Integrated Stress Response and Necroptosis Drive Epithelial Dysfunction in Crohn’s Disease: Repurposing Cancer Drugs for Permeability Barrier Healing Authors: Debasish Halder, Ritwika Biswas Lanford, Amin Esmaeilniakooshkghazi, Jason Ken Hou, Yaohong Wang, Seema Khurana Published: 2026 ABSTRACT Background and AimsIntestinal permeability dysfunction is a central pathogenic driver of Crohn’s disease (CD), fueling microbial translocation, chronic inflammation, and progressive tissue injury. While current therapies suppress inflammation, none directly restore epithelial barrier function. Importantly, in patients with CD, epithelial barrier healing rather than mucosal healing is associated with long-term remission and a reduced risk of disease complications. Yet permeability barrier healing remains an unaddressed therapeutic target in CD. Here, we investigated whether pharmacologic inhibition of the integrated stress response (ISR) and RIPK3-mediated necroptosis, 2 convergent pathways of epithelial injury, can promote epithelial viability, regeneration, and barrier integrity in CD.MethodsWe employed villin-1/gelsolin double knockout mice with epithelial-intrinsic ISR activation, TnfΔARE/+ mice with chronic inflammation, and CD patient-derived enteroids (PDEs). Animals and PDE were treated with integrated stress response inhibitor, RIPK3 inhibitor necrostatin-1, or US Food and Drug Administration–approved cancer drugs pazopanib and ponatinib, repurposed as potent RIPK3 inhibitors. Epithelial survival, regenerative growth (enteroid formation, budding), and barrier function (transepithelial electrical resistance) were assessed.ResultsChronic ISR activation and necroptosis were prominent in both murine models and CD PDEs, causing epithelial death, Paneth cell expansion, impaired enteroid survival, and regenerative failure. Pharmacologic inhibition with integrated stress response inhibitor, necrostatin-1, pazopanib, or ponatinib restored villus architecture, reduced inflammation, enhanced epithelial survival and regeneration, and significantly improved transepithelial electrical resistance.ConclusionISR activation and RIPK3-mediated necroptosis converge to drive epithelial injury and barrier dysfunction in CD. Repurposing pazopanib and ponatinib offers a potentially translatable approach to restore barrier integrity in CD. Introduction Crohn’s disease (CD) is a chronic, relapsing inflammatory disorder of the gastrointestinal tract characterized by transmural inflammation, progressive mucosal damage, and the development of complications such as strictures, fistulae, and abscesses. Despite major therapeutic advances with biologics and small molecules, most current CD therapies primarily target immune pathways and not reepithelialization of the damaged gut. Although inflammation control improves symptoms, it achieves mucosal healing or, more appropriately, endoscopic remission (ER), defined as the absence of inflammation, ulcers, or bleeding in the gastrointestinal tract, in a modest 15%–20% of patients with CD.1 While ER is an important treatment goal in CD associated with improved clinical outcomes, nevertheless, a vast majority of patients with CD remain vulnerable to disease progression and intestinal complications.1 More notably, the presence of mucosal healing in CD does not always correspond with clinical remission or the resolution of a patient’s symptoms.2 Additionally, a proportional clinical benefit is not received by most patients with CD and instead are exposed to the risks of potent immunosuppressive therapies. A defining feature of CD pathogenesis is intestinal permeability dysfunction.3 Under homeostatic conditions, the intestinal epithelium provides a dynamic and selective permeability barrier that separates luminal microbes from the mucosal immune system. Loss of barrier integrity permits the translocation of commensal and pathogenic bacteria, as well as antigens or other foreign substances from the gut lumen into the underlying tissue, triggering an exaggerated and uncontrolled immune response.4 In CD, this sets in motion a vicious cycle where microbial translocation drives inflammation, inflammation induces further epithelial injury, and the subsequent chronic tissue damage leads to fibrostenosis and/or fistulization, which ultimately increases the need for hospitalization and surgery.5 Epidemiologic and prospective studies have demonstrated that permeability dysfunction is not merely a consequence of inflammation but an initiating factor in CD.3,6 Specifically, asymptomatic individuals with increased intestinal permeability are at significantly higher risk of developing CD, providing causal evidence that permeability barrier dysfunction contributes to disease initiation.6 Relatives of patients with CD often display increased intestinal permeability even before clinical disease onset, supporting a heritable component of permeability barrier defects as well as identifying barrier dysfunction as a primary defect in CD.7,8 Together, these findings highlight the etiologic and therapeutic importance of epithelial permeability barrier restoration in CD.5,9 Therapeutic goals in CD have historically been framed around symptom control, corticosteroid-free remission, and mucosal healing.10 Indeed, the Selecting Therapeutic Targets in INflammatory Bowel Disease II consensus reinforced “deep remission,” defined as clinical remission plus ER, as the gold standard treatment target in CD.11 Achieving ER in CD has been correlated with reduced hospitalization and lower surgical risk.10,12 However, emerging evidence shows that compared to ER, barrier healing (BH) is a superior predictor of sustained remission and has more favorable long-term outcomes in CD.13,14 The prospective Endoscopic and Barrier Healing in Crohn’s Disease Trial demonstrated that BH outperformed ER and histologic remission in predicting adverse outcomes such as intestinal complications, hospitalization, and need for surgery.2 BH therefore represents a more meaningful therapeutic end point in CD. While ER reflects inflammation control and mucosal healing, BH captures restoration of the fundamental epithelial permeability defect that drives CD pathology.3 Despite this, no approved CD therapy directly targets epithelial barrier repair. This therapeutic gap underscores the urgent need for epithelial barrier-focused interventions that complement existing immunosuppressive strategies in CD. The integrated stress response (ISR) is a highly conserved adaptive pathway that regulates cellular responses to diverse stressors including nutrient deprivation, endoplasmic reticulum stress, mitochondrial dysfunction, and microbial and viral infections.15 Central to ISR signaling is phosphorylation of eukaryotic initiation factor 2α (peIF2α), which transiently attenuates global protein translation while allowing selective translation of stress-adaptive genes.15 Acute ISR activation is protective, conserves resources, and promotes cell survival.15 In contrast, chronic or unresolved ISR signaling promotes autophagy or programmed cell death via apoptosis or necroptosis, shifting the balance from adaptation to injury.15,16 Persistent ISR activation is implicated in the etiology of several human diseases, and manipulation of ISR is emerging as a promising therapeutic approach.17 Resolution of ISR requires dephosphorylation of peIF2α by GADD34-protein phosphatase 1 catalytic subunit phosphatase complex, a process sensitive to cellular actin dynamics.18 Prior work from our group has shown that villin-1 and gelsolin, 2 intestinal epithelial cell (IEC) actin-severing proteins, regulate protein phosphatase 1 catalytic subunit activity.16 Their loss leads to sustained ISR activation, mitochondrial stress, and IEC death via necroptosis.16 Double knockout (DKO) mice lacking villin-1 and gelsolin develop spontaneous ileocolitis recapitulating human CD pathology due to chronic ISR signaling, induction of the autophagy protein immunity-related GTPase family M protein 1 (IRGM1), and IRGM1-induced IEC death by necroptosis.16 Notably, similar molecular defects are seen in patients with CD, including reduced villin-1/gelsolin expression, chronic ISR activation, and IRGM1-induced IEC death by necroptosis, underscoring the relevance of this pathway to the pathophysiology of CD.16 These findings suggest that therapeutic targeting of the ISR could restore epithelial homeostasis in CD. Necroptosis is a pro-inflammatory form of programmed cell death that is mediated by receptor-interacting protein kinases RIPK1 and RIPK3 and executed by mixed lineage kinase domain like pseudokinase.19,20 Beyond depleting IECs, necroptosis leads to the release of damage-associated molecular patterns, which augment intestinal inflammation and epithelial injury.16 Moreover, independent of its role in necroptosis, RIPK3 is a key regulator of inflammation regulating the activation of nuclear factor kappa-light-chain-enhancer of activated B cells and the NLRP3 inflammasome.21,22 RIPK3 and mixed lineage kinase domain like pseudokinase levels are increased in inflamed tissues from inflammatory bowel disease patients.23 Additionally, we have shown that patients with CD display not only an upregulation of RIPK3 expression but also higher levels of RIPK3-mediated necroptosis and increased release of the damage-associated molecular pattern high mobility group box 1.16 The dual roles of RIPK3 in necroptosis and inflammation suggest that pharmacologic inhibition of RIPK3 could attenuate IEC death and pro-inflammatory signaling in CD. Tumor necrosis factor (TNF)-α is a major contributor to intestinal inflammation that is linked to the pathogenesis of CD.24 Moreover, genetic studies have identified TNF-α as a CD susceptibility locus, while anti-TNF therapy is a mainstay of CD treatment.24,25 ISR activation modulates pro-inflammatory cytokine production; however, it is not known if pro-inflammatory cytokines like TNF-α can act as stressors to activate ISR signaling. In this study, we used 2 complementary murine models of CD, the villin-1/gelsolin DKO mice, which model epithelial-intrinsic ISR dysregulation, and tumor necrosis factor AU-rich element deletion mutant mouse (TnfΔARE/+) mice, which develop chronic inflammation due to TNF-α dysregulation, together with CD patient-derived enteroids (PDEs), to test our hypothesis that ISR activation in response to intrinsic (villin-1/gelsolin loss) or extrinsic (TNF-α) stressors and downstream of it necroptosis, are the main drivers of epithelial dysfunction in CD. Furthermore, using these in vivo and ex vivo murine and human models of CD, we evaluate our hypothesis that pharmacologic inhibition of ISR and necroptosis can restore epithelial homeostasis and barrier integrity. Our findings establish a mechanistic framework linking ISR and necroptosis to permeability dysfunction in CD and demonstrate the translational potential of repurposed RIPK3 inhibitors as novel therapies to achieve mucosal and BH in CD. Methods Mouse Models of CD The Institutional Animal Care and Use Committee of the University of Houston approved all animal experiments. DKO mice lacking villin-1 and gelsolin were generated as previously described.26 TnfΔARE/+ mice, which develop chronic ileitis due to dysregulated TNF-α expression, were obtained from Dr Fabio Cominelli (Case Western Reserve University).27 Wild-type (WT) littermates of DKO and the Tnf+/+ mice served as controls. Both male and female mice were studied at 20 and 35 weeks of age, corresponding to peak spontaneous ileitis in TnfΔARE/+ and DKO mice, respectively.16,28 Intestinal Mouse and Human Enteroid Culture Human CD and non-CD control samples were obtained under protocols approved by the Institutional Review Boards of Baylor College of Medicine and University of Houston. Crypts from murine and human distal ileum were isolated by incubation in 2 mM EDTA, and enteroids were derived as described previously.29 The organoid forming capacity of stem cell or the organoid forming efficiency (percentage) is defined as a percentage of the number of organoids per number of plated stem cells at day 7, when 1 cell is added per well.30 The percentage of budded enteroids was calculated as: (number of enteroids with ≥1 crypt-like bud/total number of enteroids counted) × 100 on day 7.31 Crypt buds per enteroid, a measure of stem cell–driven proliferation and regeneration, was also quantified as described previously.31 At least 50 enteroids per sample were assessed for all quantitative assays. Epithelial cell mortality was assessed using Hoechst 33342 or propidium iodide (PI) staining. Imaging was performed on a Fluoview FV2000 laser scanning confocal microscope with a 60× NA 1.35 objective. In Vivo and Ex Vivo Treatments Mice received intraperitoneal injections of Necrostatin-1 (Nec-1; 1.65 mg/kg) or integrated stress response inhibitor (ISRIB; 0.25 mg/kg) every other day starting at 1 month of age.32 Following treatment, TnfΔARE/+ mice were euthanized at 6 months of age and DKO at 8 months of age and ileal tissue collected for histology, immunostaining, and protein analysis. Vehicle-treated animals (0.3% DMSO) served as controls. Ex vivo, DKO, TnfΔARE/+ mouse enteroids, and CD PDEs were treated with ISRIB (50 nM, Cayman Chemicals), Nec-1 (10 μM), pazopanib (0.5 μM), or ponatinib (2 nM) for 7 days and compared with vehicle (DMSO)-treated controls. Histology and Immunostaining Distal ileal specimens were fixed in 4% paraformaldehyde and embedded in paraffin. Five-micrometer sections were stained with hematoxylin and eosin to evaluate villus architecture, crypt morphology, and inflammatory infiltration. Histological evaluation of intestinal inflammation was scored as described previously.33 For immunohistochemistry, sections were deparaffinized, antigen-retrieved, and incubated with antibodies to phospho-eIF2α, IRGM1, or lysozyme. Images were captured with an Olympus BX53 microscope. For immunoblotting, protein lysates from distal ileum were prepared using radioimmunoprecipitation assay buffer with proteases and phosphatase inhibitors. Equal protein amounts were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to polyvinylidene difluoride membranes, and probed with antibodies against eIF2α, peIF2α, IRGM1, and β-actin. Bands were visualized using chemiluminescence (Thermo Fisher). Transepithelial Electrical Resistance CD PDEs were seeded on Transwell inserts (0.4 μm pore size; Corning) coated with 20 μg/mL collagen type 1 (rat tail) in 0.01% acetic acid. Enteroids were dissociated to single cells using TrypIE (Gibco, Thermo Fisher Scientific) for 10 minutes at 37 °C and cultured as monolayers until confluent. Transepithelial electrical resistance (TEER) was measured using an EVOM2 epithelial voltohmmeter (World Precision Instruments). Each insert was measured 3 times; average values were corrected for background and normalized to surface area to calculate net resistance (Ω.cm2). Statistical Analysis All experiments were performed with at least 3 biological replicates. Data are presented as mean ± standard error of the mean. Statistical analyses were conducted using GraphPad Prism v9. Comparisons between 2 groups were performed using unpaired 2-tailed Student t-test; comparison of >2 groups used one-way analysis of variance (ANOVA) with Tukey post hoc test. A P value of <.05 was considered statistically significant. Results ISR Activation Is a Convergent Hallmark of Epithelial Injury in CD Models To determine whether intrinsic epithelial stress or chronic inflammation (extrinsic stressor) converge on epithelial dysfunction and ISR activation, we examined villin-1/gelsolin DKO mice, which model epithelial-intrinsic ISR dysregulation, and TnfΔARE/+ mice, which develop chronic ileitis due to TNF dysregulation.16,27 Both strains developed spontaneous ileitis and transmural inflammation resembling CD, as shown previously.16,27 Histomorphological analysis of distal ileum revealed profound epithelial damage in DKO and TnfΔARE/+ mice, including villus blunting, villus fusion, and crypt elongation, accompanied by thickening of the muscularis propria (Figure 1A and B). Inflammatory cell infiltration was significantly elevated, as quantified by histological scoring (P < .0001, n = 6 mice per group). Paneth cell hyperplasia and displacement were observed in both mouse models, mirroring phenotypes characteristic of human CD (Figure 1A and B; P < .0001, n = 6 mice per group).34 These findings are consistent with expansion of the proliferative compartment, a change in IEC lineage, structural compromise of epithelial integrity, and chronic inflammatory remodeling.Figure 1ISR activation is a hallmark of epithelial injury in murine models of CD. (A and B) Representative hematoxylin and eosin (H&E)–stained ileal sections from distal ileum of villin-1/gelsolin double knockout (DKO) and TnfΔARE/+ mice and their wild-type (WT) littermates which include Tnf+/+ mice. DKO and TnfΔARE/+ mice show ileal wall thickening, villus blunting, villus fusion, and crypt elongation compared to their respective control mice. Histology scores, measuring intestinal inflammation, were significantly elevated in DKO and TnfΔARE/+ mice compared to their control mice. (C) Immunohistochemistry and immunoblots show upregulation of phospho-EIF2α (pEIF2α) and IRGM1 expression levels in the TnfΔARE/+ mice without any notable change in total eIF2α levels, confirming chronic ISR activation. Densitometric quantification of immunoblots for pEIF2a normalized to total eIF2α and IRGM1 normalized to β-actin is provided in the Results. Data are presented as mean ± SEM from ≥3 independent experiments. Statistical significance was determined using 2-tailed Student t-test, ∗∗∗∗P < .0001, n = 6 mice per group. SEM, standard error of the mean. ISR activation is a hallmark of epithelial injury in murine models of CD. (A and B) Representative hematoxylin and eosin (H&E)–stained ileal sections from distal ileum of villin-1/gelsolin double knockout (DKO) and TnfΔARE/+ mice and their wild-type (WT) littermates which include Tnf+/+ mice. DKO and TnfΔARE/+ mice show ileal wall thickening, villus blunting, villus fusion, and crypt elongation compared to their respective control mice. Histology scores, measuring intestinal inflammation, were significantly elevated in DKO and TnfΔARE/+ mice compared to their control mice. (C) Immunohistochemistry and immunoblots show upregulation of phospho-EIF2α (pEIF2α) and IRGM1 expression levels in the TnfΔARE/+ mice without any notable change in total eIF2α levels, confirming chronic ISR activation. Densitometric quantification of immunoblots for pEIF2a normalized to total eIF2α and IRGM1 normalized to β-actin is provided in the Results. Data are presented as mean ± SEM from ≥3 independent experiments. Statistical significance was determined using 2-tailed Student t-test, ∗∗∗∗P < .0001, n = 6 mice per group. SEM, standard error of the mean. Immunoblot analysis of intestinal tissue from TnfΔARE/+ mice revealed a marked increase in phosphorylated eIF2α (peIF2α), a marker of ISR activation, compared to Tnf+/+ controls, while total eIF2α remained unchanged (Figure 1C). Densitometric normalization to total eIF2α and β-actin demonstrated an approximately 2.5-to-3.0-fold increase, respectively, in peIF2α in TnfΔARE/+ mice relative to WT mice. IRGM1 protein expression was also robustly and reproducibly increased in TnfΔARE/+ mice relative to WT, with densitometric analysis indicating an approximately 4-fold induction after normalization to β-actin. The robust induction of IRGM1 in TnfΔARE/+ mice suggests activation of epithelial innate defense/autophagy programs in response to chronic TNF-driven stress, consistent with inflammation-linked remodeling of mucosal homeostasis. These results extend our prior findings in DKO mice and patients with CD, demonstrating that despite distinct initiating stressors, epithelial intrinsic vs extrinsic and inflammatory, chronic ISR activation, and IRGM1 induction are convergent hallmarks of epithelial injury in CD.16 Chronic ISR Impairs Epithelial Viability and Regenerative Capacity in Murine and CD Patient-Derived Enteroids To directly assess epithelial cell-intrinsic responses, we derived enteroids from the distal ileum of DKO and TnfΔARE/+ mice, as well as from ileal pinch biopsies from patients with CD and non-CD controls. Phase contrast imaging and PI staining revealed significantly increased IEC death in DKO and TnfΔARE/+ enteroids compared with their respective controls (Figure 2A and B). Quantification showed markedly elevated PI-positive cells in DKO and TnfΔARE/+ enteroids, confirming loss of membrane integrity. Necroptosis inhibition with Nec-1 rescued IEC survival in TnfΔARE/+ mouse enteroids, whereas pan-caspase inhibition (Z-VAD-FMK) did not, confirming necroptosis rather than apoptosis as the dominant mode of cell death (Supplementary Figure 1), which is similar to our findings in the DKO mice and CD PDEs.16 Enteroid forming efficiency, a measure of intestinal stem cell (ISC) regenerative potential, was profoundly reduced in both DKO and TnfΔARE/+ mouse enteroids relative to control enteroids (P < .0001, n = 4 mice per group). Enteroids from these murine models exhibited diminished complexity, with fewer crypt-like buds per enteroid and a reduced percentage of budded enteroids (Figure 2A and B). These findings reflect defects in ISC-driven regeneration and epithelial renewal. Strikingly, CD PDEs recapitulated the murine phenotypes, showing impaired survival, decreased enteroid forming efficiency, reduced bud number, and lower proportion of budded enteroids compared to non-CD controls (Figure 2C; P < .0001, n = 8 patients per group).Figure 2Chronic ISR activation impairs epithelial viability, regeneration, and growth in mouse and human models of CD. (A and B) Representative phase-contrast and fluorescence images of PI and Hoechst 33342 labeled enteroids from DKO, TnfΔARE/+, and control mice (WT and Tnf+/+, respectively) show increased intestinal epithelial cell (IEC) mortality in both murine models of CD. (A and B) Enteroids seeded at day 1 and analyzed at day 7 show significant inhibition of enteroid-forming efficiency (percentage of plated crypts forming viable enteroids) in controls vs DKO and TnfΔARE/+ mice. Quantification of crypt buds per enteroid at day 7 shows reduced budding in DKO and TnfΔARE/+ mouse enteroids compared to their respective controls. Percentage of budded enteroids is significantly decreased in DKO and TnfΔARE/+ mice at day 7 relative to controls. (C) CD patient-derived enteroids (PDEs) exhibit significantly increased mortality and inhibition of enteroid-forming efficiency, number of enteroids budded, and total number of budded enteroids formed compared to PDEs from non-CD controls. (D) Representative confocal images show Paneth cell expansion in DKO and TnfΔARE/+ mouse enteroids. Quantification of Paneth cell numbers reveals a significant increase in both DKO and TnfΔARE/+ mouse enteroids compared to their respective control mouse enteroids. (E) Representative confocal images of enteroids immunoassayed for peIF2α show significant upregulation of peIF2α expression in the TnfΔARE/+ mouse enteroids compared to Tnf+/+ mouse enteroids. Data obtained from 4 mice per group and 8 patients per group and expressed as mean ± SEM from ≥3 independent experiments. Statistical significance was assessed using one-way ANOVA with Tukey post hoc test; ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. At least 50 enteroids per sample were analyzed for the quantitative analysis. ANOVA, analysis of variance; SEM, standard error of the mean. Chronic ISR activation impairs epithelial viability, regeneration, and growth in mouse and human models of CD. (A and B) Representative phase-contrast and fluorescence images of PI and Hoechst 33342 labeled enteroids from DKO, TnfΔARE/+, and control mice (WT and Tnf+/+, respectively) show increased intestinal epithelial cell (IEC) mortality in both murine models of CD. (A and B) Enteroids seeded at day 1 and analyzed at day 7 show significant inhibition of enteroid-forming efficiency (percentage of plated crypts forming viable enteroids) in controls vs DKO and TnfΔARE/+ mice. Quantification of crypt buds per enteroid at day 7 shows reduced budding in DKO and TnfΔARE/+ mouse enteroids compared to their respective controls. Percentage of budded enteroids is significantly decreased in DKO and TnfΔARE/+ mice at day 7 relative to controls. (C) CD patient-derived enteroids (PDEs) exhibit significantly increased mortality and inhibition of enteroid-forming efficiency, number of enteroids budded, and total number of budded enteroids formed compared to PDEs from non-CD controls. (D) Representative confocal images show Paneth cell expansion in DKO and TnfΔARE/+ mouse enteroids. Quantification of Paneth cell numbers reveals a significant increase in both DKO and TnfΔARE/+ mouse enteroids compared to their respective control mouse enteroids. (E) Representative confocal images of enteroids immunoassayed for peIF2α show significant upregulation of peIF2α expression in the TnfΔARE/+ mouse enteroids compared to Tnf+/+ mouse enteroids. Data obtained from 4 mice per group and 8 patients per group and expressed as mean ± SEM from ≥3 independent experiments. Statistical significance was assessed using one-way ANOVA with Tukey post hoc test; ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. At least 50 enteroids per sample were analyzed for the quantitative analysis. ANOVA, analysis of variance; SEM, standard error of the mean. Confocal imaging revealed Paneth cell expansion in the DKO and TnfΔARE/+ mouse enteroids, consistent with the in vivo phenotype (Figure 2D). In addition, TnfΔARE/+ enteroids exhibited sustained peIF2α expression compared to Tnf+/+ control enteroids (Figure 2E). Collectively, these results demonstrate that chronic ISR activation compromises epithelial survival, reduces ISC regenerative capacity, and drives Paneth cell lineage expansion across murine and human models of CD. Together with our prior work, these data establish necroptosis as a principal driver of IEC death and chronic ISR as a central regulator of regenerative failure in CD.16 Pharmacological Inhibition of ISR or Necroptosis Restores Epithelial Architecture In Vivo We next tested whether targeting ISR or necroptosis could mitigate epithelial pathology in vivo. DKO and TnfΔARE/+ mice were treated with Nec-1 (RIPK 1/3 inhibitor) or ISRIB, a small-molecule inhibitor of ISR that binds to eIF2β and restores protein synthesis even when eIF2α is phosphorylated.35 Essentially, ISRIB “overrides” the ISR brake on translation.36 Histological analysis of DKO mouse ileum revealed significant restoration of mucosal architecture in the Nec-1- and ISRIB-treated mice compared to vehicle-treated controls (Figure 3A and B, P < .0001, n = 6 mice per group). Drug-treated mice exhibited reduced villus fusion, normalized crypt length, and decreased muscularis propria thickening (P < .0001, n = 6 mice per group). Importantly, Paneth cell hyperplasia was significantly reduced following Nec-1 or ISRIB treatment of DKO mice, indicating correction of lineage imbalance. Histological scores confirmed marked suppression of intestinal inflammation (P < .0001, n = 6 mice per group). We next asked whether ISR inhibition would provide similar benefit in an inflammation-driven model of Crohn’s-like ileitis. In TnfΔARE/+ mice, treatment with ISRIB significantly reduced histological scores compared with vehicle controls (Figure 3C, P < .0001, n = 6 mice per group). Additionally, ISRIB treatment led to significant reduction in villus fusion, crypt elongation, and ileal thickening, thus restoring the ileal tissue architecture (P < .0001, n = 6 mice per group). This was accompanied by a significant reduction in the number of Paneth cell counts per crypt, mirroring the effects of ISRIB in DKO mice (P < .0001, n = 6 mice/group). Together, these findings demonstrate that chronic ISR activation and necroptotic IEC death are tractable therapeutic vulnerabilities and that pharmacologic inhibition of either pathway can restore epithelial homeostasis in vivo.Figure 3In vivo pharmacologic inhibition of ISR and necroptosis restore epithelial homeostasis in murine models of CD. DKO (A and B) and TnfΔARE/+ (C) mice were administered ISRIB (0.25 mg/kg, intraperitoneal) or Necrostatin-1 (Nec-1; 1.65 mg/kg; intraperitoneal) as described in the Methods. Representative H&E-stained ileal sections show restoration of tissue architecture in the DKO, and TnfΔARE/+ mice treated with Nec-1 and ISRIB compared to vehicle-treated controls. Both treatments significantly improved the histologic score showing inhibition of intestinal inflammation. Other indices of tissue histomorphology, such as thickening of the muscularis propria, crypt elongation, number of fused villi, and total numbers of Paneth cells per crypt, were also significantly reduced by Nec-1 and ISRIB treatment compared to vehicle-treated controls. Data are presented as mean ± SEM from ≥3 independent experiments. Statistical analysis was performed using one-way ANOVA with Tukey post hoc test; ∗∗∗∗P < .0001; n = 6 mice per group. ANOVA, analysis of variance; H&E, hematoxylin and eosin; SEM, standard error of the mean. In vivo pharmacologic inhibition of ISR and necroptosis restore epithelial homeostasis in murine models of CD. DKO (A and B) and TnfΔARE/+ (C) mice were administered ISRIB (0.25 mg/kg, intraperitoneal) or Necrostatin-1 (Nec-1; 1.65 mg/kg; intraperitoneal) as described in the Methods. Representative H&E-stained ileal sections show restoration of tissue architecture in the DKO, and TnfΔARE/+ mice treated with Nec-1 and ISRIB compared to vehicle-treated controls. Both treatments significantly improved the histologic score showing inhibition of intestinal inflammation. Other indices of tissue histomorphology, such as thickening of the muscularis propria, crypt elongation, number of fused villi, and total numbers of Paneth cells per crypt, were also significantly reduced by Nec-1 and ISRIB treatment compared to vehicle-treated controls. Data are presented as mean ± SEM from ≥3 independent experiments. Statistical analysis was performed using one-way ANOVA with Tukey post hoc test; ∗∗∗∗P < .0001; n = 6 mice per group. ANOVA, analysis of variance; H&E, hematoxylin and eosin; SEM, standard error of the mean. Repurposed Cancer Drugs With RIPK3 Inhibitory Properties Rescue Epithelial Survival and Regeneration in CD Several US Food and Drug Administration (FDA)-approved cancer drugs, including pazopanib, ponatinib, sorafenib, dabrafenib, ibrutinib, quizartinib, and regorafenib, have been found to selectively inhibit RIPK3.37 Given the translational limitations of ISRIB and Nec-1, we tested FDA-approved cancer drugs pazopanib and ponatinib, both identified as potent RIPK3 inhibitors at suboncologic doses.38 Ex vivo treatment of DKO and TnfΔARE/+ mouse enteroids and CD PDEs with ISRIB (150 nM), Nec-1 (10 μM), pazopanib (0.5 μM), or ponatinib (2 nM) markedly restored enteroid-forming efficiency, crypt budding, and increased the number of budded enteroids, while also reducing IEC mortality compared with their respective vehicle (DMSO)-treated controls (Figure 4A–C; P < .0001, enteroids were derived from n = 4 mice and n = 6 patients; Supplementary Figure 2). The magnitude of rescue with pazopanib and ponatinib was comparable to that observed with ISRIB and Nec-1, indicating comparable efficacy. These results show that pazopanib and ponatinib can rescue epithelial viability and regenerative capacity in both murine and human CD enteroids, highlighting their potential as clinically translatable mucosal healing therapies in CD.Figure 4Repurposed cancer drugs with potent RIPK3 inhibiting properties rescue epithelial viability and regeneration. (A–C) Representative phase-contrast and PI stained-merged images of DKO and TnfΔARE/+ and CD PDEs treated with ISRIB, Nec-1, pazopanib, ponatinib, or vehicle (DMSO), show significant inhibition of IEC mortality. Enteroids exogenously treated with these inhibitors also showed significantly improved enteroid forming efficiency, number of enteroids with buds, and the number of buds per enteroid compared to vehicle-treated control enteroids. Data are from 4 mice per group and 8 patients per group, represented as mean ± SEM from ≥3 independent experiments. Statistical significance was assessed using one-way ANOVA with Tukey post hoc test; ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. ANOVA, analysis of variance; SEM, standard error of the mean. Repurposed cancer drugs with potent RIPK3 inhibiting properties rescue epithelial viability and regeneration. (A–C) Representative phase-contrast and PI stained-merged images of DKO and TnfΔARE/+ and CD PDEs treated with ISRIB, Nec-1, pazopanib, ponatinib, or vehicle (DMSO), show significant inhibition of IEC mortality. Enteroids exogenously treated with these inhibitors also showed significantly improved enteroid forming efficiency, number of enteroids with buds, and the number of buds per enteroid compared to vehicle-treated control enteroids. Data are from 4 mice per group and 8 patients per group, represented as mean ± SEM from ≥3 independent experiments. Statistical significance was assessed using one-way ANOVA with Tukey post hoc test; ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. ANOVA, analysis of variance; SEM, standard error of the mean. Repurposed RIPK3 Inhibitors Restore Permeability Barrier Integrity in CD PDE Finally, we evaluated whether these pharmacological interventions improved functional barrier integrity, as measured by TEER. Treatment with pazopanib, ponatinib, ISRIB, or Nec-1 significantly improved TEER values compared to vehicle-treated controls (Figure 5, P < .001, n = 6 patients per group). In CD PDE monolayers, pazopanib and ponatinib significantly improved TEER, directly demonstrating functional recovery of permeability barrier integrity. These results directly demonstrate that repurposed RIPK3 inhibitors not only promote mucosal healing but also promote BH in human CD epithelium.Figure 5Pharmacologic blockade of ISR, necroptosis, and RIPK3 signaling restores epithelial permeability barrier function in CD patient-derived enteroids. Transepithelial electrical resistance (TEER, Ω.cm2) was measured in CD PDE treated with vehicle (DMSO), Nec-1, ISRIB, pazopanib, or ponatinib. All agents significantly improved TEER compared with vehicle-treated CD enteroids, indicating restoration of epithelial barrier integrity. Data represent mean ± SEM from ≥3 independent experiments; ∗∗∗P < .001. SEM, standard error of the mean. Pharmacologic blockade of ISR, necroptosis, and RIPK3 signaling restores epithelial permeability barrier function in CD patient-derived enteroids. Transepithelial electrical resistance (TEER, Ω.cm2) was measured in CD PDE treated with vehicle (DMSO), Nec-1, ISRIB, pazopanib, or ponatinib. All agents significantly improved TEER compared with vehicle-treated CD enteroids, indicating restoration of epithelial barrier integrity. Data represent mean ± SEM from ≥3 independent experiments; ∗∗∗P < .001. SEM, standard error of the mean. Across murine models and human CD PDEs, we show that (1) chronic ISR activation and necroptosis are convergent hallmarks of epithelial dysfunction; (2) these pathways drive IEC death, regenerative failure, IEC lineage defects, and loss of epithelial barrier integrity; and (3) pharmacologic inhibition with ISRIB, Nec-1, pazopanib, or ponatinib rescues epithelial survival, regenerative growth, and barrier integrity. These findings position ISR and RIPK3 inhibition as promising therapeutic strategies to achieve BH in CD. Discussion Our results identify ISR activation as a unifying hallmark of epithelial stress in CD, regardless of whether the initiating insult is intrinsic (loss of villin-1/gelsolin) or extrinsic (TNF-α dysregulation). TnfΔARE/+ mice display chronic peIF2α phosphorylation and IRGM1 induction, molecular features we previously identified in the DKO mice and human CD epithelium.16 By regulating mitochondrial fission and membrane polarization, IRGM influences mitochondrial quality control However, overexpression of IRGM1 causes mitochondrial depolarization and sensitizes IECs to necroptotic death, as shown by us previously in DKO mice and patients with CD.16,39 In acute stress, the ISR restores homeostasis and promotes survival.15 Our data demonstrate that chronic ISR activation in the gastrointestinal tract is maladaptive, locking epithelial cells in a state of translational arrest and mitochondrial stress that culminates in necroptotic cell death and impairment of epithelial regeneration.16 Mechanistically, our data support a model in which chronic ISR activation and RIPK3-mediated necroptosis form a self-reinforcing epithelial injury loop in CD. These shared molecular defects across murine and human models support a convergent epithelial stress axis driving barrier dysfunction in CD and offer a tractable therapeutic target. In this study, enteroid assays provided critical functional insights into how ISR and necroptosis compromise epithelial homeostasis and function. Parallel defects in enteroid formation, budding, and regenerative capacity across murine and CD PDEs underscore the role of chronic ISR activation in compromising ISC viability and epithelial renewal. In response to injury, stem cells adopt new fates and lineages to support regeneration of the damaged tissue.40 While acute ISR activation promotes adaptive epithelial repair, chronic ISR signaling is implicated in defective epithelial regeneration.41 Our findings demonstrate that in CD, sustained ISR activation maintains IECs in a chronic stress state that inhibits ISC renewal, thus contributing directly to defective mucosal and BH. Together with our prior work, in this study, we show that aberrant Paneth cell expansion is a conserved feature of epithelial stress response in CD.16 Paneth cells provide essential niche signals for stem cell maintenance, and antimicrobial peptides and their expansion may represent a compensatory but dysfunctional response to epithelial injury. These findings reinforce the concept that chronic ISR disrupts epithelial homeostasis at multiple levels by depleting viable IECs but also by inhibiting ISC renewal and disrupting lineage specification. ISR activation has been linked to cellular metabolism and ISR modulates stem cell fate decisions in response to serine metabolism.40 This raises the intriguing possibility that dietary interventions could synergize with ISR-targeted therapies to promote mucosal and BH in CD. Together with our previous findings, data provided here show that in CD, pathologic IEC death in the intestine is mediated primarily by necroptosis and not apoptosis.16 In the small intestine, enterocytes and enterocyte progenitor cells are particularly sensitive to necroptosis, a caspase-independent lytic form of programmed cell death.16,42,43 We postulate that the increased susceptibility of the enterocyte progenitors to necroptosis contributes to impaired epithelial regeneration identified in the DKO, TnfΔARE/+, and CD PDEs. The link between impaired gut barrier and inflammatory bowel disease was established >30 years ago, but the lack of clinical translation is in part due to the lack of identifiable therapeutic targets. Previous studies have shown that growth factors such as EGFR and R-spondin1 can restore the damaged intestinal epithelium; however, both are mitogens and could increase the risk of developing colorectal cancer.44,45 Larazotide acetate was identified as a BH therapy; however, its barrier-protective effects were not recapitulated in clinical trials.46 A major advance of this work is the demonstration that pharmacologic inhibition of ISR or necroptosis restores epithelial homeostasis by promoting regenerative growth of the epithelium in both murine and human models of CD and by restoring barrier integrity in CD PDEs. In vivo, both ISRIB and Nec-1 significantly improved villus architecture, reduced inflammation, and normalized Paneth cell numbers in DKO and TnfΔARE/+ mice. Ex vivo, ISRIB and Nec-1 improved IEC viability, ISC renewal, and regeneration of the crypt-villus-like enteroid architecture. Notably, ISRIB promotes lung repair and reduces fibrosis by accelerating the differentiation of alveolar epithelial cells after injury.47 This implies the potential of ISRIB to restore epithelial homeostasis across different tissues. It also indicates that given prophylactically, ISRIB could reduce or prevent fibrostenotic strictures in CD. Treatment of stricturing and fistulizing CD is one of the hardest challenges in CD care.48 Strictures are caused by chronic inflammation, and no approved antifibrotic drug exists yet. Moreover, once fibrosis is established in the gut, treatment has limited impact. While anti-TNF biologics can reduce inflammatory components of a stricture, they do not reverse scar tissue and surgical management remains the only option. Notably, the Endoscopic and Barrier Healing in Crohn’s Disease Trial demonstrated that in patients with CD, BH is associated with a decreased risk of developing intestinal complications.2 Consistent with that, we propose that inhibition of ISR has the potential to prevent the initiation and development of fibrostricturing CD. ISRIB, though investigational, has shown safety and oral bioavailability in preclinical studies, making it attractive for translational development for CD.49 Pazopanib and ponatinib are FDA-approved with well-characterized pharmacokinetics.38 While pazopanib and ponatinib have pleiotropic kinase targets at oncologic doses, multiple independent proteomic screens demonstrate RIPK3 selectivity at suboncologic concentrations used here, supporting the epithelial-protective mechanism proposed in this study.38,50 At micromolar concentrations, ponatinib has been shown to inhibit proliferation and induce apoptosis in cancer cells.51 However, it is now clear that at suboncologic concentrations ponatinib inhibits necroptosis and not the cell cycle.52 We show that at suboncologic concentrations, pazopanib and ponatinib effectively restore epithelial viability, epithelial regenerative capacity, and TEER in CD PDEs. Unlike investigational RIPK3 inhibitors like GSK’872 and GSK843, which can paradoxically trigger apoptosis, pazopanib and ponatinib have been shown to suppress both necroptosis and RIPK3-driven inflammatory signaling, making them ideal candidates for CD therapy.53 We postulate that the dual effect of pazopanib and ponatinib on both necroptosis and inflammation may be critical for effective restoration of barrier integrity in CD. ER in CD is associated with a decreased risk of disease progression and a more favorable long-term disease outcome.10 Multiple studies have used confocal endomicroscopy to assess the structure and function of the intestinal permeability barrier in vivo and demonstrated that impaired barrier integrity in patients with CD is linked with persistence of clinical symptoms and relapsing disease behavior.2,13,14,54 These findings have led to the idea that mucosal healing or ER is not sufficient to improve treatment outcomes and that restoration of barrier integrity is essential to achieve sustained clinical remission and to reduce intestinal complications in CD. Despite that, there are no clinically validated methods for early diagnosis, risk stratification, or targeted intervention based on impaired permeability and there are no BH therapies. Our study introduces a paradigm shift in CD therapy, by focusing on BH, an unaddressed yet decisive driver of clinical outcomes in CD. We demonstrate functional restoration of barrier integrity using TEER. Prior studies have linked RIPK3-mediated necroptosis to alterations in cell-cell junctional proteins (eg, E-cadherin, occluding, Zonulin-1), but whether RIPK3 inhibition directly stabilizes junctional architecture in CD remains to be determined.55 Integrating junctional protein imaging and transcriptomics into future PDE assays will clarify how ISR and RIPK3 inhibition mechanistically restore epithelial barrier structure. Our work also opens several avenues for further investigation. First, development of gut-restricted RIPK3 inhibitors could maximize efficacy while minimizing systemic exposure. Second, incorporating BH as a therapeutic end point in clinical trials will be critical for aligning CD therapy with outcomes most predictive of long-term health.2 Additionally, biomarkers of ISR activation and RIPK3 signaling, such as peIF2α or RIPK3 expression in mucosal biopsies, could be used to stratify patients and guide therapy. Identifying patients with persistent barrier dysfunction will allow targeted use of barrier-restoring agents. Finally, integration of PDE-based assays and TEER measurements into personalized medicine pipelines could enable ex vivo prediction of treatment response in CD. This study has a few limitations. First, while pazopanib and ponatinib rescue epithelial regeneration and barrier integrity, their precise epithelial-intrinsic targets beyond RIPK3 were not dissected here. Second, our models primarily address epithelial-driven pathology and may not fully capture immune-mediated contributions to CD in vivo. Third, we did not directly measure tight junction or adheres junction protein remodeling, which will be essential in future studies to mechanistically link ISR/RIPK3 inhibition to structural barrier repair. Finally, long-term dosing and gut-restricted delivery strategies must be explored to minimize systemic exposure and optimize translational feasibility. Conclusion Our study defines ISR activation and RIPK3-mediated necroptosis as convergent mechanisms driving epithelial dysfunction in CD. By demonstrating that pharmacological inhibition of these pathways restores epithelial regeneration and permeability barrier integrity, we provide proof-of-concept for targeting epithelial stress response and necroptosis as a therapeutic strategy in CD. Repurposing pazopanib and ponatinib offers a potentially translatable approach to achieve BH, a treatment end point superior to endoscopic healing for predicting long-term disease outcomes. Incorporating BH into therapeutic targets and clinical trial design has the potential to fundamentally reshape CD management, moving beyond inflammation control toward sustained restoration of epithelial homeostasis.