Breaking the Barriers to Remyelination in Multiple Sclerosis

Abstract

Chronically demyelinated axons are rendered susceptible to degeneration through loss of trophic support from oligodendrocytes and myelin, and this process underlies disability progression in multiple sclerosis (MS). Promoting remyelination is a promising neuroprotective therapeutic strategy, but to date, has not been achieved through simply promoting oligodendrocyte precursor cell (OPC) differentiation, and it is clear that a detailed understanding of the molecular mechanisms underlying failed remyelination is required to guide future therapeutic approaches. In MS, remyelination is impaired by extrinsic inhibitory cues in the lesion microenvironment including secreted effector molecules released from compartmentalized immune cells and reactive glia, as well as by intrinsic defects in oligodendrocyte lineage cells, most notably increased metabolic demands causing oxidative stress and accelerated cellular senescence. Promising advances in our understanding of the cellular and molecular mechanisms underlying these processes offers hope for strategically designed interventions to facilitate remyelination thereby resulting in robust clinical benefits.

Failed Remyelination in MS: Impenetrable Blockade or Surmountable Barrier?

The CNS has the capacity to regenerate myelin sheaths after injury through the proliferation, migration, and differentiation of a population of adult progenitor cells referred to as oligodendrocyte precursor cells (OPCs), or, to a lesser extent, by the generation of new myelin sheaths by existing oligodendrocytes (OLs)[1]. However, remyelination is impaired in multiple sclerosis (MS). Demyelinated axons are vulnerable to physiologic stress and undergo degeneration due to increased energetic demands and lack of trophic support. Therefore, remyelination is considered a putative neuroprotective strategy in MS, but it has been difficult to translate success in preclinical models to positive clinical trials. In this review, we outline obstacles to remyelination, discuss remyelinating therapies that are in clinical trials, and highlight novel strategies to improve remyelination in preclinical models. Finally, we discuss future directions, including the need to target the effects of inflammation and gliosis on OPC differentiation and the diversion of OL lineage cells into immune or senescent profiles that no longer support remyelination.

Barriers to Remyelination in MS

Failed remyelination in MS could be due to disruptions in any stage of OL lineage cell recruitment, maturation, or axon engagement. While some studies have suggested that OPC recruitment to lesion areas is impaired during inflammatory demyelination [2], analyses of human MS patient tissue have noted that OPCs and premyelinating OLs are present in lesions but fail to mature [3]. Moreover, there is evidence that demyelinated axons themselves become less receptive to remyelination, possibly by altered expression of cell adhesion molecules such as the polysialylated neuronal cell adhesion molecule (PSA-NCAM), which inhibits myelination in development [4]. Thus, MS pathology affects myelination at multiple stages of OPC differentiation and myelin sheath extension.

Failed remyelination may reflect a reduced intrinsic ability of OPCs to differentiate into myelinating oligodendrocytes or a result of inhibitory cues in the lesion microenvironment. Studies demonstrating regional heterogeneity in OPC capacity for remyelination have provided support for both hypotheses. Single cell electrophysiological recordings in the mouse brain have shown that OPCs exhibit region-specific differences in ion channel expression, with OPCs in myelinating regions having higher levels of NMDA receptors than those in non-myelinating regions [5]. Given that NMDA receptors have previously been shown to promote OPC maturation via mTOR signaling, this suggests that subpopulations of OPCs exhibit intrinsic differences in remyelination capacity. In contrast, regional differences in remyelination following cuprizone-mediated demyelination may be the result of differing levels of microglial and astrocytic activation [6]. For this reason, both cell-autonomous and non-autonomous influences on remyelination should be considered (summarized in Figure 1).

Figure 1 -. Mechanisms of failed remyelination.

Remyelination is influenced by intrinsic and extrinsic factors that affect OPC proliferation, recruitment, maturation, and myelin sheath extension. Extrinsic factors include pro-inflammatory cytokines and inhibitory cues secreted by immune cells and resident glia, debris from autoimmune attack of the myelin sheath, composition and stiffness of the ECM, and axonal receptivity. Intrinsically, OPC capacity for proliferation and maturation is affected by various surface receptors. In addition, the normal aging process can elicit a senescent phenotype associated with pro-inflammatory secretions, mitochondrial stress, and altered gene expression, culminating in reduced responsiveness to pro-remyelination cues.

Cell-autonomous influences on OPC differentiation

The intrinsic capacity of OPCs for differentiation and proliferation may be altered in MS. It is well appreciated that remyelination decreases with age. In vitro studies of rodent OPCs have shown that aged cells differentiate less and are less responsive to signals such as thyroid hormone T3 that normally promote differentiation [7]. RecentFurther, intrinsic drivers of OPC differentiation may take on inhibitory roles in the context of demyelination. It was recently shown that in mice, poly(ADP) polymerase 1 (PARP1) inhibits a transcriptional repressor of myelin mRNAs [8]. PARP1 knockout mice showed deficits in OPC differentiation and myelination, suggesting that promoting PARP1 activity may be a viable pro-remyelination strategy [8]. However, PARP1 is also a mediator of the parthanatos cell death pathway, and increased PARP1 activation has been observed in OLs in MS lesions and in cuprizone-fed mice [9]. PARP1 inhibition was found to be protective against demyelination and OL death in the cuprizone model [9]. Thus, intrinsic pathways known to promote OPC differentiation in development may have distinct roles during inflammatory demyelination that must be considered when selecting therapeutic targets.

Effects of the microenvironment on OPC differentiation

There are multiple components of the lesion microenvironment known to suppress OPC maturation. Myelin debris produced by immune-mediated damage to axons inhibits OPC differentiation as well as axonal repair, and stimulating microglial clearance of this debris can promote OPC maturation[10]. The composition of the ECM is also altered in MS lesions [11]. Some of these components, notably hyaluronan, fibrinogen, and chondroitin sulfate proteoglycans are known to suppress remyelination in vivo. Fibrinogen, a coagulation factor that enters the parenchyma in MS following blood-brain-barrier disruption, activates bone morphogenic protein (BMP) signaling in OPCs. Pharmacological blockade of the BMP receptor type I in the presence of fibrinogen prevented OPCs from becoming astrocytes and promoted their differentiation in experimental autoimmune encephalomyelitis (EAE) [12]. In addition, ECM stiffness, which increases with age, is increasingly appreciated as a factor influencing OPC differentiation. OPCs express PIEZO1, a mechanosensitive ion channel that senses changes in ECM stiffness. Reducing PIEZO1 expression in aged mice using short hairpin RNA increased OPC proliferation and maturation, highlighting the importance of a permissive ECM environment [13].

Pro-inflammatory cytokines present in the lesion microenvironment can have opposing effects on remyelination. Interleukin-1β can promote remyelination by stimulating the production of insulin growth factor, a pro-remyelination signal, by microglia and astrocytes [14]. For other cytokines, such as tumor necrosis factor α (TNFα), the effect on remyelination is dependent on the receptor that initiates the signaling cascade. Interactions between transmembrane TNFα and TNF receptor 2 stimulate OPC proliferation. In contrast, soluble TNFα signaling through TNF receptor 1 impairs remyelination by promoting Th1 cytokine production and peripheral immune cell infiltration [15]. Other cytokines, such as interferon gamma (IFNγ), also can impair remyelination. Recently, our group and others have shown that OPCs exposed to IFNγ exhibit increased expression of genes associated with antigen presentation, as do OPCs in human MS patients [16]. These inflammatory OPCs (iOPCs) can phagocytose, process, and present antigen to both CD4+ and CD8+ T cells in vitro, potentially stimulating inhibitory pro-inflammatory cytokine production and cytotoxic cell death.

Recent evidence suggests that OPCs activate an intrinsic senescence program in the aged brain[17]. While neural progenitor cells (NPCs) normally promote OPC differentiation, they acquire a senescent phenotype in lesion areas of primary progressive (PPMS) MS patients. These NPCs secrete high mobility group box protein 1 (HMGB1), which initiates genetic and epigenetic changes in OPCs, culminating in cell senescence. Antibody blockade of HMGB1 promoted OPC differentiation in vitro [18]. In addition to intrinsic changes in OPC proliferation and differentiation, cell senescence is characterized by the release of cytokines, chemokines, proteases, and extracellular matrix (ECM) components, referred to as senescence-associated secretory phenotype[19]. Thus, senescent OPCs could contribute to the inhibitory microenvironment, further dampening remyelination capacity.

Breaking the Barriers: Remyelination Therapies in Clinical Trials

Some of the candidate drugs have been tested in MS animal models and moved to human clinical trials. We summarize past and present clinical trials in Table 1 and further discuss completed trials with published results. Figure 2 demonstrates some of the remyelinating therapies that have been evaluated in MS clinical trials.

Table 1.

Remyelinating drugs in clinical trials for MS

Drug class	Drug	Target/mechanism	Phase	Trial #	Status	MS Type
Small molecule antagonist	Clemastine Fumarate	mAChR (M1) and Histamine (H1)	II	NCT02040298	Ongoing	RMS
				NCT02521311	Recruiting	Optic neuritis
	GSK239512	Histamine (H3)	II	NCT01772199	Completed	RMS
	BN201	NRDG1 growth factor	I	NCT03630497	Completed	Healthy subjects
	PIPE-307	mAChR (M1)	I	NCT04725175	Recruiting	MS
Small molecule agonists	Bexarotene (IRX4204)	RXR	II	ISRCTN14265371	Completed	RMS
	Pioglitazone (Actos)	PPARγ	I	NCT00242177	Completed	RMS
	CHS-131 (INT-131)	PPARγ	II	NCT02638038	Completed	RMS
Hormones	Liothyrinine (T3)	Thyroid hormone receptor	I	NCT02760056 NCT02506751	Completed	MS
	Domperidone	Increases prolactin levels	II	NCT02308137, NCT02493049	Completed	PMS
	Bazedoxifene	Estrogen receptor modulator	II	NCT04002934	Recruiting	RMS
	Erythropoietin	Neuroprotective	II	NCT01144117	Recruiting	PMS
	Testosterone (Nebido®)	Androgen receptor	II	NCT03910738	Recruiting	RMS
Metabolic modulators	Simvastatin	Cholesterol metabolism	II	NCT00647348	Completed	PMS
	Metformin	Increases mitochondrial function	I/II	NCT04121468	Recruiting	Ped. MS
	Olesoxime (TRO19622)	Cholesterol-like neuroprotective agent	I	NCT01808885	Completed	RMS
	Fasting diet	Metabolism	NA	NCT01538355	Completed	MS
	Ketogenic diet			NCT03508414	Not recruiting	MS
	Biotin (MD1003)	Cellular energy, myelin synthesis	III	NCT02936037	Completed	MS
	Quetiapine fumarate	Cholesterol biosynthesis	I, II	NCT02087631	discontinued	RMS/PMS
Blocking Antibodies	Opicinumab (BIIB033)	LINGO-1	II	NCT01721161	Completed	Optic neuritis
	Pepinemab (VX15/2503)	Semaphorin 4D	I	NCT01764737	Completed	MS
	rHIgM22	Oligodendroglia	I	NCT01803867 NCT02398461	Completed	MS
	Ozanezumab (GSK1223249)	Nogo-A	I	NCT01424423 NCT01435993	Completed	RMS
	Elezanumab	RGMa	II	NCT03737851 NCT03737812	Completed	MS
Nanoparticle	Gold nanoparticles (CNM-Au8)	Nanocatalyst to transfer and/or receive electrons	II	NCT03536559	Recruiting	RMS, Optic neuritis
Combination therapy	Opicinumab DMT	LINGO-1 and immunosuppression	II	NCT01864148 NCT03222973	Completed	MS
	Pioglitazone Clemastine Dantrolene Pirfenidone	Multi-target	I/II	NCT03109288	Recruiting	MS

Abbreviations: Ref., Reference; ID, identification; RMS, Relapsing MS; Ped. MS, pediatrics MS; PMS, Progressive MS; mAChR; Muscarinic acetylcholine receptors, RXR; Retinoic acid receptor; PPAR, Peroxisome proliferator-activated receptor; NA, Not applicable; RGMa, Repulsive guidance molecule-a; DMT, Disease modifying therapies; IFN, Interferon.

Figure 2 -. Targets of clinical remyelination therapies.

Schematic of clinical remyelination strategies targeting intrinsic and extrinsic influences on OPC maturation and survival. OPC, Oligodendrocyte progenitor cells; M1 muscarinic receptors; H1 histamine receptors; RXR, retinoic acid receptor; PPAR, peroxisome proliferator-activated receptor; AMPK, AMP-activated kinase; TR, Thyroid receptor.

Small molecule agonists and antagonists

Using high-throughput screening methods in vitro, multiple research groups have identified potential remyelinating agents that promote OPC growth and differentiation [20]. A group of candidate drugs for remyelination are small-molecule antagonists that block remyelination-inhibiting receptors such as muscarinic acetylcholine receptors (mAChR). Clemastine fumarate is a mAChR/histamine H1 receptor antagonist that significantly promoted remyelination in preclinical studies[21–23]. Clemastine was quickly translated into a phase II trial (NCT02040298), in which a modest decrease in the visual evoked potential (VEP) latency was reported as the main outcome measure in treated patients, suggesting that clemastine could promote remyelination in MS. Another example is GSK239512, a potent, brain penetrant H3 receptor antagonist that was found to boost OPC differentiation in vitro and in vivo [24]. The remyelinating efficacy of GSK239512 was recently tested in people with relapsing-remitting MS (R) MS in combination with either Interferon-beta 1a (Avonex) or Glatiramer acetate (Copaxone). The drug was well tolerated, but the trial failed to meet its primary and secondary endpoints (NCT01772199).

Targeting pro-remyelination receptors is another strategy that has been tested in MS clinical trials. The retinoid X receptor (RXR) gamma family of nuclear receptors was identified as positive regulators of remyelination in rodents[25]. Oligodendrocytes express RXR during active remyelination, suggesting that RXR agonists could potentially promote remyelination. Bexarotene is an RXR agonist which was shown to restore the phagocytic capacity of blood monocyte-derived macrophages derived from MS patients [26]. The pro-remyelination activity of Bexarotene has been tested in people with RMS as an add-on therapy to dimethyl fumarate (ISRCTN14265371). Bexarotene was poorly tolerated with hematological and endocrine side effects, and the primary outcome was not reached. However, the findings from MTR and VEP P100 latency suggested that RXR agonism may promote remyelination of chronic lesions, particularly in the grey matter.

Peroxisome proliferator activated receptor (PPAR) is a nuclear receptor that functions through heterodimeric association with RXR. Once activated, the PPAR-RXR complex induces transcription of genes associated with OPC differentiation[27]. Pioglitazone is an FDA-approved PPARγ agonist used to treat type 2 diabetes. Preclinical studies demonstrate that pioglitazone delays the onset and reduces the severity of clinical symptoms in EAE [28]. Mechanistically, pioglitazone inhibits proinflammatory response of monocytes and promotes phagocytosis of myelin debris [26]. In a phase I clinical trial (NCT00242177), the therapeutic potential of oral pioglitazone was evaluated in people with RMS. Pioglitazone treatment led to clinical improvement without adverse events in this small pilot study. However, MRIs carried out after treatment showed no significant change in overall brain atrophy or demyelination. Another highly selective PPARγ agonist is CHS-131, which crosses the blood-brain-barrier and was shown to be clinically effective in EAE. CHS-131 has recently been tested in people with RMS and was found to be well tolerated and significantly decreased the incidence of new gadolinium contrast-enhancing lesions compared with placebo (NCT02638038).

Antibodies

Antibodies targeting remyelination inhibitors have been shown to promote remyelination and functional improvement in MS experimental models. One of the therapeutic targets is Nogo-A, a transmembrane protein mostly expressed by oligodendrocytes that inhibits axonal growth in the injured CNS [29]. Ozanezumab (GSK1223249) is a humanized monoclonal antibody targeting Nogo-A that promotes OPC differentiation. Two phase I studies for ozanezumab were conducted in RMS patients (NCT01424423 and NCT01435993). No safety issues were reported, and the trial details were published by GlaxoSmithKline (GSK) online.

Opicinumab (also known as BIIB033), is an antibody developed to neutralize LINGO-1, a transmembrane cell surface glycoprotein known to inhibit remyelination[30]. Studies in vitro and in animal models revealed that antibody blockade or genetic deletion of LINGO-1 promotes axonal remyelination[30,31]. The remyelination efficacy of opicinumab was tested in a phase II study in patients with acute optic neuritis (NCT01721161). The trial did not reach its primary endpoint as measured by retinal nerve fiber thickness preservation. However, the velocity of nerve conduction measured by VEP was improved in opicinumab treatment group compared with placebo group and reached significance in the per protocol subset but not in the intent to treat study group. In another phase II trial, the safety and efficacy of opicinumab and interferon beta-1a (Avonex) combination therapy was investigated in patients with RMS and SPMS (NCT01864148). The study did not achieve its primary outcome of a dose linear improvement in disability in treated patients compared with placebo. A third phase II study in RMS was recently completed (NCT03222973) and also did not achieve its primary endpoint.

Another target is semaphorin 4D (SEMA4D), which has been shown to inhibit differentiation and migration of OPCs [32]. Pepinemab (VX15/2503) is a high-affinity humanized IgG4 anti-SEMA4D antibody. A phase I study investigating the safety of VX15/2503 demonstrated that the drug is well tolerated by MS patients (NCT01764737). The possible clinical benefits of VX15/2503 need to be evaluated in future studies enrolling patients with MS.

Elezanumab is another humanized monoclonal antibody that has been tested in clinical trials. Elezanumab blocks repulsive guidance molecule-a (RGMa), a potent inhibitor of axonal growth that is found in chronic lesions of PMS patients. In preclinical models, elezanumab was shown to promote axon regeneration, neuroprotection, and remyelination[33]. In a phase II clinical trial, elezanumab was shown to be well-tolerated in patients with PMS and RMS. Elezanumab efficacy as a remyelination agent is currently being investigated in PMS (NCT03737812) and RMS (NCT03737851).

rHIgM22 is a recombinant human IgM antibody which binds to the surface of oligodendrocytes and promotes OPC proliferation and survival in animal models[34]. Additionally, rHIgM22 stimulates phagocytosis and clearance of damaged myelin[35]. Two phase I studies were completed in 2015 (NCT01803867) and 2018 (NCT02398461) evaluating the safety and tolerability of single rHIgM22 in patients with MS. Results of the 2015 clinical trial showed that rHIgM22 reported no adverse effects or discontinuations. rHIgM22 was measurable in the CSF of patients. However, no statistically significant changes in any of the pharmacodynamic or various clinical measures were observed.

Hormones

Several lines of evidence indicate that hormones influence oligodendrogenesis under physiological conditions. We have shown that thyroid hormones play a critical role in regulating oligodendrocyte differentiation during development [36]. Two small phase I clinical trials have been done to evaluate the safety of Liothyronine (L-T3), a short-acting thyroid hormone, in people with MS (NCT02760056, NCT02506751). Liothyronine was well tolerated, and a future phase II clinical trial is warranted to investigate the efficacy of L-T3 as a remyelination therapy. Sobetirome (GC-1) is a thyroid hormone receptor β-specific agonist that has shown promise for myelin repair preclinically [37].

Another example is prolactin, a hormone produced in the pituitary gland that triggers lactation post-partum. Work done in animals suggests that increased systemic prolactin levels may improve remyelination[38]. A peripheral dopamine receptor blocker, domperidone, improves myelin repair in mice via increasing the levels of prolactin. A clinical trial that assessed whether domperidone can reduce the progression of disability in secondary progressive MS (SPMS) (NCT02308137) found that the treatment was well tolerated but did not reduce disability progression.

Metabolic modulators

The biosynthesis of fatty acids or cholesterol is shown to influence remyelination [39]. In fact, several known pro-remyelinating drugs were shown to inhibit enzymes involved in cholesterol biosynthesis, including emopamil binding protein (EBP). This led to the accumulation of 8,9-unsaturated sterols, which promoted OPC differentiation and remyelination. Targeting biosynthesis and metabolism might be a viable therapeutic strategy.

Statins lower cholesterol by inhibiting 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase, a key enzyme that catalyzes the synthesis of cholesterol and other isoprenoids metabolic. Previous preclinical studies provided conflicting results on the efficacy of simvastatin for MS treatment[40,41]. A phase II clinical trial in patients with SPMS showed that high-dose simvastatin is safe and significantly reduced the rate of whole brain atrophy measured by serial volumetric MRI (NCT00647348).

Biotin is involved in a wide range of metabolic processes for fatty acid synthesis and ATP production. High-dose biotin acts as a cofactor for acetyl-CoA carboxylase and induces myelin formation by oligodendrocytes[42]. A phase II clinical trial was designed to assess the safety and efficacy of high-dose biotin (MD1003) in progressive forms of MS (NCT02936037). No significant improvement of disability or walking speed was found in people with PMS treated with high-dose biotin.

Quetiapine fumarate is an antipsychotic medication used to treat bipolar disorder, depression, and schizophrenia. Quetiapine stimulates cholesterol biosynthesis in OPCs[43] and enhances remyelination in animal models including cuprizone-induced demyelination[44]. A recent clinical trial in people with RMS and progressive MS (PMS) reported that quetiapine was intolerable even at the lowest doses tested (NCT02087631). Participants discontinued treatment due to sedative side effects and further trials were discouraged.

Novel Remyelinating Strategies in Preclinical Studies

The failure of several promising pro-remyelinating drugs in recent clinical trials illustrates the challenge of designing translatable pre-clinical studies. Rodent models of MS are limited in their ability to recapitulate human disease. EAE reproduces the inflammatory events that occur following autoimmune attack of myelin; however, remyelination is restricted in this model due to extensive axonal damage. In contrast, cuprizone-mediated demyelination allows remyelination to be studied over a predictable time course and region of the CNS but lacks the long-term inflammatory events known to occur in MS [45]. The combined adoptive transfer-cuprizone (ATCPZ) model incorporates both the inflammatory and remyelinating aspects of MS more efficiently than in EAE alone. In AT-CPZ, mice are fed cuprizone for 4 weeks to induce demyelination of the corpus callosum and then myelin-reactive CD4+ T cells that are polarized to a Th17 profile are adoptive transferred [46]. The T cells migrate to the site of demyelination and inhibit spontaneous remyelination that normally is robust following cuprizone cessation. An advantage of this model as compared to EAE is the marked axon preservation in the callosum, and therefore AT-CPZ may prove to be a better model for testing drugs that promote remyelination in an inflammatory environment.

Experimental dual-acting therapies

The need to consider both intrinsic OPC/OL remyelination capacity as well as extrinsic influences on remyelination is reflected by a variety of combinatorial approaches in recent preclinical studies (summarized in Table 2). Sephin1 has been shown to promote the integrated stress response, a cytoprotective program activated under inflammatory conditions. Sephin1 promoted remyelination following cuprizone in mice expressing IFNγ in the CNS by altering OPC response to inflammation. Bazedoxifene, a small molecule known to promote OPC proliferation and differentiation, further promoted remyelination, suggesting these approaches are complementary [47]. Similarly, a CXCR7 antagonist was shown to significantly reduce clinical EAE scores. Subsequent experiments demonstrated that the antagonist blocked peripheral immune infiltration in EAE and promoted OPC maturation after cuprizone-mediated demyelination [48]. In addition, ursolic acid, known to suppress inflammation, also promoted remyelination in cuprizone-fed mice and chronic EAE. By activating PPARγ, ursolic acid stimulated OPC maturation and astrocyte secretion of neurotrophic factors, indicating it may have pro-regenerative potential [49].

Table 2.

Experimental and repurposed remyelinating therapies

Promoting the recruitment or intrinsic differentiation of OPCs
Drug Class	Drug	Target/mechanism	Model tested
Antagonist	(±)U-50488	Kappa opioid receptor	EAE
	Pranlukast	G-protein coupled receptor Gpr17	LPC
	Yhhu4952	Jagged1-Notch1	Cuprizone, EAE
	Mirl46a	Toll-like receptor 2/Interleukin-1 receptor-associated kinase 1	EAE
	GlNAc	Platelet-derived growth factor receptor alpha endocytosis	Cuprizone
	Anacardic acid	Histone acetyl transferase	Cuprizone, EAE
	Nalfurafine	Kappa opioid receptor	EAE
	Vitamin C (As-2P)	Hypoxia-inducible factor?	EAE
Agonist	Clobetasol	Glucocorticoid receptor	LPC
	Miconazole	MAPK	EAE, LPC
Blocking Antibody	Antibody 11C7	Nogo-A	EAE
Blocking extrinsic inhibitors of remyelination
Antagonist	Nimodipine	1.2 voltage-gated L-type calcium channel	EAE
	Ac-4,4-diF-GlcNAc	CSPG biosynthesis	EAE
	CZ-7	Enhanced myelin debris clearance	Cuprizone
	Cambinol	Sphingomyelinase 2	Cuprizone
	S3	Hyaluronidase	LPC
	Fasudil	Rho kinase	Cuprizone
	LDN-212854	Type I BMP receptor	EAE
Agonist	Tuftsin	Enhanced microglia phagocytosis via tuftsin receptor	EAE
	Agathisflavone	Estrogen receptor	LPC
	Ganoderic acid A	Farnesoid-X-receptor	EAE
Combination and dual-acting therapies
Metabolic Modulator	Intermittent fasting	Metabolic reprogramming of immune and OL lineage cells	Cuprizone, EAE
	Calorie restriction	Metabolic reprogramming of immune and OL lineage cells	Cuprizone
	Metformin	Phosphorylation of CREB-binding protein and inhibition of autophagy	LPC
Agonist	Ursolic acid	Triterpenoid with anti-inflammatory and pro-myelination effects via PPARy activation	EAE
Antagonist	ACT-1004-1239	CXCR7	EAE
Combinatorial	Sephin 1 +/− bazedoxifene	Integrated stress response enhancement +/− estrogen receptor agonist	CNS delivery of IFNγ, EAE
	Apo-transferrin + thyroid hormones	OPC proliferation + thyroid receptor agonist	Cuprizone

Pro-myelinating effects of known anti-inflammatory drugs

Preclinical studies have investigated the pro-remyelination potential of existing drugs that are already known to have anti-inflammatory effects. Metformin, a diabetes drug, is already in clinical trials for MS due to its anti-inflammatory effects on peripheral immune cells in EAE [50]. More recently, metformin has been shown to increase mitochondrial function via AMP-activated protein kinase (AMPK) signaling pathway and the responsiveness of aged OPCs to pro-differentiation signals [17]. In a model of juvenile focal demyelination, metformin promotes OPC proliferation and differentiation, suggesting that it might have dual anti-inflammatory and pro-regeneration effects [51]. Whether combinatorial approaches such as these are more successful in clinical trials than those that exclusively target OPC proliferation/maturation remains to be seen.

Dual-acting potential of dietary interventions

Dietary interventions such as intermittent fasting and caloric restriction have demonstrated potent anti-inflammatory effects in neurodegenerative diseases by ameliorating CNS metabolic dysfunction [52]. Animal studies indicate that these favorable effects in MS models are partially due to changes in the gut microbiome that affect peripheral inflammation. Fasting diets altered the composition of the gut flora in mice, resulting in fewer Th17 cells and more Treg cells, and fecal transplants from fasted mice ameliorated EAE severity [53]. However, evidence also suggests that fasting diets and caloric restriction might have direct effects on remyelination. OPC maturation and survival were increased in a model of EAE following a fasting-mimicking diet (FMD) consisting of 3 days of low-calorie and low-protein intake repeated every 7 days. To ensure that this effect was not due to reduced immune cell infiltration and activation, the experiment was repeated in the cuprizone model, and FMD was still shown to promote remyelination [54]. However, levels of microgliosis and astrogliosis were not assessed, leaving it unclear whether FMD directly or indirectly promotes remyelination.

A study of mice who were given a calorie-restricted (CR) diet for four weeks following 3 months of cuprizone treatment found that remyelination was increased in CR mice compared to the unrestricted controls. The CR group also had reduced gliosis in the corpus callosum, indicating that this dietary intervention promotes OPC survival by preventing glia from taking on an inhibitory, inflammatory phenotype [55]. Future research into metabolic changes in OPCs and OLs during inflammatory demyelination could help elucidate whether dietary interventions can directly influence remyelination in addition to their favorable anti-inflammatory effects. In any case, FMD and ketogenic diet for MS have moved into clinical trials, highlighting their promise in preclinical studies.

Concluding Remarks and Future Directions

Remyelination therapy for MS remains challenging to achieve in practice. Pro-differentiation drugs have shown promise experimentally, but have had minimal clinical benefit in drug trials, suggesting that remyelination strategies targeting OPC intrinsic factors are limited by inhibitory cues from the peripheral blood infiltrate into the local microenvironment including cytokines and fibrinogen [11,12]. Further, soluble factors released by inflammatory/neurotoxic glia [56], and the accumulation of ECM proteins negatively affect OPC growth, survival, and differentiation, and may even direct OPCs toward an inflammatory phenotype [16] that might further restrict remyelination. In addition, both OPC recruitment and differentiation are impaired by aging, which induces intrinsic senescence programs and decreases the efficiency of remyelination. To break these barriers, a combination therapy targeting both OPCs and the inhibitory factors associated with the lesion microenvironment is likely needed to facilitate remyelination.