Compounds

1 S,3R-RSL3 (Tocris: 6118), ferric ammonium citrate (Sigma Aldrich), UAMC-3203 and UAMC-3206 (provided by the Laboratory of Medicinal Chemistry, University of Antwerp, Belgium) [10], zVAD-fmk (Enzo Life Sciences: ALX-260-020-M001), Necrostatin-1s (Nec-1s, Abcam: ab221984), ferrostatin-1 (Fer-1, Sigma Aldrich: SML0583), Deferoxamine (DFO, Sigma Aldrich: D9533), vitamin E (α-tocopherol; Sigma Aldrich: T3251-5G).

Mouse strains

Wild-type (WT) mice (C57BL/6J, Jackson Laboratories) and GPX4Tg/+ mice [17] overexpressing GPX4 were used for isolation of primary bone marrow-derived macrophages (BMDMs). To study the in vivo stability of UAMC-3203, female WT mice (3 months of age, C57BL/6J, Jackson Laboratories) were treated with 12.35 mg/kg/day UAMC-3203 or saline solution via subcutaneously implanted osmotic minipumps (Alzet pump type 1004, Charles River) for 2 weeks. For atherosclerosis studies, female ApoE−/−Fbn1C1039G+/− mice (aged 6–8 weeks) were fed a Western-type diet (WD; TD.88137 supplemented with 21% fat and 0.2% cholesterol, Envigo) for 12 or 20 weeks. Mice were divided over 2 treatment regimens with their respective controls (Fig. 1). After 4 or 12 weeks WD, mice were treated with 12.35 mg/kg/day UAMC-3203 or saline solution via subcutaneous osmotic minipumps (Alzet pump type 1004, 4 weeks, Charles River) for 8 weeks in total. The first osmotic minipump, containing saline solution or UAMC-3203, was replaced after 4 weeks by a second minipump to reach a total of 8 weeks of treatment in all groups. Animals were housed in a temperature-controlled room with a 12 h light/dark cycle and had free access to water and food. Cases of sudden death and head tilt were documented. The subcutaneous implantation of osmotic minipumps was performed under anesthesia (isoflurane 3–4% and O2 in a ventilated gas mixture) and on a heating element to prevent hypothermia. After shaving and disinfecting the site of implantation, an incision was made on the upper dorsal side. A subcutaneous pocket was made using a hemostat. Subsequently, the filled pump was inserted in the pocket (on opposite sides for pump 1 and pump 2) and the wound was closed with interrupted sutures (size 5 − 0 Ethilon™ II suture on a 19 mm 3/8ths of a circle cutting needle). At the end of the study, an overdose of sodium pentobarbital (250 mg/kg, i.p.) was administered and tissues were collected for histological analysis. Further, blood samples were collected via the retro-orbital plexus. Plasma levels of total cholesterol were measured using a commercially available kit (Randox laboratories).

Fig. 1

Atherosclerosis study with UAMC-3203. Treatment regimen of ApoE−/−Fbn1C1039G+/− mice with western-type diet (WD) and osmotic minipumps containing saline solution or UAMC-3203

Histological analyses

The thoracic and abdominal aorta were stained en face with Oil Red O to determine lipid burden. The proximal ascending aorta and right common carotid artery (RCCA) were fixed in 4% formaldehyde (pH 7.4) for 24 h, dehydrated overnight in 60% isopropanol and subsequently embedded in paraffin. Serial cross-sections (5 μm) of the proximal ascending aorta and serial longitudinal sections (5 μm) of the RCCA were prepared at random for histological analyses. Atherosclerotic plaque thickness and necrotic core area (defined as acellular areas with a threshold of 3000 µm2) were analyzed on hematoxylin-eosin (H&E) stained sections. Plaque thickness was measured on 10 random locations over the full length of a longitudinal section of the RCCA and mean plaque thickness was calculated. Total collagen was measured on Sirius red stained sections. For immunohistochemistry, the following antibodies were used: anti-Mac3 (BD Pharmingen, 550292); anti-α-smooth muscle actin (α-SMA; Sigma, 12547), anti-heme-oxygenase 1 (HMOX1; Enzo Life Sciences, ADI-SPA-896), anti-Ferritin heavy chain (FTH; Abcam, ab185781) and anti-TER119 (BD Pharmingen, 550565). Immunohistochemical staining for HMOX1, FTH and TER119 were always performed on serial sections. Images were acquired with an Olympus BX43 microscope. Plaques were manually delineated in Image J software (National Institutes of Health) to establish the region of interest (ROI). Further analyses within the ROIs were performed using color thresholding or manual counting (apoptotic cells). For all histological analyses, the observer was blinded to the received treatment (UAMC-3203 or saline).

Cell culture

Bone marrow-derived macrophages (BMDMs) were isolated by flushing femurs of C57BL/6J mice with RPMI medium containing L-glutamine and HEPES (Gibco Life Technologies) and supplemented with 10 U/mL heparin (Sigma). The cell suspension was pipetted through a 40 μm cell strainer (Greiner Bio-One: Easystrainer™ 542,040), washed twice and resuspended in RPMI medium with Glutamax™. The RPMI medium was supplemented with 10% FBS, 1% penicillin-streptomycin (10.000 U/mL, Gibco Life Technologies) and 0.2% polymyxin B (10.000 U/mL, Fagron) and 15% L-cell conditioned medium (LCCM) containing monocyte colony stimulating factor (M-CSF).

Recently expired (up to 5–7 days) concentrates of human erythrocytes or blood platelets were obtained from the Belgian Red Cross. Erythrocyte concentrates were stored at 4 °C for maximum 2 months. Before use, erythrocyte concentrates were freshly diluted 1:1 with PBS, centrifuged for 5 min at 201×g and washed with PBS. The number of intact erythrocytes was counted to ensure that macrophages were treated with the same number of erythrocytes in each experiment. Platelet concentrates were kept shaken at room temperature for maximum 2 months and were centrifuged for 10 min at 1258×g and washed twice with PBS. Erythrocyte concentrates from different donors and pooled platelet concentrates were used between experiments to account for the natural variation that exists between humans.

At day 7, 0.5 × 106 BMDMs were incubated with 4 × 107 erythrocytes (BMDM:erythrocyte ratio = 1:80) for 0.5, 1, 2 and/or 4 h, depending on the experimental set-up. BMDMs treated with vehicle (DMEM supplemented with 10% FBS, 1% penicillin-streptomycin and 0.2% polymyxin B) or with 108 platelets (BMDM:platelet ratio = 1:200) served as controls. After phagocytosis of erythrocytes or platelets, BMDMs were washed twice with PBS to remove free erythrocytes or platelets. Remaining erythrocytes in the culture medium or attached to (though not internalized by) BMDMs were removed with Erythrocyte Lysis Buffer (Abcam, ab204733). Finally, BMDMs were detached with 0.25% trypsin-EDTA (Thermo Fischer Scientific, 25200072) and analyzed by flow cytometry using a BD Accuri C6 flow cytometer.

To evaluate cell viability, a neutral red assay was performed as previously described [18]. Briefly, cells were incubated with 0.1% neutral red solution at 37 °C in 95% air/5% CO2. Subsequently, cells were washed with PBS and neutral red was extracted by adding 0.05 M NaH2PO4 in 50% ethanol. After 3 min, optical density was read at 540 nm using a microplate reader.

Flow cytometry

All sample analyses were performed on a BD accuri C6 flow cytometer. For each measurement, a gating strategy was applied based on fluorophore minus one (FMO) controls. Because both the treatment of BMDMs and the combination of fluorophores can result in background signals, there was always one FMO control per treatment per fluorophore included. Thus, for each treatment replicates were included to serve as FMO controls. FMO controls were treated in the same way as the samples and stained with all fluorophores except for the one they are controlling for. Debris was always gated out and at least 20,000 cells were measured.

Analysis of macrophage surface markers

Detached BMDMs were first treated with mouse anti-CD16/CD32 antibody solution (eBioscience, 14-0161-85; 1:50 dilution in FACS buffer [PBS with 0.1% bovine serum albumin and 0.05% sodium azide]) for 10 min at 4 °C to block nonspecific binding sites. After centrifugation (5 min, 201×g), BMDMs were resuspended in mouse anti-F4/80-FITC (Biolegend, 123,108; 1:50 dilution in FACS buffer) or mouse anti-F4/80-AF647 (Biolegend, 123,122; 1:500 dilution in FACS buffer) antibody solutions and incubated for 30 min at 4 °C.

Quantification of phagocytosis and cell death

Erythrocytes and platelets were stained with Cell Trace Far Red (CTFR; Thermo Fischer Scientific, C34564) according to the manufacturer’s protocol. Briefly, erythrocytes and platelets were counted and diluted in PBS. Subsequently, 1:1000 or 1:10,000 CTFR was added to the erythrocytes and platelets, respectively, and incubated for 20 min at 37 °C. After addition of an equal volume of DMEM, erythrocytes and platelets were centrifuged at 201×g and 1258×g, respectively, and resuspended in DMEM. Detached BMDMs were blocked and stained with F4/80-FITC macrophage surface marker as described above. Thereafter, BMDMs were washed once with FACS buffer. Propidium iodide (PI) was added to the samples just before measuring. The following gating strategy was used to identify macrophages with phagocytosed erythrocytes or platelets and dead/dying macrophages: F4/80-FITC positive in FL-1 channel for macrophages; F4/80-FITC positive in FL-1 channel and CTFR-positive in FL-4 channel for macrophages with phagocytosed CTFR-labelled erythrocytes or platelets; F4/80-FITC positive in FL-1 channel and PI-positive in FL-2 channel for dead/dying macrophages.

Quantification of lipid peroxidation

Detached BMDMs were incubated with 5 µM Bodipy C11 581/591 (Thermo Fischer Scientific, D3861) solution in FACS buffer for 30 min at 37 °C. Next, BMDMs were washed once with PBS and subsequently blocked for nonspecific binding and stained with F4/80-AF647 macrophage surface marker, as described above. After one additional washing step with FACS buffer, samples were immediately measured. The following gating strategy was used to characterize lipid peroxidation in macrophages: F4/80-AF647 positive in FL-4 channel for macrophages; F4/80-AF647 positive in FL-4 channel and Bodipy C11 positive in FL-1 channel for oxidized phospholipid (PL) fraction in macrophages; F4/80-AF647 positive in FL-4 channel and Bodipy C11 positive in FL-2 channel for reduced PL fraction in macrophages. The ratio of the mean fluorescence in FL-1 channel over the mean fluorescence in FL-2 channel was calculated to quantify lipid peroxidation. Positive controls consisted of BMDMs treated with 1 S,3R-RSL3.

Quantification of labile iron pool

Detached BMDMs were resuspended in 5 µM FeRhoNox-1 (Sigma-Aldrich, SCT030) and incubated for 1 h at 37 °C. After two washing steps with PBS, samples were immediately measured and labile iron was quantified in the FL-2 channel. Positive controls consisted of BMDMs treated with 1 S,3R-RSL3 and ferrous sulfate heptahydrate.

Quantification of apoptosis

Detached BMDMs were fixed with 4% paraformaldehyde for 1 h at room temperature. After washing with PBS, BMDMs were permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate solution for 2 min on ice. BMDMs were washed again and incubated with TUNEL reaction mixture using an in situ cell death detection kit (Roche, fluorescein, 1168479591) for 1 h at 37 °C. The samples were washed twice with FACS buffer and measured in the FL-1 channel. Positive controls consisted of BMDMs treated with TNFα and cycloheximide.

Confocal microscopy

Erythrocytes were stained with CTFR as described above. After phagocytosis, calcein green AM was added to the samples to stain BMDMs. Images were obtained using a microlens-enhanced dual spinning disk confocal microscope (UltraVIEW VoX; PerkinElmer) equipped with solid state 480 and 640 nm diode lasers for excitation of calcein green AM and CTFR, respectively. Images were processed using the reconstruction facilities of Volocity software (PerkinElmer).

Western blotting

Cells were lysed with Laemmli sample buffer (Bio-rad, 1610737) containing β-mercaptoethanol (Sigma-Aldrich) and boiled for 5 min. To separate nuclear and cytosolic fractions, a nuclear extraction kit (Abcam, ab113474) was used following the manufacturer’s protocol. Samples were loaded on 4–12% Bis-Tris gels (Life Technologies, NW04125BOX) for electrophoresis and transferred to Immobilon-FL PVDF membranes (Millipore, IPFL00010) according to standard procedures. Subsequently, membranes were blocked for 1 h in Odyssey Li-COR blocking buffer. After blocking, membranes were probed overnight at 4 °C with primary antibodies diluted in Odyssey Li-COR blocking buffer followed by 1 h incubation with IRDye-labeled secondary antibodies at room temperature. Membranes were visualized with an Odyssey SA infrared imaging system (Li-COR Biosciences). The following primary antibodies were used: mouse anti-HMOX1 (Abcam, ab13248), rabbit anti-FTH1 (Abcam, ab183781), rabbit anti-GPX4 (Abcam, ab125066) and mouse anti-β-actin (Abcam, ab8226). IRDye® 800CW goat anti-rabbit (Li-Cor, Li 926-32211) or IRDye® 680RD goat anti-mouse (Li-Cor, LI 926-68070) were purchased from Li-COR Biosciences.

Liquid chromatography-tandem mass spectrometry (LC-MS/MS)

A LC-MS/MS method was optimized and validated for the determination of UAMC-3203, as described previously [19]. UAMC-3203 detection in plasma, and tissue homogenates (lysed in PBS using a Precellys 24 Tissue Homogenizer (Bertin Instruments)) was performed on an Agilent 1200 series LC system connected to a 6410 triple quadrupole mass spectrometer from Agilent Technologies with electrospray ionization (ESI) interface operated in positive ionization mode. Chromatographic separation was carried out on a Kinetex Biphenyl column (100 × 2.1 mm, 2.6 μm; Phenomenex). The mobile phase consisted of (A) ultrapure water with 0.1% formic acid and (B) acetonitrile/ultrapure water (90/10) with 0.1% formic acid, in gradient at 0.3 mL/min. The ESI source parameters were gas temperature 350 °C, gas flow 10 L/min, nebulizer pressure 35 psi, and capillary voltage 4000 V. Data acquisition was done in multiple reaction monitoring mode (MRM). Confirmation of UAMC-3203 was done using three MRM transitions; the most abundant transition was used as a quantifier (Q) and the other two were used as a qualifier (q). Qualifier/quantifier ratios (q/Q) were calculated for each sample and had to be within ± 20% of the q/Q ratio observed in the calibrators. In addition, the retention time of the compound in samples could not deviate > 10% of the retention time observed in the calibrators.

UAMC-3203 and nordiazepam-D5 (Cerilliant Corporation) as internal standard (IS) were diluted in LC-MS grade methanol (Fisher Scientific). A volume of 100 µL sample was spiked with 20 µL IS (200 ng/mL), followed by the addition of 150 µL acetonitrile for plasma. For tissue homogenates, 500 µL acetonitrile with 0.1% formic acid was added. Subsequently, the mixture was vortexed (2 min) and centrifuged (10 min, 9168×g for plasma, 17,968×g for tissue). The supernatant of plasma was then transferred to a 2 mL tube with a 0.20 μm centrifugal filter (VWR). The supernatant of tissue was evaporated under a stream of nitrogen at 40 °C, reconstituted in 100 µL acetonitrile/ultrapure water (90/10) with 0.1% formic acid, and transferred to a 2 mL tube with a 0.20 μm centrifugal filter. All samples were then centrifuged (5 min 9168×g for plasma, 10 min 17,968×g for tissue), after which the final extract was transferred to an autosampler vial with a glass insert. Seven-level calibration curves were prepared in blank mouse plasma, covering a linear range from 10 to 700 ng/mL. Five-level calibration curves were prepared in blank homogenized mouse tissue matrix covering a linear range from 20 to 4000 ng/mL. The measured concentrations in ng/mL were further normalized using the weight of the tissue used for homogenization to obtain final concentrations of UAMC-3203 expressed in µg/g.

Statistical analyses

Statistical analyses were performed using Graphpad Prism 9 and SPSS software (version 27, SPSS Inc.). All data are expressed as the mean ± SEM. Datapoints in the graphs represent n samples from independent experiments or individual mice (i.e. biological repeats). Statistical tests and number of replicates are specified in the text and figure legends. If a two-way ANOVA was used, a significant main effect was further analyzed by performing a simple main effect analysis, including Bonferroni correction for multiple comparisons. Differences were considered significant when p < 0.05.