Characterization of the Ah Receptor-associated Protein, ARA9

Abstract

The unliganded aryl hydrocarbon receptor (AHR) is found in a complex with other proteins including the 90-kDa heat shock protein (Hsp90) and a 37-kDa protein we refer to as ARA9. We found that the three tetratricopeptide repeats found in the COOH terminus of ARA9 are necessary and sufficient for interaction with the AHR complex. Conversely, the AHR's "repressor"/Hsp90 binding domain is required for interaction with ARA9. Because ARA9 closely resembles the 52-kDa FK506-binding protein (FKBP52), found in the unliganded glucocorticoid receptor (GR) complex, we compared the binding specificities of ARA9 and FKBP52 for AHR and GR. In co-immunoprecipitation experiments, ARA9 specifically associated with AHR-Hsp90 complex but not with GR-Hsp90 complexes. In addition, ARA9 showed a greater capacity than FKBP52 to associate with AHR-Hsp90 complexes. The biological importance of this interaction was suggested by the observation that in a yeast expression system ARA9 expression enhanced the response of AHR to the agonist beta-napthoflavone, decreasing the EC50 by greater than 5-fold and increasing the maximal response 2.5-fold. In contrast, co-expression of FKBP52 had no effect on AHR signaling. In addition, although ARA9 contains a domain similar to that found in other FK506-binding proteins, ARA9 binding to 3H-FK506 could not be detected. Finally, we have characterized the developmental and expression pattern of ARA9 in the developing mouse embryo and mapped the ARA9 locus to human chromosome 11q13.3.

Full text

Characterization of the Ah Receptor-associated Protein, ARA9*
(Received for publication, August 10, 1998, and in revised form, September 2, 1998)
Lucy A. Carver‡§¶, John J. LaPres‡¶, Sanjay Jaini, Elizabeth E. Dunham‡,
and Christopher A. Bradfield‡**
From the ‡McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, Wisconsin 53706
and the iDepartment of Molecular Pharmacology and Biological Chemistry, Northwestern University Medical School,
Chicago, Illinois 60611
The unliganded aryl hydrocarbon receptor (AHR) is
found in a complex with other proteins including the
90-kDa heat shock protein (Hsp90) and a 37-kDa protein
we refer to as ARA9. We found that the three tetratri-
copeptide repeats found in the COOH terminus of ARA9
are necessary and sufficient for interaction with the
AHR complex. Conversely, the AHR’s “repressor”/Hsp90
binding domain is required for interaction with ARA9.
Because ARA9 closely resembles the 52-kDa FK506-bind-
ing protein (FKBP52), found in the unliganded glucocor-
ticoid receptor (GR) complex, we compared the binding
specificities of ARA9 and FKBP52 for AHR and GR. In
co-immunoprecipitation experiments, ARA9 specifically
associated with AHR-Hsp90 complex but not with GR-
Hsp90 complexes. In addition, ARA9 showed a greater
capacity than FKBP52 to associate with AHR-Hsp90
complexes. The biological importance of this interaction
was suggested by the observation that in a yeast expres-
sion system ARA9 expression enhanced the response of
AHR to the agonist
b
-napthoflavone, decreasing the
EC
50
by greater than 5-fold and increasing the maximal
response 2.5-fold. In contrast, co-expression of FKBP52
had no effect on AHR signaling. In addition, although
ARA9 contains a domain similar to that found in other
FK506-binding proteins, ARA9 binding to
3
H-FK506
could not be detected. Finally, we have characterized
the developmental and expression pattern of ARA9 in
the developing mouse embryo and mapped the ARA9
locus to human chromosome 11q13.3.
The AHR
1
is a ligand-activated transcription factor that
mediates the adaptive and toxic responses to environmental
pollutants like dioxin (1). Upon binding agonist, the AHR
dimerizes with a structurally related protein, ARNT, and this
complex interacts with enhancer elements upstream of target
promoters and up-regulates the transcription of a variety of
xenobiotic metabolizing enzymes. The AHR and ARNT are both
members of the basic helix loop helix-PAS superfamily. The
helix loop helix domain serves as a dimerization surface to
allow AHR-ARNT interactions and also positions the basic
a
-helix within the major groove of B-DNA to allow specific
interactions with target enhancer elements (2–4). The PAS
domain, a region of ;250 amino acids, functions as a dimeriza-
tion surface, harbors a repressor region, and also domains that
are required for binding agonists and forming interactions with
the molecular chaperone Hsp90 (2, 5, 6).
Recently, yeast two-hybrid screens and biochemical ap-
proaches have identified a 37-kDa protein that is associated
with the AHR (7–9). We refer to this “Ah receptor-associated
protein” as ARA9 while other groups refers to it as AIP or XAP2
(7, 8). The protein ARA9 displays structural similarity to the
glucocorticoid receptor-associated immunophilin FKBP52. In
its NH
2
-terminal half, ARA9 displays amino acid sequence
identity to a region in FKBP52 that is known to harbor both
peptidylprolyl cis-trans isomerase activity and a high affinity
binding site for the immunosuppressant macrolide, FK506. In
its COOH-terminal half, ARA9 harbors amino acid sequence
identity to a region in FKBP52 that harbors multiple TPR
motifs (7). In an effort to understand the role of ARA9 in AHR
signaling, we have defined the domains within these two pro-
teins that are required for their interaction and also demon-
strate that ARA9 expression affects both the potency and effi-
cacy of AHR agonists in the yeast Saccharomyces cerevisiae.We
have also detailed the developmental and tissue specific ex-
pression pattern of ARA9 in the developing mouse embryo and
defined its chromosomal location.
MATERIALS AND METHODS
Oligonucleotide Sequences—The following sequences were used:
OL286, CGGGATCCAAGGAATTCAGCAAGCCACTGCAGG; OL287,
CGGGATCCGATGGCTCATCTGCTTCTGTTGCC; OL303, CGGGATC-
CCAATGGACTCCAAAGAATCATTAACTCC; OL304, CGGGATCGGC-
AGTCACTTTTGATGAAACAGAAG; OL803, CGGAGATCTGAGGCTA-
TGCTTCTGTCTCCACCT; OL813, GCGGAATTCATCTTGCAGCTGC-
AGCAGTGGGCTTGG; OL814, CTCTTCGTCTGTCATGGCCCATGG;
OL824, GCGGAATTCAGGAAAATGGCGCTAGCCGGAAGC; OL838,
GGTAGATCTGGTAAGGCAGGGCCAAGTGCTCC; OL867, GGTACA-
GATCTAACGATGGCGGATATCATCGCACGCCTC; OL869, GGCGA-
ATTCACGATGTGCTGCGGTGTTGCACAGATGCG; OL963, GCCGA-
ATTCACGATGGCCGCCGAGGAGATGAAGGC; OL1066, CGGGAAT-
TCGAGGCTATGCTTCTGTCTCCACCT; OL1151, GCGAATTCGCCA-
CCATGGCGGATATCATCGCAAGACTCCGG; OL1152, GCGAATTCG-
CCACCATGCTGAAGGTGGAGAGCCCTGGC.
Strains and Plasmids—S. cerevisiae strain L40 (MAT a,his3200,
trp1–901, leu2–3, 112, ade2, LYS2::(lexAop)
4
-HIS3, URA3:: (lexAop)
8
-
lacZ, gal80) was used in both interaction and pharmacology experi-
ments (10). The GR yeast expression vector, pC7-rGR, was a generous
gift of Dr. Didier Picard (Univeristy of Geneva, Geneva, Switzerland).
Plasmid pBTM116 is a 2-
m
mTRP-marked ADH1 driven expression
vector containing the full-length Escherichia coli LexA cDNA, followed
by a polylinker for generation of fusion proteins (11). The plasmid
pYPGE2 is a 2-
m
mTRP-marked expression vector driven by the PGK1
promoter (12). The plasmid pYX242 (Novagen, Madison, WI) is a 2-
m
m
* This work was supported by the Burroughs Wellcome Foundation
and the National Institutes of Health Grants ES05703, CA07175, and
ES07015. The costs of publication of this article were defrayed in part
by the payment of page charges. This article must therefore be hereby
marked “advertisement” in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
§ Current address: Dept. of Biology, University of California, San
Diego, 9500 Gilman Dr., La Jolla, CA 92093-0349.
¶Contributed equally to the results of this report.
** To whom correspondence should be sent: McArdle Laboratory for
Cancer Research, 1400 University Ave., Madison, WI 53706-1599. Tel.:
608-262-2024; Fax: 608-262-2824; E-mail: bradfield@oncology.wisc.edu.
1
The abbreviations used are: AHR, Ah receptor; PAS, PER, ARNT,
SIM homology domain; GR, glucocorticoid receptor; Hsp90, 90-kDa heat
shock protein; PAGE, polyacrylamide gel electrophoresis; FKBP,
FK506-binding protein; MOPS, 4-morpholinepropanesulfonic acid;
b
-NF,
b
-napthaflavone.
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 273, No. 50, Issue of December 11, pp. 33580–33587, 1998
© 1998 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
This paper is available on line at http://www.jbc.org33580

LEU-marked expression plasmid with a TP1 promoter. The plasmid
pGAD424 is a 2-
m
mLEU-marked vector for construction of GAL4
transcriptional activation domain fusion proteins (13). The plasmid
pBTMAHRND166 (PL703) was constructed by subcloning the EcoRI
fragment of pEGAHRND166 (PL700) into the EcoRI site of pBTM116
(14). The plasmid pGRNLxC (PL474) is a bacterial expression vector
which contains the full-length GR cDNA with the DNA-binding domain
replaced in context with the bacterial LexA DNA-binding domain
(15). The plasmid pY2NLxC (PL740) was constructed by amplifying
PL474 with the primers OL303 and OL304. The product was cloned
into PGEM-Teasy (Promega) (PL734), and the BamHI fragment was
subcloned into the BamHI site of pYPGE2 to yield PL740. The plas-
mids pAHRND130, pAHRND287, pSportAHR, pSportARNT, pSpor-
tAHRCD237, pSportAHRCD313, pSportAHRCD425, pSportAHRCD458,
and pSportAHRCD516 have all been previously described (5, 7, 16). The
full-length ARA9 cDNA was amplified from PL580 (7) with OL824 and
OL813 and cloned into the EcoRI site of pSV-SPORT (Life Technologies,
Inc.) to make pSportARA9 (PL613). The plasmid pYXARA9 (PL810)
was constructed by subcloning the full-length ARA9 cDNA from PL613
into the EcoRI site of pYX242. The plasmid pET17b-ARA9 (PL799) was
constructed by amplifying PL580 with OL867 and OL838 followed by
BglII digestion and the resultant fragment subcloned into the BamHI
site of the pET-17b vector (Novagen). The plasmid pET17b-FKBP52
(PL947) was constructed by amplifying FKBP52 from pGEM7Zf-
FKBP52 (17) with OL963 and OL1066 and subcloned into the PGEM-T
easy vector. The resulting clone was digested with EcoRI and subcloned
into the EcoRI site of the pET-17b vector to generate PL947. The
carboxyl-terminal deletion of ARA9 was constructed by amplifying
PL580 with OL1151 and OL814 and cloning of the fragment into
PGEM-T easy. The resulting plasmid was digested with EcoRI and the
fragment was subcloned into the corresponding site of pET-17b to
create pARA9DC (PL1005). The amino-terminal deletion of ARA9 was
created in a similar fashion using OL1152 and OL813 to create
pARA9DN (PL1004). Plasmids pYXFKBP52 and pGADFKBP52 were
designed to express the full-length FKBP52 and an FKBP52-GAL4-
TAD fusion in yeast. They were constructed by amplifying the FKBP52
cDNA from pGEM7Zf-FKBP52 with OL963 and OL803 and cloning it
into PGEM-Teasy. The BamHI fragment was then subcloned into the
BamHI site of pYX242 or pGAD424. The plasmid pSportAHRD290–
491, PL248, was constructed by subcloning the NotI/HindIII fragment
of the murine AHR into the corresponding sites of pSportAHRCD516.
Co-immunoprecipitation Assays—For AHR domain mapping, pro-
teins were expressed in vitro using the TNT-coupled rabbit reticulocyte
lysate system (Promega). In these experiments, the full-length ARA9
protein was expressed in the presence of [
35
S]methionine for detection
by autoradiography. In our hands, this expression system typically
produces approximately 1 fmol of
35
S-labeled protein per 5
m
l of reac-
tion. Protein complexes were formed by mixing 5
m
l of the appropriate
in vitro translated proteins in a 1.5-ml microcentrifuge tube containing
500
m
l of MENG buffer (25 mMMOPS, 0.025% NaN
3
,1mMEGTA, 10%
(v/v) glycerol, 15 mMNaCl, 1 mMdithiothreitol, 0.1% (v/v) Nonidet P-40,
pH 7.5), then incubated for 90 min at 4 °C. Immunoprecipitations were
performed as described previously, using either BEAR-1 or G1295 as
the primary antibody (7, 18).
Bacterial Expression—The ARA9 and FKBP52 proteins tagged with
the T7 peptide (PL799, PL1004, PL1005 and PL947) were expressed in
BL21(DE3) bacteria (Novagen). The T7 tag is an 11-amino acid peptide,
MASMTGGQQMG, that facilitates detection and purification of the
recombinant protein using anti-T7 antibodies (Novagen). Saturated
cultures were used to inoculate Luria broth growth media supple-
mented with 0.4 mMisopropyl-1-thio-
b
-D-galactopyranoside and 100
m
g/ml ampicillin. The bacteria were allowed to grow an additional4hat
37 °C and pelleted by centrifugation at 5000 3gfor 5 min. The pellet
was resuspended in bind/wash buffer (4.29 mMNa
2
HPO
4
, 1.47 mM
KH
2
PO
4
, 2.7 mMKCl, 0.137 MNaCl, 0.1% (v/v) Tween, and 0.002% (w/v)
NaN
3
). The resulting mixture was disrupted using 4 315-s sonic pulses
on ice. Insoluble debris was removed by centrifugation at 12,000 3gfor
20 min at 4 °C. The supernatant was passed through a 0.45-
m
m filter
and protein concentrations determined by the BCA protein assay
(Pierce). The bacterial expression of FKBP52 and ARA9 was confirmed
by SDS-PAGE followed by both Coomassie stain and Western blot
analysis with an anti-T7 peptide antibody (Novagen). Immunoprecipi-
tations of complexes with either AHR or Hsp90 were performed in
MENG buffer as above, using anti-T7-agarose (Novagen). The com-
plexes were then precipitated and analyzed by SDS-PAGE and
autoradiography.
Purified ARA9 was prepared from BL21 cells that had been induced
by the strategy described above. Following induction, the bacteria were
pelleted by centrifugation (5,000 3g) and resuspended in MENG
buffer. The resulting mixture was disrupted using 4 315-s sonic pulses
on ice and cleared by centrifugation (17,000 3g). The resulting super-
natant was cut with 35–45% ammonium sulfate and centrifuged at
17,000 3g. The pellet was washed in ammonium sulfate (45%) and
resuspended in MENG buffer. The resulting solution was purified using
high performance liquid chromatography on a POROS 10 HQ column
using NaCl gradient from 200–800 mM(PerSeptive Biosystems,
Cambridge MA).
Pharmacology in Yeast—AHR dose-response curves were performed
in the S. cerevisiae strain L40 as described previously (14). Two days
after plasmid transformation, the colonies were replica plated on media
with various concentrations of
b
-NF spread on the surface. The replica
plates were then incubated for 2 days at 30 °C. For each dose of
b
-NF,
three colonies were suspended in 100
m
l of Z-buffer (60 mMNa
2
HPO
4
,
40 mMNaH
2
PO
4
,10mMKCl, 1 mMMgSO
4
,35mM2-mercaptoethanol)
and cell density was estimated by reading the A
600
. Twenty microliters
of the remaining cell suspension was added to 130
m
l of Z-buffer, 25
m
l
of 0.1% SDS (w/v), 25
m
l of CHCl
3
and disrupted at high speed for 10 s.
Thirty microliters of a 4 mg/ml o-nitrophenyl-
b
-D-galactoside solution
was added and the mixture incubated for 2–5 min at 30 °C. The reaction
was stopped by adding 75
m
lof1MNaCO
3
, the cell debris removed by
centrifugation and the A
420
was determined.
b
-Galactosidase units
were determined using the following formula (A
420
/(A
600
of 1/10 dilution
of cells 3volume of culture 3time of incubation)) 31,000. To examine
expression levels of FKBP52 and ARA9 in yeast, we performed Western
blot analysis on cell extracts as described previously (19).
3
H-FK506 Binding Assay—The ARA9 and FKBP52 proteins were
purified from 4
m
g of BL21(DE3) bacterial extract by incubation in the
presence of 20
m
l of T7-agarose in bind/wash buffer for1hat4°C.
Samples were washed twice in bind/wash buffer, followed by one wash
in 10 mMTris-HCl, pH 7.5. The resulting protein-agarose pellet was
resuspended in 50
m
lof10mMTris-HCl in the presence of
3
H-FK506 (84
Ci/mmol, NEN Life Science Products Inc.) at 4 °C for 3 h. The samples
were washed three times in ice-cold 10 mMTris-HCl, pH 7.5, and the
radioactivity quantified by liquid scintillation counting. To demonstrate
specificity, the incubation was also performed in the presence and
absence of unlabeled FK506, rapamycin, or cyclosporin-A as competitor.
In Situ Hybridization of Mouse Embryonic Tissue—The 1.1-kilobase
EcoRI cDNA fragment of PL613 was subcloned into pbSK in the T7
orientation for the generation of sense (NotI digested) and antisense
(EcoRI digested) riboprobes. Both sense and antisense riboprobes were
generated with 80
m
Ci of [
35
S]UTP (Amersham, .1000 Ci/mmol) as the
radioactive ribonucleotide and subjected to alkaline hydrolysis for 9
min at 60 °C as described (20). Embryos (sagittal sections) or tissues
from adult mice (cross-sections) were cut at 5
m
m and mounted on
“Superfrost-plus”slides (Fisher, Pittsburgh, PA). The processing, hy-
bridization, washing conditions, development, and photography of the
slides were performed as described previously (21). Hoechst dye (Boeh-
ringer Manheim) was used as the counterstain.
Chromosomal Localization—Chromosomal localization was per-
formed by a commercial service (SeeDNA Biotech Inc., North York,
Ontario, Canada) using the 1.1-kilobase EcoRI fragment from PL613 as
a probe. Briefly, lymphocytes isolated from human blood were cultured
in
a
-minimal essential medium supplemented with 10% fetal calf se-
rum and phytohemagglutinin at 37 °C for 72 h. The lymphocyte cul-
tures were treated with 5-bromodeoxyuridine (0.18 mg/ml) to synchro-
nize the cell population. The synchronized cells were washed three
times with serum-free medium to release the block and recultured at
37°Cfor6hin
a
-minimal essential media with thymidine (2.5
m
g/ml).
Cells were harvested and slides were made using a hypotonic treatment
procedure, then fixed and air-dried. The probe was biotinylated with
dATP using nick labeling. The procedure for FISH detection was per-
formed as described previously (22, 23). The chromosomal assignment
was confirmed using a diagnostic polymerase chain reaction strategy
and a panel of DNA samples obtained form human-hamster somatic cell
hybrids (Bios, New Haven, CT). Using oligonucleotides OL814 and
OL869 a ;600-base pair fragment of the human ARA9 gene was am-
plified from 200 ng of hybrid DNA. To determine the presence of the
human gene in the hamster DNA background, each polymerase chain
reaction was then analyzed by electrophoresis on a 1% agarose gel and
visualized with ethidium bromide for detection of the diagnostic band
(24).
RESULTS
AHR Domains Required for ARA9 Interaction—In order to
map the region of the AHR required for interaction with ARA9,
ARA9 Characterization 33581

co-immunoprecipitation assays were performed using proteins
that had been produced by in vitro translation in reticulocyte
lysates.
35
S-Labeled ARA9 and a series of deletions of the AHR
were mixed and immunoprecipitated with antibodies specific to
the basic region (G1295) or the PAS domains (BEAR-1) of the
AHR (Fig. 1). We observed that an NH
2
-terminal deletion of
130 amino acids (AHRND130) but not 287 residues
(AHRND287) was able to form complexes with ARA9. COOH-
terminal deletions of the AHR of up to 313 amino acids
(AHRCD313) also associated with ARA9, while COOH-termi-
nal deletions of 425 amino acids (AHRC 425) or greater did not
associate with the AHR. This defines the approximate bound-
aries of the AHR domain required for ARA9 interaction as lying
between amino acids 130 and 491. These approximate bound-
aries are supported by the observation that an AHR construct
containing an internal deletion in this region, AHRD290–491,
does not co-immunoprecipitate ARA9 in parallel experiments.
ARA9 Domains Required for Interaction with AHR and
Hsp90—To map the region of ARA9 that is required for inter-
action with the AHR, two ARA9 mutants that had been T7-
tagged were expressed in bacteria and affinity purified. The
ARA9DC mutant contains the NH
2
-terminal amino acids 1–174
corresponding to the FKBP12 homology domain. The ARA9DN
mutant contains the carboxyl-terminal portion of ARA9 (amino
acids 154–330) and includes the three TPR domains (Fig. 2 (7)).
These proteins were tested for their ability to interact with
35
S-labeled AHR and Hsp90 in co-immunoprecipitation assays
using anti-T7 tag antibodies. In these assays, the COOH-ter-
minal TPR domains of ARA9 were capable of precipitating both
Hsp90 and AHR, while the NH
2
-terminal FKBP domain
showed no activity above background (Fig. 2, Aand B). SDS-
PAGE analysis indicated that the two mutant ARA9 proteins
were expressed to equivalent degrees, arguing against differ-
ences in protein expression as an explanation of the differential
interaction (Fig. 2C).
Specificity of ARA9 for AHR—The structural similarity be-
tween ARA9 and FKBP52 led us to compare the relative ca-
pacity of these two factors to bind AHR. To determine their
relative binding, T7-tagged ARA9 or FKBP52 (expressed in
bacteria) were incubated in the presence of
35
S-labeled AHR
(expressed in reticulocyte lysates). The AHR-ARA9 or AHR-
FKBP52 associations were determined by co-immunoprecipita-
tion with T7-agarose. In this system, ARA9 was approximately
four times more efficient at precipitating AHR than was
FKBP52 (Fig. 3A). To demonstrate that these results were not
due to an inability of FKBP52 to function in this system, or to
a lower expression level, we performed a positive control and
measured the capacity of ARA9 and FKBP52 to associate with
Hsp90. We observed that under the same conditions as those
used to assay AHR interactions, FKBP52 exhibited a 4-fold
greater capacity to interact with
35
S-labeled Hsp90 than did
ARA9 (Fig. 3B).
In a second experiment, we analyzed the relative capacities
of AHR and GR to form higher order complexes that included
ARA9 and Hsp90. Our previous work using anti-Hsp90 anti-
bodies demonstrated that ARA9 has a much higher affinity for
AHR-Hsp90 complexes than for Hsp90 alone. Therefore, we
performed co-immunoprecipitation assays with anti-Hsp90 an-
tibodies to determine if ARA9 also had a higher affinity for
GR-Hsp90 complexes. In contrast to AHR, GR was incapable of
increasing the ARA9-Hsp90 interaction (Fig. 4).
FK506 Binding—Since the NH
2
-terminal portion of ARA9
contains an “FKBP12 homology domain,” it was of interest to
determine if ARA9 was also capable of binding the prototype
ligand, FK506. To this end, bacterial extracts containing either
T7-tagged ARA9 or T7-tagged FKBP52 were assayed for their
capacity to bind
3
H-FK506. The extract containing FKBP52
exhibited significant binding to
3
H-FK506 (Fig. 5). Scatchard
analysis revealed that the affinity of this binding displayed a
K
D
of approximately 5 310
28
M(data not shown). These
observations are in agreement with that previously reported
for FKBP52 (25, 26). Structure-activity analysis indicated that
this interaction was specific since FK506 binding to FKBP52
could be displaced by co-incubation with unlabeled FK506 or
rapamycin. In addition, cyclosporin-A, a macrolide specific for
the cyclophilin class of immunophilins, did not compete for
these FK506-binding sites (Fig. 5). In contrast to FKBP52, we
could not detect
3
H-FK506 binding to ARA9 at several concen-
trations of
3
H-FK506 (Fig. 5 and data not shown). In an effort
to increase sensitivity of the FK506 binding assay, the experi-
ment was repeated using 2
m
g of purified ARA9. Using this
purified preparation we were still unable to detect binding of
3
H-FK506 to ARA9 (Fig. 5).
ARA9 Enhances Ligand Responsiveness of AHR—Given the
sequence similarities between ARA9 and FKBP52, we asked if
either protein could influence AHR signaling in vivo. To this
end, we employed a LexA-AHR chimera and a LexA-operator-
driven
b
-galactosidase reporter gene in the yeast S. cerevisiae
(14).
b
-Napthoflavone dose-response curves were constructed
for the LexA-AHR in the presence and absence of ARA9 (Fig. 6).
Co-expression of ARA9 significantly enhanced the response of
AHR to
b
-NF, decreasing the EC
50
approximately 5-fold when
ARA9 was present (Fig. 6). In addition, the maximal response
to
b
-NF was increased approximately 2.5-fold in the presence of
ARA9 (Fig. 6). In parallel experiments a similar shift in AHR
dose-response could not be elicited by FKBP52, nor could either
factor alter the response of a GR-LexA chimera to the GR
ligand, deoxycorticosterone (Fig. 6). To rule out the possibility
that these results were due to lack of expression of either
protein in our yeast model system, yeast extracts were sub-
jected to Western blot analysis and a high level of both ARA9
FIG.1.Co-immunoprecipitation of ARA9 with AHR deletions.
Left, full-length AHR and its deletions are depicted on the left. These
proteins were combined with full-length ARA9 that had been expressed
in the presence of [
35
S]methionine. All proteins were expressed in
reticulocyte lysates. Middle, IP ARA9, a qualitative interpretation of
the immunoprecipitation is provided as plus (efficient immunoprecipi-
tation of ARA9) or minus (no immunoprecipitation). Right, protein
complexes were precipitated with the AHR specific antibodies BEAR-1
(top panel) or G1295 (bottom panel) that had been coupled to Protein
A-Sepharose. Preliminary experiments demonstrated that the corre-
sponding deletions were equally immunoprecipitated with these anti-
bodies (data not shown). Proteins were separated on 10% SDS-PAGE,
and the co-immunoprecipitated ARA9 was visualized by autoradiogra-
phy. “Input” is a loading control representing 100% of the radiolabeled
ARA9 used in the assay. “PI” is the preimmune control. The ARA9-
binding domain deduced from these experiments is described by the
bold line.
ARA9 Characterization33582

and FKBP52 were detected (Fig. 6, inset).
In Situ Hybridization—High levels of ARA9 mRNA were
detected in murine embryonic development as early as E9.5
days (Fig. 7). At E9.5 days, ARA9 mRNA is ubiquitously ex-
pressed with higher levels observed in the neuroepithelium,
trigeminal ganglion, branchial arches 1 and 2, hepatic primor-
dia, and the primitive gut. In the E13.5 day embryo, ARA9
mRNA remained widespread, with highest levels observed in
the derivatives of the branchial arches; thymus, lung, liver,
intestines, urogenital sinus, and genital tubercle. Examination
of adult lymphoid tissues also revealed high levels of ARA9
mRNA in the thymic cortex and the splenic white pulp.
Chromosomal Localization of ARA9—To determine the chro-
mosomal localization of the human gene encoding ARA9, 49,6-
diamidino-2-phenylindole banding was used to identify the spe-
cific chromosome and the detailed position was determined
based on the summary of 10 photos. These data localized ARA9
to chromosome 11, region q13.3. Another gene that maps to this
region encodes FKBP13 (11q13.1–13.3) (27). Since, the FKBP13
gene is localized to the same region of chromosome 11 (see
“Discussion” below) as the ARA9 gene and since a full-length
ARA9 cDNA probe was used in the FISH analysis we confirmed
the chromosome 11 localization of ARA9 by an independent
polymerase chain reaction protocol. Using polymerase chain
reaction probes directed to the 39end of the ARA9 cDNA
(outside of the region of sequence homology between ARA9 and
FKBP13) we analyzed a somatic cell hybrid panel of genomic
DNA derived from human-hamster fusions (Bios). This screen
confirmed ARA9’s localization to chromosome 11 (data not
shown).
DISCUSSION
The AHR and GR signaling pathways display a number of
mechanistic similarities (1, 28). In the absence of agonist, both
the AHR and GR are primarily found in a cytosolic complex
with a dimer of Hsp90 and a number of smaller proteins (29–
32). Upon binding their respective ligands, the AHR and GR
complexes undergo changes in their oligomeric state, as well as
an increased affinity for the nuclear compartment of the cell
(32–34). While the GR homodimerizes in the nucleus, the AHR
heterodimerizes with its cognate partner, ARNT. The resulting
dimers bind their respective enhancer elements and activate
transcription of target genes (1, 35). These similarities have
prompted us to compare the signal transduction and oligomeric
composition of the AHR and GR in the hopes that we can
understand aspects of these pathways that are fundamental to
both basic helix loop helix-PAS proteins and members of the
FIG.4. Effects of AHR and GR on ARA9-Hsp90 interaction.
Full-length ARA9, AHR, and GR were translated in reticulocyte lysates
in the presence of [
35
S]methionine. The proteins were incubated in the
presence of Hsp90 antibody and precipitated with Protein A-Sepharose.
The proteins were separated by SDS-PAGE gels and visualized by
autoradiography.
FIG.2.Characterization of ARA9-AHR interaction and PAS specificity. A, radiolabeled AHR was incubated in the presence of T7-tagged
proteins that had been expressed in bacteria. FL is the full-length ARA9, DNis the NH
2
-terminal truncation of ARA9, DCis the COOH-terminal
truncation of ARA9. Ctrl is bacteria lysate from untransformed BL21 cells. All complexes were precipitated with T7-tagged agarose. Proteins were
separated on 7.5% gel and visualized by autoradiography. B, radiolabeled Hsp90 was used in a similar assay described in A.C,4
m
g of total extracts
used in Awere separated on a 10% gel and transferred to nitrocellulose. Western blot analysis was preformed using a T7 antibody conjugated to
alkaline phosphatase (Novagen). D, diagram of ARA9 in which the FKBP12 homology domain is labeled FKBP and the domain implicated in
binding AHR and Hsp90 is noted by the solid line.
FIG.3.ARA9 and FKBP52 co-immunoprecipitation of AHR and
Hsp90. A, top: AHR was translated in reticulocyte lysates in the pres-
ence of [
35
S]methionine. Labeled proteins were then incubated in the
presence of T7-tagged ARA9, T7-tagged FKBP52, or untransformed
BL21 extract and complexes were separated and visualized as above.
Input lanes represent 25% of total protein (AHR or Hsp90) used in
experiments.Bottom, the bands radioactivity present in the AHR or
Hsp90 bands was quantified using a PhosphorImager to make compar-
isons of relative coimmunoprecipitation. The bars represent the means
from three independent experiments. The error bars indicate the stand-
ard error of the mean. B, Hsp90 was used in a similar experiment
described in A.
ARA9 Characterization 33583

steroid/thyroxin receptor superfamily.
Biochemical studies demonstrate that Hsp90 association is
correlated with each receptor’s capacity to bind ligand, as well
as their repressed DNA binding state (36–40). Data from a
yeast model system support these observations and has been
used to demonstrate that Hsp90 is essential for agonist-in-
duced signal transduction in vivo (14, 41, 42). In addition to
Hsp90, the GR-Hsp90 complex has also been shown to contain
a molecule of FKBP52, Cyp40, or PP5 (32, 43–46). FKBP52 and
Cyp40 are often referred to as immunophilins because they
bind the immunosuppressant macrolides, FK506 or cyclo-
sporin-A, respectively (47). In L-cell cytosol, FKBP52- and PP5-
containing GR complexes comprise 52 and 35% of the total GR
complexes, while CyP40 is found in only a small percentage of
GR complexes (46, 48).
Recently, it has been observed that the AHR-Hsp90 complex
contains a 37-kDa protein, ARA9, which displays significant
amino acid sequence similarity to FKBP52 (7–9). This similar-
ity prompted us to perform a detailed analysis of the AHR-
ARA9 interaction. In our first series of experiments, we at-
tempted to define the protein domains required to form the
ARA9-Hsp90-AHR complex. To accomplish this, we performed
deletion analyses on both ARA9 and AHR and measured the
capacity of these mutants to form higher order complexes in
vitro. Reticulocyte lysates formed the common milieu for these
experiments, since they have been shown to properly fold AHR
and are known to contain high levels of Hsp90. In this system,
we observed that the TPR domains of ARA9 were both neces-
sary and sufficient to form complexes that contained both AHR
and Hsp90 (Fig. 2, Aand B). Conversely, the NH
2
-terminal half
of ARA9 did not interact with either protein. In the reciprocal
domain mapping experiments, we observed that ARA9 inter-
acts within the AHR’s repressor domain (Fig. 1). This region of
the AHR has previously been shown to repress receptor activity
and harbor the domains required for both Hsp90 and ligand
binding (5, 49–51).
The domain mapping studies suggest a tertiary complex
where ARA9 interacts with Hsp90 through its TPR domains
and Hsp90 interacts with AHR through its repressor domain.
Given the influence of AHR on ARA9 association (Fig. 4 (7)),
the linear model predicts that AHR directly alters the structure
or activity of Hsp90, which in turn allows ARA9 binding. Al-
FIG.5.
3
H-FK506 binding.
3
H-FK506
binding assays were performed as de-
scribed under “Materials and Methods” in
the presence of 5.0 310
28
M
3
H-FK506.
The binding was done in the presence of
100 3molar cold competitor (FK506, cy-
closporin-A, or rapamycin). DMSO is the
vehicle control. Assays were done with
extracts from BL21 bacteria expressing
no exogenous protein BL21 ctrl (M),
FKBP52 (p),or ARA9 (f). Additionally
purified ARA9 protein was also used, pure
ARA9 (o).
FIG.6.ARA9 influences AHR signaling in yeast. The yeast strain L40 expressing LexA fusion proteins of the AHR (left)orGR(right) were
transformed with either ARA9 or FKBP52. The yeast transformations were replica plated onto various concentrations of the corresponding ligands
(
b
-NF for the AHR and deoxycorticosterone (DOC), for the GR). The yeast were allowed to grow for 2 days and the
b
-galactosidase activity was
measured by the o-nitrophenyl-B-D-galactoside assay. Values are the average of triplicate determinations (mean 6S.E.). These results are
representative of at least two independent experiments. Inset, Western blot analyses of ARA9 and FKBP52 expression in yeast. Lane 1, FKBP52
protein expression in L40 yeast (commercial antibody from Affinity Bioreagents Inc.). Lane 2, ARA9 in L40 yeast. Lane 3, untransformed L40 yeast
(1066 s, an antibody generated from a amino-terminal peptide of ARA9 (64)). Below, schematics of the AHR and GR constructs. Lex is the LexA
DNA-binding domain. LBD is the ligand-binding domain. TAD is the transcriptionally active domain.
ARA9 Characterization33584

though this “linear” interaction model has considerable exper-
imental support, it is premature to rule out a more complex
model where all proteins are involved in bipartite interactions,
i.e. each protein contacts each other protein. For such a model
to be correct, the AHR must directly contact ARA9 or additional
bridging proteins must exist (52). Although this has not yet
been shown, it may be that direct ARA9-AHR interaction does
occur in vivo and simply cannot be detected by the analytical
approaches currently in use. If such an interaction does exist, it
will be difficult to demonstrate since Hsp90 is required for
proper folding of AHR. If the bipartite model was correct, then
AHR could have an effect on ARA9 association function
through direct contact rather than by influencing the structure
of Hsp90.
Although ARA9 and FKBP52 share significant sequence ho-
mology and form similar oligomeric complexes, our data sug-
gests that they also have distinct cellular roles and unique
biochemical properties. A number of observations support the
idea that these proteins harbor interaction specificity for their
cognate receptors: 1) ARA9 has a greater affinity than FKBP52
for the AHR-Hsp90 complex; 2) AHR enhances the stability of
ARA9-Hsp90 interactions, but does not enhance Hsp90-
FKBP52 interactions; 3) GR does not enhance ARA9-Hsp90
interactions; 4) ARA9 expression significantly augments AHR
signaling in yeast, yet has no influence on GR signaling. An
additional distinguishing feature of ARA9 is in its capacity to
bind macrolides such as FK506. The conclusion that ARA9 does
not bind FK506 was based upon our inability to detect
3
H-
FK506 binding to ARA9, while binding to FKBP52 was easily
measured. Although it is difficult to prove a negative result, the
lack of FK506 binding to ARA9 was anticipated based upon the
observation that of the 14 residues in FKBP12 thought to
contact FK506, only 5 are conserved in ARA9 (7). Taken in
sum, these results clearly document unique biochemical prop-
erties of ARA9 and FKBP52, as well as suggest that each
protein plays an important role in signal transduction by dis-
tinct classes of ligand activated transcription factors.
The mechanism by which immunophillins influence GR sig-
naling is still somewhat unclear. Other than physical associa-
tion, the evidence for a functional role of these proteins is based
upon a number of experimental lines of evidence. First, the
importance of FKBP52 and Cyp40 in GR signal transduction is
suggested by experiments where FK506 or cyclosporin-A have
been shown to augment GR signal transduction in mammalian
cells and yeast (53–56). Second,yeast strains devoid of the
CYP40 homologue CPR7 exhibit deficient glucocorticoid recep-
tor signaling (57). Interestingly, recent evidence has indicated
that the CPR7 dependent activity of GR can be rescued by a
FIG.7.In situ hybridization of ARA9 in E9.5, E13.5 mice, and adult thymus and spleen. In situ hybridization was performed using a
full-length ARA9 probe as described under “Materials and Methods.”
ARA9 Characterization 33585

mutant cDNA that harbors only the TPR domains and is devoid
of any cyclosporin-A binding or peptidylprolyl cis-trans isomer-
ase activity (58). Although the idea that peptidylprolyl cis-trans
isomerase activity is unrelated to GR signaling is not new, the
dominance of the TPR domain in this rescue experiment sug-
gests that immunophillins and their corresponding immuno-
suppressants may be acting by multiple mechanisms.
In a similar manner, we have demonstrated that expression
of ARA9 has a functional consequence on AHR signal trans-
duction. Co-expression of ARA9 in a yeast model system sig-
nificantly increases both the sensitivity and the maximal re-
sponse of AHR to agonist (Fig. 6). Although the molecular basis
for this effect in yeast has yet to be determined, it is supported
by recent experiments in mammalian cell culture (8, 9). Inter-
estingly, the effect of ARA9 on the
b
-NF dose–response curve of
AHR is what would be predicted under conditions of increased
receptor number (59). Put in its broadest context, our data are
consistent with ARA9 increasing the number of receptors that
bind agonist at a given dose or the number of receptors that can
progress through a rate-limiting step in signal transduction.
Although Western blots to approximate AHR expression in
yeast extracts do not indicate any influence of ARA9 on total
AHR protein levels (data not shown); these results should not
be over-interpreted since we do not yet know what percentage
of AHR is properly folded under each condition. In this regard,
it is important to note that the data from the GR system
suggests that FKBP52 influences GR trafficking to the nucleus
and does not necessarily increase total receptor number (60). A
similar mechanism may be at play for ARA9-AHR.
It is tempting to speculate that the physical interaction be-
tween the AHR and ARA9 is mechanistically related to their
functional interaction at the level of enhanced AHR signal
transduction (Fig. 6) (8, 9). In an effort to present a more
balanced discussion of this point, it is important to consider a
series of experiments from the GR field. Recent evidence has
suggested that FK506 may influence GR signaling in yeast and
mammalian cells by inhibiting a membrane transporter with
specificity to pump certain glucocorticoids out of the cell (61,
62). These results suggest that FK506’s action may not be
mediated by direct interaction with the FKBP52-GR complex,
but rather may be mediated through interaction with the yeast
transporter, Pdr5p, or with a yeast FKBP that influences this
transporters activity. By extension, one could argue that
ARA9’s affect on AHR signal transduction may be mediated
indirectly through inhibition of a drug-transporter like Pdr5p.
Although we have not yet formally ruled out the possibility that
ARA9 inhibits “
b
-NF transporters,” the pharmacological profile
argues against this. If ARA9 expression was inhibiting a trans-
porter and increasing the intracellular concentration of
b
-NF,
we would predict a “left-shift” in the dose-response curve, but
not an increase in the maximal response (i.e. an increase in
apparent potency, but not in apparent efficacy of the agonist).
Additionally, since we did not use FK506 in our experiments,
competition for the pump’s activity by the macrolide can be
ruled out.
Finally, the developmental expression of ARA9 also has a
number of important implications. A comparison of the devel-
opmental profiles of AHR and ARA9 suggests that ARA9’s
function is probably not restricted to AHR signal transduction
(compare Fig. 7 to results in Ref. 63). For example, ARA9 is
highly expressed at E9.5, before high levels of AHR are de-
tected, and ARA9 is more widely expressed than the AHR at
later time points like E13.5 and into adulthood (63). The ob-
servation that AHR and ARA9 expression are coincident at a
number of sites (e.g. thymus, liver, and lung) supports the
biological relevance of the interaction. Moreover, if ARA9 aug-
ments AHR signaling, its level of expression may play an
important role in the tissue specific sensitivity of certain tis-
sues like thymus.
REFERENCES
1. Schmidt, J. V., and Bradfield, C. A. (1996) Annu. Rev. Cell Dev. Biol. 12, 55–89
2. Fukunaga, B. N., Probst, M. R., Reisz-Porszasz, S., and Hankinson, O. (1995)
J. Biol. Chem. 270, 29270–29278
3. Durrin, L. K., and Whitlock, J. P., Jr. (1987) Mol. Cell. Biol. 7, 3008–3011
4. Murre, C., McCaw, P. S., and Baltimore, D. (1989) Cell 56, 777–783
5. Dolwick, K. M., Swanson, H. I., and Bradfield, C. A. (1993) Proc. Natl. Acad.
Sci. U. S. A. 90, 8566 – 8570
6. Whitelaw, M. L., Gottlicher, M., Gustafsson, J. A., and Poellinger, L. (1993)
EMBO J. 12, 4169–4179
7. Carver, L. A., and Bradfield, C. A. (1997) J. Biol. Chem. 272, 11452–11456
8. Ma, Q., and Whitlock, J. P., Jr. (1997) J. Biol. Chem. 272, 8878–8884
9. Meyer, B. K., Pray-Grant, M. G., Vanden Heuvel, J. P., and Perdew, G. H.
(1998) Mol. Cell. Biol. 18, 978–988
10. Vojtek, A. B., Hollenberg, S. M., and Cooper, J. A. (1993) Cell 74, 205–214
11. Bartel, P. L., Chien, C., Sternglanz, R., and Fields, S. (1993) in Cellular
interactions in Development: A Practical Approach (Hartley, D. A., ed) pp.
153–179, IRL Press, Oxford
12. Brunelli, J. P., and Pall, M. L. (1993) Yeast 9, 1299–1308
13. Chien, C., Bartel, P. L., Sternglanz, R., and Fields, S. (1991) Proc. Natl. Acad.
Sci. U. S. A. 88, 9578 –9582
14. Carver, L. A., Jackiw, V., and Bradfield, C. A. (1994) J. Biol. Chem. 269,
30109–30112
15. Godowski, P. J., Picard, D., and Yamamoto, K. R. (1988) Science 241, 812–816
16. Poland, A., Palen, D., and Glover, E. (1994) Mol. Pharmacol. 46, 915–921
17. Lebeau, M. C., Massol, N., Herrick, J., Faber, L. E., Renoir, J. M., Radanyi, C.,
and Baulieu, E. E. (1992) J. Biol. Chem. 267, 4281–4284
18. Schmidt, J. V., Su, G. H.-T., Reddy, J. K., Simon, M. C., and Bradfield, C. A.
(1996) Proc. Natl. Acad. Sci. U. S. A. 93, 6731–6736
19. Hogenesch, J. B., Chan, W. C., Jackiw, V. H., Brown, R. C., Gu, Y. Z.,
Pray-Grant, M., Perdew, G. H., and Bradfield, C. A. (1997) J. Biol. Chem.
272, 8581–8593
20. Wilkinson, D. G., and Nieto, M. A. (1993) Methods Enzymol. 225, 361–373
21. Jain, S., Maltepe, E., Lu, M. M., Simon, C., and Bradfield, C. A. (1998) Mech.
Dev. 73, 117–123
22. Heng, H. H., Squire, J., and Tsui, L. C. (1992) Proc. Natl. Acad. Sci. U. S. A. 89,
9509–9513
23. Heng, H. H., and Tsui, L. C. (1993) Chromosoma 102, 325–332
24. Le Beau, M. M., Carver, L. A., Espinosa, R., Schmidt, J. V., and Bradfield, C. A.
(1993) Cytogenet. Cell Genet. 66, 172–176
25. Peattie, D. A., Harding, M. W., Fleming, M. A., DeCenzo, M. T., Lippke, J. A.,
Livingston, D. J., and Benasutti, M. (1992) Proc. Natl. Acad. Sci. U. S. A.
89, 10974–10978
26. Tai, P. K., Albers, M. W., Chang, H., Faber, L. E., and Schreiber, S. L. (1992)
Science 256, 1315–1318
27. DiLella, A. G., Hawkins, A., Craig, R. J., Schreiber, S. L., and Griffin, C. A.
(1992) Biochem. Biophys. Res. Commun. 189, 819–823
28. Evans, R. M. (1988) Science 240, 889–895
29. Prokipcak, R. D., Faber, L. E., and Okey, A. B. (1989) Arch. Biochem. Biophys.
274, 648–658
30. Pollenz, R. S., Sattler, C. A., and Poland, A. (1994) Mol. Pharmacol. 45,
428–438
31. Perdew, G. H. (1992) Biochem. Biophys. Res. Commun. 182, 55–62
32. Pratt, W. B. (1993) J. Biol. Chem. 268, 21455–21458
33. Henry, E. C., and Gasiewicz, T. A. (1993) Biochemistry 294, 95–101
34. Hankinson, O. (1995) Annu. Rev. Pharmacol. Toxicol. 35, 307–40
35. Yamamoto, K. R. (1985) Annu. Rev. Genet. 19, 209–252
36. Perdew, G. H. (1988) J. Biol. Chem. 263, 13802–13805
37. Denis, M., Cuthill, S., Wikstrom, A. C., Poellinger, L., and Gustafsson, J. A.
(1988) Biochem. Biophys. Res. Commun. 155, 801–807
38. Howard, K. J., and Distelhorst, C. W. (1988) J. Biol. Chem. 263, 3474–3481
39. Sanchez, E. R., Toft, D. O., Schlesinger, M. J., and Pratt, W. B. (1985) J. Biol.
Chem. 260, 12398–12401
40. Pongratz, I., Mason, G. G., and Poellinger, L. (1992) J. Biol. Chem. 267,
13728–13734
41. Picard, D., Khursheed, B., Garabedian, M. J., Fortin, M. G., Lindquist, S., and
Yamamoto, K. R. (1990) Nature 348, 166–168
42. Whitelaw, M. L., McGuire, J., Picard, D., Gustafsson, J. A., and Poellinger, L.
(1995) Proc. Natl. Acad. Sci. U. S. A. 92, 4437–4441
43. Chen, M. S., Silverstein, A. M., Pratt, W. B., and Chinkers, M. (1996) J. Biol.
Chem. 271, 32315–32320
44. Johnson, J. L., and Toft, D. O. (1995) Mol. Endocrinol. 9, 670–678
45. Smith, D. F., Sullivan, W. P., Marion, T. N., Zaitsu, K., Madden, B.,
McCormick, D. J., and Toft, D. O. (1993) Mol. Cell. Biol. 13, 869–876
46. Silverstein, A. M., Galigniana, M. D., Chen, M. S., Owens-Grillo, J. K.,
Chinkers, M., and Pratt, W. B. (1997) J. Biol. Chem. 272, 16224–16230
47. Kay, J. E. (1996) Biochem. J. 314, 361–385
48. Owens-Grillo, J. K., Hoffmann, K., Hutchison, K. A., Yem, A. W., Deibel, M. R.,
Jr., Handschumacher, R. E., and Pratt, W. B. (1995) J. Biol. Chem. 270,
20479–20484
49. Perdew, G. H., and Bradfield, C. A. (1996) Biochem. Mol. Biol. Int. 39, 589–593
50. Dolwick, K. M., Schmidt, J. V., Carver, L. A., Swanson, H. I., and Bradfield,
C. A. (1993) Mol. Pharmacol. 44, 911–917
ARA9 Characterization33586

51. Coumailleau, P., Poellinger, L., Gustafsson, J. A., and Whitelaw, M. L. (1995)
J. Biol. Chem. 270, 25291–25300
52. Silverstein, A. M., Grammatikakis, N., Cochran, B. H., Chinkers, M., and
Pratt, W. B. (1998) J. Biol. Chem. 273, 20090–20095
53. Ning, Y. M., and Sanchez, E. R. (1993) J. Biol. Chem. 268, 6073–6076
54. Ratajczak, T., Mark, P. J., Martin, R. L., and Minchin, R. F. (1996) Biochem.
Biophys. Res. Commun. 220, 208–212
55. Renoir, J. M., Le Bihan, S., Mercier-Bodard, C., Gold, A., Arjomandi, M.,
Radanyi, C., and Baulieu, E. E. (1994) J. Steroid Biochem. Mol. Biol. 48,
101–110
56. Tai, P. K., Albers, M. W., McDonnell, D. P., Chang, H., Schreiber, S. L., and
Faber, L. E. (1994) Biochemistry 33, 10666–10671
57. Duina, A. A., Chang, H. C., Marsh, J. A., Lindquist, S., and Gaber, R. F. (1996)
Science 274, 1713–1715
58. Duina, A. A., Marsh, J. A., Kurtz, R. B., Chang, H. C., Lindquist, S., and Gaber,
R. F. (1998) J. Biol. Chem. 273, 10819–10822
59. Nickerson, M. (1956) Nature 178, 697
60. Czar, M. J., Lyons, R. H., Welsh, M. J., Renoir, J. M., and Pratt, W. B. (1995)
Mol. Endocrinol. 9, 1549–1560
61. Kralli, A., Bohen, S. P., and Yamamoto, K. R. (1995) Proc. Natl. Acad. Sci.
U. S. A. 92, 4701– 4705
62. Kralli, A., and Yamamoto, K. R. (1996) J. Biol. Chem. 271, 17152–17156
63. Jain, S., and Bradfield, C. A. (1998) Mech. Dev. 73, 117–123
64. Poland, A., Glover, E., and Bradfield, C. A. (1991) Mol. Pharmacol. 39, 20–26
ARA9 Characterization 33587