0888-8809/03/$15.00/0
Printed in U.S.A.
Molecular Endocrinology 17(6):1019–1026
Copyright © 2003 by The Endocrine Society
doi: 10.1210/me.2002-0399
Liver X Receptors Interact with Corepressors to
Regulate Gene Expression
XIAO HU, SUZHEN LI, JUN WU, CHUNSHENG XIA,
AND
DEEPAK S. LALA
Department of Biotechnology, Pharmacia Corp., St. Louis, Missouri 63017
Liver X receptors (LXRs) are members of the nuclear receptor superfamily that regulate gene expression in response to oxysterols and play a critical role in cholesterol homeostasis by regulating
genes that are involved in cholesterol transport,
catabolism, and triglyceride synthesis. Oxysterols
and synthetic agonists bind LXRs and activate
transcription by recruiting coactivator proteins.
The role of LXRs in regulating target gene expression in the absence of ligand is unknown. Here we
show that LXRs interact with corepressors, N-CoR
(nuclear receptor corepressor) and SMRT (silent
mediator of retinoic acid receptor and thyroid receptor), which are released upon binding agonists.
The LXR-corepressor interaction is isoform selective, wherein LXR␣ has a very strong interaction
with corepressors and LXR␤ only shows weak interaction. LXRs also exhibit a preference for interacting with N-CoR vs. SMRT. Similar to other nuclear receptors, mutations in the LXR helix 3 and 4
region abolish corepressor interaction. Using a
transient transfection assay, we demonstrate that
LXR represses transcription that can be further
increased by cotransfecting N-CoR into cells.
Chromatin immunoprecipitation experiments further indicated that N-CoR is recruited onto endogenous LXR target genes, and addition of LXR agonists releases N-CoR from their promoters.
Collectively, these results suggest that corepressors play an important role in regulating LXR target
gene expression. (Molecular Endocrinology 17:
1019–1026, 2003)
N
Both N-CoR and SMRT repress transcription by recruiting histone deacetylases through their respective
repression domains (5–9). The receptor interaction domains of N-CoR and SMRT contain small peptide motifs (CoRNR boxes) that are both necessary and sufficient for interactions with TR, RAR, and, under some
conditions antagonist bound estrogen receptor (10–
14). The specificity of nuclear receptor-corepressor
interaction is determined by the individual nuclear receptor interacting with specific CoRNR boxes within a
preferred corepressor (15, 16). Initial mutagenesis
studies suggested that corepressors and coactivators
interact with the nuclear receptor ligand binding domain (LBD) via similar yet distinct amino acid sequences (10–12). Additionally, recent x-ray crystallographic studies have shown that corepressors bind to
the coactivator binding groove on the LBD of the receptor (17).
LXR␣ (NR1H3) and LXR␤ (NR1H2) are members of
the nuclear receptor superfamily that heterodimerize
with the retinoid X receptor (RXR) and are activated via
modified cholesterol molecules that function as endogenous ligands (18). The identification of synthetic
LXR specific ligands has greatly facilitated the discovery of LXR regulated genes (19). Both LXR␣ and LXR␤
have been shown to regulate several important genes
in reverse cholesterol transport, including the ATPdependent cholesterol transporter (ABCA1), cholesteryl ester transfer protein, apolipoprotein E, and
lipoprotein lipase in vitro and in vivo (20–24). Consistent with their ability to activate reverse cholesterol
transport, full agonists of LXR have also been reported
UCLEAR RECEPTORS ARE a class of DNA binding transcription factors that regulate gene expression in response to ligand and play very important
roles in many biological and pathological processes.
Consistent with this, these proteins serve as important
drug targets, and indeed several members of this class
are targets for existing drugs. Agonist-occupied nuclear receptors, when bound to their cognate DNA
response elements, generally activate transcription by
recruiting coactivators (1). Antagonist occupied nuclear receptors, on the other hand, either block transcriptional activation or actively repress transcription
(1, 2). A subset of nuclear receptors, including thyroid
receptor (TR) and retinoic acid receptor (RAR), actively
repress transcription in the unliganded state. This is
most likely due to their ability to interact with corepressor proteins, such as N-CoR and silent mediator
of retinoic acid receptor and thyroid receptor (SMRT)
in the absence of ligand (3, 4). These corepressors are
large proteins that share extensive homology in the
repression domain at the N-terminal region and in the
receptor interaction domain at the C-terminal region.
Abbreviations: ABCA1, ATP-dependent cholesterol transporter; ALPHA, amplified luminescent proximity homogeneous assay; ChIP, chromatin immunoprecipitation; CoRNR,
corepressor nuclear receptor interaction motifs in N-CoR and
SMRT; GST, glutathione-S-transferase; LBD, ligand binding
domain; LXR, liver X receptor; LXRE, LXR response element;
N-CoR, nuclear receptor corepressor; PPAR, peroxisome
proliferator-activated receptor; RAR, retinoic acid receptor;
SERMs, selective estrogen receptor modulators; SMRT, silent mediator of retinoic acid receptor and thyroid receptor;
SREBP1c, sterol response element binding protein; TR, thyroid receptor; VP, viral protein.
1019
1020 Mol Endocrinol, June 2003, 17(6):1019–1026
to increase HDL in mice. LXR agonists have the potential to eliminate the accumulation of lipid by direct
activation of key genes in macrophages to eventually
reduce atherosclerotic lesions (25, 26). In addition to
the beneficial effects on cholesterol efflux, LXR ligands
that function as full agonists also increase gene expression of sterol response element binding protein
(SREBP1c), a key gene that activates transcription of
major genes involved in triglyceride synthesis (e.g.
fatty acid synthase) (19, 27). Consistent with this, administration of LXR full agonists causes a dramatic
increase in triglyceride synthesis in rodents. All the
above studies were done using either natural LXR
ligands (oxysterols) or synthetic LXR agonists. The
exact role of LXRs in the absence of ligands is still
unknown. While knockout studies have indicated that
the expression levels of several LXR target genes are
reduced in liver, it is intriguing that at least some
genes, for example, SREBP-1, appear to be increased
in white adipose tissue of LXR␣/␤ double knockout
mice (27). This suggests that removal of LXR can, at
least under certain circumstances, lead to derepression of its target genes. A potential explanation for the
tissue-selective responses might be that LXRs selectively interact with corepressors in different tissues
perhaps as a consequence of varying levels of endogenous ligands in different tissues. Consistent with this
hypothesis, SREBP-1 expression is higher in liver than
in white adipose when wild-type mice were fed with
chow diets (27). Finally, other RXR partners, such as
RAR, TR, and peroxisome proliferator-activated receptor (PPAR)␦ interact with corepressors in the absence of agonists. Together, these data prompted us
to explore the role of corepressors in LXR-mediated
gene regulation.
In this report, we show that LXRs are capable of
interacting with the corepressors N-CoR and SMRT,
although the interaction with SMRT is much weaker.
The corepressor interaction surface on LXR is similar
to that identified in TR and RAR. In the absence of
ligand, LXRs repress transcription of reporter genes,
and this repression can be further increased by cotransfecting N-CoR into the cells. We also show by
chromatin immunoprecipitation (ChIP) experiments,
that N-CoR is associated with endogenous LXR target
gene promoters in the absence of ligand, and addition
of a LXR agonist releases N-CoR binding. These results suggest that corepressors play a key role in
controlling LXR modulation of its target genes.
RESULTS
In an attempt to understand the molecular mechanisms involved in LXR function, we used a mammalian
two-hybrid assay to investigate if LXRs interact with
corepressors in the absence of ligand. The receptor
interaction domain of N-CoR was fused to a GAL4
DNA binding domain and full length LXRs were fused
to a VP16 activation domain. As shown in Fig. 1A, both
Hu et al. • LXRs Interact with Corepressors
LXR␣ and LXR␤ interact with the C-terminal receptor
interaction domain of N-CoR. The magnitude of interactions is comparable to that of RAR and N-CoR (Ref.
15; and data not shown). These interactions were observed in the absence of ligand and LXR agonists,
either its endogenous ligand 22R-hydroxycholesterol
or the synthetic ligand T0901317, almost completely
released the corepressor. It is also worth noting that
the interaction of N-CoR with LXR␣ is much stronger
than that with LXR␤, suggesting a difference in the
ability of the two isoforms to recruit corepressors.
The interaction domains present within corepressors
contain two CoRNR box motifs and each CoRNR motif is
necessary and sufficient for interaction with unliganded
receptors (10–12). Different receptors exhibit distinct
preferences for these two CoRNR motifs, with RXR and
PPAR showing preference toward CoRNR2, and RAR
toward CoRNR1 (15). We therefore tested whether LXRs
exhibit similar preferences with respect to these interaction motifs. As shown in Fig. 1B (top and bottom), in
contrast to RAR␣, which strongly interacts with
CoRNR1, and PPAR␦, which only interacts with
CoRNR2, LXR␣ and LXR␤ interact with both CoRNR
motif peptides. However, the interactions with the two
CoRNR motif peptides are not equivalent. Whereas the
magnitude of the LXR CoRNR1 interaction is comparable to that of RAR␣, the interaction with CoRNR2 is much
stronger. Under similar conditions, LXR␣ showed more
than a 1000-fold interaction with CoRNR2, whereas
PPAR␦ only showed about a 40-fold interaction. Moreover, and consistent with Fig. 1A, LXR␣ displayed a
tighter association with both CoRNR motif peptides than
LXR␤, further confirming the observation that the two
isoforms exhibit unique preferences for their interactions
with corepressor motifs.
Having established an interaction between LXR and
corepressors in cells, we next used biochemical peptide recruitment assays to demonstrate a direct interaction between LXR and corepressor peptides. A
biotinylated CoRNR1 peptide was incubated with
bacterially produced glutathione-S-transferase (GST)tagged LXR␣ protein in the presence or absence of
T0901317. The interaction was detected using a labeled GST antibody in an amplified luminescent proximity homogeneous assay (ALPHA). When LXR␣ protein binds the peptide, the luminescent dye on the
antibody is in close proximity to the dye associated
with the peptide (via biotin-streptavidin), and a signal
is released. As shown in Fig. 1C, the signal derived
from a control peptide is not altered by adding
T0901317. In contrast, this signal is much higher (4–
5⫻) when LXR␣ LBD is incubated with the CoRNR1
peptide, indicating a direct and specific interaction
between CoRNR1 and LXR␣. Addition of T0901317
reduces this signal to background level, indicating the
interaction between CoRNR1 peptide and LXR␣ is
disrupted by ligand. These biochemical experiments
demonstrate a direct interaction of LXR␣ and corepressor proteins in the absence of ligand and a liganddependent release of corepressors from LXR.
Hu et al. • LXRs Interact with Corepressors
Mol Endocrinol, June 2003, 17(6):1019–1026
1021
Fig. 1. LXRs Interact with Corepressor N-CoR
A, The receptor interaction domain (ID) of N-CoR interacts with LXRs in mammalian two-hybrid assay. Ligand used, 22R-OHC
(22R-hydroxycholesterol, 100 ␮M), T0901317 (10 ␮M). B, Each CoRNR box peptide interacts with LXRs. Top, interaction with
CoRNR1. Bottom, interaction with CoRNR2. RAR␣ and PPAR␦ were included for comparison. C, LXR␣ interacts directly with
CoRNR peptide in vitro in ALPHA screen. D, LXR␣ (Top) and LXR␤ (Bottom) interact differentially with N-CoR and SMRT.
Specific nuclear receptors also display selective preferences for the different corepressors. For example, TR
prefers N-CoR, whereas RAR interacts better with
SMRT. We therefore tested whether LXRs also exhibit
any preferences in regard to corepressor interactions. As
shown in Fig. 1D, SMRT, compared with N-CoR,
showed a much weaker interaction with both LXR␣ (Top)
and LXR␤ (Bottom). Under similar conditions, RAR␣ has
been shown to interact better with SMRT (15, 28).
The above data suggest that LXRs can recruit corepressors in the absence of ligand and are highly selective in nature. Based on these observations, we
speculated that LXRs would repress transcription
when ligand is not available. To test this hypothesis,
we used a construct containing GAL4 DBD fused with
LXR LBD. When cotransfected with a reporter containing GAL4 binding sites, both LXRs repressed reporter
gene expression in HEK293 cells (Fig. 2A). The fold
repression mediated by LXR␣ is comparable to that of
RAR␣, a well-known receptor that also represses transcription in its unliganded state. Consistent with its
weak interaction with corepressors, LXR␤ showed relatively weaker repression. It should be noted that
lipoprotein-free serum was used in these experiments.
When normal serum was used, little repression was
detected presumably due to the presence of some
LXR ligands in normal serum. To further demonstrate
that corepressors play a role in the repression, we
introduced exogenous corepressors into cells. As
shown in Fig. 2B, cotransfection of N-CoR further
increased the repression by LXR␣.
It has been previously shown that the corepressor
interaction surface on nuclear receptors consists of
helices 3–5 within the LBD, a region that is also in-
1022
Mol Endocrinol, June 2003, 17(6):1019–1026
Hu et al. • LXRs Interact with Corepressors
Fig. 3. The Corepressor Interaction Surface on LXR Is Similar to that of TR
Top, Sequence alignments of H3–5 region of TR and LXR␣.
The two mutations (V269R and I287R) are shown below the
alignment. Bottom, Mutations in H3–5 abolish corepressor
interaction.
Fig. 2. LXRs Repress Transcription
A, GAL4 fusions of LXR LBDs repress transcription on a
GAL4 response element-luciferase reporter construct. B, NCoR increases the repression by GAL4-LXR␣.
volved in interactions with coactivator proteins. Several critical surface residues within this region have
been shown to be involved in both coactivator and
corepressor interactions (10, 29). To test whether similar amino acids present within LXRs are also critical
for their interactions with corepressors, we made mutations in two residues, Val 269 and Ile287 within the
LXR␣ LBD. As shown in Fig. 3, mutations in either of
these two residues dramatically decreased the interaction of LXR␣ with N-CoR. These mutations did not
cause a decrease in expression levels or destruction of
structural integrity as these mutations did not alter the
ability of LXR␣ to interact with RXR (data not shown).
These experiments establish that LXR LBDs interact
with corepressors in mammalian two-hybrid assay and
repress transcription when fused to a GAL4 DBD in the
absence of ligand. To extend the relevance of these
observations to a more physiological setting, we investigated whether corepressors modulate LXR function when LXR is bound to its response elements
(LXREs). Three copies of LXREs from either the human
ABCA1 or SREBP1c promoter were cloned upstream
of a synthetic minimal promoter and a luciferase gene.
When transfected into HEK293 cells, these elements
showed higher levels of transcription activity as compared with vector in the absence of ligand. The activation function 1 (AF1) region of LXRs could possibly
contribute this basal activity. It should be pointed out
that HEK293 cells express LXR␣ and several coactivators and corepressors as detected by Taqman analysis (data not shown). Northern analysis also indicated
that LXR␣ gene is expressed at high levels in both liver
and kidney (30). The presence of functional LXR proteins in HEK293 cells is also demonstrated by the
ability of T0901317 to activate these LXRE-containing
reporters (Fig. 4A). These data suggest that endogenous LXR can activate the LXRE driven promoters in a
ligand-dependent manner. When an expression construct of N-CoR was cotransfected along with these
LXRE reporters, the reporter activities were significantly decreased (Fig. 4B). In contrast, cotransfection
of N-CoR had no effect on the expression of the control reporter lacking LXREs. This effect of N-CoR on
LXRE-containing reporters but not on a control reporter lacking an LXRE suggests that N-CoR is recruited to the LXRE, most likely via endogenous LXR.
Consistent with the weak interaction of SMRT with
LXR, cotransfection of SMRT had little effect on the
LXRE driven promoter under these conditions.
Next, we investigated whether LXRs recruit corepressors on endogenous LXR target genes. It has
been shown that LXRs regulate expression of ABCA1
and SREBP1c gene expression in human macrophage
cell line THP1 and liver cell line HepG2 respectively
(31, 32). To determine if these genes are regulated by
LXR in HEK293 cells, we used real-time PCR (Taqman)
Hu et al. • LXRs Interact with Corepressors
Mol Endocrinol, June 2003, 17(6):1019–1026
1023
ers, was released from the promoters after addition of
T0901317. Our results suggest that corepressors are
recruited to target gene promoters by LXR in the absence of agonists. Upon addition of full agonists such
as T0901317, the corepressors are released, leading
to transcriptional activation of LXR target genes.
DISCUSSION
Fig. 4. N-CoR Decreases Transcription of LXRE-Containing
Reporters
A, T0901317 activates LXRE-containing reporters. B,
Corepressor preferences of LXRs. N-CoR but not SMRT decreases basal activity of LXREs derived from ABCA1 and
SREBP1c promoters. Three copies of LXREs from either
ABCA1 or SREBP1c promoter were cloned upstream of a
luciferase reporter. The transcriptional activity of each reporter was monitored under various cotransfection conditions. “Relative Luc Activity” here is relative to the basal
activity of the vector.
to examine their transcript levels. As shown in Fig. 5A,
when HEK293 cells are treated with T0101317, both
ABCA1 and SREBP1c genes are induced, further indicating that LXR is functional in these cells and capable of activating its target genes.
Having established that LXR regulation of ABCA1
and SREBP1c is well preserved in HEK293 cells, we
next used ChIP assay to determine if corepressors are
recruited to these two genes by LXR. Due to the lack
of reliable N-CoR antibodies, we first transfected
HEK293 cells with Flag-tagged N-CoR and then used
a Flag antibody to immunoprecipitate chromatin DNA.
As shown in Fig. 5B, LXR␣ bound to the promoters of
both ABCA1 and SREBP1c genes, and addition of
LXR agonist T0901317 had no significant effect on the
binding of LXR␣ to its target promoters. In contrast to
LXR␣, N-CoR, which was also bound to the promot-
LXRs were initially classified as orphan receptors because their ligands had not been identified. The role of
LXRs in lipid regulation was gradually established after
knockout studies and the discovery of oxysterols as
endogenous LXR agonists. Subsequent studies have
also demonstrated a role for LXR in atherosclerosis
because the absence of the receptors in macrophages
leads an increase in foam cell formation (33). The LXR
agonist T0901317 has also been shown to reduce
atherosclerotic lesion size in genetic models of atherosclerosis (25, 26). Collectively, these data suggest
that LXRs may serve as novel targets for the treatment
of atherosclerosis and other lipid disorders.
LXRs belong to a subgroup of the nuclear receptors
that function as heterodimers with RXR. Although
many receptors in this class, such as TR and RAR
have been shown to associate with corepressors in
the absence of their cognate ligands, the role of LXR in
the absence of its agonists is unknown. Here we report
that LXRs, like TR and RAR interact with corepressors,
especially N-CoR via their LBDs in the absence of
ligand and are released upon addition of an agonist.
The corepressors N-CoR and SMRT contain two
CoRNR boxes (CoRNR1 and CoRNR2) that control the
recruitment of nuclear receptors and these receptors
exhibit distinct preferences for CoRNR1 vs. CoRNR2.
For example, RAR interacts with CoRNR1, whereas
PPARs almost exclusively bind CoRNR2. Here we
show that LXRs are capable of interacting with both
CoRNR boxes. Consistent with their interactions with
corepressors, the LBDs of LXRs repress transcription
when fused to the GAL4 DNA binding domain. It is
noteworthy that LXR interactions with CoRNR2 are
stronger than with CoRNR1, a property that is similar
to that of TR. Mutating two homologous residues
within LBD that are involved in TR-corepressor interactions also abolished LXR-corepressor interactions
(Fig. 3). Moreover, LXRs also display a much greater
preference for N-CoR over SMRT and cotransfection
of N-CoR, but not SMRT, decreases the basal activity
of LXREs (Fig. 4B), suggesting N-CoR is the major
corepressor that binds unliganded LXRs. The corepressor preference by LXR could affect LXR functions
in different tissues as N-CoR and SMRT expression
patterns do not overlap (34). Our ChIP experiments
also showed that N-CoR is recruited to endogenous
LXR target gene promoters in the absence of LXR
ligand. These data further confirm that corepressors
are involved in regulating LXR target gene expression
and are likely to be physiologically relevant.
1024 Mol Endocrinol, June 2003, 17(6):1019–1026
Hu et al. • LXRs Interact with Corepressors
Fig. 5. N-CoR Is Recruited to Endogenous ABCA1 and SREBP1c Genes in HEK293 Cells
A, LXR agonist induces ABCA1 and SREBP1c genes in HEK293 cells. Expression levels of ABCA1 and SREBP1c were
examined by Taqman analysis. B, ChIP analysis of ABCA1 and SREBP1c genes. HEK293 cells were transfected with Flag-tagged
N-CoR and treated with LXR specific agonist, T0901317, at 10 ␮M. The occupancy of ABCA1 and SREBP1c promoter was
monitored by PCR using specific primers surrounding the LXRE in each gene. The numbers (relative to transcription start site) in
the diagram indicate the position of primers.
Oxysterols are thought to serve as endogenous LXR
ligands and accumulation of cholesterol leads to activation of LXR target genes (35). Our results indicate
that LXRs bind corepressors and repress transcription
when LXR is not occupied by an agonist ligand. This
discovery suggests a model that, in cells when cholesterol (oxysterols) levels are low, LXRs bind to the
target gene promoters and actively repress transcription of target genes. For example, unliganded LXR
may repress ABCA1 expression, leading to a decreased cholesterol efflux. As a consequence, repression by LXR may be important to regulate cellular
cholesterol to steady state levels to support normal
cell function. Thus, repression by LXR provides another layer of control in regulating intracellular cholesterol levels. When cholesterol levels are low, LXR
would also inhibit SREBP1c and genes involved in
fatty acid synthesis, leading to decreased esterification of cholesterols to preserve free cholesterols.
When cholesterol levels are high, LXR would release
corepressors and instead recruit coactivators, leading
to activation of its target genes, such as ABCA1 to
pump cholesterol out of cells (Fig. 6).
Our finding further solidifies the role of LXR as a
cholesterol sensor. The basal repression together with
LXR␣ autoregulation provides a very sensitive system
for controlling cholesterol levels in human cells. When
intracellular cholesterol levels are low, this system decreases the cholesterol efflux pathway to preserve
cholesterol. When cholesterol levels are high, this system can respond very quickly to amplify the efflux
pathway to remove cholesterol.
Hu et al. • LXRs Interact with Corepressors
Mol Endocrinol, June 2003, 17(6):1019–1026
1025
scribed (37, 38). After transfection, compounds were added
in DMEM plus 10% lipoprotein-free FBS (Intracell, Frederick,
MD). Cells were harvested 24 h later for measuring luciferase
and ␤-galactosidase activities.
ALPHA Screen
ALPHA Screen assay is a beads-based, time-resolved amplified luminescent proximity homogeneous assay. GSTLXR␣-LBD expressed in bacterial cells was incubated with
250 nM biotinylated CoRNR1 with and without 10 ␮M LXR
agonist T0901317 overnight at 4 C. An inactive biotinylated
peptide was used as a control. The assay was performed in
384-well, 25 ␮l format in buffer containing 50 mM Tris-HCl, pH
7.5; 150 mM NaCl; 2 mM MgCl2; 1 mM dithiothreitol; 20 nM
streptavidin donor beads; and 20 nM anti-GST antibody acceptor beads. The assay plates were read on a Fusion microplate reader (Packard BioScience, Meriden, CT).
Fig. 6. Model of LXR Function under Different Cellular Cholesterol Levels
Left, When cholesterol levels are low, corepressors are
recruited to LXR target genes and as a result, expression of
target genes are inhibited. Right, When cellular cholesterol
levels are high, corepressors are released and coactivators
are recruited, leading to induction of LXR target genes.
The positive effect of LXR agonists on cholesterol
efflux would be expected to reduce atherosclerotic
risk and recent reports indicate that LXR agonists inhibit atherosclerosis (25, 26). On the other hand, LXR
agonists also increase liver SREBP1c expression and
triglyceride synthesis, which would have a deleterious
effect on metabolic balance. This raises concerns regarding the use of LXRs as potential drug targets.
Perhaps a solution to this would be to design selective
LXR ligands that activate cholesterol efflux in peripheral tissues but exert a minimal effect on triglyceride
synthesis in the liver. Studies on selective estrogen
receptor modulators (SERMs) indicate that SERMs
can function as agonists in one cell type but as antagonists in others depending on the corepressor levels in
the two cell types (36). The discovery that LXRs interact with corepressors provides greater insight into the
molecular mechanisms involved in LXR function. In
analogy to SERMs, it also suggests that it may be
possible to design selective LXR modulators that exploit the intracellular levels of coactivators and corepressors within different cells
Taqman Analysis
RNAs were isolated using QIAGEN (Valencia, CA) RNeasy
mini kit. Real-time PCR was performed as described on ABI
Prism 7700 using Cyclophilin as a control. Relative expression was determined using the comparative CT method.
ChIP
ChIP experiments were performed according to Manufacturer’s instruction (Upstate Biotechnology, Inc., Lake Placid, NY)
with minor modifications. Basically, cells were cross-linked
and chromatin templates were broken into approximately
500-bp fragments by sonication. The sonicated fragments
were then immunoprecipitated overnight using various antibodies. The LXR antibody was raised in rabbit against a
specific peptide (peptide sequence: RAEPPSEPTEIRPQKRKK) in the N-terminal region of LXR␣. This antibody
does not cross-react with LXR␤. The Flag antibody was
purchased from Sigma (St. Louis, MO). The immunoprecipitated complexes were then reverse-cross-linked overnight
and eluted. DNA was then purified using QIAGEN PCR purification kit. PCR was done using primers surrounding the
LXREs in ABCA1 and SREBP1c promoters. The primers are,
for ABCA1, forward, 5⬘ GCGGCTGAACGTCGCCC, reverse,
5⬘GGGTCGGCTCGGCTCTG; for SREBP1c, forward, 5⬘TCAGGGTGCCAGCGAACC, reverse, 5⬘GCTCGAGTTTCACCCCGC.
Acknowledgments
We thank Dr. Mitchell Lazar for plasmids and Helen Hartman for helpful discussion on chromatin immunoprecipitation
experiments. We acknowledge the excellent technical assistance provided by Charles Bolten on Taqman analysis.
MATERIALS AND METHODS
Constructs
LXR mutations were made by site-directed mutagenesis using QuikChange kit (Stratagene, La Jolla, CA). VP-LXR␣ and
VP-LXR␤ were constructed by inserting coding sequences of
LXR␣ and LXR␤, respectively, into pACT vector (Promega
Corp., Madison, WI). The other constructs were described
(10, 37).
Received November 27, 2002. Accepted March 17, 2003.
Address all correspondence and requests for reprints to:
Deepak S. Lala, Ph.D., or Xiao Hu, Ph.D., Mail Zone AA3G,
700 Chesterfield Parkway North, Chesterfield, Missouri
63017. E-mail: [email protected] or Xiao.Hu@
pharmacia.com.
REFERENCES
Cell Culture and Transfection
The 293 cells were maintained in DMEM plus 10% FBS.
Transfection conditions were performed in 96-well as de-
1. Glass CK, Rosenfeld MG 2000 The coregulator exchange
in transcriptional functions of nuclear receptors. Genes
Dev 14:121–141
1026 Mol Endocrinol, June 2003, 17(6):1019–1026
2. Germain P, Iyer J, Zechel C, Gronemeyer H 2002 Coregulator recruitment and the mechanism of retinoic acid
receptor synergy. Nature 415:187–192
3. Horlein AJ, Naar AM, Heinzel T, Torchia J, Gloss B,
Kurokawa R, Ryan A, Kamei Y, Soderstrom M, Glass CK,
Rosenfeld MG 1995 Ligand-independent repression by
the thyroid hormone receptor mediated by a nuclear
receptor co-repressor. Nature 377:397–404
4. Chen JD, Evans RM 1995 A transcriptional co-repressor
that interacts with nuclear hormone receptors. Nature
377:454–457
5. Huang EY, Zhang J, Miska EA, Guenther MG, Kouzarides
T, Lazar MA 2000 Nuclear receptor corepressors partner
with class II histone deacetylases in a Sin3-independent
repression pathway. Genes Dev 14:45–54
6. Guenther MG, Lane WS, Fischle W, Verdin E, Lazar MA,
Shiekhattar R 2000 A core SMRT corepressor complex
containing HDAC3 and TBL1, a WD40-repeat protein
linked to deafness. Genes Dev 14:1048–1057
7. Jones PL, Sachs LM, Rouse N, Wade PA, Shi YB 2001
Multiple N-CoR complexes contain distinct histone
deacetylases J Biol Chem 276:8807–8811
8. Li J, Wang J, Nawaz Z, Liu JM, Qin J, Wong J 2000 Both
corepressor proteins SMRT and N-CoR exist in large
protein complexes containing HDAC3. EMBO J 19:
4342–4350
9. Wen YD, Perissi V, Staszewski LM, Yang WM, Krones A,
Glass CK, Rosenfeld MG, Seto E 2000 The histone
deacetylase-3 complex contains nuclear receptor corepressors. Proc Natl Acad Sci USA 97:7202–7907
10. Hu X, Lazar MA 1999 The CoRNR motif controls the
recruitment of corepressors by nuclear hormone receptors. Nature 402:93–96
11. Perissi V, Staszewski LM, McInerney EM, Kurokawa R,
Krones A, Rose DW, Lambert MH, Milburn MV, Glass CK,
Rosenfeld MG 1999 Molecular determinants of nuclear
receptor-corepressor interaction. Genes Dev 13:
3198–3208
12. Nagy L, Kao HY, Love JD, Li C, Banayo E, Gooch JT,
Krishna V, Chatterjee K, Evans RM, Schwabe JW 1999
Mechanism of corepressor binding and release from nuclear hormone receptors. Genes Dev 13:3209–3216
13. Webb P, Anderson CM, Valentine C, Nguyen P, Marimuthu A, West BL, Baxter JD, Kushner PJ 2000 The
nuclear receptor corepressor (N-CoR) contains three isoleucine motifs (I/LXXII) that serve as receptor interaction
domains (IDs). Mol Endocrinol 14:1976–1985
14. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000
Cofactor dynamics and sufficiency in estrogen receptorregulated transcription. Cell 103:843–852
15. Hu X, Li Y, Lazar MA 2001 Determinants of CoRNRdependent repression complex assembly on nuclear hormone receptors. Mol Cell Biol 21:1747–1758
16. Cohen RN, Brzostek S, Kim B, Chorev M, Wondisford FE,
Hollenberg AN 2001 The specificity of interactions between nuclear hormone receptors and corepressors is
mediated by distinct amino acid sequences within the
interacting domains. Mol Endocrinol 15:1049–1061
17. Xu HE, Stanley TB, Montana VG, Lambert MH, Shearer
BG, Cobb JE, McKee DD, Galardi CM, Plunket KD, Nolte
RT, Parks DJ, Moore JT, Kliewer SA, Willson TM, Stimmel JB 2002 Structural basis for antagonist-mediated
recruitment of nuclear co-repressors by PPAR␣. Nature
415:813–817
18. Janowski BA, Willy PJ, Devi TR, Falck JR, Mangelsdorf
DJ 1996 An oxysterol signalling pathway mediated by the
nuclear receptor LXR ␣. Nature 383:728–731
19. Schultz JR, Tu H, Luk A, Repa JJ, Medina JC, Li L,
Schwendner S, Wang S, Thoolen M, Mangelsdorf DJ,
Lustig KD, Shan B 2000 Role of LX Rs in control of
lipogenesis. Genes Dev 14:2831–2838
20. Venkateswaran A, Laffitte BA, Joseph SB, Mak PA, Wilpitz DC, Edwards PA, Tontonoz P 2000 Control of cellular
cholesterol efflux by the nuclear oxysterol receptor LXR
␣. Proc Natl Acad Sci USA 97:12097–12102
21. Laffitte BA, Repa JJ, Joseph SB, Wilpitz DC, Kast HR,
Mangelsdorf DJ, Tontonoz P 2001 LXRs control lipid-
Hu et al. • LXRs Interact with Corepressors
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
inducible expression of the apolipoprotein E gene in
macrophages and adipocytes. Proc Natl Acad Sci USA
98:507–512
Chawla A, Boisvert WA, Lee CH, Laffitte BA, Barak Y,
Joseph SB, Liao D, Nagy L, Edwards PA, Curtiss LK,
Evans RM, Tontonoz P 2001 A PPAR ␥-LXR-ABCA1
pathway in macrophages is involved in cholesterol efflux
and atherogenesis. Mol Cell 7:161–171
Luo Y, Tall AR 2000 Sterol upregulation of human CETP
expression in vitro and in transgenic mice by an LXR
element. J Clin Invest 105:513–520
Zhang Y, Repa JJ, Gauthier K, Mangelsdorf DJ 2001
Regulation of lipoprotein lipase by the oxysterol receptors, LXR␣ and LXR␤. J Biol Chem 276:43018–43024
Joseph SB, McKilligin E, Pei L, Watson MA, Collins AR,
Laffitte BA, Chen M, Noh G, Goodman J, Hagger GN,
Tran J, Tippin TK, Wang X, Lusis AJ, Hsueh WA, Law RE,
Collins JL, Willson TM, Tontonoz P 2002 Synthetic LXR
ligand inhibits the development of atherosclerosis in
mice. Proc Natl Acad Sci USA 99:7604–7609
Grefhorst A, Elzinga BM, Voshol PJ, Plosch T, Kok T,
Bloks VW, van der Sluijs FH, Havekes LM, Romijn JA,
Verkade HJ, Kuipers F 2002 Stimulation of lipogenesis by
pharmacological activation of the liver X receptor leads
to production of large, triglyceride-rich very low density
lipoprotein particles. J Biol Chem 277:34182–34190
Repa JJ, Liang G, Ou J, Bashmakov Y, Lobaccaro JM,
Shimomura I, Shan B, Brown MS, Goldstein JL, Mangelsdorf DJ 2000 Regulation of mouse sterol regulatory
element-binding protein-1c gene (SREBP-1c) by oxysterol receptors, LXR␣ and LXR␤. Genes Dev 14:
2819–2830
Cohen RN, Putney A, Wondisford FE, Hollenberg AN
2000 The nuclear corepressors recognize distinct nuclear receptor complexes. Mol Endocrinol 14:900–914
Darimont BD, Wagner RL, Apriletti JW, Stallcup MR,
Kushner PJ, Baxter JD, Fletterick RJ, Yamamoto KR
1998 Structure and specificity of nuclear receptor-coactivator interactions. Genes Dev 12:3343–3356
Willy PJ, Umesono K, Ong ES, Evans RM, Heyman RA,
Mangelsdorf DJ 1995 LXR, a nuclear receptor that defines a distinct retinoid response pathway. Genes Dev
9:1033–1045
Costet P, Luo Y, Wang N, Tall AR 2000 Sterol-dependent
transactivation of the ABC1 promoter by the liver X
receptor/retinoid X receptor. J Biol Chem 275:
28240–28245
Yoshikawa T, Shimano H, Amemiya-Kudo M, Yahagi N,
Hasty AH, Matsuzaka T, Okazaki H, Tamura Y, Iizuka Y,
Ohashi K, Osuga J, Harada K, Gotoda T, Kimura S,
Ishibashi S, Yamada N 2001 Identification of liver X
receptor-retinoid X receptor as an activator of the sterol
regulatory element-binding protein 1c gene promoter.
Mol Cell Biol 21:2991–3000
Tangirala RK, Bischoff ED, Joseph SB, Wagner BL, Walczak R, Laffitte BA, Daige CL, Thomas D, Heyman RA,
Mangelsdorf DJ, Wang X, Lusis AJ, Tontonoz P, Schulman IG 2002 Identification of macrophage liver X receptors as inhibitors of atherosclerosis. Proc Natl Acad Sci
USA 99:11896–11901
Misiti S, Schomburg L, Yen PM, Chin WW 1998 Expression and hormonal regulation of coactivator and corepressor genes. Endocrinology 139:2493–2500
Peet DJ, Turley SD, Ma W, Janowski BA, Lobaccaro JM,
Hammer RE, Mangelsdorf DJ 1998 Cholesterol and bile
acid metabolism are impaired in mice lacking the nuclear
oxysterol receptor LXR ␣. Cell 93:693–704
Shang Y, Brown M 2002 Molecular determinants for the
tissue specificity of SERMs. Science 295:2465–2468
Li Y, Bolten C, Bhat BG, Woodring-Dietz J, Li S, Prayaga
SK, Xia C, Lala DS 2002 Induction of human liver X
receptor alpha gene expression via an autoregulatory
loop mechanism. Mol Endocrinol 16:506–514
Wu J, Xia C, Meier J, Li S, Hu X, Lala DS 2002 The
hypolipidemic natural product guggulsterone acts as an
antagonist of the bile Acid receptor. Mol Endocrinol 16:
1590–1597