Rifaximin toxicity on hepatocellular fatty degeneration via activation of human pregnane X receptor

Rifaximin is a semi-antibiotics derived from rifamycin, which is minimally absorbed in the circulation and excreted into feces with pro-drug form. It has been approved by FDA on clinical therapy of traveler's diarrhea (2004) and hepatic hepatophelothapy (2010), in addition to being processed in the phase III clinical trial of remedy of irritable bowel disease. Previous report demonstrates rifaximin is gut-specific agonist of human pregnone X receptor (PXR), and short-term administration of rifaximin exerts the protective and therapeutic role on inflammatory bowel disease via cross-inhibition of NF-kB pathway through human PXR activation. PXR is a ligand-activated transcription factor important for its induction of drug transport and metabolism, and mediate endobiotic balance in hepatic steatosis, lipid homeostasis, bile acid homeostasis and steroid homeostasis. Although rifaximin is well known as well-tolerant drug without drug-induced hepatotoxicity addressed hitherto, the effect of long-term administration of rifaximin and subsequent function on PXR-involved body metabolism and physiology is still underdetermined. In the current study, human PXR inserted mice which lack mouse PXR was treated with control and rifaximin for 1 week, 1 month, 3 month till 6 month respectively by food intake, coupling with wild type and Pxr-null mice treated with rifaximin for 6 month as parallel cohorts. Histological analysis revealed time-dependent intense hepatocellular fatty degeneration in hPXR mice, while not in WT and Pxr-null mice. Hepatic triglyceride level was massively increased when rifaximin treated for 3 and 6 month in hPXR mice concomitant with slight increase of cholesterol and decreased free fatty acid level in liver. hPXR target genes including phase I, II, III enzyme was induced in small intestine and increased in liver, coupling with significant up-regulation of hepatic genes related to triglyceride synthesis and lipid accumulation. However, metabolomic analysis of serum and liver revealed no accumulation of rifaximin in liver even after 6 month treatment, though several nucleotide phosphates (AMP, ADP, ATP etc.) were up-regulated upon occurrence of fatty liver. In contrast, microarray and real-time PCR analysis of small intestine show the massive increase of genes involved fatty acid uptake and triglyceride, cholesterol transportation, including CD36, Dgat2 and ApoA4 etc. Lipid contents measurement of intestinal epithelial cell revealed the increase of triglyceride, fatty acid and cholesterol in hPXR mice while not in WT and Pxr-null mice. Thus, our findings suggested that long-term administration of rifaximin leads to hepatocellular fatty degeneration via hPXR-involved activation of lipid metabolism in small intestine and subsequent effect on lipid accumulation in liver, thus suggesting that human PXR is a key factor on long-term clinical therapy of rifaximin.