Multiple sclerosis (MS) is an inflammatory demyelinating autoimmune disease of the central nervous system (CNS) that typically manifests with episodes of new/recurrent neurological symptoms (i.e., relapses) followed by either complete remission or residual disability (relapsing–remitting, RR, course) [1]. In 10–15% of cases, progressive disability accrual, irrespective of relapses, occurs from disease onset: these cases are defined as primary progressive (PP) [2]. After an average of 10 to 20 years, most of the patients with RR-onset evolve towards a secondary progressive (SP) course which is characterised by subtle and progressive accumulation of disability independent of new focal inflammation (i.e., relapses and/or inflammatory lesions at magnetic resonance imaging—MRI) [2].

Distinctive clinical and radiological features characterise RR- and SP-MS: (i) RR-MS is characterised by the development of new lesions promoted by adaptive immunity, which may affect different areas of the CNS and are usually associated with the onset of new neurological symptoms congruous with the anatomical area involved (i.e., relapses); (ii) SP-MS is characterised by the persistence of chronic inflammation within the CNS compartment which acts beyond a closed or repaired blood–brain barrier (BBB), i.e., compartmentalisation of the inflammatory response [3]. This latter phenomenon contributes to ongoing demyelination and axonal damage within pre-existent inflammatory lesions and is associated with progressive deterioration in the already impaired functional system(s), with rare occurrence (or absence) of new lesions/symptoms. However, RR- and SP-MS might plausibly be placed at opposite ends of a continuum in the spectrum of disease mechanisms, with the inflammatory and degenerative phenomena underlying MS progression showing complex mutual interactions, each contributing to a variable extent across different stages of the disease [4].

Disease-modifying treatments (DMTs) currently approved for MS mostly target the immune system acting in the peripheral compartment, thus preventing the occurrence of new inflammatory lesions, whereas their potential impact on compartmentalised inflammation is still debated. Furthermore, no DMTs were proven to be effective in halting degenerative phenomena, nor in promoting re-myelination.

2. Insights into MS Pathogenesis: Onset of Autoimmunity

MS was traditionally considered a T-cell-mediated autoimmune disorder, based on preclinical data from animal models of the disease (experimental autoimmune encephalomyelitis—EAE) and evidence for T-cell infiltration in inflammatory lesions and normal-appearing white matter of autoptic and biopsy CNS specimens from affected individuals, with an association between CD8+ T cells number and axonal damage ^{[5][6][7][8][9][10][11][12]}[5,6,7,8,9,10,11,12]. Furthermore, the identification of expanded T-cell clones in the brain parenchyma, cerebrospinal fluid (CSF), and peripheral blood of MS patients detected using T-cell receptor (TCR) analyses reinforced the hypothesis that inflammatory infiltrates were constituted by pathogenetic expanded T-cell clones reactive to myelin antigens ^{[13][14][15][16]}[13,14,15,16]. Circulating CD4+ T cells from MS patients were indeed demonstrated to recognise myelin basic protein (MBP), proteolipid protein (PLP), and myelin oligodendrocyte glycoprotein (MOG), even if the same phenomenon was also observed in healthy individuals; evidence regarding potential differences between these groups in frequency and avidity of cell interactions is conflicting ^[17][18][17,18]. A contribution of B cells to MS pathogenesis was also suggested by preclinical and clinical evidence, and their role was recently re-evaluated with the observation of a remarkable therapeutic effect of B-cell-depleting strategies [19]. Innate immunity cells, including CNS-resident microglia, contribute to MS pathogenesis, and neurodegenerative phenomena possibly, at least in part, independent of inflammation play a role in advanced disease [3]. Although MS aetiology is unknown, several environmental and genetic risk factors were identified ^[20][21][22][20,21,22]. Class II major histocompatibility complex (MHCII) represents the major genetic risk factor, accounting for 20–30% of individual genetic susceptibility ^[22][23][22,23]. MHCII genes encode membrane glycoproteins that are expressed by professional antigen-presenting cells (APC), such as dendritic cells, macrophages, and B cells [24]. The MHCII complex plays a key role in the development of both primary and secondary T CD4-mediated immune response as it presents in its context small peptide antigens processed by professional APC to MHC-restricted T cells [25]. In addition to MHC, several other genes were associated with the risk of developing MS on the basis of a complex genetic background, as confirmed by genome-wide association studies (GWAS) that uncovered more than 200 genetic susceptibility variants which could jointly account for ~48% of the estimated heritability for MS [26]. Enrichment for MS susceptibility loci was mostly related to genes involved in immune system function and regulation, and it was apparent in many different immune cell types and tissues, including microglia, highlighting, overall, the relevance of adaptive and innate immune cells in MS pathogenesis.

2.1. Primary Autoimmune Response in the Peripheral Compartment

The mechanisms which trigger MS are still debated and two plausibly complementary pathogenetic models were proposed to explain the initiation of the autoimmune response, suggesting that the primum movens might take place either in the periphery (“outside-in” model) or within the CNS (“inside-out” model) [27]. The “outside-in” (or CNS-extrinsic/peripheral) model resembles the pathogenetic mechanism underlying EAE in which the disease is induced by external immunisation obtained through the inoculation of myelin-specific antigens in combination with an adjuvant [28]. According to this model, activation and expansion of CNS antigen-specific CD4+ T cells occur in the periphery and may be induced by an encounter with exogenous antigens sharing structural motifs with myelin antigens, hence capable of eliciting an autoimmune response based on molecular mimicry ^[29][30][29,30]. Several infective agents were suggested as potential exogenous triggers of the autoimmune reaction, with Epstein–Barr virus (EBV) the most plausible candidate ^[31][32][31,32]. On the other hand, the “inside-out” model suggests that the autoimmune reaction is triggered by an “internal event” occurring within the CNS and generating myelin debris. CNS-derived soluble antigens may then be drained via lymphatic pathways to peripheral lymphoid organs, where they might be presented to T and B cells eliciting the autoimmune reaction ^[33][34][33,34]. Different hypotheses suggest that the CNS-intrinsic event might be triggered by a CNS viral infection or by primary neurodegeneration [11].

2.2. Secondary Autoimmune Response in the Central Nervous System

Once their priming has occurred, autoreactive T cells might be further expanded in peripheral lymphoid tissues after the encounter with their cognate antigen ^[22][35][22,35]. In the absence of neuroinflammation, T cells patrol the CNS crossing the BBB at the level of CNS post-capillary venules and reaching perivascular or subarachnoid spaces [36]. In such areas, putative cross-reactive T-cell clones may be further activated by CNS antigens (sharing structural motifs with their cognate antigen) which, after being drained from the parenchyma, are processed and presented by tissue-resident APCs. This mechanism promotes a secondary immune response within the CNS characterised by cell proliferation, recruitment of pro-inflammatory cells, and formation of perivascular cuffs around post-capillary and pial venules [37]. Upon activation, adaptive immune cells cross the glia limitans and infiltrate the CNS parenchyma where they produce pro-inflammatory cytokines and chemokines that determine a breakdown of the BBB. This causes further recruitment of adaptive and innate immune cells, which ultimately promote the formation of demyelinating lesions and tissue injury (Figure 1) ^[37][38][37,38]. Further damage to CNS tissue might derive from uncovering and release in the extracellular space of several autoantigens which, in turn, might enhance the autoimmune response in a mechanism defined as epitope spreading [39]. The formation of new inflammatory lesions is frequent in RR-MS, in which subsequent “inflammatory waves” of autoimmune cells invade the CNS, and it is usually associated with the onset of new clinical symptoms (relapses) reflecting the impairment of the area involved. In early MS, relapses usually resolve without sequelae, thanks to functional compensation by the huge number of residual nerve fibres. The number and frequency of relapses occurring shortly after disease onset predict the accumulation of long-term disability and the achievement of pre-defined disability milestones, suggesting that in the RR phase acute inflammation is the main driver of disease activity and tissue damage [40]. Accordingly, disability worsening in RR-MS is mainly due to incomplete recovery from relapses, and no intercurrent disability progression is observed [2]. Acute inflammatory lesions are characterised by predominant demyelination with axonal injury but moderate axonal loss and gliosis and can be visualised by MRI as new hyperintense T2 lesions in the white matter of the CNS; acute lesions show gadolinium enhancement in the first weeks of their development, mirroring the breakdown of the BBB [41]. After the resolution of acute oedema, hyperintense T2 lesions shrink and evolve into iso- or hypointense T1 lesions, depending on the grade of residual tissue damage, the latter case being defined as “black holes” [42].

2.3. Contribution of B Cells and Humoral Response to Acute CNS Injury

The role of B cells and humoral immunity in the pathogenesis of MS was considered less prominent compared with that of T cells, possibly due to the lack of consistency in detecting antibodies specific to CNS self-antigens in brain lesions or CSF [43]. However, their contribution was recently reinforced by the observation of the remarkable effectiveness of monoclonal antibodies (mAbs) targeting this cell population [44]. Evidence for oligoclonal antibody production in the CSF dates to the 1940s [45], and an oligoclonal pattern of intrathecal production of immunoglobulins (OCBs) is detected in the CSF of the vast majority of MS patients; OCBs are plausibly produced by a restricted number of plasma cell clones recruited in the CNS during the secondary autoimmune response [46]. Experimental data derived from one of the animal models of MS more similar to humans—EAE induced by immunisation with myelin-antigens in marmosets—show that a secondary antigen-specific humoral immune response is elicited at a perivenular level by cellular immunity [47]. In the animal model, breakdown of the BBB in the surrounding capillaries is promoted by diffusive molecules secreted by immune cells aggregated as perivascular cuffs: in this setting, circulating antibodies specific to myelin antigens contribute to myelin damage [48]. However, differently from this model, circulating antibodies against CNS antigens are rarely detected in humans [49]; nevertheless, anatomopathological studies showing the deposition of immunoglobulins (mainly IgG) and complement C9neo antigen at sites of active myelin destruction in pattern II lesions suggest a possible contribution of the humoral response to tissue injury [50]. Pleiotropic roles of B cells independent of antibody secretion may be relevant to MS pathogenesis, such as antigen-presenting function or secretion of pro-inflammatory cytokines (Figure 1) [51]. Clonally expanded B-cell clones were detected in the meninges, parenchyma, and CSF of MS patients; they may indefinitely persist within perivascular and leptomeningeal inflammatory infiltrates where they organise in follicle-like tertiary lymphoid structures of aggregated plasma cells, B cells, T cells, and follicular dendritic cells ^[52][53][54][52,53,54]. Recent data suggest that B-cell accumulation within inflammatory infiltrates may be critical to the chronicisation of the intrathecal autoimmune response, acting as professional APCs for definite CNS autoantigens [55].