A Truncated Human Peroxisome
Proliferator-Activated Receptor a
Splice Variant with Dominant
Negative Activity
Philippe Gervois, Inés Pineda Torra, Giulia Chinetti,
Thilo Grötzinger, Guillaume Dubois, Jean-Charles Fruchart,
Jamila Fruchart-Najib, Eran Leitersdorf*, and Bart Staels
U.325 INSERM
Département d’Athérosclérose
Institut Pasteur de Lille
59019 Lille, France
Faculté de Pharmacie
Université de Lille II
59006 Lille, France
The peroxisome proliferator-activated receptor a
(PPARa) plays a key role in lipid and lipoprotein
metabolism. However, important inter- and intraspecies differences exist in the response to
PPARa activators. This incited us to screen for
PPARa variants with different signaling functions.
In the present study, using a RT-PCR approach a
variant human PPARa mRNA species was identified, which lacks the entire exon 6 due to alternative splicing. This deletion leads to the introduction
of a premature stop codon, resulting in the formation of a truncated PPARa protein (PPARatr) lacking part of the hinge region and the entire ligandbinding domain. RNase protection analysis
demonstrated that PPARatr mRNA is expressed in
several human tissues and cells, representing between 20–50% of total PPARa mRNA. By contrast,
PPARatr mRNA could not be detected in rodent
tissues. Western blot analysis using PPARa-specific antibodies demonstrated the presence of an
immunoreactive protein migrating at the size of in
vitro produced PPARatr protein both in human hepatoma HepG2 cells and in human hepatocytes.
Both in the presence or absence of 9-cis-retinoic
acid receptor, PPARatr did not bind to DNA in gel
shift assays. Immunocytochemical analysis of
transfected CV-1 cells indicated that, whereas
transfected PPARawt was mainly nuclear localized,
the majority of PPARatr resided in the cytoplasm,
with presence in the nucleus depending on cell
culture conditions. Whereas a chimeric PPARatr
protein containing a nuclear localization signal
0888-8809/99/$3.00/0
Molecular Endocrinology
Copyright © 1999 by The Endocrine Society
cloned at its N-terminal localized into the nucleus
and exhibited strong negative activity on PPARawt
transactivation function, PPARatr interfered with
PPARawt transactivation function only under culture conditions inducing its nuclear localization.
Cotransfection of the coactivator CREB-binding
protein relieved the transcriptional repression of
PPARawt by PPARatr, suggesting that the dominant negative effect of PPARatr might occur
through competition for essential coactivators. In
addition, PPARatr interfered with transcriptional
activity of other nuclear receptors such as PPARg,
hepatic nuclear factor-4, and glucocorticoid receptor-a, which share CREB-binding protein/p300 as a
coactivator. Thus, we have identified a human
PPARa splice variant that may negatively interfere
with PPARawt function. Factors regulating either
the ratio of PPARawt vs. PPARatr mRNA or the
nuclear entry of PPARatr protein should therefore
lead to altered signaling via the PPARa and, possibly also, other nuclear receptor pathways. (Molecular Endocrinology 13: 1535–1549, 1999)
INTRODUCTION
Peroxisome proliferator (PP)-activated receptors
(PPARs) are ligand-activated transcription factors belonging to the superfamily of nuclear receptors (1, 2).
After activation, PPARs heterodimerize with the 9-cisretinoic acid receptor (RXR) as a preferential partner. The
PPAR/RXR complex subsequently binds to DNA on specific response elements termed peroxisome proliferator
response elements (PPREs), located in regulatory regions of target genes, thereby modulating their transcriptional activity. PPREs consist of the juxtaposition of two
1535
MOL ENDO · 1999
1536
derivatives of a canonical hexamer sequence PuGGTCA
spaced by one or two nucleotides and commonly called
direct repeat (DR) 1 or 2. Three distinct PPARs, a, b, and
g, have been described in several species such as Xenopus (3), rat (4), mouse (5–10), hamster (11), and human
(12–17), each encoded by different genes and exhibiting
distinct tissue distribution patterns.
Mouse PPARa was the first identified PPAR family
member, and its expression is principally detected in
tissues exhibiting high rates of b-oxidation, i.e. in liver,
kidney, heart, and muscle (5). In rodents, PPARa mediates a pleiotropic response to a wide variety of compounds, termed peroxisome proliferators (PPs), leading to peroxisome proliferation, hepatomegaly, and
possibly hepatocarcinogenesis (see Ref. 18 for review). PPs include various xenobiotics such as plasticizers, herbicides, some dietary compounds (fatty acids), pharmacological drugs (e.g. fibrates), and
eicosanoids, some of which have been shown to be
PPARa ligands (19–24). The implication of PPARa in
peroxisome proliferation in mice has been unequivocally demonstrated in PPARa knockout mice that become resistant to PPs (25). Interestingly, PPARa also
mediates the hypolipidemic effect of PPs by regulating
the transcription of key genes in lipoprotein metabolism. For instance, the hypolipidemic fibrates act by
repressing the transcription of the apo A-I and apo A-II
(26), apo A-IV (27), apo C-III (28, 29), hepatic lipase
(30), and LCAT (lecithin-cholesterol acyltransferase)
(31) genes in rodents, an effect that is mediated by
PPARa (32). Moreover, fibrates induce the transcription of the lipoprotein lipase gene in rat liver (33).
PPARa also plays an important role in intracellular lipid
metabolism by regulating gene expression of peroxisomal b-oxidation enzymes such as acyl-coenzyme A
(CoA) oxidase (34, 35), multifunctional enzyme (36),
and 3-ketoacyl-CoA thiolase (37) in rodents. Moreover, PPREs have been identified in the 59-upstream
region of several extraperoxisomal genes such as microsomal CYP4A6 (38) and 3-hydroxy-3-methylglutaryl-CoA synthase, a key mitochondrial enzyme in
ketogenesis (39). In man also, PPARa mediates fibrate
action on lipoprotein metabolism since functional PPREs
have been identified in the regulatory sequences of
genes implicated in lipid transport such as apo A-I (40),
apo A-II (41), apo C-III (28), and lipoprotein lipase (42).
Thus, PPARa can be considered as a major regulator of
intra- and extracellular lipid metabolism.
Although most of the PPARa functions seem to be
conserved, some important differences in response
exist. Although treatment with fibrates as well as with
other PPs leads in rodents to a marked peroxisome
proliferation and hepatomegaly (43, 44), such a response has never been shown to occur in man (45–
47). Whereas in the majority of fibrate-treated patients
plasma high-density lipoprotein-cholesterol concentrations increase, a significant fraction of patients respond to fibrates with no change or even with a decrease in their plasma high-density lipoproteincholesterol levels (48, 49). Such different responses to
fibrates among species and within a species among
individuals may be linked to variations in the PPARa
Vol 13 No. 9
signaling pathways, possibly due to the presence of
different PPARa variants.
In this report, we therefore searched for the existence of different PPARa variants in human tissues.
Using a RT-PCR approach, we identified, in addition to
the full-length transcript, a transcript lacking exon 6
due to alternative splicing. This splicing leads to the
generation of a premature stop codon, giving rise to a
C-terminal truncated protein (hPPARatr), which lacks
part of the hinge region and the entire ligand-binding
domain (LBD). We furthermore demonstrate that the
variant transcript is expressed in various cell lines and
tissues of human origin but not in rodents. Furthermore, results from Western blot experiments indicate
both in vitro and in vivo expression of hPPARatr protein. Moreover, we show that hPPARatr has a repressive activity on the hPPARawt transactivation function,
only when it translocates to nucleus. Together, these
data indicate that hPPARatr, if produced in sufficient
amounts in vivo, may play a regulatory role on
hPPARawt signaling function.
RESULTS
The hPPARa Gene Gives Rise to Two
Distinct Transcripts
To determine the existence of different hPPARa variants, RT-PCR experiments were performed using a set
of specific primers covering the entire hPPARa-coding
region (Fig. 1A). These primers were designed to hybridize to the extremities of putative exons in the
hPPARa gene deduced from sequence comparison of
the mouse, rat, and human cDNAs and the mouse
gene of PPARa (5, 4, 13, 14, 50). RT-PCR analysis was
first performed on RNA extracted from HepG2 cells.
For each primer pair tested, a product of the expected
size was detected (Fig. 1B). Surprisingly, using primer
pairs 35/83, 45/83, and 15/73, a second, approximately 200-bp shorter product was obtained in the
same PCR reaction (Fig. 1B). However, whenever
primer combinations including a putative exon 6
primer were used, only one fragment of a size corresponding to the hPPARawt transcript was obtained.
These observations suggest the existence of two transcripts for hPPARa, the shorter one lacking at least
one portion of approximately 200 bp around the putative exon 6 region. To determine the identity of the
shorter transcript, the shorter PCR fragment was isolated, cloned, and sequenced (Fig. 2A). Sequence
analysis revealed that the hPPARa variant lacked 203
bp between bp 632–834 (numbering based on the
nucleotide sequence described in Ref. 14), resulting in
a frame shift introducing a premature stop codon 39 of
the deletion (Fig. 2A). The shorter transcript should
result in the production of a truncated form of hPPARa
containing the first 170 N-terminal amino acids of the
hPPARawt protein and 4 different C-terminal amino
acids (Fig. 2B), including the ligand-independent
transactivation and DNA-binding domains and part of
the hinge region. To determine whether this deleted
A Dominant Negative Human PPARa Isoform
1537
Fig. 1. The hPPARa Gene Gives Rise to Two Distinct Transcripts
A, Schematic representation of the hPPARa coding sequence and position of the primers (arrows) used for RT-PCR analysis.
Numbers below arrows indicate the first nucleotide of the 59-termini of primers. Numbers above the schematic coding sequence
represent the number of the first 59-nucleotide of corresponding primers. B, RT-PCR analysis on RNA isolated from HepG2 cells.
Total RNA was extracted, reverse transcribed, and amplified using the indicated pairs of primers as described in Materials and
Methods. Control corresponds to PCR reaction without RT product.
transcription variant was generated by alternative
splicing of exon 6, the intron-exon boundaries of the
hPPARa gene were determined by direct sequencing
of a hPPARa gene containing BAC clone. Sequence
comparison showed that the deleted fragment localized exactly at the boundaries of exon 6, indicating
that it is generated by an alternative splicing event
skipping exon 6 (Fig. 2C). Therefore, the hPPARa gene
gives rise in HepG2 cells to two transcripts of distinct
size, which are generated by alternative splicing.
The Variant hPPARa Transcript Is Widely
Expressed in Human Cell Lines and Tissues, but
Not in Rodents
RT-PCR analysis was performed next to test whether
the alternative splicing underlying formation of the
hPPARa variant transcript occurs in different human
cell lines and tissues. RT-PCR analysis was therefore
performed using the 35 sense primer and either the 73
or 83 antisense primers that cover the alternatively
spliced region. Two PCR products of the expected
size for wt and variant transcripts were detected in all
cells analyzed, indicating a wide distribution of both
messages in different human transformed cell lines
(Fig. 3A). Furthermore, both wt and variant transcripts
were detected in human tissues such as liver and
adipose tissue (Fig. 3B). Therefore, both hPPARa transcripts appear to be widely distributed in different
human tissues and cells.
We next tested whether the variant transcript is also
expressed in liver and in adipose tissue of two rodent
species, namely rats and mice. RT-PCR was carried
out on RNA isolated from these tissues using the conserved primer pair 35/73 (5, 4, 13, 14) and resulted in
detection of only one fragment of predicted size of 922
bp for the PPARawt transcript (Fig. 3C). Thus, the
PPARa variant transcript appears to be specifically
expressed in man but not in rodents.
To quantify the relative level of variant and wt hPPARa mRNA, a RNase protection assay was developed that allows simultaneous quantification of both
mRNA species. Using a riboprobe that includes exon
MOL ENDO · 1999
1538
Vol 13 No. 9
Fig. 2. The hPPARawt and hPPARatr Transcripts are Generated by an Alternative Splicing Event Skipping Exon 6
A, Sequence alignment of hPPARawt and hPPARatr around the exon 6 region. Sequence of the hPPARatr contains an in-frame
TGA (boldface). B, Schematic representation of the splicing events Swt and Str generating hPPARawt and hPPARatr transcripts,
respectively. The variant transcript lacks exon 6 resulting in a frame shift generating a premature TGA termination codon.
Predicted protein structure of hPPARawt and hPPARatr is represented at the bottom of the figure (DBD, DNA-binding domain;
LBD, ligand-binding domain). Numbers above the schematic protein structure correspond to amino acids delimiting PPARa
domains. C, Intron-exon boundaries of the hPPARa and mPPARa exon 6.
A Dominant Negative Human PPARa Isoform
1539
Fig. 3. The PPARatr Transcript Is Widely Expressed in Human Cell Lines and Tissues
Expression of the PPARawt and PPARatr transcripts was analyzed by RT-PCR analysis as described in Materials and Methods.
Primer pairs used are indicated at the bottom of each panel. A, Expression of the PPARawt and PPARatr transcript in various
human cell lines. B, Expression of the PPARawt and PPARatr transcript in two different human liver and adipose tissue (Ad. tissue)
samples. C, Expression of the PPARawt transcript in rat and mouse liver and adipose tissue (Ad. tissue).
Fig. 4. hPPARawt and hPPARatr mRNA Levels Vary between
Individuals and Tissues
RNase protection assay was used to assess the expression of hPPARawt and hPPARatr mRNAs in human liver and
adipose tissue from different patients. Expression of 36B4
was determined and used as internal control. Quantification
of hPPARawt and hPPARatr transcripts was performed as
described in Materials and Methods. R.A.U., Relative absorbance units.
6 and part of exon 7 of hPPARa, protected fragments
corresponding to wt and variant hPPARa mRNA were
detected in human liver and adipose tissue (Fig. 4).
When liver samples from different subjects were compared, absolute mRNA levels of hPPARawt and
hPPARatr varied from 2.42 to 6.97 relative absorbance
units (R.A.U.) and from 1.57 to 2.67 R.A.U., respectively. In adipose tissue, the hPPARawt transcript level
ranged from 1.78 to 2.00 R.A.U. whereas the hPPARatr
varied from 1.18 to 1.60 R.A.U. Moreover, in liver the
wt-variant mRNA ratio also varied among individuals
from 1:1 to 4:1, whereas it was almost constant and
near 1:1 in each adipose tissue sample analyzed. Interestingly, whereas hPPARawt levels in liver varied
substantially between individuals, hPPARatr levels appeared to be more constant and similar in liver and in
adipose tissue. These results indicate that both transcripts are widely expressed in human cell lines and
tissues and that the wt-variant PPARa mRNA ratio
seems to vary substantially among individuals in human liver. On the contrary, expression level of each
PPARa transcript appears fairly constant among individuals in adipose tissue.
hPPARatr Protein Can Be Produced in Vitro and
in Vivo
To determine whether the alternatively spliced
hPPARa mRNA could give rise to the production of a
protein, in vitro and in vivo translation experiments
were performed next. To detect effective synthesis of
both hPPARawt and hPPARatr protein, we performed
Western blot analysis. An anti-hPPARa antibody was
generated using as antigen a peptide that covers a
part of the A/B domain with 95% identity between
mouse and human PPARa. This polyclonal anti-
MOL ENDO · 1999
1540
PPARa antibody recognizes specifically the PPARa
subtype (51). A specific signal was observed for
both PPARa forms from in vitro translated plasmids
(Fig. 5A). Cos-1 cells transfected with either
pSG5hPPARawt or pSG5hPPARatr produced proteins
of the expected size, respectively (Fig. 5A). Although
the size of the detected signal in SDS-PAGE was
larger than theoretically calculated, the PPARs synthesized either with the in vitro translation rabbit reticulocyte system or in vivo in Cos-1 cells were of similar
size. Furthermore, when whole-cell protein extracts
from human hepatoma HepG2 cells and from human
primary hepatocytes were analyzed for PPARa protein
expression, specific bands were detected, the size of
which was comparable to those using in vitro translated hPPARawt and hPPARatr protein (Fig. 5, B and
D). To ensure that the shorter protein detected in
HepG2 cells and in human primary hepatocytes corresponds to hPPARatr, the membranes were reprobed
using an antibody raised against the C-terminal extremity of hPPARa, which is lacking in hPPARatr. As
expected, a specific band was observed for
hPPARawt, whereas no signal was detected for
hPPARatr (Fig. 5, C and E), strongly suggesting that
hPPARatr protein may be produced in both HepG2
cells and human primary hepatocytes. These results
indicate that the hPPARatr transcript may lead to the in
vivo expression of a corresponding protein containing
the ligand-independent transactivation and the DNAbinding domains, but lacking the ligand-dependent
transactivation domain.
Vol 13 No. 9
The hPPARatr Isoform Does Not Bind to a PPRE
Since hPPARatr still contains the DNA-binding domain, we next sought to determine whether it could
bind a PPRE-containing DNA fragment. Therefore, gel
retardation assays were performed using in vitro translated hPPARawt and hPPARatr and the apo A-II PPREcontaining oligonucleotide as probe (41). mRXRa,
hPPARawt, and hPPARatr alone did not bind to the
labeled apoA-II PPRE site (Fig. 6). However, in the
presence of mRXRa, hPPARawt formed a DNA-protein
complex. This binding was specific since excess
amounts of cold probe could compete for PPAR/RXR
binding. By contrast, even in the presence of mRXRa,
hPPARatr did not bind to DNA (Fig. 6). The same
results were obtained using different PPREs of the
ACO or the apo A-I genes (data not shown). These
results demonstrate that hPPARatr protein is not able
to interact with DNA either as dimer with RXR or even
as monomer. However, it cannot be excluded that
hPPARatr heterodimerizes with mRXRa in solution,
leading to the formation of an inactive protein complex. To test this mechanism, hPPARawt was incubated with increasing amounts of hPPARatr in the
presence of mRXRa. The hPPARawt/mRXRa complex
could be detected, but signal intensity did not vary for
a constant amount of hPPARawt even when large
amounts of hPPARatr were added (data not shown).
These results indicate that the hPPARatr form does not
bind to DNA and is unable to heterodimerize efficiently
with mRXRa in vitro.
Fig. 5. hPPARatr Protein Is Produced in Vitro and in Vivo
A, Western blot analysis of in vitro translated protein using the TNT-coupled reticulocyte lysate system and of total protein
extract from Cos-1 cells transfected with either pSG5hPPARawt or pSG5hPPARatr. B, Total protein extract from HepG2 cells.
Immunodetection was performed using PPARa rabbit polyclonal antibody developed against an N-terminal PPARa portion. C,
Immunoblot used in panel B was stripped and reprobed using a PPARa rabbit polyclonal antibody raised against the C-terminal
part of hPPARa. D, Total protein extract from human primary hepatocytes (HH). Immunodetection was performed using PPARa
rabbit polyclonal antibody developed against a N-terminal PPARa portion. E, Immunoblot used in panel D was stripped and
reprobed using a PPARa rabbit polyclonal antibody raised against the C-terminal part of hPPARa. Protein extracts were prepared
as described in Materials and Methods. Total protein extract was separated by SDS-PAGE and analyzed by immunoblot assay.
A Dominant Negative Human PPARa Isoform
1541
Fig. 7. hPPARatr Affects Transcriptional Activity of
hPPARawt
HepG2 cells were transfected with 1 mg Luc reporter construct driven by the heterologous TK promoter containing
three copies of apo A-II PPRE (J3). Increasing amounts (13,
53) of pSG5hPPARatr were transfected in the presence or in
the absence of pSG5hPPARawt (200 ng). FCS1 and FCS2
correspond to two different batches of FCS. Cotransfection
with pSG5 empty vector allows determination of the basal
level of expression of the J3-TK-Luc reporter vector, which
was set at 1. Empty vector plasmid was added to obtain
equal amounts of DNA per well.
Fig. 6. hPPARatr Does Not Bind to a PPRE
Gel retardation assays were performed on end-labeled apo
A-II J site oligonucleotide (which contains the apo A-II PPRE)
in the presence of unprogrammed reticulocyte lysate (unprog
lys), in vitro translated hPPARawt, hPPARatr, or mRXRa, in
the presence (1) or the absence (2) of 100-fold molar excess
of unlabeled apo A-II J site (competition) oligonucleotide as
described in Materials and Methods.
hPPARatr Affects Transcriptional Activity
of hPPARawt
Given the structure of the hPPARatr protein (without
transactivation domain), it is tempting to speculate
that this isoform may interfere with wild-type function
in a dominant negative manner. To test the transactivation activity of both hPPARatr and hPPARawt and
possible mutual interference, transient cotransfection
experiments were performed in HepG2 cells using a
apoA-II PPRE containing luciferase vector as a reporter (J3-TK-luc). Basal level of luciferase activity was
unchanged by cotransfection with hPPARatr whereas
hPPARawt activated transcription of the J3-TK-luc
(Fig. 7). When increasing amounts of the hPPARatr
were cotransfected with constant amounts of
hPPARawt, a slight decrease of J3-TK promoter activation by hPPARawt was observed in the presence of
high amounts of hPPARatr (Fig. 7). Surprisingly, the
repressive activity of hPPARatr on hPPARawt transactivation function was influenced by cell culture conditions using two different batches of FCS (see FCS1
and FCS2 in Fig. 7). Indeed, whereas neither batch of
FCS tested influenced basal or hPPARawt-stimulated
reporter activity, hPPARatr-repressive activity on
hPPARawt was more pronounced depending on the
FCS tested (Fig. 7). These results indicate that the
hPPARatr isoform is unable to transactivate a luciferase reporter construct driven by the PPRE-containing thymidine kinase (TK) promoter and that it can
efficiently repress hPPARawt activity depending on the
presence of certain serum factors.
Nuclear Localized hPPARatr Is a Potent
Repressor of Nuclear Receptor Activity
To determine whether the repressive activity of
hPPARatr on hPPARawt was linked to its subcellular
localization, immunocytochemistry studies were performed next. When CV-1 cells were transfected with
the pSG5hPPARawt vector, hPPARawt was predominantly detected in the nucleus with only minor staining
in the cytoplasm (Fig. 8B). Signals were specific since
pSG5-transfected cells did not give a significant fluorescence signal (Fig. 8A). In pSG5hPPARatr-transfected cells, the fluorescence signal was predominantly detected in the cytoplasm when cells were
cultured in FCS1 (Fig. 8C), which yielded low repressive activity. By contrast, a stronger signal was observed in the nucleus when cells were cultured in FCS2
(Fig. 8D) in which hPPARatr behaved as a strong repressor. These differences in subcellular localization were
confirmed by subcellular fractionation followed by immunoblotting analysis (data not shown). These results indicate that hPPARawt is predominantly localized in the
nucleus and that the extent of the repressive effect of
hPPARatr is correlated with nuclear localization.
To demonstrate unequivocally that the subcellular
localization influenced hPPARatr-repressive activity,
the SV40 large T antigen nuclear localization signal
(NLS) (52) was inserted 59 of the hPPARatr cDNA to
MOL ENDO · 1999
1542
Vol 13 No. 9
Fig. 8. Serum Factors Induce Nuclear Localization of hPPARatr
Subcellular localization of the hPPARawt and the hPPARatr protein by immunocytochemistry. A, CV-1 cells transfected with
pSG5 in FCS1; B, CV-1 cells transfected with pSG5hPPARawt in FCS1; C, CV-1 cells transfected with pSG5hPPARatr in FCS1;
D, CV-1 cells transfected with pSG5hPPARatr in FCS2; E, CV-1 cells transfected with pSG5-NLS in FCS1; F, CV-1 cells
transfected with pSG5-NLShPPARatr in FCS1.
produce a chimeric NLShPPARatr protein that should
be able to translocate into the nucleus. To determine
the subcellular localization of this protein, CV-1 cells
were transfected with pSG5-NLShPPARatr and subjected to immunocytochemistry analysis. A strong signal was detected inside the nucleus, indicating a massive translocation of NLShPPARatr into the nucleus
(Fig. 8F). The signal was specific since transfection
with pSG5-NLS resulted in undetectable fluorescence
intensity (Fig. 8E). Thus, the NLS system seems to be
able to efficiently direct hPPARatr into the nucleus.
Next, we analyzed the transcriptional activity of this
NLS-hPPARatr by transient cotransfection assays in
HepG2 cells (cultured in FCS1-containing medium)
using a Luc reporter construct that contains the apo
A-II PPRE (J site) in three copies in front of the heterologous TK promoter (J3-TK-Luc). The J3-TK-Luc vector was not activated by NLS-hPPARatr either in the
presence or in the absence of Wy 14,643 (Fig. 9A).
Cotransfection of hPPARawt resulted in a stronger
induction of Luc activity, which was repressed when
NLS-hPPARatr was cotransfected at a 1:1 ratio. Western blot analysis cell extracts of pSG5hPPARawt- and
pSG5hPPARatr-cotransfected cells at a 1:1 ratio demonstrated that both proteins were at approximately
equimolar levels (Fig. 9A, inset). Interestingly, this transcriptional activation was repressed by increasing
amounts of NLS-hPPARatr (Fig. 9A). The repressive
effect of NLS-hPPARatr was similar both in the absence and in the presence of Wy 14,643. To ensure
that the repressive effect of NLS-hPPARatr was not
due to a saturation effect of the protein import pathway caused by high levels of NLS peptide, the pSG5NLS expression vector was cotransfected in increasing amounts against a constant amount of the
hPPARawt. As shown in the right part of Fig. 9A, the
induction of Luc activity by hPPARawt was unaffected,
indicating that there was no effect of the NLS peptide
itself on hPPARawt-induced Luc activity. Consequently, these experiments confirm that nuclear localized hPPARatr has a repressive activity on hPPARawt
transactivation function, independent of the presence
of the NLS peptide. Given the fact that the nuclear
translocation of hPPARatr has to be induced, we can
consider that it acts as a negative modulator on
hPPARawt function. Finally, to define the specificity of
the negative effect of hPPARatr, we assessed its ability
to alter the transactivation of other nuclear receptor
family members, namely PPARg, hepatic nuclear factor-4 (HNF-4), related orphan receptor-a (RORa), and
glucocorticoid receptor-a (GRa). HepG2 cells were
transfected with the appropriate response elementdriven reporter plasmids in the presence of expression
vectors expressing either PPARg, HNF-4, RORa, and
GRa in the presence of hPPARatr (Fig. 9B). Cotransfection of hPPARatr almost completely repressed
hPPARawt and HNF-4, whereas PPARg and GRa activity was less affected and RORa activity was only
marginally repressed. These experiments indicate that
hPPARatr, if synthetized in significant amounts in vivo,
interferes with several nuclear receptor pathways, albeit to a different extent, and suggest that one mechanism through which hPPARatr exerts its inhibitory
effect may be by sequestration of a common coactivator necessary for nuclear receptor transcriptional
activity.
Dominant Negative Effect of hPPARawt May
Occur through Titration of the Coactivator CREBBinding Protein (CBP)
Since it was reported that CBP or p300 is a coactivator
of PPARa, PPARg, HNF-4, and GRa, as well as RORa
(53–57), we hypothesized that titration of CBP could
A Dominant Negative Human PPARa Isoform
1543
Fig. 10. Dominant Negative Effect of hPPARawt May Occur
through Titration of the Coactivator CBP
HepG2 cells, cultured in FCS2-containing medium, were
transfected with 1 mg of J3-TK-Luc, 200 ng of
pSG5hPPARawt, and 200 ng of pSG5hPPARatr. Increasing
amounts of pRSVCBP (0.2 mg, 0.4 mg, 1 mg) were added as
indicated. Empty vector plasmid was added to obtain equal
amounts of DNA per well. Activity of the reporter plasmid
J3-TK-Luc was set at 1.
Fig. 9. Nuclear Localized hPPARatr Is a Potent Repressor of
Nuclear Receptor Activity
A, HepG2 cells, cultured in FCS1-containing medium, were
cotransfected with 1 mg of Luc reporter construct driven by
the heterologous TK promoter containing three copies of the
apo A-II PPRE (J3). Activity of the reporter plasmid J3-TK-Luc
was set at 1. Increasing amounts (13, 53, 103) of pSG5NLShPPARatr were transfected in the presence or in the
absence of pSG5hPPARawt (200 ng). Proteins extracted from
HepG2 cells transfected at a 1:1 ratio of pSG5hPPARawt:
pSG5-NLShPPARatr were used for immunoblot analysis (inset). Increasing amounts of pSG5-NLS (13, 53, 103) were
added to constant amounts of pSG5hPPARawt (right of the
figure). Cells were treated with either Wy 14,643 (1 mM) or
vehicle (dimethylsulfoxide), and luciferase activity was measured as described in Materials and Methods. B, HepG2 cells,
cultured in FCS2-containing medium, were cotransfected
with 1 mg of J3-TK-Luc, and 200 ng of pSG5, pSG5hPPARa,
pSG5hPPARg, or pSG5HNF-4. pSG5RORa or GRa (200 ng)
was transfected in the presence of 1 mg of RORE-TK-Luc and
MMTV-TK-Luc, respectively. Increasing amounts (13, 53) of
pSG5hPPARatr were cotransfected as indicated. hPPARa,
PPARg, and GRa were activated using Wy 14,643 (1 mM),
BRL 49653 (0.1 mM) and dexamethasone (1 mM), respectively.
Empty vector plasmid was added to obtain equal amounts of
DNA per well. Activity of the reporter plasmid J3-TK-Luc,
RORE-TK-Luc, and MMTV-TK-Luc was set at 1.
constitute a mechanism by which hPPARatr might exert its inhibitory effect. To test this possibility, HepG2
cells, cultured in FCS2-containing medium, were
transfected with increasing amounts of CBP expression vector in the presence of hPPARawt alone or both
hPPARawt and hPPARatr (Fig. 10). As expected, transcriptional activity of hPPARawt was markedly enhanced by increasing amounts of CBP. Interestingly,
transcriptional repression of hPPARawt activity by
hPPARatr was restored by CBP cotransfection to levels attained by hPPARawt in the absence of hPPARatr.
These data indicate that hPPARatr may exert its dominant negative activity through a mechanism involving
the titration of common coactivator, such as CBP.
DISCUSSION
PPARa plays an important role in lipid and lipoprotein
metabolism by regulating a number of genes implicated in b-oxidation and plasma lipid transport (18). It
has been well documented that treatment with hypolipidemic drugs results in distinct effects depending on
species (humans vs. rodents) (18). Furthermore,
among patients, differential responses to these drugs
have been reported as well (48, 49). In an attempt to
determine whether these phenomena could be linked
to differences in expression level or to the existence of
variant forms of hPPARa, we performed expression
analysis of PPARa in humans. In this report we iden-
MOL ENDO · 1999
1544
tified a PPARa-truncated protein (hPPARatr) encoded
by a mRNA derived from an alternative splicing event,
results that extend observations published while this
work was in progress (58). We furthermore show that
the variant transcript appears to be present in all human cells and tissues expressing hPPARawt, but not in
rodent liver. The hPPARatr transcripts might result in
both in vitro and in vivo production of a protein as
shown by immunoblot analysis of extracts from transfected cells or from HepG2 cells and human primary
hepatocytes. Moreover, immunocytochemistry studies of transfected cells revealed that whereas transfected hPPARawt is constitutively imported into the
nucleus, hPPARatr cellular localization is influenced by
cell culture conditions. Furthermore, transient transfection studies with an NLS-containing hPPARatr expression vector showed that this variant can alter
hPPARawt transactivation capacities once translocated
in the nucleus. Finally, cotransfection experiments with
CBP suggested cofactor competition as a potential
mechanism of hPPARatr dominant negative activity.
PPARa function may be regulated at the transcriptional, the posttranscriptional, and at the protein level.
Transcriptional regulation of PPARa expression has
been shown to occur in rodents. In rat, PPARa expression is regulated by hormones, such as glucocorticoids and insulin, or by physiological stimuli such as
stress (59–61). However, very little is known about the
regulation of PPARa gene expression in man. In the
present study, we analyzed the variation of hPPARawt
expression in human liver, the principal site of PPARa
expression, among several individuals. Our results indicate that hPPARawt mRNA levels in the liver vary
substantially (from 1 to 3) among individuals. These
data suggest that PPARa is strongly regulated at the
gene level in human liver by genetic and/or environmental factors. By contrast, in adipose tissue, the level
of expression of hPPARa mRNA appears to be lower
and more constant among the individuals analyzed. It
will be interesting to determine whether transcriptional
regulation of PPARa also occurs through the use of
alternative promoters, as has been described for another member of the PPAR family, PPARg, which results both in mouse and man (9, 62, 63) in the production of two distinct proteins, PPARg1 and PPARg2,
with distinct activation capacities due to differences
only in their N-terminal amino acid sequence (64).
Our results furthermore demonstrate that the
hPPARa variant, identified in this and a previous study
(58), is generated by a posttranscriptional mechanism
involving alternative splicing resulting in the skipping
of exon 6. Generation of isoforms by alternative splicing has already been reported for other nuclear receptors such as the estrogen receptor (65), the glucocorticoid receptor (GR) (66), the thyroid hormone receptor
(67), or the vitamin D receptor (68). Alternative splicing
of these receptors results in the production of isoforms
with repressive activity on the wild-type (wt) receptor.
Therefore, the use of alternative splicing may constitute a general mechanism to diversify and to adapt
receptor action upon physiological changes.
Vol 13 No. 9
Alternative splicing could regulate wt receptor activity both at the gene expression level and at the
protein level. Our results obtained from the quantitative RNase protection assay demonstrate that the ratio
of wt to variant PPARa mRNA varies substantially in
liver among individuals. The variation in this ratio appears, however, to be mainly due to variation in
hPPARawt mRNA levels. Nevertheless, factors that influence this splicing event may interfere with the
PPARa-signaling pathway by posttranscriptional regulation of the RNA level of hPPARawt. By contrast, in
adipose tissue, wt-variant mRNA levels appear to be
fairly constant and close to 1:1. Therefore, high levels
of hPPARatr mRNA may result in lesser activity of
PPARa and its ligands, such as fibrates, in this tissue.
In addition, if present in significant amounts in vivo,
hPPARatr may interfere with PPARa-signaling pathways at the protein level. Immunoblot experiments
revealed that both wt and truncated PPARa proteins
can be produced in vivo by HepG2 cells and by human
primary hepatocytes. Moreover, transient expression
assays suggest that hPPARatr may display a repressive effect on hPPARawt transactivation function. As
shown by immunocytochemistry experiments, modulation of this effect might be explained by the amount
of hPPARatr that translocates into the nucleus. Furthermore, when hPPARatr protein is fused to a NLS
peptide, nuclear translocation of hPPARatr is induced,
and its repressive effect on hPPARawt transactivation
function is strongly enhanced. This repressive activity
of nuclear hPPARatr could be due to several mechanisms, such as competition for DNA binding, sequestering of RXR by the formation of inactive heterodimers, or titration of nuclear factors necessary for
transcription activation (69). The first two mechanisms
can be excluded since, as shown by gel retardation
assays, hPPARatr does not bind to a PPRE sequence
nor does it inhibit DNA binding of the PPARawt/RXRa
complex even when added at a 4-fold excess. These
data are in line with recently published findings that
demonstrate that the C-terminal part of PPARa is required for heterodimerization and DNA binding (70).
The third mechanism of inhibitory effect on hPPARawt
transactivation might be due to the titration of a cofactor binding to the N-terminal half of the protein,
thereby limiting the amounts of common coactivators
available for transactivation by hPPARawt and other
nuclear receptors. Such a mechanism is probable
since cotransfection of CBP could completely revert
the inhibition of hPPARawt transcriptional activity by
hPPARatr. Furthermore, hPPARatr was also found to
exert transcriptional repressive activity on nuclear receptors such as PPARg, HNF-4, GRa, and RORa,
which all share CBP/p300 as a common coactivator
(53–57). Interestingly, hPPARatr interfered only marginally with RORa transcription activity, suggesting
that a selectivity of hPPARatr toward certain nuclear
receptors exists. The hPPARatr protein structure consists of the A/B and C domains. Similarly, an estrogen
receptor isoform with identical structure has been previously described also being derived from alternative
splicing and interfering with wt receptor function (71).
A Dominant Negative Human PPARa Isoform
Although most cofactors thus far identified interact
with the LBD, the existence of cofactors binding to the
A/B domain has been also demonstrated, such as for
thyroid hormone receptor (72). Thus, the expression
level of hPPARatr, as well as the selectivity of interaction with cofactors such as CBP, will determine the
specificity of hPPARatr action on transcription signaling by hPPARawt and other transcription factors. However, hPPARatr dominant negative action was more
effective on hPPARawt, suggesting that hPPARatr may
also compete specifically for cofactors other than CBP
alone binding the N terminus of hPPARa.
To exert its repressive activity, hPPARatr needs to
be transported into the nucleus. Transport of transcription factors and nuclear receptors into the nucleus is an essential step in target gene transcription
regulation. In eukaryotes, the nuclear membrane constitutes a barrier and thereby offers a way to regulate
gene expression. Transport of nuclear proteins across
the nuclear pore complex (73) is a regulated process,
but the mechanisms by which proteins enter the nucleus are not yet fully understood. To date, the best
studied mechanism is the NLS-mediated system.
Many nuclear proteins carry NLS. Given the fact that
hPPARatr was observed both in the cytoplasm and in
the nucleus, with higher intensity of hPPARawt in the
nucleus, the constitutive NLS of hPPARawt is probably
comprised in the E/F domain. However, immunocytochemistry experiments, showing that the cellular distribution of hPPARatr is influenced by cell culture conditions, suggest the existence of an inducible NLS in
hPPARatr that could influence receptor trafficking.
Such a regulated nuclear entry has been demonstrated for a number of proteins, including GR (74), for
which two NLS appear to mediate regulated nuclear
import. Certain proteins, such as mitogen-activated
protein kinase kinase, are able to translocate rapidly
into the nucleus upon mitogenic stimulation and thus
to regulate gene expression through phosphorylation
of transcription factors (75, 76). It is therefore possible
that the nuclear translocation of hPPARatr may constitute a specific and controlled mechanism that modulates its activity. Several mechanisms of induced nuclear translocation may be suggested. First, protein
phosphorylation constitutes a mechanism by which a
cryptic NLS may be unmasked. Such mechanism has
been shown to occur for the Xenopus nuclear factor
Xnf7 (77) implicated in gene regulation during development. Second, PPARa might show high affinity for a
cytoplasmic retention protein. Such cytoplasmic retention occurs for NF-KB, which is sequestered in the
cytoplasm by its inhibitor IK-B. Under activation by a
variety of cytokines and mitogenic factors (e.g. interleukin-1, tumor necrosis factor-a), the NFK-B/IK-B
complex dissociates and NFK-B is imported into the
nucleus (78). Finally, hPPARatr might be translocated
to the nucleus as a part of a cell cycle-dependent
process as shown for viral Jun (79). In this respect, it
is interesting to note that, upon transfection with
hPPARatr, a small number of cells consistently stained
more pronouncedly in the nucleus. It will therefore be
of interest to determine whether nuclear import of
1545
hPPARatr is subject to regulation, for instance during
different stages of the cell cycle.
In conclusion, we have described a PPARa variant
transcript that is expressed in human cell lines and
tissues, but not in rodents. This transcript may give
rise to the production of a truncated receptor form of
hPPARa in human liver cells. In addition, upon nuclear
translocation, hPPARatr becomes a potent inhibitor of
hPPARawt as well as other transcription factors via a
mechanism implicating CBP sequestration. The generation of PPARa variant transcript by alternative
splicing may regulate PPARa signaling both at the
mRNA and at the protein level. Regulation of variant
transcript generation, together with its powerful inducible repressive effect on hPPARawt function upon induction of nuclear translocation, might represent a
new approach for modulation of nuclear receptor
function. It will be of interest to determine whether
hPPARatr also contributes to the species-specific differential response to PPARa activators. Moreover, differences in the level of expression of hPPARatr might
be implicated in the heterogeneity in response to fibrates among different patients. Further studies are
warranted to determine the physiological role of
hPPARatr protein in humans.
MATERIALS AND METHODS
Tissue and Cell RNA Extraction and RT-PCR Analysis
Total cellular RNA from tissues and cells was prepared by the
guanidinium thiocyanate/phenol-chloroform method (80). For
analysis of hPPARawt expression by RT-PCR, total RNA was
reverse transcribed using random hexamer primers (Amersham Pharmacia Biotech, Buckinghamshire, UK) and Superscript reverse transcriptase (Life Technologies, Inc., Paisley,
UK), and the resulting cDNA product was subsequently PCR
amplified (35 cycles of 1 min at 94 C, 1 min at 55 C, 1 min at
72 C) using the following primers: oligo 15, 59-GAA GTT CAA
GAT CAA AGT GCC AGC-39; oligo 35, 59-TCT GAA GAG TTC
CTG CAA GAA ATG G-39; oligo 45, 59-ACA CGC TTT CAC
CAG CTT CGA CCC-39; oligo 65, 59-GAC GAA TGC CAA
GAT CTG AGA AAG C-39; oligo 33, 59-GTG ATG ACC GAG
CCA TCT GAG C-39; oligo 63, 59-TGT ATT GTT ACT GGC
CTT TCC TGA GAG G-39; oligo 73, 59-AGC ATC CCG TCT
TTG TTC ATC-39; oligo 83, 59-CGT CTC CTT TGT AGT GCT
GTC AGC-39. Each primer corresponds to a region well conserved between the mouse and human PPARa-coding sequence (5, 13, 14). The resulting products were separated
alongside suitable molecular markers on a 1.5% agarose gel
and visualized by ethidium bromide staining.
Cloning and Construction of Recombinant Plasmids
The hPPARawt cDNA containing the entire open reading
frame of wt hPPARa was excised from the pCMX vector (14)
by NotI digestion and blunt ending followed by NruI digestion, and was subsequently cloned into the blunt-ended
BamHI site of the pSG5 expression vector (Stratagene, La
Jolla, CA), giving pSG5hPPARawt. The hPPARatr-containing
pSG5 expression vector was constructed by insertion of the
shorter fragment, lacking exon 6, produced by RT-PCR amplification using the 35/73 pair of primers (Fig. 1). The PCR
fragment was digested with MluI and SphI and inserted into
the MluI and SphI sites of the pSG5hPPARawt vector result-
MOL ENDO · 1999
1546
ing in the pSG5hPPARatr construct. To increase translation
efficiency of both hPPARawt and hPPARatr, a Kozak consensus sequence (81) was inserted at the 59-ATG of both
pSG5hPPARawt and pSG5hPPARatr constructs using a PCR
method. Briefly, the sense primer 59-C CAT GGA TCC ACC
ATG GTG GAC ACG GAA AGC-39, which includes a BamHI
site upstream from the Kozak consensus sequence, and the
antisense primer 73 were used to amplify a fragment of 1026
bp, which was subsequently digested with BamHI and MluI.
After agarose gel purification, this fragment was inserted into
the BamHI and MluI sites of the pSG5hPPARawt or
pSG5hPPARatr plasmids. To introduce a nuclear localization
signal (NLS) into hPPARatr, a cDNA encoding the SV40 large
T antigen NLS (52) was inserted in frame 59 of the hPPARatr
cDNA. The Kozak consensus containing pSG5hPPARatr was
digested with NcoI, blunt ended, and then BglII digested. The
resulting fragment was inserted into pSG5-NLS linearized by
BamHI digestion followed by blunt ending and BglII digestion
yielding pSG5-NLShPPARatr.
Determination of Intron-Exon Boundaries and
DNA Sequencing
The BAC clone containing the hPPARa gene was a kind gift
from B. Wilkison (Glaxo Wellcome Inc., Research Triangle
Park, NC). Sequencing reactions were performed using the
Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin
Elmer Corp., Foster City, CA) and an automatic ABI prism 377
sequencer (Perkin Elmer Corp.). 59- And 39-intron-exon
boundaries of exon 6 were sequenced using the oligonucleotides 59-GAC CCG GGC TTT GAC CTT GTT G-39 and 59GAC GAA TGC CAA GAT CTG AGA AAG C-39, respectively,
as primers.
RNase Protection Assay
The hPPARa antisense probe used was designed to distinguish between the wt and the variant transcripts. A pBSSKhPPARa clone containing the full-length cDNA sequence was
SmaI and SacI digested, and the resulting fragment (which
spans a region located between bp 800–983 according to
nucleotide numbering used in Ref. 14 was isolated and subcloned into a pBSKS vector (Stratagene). The resulting
hPPARa riboprobe covers 54 bp of vector sequence plus 152
bp that are common to both hPPARawt and hPPARatr and 36
bp of exon 6 that are only present in the wt transcript. A 36B4
riboprobe was used as an internal control (82). Therefore, the
36B4 cDNA was PstI/Sau3AI digested and subcloned in the
PstI/BamHI sites of the pBSKS vector. The construct was
further linearized with XhoI. This control probe contains 140
bp of vector sequence and 170 bp of 36B4 sequence. Identity
and orientation of both clones were confirmed by sequencing
analysis. Both PPARa and 36B4 riboprobes were subsequently produced by in vitro transcription of the respective
XhoI-linearized pBSKS plasmid in the presence of [32P]-CTP
(800 Ci/mmol) using T3 RNA Polymerase and the RNA transcription Kit (Stratagene). Different molar ratios of [32P]CTP to
cold CTP were used (1:166 for 36B4 and 1:0 for hPPARa) to
synthesize each riboprobe, which allowed signal comparison
of the corresponding mRNA on the same gel. The RNase
protection assay was carried out using the HybSpeed RNase
protection Kit (Ambion, Inc. Austin, TX). Twenty micrograms
of total RNA were hybridized simultaneously to hPPARa (6 3
104 cpm) and 36B4 (103 cpm) antisense probes. RNA samples were electrophoresed on a 5% denaturing polyacrylamide gel and visualized by autoradiography. Protected fragments representing hPPARawt, hPPARatr, and 36B4 mRNA
were quantified on a GS525 Phosphorimager (Bio-Rad Laboratories, Inc. Hercules, CA). The radioactivity was corrected
for differences in the radiolabeled nucleotide content between hPPARawt and hPPARatr and subsequently normalized to the internal 36B4 control.
Vol 13 No. 9
Cell Culture and Transfections
Human hepatoma Hep G2 and Cos-1 cells were obtained
from the European Collection of Animal Cell Culture (Porton
Down, Salisbury, UK). Cells were grown in DMEM, supplemented with 2 mM glutamine and 10% (vol/vol) FCS, in a 5%
CO2 humidified atmosphere at 37 C. Medium was changed
every other day. Stimuli were dissolved in dimethylsulfoxide.
Control cells received vehicle only. All transfections were
performed with a mixture of plasmids, which contained in
addition to the reporter (1 mg) and expression vector (0.2–2
mg), 1 mg of Cytomegalovirus-driven b-galactosidase expression vector as control for transfection efficiency. All samples
were complemented to an equal amount of plasmid DNA
using empty pSG5 vector. Cells were transfected at 50–60%
confluence in 60-mm dishes by the calcium phosphate coprecipitation procedure. After a 4-h incubation period, cells
were washed with PBS and then refed with fresh medium and
treated with Wy 14,643 (Chemsyn, Lenexa, KS), vehicle, or
dexamethasone (Sigma, St. Louis, MO) as indicated. Cells
were harvested after 24 h incubation. The luciferase activity in
cell extracts was determined using a luciferase assay system
(Promega Corp., Madison, WI) following the supplier’s instruction. Transfection experiments were performed in triplicate and repeated at least three times.
Protein Extract Preparation
Total cellular extracts were made from 5–10 3 106 cells. Cells
were washed twice with ice-cold PBS, scraped off in 5 ml
ice-cold PBS, and collected by centrifugation for 5 min at 800
rpm at 4 C. The pellet was resuspended in 100 ml ice-cold
lysis buffer (1% Nonidet P-40, 0.5% sodium desoxycholate,
0.1% SDS in PBS), and protease inhibitors were freshly
added (5 mg/ml leupeptin, 5 mg/ml pepstatin, 5 mg/ml EDTANa2, 1 mM benzamidine, 5 mg/ml aprotinin, and 0.5 mM phenylmethylsulfonyl fluoride). Cells were immediately transferred into 1.5-ml tubes, vigorously but quickly (10 sec)
vortexed, and allowed to swell on ice for 15 min. The cell
extract was then obtained by centrifugation (5 min at 13000
rpm at 4 C), and the supernatant was transferred into new
tubes, aliquoted, and stored at 280 C. The samples were
continuously kept at 4 C during the entire extraction
procedure.
SDS-PAGE and Western Blotting
Electrophoresis of samples (100 mg of total protein) and
prestained mol wt markers was performed on 10% SDSpolyacrylamide gels (Minigel system, Bio-Rad Laboratories,
Inc.) under reducing conditions (sample buffer containing 10
mM dithiothreitol). Proteins were electrophoretically blotted
onto a nitrocellulose membrane. Nonspecific binding sites
were blocked with 10% skim milk powder in TNT buffer (20
mM Tris, 55 mM NaCl, 0.1% Tween 20), overnight at 4 C. The
membrane was probed with a PPARa rabbit polyclonal antibody developed against either a N-terminal PPARa peptide
(amino acids 10–56) or a C-terminal PPARa peptide and
diluted in 5% skim milk-TNT, for 4 h at room temperature.
After washes, membranes were incubated with peroxidaseconjugated antirabbit antibody (Diagnostics Pasteur, MarnesLa-Coquette, France) diluted 1:4000 and visualized using
a chemiluminescent system (ECL, Amersham Pharmacia
Biotech, Buckinghamshire, UK).
Immunocytochemistry
Cells were cultured on microscopy coverslips of 14 mm diameter (sterilized with ethanol) under standard conditions.
Cells were washed with a solution of 0.01 M Tris, 0.5 M NaCl
(pH 7.4) and then fixed with 3% paraformaldehyde, 2% su-
A Dominant Negative Human PPARa Isoform
crose in PBS, pH 7.5, for 15 min at room temperature. After
three washes in Tris-NaCl, cells were incubated with 0.1 M
lysine for 4 h at room temperature to avoid nonspecific fluorescent signals. Cells were permeabilized using methanolacetone (1:1 vol/vol) for 5 min at room temperature. The
nonspecific staining was blocked by incubation in Tris-NaCl,
0.5% ovalbumin (TNO), and 1% of normal goat serum for 30
min at room temperature. The preparations were incubated
overnight at 4 C with primary antibody (N-terminal PPARa
antibody) diluted in TNO and followed by three washes with
TNO (3 3 30 min) and by incubation with secondary fluorescein isothiocyanate-conjugated antirabbit IgG antibody (dil.1:
100 in TNO) for 2 h at room temperature. After successive
washes in TNO, Tris-NaCl, and distilled water, the preparations were assembled on microscope slides for analysis with
a Leitz DMR fluorescence microscope (Leica Corp., Nussloch, Germany).
In Vitro Translation and Gel Retardation Assays
pSG5hPPARawt, pSG5hPPARatr and pSG5mRXRa were in
vitro transcribed with T7 polymerase and translated using the
rabbit reticulocyte lysate sytem (Promega Corp.). For gel
retardation assay, a synthetic double stranded oligonucleotide, which spans nucleotides 2737 to 2715 of the human
apo A-II gene upstream regulatory sequences and contains a
PPRE, was end labeled and used as probe (41). The PPAR
and RXR proteins were incubated in a total volume of 20 ml
for 15 min on ice with 2.5 mg of poly-dI-dC and 1 mg of herring
sperm DNA in TM buffer (10 mM Tris-HCl, pH 7.9, 40 mM KCl,
10% glycerol, 0.05% Nonidet P-40, and 1 mM dithiothreitol).
The 32P-radiolabeled oligonucleotide was then added, and
the mixture was incubated 20 min on ice. For competition
experiments, 100-fold molar excess of cold probe was added
just before the labeled oligonucleotide. The complexes were
resolved on 5% polyacrylamide gels in 0.253 TBE buffer (90
mM Tris-borate, 2.5 mM EDTA, pH 8.3) at 4 C. Gels were dried
and exposed overnight at 270 C to x-ray film (XOMAT-AR,
Eastman Kodak, Rochester, NY).
Acknowledgments
Helpful discussions with Gérard Torpier, Ngoc Vu-Dac, and
Piet De Vos are gratefully acknowledged. We thank Odile
Vidal, Lluis Fajas, and Isabelle Saves for expert technical
assistance. The BAC clone, hepatoma cell line RNA, and
human liver samples were kind gifts from B. Wilkison, H.
Will/S.F. Chang, and U. Beisiegel/V. Kosykh, respectively. We
thank R. Mukherjee for providing a hPPARa cDNA clone. The
CBP clone is the kind gift of D. Hum. M. Dauça is acknowledged for providing anti-PPARa antibody raised against the
C-terminal part of PPARa. We are grateful to J. J. Berthelon
for providing BRL 49653.
Received July 13, 1998. Re-revision received April 22,
1999. Accepted May 26, 1999.
Address requests for reprints to: Dr. Bart Staels, U.325
INSERM, Département d’Athérosclérose, Institut Pasteur,
1 Rue Calmette, 59019 Lille, France. E-mail Bart.Staels@
pasteur-lille.fr.
This research was sponsored by grants from INSERM,
ARCOL (Comité Français de Coordination des Recherches
sur l’Athérosclérose et le Cholestérol), Biomed 2 Concerted
action (Grant PL963324), Fondation pour la Recherche Médicale, the Région Nord-Pas de Calais, and the European Community (Grant ERBFMBICT983214).
* On sabbatical leave from the Center for Research, Prevention and Treatment of Atherosclerosis, Department of
Medicine, Hadassah University Hospital, Jerusalem, Israel.
1547
REFERENCES
1. Tsai M-J, O’Malley B 1994 Molecular mechanisms of
action of steroid/thyroid receptor superfamily members.
Annu Rev Biochem 63:451–486
2. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P,
Schutz G, Umesono K, Blumberg B, Kastner P, Mark E,
Chambon P, Evans RM 1995 The nuclear receptor
superfamily: the second decade. Cell 83:835–839
3. Dreyer C, Krey G, Keller H, Givel F, Helftenbein G, Wahli
W 1992 Control of the peroxisomal b-oxidation pathway
by a novel family of nuclear hormone receptors. Cell
68:879–887
4. Göttlicher M, Widmark E, Li Q, Gustafsson JA 1992 Fatty
acids activate chimera of the clofibric acid-activated receptor and the glucocorticoid receptor. Proc Natl Acad
Sci USA 89:4653–4657
5. Isseman I, Green S 1990 Activation of a member of the
steroid hormone receptor superfamily by peroxisome
proliferators. Nature 347:645–650
6. Chen F, Law SN, O’Malley BW 1993 Identification of two
mPPAR related receptors and evidence for the existence
of five subfamily members. Biochem Biophys Res Commun 196:671–677
7. Zhu Y, Alvares K, Huang Q, Rao MS, Reddy JK 1993
Cloning of a new member of the peroxisome proliferator
activated receptor gene family from mouse liver. J Biol
Chem 268:26817–26820
8. Kliewer SA, Forman BM, Blumberg B, Ong ES,
Borgmeyer U, Mangelsdorf DJ, Umesono K, Evans RM
1994 Differential expression and activation of a family of
murine peroxisome proliferator-activated receptors. Proc
Natl Acad Sci USA 91:7355–7359
9. Tontonoz P, Hu E, Graves RA, Budavari AI, Spiegelman
BM 1994 mPPARg2: tissue-specific regulator of an adipocyte enhancer. Genes Dev 8:1224–1234
10. Amri E-Z, Bonino F, Ailhaud G, Abumrad NA, Grimaldi PA
1995 Cloning of a protein that mediates transcriptional
effects of fatty acids in preadipocytes. J Biol Chem
270:2367–2371
11. Aperlo C, Pognonec P, Saladin R, Auwerx J, Boulukos K
1995 Isolation and characterization of the hamster peroxisomal proliferator activated receptor hPPARg, a
member of the nuclear hormone receptor superfamily.
Gene 162:297–302
12. Schmidt A, Endo N, Rutledge SJ, Vogel R, Shinar D,
Rodan GA 1992 Identification of a new member of the
steroid hormone receptor superfamily that is activated by
a peroxisome proliferator and fatty acids. Mol Endocrinol
6:1634–1641
13. Sher T, Yi HF, McBride W, Gonzalez FJ 1993 cDNA
cloning, chromosomal mapping, and functional characterization of the human peroxisome proliferator activated
receptor. Biochemistry 32:5598–5604
14. Mukherjee R, Jow L, Noonan D, McDonnell DP 1994
Human and rat peroxisome proliferator activated receptors (PPARs) demonstrate similar tissue distribution but
different responsiveness to PPAR activators. J Steroid
Biochem Mol Biol 51:157–166
15. Greene ME, Blumberg B, McBride OW, Yi HF, Kronquist
K, Kwan K, Hsieh L, Greene G, Nimer SD 1995 Isolation
of the human peroxisome proliferator activated receptor
gamma cDNA: expression in hematopoietic cells and
chromosomal mapping. Gene Expression 4:281–299
16. Elbrecht A, Chen Y, Cullinan CA, Hayes N, Leibowitz MD,
Moller DE, Berger J 1996 Molecular cloning, expression
and characterization of human peroxisome proliferator
activated receptors g1 and g2. Biochem Biophys Res
Commun 224:431–437
17. Lambe KG, Tugwood JD 1996 A human peroxisomeproliferator activated receptor-g is activated by inducers
MOL ENDO · 1999
1548
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
of adipogenesis, including thiazolidinedione drugs. Eur
J Biochem 239:1–7
Schoonjans K, Staels B, Auwerx J 1996 The peroxisome
proliferator activated receptors (PPARs) and their effects
on lipid metabolism and adipocyte differentiation. Biochim Biophys Acta 1302:93–109
Forman BM, Tontonoz P, Chen J, Brun RP, Spiegelman
BM, Evans RM 1995 15-Deoxy-D12,14 prostaglandin J2
is a ligand for the adipocyte determination factor PPARg.
Cell 83:803–812
Kliewer SA, Lenhard JM, Willson TM, Patel I, Morris DC,
Lehman JM 1995 A prostaglandin J2 metabolite binds
peroxisome proliferator-activated receptor g and promotes adipocyte differentiation. Cell 83:813–819
Devchand PR, Keller H, Peters JM, Vazquez M, Gonzalez
FJ, Wahli W 1996 The PPARa-leukotriene B4 pathway to
inflammation control. Nature 384:39–43
Forman BM, Chen J, Evans RM 1997 Hypolipidemic
drugs, polyunsaturated fatty acids, and eicosanoids are
ligands for peroxisome proliferator-activated receptors a
and d. Proc Natl Acad Sci USA 94:4312–4317
Kliewer SA, Sundseth SS, Jones SA, Brown PJ, Wisely
GB, Koble CS, Devchand P, Wahli W, Willson TM, Lenhard JM, JML 1997 Fatty acids and eicosanoids regulate
gene expression through direct interactions with peroxisome proliferator-activated receptors a and g. Proc Natl
Acad Sci USA 94:4318–4323
Lehmann JM, Lenhard JM, Oliver BB, Ringold GM,
Kliewer SA 1997 Peroxisome proliferator-activated receptors a and g are activated by indomethacin and other
non-steroidal anti-inflammatory drugs. J Biol Chem
272:3406–3410
Lee SST, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz
DL, Fernandez-Salguero PM, Westphal H, Gonzalez FJ
1995 Targeted disruption of the a isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol 15:3012–3022
Staels B, Van Tol A, Andreu T, Auwerx J 1992 Fibrates
influence the expression of genes involved in lipoprotein
metabolism in a tissue-selective manner in the rat.
Arterioscler Thromb Vasc Biol 12:286–294
Staels B, van Tol A, Verhoeven G, Auwerx J 1990 Apolipoprotein A-IV messenger ribonucleic acid abundance
is regulated in a tissue-specific manner. Endocrinology
126:2153–2163
Hertz R, Bishara-Shieban J, Bar-Tana J 1995 Mode of
action of peroxisome proliferators as hypolipidemic
drugs, suppression of apolipoprotein C-III. J Biol Chem
270:13470–13475
Staels B, Vu-Dac N, Kosykh V, Saladin R, Fruchart JC,
Dallongeville J, Auwerx J 1995 Fibrates down-regulate
apolipoprotein C-III expression independent of induction
of peroxisomal acyl co-enzyme A oxidase. J Clin Invest
95:705–712
Staels B, Peinado-Onsurbe J, Auwerx J 1992 Downregulation of hepatic lipase gene expression and activity
by fenofibrate. Biochim Biophys Acta 1123:227–230
Staels B, van Tol A, Skretting G, Auwerx J 1992 Lecithin:
cholesterol acyltransferase gene expression is regulated
in a tissue-selective manner by fibrates. J Lipid Res
33:727–735
Peters JM, Hennuyer N, Staels B, Fruchart JC, Fievet C,
Gonzalez FJ, Auwerx J 1997 Alterations in lipoprotein
metabolism in PPARa-deficient mice. J Biol Chem
272:27307–27312
Staels B, Auwerx J 1992 Perturbation of developmental
gene expression in rat liver by fibric acid derivatives:
lipoprotein lipase and a-fetoprotein as models. Development 115:1035–1043
Osumi T, Wen JK, Hashimoto T 1991 Two cis-acting
regulatory elements in the peroxisome proliferatorresponsive element enhancer region of rat acyl-CoA
Vol 13 No. 9
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
oxidase gene. Biochem Biophys Res Commun
175:866–871
Tugwood JD, Isseman I, Anderson RG, Bundell KR,
McPheat WL, Green S 1992 The mouse peroxisome
proliferator activated receptor recognizes a response element in the 59 flanking sequence of the rat acyl CoA
oxidase gene. EMBO J 11:433–439
Bardot O, Aldridge TC, Latruffe N, Green S 1993 PPARRXR heterodimer activates a peroxisome proliferator response element upstream of the bifunctional enzyme
gene. Biochem Biophys Res Commun 192:37–45
Hijikata M, Ishii N, Kagamiyama H, Osumi T, Hashimoto
T 1990 Rat peroxisomal 3-ketoacyl-CoA thiolase gene:
occurrence of two closely related but differentially regulated genes. J Biol Chem 265:4600–4606
Muerhoff AS, Griffin KJ, Johnson EF 1992 The peroxisome proliferator activated receptor mediates the induction of CYP4A6, a cytochrome P450 fatty acid v-hydroxylase by clofibric acid. J Biol Chem 267:19051–19053
Rodriguez JC, Gil-Gomez G, Hegardt FG, Haro D 1994
Peroxisome proliferator activated receptor mediates induction of the mitochondrial 3-hydroxy-3-methylglutarylCoA synthase gene by fatty acids. J Biol Chem
269:18767–18772
Vu-Dac N, Schoonjans K, Laine B, Fruchart JC, Auwerx
J, Staels B 1994 Negative regulation of the human apolipoprotein A-I promoter by fibrates can be attenuated by
the interaction of the peroxisome proliferator-activated
receptor with its response element. J Biol Chem
269:31012–31018
Vu-Dac N, Schoonjans K, Kosykh V, Dallongeville J, Fruchart JC, Staels B, Auwerx J 1995 Fibrates increase
human apolipoprotein A-II expression through activation
of the peroxisome proliferator-activated receptor. J Clin
Invest 96:741–750
Schoonjans K, Peinado-Onsurbe J, Lefebvre AM, Heyman R, Briggs M, Cayet D, Deeb S, Staels B, Auwerx J
1996 PPARa and PPARg activators direct a tissuespecific transcriptional response via a PPRE in the lipoprotein lipase gene. EMBO J 15:5336–5348
Svoboda DJ, Azarnoff DL, Hignite CE 1966 Response of
hepatic microbodies to a hypolipidemic agent, ethylchlorophenoxyisobutyrate (CPIB). J Cell Biol 30:442–450
Moody DE, Reddy JK 1978 Hepatic peroxisome (microbody) proliferation in rats fed plasticizers and related
compounds. Toxicol Appl Pharmacol 45:487–504
De La Iglesia FA, Lewis JE, Buchanan RA, Marcus EL,
McMahon G 1982 Light and electron microscopy of liver
in hyperlipoproteinemic patients under long-term gemfibrozil treatment. Atherosclerosis 43:19–37
Blumcke S, Schwartzkopff W, Lobeck H, Edmondson
NA, Prentice DE, Blane GF 1983 Influence of fenofibrate
on cellular and subcellular liver structure in hyperlipidemic patients. Atherosclerosis 46:105–116
Hanefeld M, Kemmer C, Kadner E 1983 Relationship
between morphological changes and lipid-lowering action of p-chlorphenoxyisobutyric acid (CPIB) on hepatic
mitochondria and peroxisomes in man. Atherosclerosis
46:239–246
Chandler HA, Batchelor AJ 1994 Ciprofibrate and lipid
profile. Lancet 344:128–129
McLeod AJ, Warren RJ, Armitage M 1994 Ciprofibrate
and lipid profile. Lancet 344:955
Gearing KL, Crickmore A, Gustafsson JA 1994 Structure
of the mouse peroxisome proliferator activated receptor
a gene. Biochem Biophys Res Commun 199:255–263
Chinetti G, Griglio S, Antonucci M, Pineda Torra I, Delerive P, Majd Z, Fruchart J-C, Chapman J, Najib J, Staels
B 1998 Activation of proliferator-activated receptors a
and g induces apoptosis of human monocyte-derived
macrophages. J Biol Chem 273:25573–25581
A Dominant Negative Human PPARa Isoform
52. Kalderon D, Roberts BL, Richardson WD, Smith AE 1984
A short amino acid sequence able to specify nuclear
location. Cell 39:499–509
53. Dowell P, Ishmael JE, Avram D, Peterson VJ, Nevrivy DJ,
Leid M 1997 p300 functions as a coactivator for the
peroxisome proliferator-activated receptor a. J Biol
Chem 272:33435–33443
54. Mizukami J, Taniguchi T 1997 The antidiabetic agent
thiazolidinedione stimulates the interaction between
PPARg and CBP. Biochem Biophys Res Commun
240:61–64
55. Yoshida E, Aratani S, Itou H, Miyagishi M, Takiguchi M,
Osumu T, Murakami K, Fukamizu A 1997 Functional
association between CBP and HNF4 in transactivation.
Biochem Biophys Res Commun 241:664–669
56. Sheppard KA, Phelps KM, Williams AJ, Thanos D, Glass
CK, Rosenfeld MG, Gerritsen ME, Collins T 1998 Nuclear
integration of glucocorticoid receptor and nuclear factorkappaB signaling by CREB-binding protein and steroid
receptor coactivator-1. J Biol Chem 273:29291–29294
57. Lau P, Bailey P, Dowhan DH, Muscat GE 1999 Exogenous expression of a dominant negative RORalpha1 vector in muscle impairs differentiation: RORalpha1 directly
interacts with p300 and myoD. Nucleic Acids Res
27:411–420
58. Palmer CNA, Hsu M-H, Griffin KJ, Raucy JL, Johnson EF
1998 Peroxisome proliferator activated receptor-a expression in human liver. Mol Pharmacol 53:14–22
59. Lemberger T, Staels B, Saladin R, Desvergne B, Auwerx
J, Wahli W 1994 Regulation of the peroxisome proliferator-activated receptor a gene by glucocorticoids. J Biol
Chem 269:24527–24530
60. Steineger HH, Sorensen HN, Tugwood JD, Skrede S,
Spydevold O, Gautvik KM 1994 Dexamethasone and
insulin demonstrate marked and opposite regulation of
the steady-state mRNA level of the peroxisomal proliferator-activated receptor (PPAR) in hepatic cells. Hormonal modulation of fatty acid-induced transcription. Eur
J Biochem 225:967–974
61. Lemberger T, Saladin R, Vazquez M, Assimacopoulos F,
Staels B, Desvergne B, Wahli W, Auwerx J 1996 Expression of the peroxisome proliferator-activated receptor a
gene is stimulated by stress and follows a diurnal rhythm.
J Biol Chem 271:1764–1769
62. Zhu Y, Qi C, Korenberg JR, Chen X-N, Noya D, Rao MS,
Reddy JK 1995 Structural organization of mouse peroxisome proliferator activated receptor g (mPPARg) gene:
alternative promoter use and different splicing yield two
mPPARg isoforms. Proc Natl Acad Sci USA
92:7921–7925
63. Fajas L, Auboeuf D, Raspe E, Schoonjans K, Lefebvre
AM, Saladin R, Najib J, Laville M, Fruchart JC, Deeb S,
Vidal-Puig A, Flier J, Briggs MR, Staels B, Vidal H,
Auwerx J 1997 Organization, promoter analysis and expression of the human PPARg gene. J Biol Chem
272:18779–18789
64. Werman A, Hollenberg A, Solanes G, Bjorbaek C, VidalPuig AJ, Flier JS 1997 Ligand-independent activation
domain in the N terminus of peroxisome proliferatoractivated receptor g. J Biol Chem 272:20230–20235
65. Fuqua SAW, Fitzgerald SD, Craig Allred D, Elledge RM,
Nawaz Z, McDonnell DP, O’Malley BW, Greene GL,
McGuire WL 1992 Inhibition of estrogen receptor action
by a naturally occurring variant in human breast tumor.
Cancer Res 52:483–486
1549
66. Oakley RH, Sar M, Cidlowski JA 1996 The human glucocorticoid receptor b isoform. Expression, biochemical
properties and putative function. J Biol Chem
271:9550–9559
67. Koenig DJ, Lazar MA, Hodin RA, Brent GA, Larsen R,
Chin WW, Moore DD 1989 Inhibition of thyroid hormone
action by a non hormone binding c-erbA protein generated by alternative mRNA splicing. Nature 337:659–661
68. Ebihara K, Masuhiro Y, Kitamoto T, Suzawa M, Uematsu
Y, Yoshizawa T, Ono T, Harada H, Matsuda K, Hasegawa
T, Masushige S, Kato S 1996 Intron retention generates
a novel isoform of the murine vitamin D receptor that acts
in a dominant negative way on the vitamin D signaling
pathway. Mol Cell Biol 16:3393–3400
69. Yen PM, Chin WW 1994 Molecular mechanisms of dominant negative activity by nuclear hormone receptors.
Mol Endocrinol 8:1450–1454
70. IJpenberg A, Jeannin E, Wahli W, Desvergne B 1997
Polarity and specific sequence requirements of peroxisome proliferator-activated receptor (PPAR)/retinoid X
receptor heterodimer binding to DNA. J Biol Chem
272:20108–20117
71. Satoshi I, Hoshino SJ, Miyoshi H, Akishita M, Hosoi T,
Orimo H, Ouchi Y 1996 Identification of a novel isoform
of estrogen receptor, a potential inhibitor of estrogen
action, in vascular smooth muscle cells. Biochem Biophys Res Commun 219:766–772
72. Hollenberg AN, Monden T, Madura JP, Lee K, Wondisford FE 1996 Function of nuclear co-repressor protein on
thyroid hormone response elements is regulated by the
receptor A/B domain. J Biol Chem 271:28516–28520
73. Feldherr CM, Kallenbach E, Schultz N 1984 Movement of
a karyophilic protein through the nuclear pores of
oocytes. J Cell Biol 99:2216–2222
74. Picard D, Yamamoto KR 1987 Two signals mediate hormone-dependent nuclear localization of the glucocorticoid receptor. EMBO J 6:3333–3340
75. Seger R, Krebs EG 1995 The MAPK signalling cascade.
FASEB J 9:726–735
76. Jaaro H, Rubinfeld H, Hanoch T, Seger R 1997 Nuclear
translocation of mitogen-activated protein kinase kinase
(MEK1) in response to mitogenic stimulation. Proc Natl
Acad Sci USA 94:3742–3747
77. Miller M, Reddy BA, Kloc M, Li XX, Dreyer C, Etkin LD
1991 The nuclear-cytoplasmic distribution of the Xenopus nuclear factor, xnf7, coincides with its state of phosphorylation during early development. Development
113:569–575
78. Baeurle PA, Baltimore D 1988 IK-B: a specific inhibitor of
NFK-B transcription factor. Science 242:540–546
79. Chida K, Vogt PK 1992 Nuclear translocation of viral jun
but not cellular jun is cell cycle dependent. Proc Natl
Acad Sci USA 89:4290–4294
80. Chomczynski P, Sacchi N 1987 Single step method for
RNA isolation by acid guanidinium-thiocyanate-phenolchloroform extraction. Anal Biochem 162:156–159
81. Kozak M 1986 Point mutation define a sequence flanking
the AUG initiation codon that modulates translation by
eukaryotic ribosomes. Cell 44:283–292
82. Masiakowski P, Breathnach R, Bloch J, Gannon F, Krust
A, Chambon P 1982 Cloning of cDNA sequences of
hormone-regulated genes from MCF-7 human breast
cancer cell line. Nucleic Acids Res 10:7895–7903