 |
INTRODUCTION |
The acyl-CoA-binding protein
(ACBP)1 is an intracellular
lipid-binding protein that selectively binds medium and long chain acyl-CoA esters with high affinity. The protein has been found in all
eukaryotes investigated to date from mammals to yeasts and plants and
is structurally and functionally highly conserved through evolution
(1). However, with the exception of the recently identified
testis-specific protein known as endozepine-like peptide, ACBPs diverge
significantly at a structural and functional level from other mammalian
lipid-binding proteins. Whereas members of the fatty acid-binding
protein family have a so-called 
-barrel structure and bind a broad
spectrum of fatty acids and fatty acid derivatives, ACBP has a four
-helix bundle structure and binds specifically acyl-CoA esters by
docking the hydrocarbon chain into the lipophilic pocket and using the
CoA moiety as a lid (1).
In vitro investigations have shown that ACBP is able to
protect efficiently acyl-CoA esters from hydrolysis by thioesterases and that it can function both as acceptor and donor of acyl-CoA esters
(2). Thus, ACBP relieves acyl-CoA inhibition of long chain acyl-CoA
synthetase, acetyl-CoA carboxylase, and adenine nucleotide transferase
(2) and regulates acyl-CoA:cholesterol acyltransferase activity (3).
Furthermore, ACBP is able to transport acyl-CoA esters and donate these
to mitochondrial -oxidation (4-6), microsomal glycerolipid
synthesis (4), and phospholipid synthesis (7, 8). Overexpression of
ACBP in yeast significantly increases the acyl-CoA pool size,
indicating that ACBP can generate an intracellular acyl-CoA pool
(9, 10).
The mammalian ACBP gene is a typical housekeeping gene (11), and ACBP
appears to be ubiquitously expressed from early stages of mammalian
embryogenesis (12) as well as in adult tissues (reviewed in Ref. 13).
However, the level of ACBP differs markedly among different cell types.
Cell types with a high level of ACBP expression include hepatocytes,
steroidogenic cells, and adipocytes.
The expression of the mammalian ACBP gene is regulated with the feeding
status. Fasting of rats results in a significant decrease in ACBP
mRNA and protein in the liver, whereas the level in heart and
kidney is unaffected (6, 14). In contrast, conditions that induce
de novo fatty acid synthesis appear to induce ACBP expression (15). Similarly, ACBP expression is induced during in
vitro differentiation of 3T3-L1 preadipocytes (16), a process that
is accompanied by a marked triglyceride accumulation and de
novo fatty acid synthesis. The proximal promoter of the human ACBP
gene has been shown to contain a sterol regulatory element that was
activated by sterol regulatory element-binding protein (SREBP-1)/adipocyte determination and differentiation factor 1 in
transient transfections (17). This transcription factor has been shown
recently to be involved in the coordinated induction by insulin of
genes involved in lipogenesis (18, 19), and it is therefore possible
that SREBP-1 regulates ACBP in response to fasting and feeding.
Interestingly, ACBP expression is not only increased by conditions that
favor de novo fatty acid synthesis. High fat feeding of rats
leads to elevated levels of ACBP in the liver (6), and various
peroxisome proliferators, which are known as potent inducers of liver
mitochondrial and peroxisomal -oxidation, induce ACBP mRNA and
protein expression in the liver (14, 20) and ACBP protein expression in
isolated hepatocytes (21). The stimulation of liver lipid catabolism by
these compounds is known to be mediated by an activation of the
peroxisome proliferator-activated receptor (PPAR) (22, 23),
suggesting that ACBP may be a PPAR target gene.
Although ACBP expression has been reported to increase in the liver by
prolonged exposure to PPAR activators, a functional PPAR-response
element (PPRE) has never been identified in the ACBP gene, and it is
unknown whether the ACBP gene is a direct target of the PPARs. In this
report we demonstrate unequivocally that ACBP is a PPAR
target gene,
which is activated by PPAR-selective ligands in adipose tissue as
well as in adipocyte cell lines, and we demonstrate the existence of a
functional PPRE in intron 1 of the rat ACBP gene. This PPRE confers
PPAR and PPAR responsiveness to the promoter and is functionally
conserved between humans and rodents. In adipocytes, the intronic PPRE
mediates transcriptional activation in response to treatment with the
PPAR-selective ligand BRL49653. Chromatin immunoprecipitation shows
that the PPRE binds PPAR/RXR in vivo. These results show
that ACBP expression is directly regulated by PPAR and possibly also
PPAR through the intronic PPRE.
| |
MATERIALS AND METHODS |
Animal Experiments--
C57BL-Ks
db/db mice at the age of 12 weeks were
divided into groups of 6 animals and dosed once daily for 10 days with
rosiglitazone (1.0, 3, or 10 mg/kg/dosing) or vehicle (0.2% CMC + 0.4% Tween 80 in saline) by oral gavage. At the end of the dosing
period, animals were killed by decapitation, and white adipose tissue (epididymal fat pads) was removed and frozen in liquid nitrogen, and
RNA was isolated by RNazol (BioSite, Täby, Sweden) according to
the manufacturer's instructions. Equal amounts of total RNA from all
animals in each group were pooled. All animal experiments were
conducted in accordance with the Danish law.
Quantitative PCR--
Pooled total RNA isolated from the white
adipose tissue was DNase-treated, and three independent reverse
transcription reactions were performed using Superscript II reverse
transcriptase (Invitrogen) following the manufacturer's instructions.
mRNA expression levels were determined using real time fluorescent
detection in a Lightcycler instrument (Roche Molecular Biochemicals)
and the following primer combinations: 5'-AGCCAACTGATGAAGAGATG-f-3' and
5'-AGGCATTATGTCCTCACAGG-r-3' for ACBP, 5'-ATGCCTTTGTGGGAACCTGG-f-3';
5'-CCCAGTTTGAAGGAAATCTCGG-r-3' for adipocyte lipid-binding protein
(aP2). mRNA expression levels were determined twice in each 1st
strand synthesis reaction and normalized to the expression levels of
18 S rRNA as described by the vendor (Applied Biosystems).
Cell Culture--
3T3-L1, NIH-3T3, and 293 cells were cultured
in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen) containing
4500 mg/liter glucose supplemented with 100 µg/ml streptomycin, 62.5 µg/ml penicillin, 8 µg/ml biotin, and 8 µg/ml pantothenic acid.
Standard 3T3-L1 and NIH-3T3 culture media contained 10% calf serum
(Sigma). Standard 293 media contained 10% fetal calf serum (FCS)
(Invitrogen). Medium was exchanged every other day.
Differentiation of 3T3-L1 cells was obtained by exposing 2-day
post-confluent cells (designated day 0 cells) to DMEM containing 10%
FCS (Invitrogen) supplemented with 1 µM dexamethasone
(Sigma), 0.5 mM 3-isobutyl-1-methylxanthine (Aldrich), and
1 µg/ml insulin (Roche Molecular Biochemicals). At day 2 cells were
fed DMEM containing FCS and 1 µg/ml insulin, thereafter cells were
maintained in DMEM and FCS.
Northern Blotting--
RNA was isolated (24) from 3T3-L1 cells
at day 
2, 0, 1-4, 6, 8, and 10 of differentiation. At day 10, cells
were exposed to Me2SO, 1 µM BRL49653 in
Me2SO, 15 µg/ml cycloheximide, or 1 µM
BRL49653 + 15 µg/ml cycloheximide. RNA was isolated after 2, 6, and
12 h of incubation.
RNA was analyzed by Northern blotting with a 32P-labeled
rat ACBP cDNA fragment as probe. Signals were quantified by
PhosphorImaging. The membranes were stripped and reprobed with a
labeled mouse adipocyte lipid-binding protein (ALBP) cDNA probe and
finally with the human 28 S rRNA probe.
Western Blotting--
3T3-L1 cells from day 2, 0, 1-4, 6, 8, and 10 were lysed in 0.5 ml of 2.5% SDS sample buffer per 10-cm dish.
Lysates were subjected to SDS-PAGE. Approximately 20 µg of cellular
protein was loaded per lane. The separated proteins were transferred to a polyvinylidene difluoride membrane and stained with Ponceau S for
control of equal loading. The membranes were blocked in 5% (w/v)
nonfat dry milk, incubated with affinity-purified rabbit anti-mouse
ACBP for 1 h, and horseradish peroxidase-conjugated secondary
antibody (Dako) for another hour. Immunoreactive protein bands were
detected by enhanced chemiluminescence (Amersham Biosciences).
Electrophoretic Mobility Shift Assay--
Nuclei were purified
from 3T3-L1 preadipocytes or adipocytes by a modification (25) of the
procedure of Dignam et al. (26). Nuclear extracts were
prepared as described by Lavery and Schibler (27) using a 1× NUN
solution (0.3 M NaCl, 1 M urea, 1% Nonidet P-40, 25 mM HEPES, pH 7.9, and 1 mM
dithiothreitol). Protein concentrations were determined using Bradford
protein assay reagent (Bio-Rad).
Rat liver nuclear extract was prepared from 3-month-old Sprague-Dawley
rats weighing ~300 g according to the procedure described by Gorski
et al. (28) except that the second purification of the
nuclei was omitted, and protease inhibitors (1 µg/µl leupeptin, 1 µg/µl antipain, 1 µg/µl pepstatin, and 0.01 TIU/ml aprotinin) were added to all buffers just prior to use.
In vitro translations were performed using TNT
kit according to the recommendations of the manufacturer (Promega).
Double-stranded oligonucleotides corresponding to the rat
ACBP intron 1 DR-1, rat ACBP upstream DR-1, and human ACBP intron 1 DR-1 were labeled using [-32P]ATP and polynucleotide
kinase (Roche Molecular Biochemicals).
Nuclear extracts (2-4 µg) or in vitro translated proteins
were incubated 20 min on ice in binding buffer (10 mM Tris,
pH 8.0, 40 mM KCl, 1 mM dithioerythritol, 4%
glycerol, and 0.05% Nonidet P-40, 2.4 µg of poly(dI-dC)).
Subsequently 2 × 105 cpm of 32P-labeled
oligonucleotide was added, and the mixture was incubated for 20 min at
room temperature. For competition assays, 10-fold excess of a
homologous or heterologous competitor was mixed with the labeled probe
and added to the preincubation mixture. Free DNA and DNA-protein
complexes were resolved by electrophoresis in 0.5× TBE, 5%
polyacrylamide gels.
Plasmids--
Plasmids used in transient transfections were
pTK-3x- PPRE-luc (29), pTK-3xACBP-PPRE-luc, and pTK-luc. The plasmid
pTK3xACBP-PPRE was made from pTK-3xPPRE-luc by replacement of the three
copies of ACO PPRE with three copies of the rat ACBP intronic PPRE
inserted in the same orientation and with the same spacing as the
original ACO PPREs. Rat ACBP promoter reporter constructs
rACBP(392/+1)-luc, rACBP(1512/+1)-luc, and rACBP(1535/+1)-luc
were constructed by inserting the respective promoter fragments in the
pGL3-basic vector (Promega). Rat ACBP promoter construct
rACBP(392/+979)-luc, rACBP(1512/+979)-luc,
rACBP(1535/+979)-luc, rACBP(392/+979)
PPRE-luc, rACBP(1512/+979)PPRE-luc, and rACBP(1535/+979)-luc were
constructed by inserting the respective promoter/exon 1 and intron
1 sequence in pGL3-basic so that the reading frame of ACBP exon 2 was fused in-frame with that of luciferase. The intronic PPRE
(GGGACAGAGGTCA) was mutated to GTTTTTTTTGTCA in the PPRE
constructs. Human ACBP promoter reporter constructs
hACBP(516/+5)-luc, hACBP(516/+1136)-luc, and
hACBP(516/+1136)PPRE-luc were constructed similarly. The intronic
PPRE (GGGACAGAGGTCG) was mutated to GTTTTTTTTGTCG in the PPRE
constructs. For expression of PPARs and RXR, the expression plasmids
pSG5-PPAR (30), pSG5-PPAR
(31), pSPORT-PPAR2 (32), and
pCMX-mRXR (33) were used.
pCMV--galactosidase-control (Promega) was used for normalization,
and pBluescriptKS (Stratagene), pSG5 (Stratagene), and/or pcDNA
were used to adjust to equal DNA load and promoter load, respectively.
Transient Transfections--
NIH-3T3 and 293 cells were
transfected at 50-70% confluency in 60-mm dishes or 6-well plates
using the DC-Chol lipofection procedure (34) and a total of 5 or 2.5 µg, respectively, of DNA/plate/well. Following 6 h of
incubation with DNA mixture, the medium was changed to DMEM and 10%
resin-charcoal-stripped calf serum (for NIH-3T3) or FCS (for 293)
supplemented with PPAR activator or vehicle (Me2SO) alone.
3T3-L1 cells were transfected at day 4 of differentiation using the
LipofectAMINE Plus procedure (Invitrogen). Transfections were performed
in 12-well plates with a total of 1 µg/well. After 3 h of
incubation with the DNA mixture, medium was changed to serum-free DMEM
supplemented with PPAR activator or vehicle (Me2SO) alone.
Cells were harvested 20 h later in lysis buffer (Tropix), and the
lysates were stored at 80 °C. All transfections were performed as
triplicates. Luciferase and -galactosidase assays were performed as
described previously (35).
The PPAR activators Wy14643 (Calbiochem), BRL49653 (Novo Nordisk A/S),
and activator L-165041 (Merck) were used in the indicated concentrations to activate PPAR, PPAR, and PPAR, respectively. Serum was stripped with AG-1X-8 resin and activated charcoal powder as
described previously (35).
Chromatin Cross-linking--
Day 4 3T3-L1 adipocytes were
removed from plates by trypsinization and resuspended in DMEM + 10%
FCS (18 ml per four 15-cm plates). Cells were fixed in vivo
for 10 min at room temperature by addition of 2 ml of formaldehyde
solution (11% formaldehyde, 0.1 M NaCl, 1 mM
EDTA, 0.5 mM EGTA, and 50 mM HEPES, pH 8.0). Fixation was stopped by addition of glycine to 0.125 M
final concentration, and cells were collected by centrifugation
(1000 × g, 4 °C for 5 min), washed once in
phosphate-buffered saline, and incubated in 10 ml of Triton lysis
buffer for 10 min at 4 °C (0.25% Triton X-100, 10 mM
EDTA, 0.5 mM EGTA, 10 mM Tris-HCl, pH 8.0).
Chromatin was collected by centrifugation (1000 × g,
4 °C, 5 min), washed once in 10 ml of NaCl washing buffer (0.2 M NaCl, 10 mM EDTA, 0.5 mM EGTA, 10 mM Tris-HCl, pH 8.0), resuspended in 3 ml of resuspension buffer (10 mM EDTA, 0.5 mM EGTA, 10 mM Tris-HCl, pH 8.0), and transferred to 15-ml tubes.
Shearing of chromatin was done by sonicating each sample 7 times for
30 s at 0 °C using a Branson 250 sonicator (output control set
at 5). Samples were adjusted to 0.5% sarcosyl and swirled for 10 min.
Cell debris was collected by centrifugation for 5 min at 13,000 × g. Samples were adjusted to 1.42 g/cm3 CsCl and
brought to a volume of 4 ml with 1.42 g/cm3 CsCl in
resuspension buffer. The cross-linked chromatin complexes were
separated from free protein, lipids, DNA, and RNA by isopycnic ultracentrifugation for 72 h at 20 °C, 40,000 rpm in a Beckman SWTi60 rotor. Fractions of 0.5 ml were collected from the bottom of the
gradient with syringe and needle, and the samples containing chromatin
were identified on a 0.8% agarose gel. The samples containing chromatin were pooled and dialyzed overnight at 4 °C against
dialysis buffer (5% glycerol, 1 mM EDTA, 0.5 mM EGTA, 10 mM Tris-HCl, pH 8.0). Dialyzed
chromatin was sonicated again for 30 s as above. The cross-linked
chromatin was aliquoted and stored at 80 °C.
Chromatin Immunoprecipitation--
120 µl of protein A/G beads
(Santa Cruz Biotechnology) were washed three times in
immunoprecipitation (IP) dilution buffer (1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 300 mM NaCl and 1:50 Complete protease inhibitor mixture (Roche
Molecular Biochemicals)). Beads were then incubated with 10 µg of
sheared salmon sperm DNA and 60 µg of bovine serum albumin for 30 min at 4 °C in IP dilution buffer, washed three times in IP dilution buffer, and finally resuspended in IP dilution buffer. To 200-µl aliquots of isolated cross-linked chromatin an equal volume of 2-fold
concentrated IP dilution buffer was added, and the chromatin was
incubated with half of the protein A/G beads for 1 h at 4 °C on
a rotating wheel. Beads were removed by centrifugation (13,000 × g, 20 s, 4 °C) and the supernatant was incubated
overnight at 4 °C with 5 µl of antibody (PPAR polyclonal (Santa
Cruz Biotechnology) or pan-RXR polyclonal antibody (Santa Cruz
Biotechnology)) on a rotating wheel. Complexes containing PPAR and
RXR were immunoprecipitated by adding the other half of the beads and
incubating 11/2 h at 4 °C on a rotating wheel. Immune
complexes were washed 5 min on a rotating wheel with buffer 1 (1%
Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH
8.1, and 150 mM NaCl), twice with buffer 2 (1% Triton
X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 500 mM NaCl), and twice with TE buffer. Immune complexes were
eluted from the beads with 200 µl of 1% SDS, 0.1 M
NaHCO3 for 15 min. The elution was repeated; eluates were
combined; 16 µl of 5 M NaCl was added, and chromatin was
decross-linked for 4 h at 65 °C. DNA was purified by
phenol/chloroform extraction, precipitated with 20 µg of glycogen as
carrier, and finally dissolved in 50 µl of TE buffer. For input control, 200 µl of input chromatin was decross-linked,
phenol/chloroform-extracted, precipitated, and dissolved in 100 µl of TE.
Real Time PCR on Immunoprecipitated DNA--
Precipitated DNA
was quantified using the GeneAmp 5700 Sequence Detection system (PE
Biosystem) and real time PCR kit (Eurogenetics). Primer sets were
designed to amplify the intronic PPRE of the ACBP gene, the PPRE of the
adipocyte lipid-binding protein promoter (32), and the PPRE of the
lipoprotein lipase promoter (36), respectively. As a negative
background control, a primer set located ~8 kb downstream of the
intronic PPRE of the ACBP gene was used. Primers were designed so that
all PCR products were between 81 and 87 bp. Primers were as follows:
ACBP intron 1 PPRE, forward 5'-TCCCACTTGCCTCTCCCTAA-3' and
reverse 5'-CAGCTGGTCCCTTCCTACAGG-3'; LPL PPRE, forward
5'-CCTCCCGGTAGGCAAACTG-3' and reverse 5'-AACGGTGCCAGCGAGAAG-3'; ALBP
PPRE (ARE7), forward 5'-GAGAGCAAATGGAGTTCCCAGA-3' and reverse 5'-TTGGGCTGTGACACTTCCAC-3'; and background control, forward
5'-ACACCACTGGCCGTGATGTT-3' and reverse 5'-CATCGGCGTACTCTGCTGTG-3'. The
PCR amplification was carried out as follows: initial denaturation at
95 °C for 10 min, followed by 40 cycles of denaturation at 96 °C
for 15 s and combined annealing and extension at 60 °C for
60 s. The GeneAmp 5700 software was used to perform analysis of
the real time fluorescence signal from SyBR Green I bound to
double-stranded DNA. A threshold cycle was determined for each
sample, using the exponential growth phase and base-line data of the
fluorescent amplification plots. Dissociation curves were
subsequently used to identify PCR products. To correct for differences
in efficiency of the different PCRs, all PCR signals from
immunoprecipitated DNA were normalized to PCR signals from
non-precipitated input DNA. The normalized signal from the PCR obtained
with the background control primer set was arbitrarily set to 1, and
other PCR signals were expressed as fold above background.
| |
RESULTS |
ACBP Expression Is Increased in Adipose Tissue of
db/db Mice Treated with BRL49653--
Peroxisome
proliferators have been shown to increase the expression of ACBP in rat
liver (14, 20), and we have shown previously (16) that ACBP expression
is significantly increased during in vitro adipocyte
differentiation, a process during which the expression of a number of
PPAR target genes is up-regulated. To investigate whether ACBP
expression can be induced by PPAR ligands in vivo, 12 week-old db/db mice were fed either vehicle or BRL49653 at
different concentrations for 10 days. The db/db model is
commonly used to study the antidiabetic effects of
thiazolidinediones, and ACBP mRNA levels are similar in
adipose tissue of db/db mice and wild type C57BL6 mice
(results not shown). Adipose tissue was isolated, and the expression of
ACBP was compared with that of the ALBP, which is a well established
PPAR target gene (32). Both ACBP and ALBP are expressed at very high
levels in adipocytes (37, 38); however, despite these high basal
levels, the expression of ACBP as well ALBP mRNA was significantly
and dose-dependently increased by BRL49653 (Fig.
1). This suggests that ACBP might be a
PPAR target gene.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1.
Expression of ACBP is induced in parallel
with ALBP in adipose tissue of db/db mice treated with
the PPAR ligand BRL49653. Twelve-week-old
db/db mice were dosed daily with BRL49653 (1, 3, or 10 mg/kg/dosing) or vehicle for 10 days. RNA was isolated from epididymal
fat pads, and after reverse transcription, mRNA expression levels
were determined using real time PCR. Each bar represents the
average RNA level and S.E. of the mean from six animals. Values are
normalized to expression levels of 18 S rRNA. RNA level of
vehicle-treated animals was set to 1.
|
|
ACBP Expression Is Induced by PPAR Ligands in the Absence of
Protein Synthesis--
We have reported previously (16) that ACBP
mRNA and protein are induced during adipocyte differentiation of
3T3-L1 cells. Fig. 2A confirms
this observation using standard differentiation conditions instead of
the limited differentiation mixture that was used in the original
publication. To investigate whether the ACBP gene is directly regulated
by PPAR, we added the PPAR-selective ligand BRL49653 to fully
differentiated 3T3-L1 cells, which express high amounts of PPAR. The
PPAR ligand significantly induced the expression of ACBP in parallel
with ALBP (Fig. 2B). The observation that no significant
increase is observed at the 6-h but only at the 12-h time point
probably reflects the fact that the transcripts of both ACBP and ALBP
are highly abundant already in the differentiated adipocytes. Addition
of cycloheximide did not prevent induction of either ACBP or ALBP by
the PPAR ligand, indicating that ACBP similar to ALBP is a direct
PPAR target gene.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 2.
Expression of ACBP is induced in parallel
with ALBP by the PPAR ligand BRL49653 in
mature adipocytes in the absence of protein synthesis. 3T3-L1
cells were cultured and differentiated following the standard
differentiation protocol. A, RNA and protein extracts were
prepared at different times during adipocyte differentiation. RNA was
analyzed by Northern blotting using 32P-labeled rat ACBP
cDNA as probe. Protein extracts were analyzed by Western blotting
using affinity-purified rabbit anti-rat ACBP antibody. B, at
day 10, BRL49653 (1 µM) was added in the presence or
absence of cycloheximide (CHX) and incubated
(Inc. and Incub) for 2, 6, and 12 h. RNA was
isolated, and the expression of ACBP and ALBP was determined by
northern blotting. Signals were quantified by PhosphorImaging and
normalized to 28 S rRNA.
|
|
Intron 1 Contains a Potential PPRE, Which Is Conserved between Rat,
Mouse, and Humans--
We and others (39, 40) have reported the
identification of potential regulatory elements in the promoter of the
rat ACBP gene, but functional elements have not yet been described. We reported the identification of a potential PPRE in the rat ACBP gene at
position 1525 and showed that it bound PPAR/RXR heterodimers in
electrophoretic mobility shift assays (39). However, cloning and
sequencing of the mouse and human ACBP genes showed that it was neither
conserved in the human nor in the mouse ACBP sequence (Fig.
3A). Careful analysis of the
entire 10.9 kb that had been sequenced from the rat ACBP gene (from
~2.2 kb upstream of exon 1 to 0.2 kb downstream of the last exon)
revealed an almost perfect DR-1 (direct repeat of AGGTCA with 1 bp
spacing) in intron 1. This DR-1 conformed better to the PPRE consensus,
both with respect to core sequence as well as with respect to its
5'-flanking sequence, than any other sequence within the 10.9 kb.
Importantly, this DR-1 is well conserved between rat, mouse, and human
(Fig. 3A). Unlike most PPREs, the intronic DR-1 has a
"G" as spacer between the two direct repeats. Interestingly the
ARE7 PPRE, which is the most important of the two PPREs in the enhancer
of ALBP (41), shows considerable sequence identity with the ACBP DR-1
and has a G as spacer as well.
View larger version (45K):
[in this window]
[in a new window]
|
Fig. 3.
Potential PPREs in the rat ACBP gene.
A, structure of the rat ACBP promoter region and intron 1 with two potential PPREs indicated. The potential PPRE in intron 1 is
conserved between human and rodents, whereas the upstream potential
PPRE reported previously (Elholm et al. (39)) is only
present in the rat gene. The numbers, 1535, 1512, 392, +1 and
+979, refer to position relative to translation start site and indicate
the extension of the promoter fragments cloned into the
promoter-reporter constructs. B, comparison of the potential
PPREs of the rat, mouse, and human ACBP genes with well characterized
functional PPREs. The potential upstream PPRE of the rat gene is shown
in italics. The indicated consensus is from Ref. 53.
|
|
To investigate whether the intronic DR-1 could bind PPAR·RXR
complexes in vitro, we performed electrophoretic mobility
shift assays using adipocyte and hepatocyte nuclear extracts as well as
in vitro translated proteins. As seen in Fig.
4A, the intronic DR-1 of the
rat and human ACBP genes gives rise to retarded bands with adipocyte
and hepatocyte nuclear extracts. The bands comigrate with the band
formed with the previously identified upstream DR-1 in the rat
promoter. The band formed with adipocyte nuclear extracts comigrates
with in vitro translated PPAR/RXR and is supershifted by PPAR as well as RXR antibodies (Fig. 4B). In
keeping with the very low levels of PPAR expression in
preadipocytes, the retarded band is not formed with preadipocyte
nuclear extract.
View larger version (56K):
[in this window]
[in a new window]
|
Fig. 4.
The potential PPRE in intron 1 binds
PPAR·RXR complexes in vitro.
Electrophoretic mobility shift assays using the intronic DR-1 and
in vitro translated proteins or nuclear extracts.
A, protein complexes from adipocyte and rat liver nuclear
extracts bind to the intronic DR-1 of both the rat and the human ACBP
gene as well as to the upstream DR-1 of the rat gene. B, the
intronic DR-1 binds a protein complex from 3T3-L1 adipocytes but not
preadipocytes. This complex comigrates with in vitro
translated PPAR/RXR and is effectively supershifted by antibodies
against PPAR and RXR.
|
|
The Intronic DR-1 Is a Functional PPRE in the Context of a
Heterologous Promoter--
To investigate whether the DR-1 in intron 1 of the rat ACBP gene could function as a PPRE upstream of a
heterologous promoter, we cloned three copies of this DR-1 in front of
the thymidine kinase gene basal promoter. This was done by replacing
the three copies of the rat acyl-CoA oxidase (ACO) PPRE in
TK-3xPPRE-luc (29) with three copies of the rat ACBP DR-1 keeping
orientation and spacing of the DR-1s identical to that of the ACO
PPREs. The resulting constructs were transfected into NIH-3T3 cells, a
cell line that we routinely use for investigating activation by
specific PPAR subtypes because it has low endogenous levels of all
PPARs. As shown in Fig. 5, the
multimerized rat ACBP DR-1 functions as a PPAR-, PPAR-, and
PPAR-responsive PPRE in NIH-3T3 cells.
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 5.
The DR-1 in intron 1 of the ACBP gene is a
functional PPRE in the context of a heterologous promoter. NIH-3T3
cells were transfected with either TK-luc reporter, 3×
rACBP-PPRE-TK-luc or 3× rACO-PPRE-TK-luc, and pSV40-mPPAR,
pSV40-mPPAR2, pSV40-mPPAR, or pCMV-HNF4 expression vectors as
indicated. pCMV-RXR was cotransfected with all PPAR constructs and
pCMV--galactosidase expression vector as control. Following
transfection cells were treated for 24 h with 1 µM
BRL49653, 100 µM Wy14643, 1 µM L-165041, or
Me2SO vehicle as indicated. All transfections were
performed as triplicate DC-Chol lipofections in 6-cm plates with a
total of 5 µg of DNA/plate. Empty expression vectors were added to
compensate for promoter load. Luciferase values have been normalized to
-galactosidase values. Transfection values are shown as fold
induction compared with a transfection with TK-luc reporter and
normalization vector alone. Transfections were done in triplicate.
Standard deviations are indicated. The results are representative of
three or more independent experiments.
|
|
Interestingly, the ACBP PPRE appeared to be much less efficient than
the ACO PPRE in mediating PPAR transactivation, whereas the
efficiency of mediating PPAR transactivation was similar for the
ACBP and ACO PPREs. PPAR transactivation was mediated more
efficiently by the ACBP PPRE than by the ACO PPRE. Although the ACBP
PPRE is a relatively poor PPRE for PPAR transactivation, PPAR
still transactivates the 3xACBP-TK-luc severalfold better than
PPAR2. Whether this is due to a much higher expression level of
PPAR or to a higher transactivation potential of PPAR in these
cells is not clear.
Because electrophoretic mobility shift assays showed that in
vitro translated HNF4 bound strongly to the ACBP DR-1 (results not shown), we tested whether this site could mediate HNF4
transactivation. However, when assayed in the context of the thymidine
kinase gene basal promoter, HNF4 was unable to direct
transcriptional activation through this DR-1 element and gave only a
very minor activation through the ACO PPRE. Parallel transfections with
an HNF4 reporter construct showed that HNF4 was indeed
transcriptionally active in NIH-3T3 cells (data not shown).
The Intronic DR-1 Confers PPAR and PPAR Responsiveness to the
Rat ACBP Promoter--
To investigate the functionality of the rat
intronic PPRE in situ, we prepared constructs in
which the rat ACBP promoter regions, exon 1 and intron 1, were fused
with the luciferase gene such that the luciferase gene was inserted
in-frame with the ACBP reading frame of exon 2 (Fig.
6A). The NIH-3T3 cell line was
used due to their low levels of endogenous PPARs. The
rACBP(392/+1)-luc as well as the rACBP(392/+979)-luc reporter
constructs were highly active in NIH-3T3 cells. However, cotransfection
of PPAR/RXR and addition of the PPAR-selective ligand Wy14643
significantly increased luciferase expression in NIH-3T3 cells.
Similarly, cotransfection with PPAR/RXR and addition of the
PPAR-selective ligand BRL49653 increased luciferase expression,
whereas cotransfection with PPAR/RXR and addition of the
PPAR-selective ligand L-165041 had no effect on promoter activity.
HNF4 was also in the ACBP promoter context unable to activate
transcription in NIH-3T3 cells. Mutation of the intronic PPRE abolished
the PPAR-mediated transactivation demonstrating that this PPRE is a
functional PPRE in the natural gene context as well as when linked to a
heterologous promoter. Thus, the intronic PPRE confers PPAR and
PPAR responsiveness to the ACBP promoter. However, despite the fact
that the multimerized intronic PPRE in the thymidine kinase gene basal
promoter context is responsive to PPAR in NIH-3T3 cells, PPAR is
unable to activate transcription from the ACBP promoter via this
PPRE.
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 6.
The DR-1 in intron 1 but not the upstream
DR-1 of the rat ACBP gene is a functional PPRE in
situ, which is activated by
PPAR/RXR and
PPAR2/RXR. A,
the seven different constructs used for analysis of the rat intron 1 PPRE and upstream DR-1, in rACBP(392/+979)PPRE-luc the rACBP
intron 1 PPRE, have been mutated. B, NIH-3T3 cells were
transfected with either rACBP(392/+1)-luc, rACBP(392/+979)-luc, or
rACBP(392/+979)PPRE-luc reporter constructs and pSV40-mPPAR
pSV40-mPPAR2, pSV40-mPPAR, or pCMV-HNF4 expression vectors as
indicated. pCMV-RXR was cotransfected with all PPAR constructs, and
pCMV--galactosidase expression vector was used as control. Following
transfection, cells were treated for 24 h with 1 µM
BRL49653, 100 µM Wy14643, 1 µM L-165041, or
Me2SO vehicle as indicated. Transfection and normalization
performed as indicated in Fig. 5. Normalized luciferase values are
shown as fold induction compared with a transfection with reporter
alone. C, NIH-3T3 cells were transfected with either
rACBP(1512/+1)-luc, rACBP(1512/+979)-luc, or rACBP(1535/+1)-luc
and rACBP(1535/+979)-luc reporter constructs and pSV40-mPPAR
expression vector as indicated. pCMV-RXR was cotransfected with all
PPAR constructs, and pCMV--galactosidase expression vector was used
as control. Following transfection cells were treated for 24 h
with 100 µM Wy14643 or Me2SO vehicle as
indicated. The 1535 clone but not the 1512 clone contain the
upstream PPRE. Transfection and normalization were performed as
indicated in Fig. 5. Normalized luciferase values are shown as fold
induction compared with a transfection with reporter alone.
Transfections were done in triplicate. Standard deviations are
indicated. The results are representative of three or more independent
experiments.
|
|
The functionality of the previously identified potential PPRE upstream
of the rat ACBP gene has not been investigated. We therefore accessed
whether this element could contribute to the PPAR responsiveness of the
rat promoter. As seen in Fig. 6C, this DR-1 was unable to
mediate PPAR/RXR transactivation and did not cooperate with the
intronic PPRE in mediating PPAR/RXR transactivation. Thus, the
upstream DR-1 is not a functional PPRE in transient transfections.
The Intronic PPRE Is Functionally Conserved in the Human
ACBP Gene--
Electrophoretic mobility shift assay showed that the
intronic DR-1 in the human ACBP gene was able to bind PPAR/RXR
and PPAR/RXR. To investigate whether the functionality of the
intron 1 PPRE was conserved in the human ACBP gene, we prepared
reporter plasmids similar to the rat constructs by fusing the
respective human ACBP promoter region, exon 1 and intron 1, sequences
with the luciferase gene (Fig.
7A). These constructs were
cotransfected with human PPAR expression plasmids into the human
embryonic kidney cell line 293, which has low levels of endogenous
PPARs. As for the corresponding rat constructs, the basal activity of
this reporter constructs was very high. However, despite the high basal
activity, hPPAR/RXR as well as hPPAR/RXR were able to
activate further luciferase expression from this construct, whereas
cotransfection with PPAR/RXR only marginally increased the
promoter activity (Fig. 7B). PPAR-mediated transactivation
was abolished by mutating the PPRE, indicating that the intronic PPRE
is functionally conserved between rodents and humans. Similar to the
transfections with the rat constructs the relative activation by
PPAR and PPAR cannot be directly compared.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7.
The intronic PPRE is functionally conserved
between rats and humans. A, the three different
constructs used for analysis of the human intron 1 PPRE, and in
hACBP-(3516/+1136)PPRE-luc the hACBP intron 1 PPRE has been
mutated. B, 293 cells were transfected with either
hACBP(516/+5)-luc, hACBP(516/+1136)-luc, or
hACBP(516/+1136)PPRE-luc reporter constructs and pCMV-hPPAR,
pCMV-hPPAR, or pCMV-hPPAR and pCMV-mRXR expression vectors
where indicated. pCMV--galactosidase expression vector was
cotransfected as control. Following transfection cells were treated for
24 h with media containing 1 µM BRL49653, 100 µM Wy14643, 1 µM L-165041, or
Me2SO vehicle where indicated. Transfection and
normalization were performed as indicated in Fig. 5 but with a total of
2.5 µg of DNA/well. Normalized luciferase values are shown as fold
induction compared with a transfection with reporter alone.
Transfections were done in triplicate. Standard deviations are
indicated. The results are representative of three or more independent
experiments.
|
|
The Intronic PPRE Mediates Thiazolidinedione Responsiveness
of the ACBP Promoter in Adipocytes--
If the intronic PPRE functions
as a PPAR-response element it should instigate
thiazolidinedione activation of the ACBP promoter in
adipocytes, which express high levels of PPAR. We therefore transiently transfected 3T3-L1 adipocytes with the
rACBP(392/+979)-luc or rACBP(392/+979)PPRE-luc constructs and
incubated the cells with or without the thiazolidinedione BRL49653. As
seen in Fig. 8A, BRL49653
efficiently activated the rACBP(392/+979)-luc construct but not the
rACBP(392/+979)PPRE-luc construct. Similarly, BRL49653 activated
the longer reporter constructs prACBP(1535/+979)-luc and
prACBP(1512/+979)-luc but not the corresponding constructs where
the intronic PPRE had been mutated (Fig. 8B). This clearly indicates that this PPRE mediates activation by endogenous PPAR in
adipocytes and that the activation of the ACBP promoter by thiazolidinedione is dependent on the intronic PPRE. In keeping with
the cotransfection experiments using NIH-3T3 cells (Fig. 6C), the potential PPRE between 1535 and 1512 was unable
to function as a PPRE in transient transfections of adipocytes.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 8.
The intronic PPRE is necessary
for ligand-dependent activation of ACBP promoter constructs
by endogenous PPAR in adipocytes. 3T3-L1
cells were cultured and differentiated following the standard
differentiation protocol. At day 4 cells were transfected with
rACBP(392/+979)-luc or rACBP(392/+979)PPRE-luc,
rACBP(1512/+979)-luc, rACBP(1512/+979)PPRE-luc,
rACBP(1535/+979)-luc, or rACBP(1535/+979)PPRE-luc reporter
constructs as indicated and pCMV--galactosidase expression vector
using LipofectAMINE Plus. Cells were treated for 24 h with media
containing 1 µM BRL49653 or Me2SO as control.
Luciferase values have been normalized to -galactosidase values, and
the normalized luciferase values are shown as fold induction compared
with the Me2SO control. Transfections were done in
triplicate. Standard deviations are indicated. The results are
representative of three or more independent experiments.
|
|
PPAR and RXR Bind to the Intronic PPRE on Chromatin in
Adipocytes--
Finally, to confirm that the intronic PPRE was also
functional in vivo in the chromatin context, we isolated
fragmented formaldehyde cross-linked chromatin from mature 3T3-L1
adipocytes and subjected this to chromatin immunoprecipitation using
PPAR-selective and pan-RXR antibodies. PPAR and RXR
cross-linking to chromatin was verified using Western blotting (data
not shown). Following DNA extraction of the immunoprecipitated
chromatin, real time PCR was used to determine the relative occupancy
of different PPREs compared with a control fragment located ~8 kb
downstream of the intronic PPRE. As shown in Fig.
9, PPAR and pan-RXR antibodies efficiently precipitated DNA fragments spanning the ACBP PPRE as well
as fragments covering the ALBP and lipoprotein lipase (LPL) PPREs. The
relative occupancies of the three PPREs were similar in the
immunoprecipitates, indicating that these PPREs recruit PPAR/RXR
with equal efficiency to their cognate sites embedded in chromatin.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 9.
PPAR and RXR
associates with the intronic PPRE on chromatin in adipocytes.
3T3-L1 cells were cultured and differentiated following the standard
differentiation protocol. Chromatin was isolated from day 4 adipocytes
and used for chromatin immunoprecipitation with PPAR- and
RXR-specific antibodies, respectively. The recovered DNA was quantified
by real time PCR using primer sets amplifying the ACBP, ALBP, or LPL
PPRE, respectively. A fragment located ~8 kb downstream of the
intronic ACBP PPRE was used as background control. Results are
expressed as relative occupancy of the respective PPREs compared with a
control fragment. The figure illustrates the results from a
representative experiment.
|
|
| |
DISCUSSION |
ACBP is an evolutionarily highly conserved protein that is thought
to play a role in the intracellular transport of medium to long chain
acyl-CoA esters. The structure and biochemical properties of ACBP are
well described, but the precise function of this protein in
vivo remains unresolved. Because it is highly conserved from yeast
to mammals and is ubiquitously expressed in mammals, it is likely that
ACBP carries out basic cellular functions. In keeping with this, the
mammalian ACBP gene displays all the hallmarks of a housekeeping gene
(11). However, expression levels vary considerably between different
cell types and in response to different metabolic conditions,
indicating that during evolution ACBP may have acquired additional
specialized functions. Induction of ACBP expression appears to be
correlated at least to some extent with increased lipogenesis. In
addition, induction of ACBP expression by PPAR activators, which
induce mitochondrial and peroxisomal -oxidation in liver, has been
reported. However, the molecular mechanisms underlying the regulation
of ACBP expression in mammalian cells have remained largely unknown.
The only functional regulatory elements described to date in a
mammalian ACBP gene is an SREBP-binding site and an NFY-binding site
identified in the proximal promoter of the human gene (17). We have
shown that these elements, at least in transient transfections, are
functionally conserved in the rat ACBP
gene,2 but they remain to be
characterized in a chromatin context.
Previous observations suggested that the ACBP gene might also be a PPAR
target gene. First, PPAR ligands induce ACBP expression in the liver
(14, 20). Second, ACBP expression is significantly induced during the
course of adipocyte differentiation (16), and the induction parallels
that of ALBP, a well known PPAR target gene. In this report we show
that the expression of ACBP in adipose tissue is significantly induced
in a dose-dependent manner when db/db mice are
fed a diet containing the PPAR-specific ligand BRL49653.
Furthermore, we show that the expression of ACBP is induced in 3T3-L1
adipocytes by BRL49653 independently of protein synthesis, indicating
that ACBP is a direct PPAR target gene.
In keeping with this notion we identify a functional PPRE in
intron 1 of the rat ACBP gene and show that it mediates PPAR/RXR transactivation in NIH-3T3 cells. The PPRE is functionally conserved in
the human ACBP gene. The previously identified potential PPRE at 1525
in the rat gene is neither conserved in the human nor in the mouse gene
and is not functional in the rat promoter context. The intronic PPRE
mediates induction by endogenous PPAR in murine adipocytes and
confers responsiveness to the PPAR-selective ligand BRL49653.
Finally, we have used chromatin immunoprecipitation to demonstrate that
the intronic PPRE efficiently binds PPAR/RXR in the chromatin
context in adipocytes. The relative occupancy of the intronic PPRE was
similar to that of the well characterized PPREs on the ALBP and LPL
genes, respectively. Thus, the PPRE in intron 1 of the ACBP gene is a
bone fide PPAR-response element. To our knowledge this is
the first reported chromatin immunoprecipitation assay determining
PPAR/RXR binding to a PPRE.
PPAR and the CCAAT/enhancer-binding protein (C/EBP) are known
to be the major and determining adipogenic transcription factors
(reviewed in Ref. 42). The finding that ACBP is a direct PPAR target
gene indicates that PPAR plays a central role in the up-regulation
of ACBP expression during adipogenesis. SREBP-1 may be less important
in the induction during differentiation, but it is likely that it
modulates the expression of ACBP at the differentiated stage, like it
modulates a number of other adipocyte genes (reviewed in Ref. 43). We
have identified previously a potential C/EBP-responsive element at
position 791 in the rat ACBP gene (39). This element bound proteins
from rat liver nuclear extracts, which express high levels of C/EBP,
and we therefore suggested that it might be involved in the activation
of ACBP expression during adipocyte differentiation. However, this
element is neither conserved in the mouse nor in the human ACBP gene
(data not shown), and transfection with rat ACBP promoter constructs encompassing the region from 2310 to +979 showed that this region of
the promoter was not activated by coexpression of C/EBP (results not
shown). Thus, the previously identified potential C/EBP-response element in the rat gene does not seem to be functional, and no other
functional C/EBP-response elements seems to be present in the region
2310 to +979 of the rat ACBP gene
The results presented in this paper show that the intronic PPRE also
confers PPAR responsiveness to the ACBP promoter constructs, indicating that the ACBP gene may be a natural PPAR target gene. This is supported by the fact that ACBP expression is induced in the
liver following prolonged exposure to PPAR ligands. However, the
finding that ACBP expression is down-regulated in the liver by fasting
when most PPAR target genes are induced (44, 45) indicates that if
ACBP is a direct PPAR target gene, PPAR induction during fasting
must be hampered by the absence of other transcription factors like
SREBP-1, for example, or overruled by the presence of inhibitory
transcription factors.
Interestingly, the sequence of the ACBP PPRE deviates from classical
PPAR-responsive elements by having a G as spacer between the two
repeats. When directly compared in transient transfections, the ACBP
PPRE was much less efficient in mediating PPAR-dependent transactivation than the ACO PPRE, a "classical" PPAR target. This is in keeping with the finding that an "A" between the two repeats gives a stronger binding of PPAR/RXR (46). Our results indicate that an A in this position is far less critical for
PPAR/RXR than for PPAR/RXR binding and function. This
observation is in keeping with the PPAR-responsive ARE7 PPRE of the
ALBP enhancer also having a G in this position and showing preference
for PPAR binding (41). Interestingly, the sequences of the ARE7 PPRE and the ACBP PPRE including their 5'-flanking sequences are strikingly similar and different from that of PPREs of most PPAR-responsive genes. Thus, PPAR may be able to transactivate the ACBP gene, but
ACBP does not appear to belong to the classical PPAR target genes.
PPAR/RXR were able to bind to the intronic ACBP PPRE in
electrophoretic mobility shift assays, and the element efficiently mediated PPAR/RXR transactivation of a heterologous promoter in
NIH-3T3 cells. Interestingly, however, PPAR/RXR was unable to
transactivate natural ACBP promoter constructs in this cell type. HNF4
bound strongly to the intronic PPRE in electrophoretic mobility shift
assays but was unable to transactivate via this element in the
thymidine kinase gene promoter context as well as in the natural ACBP
promoter context in NIH-3T3 cells. Thus, at least under the conditions
used for our investigations the rat ACBP gene is neither a PPAR nor
an HNF4 target gene. Whether HNF4 and PPAR/RXR by binding to the
intronic PPRE interfere with PPAR/RXR binding and transactivation
(47, 48), and thereby inhibit the transcription of the ACBP gene,
remains to be established.
Regulatory elements are frequently observed in intronic regions;
however, to our knowledge this is the first functional PPRE to be
reported in an intronic region. Interestingly, the intronic PPRE is
located in a highly conserved region of intron 1 (49) and overlaps with
an alternative splice donor in the human ACBP gene (50). We have
recently found that this alternative splicing is neither conserved in
rats nor mice,3 indicating
that the conservation of the area between humans and rodents cannot be
explained by functional importance of such an alternative splicing. It
is therefore likely that the PPRE and possibly other regulatory
elements yet to be identified have contributed to the sequence
conservation of this region in intron 1.
The finding that PPARs directly regulate the expression of ACBP is
particularly interesting in the light of our recent results showing
that acyl-CoA esters function as PPAR antagonists in vitro (51, 52) and that overexpression of lipid-binding proteins including
ACBP decrease activation of PPARs by exogenous fatty acids (35). How
ACBPs repress PPAR-mediated transactivation induced by exogenous fatty
acids is unknown. Our data would be in keeping with a model in which
ACBP increase the esterification of exogenously added fatty acids,
thereby decreasing their availability as PPAR ligands. Whether these
effects of ACBP on PPAR transactivation are physiologically relevant
are unclear at this point. In another study we have shown that
antisense ACBP inhibits the ability of 3T3-L1 preadipocytes to undergo
differentiation and to induce key adipogenic transcription factors like
PPAR (38). In the present report we show that ACBP is also a PPAR
target gene, indicating that ACBP acts both upstream and downstream of
PPAR in the adipogenic process. Thus, in the context of adipocyte
differentiation ACBP appears to function in a positive feedback loop to
increase its own expression.
In summary, we have demonstrated that the ACBP gene is a novel
PPAR/RXR target gene and that PPAR/RXR activates transcription through an intronic PPRE in humans and rodents. This PPRE may also
mediate transactivation by PPAR/RXR, but final proof for that awaits
investigations of the interaction of endogenous PPAR/RXR with this
element in a chromatin context.
through an Intronic
Response Element Functionally Conserved between Humans and Rodents*
,
rd, Denmark
TOP
ABSTRACT
INTRODUCTION
