1 Therapeutic inhibition of ferroptosis in neurodegenerative disease 1 Sean K Ryan1, Cathryn L. Ugalde2, Anne-Sophie Rolland3, John Skidmore2, David Devos3, and 2 Timothy R. Hammond1* 3 1. Sanofi, Rare and Neurologic Diseases, Cambridge, MA, USA. 4 2. The ALBORADA Drug Discovery Institute, University of Cambridge, Island Research 5 Building, Cambridge Biomedical Campus, Hills Road, Cambridge, CB2 0AH, UK 6 3. Department of Medical Pharmacology, Expert Center of Parkinson’s Disease ALS and 7 neurogenetic, University of Lille, LilNCog, Lille Neuroscience & Cognition, Inserm, 8 INSERM UMR-S1172, CHU de Lille LICEND COEN Center NS-PARK Network, 9 France 10 * Corresponding Author 11 Keywords: Ferroptosis, neurodegeneration, therapeutics, Parkinson’s disease 12 Abstract 13 Iron accumulation has been associated with the etiology and progression of multiple 14 neurodegenerative diseases. The exact role of iron in these diseases is not fully understood, but an 15 emerging mechanism of iron-dependent form of regulated cell death called ferroptosis could be 16 key. While there is substantial pre-clinical and clinical evidence for ferroptosis in disease, there 17 are many remaining questions regarding the role of ferroptosis inhibition in treatment, including 18 which proteins to target, what the clinically relevant biomarkers are, and which patients might 19 benefit most. Clinical trials with iron- and ferroptosis-targeted therapies are beginning to provide 20 some answers, but there is growing interest in developing new ferroptosis inhibitors. Here we 21 describe newly identified ferroptosis targets, opportunities and challenges in neurodegenerative 22 disease, and key considerations for progressing to the clinic. 23 2 24 Iron dysregulation in neurodegeneration 25 Neurodegenerative diseases (NDDs) represent a heterogenous group of conditions 26 associated with the progressive damage and loss of neurons. The mechanisms underlying 27 neurodegeneration are complex and poorly understood. Accordingly, clinical trials have had poor 28 success rates, with a below average likelihood of final approval for targets that have entered phase 29 I trials [1]. However, recent advances in genome wide association studies (GWAS), genomic and 30 proteomic approaches, and animal and cellular models of disease have led to identification of new 31 targets and recently approved treatments in Alzheimer’s (AD), amyotrophic lateral sclerosis 32 (ALS), and other NDDs. 33 A shared pathological feature of several NDDs is iron accumulation, which occurs in 34 affected brain regions and correlates with progression[2-4]. While the precise role of iron across 35 the spectrum of NDDs is unclear, interest in it as a driver of neurodegeneration is increasing due 36 to advanced imaging approaches to track brain iron levels in patients as well as emerging genetics 37 revealing that mutations in iron storage and handling genes cause adult onset neurodegeneration 38 [5]. For example, individuals with neuroferritinopathy (see glossary) suffer from adult-onset 39 basal ganglia degeneration, Parkinsonism, dystonia, and dementia as a result of a loss-of-function 40 mutation in the gene FTL which encodes the ferritin light chain [6]. Restoring normal brain iron 41 balance can be difficult given the complexity of brain iron import, export, and storage. Brain 42 penetrant iron chelators have been used in the clinic with mixed results, and it is thought that, in 43 addition to potential benefits, these chelators might disrupt important iron-dependent physiological 44 functions in cells [7, 8]. 45 3 The mechanism of iron-driven neurotoxicity is still not well understood, but a new form of 46 iron-mediated cell death called ferroptosis has emerged as a key target. The discovery of iron 47 accumulation in NDD dates back to the early 20th century when the neurodegeneration with brain 48 iron accumulation (NBIA) disorder Pantothenate kinase-associated neurodegeneration (PKAN), 49 also known as Hallervorden‒Spatz disease, was first described [9]. However, it was only recently 50 that the term ferroptosis was coined to describe a regulated mechanism of iron-dependent death 51 [10]. Ferroptosis has since been linked to neurodegeneration [11-14] , and it can be induced in 52 several brain cell types including neurons and microglia [15-17]. Others have reviewed the basic 53 biology and disease implications of ferroptosis [11, 18]. In this review, we will discuss the process, 54 regulation, and emerging molecular targets of ferroptosis, evidence for its role in 55 neurodegeneration, inducers and inhibitors, and lessons from the clinic that will guide therapeutic 56 development in the future. 57 58 59 Mechanisms and regulation of ferroptosis 60 Molecular features of ferroptosis 61 Ferroptosis is an iron-dependent form of cell death distinct from apoptosis [19]. While 62 ferroptosis is a highly conserved process that has molecular regulators, whether it is a programmed 63 form of cell death is not clear. The importance of the pathway in normal physiology have not been 64 well explored. Ferroptosis is driven by three key mechanisms: iron dyshomeostasis, lipid 65 peroxidation, and loss of antioxidant defenses [10, 11]. Ferrous iron (Fe2+) accumulation or 66 redistribution in cells can trigger lipid peroxidation either enzymatically or non-enzymatically 67 through a Fenton reaction [20-23]. Additionally, excess iron may be released through 68 4 ferritinophagy, a process in which ferritin undergoes autophagic clearance, a mechanism which 69 itself has been implicated in multiple NDDs [24]. This process is regulated by NCOA4 which, in 70 turn, is regulated by HIF-1α, HIF-2α, and ATM [25-27]. Outside of TfR1-mediated iron uptake, 71 ZIP8 and ZIP14 can also increase the labile iron pool by transporting ferrous iron into the cell 72 (Fig. 1) [3]. Phospholipid bilayers are highly susceptible to iron-driven oxidation [19] especially 73 those containing polyunsaturated fatty acids (PUFAs), such as arachidonic acid due to their bis-74 allylic hydrogen [28, 29]. This hydrogen has a particularly weak bond, making it highly 75 susceptible to oxidation [30]. Recent evidence in a pancreatic cancer cell line and two neuronal 76 cell lines indicates that lipid peroxidation starts in the endoplasmic reticulum (ER), spreads to 77 mitochondria and then the plasma membrane (PM). However, only lipid peroxidation in the ER 78 and PM was sufficient to induce ferroptosis [31]. These cellular compartment findings need to be 79 validated in more cell types and in vivo to determine their generalizability. To counteract lipid 80 peroxidation, cells employ several antioxidant defense mechanisms. The enzyme glutathione 81 peroxidase 4 (GPX4) reduces lipid peroxides to alcohols [32] using glutathione that is produced 82 with cysteine mainly generated from cystine imported by the glutamate-cystine antiporter, 83 system Xc- , encoded by the genes SLC7A11 (xCT) and SLC3A2 [10, 11]. Three other anti-84 ferroptosis defense mechanisms can work in parallel or independently to GPX4. Ferroptosis 85 suppressor protein 1 (FSP1) generates reduced forms of coenzyme Q10 (CoQ10) and vitamin K to 86 suppress lipid peroxidation [33-36], GTP cyclohydrolase 1 (GCH1) produces the antioxidant 87 tetrahydrobiopterin (BH4), and dihydroorotate dehydrogenase (DHODH) potentially targets and 88 reduces CoQ10 in the mitochondria [37, 38]. However, a recent study suggests DHODH is not 89 involved in ferroptosis suppression, rather the inhibitors of DHODH, leading to ferroptosis, work 90 5 through FSP1 [39]. All three of these systems have been reported to be dysregulated in 91 neurodegenerative disease [40]. 92 93 Functional genomic and small molecule screens to identify regulators of ferroptosis 94 The molecular underpinnings of ferroptosis have emerged from a series of large scale 95 genetic and pharmacological screens. These studies showed that ferroptosis is a regulated form of 96 cell death affecting specific classes of lipids and redox related signaling pathways [17, 28, 33, 34, 97 37, 41-48]. The PUFA synthesis pathway has emerged as a strong regulator of ferroptosis 98 regardless of cell type. One of the most potent regulators of ferroptosis is acyl-CoA synthetase 99 long chain family member 4 (ACSL4) which has been a shared hit in most screens [17, 28, 42-45]. 100 This enzyme attaches CoA groups onto fatty acids including arachidonic and linoleic acid. The 101 modified PUFAs are then integrated into the plasma membrane by lysophosphatidylcholine 102 acyltransferase 3 (LPCAT3), which has also been identified in several screens [29, 43, 45]. Other 103 identified targets include redox-regulators FSP1 [33, 34, 41], DHODH [38], and cytochrome P450 104 oxidoreductase (POR) [46], lipid re-modeler GCH1 [37], ER homeostasis gene SLC39A7, which 105 encodes the zinc transporter ZIP7 [47], SEC24B [17], and cysteine producer cysteinyl-tRNA 106 synthetase (CARS) [48] (Fig. 1). Many other proposed ferroptosis inducers regulate oxidative 107 stress, lipid metabolism, and ER stress which have all been implicated in various NDDs [49-51]. 108 Care must be taken not to overinterpret the importance of individual targets from these 109 screens given that most were identified in immortalized or cancer cell lines. Additional validation 110 in different animal species, including in vivo gene targeting will be important to translate these hits 111 into drug targets. Nevertheless, development of selective and potent inhibitors of these ferroptosis 112 6 regulators will allow validation of these targets in different preclinical disease models and eventual 113 advancement to the clinic. 114 115 116 Ferroptosis and neurodegeneration in the human brain 117 As neurons utilize iron to meet their high energy demands and contain high levels of 118 PUFAs, they are ripe for sensitivity to ferroptosis [52, 53]. In fact, there is increasing evidence of 119 ferroptosis pathway activation in AD, ALS, and PD [2, 54-68] and mutations in ferroptosis-related 120 genes are associated with multiple NDDs. Here we describe the studies linking iron, lipid 121 peroxidation, lipid repair mechanisms, and ferroptosis in AD, ALS, and PD, and provide insight 122 into how these findings may support the strategic targeting of ferroptosis in these diseases. 123 124 Alzheimer’s disease 125 Alzheimer’s disease (AD) is a slow progressing, neurodegenerative disease pathologically 126 identified by the presence of extracellular senile plaques in the CNS composed of amyloid-β; a 127 cleavage fragment of membrane-bound amyloid precursor protein (APP), and intracellular 128 neurofibrillary tangles composed of hyperphosphorylated tau protein. Disease progression often 129 spans decades, with patients first presenting with mild cognitive impairment (MCI) that progresses 130 at varying rates to AD. Elevations in ferritin and iron, the latter which also positively correlates 131 with transferrin receptor levels, have all been reported in human AD tissue [69, 70], and there is a 132 positive correlation between total iron in postmortem brain samples and the rate of cognitive 133 decline in the years since diagnosis [70], providing compelling evidence that iron dyshomeostasis 134 7 is linked to disease progression. Several studies have also reported altered antioxidant function in 135 AD. xCT, the light-chain subunit of Xc-, is upregulated and GSH levels correlate with the level of 136 cognitive decline in AD brain and are lower compared to age-matched control tissue. However, 137 cell-type specific expression of xCT in the CNS is not fully known. In fact, several studies suggest 138 it is enriched astrocytes [71]. Products of lipid peroxidation are also elevated in AD brain compared 139 to controls [69, 72], further suggesting an impairment in anti-ferroptotic defense. The cause of this 140 impairment is still unclear, but recent studies of familial AD have suggested that presenilin 141 mutations could contribute to ferroptosis susceptibility by suppressing GPX4 expression [73]. 142 143 Amyotrophic lateral sclerosis 144 Amyotrophic lateral sclerosis (ALS) is a rapidly progressive NDD characterized by the 145 loss of motor neurons in the brain and spinal cord and presence of intracellular ubiquitylated 146 proteinaceous inclusions. Numerous data implicate oxidative stress and mitochondrial dysfunction 147 in the loss of neurons in disease; an observation that is supported by genetic evidence whereby 148 inherited disease-causing mutations have been linked to proteins directly associated with oxidative 149 stress pathways, including superoxide dismutase I (SOD1) and TAR DNA-binding protein 43 150 (TDP-43) [74]. Multiple studies suggest antioxidant capacity is limited in ALS tissue. For 151 example, GSH depletion has been observed in disease-relevant brain areas [75], with GSH levels 152 inversely correlating with the time interval between diagnosis and imaging [76]. Downregulation 153 of GPX4 [68] and increased lipid peroxide product 4-hydroxynonenal (4-HNE) has been reported 154 in spinal cord tissue of patients at post-mortem [77]. Lipid peroxidation is also elevated in the 155 blood from patients affected by ALS ante mortem [78], providing evidence that these disease-156 associated changes occur prior to the onset of late-stage disease. Together with studies reporting 157 8 abnormal iron deposition in disease-associated brain regions in ALS patients [79, 80], there is 158 collectively strong evidence that pathways relevant to ferroptosis occurs in ALS. In the SOD1G93A 159 mouse model, overexpression of GPX4 or the ferroptosis suppressor SPY1 provides protection 160 from disease, significantly extending mouse lifespan. These findings were recapitulated in other 161 mouse and human neuronal cell line models of ALS [68, 81, 82]. Together the data implicate 162 deficiencies in ferroptosis preventing pathways in the development of this disease. 163 Parkinson’s disease 164 Parkinson’s disease (PD) is a neurodegenerative disease associated with the presence of 165 intracellular inclusions that contain misfolded α-synuclein, called Lewy bodies (LB) and the 166 progressive loss of dopaminergic neurons in the substantia nigra (SN) that causes significant motor 167 disability [83]. Misfolded α-synuclein possesses pathogenic properties including the capacity to 168 spread between cells via seeded propagation and cause neurotoxicity [84]. Several disease-169 associated SNCA point mutations (such as A53T and A30P) lead to an increased propensity to 170 misfold compared to wild-type protein [85]. α-Synuclein has well established links to iron and 171 ferroptosis-associated mechanisms, including switching iron uptake from TfR1-mediated to 172 DMT1-mediated [86, 87] (Fig. 2). In addition, the amount of soluble α-synuclein correlates with 173 levels of lipid peroxidation [54], which others have shown is increased in PD [54, 64, 88], and α-174 synuclein can increase cytosolic dopamine which can produce oxidative stress [89]. SNCA 175 triplication is also associated familial PD, and iPSC-derived neurons generated from a patient with 176 SNCA triplication are more susceptible to iron-dependent lipid-peroxidation and ferroptosis [54, 177 64, 88]. In another study, patient iPSC-derived dopaminergic neurons with SNCA triplication were 178 more vulnerable to ferroptosis and exhibited increased lipid peroxidation, which was partially 179 blocked by α-synuclein knockdown. Lipid profiling of these neurons revealed that SNCA or ACSL4 180 9 knockdown reduced ether-linked phospholipids, which are linked to ferroptosis [55], but the role 181 α-synuclein plays in this process is still unclear. 182 Several other Parkinson’s-associated or causal genes are also linked to ferroptosis 183 including SLC7A11, DJ-1, and PLA2G6 (Fig. 2). Methylation changes in the glutamate-cystine 184 antiporter gene SLC7A11 have been linked to PD. A recent study involving roughly 1,000 case 185 and control blood samples identified a hypermethylated CpG, cg06690548, in the promoter region 186 of SLC7A11 that was associated with disease and caused down regulation of SLC7A11. The 187 resultant SLC7A11 loss-of-function reduces glutathione production, impairing GPX4 function and 188 increasing ferroptosis. Pharmacological inhibition of Xc- is one of the primary methods to induce 189 ferroptosis in vitro [10] and impaired GSH production in PD [90] could be a direct result of altered 190 SLC7A11 expression or function, even in idiopathic cases. In one patient brain analysis, SLC7A11 191 expression was significantly reduced in the substantia nigra of PD patients, while GPX4 expression 192 was increased, suggesting a compensatory mechanism [91]. However, in an MPTP model of PD, 193 Slc7a11-/- mice showed no difference in loss of dopaminergic neurons in the substantia nigra 194 compared to littermate controls [92], and Slc7a11 knockout even protected against neuron loss in 195 the 6-OHDA PD model and did not cause any changes in glutathione levels. A major reduction in 196 extracellular glutamate was found in the knockout mice, highlighting a need to better understand 197 the role of Xc- in cystine uptake and glutamate transport in neurodegeneration [93]. 198 199 Mutations in DJ-1, also known as Parkinsonism associated deglycase (PARK7), cause early 200 onset Parkinson’s disease [94]. Disease-causing mutations lead to loss of function and may affect 201 DJ-1’s ability to dimerize as well as its ATP synthase activity [95, 96]. In addition, DJ-1 has also 202 been linked to oxidative stress protection, but its exact mechanism is unknown [97]. Disease-203 10 causing mutations in DJ-1, including M26I, E64D, A104T, and L166P, increase ferroptosis 204 susceptibility [98]. DJ-1 is involved in the transsulfuration pathway, which can metabolize 205 cystathionine into cysteine for production of glutathione [48] in cases where cystine levels are low. 206 DJ-1 inhibits adenosylhomocysteinase like 1 (AHCYL1) which, in turn, inhibits S-adenosyl 207 homocysteine hydrolase (SAHH) an enzyme involved in the transsulfuration pathway to generate 208 cysteine. Loss of function of DJ-1 allows AHCYL1 to repress SAHH, thus preventing cysteine 209 production through the transsulfuration pathway. This would be expected to induce ferroptosis 210 only under conditions where sufficient cystine cannot be imported through system xCT [98]. 211 However, in a mouse model, neuron-specific knockouts of Gpx4 and DJ-1 resulted in mild PD-212 like phenotypes but no loss of dopaminergic neurons [99]. Repression may also be necessary in 213 non-neuronal cells to induce a more severe phenotype. 214 215 Loss of function mutations in the PLA2G6 gene can cause one of the NBIA diseases 216 PLA2G6-associated neurodegeneration, PLAN, which is characterized by Parkinsonism [100]. 217 PLA2G6 encodes phospholipase A2 group VI, a phospholipase that hydrolyzes fatty acids from 218 phospholipids. It has an affinity for arachidonic acid and has been shown to remove 15-HpETE 219 from phosphatidyl ethanolamine, preventing ferroptosis induction. Fibroblasts from PD patients 220 with PLA2G6 mutations have increased susceptibility to ferroptosis, and genetically-engineered 221 mice harboring a mutation in the orthologous gene Pnpla9 developed a Parkinson’s-like phenotype 222 as well as accumulation of 15-HpETE-PE [101]. Vitamin E, a lipophilic antioxidant and known 223 ferroptosis inhibitor, reduced lipid peroxidation and iron accumulation in cultured cells from 224 PLAN patients [102]. 225 226 11 These gene mutations, in addition to those directly involved in iron regulation, converge 227 on known ferroptosis regulatory pathways and suggest a causal neurodegenerative mechanism for 228 ferroptosis in Parkinson’s. However, there is a need for further investigation into the complexities 229 of both in vitro and in vivo models to better understand the caveats of each approach and 230 translatability to human disease. Overall, it is still unclear whether these mutations can also trigger 231 non-ferroptotic mechanisms to influence disease onset and progression. 232 Ferroptosis regulation by small molecules 233 Ferroptosis inducers 234 To induce ferroptosis experimentally, several small molecules have been identified that 235 target different parts of the pathway. Erastin was initially reported as a tool compound causing 236 gain-of-function of mitochondrial voltage-dependent anion channels (VDACs) triggering the 237 generation of oxidative species [103]. Later studies suggested that VDAC modulation is not 238 sufficient for triggering ferroptosis, and Stockwell proposed that in Jurkat T lymphocytes erastin 239 also targets the SLC7A5/SLC3A2 system-L neutral amino acid transporter whilst inhibiting the 240 Xc– cystine-glutamate antiporter in trans [10]. Subsequent work suggested a more direct inhibition 241 of system Xc– in Ht-1080 and Calu-1 cells [104]. It is generally thought that this latter mechanism 242 is most important, with a blockade of cystine import leading to glutathione deficiency triggering 243 ferroptosis. 244 Like erastin, RAS-selective lethal 3 (RSL3) was discovered through a synthetic lethality 245 cell-based screen [105]. RSL3 is a chiral molecule with an electrophilic warhead, properties that 246 were used in an affinity-based proteomics experiment to show that the potent stereoisomer 247 selectively binds to a small number of protein targets, of which GPX4 is the most prominent. RSL3 248 has also been shown to inhibit GPX4 function in lysates from COH-BR1 GPX4 over-expressing 249 12 cells and thus is well characterized as a cell-active, selective blocker of this essential antioxidant 250 peroxidase [106]. Apparent cellular sensitivity to RSL3 can vary greatly. In some cell types 251 concentrations below 10nM are sufficient to trigger ferroptotic cell death, whereas in other cells 252 micromolar concentrations are required. These differences may reflect, inherent sensitivity to the 253 pathway in different cell types, the impact of different media with varying antioxidant content, or 254 off target effects at the highest doses [55]. There are many other pharmacological inducers of 255 ferroptosis that target either GPX4 ( ML162, ML210, FIN56) or system Xc– (sulfasalazine, and 256 glutamate) [11]. Glutamate is especially interesting considering glutamate-mediated toxicity is 257 strongly connected to neurodegeneration [107]. 258 Ferroptosis inhibitors 259 The first reported inhibitors of ferroptosis, Ferrostatin-1 and Liproxstatin-1 [10, 108], were 260 discovered by phenotypic screening. The mechanism of these radical trapping antioxidants 261 (RTA) has been studied in detail including their increased activity in lipid bilayers [109, 110]. 262 These compounds have been useful tools for cell biology since they are very potent inhibitors of 263 ferroptosis with low nanomolar potency in vitro. However, since they operate through a non-264 specific mechanism rather than inhibiting a specific protein target, combined with the long history 265 of ineffective antioxidants in neurodegeneration [111], their translational and mechanistic potential 266 has been limited. It is possible that specific RTAs with limited off target activity and improved in 267 vivo pharmacology could show efficacy in the clinic, but that remains to be seen. Another set of 268 compounds that inhibit ferroptosis are probucol analogues, which act as GPX4 activators [112]. 269 An alternative approach to blocking the radical chain reaction responsible for PUFA 270 oxidation is stabilization of the fatty acid substrates through supplementation of natural PUFAs 271 with their deuterated analogues or to directly target PUFA synthesis. To reduce PUFA 272 13 susceptibility to oxidation, D-PUFAs were developed, and these are converted into radicals much 273 more slowly due to the kinetic isotope effect. D-PUFAs are brain permeable and are incorporated 274 into cells in the brain. They have shown some promising effects in a number of animal models of 275 neurodegeneration and have progressed to the clinic [113]. D-PUFAs have been shown to block 276 both erastin- and RSL3-induced ferroptosis (Box 1) as well as α-synuclein oligomer induced 277 toxicity in cellular assays [30, 88]. 278 Alternatively, several protein targets have been identified in genetic screens that increase 279 ferroptosis susceptibility by controlling PUFA production and incorporation into the cell 280 membrane. One such target, the enzyme ACSL4, activates PUFAs by forming acyl-CoA 281 derivatives which are then converted into phospholipids by lysophosphatidylcholine acyl 282 transferase 3. ACSL4 knockout effectively blocks ferroptosis in vitro [28], but very few potent 283 ACSL4 inhibitors exist. A series of PPAR-ɣ agonists – rosiglitazone, pioglitazone and troglitazone 284 have been shown to inhibit ACSL4 with IC50 values of around 1 µM [114]. These glitazones have 285 been shown to prevent RSL3-induced ferroptosis and lipid peroxidation in Pfa1 cells [28]. 286 Glitazones can have antioxidant properties, so ferroptotic inhibition may be due to off-target 287 effects [115]. Recently, a series of azole antifungal agents were also found to be ACSL4 inhibitors. 288 Of these the most potent, sertaconazole, has an IC50 of 280 nM against ACSL4 and was shown to 289 block RSL3 induced lipid peroxidation and protect cells from ferroptosis in LUHMES cells (Fig. 290 1) [116]. Downstream of ACSL4 and LPCAT3 is arachidonate 15-lipoxygenase, 15-LOX, which 291 can oxidize arachidonic acid into the ferroptotic lipid peroxide 15-HpETE as well as oxidizing 292 arachidonate esters, and while not identified in genetic screens, ALOX15 knockdown prevents 293 ferroptosis in some cell lines [16, 21, 22, 29]. However, there is controversy over its true role in 294 ferroptosis and the specificity of existing 15-LOX inhibitors has been questioned [23, 30, 109, 295 14 117]. More ALOX15 knockout data in vitro and in vivo will be needed to validate the target and 296 binding assays will be important to determine the specificity of putative inhibitors. 297 Lastly, ferroptosis can be prevented by targeting iron directly. Iron chelators bind to iron 298 in cells and promote excretion from the body [118]. Extensive preclinical data in a range of animal 299 models supports iron chelation as a therapeutic approach [119, 120]. Iron chelators prevent 300 ferroptosis by blocking iron-mediated Fenton reaction-driven lipid peroxidation, but they also 301 block other iron-dependent physiological mechanisms, which can lead to off-target side effects 302 and toxicity. There is a great need in the neurodegeneration field for CNS-penetrant ferroptosis 303 inhibitors with properties suitable for clinical studies in order to establish the potential of this 304 mechanism for the treatment of disease. Notably, several iron chelators and RTAs have gone into 305 the clinics and failed. 306 307 Ferroptosis in the clinic: past, present, and future 308 A number of agents under clinical investigation have mechanisms of action related to iron 309 biology or antioxidant activity that give them the potential to block ferroptosis. The most widely 310 used in the clinic are iron chelators (Fig. 1). Of the many iron chelators tested in ferroptosis related 311 preclinical studies [120], deferoxamine (DFO) and deferiprone (DFP) have been studied in 312 humans. However, the potential off-target effects as well as the wide range of homeostatic 313 functions that iron performs provides significant challenges to overcome for iron chelators to be 314 effective treatments. 315 DFO, which is not blood-brain barrier penetrant, was tested in Alzheimer’s disease using 316 intramuscular administration in a 2-year single-randomized blind study with 48 patients [121]. It 317 was reported that the treatment reduced the rate of clinical progression of AD dementia by half 318 15 compared with controls, although the biological mechanism is still unclear given the lack of CNS 319 exposure. DFP is a blood-brain barrier penetrant iron chelator which redistributes captured iron by 320 reducing it in overloaded areas and redeploying it [122], a process called conservative iron 321 chelation. The ability of DFP to reduce iron levels only in overloaded areas has now been 322 demonstrated in PD and ALS [123-125]. These studies demonstrated by MRI that iron was reduced 323 only in overloaded areas associated with degeneration but not in healthy areas of the central 324 nervous system. 325 Two single-center pilot studies involving respectively 40 and 22 patients with early PD, 326 showed a positive symptomatic effect of DFP at 30 mg/kg/day in patients who were also receiving 327 dopaminergic treatments [123, 126]. On the other hand, a recent European multicenter study 328 showed that deferiprone at 30 mg/kg/day in 372 de novo dopaminergic treatment-naive patients 329 worsened their UPDRS Part III motor score [124]. These studies showed that it might be critical 330 to maintain dopaminergic therapy while dosing iron chelators because iron chelation can interfere 331 with dopamine synthesis by depriving tyrosine hydroxylase of its iron cofactor. A new trial is 332 underway to test that hypothesis (NCT02728843, Table 1). 333 DFP has also been tested in other neurodegenerative diseases. In ALS, DFP treatment for 334 three months in 23 patients with sporadic disease delayed functional decline compared with the 335 previous 3-month period without treatment [125]. A multicenter, parallel-group, placebo-336 controlled, randomized clinical trial of about 200 ALS patients for 12 months is due to end in late 337 2023 (Table 1). In AD, a randomized, placebo-controlled clinical trial with DFP is underway in 338 2023 (NCT03234686, Table 1), but as yet no results have been released. 339 In addition to iron targeting therapeutics, several lipid peroxidation inhibitors have been 340 tested in clinical trials. General antioxidants, some of which inhibit ferroptosis in vitro, have been 341 16 tested in the clinic without much success (Table 1). One Phase IIb trial administered GSH, the 342 cofactor of GPX4, intranasally for three months in 45 PD patients. The cohorts receiving either 343 dose of GSH did not improve beyond the placebo group [127]. Another study aimed to increase 344 GSH activity indirectly by administering N-acetyl cysteine as a daily oral supplement or by 345 intravenous administration, but given the low brain penetrance of N-acetyl cysteine, high doses 346 were required to increase GSH activity (Table 1) [128]. Several trials have also evaluated CoQ10, 347 which is reduced by FSP1, but low doses were not sufficient to have a clinical effect (Table 1) 348 [33, 129]. It would be beneficial to understand the pharmacokinetic and pharmacodynamic 349 properties of CoQ10 to determine if sufficient brain exposure was achieved. Another emerging, 350 potent anti-ferroptotic molecule is copper(II)-diacetyl-bis(N4-methylthiosemicarbazone) 351 (CuATSM) (Fig. 1), developed for ALS with radical-trapping, antioxidant and anti-ferroptotic 352 properties that localizes to the mitochondria [130, 131]. A randomized, double-blind, placebo-353 controlled phase II study of 80 ALS patients at four clinical sites in Australia (NCT04082832) is 354 underway. Finally, the reported 15-LOX inhibitors vatiquinone and utreloxastat (Fig. 1) from PTC 355 bio have been tested in Friedreich’s Ataxia (FA) and ALS, respectively. A Phase III MOVE-FA 356 trial was recently completed for vatiquinone and a phase II/III clinical trial for utreloxastat in 258 357 ALS patients (NCT05349721) recently began. 358 One indirect ferroptosis inhibitor, an NRF2 activator small molecule called omaveloxolone 359 (Fig.1), was recently approved for use in FA [132]. FA is a neurological disease caused by 360 expansion repeats in the FXN gene that cause reduced expression of frataxin, an iron handling 361 protein found in the mitochondria [133]. Ferroptosis has been implicated in FA with iron 362 accumulation, oxidative stress, reduced glutathione, and increased susceptibility to cell death in 363 FA patient iPSC-derived neurons, increased lipid peroxidation in cerebellar granule cells in a 364 17 humanized mouse model, and increased levels of the lipid peroxide marker malondialdehyde in 365 patient plasma [134-136]. NRF2 is a transcription factor and master regulator of cellular 366 antioxidant pathways. Activation leads to altered expression of pathways that improve 367 mitochondrial function and increase glutathione levels. In addition to omaveloxolone, NRF2 can 368 be activated by dimethyl fumarate (DMF) and monomethylfumarate (MMF) (Fig. 1) [137]. NRF2 369 also regulates genes in ferroptosis related pathways including FTH1 and FTL in the iron 370 metabolism pathway and SLC7A11 and GPX4 in the glutathione/antioxidant pathway [49]. In 371 addition, pharmacological and genetic inhibition of NRF2 increased sensitivity to ferroptosis, 372 whereas activation of NRF2 prevented ferroptosis in pre-clinical models [138-140]. Considering 373 the evidence, omaveloxolone may, in part, be effective in FA by inhibiting ferroptosis, although 374 more studies are required to confirm this. 375 As more anti-ferroptotic therapies emerge, the development of specific imaging and biological 376 ferroptosis biomarkers will be important. Specific biomarkers will not only allow direct 377 measurement of target engagement in the brain but could also have surrogate value for the 378 prognosis of the disease, patient identification, and prediction of therapeutic response. Given that 379 ferroptosis is defined by iron accumulation and high levels of lipid peroxidation a number of 380 existing approaches may be applicable. Iron accumulation can be readily monitored by T2* and 381 QSM sequences in magnetic resonance imaging (MRI) [141, 142] and ferritin levels in serum and 382 plasma by ELISA [143, 144]. Similarly, lipid peroxidation can be analyzed with 4-hydroxy-2,3-383 nonenal, malonaldehyde in patient blood and CSF [143, 145] but these approaches have not been 384 validated in the context of ferroptosis. 385 386 Concluding Remarks and Future Perspectives 387 18 Collectively, there is compelling human evidence that ferroptosis plays an important role 388 in neurodegenerative disease, but it remains to be seen whether ferroptosis actively drives disease 389 pathogenesis or is a secondary consequence. Nevertheless, new potent ferroptosis inhibitors could 390 be important disease modifying therapies to delay disease progression, particularly in iron 391 accumulating diseases and potentially in the more common NDDs. Linking genetic and patient 392 data will help us understand this relationship, especially by leveraging our understanding of 393 monogenic iron mishandling diseases to understand more prevalent sporadic diseases featuring 394 iron accumulation. One challenge for the field is the translatability of cellular and animal models 395 of neurodegeneration. These typically represent one aspect of the disease like protein misfolding, 396 disrupted gene translation, or altered redox state, but rarely replicate all the key hallmarks of 397 neurodegeneration and/or ferroptosis [146, 147]. Developing and understanding the strengths and 398 weaknesses of these models in the context of ferroptosis will be critical for drug development and 399 testing before the clinic. 400 Another important aspect to understanding the role of ferroptosis in human disease will be 401 defining the cellular responses of different brain cell types to iron overload and how it leads to 402 ferroptosis. In particular, further investigation into the role of microglia and the inflammation 403 response may reveal new insight, given the prevalence of iron accumulating microglia in brain 404 tissue, sensitivity of microglia to ferroptotic stimuli, and microglial ferroptotic signatures in MS, 405 PD, and other diseases [17, 148]. Beyond microglia, genetic screens have shown that there could 406 be cell-specific regulation of ferroptosis, and a comprehensive meta-analysis of existing and future 407 datasets would be helpful to understand how to inhibit ferroptosis in specific tissues and diseases. 408 More screening of CNS specific cell types is also needed to identify shared and unique pathways 409 in the brain. 410 19 Interest in activating or inhibiting ferroptosis in different diseases has led to an explosion 411 of research in recent years in both cancer [149] and neurodegeneration that is further defining this 412 unique form of cell death. The next step will be finding the best therapeutic strategies to translate 413 these findings to the clinic. In addition, designing clinical trials to ensure the highest chance of 414 success is crucial, and picking the right patient subpopulation, drug dose, or disease timing could 415 be critical for the development of new ferroptosis inhibitors. These early trials have laid the 416 groundwork for identifying those parameters that could lead to success in the future. Identification 417 of novel molecular targets will allow rational drug design and could lead to brain penetrant small 418 molecules and biologics for iron accumulating diseases. 419 Outstanding Questions: 420 What are the appropriate biomarkers/molecular targets for ferroptosis? 421 How do you reduce iron load without disturbing homeostatic functions? 422 How can we better translate pre-clinical models to disease? 423 Can an effective, targeted RTA / iron chelator be developed? 424 What is the relative importance of different CNS cell types in ferroptosis-mediated 425 neurodegeneration? 426 What are the best anti-ferroptotic targets for neuroprotection? 427 Are there normal physiologic functions of ferroptosis? If so, what are they, and are they necessary 428 in the brain? 429 Do sublethal levels of iron dysregulation and lipid peroxidation contribute to induction or 430 progression of disease? 431 432 20 433 434 435 436 Acknowledgements 437 JS and CLU thank Alzheimer’s Research UK (grant: ARUK-2020DDI-CAM) and acknowledge support from 438 the ALBORADA Trust. The ALBORADA Drug Discovery Institute is core funded by Alzheimer’s Research UK 439 (registered charity no. 1077089 and SC042474). 440 441 References 442 443 1. Mullard, A., Parsing clinical success rates. Nature Reviews Drug Discovery, 2016. 15(7): p. 447-444 447. 445 2. Sato, T., et al., Aging is associated with increased brain iron through cortex-derived hepcidin 446 expression. eLife, 2022. 11: p. e73456. 447 3. Tian, Y., et al., Iron Metabolism in Aging and Age-Related Diseases. International Journal of 448 Molecular Sciences, 2022. 23(7): p. 3612. 449 4. Salami, A., et al., Elevated neuroinflammation contributes to the deleterious impact of iron 450 overload on brain function in aging. NeuroImage, 2021. 230: p. 117792. 451 5. Spaull, R.V.V., et al., Towards Precision Therapies for Inherited Disorders of Neurodegeneration 452 with Brain Iron Accumulation. Tremor and Other Hyperkinetic Movements, 2021. 11(1): p. 51. 453 6. Reinert, A., et al., Iron concentrations in neurons and glial cells with estimates on ferritin 454 concentrations. BMC Neurosci, 2019. 20(1): p. 25. 455 7. Crichton, R., Iron Deficiency, Iron Overload and Therapy, in Iron Metabolism. 2016. p. 376-417. 456 8. Andreini, C., et al., The human iron-proteome. Metallomics, 2018. 10(9): p. 1223-1231. 457 9. Foley, P.B., D.J. Hare, and K.L. Double, A brief history of brain iron accumulation in Parkinson 458 disease and related disorders. J Neural Transm (Vienna), 2022. 129(5-6): p. 505-520. 459 10. Dixon, Scott J., et al., Ferroptosis: An Iron-Dependent Form of Nonapoptotic Cell Death. Cell, 2012. 460 149(5): p. 1060-1072. 461 11. Stockwell, B.R., Ferroptosis turns 10: Emerging mechanisms, physiological functions, and 462 therapeutic applications. Cell, 2022. 185(14): p. 2401-2421. 463 12. Seiler, A., et al., Glutathione peroxidase 4 senses and translates oxidative stress into 12/15-464 lipoxygenase dependent- and AIF-mediated cell death. Cell Metab, 2008. 8(3): p. 237-48. 465 13. Wirth, E.K., et al., Cerebellar Hypoplasia in Mice Lacking Selenoprotein Biosynthesis in Neurons. 466 Biological Trace Element Research, 2014. 158(2): p. 203-210. 467 14. Wirth, E.K., et al., Neuronal selenoprotein expression is required for interneuron development and 468 prevents seizures and neurodegeneration. The FASEB Journal, 2010. 24(3): p. 844-852. 469 15. Chen, L., et al., Overexpression of ferroptosis defense enzyme Gpx4 retards motor neuron disease 470 of SOD1G93A mice. Scientific Reports, 2021. 11(1): p. 12890. 471 16. Kapralov, A.A., et al., Redox lipid reprogramming commands susceptibility of macrophages and 472 microglia to ferroptotic death. Nat Chem Biol, 2020. 16(3): p. 278-290. 473 21 17. Ryan, S.K., et al., Microglia ferroptosis is regulated by SEC24B and contributes to 474 neurodegeneration. Nature Neuroscience, 2022: p. 1-15. 475 18. Costa, I., et al., Molecular mechanisms of ferroptosis and their involvement in brain diseases. 476 Pharmacology & Therapeutics, 2023. 244: p. 108373. 477 19. Wiernicki, B., et al., Excessive phospholipid peroxidation distinguishes ferroptosis from other cell 478 death modes including pyroptosis. Cell Death Dis, 2020. 11(10): p. 922. 479 20. Stoyanovsky, D.A., et al., Iron catalysis of lipid peroxidation in ferroptosis: Regulated enzymatic or 480 random free radical reaction? Free Radic Biol Med, 2019. 133: p. 153-161. 481 21. Wenzel, S.E., et al., PEBP1 Wardens Ferroptosis by Enabling Lipoxygenase Generation of Lipid 482 Death Signals. Cell, 2017. 171(3): p. 628-641 e26. 483 22. Anthonymuthu, T.S., et al., Resolving the paradox of ferroptotic cell death: Ferrostatin-1 binds to 484 15LOX/PEBP1 complex, suppresses generation of peroxidized ETE-PE, and protects against 485 ferroptosis. Redox Biol, 2021. 38: p. 101744. 486 23. Shah, R., M.S. Shchepinov, and D.A. Pratt, Resolving the Role of Lipoxygenases in the Initiation and 487 Execution of Ferroptosis. ACS Central Science, 2018. 4(3): p. 387-396. 488 24. Guo, F., et al., Autophagy in neurodegenerative diseases: pathogenesis and therapy. Brain 489 Pathology, 2018. 28(1): p. 3-13. 490 25. Wu, H., et al., ATM orchestrates ferritinophagy and ferroptosis by phosphorylating NCOA4. 491 Autophagy, 2023. 19(7): p. 2062-2077. 492 26. Li, X., et al., NCOA4 is regulated by HIF and mediates mobilization of murine hepatic iron stores 493 after blood loss. Blood, 2020. 136(23): p. 2691-2702. 494 27. Mancias, J.D., et al., Quantitative proteomics identifies NCOA4 as the cargo receptor mediating 495 ferritinophagy. Nature, 2014. 509(7498): p. 105-109. 496 28. Doll, S., et al., ACSL4 dictates ferroptosis sensitivity by shaping cellular lipid composition. Nat Chem 497 Biol, 2017. 13(1): p. 91-98. 498 29. Kagan, V.E., et al., Oxidized arachidonic and adrenic PEs navigate cells to ferroptosis. Nature 499 Chemical Biology, 2017. 13(1): p. 81-90. 500 30. Yang, W.S., et al., Peroxidation of polyunsaturated fatty acids by lipoxygenases drives ferroptosis. 501 Proc Natl Acad Sci U S A, 2016. 113(34): p. E4966-75. 502 31. Krusenstiern, A.N.v., et al., Identification of essential sites of lipid peroxidation in ferroptosis. 503 Nature Chemical Biology, 2023: p. 1-12. 504 32. Forcina, G.C. and S.J. Dixon, GPX4 at the Crossroads of Lipid Homeostasis and Ferroptosis. 505 Proteomics, 2019. 19(18): p. e1800311. 506 33. Doll, S., et al., FSP1 is a glutathione-independent ferroptosis suppressor. Nature, 2019. 575(7784): 507 p. 693-698. 508 34. Bersuker, K., et al., The CoQ oxidoreductase FSP1 acts parallel to GPX4 to inhibit ferroptosis. 509 Nature, 2019. 575(7784): p. 688-692. 510 35. Jin, D.-Y., et al., A genome-wide CRISPR-Cas9 knockout screen identifies FSP1 as the warfarin-511 resistant vitamin K reductase. Nature Communications, 2023. 14(1): p. 828. 512 36. Mishima, E., et al., A non-canonical vitamin K cycle is a potent ferroptosis suppressor. Nature, 513 2022: p. 1-6. 514 37. Kraft, V.A.N., et al., GTP Cyclohydrolase 1/Tetrahydrobiopterin Counteract Ferroptosis through 515 Lipid Remodeling. ACS Central Science, 2019. 6(1): p. 41-53. 516 38. Mao, C., et al., DHODH-mediated ferroptosis defence is a targetable vulnerability in cancer. 517 Nature, 2021. 593(7860): p. 586-590. 518 39. Mao, C., et al., Reply to: DHODH inhibitors sensitize to ferroptosis by FSP1 inhibition. Nature, 2023. 519 619(7968): p. E19-E23. 520 40. Li, J., et al., Ferroptosis: past, present and future. Cell Death & Disease, 2020. 11(2): p. 88. 521 22 41. Yoshioka, H., et al., Identification of a Small Molecule That Enhances Ferroptosis via Inhibition of 522 Ferroptosis Suppressor Protein 1 (FSP1). ACS Chemical Biology, 2022. 523 42. Zhang, H.-L., et al., PKCβII phosphorylates ACSL4 to amplify lipid peroxidation to induce ferroptosis. 524 Nature Cell Biology, 2022. 24(1): p. 88-98. 525 43. Dixon, S.J., et al., Human Haploid Cell Genetics Reveals Roles for Lipid Metabolism Genes in 526 Nonapoptotic Cell Death. ACS Chemical Biology, 2015. 10(7): p. 1604-1609. 527 44. Zou, Y., et al., A GPX4-dependent cancer cell state underlies the clear-cell morphology and confers 528 sensitivity to ferroptosis. Nature Communications, 2019. 10(1): p. 1617. 529 45. Zou, Y., et al., Plasticity of ether lipids promotes ferroptosis susceptibility and evasion. Nature, 530 2020. 585(7826): p. 603-608. 531 46. Zou, Y., et al., Cytochrome P450 oxidoreductase contributes to phospholipid peroxidation in 532 ferroptosis. Nat Chem Biol, 2020. 16(3): p. 302-309. 533 47. Chen, P.-H., et al., Zinc transporter ZIP7 is a novel determinant of ferroptosis. Cell Death & Disease, 534 2021. 12(2): p. 198. 535 48. Hayano, M., et al., Loss of cysteinyl-tRNA synthetase (CARS) induces the transsulfuration pathway 536 and inhibits ferroptosis induced by cystine deprivation. Cell Death & Differentiation, 2016. 23(2): 537 p. 270-278. 538 49. Dodson, M., R. Castro-Portuguez, and D.D. Zhang, NRF2 plays a critical role in mitigating lipid 539 peroxidation and ferroptosis. Redox Biology, 2019. 23: p. 101107. 540 50. Sharon, R., et al., Altered Fatty Acid Composition of Dopaminergic Neurons Expressing α-Synuclein 541 and Human Brains with α-Synucleinopathies*. Journal of Biological Chemistry, 2003. 278(50): p. 542 49874-49881. 543 51. Ghemrawi, R. and M. Khair, Endoplasmic Reticulum Stress and Unfolded Protein Response in 544 Neurodegenerative Diseases. International Journal of Molecular Sciences, 2020. 21(17): p. 6127. 545 52. Ndayisaba, A., C. Kaindlstorfer, and G.K. Wenning, Iron in Neurodegeneration – Cause or 546 Consequence? Frontiers in Neuroscience, 2019. 13: p. 180. 547 53. Liu, J.J., et al., Pathways of polyunsaturated fatty acid utilization: Implications for brain function 548 in neuropsychiatric health and disease. Brain Research, 2015. 1597: p. 220-246. 549 54. Fu, Y., et al., Increased unsaturated lipids underlie lipid peroxidation in synucleinopathy brain. Acta 550 Neuropathologica Communications, 2022. 10(1): p. 165. 551 55. Mahoney-Sanchez, L., et al., Alpha synuclein determines ferroptosis sensitivity in dopaminergic 552 neurons via modulation of ether-phospholipid membrane composition. Cell Reports, 2022. 40(8): 553 p. 111231. 554 56. Lee, H.-J., et al., Brain Arachidonic Acid Cascade Enzymes are Upregulated in a Rat Model of 555 Unilateral Parkinson Disease. Neurochemical Research, 2010. 35(4): p. 613-619. 556 57. Emre, C., et al., Age-related changes in brain phospholipids and bioactive lipids in the APP knock-557 in mouse model of Alzheimer’s disease. Acta Neuropathologica Communications, 2021. 9(1): p. 558 116. 559 58. Jové, M., et al., New insights into human prefrontal cortex aging with a lipidomics approach. 560 Expert Review of Proteomics, 2021. 18(5): p. 333-344. 561 59. Tu, J., et al., Absolute quantitative lipidomics reveals lipidome-wide alterations in aging brain. 562 Metabolomics, 2017. 14(1): p. 5. 563 60. Pamplona, R., et al., Redox lipidomics to better understand brain aging and function. Free Radical 564 Biology and Medicine, 2019. 144: p. 310-321. 565 61. Chen, C.T., et al., Regulation of brain polyunsaturated fatty acid uptake and turnover. 566 Prostaglandins, Leukotrienes and Essential Fatty Acids, 2008. 79(3-5): p. 85-91. 567 62. Chappus-McCendie, H., et al., Omega-3 PUFA metabolism and brain modifications during aging. 568 Progress in Neuro-Psychopharmacology and Biological Psychiatry, 2019. 94: p. 109662. 569 23 63. Meng, Q., et al., Ethyl-eicosapentaenoate modulates changes in neurochemistry and brain lipids 570 induced by parkinsonian neurotoxin 1-methyl-4-phenylpyridinium in mouse brain slices. European 571 Journal of Pharmacology, 2010. 649(1-3): p. 127-134. 572 64. Dexter, D., et al., LIPID PEROXIDATION AS CAUSE OF NIGRAL CELL DEATH IN PARKINSON'S DISEASE. 573 The Lancet, 1986. 328(8507): p. 639-640. 574 65. Bouchaoui, H., et al., ACSL4 and the lipoxygenases 15/15B are pivotal for ferroptosis induced by 575 iron and PUFA dyshomeostasis in dopaminergic neurons. Free Radical Biology and Medicine, 2023. 576 195: p. 145-157. 577 66. Xu, J., et al. Impaired Iron Status in Aging Research. International Journal of Molecular Sciences, 578 2012. 13, 2368-2386 DOI: 10.3390/ijms13022368. 579 67. Ghadery, C., et al., R2* mapping for brain iron: associations with cognition in normal aging. 580 Neurobiology of Aging, 2015. 36(2): p. 925-932. 581 68. Wang, T., et al., Ferroptosis mediates selective motor neuron death in amyotrophic lateral 582 sclerosis. Cell Death Differ, 2022. 29(6): p. 1187-1198. 583 69. Ashraf, A., et al., Iron dyshomeostasis, lipid peroxidation and perturbed expression of 584 cystine/glutamate antiporter in Alzheimer’s disease: Evidence of ferroptosis. Redox Biology, 2020. 585 32: p. 101494. 586 70. Ayton, S., et al., Brain iron is associated with accelerated cognitive decline in people with Alzheimer 587 pathology. Molecular Psychiatry, 2020. 25(11): p. 2932-2941. 588 71. Fournier, M., et al., Implication of the glutamate–cystine antiporter xCT in schizophrenia cases 589 linked to impaired GSH synthesis. npj Schizophrenia, 2017. 3(1): p. 31. 590 72. Dahlmanns, M., et al., Glial Glutamate Transporter-Mediated Plasticity: System xc-/xCT/SLC7A11 591 and EAAT1/2 in Brain Diseases. Frontiers in Bioscience-Landmark, 2023. 28(3): p. 57. 592 73. Greenough, M.A., et al., Selective ferroptosis vulnerability due to familial Alzheimer's disease 593 presenilin mutations. Cell Death Differ, 2022. 29(11): p. 2123-2136. 594 74. Hemerková, P. and M. Vališ, Role of Oxidative Stress in the Pathogenesis of Amyotrophic Lateral 595 Sclerosis: Antioxidant Metalloenzymes and Therapeutic Strategies. Biomolecules, 2021. 11(3): p. 596 437. 597 75. In Young, C., et al., Reduction in cerebral antioxidant, glutathione (GSH), in patients with ALS: A 598 preliminary study (P6.105). Neurology, 2015. 84(14 Supplement): p. P6.105. 599 76. Andronesi, O.C., et al., Imaging Neurochemistry and Brain Structure Tracks Clinical Decline and 600 Mechanisms of ALS in Patients. Frontiers in Neurology, 2020. 11: p. 590573. 601 77. Pedersen, W.A., et al., Protein modification by the lipid peroxidation product 4-hydroxynonenal in 602 the spinal cords of amyotrophic lateral sclerosis patients. Annals of Neurology, 1998. 44(5): p. 819-603 824. 604 78. Siciliano, G., et al., Impaired oxidative metabolism and lipid peroxidation in exercising muscle from 605 ALS patients. Amyotrophic Lateral Sclerosis and Other Motor Neuron Disorders, 2002. 3(2): p. 57-606 62. 607 79. Kasarskis, E.J., et al., Aluminum, calcium, and iron in the spinal cord of patients with sporadic 608 amyotrophic lateral sclerosis using laser microprobe mass spectroscopy: a preliminary study. 609 Journal of the Neurological Sciences, 1995. 130(2): p. 203-208. 610 80. Acosta-Cabronero, J., et al., Quantitative Susceptibility MRI to Detect Brain Iron in Amyotrophic 611 Lateral Sclerosis. Radiology, 2018. 289(1): p. 195-203. 612 81. Chen, L., et al., Overexpression of ferroptosis defense enzyme Gpx4 retards motor neuron disease 613 of SOD1G93A mice. Sci Rep, 2021. 11(1): p. 12890. 614 82. Wang, D., et al., SPY1 inhibits neuronal ferroptosis in amyotrophic lateral sclerosis by reducing lipid 615 peroxidation through regulation of GCH1 and TFR1. Cell Death & Differentiation, 2023. 30(2): p. 616 369-382. 617 24 83. Greffard, S., et al., Motor Score of the Unified Parkinson Disease Rating Scale as a Good Predictor 618 of Lewy Body–Associated Neuronal Loss in the Substantia Nigra. Archives of Neurology, 2006. 619 63(4): p. 584-588. 620 84. Ugalde, C.L., et al., Pathogenic mechanisms of prion protein, amyloid-β and α-synuclein misfolding: 621 the prion concept and neurotoxicity of protein oligomers. Journal of Neurochemistry, 2016. 139(2): 622 p. 162-180. 623 85. Li, J., V.N. Uversky, and A.L. Fink, Effect of Familial Parkinson's Disease Point Mutations A30P and 624 A53T on the Structural Properties, Aggregation, and Fibrillation of Human α-Synuclein. 625 Biochemistry, 2001. 40(38): p. 11604-11613. 626 86. Wise, R.M., et al., Interactions of dopamine, iron, and alpha-synuclein linked to dopaminergic 627 neuron vulnerability in Parkinson's disease and Neurodegeneration with Brain Iron Accumulation 628 disorders. Neurobiology of Disease, 2022. 175: p. 105920. 629 87. Chen, B., et al., Interactions between iron and alpha-synuclein pathology in Parkinson's disease. 630 Free Radic Biol Med, 2019. 141: p. 253-260. 631 88. Angelova, P.R., et al., Alpha synuclein aggregation drives ferroptosis: an interplay of iron, calcium 632 and lipid peroxidation. Cell Death Differ, 2020. 27(10): p. 2781-2796. 633 89. Duce, J.A., et al., Post translational changes to α-synuclein control iron and dopamine trafficking; 634 a concept for neuron vulnerability in Parkinson’s disease. Molecular Neurodegeneration, 2017. 635 12(1): p. 45. 636 90. Bjørklund, G., et al., The glutathione system in Parkinson’s disease and its progression. 637 Neuroscience & Biobehavioral Reviews, 2021. 120: p. 470-478. 638 91. Do Van, B., et al., Ferroptosis, a newly characterized form of cell death in Parkinson's disease that 639 is regulated by PKC. Neurobiol Dis, 2016. 94: p. 169-78. 640 92. Bentea, E., et al., Aged xCT-Deficient Mice Are Less Susceptible for Lactacystin-, but Not 1-Methyl-641 4-Phenyl-1,2,3,6- Tetrahydropyridine-, Induced Degeneration of the Nigrostriatal Pathway. 642 Frontiers in Cellular Neuroscience, 2021. 15: p. 796635. 643 93. Massie, A., et al., Dopaminergic neurons of system xc–-deficient mice are highly protected against 644 6-hydroxydopamine-induced toxicity. The FASEB Journal, 2011. 25(4): p. 1359-1369. 645 94. Stephenson, S.E.M., et al., Familial early onset Parkinson's disease caused by a homozygous 646 frameshift variant in PARK7: Clinical features and literature update. Parkinsonism & Related 647 Disorders, 2019. 64: p. 308-311. 648 95. Kumar, A., D. Mukherjee, and P. Satpati, Mutations in Parkinson’s Disease Associated Protein DJ-1 649 Alter the Energetics of DJ-1 Dimerization. Journal of Chemical Information and Modeling, 2019. 650 59(4): p. 1497-1507. 651 96. Chen, R., et al., Parkinson’s disease protein DJ-1 regulates ATP synthase protein components to 652 increase neuronal process outgrowth. Cell Death & Disease, 2019. 10(6): p. 469. 653 97. Repici, M. and F. Giorgini, DJ-1 in Parkinson’s Disease: Clinical Insights and Therapeutic 654 Perspectives. Journal of Clinical Medicine, 2019. 8(9): p. 1377. 655 98. Cao, J., et al., DJ-1 suppresses ferroptosis through preserving the activity of S-adenosyl 656 homocysteine hydrolase. Nat Commun, 2020. 11(1): p. 1251. 657 99. Schriever, S.C., et al., Alterations in neuronal control of body weight and anxiety behavior by 658 glutathione peroxidase 4 deficiency. Neuroscience, 2017. 357: p. 241-254. 659 100. Ferese, R., et al., Heterozygous PLA2G6 Mutation Leads to Iron Accumulation Within Basal Ganglia 660 and Parkinson's Disease. Frontiers in Neurology, 2018. 9: p. 536. 661 101. Sun, W.Y., et al., Phospholipase iPLA2beta averts ferroptosis by eliminating a redox lipid death 662 signal. Nat Chem Biol, 2021. 17(4): p. 465-476. 663 102. Villalón-García, I., et al., Vitamin E prevents lipid peroxidation and iron accumulation in PLA2G6-664 Associated Neurodegeneration. Neurobiology of Disease, 2022. 165: p. 105649. 665 25 103. Yagoda, N., et al., RAS-RAF-MEK-dependent oxidative cell death involving voltage-dependent 666 anion channels. Nature, 2007. 447(7146): p. 864-8. 667 104. Dixon, S.J., et al., Pharmacological inhibition of cystine–glutamate exchange induces endoplasmic 668 reticulum stress and ferroptosis. eLife, 2014. 3: p. e02523. 669 105. Yang, W.S. and B.R. Stockwell, Synthetic Lethal Screening Identifies Compounds Activating Iron-670 Dependent, Nonapoptotic Cell Death in Oncogenic-RAS-Harboring Cancer Cells. Chemistry & 671 Biology, 2008. 15(3): p. 234-245. 672 106. Yang, Wan S., et al., Regulation of Ferroptotic Cancer Cell Death by GPX4. Cell, 2014. 156(1-2): p. 673 317-331. 674 107. Lewerenz, J. and P. Maher, Chronic Glutamate Toxicity in Neurodegenerative Diseases—What is 675 the Evidence? Frontiers in Neuroscience, 2015. 9. 676 108. Angeli, J.P.F., et al., Inactivation of the ferroptosis regulator Gpx4 triggers acute renal failure in 677 mice. Nature Cell Biology, 2014. 16(12): p. 1180-1191. 678 109. Zilka, O., et al., On the Mechanism of Cytoprotection by Ferrostatin-1 and Liproxstatin-1 and the 679 Role of Lipid Peroxidation in Ferroptotic Cell Death. ACS Cent Sci, 2017. 3(3): p. 232-243. 680 110. Miotto, G., et al., Insight into the mechanism of ferroptosis inhibition by ferrostatin-1. Redox Biol, 681 2020. 28: p. 101328. 682 111. Barnham, K.J., C.L. Masters, and A.I. Bush, Neurodegenerative diseases and oxidative stress. 683 Nature Reviews Drug Discovery, 2004. 3(3): p. 205-214. 684 112. Bueno, D.C., et al., New Probucol Analogues Inhibit Ferroptosis, Improve Mitochondrial 685 Parameters, and Induce Glutathione Peroxidase in HT22 Cells. Molecular Neurobiology, 2020. 686 57(8): p. 3273-3290. 687 113. Shchepinov, M.S., Polyunsaturated Fatty Acid Deuteration against Neurodegeneration. Trends in 688 Pharmacological Sciences, 2020. 41(4): p. 236-248. 689 114. Kim, J.-H., T.M. Lewin, and R.A. Coleman, Expression and Characterization of Recombinant Rat 690 Acyl-CoA Synthetases 1, 4, and 5: SELECTIVE INHIBITION BY TRIACSIN C AND 691 THIAZOLIDINEDIONES*. Journal of Biological Chemistry, 2001. 276(27): p. 24667-24673. 692 115. Garg, R., et al., Troglitazone Reduces Reactive Oxygen Species Generation by Leukocytes and Lipid 693 Peroxidation and Improves Flow-Mediated Vasodilatation in Obese Subjects. Hypertension, 2000. 694 36(3): p. 430-435. 695 116. Marteau, R., et al., Repositioning of FDA-Approved antifungal agents to interrogate Acyl-CoA 696 synthetase long chain family member 4 (ACSL4) in ferroptosis. Biochemical Pharmacology, 2022. 697 204: p. 115239. 698 117. Jiang, X., B.R. Stockwell, and M. Conrad, Ferroptosis: mechanisms, biology and role in disease. Nat 699 Rev Mol Cell Biol, 2021. 22(4): p. 266-282. 700 118. Mobarra, N., et al., A Review on Iron Chelators in Treatment of Iron Overload Syndromes. 701 International journal of hematology-oncology and stem cell research, 2016. 10(4): p. 239-247. 702 119. Weinreb, O., et al., Targeting dysregulation of brain iron homeostasis in Parkinson's disease by 703 iron chelators. Free Radical Biology and Medicine, 2013. 62: p. 52-64. 704 120. Devos, D., et al., Conservative iron chelation for neurodegenerative diseases such as Parkinson’s 705 disease and amyotrophic lateral sclerosis. Journal of Neural Transmission, 2020. 127(2): p. 189-706 203. 707 121. McLachlan, D.R.C., et al., Intramuscular desferrioxamine in patients with Alzheimer's disease. The 708 Lancet, 1991. 337(8753): p. 1304-1308. 709 122. Cabantchik, Z.I., et al., Regional siderosis: a new challenge for iron chelation therapy. Frontiers in 710 Pharmacology, 2013. 4: p. 167. 711 123. Devos, D., et al., Targeting Chelatable Iron as a Therapeutic Modality in Parkinson's Disease. 712 Antioxidants & Redox Signaling, 2013. 21(2): p. 195-210. 713 26 124. Devos, D., et al., Trial of Deferiprone in Parkinson’s Disease. New England Journal of Medicine, 714 2022. 387(22): p. 2045-2055. 715 125. Moreau, C., et al., Could Conservative Iron Chelation Lead to Neuroprotection in Amyotrophic 716 Lateral Sclerosis?© Caroline Moreau et al. 2018; Published by Mary Ann Liebert, Inc. This Open 717 Access article distributed under the terms of the Creative Commons License 718 (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and 719 reproduction in any medium, provided the original work is properly cited. Antioxidants & Redox 720 Signaling, 2018. 29(8): p. 742-748. 721 126. Martin-Bastida, A., et al., Brain iron chelation by deferiprone in a phase 2 randomised double-722 blinded placebo controlled clinical trial in Parkinson’s disease. Scientific Reports, 2017. 7(1): p. 723 1398. 724 127. Mischley, L.K., et al., Phase IIb Study of Intranasal Glutathione in Parkinson’s Disease. Journal of 725 Parkinson's Disease, 2017. 7: p. 289-299. 726 128. Holmay, M.J., et al., N-acetylcysteine Boosts Brain and Blood Glutathione in Gaucher and 727 Parkinson Diseases. Clinical Neuropharmacology, 2013. 36(4). 728 129. Investigators, T.P.S.G.Q.E., et al., A Randomized Clinical Trial of High-Dosage Coenzyme Q10 in 729 Early Parkinson Disease: No Evidence of Benefit. JAMA Neurology, 2014. 71(5): p. 543-552. 730 130. Zilka, O., J.-F. Poon, and D.A. Pratt, Radical-Trapping Antioxidant Activity of Copper and Nickel 731 Bis(Thiosemicarbazone) Complexes Underlies Their Potency as Inhibitors of Ferroptotic Cell Death. 732 Journal of the American Chemical Society, 2021. 143(45): p. 19043-19057. 733 131. Donnelly, P.S., et al., An impaired mitochondrial electron transport chain increases retention of 734 the hypoxia imaging agent diacetylbis(4-methylthiosemicarbazonato)copperII. Proceedings of the 735 National Academy of Sciences, 2012. 109(1): p. 47-52. 736 132. Lynch, D.R., et al., Safety and Efficacy of Omaveloxolone in Friedreich Ataxia (MOXIe Study). Annals 737 of Neurology, 2021. 89(2): p. 212-225. 738 133. Lynch, D.R., et al., Safety, pharmacodynamics, and potential benefit of omaveloxolone in 739 Friedreich ataxia. Annals of Clinical and Translational Neurology, 2019. 6(1): p. 15-26. 740 134. Codazzi, F., et al., Friedreich ataxia-induced pluripotent stem cell-derived neurons show a cellular 741 phenotype that is corrected by a benzamide HDAC inhibitor. Hum Mol Genet, 2016. 25(22): p. 742 4847-4855. 743 135. Abeti, R., et al., 'Mitochondrial energy imbalance and lipid peroxidation cause cell death in 744 Friedreich's ataxia'. Cell Death Dis, 2016. 7: p. e2237. 745 136. Emond, M., et al., Increased levels of plasma malondialdehyde in Friedreich ataxia. Neurology, 746 2000. 55(11): p. 1752-1753. 747 137. Majkutewicz, I., Dimethyl fumarate: A review of preclinical efficacy in models of 748 neurodegenerative diseases. European Journal of Pharmacology, 2022. 926: p. 175025. 749 138. Anandhan, A., et al., NRF2 controls iron homeostasis and ferroptosis through HERC2 and VAMP8. 750 Science Advances, 2023. 9(5): p. eade9585. 751 139. Zhao, T., et al., Regulating Nrf2-GPx4 axis by bicyclol can prevent ferroptosis in carbon 752 tetrachloride-induced acute liver injury in mice. Cell Death Discovery, 2022. 8(1): p. 380. 753 140. Cheng, H., et al. Neuroprotection of NRF2 against Ferroptosis after Traumatic Brain Injury in Mice. 754 Antioxidants, 2023. 12, DOI: 10.3390/antiox12030731. 755 141. Moreau, C., et al., Iron as a therapeutic target for Parkinson's disease. Movement Disorders, 2018. 756 33(4): p. 568-574. 757 142. Kuchcinski, G., et al., Quantitative susceptibility mapping demonstrates different patterns of iron 758 overload in subtypes of early-onset Alzheimer’s disease. European Radiology, 2023. 33(1): p. 184-759 195. 760 27 143. Devos, D., et al., A ferroptosis–based panel of prognostic biomarkers for Amyotrophic Lateral 761 Sclerosis. Scientific Reports, 2019. 9(1): p. 2918. 762 144. Nadjar, Y., et al., Elevated Serum Ferritin Is Associated with Reduced Survival in Amyotrophic 763 Lateral Sclerosis. PLOS ONE, 2012. 7(9): p. e45034. 764 145. Simpson, E.P., et al., Increased lipid peroxidation in sera of ALS patients. Neurology, 2004. 62(10): 765 p. 1758-1765. 766 146. Yokoyama, M., et al., Mouse Models of Alzheimer’s Disease. Frontiers in Molecular Neuroscience, 767 2022. 15: p. 912995. 768 147. Khan, E., I. Hasan, and M.E. Haque, Parkinson’s Disease: Exploring Different Animal Model 769 Systems. International Journal of Molecular Sciences, 2023. 24(10): p. 9088. 770 148. Absinta, M., et al., A lymphocyte–microglia–astrocyte axis in chronic active multiple sclerosis. 771 Nature, 2021: p. 1-6. 772 149. Lei, G., L. Zhuang, and B. Gan, Targeting ferroptosis as a vulnerability in cancer. Nature Reviews 773 Cancer, 2022. 22(7): p. 381-396. 774 150. Monti, D.A., et al., N-Acetyl Cysteine May Support Dopamine Neurons in Parkinson's Disease: 775 Preliminary Clinical and Cell Line Data. PLoS ONE, 2016. 11(6): p. e0157602. 776 151. Coles, L.D., et al., Repeated-Dose Oral N-Acetylcysteine in Parkinson's Disease: Pharmacokinetics 777 and Effect on Brain Glutathione and Oxidative Stress. The Journal of Clinical Pharmacology, 2018. 778 58(2): p. 158-167. 779 152. Monti, D.A., et al., N-Acetyl Cysteine Is Associated With Dopaminergic Improvement in Parkinson's 780 Disease. Clinical Pharmacology & Therapeutics, 2019. 106(4): p. 884-890. 781 782 Declaration of Interest 783 S.K.R. and T.R.H. were employees of Sanofi at the time that the review was conducted. The 784 authors declare no other competing interests. 785 786 Glossary 787 α-synuclein: The major misfolded protein in Lewy bodies, which are thought to cause 788 Parkinson’s disease. 789 CoA: A coenzyme that can form acyl CoA derivatives when reacting with an acid. The 790 generation of fatty acid acyl CoA derivatives allows them to be converted to phospholipids that 791 are incorporated into membranes. 792 CpG: Sites in the genome where a cystine is followed by a guanine. 793 D-PUFA (deuterated-poly unsaturated fatty acid): A poly unsaturated fatty acid that has a 794 number of hydrogen atoms replaced with deuterium atoms. 795 Ferritinophagy: Autophagy of the protein ferritin. 796 glutamate-cystine antiporter (Xc-): A heterodimer composed of SLC7A11(xCT) and SLC3A2 797 (4F2hc) that imports cystine into the cell in exchange for glutamate. 798 iron chelators: Molecules that bind free iron, making it inaccessible for cellular functions. 799 28 labile iron pool: Free, reduced ferrous iron in the cell. 800 MPTP: 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine a neurotoxin that localizes to the 801 mitochondria. Utilized to induce a Parkinson’s disease model. 802 N-acetyl cysteine: Antioxidant that can increase glutathione levels. 803 neurodegeneration with brain iron accumulation (NBIA): Set of 10 rare inherited diseases 804 that all present with excessive brain iron accumulation. 805 Neuroferritinopathy: NBIA disorder caused by mutations in the FTL gene. 806 Pantothenate kinase-associated neurodegeneration (PKAN): NBIA disorder caused by 807 mutations in the PANK2 gene. 808 Pharmacodynamic: The effect of the drug on the body. 809 Pharmacokinetic: Kinetics of the drug as it is metabolized by the body. 810 plaques: Buildup of misfolded amyloid proteins. 811 polyunsaturated fatty acids (PUFA): Fatty acids with at least 2 double bonds. 812 PPAR-ɣ: Peroxisome proliferator-activated receptor-gamma. Receptor for glitazones. 813 Presenilin: Transmembrane proteins that are components of the gamma-secretase 814 intramembrane protease protein complex. Mutations in both genes, PSEN1 and PSEN2, cause 815 familial Alzheimer’s disease. 816 QSM: Quantitative susceptibility mapping. A version of magnetic resonance imaging (MRI) that 817 allows for spatial measurement of analytes including iron. 818 Radical-trapping antioxidants (RTA): Antioxidants that react with propagating radicals to 819 prevent further spreading of oxidation. 820 T2*: MRI that highlights fat and water. 821 Transsulfuration: Biological pathway to convert methionine to cysteine. 822 UPDRS Part III motor score: Standardized scale for measuring motor capabilities in patients 823 with Parkinson’s disease.824 29 825 Table 1: Clinical trials with anti-ferroptotic therapeutics 826 Action mechanism Molecules Pathological population Clinical trial phase & design Outcome PubMed reference or Clinical trial ID Anti-ferroptotic mechanism: reduction of the ferroptotic death effector (reduction of intracellular iron), Fenton reaction, and 15-LOX activity Iron chelation Deferoxamine (intramuscular) 48 early Alzheimer Disease Phase II single blind placebo- controlled Significant improvement: reduction in the rate of decline of daily living skills with 125 mg twice daily, 5 days per week, [121] Conservative iron chelation Deferiprone (oral) 40 early PD with dopaminergic treatment Phase II double-blind, placebo- controlled Significant improvement: ~ 3 points of UPDRS part III compared with placebo with 30 mg/kg/d [123] Deferiprone (oral) 22 early PD with dopaminergic treatment Phase II double-blind, placebo- controlled Non-significant improvement: ~2 points of UPDRS part III compared with placebo with 30 mg/kg/d [126] Deferiprone (oral) 372 de Novo PD without dopaminergic treatment Phase II double-blind, placebo- controlled Worsening: ~ 5.8 points compared with placebo with 30 mg/kg/d [124] Deferiprone (oral) 140 PD with dopaminergic treatment Phase II double-blind, placebo- controlled Non-significant improvement: ~ 5 points of improvement compared with placebo in a subgroup of 26 patients with 600 mg BID NCT02728843 Deferiprone (oral) 23 early ALS Phase II Open label Significant improvement: ~2 points of the ALSFRS-r with 30 mg/kg/d [125] Deferiprone (oral) 240 early ALS Phase II double-blind, placebo- controlled Upcoming NCT03293069 Deferiprone (oral) 171 Prodromal and Mild Alzheimer's Disease Phase II double-blind, placebo- controlled Upcoming NCT03234686 SP420 (oral) Planned population: PD Phase II double-blind, placebo- controlled Planned - Siderophore ATH 434 (oral) Multisystem Atrophy Phase II double-blind, placebo- controlled Upcoming NCT05109091 Antiferroptotic action: increase glutathione activity (directly or indirectly by increasing N-acetyl cysteine) to increase GPX4 activity Glutathione increase Glutathione (intranasally) 45 PD Phase II double-blind, placebo- controlled Non-significant improvement: ~ 2.2 points of UPDRS part III compared with placebo with highest dose (600 mg/d) [127] N-acetyl cysteine (intravenous) 3 PD, 3 Gaucher Disease, and 3 healthy controls Phase I open label Single high dose of 150 mg/ kg increased brain glutathione level measured by H-MRS [128] N-acetyl cysteine (oral) 51 PD patients Phase II randomized open label Significant improvement: ~ 12.9% of UPDRS part III compared with placebo [150] N-acetyl cysteine (oral) 5 PD and 3 healthy controls Phase II open-label Repeated low dose of 6000 mg NAC/day over 4 weeks did not change brain glutathione level measured by H- MRS [151] N-acetyl cysteine (intravenous + oral) 42 PD Phase II randomized open label Significant improvement of 2.88 points of UPDRS part III compared with placebo with after 3 months of intravenous infusions (50 mg/kg) + oral doses (500 mg twice per day) [152] Antiferroptotic action: increase coenzyme Q10 to increase Ferroptosis Suppressor Protein 1 activity (FSP1) 30 Coenzyme Q10 (oral) 600 PD untreated Phase II double-blind, placebo- controlled No significant improvement with 1200 mg/d, or 2400 mg/d in association with 1200 IU/d of vitamin E [129] Coenzyme Q10 (oral) 180 prodromal PD Phase II double-blind, placebo- controlled Upcoming NCT04152655 Antiferroptotic action: increase vitamin E to reduce lipid peroxidation Palm oil-derived vitamin E, tocotrienol 100 PD Phase II double-blind, placebo- controlled Upcoming NCT04491383 Antiferroptotic action: Nuclear factor E2 related factor 2 (Nrf2) activator: acting on iron/metal metabolism, intermediate metabolism, and GSH synthesis/metabolism Omaveloxolone (oral) 155 Friedreich ataxia Double-blind, placebo-controlled Significant improvement: -2.40 on mFARS compared with placebo [132] Antiferroptotic action: Lox-15 inhibitor to reduce lipid peroxidation Utreloxastat 258 ALS Double-blind, placebo-controlled Planned NCT05349721 Antiferroptotic action: Copper deliver to mitochondria 80 ALS Double-blind, placebo-controlled Planned NCT04082832 The reference is given. If the study is not yet published the Clinical trial ID is given. UPDRS part III means Movement Disorders 827 Society Unified Parkinson Disease Rating Scale part III (motor part); ALSFRS-R means ALS Functional Rating Scale revised; d 828 means day; MRI means magnetic resonance imaging; 1H-MRS means proton magnetic resonance spectroscopy; IU means 829 International Unit; mFARS means modified Friedreich's Ataxia Rating Scale 830 831 31 832 Figure 1: Sub-cellular localization of genetic and pharmacological targets 833 Targets identified in genetic screens (green) include 1.) membrane-localized regulators: FSP1 reduces the antioxidant CoQ10, 834 preventing lipid peroxidation. ACSL4 and LPCAT3 integrate PUFA into the lipid membrane. 2.) mitochondria-localized regulator: 835 VDAC2/3 prevents oxidative stress. 3.) ER-localized regulators: SEC24B is involved in COPII-vesicle shuttling. POR induces in lipid 836 peroxidation. SLC39A7 is a zinc transporter that helps maintain ER homeostasis. 4.) redox-related regulators: CARS operates in the 837 transsulfuration pathway to generate cysteine, which is necessary for glutathione (GSH) production. GCH1 produces the antioxidant 838 BH4. Pharmacological compounds (blue) includelipophilic antioxidants Vitamin E and liproxstatin-1, iron chelators, NRF2 activators 839 omaveloxolone, DMF and MMF, mitochondrial RTA CuATSM, and inhibitors of the PUFA synthesis pathway, D4-linoleic acid, 840 32 glitazones, serataconazole, vatiquinone, and utreloxastat. Glitazones, vatiquinone, and utreloxastat may also function as RTAs. Targets 841 not identified by genetic screens (grey) include reducing agents CoQ10 and GPX4, the lipoxygenase 15-LOX, ZIP8 and ZIP14, NCOA4, 842 and DHODH. 843 844 Figure 2: PD-associated genes and their roles in ferroptosis 845 PD-associated mutations can lead to the three defining features of ferroptosis in patient brains: loss of antioxidant capability, lipid 846 peroxidation, and iron dyshomeostasis. Epigenetic changes associated with loss of function in SLC7A11 have been identified in PD 847 patients. Loss of function leads to impaired cystine import and thus glutathione production, eliminating the ability of GPX4 to reduce 848 lipid peroxidation. DJ-1 loss of function can impair cysteine production through the transsulfuration pathway leading to reduced 849 glutathione production. PLA2G6 mutations prevent hydrolysis of the ferroptotic lipid peroxide 15-HpETE-PE, which can be produced 850 by Fenton reaction or 15-LOX. Iron can post-transcriptionally upregulate SNCA translation. SNCA A53T mutations can lead to increased 851 iron dyshomeostasis via inhibiting endocytosis of the transferrin-receptor leading to import of Fe through DMT1, oxidative stress by 852 increasing cytosolic dopamine, and lipid peroxidation through the Fenton reaction. GSH: glutathione, Fe: iron, TfR: Transferrin receptor, 853 IRE: Iron response element, Da: dopamine. 854 33 855