In vivo and in vitro modulation of the Eph/ephrin System as a key element in intestinal inflammation

BACKGROUND Eph (Erythropoietin-producing hepatocellular carcinoma) receptor tyrosine kinases and their cell bound ephrin ligands are a complex of proteins, characterized by bidirectional signalling. In the last years, their involvement is emerging in inflammation processes and immune responses: the Eph-ephrin system appears to play a role in cell adhesion-based responses, in the migration of immune cells and in T lymphocytes maturation and activation. Interestingly, the up-regulation of Eph-ephrin mRNA levels has been revealed in the mucosal lesions of colonic specimen from patients affected by Crohn’s disease (CD), an intestinal inflammatory condition whose etiology is still elusive but apparently involving an aberrant immune response to luminal pathogens. In fact, dysbiosis easily occurs in these patients and it seems related to the increased activity of thrombin that has been recently identified to control mucosal biofilm. AIM The aim of the present project was, first, to investigate the effects of the blockade of Eph/ephrin signalling on intestinal inflammation with a particular focus on the responses of the immune system. A Second aim was to get a deeper insight into the responses evoked by unidirectional activation of EphA/ephrin-A forward or reverse signalling in the same experimental conditions. Additional investigations were devoted to develop a murine colonoid culture to address in future studies the Eph/ephrin involvement in epithelial dysfunctions or to unravel the physiology of intestinal epithelial thrombin. A preliminary study was aimed at investigating whether exposure to high thrombin levels could alter microbiota behaviour. These two last experiments were carried out under the supervision of Dr. Nathalie Vergnolle, Director at Institut de Recherche en Santé Digestive, INSERM, Toulouse (France). METHODS In vivo study: acute colitis was induced in C57BL/6J mice by enema administration of 5mg/mouse TNBS. Mice were treated with: saline (10 ml/kg/die s.c), EphB4 (5-10-20 μg/kg/die s.c.), UniPR1331 (10-25 mg/kg/bid p.o.), EphA2 (20 μg/kg/die s.c), EphA2-Fc (30 μg/kg/die s.c.), ephrin-A1-Fc (16-50 μg/kg/die s.c.) or Sulfasalazine (50 mg/kg/die p.o.). DAI, MS, colon length and thickness, colonic IL-1β and colonic and pulmonary MPO were determined. Changes of splenic and mesenteric lymph nodes T-cells were studied through flow cytometry analysis. Colic expression levels of ephrin-B2, EphB4, ephrin-A1 and EphA2 mRNAs and their corresponding proteins were detected through reverse transcriptase-PCR and western-blot analysis, respectively. In vitro study-section 1: Splenic mononuclear cells obtained from healthy mice by percoll-gradient separation were cultured at 37°C in the absence or in the presence of the following compounds incubated for 4 hours: EphB4 (0.01-0.1 µg/ml), UniPR1331 (10-30 µM), EphA2 (0.25-1 µg/ml), ephrin-A1-Fc (0.25-1 µg/ml), EphA2-Fc (0.5-2 µg/ml), IgG1-Fc (0.3-0.6 µg/ml), Cyclosporine A (1µg/ml). Simultaneously, ionomycin (500 ng/mL) and PMA (50 ng/mL) were added to the culture media and TNFα levels were determined through ELISA kits. In vitro study-section 2: Crypts were extracted from healthy C57BL/6J mice and cultured for 6 days. After setting up a colonoid culture, EphB2 receptors were identified in growing organoids using immunocytofluorescence techniques. Microbiota study: Microbiota extracted from healthy human colon biopsies was exposed to increasing concentration of thrombin and put in contact with a Caco-2 monolayer in order to study possible bacterial adhesive changes. RESULTS AND CONCLUSION The colitis induced by the haptenating agent was characterised by the worsening of all the considered inflammatory parameters. Moreover, the analysis of ephrin-B2 gene transcripts revealed the presence of a second band together with the main isoform in TNBS-inflamed colons. Treatment with monomeric EphB4 was able to counteract dose-dependently the disease onset and the severity of the local and remote inflammatory responses evoked by TNBS colitis. The same beneficial profile was found by the blockade of the Eph/ephrin system through the administration of UniPR1331. Neither EphB4 nor UniPR1331 were able to prevent the generation of the alternative variant of ephrin-B2. Interestingly, both EphB4 and UniPR1331 were able to re-establish the CD4+/CD8+ ratio. Finally, UniPR1331 counteracted the TNF-α release by mononuclear cells, while EphB4 had an opposite effect, potentiating the production of the pro-inflammatory cytokine. Taken together these data point to identify a potential role of the EphB/ephrin-B system as a pharmacological target in intestinal inflammatory disorders and suggest that the therapeutic efficacy of its blockade apparently works through the modulation of immune responses. In contrast, the EphA/ephrin-A system does not seem to be as much involved. In fact, no correlation was found between colic inflammation and changes of ephrin-A1 or EphA2 mRNA expression in the colon. Moreover, the blockade of A-pathway through monomeric EphA2 did not change any inflammatory output, neither in vivo nor in vitro. In vivo the same irrelevant results were obtained upon EphA2-Fc treatment, thus showing that selective triggering of reverse A-signalling is not able to modulate TNBS-induced changes. On the other hand, ephrin-A1-Fc significantly reduced DAI, and mitigated colic and pulmonary MPO and MS. Moreover, a direct in vitro anti-inflammatory effect was revealed by the ability of ephrin-A1-Fc to reduce TNF-α levels released by mononuclear cells. On the whole, despite the apparently limited involvement of EphA/ephrin-A system in this murine IBD model, EphA forward signalling could possibly represent a concomitant, advantageous pathway to be activated by potential anti-inflammatory strategies. As regard the organoid culture, EphB2 has been identified as an important marker of organoid characterization being highly expressed on mature organoids. Finally, the microbiota study revealed that bacteria of thrombin‐treated biofilms attached more importantly to the Caco-2 monolayers compared to the untreated biofilm, opening the way for a novel strategy of targeting for IBD.