Fig 2: Rip3K51A/K51Amice exhibit more tissue injury, worse functional deficits and increased PEox following CCI. (A) More ipsilateral cortical tissue loss occurs in Rip3K51A/K51A animals vs. Rip3+/+ following CCI. Data are Mean ± SEM, n = 8–10/group, *p < 0.05. (B) Neurological function was assessed after CCI in Rip3+/+ and Rip3K51A/K51A mice. Line graph showing the MWM swim latency to reach the hidden (H) platform on days 1–7 and visible (V) platform on day 8 and 9 following CCI. Data are Mean ± SEM, n = 10/group, *p < 0.05. LC/MS-based quantitative assessment of ferroptotic cell death signals PE(38:4)-OOH (C), PE(38:5)-OOH (D) and PE(40:4)-OOH (E) in cerebral cortical tissue of naïve and CCI exposed Rip3+/+ and Rip3K51A/K51A mice. Contra and ipsi denote the opposite side and the same side of the injured cerebral hemisphere, respectively, in the CCI mice. Shown are the MS2 spectra of PE(38:4)-OOH (F), and PE(38:5)-OOH (G). Fragments formed during MS2 fragmentation that are attributed to polar head of PE with m/z 140.0 and 196.0 along with others are shown and associated with identification of hydroperoxy-PE species. Data are Mean ± SEM, n = 5/group, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig 3: RIP3K51Amutant and inhibitor GSK’872 promote IR and RSL3 induced ferroptosis. (A) Rip3K51A/K51A (red bars) primary BM cells were dose-dependently sensitive to the GPX4 inhibitor and ferroptosis-inducer, RSL3 (dose range: 0–500 nM). Mean ± SD, *p < 0.05 vs. respective genotype DMSO control (0 nM RSL3) group, n = 4 independent experiments. (B) Ferrostatin-1 (Fer-1, 1 µM), but not z-VAD-fmk, Nec-1s, or BafA1 reverse Rip3K51A/K51A BM cell death upon exposure to RSL3. Cell death measured by Annexin V-FITC + Propidium iodide (PI) flow cytometry, Mean ± SD, *p < 0.05 vs. respective genotype RSL3 group, n = 4 independent experiments. (C) GSK’872 synergistically enhances RSL3-induced ferroptosis in HT22 cells. Toxicity was only observed with the GSK’872 + RSL3 cotreatment, not individually. Synergistic injury was rescued completely by Fer-1. Cell death measured by LDH release, Mean ± SD, *p < 0.05 vs. vehicle, n = 3 independent experiments. (D) RIP3 inhibitor, GSK’872, synergistically enhances RSL3-induced ferroptotic cell death in wild type primary mouse small intestinal enteroid monolayers. Compared to RSL3 or GSK’872 monotreatment, cotreatment significantly increased enteroid death. The synergistic injury was rescued by anti-ferroptotic Lip-1. Cell death measured by LDH release at 22–24 h, Mean ± SD, *p < 0.05 vs. RSL3 only group, n = 3 independent experiments. (E) GSK’872 sensitized immortalized wild-type rat intestinal epithelial cells (IEC18) to ferroptosis. Lip-1 reversed the GSK’872 + RSL3 cotreatment effect. Cell death measured by LDH release at 18–20 h, Mean ± SD, *p < 0.05 vs. RSL3 only group, n = 3 independent experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig 4: PEBP1 associates with RIP3. (A) Structures of RAF (pdbid: 3c4c) and RIP3 (pdbid: 4m66) are highly conserved in sequence and tertiary structure. Catalytic regions shown enclosed in dashed region. (B) Lowest energy binding pose of PEBP1 (cyan) on RIP3 (grey) from docking simulations. Key interfacial interactions between PEBP1’s heterodimerization loop region (residues 127–150, including D144-H145) and RIP3’s a-helix C (aC), specifically R69 (spacefilling model) are highlighted. (C–F) Full Atomistic Molecular Dynamics simulations between PEBP1 with RAF, RIP3, RIP1, or RIP3K51A. Grey dotted line indicates the threshold maximum distance positive protein-protein interaction (=0.5 nm). Favorable interaction is predicted between (C) PEBP1 and RAF or (D) RIP3WT, but not (E) RIP1 or (F) RIP3K51A mutant. N = 3–4 independent simulations. (G–H) Colocalization between RIP3 and PEBP1 is reduced in Rip3K51A/K51A vs. Rip3+/+ primary bone marrow cells. (G) PEBP1_RIP3 colocalization normalized to cell number, Mean ± SD, *p < 0.05. (H): “Merge” [left panels]: PEBP1 (red), RIP3 (green), and hoechst (blue). “Colocalized objects” [right panels]: PEBP1_RIP3 colocalized objects (yellow), n = 3 independent experiments, scale: 20 µm. (I-J) FRET analysis showing close physical proximity (=10 nm) of RIP3 with PEBP1 in HT22 cells. (I) FRET effect was confirmed through acceptor (cy5) photobleaching (white arrows) and reciprocal Cy3 donor fluorophore unquenching. FRET ratio (donor/acceptor relative fluorescent intensity (RFU)) is pseudo-colored (range 0–10, violet-red); (J) Representative RFU vs. single excitation (Ex) wavelength. Cy3 vs. Cy5 emission (Em) wavelengths are indicated. N = 3 independent experiments. (K) Far western blotting demonstrating specific interaction of recombinant human PEBP1 with RIP3. Representative non-denaturing immunoblots showing PEBP1 (0.5 µg protein per incubation) binds to membrane-immobilized RIP3 (left) but not GST or BSA control proteins. No PEBP1 signal is detected if the blot is not incubated with recombinant PEBP1 (middle). Protein loading was verified by Ponceau S prior to incubation with PEBP1 (right). N = 3 independent experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig 5: PEBP1 regulates necroptotic death. (A) Human recombinant RIP3 kinase activity is specifically and potently inhibited by equimolar PEBP1 or the small molecule inhibitor, GSK’872. Human recombinant RIP1 kinase and another canonical serine/threonine kinase, MAPK14 (p38a), are not inhibited by equimolar PEBP1. PEBP1 alone has no effect on ATP concentration. Mean ± SD, *p < 0.05 vs. uninhibited enzyme, n = 3 independent experiments. (B) Representative immunoblots showing reduced PEBP1 expression is associated with greater pRIP3 (pS231/pT232) and pMLKL (pS345) levels 12 h following TNFa, z-VAD-fmk, and SM-164 (T/Z/S) necroptosis induction. L929 cells were transfected with PEBP1 or non-targeted (NT) siRNA for 48 h prior to T/Z/S treatment. Representative of 3 independent experiments. (C) PEBP1 siRNA knockdown L929 cells experienced a greater increase in necroptosis at higher TNFa (1, 5, 10 ng/mL) doses compared to NT siRNA cells. Cell death was specifically rescued by Nec-1s (see Fig. S5A). Z/S: z-VAD-fmk + SM164; cell death measured by LDH release at 16–20 h, Mean ± SD, *p < 0.05 vs. NT siRNA, n = 3 independent experiments. (D) PEBP1 knockdown CRISPR sensitizes L929 cells to necroptotic death. T/Z: TNFa (10 ng/mL) + z-VAD-fmk. Cell death measured by LDH release at 18 h, Mean ± SD, *p < 0.05 vs. non-targeted (NT) CRISPR, n = 3 independent experiments.