Screening-based repurposing identifies potassium dehydroandrographolide succinate (KDAS) as a selective suppressor of TNF-α-induced caspase-8-dependent apoptosis in colonic epithelial cells Authors: Alvaro Valero-Cortijo, Joan Antoni Shoenenberger-Arnaiz, Victor J. Yuste, Judit Ribas Published: 2026 ABSTRACT Under physiological conditions, tumor necrosis factor-α (TNF-α) favors cell survival through cytoprotective checkpoints, including a FIP200-dependent non-canonical autophagy pathway. Disruption of this checkpoint shifts TNF-α signaling toward apoptosis, a process relevant to epithelial injury in inflammatory bowel disease (IBD). Anti-TNF-α biologics are standard IBD therapy but primarily neutralize extracellular TNF-α without reducing epithelial susceptibility to death. Here, we conducted a two-step screen of 2421 clinically approved compounds in TNF-α-challenged FIP200-deficient cells to identify drugs that restore cytoprotection. Potassium dehydroandrographolide succinate (KDAS) emerged as a top candidate and conferred > 85% cytoprotection in FIP200-deficient HCT116 cells by inhibiting TNF-α-induced, caspase-8-dependent apoptosis without affecting necroptosis. Other andrographolides did not confer cytoprotection, highlighting KDAS specificity. KDAS also suppressed TRAIL-induced apoptosis and protected additional human cell lines and primary human colonic epithelial cells. Although KDAS enhanced early NF-κB and ERK signaling, pharmacological inhibition of these pathways did not abrogate cytoprotection, suggesting that they are dispensable for its protective effect. These findings identify KDAS as a candidate cytoprotective compound capable of preserving epithelial cell viability downstream of TNF-α receptor activation and indicate a pharmacological approach to limit epithelial apoptosis in TNF-α-mediated diseases such as IBD.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-55892-0. Introduction Tumor necrosis factor-α (TNF-α) is a master regulator of intestinal inflammation and epithelial injury in inflammatory bowel disease (IBD)1. Although TNF-α is widely perceived as a cytotoxic cytokine, its predominant cellular outcome under physiological conditions is survival, mediated by engagement of its two receptors, TNFR1 and TNFR22. Activation of TNFR1 rapidly triggers the assembly of a membrane-bound signaling platform known as complex I, leading to activation of mitogen-activated protein kinase (MAPK) and nuclear factor-κB (NF-κB) pathways, thereby promoting inflammatory gene expression and cell survival3. Nonetheless, this default pro-survival program is tightly controlled by multiple regulatory checkpoints that restrain the conversion of TNF-α signaling into cell death4. When these checkpoints fail, complex I is redirected toward cytosolic lethal assemblies, such as the pro-apoptotic complex II or the necrosome, which execute apoptotic or necroptotic cell death respectively4. Pharmacological disruption of these checkpoints using classical sensitizers (cycloheximide, actinomycin D, or Smac mimetics) experimentally recapitulates the TNF-α-driven epithelial cell death observed in vivo during chronic inflammation4. Canonical macroautophagy (hereafter autophagy) is a lysosome-dependent, multistep degradative process that proceeds through sequential steps of initiation, nucleation, elongation, degradation, and recycling and contributes to cellular resilience to inflammatory stress5. In intestinal epithelial cells, canonical autophagy plays a critical role in maintaining barrier integrity, limiting inflammatory signaling and protecting against cytokine-mediated injury6. Consistent with this protective function, genetic disruption of essential autophagy genes such as ATG5 or ATG16L in the intestinal epithelium leads to impaired epithelial homeostasis, exacerbated inflammatory responses, and increased susceptibility to experimental colitis, including enhanced TNF-α-driven epithelial apoptosis6. Autophagic activity is experimentally assessed by monitoring the autophagic flux via MAP1 light chain 3-II (LC3-II) accumulation and p62 turnover in the presence of lysosomal inhibitors. However, loss of canonical autophagy alone cannot fully explain TNF-α apoptotic hypersensitivity observed with genetic disruption of specific autophagy-related genes, including FIP200 and ATG97. This indicates that certain autophagy proteins exert protective functions beyond canonical autophagy. Recent evidence identifies a FIP200-dependent unconventional autophagy pathway as a critical checkpoint restraining TNF-α-driven cell death. Beyond scaffolding the Unc-51-like kinase 1 (ULK1) initiation complex in canonical autophagy, FIP200 mediates unconventional deployment of the autophagy machinery that selectively targets caspase-8-containing complexes for lysosomal degradation, suppressing TNF-α-induced apoptosis8. Loss of this unconventional autophagy checkpoint unleashes caspase-8 activation and sensitizes cells to TNF-α-induced apoptosis, a mechanism of direct relevance to intestinal inflammation8. Potassium dehydroandrographolide succinate (KDAS) is a semisynthetic derivative of the natural compound andrographolide, which is originally isolated from the medicinal herb Andrographis paniculata 9,10. Among injectable andrographolide derivatives, KDAS is considered to have the strongest anti-inflammatory and anti-pyretic activity9. However, KDAS has not yet been tested in IBD patients, nor have in vitro studies explored its effects on TNF-α-driven cell death. Based on FIP200’s role as a critical checkpoint limiting TNF-α-driven apoptosis, we hypothesized that compounds restoring survival in FIP200-deficient settings would inhibit TNF-α-induced apoptotic signaling. We therefore performed a phenotypic screening of approved drugs in FIP200-deficient models, using cell viability and cell death as primary endpoints, complemented by caspase activation and mitochondrial membrane permeabilization readouts. This two-tiered screen identified KDAS as the top-ranking protective agent. KDAS rescued human colonic HCT116 cells from TNF-α-induced, caspase-8-mediated apoptosis by attenuating initiator and executioner caspase activation and mitigating mitochondrial membrane permeabilization, without affecting TNF-α-driven necroptosis. Structurally related andrographolide analogs lacked protective activity, underscoring KDAS specificity. KDAS cytoprotection extended to TRAIL and persisted across ontogenically distinct human cells, including HeLa (cervical carcinoma), HT1080 (fibrosarcoma) and primary human colonocytes. Mechanistically, although KDAS enhanced NF-kB and ERK signaling, these pathways proved dispensable for cytoprotection. Collectively, these findings position KDAS as a selective inhibitor of TNF-α-driven apoptosis and support its repurposing to preserve epithelial integrity in IBD. Materials and methods Cell lines and culture Immortalized Fip200−/− and Fip200+/+ MEFs were kindly provided by Dr. Guan (Department of Cancer Biology, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA). The wt HCT116 and BAX−/− HCT116 cell lines were a generous gift from Dr Vogelstein (Howard Hughes Medical Institute, Sidney Kimmel Comprehensive Cancer Center, The Johns Hopkins School of Medicine, Baltimore, MD, USA). The HT-29 cell line was kindly provided by Dr. Arango and Dr. Martinez (IRBLleida, Lleida, Spain), HT1080 cells kindly provided by Dr. Gari (IRBLleida, Lleida, Spain), and HeLa cells were obtained from the ATCC (Manassas, VA, USA). MEFs, HeLa, HT1080 and HT-29 cells were maintained in DMEM (Gibco; Paisley, Scotland, UK) supplemented with 10% FCS (Gibco; Paisley, Scotland, UK). HCT116 cells were cultured in McCoy’s 5A medium (Cytiva; South Logan, UT, USA) containing 10% FCS. All media were supplemented with 5 μg/ml Plasmocin™ (InvivoGen; San Diego, CA, USA) to prevent mycoplasma contamination. Primary human colonic epithelial cells (P10768) were cultured in colonic epithelial cell medium (P60165), consisting of basal medium supplemented with 5% FBS, 1% colonic cell growth supplement, and penicillin/streptomycin (all from Innoprot; Bizkaia, Spain). Cells were maintained on collagen I-coated culture plates (P8188, Innoprot; Bizkaia, Spain). For experiments, human colonic epithelial cells passage 3 to 6 were seeded onto collagen I-coated plates with McCoy’s 5A medium containing 10% FCS and Plasmocin™. Cultures were incubated at 37 °C in a humidified atmosphere of 5% CO2. All cell lines were routinely tested for mycoplasma contamination (negative). Cell culture plastics were obtained from Avantor (Radnor, PA, USA) or from Nunc, Thermo Fisher Scientific (Roskilde, Denmark). Drugs “FDA-approved Drug Library Plus” (HY-L022P), Potassium dehydroandrographolide succinate (HY-N0677A), Trametinib (HY-10999), IKK16 (HY-13687A), z-IETD-fmk (HY-101297) were purchased from MedChemTronica (Sollentuna, Sweden). Birinapant (T6007), Necrosulfonamide (T6904), cisplatin (T1564) and TPCA-1 (T3049) were acquired from TargetMol (Wellesley Hills, MA, USA). Necrostatin-1 (2324) and UO126 (1144) were purchased from Tocris Biosciences (Bristol, GB). Cycloheximide (C7698) was from Sigma (Saint Louis, MI, USA). Actinomycin D (10-2054), Etoposide (10-1123), 5-Fluorouracil (10-1276) were purchased from Focus Biomolecules (Plymouth Meeting, PA, USA). q-VD-(OMe)-OPh (A8165) was from ApexBio (Houston, TX, USA). Irinotecan and carboplatin were kindly provided by the Hospital Pharmacy Service at the Hospital Universitari Arnau de Vilanova (Lleida, Spain). Human and mouse TNF-α (300-01A and 315-01A, respectively) and human TRAIL (31004) were purchased from Peprotech, part of Fisher Scientific (Rocky Hill, NJ, USA). Stocks were prepared either at 10 μg/ul or 100 μg/ul in distilled water + 0.1% BSA. Bafilomycin A1 (11038) was from Cayman Chemical (Ann Arbor, MI, USA). To control experiments, cells were treated with medium plus the highest concentration of solvent employed in the drug treatments. DMSO concentrations never surpassed a 0.1% in the medium. Cell viability and cell death determination Cell viability was assessed using the MTS Assay Kit (ab197010, Abcam; Rockford, IL, USA) following the manufacturer’s instructions. Briefly, cells were seeded in 96-well plates and exposed to the indicated drug treatments. Subsequently, MTS reagent was added to each well (10% of the final volume), and after 2 h incubation at 37ºC, absorbance was measured at 490 nm using an Infinite M Nano+ plate reader (Tecan; Männedorf, Switzerland). Cell viability was expressed as a percentage relative to vehicle-treated controls. For cell death analysis, adherent and floating cells were collected by trypsinization, incubated for 15 min at room temperature with 2.5 μg/ml propidium iodide (PI) from Sigma-Aldrich (St. Louis, MO, USA), and analyzed by flow cytometry. The percentage of PI-positive cells was quantified to determine the extent of cell death. Analysis of the screening data For the initial screening phase, Fip200−/− MEFs were seeded in 96-well plates at 3,000 cells/well. After 24 h, cells were treated with 12.5 ng/ml mouse TNF-α, with or without compounds from the “FDA-approved Drug Library Plus” (1:400 dilution; 5 or 25 μM final concentration), adding one compound per well and DMSO as vehicle control. After 48 h, MTS reagent was added and incubated for 2 h; absorbance was measured at 490 nm. Cell viability was normalized to vehicle-treated controls, and cellular rescue was calculated as follows:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} {\mathbf{Cell}}\;{\mathbf{Rescue}}\;{\mathbf{MTS}} & = ((\% \;{\mathrm{Viability}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{TNF}} + {\mathrm{DRUG}})}} - \% {\mathrm{Viability}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{TNF}})}} ){/} \\ & (\% \;{\mathrm{Viability}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{CONTROL}})}} - \% \;{\mathrm{Viability}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{TNF}})}} )) \times {1}00 \\ \end{aligned}$$\end{document} Compounds were classified as hits if they induced ≥ 25% cell rescue relative to the TNF-α condition in the MTS assay. In parallel, cell viability was also evaluated by quantifying Hoechst-stained nuclei. Fip200−/− MEFs, seeded and treated as described above, were washed with pre-warmed phosphate buffered saline (PBS) and fixed for 30 min in 4% paraformaldehyde (VWR; Radnor, PA, USA) in 0.1 M phosphate buffer (pH 7.4). After removing the fixative, cells were stained for 1 h with 1 μg/ml Hoechst 33342 (B2261, Sigma-Aldrich; St. Louis, MO, USA). Images were acquired on a Zeiss Fluorescence microscope, and nuclei were counted from 5 images per well using Fiji (ImageJ) software (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/, 1997–2018). Rescue was calculated using the following formula:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} {\mathbf{Cell}}\;{\mathbf{Rescue}}\;{\mathbf{Counting}} & = (({\mathrm{Nuclei}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{TNF}} + {\mathrm{DRUG}})}} - {\mathrm{Nuclei}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{TNF}})}} )/ \\ & \quad {\mathrm{Nuclei}}Fip200^{{ - / - }} {\mathrm{MEFs}}_{{({\mathrm{TNF}})}} ) \, \times { 1}00 \\ \end{aligned}$$\end{document} Compounds were considered candidates if they showed ≥ 20% cell rescue by nuclei counting. For follow-up (refinement) screening, FIP200−/− HCT116 cells were seeded in 96-well plates at 10,000 cells/well. After 24 h, cells were left untreated or treated for 48 h with 10 ng/ml Actinomycin D (DAct), or with a combination of 10 ng/ml human TNF-α and DAct, with or without the 208 candidate compounds identified in the initial screening. For the DAct-alone condition, cell viability was 83.81% ± 6.29 (mean ± SD, n = 24), and this value was subsequently used for rescue calculations. Each condition was tested in triplicate. To assess intrinsic toxicity, cells were also treated with each candidate compound alone; compound concentration was reduced if cytotoxicity exceeded 80%. Cell rescue was calculated as follows:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathbf{Cell}}\;{\mathbf{Rescue}}\;{\mathbf{MTS}} = \left( {{\mathrm{Drug}}\;{\mathrm{Rescue/Maximal}}\;{\mathrm{Rescue}}} \right) \, \times { 1}00$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} {\mathbf{Drug}}\;{\mathbf{Rescue}} & = (\% \;{\mathrm{Viability}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}} + {\mathrm{DRUG}})}} {/}\% \;{\mathrm{Viability}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}})}} ) \\ & \quad - (\% \;{\mathrm{Viability}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}})}} {/}\% \;{\mathrm{Viability}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{DACT}})}} ) \\ \end{aligned}$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathbf{Maximal}}\;{\mathbf{Rescue}} = \% \;{\mathrm{Viability}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{DACT}})}} - \% {\text{ Viability}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}})}}$$\end{document} Compounds conferring ≥ 25% cell rescue in the MTS assay were further evaluated for their ability to protect the membrane cell permeability by propidium iodide (PI) uptake. FIP200−/− HCT116 were seeded in 24-well plates at 60,000 cells/well. After 24 h, cultures were left untreated or treated with as described above, in the presence or absence of the 28 most protective candidates. Each condition was evaluated in triplicate. Cell survival was assessed as the proportion of PI-negative cells by flow cytometry, and rescue was calculated using the following formula:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} {\mathbf{Cell}}\;{\mathbf{Rescue}}\;{\mathbf{PI}} & = ((\% {\text{ PI}} - {\mathrm{negative}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}} + {\mathrm{DRUG}})}} - \% {\text{ PI}} - {\mathrm{negative}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}})}} )/ \\ & \quad (\% {\text{ PI}} - {\mathrm{negative}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{DACT}})}} - \% {\text{ PI}} - {\mathrm{negative}}FIP200^{{ - / - }} {\mathrm{HCT116}}_{{({\mathrm{TNF}} + {\mathrm{DACT}})}} )) \, \times { 1}00 \\ \end{aligned}$$\end{document} Compounds conferring ≥ 20% cell rescue in the PI uptake assay were considered the hits of this screening. Caspase activity Caspase-8 activity was determined using the Caspase-Glo® 8 Assay (G8201, Promega; Southampton, UK) according to the manufacturer’s protocol. Briefly, 50 μl of reconstituted Caspase-Glo® 8 reagent was added in each well of 96-multiwell microplates. After an incubation of 150 min at RT, luminescence was measured using a Tecan M200 luminometer (Tecan; Männedorf, Switzerland). Caspase-3 and caspase-7 activities were determined using the Ac-DEVD-afc substrate as described in11 with minor modifications. Briefly, 20 μg of NP-40-soluble protein extracts were incubated for 6 h at 35ºC with 10 μM Ac-DEVD-AFC (HY-P1005, MedchemTronica; Sollentuna, Sweden). The assay was performed in 96-multiwell microplates with a final volume of 100 μl/well containing 50 μl of 2X DEVD assay buffer (100 mM Tris–HCl pH 7.4, 4 mM EGTA, 10 mM DTT, 20% sucrose, 2 mM PMSF, 20 μM Ac-DEVD-AFC). The fluorescent signal was quantified by using a Fluorescence Tecan M200 fluorimeter (Tecan; Männedorf, Switzerland) under an excitation filter of 360 nm (40-nm bandwidth) and an emission filter of 530 nm (25-nm bandwidth). Results are expressed as “fold vs. control”, which refer to a normalization using the averaged arbitrary units of fluorescence of control populations. Western blot To obtain cytosolic extracts, cells were harvested and lysed in a buffer containing 50 mM NaCl, 25 mM Tris pH 6.8, 1 mM EDTA, 0.1% Triton X-100 (cytosolic lysis buffer) supplemented with the Protease Inhibitor Cocktail Set III, EDTA-free (539134, Calbiochem, part of Merck; Darmstadt, Germany). If a protein phosphorylation was examined, the phosphatase Inhibitor Cocktail II was added to the buffer. Cell lysates were centrifuged at 20,000×g for 30 min at 4 °C before the determination of the protein content. To explore the release of cytochrome c from mitochondria, we performed a digitonin-based cell subfractioning as previously described12. Briefly, cells treated as indicated in the figure legend, were lysed on ice for 5 min with a buffer containing 50 mM Tris/HCl pH 6.8, 150 mM NaCl, 1 mM EDTA, the protease Inhibitor cocktail set III and 0.033% digitonin (BN2006, Invitrogen; Waltham, MA, USA). To study ERK1/2 signaling, cells were rinsed once with ice-cold 1 × PBS (pH 7.2), and lysed directly in the cell culture plate in a buffer containing 100 mM Tris/ClH pH 6.8, 1% SDS, 1 mM EDTA plus the protease and phosphatase inhibitory cocktails afore mentioned. Then, cells were scrapped using a rubber policeman and transferred to an Eppendorf. Cell lysates were sonicated on ice for two rounds of 10 s before being clarified by centrifugation at 15,000 r.p.m. for 30 min. Protein extracts were centrifuged at 4 °C for 5 min at 16,000g and supernatants were collected (supernatant subfraction). The amount of protein was quantified by the DC protein assay (5000111, Bio-Rad; Hercules, CA, USA). Equal amounts of protein were loaded in the wells of SDS-polyacrylamide gels and separated in an electric field. Next, gels were electrotransferred onto Immobilon-P 0.45 μm PVDF membranes (IPVH00005, Millipore; Billerica, MA, USA) and blocked with Tris-buffered saline with 0.1% Tween 20 and 5% non-fat dry milk. Blocked membranes were probed with the following specific primary antibodies: anti-LC3B (1:1000; L7543) and anti-AIF (1:40,000; A7549) from Sigma-Aldrich (St. Louis, MO, USA), anti-p62 (1:2000; PM045) and anti-ATG14 (1:1000, PD026) from Medical and Biological laboratories Co (Tokyo, Japan). Anti-phospho-Akt -Thr308- (1:1000, 13,038), anti-phospho-Akt -Ser473- (1:1000, 4060), anti-caspase-3 (1:2000, 9665 or 1:1000, 9662), and anti-caspase-9 (1:1000; 9508) from Cell Signaling Technology (Danvers, MA, USA), anti-αFodrin (1:80,000, MAB1622) from Millipore (Billerica, MA, USA), anti-AKT (1:10,000, sc-1618) and anti-phospho-ERK1/2 (1:1000, sc-7383) from Santa Cruz Biotech (Dallas, TX, USA), anti-pan-ERK (1:4000, 610,123) from BD Transduction Laboratories™ (San Jose, CA, USA) and anti-cyt c (1:1000, 556,433) from BD Pharmingen™ (San Diego, CA, USA). After incubation with the corresponding horseradish peroxidase-conjugated secondary antibody, the immunoblots were developed with the Immobilon™ reagent (WBKLS0500, Millipore; Burlington, MA, USA). Chemiluminescent images were recorded by means of a Chemidoc XRS apparatus and analyzed with the Image Lab version 4.0.1 software from Bio-Rad. Loading of each sample was assessed by staining the membranes during 5 min in a solution containing 10% methanol, 2% acetic acid and 0.1% of Naphthol blue black (N3393, Sigma-Aldrich; St. Louis, MO, USA). Then, membranes were de-stained in a 10% methanol and 2% acetic acid solution during 10 min. Membranes were allowed to dry before capturing images with the Chemidoc XRS apparatus. Otherwise stated, the westerns presented are a representative of at least, three independent experiments. Immunofluorescence of RelA/p65 nuclear translocation HCT116 cells, treated as described in the legend of Fig. 7B, were rinsed with PBS at room temperature (RT) and fixed with 4% paraformaldehyde/PBS for 30 min at RT. After fixation, cells were washed twice with PBS, then permeabilized and blocked in a solution of 3% bovine serum albumin (BSA) and 0.1% Triton X-100 in PBS for 60 min at 4 °C. Cells were subsequently incubated overnight at 4 °C with anti-p65 primary antibody (D14E12, Cell Signaling Technology; Danvers, MA, USA) diluted 1:1000 in PBS containing 1% BSA. The next day, cells were rinsed twice with ice-cold PBS and incubated for 1 h at 4 °C in the dark with Alexa Fluor 488-conjugated anti-rabbit secondary antibodies (Invitrogen; Waltham, MA, USA) and Hoechst 33,342 nuclear stain. After a final wash with PBS at RT, cells were imaged using a Nikon Eclipse Ts2R fluorescence microscope. For each well, three representative images were captured and the average nuclear translocation of p65/RelA was quantified. Generation of FIP200 knockout HCT116 FIP200−/− HCT116 cells were generated using the Edit-R Gene Editing System (Horizon Discovery, Cambridge, UK), which utilizes algorithm-optimized, high-specificity synthetic guide RNAS and transient Cas9 expression for efficient genome editing. Three pre-designed synthetic sgRNAs targeting FIP200 (SQ-021117-01-0002) were used: sgRNA 1: GGACAAACTGCAGGACATAA, sgRNA 2: CTTTAGATGACTTTGCTCCA, sgRNA 3: TGGTGGAGGAGTTCTCGGTG. These sgRNAs were co-transfected with Edit-R EGFP Cas9 Nuclease mRNA (CAS11860; Forward: 5’-CACCGCAGGATCGTCACGATCAAAG-3′; Reverse: 5′-AAACCTTTGATCGTGACGATCCTGC-3′), utilizing DharmaFECT Duo Transfection Reagent (77T-2010-01, Dharmacon; Lafayette, CO, USA) according to the manufacturer guidelines. 24 h post-transfection, EGFP-positive cells were individually sorted using a Beckman Coulter Cytoflex SRT cell sorter. Single-cell clones were cultured in conditioned McCoy’s 5A medium containing 10% FCS and Plasmocin™ until confluence. Clonal cells were subsequently expanded and screened by Western blot to confirm knockout efficiency. JC-1 determination of mitochondrial membrane potential (ΔΨm) WT or FIP200−/− HCT116 were seeded at 104 cells well in black 96 well plates. After 24 h, cultures were treated as indicated in the figure legend 6A and 6B. 48 h after treatment, cells were incubated at 37ºC for 30 min with 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide (JC-1, HY-15534, MedchemTronica; Sollentuna, Sweden) at 1 μM final concentration. Cell medium was aspirated and replaced with phenol red free DMEM medium (21063029, Gibco; Paisley, Scotland, UK). Fluorescence intensities were measured using an Infinite M Nano + (Tecan; Männedorf, Switzerland). Emission wavelengths of 530 nm (green) and 590 nm (red) were measured at nine points per well in a grid pattern, and the average value of each well is calculated. The ratio between red/green was calculated and referred to the control ratio. Statistical analyses Statistical significance was determined using two-tailed paired Student’s t test performed in Microsoft Excel. Data represent means ± SD from at least three independent experiments (n ≥ 3). P-values are indicated in figure legends. Results Identification of clinically approved drugs that rescue from TNF-α-driven cell death. Knockout of the autophagy gene FIP200 sensitizes mouse embryonic fibroblasts (MEFs) to TNF-α-driven cell death13. After validating the absence of FIP200 by Western blot (Supplementary Fig. S1A), we assessed the basal autophagic flux in Fip200−/− MEFs using LC3-II and p62 (Supplementary Fig. S1B). Compared to Fip200+/+ MEFs, Fip200−/− MEFs exhibited a markedly reduced autophagic flux, as indicated by a smaller difference in LC3-II content between conditions with and without the lysosomotropic agent bafilomycin A1 (Supplementary Fig. S1B). Consistently, p62 accumulated in Fip200−/− MEFs relative to Fip200+/+ MEFs, regardless of bafilomycin A1 treatment (Supplementary Fig. S1B). Collectively, these data confirmed a severe autophagy defect in Fip200−/− MEFs. Functionally, FIP200 deficiency rendered MEFs more susceptible to cell death at TNF-α concentrations of 12.5 ng/ml or higher (Fig. 1A). This death was fully prevented by the pan-caspase inhibitor QVD-OPh (QVD), indicating that cell death was apoptotic in nature (Fig. 1B).Fig. 1Initial pharmacological screening of Fip200−/− MEFs sensitive to TNF-α-induced cell death using 2421 multi-agency-approved drugs. (A) Fip200−/− and Fip200+/+ MEFs were treated with increasing concentrations of mouse TNF-α (mTNF-α). (B) Fip200−/− MEFs were treated with mTNF-α in the presence of 25 μM q-VD-OPh (QVD). After 48 h, cell death in (A) and (B) was measured by propidium iodide (PI) uptake. Bar graphs represent the mean ± SD (n = 3). Statistical analysis was performed using Student’s t test ***P < 0.001. (C) Schematic representation of the two-step screening workflow applied to a library of 2421 multi-agency-approved compounds. The pipeline involved initial selection of hits in Fip200−/− MEFs by MTS assay (≥ 25% rescue, T1) and cell counting (≥ 20% rescue, T2), followed by refinement in FIP200−/− HCT116 cells using MTS assay (≥ 25% rescue; T3) and PI cytometry (≥ 20% rescue; T4). T1 to T4 correspond to Supplementary Tables S1 to S4. Initial pharmacological screening of Fip200−/− MEFs sensitive to TNF-α-induced cell death using 2421 multi-agency-approved drugs. (A) Fip200−/− and Fip200+/+ MEFs were treated with increasing concentrations of mouse TNF-α (mTNF-α). (B) Fip200−/− MEFs were treated with mTNF-α in the presence of 25 μM q-VD-OPh (QVD). After 48 h, cell death in (A) and (B) was measured by propidium iodide (PI) uptake. Bar graphs represent the mean ± SD (n = 3). Statistical analysis was performed using Student’s t test ***P < 0.001. (C) Schematic representation of the two-step screening workflow applied to a library of 2421 multi-agency-approved compounds. The pipeline involved initial selection of hits in Fip200−/− MEFs by MTS assay (≥ 25% rescue, T1) and cell counting (≥ 20% rescue, T2), followed by refinement in FIP200−/− HCT116 cells using MTS assay (≥ 25% rescue; T3) and PI cytometry (≥ 20% rescue; T4). T1 to T4 correspond to Supplementary Tables S1 to S4. To identify drugs that counteract TNF-α-driven cell death, we screened a library of 2,421 compounds approved by multiple agencies (FDA/EMA/worldwide) in a two-step workflow (Fig. 1C). In the initial screening, Fip200−/− MEFs were exposed to each drug (1:400 dilution from the library stock) for 48 h in the continuous presence of 12.5 ng/ml of TNF-α. Cell viability was determined using two complementary and orthogonal methods: quantification of cell number by Hoechst nuclear staining and metabolic activity measurement by MTS assay (Supplementary Fig. S1C & S1D, respectively). The use of these independent readouts minimized handling-related artifacts. Cell rescue in the Hoechst assay was expressed as the increase in nucleus counts in the drug plus TNF-α condition relative to TNF-α alone (Supplementary Fig. S1C). In the MTS assay, cell rescue was calculated as the difference in cell survival percentage between drug plus TNF-α and TNF-α alone (Supplementary Fig. S1D). Any compound affording ≥ 25% rescue in the MTS assay (102 compounds, Supplementary Table S1) or ≥ 20% increase in Hoechst assay (106 compounds, Supplementary Table S2) was considered a hit, yielding 208 candidates for the secondary screen. Notably, the identified hits included agents used to treat IBDs, such as sulfasalazine and other sulfonamides. Similarly, additional antibiotics, including rifamycin and four different fluoroquinolones, were also among the candidates. To assess the robustness of the primary screening workflow, assay quality metrics were calculated across screening plates. Z’-factor values ranged from 0.40 to 0.72 for Hoechst-based nuclei counting in Fip200−/− MEFs, indicating a good-to-excellent assay window. MTS-based measurements showed greater variability, with Z′-factor values ranging from 0.11 to 0.88, consistent with the intrinsic characteristics of metabolic viability assays in phenotypic screening approaches. Intra-plate variability remained below 10.2% for untreated conditions and below 27.8% for TNF-α-treated wells in Hoechst-based assays, and below 14.2% and 23.4%, respectively, in MTS-based measurements (Supplementary Tables S5 and S6). Refinement screening in human colonic HCT116 cells: validation and narrowing of candidates. For the follow-up screening, we established a FIP200-knockout derivative of the human colorectal-cancer cell line HCT116 using CRISPR/Cas9 technology. FIP200, a key regulator of autophagy, has been implicated in the lysosomal degradation of caspase-8, a mechanism that contributes to resistance against TNF-α-induced cell death7. As illustrated in Supplementary Fig. S2A, FIP200 was absent in FIP200−/− HCT116 cells compared to wild type (wt) cells. Functionally, loss of FIP200 completely abrogated ULK complex activity, shown by the absence of ATG14 phosphorylation at Ser29 (Supplementary Fig. S2A). In addition, FIP200 deficiency markedly attenuated basal autophagic flux, indicated by lowered LC3-II accumulation following co-treatment with bafilomycin A1, relative to LC3-II accumulation in wt (Supplementary Fig. S2B). Consistent with previous findings in Fip200−/− MEFs, we also noted a pronounced accumulation of p62 in FIP200−/− HCT116 cells (Supplementary Fig. S2B). Functionally, FIP200 abrogation provoked a reduction in cell viability of less than 20% when challenged with human TNF-α alone, therefore we increased cytotoxicity by a combination with inhibitors of RNA transcription (actinomycin D, DAct) or protein translation (cycloheximide, CHX). As shown in Supplementary Fig. S2C and S2D, FIP200−/− HCT116 cells challenged with 10 ng/ml TNF-α in combination with DAct or CHX underwent a significant reduction in cell survival, whereas wt HCT116 cells were viable under the same experimental conditions. Of note, FIP200 abrogation did not increase sensitivity to standard antitumor agents, including 5-fluorouracil, cisplatin, carboplatin, topotecan, irinotecan, or etoposide, or to the apoptotic drug staurosporine at 48 h of treatment (Supplementary Fig. S2E). Collectively, these data demonstrate that FIP200 loss selectively sensitizes HCT116 cells to TNF-α-driven cell death. Next, we refined the 208 primary hits by testing them in FIP200−/− HCT116 cells under continuous exposure to 10 ng/ml TNF-α plus DAct for 48 h. Cell rescue was first evaluated by MTS assay, calculated as the difference in cell viability between cultures treated with drug plus TNF-α plus DAct, and those treated with TNF-α plus DAct. For compounds showing a ≥ 20% cytotoxicity, we readjusted concentrations to stay below this threshold. After this filtering, 25 compounds remained as hits (≥ 25% increase in cell viability; Supplementary Table S3). Subsequently, the protective activity of these candidates was reassessed by determining a reduction in cell membrane permeability to propidium iodide (PI). This screening identified 15 compounds that conferred ≥ 20% cell rescue against TNF-α-induced cell death in FIP200−/− HCT116 cells (Supplementary Table S4). Assay quality analysis of the secondary validation workflow in FIP200−/− HCT116 cells yielded Z’-factor values ranging from 0.22 to 0.89 for MTS-based measurements and from 0.69 to 0.90 for PI uptake assays, supporting the robustness of the screening strategy. Intra-plate variability remained below 15% for untreated cells, below 14.7% for DAct-treated cells and below 19.9% for DAct + TNF-α-treated conditions in MTS-based assays, whereas PI uptake assays showed variability below 1.7 (untreated) and 6.8% (DAct + TNF-α) (Supplementary Tables S5 and S6). KDAS reproducibly exerted cytoprotective activity across independent experiments, with a mean rescue effect of 85.8 ± 14.3% and a coefficient of variation of 16.7% (Supplementary Table S7). KDAS, but not other andrographolides, selectively prevents TNF-α-driven cell death. Potassium dehydroandrographolide succinate (5 μM KDAS), also known as Kalii Dehydrographolidi Succinas, was identified as the compound with the strongest cytoprotective activity. KDAS, an anti-infective agent with known immunomodulatory and anti-inflammatory properties,14 has not been investigated as a potential candidate to manage IBD. As illustrated in Fig. 2A, concentration–response viability assays with KDAS (0.05–5 μM KDAS) confirmed its strong cytoprotective activity against DAct plus TNF-α-induced cell death, with effective concentrations starting at 0.05 μM. To further substantiate the protective capacity of KDAS, we employed the Smac mimetic birinapant (Bir) -known to promote cIAPs degradation via autoubiquitination- as alternative sensitizer15. As shown in Fig. 2B, KDAS efficiently rescued wt HCT116 cells challenged with TNF-α plus Bir within the same concentration range effective in DAct plus TNF-α conditions. Indeed, KDAS (5 μM) significantly reduced PI positivity in wt HCT116 cells treated with Bir plus TNF-α (from 59.1 ± 6.7% to 15.7 ± 5.9%), as well as in FIP200−/− HCT116 cells exposed to either DAct plus TNF-α (from 48.8 ± 3.2 to 14.9 ± 4.4%) or CHX plus TNF-α (from 59.9 ± 6.7% to 22 ± 4.2%) (Fig. 2C). For subsequent experiments, we selected the Bir and DAct setups -employed during the initial screening phase- to be applied indistinctly throughout the study. Given that KDAS is structurally related to other andrographolides, we next compared its protective activity with andrographolide and dehydroandrographolide, both of which scored negative in the initial Fip200−/− MEFs screen. As illustrated in Fig. 2D, only KDAS exhibited protective properties against TNF-α-driven cell death. In sum, our data indicated that KDAS acted as an effective cytoprotective agent against TNF-α-driven cell death irrespective of the employed sensitizer. The structural specificity of KDAS’s action was absent in related andrographolides, highlighting its unique potential for therapeutic repurposing in the context of TNF-α-induced cytotoxicity.Fig. 2KDAS, but not other andrographolides, rescues TNF-α-driven cell death irrespective of the sensitizing agent. (A) FIP200−/− HCT116 cells treated for 48 h with DAct alone or DAct plus TNF-α in the presence of increasing concentrations of KDAS. (B) wt HCT116 cells treated for 48 h with Bir alone or Bir plus TNF-α, in the presence of increasing concentrations of KDAS. Cell viability was determined by MTS. Bar value is the mean ± SD (n = 3). (C) wt and FIP200−/− HCT116 cells were treated for 48 h with 10 ng/ml TNF-α and Actinomycin D (DAct, 8 ng/ml), Birinapant (Bir, 60 nM) or cycloheximide (CHX, 1 μg/ml) in the presence of 5 μM KDAS. (D) FIP200−/− HCT116 cells were co-incubated for 48 h with 5 μM KDAS or 5 μM dehydroandrographolide (DA) or 5 μM andrographolide (A) in the presence of 10 ng/ml TNF-α plus Actinomycin D (DAct, 8 ng/ml). Cell death was measured by PI uptake. Bar value is the mean ± SD (n = 3). Student’s t test **P < 0.005 and ***P < 0.001. KDAS, but not other andrographolides, rescues TNF-α-driven cell death irrespective of the sensitizing agent. (A) FIP200−/− HCT116 cells treated for 48 h with DAct alone or DAct plus TNF-α in the presence of increasing concentrations of KDAS. (B) wt HCT116 cells treated for 48 h with Bir alone or Bir plus TNF-α, in the presence of increasing concentrations of KDAS. Cell viability was determined by MTS. Bar value is the mean ± SD (n = 3). (C) wt and FIP200−/− HCT116 cells were treated for 48 h with 10 ng/ml TNF-α and Actinomycin D (DAct, 8 ng/ml), Birinapant (Bir, 60 nM) or cycloheximide (CHX, 1 μg/ml) in the presence of 5 μM KDAS. (D) FIP200−/− HCT116 cells were co-incubated for 48 h with 5 μM KDAS or 5 μM dehydroandrographolide (DA) or 5 μM andrographolide (A) in the presence of 10 ng/ml TNF-α plus Actinomycin D (DAct, 8 ng/ml). Cell death was measured by PI uptake. Bar value is the mean ± SD (n = 3). Student’s t test **P < 0.005 and ***P < 0.001. KDAS alters TNF-α-driven NF-ΚB and MAPK signaling pathways. Activation of TNFR1 by TNF-α promotes the assembly of complex I, thereby activating both NF-κB and MAPK pathways, leading to inflammatory and cell survival responses16. Andrographolide and its derivatives have been previously reported as modulators of these pathways17. To determine the effects of KDAS on canonical NF-κB signaling, nuclear translocation of p65/RelA was explored by immunostaining (Supplementary Fig. S3). As shown in Fig. 3A, TNF-α alone induced a gradual accumulation of nuclear p65/RelA, peaking at 30 min. KDAS accelerated this translocation, with 69.9 ± 4.8% of cells exhibiting nuclear staining at 15 min compared to only 22.8 ± 10.0% with TNF-α alone (Fig. 3A). The subsequent decline in nuclear p65/RelA followed a similar kinetic under both conditions (Fig. 3A). These findings demonstrate that at early times KDAS modulates NF-κB signaling kinetics by expediting the nuclear import of p65/RelA in response to TNF-α. To further clarify the pathway’s function in KDAS-mediated cytoprotection, we blocked NF-κB activity with TPCA-1 or IKK16, two potent inhibitors of IκB kinase (IKK) complex18,19. Both drugs effectively repressed TNF-α-driven nuclear localization of p65/RelA in the presence of KDAS in both wt and FIP200−/− HCT116 cells (Fig. 3B). When we compared the effects of TNF-α combined with TPCA-1 or IKK16 on cell viability, we observed that FIP200−/− HCT116 cells exhibited sensitivity after 48 h of treatment, while wt HCT116 cells required treatment for 120 h (data not shown). Consequently, further experiments were performed using FIP200−/− HCT116 cells. As shown in Fig. 3C, pharmacological inhibition of NF-κB pathway in these cells with TPCA-1 or IKK16 significantly increased cell death rates, from 7.5 ± 1.1% in controls to 20.9 ± 2.1% and 17.9 ± 1.4%, respectively, highlighting a pro-survival role for this pathway. Notably, KDAS reduced cell death induced by TNF-α plus TPCA-1 (from 48.5 ± 1.6% to 11.5 ± 6.2%) or TNF-α plus IKK16 (from 62.1 ± 3.4% to 6.2 ± 0.8%). These results ruled out the NF-κB pathway as the main driver of KDAS-mediated cytoprotection. Next, we examined the effects of KDAS on the activation of extracellular signal-regulated kinases (ERK1/2, p42/p44 MAPKs) via Western blot, as these kinases are also pivotal for TNF-α-driven cell survival. As shown in Fig. 4A, TNF-α alone induced a strong and progressive phosphorylation of both ERK1 and ERK2, reaching a maximal peak at 30 min. The presence of KDAS further enhanced phosphorylation of both kinases (Fig. 4A). To test the functional relevance of ERK1/2 activation in KDAS-mediated cytoprotection, wt HCT116 cells were challenged with Bir plus TNF-α in the presence of UO126 or Trametinib, two pharmacological inhibitors of the ERK pathway20,21. As shown in Fig. 4B, both inhibitors abrogated TNF-α-driven ERK1/2 phosphorylation in the presence of KDAS. Functionally, their co-incubation potentiated cell death in response to Bir plus TNF-α (Fig. 4C), yet KDAS retained significant cytoprotective activity even when ERK1/2 was inhibited (Fig. 4C). These findings indicated that KDAS modified TNF-α signaling by increasing ERK1/2 phosphorylation. Nonetheless, its cytoprotective effect persisted even when ERK1/2 activity was blocked, suggesting that the mechanism for KDAS-mediated cytoprotection also operated independently of ERK1/2.Fig. 3KDAS-mediated cytoprotection to TNF-α-induced cell death is independent of NF-κB activation. (A) Time-course of p65/RelA nuclear localization in wt HCT116 cells treated with 10 ng/ml TNF-α, either without (none) or with KDAS (5 μM). (B) Effects of TPCA-1 (20 μM) and IKK16 (1 μM) inhibiting TNF-α-driven p65/RelA nuclear localization in the presence of KDAS after 15 min of treatment. Results for (A) and (B) represent the mean ± SD of quantification from 3 independent wells. (C) FIP200−/− HCT116 cells were preincubated with vehicle (None), or 20 μM TPCA-1, or 1 μM IKK16, then treated for 48 h with TNF-α (10 ng/ml) with or without KDAS (5 μM). Cell death was quantified by PI uptake. Bars represent mean ± SD (n = 3). Student’s t test. *P < 0.05 and ***P < 0.001.Fig. 4KDAS-mediated cytoprotection to TNF-α-induced cell death is independent of ERK1/2 activation. (A) Western blot analysis of whole-cell lysates from wt HCT116 cells treated with 10 ng/ml TNF-α, in the absence or presence of KDAS (5 μM), at various time points. Blots were probed for p-ERK and total ERK. Intensities of phosphorylated ERK2 (p-ERK2) and ERK1 (p-ERK1) were normalized to total ERK. Representative images from three independent experiments are shown. (B) Western blot of whole-cell lysates from wt HCT116 left untreated (C), or treated for 30 min with TNF-α (10 ng/ml, “T”), or with TNF-α plus KDAS (T + K), in the absence or presence of 25 μM Trametinib (Tram) or 20 μM UO126 (UO). Blots were probed for p-ERK and total ERK; NB served as loading control. Image is representative of two independent experiments. (C) wt HCT116 cells were preincubated with vehicle, or 25 μM Trametinib, or 20 μM UO126, then treated for 48 h with KDAS (5 μM) in the presence or absence of TNF-α (10 ng/ml) plus 60 nM Birinapant (Bir). Cell death was quantified by PI uptake. Bars represent mean ± SD (n = 3). Student’s t test ***P < 0.001. KDAS-mediated cytoprotection to TNF-α-induced cell death is independent of NF-κB activation. (A) Time-course of p65/RelA nuclear localization in wt HCT116 cells treated with 10 ng/ml TNF-α, either without (none) or with KDAS (5 μM). (B) Effects of TPCA-1 (20 μM) and IKK16 (1 μM) inhibiting TNF-α-driven p65/RelA nuclear localization in the presence of KDAS after 15 min of treatment. Results for (A) and (B) represent the mean ± SD of quantification from 3 independent wells. (C) FIP200−/− HCT116 cells were preincubated with vehicle (None), or 20 μM TPCA-1, or 1 μM IKK16, then treated for 48 h with TNF-α (10 ng/ml) with or without KDAS (5 μM). Cell death was quantified by PI uptake. Bars represent mean ± SD (n = 3). Student’s t test. *P < 0.05 and ***P < 0.001. KDAS-mediated cytoprotection to TNF-α-induced cell death is independent of ERK1/2 activation. (A) Western blot analysis of whole-cell lysates from wt HCT116 cells treated with 10 ng/ml TNF-α, in the absence or presence of KDAS (5 μM), at various time points. Blots were probed for p-ERK and total ERK. Intensities of phosphorylated ERK2 (p-ERK2) and ERK1 (p-ERK1) were normalized to total ERK. Representative images from three independent experiments are shown. (B) Western blot of whole-cell lysates from wt HCT116 left untreated (C), or treated for 30 min with TNF-α (10 ng/ml, “T”), or with TNF-α plus KDAS (T + K), in the absence or presence of 25 μM Trametinib (Tram) or 20 μM UO126 (UO). Blots were probed for p-ERK and total ERK; NB served as loading control. Image is representative of two independent experiments. (C) wt HCT116 cells were preincubated with vehicle, or 25 μM Trametinib, or 20 μM UO126, then treated for 48 h with KDAS (5 μM) in the presence or absence of TNF-α (10 ng/ml) plus 60 nM Birinapant (Bir). Cell death was quantified by PI uptake. Bars represent mean ± SD (n = 3). Student’s t test ***P < 0.001. KDAS attenuates TNF-α-induced caspase-8-driven apoptotic cell death. We examined the cell death subroutine induced by DAct plus TNF-α in FIP200−/− HCT116 and Bir plus TNF-α in wt HCT116. As shown in Fig. 5A, B, the pan-caspase inhibitor QVD fully prevented TNF-α-driven cell death under both sensitizers, confirming the apoptotic nature of cell death. Similarly, selective inhibition of caspase-8 with z-IETD-fmk led to a comparable rescue, supporting that TNF-α triggered caspase-8-dependent apoptotic cell death. In contrast, inhibition of RIP1 kinase activity (necrostatin-1, Nec-1) or MLKL (necrosulfonamide, NSA) did not confer equivalent cytoprotection regardless of the sensitizer employed. Notably, Nec-1 partially reduced apoptotic cell death when Bir was used as a sensitizer (Fig. 5B), indicating a detrimental role for RIP1 kinase activity in this specific setting. To confirm the protective role of KDAS in TNF-α-mediated apoptosis, we next examined the activation of initiator caspase-8 and the effector caspase-3/7. In wt HCT116 cells, treatment with Bir plus TNF-α increased caspase-8 activity by 2.7 ± 0.0-fold, which was reduced to 1.3 ± 0.0-fold by KDAS (Fig. 5C). A similar effect was observed in FIP200−/− HCT116 cells with DAct plus TNF-α (from 1.9 ± 0.0-fold to 1.3 ± 0.0-fold with KDAS, Fig. 5C). Given the stronger caspase-8 activation in Bir plus TNF-α-treated wt HCT116 cells, this condition was selected to further confirm the impact of KDAS on caspase-8 activation by Western blot. As shown in Fig. 5D, KDAS markedly reduced the processing of full-length caspase-8 into its active p18 fragment. We then examined the effect of KDAS on the activation of executioner caspase-3/7 using the fluorogenic substrate Ac-DEVD-afc. As shown in Fig. 5E, KDAS markedly reduced caspase-3/7 activity, decreasing Bir plus TNF-α-induced activation in wt HCT116 cells from 26.6 ± 0.7-fold to 8.8 ± 2.2-fold, and DAct plus TNF-α- induced activation in FIP200−/− HCT116 cells from 11.6 ± 1.7-fold to 4.9 ± 0.3-fold. In parallel, we followed processing of caspase-3 into its p19 and p17 fragments, cleavage of α-fodrin into the caspase-3-specific p120 fragment and cleavage of p23 co-chaperone into the caspase-7-specific p15 fragment. Indeed, KDAS led to a significant decrease in all these processing fragments, in Bir plus TNF-α-treated wt HCT116 cells. Altogether, these findings proved that KDAS attenuated TNF-α-driven apoptosis by interfering with both initiator caspase-8 activation and downstream caspase-3/7 executioner activity.Fig. 5KDAS Inhibits TNF-α-driven apoptotic cell death. (A) wt HCT116 cells were either untreated (control) or treated with 10 ng/ml TNF-α plus 60 nM Bir in the presence of 25 μM q-VD-OPh (QVD), 40 μM z-IETD-fmk (IETD), 25 μM necrostatin-1 (Nec-1), or 20 μM necrosulfonamide (NSA). (B) FIP200−/− HCT116 cells were untreated (control, white bars) or treated with 10 ng/ml TNF-α plus 8 ng/ml DAct (black bars) together with QVD, IETD, Nec-1, or NSA at the same concentrations. (A,B) After 48 h, cell death was measured by PI uptake. Bars represent mean ± SD (n = 3). (C) wt HCT116 cells were incubated in the absence or presence of 5 μM KDAS, left untreated (“Co”), or treated for 18 h with 60 nM Bir (“Bi”), or 10 ng/ml TNF-α plus Bir (“Bi + T”). FIP200−/− HCT116 cells in the absence or presence of 5 μM KDAS were left untreated (“Co”), or treated for 24 h with 8 ng/ml DAct (“DA”), or 10 ng/ml TNF-α plus DAct (“DA + T”). Caspase 8 (Cas-8) activity was determined using the Caspase-Glo® 8 Assay and normalized to the luminescent values from control cultures. Results are expressed as fold change over controls. Bars represent mean ± SD (n = 3). (D) Cell lysates from wt HCT116 cells treated for 18 h with the indicated drugs were immunoblotted for caspase-8. Blot is representative of three independent experiments. (E) Caspase-3/7 (DEVDase) activity was determined in cell lysates after 18 h (wt) and 24 h (FIP200−/−) treatments, and fluorescent signals normalized to untreated controls. Treated conditions matched those in (C). Bars are mean ± SD (n = 3). Student’s t test, ***P < 0.001. (F) Cell lysates from wt HCT116 cells treated for 18 h as indicated were immunoblotted for caspase-3, α-Fodrin, and p23. Blots are representative of three independent experiments. Student’s t test, ***P < 0.001. KDAS Inhibits TNF-α-driven apoptotic cell death. (A) wt HCT116 cells were either untreated (control) or treated with 10 ng/ml TNF-α plus 60 nM Bir in the presence of 25 μM q-VD-OPh (QVD), 40 μM z-IETD-fmk (IETD), 25 μM necrostatin-1 (Nec-1), or 20 μM necrosulfonamide (NSA). (B) FIP200−/− HCT116 cells were untreated (control, white bars) or treated with 10 ng/ml TNF-α plus 8 ng/ml DAct (black bars) together with QVD, IETD, Nec-1, or NSA at the same concentrations. (A,B) After 48 h, cell death was measured by PI uptake. Bars represent mean ± SD (n = 3). (C) wt HCT116 cells were incubated in the absence or presence of 5 μM KDAS, left untreated (“Co”), or treated for 18 h with 60 nM Bir (“Bi”), or 10 ng/ml TNF-α plus Bir (“Bi + T”). FIP200−/− HCT116 cells in the absence or presence of 5 μM KDAS were left untreated (“Co”), or treated for 24 h with 8 ng/ml DAct (“DA”), or 10 ng/ml TNF-α plus DAct (“DA + T”). Caspase 8 (Cas-8) activity was determined using the Caspase-Glo® 8 Assay and normalized to the luminescent values from control cultures. Results are expressed as fold change over controls. Bars represent mean ± SD (n = 3). (D) Cell lysates from wt HCT116 cells treated for 18 h with the indicated drugs were immunoblotted for caspase-8. Blot is representative of three independent experiments. (E) Caspase-3/7 (DEVDase) activity was determined in cell lysates after 18 h (wt) and 24 h (FIP200−/−) treatments, and fluorescent signals normalized to untreated controls. Treated conditions matched those in (C). Bars are mean ± SD (n = 3). Student’s t test, ***P < 0.001. (F) Cell lysates from wt HCT116 cells treated for 18 h as indicated were immunoblotted for caspase-3, α-Fodrin, and p23. Blots are representative of three independent experiments. Student’s t test, ***P < 0.001. KDAS preserves mitochondrial function and prevents pro-apoptotic factors release in response to TNF-α Mitochondrial permeabilization can trigger inflammation and alter the extent of both extrinsic and intrinsic apoptotic signaling cascades22. In this sense, BAX is considered a pivotal gatekeeper for mitochondrial outer membrane permeabilization23. BAX deficiency was confirmed by Western blot in BAX−/− HCT116 cells (Supplementary Fig. S4). BAX−/− HCT116 cells exhibited delayed kinetics of cell death in response to staurosporine or TNF-α, either in the presence of Bir or CHX (Supplementary Fig. S4). These findings supported the view that mitochondrial permeabilization contributed to TNF-α-driven cell death in HCT116 cells. To evaluate the protective effects of KDAS on mitochondrial function, we monitored the mitochondrial membrane potential (ΔΨm) using JC-1 staining. In wt HCT116 cells, Bir and Bir plus TNF-α markedly reduced JC-1 staining to 0.60 ± 0.03 and 0.44 ± 0.01-fold, respectively, whereas KDAS reversed this decrease to 0.82 ± 0.03 and 0.77 ± 0.09-fold, respectively (Fig. 6A). Similarly, KDAS counteracted the DAct plus TNF-α-mediated reduction in JC-1 staining, restoring values from 0.68 ± 0.01 to 0.90 ± 0.00-fold (Fig. 6B). Because mitochondrial depolarization is often paralleled by outer membrane permeabilization, we next examined the cytosolic cytochrome c (cyt c) and AIF release. Western blot analysis revealed that KDAS prevented the early cytosolic accumulation of cyt c and AIF following Bir plus TNF-α (18 h) or DAct plus TNF-α (24 h), and this protective effect persisted at 48 h of treatment (Fig. 6C and 6D, respectively). Together, these findings indicated that KDAS preserved mitochondrial integrity by maintaining ΔΨm and preventing the release of pro-apoptotic mitochondrial factors during TNF-α-induced extrinsic apoptosis.Fig. 6KDAS preserves ΔΨm and prevents cytochrome c and AIF release during TNF-α-induced apoptosis. (A) wt HCT116 cells untreated or treated with 5 μM KDAS were challenged for 48 h with 10 ng/ml TNF-α, 60 nM Birinapant (Bir), or 60 nM Bir plus 10 ng/ml TNF-α. (B) FIP200−/− HCT116 cells untreated or treated with 5 μM KDAS were challenged for 48 h with 10 ng/ml TNF-α, 10 ng/ml Actinomycin D (DAct), or 10 ng/ml DAct plus 10 ng/ml TNF-α. (A, B) After staining cells with JC-1, ΔΨm was determined by the ratio between red (high ΔΨm) and green (low ΔΨm) signals and normalized to the values obtained in controls. Results are the average of 3 independent experiments with at least 3 independent wells per experiment. Bars represent mean ± SD (n = 3). Student’s t test, **P < 0.005 and ***P < 0.001. (C) Mitochondrial lysates from wt HCT116 cells treated for 18 h and 48 h, were immunoblotted against cytochrome c and AIF. (D) Mitochondrial lysates from FIP200−/− HCT116 cells treated for 24 h and 48 h, were immunoblotted against cytochrome c and AIF. (C, D) Blots are representative of three independent experiments. KDAS preserves ΔΨm and prevents cytochrome c and AIF release during TNF-α-induced apoptosis. (A) wt HCT116 cells untreated or treated with 5 μM KDAS were challenged for 48 h with 10 ng/ml TNF-α, 60 nM Birinapant (Bir), or 60 nM Bir plus 10 ng/ml TNF-α. (B) FIP200−/− HCT116 cells untreated or treated with 5 μM KDAS were challenged for 48 h with 10 ng/ml TNF-α, 10 ng/ml Actinomycin D (DAct), or 10 ng/ml DAct plus 10 ng/ml TNF-α. (A, B) After staining cells with JC-1, ΔΨm was determined by the ratio between red (high ΔΨm) and green (low ΔΨm) signals and normalized to the values obtained in controls. Results are the average of 3 independent experiments with at least 3 independent wells per experiment. Bars represent mean ± SD (n = 3). Student’s t test, **P < 0.005 and ***P < 0.001. (C) Mitochondrial lysates from wt HCT116 cells treated for 18 h and 48 h, were immunoblotted against cytochrome c and AIF. (D) Mitochondrial lysates from FIP200−/− HCT116 cells treated for 24 h and 48 h, were immunoblotted against cytochrome c and AIF. (C, D) Blots are representative of three independent experiments. KDAS suppresses TRAIL-induced apoptosis but fails to prevent TNF-α-driven necroptosis TNF-related apoptosis-induced ligand (TRAIL) is a member of the TNF superfamily that triggers extrinsic apoptosis through Death Receptor 4 (DR4) and DR524. As shown in Supplementary Fig. S5A, treatment of FIP200−/− HCT116 cells with growing concentrations of TRAIL (10 to 100 ng/ml) led to a progressive reduction of cell viability in the presence of DAct, an effect that was reversed by co-treatment with KDAS. Using a fixed concentration of TRAIL (50 ng/ml), we next examined cell death by PI staining with either DAct (Supplementary Fig. S5B) or Bir (Supplementary Fig. S5C) as sensitizers. In both experimental settings, KDAS reduced TRAIL-induced cell death. In contrast, KDAS failed to reduce apoptotic cell death elicited after challenging wt HCT116 cells with several anti-tumor agents for 72 h (Supplementary Fig. S5D). TNF-α-induced cell death proceeds primarily through apoptosis, but in conditions where caspase-8 is inhibited, the pathway switches to necroptosis25. The HT-29 colorectal adenocarcinoma cell line is a classical model for necroptosis in response to TNF-α combined with a Smac mimetic and a pan-caspase inhibitor25. Consistent with this, co-treatment with QVD and Bir plus TNF-α induced 95 ± 0.5% cell death in HT29 cells (Supplementary Fig. S5E). Importantly, this cytotoxicity was reversed by Nec-1 or NSA (Supplementary Fig. S5E), confirming that cell death occurred through necroptosis. As shown in Supplementary Fig. S5F, KDAS was unable to reverse necroptotic cell death of HT29 cells in response to Bir plus TNF-α. Collectively, these data demonstrated that KDAS selectively repressed death-ligand-induced, caspase-8-depedent apoptosis but did not prevent TNF-α-driven necroptosis. KDAS halts TNF-α-driven cell death in ontogenically unrelated cell lines and human colonic epithelial cells To explore whether KDAS exhibited protective properties beyond MEFs or HCT116 cells, we extended our analysis to two additional ontogenically unrelated human cell models: HeLa (cervical carcinoma) and HT1080 (fibrosarcoma). Both cell lines were intrinsically resistant to TNF-α-driven cell death and required sensitizing agents to trigger cytotoxicity (Fig. 7A, B, respectively). Inhibition of IKK complex disrupts two checkpoints in the TNF-α signaling cascade: the IKK checkpoint and the downstream NF-κB-mediated transcriptional checkpoint, thereby sensitizing cells to TNF-α-driven cell death. Alongside conventional sensitizers, such as CHX, DAct, or Bir, we also tested pharmacological IKK inhibitors including IKK16 and TPCA-1. As shown in Fig. 7A, KDAS significantly diminished cell death in HeLa cells exposed to TNF-α plus DAct from 58.7% ± 2.7 down to 19.6% ± 2.1. Similarly, when TNF-α was combined with IKK16, KDAS lowered cell death from 24% ± 0.9 to 11.2% ± 2.3. In the HT1080 fibrosarcoma model, KDAS reduced cell death in response to TNF-α plus DAct from 64.5% ± 3.2 to 37.4% ± 4.3, TNF-α plus IKK16 from 43.6% ± 1 to 26% ± 4.4, and TNF-α plus TPCA-1 from 55.2% ± 4.1 to 30.5% ± 4.5 (Fig. 7B). To further assess the cytoprotective potential of KDAS, we examined its effects in primary human colonocytes. Colonocytes exhibited high refractoriness to TNF-α-induced cell death, remaining viable even at concentrations up to 200 ng/ml of TNF-α (Supplementary Fig. S6A). Among CHX, DAct, Bir, and IKK16, only IKK16 significantly reduced cell viability by TNF-α, with the effect plateauing at 100 ng/ml of TNF-α (Supplementary Fig. S6A). When 100 ng/ml TNF-α was combined with either IKK16 or TPCA-1, cell viability diminished to 51% ± 1.7 and 54.4% ± 1.8, respectively (Supplementary Fig. S6B). Addition of KDAS restored cell viability to 81.5% ± 3.0 and 80.1% ± 5.6, respectively (Supplementary Fig. S6B). Moreover, as shown in Fig. 7C, PI uptake evidenced that after 48 h, KDAS reduced the percentage of PI-positive colonocytes from 30.0% ± 3.4 to 9.0% ± 1.1 in the TNF-α plus IKK16 condition, and from 34.8% ± 3.0 to 11.1% ± 0.1 in the TNF-α plus TPCA-1 condition. Collectively, these results demonstrated that KDAS provided robust cytoprotection against TNF-α-driven apoptotic cell death across diverse human cell types, including primary human colonocytes. This broad cytoprotective activity highlighted KDAS as a promising candidate for further investigation in TNF-α-mediated inflammatory diseases.Fig. 7KDAS confers cytoprotection against TNF-α-driven cell death in human cell lines of different ontogenies and human colonic cells. (A) HeLa cells were treated with TNF-α (10 ng/ml) in the absence or presence of 5 μM KDAS, either alone (control) or combined with Actinomycin D (10 ng/ml, DAct), or IKK16 (500 nM). (B) HT1080 cells were treated with TNF-α (100 ng/ml) in the absence or presence of 5 μM KDAS, either alone (control) or combined with Actinomycin D (10 ng/ml, DAct), IKK16 (1 μM), or TPCA-1 (10 μM). (C) Primary human colonic cells preincubated with or without 5 μM KDAS, were treated for 48 h with IKK16 (1 μM) or TPCA-1 (5 μM) in the presence or absence of TNF-α (100 ng/ml). Cell death was quantified by PI uptake and expressed as the mean ± SD (n = 3). Student’s t test, ***P < 0.001. KDAS confers cytoprotection against TNF-α-driven cell death in human cell lines of different ontogenies and human colonic cells. (A) HeLa cells were treated with TNF-α (10 ng/ml) in the absence or presence of 5 μM KDAS, either alone (control) or combined with Actinomycin D (10 ng/ml, DAct), or IKK16 (500 nM). (B) HT1080 cells were treated with TNF-α (100 ng/ml) in the absence or presence of 5 μM KDAS, either alone (control) or combined with Actinomycin D (10 ng/ml, DAct), IKK16 (1 μM), or TPCA-1 (10 μM). (C) Primary human colonic cells preincubated with or without 5 μM KDAS, were treated for 48 h with IKK16 (1 μM) or TPCA-1 (5 μM) in the presence or absence of TNF-α (100 ng/ml). Cell death was quantified by PI uptake and expressed as the mean ± SD (n = 3). Student’s t test, ***P < 0.001. Discussion IBD, encompassing Crohn’s disease and ulcerative colitis, is characterized by chronic intestinal inflammation in which excessive epithelial apoptosis contributes to barrier disruption and disease progression26. Autophagy is essential for intestinal homeostasis, and genetic or functional impairment of this pathway predisposes to chronic inflammation and epithelial dysfunction6. FIP200, a core component of the ULK complex or autophagy-initiating complex, is required for canonical autophagosome formation and maintenance of basal autophagic flux27. Beyond its canonical role, FIP200 also participates in a non-canonical regulatory checkpoint that limits TNF-α-induced caspase-8 activation7. TNF-α exerts pleiotropic effects in this context by promoting either survival or cell death, depending on the integrity of intracellular regulatory checkpoints that restrict the formation of lethal assemblies downstream of TNF receptor activation4,8. Loss of the FIP200-dependent unconventional autophagy checkpoint sensitizes cells to TNF-α-driven apoptosis, providing a mechanistically grounded model for investigating drugs capable of restoring epithelial resilience. To exploit this vulnerability, we implemented a two-tier screening strategy using TNF-α-sensitive cell lines rendered autophagy-deficient through FIP200 ablation. Notably, the FIP200−/− HCT116 cell line was generated in our laboratory and, to our knowledge, has not previously been characterized for its response to TNF-α, thereby providing a novel platform to interrogate death-survival decisions in intestinal epithelial cells. Using FIP200-deficient murine MEFs and human HCT116 cells, we screened clinically approved agents for protection against TNF-α-induced apoptosis. FIP200 ablation reduced basal autophagic flux and selectively increased susceptibility to TNF-α-driven apoptosis without enhancing sensitivity to classical chemotherapeutic agents. Using this system, we identified several compounds capable of rescuing TNF-α-induced cytotoxicity, including potassium dehydroandrographolide succinate (KDAS), D-alpha-Tocopherol (vitamin E), apremilast, dehydrocholic acid, danofloxacin mesylate, and cromolyn sodium. The protective activity of D-alpha-Tocopherol and apremilast is consistent with their reported anti-IBD properties in animal studies28,29 or clinical assays30, thereby validating the screening platform. Among all candidates, KDAS displayed the greatest protective profile. KDAS is the potassium salt form of dehydroandrographolide succinate, a semi-synthetic derivative of andrographolide, the principal diterpene lactone from Andrographis paniculata. It is clinically marketed in China as Chuanhuning and is used parenterally for the treatment of upper respiratory tract infections, pediatric pneumonia, and bacillary dysentery31–33. KDAS exhibits improved aqueous solubility and bioavailability compared with andrographolide34, resulting in rapid systemic exposure and a short elimination half-life14. Pharmacovigilance data indicate a generally favorable safety profile, with predominantly mild adverse reactions and rare severe hypersensitivity events35–37. Despite its clinical use, mechanistic studies on KDAS remain scarce. Its antipyretic and anti-inflammatory properties in patients are largely inferred from “in vitro” studies reporting antiviral and antibacterial activities38, rather than from detailed investigations of intracellular signaling pathways. Moreover, the biological effects of KDAS are frequently assumed to mirror those of other andrographolides. However, emerging evidence indicates that structural modifications introduced during semi-synthesis confer distinct functional properties39. In agreement with this idea, neither andrographolide nor dehydroandrographolide reverted TNF-α-driven cell death in our experimental models, underscoring a distinct structure–activity relationship. Mechanistically, KDAS was identified in FIP200-null cells characterized by retaining intact NF-κB and MAPK signaling, allowing evaluation of cell death without interfering with these cellular routes. Notably, these pathways are responsive to the treatment with andrographolide derivatives17. KDAS accelerated early TNF-α-induced nuclear translocation of p65/RelA and enhanced ERK1/2 phosphorylation. However, its cytoprotective effect persisted despite pharmacological inhibition of IKK activity, indicating that canonical NF-κB transcriptional signaling is not required for KDAS-mediated rescue. Moreover, KDAS retained cytoprotective activity in the presence of DAct or CHX, which block de novo transcription and protein synthesis, respectively. These findings indicate that KDAS-mediated cytoprotection does not rely on newly synthesized NF-κB-dependent survival factors, but instead operates upstream of transcriptional survival programs. A similar conclusion was reached for the MAPK pathway. Although KDAS enhanced TNF-α-driven ERK1/2 phosphorylation, pharmacological inhibition of MEK with UO126 or trametinib effectively suppressed ERK activation without abolishing KDAS-mediated cytoprotection. While ERK blockade increased TNF-α-induced cytotoxicity, consistent with its established pro-survival role40, KDAS retained significant rescue capacity under these conditions. Together these observations indicate that neither NF-κB nor ERK signaling is required for KDAS-mediated cytoprotection. KDAS reduces caspase-8 activation and downstream apoptotic events, indicating interference with early stages of extrinsic apoptotic signaling. However, the precise molecular target within the TNFR1 signaling cascade remains to be defined. TNF-α-induced apoptosis in our system is most consistent with TNFR1-mediated signaling, as soluble TNF-α predominantly activates TNFR1, which initiates caspase-8 activation through death-domain-dependent signaling complexes, whereas TNFR2 primarily modulates this pathway indirectly3. Notably, KDAS exerts its protective effect in both FIP200-deficient and wt cells, suggesting that its activity does not depend on restoration of FIP200-dependent unconventional autophagy. Consistent with this, pharmacological inhibition of autophagy using SAR405, a broadly employed inhibitor of the lipid kinase VPS3441, did not abolish KDAS-mediated cytoprotection, supporting an autophagy-independent mechanism (Supplementary Fig. S7). While these findings provide mechanistic insight into the level at which KDAS interferes with TNF-α-induced apoptosis, the precise molecular interactions involved remain unresolved and will require further investigation. From a translational perspective, KDAS-mediated protection of epithelial cells may contribute to preserving barrier integrity under inflammatory conditions. In line with this possibility, dehydroandrographolide succinate, the unsalted form of KDAS, has recently been reported to exert protective effects in a dextran sodium sulfate (DSS)-induced murine model of colitis42, supporting the translational potential of this compound class in IBD. Notably, that study focused on macrophage polarization, NF-κB-driven inflammatory responses, and gut microbiota remodeling, rather than direct regulation of epithelial apoptosis or TNF-α-induced cell death. These observations suggest that related compounds may exert complementary activities across both immune and epithelial compartments. Although epithelial barrier function was not directly assessed in the present study, further evaluation in more physiologically relevant experimental systems, including intestinal organoids, patient-derived epithelial monolayers from IBD and non-IBD donors under inflammatory cytokine challenge, epithelial barrier assays, and in vivo models of colitis, will be important to determine whether KDAS-mediated cytoprotection translates into improved barrier integrity, modulation of inflammatory responses, and altered epithelial-immune interactions under inflammatory conditions. In epithelial cells, extrinsic apoptosis frequently requires mitochondrial amplification to achieve full caspase activation43,44. Consistent with this, Bax deficiency delayed TNF-α-induced cell death in HCT116 cells. KDAS preserved mitochondrial membrane potential and prevented cytosolic release of cytochrome c and apoptosis-inducing factor following TNF-α challenge, suggesting interruption of the apoptotic cascade prior to mitochondrial outer membrane permeabilization. By reducing caspase-8 activation, KDAS limits downstream apoptotic signaling and maintains membrane integrity. Moreover, KDAS suppressed TRAIL-induced apoptosis, a DR4/DR5 ligand that also converges on caspase-8 activation. TRAIL has been implicated in inflammatory intestinal epithelial lesions, where it contributes to enterocyte apoptosis45. In addition, intestinal fibroblasts from Crohn’s disease patients display increased sensitivity to TRAIL-mediated apoptosis, a process linked to tissue remodeling and the development of strictures and fistulas46. These observations suggest that KDAS interferes with a common step within the extrinsic apoptotic signaling, although the precise molecular target remains to be determined. In contrast, KDAS did not prevent TNF-α-driven necroptosis in HT29 cells. Moreover, KDAS also failed to prevent apoptosis induced by several chemotherapeutic agents, broadly employed in colon adenocarcinoma treatment (including irinotecan, topotecan, and cisplatin), likely because these drugs induce cell death independently of caspase-8 activation. This selectivity indicates that KDAS does not broadly suppress cell death but preferentially reduces death-ligand-mediated, caspase-8 activation. Such specificity may be advantageous in limiting inflammatory epithelial injury without interfering with intrinsic apoptosis triggered by cytotoxic therapies. In summary, KDAS selectively protects cells rendered vulnerable to TNF-α-driven apoptosis by genetic abrogation of FIP200 or by pharmacological disruption of survival checkpoints. KDAS consistently reduces caspase-8 activation and downstream mitochondrial apoptotic events, suggesting interference with early stages of extrinsic apoptotic signaling. However, the precise molecular target of KDAS within the TNFR1 signaling cascade remains to be defined. Further studies addressing TNFR1 complex assembly, ubiquitination dynamics, and DISC formation will be required to fully elucidate its mechanism of action. Given the contribution of excessive epithelial apoptosis to mucosal barrier breakdown in moderate-to-severe IBD, pharmacological reinforcement of epithelial resilience during inflammatory bursts may represent a complementary therapeutic strategy. Further studies in more physiologically relevant systems will be required to determine whether KDAS-mediated restraint of caspase-8 activation can preserve intestinal mucosal integrity under inflammatory conditions. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2