Caspase-8^flox/flox mice were kindly provided by Hedrick’s laboratory (UCSD) (14) and crossed with LysMCre transgenic mice (15) to generate cohoused caspase-8^komac littermates (caspase-8^{flox/flox-LysM-Cre}), recipient Ldlr^-/- female mice were obtained from The Jackson Laboratory. All mice used were maintained on a C57BL/6J genetic background. Animal procedures were performed in compliance with the ethical guidelines approved by Ethics Committee for the Use of Laboratory Animals of the University of Burgundy (Protocol No. 2017032015521461_v1#8381). Mice were fed either a standard chow diet (A3, Safe) or a Western diet (U8958, Safe) starting at 8 weeks of age.

Eight weeks old Ldlr^-/- female recipient mice were irradiated with a dose of 11 Gy. Subsequently, the mice received a tail vein injection of 2x10⁶ bone marrow cells isolated from caspase-8^flox/flox and caspase-8^komac male mice. A recovery period of 4 weeks followed the injection, during which the drinking water was supplemented with enrofloxacin antibiotic (100 mg/L), renew on a weekly basis.

After the recovery period, the mice were fed with a western type diet (U8958, Safe) containing: 7% PM 205B SAFE; 1% PV 200 SAFE; 17% casein acid; 0.3% methionine; 34% saccharose; 14.5% corn starch; 0.2% cholesterol; 5% cellulose; 21% dairy butter for 12 weeks. At the end of the dietary intervention, mice were anesthetized with isoflurane and blood samples were collected via intracardiac puncture. The mice were then euthanized by cervical dislocation and the bones, heart, aorta and carotid arteries were harvested. Heart was perfused with PBS, fixed in 4% paraformaldehyde (PFA embedded in paraffin and sectioned for quantification of aortic valves with hematoxylin-eosin (HE) staining. Quantitative analysis was performed using NDP.View2 software, version 2.7.25 from Hamamatsu by three independent investigators blinded to the experimental conditions. Aorta was fixed in 4% PFA for en face staining.

Casp8^flox/flox , Casp8^komac and wild type C57BL/6J mice were anesthetized under isoflurane and euthanized by cervical dislocation. Femurs and tibias were harvested, and bone marrow cells were flushed out using RPMI 1640 medium supplemented with 5% FBS and 1% penicillin-streptomycin. Cells were seeded at a density of 5x10⁵ in 12-well plates or 1x10⁶ cells in 6-well plates. Differentiation into macrophages was induced by adding recombinant M-CSF (130-096-492, Miltenyi) for 5 to 6 days until complete differentiation.

Plasma lipid parameters were assessed using a Victor² 1420 Multilabel Counter (PerkinElmer Life Science, Boston, MA). Cholesterol and triglyceride levels were quantified through enzymatic assays (Diasys; Catalog No. 113009910026, 157109910026), following the manufacturer’s instructions. To analyze lipoprotein fractions, pooled plasma samples from Casp8^flox/flox and Casp8^komac mice were prepared. The separation of lipoproteins was performed using an FPLC system (UPC-900, P-920, Frac-900, AKTA). Cholesterol content in each fraction was subsequently determined using an enzymatic cholesterol assay, as previously described.

White blood cell counts were determined from blood samples collected at the time of sacrifice. A 50 µL aliquot was analyzed using a Scil vet abc Plus+ hematology analyzer (Scil).

Erythrocytes were lysed using ACK lysis buffer, and the remaining nucleated cells were harvested. Cells were washed with ice-cold PBS and blocked with Fc receptor blocker (Murine TruStain FcX, Biolegend, diluted 1:50) for 15 minutes. Cells were then incubated with specific antibodies for 20 minutes at 4°C for white blood cell phenotyping. The following antibodies were used: CD45-BUV395 (BD, 564279), CD11b-BV510 (Biolegend, 101245), Ly6G-APC-Cy7 (Biolegend, 127624), Ly6C-PE (Biolegend, 128008), CX3CR1-BV711 (Biolegend, 149031), and CD192-PE-Cy7 (Biolegend, 150612), following the manufacturers’ protocols. After washing, fluorescence was detected using a BD LSR Fortessa X-20, and the data were analyzed using FlowJo software (v10.9.0). The gating strategy is depicted Supplementary Figure S1 .

Cytokines were analyzed in the murine plasma, using the Mouse Pro-Inflammatory Panel 1 V-Plex (MSD, Rockville, MD, US). The kit was run according to the manufacturer’s guidelines and the chemi-luminescence signal was measured on a Sector Imager 2400 (MSD).

After 5 to 6 days of differentiation, macrophages derived from C57BL/6J mice were either treated or left untreated with the caspase-8 specific inhibitor Z-IETD-FMK (10 µM; Invivogen) for 2 hours, followed by exposure to 7-ketocholesterol (40 µM; Sigma, 700015P) for either 8 or 18 hours. For Casp8^flox/flox and Casp8^komac mice, macrophages were differentiated and treated with ox-LDL (50 µg/mL) for 18 hours. Cells were then harvested for subsequent western blot or flow cytometry analysis.

A TUNEL assay was performed on murine valves using the Apoptag Peroxidase In Situ Apoptosis Detection Kit (Sigma, S700), following the manufacturer’s protocol. Two independent investigators, blinded to the experimental conditions, quantified cell death. Additionally, apoptosis in bone-marrow-derived macrophages (BMDMs) was assessed using the FITC Annexin V Apoptosis Detection Kit (BD, 556547). After treatment, 5x10⁵ cells were washed with PBS and resuspended in 1X Annexin V binding buffer containing CaCl₂. Each sample was incubated with 5 µL of FITC Annexin V and 5 µL of Propidium Iodide (PI) for 15 minutes at room temperature, in the dark. Finally, 400 µL of 1X binding buffer was added, and fluorescence was measured using a BD LSR Fortessa. Data were analyzed with FlowJo software (v10.9.0).

Following treatment, 1x10⁶ cells were harvested in 100 µL of RIPA buffer (Fisher, 10230544) supplemented with a protease inhibitor (Fisher, 15634189) and Halt phosphatase inhibitor cocktail (Fisher, 10668304). After centrifugation at 15,000 g for 10 minutes, protein quantification was performed using a BCA assay. A total of 10 to 20µg of protein was mixed with loading buffer and reducing agent (Invitrogen, 2550989-2550991). SDS-PAGE was conducted, and proteins were transferred onto a 0.2 µm PVDF membrane (Bio-Rad, 1704272) using a Bio-Rad transblot system. Membranes were blocked with PBS containing 5% BSA and 0.1% Tween-20 for 1 hour at room temperature, followed by incubation overnight at 4°C with primary antibodies (diluted 1:1000 in PBS-5% BSA-0.1% Tween-20). After washing with PBS-0.1% Tween-20, membranes were incubated with HRP-conjugated secondary antibodies (1:2000) for 2 hours. Detection was performed using Pierce ECL substrate (Fisher, 10005943) or SuperSignal West Atto Ultimate Sensitivity Substrate (Fisher, 171181819). The following primary antibodies were used: anti-caspase-3 (Cell Signaling, 9661), anti-caspase-7 (Cell Signaling, 9492), anti-caspase-8 (Cell Signaling, 4790), anti-RIP (Cell Signaling, 3493), anti-RIP3 (Abcam, ab62344), anti-phospho-MLKL (Abcam, ab196436), anti-MLKL (Abcam, ab184718), anti-caspase-1 (Abcam, ab179515), anti-GSDMD (Novus, NBP2-33422) and anti-actin (Abcam, ab49900). HRP-conjugated secondary antibodies included anti-goat (Abcam, ab97100) and anti-rabbit (Cell Signaling, 7074). Caspase 7 and 8 were sequentially revealed from the same membranes after an intermediate stripping step. The same process was applied for pMLKL and MLKL. Non-specific bands are marked with an asterisk. For RIPK1 and RIPK3, the membranes were sectioned at the appropriate molecular weights before hybridization and detection. Uncropped images of the membranes are provided in Supplementary Figure S2 .

Serial sections of mouse heart tissue were prepared and subjected to immunohistochemical staining to evaluate MLKL and phosphorylated-MLKL (pMLKL) expression. Tissue sections were incubated with primary antibodies 1 hour at 1/200 in TBST-BSA 1%: anti-p-MLKL (NBP2-66953, Novus Biologicals) and anti-MLKL (3D4C6, Proteintech). After washing, sections were incubated with ImmPRESS^® polymer reagents for IHC (Vector Labs, MP-7401, MP-2400) followed by detection using Vector NovaRED^® substrate kit, peroxidase HRP (Vector Labs, SK-4800) as a chromogen. Sections were then counterstained with hematoxylin, dehydrated and mounted. Stained slides were scanned using a high-resolution digital slide scanner. Image analysis was performed using Aperio ImageScope software (v12.4.6.5003). Unsupervised quantification of chromogen-positive staining and total cell count within atheroma plaque was performed using ImageJ software (v1.53V with Java 1.8.0_322).

Data from one representative experiment from 2 to 7 independent experiments were analyzed using GraphPad Prism 10. The number of biological replicates is described in the corresponding figure legend. Results are presented as mean ± SEM. Normality of data distribution was assessed using the Shapiro-Wilk test. If the data followed a normal distribution, statistical comparisons were performed using unpaired or paired t-test as appropriate. If normality was not met, Mann-Whitney U tests (for unpaired comparisons) or Wilcoxon signed-rank tests (for paired comparisons) were applied. A p-value < 0.05 was considered statistically significant.

Ldlr^-/- mice were transplanted with bone marrow from either Casp8^komac or Casp8^flox/flox cohoused littermate donor mice and subjected to a western-type diet for 12 weeks. Following the dietary period, mice were sacrificed to assess the progression of atherosclerotic lesions ( Figure 1A ). Quantification of lesions in the aortas revealed no significant differences of their number and extent between two groups ( Figure 1B ). However, aortic valve plaques demonstrated a marked increase in lesion size in Casp8^komac mice as compared to control littermates. The absolute necrotic core area was also significantly larger in the Casp8^komac group ( Figure 1C ). To further evaluate whether this increased size of the necrotic core within atheroma plaque could be explained by enhanced necrotic cell death following deficiency of caspase-8, TUNEL assays were performed on the aortic valves. Casp8^komac mice exhibited an increased number of TUNEL-positive cells compared to controls ( Figure 1D ). Noteworthy, while TUNEL staining is typically used to detect apoptotic cells, it also identifies necroptotic and necrotic cells, given that DNA fragmentation occurs in these cell death pathways as well (16, 17). Collectively, these results suggest that ablation of caspase-8 promotes increased cell death, leading to greater necrotic core formation and exacerbated necrotic damage within atheroma plaques.

To explore how caspase-8 deficiency influences macrophage cell death pathways, bone marrow-derived macrophages (BMDMs) were generated from Casp8^komac and Casp8^flox/flox mice and exposed to oxLDL to simulate an atherosclerotic environment. After 18 hours of stimulation with 50 µg/mL oxLDL, flow cytometry revealed no significant differences in apoptotic cell death (Annexin V⁺/PI^-) between the two groups. However, Casp8^komac BMDMs showed a marked increase in necrotic cell populations (Annexin V⁺/PI⁺) compared to Casp8^flox/flox BMDMs ( Figure 2A ). These findings were corroborated by similar experiments using 7-ketocholesterol, a major oxysterol found in atherosclerotic plaques. To mitigate potential compensatory mechanisms arising from genetic knockout, we employed the caspase-8- inhibitor Z-IETD-FMK (10 µM, administered 2 hours before oxysterol exposure), which yielded comparable results. Despite no significant change in apoptosis, necrosis was substantially increased in BMDMs treated with both the caspase-8 inhibitor and 7-ketocholesterol compared to those exposed to 7-ketocholesterol alone ( Figure 2B ). Western blot analyses were performed to validate these observations by assessing key markers of apoptotic and necroptotic pathways. Exposure to 7-ketocholesterol activated apoptotic signaling, as indicated by increased levels of cleaved caspase-3, caspase-7, and caspase-8 in BMDMs (17). However, no significant changes in necroptotic markers were observed at earlier time points, consistent with prior reports suggesting that MLKL phosphorylation precedes necroptosis by several hours (18). Notably, combined treatment with the caspase-8 inhibitor and 7-ketocholesterol resulted in a reduction of apoptotis pathway, as evidenced by inhibited cleavage of caspase-3 and caspase-7. By contrast, the levels of phosphorylated MLKL (pMLKL) and the pMLKL/MLKL ratio were elevated, indicating enhanced necroptotic signaling ( Figures 2C, D ). By contrast, in vivo, whereas TUNEL assay revealed an increase in necrotic death in atheroma plaques from Ldlr^-/- mice transplanted with Casp8^komac or Casp8^flox/flox bone marrow ( Figure 1D ), MLKL and pMLKL staining did not show any significant difference in pMLKL/MLKL ratio ( Supplementary Figure S3 ). Thus, the early shift from apoptosis to necroptosis caused by caspase 8 inactivation within the plaque microenvironment may ultimately lead to accumulation of dead cell corpses at advanced stages of atheroma development.

Additionally, caspase-8 is known to regulate pyroptosis (19, 20). This pathway depends on caspase-1-dependent cleavage of gasdermin-D. Exposure of BMDMs to 7-ketocholesterol was associated to an increase in the level of the active cleaved form of caspase-1, yet no change was observed in gasdermin-D cleavage ( Supplementary Figure S4 ). Furthermore, pharmacological inhibition of caspase-8 in the presence of 7-ketocholesterol did not enhance caspase-1 cleavage but rather led to its reduction ( Supplementary Figure S4 ). These findings do not support the occurrence of pyroptotic cell death when caspase-8 is inhibited in macrophages exposed to 7-ketocholesterol or other components of the plaque microenvironment.

In summary, both genetic deletion and pharmacological inhibition of caspase-8 in macrophages shift the balance of cell death regulation from apoptosis towards necroptosis and accumulation of dead cell corpses under atherogenic conditions.

Since disturbances of plasma lipid and myeloid cell profiles play a central role in the onset of atherosclerosis, these parameters were investigated in Casp8^komac and Casp8^flox/flox mice fed with an atherogenic diet for 12 weeks. Mice grafted with Casp8^komac bone marrow exhibited lower levels of total cholesterol and triglycerides compared to mice transplanted with bone marrow from Casp8^flox/flox controls ( Figure 3A ). Gel exclusion chromatography further confirmed a significant reduction in cholesterol levels within the VLDL and LDL fractions ( Figure 3A ). Therefore, the atherogenic phenotype associated with Casp8^komac bone marrow engraftment ( Figures 1B, C ) cannot be explained by the plasma lipid profile. Blood cell counts were also performed on these mice. Caspase-8 ablation in myeloid cells is associated with a rise in total white blood cells, with an increased count of lymphocytes whereas the number of monocytes and granulocytes were not significantly altered ( Figure 3B ). Flow cytometry analysis of blood monocytes revealed that myeloid caspase-8 deficiency led to a reduced proportion of inflammatory Ly6C^Hi monocytes, while the percentage of patrolling Ly6C^Med/Lo monocytes was increased. The decrease in inflammatory monocytes was confirmed by the reduced proportion of CCR2^HiLy6C^Hi monocytes, alongside an unchanged percentage of non-classical CX3CR1^HiLy6C^Lo monocytes ( Figure 3C ). Overall, phenotypic analysis of blood monocytes from Casp8^komac mice compared to controls showed an increase in patrolling monocytes and a reduction in inflammatory monocytes. The lack of an inflammatory profile in mice transplanted with Casp8^komac bone marrow was further supported by plasma cytokine measurements, which revealed no elevation of inflammatory cytokines typically linked to atherogenesis, as compared to control grafted mice ( Figure 3D ).These findings further confirm that the enhanced atheroma plaque development observed in Casp8^komac bone marrow-grafted Ldlr^-/- mice is strictly due to increased macrophage cell death and related necrotic formation, rather than an aggravated inflammatory profile. Only an elevation in plasma MCP-1 levels was observed, which may contribute to the recruitment of monocytes into atheromatous lesions.