toxicological sciences 132(1), 142–150 2013

doi:10.1093/toxsci/kfs321
Advance Access publication November 14, 2012

Expression of Human CAR Splicing Variants in BAC-Transgenic Mice

Yu-Kun Jennifer Zhang, Hong Lu, and Curtis D. Klaassen1
Department of Pharmacology, Toxicology and Therapeutics, University of Kansas Medical Center, Kansas City, Kansas 66160

1
To whom correspondence should be addressed at Department of Internal Medicine, University of Kansas Medical Center, 3901 Rainbow Blvd, Kansas City,
KS 66160. Fax: (913) 588-9197. E-mail: CKLAASSE@kumc.edu.

Received May 10, 2012; accepted November 7, 2012

heterodimerizes with the retinoid X receptor, binds to the pheno-

The nuclear receptor constitutive androstane receptor (CAR) is
a key regulator for drug metabolism in liver. Human CAR (hCAR)
transcripts are subjected to alternative splicing. Some hCAR spli-
cing variants (SVs) have been shown to encode functional proteins
by reporter assays. However, in vivo research on the activity of these
hCAR SVs has been impeded by the absence of a valid model. This
study engineered an hCAR-BAC-transgenic (hCAR-TG) mouse
model by integrating the 8.5-kbp hCAR gene as well as 73-kbp
upstream and 91-kbp downstream human genomic DNA into the
genome of CAR-null mice. A  series of experiments demonstrate
that (1) the expression of major hCAR mRNA SVs, SV0-4, in livers
of hCAR-TG mice is comparable to that in human livers; (2) the
hCAR SVs are predominantly expressed in liver, which resembles
the tissue distribution of CAR in humans, but diverges from that in
mice; and (3) major hCAR mRNA SVs increase markedly in post-
natal livers of hCAR-TG mice, which mimics the ontogeny of CAR
mRNA in humans. Thus, the transgene likely contains all the func-
tional regulatory elements controlling proper spatial and temporal
expression of the hCAR gene. Moreover, hCAR-TG mice respond
to the hCAR-specific agonist 6-(4-chlorophenyl)imidazo[2,1-b]
[1,3]thiazole-5-carbaldehyde O-(3,4-dichlorobenzyl)oxime instead
of the mouse CAR agonist 1,4-bis[2-(3,5-dichloropyridyloxy)]ben-
zene, as well as the common CAR activator, phenobarbital, sug-
gesting that hCAR is fully functional in livers of transgenic mice.
In summary, the hCAR-TG mice developed by this study represent
a valid model for studying in vivo function and regulation of hCAR
and its splicing variants.
Key Words: CAR; splicing variant; in vivo; BAC; transgenic mice.
Nuclear receptor constitutive androstane receptor (CAR;
NR1I3) plays an important role in regulating the metabolism of
xenobiotics and nutrients in the liver. Compared with many other
nuclear receptors, CAR is unique in that it is constitutively active
and can be further activated by many endogenous and exogen-
ous chemicals (Swales and Negishi, 2004), including steroid
hormones, pharmaceuticals, as well as environmental, dietary,
and occupational chemicals (Hernandez et al., 2009). Upon acti-
vation, CAR translocates from the cytoplasm into the nucleus,
barbital (PB) responsive enhancer module elements, and induces
the transcription of numerous phase I and II enzymes and trans-
porters to facilitate the elimination of chemicals (Honkakoski
et  al., 1998). Examples of genes regulated by CAR are phase
I enzymes such as cytochrome P450s (Cyps) (Huang et al., 2004;
Petrick and Klaassen, 2007), phase II enzymes such as UDP-
glucuronosyltransferase 1a1 (Huang et al., 2004; Sugatani et al.,
2001), glutathione S-transferase a1 (Huang et  al., 2004), and
sulfotransferase 2a1 (Assem et al., 2004), as well as membrane
transporters such as multidrug resistance-associated protein 4
(Assem et al., 2004; Petrick and Klaassen, 2007). Because of its
constitutive activity and broad influence on xenobiotic metabol-
ism, the hepatic expression level of CAR could potentially affect
the capability of an individual in eliminating drugs from the body
(Lamba et al., 2003; Vyhlidal et al., 2006).
Alternative splicing is a major source of protein diversity
in higher eukaryotes and plays critical roles in differentiation,
development, and disease. In humans, 35 splicing variants
(SVs) of CAR mRNA have been identified (Ensembl:
ENSG00000143257). In addition to the first-cloned SV0
(CAR1), SV1 (CAR2), and SV2 (CAR3) are the most abundant
human CAR (hCAR) SVs in human livers (Lamba et al., 2004).
Although most of hCAR SVs do not encode functional proteins,
several in vitro assays showed that some hCAR SVs encode
proteins that can be activated by chemicals that do not activate
the constitutively active wild-type (WT) hCAR (SV0 or CAR1)
(DeKeyser et al., 2009; Dring et al., 2010). For example, SV1
responds to the common plasticizer di(2-ethylhexyl)phthalate
(DeKeyser et al., 2009), whereas SV2 exhibits low basal activity,
but it is strongly activated by the prototypical hCAR agonist
6-(4-chlorophenyl)imidazo[2,1-b] [1,3]thiazole-5-carbaldehyde
O-(3,4-dichlorobenzyl)oxime (CITCO) (Auerbach et al., 2005).
However, the in vivo function and expression profiles of these
hCAR SVs remain unknown, mainly due to the lack of a valid
animal model.
Human hepatoma cell lines, human primary hepatocytes,
and human liver tissues are model systems that are commonly

© The Author 2012. Published by Oxford University Press on behalf of the Society of Toxicology.
All rights reserved. For permissions, please email: journals.permissions@oup.com
 In Vivo Study of Human Car Splicing Variants 143
used to study the regulation and function of human genes.
Significant disadvantages of these systems include compro-
mised regulatory properties (cell lines), lack of in vivo envir-
onment (cell lines and primary hepatocytes), and limited
accessibility (primary hepatocytes and human liver tissues).
Laboratory mice provide useful models of human development
and drug metabolism, but hCAR is different from mouse CAR
(mCAR) in regard to ligand specificity and pre-mRNA alter-
native splicing. mCAR has a specific ligand, 1,4-bis[2-(3,5-
dichloropyridyloxy)]benzene (TCPOBOP) (Nims et al., 1999),
and does not respond to the hCAR-specific ligand CITCO. In
comparison to the 35 hCAR SVs, mice only express 2 SVs of
CAR mRNA (Choi et al., 1997). hCAR transgenic mice pro-
vide a promising tool to study in vivo function and regulation of
hCAR. Whereas two lines of hCAR-transgenic mice have been
engineered, they either express only WT (SV0) hCAR (Zhang
et al., 2002) or lack the promoter and 5′-untranslated region of
hCAR gene (Scheer et al., 2008). Therefore, the purpose of this
study was to engineer a new line of hCAR-transgenic mice by
integrating the entire hCAR gene encoding region as well as
73-kbp upstream and 91-kbp downstream of human genomic
DNA into the mouse genome. Characterization of these hCAR-
transgenic mice indicates that they are suitable for studying the
developmental regulation and in vivo transcriptional function
of this human nuclear receptor.
Materials and Methods
Chemicals and reagents.  CITCO was obtained from Enzo Life Science
(Farmingdale, NY). TCPOBOP and PB were obtained from Sigma Aldrich (St
Louis, MO). All other reagents were obtained from Sigma Aldrich.
Generation of hCAR-BAC-transgenic mice.  The hCAR-BAC-transgenic
(hCAR-TG) mouse founders were generated by the Transgenic and Gene-
Targeting Institutional Facility at the University of Kansas Medical Center.
Briefly, bacterial artificial chromosome (BAC) RP11-297K8 DNA, constructed
by the Sanger Center Chromosome 1 Mapping Group, was linearized with
endonuclease NotI and microinjected at the concentration of 0.1 ng/µl into pro-
nuclei of fertilized eggs of FVB × C57BL/6 mice. Two founder lines (no. 22
and no.  25)  were generated, as detected by PCR analysis of genomic DNA
from tail biopsies.
The transgenic founders were bred with CAR-null mice (provided by Dr.
Ivan Rusyn, University of North Carolina, Chapel Hill, NC) on the C57BL/6
background to generate hCAR+/−/mCAR+/− F1 hybrids. The F1 offspring
were further crossed with CAR-null mice to generate the hCAR+/−/mCAR−/−
F2 generation that expressed only hCAR but not mCAR as determined by
PCR analysis. The F2 generation of hCAR+/−/mCAR−/− mice were bred with
their littermates to generate homozygous hCAR+/+/mCAR−/−. The zygosity
of hCAR genomic DNA in hCAR-TG mice was determined by SYBR Green
real-time quantitative PCR (real-time qPCR) as described (Haurogne et al.,
2007). PCR primers (5′-cagcgcattaaagtggtagcc-3′ and 5′-cccttgcagccct-
cacaagt-3′), targeting the 3′-end of the hCAR gene and resulting in a 201-bp
band, were used to determine the zygosity of the hCAR gene in hCAR-TG
mice. The mouse gap junction channel protein alpha 5 (mGJA5; GenBank:
NM_008121) was used as the endogenous reference. Real-time qPCR prim-
ers for mGJA5 were 5′-accatggaggtggccttca-3′ and 5′-catgcagggtatccag-
gaaga-3′. Genomic DNA from the F1 generation of transgenic mice, which
is known to be heterozygous, was used as the quantity control. Genomic
DNA from mCAR−/− mice served as the negative control. The homozygous
hCAR-TG mice from founder no. 22, as well as their WT littermates, were
used for this study.
Animals and treatments.  All procedures were conducted in accordance
with the NIH Guidelines for the Care and Use of Laboratory Animals and were
approved by the University of Kansas Medical Center Institutional Animal
Care and Use Committee. Mice were kept in a temperature/humidity controlled
facility. Food and water were available ad libitum. Light cycles were on a
14/10-h light/dark cycle with the light phase starting at 0600 h. For the devel-
opmental study, livers were collected from the hCAR-TG mice that were 2 days
before birth (both male and female) and 5, 10, 15, 20, and 60 days (male only)
of age. Livers were also collected from 60-day-old female hCAR-TG mice for
the gender divergence study. For the tissue distribution study, the following
organs were harvested from 2-month-old male hCAR-TG mice: liver, kidney,
stomach, duodenum, jejunum, ileum, colon, lung, heart, testis, white adipose
tissue, brown adipose tissue, and muscle. For the inducer study, male and/or
female hCAR-TG and WT mice (2 months old; n = 3–5) were dosed daily by
ip injection with corn oil (vehicle), CITCO (20 mg/kg/d), or PB (100 mg/kg/
day; male only) for 3 days, and livers were removed 24 h after the last dose.
TCPOBOP (3 mg/kg) was administrated by a single ip injection, and livers
were removed 72 h later.
qRT-PCR.  Total RNA was prepared from various tissues of WT and
hCAR-TG mice using RNA-Bee reagent (Tel-Test, Inc., Friendswood, TX)
according to the manufacturer’s protocol. RNA was dissolved in diethyl pyro-
carbonate–treated deionized water, and RNA concentrations were determined
spectrophotometrically at 260 nm. RNA samples from five male Caucasian
human livers were available in Dr. Klaassen’s laboratory (Xenotech, Lenexa,
KS). RNA was reverse transcribed into cDNA with multiscript reverse tran-
scriptase using High Capacity cDNA Reverse Transcription kits from Applied
Biosystems (Foster City, CA). Primers for detecting individual SVs of hCAR
were described previously (Jinno et al., 2004). The following primers were
used to detect respective gene expression: Cyp2b10 (GenBank accession
number AF128849), 5′-aaggagaagtccaaccagca-3′ and 5′-ctctgcaacatggggg-
tact-3′; mCAR (GenBank accession number NM_009803), 5′-ctcaaggaaa-
gcagggtcag-3′ and 5′-agttcctcggcccatattct-3′; Cyp3a11 (GenBank accession
number NM_007818), 5′-acaaacaagcagggatggac-3′ and 5′-ggtagaggag-
caccaagctg-3′; and glyceraldehyde-3-phosphate dehydrogenase (Gapdh)
(GenBank accession number M32599), 5′-aactttggcattgtggaagg-3′ and
5′-ggatgcagggatgatgttct-3′. The Power SYBR Green Master Mix (Applied
Biosystems) was used for real-time PCR analysis. Differences in gene
expression between groups were calculated using cycle threshold (Ct) values,
which were normalized to Gapdh mRNA.
Immunoblot analysis.  Nuclear and cytoplasmic fractions were prepared
from three individual mouse livers in each treatment group using NE-PER
nuclear protein extraction kit (ThermoFisher Scientific Inc., Rockford, IL).
Thirty micrograms of nuclear protein and 100 μg of the cytosol protein were
separated by SDS-polyacrylamide gel electrophoresis and electrophoretically
transferred to a polyvinylidene fluoride membrane. Nuclear proteins were
probed with an anti-CAR1/2 (M-150) polyclonal antibody (Santa Cruz
Biotechnology, Inc., Santa Cruz, CA) and anti-TATA-binding protein (TBP)
monoclonal antibody (ThermoFisher Scientific Inc.). Cytoplasmic proteins
were probed with rabbit anti-rat CYP2B1/2 polyclonal antibody (Xenotech)
and rabbit β-actin polyclonal antibody (Abcam, Cambridge, MA). Secondary
anti-rabbit or mouse antibodies were diluted 10,000-fold. Immunoreactions
were detected by Pierce ECL Western blotting substrate (ThermoFisher
Scientific Inc.). The density of blots was quantified by Quantity One 1-D
Analysis Software (Bio-Rad Laboratories, Hercules, CA). The individual blot
densities of CAR and Cyp2b10 were normalized to the nuclear and cytosol
loading control, TBP, and β-actin, respectively.
Statistical analysis.  Statistical significance was assessed by two-way
ANOVA when two variables were examined or Student’s t-test when one vari-
able was tested. The criterion for statistical significance was p < 0.05.
144 Zhang, Lu, And Klaassen

Results Comparison of Hepatic Expression of Major SVs of hCAR

Generation and Characterization of hCAR-TG Mice
BAC clone RP11-297K8 containing a 173-kbp DNA frag-
ment from human chromosome 1 (Fig.  1A) was used to
generate hCAR-TG mice. In addition to the hCAR gene, RP11-
297K8 encompasses other human gene loci such as ubiquitin
specific protease 21 (USP21), protoporphyrinogen oxidase
(PPOX), and apolipoprotein A2 (APOA2). Further details about
RP11-297K8 can be found on NCBI website http://www.ncbi.
nlm.nih.gov/clone/270914/. The transgenic founders were gen-
erated and bred with CAR-null mice on the C57BL/6 back-
ground for two generations to obtain the hCAR+/+/mCAR−/− F2
generation that expresses only hCAR but not mCAR (Fig. 1B).
Homozygous hCAR-TG mice were identified by real-time
qPCR as described (Haurogne et al., 2007). The primer designed
for the hCAR gene was able to detect one Ct difference for two-
fold DNA quantities between heterozygous and homozygous,
whereas the reference gene mGJA5 showed similar Ct values
in all examined samples (Fig. 1C). The homozygous hCAR-TG
mice from line no. 22 were used in this study and their WT lit-
termates were used as controls.
mRNA in hCAR-TG Mice and Humans
Major SVs of hCAR mRNA reported in a previous study
(Jinno et  al., 2004) were determined in livers of male adult
hCAR-TG mice and compared with those in human livers. As
shown in Figure 2, the WT hCAR SV0 was the predominant iso-
form (~60%), followed by SV2 (~30%), and SV1 (~8%) in both
human and hCAR-TG mouse livers. Marked interindividual dif-
ferences in hCAR SV0 were observed in human liver samples.
The expressions of SV0, SV1, SV2, and SV3 were comparable
in human livers and hCAR-TG mouse livers, whereas SV4 was
expressed lower in hCAR-TG mice than in humans.
Comparison of Prototypical CAR-Target Genes Induced by
Species-Specific CAR Activators in the Liver
Both WT and hCAR-TG mice (male) were subjected to the
following four treatments: corn oil (vehicle control), TCPOBOP
(mCAR-specific ligand), CITCO (hCAR-specific ligand), and
PB (nonligand activator for both mCAR and hCAR). As shown
in Figure 3A, hCAR and mCAR mRNAs were only detected in
livers of hCAR-TG and WT mice, respectively. The prototyp-
ical CAR-target gene Cyp2b10 was induced by TCPOBOP and

FIG. 1.  Generation of hCAR-TG mice. (A) The BAC clone RP11-297K8, containing about 73-kbp human DNA upstream of CAR gene and 91-kbp down-
stream of CAR gene, was integrated into mouse genome to generate the hCAR-TG mice. Abbreviations of genes surrounding hCAR gene locus were shown. For
detailed information about RP11-297K8, please refer to NCBI http://www.ncbi.nlm.nih.gov/clone/270914/. (B) PCR analysis of transgenic mice that contain
hCAR but not mCAR using hCAR− and mCAR-specific primers. L, DNA ladder; P, positive plasmid control; WT, wild-type mice; −/−, CAR-null mice; 1 and 2
are two hCAR-TG/mCAR-null mice. (C) Real-time quantitative PCR analysis to identify homozygous hCAR-TG mice. The mouse gap junction channel protein
alpha 5 (mGJA5; GenBank: NM_008121) was used as the endogenous reference. A, B, C, and D represent four hCAR-TG mice that are either homozygous (+/+)
or heterozygous (+/−) of hCAR gene.
 In Vivo Study of Human Car Splicing Variants 145
FIG. 2.  Expression of major hCAR mRNA SVs in livers of humans and
hCAR-TG mice. Major SVs of hCAR mRNA were analyzed in livers of five
male Caucasians and five male hCAR-TG mice (8 weeks old). The human liver
RNA samples were purchased from Xenotech.
PB in WT mice, but not by CITCO. On the contrary, Cyp2b10
mRNA was increased by CITCO and PB in livers of hCAR-
TG mice, but not by TCPOBOP. Cyp3a11 is a shared target
gene by CAR and pregnane X receptor. Cyp3a11 mRNA was
predominantly induced by TCPOBOP and PB in WT mice,
and by CITCO and PB in livers of hCAR-TG mice. However,
TCPOBOP also slightly increased Cyp3a11 mRNA in hCAR-
TG mice, suggesting off-target induction of Cyp3a11 by this
mCAR-specific ligand in mouse liver. Interestingly, the induc-
tion of CAR target genes by PB appears to be more potent in
livers of hCAR-TG mice than that in WT mice. Moreover,
TCPOBOP in WT mice and CITCO in hCAR-TG mice both
increased nuclear accumulation of mCAR and hCAR proteins,
respectively, as well as markedly induced Cyp2b10 protein
(Fig.  3B). In summary, the above data demonstrate that the
hCAR gene is functional in livers of the transgenic mice.
Ontogenetic Expression of Major hCAR SVs, Cyp2b10, and
Other Human Genes in Livers of hCAR-TG Mice
The expression of hCAR SVs was determined in livers of
male hCAR-TG mice at ages of 2 days before birth (−2 days),
5, 20, and 60 days. hCAR mRNA SV0, 1, 2, 3, and 4 were detect-
able in fetal livers of hCAR-TG mice. The expression of these
hCAR SVs was two- to seven-fold higher in neonates (5-day-old
hCAR-TG mice) livers than in fetal livers (Fig. 4A). Whereas
the expression of SV1, SV2, SV3, and SV4 remained stable
from 5 to 60 days of age, the WT SV0 further increased in adult
mice (60 days old) compared with adolescents (20 days old).
Cyp2b10 is the prototypical target gene of CAR in mouse
livers. The mRNA of Cyp2b10 was determined along with
hCAR SV0 in male hCAR-TG mice at ages of 2 days before
birth, 5, 10, 15, 20, and 60 days. As shown in Figure 4B (left
panels), two major increases of hCAR SV0 were observed at
5 days (newborn) and 60 days (adult). In comparison, Cyp2b10
mRNA constantly increased along with liver development in
the hCAR-TG mice. Specifically, hCAR SV0 was 2.9-fold and
Cyp2b10 was 7.2-fold higher in livers of 5-day than those in
−2-day-old mice. At 60  days of age, hepatic hCAR SV0 and
Cyp2b10 mRNA were both ~1.9-fold higher than those at
20 days of age.
In addition to hCAR, other human genes on the BAC construct,
such as apolipoprotein A2 (APOA2) and protoporphyrinogen
oxidase (PPOX), were also expressed in livers of the hCAR-TG
mice (Fig.  4B, right panels). APOA2 encodes an abundant
apolipoprotein, which stabilize the high-density lipoprotein
particles. Human APOA2 (hAPOA2) mRNA gradually
increased in hCAR-TG mice during liver development and
maturation. In contrast, the mRNA of human PPOX (hPPOX),
the enzyme that catalyzes the seventh step in heme production,
was expressed in fetal livers and markedly reduced in neonates
and progressively decreased to a minimal level in adults.
Tissue Distribution of hCAR SVs in hCAR-TG Mice and
Species-Specific Activation of CAR in Jejunum
The expression of hCAR SVs in major organs in male
hCAR-TG mice was shown in Figure  5A. The SVs of hCAR
mRNA were predominantly expressed in liver. The expression
of hCAR mRNA SVs in kidney and colon were ~2% of that
in liver, even lower in small intestine, and almost undetectable
in adipose tissues and muscle. In organs, where hCAR SVs
were detected, SV0 and SV2 were more abundant than other
SVs. In order to determine the function of hCAR in organs that
express low levels of hCAR in the transgenic mice, mRNAs of
hCAR, mCAR, and Cyp2b10 were determined in jejunums of
WT and hCAR-TG mice that were treated with species-specific
ligands of CAR or the common CAR activator, PB. As shown
in Figure 5B, hCAR mRNA was expressed at a very low level
in jejunums of the hCAR-TG mice, and neither CITCO nor
PB induced the CAR-target gene, Cyp2b10. In comparison,
WT mice expressed much more mCAR in jejunum. The
mCAR-specific ligand, TCOPOBOP, efficiently induced
Cyp2b10 mRNA in jejunums of WT mice. Interestingly, PB
was not able to increase Cyp2b10 expression in jejunums of
WT mice. Moreover, the basal expression of the CAR-target
gene Cyp2b10 was also higher in jejunums of WT mice than
those in hCAR-TG mice, which is consistent with the higher
expression of CAR in WT mice than in hCAR-TG mice. These
data indicate that, in contrast to the liver, in jejunums of the
hCAR-TG mice, hCAR is expressed at very low levels and the
CAR-target genes are not inducible by hCAR activators.
Gender Divergence of hCAR mRNA Expression and
Transcriptional Function
Male and female hCAR-TG mice expressed similar levels
of each of the five SVs of hCAR mRNA in their livers (Fig. 6,
upper panel). Female hCAR-TG mice expressed about onefold
more Cyp2b10 mRNA than male hCAR-TG mice, although
the difference was not statistically significant. The induction
146 Zhang, Lu, And Klaassen

FIG. 3.  Comparison of prototypical CAR-target genes induced by CAR activators in the liver. Male adult wild-type (WT) and hCAR-TG (TG) mice were
treated with corn oil (CO), TCPOBOP (TCP; mCAR agonist), CITCO (CIT; hCAR agonist), and PB. (A) Messenger RNA of hCAR, mCAR, and two CAR-target
genes, Cyp2b10 and Cyp3a11, in the liver. (B) Western blotting showing increased accumulation of CAR protein in the nucleus and markedly increased expression
of Cyp2b10 protein in cytoplasm after treatment with respective CAR agonist in both genotypes. Representative blots were shown on the left, and blot density
analysis of 3 individual samples of each group was presented on the right. Values were presented as mean ± SEM; n = 3–5 for mRNA data and n = 3 for protein data.
Student’s t-test (two sided) was performed. *Statistical differences between inducer-treated and corn oil–treated control groups within the same genotype, p < 0.05.

of Cyp2b10 by TCPOBOP in livers of WT mice and its induc-
tion by CITCO in hCAR-TG mice were both more profound
in females than in males (Fig. 6, lower panel). The hCAR-TG
mice appear to retain the female-predominant pattern of CAR
activation in liver.
Discussion
CAR together with its closest mammalian relative, the
pregnane X receptor, are nuclear receptors that regulate hepatic
genes that are responsible for the biotransformation and
distribution of drugs. Recently, numerous SVs of hCAR mRNA
have been identified (Lamba et  al., 2004) and some of them
have been shown to have transcriptional activity by in vitro
assays (DeKeyser et al., 2009; Dring et al., 2010; Omiecinski
et al., 2011). To obtain insights into the in vivo functions and
expression profiles of hCAR SVs, as well as to overcome the
marked species differences between human and mouse, this
study engineered a new line of hCAR transgenic mice utilizing
the BAC-transgenic technology.
Alternative splicing of mRNA is a common phenomenon in
mammals. The proper transcription and alternative splicing of
genes require the presence of not only the coding region of the
gene, but also the promoter, untranslated regions, and even distant
regulatory elements (Luco et al., 2010, 2011). The first hCAR-
transgenic mouse was engineered by integrating hCAR cDNA
(SV0) into the mouse genome (Zhang et al., 2002). Later, another
line of hCAR-transgenic mice was developed by knockin of the
 In Vivo Study of Human Car Splicing Variants 147
FIG.  4.  Developmental expression of hCAR SVs, Cyp2b10, and other
human genes in livers of male hCAR-TG mice. (A) The expression pattern of
major hCAR mRNA SVs during liver development. (B) Developmental expres-
sion profiles of hCAR SV0, the prototypical CAR target gene, Cyp2b10, as well
as human genes hAPOA2 and hPPOX in livers of hCAR-TG mice. Values are
presented as mean ± SEM; n = 3–5.
coding region (exons 2–9) of hCAR gene into the mCAR gene
allele, where hCAR replaced the endogenous mCAR gene and
was driven by the endogenous mCAR promoter (Scheer et  al.,
2008). Obviously, neither of these lines of hCAR-transgenic
mice contains the necessary human regulatory elements for the
hCAR gene. In comparison, the hCAR-TG (hCAR+/+/mCAR−/−)
mice described in this study contain the full-length hCAR gene,
including the noncoding exon 1, as well as flanking fragments
(73-kbp upstream and 91-kbp downstream) from the human gen-
ome (Fig. 1). Therefore, the hCAR-TG mice engineered by this
study are more likely to recapitulate the in vivo transcriptional
regulation of the CAR gene in humans.
The hepatic expression profiles of major SVs of hCAR in the
hCAR-TG mice, namely SV0, SV1, SV2, and SV3, were com-
parable to those in humans, suggesting that the pre-mRNA pro-
cessing of hCAR in hCAR-TG mice resembles that in humans
(Fig. 2). SV4, a less-investigated hCAR SV, is expressed at a
higher level in human livers than in hCAR-TG mice livers. This
might be due to the small sample size we used (n = 5), consider-
ing the significant interindividual differences in hCAR expres-
sion observed in both humans and hCAR-TG mice. The hCAR
knockin mice, in which the hCAR gene is regulated by the
mCAR promoter, were also indicated to express hCAR SVs 2,
3, 4, and 9, although their expression levels were not specified
(Scheer et al., 2008). Later work demonstrated that the hCAR-
knockin mice expressed hCAR SV0 (CAR1), SV1 (CAR2), and
SV2 (CAR3) at similar levels to those in human livers (Ross
et al., 2010). The data from the hCAR knockin mice suggest
that hCAR gene-coding region contains necessary elements for
the pre-mRNA editing. However, to what extent these elements
within the hCAR gene and those outside of the coding region
contribute to hCAR alternative splicing is unknown. Comparing
the expression profiles of a full spectrum of hCAR SVs in the
hCAR-TG mice developed by this study and the hCAR knockin
mice will shed light on this topic.
Species-specific activation of CAR by certain chemicals is
a well-known phenomenon. In agreement with previous stud-
ies (Scheer et al., 2008), hCAR-TG mice expressed increased
amounts of Cyp2b10 mRNA and protein in liver in response
to the hCAR-specific ligand, CITCO, and the common CAR
activator, PB, but not to the mCAR-specific ligand, TCPOBOP
(Fig.  3). In WT mice, the opposite is the case; CAR is acti-
vated by TCPOBOP and PB, but not by CITCO. Therefore, the
hCAR-TG mice appear to mimic the responses of humans to
various CAR activators.
Differences in the potency of induction by CAR activators
were observed between WT and hCAR-TG mice (Fig. 3A). The
induction of Cyp2b10 and 3a11 by CITCO (20 mg/kg/day for 3
continuous days) in livers of hCAR-TG mice was ~40% lower
than that by TCPOBOP (3 mg/kg for a single dose) in WT mice.
In contrast, PB (100 mg/kg/day for 3 continuous days) induced
both Cyp2b10 and Cyp3a11 to higher levels in hCAR-TG mice
than in WT mice. At dosages used in this study, TCPOBOP
is a more effective activator for mCAR in WT mice, whereas
PB, instead of CITCO, is the stronger activator for hCAR in
hCAR-TG mice. Higher doses of CITCO could be tested in the
hCAR-TG mice in future studies. However, 50 mg/kg CITCO
was shown to nonspecifically induce Cyp2b10 in WT mice in a
previous report (Scheer et al., 2008).
It has been reported that CAR mRNA was expressed at
low levels in fetal livers but increased after birth in humans
(Huang et al., 2003; Vyhlidal et al., 2006). Consistent with the
ontogenetic expression pattern of CAR in human livers, major
SVs of hCAR mRNA were expressed at low levels in livers of
hCAR-TG fetus and markedly increased in postnatal livers of
hCAR-TG mice (Fig. 4A). The WT hCAR mRNA SV0 but not
the other hCAR SVs further increased when the hCAR-TG mice
matured at 60 days of age. The hepatic expression of Cyp2b10
mRNA, the prototypical CAR target gene, constantly increased
during the maturation of hCAR-TG mice, including the two
time points when hCAR SV0 increased. These data suggest that
148 Zhang, Lu, And Klaassen

FIG. 5.  Tissue distribution of hCAR SVs in adult male hCAR-TG mice and species-specific activation of CAR in jejunum. (A) Tissue distribution of various
hCAR SVs. Total hCAR was the sum of all five SVs in each tissue. Values are presented as mean ± SEM; n = 4–5. (B) The expression of hCAR and mCAR as well
as CAR-target gene, Cyp2b10, in jejunums of wild-type (WT) mice treated with TCPOBOP (TCP) and PB, as well as hCAR-TG (TG) mice treated with CITCO
(CIT) and PB. Values were presented as mean ± SEM; n = 3–5. Student’s t-test (two sided) was performed. *Statistical differences between inducer-treated and
corn oil–treated control groups within the same genotype, p < 0.05.

the expression of CAR-target genes during liver maturation
could be affected not only by the amount of CAR, but also by
factors that could potentially activate CAR, as well as by other
transcription factors.
In addition to hCAR, other human genes on the BAC clone
were also expressed in livers of the hCAR-TG mice (Fig. 4B).
The hPPOX gene, encoding an enzyme involved in heme
production, was expressed higher in livers of hCAR-TG mice
before birth. The mRNA of the lipid metabolism-related gene,
APOA2, in contrast, increased after birth. It is known that
the hematopoietic cell transcriptional signature is dominant,
whereas most metabolic pathways are underexpressed in fetal
livers (Lee et al., 2012). The developmental expression profiles
of hPPOX, hAPOA2, hCAR, and Cyp2b10 in the hCAR-TG
mice reflected the transition of fetal liver from a hematopoietic
organ to a metabolic one around birth.
In mice, CAR mRNA is detected in several tissues with the
highest expression in liver and small intestine (Petrick and
Klaassen, 2007). In comparison, CAR mRNA is expressed at
a much higher level in liver than other tissues in humans (Baes
et al., 1994). In the hCAR-TG mice, the major SVs of hCAR
mRNA were highly expressed in liver, but only at low levels in
small intestine and other tissues (Fig. 5). Thus, the tissue dis-
tribution of hCAR in the hCAR-TG mice resembles the human
pattern instead of the mouse pattern. Moreover, the low expres-
sion of hCAR in jejunum was not sufficient to respond to the
hCAR-specific agonist, CITCO, to induce Cyp2b10 expression
in the hCAR-TG mice. In summary, the temporal and spatial
 In Vivo Study of Human Car Splicing Variants 149
FIG.  6.  Gender differences in hepatic expression of hCAR mRNA SVs
and Cyp2b10 in hCAR-TG mice. Upper panel: major hCAR SVs and prototypi-
cal CAR target gene, Cyp2b10, in livers of naïve male and female hCAR-TG
mice (8 weeks old). Lower panel: the induction of Cyp2b10 mRNA in wild-
type (WT) and hCAR-TG (TG) mice by mCAR agonist, TCPOBOP (TCP),
and hCAR agonist CITCO (CIT), respectively, in both genders. Control groups
were treated with corn oil (CO). Values are presented as mean ± SEM; n = 4–5.
Two-way ANOVA was performed with treatment and gender as main factors.
Significant main effects and interactions were followed by Student-Newman-
Keuls comparisons to assess the differences between groups. *Statistical differ-
ences between inducer-treated and corn oil–treated control groups; #Statistical
differences between male and female mice; p < 0.05.
expression patterns of the hCAR gene indicate that the trans-
gene likely contains all the functional regulatory elements
controlling the humanized expression of the hCAR gene in the
hCAR-TG mice.
Female-predominant expression and activity of CAR have
been documented in mice (Ledda-Columbano et al., 2003). In
humans, hepatic mRNA expression of both CAR and CYP2B6,
the human homologue of the mouse Cyp2b10, tend to be higher
in females than in males in certain ethnicities but not others
(Lamba et al., 2003). Marked interindividual variation in CAR
mRNA expression was observed in humans. The lower induc-
tion of Cyp2b10 by CAR activators in males was postulated to
be due to the male predominant hormone, androstanes, which
function as endogenous antagonists of CAR (Forman et  al.,
1998). This study shows that the five SVs of hCAR mRNA
were expressed at similar levels in livers of male and female
hCAR-TG mice, whereas Cyp2b10 mRNA tended to be higher
in females (Fig. 6). After CITCO treatment, Cyp2b10 was more
profoundly induced in livers of female than in male hCAR-TG
mice, which resembles the female-predominant induction of
Cyp2b10 by TCPOBOP in WT mice. Therefore, the activation
of CAR is female-predominant in the hCAR-TG mice.
In summary, this study established a novel hCAR-transgenic
mouse model that resembles the hepatic expression profiles of
major hCAR SVs in humans in aspects including expression
level, ontogeny, tissue distribution, and ligand specificity. This
model is extremely useful for in vivo research on the function
and regulation of this important human xenobiotic sensor, espe-
cially, for studies on in vivo roles of hCAR SVs in metabolizing
environmental toxicants and pharmaceutical drugs.
Funding
National Institute of Health [DK-081461, ES-019487, and
ES-009649 to C.D.K.; and 5P20RR024214-05 to Y.K.Z.].
Acknowledgments
The authors would like to thank the Transgenic and Gene-
targeting Institutional Facility and Dr. Kenneth Peterson in the
Biochemistry Department at KUMC for helping to develop the
transgenic mice, Drs. Dan Li, Andrew Lickteig, and Jie Liu for
their assistance in harvesting tissues, as well as colleagues in Dr.
Klaassen’s laboratory for critically reviewing the manuscript.
References
Assem, M., Schuetz, E. G., Leggas, M., Sun, D., Yasuda, K., Reid, G., Zelcer,
N., Adachi, M., Strom, S., Evans, R. M., et al. (2004). Interactions between
hepatic Mrp4 and Sult2a as revealed by the constitutive androstane receptor
and Mrp4 knockout mice. J. Biol. Chem. 279, 22250–22257.
Auerbach, S. S., Stoner, M. A., Su, S., and Omiecinski, C. J. (2005). Retinoid X
receptor-alpha-dependent transactivation by a naturally occurring structural
variant of human constitutive androstane receptor (NR1I3). Mol. Pharmacol.
68, 1239–1253.
Baes, M., Gulick, T., Choi, H. S., Martinoli, M. G., Simha, D., and Moore, D.
D. (1994). A new orphan member of the nuclear hormone receptor superfam-
ily that interacts with a subset of retinoic acid response elements. Mol. Cell.
Biol. 14, 1544–1552.
Choi, H. S., Chung, M., Tzameli, I., Simha, D., Lee, Y. K., Seol, W., and Moore,
D. D. (1997). Differential transactivation by two isoforms of the orphan
nuclear hormone receptor CAR. J. Biol. Chem. 272, 23565–23571.
DeKeyser, J. G., Stagliano, M. C., Auerbach, S. S., Prabhu, K. S., Jones, A. D.,
and Omiecinski, C. J. (2009). Di(2-ethylhexyl) phthalate is a highly potent
agonist for the human constitutive androstane receptor splice variant CAR2.
Mol. Pharmacol. 75, 1005–1013.
Dring, A. M., Anderson, L. E., Qamar, S., and Stoner, M. A. (2010). Rational
quantitative structure-activity relationship (RQSAR) screen for PXR and
CAR isoform-specific nuclear receptor ligands. Chem. Biol. Interact. 188,
512–525.
Forman, B. M., Tzameli, I., Choi, H. S., Chen, J., Simha, D., Seol, W., Evans,
R. M., and Moore, D. D. (1998). Androstane metabolites bind to and deacti-
vate the nuclear receptor CAR-beta. Nature 395, 612–615.
150 Zhang, Lu, And Klaassen
Haurogne, K., Bach, J. M., and Lieubeau, B. (2007). Easy and rapid method of
zygosity determination in transgenic mice by SYBR Green real-time quanti-
tative PCR with a simple data analysis. Transgenic Res. 16, 127–131.
Hernandez, J. P., Mota, L. C., and Baldwin, W. S. (2009). Activation of CAR
and PXR by dietary, environmental and occupational chemicals alters
drug metabolism, intermediary metabolism, and cell proliferation. Curr.
Pharmacogenomics Person. Med. 7, 81–105.
Honkakoski, P., Zelko, I., Sueyoshi, T., and Negishi, M. (1998). The nuclear
orphan receptor CAR-retinoid X receptor heterodimer activates the pheno-
barbital-responsive enhancer module of the CYP2B gene. Mol. Cell. Biol.
18, 5652–5658.
Huang, W., Zhang, J., Chua, S. S., Qatanani, M., Han, Y., Granata, R., and
Moore, D. D. (2003). Induction of bilirubin clearance by the constitutive
androstane receptor (CAR). Proc. Natl. Acad. Sci. U.S.A. 100, 4156–4161.
Huang, W., Zhang, J., Wei, P., Schrader, W. T., and Moore, D. D. (2004). Meclizine
is an agonist ligand for mouse constitutive androstane receptor (CAR) and an
inverse agonist for human CAR. Mol. Endocrinol. 18, 2402–2408.
Jinno, H., Tanaka-Kagawa, T., Hanioka, N., Ishida, S., Saeki, M., Soyama, A.,
Itoda, M., Nishimura, T., Saito, Y., Ozawa, S., et al. (2004). Identification of
novel alternative splice variants of human constitutive androstane receptor
and characterization of their expression in the liver. Mol. Pharmacol. 65,
496–502.
Lamba, J. K., Lamba, V., Yasuda, K., Lin, Y. S., Assem, M., Thompson, E.,
Strom, S., and Schuetz, E. (2004). Expression of constitutive androstane
receptor splice variants in human tissues and their functional consequences.
J. Pharmacol. Exp. Ther. 311, 811–821.
Lamba, V., Lamba, J., Yasuda, K., Strom, S., Davila, J., Hancock, M. L.,
Fackenthal, J. D., Rogan, P. K., Ring, B., Wrighton, S. A., et  al. (2003).
Hepatic CYP2B6 expression: Gender and ethnic differences and relationship
to CYP2B6 genotype and CAR (constitutive androstane receptor) expres-
sion. J. Pharmacol. Exp. Ther. 307, 906–922.
Ledda-Columbano, G. M., Pibiri, M., Concas, D., Molotzu, F., Simbula,
G., Cossu, C., and Columbano, A. (2003). Sex difference in the prolif-
erative response of mouse hepatocytes to treatment with the CAR ligand,
TCPOBOP. Carcinogenesis 24, 1059–1065.
Lee, J. S., Ward, W. O., Knapp, G., Ren, H., Vallanat, B., Abbott, B., Ho, K.,
Karp, S. J., and Corton, J. C. (2012). Transcriptional ontogeny of the devel-
oping liver. BMC Genomics 13, 33.
Luco, R. F., Allo, M., Schor, I. E., Kornblihtt, A. R., and Misteli, T. (2011).
Epigenetics in alternative pre-mRNA splicing. Cell 144, 16–26.
Luco, R. F., Pan, Q., Tominaga, K., Blencowe, B. J., Pereira-Smith, O. M., and
Misteli, T. (2010). Regulation of alternative splicing by histone modifica-
tions. Science 327, 996–1000.
Nims, R. W., Beebe, L. E., Utermahlen, W. E. Jr., Thomas, P. E., and Lubet, R.
A. (1999). Induction of hepatic cytochromes P450 by dietary 1,4-bis[2-(3,5-
dichloropyridyloxy)]benzene (TCPOBOP) in the B6C3F1 mouse: Dose-
dependence and pharmacodynamics. Int J Toxicol 18, 123–130.
Omiecinski, C. J., Coslo, D. M., Chen, T., Laurenzana, E. M., and Peffer, R. C.
(2011). Multi-species analyses of direct activators of the constitutive andros-
tane receptor. Toxicol. Sci. 123, 550–562.
Petrick, J. S., and Klaassen, C. D. (2007). Importance of hepatic induction of
constitutive androstane receptor and other transcription factors that regulate
xenobiotic metabolism and transport. Drug Metab. Dispos. 35, 1806–1815.
Ross, J., Plummer, S. M., Rode, A., Scheer, N., Bower, C. C., Vogel, O.,
Henderson, C. J., Wolf, C. R., and Elcombe, C. R. (2010). Human constitu-
tive androstane receptor (CAR) and pregnane X receptor (PXR) support the
hypertrophic but not the hyperplastic response to the murine nongenotoxic
hepatocarcinogens phenobarbital and chlordane in vivo. Toxicol. Sci. 116,
452–466.
Scheer, N., Ross, J., Rode, A., Zevnik, B., Niehaves, S., Faust, N., and Wolf,
C. R. (2008). A novel panel of mouse models to evaluate the role of human
pregnane X receptor and constitutive androstane receptor in drug response.
J. Clin. Invest. 118, 3228–3239.
Sugatani, J., Kojima, H., Ueda, A., Kakizaki, S., Yoshinari, K., Gong, Q. H.,
Owens, I. S., Negishi, M., and Sueyoshi, T. (2001). The phenobarbital response
enhancer module in the human bilirubin UDP-glucuronosyltransferase
UGT1A1 gene and regulation by the nuclear receptor CAR. Hepatology 33,
1232–1238.
Swales, K., and Negishi, M. (2004). CAR, driving into the future. Mol.
Endocrinol. 18, 1589–1598.
Vyhlidal, C. A., Gaedigk, R., and Leeder, J. S. (2006). Nuclear receptor expres-
sion in fetal and pediatric liver: Correlation with CYP3A expression. Drug
Metab. Dispos. 34, 131–137.
Zhang, J., Huang, W., Chua, S. S., Wei, P., and Moore, D. D. (2002). Modulation
of acetaminophen-induced hepatotoxicity by the xenobiotic receptor CAR.
Science 298, 422–424.