668 - Annexin A1-mediated therapeutic effect of extracellular vesicles derived from thrombin-preconditioned WJ-MSC in experimental necrotizing enterocolitis

Publication Number: 1645.668

Choleun Kwon, Sungkyunkwan University, SEOUL, Seoul-t'ukpyolsi, Republic of Korea; Sein Hwang, Samsung Medical Center, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Young Eun Kim, Samsung medical center, Gangnam-gu, Seoul-t'ukpyolsi, Republic of Korea; Misun yang, samsung medical center, gangnam, Seoul-t'ukpyolsi, Republic of Korea; So Yoon Ahn, Sungkyunkwan university, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Se In sung, Sungkyunkwan university, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Yun Sil Chang, Samsung Medical Center, Seoul, Seoul-t'ukpyolsi, Republic of Korea

Background: Necrotizing enterocolitis (NEC) is a life-threatening intestinal inflammatory disease primarily affecting premature infants. We previously demonstrated that extracellular vesicles derived from thrombin-preconditioned Wharton's Jelly mesenchymal stem cells (thEVs) show enhanced therapeutic efficacy compared to naïve EVs in NEC models.

Objective: We hypothesized that specific factors increased in thEVs mediate anti-inflammatory and regeneration effect. This study aimed to identify upregulated key factors of thEVs in NEC models.

Design/Methods: Proteomics: LC-MS/MS was performed to compare profiles of thEVs and naïve EVs. In vivo: NEC was induced in postnatal day 4 (P4) ICR mice with formula feeding, oral LPS, hypoxia, and hypothermia, and were intraperitoneally treated with saline (NEC) or thEV (NEC-thEV). At P7, pups were sacrificed for small intestinal RNA sequencing (n = 3/group). In vitro: LPS-induced peritoneal macrophages/IEC6 co-culture was treated with PBS, thEVs, or Annexin A1 (ANXA1) mimetic peptide Ac2-26.

Results: In proteomic analysis, ANXA1 was 3.8-fold greater in thEVs than naïve EVs. In vivo, ANXA1-FPR2 was the most prominent interaction observed in NEC-thEV. While the changes were not statistically significant, thEV treatment showed a trend toward reduced pro-inflammatory cytokines (TNF, IL-1β), increased anti-inflammatory Tgf-β, and restored β-catenin levels. In vitro, thEVs and Ac2-26 significantly reduced LPS-induced pro-inflammatory cytokines (IL-1α, IL-1β, IL-6) and restored proliferation marker Ki67 (p < 0.05), with a trend toward increased β-catenin levels.

Results: In proteomic analysis, ANXA1 was 3.8-fold greater in thEV than naïve EVs. In vivo, ANXA1-FPR2 was the most prominent interaction observed in NEC-thEV. In vitro, thEV and Ac2-26 significantly reduced LPS-induced pro-inflammatory cytokines (IL-1α, IL-1β, IL-6) and restored proliferation marker Ki67 (p < 0.05), with a trend toward increased β-catenin levels. Crucially, EVs from ANXA1-knockdown MSCs showed a markedly diminished therapeutic effect, confirming that ANXA1 is a key mediator of the protective functions of thEV.

Conclusion(s): ANXA1-mediated anti-inflammatory and regenerative effects implied in vivo were confirmed in vitro. These suggest a crosstalk between Annexin A1-FPR2 and Wnt/β-catenin pathway, supporting the pleiotropic role of Annexin A1 in NEC.

Figure 1: Multi-omics analysis revealed that ANAX1-FPR2 was the prominent therapeutic axis in NEC.
The therapeutic effects of thMSC-EVs on necrotizing enterocolitis (NEC) were evaluated in an in vivo mouse model.
(A) Proteomic analysis using LC-MS/MS identified 64 ligands by comparing thEV and Naive EV
(B) Histological analysis using H&E stain and a TUNEL assay revealed that thMSC-EVs significantly attenuated intestinal necrosis and apoptosis, particularly in the intestinal crypts. The histological damage score, which was elevated in the NEC group, was significantly reduced in the thMSC-EV-treated group.
(C) A PCA plot shows that samples from the CON, NEC, and NEC+thEV groups form distinct clusters based on their gene expression profiles.
(D) A volcano plot shows the differentially expressed genes (DEGs) in the NEC group compared to the CON group, and in the NEC+thEV group compared to the NEC group. Transcriptomic analysis showed that thMSC-EVs fundamentally altered the intestinal immune environment by suppressing the heightened mucosal immune response and promoting tissue repair pathways.
(E) A bubble plot shows the enrichment of biological processes in each experimental group. The NEC group showed a high enrichment in inflammatory response pathways.
(F) GO Module scoring shows that NEC induce inflammatory pathway and treatment of EV can relieve them.
(G) A multi-omics ligand-receptor analysis was performed, integrating proteomic data from the EVs and transcriptomic data from the intestinal tissue. This analysis identified the interaction between ANXA1 from thMSC-EVs and its receptor, Formyl Peptide Receptor 2 (FPR2), on intestinal cells as the most prominent therapeutic axis.

Figure 2: ANXA1 Functionally Mediates the Cytoprotective and Anti-inflammatory Effects of thMSC-EVs in In Vitro NEC Models
Functional validation confirms that ANXA1 is a key mediator of the therapeutic effects of thMSC-EVs in in vitro NEC models.
(A) Ki67 staining shows NEC diminished Ki67 positive IEC6 cells. thEV and Ac2-26 can recover them. A bar graph represents the CCK-8 cell viability assay results for IEC-6 cells treated with conditioned medium (CM) from LPS-induced peritoneal macrophages. Each group had n=16. A bar graph also shows the levels of inflammatory cytokines in LPS-induced peritoneal macrophages. In this model, co-treatment with Ac2-26 markedly reduced the secretion of pro-inflammatory cytokines, replicating the potent anti-inflammatory effect of thMSC-EVs. Each cytokine group had n=3. Statistical significance was defined as p  < 0.01 vs. CON (*), p  < 0.01 vs. LPS (#), and p  < 0.01 vs. LPS + thEV ($). All data are presented as the mean ± SEM and were analyzed using one-way ANOVA and the post-hoc Tukey test.
(B) ANXA1 was shown to perform a dual therapeutic mechanism by modulating key signaling pathways. It suppressed the inflammatory TLR4/NF-kB pathway while simultaneously activating the regenerative Wnt/β-catenin pathway, which is essential for intestinal epithelial repair. Each group had n=3. Statistical significance was defined as p  < 0.01 vs. CON (*), p  < 0.01 vs. LPS (#), and p  < 0.01 vs. LPS + thEV ($). All data are presented as the mean ± SEM and were analyzed using one-way ANOVA and the post-hoc Tukey test.
(C) A bar graph shows the results from a CCK-8 cell viability assay, TUNEL assay, and a Western blot for ZO-1 protein expression in IEC-6 cells. In the hyperosmotic, ischemic, and hypothermic (HIT) stress model, an ANXA1 mimetic peptide (Ac2-26) replicated the effects of thMSC-EVs by restoring cell viability and the expression of the key barrier integrity protein ZO-1. The CCK-8 assay had n=16 per group, the TUNEL assay had n=10, and the Western blot analysis had n=3. Statistical significance was defined as p  < 0.01 vs. CON (*), p  < 0.01 vs. HIT (#), and p  < 0.01 vs. HIT + thEV ($).

Figure 3: Knockdown of ANXA1 Abrogates the Therapeutic Efficacy of MSC-EVs
(A) The figure shows that the ANXA1 knockdown in MSCs was confirmed using ELISA, PCR, and Western blot analyses.
(B) EVs derived from ANXA1-knockdown MSCs (siANXA1-MSC-EVs) were tested in both HIT and inflammatory models. The therapeutic effects of siANXA1-MSC-EVs were substantially diminished compared to control EVs. Specifically, siANXA1-MSC-EVs failed to effectively restore cell viability and their ability to suppress cytokine secretion by LPS was greatly reduced. This confirms that ANXA1 is a functionally indispensable key factor for the enhanced therapeutic efficacy of thMSC-EVs. _x000B_