Activation of PPAR-gamma signaling as a novel target to limit nfkb-dependet in flammation in cystic fibrosis biliar epithelium

Cystic fibrosis-associated liver disease (CFLD) is a chronic cholangiopathy that negatively impacts the quality of life and survival of patients with Cystic Fibrosis (CF). CF is a disease of secretory epithelia caused by genetically-determined defective function of CFTR (cystic fibrosis conductance regulator), a cAMP-activated Cl- channel that, in the liver, is uniquely expressed in the biliary epithelium. The pathogenesis of CFLD is thought to be related to the ductal cholestasis caused by the defective channel function of CFTR. However, a number of evidence indicate that, in association with the genetic defect, other pathogenetic factors concur to determine the liver phenotype. During the first year of my doctorate we have demonstrated that in CFTR-defective biliary epithelium, exposed to bacterial-derived endotoxins, TLR4/NF-kB-dependent innate immune responses are increased, indicating a loss of “endotoxin tolerance”. The current standard of care for CFLD is limited to the administration of choleretic agents. Based on our novel interpretation of the pathogenesis of CFLD, a rational therapeutic approach to the management of CFLD should be to target the exaggerated inflammatory response of the biliary epithelium. Nuclear receptors (NRs), ligand-activated transcription factors that regulate several intracellular metabolic functions by controlling the expression of specific target genes, are now emerging as important regulators of inflammation in several conditions. In this work we sought to investigate the ability of NRs to modulate the TLR4/NF-kB-dependent innate immune responses in CF biliary epithelia. Our data indicate that cholangiocytes express several subtypes of NRs, including Peroxisome Proliferator Activated Receptor (PPAR) subgroup α, γ and β/δ, Liver X Receptors (LXR) subgroup β, Farnesoid X Receptor (FXR) and Vitamin D Receptor (VDR). Among these NRs, PPAR-γ was barely expressed in wild type (WT) cholangiocytes , but its expression, both at the gene and protein levels, was strongly and significantly more expressed in CF cells. In spite of its increased expression in the nucleus, PPAR-γ was not transcriptionally active in CF cholangiocytes; in fact the basal level of expression of PPAR-γ target genes (Acaa1b, Angptl4 and Hmgcs2) was similar to WT cholangiocytes. On the other hand, treatment of CF cholangiocytes with the PPAR-γ agonist pioglitazone significantly increased the expression of its target genes, indicating that the transcriptional activity of nuclear PPAR-γ could be readily activated in the presence of proper agonists. Consistent with the higher expression level of the receptor, the increase in PPAR-γ target genes stimulated by pioglitazone was higher in CF than in WT cholangiocytes. The observation that PPAR-γ, while inactive at baseline levels, was able to respond to exogenous stimulation, indicates that a defect in the availability of endogenous PPAR-γ activators might impair PPAR-γ function in CF and increase the receptor expression as counter-regulatory mechanism. Thus, to study the availability of endogenous PPAR-γ activators in CF, we performed lipidomic analysis of extracts from cultured cholangiocytes, and quantified the major ω-3 and ω-6 polyunsaturated fatty acids (PUFAs), i.e. the precursors of the active form of PPAR-γ ligands. We detected an increased ratio between arachidonic acid and docosahexaenoic acid (AA/DHA) and between arachidonic acid and linoleic acid (AA/LA) in CF cells compared to their controls. LA, the precursor of AA, is the main source of pro-inflammatory mediators, while DHA is an important precursor of anti-inflammatory eicosanoids that are able to activate PPAR-γ. This finding might explain, at least in part, the increased receptor expression in CF biliary cells. Having shown that CF cells express higher amounts of PPAR-γ, that can be activated by exogenous ligands, but it is not transcriptionally active because of the lack of endogenous ligands, we next tested if exogenous activation of PPAR-γ by synthetic ligands, might decrease the TLR4/NF-kB-driven inflammation in CFTR-defective biliary epithelium. We found that in Cftr-KO cells, in the presence of pioglitazone, nuclear translocation of p65/NF-kB, as well as its transcriptional activity, were significantly inhibited, both before and after stimulation with bacterial LPS. Moreover, the gene expression and protein secretion of several NF-kB-dependent pro-inflammatory cytokines, such as LIX (LPS-induced CXC chemokine), MCP-1 (monocyte chemotactic protein-1), MIP-2 (macrophage inflammatory protein 2), G-CSF (granulocyte colony-stimulating factor) and KC (keratinocyte chemo-attractant) were significantly inhibited both at baseline and after LPS stimulation. The anti-inflammatory effect of pioglitazone was the result of a direct activation of PPAR-γ receptor, as shown by reversal of pioglitazone-induced protection in the presence of the PPAR-γ antagonist GW9662. Finally, to understand the mechanism by which ligand-activated PPAR-γ blocks inflammation in CF cholangiocytes we analyzed the effect of pioglitazone on NF-kB activation pathway steps, and found that pioglitazone up-regulated IkB-α, a negative regulator of NF-kB. In conclusion, our results indicate that a decreased PPAR-γ function caused by a decreased availability of PPAR-γ endogenous agonists might be responsible for the chronic inflammatory state of CFTR-defective cholangiocytes. Stimulation of PPAR-γ signaling by the agonist pioglitazone is able to decrease NF-kB-dependent inflammatory responses by increasing IkB-α, thus may represent a novel strategy to limit inflammation in CFLD. Studies in mice models of liver damage in CFLD have been planned and may represent a proof of concept for anti-inflammatory treatment in CF.