Role of lipoic acid in multiple sclerosis

Abstract Lipoic acid (LA) is an endogenous antioxidant that exists widely in nature. Supplementation with LA is a promising approach to improve the outcomes of patients with multiple sclerosis (MS). This systematic review aimed to provide a comprehensive overview of both in vitro and in vivo studies describing the pharmacokinetics, efficacy, safety, and mechanism of LA in MS‐related experiments and clinical trials. A total of 516 records were identified by searching five databases, including PubMed, Web of Science, Embase, Scopus, and Cochrane Library. Overall, we included 20 studies reporting LA effects in cell and mouse models of MS and 12 studies reporting LA effects in patients with MS. Briefly, cell experiments revealed that LA protected neurons by inhibiting the expression of inflammatory mediators and activities of immune cells. Experimental autoimmune encephalomyelitis mouse experiments demonstrated that LA consistently reduced the number of infiltrating immune cells in the central nervous system and decreased the clinical disability scores. Patients with MS showed relatively stable Expanded Disability Status Scale scores and better walking performance with few adverse events after the oral administration of LA. Notably, heterogeneity of this evidence existed among modeling methods, LA usage, MS stage, and trial duration. In conclusion, this review provides evidence for the anti‐inflammatory and antioxidative effects of LA in both in vitro and in vivo experiments; therefore, patients with MS may benefit from LA administration. Whether LA can be a routine supplementary therapy warrants further study.


| INTRODUC TI ON
Multiple sclerosis (MS) is a disabling autoimmune disease of the central nervous system (CNS) characterized by demyelination and neurodegeneration. 1 It affects approximately 2.5 million people worldwide and poses a growing burden to society. 2,3 Relapsing-remitting MS (RRMS) is the most common initial course featuring alternate relapse and remission, and disability is aggravated gradually with illness development. 4 After approximately 20 years, around 90% of RRMS patients will develop secondary progressive MS (SPMS) characterized by progressive neurodegeneration without any definite remission periods. 5,6 In addition to SPMS, progressive MS (PMS) also includes primary progressive MS featuring inapparent clinical relapses from the onset. At present, both immune and nonimmune mechanisms are believed to be involved in MS pathogenesis.
The "outside-in" hypothesis proposes that the inflammatory demyelinating process begins in the subarachnoid space and cortex and extends into the white matter. 7,8 In this model, the invasion of peripheral immune cells disrupts the blood-brain barrier (BBB) integrity and contributes to the prolonged presence of inflammatory activity.
In RRMS, the interaction of monocytes and brain endothelial cells (ECs) produces massive reactive oxygen species (ROS), leading to the loss of tight junctions and migration of monocytes. 9 For T cells, the mutual recognition of lymphocyte function-associated antigen-1 (LFA-1), intercellular cell adhesion molecule-1 (ICAM-1), very late antigen-4 (VLA-4), and vascular cell adhesion molecule-1 (VCAM-1) permits them to cross the BBB. The release of matrix metalloprotein-9 (MMP-9) by T cells is also essential for the migration process.
Notably, infiltrated T cells can recruit macrophages, microglia, and astrocytes by secreting mediators, including tumor necrosis factorα (TNFα), interferonγ (IFNγ), and interleukins-17 (IL-17). 10,11 These abnormally activated immune cells target neurons and the myelin sheath and drive MS relapse and progression. Therefore, several disease-modifying therapies (DMTs) can decrease relapse rates by immunomodulation. However, the potential risks of serious adverse events (AEs) and COVID-19 infection limit its clinical use to some extent. 12,13 In PMS, inflammation is compartmentalized and mainly driven by the activities of innate microglia, astrocytes, and B cells. 11 Unfortunately, the efficacy of DMTs for PMS tends to be disappointing, motivating the search for a new treatment option. 14 Oxidative stress is another crucial driver of MS once the autoimmune system has caused damage to the CNS. 10 It occurs when an imbalance exists between excessive production of free radicals and insufficient biological ability to remove them. 15 The CNS is quite sensitive and vulnerable to oxidative stress because of its high oxygen consumption and lipid abundance. Oxidizing substances, such as ROS and nitrogen species, are usually produced by activated macrophage and microglial structures, causing damage to lipids, proteins, and DNA. Consequently, the CNS is variously disrupted through processes such as increased BBB permeability, myelin phagocytosis, and neurodegeneration. 16,17 In the plasma of MS patients, the levels of antioxidants and total antioxidant capacity are decreased. 18,19 Autopsy studies have also widely detected the damage induced by oxidative stress in cerebrospinal fluid and CNS tissues. 20,21 Therefore, oxidative stress may be another hopeful therapeutic target of MS. At present, many antioxidant compounds have improved serological indicators in MS patients. 22 Vitamin D decreased the relapse rates as an antioxidant in RRMS patients. 23 However, the findings of the efficacy of antioxidants tend to be conflicting and confusing, strongly suggesting that the effect of using a single antioxidant is limited. Considering the above, an ideally effective medicine must possess the ability to prevent multiple pathogenic factors and outstanding BBB permeability.
Lipoic acid (LA), also known as thioctic acid, has become a hopeful complementary therapy in MS to target both inflammation and oxidative stress. LA is a double-sulfhydryl natural antioxidant with two enantiomers according to optical rotation: R-LA and S-LA.
Overall, R-LA exists widely in plants and animals, whereas S-LA is artificially synthesized to compose the racemic mixture (1:1 R/S-LA). 24 In the human body, R-LA is synthesized de novo by cysteine and fatty acids in small amounts; thus, it primarily depends on exogenous supplements such as organ meat, broccoli, and fruits. 25 For individuals, the racemic form can be absorbed rapidly after oral administration and participate in various biological metabolic pathways. First, it contributes to the synthesis of vitamin C and vitamin E. 26 Second, it is reduced to dihydro-LA (DHLA), and DHLA is involved in the biosynthesis of intracellular glutathione (GSH) and coenzyme Q10. 27,28 Third, R-LA plays a crucial role in mitochondrial energy production as a cofactor for some enzymatic complexes in the Krebs cycle. 29 When other metabolic pathways are saturated, redundant LA (nearly 10%) will be excreted through the kidneys. 30 Over the past two decades, whether LA improves the quality of life of patients with MS has been intensively studied. 31 In mouse models of experimental autoimmune encephalomyelitis (EAE), LA increased the population of mature oligodendrocytes and alleviated neurological symptoms, suggesting that LA might protect and promote neuronal regeneration. 32,33 However, the results of alleviated neurological symptoms were inconsistent for different administration pathways, timing, and dosage, making the evidence somewhat fragile. In patients with MS, LA reduced the Expanded Disability Status Scale (EDSS), although the between-group difference was not statistically significant. 34,35 The confusing result regarding whether LA could improve patient outcomes probably resulted from the short trial duration. Additionally, the annualized percent brain volume change was less after 2 years of supplementation in the LA group, indicating that LA might prevent neuronal death and reduction. 36 More importantly, few AEs were reported when using LA as an oral preparation for 2 years. In summary, LA shows strong antioxidative and anti-inflammatory effects in MS, which makes it a potential candidate for complementary and long-term therapy.
To date, no study has systematically summarized the current findings of LA in MS, and some results appear to be controversial. A good review of both achievements and limitations will contribute to determining reliable evidence and research trends for future studies.
In this review, we aimed to provide comprehensive insight into the role of LA in MS, including the aspects of pharmacokinetics, efficacy, safety, and mechanism, in both in vitro and in vivo experiments. We hope that our work will contribute to the development of new drugs and combination therapy for patients with MS.

| Search strategy
According to the guidelines of the 2009 Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement, 37 English-language studies published from inception up to July 1, 2021, were collected by searching five databases: PubMed (Medline), EMBASE, Web of Science, Scopus, and Cochrane Library.
Identified search terms included ("multiple sclerosis" OR "MS") AND ("lipoic acid" OR "Thioctacid" OR "LA"). Additional records were identified manually through other sources, such as any related review papers and reference lists of all included studies to avoid missing relevant studies in the initial search. The whole search process was conducted by two authors independently (H.S.X. & X.F.Y.).

| Study selection
After removing duplicates, all the studies were screened for eligibility

| Data extraction
For the included studies, we collected the following information into a spreadsheet in Excel: (1) subject characteristics including age, sex, EDSS score, and MS duration; (2) MS-related model establishment in the preclinical experiments; (3) LA dosage (4) endpoints including efficacy, safety, pharmacokinetics, and mechanism; and (5) first author's name, publication date, study design, and follow-up duration.
For detailed data not shown in the full text, the e-mails were sent to the corresponding authors for help.

| DISCUSS ION
We obtained 516 potential records in the initial systematic search.
After the removal of duplicates, 143 studies were screened based on the title and abstract, leading to 59 full-text studies screened for eligibility. In this process, 27 articles were excluded because of irrelevant endpoints, nonoriginal studies, and combined antioxidant supplements. Finally, 32 intervention studies were included in this systematic review to investigate the effects of LA on efficacy, safety, pharmacokinetics, and mechanism. An overview of the study selection is presented in Figure 1.

| LA pharmacokinetics and transportation to the brain
A rat experiment found that the duodenum was the best portion of the intestine for LA absorption and that R-LA showed a higher absorption percentage than S-LA. 40 Notably, two vital pathways are involved in the process of LA crossing the intestinal barrier: Na+/multivitamin F I G U R E 1 Preferred Reporting Items for Systematic Reviews flowchart (SMVT) and monocarboxylic acid (MCT) transporters. 41,42 Under equilibrium conditions, human SMVT can simultaneously bind and transport two LA molecules into the mesenteric vein, and human MCT transports LA in an energy-and low-pH-dependent manner. In the patients with RRMS/SPMS and healthy volunteers, the pharmacokinetic parameters showed no significant difference, suggesting that the MS status did not influence LA metabolism. 43 In 54 patients with SPMS, pharmacokinetics showed no significant difference between the baseline and 1 year later, suggesting that the oral administration of LA was stable for long-term use. 44 In patients with MS, three studies found that the time to reach the peak concentration of R-LA was much shorter than that of the racemic form, indicating the quicker absorption of the R-configuration. 33 Notably, human SMVT and MCT are also expressed in brain microvessels and contribute to the transportation of LA across the BBB. 46,47 In an in vitro experiment, LA showed the ability to cross the BBB and exert beneficial effects on the viability of astrocytes. 48 Besides, a rat experiment found that 14 C-labeled LA reached peak levels in the cortex, spinal cord, and sciatic nerve after one-half hour of oral administration, indicating that LA was taken up by both the CNS and peripheral nerves. 49 LA was also measured in the rat brain cortex, cerebellum, striatum, and hippocampus after intravenous and intraperitoneal administration. 50,51 Notably, a recent rat experiment found that the LA did not cross the BBB as easily as supposed after the correction for blood volume, which emphasized that the permeability of the BBB might be greatly influenced by cerebral blood flow. 52

| Role of LA in cell experiments
Human peripheral blood mononuclear cells (PBMCs) are isolated from peripheral blood and feature round nuclei. They mainly comprise lymphocytes, monocytes, and NK cells. 53 Most PBMCs are naïve without immune effects. Importantly, the largest fraction, T cells, will develop into diverse subsets of Th1, Th2, Th17, or regulatory T cells (Treg cells) after activation by different cytokines. 54,55 Monocytes in PBMCs can also be activated by proinflammatory

| Role of LA in animal experiments
EAE is a reliable murine model that can well simulate the occurrence and development of MS. 65 In the 11 included studies, EAE induction was accomplished using three methods (Table 1) protein content of CNS myelin and was recently proven to correlate with the severity of disease in MS patients. 66 In addition to the autoimmune component, the oral administration of cuprizone can cause whole-brain demyelination and gliosis and was used to establish the murine model in one study. Compared with the two previous methods, cuprizone induction is easier to operate but time-consuming (5 vs. 2 weeks). More importantly, female mice are more resistant to cuprizone induction, and estradiol/progesterone can protect against cuprizone-induced demyelination. 67,68 The LA dosage was 5-100 mg/kg per day, and the mode of administration F I G U R E 2 Lipoic acid protects the central nervous system by immunomodulation and antioxidation. In the periphery, LA prevents inflammatory cells from crossing the BBB by inhibiting the expression of LFA-1, ICAM-1, VLA-4, VCAM-1, and MMP-9 and protects brain endothelial cells. In the CNS, LA modulates autoimmunity by inhibiting the activity of T cells/microglia and decreasing the expression of TNFα and IFNγ, and LA reduces oxidative stress by neutralizing ROS and NO

Illness duration (years ± SD) LA dosage (orally/day) Serious adverse events
Yadav et al 33  Note: "-" indicates decreased expression or event compared with non-LA group, "+" indicates increased expression or enhanced activity, and "=" indicates no statistical difference.
included intramuscular injection (n = 5), intraperitoneal injection (n = 3), subcutaneous injection (n = 1), oral administration (n = 1), and general injection (n = 1). Only three studies indicated that a racemic form of LA was used, and the remaining eight studies did not report the specific form.

| LA mechanisms in EAE mice
Nuclear factor erythroid-2 related factor 2 (Nrf2) is a redox-sensitive transcription factor existing in the cytoplasm, and it will translocate to the nucleus when oxidative stress occurs. 77 In the nucleus, it binds to antioxidant response elements and initiates the transcription of over 200 detoxification genes. 78 In the rat brain, LA promoted Nrf2 translocation and the superoxide dismutase (SOD) activity to defend against oxidative stress. 79 Besides, LA upregulated the expression of Nrf2 and its downstream hemeoxygenase-1 to alleviate neuronal cell apoptosis. 80 In EAE mice, LA increased the expression of GSH and SOD to enhance antioxidant system activity ( Figure 2). Meanwhile, LA decreased the levels of ROS and lipid peroxidation in EAE mice. 32 These findings suggested that the LA-Nrf2-antioxidative system pathway might be involved in neurological improvement in EAE mice.

PK
Open label, RCT NA 1200mg oral lipoic acid can achieve therapeutic serum levels.

| Role of LA in clinical trials
Twelve studies based on clinical trials were included to describe the role of LA in patients with MS ( LA appears quite safe, with a compliance rate of 80% to 97% in MS patients. 44 The most common adverse events are gastrointestinal intolerance and rash. Notably, the oral administration of 600 mg of R-LA showed approximately half less gastrointestinal discomfort than 1200 mg of the racemic form, while the bioavailability was nearly equivalent. 98 Taking LA after meals and enteric coating can also improve tolerability. 99 Furthermore, one case of maculopapular rash with fever was reported after 2400 mg/day for 1 week, and the symptoms were resolved only by stopping intake. 100 The other three cases of rash were found in two double-blind RCTs, but they seemed to be milder and only affected the skin. 36,101 Consistently, in patients with diabetic polyneuropathy or other CNS diseases, few AEs were found, and some studies did not set AEs as an endpoint. [102][103][104] Consistent with our results, a meta-analysis including 71 placebo-controlled clinical studies found that LA was associated with no increased risk of AEs, even with pregnancy status. 105 For future studies, seeking the balance of effective dose and fewer AEs will be necessary.

| LA mechanisms in MS patients
LA showed mixed antioxidative and anti-inflammatory effects ( Figure 2). A 12 weeks double-blind RCT reported an apparent improvement in serum total antioxidant capacity in 52 RRMS subjects. 106 However, the specific approach remained confusing because no difference was found in the serum GSH:GSSG ratio, superoxide dismutase, and GSH peroxidase activity. 34,107 Notably, LA reduced the content of asymmetric dimethylarginine (a major endogenous inhibitor of endothelial NO synthase) in the blood of MS patients. 108 In future, considering that LA can chelate heavy metals, including iron and copper, investigating whether LA can function to prevent gadolinium-related contrast magnetic resonance imaging (MRI) will be interesting. 109 Additionally, another key point is whether improved peripheral antioxidant capacity can exactly reflect the redox status in the CNS, a topic that warrants investigation. In summary, LA is one of the most promising antioxidants to alleviate oxidative stress in the CNS because of its high water and fat solubility.
Regarding immunomodulation, LA protects the BBB from disruption by peripheral inflammatory cells. On the one hand, MMP-9 released by T cells degrades components of the extracellular matrix, 110 and ROS produced during monocyte binding to ECs result in the loss of tight junctions. 9 On the contrary, when endothelial cells are activated by TNFα/IFN-γ, ICAM-1 is overexpressed and binds to LFA-1, initiating cytoskeletal rearrangement in brain ECs. 111,112 These events that disrupt the BBB explain the finding that high levels of MMP-9/ICAM-1 are present before the appearance of new MRI-based gadolinium-enhancing lesions in MS patients. [113][114][115] In response, a two-week LA supplement decreased the levels of serum MMP-9 and ICAM-1 in 33 patients. 100 Another twelve weeks of LA supplementation decreased the levels of serum IFNγ and IL-4 in 46 RRMS patients, but the serum TNFα and IL-6 levels showed no difference. 35 Notably, a recent study found that the oral administration of meglumine cAMP promoted BBB integrity, suggesting that LA may maintain the normal functioning condition of ECs through a similar effect. 116 In summary, by stabilizing the BBB, LA can disturb inflammation progression in the CNS. Additionally, using LA orally appears to benefit MS patients and help to reduce relapse tendency. Future studies should note the clinical heterogeneity of a relatively short trial duration, LA forms, and different MS stages.

| CON CLUS IONS
We comprehensively summarized the current findings of LA regarding pharmacokinetics, efficacy, safety, and mechanisms in To date, the achievements of LA supplementation are exciting, but the evidence is not sufficiently strong, being limited primarily by the short trial duration and insufficient study quantity. Thus, multicenter and long-term controlled studies are encouraged to determine the strength of LA orally, an appropriate dose for long-term usage, and the most suitable combination therapy. As our understanding of the role of LA improves, we hope to uncover the best treatment regimens for MS patients.

ACK N OWLED G M ENTS
The authors sincerely thank Dexing Xun (the scientific illustrator from Nanchang University School of Pharmacy) for the artwork in Figure 2.

CO N FLI C T S O F I NTE R E S T
The authors declare no financial or other conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Hongsheng Xie contributed to investigation, data curation, and writing (original draft). Xiufang Yang contributed to investigation, data curation, and writing (editing). Yuan Cao involved in data curation. Xipeng Long contributed to validation and resources.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analyzed in this study.