Society: AASLD

Aims: The mechanism of gut microbiota involved in the etiopathogenesis of primary cholangitis (PBC) remains unclear. Here, we investigated the role of gut microbiota in liver immune response in PBC.
Methods: Female C57BL6/J mice (n=8 per group) were gavaged with placebo, feces from healthy controls or PBC patients after antibiotic cocktails. Alpha-naphthylisothiocyanate (ANIT) was given to induce cholestasis. The liver, mesenteric lymph node (MLN) and spleen of mice were cultured under anaerobic and aerobic conditions. Serum, intestinal and liver tissues were isolated for ELISA, histology, real-time PCR and flow cytometry analysis. Fecal and liver biopsies from PBC patients (n = 14) and healthy controls (n = 5) were collected to confirm the distribution of Klebsiella pneumonia and T helper 17 (Th17) infiltration. The liver biopsies from early and advanced PBC patients (n = 13) were isolated for RNA-seq. Immortalized Human Intrahepatic Biliary Epithelial Cells (HIBEpiCs) were used for investigation the effect of K. pneumonia on the chemotaxis of Th17.
Results: More severe liver injuries and Th17 cell infiltration were found in mice gavaged by feces from PBC patients than control. Denser Th17 cells was observed in liver of PBC group after ANIT treatment (P < 0.05). K. pneumonia was isolated from the liver, spleen and MLN of the PBC group mice. Interestingly, administration of feces from PBC patients increased serum endotoxin and impairment of the intestinal epithelial barrier (P < 0.05). Antibiotic treatment significantly decreased Th17 infiltration, intestinal epithelial barrier and liver injury induced by PBC-derived microbiota. Besides, K. pneumoniae monocolonization can induce K. pneumoniae translocation and inflammatory cell infiltration in liver. Furthermore, PBC patients showed increased of CCL20 expression and infiltration Th17 cells in liver, which was positively correlated with disease progression. Notably, the abundance of K. pneumoniae was increased in the liver and fecal of PBC patients. In vitro, the bacterial lysate of K. pneumoniae upregulated cholangiocyte senescence and CCL20 in HIBEpiCs in a concentration-dependent manner.
Conclusions: We demonstrate that K. pneumoniae translocation to liver increased CCL20 secretion promote Th17 immune response in PBC.

Background: Primary biliary cholangitis (PBC) is a cholestatic liver disease that targets cholangiocytes and is characterized by enhanced biliary senescence, ductular reaction (DR), autoimmune response, liver inflammation and fibrosis that is triggered by the biliary release of senescence-associated secretory phenotype (e.g., TGFβ1). Heterozygous 162 nt AU-rich element (ARE) region deletion in the 3’ untranslated region of the interferon-gamma gene (ARE-Del^+/-) in mice at 20 wk of age display distinct pathological features that are characteristic of human PBC. We have previously demonstrated that melatonin therapy ameliorates damage in mouse models of primary sclerosing cholangitis via decreased expression of the MT1 melatonin receptor and selected circadian genes including CLOCK. Thus, we aimed to evaluate the therapeutic effects of melatonin in a mouse model of PBC.
Methods: We measured biliary senescence (by SA-β-gal staining), and the biliary immunoreactivity of MT1/CK19 and CLOCK/CK19 by double immunofluorescence (IF) in liver sections from healthy controls (2 female and 2 male) and late-stage PBC patients (1 male and 3 female). In vivo, male, and female wild-type (WT) and ARE-Del^+/- mice at 20 wk of age had free access to drinking water containing melatonin (0.03%) for 1 wk (daily intake 1.2 mg). We evaluated (i) liver damage by H&E staining; (ii) DR by immunohistochemistry (IHC) for CK19; (iii) biliary senescence by double IF for p16/CK19 and SA-β-gal staining; (iv) liver inflammation by IHC for CD68; and (v) liver fibrosis by Sirius Red staining in liver sections from WT and ARE-Del^+/- mice treated with/without melatonin. Liver autoimmune response was evaluated in mouse liver sections by IHC for CD3⁺ T cell and CD20⁺ B cell infiltration. The biliary immunoreactivity of MT1 and CLOCK was measured by IF in liver sections co-stained with CK19.
Results: Enhanced biliary senescence as well as increased immunoreactivity of MT1 and CLOCK were detected in liver sections from late-stage PBC patients compared to healthy controls. There was enhanced expression of MT1 and CLOCK and increased liver damage, DR, and biliary senescence as well as autoimmune response, liver inflammation and liver fibrosis in ARE-Del^+/- compared to WT mice, phenotypes that were decreased in melatonin-treated ARE-Del^+/- mice compared to vehicle-treated mice.
Conclusion: Melatonin administration to ARE-Del^+/- mice ameliorates PBC phenotypes through reduced MT1 and CLOCK expression. Restoration of melatonin-dependent signaling via melatonin treatment may be therapeutic for patients with PBC.

Background: PSC is devasting cholestatic liver disease without curative treatment. In PSC patients and cholestatic mouse models, MC presence and activation significantly increases. MC injection induces PSC phenotypes increasing ductular reaction (DR), biliary senescence, and hepatic fibrosis in wild-type (WT) or MC-deficient mice. NFkB promotes inflammation, and, in MCs, IL-33 induces NFkB translocation causing MC activation and subsequent inflammatory cytokine release. In rodents, inhibition of NFkB by caffeic acid phenethyl ester (CAPE) decreases cholestatic phenotypes. Aim: To determine the effects of inhibition of MC-specific NFkB on PSC phenotypes. Methods: 12 wk male HDC^-/- mice (balb/c (WT)) were injected via tail vein with cultured mouse MCs (mMCs) pretreated with vehicle or CAPE (20 mM) 1x prior to sacrifice 3 days later. Serum, liver, and cholangiocyte supernatants were collected. Liver damage was assessed by serum AST. Serum HA content and serum/cholangiocyte IL-33 secretion was measured by EIA. Downstream of IL-33, we measured IRAK4 gene expression in total liver by qPCR. DR and senescence were determined by semiquantitative CK-19 immunohistochemistry and qPCR for p21 in total liver mRNA, respectively. Hepatic fibrosis and hepatic stellate cell (HSC) activation were evaluated by semiquantitative Fast Green/Sirius Red staining, immunofluorescence for desmin (co-stained with CK-19) and qPCR for collagen-1a and a-SMA in total liver. mMCs were evaluated for NFkB signaling components, p105 and p65 (RelA) by immunostaining and immunofluorescence, respectively. 3D human organoids were generated from normal primary isolated human cholangiocytes, HSCs, and human MCs (hMCs) pretreated with vehicle or CAPE (20 mM); organoids developed for 72 hrs. HA content was measured by EIA and chymase (Cma1) gene expression by qPCR. Results: mMCs express p105 and RelA and injection of vehicle-mMCs into HDC^-/- mice increased AST, senescence, and fibrosis compared to WT or HDC^-/- mice. These parameters were reduced in HDC^-/- mice injected with mMC-CAPE. No significant changes were found in DR across all groups. Serum HA and IL-33 increased in HDC^-/- mice treated with vehicle-mMCs that decreased in HDC^-/- mice injected with mMC-CAPE. Biliary IL-33 increased in vehicle-mMC injected HDC^-/- mice, whereas inhibition of mMC-NFkB reduced IL-33/IRAK signaling. Normal 3D human organoids generated with vehicle hMCs had increased HA content and chymase gene expression that were reduced when hMCs were treated with CAPE prior to organoid formation. Conclusion: Injected mMCs (i) promote hepatic damage resembling PSC phenotypes and (ii) increase hepatic IL-33 signaling in HDC^-/- mice. Inhibition of MC-NFkB signaling reduces damaging phenotypes by decreasing biliary and hepatic IL-33 signaling. Modulation of inflammatory MC mediators may be key to reduce PSC phenotypes.

Background: Primary sclerosing cholangitis (PSC) targets bile ducts but is complicated by vascular dysfunction. Portal angiogenesis occurs in PSC, but the pathological role is undefined. Endothelin (ET) is a vasopeptide found in 3 isoforms (ET-1, 2 and 3), and, in cholestatic rats, ET-1 expression is increased in cholangiocytes. ET-1 and ET-2 signal via ET-A, and we previously found that ET-A inhibition reduces ductular reaction (DR), angiogenesis and fibrosis in the multidrug resistance 2 knockout (Mdr2^-/-) mouse model of PSC. We aimed to delineate the mechanism by which ET-A regulates biliary damage and vascular dysfunction.
Methods: We evaluated ET-1, ET-2 and ET-A expression in female and male human controls and late-stage PSC samples, and male 12-wk old wild-type (WT) and Mdr2^-/- mouse liver sections, bile and isolated cholangiocytes. Additional WT and Mdr2^-/-mice were treated with Ambrisentan (ET-A antagonist, 5 mg/kg/d for 7d). Portal vasculature was evaluated by CD31 and vWF staining, and transmission electron microscopy. Angiopoietin-1 (Angpt-1, vessel stabilizing) and Angpt-2 (angiogenic) mRNA levels were measured by qPCR. Angiogenesis (CD31) was correlated with DR, immune cell infiltration and fibrosis by Pearson correlation. Cholangiocyte supernatants were run on an angiogenesis array and transforming growth factor (TGF)-β1 EIA, isolated cholangiocytes were run on angiogenic factor PCR array and select factors were confirmed by staining. ET-A connection with select angiogenic factors was verified by Ingenuity Pathway Analysis (IPA). In vitro, human control (n=3) and late-stage (n=3) male and female PSC primary cholangiocytes were treated with Ambrisentan (470 nM) or control for 24 hrs; supernatants were collected and placed on human umbilical vein endothelial cells (HUVECs) for 24 hrs prior to evaluating migration.
Results: Mdr2^-/- mice had increased expression of ET-1, ET-2 and ET-A, whereas only ET-1/ET-A was enhanced in human PSC. Mdr2^-/- mice had portal angiogenesis with dilated vessels that lacked primary cilia, indicating damage, but these parameters were reversed by Ambrisentan. Angpt1 was unchanged, and Angpt2 increased, in Mdr2^-/- mice, and Ambrisentan increased Angpt1 but reduced Angpt2. Angiogenesis positively correlated with DR, immune cell infiltration and fibrosis in Mdr2^-/- mice. Mdr2^-/- mice had increased biliary secretion of IL-6 and TGF-β1, and increased expression of midkine (Mdk); these angiogenic factors were decreased by Ambrisentan. IPA linked ET-A activation to IL-6, TGF-β1 and Mdk. In vitro, PSC cholangiocyte supernatants induced HUVEC migration, which was blocked by Ambrisentan.
Conclusion: ET-1/ET-A may drive DR and angiogenesis in PSC, and portal angiogenesis (induced by biliary-derived angiocrine factors) may be pathological in PSC. Targeting ET-A may be therapeutic for PSC patients with vascular dysfunction.

Background: Gonadotropin-releasing hormone (GnRH) is synthesized from neurons within the hypothalamus as well as peripheral systems such as the biliary epithelium. Biliary secretion of GnRH is increased in cholestatic models such as bile duct ligation (BDL) and Mdr2^-/- (model of primary sclerosing cholangitis, PSC). We have previously shown that: (i) GnRH triggers biliary damage in normal and cholestatic animals by paracrine/autocrine GnRHR1-mediated signaling; and (ii) knockdown of GnRH ameliorates PSC phenotypes in Mdr2^-/- mice through downregulation of miR-200b. Aim: Evaluate the role of Cetrotide® (Cetrorelix acetate, FDA approved GnRHR1 inhibitor used for the inhibition of premature luteum hormone surges in women undergoing controlled ovarian stimulation) in the phenotypes of primary biliary cholangitis (PBC). Methods: In in vivo experiments we used 2 different PBC mouse models: (i) dominant-negative TGF-β receptor II (dnTGFβRII 12 wk of age, early stage, and 32 wk of age, late stage); and (ii) a murine model of autoimmune cholangitis (24 wk of age, late stage PBC, named ARE-Del^-/-), generated by altering the AU-rich element by deletion of the interferon gamma 3' untranslated region. At the selected age, male mice (n=2 per group) were treated with Cetrorelix acetate (10 mg/kg WT per day) by IP implanted minipumps for 2 wk. Thereafter, we collected serum, liver tissue and isolated cholangiocytes. We evaluated: (i) by immunofluorescence (IF)/IHC the hepatic localization of GnRHR1 in mouse and human (health controls and late-stage PBC), (n=2) each PBC and healthy controls, liver sections co-stained with markers of cholangiocytes (CK19), hepatic stellate cells (HSCs, desmin), hepatocytes (HNF4a) and macrophages (CD68); and (ii) GnRHR1 mRNA expression in mouse and human total liver by qPCR; and (iii) GnRH levels in human serum (n=8) healthy controls and (n=40) PBC. We also measured liver damage by H&E staining and ductular reaction (DR) by IHC for CK19 in liver sections. Results: We observed enhanced liver damage, DR, in PBC mouse models compared to their respective controls, phenotypes that were decreased in PBC mouse models treated with Cetrorelix. We demonstrated: (i) the presence of GnRHR1 in mouse and human cholangiocytes, HSCs, hepatocytes and macrophages, whose expression increased in PBC mouse models and human late-stage PBC samples; and (ii) increased GnRH levels in human serum. GnRHR1 immunoreactivity and mRNA expression decreased in mice treated with Cetrorelix compared to controls. Summary/Conclusion: We demonstrated enhanced expression of the GnRH/GnRHR1 axis in mouse PBC models and late-stage PBC samples. Inhibition of GnRHR1 with Cetrorelix may be an important target for the management of PBC phenotypes.

Lindsey Kennedy

Appearances