Hepatic drug metabolism is one of the mechanisms whereby organisms protect themselves from exposure to environmental chemicals and either activates or deactivates [34
] therapeutic drugs. CYP2C9
is an important drug-metabolizing enzyme which metabolizes ~16% of all clinical drugs as well as some endogenous compounds. CYP2C9
is up regulated by drugs and other xenobiotics primarily via two xenobiotic sensing receptors, CAR (NR1I3) and PXR (NR1I2). PXR is one of the most important xenobiotic sensing receptors since it is promiscuous, interacting with a wide array of ligands including clinical drugs such as rifampicin, paclitaxol, and dietary supplements such as St. John’s Wort. PXR regulates the response of a variety of CYP enzymes, drug transporters, and phase II drug metabolizing enzymes in response to environmental stimuli. Many of the CYP enzymes and phase II enzymes are found in high concentrations in liver, where HNF4α, a hepatic enriched factor, regulates both their constitutive expression and enhances their response to the xenobiotic sensing receptors CAR and PXR [41
]. In fact, for a number of these enzymes such as CYP2C9 and CYP3A4, the receptors HNF4α and CAR have been reported to synergistically activate their promoter activity in cell lines, and studies in primary hepatocytes [34
] and HNF4α knockout mice have confirmed that HNF4α plays a central role in regulating the response of drug metabolizing enzymes to PXR and CAR [41
It has become apparent that cross talk with other transcription factors, coactivators and growth signals is necessary to explain the interplay between different receptors [45
]. In these and recent studies [37
], we performed yeast two-hybrid screens to identify receptors/and or co activators which might explain the interplay between these receptors. One of the co activators identified in a yeast two-hybrid screen using HNF4α as bait was the co activator NCOA6 which was subsequently shown to interact with the receptor CAR using protein-protein interaction techniques principally through the first LXXLL motif with some interaction through the second LXXLL motif [37
]. The present study shows that the xenoreceptor PXR also interacts with NCOA6 through the first LXXLL motif. Mammalian two hybrid assays further confirm that NCOA6 interacts with PXR and HNF4α principally through its first LXXLL motif although a smaller interaction is seen with the second LXXLL motif.
Exogenous NCOA6 has little effect on CYP2C9 or CYP3A4 promoter activity in the presence or absence of PXR, but increases HNF4α activation and the synergistic activation by PXR and HNF4α particularly in the presence of the PXR ligand rifampicin. Silencing NCOA6 decreased the synergistic activation of the CYP2C9 promoter by PXR and HNF4α, but surprisingly did not affect that of the CYP3A4 promoter. Parallel effects were seen on CYP2C9 and CYP3A4 mRNA. shNCOA6 preferentially affected the synergistic increase in CYP2C9 mRNA by HNF4α and PXR but had no effect on CYP3A4 mRNA induction.
There are many examples of crosstalk between CAR or PXR and HNF4α. Cross-talk between PXR and HNF4α has been reported for the human sulfotransferase SULT2A1 whereby this enzyme is up regulated by HNF4α and down regulated by PXR [47
]. Rifampicin induction of CYP3A4
has been proposed to involve cross talk with HNF4α and PXR via various co activators such as PGC-1α, SRC-1, and the small heterodimer partner (SHP) may compete with HNF4α and SRC-1 for PXR [48
]. The observed differential regulation of CYP2C9
could be due the fact that at the promoter level, the constructs of CYP2C9
are of similar length and the distance between the PXR binding site and the HNF4a binding site is about 1.6 kbp, which could be a possible explanation for the synergistic activation of the promoter constructs. But at the chromatin level, though CYP2C9
is about 1.6 kbp, the distance between the binding sites is about 8 kbp for CYP3A4
. This long distance between the binding sites could prevent the proper loop formation required for the synergistic activation. Our data with mRNA and silencing of NCOA6 strongly supports this hypothesis. Negishi and coworkers [45
] reported that the signal molecule EGR1 (early growth factor) binds to the proximal promoter of CYP2B6
and coordinates interaction with a nearby HNF4α site which cross-talks with a CAR binding at a distal enhancer site causing synergistic induction by drugs. A later paper from the same group provided evidence that EGR1 is involved in a loop formation between the distal enhancer and the proximal sites [46
]. Similarly, our studies are consistent with the hypothesis that NCOA6 may assist in loop formation between the distal PXR sites and the proximal HNF4α sites of the CYP2C9
promoter as depicted in the graphical abstract.