Induced pluripotent stem cells‐based disease modeling, drug screening, clinical trials, and reverse translational research for amyotrophic lateral sclerosis

It has been more than 10 years since the hopes for disease modeling and drug discovery using induced pluripotent stem cell (iPSC) technology boomed. Recently, clinical trials have been conducted with drugs identified using this technology, and some promising results have been reported. For amyotrophic lateral sclerosis (ALS), a devastating neurodegenerative disease, several groups have identified candidate drugs, ezogabine (retigabine), bosutinib, and ropinirole, using iPSCs‐based drug discovery, and clinical trials using these drugs have been conducted, yielding interesting results. In our previous study, an iPSCs‐based drug repurposing approach was utilized to show the potential of ropinirole hydrochloride (ROPI) in reducing ALS‐specific pathological phenotypes. Recently, a phase 1/2a trial was conducted to investigate the effects of ropinirole on ALS further. This double‐blind, randomized, placebo‐controlled study confirmed the safety and tolerability of and provided evidence of its ability to delay disease progression and prolong the time to respiratory failure in ALS patients. Furthermore, in the reverse translational research, in vitro characterization of patient‐derived iPSCs‐motor neurons (MNs) mimicked the therapeutic effects of ROPI in vivo, suggesting the potential application of this technology to the precision medicine of ALS. Interestingly, RNA‐seq data showed that ROPI treatment suppressed the sterol regulatory element‐binding protein 2‐dependent cholesterol biosynthesis pathway. Therefore, this pathway may be involved in the therapeutic effect of ROPI on ALS. The possibility that this pathway may be involved in the therapeutic effect of ALS was demonstrated. Finally, new future strategies for ALS using iPSCs technology will be discussed in this paper.


| INTRODUC TI ON
More than 15 years have passed since the development of induced pluripotent stem cell (iPSC) technology by Shinya Yamanaka and his colleagues (Takahashi et al., 2007;Takahashi & Yamanaka, 2006), and yet, amazingly, interest in the application of iPSC technology to medicine and related fields continues to grow.In the field of iPSC-based regenerative medicine, the First in Human (FIH) clinical trials have begun for many diseases (Okano & Sipp, 2020;Yamanaka, 2020).The field of disease and drug discovery research using iPSC technology has become capable of deep phenotypic analysis beyond the hierarchy with the advancement of genome editing technology, organoid technology, single cell analysis technology, imaging technology, artificial intelligence technology, and so on.This has made it possible to mimic the phenomena that occur in vivo, develop drugs with a high probability of success based on the pathological mechanism of disease, elucidate the mechanism of action of developed drugs, validate administration methods, obtain proofof-concept (PoC), and stratify and select patients for treatment.We are entering a very exciting era.Under these circumstances, research on amyotrophic lateral sclerosis (ALS), a neurological intractable disease, is undoubtedly one of the most advanced in pathological analysis and drug discovery research using iPSCs.Three research groups worldwide, including ours, have been conducting pathological analysis, drug discovery, clinical trials, and reverse translation of the intractable neurological disease ALS, using iPSC technology.In this paper, we would like to present some aspects of the project and future challenges.

| DIS E A S E MODELING AND DRUG SCREENING FOR AL S US ING iPSC TECHNOLOGY
ALS is one of the most incurable neurological diseases and is a progressive disease characterized by muscle atrophy and loss of muscle strength due to damage to spinal motor neurons (MNs; lower motor neurons [LMNs]).Symptoms include difficulty walking, bulbar signs (dysarthria and dysphagia), and breathing problems, and although the patient's consciousness and perception are normal, the quality of life is significantly reduced.
The disease, known as ALS, was initially described by François Aran in 1850 (Aran, 1850) and later named by Jean-Martin Charcot in 1874.However, until 1993, when the superoxide dismutase 1 (SOD1) gene was first identified as the causative gene, there had been no clues for therapy.However, in 2023, 30 years after the first report of SOD1 mutations in familial ALS (FALS), the FDA approved QALSODY™ (tofersen) 100 mg/15 mL injection, a nucleic acid medicine based on antisense oligonucleotides (ASOs; Blair, 2023), for the treatment of this disease based on the results of the preceding clinical trial (Miller et al., 2022).It is now understood that 5%-10% of all ALS cases are familial (FALS), caused by mutations in a single gene, while the remaining approximately 90% are caused by environmental factors, aging, multiple genetic factors, etc. (sporadic ALS [SALS]).FALS, a hereditary form of the disease, has seen significant progress in gene identification.In a study conducted by Gregory and colleagues in 2020, they identified nearly 30 genes associated with FALS.Notably, SOD1, which is a causative gene for FALS identified by Rosen et al. in 1993, a crucial milestone in understanding was confirmed on the genetic basis of FALS.QALSODY™ (tofersen) can be applied to only a small fraction of ALS patients with SOD1 mutation.Riluzole (which inhibits glutamate neurotransmission), edaravone (which scavenges free radicals) and are already approved drugs for ALS, including both FALS and SALS patients, but their efficacy is limited, and there is an urgent need to elucidate the underlying pathology and develop therapeutic agents (Jaiswal, 2019).In 2022, in addition to riluzole and edaravone, a new combination therapy, Relyvrio™ (consisting of sodium phenylbutyrate and taurursodiol), was approved for use in the United States (Heo, 2022).The therapeutic targets of these drugs appear to be secondary and nonspecific neuronal protective effects, which are downstream of the primary pathogenesis of ALS rather than oxidative stress and mitochondrial function preservation that has been previously assumed, and they seem to inhibit the progression of ALS pathology.

Degeneration of MNs in ALS can be broadly classified into (i)
the Dying Forward mechanism, in which MNs are hyperexcited, and (ii) the Dying Back mechanism, in which abnormalities and degeneration of nerve axons precede cell death, both of which are associated with FALS and SALS.Studies have shown that the causes of ALS include degeneration of neurites, especially axons, which occurs after the onset of the disease; damage due to overexcitation of neurons; abnormal subcellular localization of key molecules; accumulation of abnormal proteins; impairment of mitochondrial function; increased oxidative stress, which damages cells; and increased series of abnormalities such as increased cell-damaging oxidative stress, increased inflammation, and increased apoptosis (neuronal cell death) have been postulated.Mice with ALS-causing genes have been used as experimental models of ALS, but there is still no model that adequately reproduces the actual human condition.
In fact, many drugs have been developed using ALS model mice (transgenic mice with mutant human SOD1), but few of them have unfortunately shown efficacy in human ALS patients.This may be due to differences of biological nature between humans and mice, which could play a crucial role in the pathogenesis of ALS, such as genomic information including introns, alternative splicing, distal enhancers, non-coding genes, transposons, and other repetitive sequences between humans and mice, differences in neural circuits amyotrophic lateral sclerosis (ALS), clinical trials, induced pluripotent stem cell (iPSC), motor neurons (MNs), ropinirole hydrochloride (ROPI), sterol regulatory element-binding protein 2 (SREBP2) and cell environment that make up the primary motor cortex and spinal cord, differences in immune regulation, and drug metabolism systems, etc.To overcome these limitations of mouse models of ALS, there has been increasing interest in human iPSC technologies for disease modeling and drug development for ALS (Okano et al., 2020;Okano & Morimoto, 2022).Several groups have identified candidate drugs for ALS using iPSCs-based drug screening with drug repositioning, such as ezogabine (retigabine), bosutinib, and ropinirole, and concluded early clinical trials.

| E ZOG AB INE (RE TI G AB INE ): Kv7 P OTA SS IM CHANNEL AC TIVATOR
Ezogabine (retigabine) has been studied for its potential therapeutic effects in ALS, initially identified by Brian Waigner and Clifford Woolf at Boston Children's Hospital and Harvard Stem Cell Institute.Previous research, including our own and other groups, has demonstrated that MNs derived from iPSCs of ALS patients with SOD1 and fused-in sarcoma (FUS) mutations, exhibit hyperexcitability, as observed through microelectrode array (MEA) recording (Kiskinis et al., 2014).This hyperexcitability is believed to contribute to the vulnerability and subsequent cell death of ALS MNs.Therefore, targeting the suppression of hyperexcitability and subsequent degeneration in MNs could be a promising approach for ALS treatment, similar to the case of agonists of a G protein-coupled receptor (GPCR) that is coupled with Gi, such as ropinirole.In this context, the Kv7.2/7.3 ion channel activator retigabine, used as an antiepileptic drug based on its ability to regulate hyperexcitability, is an excellent target, providing a rationale for the use of this drug in a clinical trial in ALS (Wainger et al., 2014(Wainger et al., , 2021)).This double-blind, placebo-controlled phase II randomized trial (clinicaltrials.govidentifier: NCT02450552) enrolled 65 participants and demonstrated that ezogabine (a form of retigabine) reduced the excitability of both upper and lower MNs in ALS patients.However, it remains uncertain whether ezogabine can effectively slow down the progression of ALS.
There is an increasing interest in a safer, more potent, and selective Kv7.2/7.3 ion channel opener, QRL-101 for the new clinical trial program that started in 2023 (clinicaltrials.govidentifier: NCT05667779).

| BOSUTINIB
There is an increasing interest in the potential of bosutinib, a tyrosine kinase inhibitor originally developed for chronic myelogenous leukemia, as a candidate therapeutic agent for ALS.The research groups led by Brigitte van Zundert at Universidad Andres Bello in Chile and Haruhisa Inoue at Kyoto University in Japan independently found that bosutinib, through inhibition of the Src/c-Abl kinase pathway, could improve the survival of ALS MNs, as demonstrated using human SOD1 G93A transgenic ALS model mice model (Rojas et al., 2015) or SOD1 iPSC-derived MNs for their survival rate (Imamura et al., 2017).Notably, bosutinib exhibited autophagy-promoting properties and reduced the accumulation of misfolded SOD1 protein in iPSCs-derived MNs.Furthermore, it showed increased survival in both FALS patients with TDP-43 mutation and SALS patients.Recent findings from the iDReAM study (clinicaltrials.govidentifier: NCT04744532), a multicenter, open-label, dose-escalation phase 1 trial, provided insights into the safety, tolerability, and exploratory efficacy of bosutinib in ALS patients (Imamura et al., 2019(Imamura et al., , 2022)).In this study, following the enrollment of 20 patients, 13 individuals underwent bosutinib treatment, and 12 were included in the safety and efficacy analyses to evaluate its effects further.The study presented promising outcomes regarding the safety and tolerability of bosutinib, as well as encouraging trends in its therapeutic effects.However, it is important to note that this study was conducted as an open-label trial within a limited timeframe and with a small patient population.

| ROPINIROLE
Ropinirole was identified as a candidate anti-ALS drug using iP-SCs-technology by our group at Keio University in Japan (Fujimori et al., 2018).At the initial stage of iPSC-based drug development, developed experimental methods to generate human spinal cord MNs using human iPSCs derived from healthy controls and FALS patients with FUS and TARDBP gene mutations and to search for superior and novel therapeutic agents (Ichiyanagi et al., 2016).Interestingly, iPSCs-MNs derived from ALS patients exhibited ALS-related phenotypes in a dish, such as neurite retraction, stress granule formation, mislocalization of FUS/transactivating response element DNA-binding protein of 43 kDa (TDP-43, encoded by TARDBP gene) proteins, and MN degeneration.Taking advantage of these in vitro properties of ALS iPSCs-MNs, we performed phenotypic screening to identify drugs that suppress these ALS-related phenotypes using an existing (FDA-approved) drug library.In the first screening, FUS ALS iPSCs-MNs were used and 95 drugs were selected, and then in the secondary screening, 9 drugs were selected from TARDBP ALS iPSCs-MNs.Among them, after considering the drug's ability to cross the blood-brain barrier and its safety, including side effects, we selected ropinirole hydrochloride (ROPI), a drug already marketed as a treatment (agonist) for Parkinson's disease (an incurable neurological disease that causes slow movement, stiffness, and tremors in the body and limbs), as the most promising drug for ALS.Research using these cells showed that a drug already on the market for treating Parkinson's disease was best suited for ALS.Further in vitro studies showed the following findings.First, the anti-ALS effect of ROPI is mediated by both dopamine D2 receptor (D2R)-dependent and independent mechanisms.D2R is a Gi-coupled GPCR that downregulates cAMP levels and is expected to suppress the hyperexcitability phenotype of neurons.Indeed, ROPI suppressed the hyperexcitability of ALS patient-derived iPSCs-MNs in a dish.Interestingly, among various dopamine receptors, D2R was selectively expressed in hiPSCs-derived MNs and MNs in the postmortem spinal cord tissue.These results are consistent with the D2R-dependent anti-ALS action model of ROPI.Regarding the D2R-independent mechanisms, ROPI is expected to scavenge ROS, protect mitochondria, and prevent apoptosis through its oxindole structure with antioxidant activity and its localization to the inner mitochondrial membrane (Okano et al., 2020).Second, the anti-ALS effect of ROPI is mediated by both D2R-dependent and independent mechanisms.As mentioned above, ROPI suppressed the hyperexcitability of iPSCs-MNs derived from FALS.Thus, it is likely to suppress the death-promoting mechanisms of ALS MNs degeneration.ROPI also suppressed the retrograde degeneration and poor axon extension phenotype of ALS iPSCs-MNs.Third, in vitro results suggest that ROPI is superior to existing anti-ALS drugs (riluzole and edaravone).
Finally, among patients with SALS, approximately 70% were ROPI responders in the in vitro assay.

| ROPALS TRIAL (INVESTIGATOR-INITIATED CLINICAL TRIALS: TRANSLATIONAL RESEARCH)-ROPI CAN BE USED SAFELY IN ALS PATIENTS AND MAY SLOW DISEASE PROGRESSION
In this single-center, randomized, double-blind, placebo-controlled clinical trial, the primary endpoints were safety and tolerability and the secondary endpoints included the therapeutic effects of ROPI.
Twenty patients with SALS, age 20 to 80, grade 1 or 2 of ALS severity classification, ALS Functional Rating Scale-Revised (ALSFRS-R) score ≥2 points on all items participated in this trial (Morimoto et al., 2023).
The ALSFRS-R score represents the severity of ALS symptoms and is scored out of 12 for 4 items (respiratory function, bulbar function, fine motor function, and gross motor function) for a total of 48 points.
Thirteen patients were initially administered ROPI, while 7 patients received a placebo for 6 months.Subsequently, all patients received ROPI for an additional 6 months.It is worth noting that all patients were able to tolerate the maximum prescribed dose of ROPI (16 mg) without discontinuing oral medication due to any adverse effects.This observation solidifies the conclusion that ROPI can be safely used for treating ALS patients.
Moreover, the impact of ROPI on ALS was evaluated by assessing the ALSFRS-R score, which gauges the severity of ALS symptoms throughout the entire 12-month duration of the study.The results suggested that ROPI might potentially decelerate the progression of motor symptoms.Furthermore, when analyzing the survival period in terms of time until death or specific disease progression, the ROPI group displayed a potential delay in disease advancement of approximately 27.9 weeks (equivalent to around 7 months) over the course of the 1-year study.This finding holds significance for patients, considering that the average time from ALS onset to death typically ranges between 2 and 4 years.
However, it should be noted that the ROPALS trial is a small-sized trial and that the efficacy of ROPI in ALS should be definitively investigated in larger-scale, multi-center, and international trials.

| LIMITATI ON S FOR THE RE SULTS OF THE ROPAL S TRIAL
As described above, the ROPALS trial has some limitations that require further validation, including small sample sizes (N = 20) and high attrition rates observed during the open-label extension period.Furthermore, the results regarding efficacy in this study are mainly from the continuous dosing period, the period during which both active drug and placebo groups are taking ROPI.To address this point, we compared the ALSFRS-R transition data for the true placebo group that did not take ROPI for 1 year with the actual drug group data from the ROPALS trial using a large international database of ALS patient information (Pooled Resource Open-Access ALS Clinical Trials Database) (Atassi et al., 2014), leveraging the power of Big Data.As a result, we concluded that the ROPI group significantly reduced the decline in ALSFRS-R (Morimoto et al., 2023).

| RE VER S E TR ANS L ATIONAL RE S E ARCH IN THE ROPAL S TRIAL
In the ROPALS trial, concurrent with clinical trials, reverse translational research was undertaken from the following perspectives, including the evaluation of the drug effect in vitro and clinically using MNs derived from patient-specific iPSCs (Morimoto et al., 2023).
Moreover, potential biomarkers were explored in blood and spinal fluid samples, with the aim of distinguishing responders to ROPI and assessing ALS disease progression.Concurrently with the clinical drug administration study, iPSCs were generated from all 20 patients in the ROPALS trial, and the impact of ROPI was examined by analyzing the neurite length of these MNs over time.The results showed that MNs from patients exhibited pathogenic phenotypes, such as shorter and thinner neurites, compared to those of healthy individuals, and ROPI treatment rejuvenated their functionality.
Most notably, patients whose iPSC-derived MNs responded robustly to ROPI demonstrated higher efficacy in the clinical study, as indicated by the ALSFRS-R trend.These results suggest that iPSCs-MNs models may serve as predictive markers for drug effectiveness.
Comprehensive transcriptome analysis by time-series RNA-seq using iPSCs-MNs derived from ALS patients by changes in gene expression after ROPI treatment suggested the presence of 29 gene groups whose expression was downregulated by ROPI.Interestingly, these 29 gene groups were significantly enriched in a group of enzymes involved in the cholesterol synthetic pathway.Cholesterol is synthesized from acetate in a complex and energy-consuming process involving more than 20 reactions in 5 major steps, and ROPI has been shown to largely suppress gene expression of many of the enzymes involved in these steps.Since cholesterol cannot be easily degraded, excess cholesterol is metabolized to either more soluble forms, i.e. oxysterols, or more hydrophobic forms, i.e. cholesteryl esters (CEs), to maintain intracellular cholesterol levels.CEs are the storage form of cholesterol and accumulate in cytoplasmic lipid droplets along with other neutral lipids such as triglycerides and fatty acids.They can be used to supply metabolic energy or membrane components as needed.
Cholesterol is esterified by acyl-coenzyme A: cholesterol acyltransferase 1 (ACAT1, also known as SOAT1, sterol O-acyltransferase), an enzyme that is mainly expressed by neurons in the brain, suggesting that neurons may be more sensitive to elevated cholesterol levels.In addition to cholesterol, 24(S)-hydroxycholesterol (24S-OHC) is a substrate for ACAT1, which generates oxysterol esters.
In relation to our research, the analysis of transcription factors indicated that a group of 29 genes involved in cholesterol biosynthesis is regulated by sterol regulatory element-binding protein 2 (SREBP2), which is a transcription factor bound to the cell membrane responsible for activating genes involved in cholesterol synthesis (Brown & Goldstein, 1997).SREBP2 is located in cellular compartments where cholesterol biosynthesis occurs, such as the endoplasmic reticulum (ER).Under conditions of cholesterol depletion, SREBP2 is transported from the ER to the Golgi complex, where it undergoes cleavage by proteases site 1 protease (S1P) and site 2 protease (S2P).
Subsequently, the N-terminal domain of SREBP2 is released into the cytoplasm, where it interacts with importin-β and is carried into the nucleus (Sato et al., 1999).Inside the nucleus, SREBP2 activates the expression of more than 30 genes related to cholesterol biosynthesis, including HMG-CoA reductase (HMGCR) and those inhibited by ROPI treatment (Horton et al., 2002).
Correspondingly, in the ALS patients-derived iPSC-MNs, significant improvement in normalized maximum neurite length was observed when inhibiting the activation of SREBP2 with fatostatin or inhibiting HMGCR with statin compounds, particularly with highly brain-permeable lipophilic statins like atorvastatin and pitavastatin (Morimoto et al., 2023).Thus, the suppression of the SREBP2dependent cholesterol biosynthetic pathway (CBP) was likely to be involved in the therapeutic effect of ROPI in ALS (Figure 1).

| LIPID ME TABOLIS M AND AL S
Recent studies have shown dysmetabolism of various lipid classes, such as fatty acids, glycerolipids, glycerophospholipids, sphingolipids, and sterols, in ALS patients (Agrawal et al., 2022;Liu et al., 2020;Nabizadeh et al., 2022).Lipids are believed to play a significant role in ALS pathogenesis and hold promise as both biomarkers and therapeutic targets.ALS patients often exhibit hypermetabolism at the body level, such as fatty acid oxidation (β-oxidation leading to excess production of acetyl-CoA and cholesterol synthesis in mitochondria, F I G U R E 1 Mechanism of action of ropinirole hydrochloride (ROPI) in amyotrophic lateral sclerosis and bridging between clinical and cellular phenotypes in the ROPALS study.ROPI is thought to prevent motor neuron degeneration by inhibiting the CBP, similar to the SREBP2 activation inhibitor fatostatin and the HMG-CoAR inhibitors statins.In addition, based on the analysis of the clinical and basic parts of the ROPALS study, iPS cell-derived motor neurons (MNs) generated from patients whose symptoms were well improved by ROPI (the responder group) had particularly high expression of genes related to CBP, and the abnormal gene expressions were also improved by ROPI treatment compared to the suboptimal responder group.In addition, the motor neuron phenotype was also better improved in the responder group, suggesting that iPS cell-derived MNs could be a surrogate marker for the effect of ropinirole and could also be used as a stratification marker.HMG-CoAR, hydroxymethylglutaryl-CoA reductase; iPS, induced pluripotent stem; ROPALS, ROPI drug for ALS; SREBP2, sterol regulatory element-binding protein 2. mainly in skeletal muscles) (Ludolph et al., 2023), although the exact cause remains unknown.
Alterations in cholesterol and its metabolites, among other lipids, have been observed in the plasma and cerebrospinal fluid (CSF) of ALS patients, with several studies reporting hyperlipidemia (Hartmann et al., 2022).Further research is needed to fully understand how lipid dysmetabolism contributes to ALS pathogenesis and to develop effective treatments.Interestingly, the anti-ALS effects of ROPI (ropinirole) appear to involve the regulation of lipid metabolism beyond the cholesterol biosynthetic system.Specifically, ROPI treatment has been shown to increase eicosanoid biosynthesis while decreasing lipid peroxidation, suggesting a potential role for ferroptosis in ALS motor neuron cell death (Fujimori et al., 2018).
The ROPALS study, which analyzed blood and CSF samples from ALS patients, identified CSF lipid peroxide as a potential marker for assessing the efficacy of ROPI treatment and disease progression (Morimoto et al., 2023).This finding further supports the involvement of lipid dysmetabolism in ALS and highlights the potential of targeting lipid-related pathways for therapeutic intervention.
Up until now, our research has primarily focused on studying cholesterol metabolism and lipid abnormalities in MNs as related to ALS.However, it is important to note that neurons themselves do not store lipid droplets; instead, they receive lipids from astrocytes' lipid droplets during energy-demanding situations (Agrawal et al., 2022).This raises the possibility that abnormalities in lipid metabolism within glial cells (astrocytes, oligodendrocytes, and microglia) may play a role in the development of ALS.Interestingly, studies have already shown that lipid metabolism irregularities in glial cells are closely linked to the pathogenesis of other neurodegenerative diseases, such as Alzheimer's disease.For example, astrocytes carrying the APOE4 gene, a major risk factor for late-onset Alzheimer's, exhibit increased cholesterol synthesis despite cholesterol accumulation in lysosomes.Additionally, dysregulation of the matrisome, which includes extracellular matrix proteins and associated factors, in astrocytes co-cultured with neurons is associated with increased chemotaxis, glial activation, and lipid biosynthesis.Similar alterations in astrocyte matrisome signaling have been observed in the human brain (Julia et al., 2022).
As mentioned earlier, excess cholesterol is converted into oxysterols or CEs when cholesterol levels surpass a certain threshold, and it is not yet clear if changes in CE levels occur before the onset of symptoms and neurodegeneration.However, the accumulation of CE itself could potentially contribute to neurodegeneration.One of the mechanisms by which CEs cause neuronal damage is reported to be ER dysfunction and the accompanying regulated IRE1-dependent decay (RIDD)-mediated pro-death ubiquitinproteasome system (UPR) signaling and global protein synthesis inhibition (Urano et al., 2019).
Male SALS patients have been found to exhibit elevated levels of CE and triacylglycerols in the gray matter of the spinal cord, along with reduced levels of HMGCR, a key enzyme in cholesterol synthesis.Such ALS-related reductions in HMGCR have been suggested to be part of a compensatory response to lower lipid levels to mitigate lipid cacostasis (Dodge et al., 2020).The toxicity of CE may be attributed to one of its by-products, lysophosphatidylcholine (lyso-PC), as evidenced by a decrease in the survival of human MNs derived from iPSCs when exposed to lyso-PC in vitro and accumulation of lyso-PC in the spinal cords of SOD1 G93A mice and ALS patients.Moreover, when SREBP2 (a transcription factor crucial for cholesterol synthesis) was overexpressed in the spinal cord of normal mice to mimic CE accumulation, it led to ALS-like lipid pathology, motor neuron death, astrogliosis, paralysis, and reduced survival, indicating that CNS neutral lipid accumulation alone is sufficient to induce an ALS phenotype.Thus, we believe that ROPI exerts its therapeutic effect on ALS pathomechanism by reducing toxic lipids (lyso-PC and CE) by suppressing the expression of SREBP2 and thus HMGCR.
Several genes associated with ALS have been found to regulate lipid homeostasis, further implicating the role of lipids in the disease.
Notably, Dodge and colleagues demonstrated that lipid dysregulation is a prominent feature of ALS and inducing it in wild-type mice is enough to trigger an ALS-like phenotype.Changes in SREBP2, HMGCR, and cholesterol precursors in the context of the disease are likely part of a compensatory response to counteract CE accumulation in the CNS.Importantly, Dodge et al. (2020) showed that increasing the expression of the transcriptionally active form of SREBP2, a key driver of cholesterol synthesis, in wild-type mice led to ALS-like lipid cacostasis in the spinal cord, astrogliosis, MN death, paralysis, and reduced lifespan.
Levels of the cholesterol metabolite 25-hydroxycholesterol (25OHC) showed a substantial association with disease severity and progression in patients with ALS, suggesting its potential use as a biomarker (Kim et al., 2017;Vejux et al., 2018).In NSC34 motor neuron-like cell lines carrying the human G93A mutant SOD1 gene (mSOD1-G93A), 25OHC induced motor neuron (MN) death or apoptosis via the glycogen synthase kinase-3β and liver X receptor pathways, a process that could be alleviated by riluzole treatment.In addition, an increase in the expression of enzymes associated with 25OHC synthesis was found in the brains of early symptomatic mSOD1 G93A mutant mice, suggesting a possible link to ALS pathology (Kim et al., 2017).
Interestingly, the presence of ALS-associated mutations in serine palmitoyltransferase long chain base subunit 1 (SPTLC1) has been found to affect the activity and/or substrate specificity of serine palmitoyltransferase (SPT).This results in the accumulation of toxic lipids (Lone et al., 2022).Elimination of SPTLC1 mutants has shown promise in reducing the accumulation of these potentially harmful lipids, and lowering toxic lipid levels has shown beneficial effects in restoring cellular phenotypes, suggesting a potential link between lipid dysfunction and ALS pathogenesis.Of note, Agrawal et al. (2022) highlighted that specific toxic lipids, such as ceramides and arachidonic acid, have particularly detrimental effects on MNs, while reducing their accumulation has shown beneficial results in various ALS models.In addition, increased levels of potentially beneficial lipids, such as glucosylceramides, and activation of S1P-mediated signaling have been identified Overall, these findings suggest that TDP-43 plays a role in cholesterol homeostasis and implicates cholesterol dysmetabolism in ALS pathology.

| NE W TOPI C S IN AL S RE S E ARCH
There has been rapid progress in research into the pathogenesis of ALS and the development of treatments using new technologies, including iPSCs technology, single-cell RNA-seq, and antisense oligonucleotide technology.We would like to introduce a few studies that deserve special attention.

| Cryptic RNA splicings of the targets of TDP-43
The RNA-binding protein called TDP-43 is not only a causative gene for FALS but also has also been found to disappear from the nucleus of MNs in SALS.It forms fibrous aggregates consisting of RNA and proteins in the cytoplasm, playing a crucial role in the pathogenesis of the disease.In the nucleus, TDP-43 is involved in various aspects of transcriptional and post-transcriptional regulation (such as splicing regulation, stabilization of long introns, binding on long non-coding RNA, and miRNA biogenesis) (Warraich et al., 2010).Therefore, if the TDP-43 protein changes its localization from the nucleus to the cytoplasm in ALS MNs, the loss of the nuclear function of TDP-43 could be a biomarker for ALS and be associated with the appearance of ALS pathology.Among them, cryptic splicing-polyadenylation of the stathmin-2 (STMN2) gene, encoding a protein required for axonal regeneration, is caused by TDP-43 mislocalization and is likely to be involved for axonal phenotypes and vulnerability of ALS MNs (Klim et al., 2019;Melamed et al., 2019).It is remarkable that ASOs targeted for TDP-43 binding sites within human STMN2 gene suppressed cryptic splicing, which restored axonal regeneration and STMN2-dependent lysosome trafficking in TDP-43-deficient human MNs (Baughn et al., 2023).This suggests the merit of ASO targeting STMN2 cryptic splicing for clinical application.TDP-43 also acts as a main repressor of the inclusion of UNC13A cryptic exons, which is a genetic risk factor for ALS and frontotemporal dementia (FTD) (Brown et al., 2022;Koike et al., 2023;Ma et al., 2022;Mehta et al., 2023).How the inclusion of UNC13A cryptic exons is involved in the pathogenesis of these ALS/FTD remains to be elucidated.

| Large scale library of iPSCs-MNs from ALS patients
The use of iPSCs presents a promising avenue for gaining insights into the underlying mechanisms of neurological diseases.ALS, especially SALS, is a heterogeneous disease, and it is essential to establish and analyze iPSCs from a large number of ALS patients in order to understand and stratify the full picture of the disease, including its diversity.However, a significant limitation in this field is the scarcity of well-curated iPSC lines encompassing a diverse range of participants.To address this issue, Answer ALS has successfully generated an extensive collection of over 1000 iPSC lines derived from both healthy individuals and those diagnosed with ALS, accompanied by valuable clinical and whole-genome sequencing data.(Workman et al., 2023).
Also in Japan, our group at Keio University established a large iPSCs cohort more than 200 derived from FALS, SALS, and healthy control subjects with clinical information, collaborated with the Japanese Consortium for Amyotrophic Lateral Sclerosis Research (JaCALS) at Aichi Medical University (Nakamura et al., 2023) and Japan Familial ALS Trial-ready registry (J-FAST) at Tohoku University (Suzuki et al., 2023).

| Single-cell transcriptome analysis of human spinal cord tissues
To comprehend the structure and functionality of a healthy spinal cord, as well as the underlying mechanisms of diseases such as ALS and spinal cord injury, it is crucial to understand the diverse cellular

| FUTURE ISSUE S AND PER S PEC TIVE S
As described in this review, the development of drug discovery research for ALS using iPSCs is progressing, including clinical trials, but the following issues remain (Figure 2): 1.The molecular mechanisms of the dying forward and dying back mechanisms that lead to MN degeneration in ALS are not yet uniquely understood, and there are many unknowns.The analysis will be advanced with a focus on the neuronal axon-regulatory and MNs-resilience factors, such as Phox2B (Mitsuzawa et al., 2021).
2. Phenotypic analysis of ALS using iPSCs has so far focused on spinal cord-type lower MNs and has not analyzed upper MNs themselves, the interaction between upper MNs and lower MNs, and the synapse formation and interaction between lower MNs and muscles.For this analysis, it would be meaningful to use the assembloid constructed by the organoids of the cerebral cortex, spinal cord, and muscles derived from iPSCs (Kelley & Pașca, 2022).
3. The cellular phenotypes and drug discovery assay systems for ALS iPSCs-MNs do not include glial cells, which prevents the analysis of cell non-autonomous factors related to ALS.Any glial cell (e.g.astrocytes, oligodendrocytes, and microglia) is thought to lose its support function for MNs or gain injurious properties against MNs (Taylor et al., 2016).All these glial cells can now be induced from iPSCs, and their effects on MNs and ALS pathology can be studied.
4. The disease is diverse from case to case, and it is necessary to stratify the SALS patients and find the optimal treatment for each group.This stratification could be achieved by introducing functional epigenomic analysis into iPSC technology: comprehensive sequencing analysis using Native Elongating Transcript Cap Analysis of Gene Expression and Mutation Effect prediction on ncRNA transcription methods, as well as genetics and machine learning approaches.Approaches, such as the Polygenic Risk Score, could be incorporated to stratify patients based on genetic signatures (Okano & Morimoto, 2022).The worldwide availability of iPSCs derived from ALS patients with whole genome sequences and other data (e.g., Answer ALS) is increasing, and epigenomic analysis using these iPSCs will help us to better understand the stratification of SALS and its pathogenesis.5.In approximately 30% of patients with SALS, the phenotype does not emerge in iPSC-derived MNs.In such cases, the direct induction of fibroblasts into MNs may be effective (iMN method).The phenotypic analysis of SALS using the iMN method, in combination with item 4 above, should facilitate the stratification of patients with SALS.The RNA-binding protein TDP-43 is associated with both familial and sporadic ALS.In ALS, TDP-43 translocates from the nucleus to the cytoplasm, forms fibrous aggregates, and contributes to the pathogenesis of the disease.TDP-43 is involved in several regulatory processes in the nucleus.Mislocalization of TDP-43 is associated with cryptic splicing polyadenylation of the stathmin-2 (STMN2) gene, affecting axonal regeneration and vulnerability of ALS MNs.Antisense oligonucleotides (ASOs) targeting STMN2 cryptic splicing show promise in restoring axonal regeneration.TDP-43 also suppresses the inclusion of cryptic Unc-13 homolog A (UNC13A) exons, a genetic risk factor for ALS and frontotemporal dementia (FTD).(b) Large-scale library of iPSCs-MNs from ALS patients.iPSCs offer promising insights into neurological diseases such as ALS.However, the lack of well-curated iPSC lines is a limitation.Answer ALS, JaCALS, and J-FAST have generated large scale iPSC lines from healthy and ALS patients with comprehensive clinical and molecular summary.These resources advance ALS research and highlight the importance of addressing variability in future studies.(c) Single-cell transcriptome analysis of human spinal cord tissues.Single-cell transcriptomics reveals a cellular taxonomy of the adult human spinal cord.Specifically, they identify and categorize a variety of neuronal and glial clusters based on anatomical location and specific genes associated with their vulnerability in ALS motor neurons.ALS, amyotrophic lateral sclerosis; iPSC, induced pluripotent stem cell; JaCALS, Japanese Consortium for Amyotrophic Lateral Sclerosis Research; J-FAST, Japan Familial ALS Trialready registry; TDP-43, TAR DNA-binding protein of 43 kDa.Supervision; writing -review and editing.Shinichi Takahashi: Funding acquisition; supervision; writing -original draft; writingreview and editing.
as potential protective factors in ALS.Regarding the relationship between the ALS-associated protein TDP-43 and cholesterol metabolism, Ho et al. (2021) demonstrated that TDP-43 directly binds to the mRNA of SREBF2 and several other mRNAs encoding proteins involved in cholesterol biosynthesis and uptake, including HMGCR, 3-Hydroxy-3-Methylglutaryl-CoA Synthase 1 (HMGCS1), and low-density lipoprotein receptor (LDLR).Depletion of TDP-43 in oligodendrocytes resulted in reduced SREBP2 and cholesterol levels both in vitro and in vivo, which could be rescued by the reintroduction of SREBP2 or LDLR.In contrast, Egawa et al. (2022) reported that TDP-43 overexpression in MNs impaired SREBP2 transcriptional activity, resulting in the inhibition of cholesterol biosynthesis and decreased cholesterol levels observed in the spinal cord of TDP-43-overexpressing ALS model mice and in the CSF of ALS patients.These discrepancies in results may reflect differences in cell types, including lipid utilization capacity.It is also assumed that the pathogenesis of TDP-43 may differ depending on the stage of ALS, making it more important to use patient-derived cells, where simple knockout or overexpression of TDP-43 would be difficult to recapitulate the pathogenesis of ALS.
In a notable contribution, Clive N Svendsen's research group at Cedar Sinai Medical Center has provided a comprehensive summary of cell marker expression and gene activity within MN cultures.These cultures were established using the MN differentiation protocol and comprised samples derived from 92 healthy control subjects and 341 individuals with ALS.It is worth noting that this study represents the largest dataset of iPSCs differentiated into MNs to date.The findings of this research shed light on key factors affecting cellular composition and variability, specifically highlighting the influence of cell type and the participants' sex.It is imperative to carefully control for these factors in future investigations to ensure reliable and meaningful results.This study serves as an important foundation for further exploration and emphasizes the significance of meticulously addressing sources of variability in subsequent iP-SC-based studies of neurological diseases composition and taxonomy of the spinal cord.Researchers from Columbia University(Yadav et al., 2023) and Stanford University(Gautier et al., 2023) have conducted a study aimed at establishing a cellular taxonomy of the adult human spinal cord using singlenucleus RNA sequencing in conjunction with spatial transcriptomics and antibody validation.The team, led by Yadav et al. (2023), successfully identified and categorized 29 glial clusters and 35 neuronal clusters, primarily based on their anatomical location.Additionally, the researchers characterized spinal MNs from individuals with ALS to demonstrate the applicability of this resource to the study of ALS.They found that human MNs are characterized by specific genes associated with cell size and cytoskeletal structure, shedding light on the molecular basis of their selective vulnerability in ALS.The study also provides a web resource that can facilitate further investigations into the biology of the human spinal cord.

6.
The current process of inducing differentiation of MNs from iPSCs and determining therapeutic response to drugs, including ROPI, is time-consuming, while the time the patient's symptoms progress.To overcome this, it will be important to enhance stratification by genomic analysis to predict drug responsiveness and to combine rapid induction of differentiation into MNs with biomarkers related to ALS to determine drug responsiveness quickly.AUTH O R CO NTR I B UTI O N SHideyukiOkano: Conceptualization; funding acquisition; project administration; supervision; writing -original draft; writing -review and editing.Satoru Morimoto: Funding acquisition; visualization; writing -original draft; writing -review and editing.Chris Kato: Writing -original draft; writing -review and editing.Jin Nakahara: F I G U R E 2 New topics in ALS research.(a) Cryptic RNA splicing of TDP-43 targets.