
`potently inhibit crucial steps in the pathogenesis of MS,
`especially lymphocyte migration. Currently, the most
`promising candidate is natalizumab [52], a humanized
`monoclonal antibody against the a4 chain of a4b1 integrin
`that has been effective in Phase II clinical trials [53].
`‘Humanized’ means that a murine antibody clone has been
`grafted to a human IgG4 framework at the complementary
`determining region to reduce its immunogenicity [52]. Two
`large-scale, Phase III clinical trials in RR MS have been
`initiated recently and small-molecule antagonists, which
`offer advantages over monoclonal antibodies (oral avail-
`ability and lack of antigenicity) will certainly be explored
`in the future [54].
`
`Statins
`Statins are effective lipid-lowering agents. Recent findings
`indicate that they have additional immunomodulatory
`effects in vivo in the animal model experimental auto-
`immune encephalomyelitis (EAE), and in vitro [55– 57].
`One proposed immunomodulatory mechanism of statins is
`based on the selective inhibition of the adhesion molecule
`LFA-1, an integrin that is involved in inflammation [58].
`Furthermore, statins reduce IFN-g-induced MHC class II
`expression by blocking transcription of the class II
`
`transactivator IV promoter [55]. In addition, statins
`curtail T-cell proliferation, lower expression of activation
`surface markers and induce production of the cytokine
`IL-4 [56]. Currently, simvastatin is being tested in a Phase
`II clinical trial in MS. The oral administration, extensive
`safety data, possible effects synergistic with IFN-b and
`simultaneous treatment of co-morbidity make statins
`particularly attractive candidate agents.
`
`Neurotrophic factors
`Neurotrophic factors are secreted proteins that regulate
`the survival and differentiation of nerve cells [59]. They act
`via specific neurotrophin receptors [60]. Neurotrophic
`factors have been observed to shift the CNS cytokine
`balance from Th1 to Th2 by an undefined mechanism [61].
`In addition, they might promote survival of neurons in MS
`lesions [6] and pharmacological neuroprotection by the
`exogenous application of neurotrophic factors provides
`a promising therapeutic approach [59,62]. Insulin-like
`growth factor 1 is currently explored in a Phase I/II study.
`Other attractive candidates are brain-derived neuro-
`trophic factor, glial growth factor and ciliary neurotrophic
`factor [63].
`
`Neuroprotective agents
`Recent evidence indicates that axonal and neuronal
`degeneration occur early in MS and – as the disease
`evolves – predominate the underlying pathogenetic
`mechanisms [4,17,24]. This paradigm shift has obvious
`therapeutic implications. In addition to neurotrophic
`factors, other chemically defined neuroprotective agents
`that save neurons from toxic stress [64], such as riluzole (a
`þ
`channel activator used in amyotrophic lateral
`potent K
`sclerosis) are either being investigated or are likely to be
`tested in clinical trials.
`
`Similarities and peculiarities
`Given the complexity of MS pathogenesis, there are
`multiple sites where immunomodulatory agents, either
`alone or in combination, might be effective. Current
`knowledge of the sites of action obtained from in vitro
`and in vivo data is summarized in Fig. 3. Although the
`
`Fig. 2. Structures of immunomodulators. Interferon b (IFN-b), glatiramer acetate (GA) and mitoxantrone are currently approved for use in multiple sclerosis (MS). (a) Gluco-
`corticosteroids are derived from cortisol, a naturally occurring adrenal hormone. Methylprednisolone is commonly used to treat acute relapses in MS. (b) IFN-b contains
`166 residues that form five a-helices (A –E). Two preparations of recombinant IFN-b are approved for MS treatment. IFN-b1a is produced in Chinese hamster ovary cells and
`is pharmacologically identical to the natural form (i.e. it is glycosylated by oligosaccharides at Asn80). Compared with natural IFN-b, IFN-b1b is not glycosylated, it lacks
`Met1 (i.e. IFN-b1b has 165 residues) and there is a Cys to Ser substitution at residue 17. In both forms of recombinant IFN-b, there is a disulfide bridge between Cys31 and
`Cys141. Structurally important amino acids are shown in green. (c) GA is the acetate salt of a standardized, randomized mixture of synthetic polypeptides (average length
`45 –100 amino acids) that consist of L-glutamic acid, L-lysine, L-alanine and L-tyrosine in the molar ratio of 0.14:0.34:0.43:0.09. An example sequence is shown. (d) Intra-
`venous immunoglobulins (IVIgs) are pooled, purified, human Igs prepared by cold ethanol fractionation of human plasma derived from 3000 –10 000 donors. Igs share a
`common Y-shaped structure, which consists of two heavy chains (50 , hsp sp ¼ 0.25 . kDa each) and two light chains (25 , hsp sp ¼ 0.25 . kDa each) connected by disul-
`fide bridges. The variable region is responsible for antigen recognition. The five heavy-chain isotypes determine the immunoglobulin classes. IVIgs contain , 95% IgG,
`2.5% IgA and a minority of IgM. (e) Mitoxantrone is an anthracenedione derivative that is related to the anthracyclins doxorubicine and daunorubicine. It interacts with
`topoisomerase-2 and causes single and double string breaks by intercalating with DNA. (f) Azathioprine is a purine analogue that is metabolized rapidly to the cytotoxic
`and immunosuppressant derivatives 6-mercaptopurine and thioinosine acid, the latter competes with DNA nucleotides. (g) Cyclophosphamide is an alkylating agent of the
`nitrogen mustard group; its active metabolites, formed by the activity of hepatic cytochrome P450, induce DNA-string breaks. (h) Methotrexate interferes with DNA syn-
`thesis by inhibiting dihydrofolate reductase and, thus, thymidine biosynthesis. It reduces 1-carbon transfers to purines. (i) Monoclonal antibodies against adhesion mol-
`ecules inhibit homing of T cells to the CNS. To reduce immunogenicity, newer approaches use ‘humanized’ antibodies, which are chimeras of murine variable regions and
`human constant regions connected at the complementary determining region. (j) Statins inhibit 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase. Statins are
`effective lipid-lowering agents because HMG-CoA reductase is essential to the cholesterol biosynthesis pathway, but they also have immunomodulatory properties. (k) The
`structure of neurotrophic factors is dominated by four antiparallel pairs of b strands (green arrows) held in place by three disulfide bridges. Neurotrophic factors are active
`as homodimers and act via specific neurotrophin receptors. Brain-derived neurotrophic factor (BDNF) is shown, with the position of the cysteine residues that form the dis-
`ulfide bridges in green. (l) Neuroprotective agents comprise a heterogeneous group that is characterized by their ability to exert protection from neurotoxic stress. Riluzole
`þ
`acts by activating K
`channels.
`
`http://tips.trends.com
`
`5
`
`

`

`136
`
`Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`(c) T-cell migration, adhesion and penetration
`Steroids, IFN-β, IVIgs, anti-adhesion molecule
`antibodies, statins
`
`(e) Release of pro-inflammatory cytokines
`Steroids, IFN-β, IVIgs
`
`(h) Effector cell activation: macrophages
`Steroids, IFN-β, mitoxantrone
`
`(d) T-cell reactivation
`IFN-β, IVIgs, azathioprine
`
`(i) Effector cell activation: T cells
`Steroids, IFN-β, IVIgs, azathioprine
`
`(k) Axonal damage
`Neurotrophins, neuroprotectíve agents
`
`Central nervous system
`
`Reactivation
`
`Inflammation
`
`Demyelination
`
`Axonal damage
`
`(a) Antigen-presenting cell
`Steroids, IFN-β, IVIgs, mitoxantrone,
`statins
`
`Periphery
`
`Activation
`
`Migration,
`adhesion,
`penetration
`
`Antigen-
`presenting
`cell
`
`Autoreactive
`T cell
`
`Th1
`
`(b) Autoreactive T cell
`IFN-β, GA, mitoxantrone,
`azathioprine, cyclophosphamide,
`methotrexate, statins
`
`Bystander
`suppression
`effect
`
`(g)
`
`GA
`
`TCR
`MHC-II
`
`Th2
`
`GA-reactive
`regulatory T cell
`
`IL-4
`CNS Ag
`
`TCR
`MHC-II
`
`Effector cells
`
`Target cells
`
`(f) Cytokine shift
`Th1 Th2
`Steroids, IFN-β,
`GA-reactive T cells,
`cyclophosphamide,
`statins, neurotrophic
`factors
`
`(l) Demyelination
`• Remyelination as neuroconstructive approach?
`• Stem cells as source?
`• Schwann cell preparations?
`
`(j) Effector cell activation: B cells,
`plasma cells and demyelinating antibodies
`IVIgs, mitoxantrone, azathioprine
`
`TRENDS in Pharmacological Sciences
`
`Fig. 3. Immunomodulatory agents have common sites of action in multiple sclerosis (MS) and interfere at other sites individually. (a) Antigen presentation by antigen-pre-
`senting cells is inhibited. (b) Proliferation of autoreactive T cells is blocked or tolerance is induced. (c) T-cell migration, adhesion and penetration into the CNS are abro-
`gated. (d) T-cell reactivation in the CNS is blocked. (e) Release of pro-inflammatory cytokines by autoreactive T cells is diminished. (f) Several agents induce a shift from
`pro-inflammatory Th1 to anti-inflammatory Th2 cytokines. (g) Uniquely, GA induces regulatory T cells that cross-react to both GA and CNS Ags. After reactivation inside
`the CNS, these regulatory T cells exert a local bystander suppression effect through the release of Th2 cytokines. (h) Activation of macrophages or secretion of pro-inflam-
`matory mediators are curtailed. (i) Apoptosis of autoreactive T cells is induced. (j) Activation of B cells and their differentiation to plasma cells is blocked; demyelinating
`antibodies are neutralized by several mechanisms. (k) Axonal damage is reduced by neuroprotective approaches. (l) Therapeutical approaches that target demyelination
`are still at an experimental stage. Abbreviations: Ag, antigen; GA, glatiramer acetate; IFN, interferon; IL, interleukin; IVIgs, intravenous immunoglobulins; MHC, major histo-
`compatibility complex; TCR, T-cell receptor; Th1, T helper 1.
`
`immunomodulatory agents presented share some common
`features, many differ markedly in their mechanisms and
`sites of action. General properties to be considered are
`nonspecific versus specific antigen effects (e.g. IFN-b
`versus GA), polydirectional versus unidirectional immuno-
`modulatory properties (e.g. the pleiotropic effects of IFN-b
`versus
`the monoclonal antibody directed against
`a4-integrin)
`and
`pure
`immunomodulators
`versus
`immunosuppressants with additional immunomodulatory
`properties (e.g. IFN-b versus mitoxantrone).
`The interactions between the various drugs are still
`widely unknown [65,66]. However, gaining more knowl-
`edge of the individual targets might allow combination
`therapies to be developed that have either additive or
`synergistic effects.
`
`Concluding remarks
`Although an increasing arsenal of immunomodulatory
`agents is available, many questions remain about each
`single approach [67]. These include when to initiate and
`when to stop treatment, the optimal dose, frequency and
`route of administration, the long-term effects of treatment,
`the occurrence and relevance of neutralizing antibodies to
`biological agents, and the cost utility.
`
`http://tips.trends.com
`
`6
`
`It is likely that differential therapy in MS will gain in
`importance in the near future. As evidence mounts that
`MS comprises several distinct subforms [5], the choice of
`treatment for individual patients will, ideally, be determined
`by the knowledge of the specific underlying pathomechan-
`ism and the respective optimal drug or combination of drugs.
`The goal is tailor-made immunotherapy.
`Recent evidence indicates that autoimmune cells have
`neuroprotective properties that are mediated, at least in
`part, by neurotrophic factors [6,7]. Based on this, a novel
`therapeutic approach would be to ‘import’ neurotrophic
`factors into the CNS via immune cells [68]. However, these
`findings also provide an impetus to shift the rationale of
`nonselective immunosuppressive therapies towards a more
`selective immunomodulatory regimen, which would pre-
`serve or even enhance the neuroprotective functions of
`autoimmune T cells [6].
`In a contrasting approach, bone marrow transplan-
`tation (the most aggressive strategy to eradicate a
`disordered immune system) is being tested clinically
`[69,70], but this remains controversial [71].
`Taking neurobiological approaches further, the next
`step on the horizon (which is being investigated in
`animal models and in vitro) is neuroregeneration. This
`
`

`

`Review
`
`TRENDS in Pharmacological Sciences Vol.24 No.3 March 2003
`
`137
`
`Box 1. Current clinical trials in multiple sclerosis
`
`http.//www.nationalmssociety.org/pdf/research/agents.pdf
`http.//www.nationalmssociety.org/pdf/research/clinicaltrials.pdf
`
`neuroconstructive therapy encompasses inhibition of scar
`formation and promotion of axonal regrowth and remye-
`lination [72 – 74]. The contentious issue of the use of
`embryonic stem cells as source of progenitors for remye-
`lination must be discussed in this context. Alternatively,
`retrieval of multipotential neuronal stem cells from the
`adult brain has been suggested [75]. Recently, oligo-
`dendrocyte progenitors have been identified, even in
`chronic MS lesions [76]. Transplantation of peripheral
`Schwann cells is being studied in Phase II clinical trials
`and defined chemical agents for CNS myelination, such as
`eliprodil (a high-affinity ligand of a so-called s-receptor
`that mediates neuroprotection) are being explored experi-
`mentally [77]. Because activated immune cells commonly
`release glutamate, excitotoxicity induced by glutamate is
`thought to be involved in the formation of MS lesions and
`axonal damage. Hence, glutamate receptor antagonists
`are promising candidates [78].