| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Endocrinology & Reproduction (H.T.B., A.L.M.B., A.O.B.) and Pathology (J.T.), Erasmus University, Rotterdam, The Netherlands; and Pathologie Molèculaire des Recepteurs Nucléaires, INSERM U-439 (J.M.L., C.S.), and Centre de Biochimie Structurale, UMR 9955, Faculté de Pharmacie, CNRS-INSERM-Université Montpellier I (L.C.), Montpellier, France
Address all correspondence and requests for reprints to: Dr. H. T. Brüggenwirth, Department of Endocrinology & Reproduction, Faculty of Medicine and Health Sciences, Erasmus University, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. E-mail: bruggenwirth{at}endov.fgg.eur.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male sex differentiation and development proceed under direct control
of the male sex hormones testosterone and 5
-dihydrotestosterone, and
the actions of both androgens are mediated by the AR. Mutations in the
AR gene of 46,XY individuals are associated with the androgen
insensitivity syndrome (AIS), a disorder of sex differentiation. Many
abnormalities have been described, causing a wide spectrum of
phenotypes, ranging from subjects with an external female phenotype and
absence of Mullerian and Wolffian duct derivatives, which is the
complete form of AIS, to a phenotype with ambiguous genitalia, called
partial AIS (10). The most frequently reported defects are point
mutations in the ligand- and DNA-binding domains of the AR (10, 11).
In the present study a mutation in exon 2 of a subject with complete AIS is reported. The alanine residue at position 564 in the DBD was substituted into aspartic acid (mutant A564D). The effect of the A564D mutation on AR function was investigated, as was the effect of an asparagine substitution (mutant A564N) and a leucine substitution (mutant A564L) at this same position. These studies were completed with molecular modeling.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-32P]ATP (SA, 3000
Ci/mmol) was obtained from Amersham (Little Chalfont, UK).
17ß-Hydroxy-17-[3H]methyl-4,9,11-estratrien-3-one
([3H]R1881; SA, 85 Ci/mmol), and unlabeled R1881 were
purchased from New England Nuclear-DuPont de Nemours (s’Hertogenbosch,
The Netherlands). The double stranded probe, containing an ARE, derived
from the tyrosine aminotransferase (TAT) promoter
5'-TCGACTGTACAGGATGTTCTAGCTACT-3' (half-sites
are underlined) was obtained from Promega (Woerden, The
Netherlands). The 27-bp oligonucleotides, used to produce two other
double stranded probes,
5'-TCGACGTTACAAACTGTTCTAGCTACT-3' and
5'-TCGACGGTACAGTTTGTTCTAGCTACT-3' (half-sites
are underlined), containing, respectively, the strongest ARE
from the mouse mammary tumor virus (MMTV) promoter (12) and a consensus
ARE (13), were obtained from Pharmacia Biotech Benelux.
Clinical data
The patient exhibiting female external genitalia, atrophic
epididymides, and vasa deferentia was diagnosed as having AIS at the
age of 1 yr in the absence of a positive family history. A blind ending
vagina was present, the uterus was absent, and testes with a normal
histology for a boy of this age were present in the inguinal canal.
Testosterone synthesis disorders were excluded as a cause of the 46,XY
sex reversal. Genital skin fibroblasts (GSF) were obtained from the
index patient for Scatchard analysis and structural analysis of the AR
gene. AR sequence analysis of relatives was performed on white blood
cell genomic DNA.
Mutation detection
PCR-single strand conformation polymorphism analysis and direct
sequencing were performed as described previously (14).
Ligand binding study
For determination of ligand binding characteristics of the AR of
the AIS subject, GSF were incubated with serial dilutions
[3H]R1881 (0.02, 0.05, 0.3, 1.0, and 3.0 nM,
respectively) in serum-free medium. The binding assay was performed as
described previously (14).
Western blot analysis
AR protein, derived from GSF or transiently transfected Chinese
hamster ovary (CHO) cells was immunoprecipitated and analyzed by
Western immunoblotting according to the method of Ris-Stalpers et
al. (15).
Construction of AR expression vectors
pAR(0), a human wild-type AR complementary DNA expression
plasmid, was described previously (1). Expression plasmids encoding the
various mutants, pAR(A564D), pAR(A564N), and pAR(A564L), respectively,
were constructed by site-directed mutagenesis. The
KpnI-AspI-digested fragment of pAR(0) was
exchanged with mutated KpnI-AspI fragments,
generated in two separate PCR amplifications (16). Sense primer 470A
(14), located upstream of the KpnI site in exon 1, was
combined with an antisense primer, containing the mutation (antisense
primers: construct A564D, 5'-CATGTGAGAtCTCCATAGTGACAC-3'; construct
A564N, 5'-CATGTGAGAttTCCATAGTGACACCC-3'; construct A564L,
5'-CATGTGAGAagTCCATAGTGACACCC3'), and a sense primer, introducing the
mutation (sense primers: construct A564D,
5'-GTGTCACTATGGAGaTCTCACATG-3'; construct A564N,
5'-GGTGTCACTATGGAaaTCTCACATGTGG-3'; construct A564L,
5'-GGTGTCACTATGGActTCTCACATGTGG-3'), was used in combination with an
antisense primer, 14NB (14), located downstream of a unique
AspI site in exon 4. One microliter of both PCR products was
used as template in a second PCR reaction using primers 470A and 14NB.
The resulting PCR fragment was digested with KpnI and
AspI, and exchanged for the corresponding wild-type fragment
in pAR0.
pSG5AR(0), a human wild-type AR complementary DNA expression vector (provided by Dr. A. C. B. Cato, Karlsruhe, Germany) was used for transient transfection of COS-1 cells. pSG5AR(A564D), pSG5AR(A564N), and pSG5AR(A564L) were constructed by exchanging the 472-bp KpnI-AspI insert from pSG5AR(0) and the KpnI-AspI fragment from the pSVAR plasmids, encoding the variant ARs.
MMTV-LUC reporter plasmid, cytomegalovirus (CMV)-LUC, the CMV-(ARE)3-LUC reporter plasmids, and pJH4-(ARE)2-TATA-LUC, containing the TATA box and an Sp1 site derived from the Oct-6 gene promoter have been described previously (17, 18, 19).
Cell culture conditions and transfections
GSF and COS-1 cells were cultured as previously described (20).
CHO cells were cultured according to the COS-1 cell culture protocol.
The CHO cells used for transcription activation studies and promoter
interference assays were plated in 7- or 11-cm2 (promoter
interference assay) wells and grown for 24 h. Cells were
cotransfected, using the calcium-phosphate method, with AR expression
plasmid (10 ng/ml precipitate) and reporter plasmid (2 µg/ml
precipitate) (21). Carrier DNA (pTZ19) was added to a total of 20 µg
DNA/ml precipitate, and 90 µl precipitate were added per well. In the
promoter interference assay, 300 ng AR expression plasmid and 30 ng
reporter plasmid [CMV-LUC or CMV-(ARE)3-LUC],
respectively, were added per ml precipitate. pTZ19 was added to a total
of 20 µg DNA/ml precipitate, and 250 µl precipitate were added to
11-cm2 wells. The transfection and luciferase assay were
performed as described previously (22). Both transcription activation
studies and the promoter interference assay were performed at least
three times in triplicate, using three independent isolates of
expression plasmid. In the case of transcription activation studies,
luciferase activity was expressed relative to basal activity measured
after culture in the absence of hormone. For promoter interference
studies, luciferase activity in cells, transfected with
CMV-(ARE)3-LUC and AR expression plasmid and cultured in
the absence of hormone was set at 100%. Inhibition of promoter
activity in the presence of hormone was expressed relative to this
100% activity. CHO cells used for expression studies by Western
blotting were also transiently transfected by the calcium-phosphate
method. To this end, cells were plated in 175-cm2 culture
flasks and transiently transfected with 20 µg expression plasmid.
COS-1 cells were plated in 80-cm2 culture flasks and
transiently transfected with 9.4 µg expression plasmid, using the
diethylaminoethyl-dextran method (23). Twenty-four hours before
harvesting, CHO and COS-1 cells were washed and incubated with medium
containing 1 nM R1881.
Gel retardation assays
Transfected COS-1 cells were collected in 5 ml PBS, and the
pellet was resuspended in extraction buffer [10 mM
NaH2PO4 (pH 7.4), 0.4 M KCl, 1
mM EDTA, 10% (vol/vol) glycerol, 0.5 mM
bacitracin, 0.5 mM leupeptin, 0.6 mM
phenylmethylsulfonylfluoride, and 10 mM dithiothreitol]
and subjected to four freeze-thaw cycles, followed by 10-min
centrifugation at 400,000 x g at 4 C in a TLA120.2
rotor (Beckman, Fullerton, CA) in a Beckman Optima TLX ultracentrifuge.
The TAT ARE-containing probe
(5'-TCGACTGTACAGGATGTTCTAGCTACT-3') (halfsites
are underlined) was obtained from Promega. Two other probes
were produced by annealing a 27-bp oligonucleotide with an
oligonucleotide of complementary sequence. One of them
(5'-TCGACGTTACAAACTGTTCTAGCTACT-3') (half-sites
are underlined) contains the strongest ARE from the MMTV
promoter (12), and the other probe
(5'-TCGACGGTACAGTTTGTTCTAGCTACT-3') (half-sites
are underlined) contains a consensus ARE (13). The
ARE-containing probes were end labeled using T4 polynucleotide kinase,
and double stranded probe was purified from a 4% acrylamide gel in
0.5 x TBE (1 x TBE = 50 mM Tris base, 50
mM boric acid, and 1 mM EDTA, pH 8.6). Cellular
extracts were incubated in binding buffer [10 mM HEPES (pH
7.9), 60 mM KCl, 1 mM dithiothreitol, 1
mM EDTA, and 4% Ficoll] and 1 µg
polydeoxyinosinic-deoxycitidylic acid in the absence or presence of the
polyclonal AR antibody Sp197 (10-fold diluted) (24). After an
incubation period of 10 min on ice, 2 µl purified DNA probe (50,000
cpm/µl) were added, and incubation was continued for 20 min at room
temperature. The 20-µl sample was separated on a 4% polyacrylamide
gel in 0.5 x TBE. Gels were fixed for 10 min in 10% acetic
acid-10% methanol, and subsequently dried and exposed.
Molecular modeling
The crystal structure of the rat GR DBD bound to a
glucocorticoid response element was used as a basis upon which the
three-dimensional (3-D) AR models were built. The 3-D model of AR bound
to a glucocorticoid response element has previously been described
(25). The A564D, A564N, and A564L mutants were built according to the
same strategy as that used to build the wild-type model. Briefly, the
side-chains of the AR mutants that were substituted in the GR model
were placed in energetically favorable conformations using the SMD
program (26). The whole system was then energy minimized with the AMBER
program (Pearlman et al., 1991, University of California,
San Francisco, CA). During the optimization process, the
oligonucleotide was kept frozen to prevent unrealistic deviation from
the initial crystal structure. Moreover, positional restraints on the
backbone and on side-chains of conserved residues were applied and
gradually released during the optimization. Figure 5A
was generated
with the Insight II viewer (Biosym Technologies, San Diego, CA), and
Fig. 5B was generated using the program MOLSCRIPT (27).
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The numbering of amino acid residues
throughout the text is based on a total number of 910 amino acid
residues in the human AR (1). The mutation created a BglII
site, which was used to investigate the segregation of this mutation in
the family of the index patient. The mother and grandmother of the
index patient were heterozygous carriers of this AR mutation (results
not shown).
|
Transcriptional activity of AR A564D
AR A564D was inactive in CHO cells cotransfected with
(ARE)2-TATA-LUC, in contrast to the wild-type AR (Fig. 2A). The promoter of this reporter
construct contains a TATA box and an Sp1 site derived from the Oct-6
gene promoter, and two AREs (19). In general, comparable levels of
wild-type and mutant receptor protein were expressed in transiently
transfected CHO cells, as verified by SDS-PAGE and immunoblotting (for
example, see Fig. 3C). AR A564D was also
unable to activate transcription from the complex MMTV promoter in CHO
cells (Fig. 2B).
|
|
). The AR could
sequester factors that are essential for the transcriptional activity
of the CMV promoter (squelching). However, no reduction of luciferase
expression was seen in cells cotransfected with CMV-LUC (Fig. 2C).
Transcriptional activities of AR A564N and AR A564L
To investigate whether the inactivity of AR A564D was caused by
steric hindrance or by a conformational change due to the introduction
of a negative charged amino acid residue, the alanine residue was
replaced by either a leucine residue (A564L) or an asparagine residue
(A564N). Leucine has, like the aspartic acid residue, a larger
side-chain, but is a neutral amino acid, as is the alanine residue
present in the wild-type receptor. Asparagine has also a larger
side-chain, but is a polar amino acid residue. CHO cells were
transiently cotransfected with AR expression plasmids and
(ARE)2-TATA-LUC. Wild-type AR and AR A564L showed
comparable activation of the minimal promoter at increasing amounts of
R1881, whereas AR A564N showed strongly reduced transcription
activation compared with the wild-type AR (Fig. 3A). On the more
complex MMTV promoter, AR A564L showed activity comparable to that of
the wild-type receptor, and AR A564N displayed a low level of
hormone-induced transcription activation (Fig. 3B). All proteins were
expressed, and in general, expression levels were comparable (Fig. 3C).
In vivo DNA binding of AR A564N and AR A564L
DNA binding was studied in CHO cells cotransfected with
CMV-(ARE)3-LUC (Fig. 3D). In contrast to the wild-type AR,
luciferase expression was not lowered after cotransfection of cells
with AR A564N and culture in the presence of hormone. AR A564L showed
38% inhibition, which was significantly different (P
< 0.05) from the 48% inhibition observed for the wild-type receptor.
Protein expression levels were identical for all mutant receptors (see
also Fig. 3C).
In vitro DNA binding comparing different AREs
The DNA-binding capacities of the wild-type and the different AR
mutants were tested in vitro in gel retardation assays,
using probes containing various AREs. AR was produced in transfected
COS-1 cells. The amount of receptor protein was checked by Western
blotting, followed by immunostaining. Comparable amounts of AR in COS-1
cellular extracts were incubated with a 32P-labeled probe
in either the absence or presence of polyclonal antibody Sp197, which
stabilizes AR dimers bound to the DNA (24). Wild-type AR and AR A564L
did bind to the ARE derived from the TAT promoter (Fig.
A, lanes 3 and 9),
whereas binding of AR A564D and AR A564N could not be detected (Fig. 4A, lanes 5 and 7). Gel shift assays were also performed with two other
probes. One of the probes contained the strongest ARE from the MMTV
promoter (12), which was also present in (ARE)2-TATA-LUC.
The other probe contained a consensus ARE (13), which was also cloned
behind the constitutively active CMV promoter, that was used for
promoter interference studies. Wild-type AR was able to shift the
probes in the presence of antibody, indicative of specific DNA binding
(Fig. 4B, lane 3, and Fig. 4C, lane 3). Neither probe could be shifted
with AR A564D (Fig. 4B, lane 5, and Fig. 4C, lane 5) or AR A564N (Fig. 4B, lane 7, and Fig. 4C, lane 7), although A564N showed transcription
activation on a complex MMTV promoter and even on a minimal ARE
promoter. AR A564L interacted with both probes, although less
efficiently than the wild-type AR, which is in agreement with the
results of the promoter interference assay (Fig. 4B, lane 9, and 4C,
lane 9).
|
and 5, A and B). The C-Cß bond of
the alanine residue at position 564 is directed toward the cysteine
residue at position 610 in the protein core. The backbone of residue
564 is hydrogen bonded with the backbone of histidine residue 561, as
both residues belong to a ß-hairpin (Fig. 5B). They are located at
the same side of the hairpin at facing positions. The histidine residue
at position 561 is involved in direct contacts with DNA and
participates in ARE recognition (Fig. 5B). Molecular modeling showed
that in the A564D mutant, the aspartic acid residue is still buried.
However, burying of charged residues is unfavorable unless a
compensatory charge forms a salt bridge and is also buried at the same
site. The only way AR A564D can take a more favorable conformation is
by breaking of the ß-hairpin and the hydrogen bonding with the
histidine residue at position 561. This most likely affects the
conformation of the histidine residue 561, resulting in disturbed ARE
recognition. The asparagine residue in A564N should lead to smaller
perturbations because it will remain buried, although asparagine is a
polar residue. The modeling showed that hydrogen bonding of asparagine
with threonine 566 may roughly compensate for the unfavorable burying
of polar atoms. Modeling of the A564L mutant showed that the larger
leucine side-chain can be accommodated without difficulties in the
hydrophobic pocket. Burying of the larger hydrophobic surface may even
provide additional stability to the AR. Therefore, the leucine mutant
was not expected to significantly perturb DNA recognition. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-helical region that is exposed to the major groove of DNA and
downstream of cysteine 560, histidine 561, and tyrosine 562, which are
involved in specific as well as nonspecific contacts with the phosphate
backbone of the DNA (28). Substitution of the latter residues by
nonconservative amino acids in the GR resulted in loss of function
(in vivo), and in vitro reduced DNA binding
affinity was seen (35). Warriar et al. (36) substituted the
cysteine residue at position 560 in the human AR by a serine residue.
Although this is a relatively conservative change, decreased DNA
binding and transcription activation were observed, which were
attributed to the instability of the AR mutant-DNA complex.
We showed that AR A564D displayed defective transcription activation.
Specific binding to DNA, which was studied in vitro by gel
retardation assays with oligonucleotide probes containing different
AREs and in vivo with a promoter interference assay, could
not be detected. From 3-D modeling studies, it became clear that the
alanine residue at position 564 is buried in a hydrophobic cluster
(Figs. 1 and 5A). A hydrogen bond is formed between the backbone of
alanine 564 and the backbone of histidine 561. This latter residue is
involved in direct interaction with the phosphate backbone of the DNA
and is conserved in other nuclear receptors. Yagi et al.
(37) reported a patient with hereditary 1,25-dihydroxyvitamin
D-resistant rickets caused by substitution of the conserved histidine
at position 35 of the vitamin D receptor, comparable to histidine 561
in the AR. The phenotype of the patient was caused by perturbation of
the conserved site that contacts the phosphate backbone of DNA. For AR
A564D, the computer model displayed almost the exact conformation as
that seen for the wild type, showing that there is no steric hindrance
due to the mutation. However, buried charged residues are only observed
in proteins when they can form salt bridges with residues of opposite
charge. Therefore, the modeled conformation appears unlikely. Probably,
aspartic acid 564 adopts a more favorable conformation, and as a
result, the main chain hydrogen bond between aspartic acid 564 and
histidine 561 will be broken, which has consequences with respect to
DNA binding.
To investigate the role of the alanine residue at position 564 in more detail, the residue was also replaced by an asparagine or a leucine residue. 3-D modeling predicted that introduction of an asparagine residue should have intermediate effects with respect to transcription activation. Hydrogen bonding between asparagine 564 and threonine 566 might compensate for unfavorable burying of the polar amino acid residue. Substitution by an asparagine residue resulted in a less stable AR-DNA complex, which had clear consequences for transcription activation on a minimal (ARE)2-TATA promoter and to a lesser extent on a complex MMTV promoter. DNA binding was not observed in vitro by gel retardation analysis. DNA-binding capacity remained undetectable, even when studied in whole cells. However, the functionality of AREs is determined by additional transcription factor-binding sites in the vicinity of AREs. Interaction with other proteins might stabilize the AR-DNA complex. This might explain as well the stronger activation of AR A564N on the MMTV promoter compared with its activity on the minimal (ARE)2-TATA promoter (8, 38, 39).
Modeling showed that no particular constraint resulted from the larger size of the leucine residue. It was predicted that the leucine residue, because of its larger hydrophobic surface, provides even more favorable stability to the AR. However, reduced DNA binding affinity, in vitro as well as in vivo, was observed for AR A564L. Transcription activation was comparable with activation displayed by the wild-type AR. Apparently, molecular modeling has some limitations with respect to the prediction of complex interactions, which might be explained by the fact that the model was based upon the crystal structure, which was solved for the GR DBD. In addition, functional studies were performed with the intact receptor and not only the DBD.
In conclusion, the negative charge introduced by the aspartic acid residue destabilizes the normal conformation of the AR DBD, resulting in disturbed ARE recognition, in agreement with the phenotype of the patient expressing this mutant receptor. Results from the functional assays were partially supported by predictions made by 3-D modeling. Although not predicted by molecular modeling, steric hindrance might have an impact on the DNA-binding capacities of AR A564N and AR A564L.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 10, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  P. Farla, R. Hersmus, J. Trapman, and A. B. Houtsmuller Antiandrogens prevent stable DNA-binding of the androgen receptor J. Cell Sci., September 15, 2005; 118(18): 4187 - 4198. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Dubbink, R. Hersmus, C. S. Verma, H. A. G. M. van der Korput, C. A. Berrevoets, J. van Tol, A. C. J. Ziel-van der Made, A. O. Brinkmann, A. C. W. Pike, and J. Trapman Distinct Recognition Modes of FXXLF and LXXLL Motifs by the Androgen Receptor Mol. Endocrinol., September 1, 2004; 18(9): 2132 - 2150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Buchanan, N. M. Greenberg, H. I. Scher, J. M. Harris, V. R. Marshall, and W. D. Tilley Collocation of Androgen Receptor Gene Mutations in Prostate Cancer Clin. Cancer Res., May 1, 2001; 7(5): 1273 - 1281. [Abstract] [Full Text]
|
|
|
|
 |
|
 J.-M. Lobaccaro, N. Poujol, B. Térouanne, V. Georget, S. Fabre, S. Lumbroso, and C. Sultan Transcriptional Interferences between Normal or Mutant Androgen Receptors and the Activator Protein 1--Dissection of the Androgen Receptor Functional Domains Endocrinology, January 1, 1999; 140(1): 350 - 357. [Abstract] [Full Text]
|
|
|
|
 |
|
 V. Georget, B. Térouanne, S. Lumbroso, J.-C. Nicolas, and C. Sultan Trafficking of Androgen Receptor Mutants Fused to Green Fluorescent Protein: A New Investigation of Partial Androgen Insensitivity Syndrome J. Clin. Endocrinol. Metab., October 1, 1998; 83(10): 3597 - 3603. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
D mutation,
located in the first zinc cluster. The mutation found in the index
patient’s AR is present at position 564 of the AR DBD, located near
the P box, of which the circled amino acid residues are
involved in ARE recognition. The boxes indicate amino
acid residues that interact with the phosphate backbone of DNA either
at specific (black boxes) or at nonspecific sites (open
boxes) (25, 28). The second zinc cluster contains the D box,
which is involved in dimerization with the other AR receptor. The first
part of the nuclear localization signal (NLS) is also shown
(underlined). The numbering of the various codons is
based on a total of 910 amino acid residues in the human AR (1).
, A564D). Transcription activation
by AR A564D was significantly different from that by the wild-type AR
(by Student’s t test, P = 0.05). B,
Transcriptional activity of AR A564D at the MMTV promoter.
Transcription activation properties were determined by cotransfection
of CHO cells with MMTV-LUC reporter plasmid and either the wild-type or
mutant ARA564D expression plasmid. After 24 h, the cells were
incubated with different concentrations of R1881 and incubated for
another 24 h before a luciferase assay was performed. Each
data point was tested in triplicate. Induction of
luciferase activity was calculated from three different experiments;
symbols represent the mean ± SEM (•, wild type;
, A564L). Transcription activation by AR A564N was
significantly different from that displayed by the wild-type AR (by
Student’s t test, P = 0.05). B,
Transcription activation by the wild-type receptor, AR A564N, and AR
A564L, using an MMTV-regulated reporter gene. CHO cells were
cotransfected with MMTV-LUC reporter plasmid and either the wild-type
or one of the mutant expression vectors. After 24 h, cells were
cultured with increasing concentrations of R1881 for 24 h before a
luciferase assay was performed. Each data point was
tested in triplicate. Induction of luciferase activity, presented
relative to basal activity in the absence of R1881, was calculated from
five different experiments; symbols represent the mean ±
SEM (•, wild type; 