Evidence for the involvement of xenobiotic-responsive nuclear receptors in transcriptional effects upon perfluoroalkyl acid exposure in diverse species

Abstract

Humans and ecological species have been found to have detectable body burdens of a number of perfluorinated alkyl acids (PFAA) including perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS). In mouse and rat liver these compounds elicit transcriptional and phenotypic effects similar to peroxisome proliferator chemicals (PPC) that work through the nuclear receptor peroxisome proliferator-activated receptor alpha (PPARα). Recent studies indicate that along with PPARα other nuclear receptors are required for transcriptional changes in the mouse liver after PFOA exposure including the constitutive activated receptor (CAR) and pregnane X receptor (PXR) that regulate xenobiotic metabolizing enzymes (XME). To determine the potential role of CAR/PXR in mediating effects of PFAAs in rat liver, we performed a meta-analysis of transcript profiles from published studies in which rats were exposed to PFOA or PFOS. We compared the profiles to those produced by exposure to prototypical activators of CAR, (phenobarbital (PB)), PXR (pregnenolone 16 alpha-carbonitrile (PCN)), or PPARα (WY-14,643 (WY)). As expected, PFOA and PFOS elicited transcript profile signatures that included many known PPARα target genes. Numerous XME genes were also altered by PFOA and PFOS but not WY. These genes exhibited expression changes shared with PB or PCN. Reexamination of the transcript profiles from the livers of chicken or fish exposed to PFAAs indicated that PPARα, CAR, and PXR orthologs were not activated. Our results indicate that PFAAs under these experimental conditions activate PPARα, CAR, and PXR in rats but not chicken and fish. Lastly, we discuss evidence that human populations with greater CAR expression have lower body burdens of PFAAs.

Introduction

A wide range of species including wildlife and humans have measurable amounts of perfluoroalkyl acids (PFAA) in their tissues [1]. PFAA are composed of a carbon backbone (typically ranging from 4 to 15 carbons), full substitution of hydrogen by fluorine, and a functional group (carboxylic acid in the case of perfluorooctanoic acid (PFOA) and a sulfonic acid group for perfluorooctane sulfonate (PFOS)). PFAA have a number of uses including surfactant-processing aids in the production of fluoropolymers, coatings for clothing, fabrics, upholstery and carpets, and paper products approved for food contact. PFAA are readily absorbed [2], [3], are not known to be metabolized [4], [5], [6], and are poorly eliminated, with half-lives in humans estimated at 3.8 years for PFOA [7], [8]. Production of PFOA was estimated in excess of 500 metric tons in the year 2000. Although the production of PFOS by its major manufacturer was phased out at the end of 2002, likely contributing to the decline in PFOS body burdens [9], replacement PFAA chemicals (such as PFOA) have filled its void in the consumer and industrial markets.

Many PFAAs are members of a large class of structurally heterogeneous pharmaceutical and industrial chemicals called peroxisome proliferator chemicals (PPC) [10], [11]. The peroxisome proliferator-activated receptors (PPARα, β/δ, and γ), comprise a subset of the nuclear receptor superfamily and mediate many of the adaptive consequences of PPC exposure through direct or indirect activation [12], [13]. PPC exposure results in a predictable set of responses in the rodent liver. These responses include increased expression of fatty acid β-oxidation genes, hepatocyte peroxisome proliferation, hepatomegaly, hepatocyte hypertrophy and hyperplasia, and increased incidence of liver tumors [10]. For a number of PPC including WY-14,643, bezafibrate, and di-2-ethylhexyl phthalate, these responses are abolished in PPARα-null mice [14], [15], [16], [17], [18], [19] demonstrating the importance of PPARα activation as a key event in the PPARα mode of action [10].

PFOA exposure in rats increases liver tumor incidence. Carcinogenesis is thought to arise through activation of PPARα and induction of responses associated with a typical PPC (summarized in ref. [11]). PFOA and PFOS can activate PPARα and to a lesser extent, PPARβ and PPARγ in trans-activation assays [20], [21], [22]. Most of the gene expression changes (∼85%) in the mouse liver after PFOA exposure were shown to be PPARα-dependent as transcriptional levels were altered in wild-type mice but not PPARα-null mice [23], [24]. Wild-type but not PPARα-null mice exposed to PFOA exhibited a number of pathological and histological features associated with PPC-like effects [25]. Another PFAA, perfluorodecanoic acid (PFDA) was shown to induce a typical marker gene of PPC exposure (Cyp4a14) in wild-type but not PPARα-null mice [26] and PFDA down-regulated a number of bile acid transporters in wild-type but not PPARα-null mice [27]. Taken together, many of the hepatic responses associated with exposure to PFOA and other PFAAs are likely due to activation of PPARα.

In addition to activation of PPARα, PFAA may also mediate effects in the liver through other nuclear receptors. Increases in liver to body weights were observed in PFOA-exposed PPARα-null mice [28], [29], the biological basis for which is unknown. The transcript profiles of livers from rats treated with PFOA or PFOS were compared to a toxicogenomics reference database of over 600 compounds. As expected, PFOA and PFOS exhibited gene expression changes similar to other PPC. However, PFOA and to a lesser extent PFOS, increased the expression of marker genes under control of other nuclear receptors [30]. These candidate nuclear receptors include constitutive activated/androstane receptor (CAR) and pregnane X receptor (PXR) which regulate the expression of xenobiotic metabolizing enzymes (XME) in response to exposure to drugs and environmental chemicals [31], [32]. Activators of CAR and PXR such as phenobarbital (PB) and pregnenolone 16 alpha-carbonitrile (PCN), respectively regulate an overlapping set of XMEs and growth regulatory genes in the rat liver [33]. Like PPC, exposure to CAR or PXR activators can lead to increases in liver weight and hepatocyte hyperplasia that are abolished in mice nullizygous for the individual receptors [34], [35], [36], [37]. Chronic exposure to the CAR activators PB or 1,4-bis-[2-(3,5-dichloropyridyloxy)] benzene (TCPOBOP) leads to increases in liver cancer in wild-type but not CAR-null mice [36], [37]. By comparing the transcript profiles in wild-type and PPARα-null mice after PFOA exposure to those of wild-type and CAR-null mice exposed to CAR activators, our group showed that the PPARα-independent genes regulated by PFOA largely overlapped with those regulated by CAR [23], [24]. PFDA was shown to regulate the CAR target gene and protein Cyp2b10 in wild-type but not CAR-null mice [26]. Taken together, there is growing evidence that PFAAs including PFOA activates multiple nuclear receptors in mouse liver including PPARα and CAR.

There is interest in understanding the potential toxicological effects of increasing body burdens of PFAAs on wild species. A number of studies have examined the effects of PFAA exposure on liver gene expression in birds and fish. Chickens have been used as a surrogate species to study the effects of PFAAs in wild birds. The chicken genome does not encode PXR or CAR genes but rather the chicken X receptor (CXR) gene, which has about equal sequence and functional similarity to PXR and CAR. CXR acts as the main xenobiotic sensing nuclear receptor in the chicken and possibly other birds [38]. Microarray studies using the Affymetrix chicken gene chip showed that chickens exposed to PFOA or PFOS exhibited alterations of some XMEs in liver [39]. A microarray study on the gene expression in the livers of wild common cormorants in which the levels of environmental contaminants including PFAAs, dioxins, and PCBs were measured showed correlations between PFOS and a number of genes but few XMEs [40]. Despite the evidence that chickens and other bird species possess a PPARα gene and can respond modestly to the effects of a classical PPARα activator ciprofibrate [41], there was no evidence in these microarray studies that exposure to PFAAs led to activation of PPARα.

Some fish species possess a PXR but not a CAR gene [42], [43]. Compared to mammals, there is little information about what genes PXR is regulating in fish. However, there were correlations between activation of PXR and the PXR-regulated genes MDR1 and CYP3A in fish [43]. Two microarray studies have been recently published revealing the gene expression changes in the livers of common carp or rare minnow after PFAA exposure. Both studies observed alterations in a number of XMEs after exposure [44], [45]. Studies in which different species of fish were exposed to activators of PPARα revealed species differences in PPARα-like responses. Catfish, bluegill, rainbow trout and zebrafish were responsive; medaka were marginally responsive; and common carp and sea bass were non-responsive [46], [47], [48], [49], [50], [51]. In fathead minnows, there were minor increases in palmitoyl-CoA oxidase activity (an indicator of PPARα activation) in livers after exposure to PFOA; these increases were observed at one low concentration but not higher concentrations [52]. The microarray studies in rare minnow and common carp did not show evidence for activation of PPARα by PFAAs [44], [45]. Overall, studies in the livers of birds and fish provide some marginal evidence that CAR/PXR-regulated responses are playing a role in gene expression changes after exposure to PFAAs.

Understanding the mechanism of chemical toxicity in the rat is hampered by the lack of genetic models abundantly available in mice. However, toxicogenomic analyses in which global gene expression profiles of chemicals with different modes of action are directly compared can be invaluable to make predictions as to the molecular basis for toxicity. Given that the microarray profiles in the rat liver after exposure to PFOA and PFOS indicated the involvement of CAR and PXR [30], we compared the profiles produced by PFOA and PFOS from available studies to known activators of CAR, PXR, and PPARα in rat liver. Additionally, we reanalyzed and reviewed microarray data from birds and fish after PFAA exposure to determine if transcriptional responses could be attributable to activation of xenobiotic-responsive nuclear receptors. Finally, we make correlations between expression of CAR and regulated genes in the livers of different human populations and levels of PFAAs in their blood.