Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun;19(6):719-730.
doi: 10.1038/s41589-022-01249-3. Epub 2023 Feb 6.

Identification of essential sites of lipid peroxidation in ferroptosis

A Nikolai von Krusenstiern  1 , Ryan N Robson  2 , Naixin Qian  3 , Baiyu Qiu  1 , Fanghao Hu  3 , Eduard Reznik  1 , Nailah Smith  1 , Fereshteh Zandkarimi  3 , Verna M Estes  2 , Marcel Dupont  1 , Tal Hirschhorn  1 , Mikhail S Shchepinov  4 , Wei Min  5 , K A Woerpel  6 , Brent R Stockwell  7   8
Affiliations

Identification of essential sites of lipid peroxidation in ferroptosis

A Nikolai von Krusenstiern et al. Nat Chem Biol. 2023 Jun.

Abstract

Ferroptosis, an iron-dependent form of cell death driven by lipid peroxidation, provides a potential treatment avenue for drug-resistant cancers and may play a role in the pathology of some degenerative diseases. Identifying the subcellular membranes essential for ferroptosis and the sequence of their peroxidation will illuminate drug discovery strategies and ferroptosis-relevant disease mechanisms. In this study, we employed fluorescence and stimulated Raman scattering imaging to examine the structure-activity-distribution relationship of ferroptosis-modulating compounds. We found that, although lipid peroxidation in various subcellular membranes can induce ferroptosis, the endoplasmic reticulum (ER) membrane is a key site of lipid peroxidation. Our results suggest an ordered progression model of membrane peroxidation during ferroptosis that accumulates initially in the ER membrane and later in the plasma membrane. Thus, the design of ER-targeted inhibitors and inducers of ferroptosis may be used to optimally control the dynamics of lipid peroxidation in cells undergoing ferroptosis.

PubMed Disclaimer

Figures

Extended Data Fig. 1
Extended Data Fig. 1. Dose-response of D-PUFAs, peroxisome and mitochondrial staining of D-PUFA-treated cells, and mitochondrial/ER quantification of D-PUFAs.
A. Dose-response curves of HT-1080 cells pretreated with varying concentrations of D-PUFAs and then treated with FINs. Data are plotted as mean ± SEM, n=3 biologically independent samples. B. HT1080 cells treated for 24 hours with 20 μM ARA-d6 and CellLight Peroxisome-GFP, and imaged by fluorescence and SRS imaging. C. HT-1080 cells treated for 24 hours with 20 μM DHA-d10, and then stained with MitoTracker Red CMXRos and imaged by fluorescence and SRS imaging. D. High resolution SRS imaging of HT-1080 cells treated for 24 hours with 20 μM DHA-d10 with DGAT inhibitors PF-06424439 (1 μM) and A922500 (1 μM), then stained with MitoTracker Red CMXRos and ERTracker Green and imaged by fluorescence and SRS imaging. E. Quantification of arachidonic acid-d11 in mitochondrial and ER fractions isolated from HT1080 cells treated at 20 μM for 24 hours. Values determined by high resolution mass spectrometry and plotted as normalized to internal standard. Data are plotted as mean of 3 biological replicates ± SEM. F. Western blotting of mitochondrial and ER fractions stained for PDI as ER marker and Cytochrome C as mitochondrial marker, representative of four experiments. Results indicate no mitochondrial contamination of ER fraction, but indicate ER contamination of mitochondrial fraction.
Extended Data Fig. 2
Extended Data Fig. 2. Knockdown of PUFA-related genes shows some impact on D-PUFA potency, but no observable decrease in incorporation
A. Dose-response curves of stable non-targeting (NT) or ACSL knockdown HT-1080 cells pretreated with EtOH or D-PUFAs and then treated with RSL3. Data are represented as mean ± SEM, n=3. B. Dose-response curves of stable nontargeting (NT) or AGPAT3 knockdown HT-1080 cells pretreated with EtOH or D-PUFAs and then treated with RSL3 or IKE. Data are represented as mean ± SEM, n=3. C. qPCR data of stable shRNA knockdowns. Data are represented as mean ± SD of three technical replicates. D. SRS images of shNT, shACSL5, and shAGPAT3 HT-1080 cells treated with DHA-d10 (10 μM) (left), and quantification of their signal intensity (right), Data are plotted as mean ± SEM, n=3.
Extended Data Fig. 3
Extended Data Fig. 3. Thimerosal, Miltefosine, and N-Ethyl Maleimide do not impact anti-ferroptotic potency of D-PUFAs, and knockdown of ER-phagy related genes SEC62, RTN3, and FAM134B did not result in apparently altered ER area
A. Dose-response curves of HT-1080 cells treated with vehicle (water) or 400 nM thimerosal, and subsequently treated with vehicle (EtOH) or PUFAs 4 hours later, followed by varying concentrations of IKE and RSL3 24 hours later. Data are represented as mean ± SEM, n=3. B. Dose-response curves of HT-1080 cells treated with vehicle (water) or 7.5 μM miltefosine, and subsequently treated with vehicle (EtOH) or PUFAs 4 hours later, followed by varying concentrations of RSL3 24 hours later. Data are represented as mean ± SEM, n=3. C. Dose-response curves of HT-1080 cells treated with D-PUFAs and either pretreated, cotreated, or post-treated with vehicle (EtOH) or 4.5 μM NME, and subsequently treated with varying concentrations IKE and RSL3. In the post-treatment experiments, media containing PUFAs was removed before NME was added. Data are represented as mean ± SEM, n=3. D. Dose-response curves of stable shNT, shSEC62, shRTN3, and shFAM134B HT-1080 cells treated with varying doses of IKE, RSL3, FIN56, and FINO2. Data are represented as mean ± SEM, n=3. E. ER area of ER-Phagy knockdown cell lines as compared to control. Cells were stained with ERTracker Blue-White, imaged with confocal microscopy, and their ER areas were measured using the CellProfiler software. Individual ER areas are shown for each cell, as well as the mean ± SD. Sample sizes (number of cells) are as follows: shNT n=123, shFAM134B n=98, shRTN3 n=60, shSEC62 n=26. F. qPCR analysis of knockdowns of ER-phagy genes in HT-1080 cells. Data are represented as mean of 3 technical replicates ± SD.
Extended Data Fig. 4
Extended Data Fig. 4. Overexpression and targeting of organelle-phagy related genes did not result in apparently altered ER area.
A. Confocal fluorescence images of HT-1080 cells overexpressing GFP, GFP-PLA2G16, GFP-FUNDC1 (as well as the cytoplasmic N-terminal FUNDC1 sequence), with and without c terminal- cytochrome b5 ER targeting signals, stained with ERTracker Red. GFP channels show protein distribution as all overexpressed proteins are tagged with GFP. Representative images of at least 6 images per sample are shown. B. Dose-response curves of overexpression cell lines with varying doses of RSL3. Data are represented as mean ± SEM, n=3 biologically independent samples. C. ER area overexpression cell lines as compared to control. Cells were stained with ERTracker Red, imaged with confocal microscopy, and their ER areas were measured using the CellProfiler software. Violin quartile plots are shown. Sample sizes (number of cells) are as follows: GFP n=358, GFP-cyb5 n=478, PLA2G16 n=353, PLA2G16-cyb5 n=292, FUNDC1 n=354, FUNDC1-cyb5 n=212, Nterm-FUNDC1-cyb5 n=399.
Extended Data Fig. 5
Extended Data Fig. 5. Subcellular localization and effect on ferroptosis of myristic acid and cholesterol, overexpression and distribution of ACSL4.
A. Structure and SRS image of myristic acid-d27 (20 μM) in HT-1080 cells. B. Structure and SRS image of cholesterol-d6 (20 μM) in HT-1080 cells. C. Fluorescence and confocal imaging of cholesterol-d6 to evaluate its subcellular localization. D. Solution Raman spectra of the FAs and cholesterol used in these experiments. E. Dose-response curves of HT-1080 cells pretreated with either MA-d27 or cholesterol-d6 (20 μM) followed by varying concentrations of RSL3. Data are represented as mean ± SEM, n=3. F. Western blot of HT-1080 cells overexpressing GFP (control) or GFP-ACSL4. ACSL4 antibody was used, with actin as a control. G. Immunofluorescence staining of HT1080 cells labeled for ACSL4 (anti-ACSL4 antibody), ER (anti-calnexin antibody), and nucleus (DAPI). Individual channels and overlay is shown. H. Western blotting of mitochondrial and ER HT1080 fractions stained for ACSL4 PDI (ER), and Cytochrome C (mitochondria), indicating presence in both fractions, representative of two experiments.
Extended Data Fig. 6
Extended Data Fig. 6. FINO2-0 and FINO2-2 accumulate in the ER, and FINO2-1/FINO2-3/FINO2-4 do not accumulate in the plasma membrane.
A. Structure of FINO2-0. B. Dose-response curve of HT-1080 cells treated with FINO2-0 ± fer-1. Data are represented as mean ± SEM, n=3. C. Confocal fluorescence images of HT-1080 cells treated for 3 hours with 3 μM FINO2-1 alone or with 3 μM fer-1. D. Confocal fluorescence image of HT-1080 cells treated with 3 μM FINO2-0. F. Confocal fluorescence imaging of HT-1080 cells treated with FINO2-1 (3 μM) and fer-1 (3 μM) for 3 hours, co-stained with BODIPY TR ceramide. G. Structure of FINO2-2. H. Dose-response curve of HT-1080 cells treated with FINO2-2 ± fer-1. Data are represented as mean ± SEM, n=3. I. SRS and fluorescence imaging of HT-1080 cells treated with 20 μM FINO2-2 and 2 μM fer-1, and stained with Nile Red and Lysotracker green. J. Confocal fluorescence image of HT-1080 cells treated with 3 μM FINO2-1 and 3 μM fer-1, and costained with CellMask Deep Red. K. Confocal fluorescence image of HT-1080 cells treated with 3 μM FINO2-3 or 3 μM FINO2-4, and 3 μM fer-1, and co-stained with CellMask Deep Red.
Extended Data Fig. 7
Extended Data Fig. 7. Analogs of FINO2 redistribute throughout the cell.
A. Structures of analogs of FINO2. B. Confocal fluorescence images of HT-1080 cells treated with FINO2-5 (3 μM) or FINO2-6 (3 μM) and fer-1 (3 μM), and co-stained with LysoTracker Red. C. Dose-response curves of HT-1080 cells treated with fixed concentrations of FINO2 analogs (10 μM) and varying concentrations of ferroptosis inhibitors. Lysosome-directed ferrostatin previously published by Gaschler et al.27 Data are represented as mean ± SEM, n=3. D. Confocal fluorescence image of HT-1080 cells treated with FINO2-7 (3 μM) and fer-1 (3 μM). E. Dose-response curves of HT-1080 cells comparing treatment with FINO2-1 and FINO2-3 or FINO2-4. Data are represented as mean ± SEM, n=3.
Extended Data Fig. 8
Extended Data Fig. 8. Ferroptosis induced by DHODH inhibition is amplified by ER peroxidation, ferroptosis induced by IKE results in ER membrane peroxidation followed by PM peroxidation, early perinuclear lipid peroxidation occurs in the ER.
A. Brequinar (BQR) induces ferroptosis in HT1080s as rescued by fer-1, and cotreatment with BQR (500 μM) increases sensitivity to RSL3 in HT1080s. Data are represented as mean ± SEM, n=3. B. Dose-response curve of HT1080 cells cotreated with 1 μM of FINO2-1, FINO2-3, or FINO2-4. Data are represented as mean ± SEM, n=3. C. HT-1080 cells treated with DMSO or IKE (10 μM) for 6 hours and stained with C11 BODIPY (oxidized and reduced overlay), CellMask Deep Red, and ERTracker Blue-White. D. HT-1080 cells treated with DMSO or IKE (10 μM) for 10 hours and stained with C11 BODIPY (oxidized and reduced overlay), CellMask Deep Red, and ERTracker Blue-White. E. Correlation (Manders coefficient) of C11 BODIPY oxidized signal with ERTracker Blue-White or MitoTracker Deep Red in HT1080 cells treated with either RSL3, IKE, FINO2, or FIN56 at designated timepoints. Data are represented as mean ± SEM, each individual point represents an image of multiple cells. Number of images for each condition are RSL3 2 hour (n=3), 5 hour (n=2); IKE 2 hour (n=7), 5 hour (n=5), FINO2 2.5 hour (n=6), 4.5 hour (n=9), FIN56 2 hour (n=6), 4 hour (n=5), 5 hour (n=2). Ordinary one way ANOVA with Tukey’s test for multiple comparisons was used with p values of: RSL3 (0.0016, 0.6120), IKE (0.0023, 0.1279), FINO2 (<0.0001, <0.0001), FIN56 (0.0034, 0.0005, >0.9999). F. Correlation (Manders coefficient) of C11 BODIPY oxidized signal with ERTracker Blue-White or MitoTracker Deep Red in HT1080 cells treated with either RSL3 (1 μM), BQR (1 mM) or both at designated timepoints. Data are represented as mean ± SEM, each individual point represents an image of multiple cells. Number of images for each condition are RSL3 1.5 hour (n=4), 2 hour (n=2); BQR 2 hour (n=3), 2.5 hour (n=3); RSL3 + BQR 1.5 hour (n=2), 2 hour (n=2). Two-sided unpaired t test was used with p values of: RSL3 (<0.0001, 0.0178), BQR (0.0006, 0.1011), RSL3 + BQR (0.0117, 0.2027). For all panels, GraphPad Prism (GP) P value style of 0.1234 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****).
Figure 1.
Figure 1.. Exogenous deuterated polyunsaturated fatty acids rescue against ferroptosis and accumulate perinuclearly and in puncta.
A. Structures of polyunsaturated fatty acids deuterated at their bis-allylic sites. B. Rescue of HT-1080 cells treated with each of four classes of ferroptosis inducers after a 24-hour pre-treatment with D-PUFAs. Data are represented as mean ± SEM, n=3. C. SRS images of HT-1080 cells treated for 24 hours with 20 μM DHA-d10. The protein (CH3) and lipid (CH2) cell-intrinsic vibrational frequencies are shown as well for comparison. Red arrow points to lipid droplets and white arrow points to perinuclear accumulation. D. SRS imaging of HT-1080 cells treated for 24 hours with the three different D-PUFAs used in this study, ARA-d6 (80 μM), EPA-d8 (20 μM), and DHA-d10 (20 μM). Red arrow points to lipid droplets and white arrow points to perinuclear accumulation. E. SRS imaging of Panc-1, N27, and HT-22 cells treated for 24 hours with 20 μM DHA-d10.
Figure 2.
Figure 2.. D-PUFA accumulation in lipid droplets does not play a role in inhibition of ferroptosis.
A. SRS and fluorescence imaging of HT-1080 cells treated with 20 μM DHA-d10, and then stained with Nile Red (shown in green for consistency with SRS image) and Lysotracker Green (in red). B. Heatmap showing relative incorporation of deuterated DHA into HT-1080 triglycerides, as measured by LC-MS. Vehicle showed no deuterated incorporation, whereas DHA-d10 had varying levels of incorporation into triglycerides and cholesterol esters, with different sum total numbers of carbons and double bonds. Data shown are an average of absolute signal intensity of three biological replicates. C. SRS and fluorescence imaging of HT-1080 cells treated with DHA-d10 (20 μM) ± cotreatment with DGAT inhibitors PF-06424439 (1 μM) and A922500 (1 μM), and then stained with Nile Red. D. Rescue of HT-1080 cells from erastin or RSL3 lethality with pre-treatment of DHA-d10 ± co-treatment of DGAT inhibitors. Data are represented as mean ± SEM, along with individual data points, n=3. Statistics performed using two-sided unpaired t test.
Figure 3.
Figure 3.. Anti-ferroptotic D-PUFAs incorporate into ER phosphatidylethanolamine phospholipids and ether phospholipids.
A. SRS and fluorescence imaging of HT-1080 cells treated with DHA-d10 (20 μM) ± co-treatment with DGAT inhibitors PF-06424439 (1 μM) and A922500 (1 μM), and then stained with ERTracker Green. B. SRS and fluorescence imaging of HT-1080 cells treated with DHA-d10 (20 μM) and then stained with ERTracker Green and BODIPY TR ceramide (a Golgi stain). White arrows indicate the Golgi region. C. Heatmap showing incorporation of deuterated DHA into HT-1080 phospholipids as measured by LC-MS. Vehicle shows no deuterated incorporation, whereas DHA-d10 has varying incorporation depending on the lipid species and fatty acid compositions. PE: phosphatidylethanolamine, PI:phosphatidylinositol, PS:phospatidyserine, PC: phosphatidylcholine. Data shown are an average of absolute signal intensity of three biological replicates.
Figure 4.
Figure 4.. Enriching the ER and PM with pro- or anti-ferroptotic lipids modulates cell sensitivity to ferroptosis.
A. Structures of deuterated fatty acids. Of note, the PUFAs are deuterated at non-bis-allylic positions and therefore do not inhibit ferroptosis. B. SRS images of HT-1080 cells treated with each of the deuterated fatty acids. C. Effect of pre-treatment (24 hours) with fatty acids on ferroptosis induced by RSL3 in HT-1080 cells, as compared to ethanol vehicle. Data are represented as mean ± SEM, n=3. Two-sided unpaired t test was performed with p values of 0.00001, 0.000053, 0.010389, 0.012767. D. Increase in sensitivity to ferroptosis in cells overexpressing GFP-ACSL4 when treated with equivalent dose of PUFAs, as compared to parental HT-1080 cells. Data are represented as mean ± SEM, n=3. 2way ANOVA with Tukey’s multiple comparison test was performed with p values of 0.5037, 0.0004, <0.0001, <0.0001, <0.0001, <0.0001, <0.0001, 0.0001, 0.0033. E. Representative imaging of cells stained with ERTracker Green and FM 4-64 to label ER and PM, respectively, and the resulting masks generated in MATLAB. F. Quantification of relative fatty acid incorporation as CD signal to general lipid CH2 signal ratio within the PM and ER masks as shown in 4E. Data are represented as mean ± SEM with each point representing an individual cell. Two-sided unpaired t test resulted in ARA n=37 p<0.0001, OA n=41 p<0.0001, MA n=26 p=0.016. For all panels, GraphPad Prism (GP) P value style of 0.1234 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****).
Figure 5.
Figure 5.. FINO2 analogs can induce ferroptosis through accumulation in the ER, lysosome, or mitochondria.
A. Structures of FINO2 and analog FINO2-1 with an added fluorescent naphthalimide label. B. Dose response curves of FINO2 and FINO2-1, ± fer-1 (2 μM). Data are represented as a mean ± SEM, n=3. C. Confocal fluorescence imaging of HT-1080 cells treated with FINO2-1 (3 μM) and fer-1 (3 μM) for 3 hours, co-stained with ERTracker Red. D. Confocal fluorescence imaging of HT-1080 cells treated with FINO2-1 (3 μM) and fer-1 (3 μM) for 3 hours, co-stained with BODIPY TR ceramide. E. Structures of FINO2analogs FINO2-3 and FINO2-4. F. Confocal fluorescence imaging of HT-1080 cells treated with FINO2-3 (3 μM) and fer-1 (3 μM) for 3 hours, co-stained with MitoTracker Red CMXRos. G. Confocal fluorescence imaging of HT-1080 cells treated with FINO2-4 (3 μM) and fer-1 (3 μM) for 3 hours, co-stained with LysoTracker Red. H. Rescue of HT-1080 cells treated for 24 hours with FINO2-1 (2.5 μM), FINO2-3 (5 μM), or FINO2-4 (5 μM) by DFO (left, 10 μM cotreatment), fer-1 (center, 2 μM cotreatment), or DHA-d10 (right, 20 μM 24-hour pretreatment). Data are represented as mean ± SEM, n=3 biologically independent samples.
Figure 6.
Figure 6.. Ferroptosis induced by RSL3, FIN56, FINO2, or IKE results in ER peroxidation followed by PM peroxidation.
A. Quantification of C11 BODIPY oxidized:reduced ratio within the ER and PM of HT-1080 cells treated with DMSO, RSL3 (0.5 μM), FIN56 (10 μM), or FINO2 (10 μM) at 2 and 5 hours. CellMask Deep Red (PM) and ERTracker Blue-White (ER) were used to select regions of interest in CellProfiler within which C11 BODIPY signal was quantified. Data are represented as mean ± SEM, with each point representing a single cell. Sample sizes are 2 hours ER (n=18, 19, 25, 18), PM (n=18, 19, 25, 14) and 5 hours ER (n=21, 24, 28, 36), PM (n=21, 22, 28, 36). Brown-Forsythe and Welch ANOVA with Dunnet T3 test for multiple comparisons was used with p values of: ER 2 hours (<0.0001, 0.0007, <0.0001), PM 2 hours (0.6485, 0.9049, 0.1531), ER 5 hours (<0.0001, <0.0001, <0.0001), PM 5 hours (<0.0001, <0.0001, <0.0001) B. Quantification of C11 BODIPY oxidized:reduced ratio within the ER and PM of HT-1080 cells treated with DMSO or IKE (10 μM) at 6 and 10 hours. CellMask Deep Red (PM) and ERTracker Blue-White (ER) were used to select regions of interest in CellProfiler within which C11 BODIPY signal was quantified. Data are represented as mean ± SEM, with each point representing a single cell. Sample sizes are 6 hours ER (n=20, 32), PM (n=14, 23) and 10 hours ER (n=20, 24), PM (n=14, 19). Two-sided unpaired t test was used with p values of: ER 6 hours (<0.0001), PM 6 hours (0.4385), ER 10 hours (<0.0001), PM 10 hours (<0.0001) C. Representative images of HT-1080 cells treated with DMSO, RSL3 (0.5 μM), FIN56 (10 μM), or FINO2 (10 μM) for 2 hours and stained with C11 BODIPY (oxidized and reduced overlay), CellMask Deep Red, and ERTracker Blue-White. D. Representative images of HT-1080 cells treated with DMSO, RSL3 (0.5 μM), FIN56 (10 μM), or FINO2 (10 μM) for 5 hours and stained with C11 BODIPY (oxidized and reduced overlay), CellMask Deep Red, and ERTracker Blue-White. For all panels, GraphPad Prism (GP) P value style of 0.1234 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****).
See this image and copyright information in PMC

References

    1. Li J et al. Ferroptosis: past, present and future. Cell Death Dis. 11, (2020). - PMC - PubMed
    1. Liang C, Zhang X, Yang M & Dong X Recent Progress in Ferroptosis Inducers for Cancer Therapy. Adv. Mater 31, 1–25 (2019). - PubMed
    1. Su Y et al. Ferroptosis, a novel pharmacological mechanism of anti-cancer drugs. Cancer Lett. 483, 127–136 (2020). - PubMed
    1. Reichert CO et al. Ferroptosis mechanisms involved in neurodegenerative diseases. Int. J. Mol. Sci 21, 1–27 (2020). - PMC - PubMed
    1. Ye LF et al. Radiation-Induced Lipid Peroxidation Triggers Ferroptosis and Synergizes with Ferroptosis Inducers. ACS Chem. Biol 15, 469–484 (2020). - PMC - PubMed

Methods References

    1. Schindelin J et al. Fiji: An open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012). - PMC - PubMed
    1. McQuin C et al. CellProfiler 3.0: Next-generation image processing for biology. PLoS Biol. 16, 1–17 (2018). - PMC - PubMed
    1. Kraft VAN et al. GTP Cyclohydrolase 1/Tetrahydrobiopterin Counteract Ferroptosis through Lipid Remodeling. ACS Cent. Sci 6, 41–53 (2020). - PMC - PubMed
    1. Smith CA, Want EJ, O’Maille G, Abagyan R & Siuzdak G XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal. Chem 78, 779–787 (2006). - PubMed
    1. Core Team, R. R: A language and environment for statistical computing. (2014).

Publication types