Glutathione peroxidase 4-regulated neutrophil ferroptosis induces systemic autoimmunity

Abstract

The linkage between neutrophil death and the development of autoimmunity has not been thoroughly explored. Here, we show that neutrophils from either lupus-prone mice or patients with systemic lupus erythematosus (SLE) undergo ferroptosis. Mechanistically, autoantibodies and interferon-α present in the serum induce neutrophil ferroptosis through enhanced binding of the transcriptional repressor CREMα to the glutathione peroxidase 4 (Gpx4, the key ferroptosis regulator) promoter, which leads to suppressed expression of Gpx4 and subsequent elevation of lipid-reactive oxygen species. Moreover, the findings that mice with neutrophil-specific Gpx4 haploinsufficiency recapitulate key clinical features of human SLE, including autoantibodies, neutropenia, skin lesions and proteinuria, and that the treatment with a specific ferroptosis inhibitor significantly ameliorates disease severity in lupus-prone mice reveal the role of neutrophil ferroptosis in lupus pathogenesis. Together, our data demonstrate that neutrophil ferroptosis is an important driver of neutropenia in SLE and heavily contributes to disease manifestations.

Extended Data Fig. 1. IgG and IFNα but not CXCL11 or IL12/23 p40 present in SLE sera contributed to neutropenia.

a-b. Flow cytometry quantification of cell viability of neutrophils (a: n=3; b: n=9) in vitro cultured with (a) 5%, 10%, or 20% SLE serum for 6, 16, 24 hours respectively, or with (b) 20 % HC, SLE or RA serum respectively for 16 hours. c. Detection of the inflammatory factors in the sera from RA (n = 16), BD (n = 20) and AS (n = 18) patients vs. HCs (n = 19). d-e. Flow cytometry quantification of cell viability and lipid ROS of HC neutrophils (d: n=7; e: n=7 for anti-CXCL11 or n=4 for Ustekinumab) cultured in vitro with 20% SLE serum supplemented with anti-CXCL11 (0.1, 1, 5 µg ml⁻¹) or Ustekinumab (0.1, 1, 10 µg ml⁻¹) for 16 hours. f-g. The proportion of anti-dsDNA in total IgG correlated with SLE neutrophil counts and SLEDAI scores (n=63). h. Western blot validation of purified IgG from serum and of serum with IgG depletion. i. Flow cytometry quantification of cell viability of neutrophils (n=5) cultured in vitro with serum in the presence or absence of anti-IFNAR (10 µg ml⁻¹), or IgG depletion, for 16 hours. j-k. Serum IgG was purified or depleted by Protein A/G. (j) Ponceau S staining (upper panel) and western blot (lower panel) detection of purified IgG and of serum with depleted IgG. (k) Flow cytometry quantification of cell viability of neutrophils (n=4) with HC serum in the presence of HC or SLE IgG at different concentrations (1.2, 2.4, 3.6 g L⁻¹), or SLE serum with/without IgG depletion, for 16 hours. Data are shown as mean ± SD. *p < 0.05, **p < 0.01, ns p > 0.05. Two-tailed paired or unpaired Student’s t-test was applied.

Extended Data Fig. 2. Ferroptosis is restricted in neutrophils but not other cells in SLE and this could be reverted by addition of Ferroptosis specific inhibitors.

a. Flow cytometry quantification of cell viability of HC lymphocytes, monocytes, and neutrophils cultured with 20% HC or SLE serum for 16 hours and the proportion of apoptotic (Annexin V+ 7AAD-), necrotic (Annexin V+ 7AAD+) and live (Annexin V- 7AAD-) cells in each subset was analyzed (n=6). b. HC neutrophils were cultured with 20% HC or SLE serum, and lipid-ROS productions at different time points were detected (n=3). c. Dot plots show cell viability analyzed by lactate dehydrogenase (LDH) release. HC neutrophils (n=9) were cultured with 20% HC serum supplemented with RSL-3 (10 μM) or SLE serum supplemented with LPX-1 (1 μM) for 16 hours before analysis. d-e. Dot plots show flow cytometry quantification of the percentage of apoptotic, necrotic, and live cells. HC neutrophils (n=5) were cultured with 20% HC or SLE serum in the presence or absence of LPX-1 (1 μM) for 16 hours before analysis. f-g. HC B cells (n=6) were cultured in 20% HC or SLE serum supplemented with LPX-1 for 72 hours, and plasmacytoid dendritic cells (pDC) (n=3) were cultured for 24 hours, the level of IgG was assessed by ELISA and type 1 IFNs by flow cytometry individually. h-i. Dot plots show cell viability and lipid-ROS in HC neutrophils (n=7) cultured with 20% HC or SLE serum supplemented with ß-ME (10/50 μM) for 16 hours. Data are shown as mean ± SD. ns p > 0.05. Two-tailed paired Student’s t-test was applied.

Extended Data Fig. 3. The cooperative effects between IFNα and SLE IgG on cell death.

a-c. HC neutrophils were cultured in the presence of HC or SLE serum with or without the addition of Cl-amidine (Cl) (peptidyl arginine deiminase 4 (PAD4) inhibitor, 100 μM), or LPX-1 (1 μM) for 4 or 16 hours and NETs were assessed in SYTOX Green+ cells based on morphology (n=6). Neutrophils with DNA area greater than 400um² were considered as NETs. Dot plots show the percentage of cells forming NETs in all dead neutrophils from the indicated group. d-e. Representative fluorescent images and related quantification of NETosis. HC neutrophils (n=6) were stimulated by PMA (50 nM) with or without LPX-1(1 μM) for 4 hours. f-i. HC neutrophils were cultured with SLE IgG (3.6 g L⁻¹) and/or IFN-α (10^5 U ml⁻¹) for 4 or 16 hours and cells were stained with SYTOX Green for the detection of NETs. Dot plots show the immunofluorescence microscope quantification of NETosis in total dead neutrophils from the indicated group (4h: n=6; 16h: n=3). The scale bar represents 50 μm. Data are shown as mean ± SD. ns p > 0.05. Two-tailed paired Student’s t-test was applied.

Extended Data Fig. 4. The ferroptosis inhibitor ameliorates lupus progression with much better therapeutic effect compared to the NETosis inhibitor.

a-i. MRL/lpr mice (n=6) were treated with DMSO (0.1ml 10%), LPX-1(10 mg/kg) or CTX (20 mg/kg) every other day at week 12 for 6 weeks, DMSO (0.1ml 10%) was applied to sex-matched MRL/Mpj mice (n=5) as control. Mice were euthanized at 18 weeks of age for analysis. (a). Flow cytometry quantification of lipid ROS. (b). Representative immunofluorescent images of glomeruli stained with IgG (red), IgM (yellow), C1q (green), and DAPI (blue). (c-e) Flow cytometry quantification of plasma inflammatory factors and (f-i) plasma IgG. j-o. MRL/lpr mice (DMSO, LPX-1: n=3; Cl, Cl+LPX-1: n=4) were treated with DMSO, Cl, LPX-1, or Cl combined with LPX-1 every other day for 3 weeks starting at the age of week 12. Mice were euthanized at 15 weeks of age for analysis. (j-k) Representative images and related quantification of axillary spleens and lymph nodes. (l) Western blot analysis of cit-H3 in circulating neutrophils from mice subjected to the indicated treatment. (m) Dot plots show the ELISA assessment of serum complement 3. (n) Dot plots show the ELISA assessment of serum anti-dsDNA antibodies titers. (o) Dot plots shows the Bicinchoninic acid (BCA) assay of urine proteins. The scale bar represents 50 μm. Data are shown as mean ± SD. ns p > 0.05. Two-tailed unpaired Student’s t-test was applied.

Extended Data Fig. 5. The expression of cystine transporter SLC7A11 is not different between HC and SLE neutrophils.

a. Heatmap visualization of RNA-seq analysis on differentially expressed ferroptosis-related genes (Standardized with GAPDH) in neutrophils between new onset treatment-naïve SLE patients (n=6) and HCs (n=6). b. Western blot validation for SLC7A11 antibody. 293T cells were transfected with Slc7a11 overexpression plasmid and cells without transfection were used as control. c. Western blot assay shows the expression of cystine transporter SLC7A11 in neutrophils from HCs (n=8) and SLE patients (n=8). Data are shown as mean ± SD. ns p > 0.05. Two-tailed unpaired Student’s t-test was applied.

Extended Data Fig. 6. GPX4 reduction was observed in neutrophils but not other immune cells in SLE.

a. Flow cytometry quantification of GPX4 expressions in HCs (n=16) and SLE (n=12) neutrophils. b. GPX4 expressions in neutrophils from treatment-naïve SLE patients correlated negatively with disease activities as measured by SLEDAI (n=12). c. Flow cytometry quantification of GPX4 expressions in lymphocytes (including CD4+T, CD8+T, and B cells) and monocytes from HCs (n=11) and SLE patients (n=9). d-e. Western blot analysis of GPX4 expressions in lymphocytes and monocytes from HCs (n=11) and SLE patients (n=10). f-g. Western blot analysis of GPX4 expression in HC neutrophils, monocytes, and lymphocytes (n=7). h-i. Western blot analysis of GPX4 expression in HC neutrophils, monocytes, and lymphocytes (n=7) cultured with 20% HC or SLE serum for 30 hours. j-k. Western blot analysis of GPX4 expression in HC neutrophils (n=3) when cultured with 20% HC serum or SLE serum supplemented with Cl-amidine (Cl, 100 μM), APX-115 (APX) (pan-NADPH oxidase (NOX) inhibitor, 20 μM), and GSK2795039 (GSK) (NOX2 inhibitor, 10 μM). Data are shown as mean ± SD. ns p > 0.05. Two-tailed unpaired Student’s t-test was applied.

Extended Data Fig. 7. FcγR3β is essential for the SLE IgG-mediated GPX4 downregulation in neutrophils.

a-b. Expression correlation analysis between different TLRs or FcRs with GPX4 based on RNA-seq data. In SLE neutrophils, (a) TLR signaling pathways are not associated with GPX4 reduction. (b) Fcγr3b but not other FcRs’ expression is negatively associated with GPX4 reduction. c. Different Fc receptor expressions in HCs (n=6) and SLE (n=6) analyzed by RNA-seq. d. Western blot analysis of FcγR3β expressions in neutrophils, monocytes and lymphocytes from HCs (n=3). e. GPX4 expressions in HL60 cells after overexpression of FcγR3β(n=4). Control referred to cells without transfection. Data are presented as mean ± SD or median with interquartile range. ns p > 0.05. one-tailed or two-tailed unpaired Student’s t-test or Mann Whitney test was applied.

Extended Data Fig. 8. Mice with Gpx4 haploinsufficiency in neutrophils developed spontaneous lupus-like disease, while Gpx4 ^fl/flLysMCre⁺ mice exhibited mild autoimmunity.

a-b. Flow cytometry quantification and western blot analysis of GPX4 in neutrophils (CD45⁺CD11b⁺Ly6G&Ly6C⁺) and non-neutrophils (including monocytes and lymphocytes) from Gpx4^fl/fl (n=6) and Gpx4^fl/wtLysMCre⁺ (n=9) mice. c-d. Flow cytometry analysis of peripheral neutrophils (CD45⁺CD11b⁺Ly6G&Ly6C⁺) and monocytes (CD45⁺CD11b⁺Ly6G&Ly6C⁻) from Gpx4^fl/fl (c: n=6, d: n=10) and Gpx4^fl/wtLysMCre⁺ (c: n=9, d: n=13) mice. e. Flow cytometry quantification of lipid-ROS and cell viability in neutrophils (n=6) from Gpx4^fl/wtLysMCre⁺ mice cultured in complete RPMI 1640 basic medium in the presence or absence of LPX-1 (1μM). f. Skin lesions of Gpx4^fl/wtLysMCre⁺ mice. g. Immunofluorescent images of glomeruli in Gpx4^fl/fl mice and Gpx4^fl/wtLysMCre⁺ mice. IgG (red), IgM (yellow), C1q (green), and DAPI (blue). h. Dot plots show the proteinuria of Gpx4^fl/fl and Gpx4 ^fl/flLysMCre⁺ mice at 4 months of age assessed by BCA assay. i. ELISA assay shows the levels of serum complement 3 in Gpx4^fl/fl (n=8) and Gpx4 ^fl/flLysMCre⁺ (n=8) mice at 6 months of age. j. ELISA assay shows the levels of serum anti-dsDNA antibodies in Gpx4^fl/fl (n=8) and Gpx4 ^fl/flLysMCre⁺ (n=8) mice at 6 months of age. The scale bar represents 50 μm. Data are shown as mean ± SD, ns p > 0.05. Two-tailed unpaired or paired Student’s t-test was applied.

Extended Data Fig. 9. IFNα and SLE IgG enhanced ferroptosis by promoting binding of CREM to the Gpx4 promoter.

a. Western blot analysis of CREMα and CaMKIV in cytoplasm and nucleus of neutrophils from HCs and SLE patients. b. Dot plots show the CHIP analysis results on CREMα binding to the promoter of Gpx4 from neutrophils (n=6) with indicated treatment: IFN-α (10^5 U ml⁻¹), anti-IFNAR (10 µg ml⁻¹), SLE IgG (2.4 g L⁻¹) or SLE sera with IgG depletion. c-d. Efficiency of CREMα knockdown by siRNA (n=3) or CREMα over-expression (n=4) in HL-60 cells validated by qPCR (c) and western blot (d). e. Effect of IFN-α or SLE IgG on GPX4 expressions in HL60 cells after knockdown or overexpression of CREMα. f. Efficiency of CREMα knockdown or overexpression on ferroptosis in HL60 cells (n=4), assessed by flow cytometry using BODIPY C11. Data are shown as mean ± SD, ns p > 0.05. Two-tailed unpaired or paired Student’s t-test was applied.

Extended Data Fig. 10

The hypothetical model for neutrophil ferroptosis in SLE pathogenesis.

Fig. 1.. SLE IgG and IFNα modulate neutrophil viability.

a. The numbers of peripheral neutrophils from either healthy controls (HC) or patients with different rheumatic diseases including systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), Behcet’s disease (BD) and ankylosing spondylitis (AS) (HC: n=188, SLE: n=126, RA: n=50, BD: n=50, AS: n=50). b. The numbers of peripheral neutrophils in SLE correlated negatively with disease activities as measured by systemic lupus erythematosus disease activity index (SLEDAI) (n=98). c. The numbers of peripheral neutrophils in SLE (n=98) patients before and after treatments vs. HCs. d. Flow cytometry quantification of cell viability of neutrophils freshly isolated from peripheral blood of SLE patients before and after treatments (n=7) vs. HCs (n=11) (live cells were defined as 7AAD-AnexinV-). e. Flow cytometry quantification of cell viability of neutrophils cultured in vitro with 20 % serum from either HCs (n=5) or SLE patients (n=9) for 16 hours. f. The in vitro effect of SLE sera on neutrophil cell viability corelated with the peripheral neutrophil counts in SLE patients (n=14). g. Multiplex cytokine array of the inflammatory factors in sera from SLE patients vs. HCs (SLE=39, HC=37). h. Venn diagram showed serum factors specifically increased in SLE. i-j. Flow cytometry quantification of cell viability of neutrophils (n=7) cultured in vitro with SLE serum supplemented with blocking antibody targeting IFNα at different dosages (0.1, 1, 10 µg ml⁻¹), or with HC serum with the addition of IFNα at different dosages (10^3, 10^4, 10^5 U ml⁻¹) for 16 hours. k-l. Flow cytometry quantification of cell viability of neutrophils (n=7) from HC cultured with HC serum with the addition of SLE IgG at different dosages (1.2, 2.4, 3.6 g L⁻¹), or SLE serum with/without IgG depletion, for 16 hours. Data are shown as mean ± SD. ns p > 0.05. Two-tailed unpaired or paired Student’s t-test was applied.

Fig. 2.. Neutrophil ferroptosis is prevalent in patients with SLE.

a. Left: Representative fluorescent images indicated neutrophils death. Peripheral neutrophils from both HCs and patients with SLE were stained with propidium iodide (PI) to detect dead cells and DAPI as counterstain, and NETs were defined as cells with DNA area exceeding 400 μm². The scale bar represents 100 μm. Right: Dot plots showed the quantification (HC: n=3, SLE: n=7). b. Left: Representative electron microscopy images of neutrophils in HCs and SLE patients. Right: Quantification of neutrophil mitochondrion with cavity in HCs (n=6) and SLE patients (n=8). The scale bar represents 500 nm. c. Representative electron microscopy images of HC neutrophils cultured with 20% HC serum supplemented with dimethyl sulfoxide (DMSO) vehicle or 10 μM RSL-3, a ferroptosis inducer for 4 hours. The scale bar represents 500 nm. d. Lipid-ROS content in neutrophils from SLE patients (n=39), paired SLE samples before and after treatments (n=11), neutrophils from HCs were used as controls (n=17). Cells were incubated with BODIPY C11, a fluorescent lipid peroxidation reporter molecule that shifts its fluorescence from red to green, for 15 min before flow cytometry assessment. e. Flow analysis of lipid-ROS content in lymphocytes and monocytes from either HCs or SLE patients (n=8). f. Flow analysis of lipid-ROS content in HC lymphocytes, monocytes, and neutrophils, after cultured with 20% HC or SLE serum respectively for 16 hours (n=6). Data are shown as mean ± SD or median with interquartile range. ns p > 0.05. Two-tailed unpaired or paired Student’s t-test or Mann Whitney test were applied.

Fig. 3.. Neutrophil ferroptosis, the main form of neutrophil death in SLE, is induced by autoantibodies and IFNα.

a. HC neutrophil cell viability when cultured with 20% HC serum supplemented with RSL-3 (10 μM) or DMSO for 16 hours, and cell viabilities were assessed by flow cytometry (n=4). b. HC neutrophil cell viability when cultured in 20% HC or SLE serum supplemented with increasing dosages of various reagents, including two ferroptosis inhibitors, liproxstatin-1 (LPX-1, 10/100/1000 nM), or deferoxamine (DFO, 1/10/100 μM), two necroptosis inhibitors, necrostatin-1 (Nec-1, 10/100/1000 nM), or necrosulfonamide (NSA, 10/100/1000 nM), or apoptosis inhibitor, Z-VAD (0.1/1/10 μM) for 16 hours, and cell viabilities were assessed by flow cytometry (n=5). c. HC neutrophils were cultured in the presence of HC or SLE serum with or without addition of Cl-amidine (Cl) (an inhibitor of NETosis, 100 μM), or LPX-1 (1 μM) for 16 hours and NETs were counted in SYTOX Green+ cells by morphology (n=8). Neutrophils with DNA area greater than 400um² were considered to have undergone NETosis. Dot plots show the percentage of NETosis in all dead neutrophils in the indicated group. d. Lipid ROS production by HC neutrophils when cultured with 20% HC serum supplemented with SLE IgG, or with SLE serum with or without IgG depletion for 16 hours (n=11). e. Lipid ROS production by HC neutrophils when stimulated with IFN-α at different concentrations (10^3, 10^4, 10^5 U ml⁻¹) for 16 hours (n=6). f. Lipid ROS production by HC neutrophils when cultured with SLE serum in the absence or presence of IFNAR blocking antibody at different dosages (0.1, 1, 10 µg ml⁻¹) for 16 hours (n=6). Data are shown as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns p > 0.05; two-tailed unpaired or paired Student’s t-test were applied.

Fig. 4.. Ferroptosis inhibitors ameliorate disease progression in MRL/lpr mice.

a. Flow cytometry quantification of lipid ROS in lymphocytes, monocytes and neutrophils from MRL/Mpj (n=6), MRL/lpr (n=6) and NZB/W F1 mice (n=9). Splenic cells were incubated with BODIPY C11 for 15 min before flow cytometry assessment. b. Cell viability of circulating neutrophils from indicated mice as quantified by flow cytometry (MRL/Mpj, MRL/lpr were euthanized at 18 weeks of age for analysis, NZB/W F1 at 3 or 7 months of age). c-j. MRL/lpr mice (n=6) were treated with DMSO, LPX-1, or cytoxan (CTX) every other day for 6 weeks starting at week 12, DMSO was applied to sex-matched MRL/Mpj mice (n=5) as control. Mice were euthanized at 18 weeks of age for analysis. (c) ELISA assessment of serum anti-dsDNA antibodies titers. (d) ELISA assessment of serum complement 3. (e,f) Size and length of axillary lymph nodes. (g, h) Size and length of spleens. i. Bicinchoninic acid (BCA) assay of urine proteins. j. Periodic Acid-Schiff (PAS) staining of glomeruli. The scale bar represents 20 μm. Data are shown as mean ± SD. *or # p < 0.05, **or ## p < 0.01, ***p < 0.001, ****p < 0.0001, ns p > 0.05. Two-tailed unpaired Student’s t-test was applied.

Fig. 5.. IFN-α and SLE IgG are the main drivers of neutropenia by reducing GPX4 in neutrophils.

a-b. Western blot analysis of GPX4 expression in lymphocytes, monocytes and neutrophils from MRL/Mpj, MRL/lpr and NZB/W F1 mice. c-e. Western blot (HC=5, SLE=10) and qPCR analysis (HC=19, SLE=27) of GPX4 expression in neutrophils isolated from HCs and SLE patients. Paired SLE samples before and after effective treatments were analyzed (d). f-g. Western blot and qPCR analysis of GPX4 expression in HC neutrophils cultured with 20% HC or SLE serum for 30 hours (n=7). h. Western blot analysis of GPX4 expression in HC neutrophils cultured for 30 hours with 20% HC or SLE serum with different disease activities and neutrophil (NEU) counts. i-j. GPX4 expressions in HC Neutrophils when cultured with HC serum supplemented with SLE IgG (1.2, 2.4, 3.6 g L⁻¹), or with SLE serum with or without IgG depletion for 16 hours (n=5). k-l. GPX4 expressions in HC neutrophils when cultured with HC serum supplemented with IFNα (10^3, 10^4, 10^5 U ml⁻¹) or SLE serum supplemented with anti-IFNAR monoclonal antibody (0.1, 1, 10 µg ml-1) (n=6). Data are shown as mean ± SD. ns p > 0.05. Two-tailed unpaired or paired Student’s t-test were applied.

Fig. 6.. Mice with Gpx4 haploinsufficiency in neutrophils develop spontaneous lupus-like disease.

a. Size and length of spleen and lymph node from Gpx4^fl/wtLysMCre⁺ and wild-type (WT) mice at 12 months of age (n=6). b. Glomeruli of WT and Gpx4^fl/wtLysMCre⁺ mice at 4 months of age (PAS staining). The scale bar represents 50μm. c. Proteinuria of Gpx4^fl/fl and Gpx4^fl/wtLysMCre⁺ mice at 4 months of age (BCA assay) (Gpx4^fl/fl: female=15, male=4, Gpx4^fl/wt: female=20, male=11). d-e. The titers of circulating anti-dsDNA antibodies, complement 3 and various inflammatory factors in Gpx4^fl/fl and Gpx4^fl/wtLysMCre⁺ mice at different ages by ELISA and Multiplex cytokine detection (Gpx4^fl/fl: female=6, male=6, Gpx4^fl/wtLysMCre⁺: female (3m/6m) =16/29, male (3m/6m) =7/7). Data are shown as mean ± SD or median with interquartile range. ns p > 0.05. Two-tailed unpaired or paired Student’s t-test or Mann Whitney test were applied.

Fig. 7.. IFN-α and SLE IgG suppresses the transcription of GPX4 by promoting binding of CREM to the Gpx4 promoter.

a. Western blot analysis of nuclear accumulation of CREMα and CaMKIV in neutrophil isolated from HCs and SLE patients. b-c. SLE IgG (1.2, 2.4, 3.6 g L⁻¹) and IFNα (10^3, 10^4, 10^5 U ml⁻¹) at different concentrations increased nuclear accumulation of CREMα and CaMKIV. d. SLE serum increased DNA-binding of CREMα to Gpx4 promoter area by chromatin immunoprecipitation analysis (n=8). e. Intracellular IFIT1 and ISG15 expressions correlated positively with the enrichment of CREMα in the Gpx4 promoter region. f. The interaction between the Gpx4-promotor and CREMα by DNA pulldown assay. g. Western blot analysis of GPX4 in neutrophil isolated from wild type and Camk4^−/− MRL/lpr mice at 18 weeks of age (n=4). h. Dot plots show flow cytometry quantification of neutrophil viability in wild type (WT) and Camk4^−/− MRL/lpr mice at 18 weeks of age (n=4). i. Flow cytometry quantification of lipid ROS in neutrophils from WT and Camk4^−/− MRL/lpr mice (n=4). j-m. WT, Crem^−/− and Camk4^−/− mice were first administrated with pristane (i.p., 0.5 ml per mouse) for autoantibody induction and 2 months later i.v. with 2 x 10^9 pfu adenovirus IFNα (ad-IFNα) to overexpress IFNα in vivo. One more month after ad-IFNα administration, mice were euthanized and GPX4 expression, cell viability and lipid ROS in circulating neutrophils isolated from indicated mice were analyzed (n=4). Data are shown as mean ± SD. ns p > 0.05. Two-tailed unpaired Student’s t-test were applied.