Original article
Use of dominant negative nuclear receptors to study xenobiotic-inducible gene expression in primary cultured hepatocytes

https://doi.org/10.1016/S1056-8719(03)00002-9Get rights and content

Abstract

Introduction: To determine the feasibility of using dominant negative nuclear receptors to dissect the regulation of inducible gene expression in primary cultured hepatocytes, a series of dominant negative nuclear receptor expression plasmids were designed with truncated AF-2 subdomains. Methods: Plasmids expressing dominant negative or wild-type constitutive androstane receptor (CAR), pregnane X receptor (PXR), farnesoid X receptor (FXR), liver X receptor (LXR), or peroxisome proliferator-activated receptor α (PPARα) were transiently cotransfected into primary cultured rat hepatocytes, together with an appropriate reporter plasmid. Results: Treatment with prototypic inducers, 10−4 M phenobarbital (CAR activator), 10−5 M pregnenolone 16α-carbonitrile (PXR activator), 3×10−5 M chenodeoxycholate (FXR activator), or 10−4 M ciprofibrate (PPARα activator), significantly activated expression from the corresponding reporter plasmid. Treatment with 22(R)-hydroxycholesterol (LXR activator) only weakly activated the LXR-responsive reporter, while pregnenolone 16α-carbonitrile treatment significantly activated this reporter. Cotransfection with wild-type LXRα strongly enhanced 22(R)-hydroxycholesterol-inducible expression from the LXR-responsive reporter. Cotransfection of hepatocyte cultures with each of the dominant negative nuclear receptor plasmids significantly inhibited inducible expression of the corresponding reporter while, with one exception (LXRα), cotransfection with the wild-type receptor moderately enhanced or had little effect on reporter expression. When each dominant negative nuclear receptor was cross-examined against all inducer–reporter pairs, effects on multiple inducer–reporter pairs were frequently observed. However, in general, only cotransfection with the appropriate dominant negative inhibited inducible reporter expression to a greater extent than did cotransfection with the corresponding wild-type receptor. Discussion: We suggest that the application of dominant negative nuclear receptors has utility in transient transfection studies aimed at discerning the regulatory role of individual nuclear receptor transcription factors in inducible hepatic gene expression, provided that appropriate controls are employed.

Introduction

Members of the nuclear receptor superfamily orchestrate a wide array of transcriptional responses to endogenous and exogenous stimuli. A wealth of evidence now indicates that a great proportion of the classical xenobiotic-inducible effects on hepatocellular gene expression is mediated through “orphan” members of the nuclear receptor superfamily. In recent years, many orphan receptors have been “adopted” with the discovery of hormones, endogenous intermediates, and xenobiotics that activate orphan receptor-mediated gene transcription. For example, in the liver, the constitutive androstane receptor (CAR), the farnesoid X receptor (FXR), the liver X receptor (LXR), the pregnane X receptor (PXR), and the peroxisome proliferative receptor α (PPARα) are now known to play central roles as sensors and transcriptional mediators in highly integrated drug and/or lipid metabolism signaling networks Lu et al., 2001, Olefsky, 2001, Xie & Evans, 2001.

Nuclear receptors share the common bonds of a modular structure, including well-described DNA- and ligand-binding domains, the latter of which contains a characteristic AF-2 transcription activation function/coactivator recruitment subdomain (Olefsky, 2001). In addition, many of the known ligand-activated orphan nuclear receptors exhibit some degree of promiscuity in their ligand- and DNA-binding requirements, as well as a prerequisite for heterodimer partnership with the retinoid X receptor (RXR). These features invite a matrix of opportunities for interrupted regulation by mutations.

The term “dominant negative” in nuclear receptor biology connotes the introduction of a structural defect in the receptor which, when expressed in a cell either as a natural or engineered mutation, hobbles the endogenous receptor's ability to form active transcription factor complexes. Examples of dominant negative nuclear receptors occur in nature. For example, the deletion of the AF-2 domain in the v-ErbA avian erythroleukemia oncoprotein represents a ligand binding-defective mutant of the thyroid hormone receptor, which produces dominant negative effects on endogenous thyroid hormone receptor activity Damm et al., 1989, Desbois et al., 1991, Shen & Subauste, 2000. Mutant dominant negative forms of the glucocorticoid receptor have been implicated in the pathogenesis of familial and sporadic cases of glucocorticoid resistance, and the physiological significance of a naturally occurring mutant glucocorticoid receptor (GRβ), which is deficient in ligand binding and has dominant negative effects on the wild-type glucocorticoid receptor (GRα), has yet to be determined Bamberger et al., 1996, Kino et al., 2001.

Primary cultured hepatocytes, which preserve much of the liver's well-differentiated cellular response machinery, provide a particularly attractive model to determine the precise roles of overlapping classes of orphan nuclear receptor in the regulation of gene transcription programs in the liver. However, to date, most mechanism-based orphan receptor investigations have been performed by cotransfecting the receptor, along with an appropriate reporter plasmid, into a recipient cell line (e.g., CV1) that lacks much of the response machinery that is present in the normal hepatocyte. Another approach has been to engineer nuclear receptor “gene knockout” models in the mouse Lee et al., 1995, Peet et al., 1998, Sinal et al., 2000, Staudinger et al., 2001, Wei et al., 2000. Unfortunately, interpretations based on either cell line or knockout mouse models are fraught with the difficulties that arise from acknowledging the considerable differences in xenobiotic metabolism mechanisms that occur across cell and species lines. These practical challenges have propelled into the limelight, paradigms that feature a well-differentiated cell phenotype coupled with an ability to manipulate the expression of target genes in cells derived from species other than the mouse, including human.

For most nuclear receptor-mediated events, specific inhibitors are either not available or hampered with the limitations associated with cell toxicity. In this regard, dominant negatives have been successfully used to inhibit the activity of a large variety of endogenous proteins. Since several examples of dominant negative forms of nuclear receptors have been described, sufficient information was available to permit the convenient design of the additional dominant negatives that were used in this study. The present investigation was undertaken to determine the feasibility and specificity of dominant negative nuclear receptors, when used as reagents to study inducible gene expression in primary cultured hepatocytes.

Section snippets

Materials

Phenobarbital, pregnenolone 16α-carbonitrile, and 22(R)-hydroxycholesterol were purchased from Sigma (St. Louis, MO). Chenodeoxycholic acid was purchased from Steraloids (Newport, RI). Ciprofibrate was a gift from Sterling Winthrop Pharmaceuticals Research Division (Rensselaer, NY). Matrigel was purchased from Collaborative Biomedical Products (Bedford, MA). Vitrogen was purchased from The Collagen Corporation (Palo Alto, CA). Recombinant human insulin (Novolin R) was purchased from Novo

Results

In this study, phenobarbital, at 10−4 M, was used as the CAR activator, while a plasmid containing 2413 bp of the CYP2B1-5′-flanking was used as reporter (Kocarek et al., 1998). We (Kocarek, Schuetz, & Guzelian, 1990) and others (Sidhu & Omiecinski, 1995) have previously demonstrated that this concentration of phenobarbital effectively induces CYP2B mRNA in primary cultured rat hepatocytes, whereas higher concentrations (i.e., ∼1 mM) are necessary to achieve effective induction of CYP3A. For

Discussion

In the present study, the capacities of prototypical nuclear receptor-activating agents to induce expression from a cotransfected reporter plasmid containing a response element known to be activated by that receptor were examined. It was determined that the selected chemical agents generally exhibited high specificities toward activation of the appropriate reporters. Thus, treatment with phenobarbital, pregnenolone 16α-carbonitrile, chenodeoxycholic acid, or ciprofibrate effectively increased

Acknowledgements

This work was supported by the National Institutes of Health Grants HL50710 (T.A.K.) and ES05823 (M.R.M), and by services provided by the Cell Culture and Imaging and Cytometry Facility Cores of the National Institute of Environmental Health Sciences Center Grant P30 ES06639. We thank Drs. Bryan Goodwin, Steven Kliewer, Cary Weinberger, Jurgen Lehmann, and Eric Johnson for providing plasmid constructs used in these studies.

References (37)

  • L.C. Quattrochi et al.

    A novel cis-acting element in a liver cytochrome P450 3A gene confers synergistic induction by glucocorticoids plus antiglucocorticoids

    Journal of Biological Chemistry

    (1995)
  • Q. Shen et al.

    Dimerization interfaces of v-ErbA homodimers and heterodimers with retinoid X receptor

    Journal of Biological Chemistry

    (2000)
  • J.S. Sidhu et al.

    cAMP-associated inhibition of phenobarbital-inducible cytochrome P450 gene expression in primary rat hepatocyte cultures

    Journal of Biological Chemistry

    (1995)
  • C.J. Sinal et al.

    Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis

    Cell

    (2000)
  • T. Sueyoshi et al.

    The repressed nuclear receptor CAR responds to phenobarbital in activating the human CYP2B6 gene

    Journal of Biological Chemistry

    (1999)
  • W. Xie et al.

    Orphan nuclear receptors: the exotics of xenobiotics

    Journal of Biological Chemistry

    (2001)
  • T.C. Aldridge et al.

    Identification and characterization of DNA elements implicated in the regulation of CYP4A1 transcription

    Biochemical Journal

    (1995)
  • C.M. Bamberger et al.

    Molecular determinants of glucocorticoid receptor function and tissue sensitivity to glucocorticoids

    Endocrine Reviews

    (1996)
  • Cited by (27)

    • Activation of the Farnesoid X-receptor in breast cancer cell lines results in cytotoxicity but not increased migration potential

      2016, Cancer Letters
      Citation Excerpt :

      Inserts were removed, and gap closure measured at time 0 and 12 hours after exposure to test chemical using tScratch program [29]. To demonstrate that the observed effects were FXR-dependent, experiments were undertaken using over-expression of a dominant negative FXR mutant [30]. MCF-7 and MDA-MB-231 cells were transfected with the FXR-DN expressing plasmid using Fugene 6 (Promega, UK) according to the manufacturer's instruction, and incubated overnight to allow protein expression.

    • Upregulation of scavenger receptor class B type I expression by activation of FXR in hepatocyte

      2010, Atherosclerosis
      Citation Excerpt :

      Plasmids pCMX-FXR and pCMV-β gal were described previously [15]. pCMX-D/N FXR, a plasmid expressing dominant negative FXR, was constructed following a published protocol [16]. Lipofectamine2000, TRIzol reagent, M-MLV reverse transcriptase and all products for cell culture were purchased from Invitrogen (Carlsbad, CA, USA).

    • Farnesoid X receptor, through the binding with steroidogenic factor 1-responsive element, inhibits aromatase expression in tumor leydig cells

      2010, Journal of Biological Chemistry
      Citation Excerpt :

      The plasmids containing different segments of the rat aromatase PII sequence ligated to a luciferase reporter gene (−1037/+94 (p-1037), −688/+94 (p-688), −475/+94 (p-475), −183/+94 (p-183), and −688/+94 mut (p-688m) (SF-1 site mutant)) were previously described (27). The FXR-responsive reporter gene (FXRE-IR1) and FXR-DN (dominant negative) expression plasmids were provided by Dr. T. A. Kocarek (Institute of Environmental Health Sciences, Wayne State University) (29). The FXR expression plasmid was provided by Dr. D. J. Mangelsdorf (Southwestern Medical Center, TX).

    • Regulatory network of lipid-sensing nuclear receptors: Roles for CAR, PXR, LXR, and FXR

      2005, Archives of Biochemistry and Biophysics
      Citation Excerpt :

      LXR binds to drug-responsive enhancer elements in the chicken CYP2H1 and the human CYP2B6 and CYP3A4 5′-flanking regions [93]. One possible mechanistic explanation for these observations is binding of LXR to these elements followed by competition with the binding of the xenosensors and a resulting decreased transcriptional activity of the drug-responsive enhancers [93,102]. Thus, LXR forms a negative feedback loop on the drug-inducible CYPs catalyzing the hydroxylation reactions of bile acids which result in LXR agonists.

    View all citing articles on Scopus
    View full text