| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Biochemistry Faculty of Medicine Campus Gasthuisberg University of Leuven B-3000 Leuven, Belgium
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
SC is exclusively expressed in the epithelial cells of
many different tissues [e.g. skin, lung, intestine,
reproductive tract (3)]; its promoter is therefore highly specific for
epithelial cells. The level of expression can be regulated by a wide
variety of factors depending on the cell type of interest: the effect
of the cytokines interferon-
, tumor necrosis factor-
, and
interleukin-4 on sc expression in intestinal epithelial
cells has been extensively studied (4, 5, 6, 7). Steroid hormones have been
described to influence sc expression in epithelial tissues
of the reproductive tract (8, 9, 10), the mammary gland (11), liver (12),
and the lacrimal gland (13). In prostate epithelial cells, androgens
enhance expression and secretion of SC (8). In the acinar epithelial
cells of the lacrimal gland, androgen-stimulated sc
expression is mediated, at least partly, by a rise in the sc
mRNA content of the cell (14, 15). In primary cultures of lacrimal
acinar cells, stimulation of sc expression by androgens is
inhibited by actinomycin D as well as antiandrogens, clearly indicating
a direct regulation of transcription (16).
Androgen stimulation of gene expression is mediated by androgen receptor (AR) binding to motifs resembling the 5'-TGTTCT-3' consensus binding sequence located within enhancer elements or promoters of androgen responsive genes. Together with the progesterone and mineralocorticoid receptor, the androgen and glucocorticoid receptors form the steroid receptor (SR) subfamily of nuclear receptors having similar DNA-binding domains (DBDs), hence having the same consensus recognition sequence (17). Imperfect palindromic repeats of this 5'-TGTTCT-3' motif in which the half-sites are separated by a three-nucleotide spacer are high-affinity binding sites for the members of this nuclear receptor subfamily (18). The mechanisms by which specificity of steroid hormone action through these response elements is regulated still remain, for the most part, unrevealed. Recent reports, however, discuss the possibility of specific recognition by the AR of sequences that are not recognized by the glucocorticoid receptor (GR) (19, 20).
In our earlier work, we cloned the 5'-region of the human
sc gene (EMBL-Genbank accession numbers X95880 and
X98765) and determined the major site of transcription initiation in
prostate epithelial cells (21). Transcriptional activity of the
proximal sc promoter was studied in the human HeLa and HepG2
cell lines. Brandtzaeg and co-workers (22) recently demonstrated the
involvement of an E-box (from nucleotide (nt) -74 to -62) and an
inverted repeat sequence (from nt -64 to -47) in the basal
transcriptional activity of the proximal sc promoter.
Piskurich et al. (4) postulated that IFN- stimulation of
sc transcription in human HT-29 colon carcinoma cells is
mediated by an interferon stimulatory response element (ISRE) in the
first exon of the gene. Work in our laboratory demonstrates activation
of the ISRE by IRF-2, as well as the presence of a steroid
hormone-regulatory element in the first exon of the SC gene (23).
In this report, we describe and analyze a genomic region, located 3.3 kb upstream of the transcription initiation site, that confers androgen responsiveness to the sc promoter as well as to a heterologous SV40 promoter. Furthermore, we demonstrate that this enhancer element shows a strong androgen selectivity in conferring steroid responsiveness to the homologous proximal promoter.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). We have therefore inserted the
region from nt -3479 to -2442 in front of either the homologous
537-bp sc proximal promoter (pSC537Luc) or the heterologous
SV40 early promoter in the pGL3 promoter vector (pSV40Luc).
Transcriptional activity of these constructs (pI-IVSC537Luc and
pI-IVSV40Luc) is indeed responsive to androgens (the induction factors
are 5.2 ± 0.9 and 11.8 ± 2.2 SEM, respectively
(Fig. 1B)). Progressive deletion analysis of this fragment showed that
the presence of a region between nt -3319 and -3141 (fragment II) is
essential for androgen response in both promoter contexts. In these
experiments, the activity of the androgen response unit (ARU) seems
largely dependent on its environment, since induction factors mediated
by the different fragments range from 5.2 (pI-IVSC537Luc) to 58
(pIISV40Luc).
|
). sc Promoter fragments
ranging from 3479 bp (pSC3479) to 1949 bp (pIISC1949Luc) show
comparable responses to androgen stimulation. The 537-bp sc
promoter, however, is remarkably inert, showing no significant
androgen-stimulated transcription (pIISC537Luc). Androgen
responsiveness is, however, restored in the 86-bp sc
promoter construct (pIISC86Luc).
We have further delineated the minimal enhancer fragment by deletion
analysis. Deletion of the 62 nt 5'-part of fragment II in the
pIISV40Luc construct abolishes the androgen response in transient
transfection assays in T-47D cells (Fig. 1D, p3'IISV40Luc), whereas
deletion of the 113-bp downstream part of fragment II attenuates, but
does not destroy, the androgen response (Fig. 1D, p5'IISV40Luc). Two
copies of the 45-bp AR.NF oligonucleotide [nt -3319 to -3275], when
inserted in front of the SV40 promoter, give rise to a strong androgen
response of transcription (pAR.NFSV40Luc), indicating that this
fragment contains the elements sufficient for conferring androgen
responsiveness to a promoter.
Interaction of the AR- and GR-DBD with the sc Upstream
Enhancer
In the sc upstream enhancer, four motifs resembling the
5'-TGTTCT-3' core SR consensus recognition sequence (called
cores 1, 2, 3, and 4) are found (Fig. 2).
Band shift assays were performed with DNA probes containing these core
motifs: ARE1.2 (cores 1 and 2), ARE2.3 (cores 2 and 3), ARE3.0 (core
3), and NFsc (core 4). As a positive control, a probe containing the
C3(1) ARE motif (24, 25) was used.
|
). Furthermore, the AR-DBD binds
to the ARE1.2 probe preferentially as a dimer, since the DNA/protein
complex is localized at the same position in the gel as the
DNA/AR-DBD complex in the lane containing the C3(1) ARE (lane 14). It
has been shown previously that in the same experimental conditions a
dimer of the AR- and GR-DBD binds to the C3(1) ARE (26). In contrast
with exclusively dimeric AR-DBD binding to the ARE1.2 motif, only a
low-affinity interaction of monomeric AR-DBD was seen with the ARE 2.3
and NFsc probes. Binding of dimeric AR-DBD was also observed with the
ARE3.0 probe. In this case, however, a monomer band was always equally
pronounced at the AR-DBD concentrations that were used.
|
). Furthermore, the GR-DBD binds to these motifs only as
a monomer, since the protein-DNA complexes migrate faster than does the
AR-DBD bound to ARE1.2, whereas, if it were a dimer, it should migrate
more slowly (cf. the C3(1)ARE control, compare lanes 14 and
15). We have also performed comparative band shift assays with increasing amounts of both receptor DBDs using the ARE1.2, ARE2.3, ARE3.0, NFsc, AR.NF (which contains the cores 1, 2, 3, and 4) and, as a positive control, the C3(1) ARE probes. This revealed that the affinity of the AR-DBD for ARE1.2 or AR.NF is higher than for the ARE2.3, ARE3.0, and NFsc oligonucleotides. In this experiment, approximate values of the apparent dissociation constants (KS) for the AR-DBD interaction with the different probes were calculated: 40 ± 3 nM for the C3(1) ARE, 550 ± 30 nM for ARE1.2, and 600 ± 60 nM for the AR.NF probes. AR-DBD binding characteristics to ARE1.2 and AR.NF are identical. The KS value for binding of the AR-DBD to the ARE2.3, ARE3.0, and NFsc probes could not be determined with the amounts of AR-DBD used. Parallel experiments using the GR-DBD show very low affinity for any of the tested sc upstream enhancer sequences as compared with the C3(1) ARE. Again, due to the low affinities of the GR-DBD for these elements, the respective KS values could not be calculated with the amounts of GR-DBD used. A KS value of 77 ± 8 nM was calculated for interaction of the GR-DBD with the C3(1) ARE.
A NF-I Binding Site Flanks the sc ARE
To identify interactions of other transcription factors with
the sc upstream enhancer, we have performed in
vitro DNaseI footprinting experiments on fragment II with rat
liver nuclear extracts as a source of ubiquitous transcription factors.
A 24-bp region (nt -3295 to -3272) is protected from DNaseI digestion
(Fig. 4A). In this window, two
hypersensitive bands (the A at nt -3277 and the G at nt -3278) are
present. Protection of the sequence from digestion is abolished by
addition of an excess of cold competitor oligonucleotide containing the
NF-I recognition sequence of the adenovirus origin of replication (27,
lane 10) or a competitor oligonucleotide containing the footprinted
sequence itself (NFsc, lane 6). Sequences both downstream and upstream
of the two hypersensitive bands are affected in the same way by the
competition; therefore, a single protein is probably responsible for
the protection. High-affinity binding sites for other transcription
factors [among others HNF-5 (28), PEA-3 (29), AP2 (30), AR/GR
(24, 25), AP1 (31), and C/EBP (32)], were not able to compete for
binding to the protected region.
|
, band shift assays using nuclear extracts
from T-47D cells show a high-affinity and specific binding of a protein
to the NFsc oligonucleotide. Binding of this protein is fully competed
for by addition of a 400-fold excess of cold competitor oligonucleotide
containing the NF-I binding sequence of the adenovirus origin of
replication (NFAd). This sequence consists essentially of a partial
palindromic repeat of the 5'-TTGGC-3' NF-I core binding site
separated by five nucleotides. The NFAdmut oligonucleotide, in which
one of the two copies of the NF-I binding motif is mutated to
5'-TTTTA-3', is only moderately effective in competition
experiments, since a more than 400-fold excess of this competitor
oligonucleotide does not fully compete for binding of the factor (data
not shown). Addition of a 400-fold excess of a nonspecific
competitor does not influence the binding of the protein (Fig. 4B).
Binding of CTF/NF-I to NFsc was confirmed by supershift experiments
using a polyclonal antibody against the -subunit of CTF/NF-I
proteins (Fig. 4C). Inclusion of 1 µl of undiluted rabbit immune
serum in the incubation mixture shifts the retarded band, whereas the
same amount of undiluted preimmune serum does not. The same results
were obtained when the NFAd motif was used as a radiolabeled probe.
Mutational Analysis of Putative Regulatory Elements in Fragment
II
To asses the implication of each of the putative AR and NF-I
binding sites, we have introduced point mutations destroying the
binding of these proteins (see also Fig. 2) in them in the context of
pIISV40Luc and investigated their effects on the androgen response
(Fig. 5). Mutation of the core 1 sequence
most dramatically decreases androgen stimulation of promoter activity
(>90% reduction of induction), while mutations of cores 2, 3, or 4
have a less pronounced effect. Mutation of the NF-I binding site in
fragment II (5'-TTGTGCAC-3' instead of
5'-TTTGGCAC-3', see also Fig. 2) results in a 70% decrease
of androgen induction (pIImutNFSV40Luc). The effect of this mutation on
NF-I binding was checked in band shift assays: a
400-fold excess of cold oligonucleotide containing the mutated sequence
did not compete for NF-I binding to the wild-type NFsc oligonucleotide
(Fig. 4B). Combined mutations of core 1 and the NF-I site
(pIImut1+NFSV40Luc) or core 1 and core 2 combined with the NF-I site
(pIImut1+2+NFLuc) completely abolish the androgen response.
|
). The pSC3479Luc reporter construct is
also stimulated by androgens (up to 5-fold induction with
10-8 M R1881) and not by glucocorticoids. The
pMMTVLuc reporter vector was induced 28- and 38-fold upon stimulation
with 1 nM R1881 and dexamethasone, respectively.
|
Taken together, these functional data corroborate our in vitro finding that ARE1.2 is the predominant interaction site for the AR-DBD within the sc upstream ARU and that the same motif shows a strong preference for the AR compared with the GR.
The ARE1.2 Plays a Crucial Role in the Androgen Specificity of the
sc Upstream Enhancer
It has been postulated previously (Refs. 19, 20 ; see also
Discussion) that differential binding of the AR to direct
repeats, rather than the classical inverted repeats, might account for
androgen-specific transactivation . We have therefore increased the
palindromic nature of the ARE1.2 motif by replacing the T at position 4
by an A (5'-GGCACTttcAGTTCT-3') and investigated
its effect on AR- and GR-DBD binding (Fig. 7A). Whereas the GR-DBD is excluded from
binding to the wild-type motif, it can specifically interact with the
mutant motif. Furthermore, the GR-DBD binds exclusively as a dimer to
this element, since no monomer band can be detected at any
concentration of the GR-DBD that was used. The affinity of the AR-DBD
for the mutated motif (KS = 486 nM) is
slightly increased when compared with the wild-type element
(KS = 580 nM, Fig. 7A).
|
). The androgen responsiveness of the mutated construct is an
approximate 3-fold higher compared with the wild-type construct,
indicating that a relatively small increase in the affinity of the
receptor’s DBD for the element in vitro can cause a marked
increase in its functionality. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). By investigating the effects of
progressive deletions of this region cloned in front of the SV40
promoter, we could delineate a 178-bp region (here called fragment II)
from nt -3141 to -3319 that is not only essential for the mediation
of androgen responsiveness of transcription, but also shows a very high
stimulatory capacity (
60-fold stimulation of SV40 promoter activity;
see Fig. 1B). The presence of regions flanking fragment II generally
seems to have an inhibitory effect on its function, although clear
functional regions surrounding the core fragment II could not be
identified. Fragment I, however, located upstream of fragment II, seems
to have a strong and reproducible inhibitory effect since its deletion
causes a 2- to 5- fold increase in androgen responsiveness irrespective
of the promoter context (compare in Fig. 1B pI-IVSC537Luc and
pII-IVSC537Luc with pI-IVSV40Luc and pII-IVSV40Luc) or the presence of
fragments III and/or IV (compare in the same figure pI-IVSV40Luc,
pI-IIISV40Luc, and pI-IISV40Luc with pII-IVSV40Luc, pII-IIISV40Luc, and
pIISV40Luc). Whether this inhibitory effect is due to actual
transcription factor binding to cis-acting elements within
fragment I, or is merely the consequence of a general change in the
structural environment of the core enhancer fragment, remains to be
elucidated. We attribute the ambiguous effects of the deletions of the
fragments located downstream of fragment II to positional effects. It
has been shown previously that the mere distance between a promoter and
an isolated ARE or GRE can be a determining factor for the level of
induction that can be conferred to the promoter by the enhancer (33, 34).
In the context of the homologous sc promoter, changing the
distance between the core enhancer fragment II (from 3141 to 1949 nt
from the transcriptional start point) and the proximal promoter has no
severe effect on the stimulatory capacity of the ARU (Fig. 1C). The
537-bp sc proximal promoter, however, is remarkably
insensitive to stimulation when the enhancer is inserted immediately
upstream: promoter activity of the pIISC537Luc fragment is not
stimulated to a significantly higher level than the pSC537Luc
construct. The shorter 86-bp sc proximal promoter is,
however, highly sensitive for androgen stimulation mediated by fragment
II.
The 62-bp upstream part of fragment II was found to be necessary
and sufficient for the mediation of androgen responsiveness of proximal
promoter activity (Fig. 1D). Although the downstream 113-bp subfragment
in itself is silent, its deletion causes an approximate 4-fold drop in
androgen responsiveness. Again, whether the stimulatory effect of this
fragment is due to protein binding to cis-acting elements
within this region or is merely the consequence of environmental
changes or positional effects, remains to be determined.
Of the 62-bp upstream part of fragment II, a 45-bp region contains
sufficient elements to confer androgen responsiveness to a
promoter (Fig. 1C, pAR.NFSV40Luc). This fragment contains four SR core
binding elements as well as a consensus CTF/NF-I monomer binding motif
(Fig. 2). The implication of each of these elements in the androgen
responsiveness mediated by the enhancer was demonstrated by mutational
analysis in transient transfection assays (Fig. 5). Strikingly, three
of the SR monomer binding elements form a direct repeat in which each
element is separated by a three-nucleotide spacer. The forth motif is
located nine nucleotides downstream of the third. The CTF/NF-I binding
element resides two nucleotides downstream of the third SR monomer
binding motif and overlaps with the fourth. CTF/NF-I is known to be
implicated in the functioning of several steroid response units
described to date (35, 36, 37, 38). It recognizes a partial palindromic
sequence (5'-TTGGCN5(T/G)CCA-3')
although high-affinity binding has also been demonstrated for the
single 5'-TTGGC-3' motif (39). In the sc upstream
enhancer, only one half of the palindrome is present. The NF-I
footprint in the sc upstream enhancer shows two
hypersensitive bands downstream of the central motif (a G at position
18 and an A at position 19 in the footprint; Fig. 4A) similar to the
NF-I footprint in the ARU in the first intron of the C3(1) gene of the
rat prostatic binding protein (40). In the MMTV-LTR, the NF-I site is
located at the border of a precisely positioned nucleosome and is not
occupied by NF-I in the noninduced state (37, 41, 42). SR binding to
the MMTV-LTR nucleosomal DNA in vivo is proposed to alter
the nucleosome structure in such a way that NF-I can bind to its
cognate sequence and activate transcription of the MMTV-LTR promoter
(41). For the MMTV-LTR, it has been demonstrated that the orientation
of the ARE/GRE elements on the nucleosomal surface is of great
importance in the functionality of the element (42, 43).
All four of the SR core binding elements (each of them being the
downstream binding element in the repeated motifs ARE1.2, ARE2.3,
ARE3.0, and ARE4.0) are bound poorly by the GR-DBD in in
vitro binding assays (Fig. 3). The GR-DBD binds to
oligonucleotides containing these elements exclusively as a monomer,
even at high concentrations (
3 µM) of protein (data not
shown). The affinities of the AR-DBD for the same elements are
comparable to those of the GR-DBD except for the ARE1.2 motif
containing cores 1 and 2 (Fig. 3). The affinity of the AR-DBD for this
element is significantly higher compared with the other motifs.
Furthermore, the AR-DBD binds to this element exclusively as a dimer
even at lower concentrations of the protein (see also Fig. 7A). Since
the AR-DBD shows very poor binding to the core 2 motif within the
ARE2.3 motif, high-affinity and exclusively dimeric binding of the
AR-DBD to the ARE1.2 motif strongly suggests a high degree of
cooperativity in binding of the AR-DBD monomers. The overall affinity
of the AR-DBD for this element is, however, still considerably lower
when compared with the C3(1) ARE that was used as a positive control in
the experiment. The discrepancy between AR- and GR-DBD binding to
ARE1.2 is, however, clear and consistent.
ARE3.0 shows some degree of AR specificity in that, at higher concentrations of the AR-DBD, a dimer band appears whereas this is not the case with the GR-DBD. The presence of a monomeric band in the band shift assays, however, indicates that the dimer binding does not involve cooperativity between binding of the DBDs. Furthermore, the same KS values were calculated for AR-DBD binding to the AR.NF oligonucleotide (which contains cores 1, 2, 3, and 4) and the ARE1.2 motif (600 ± 60 nM and 550 ± 30 nM, respectively). Therefore, no cooperativity seems to exist between the ARE2.3, 3.0, or 4.0 elements and the ARE1.2 motif in the binding of the AR-DBD. In conclusion, the ARE1.2 is the strongest ARE within the sc upstream enhancer, although functional analysis of point mutations in the other elements clearly indicates their involvement in the functionality of the enhancer.
The sc Upstream Enhancer Is Androgen Specific
A good correlation exists between the mode of AR-DBD binding to a
motif in our in vitro binding assays and its implication in
the functionality of the sc upstream enhancer. Mutation of
core 1, the downstream half-site within ARE1.2, the only motif that is
specifically recognized by dimeric AR-DBD, has by far the most
dramatic effect on the functioning of the enhancer (Fig. 5).
Furthermore, the fact that the ARE1.2 element is not bound by the
GR-DBD is correlated with the fact that the sc upstream
enhancer does not confer glucocorticoid responsiveness to the
homologous proximal sc promoter in transient transfection
assays (Fig. 6).
The ARE1.2 motif (5'-GGCTCTttcAGTTCT-3') shows a
striking resemblance to the PB-ARE-2
(5'-GGTTCTtggAGTACT-3'), another motif proposed to be
specifically recognized by the AR (44). Both AREs can be considered
direct repeats of the monomer binding elements with a three-nucleotide
spacer, raising the possibility that the mechanism of AR specificity of
binding to these motifs might be that the AR is able to bind a direct
repeat, whereas the GR is not. Within the nuclear receptor core binding
sites, the nucleotides at positions 2 and 5 (a guanine and a cytosine,
respectively) are known to be essential for high-affinity binding of
any member of the nuclear receptor superfamily (45, 46). The residues
at positions 3 and 4 are known to be discriminative for binding of
members of the two subfamilies of the nuclear receptors. High-affinity
binding of a member of the SR family requires a thymidine at position
3; an adenine at this position turns the element into a binding element
for a member of the RAR/RXR subfamily of nuclear receptors (45). Since
one striking similarity between the sc ARE1.2 and the
PB-ARE-2 is the thymidine residue at position 4 in the left half-site,
it can be argued that this nucleotide is responsible for specific
AR-DBD binding as opposed to GR-DBD. Position 4 is equivalent to
position 3 of a half-site in the other orientation, which is therefore
an A in the case of the sc ARE1.2 and the PB-ARE-2. We have
provided further evidence confirming the validity of the aforementioned
hypothesis by replacing the thymidine at position 4 of the ARE1.2
element with an adenine, increasing the palindromic nature of the
repeat. Indeed, this mutation dramatically increases the affinity of
the GR-DBD for the mutated element, whereas the affinity of the AR-DBD
for the motif is hardly affected (Fig. 7A). Not only does the GR-DBD
now bind to the element, it does so exclusively as a dimer, indicative
for a highly cooperative binding of the GR-DBD to both half sites.
Furthermore, the introduction of the same point mutation in ARE1.2 in
the context of the sc upstream enhancer now allows the
enhancer to confer glucocorticoid responsiveness to the sc
proximal promoter in transient transfection assays (Fig. 7B). The
androgen specificity of the sc upstream enhancer is
therefore largely diminished essentially by converting the ARE1.2
element from an imperfect direct repeat into a partially palindromic
repeat. We believe that these findings are a further confirmation of
our hypothesis that transactivation by the AR can be mediated by AR
binding to a direct repeat of its monomer binding motifs whereas the GR
is not able to do so. Further investigation will be required, however,
to establish whether or not this assumption will prove to be generally
valid.
In conclusion, we have identified and functionally analyzed a complex and androgen-specific enhancer in the far upstream region of the human SC gene, which we believe is a likely candidate to be the key regulatory element in the steroid control of human sc expression. We have also demonstrated that the sc upstream ARU is androgen specific in vitro and in functional assays. From these findings, we believe that GR exclusion from binding to direct repeats of SR monomer binding elements, is an important mechanism that imposes androgen specificity on enhancer responsiveness.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Synthetic Oligonucleotides
Oligonucleotides used in this study were synthesized on a
Biosearch Cyclone DNA synthesizer (Milligen Corp., Bedford, MA). Next
to the T7 universal primer, the following primers were used in the
generation, by PCR amplification, of sc upstream
fragments (sc genomic sequences are in
capitals):PCR1fwd (located from nt -3319 to -3296):
5'-actcgagctcTCCTAGAACTGAAAGAGCCTTTGG-3'; PCR1rev:
5'-catatgaattcCCAAAGGCTCTTTCAGTTCTAGGA-3'; PCR2fwd (from nt
-3164 to -3141): 5'-actcgagctcGCTGAGTCCAGAGTCAGGAAAGTC-3';
PCR2rev: 5'-catatgaattcGACTTTCCTGACTCTGGACTCAGC-3'; PCR3fwd
(nt -3006 to -2985):
5'-actcgagctcGGGCAATGGACTCTCTTGGCCT-3'; PCR3rev:
5'-catatgaattcAGGCCAAGAGAGTCCATTGCCC-3'; PCR4rev (nt -2467
to -2442): 5'-gagatgaattcAAGAAATAAGTTGTGTCCAGTTGTCC-3'.
Figure 2 depicts the AR.NF, ARE1.2, ARE2.3, ARE3.0, and NFsc
oligonucleotides used in footprinting and/or band shift experiments.
The upper strand oligonucleotides all have a 5'-CTAGC-3'
extension at their 5'-ends as well as an additional A at their 3'-ends.
The lower-strand oligonucleotides all have a 5'-GATCT-3'
extension at their 5'-ends and an additional G at their 3'-ends. This
generates NheI and BglII sticky ends at the 5'-
and 3'-ends of the double- stranded oligonucleotides, respectively. The
ARE1.2 mut and NFscmut oligonucleotides are identical to ARE1.2 and
Nfsc, respectively, except for the mutation as depicted in Fig. 2. Next
to these, the following oligonucleotides were used for competition or
as radiolabeled probes in bandshift and or footprinting experiments:
NFAd (27): 5'-ATTTTGGCTACAAGCCAATATGAT-3' and
5'-ATCATATTGGCTTGTAGCCAAAAT-3'; NFAdmut:
5'-ATTTTGGCTACAATAAAATATGAT-3' and
5'-ATCATATTTTATTGTAGCCAAAAT-3'. C3(1) ARE (24, 25):
5'-aagcttACATAGTACGTGATGTTCTCAAGg-3' and
5'-tcgacCTTGAGAACATCACGTACTATGTa-3'. The sequences of the
PCR primers used for the introduction of point mutations in the
putative core SR binding motifs in the sc upstream enhancer
destroying AR or GR binding are identical to the wild-type sequence
except for the G-to-T mutation at position 2 in the core motif. They
all start at nt -3319 and all have a 5'-GGGGGA-3' extension
at their 5'-end creating a BamHI restriction site in the
amplified fragment. The oligonucleotides carrying the mutated cores 1,
2, 1+2, and the sc ARE1.2 mut oligonucleotide have their
3'-ends at nt -3290. The oligonucleotide for the introduction of the
mutation in core 3 has its 3'-end at nt -3281; PCR oligonucleotides
mutated in core 4 and the NF-I binding motif both have their 3' ends at
nt -3272.
Luciferase Reporter Constructs
The reporter construct pSC3479Luc (Fig. 1A) was made by
insertion of a NsiI fragment from a pGEM-15 construct
containing a 4.1-kb XbaI fragment of the human SC gene (from
nt -3479 to + 601) cloned downstream of the T7 RNA polymerase
promoter, in the NsiI-digested 1257pGL construct described
previously (21). The pSC2507Luc construct was made by digesting
pSC3479Luc with SpeI and NheI, followed by
self-ligation. The pSC1949Luc construct was made by digesting p3479Luc
with StuI and XbaI followed by a fill-in of the
overhanging ends and intramolecular ligation. Different sc
upstream genomic fragments were generated by PCRs on the pGEM-15
construct containing the 4.1-kb XbaI fragment. PCR products
originating from the primer combinations T7/PCR1rev (fragment I),
T7/PCR3rev; T7/PCR4rev; PCR1fwd/PCR4rev, PCR2fwd/PCR4rev, and
PCR3fwd/PCR4rev (fragment IV) were cloned in the pGEM-T cloning vector
and inserted as SacI fragments from these plasmids in the
correct orientation in the pGL3 promoter vector or in the pSC537Luc
vector (as described in Ref. 21), as appropriate. PCR products from the
primer combinations T7/PCR2rev; PCR1fwd/PCR2rev (fragment II);
PCR1fwd/PCR3rev and PCR2fwd/PCR3rev (fragment III) were inserted into
the pGEM-15 vector as EcoRI/SacI fragments. The
T7/PCR2rev PCR product was cloned from the pGEM-15 construct as a
XbaI fragment in the correct orientation in the
NheI-digested pGL3 promoter vector. The
PCR1fwd/PCR2rev, PCR1fwd/PCR3rev, and PCR2fwd/PCR3rev PCR products were
inserted as XbaI/SacI fragments into the
NheI/SacI-digested pGL3 promoter vector. Fragment
II was cloned as a EcoRI/MluI fragment in
the EcoRI/MluI-digested pSC2507Luc,
pSC1949Luc, and pSC86Luc and as a XbaI/EcoRI
fragment from a pGEM-15 subclone in pSC537Luc digested with
NheI and EcoRI. The PCR1fwd/PCR2rev fragment in
the pGL3 promoter vector was divided in a 62-bp upstream and a 113-bp
downstream fragment (Fig. 1D) by digesting the plasmid with a
combination of PstI and either KpnI (5' deletion)
or EcoRI (3' deletion) followed by intramolecular ligation.
Two copies of the AR.NF oligonucleotide were cloned in the
SmaI site of pGEM-15 and subsequently as a
XbaI/SacI fragment in the pGL3 promoter
vector.
Point mutations of cores 1, 2, 3, and 4 putative SR binding
sites and the NF-I site in the context of pIISV40Luc (as
depicted in Fig. 2) were made by PCRs using combinations of the
respective mutated forward primers (as described above) and the PCR2rev
primer. PCR products were inserted as EcoRI/BamHI
fragments into the pGEM-15 vector. From these constructs, the PCR
fragments were subsequently inserted as XbaI/SacI
fragments into the NheI/SacI-digested pGL3
promoter vector or as EcoRI/MluI fragments in the
EcoRI/MluI-digested pSC86Luc (21).
Band Shift Assays and DNaseI Footprinting Reactions
Labeling of synthetic oligonucleotides or restriction fragments
was performed by a fill-in reaction using the Klenow fragment of DNA
polymerase I in the presence of [-32P]dATP or
[-32P]dCTP (Amersham Pharmacia Biotech,
Buckinghamshire, UK) to a specific activity of 5–10 x
103 cpm/fmol. Band shift assays were performed essentially
according to De Vos et al. (48). Each binding mixture
contained 50 ng/µl poly(dI/dC) as a nonspecific competitor. In
competition experiments, a 400-fold excess of cold oligonucleotide was
included in the binding mixture and incubated on ice for 10 min before
the radiolabeled oligonucleotide was added. Samples were loaded on a
nondenaturing 5% polyacrylamide/bis-acrylamide (29/1) gel and run at
120 V for 90 min at room temperature. The gel was dried and exposed to
X-Omat-AR film (Eastman Kodak Co., Rochester, NY). For
quantitative analysis, the dried gel was exposed to a PhosphoImager
cassette and radioactivity was measured in a STORM 840 Phosphoimager
(Molecular Dynamics, Inc., Sunnyvale, CA) using the
Imagequant software provided by the manufacturer. For the calculation
of the KS values, the Fig.P program (Fig.P Software Corp.,
Durham, NC) was used. The relative amount of radioactivity in the
retarded bands was plotted as a function of the concentration of AR- or
GR-DBD. Amounts of AR- and GR-DBD ranged from 17 nM to 2.7
µM. Binding curves were fitted to a function representing
allosteric Hill kinetics.
In the supershift experiments, the antibody was added to the binding mixture and incubated at room temperature for 30 min before the radiolabeled probe was added. The mixture was then incubated an additional 30 min at room temperature.
The DNA fragment used in the footprinting experiments is an EcoRI/MluI restriction fragment from a pGEM-15 construct containing fragment II between the EcoRI and SacI sites. Footprinting reactions were performed essentially according to Lemaigre et al. (49). G and AG chemical cleavage reactions, according to Maxam and Gilbert (50), were performed on the same DNA fragment and were used as reference in each footprinting gel.
Preparation of Nuclear Extracts of the Rat AR- and GR-DBD
Rat liver nuclear extracts were prepared from the livers of
8-week-old male rats essentially as described by Wall et al.
(51) and modified by Zhang et al. (52). Protein
concentrations were determined by the Bradford method. Nuclear extracts
from T-47D cells used in band shift assays were prepared according to
Andrews and Faller (53). The DBD of the rat AR [Asp 533 to Asp 637
(54)] and GR [Ala 432 to Asn 533 (55)] were expressed in
Escherichia coli as glutathione S-transferase
(GST) fusion proteins and purified on a glutathione sepharose column to
a concentration of 1 mg/ml of more than 95% pure protein, as was
assessed by Coomassie-stained protein gels (56). The GST was removed by
thrombin digestion.
Cell Culture and Transfection Experiments
T-47D human mammary gland carcinoma cells and COS-7 monkey
kidney cells were obtained from the American Type Culture Collection (Manassas, VA). Cells were maintained in DMEM
containing 1000 mg/liter glucose, supplemented with penicillin (100
IU/ml), streptomycin (100 µg/ml), and 10% FCS (Life Technologies, Inc.) at 37 C and 5% CO2. Transient
transfections were performed by the calcium phosphate-DNA
coprecipitation method mainly according to Claessens et al.
(24). The first day, cells were plated in DMEM, supplemented with 5%
charcoal-treated FCS, in 24-well tissue culture plates (Nunc, Roskilde,
Denmark) at a density of 100,000 cells per well. The second day, cells
were transfected with 1 µg of reporter construct per well. After
4 h, the cells were incubated for 1 min in 15% glycerol in 1
x PBS. The third day, medium was replaced with or without addition of
the synthetic androgen methyltrienolone (R1881) or the synthetic
glucocorticoid dexamethasone (Dex). Cells were incubated with hormone
for 24 h. The fourth day, cells were harvested in 100 µl of
Passive Lysis Buffer (Promega Corp.) according to the
instructions of the manufacturer. Luciferase activity of 10–20 µl of
cell lysate was measured in a Microlumat LB 96P luminometer (EG&G
Berthold, Bad Wilstadt, Germany). In transfection experiments in
T-47D cells, the reporter construct was cotransfected with either the
pSV-AR0 human AR expression plasmid [as described by Brinkmann
et al. (57)] or the pRSV-GR human GR expression plasmid
(pRShGR) described by Giguerre et al. (58), as
appropriate (100 ng/well). In experiments in COS-7 cells,
cotransfections were performed with the pSG5-hAR human AR expression
plasmid [as described by Alen et al. (59)] or the pSG8-rGR
rat GR expression plasmid (60). AR and GR expression levels in
both cell lines were evaluated by a hormone binding assay using
[H3]mibolerone (Amersham Pharmacia Biotech)
or [H3]triamcinolone acetonide (Dupont New NEN, Boston, MA). Cells were transfected with 100 ng per ml
medium of the appropriate SR expression plasmids, 100 ng per ml medium
of CMV-ß-galactosidase expression plasmid, and 1.8 µg of carrier
DNA. Retention of radioactivity at increasing concentrations of the
radiolabeled hormones was compared between conditions with and without
the addition of an excess of nonradiolabeled R1881 or dexamethasone. At
saturating conditions, displaceable [H3]mibolerone
binding of 0.85 (± 0.25) and 5.6 (± 1.8) fmol per µg protein
was measured in T-47D and COS-7 cells, respectively. The capacity of
displaceable [H3]triamcinolone acetonide binding of
transfected T-47D and COS-7 cells at the same conditions were 0.33
(± 0.13) and 2.3 (± 0.17) fmol per µg protein, respectively.
Transfection efficiencies were assessed by ß-galactosidase assays on
parallel samples.
In the transfection assays using reporter plasmids, a luciferase
reporter construct driven by the steroid-sensitive MMTV promoter
(pMMTVLuc) was always included as a positive control (average
induction: 89 ± 24 SEM). Luciferase values of the
samples were normalized according to the protein concentration.
Transfection experiments were performed in triplicate and repeated at
least three times independently. In the generation of the dose-
response curves upon stimulation with R1881 and dexamethasone in T-47D
and COS-7 cells (Fig. 6), all samples were performed in duplicate and
repeated at least twice independently. In the calculation of the
SEM values, each independent experiment (in triplicate) is
considered as one. In all transfection experiments, the activities of
the reporter constructs driven by the different sc promoters
constructs did not differ significantly. No influences of the length of
the different sc promoters, of the presence of different
upstream enhancer fragments, or the cotransfections with either GR or
AR expression plasmids on the nonstimulated sc or SV40
promoter activities was seen in either T-47D or COS-7 cells. In T-47D
cells, an average luciferase value of 1,600 light units per µg of
cell lysate was measured for the nonstimulated MMTV-LTR; for the
sc promoter or SV40 promoter-driven reporter constructs,
average luciferase values were 14,000 and 9,000 light units per µg of
cell lysate, respectively. In COS-7 cells average luciferase values are
6,000 and 23,000 for the nonstimulated MMTV-LTR and sc
promoter-driven reporter constructs, respectively.
|
ACKNOWLEDGMENTS |
|---|
CTF/NF-I polyclonal antibody was the kind gift of
Dr. N. Tanese of the NYU Medical Center (New York, NY). The pSG8-rGR
plasmid was a kind gift of Dr. Stunnenberg. The authors are grateful to
H. Debruyn and R. Bollen for their excellent technical assistance and
to V. Feytons for the expert synthesis of numerous
oligonucleotides. |
FOOTNOTES |
|---|
This work was supported by Grant 3.0048.94 of the Geconcerteerde Onderzoeksactie van de Vlaamse Gemeenschap, and by grants from the Inter Universitaire Attractie Pool, the Belgian Cancer Fund, the Fonds voor Geneeskundig en Wetenschappelijk Onderzoek, and the Vlaamse Wetenschappelijke Stichting. G.V. and P.A. were supported by a scholarship from the Vlaams Institut voor de Bevordering van het Wetenschappelijk-Technologisch Onderzoek in de Industrie. F.C. is a senior assistant of the Fonds Voor Wetenschappelijk Onderzoek.
Received for publication February 23, 1999. Revision received May 11, 1999. Accepted for publication June 3, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. Mol Immunol 34:75–91[CrossRef][Medline]
induces polymeric immunoglobulin
receptor mRNA in human intestinal epithelial cells by a protein
synthesis dependent mechanism. Mol Immunol 30:413–421[CrossRef][Medline]
and gamma interferon. Scand
J Immunol 28:351–357[CrossRef][Medline]
synergistically increase
total polymeric Iga receptor levels in human intestinal epithelial
cells. Role of protein tyrosine kinases. J Immunol 156:4807–4814[Abstract]
with multiple
steroid receptors and identification of an internally deleted ELE1ß
isoform. Mol Endocrinol 13:117–128This article has been cited by other articles:
 |
|
 |
|
  J. Lai, S. A. Myers, M. G. Lawrence, D. M. Odorico, and J. A. Clements Direct Progesterone Receptor and Indirect Androgen Receptor Interactions with the Kallikrein-Related Peptidase 4 Gene Promoter in Breast and Prostate Cancer Mol. Cancer Res., January 1, 2009; 7(1): 129 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Centenera, J. M. Harris, W. D. Tilley, and L. M. Butler Minireview: The Contribution of Different Androgen Receptor Domains to Receptor Dimerization and Signaling Mol. Endocrinol., November 1, 2008; 22(11): 2373 - 2382. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Teyssier, S. Bianco, O. Lanvin, and J.-M. Vanacker The orphan receptor ERR{alpha} interferes with steroid signaling Nucleic Acids Res., September 1, 2008; 36(16): 5350 - 5361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wu, W. Zhao, J. Zhao, J. Pan, Q. Wu, Y. Zhang, W. A. Bauman, and C. P. Cardozo Identification of Androgen Response Elements in the Insulin-Like Growth Factor I Upstream Promoter Endocrinology, June 1, 2007; 148(6): 2984 - 2993. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Schauwaers, K. De Gendt, P. T. K. Saunders, N. Atanassova, A. Haelens, L. Callewaert, U. Moehren, J. V. Swinnen, G. Verhoeven, G. Verrijdt, et al. Loss of androgen receptor binding to selective androgen response elements causes a reproductive phenotype in a knockin mouse model PNAS, March 20, 2007; 104(12): 4961 - 4966. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Peng, P. J. Malloy, J. Wang, and D. Feldman Growth Inhibitory Concentrations of Androgens Up-Regulate Insulin-Like Growth Factor Binding Protein-3 Expression via an Androgen Response Element in LNCaP Human Prostate Cancer Cells Endocrinology, October 1, 2006; 147(10): 4599 - 4607. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Ye, S. J. Han, S. Y. Tsai, F. J. DeMayo, J. Xu, M.-J. Tsai, and B. W. O'Malley Roles of steroid receptor coactivator (SRC)-1 and transcriptional intermediary factor (TIF) 2 in androgen receptor activity in mice PNAS, July 5, 2005; 102(27): 9487 - 9492. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Steketee, A. C J Ziel-van der Made, H. A G M van der Korput, A. B Houtsmuller, and J. Trapman A bioinformatics-based functional analysis shows that the specifically androgen-regulated gene SARG contains an active direct repeat androgen response element in the first intron J. Mol. Endocrinol., October 1, 2004; 33(2): 477 - 491. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. L. Shaffer, A. Jivan, D. E. Dollins, F. Claessens, and D. T. Gewirth Structural basis of androgen receptor binding to selective androgen response elements PNAS, April 6, 2004; 101(14): 4758 - 4763. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. V. Nantermet, J. Xu, Y. Yu, P. Hodor, D. Holder, S. Adamski, M. A. Gentile, D. B. Kimmel, S.-i. Harada, D. Gerhold, et al. Identification of Genetic Pathways Activated by the Androgen Receptor during the Induction of Proliferation in the Ventral Prostate Gland J. Biol. Chem., January 9, 2004; 279(2): 1310 - 1322. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhang, N. Gao, S. Kasper, K. Reid, C. Nelson, and R. J. Matusik An Androgen-Dependent Upstream Enhancer Is Essential for High Levels of Probasin Gene Expression Endocrinology, January 1, 2004; 145(1): 134 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Callewaert, G. Verrijdt, V. Christiaens, A. Haelens, and F. Claessens Dual Function of an Amino-terminal Amphipatic Helix in Androgen Receptor-mediated Transactivation through Specific and Nonspecific Response Elements J. Biol. Chem., February 28, 2003; 278(10): 8212 - 8218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Reid, I. Murray, K. Watt, R. Betney, and I. J. McEwan The Androgen Receptor Interacts with Multiple Regions of the Large Subunit of General Transcription Factor TFIIF J. Biol. Chem., October 18, 2002; 277(43): 41247 - 41253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Verrijdt, K. Schauwaers, A. Haelens, W. Rombauts, and F. Claessens Functional Interplay between Two Response Elements with Distinct Binding Characteristics Dictates Androgen Specificity of the Mouse Sex-limited Protein Enhancer J. Biol. Chem., September 13, 2002; 277(38): 35191 - 35201. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Solorzano-Vargas, J. Wang, L. Jiang, H. V. Tsai, L. O. Ontiveros, M. A. Vazir, R. J. Aguilera, and M. G. Martin Multiple transcription factors in 5'-flanking region of human polymeric Ig receptor control its basal expression Am J Physiol Gastrointest Liver Physiol, August 1, 2002; 283(2): G415 - G425. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Heinlein and C. Chang Androgen Receptor (AR) Coregulators: An Overview Endocr. Rev., April 1, 2002; 23(2): 175 - 200. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Barbulescu, C. Geserick, I. Schuttke, W.-D. Schleuning, and B. Haendler New Androgen Response Elements in the Murine Pem Promoter Mediate Selective Transactivation Mol. Endocrinol., October 1, 2001; 15(10): 1803 - 1816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Grad, L. S. Lyons, D. M. Robins, and K. L. Burnstein The Androgen Receptor (AR) Amino-Terminus Imposes Androgen-Specific Regulation of AR Gene Expression via an Exonic Enhancer Endocrinology, March 1, 2001; 142(3): 1107 - 1116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-J. Lareyre, K. Reid, C. Nelson, S. Kasper, P. S. Rennie, M.-C. Orgebin-Crist, and R. J. Matusik Characterization of an Androgen-Specific Response Region Within the 5' Flanking Region of the Murine Epididymal Retinoic Acid Binding Protein Gene Biol Reprod, December 1, 2000; 63(6): 1881 - 1892. [Abstract] [Full Text]
|
|
|
|
 |
|
 H. Schjerven, P. Brandtzaeg, and F.-E. Johansen Mechanism of IL-4-Mediated Up-Regulation of the Polymeric Ig Receptor: Role of STAT6 in Cell Type-Specific Delayed Transcriptional Response J. Immunol., October 1, 2000; 165(7): 3898 - 3906. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Schoenmakers, G. Verrijdt, B. Peeters, G. Verhoeven, W. Rombauts, and F. Claessens Differences in DNA Binding Characteristics of the Androgen and Glucocorticoid Receptors Can Determine Hormone-specific Responses J. Biol. Chem., April 14, 2000; 275(16): 12290 - 12297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Verrijdt, E. Schoenmakers, A. Haelens, B. Peeters, G. Verhoeven, W. Rombauts, and F. Claessens Change of Specificity Mutations in Androgen-selective Enhancers. EVIDENCE FOR A ROLE OF DIFFERENTIAL DNA BINDING BY THE ANDROGEN RECEPTOR J. Biol. Chem., April 14, 2000; 275(16): 12298 - 12305. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
