First identified in as the second mammalian

First identified in 1982 as the second mammalian glutathione peroxidase [23], we and Stockwell's group demonstrated in 2014 that the selenoperoxidase GPX4 is the key upstream regulator of ferroptosis [12], [13]. The role of GPX4 as the main regulator in the ferroptotic process is based on its unique function to reduce complex hydroperoxides including phospholipid hydroperoxides and cholesterol hydroperoxides to their corresponding counterparts, thereby interrupting the lipid peroxidation chain reaction. In fact, early studies with conditional deletion of Gpx4 in brain and fibroblasts provided initial evidence that neurodegeneration and cell death of hippocampal neurons occurred in non-apoptotic manner entailing massive lipid peroxidation [24]. The promiscuity of GPX4 towards a number of reducing substrates is also reflected by its use of oxidizing substrates. Besides its main cofactor GSH, other low molecular thiols and even protein thiols can be readily exploited by GPX4 [25]. Recently, the molecular role of selenium in form of the 21st amino Ciclopirox selenocysteine (Sec) in GPX4 has been elucidated. Mice expressing a targeted mutation of the active Sec to Cys (GPX4_U46C) are surprisingly viable (unlike the full knockout of Gpx4 causing early embryonic death [26]) but failed to survive the weaning stage due to severe epileptic seizures [27]. Loss of a specific subset of parvalbumin–positive (PV+) GABAergic interneurons in cortex of homozygous mice was identified as the underlying reason for the seizures, implying that this specific type of inhibitory neurons requires fully functional, selenium containing GPX4 [28]. The reasons for the peculiar sensitivity of PV+ GABAergic interneurons towards cell death are not completely understood and might be of complex nature. For instance, these cells need to migrate and form inhibitory synapses early after birth, which requires a high degree of unsaturation in their membranes. Thus, they might be particularly vulnerable to cell death. Alternatively, these neurons are known to have a particularly high demand for ATP, therefore they might generate a substantial amount of oxygen radicals as a metabolic side product. Remarkably, homozygous GPX4_U46C expressing cells proliferate normally and were as resistant to cytotoxic agents as wildtype cells; however, in stark contrast these cells were exquisitely sensitive to peroxide-induced ferroptosis due to the high susceptibility of Cys containing GPX4 towards peroxide-induced irreversible overoxidation and inactivation of GPX4 in cells [27]. Hence, it emerged that selenium utilization by GPX4 not only allows cells and tissues to become highly resistant to pro-ferroptotic, pathological conditions but perhaps also to exploit hydrogen peroxide to be used as a cellular signaling molecule [29]. While the upstream mechanisms of ferroptosis including cysteine availability, GSH biosynthesis and proper functioning of GPX4 have been well established in the meantime, much less is known about potential downstream events. Genetic screens independently performed in haploid cells and fibroblasts identified acyl-CoA synthetase long-chain family member 4 (ACSL4) as an additional and essential downstream player in the ferroptotic process [6], [8]. ACSL4 is one of a number of fatty acid activating enzymes functioning by esterifying CoA to free fatty acids in an ATP dependent manner. Unlike other family members, ACSL4 shows a high preference towards long chain PUFAs such as arachidonic acid (AA) and adrenic acid (AdA). CRISPR/Cas9-mediated knockout and pharmacological inhibition of ACSL4 by thiazolidinediones conferred an unparalleled protection against ferroptosis elicited by small molecule ferroptosis inducers or the genetic inactivation of Gpx4 in fibroblasts [6]. An interdisciplinary approach combining redox global phospholipidomics, reverse genetics, bioinformatics and systems biology allowed to pinpoint 15-hydroperoxy-arachidonoyl- and 15-hydroperoxy-adrenoyl residues esterified in phosphatidylethanolamines (PE) as proximate signals of the ferroptotic death program [30]. Accordingly, genetic disruption/pharmacological inhibition of ACSL4 prevented the accumulation of these death signals in cells [6], [30]. Although identified in the haploid screen [8], knockout of lysophosphatidylcholine acyltransferase 3 (LPCAT3), an enzyme which preferably re-esterifies polyunsaturated fatty acyl-CoAs into certain lyso-phospholipids, did not yield strong protection against ferroptosis, suggesting cell-specific contexts or compensatory mechanisms by other enzymes of this family of proteins [6].