Despite the importance of intestinal stem cells (ISCs) for epithelial maintenance, there is limited understanding of how immune-mediated damage affects ISCs and their niche. We found that stem cell compartment injury is a shared feature of both alloreactive and autoreactive intestinal immunopathology, reducing ISCs and impairing their recovery in T cell–mediated injury models. Although imaging revealed few T cells near the stem cell compartment in healthy mice, donor T cells infiltrating the intestinal mucosa after allogeneic bone marrow transplantation (BMT) primarily localized to the crypt region lamina propria. Further modeling with ex vivo epithelial cultures indicated ISC depletion and impaired human as well as murine organoid survival upon coculture with activated T cells, and screening of effector pathways identified interferon-γ (IFNγ) as a principal mediator of ISC compartment damage. IFNγ induced JAK1- and STAT1-dependent toxicity, initiating a proapoptotic gene expression program and stem cell death. BMT with IFNγ–deficient donor T cells, with recipients lacking the IFNγ receptor (IFNγR) specifically in the intestinal epithelium, and with pharmacologic inhibition of JAK signaling all resulted in protection of the stem cell compartment. In addition, epithelial cultures with Paneth cell–deficient organoids, IFNγR-deficient Paneth cells, IFNγR–deficient ISCs, and purified stem cell colonies all indicated direct targeting of the ISCs that was not dependent on injury to the Paneth cell niche. Dysregulated T cell activation and IFNγ production are thus potent mediators of ISC injury, and blockade of JAK/STAT signaling within target tissue stem cells can prevent this T cell–mediated pathology.

Highlights d The crypt base region is the primary intestinal location invaded by T cells after BMT d T cell infiltration does not correlate with overall intestinal vascular architecture d Allo donor T cells rapidly access the ISC compartment and directly interact with ISCs d MAdCAM-1 localizes to crypt region vessels and directs T cells to the ISC compartment

55 years, range 4-72) with moderate-severe CGVHD following an allogeneic HCT for a primary hematologic malignancy. NIH consensus criteria (2015) were used for CGVHD grading. Patients were divided into 2 cohorts, a) those receiving an extended course of azithromycin (14 days) for CGVHD management (cohort 1, n=86) and b) those who did not (cohort 2, n=153). Patients in cohort 2 either did not receive any azithromycin (n=122) or had received an abbreviated (<14 day) course (n=31) of azithromycin post-HCT. For cohort 1, the median time to initiation of azithromycin therapy was 15 months post-HCT (range 3-68), with a 26 month median duration of azithromycin therapy (range 1-77). All patients in cohort 1 met NIH consensus criteria for BOS. Patients in cohort 2 did not exhibit BOS, but still met NIH criteria for moderatesevere CGVHD. Hematologic conditions included acute leukemia (n=139), MDS/MPD (n=44), malignant lymphomas (n=26), chronic leukemia (n=24), and multiple myeloma (n=6). Ninetyone patients exhibited moderate and 148 patients exhibited severe CGVHD. The two cohorts were balanced for CGVHD severity, with severe CGVHD noted in 69% and 63% of patients in cohorts 1 and 2 respectively. RESULTS: Decreased rates of relapse and improved survival were noted for patients treated with azithromycin for CGVHD. At 2 years, the cumulative incidence of relapse was 2% (95% CI:1-9%) for patients in cohort 1 vs 16% (95% CI: 11-23%) in cohort 2, p=0.001. Overall survival (2-year OS) in cohort 1 was 93% (95%CI: 88-99%) vs 78% (95% CI:72-85%) for cohort 2, p=0.003. Overall, 7 of 86 (8.1%) CGVHD patients treated with an extended course of azithromycin (cohort 1) have relapsed, the median time to relapse 876 days (range 379-1303) post-HCT. In comparison, 28 of 153 (18.3%) in cohort 2 have relapsed, a median 371 days (range 98-1252) post-HCT. CONCLUSION: The use of azithromycin for the management of moderate to severe CGVHD was not associated with an increased risk of relapse in patients undergoing HCT for a hematologic malignancy. Azithromycin therapy for patients with CGVHD should not be contra-indicated in this patient population.

Corticosteroids (CS) represent first-line treatment for gastrointestinal graft vs host disease (GI GVHD), and CS failure is associated with severe morbidity and mortality. While the immune system is the intended target of CS treatment, the glucocorticoid receptor (GR) is widely expressed, and there is limited understanding of the direct effects of CS on intestinal epithelium following immune-mediated damage. We thus investigated how CS treatment could impact intestinal homeostasis and regeneration following experimental bone marrow transplantation (BMT). In healthy C57BL/6 (B6) mice, in vivo administration of clinically relevant CS doses reduced Ki67 + epithelial proliferation in the ileum (p&lt;0.001; Fig. 1A) without inducing crypt loss or overt pathology. Given the numerous potential effects of systemic administration, we next utilized ex vivo small intestine (SI) organoid cultures to explore direct effects of CS on murine and human epithelium. Assessing a variety of clinically relevant CS agents, we found that methylprednisolone (MP), dexamethasone, and budesonide all decreased murine organoid size without affecting organoid number (p&lt;0.05; only MP shown; Fig. 1B). We also identified that GR-deficient (Nr3c1 -/-) organoids were significantly resistant to growth inhibition by MP (p&lt;0.05), indicating a direct GR-mediated effect of CS on intestinal epithelium leading to reduced growth. Furthermore, MP treatment significantly decreased the size of human organoids generated from primary duodenal tissue without affecting organoid numbers (p&lt;0.001). Organoid culture models were thus highly consistent with the findings from in vivo CS treatment. We next investigated CS effects on epithelial cells during immune-mediated damage. Pre-treatment of mice with 2 mg/kg MP x 7 days in vivo prior to crypt harvest and organoid culture increased organoid sensitivity to T-cell-mediating killing ex vivo (p&lt;0.05). Additionally, modeling steroid-refractory disease, GR-deficient (Nr3c1 -/-) T cells mediated greater killing of SI organoids if co-cultures were performed in the presence of MP (p&lt;0.01). We next investigated CS-mediated effects on epithelial damage in vivo, treating with MP x 7 days starting on day 7 after MHC-mismatched BMT, once GVHD had already been established. Vehicle-treated mice demonstrated GVHD-associated T cell activation, lymphocytic tissue infiltration, and ileal crypt loss compared to BM only controls, as well as increased height and Ki67 + cell frequency in residual crypts reflecting damage-induced regeneration (p&lt;0.001, Fig. 2A-C). Modeling steroid-refractory disease, systemic CS treatment failed to reduce T cell activation or lymphocytic infiltration. However, MP treatment appeared to attenuate regeneration and worsen intestinal pathology, as evidenced by exacerbated crypt loss in association with reduced crypt height and Ki67 + cell frequency (p&lt;0.01; Fig. 2A-C). Despite potential harmful side effects, CS are frequently necessary for treatment of clinical GVHD. We hypothesized that CS-mediated epithelial suppression could be mitigated by concurrent administration of agents capable of inducing tissue regeneration. Interleukin-(IL)-22 has been shown to promote epithelial proliferation and recovery following GI damage. We thus investigated whether IL-22 treatment could counterbalance CS-induced impairment of epithelial recovery in GVHD. Indeed, addition of IL-22 to MP-treated organoids promoted organoid growth without inducing toxicity/organoid loss in both murine and human SI organoid cultures (p&lt;0.001; Fig. 3A and B). Moreover, IL-22 administration in vivo with F-652, a clinical grade recombinant human IL-22 dimer, reversed MP-mediated crypt loss and reduction of crypt height and Ki67 + cell frequency in mice with GVHD (p&lt;0.001; Fig. 3C). In summary, these findings indicate that CS treatment can suppress epithelial proliferation in the intestines and exacerbate GI damage if it fails to control the pathologic immune response. However, deleterious CS side effects can be counterbalanced by promotion of epithelial regeneration, providing rationale for combining immunosuppression with tissue-supporting therapeutics such as IL-22 to optimize intestinal recovery in GVHD. Figure 1 Figure 1. Disclosures Blazar: Magenta Therapeutics: Membership on an entity's Board of Directors or advisory committees; BlueRock Therapeutics: Membership on an entity's Board of Directors or advisory committees, Research Funding; Rheos Medicines: Research Funding; Equilibre Pharmaceuticals Corp: Research Funding; Carisma Therapeutics, Inc: Research Funding; Tmunity Therapeutics: Other: Co-founder. Hanash: Evive Biotech: Ended employment in the past 24 months.

No abstract

Crypt base intestinal stem cells (ISCs) marked by Lgr5 and Olfm4 maintain the intestinal epithelium, and Paneth cells (PCs) provide an epithelial niche for ISCs in the small bowel. ISCs are reduced during gastrointestinal (GI) GVHD, but the precise mechanisms including the role of niche injury are unknown. Additionally, the specific effects of Interferon-γ (IFNγ) on intestinal epithelium in GVHD remain ill-defined. We evaluated kinetics of ISC loss by histology after allogeneic (allo) BMT using Lgr5-LacZ reporter mice. In both MHC- and miHA-mismatched models (LP&gt;B6 and B6&gt;BDF1), ISC numbers quickly recovered from pretransplant TBI conditioning in recipients of T-cell-depleted (TCD) BMT by day +10, but ISCs failed to recover in recipients of allo T cells. T-cell-induced ISC reduction was functionally validated by genetic marking of stem cell progeny and by culturing intestinal organoids from crypts isolated post-BMT. Similar to the kinetics of ISC loss, ISC-dependent organoid-forming capacity was impaired in recipients of allo T cells compared with TCD BMT recipients on day +10 (p&lt;0.05). Likewise, BMT into Olfm4 reporter mice showed significantly reduced lineage tracing from ISCs in recipients of allo T cells (Fig. 1). To better understand the potential for T cells to interact with the ISC compartment, we performed whole-mount 3-D microscopy to distinguish T cell localization within the intraepithelial and lamina propria compartments post-transplant. We found that donor T cells invading the small bowel after BMT were mostly in the crypt region, and infiltration within the lamina propria adjacent to the ISC compartment was much greater than invasion within the epithelium itself (Fig. 2). We next established an ex vivo co-culture system, to model interactions between intestinal epithelium and donor T cells and investigate mechanisms of T-cell-mediated ISC injury. Screening of effector pathways revealed no impact of perforin or FasL, but identified IFNγ as a principal mediator of ISC injury. Culture with allo T cells significantly reduced viable human and mouse intestinal organoid numbers, and this was inhibited by IFNγ neutralization. IFNγ receptor knockout (IFNγR-/-) organoids were resistant to T cells. IFNγ increased expression of Bak1 and decreased expression of Bcl2 in organoids, and induced ISC apoptosis defined by Annexin+Dapi-Lgr5+ phenotype. ISC killing was mediated by intraepithelial JAK/STAT signaling, as JAK1- and STAT1-deficient organoids were resistant, and it was inhibited by Ruxolitinib. Investigating the role of IFNγ in vivo, FACS analysis confirmed donor T cells to be the primary producers of IFNγ in crypt lamina propria, and BMT with IFNγ-/- donor T cells reduced crypt apoptosis, and preserved ISC frequencies. Moreover, BMT with recipients lacking IFNγR specifically in the intestinal epithelium significantly protected ISCs, reduced crypt apoptosis, and ameliorated GI GVHD pathology (Fig.3). Furthermore, ISCs were also protected by epithelial deletion of STAT1 and by Ruxolitinib treatment. As specific genetic manipulation of ISCs in vivo is not possible because genetic targeting of ISCs results in the same changes in their progeny, we utilized ex vivo models to determine if IFNγ kills ISCs by directly inducing their apoptosis or by damaging the PC niche. Manipulation of the niche by culturing wild-type ISCs with IFNγR-/- PCs was not protective to allo T cells. Using a niche-independent high-purity Lgr5+ ISC culture system based on combined GSK3β and HDAC inhibition, IFNγ directly induced cleaved-caspase-3+ ISC apoptosis and substantial ISC colony death, which were inhibited by Bak/Bax deficiency and by the pan-caspase inhibitor QVD. These results confirmed that IFNγ can directly induce ISC apoptosis independent of other cytotoxic effector molecules and independent of injury to the PC niche. In summary, T cells migrating to the GI tract primarily infiltrated the lamina propria adjacent to the ISC compartment, and T-cell-derived IFNγ directly targeted intestinal epithelium via JAK1/STAT1 signaling to induce ISC apoptosis in a PC-independent manner. ISC reduction and GI GVHD pathology were prevented by inhibiting the IFNγR/JAK1/STAT1 axis within the intestinal epithelium, indicating that in addition to their effects on T cells, JAK inhibitors may treat GVHD by inhibiting pathologic cytokine signaling within target organs and shielding them from allo T cells. Disclosures Hanash: Nexus Global Group LLC: Consultancy.

Crypt apoptosis and regeneration are characteristic findings in GI-GVHD. Intestinal stem cells (ISCs) are critical for maintaining the intestinal epithelium, but their frequencies are reduced in experimental GVHD, and the mechanisms driving crypt regeneration in this context are poorly understood. To better understand the impact of GVHD on individual epithelial components, we performed single cell RNA sequencing (scRNAseq) of purified small intestine crypts from B6 mice during homeostasis and five days after B6 into B6 syngeneic (syn) or B10.Br into B6 allogeneic (allo) BMT. Sequenced cells were first partitioned into distinct clusters using PhenoGraph, and the cluster identities were annotated based on their gene expression profiles. Unsupervised clustering indicated multiple subpopulations within the various recognized crypt components (Fig 1A). While secretory lineages clustered similarly across experimental conditions, there was highly distinct clustering among ISC populations and striking dissimilarity between allo and syn ISCs. Quantification by computing the subpopulation's phenotypic distance, a composite score integrating both the number and amplitude of differentially expressed genes, indicated that ISCs demonstrated the greatest transcriptional change in response to GVHD among all crypt lineages (Fig 1B). Gene Set Enrichment Analysis (GSEA) highlighted activation of the interferon-γ (IFNγ) pathway in allo ISCs, and differential gene expression indicated that STAT1 was their most highly upregulated transcription factor. We then performed MHC-mismatched allo-BMT into STAT1-floxed x villin-Cre recipients. Consistent with a role in IFNγ-related toxicity, STAT1-deficient recipients initially demonstrated reduced GVHD pathology, as well as a reduction in proliferating Ki67 + cells (Fig 2A). However, the pathology reduction in STAT1-deficient recipients was transient, while reduction in crypt proliferation persisted. STAT1-deficient recipients ultimately demonstrated increased mortality after allo-BMT, indicating a complex response to epithelial STAT1 signaling in GVHD. While IFNγ can induce epithelial apoptosis and kill organoid cultures, organoid exposure to IFNγ augmented size even at IFNγ concentrations that did not impair viability in a STAT1-dependent manner, and co-culture with allo T cells augmented organoid size (Fig 2B, C). Moreover, cell cycle analysis showed augmentation of cell cycle in ISCs after IFNγ treatment in association with upregulation of cyclin D1 (Fig 2D), and human organoids also showed increased size in response to IFNγ treatment, further suggesting that this growth promotion was not simply a secondary response to tissue injury. In addition to the IFNγ signaling, GSEA of allo ISCs indicated activation of the Myc pathway. scRNAseq data showed specific upregulation of Myc in allo ISCs (Fig 3A). Myc expression within individual ISCs indicated that greater STAT1 expression and IFNγ signaling directly correlated with Myc expression in the same ISCs, providing a potential direct link between T-cell-derived IFNγ and ISC-dependent regeneration (Fig 3B). Additionally, intestinal organoid qPCR showed that Myc expression was upregulated after IFNγ treatment, and scRNAseq of IFNγ-treated organoids indicated this Myc upregulation was restricted to the ISC/progenitor compartment. Although Myc is downstream of Wnt signaling, which is critical for ISC maintenance, expression of the representative Wnt target gene Axin2 was downregulated after IFNγ treatment, and Irf1, a representative IFNγ/STAT1 target gene, was upregulated, suggesting that IFNγ/STAT1 signaling could replace Wnt/β-catenin as a driver of ISC Myc expression. We next examined Myc function and found that treatment with the Myc inhibitor 10058-F4 suppressed IFNγ-dependent organoid growth (Fig 3C). Finally, immunofluorescent staining showed Myc protein expression in intestinal crypts after allo-BMT in a STAT1-dependent manner (Fig 3D), and Myc inhibitor treatment in vivo suppressed crypt epithelial proliferation in mice with GVHD. In summary, we found that epithelial STAT1 contributes to crypt regeneration in GVHD by transmitting T-cell-derived JAK/STAT cytokine signaling to activate Myc expression in ISCs. Clinical use of JAK inhibitors in GVHD may inhibit this regenerative response, necessitating concurrent interventions aimed at restoring it. Figure 1 Figure 1. Disclosures Hanash: Evive Biotech: Ended employment in the past 24 months.

No abstract