Ferroptosis is programmed by the coordinated regulation of glutathione and iron metabolism by BACH1

Ferroptosis is an iron-dependent programmed cell death resulting from alterations of metabolic processes. However, its regulation and physiological significance remain to be elucidated. By analyzing transcriptional responses of murine embryonic fibroblasts exposed to the ferroptosis-inducer erastin, we found that a set of genes related to oxidative stress protection was induced upon ferroptosis. We further showed that the transcription factor BACH1 promoted ferroptosis by repressing the expression of a subset of erastin-inducible genes involved in the synthesis of glutathione or metabolism of intracellular labile iron, including Gclm, Gclc, Slc7a11, Hmox1, Fth1, Ftl1, and Slc40a1. Compared with wild-type mice, Bach1-/- mice showed resistance to myocardial infarction, the seriousness of which was palliated by the iron-chelator deferasirox, which suppressed ferroptosis. Our findings suggest that ferroptosis is programmed at the transcriptional level to induce genes combating labile-iron-induced oxidative stress and executed upon disruption of the balance between the transcriptional induction of protective genes and accumulation of iron-mediated damage. BACH1 is suggested to control the threshold of ferroptosis and to be a therapeutic target for palliating myocardial infarction.


Introduction
proteins, Gclm and Gclc encoding glutamate-cysteine ligase modifier and catalytic subunits, 1 and other genes involved in the oxidative stress response (Hintze et al., 2007, Sun et al., 2 2002, Warnatz et al., 2011. We hypothesized that BACH1 might regulate ferroptosis by 3 inhibiting the expression of these genes. In addition, since BACH1 is involved in the 4 exacerbation of various diseases involving oxidative stress, such as ischemic heart disease 5 (Yano et al., 2006), hyperoxic lung injury (Ito et al., 2017), trinitrobenzene sulfonic  To understand the regulatory mechanism underlyning ferroptosis, we analyzed 10 the transcriptome response in ferroptotic cells with RNA sequencing (RNA-seq). We also 11 examined whether or not BACH1 was involved in the regulation of ferroptosis by comparing 12 ferroptosis and the expression of ferroptosis-induced genes between wild-type (WT) and 13 Bach1 -/murine embryonic fibroblasts (MEFs). Furthermore, we assessed the influence of 14 BACH1 and ferroptosis on the severity of acute myocardial infarction (AMI) in model mice. 15 We found that BACH1 promoted ferroptosis by directly repressing genes involved in the 16 synthesis of glutathione (GSH) and sequestration of free labile iron. BACH1 also increased 17 (cystine/glutamine transporter) (Sato et al., 2005, Sato et al., 2000 and a well-known 1 regulator of ferroptosis (Jiang et al., 2015). Gclm and Gclc encode glutamate-cysteine 2 ligase modifier and catalytic subunits (Fan et al., 2012, Telorack et al., 2016, both 3 considered to suppress ferroptosis by GSH synthesis (Miess et al., 2018, Stockwell et al., 4 2017). These genes for the pathway of GSH synthesis are also considered to be targets of 5 BACH1 (Warnatz et al., 2011). Indeed, the amount of BACH1 protein was decreased in 6 MEFs exposed to erastin, which was accompanied by the induction of Hmox1 (Figs 1B and 7 EV1). With the reduction in BACH1 protein, the production of its mRNA was induced (Fig   8   EV1), suggesting the presence of feedback regulation of BACH1. 9 These observations suggest that, when cells are exposed to erastin, the 10 expression of genes that counteract ferroptosis is induced in part by a reduction in BACH1 11 protein and that the amount or activity of BACH1 and the kinetics of its feedback regulation 12 may influence ferroptosis by suppressing this counteracting subprogram of ferroptosis. 13 14 BACH1 promotes ferroptosis 15 To clarify whether or not BACH1 regulates ferroptosis, we treated WT and Bach1 -/-MEFs 16 with erastin, stained them with propidium iodide (PI) and annexin V, and compared the cell 17 death by a flow cytometry analysis ( Fig EV2). Bach1 -/-MEFs showed less cell death in 1 response to erastin than WT cells (Fig 2A and B). When the erastin-treated MEFs were 2 observed with a transmission electron microscope, shrunken mitochondria, which are 3 characteristic of ferrotosis (Dixon et al., 2012), were confirmed in both WT and Bach1 -/-4 MEFs ( Fig 2C). The cell death in our experiments was inhibited by the iron chelator 5 deferoxamine (DFO) (Fig 2D-F), confirming that this death was ferroptosis. These results 6 showed that BACH1 promoted ferroptosis in MEFs.

7
It should be noted that the difference in cell death between WT and Bach1 -/-MEFs 8 became smaller as the dose of erastin increased (Fig 2A and B). This may be because 9 even Bach1 -/-MEFs lost their resistance to ferroptosis under high doses of erastin. This 10 suggests that the function of BACH1 is more meaningful for restricting ferroptosis under 11 low-stress conditions. Therefore, the reduction in BACH1 protein may be part of the early 12 ferroptosis program, and BACH1 may set the threshold for ferroptosis. Execution of 13 ferroptosis may be determined by the basal amount of BACH1 and how rapidly it is 14 degraded in response to ferroptosis inducers. 15 16 BACH1 represses the expression of genes involved in the GSH synthesis pathway BACH1 may decrease GSH by repressing the expression of genes involved in the pathway 1 of GSH synthesis. To investigate this possibility, we measured the intracellular GSH 2 concentrations in WT and Bach1 -/-MEFs. The amount of GSH was significantly higher in 3 Bach1 -/-MEFs than in WT cells (Fig 3A), suggesting that BACH1 promoted ferroptosis by 4 reducing GSH within cells.

5
By revisiting our previous data of chromatin immunoprecipitation with sequencing 6 (ChIP-Seq) of BACH1 in mouse myeloblast M1 cells (Ebina-Shibuya et al., 2017, 7 Ebina-Shibuya et al., 2016), we found peaks of BACH1 and its partner MAFK in the 8 regulatory regions of genes encoding molecules for glutathione synthesis, including Gclm, 9 Gclc, and Slc7a11 ( Fig 3B). Furthermore, by comparing the expression of these genes in 10 WT and Bach1 -/-MEFs by quantitative polymerase chain reaction (qPCR), the expression 11 of all of these genes was confirmed to be higher in Bach1 -/-MEFs than in WT cells (Fig 3C). 12 These results suggested that BACH1 bound to the regulatory regions of these genes to 13 repress their expression. 14 A comparison of the protein amounts of SLC7A11, GCLM, and GCLC in MEFs by 15 Western blotting revealed that more GCLM protein was present in Bach1 -/-MEFs than in 16 WT cells (Fig 3D and E). Although the amounts of SLC7A11 protein were similar in WT and 17 Bach1 -/-MEFs (Fig 3D), more SLC7A11 protein was present in Bach1 -/-MEFs than in WT 1 cells when they were treated with proteasome inhibitor MG132 (Fig 3E). These 2 observations suggest that the amount of SLC7A11 protein is further tuned by 3 proteasomal-mediated degradation. There were no marked differences in the amount of 4 GCLC protein with or without MG132 (Fig 3D and E). BACH1 may affect the expression of 5 GCLC protein under certain circumstances. Given these results, we surmised that BACH1 6 decreased the amount of GSH in part by repressing the expression of Gclm and Slc7a11. 7 8 BACH1 promotes ferroptosis by altering GSH 9 We next examined whether or not the resistance of Bach1 -/-MEFs against ferroptosis was 10 actually dependent on the increased expression of the genes involved in the GSH 11 synthesis pathway. Although it is not always statistically significant, knockdown of any of 12 Slc7a11, Gclm, and Gclc resulted in slight but reproducible increases in ferroptosis in both 13 WT and Bach1 -/-MEFs (Figs 4A-D and EV3A, B. Appendix Fig S1A-C). These results show 14 that the genes involved in the GSH synthesis pathway have inhibitory effects against 15 ferroptosis and suggest that BACH1 promotes ferroptosis by repressing their expression. 16 We next examined the effect of knockdown of Hmox1. WT MEFs became more sensitive to ferroptosis by knockdown of Hmox1 than cells with control knockdown (Figs 4E 1 and EV3C. Appendix Fig S1D). We thus concluded that HO-1 works as an inhibitor of 2 ferroptosis under our experimental conditions. However, the effect of HO-1 to accelerate 3 ferroptosis has also been reported (Fang et al., 2019, Kwon et al., 2015. The function of 4 HO-1 in ferroptosis might differ depending on the situations of cells.

5
Importantly, knockdown of Slc7a11, Gclm, Gclc, or Hmox1 did not decrease the 6 observed differences in ferroptosis between WT and Bach1 -/-MEFs (Figs 4A-E and 7 EV3A-C. Appendix Fig S1A-D). These results suggest that the role of BACH1 in promoting 8 ferroptosis depends on the repression of multiple genes involved in ferroptosis. To explore other target genes of BACH1 in the regulation of ferroptosis, we examined 12 genes involved in the regulation of iron metabolism (Fth1,Ftl1,Slc40a1,Tfrc,Mfn2,and 13 Fxn), heavy metal stress (Mt1), and lipoperoxidation (Gpx4). Some of these genes were 14 upregulated in response to erastin (see Fig 1A). Among these genes, ferritin genes (Fth1 15 and Ftl1) and the ferroportin gene (Slc40a1) were dramatically upregulated in Bach1 -/-16 MEFs (Fig 5A), and binding peaks of BACH1 and MAFK were observed near their 17 regulatory regions (Fig 5B). In contrast, the expression of Tfrc, Mfn2, Fxn, Mt1, and Gpx4 1 was only mildly increased in Bach1 -/-MEFs ( Fig EV4A). There were no strong binding 2 peaks of BACH1 or MAFK in the regulatory regions of these genes ( Fig EV4B). Considering 3 that both ferritin and ferroportin reduce the availability of free labile iron and are known to 4 inhibit ferroptosis (Geng et al., 2018, Wang et al., 2016, these results suggest that BACH1 5 promotes ferroptosis by repressing the transcription of ferritin and ferroportin genes. These 6 findings, along with the regulation of GSH synthesis pathway by BACH1, suggest that 7 BACH1 accelerates ferroptosis by decreasing the intracellular activity of GSH and 8 increasing the oxidative activity of labile iron ( Fig 5C). 9 BACH1 aggravates acute myocardial infarction by promoting ferroptosis 11 Finally, we tried to examine whether or not the promotion of ferroptosis by BACH1 is 12 involved in pathological changes in vivo. As there are several reports showing that 13 ferroptosis is involved in ischemia-reperfusion injury in the heart (Baba et al., 2018, Fang et 14 al., 2019, Gao et al., 2015, we used an AMI model based on left anterior descending 15 coronary artery (LAD) ligation (Abarbanell et al., 2010, Shindo et al., 2016 (Fig 6A). In this 16 model, Bach1 -/mice showed less severe injuries than WT mice as judged by the post-operative survival rate and an evaluation of the cardiac function with 1 echocardiography (Figs 6B, C and EV5A-C. Movie EV1A-D). The infarct area on 2 pathological specimens was also smaller in Bach1 -/mice than in WT mice (Fig 6D and E).

3
These results suggest that BACH1 exacerbates the pathology of AMI.

4
In order to investigate whether or not ferroptosis is involved in the pathology, we 5 observed the myocardial infarct regions using a transmission electron microscope.

6
Shrunken mitochondria were observed in both WT and Bach1 -/mice ( Fig 6F). We then 7 investigated whether or not the pathological changes could be improved by administering 8 DFX, which is a clinically used iron chelator. First, we confirmed that it inhibited ferroptosis 9 in MEFs (Figs 7A and EV5D, E). Although there was no improvement in the survival rates in 10 WT or Bach1 -/mice ( Fig 7B), an improvement in the cardiac function on echocardiography 11 was observed in the DFX group, which was more prominent in the WT mice than Bach1 -/-12 mice (Figs 7C, D and EV5F-K). The DFX group of WT mice showed a reduction in the 13 infarct area; however, no such effect was noted in Bach1 -/mice (Fig 7E and F). These 14 results suggest that BACH1 exacerbates the pathology of AMI by promoting ferroptosis.

17
While genes involved in ferroptosis are being discovered (Stockwell et al., 2017), how their 1 expression is regulated during ferroptosis remains unclear. In this study, we found that 2 many of the inhibitory genes of ferroptosis were coordinately upregulated upon induction of 3 ferroptosis with erastin ( Fig 1A). Such a coordinated response may be a mechanism for 4 restricting ferroptosis. We further showed that BACH1 directly counteracted this 5 coordinated response of genes, including Hmox1, Slc7a11,Gclm,Gclc,Fth1,Ftl1,and 6 Slc40a1 (Figs 3B, C and 5A, B), which are involved in the metabolism of GSH or labile iron.

7
The protein amount of BACH1 was reduced upon the induction of ferroptosis ( Fig 1B).  Therefore, the reduction of BACH1 protein level may trigger the coordinated induction of 10 these genes as a subprogram of the initial phase of ferroptosis program. Cells can then 11 integrate distinct signals leading to BACH1 degradation, and thus judge whether or not they 12 should undergo ferroptosis. Thus, BACH1 sets the threshold for whether or not ferroptosis 13 occurs in response to lipid peroxide synthesized. 14 NRF2 is known to activate some of the genes that are repressed by BACH1, 15 including Hmox1, Slc7a11, Gclm andGclc (Alam et al., 1999, Bea et al., 2003, Ishii et al., increases the intracellular glutathione amount, it only weakly protects cells from ferroptosis 1 (Cao et al., 2019). Other reports have shown that NRF2 can inhibit ferroptosis (Fan et al., 2 2017, Roh et al., 2017, Sun et al., 2016. Therefore, ferroptosis execution may depend on 3 the initial amounts and kinetics of the induction or reduction of these transcription factors.

4
This mechanism may extend our understanding of the regulation of ferroptosis, wherein 5 ferroptosis is a cell death programmed at the level of the gene regulatory network. 6 We showed that GSH was higher in Bach1 -/-MEFs than WT cells ( Fig 3A). Our 7 results strongly suggest that BACH1 decreases intracellular GSH by repressing the 8 expression of Gclm, Gclc, and Slc7a11 (Fig 3B and C). Indeed, the protein amount of 9 GCLM was higher in Bach1 -/-MEFs than in WT cells ( Fig 3D). However, the protein 10 amounts of GCLC and SLC7A11 were similar between WT and Bach1 -/-MEFs ( Fig 3D).

11
Cells may have additional mechanisms to tune strictly the protein amounts of GCLC and 12 SLC7A11, managing the intracellular GSH amount and maintaining homeostasis. We found 13 that SLC7A11 was further regulated by proteosomal degradation (Fig 3E). This observation 14 suggests that the decision to undergo ferroptosis may be made based upon whether or not 15 cells can induce efficiently inhibitory proteins like SLC7A11. Cells with higher amounts of 16 SLC7A11 may likely be protected from ferroptosis. Gclc and Slc7a11 may be critical factors 17 for cells, with the transcriptional regulation by BACH1 and additional layers of regulation, 1 although these points will need to be explored in further studies.

2
Reports on the function of HO-1 are conflicting, with studies conversely describing 3 it to promote or inhibit ferroptosis (Adedoyin et al., 2018, Fang et al., 2019, Kwon et al., 4 2015, Sun et al., 2016. These discrepant findings may be due to the fact that HO-1 5 degrades prooxidant heme to produce not only the radical scavengers biliverdin and 6 bilirubin but also free iron that mediates ferroptosis through Fenton reaction (Igarashi & 7 Watanabe-Matsui, 2014, Stockwell et al., 2017. Therefore, in order to allow HO-1 to 8 function effectively as an anti-oxidative stress enzyme, it is essential to suppress the 9 reactivity of labile iron derived from heme. We showed that BACH1 represses the 10 expression of the genes of ferritin and ferroportin ( Fig 5A and B), which reduce the 11 intracellular availability of labile iron. By increasing the expression of not only HO-1 but also 12 ferritin and ferroportin during the induction of ferroptosis (Fig 1A), the prooxidant activities 13 of heme and heme-derived free iron can be suppressed efficiently, thus protecting cells 14 from ferroptosis. Conversely, BACH1 represses the expression of ferritin and ferroportin in 15 addition to HO-1, thus effectively promoting ferroptosis ( Fig 5C). Based on the present and 16 previous findings, we proposed a model in which BACH1 accelerates ferroptosis by suppressing two major intracellular counteracting mechanisms against ferroptosis: the 1 GSH synthesis pathway and the system for the sequestration of labile iron ( Fig 5C).

2
In addition, we showed that ferroptosis was involved in the pathology of not only 3 ischemia-reperfusion injury (IRI) (Baba et al., 2018, Fang et al., 2019, Gao et al., 2015 but 4 also AMI. The severity of AMI was improved by the iron chelator, DFX particularly in WT  This may be explained by observations that adhesion between the cardiac infarct area and 8 chest wall was smaller and cardiac rupture occurred more frequently in the DFX group than 9 in the control group. These effects may offset the reduction in the infarct areas. Ferroptosis 10 and subsequent inflammation may prevent cardiac rupture by pleural adhesion, but this 11 issue needs to be investigated further. Nonetheless, our results here suggest that the 12 therapeutic effect of DFX is expected in AMI and IRI. Necroptosis is also reportedly 13 involved in cardiac ischemic disease (Oerlemans et al., 2012, Smith et al., 2007. Therefore, 14 the double inhibition of ferroptosis and necroptosis may lead to the more effective palliation 15 of AMI. In addition, this study suggests that Bach1 -/mice are more resistant to AMI than 16 WT mice because of their lower rate of ferroptosis than in WT mice (Figs 6 and 7). BACH1 17 may be a potential therapeutic target of AMI in the future.   MEFs (3-8 x 106 cells for each lot) were suspended in 100 µL of methanol containing the internal standards (0.2 µg/mL SAM-13C515N for positive ion mode (Pos) and 1 µg/mL 1 GSH-13C215N for negative ion mode (Neg)), and were homogenized by mixing for 30 sec 2 followed by sonication for 10 min. After centrifugation at 16,400 x g for 20 min at 4°C 3 followed by deproteinization, 3 µL of each extract was analyzed by ultra high-performance 4 liquid chromatography triple quadrupole mass spectrometry (UHPLC/MS/MS).  Appendix Table S1. The other settings are as follows: 3.5 kV (Pos) or 2.5 kV (Neg) 13 capillary voltage, 30 V cone voltage, 50 V source offset, source temperature at 150°C, 150 14 L/hr cone gas (N2) flow rate, desolvation temperature at 450°C, 1000 L/hr desolvation gas 15 flow, 0.15 min/mL collision gas flow, 7.00 bar nebulization gas (N 2 ) flow. LC separation, 16 was performed as described before (Saigusa et al., 2016), using a normal-phase column 17 (ZIC-pHILIC; 100 mm × 2.1 mm i.d., 5 µm particle size; Sequant, Darmstadt, Germany) with 1 a gradient elution using solvent A (10 mmol/L NH 4 HCO 3 , adjusted to pH 9.2 using ammonia 2 solution) and B (acetonitrile) at 300 µL/min: 99 to 70% B from 0.5 to 4.0 min, 70 to 1% B 3 from 4.0 to 6.5 min, 1% B for 2.5 min, and 99% B for 9 min until the end of the run. The 4 oven temperature was 20°C. The data were collected using the MassLynx v4.1 software 5 (Waters Corp.) and the ratio of the peak area of analyte to the IS was analyzed by Traverse 6 MS (Reifycs Inc., Tokyo, Japan).      were prepared on the Ion Chef system using an Ion PI Hi-Q Chef kit (Thermo Fisher 1 Scientific) and sequencing was performed on an Ion Proton system using with Ion PI Hi-Q 2 sequencing kit (Thermo Fisher Scientific) the PI v3 chip (Thermo Fisher Scientific). The 3 sequence data were obtained as fastq files. The sequence data was aligned to reference 4 hg19 using the RNASeqAnalysis plugin from Ion torrent suite software (Thermo Fisher 5 Scientific). Mapped reads were counted for each gene using HTSeq v 0.9.1 htseq-count.

6
The differential expression analysis was performed on edge R v 3.16.5 after removal of low 7 count lead genes using three biological replicates for each condition (less than 5 leads per 8 gene in the sample and counts per million mapped reads (CPM) of 1 or less). For all experiments, differences of data sets were considered statistically significant when 12 P-values were lower than 0.05. Statistical comparisons were performed using the t-test in 13 comparison between the two groups, and one, two, or three way ANOVA followed by 14 Tukey's test or Tukey-Kramor method in comparison among multiple groups. For the t-test, 15 student's t-test was used when the standard deviation (SD) of the groups was not 16 significantly different by f-test. Welch's t-test was used when the SD of the groups was significantly different by f-test.    Error bars of (B) represent standard deviation. The box and whisker plots of (E) show the 8 25th and 75th percentile quartiles and median values (center black line) and maximum and 9 minimum values of the data. P-value of (B) by t-test. P-value of (F) by three-way ANOVA. WT MEFs (7th, 9th, and 11th passage : P7, P9, and P11) were exposed to erastin for 10 8 hrs. qRT-PCR analysis for Bach1 and Hmox1 mRNA relative to Actb mRNA. 14 WT and Bach1 -/-MEFs (11th passage: P11) exposed to erastin for 24 hrs ( Figure 2B). Tukey-Kramer method after two-way ANOVA. P-value of (E) by Tukey's test after two-way