Professional Documents
Culture Documents
for
and Experimental
Therapeutics
ACCELERA
TED COMMUNICATION
Role of the Aryl Hydrocarbon Receptor Nuclear Translocator Protein in Aryl Hydrocarbon (Dioxin) Receptor Action
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
MARKUS R. PROBST,1 SUZANNE REISZ-PORSZASZ,1 ROSALLY V. AGBUNAG,
R.V.A., M.S.O., OH.) California 90024
MAY
S. ONG,
and OLIVER
HANKINSON
and Laboratory
of Pathology
June 21 , 1993
SUMMARY
Immunoprecipitation experiments
tracts
confirm receptor
performed on cytosolic excell line Hepa-1 ci c7 (Hepa-1) cytosolic aryl hydrocarbon (Ah)
90-kDa heat shock protein and
(XRE5).
Treatment
of cytosolic
extracts
of Hepa-1
cells with
the Ah receptor protein but reveal that it does not contain the Ah receptor nuclear translocator (ARNT) protein. These experiments confirm that the 6-S liganded form of the receptor identifled in nuclear extracts of cells treated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) contains the Ah receptor protein and ARNT but not the 90-kDa heat shock protein. The 6-S liganded Ah receptor complex activates transcription of the CYP1A 1 gene via its binding to upstream xenobiotic-responsive elements
TCDD in vitro transforms the Ah receptor complex to the XREbinding state. No such transformation occurs in a C mutant deficient in ARNT activity. When in vitro synthesized ARNT was added concomitantly with TCDD to C cytosolic extracts, it associated with the Ah receptor and restored Ah receptordependent XRE-binding activity to the extracts. Covalent crosslinking experiments in nuclear extracts of Hepa-1 and human LS18O cells treated with TCDD in vivo demonstrate that both ARNT and the Ah receptor bind directly to.the XRE core sequence.
The Ah receptor binds a variety of environmentally important carcinogens, including polycyclic aromatic hydrocarbons, heterocyclic amines, and halogenated aromatic hydrocarbons such as TCDD and polychlorinated biphenyls (1) (reviewed in Ref. 2). The liganded receptor complex activates transcription of the CYP1A1 and CYP1A2 genes and several other genes involved in xenobiotic metabolism (reviewed in Ref. 3). The CYP1A1 and CYP1A2 proteins play important roles in the metabolic activation of polycyclic aromatic hydrocarbons and heterocyclic amines, respectively, to their ultimate carcinogenic
metabolites
This work
mediates
of the lism
most,
halogenated
if not
compounds
all, of the
does not
carcinogenic
appear
and
although
toxic
effects
metabo(6).
aromatic
hydrocarbons,
of these
to be involved
The location of the unliganded Ah receptor complex has not been fully resolved (7), although the weight of evidence suggests that this moiety is cytoplasmic (8, 9) and it is found in the cytosolic fraction after conventional subcellular fractionation. The composition of the unliganded receptor complex has not been fully determined. The complex sediments at about 9 S in
sucrose the teins Ah strains) (15). gradients receptor (11, 12), After and (90 TCDD HSP9O is about kDa in (13, treatment, 280 Hepa-1 14), and the kDa in size cells Ah and receptor perhaps (10). two It contains mouse other is found proin certain
(reviewed
was supported
in Refs.
by
4 and
5).
The
Institute
Ah
receptor
Grant
also
CA28868,
National
Cancer
Department of Energy Contract DE-FCO3-87ER 60615, and National Cancer Institute Core Grant CA16042 to the UCLA Jonsson Comprehensive Cancer Center. M.R.P. was supported by a fellowship from the Swiss National Research Foundation and the Basler Freiwillige Akademische Gesellschaft and Fellowship
J920910A 1 The
2
Ah receptor
refers
from the UCLA Jonsson Cancer Center Foundation. contributions of the first two authors should be considered refers to the 90-kDa ligand-binding subunit. to any multimeric protein complex containing
equal. Ah receptor
complex
the Ah receptor.
fraction, as assessed by conventional subcellular Operationally, therefore, transformation of the complex by TCDD results in translocation of the to the nucleus. The nuclear form of the receptor transcription of the CYP1A1 gene via interaction
ABBREVIATIONS: Ah, aryl hydrocarbon; ARNT, aryl hydrocarbon receptor nuclear translocator; OTT, dithiothreitol; EMSA, electrophoretic shift assay(s); GST, glutathione-S-transferase; HSP9O, 90-kDa heat shock protein; PAGE, polyacrylamide gel electrophoresis; TCOO, tetrachlorodibenzo-p-dioxin; XRE, xenobiotic-responsive element; SOS, sodium dodecyl sulfate; EGTA, ethylene glycol bis(3-aminoethyl
mobility
2,3,7,8ether)-
N,N,N,N-tetraacetic
acid; dBrUTP,
bromo-dUTP;
Hepa-1
Hepa-lclc7. 511
512
with
(16,
Probst
et a!.
XRE
17).
sequences
The liganded
located
Ah
5 to the coding
receptor can
region
of the gene
from nuclei
fusion
proteins
be extracted
Sepharose.
antiserum with ugation, ing the
were To adsorb
diluted
immobilized antibodies,
Tris
.
associated with the 87-kDa ARNT protein (18). This Ah receptor-ARNT heterodimer is about 176 kDa in size and sediments at about 6 5 in sucrose gradients (19, 20). Elferink et al. (21) provided evidence that the form of the Ah receptor complex that binds the XRE contains two polypeptides
that bind the XRE core sequence directly. One of these polypeptides was not identified. associated although whether investigate fore fact variety protein question unliganded corresponds that with those ARNT whether This seemed both ARNT identified as the Ah receptor, We showed previously that the Ah receptor during experiments binds the ARNT to the second did not XRE directly binds the polypeptide allow (18). XRE but the ARNT to us to In this directly detected reasons, contain other was is found the XRE, ascertain paper we and including basic therethe helixby Elferink
was
immobilized immobilized
the Sepharose beads and washed consecutively with pH 7.5, and with 10 mM Tris . HC1, pH 7.5, containing
specific ARNT-1.2 antibodies were eluted with 100
10 mM
HC1,
The pH
binding
the mouse 84-kDa and 86-kDa heat shock proteins (27) was a generous gift of Dr. Steven Ulirich, National Institutes of Health (Bethesda, MD). An IgG fraction of the rabbit antiserum was prepared and used without further purification. Cell culture and cellular extracts. The mouse hepatoma cell line Hepa-1, the C- mutant c4, and the human colon adenocarcinoma cell
line LS18O were by the cultured scraping and maintained as described monolayers into (28). Cells were phosharvested of confluent ice-cold
that
recognizes
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
motifs
of transcription dimerization as to Ah
(11,
12,
and
22).
DNA ARNT
These
where binding. is a
motifs
they We component
are
function also
found
in address of the
in a
both the 9-S
factors,
whether receptor
phate-buffered saline. Cytosolic extracts were made from cells cultured in the absence of TCDD. Nuclear extracts were made from cells that had been treated with TCDD (2 nM for Hepa-1, 10 nM for L5180) in culture for 90 mm at 37 before harvesting in ice-cold phosphatebuffered saline. Protease inhibitors (200 zM phenylmethylsulfonyl fluoride, 100 tM leupeptin, and 40 units/mi aprotinin) and 1 mM DTT
were added to all extracts by the cytosolic for and nuclear extraction buffers immediately
complex.
The Hepa-1 cell line is a widely used mechanism of action of the Ah receptor.
1 line with reduced or undetectable
before
were cells
Cytosolic
receptor groups,
in Ref.
function, probably
23).
have rep-
for the
that
EMSA
had
and
been
UV cross-linking
for UV cross-linking experiments et al. (29). Nuclear extracts experiments were prepared from
HEPES, pH 7.5, for 15 mm
10 mM
on ice, centrifuged, resuspended in 1 cell pellet volume of HED (25 mM HEPES, 2 mM EDTA, 1 mM DTT, pH 7.5), and homogenized with a loose-fitting Dounce homogenizer. Cell breakage was monitored under the microscope. After centrifugation (10 mm, 14,000 x g, 4), the
supernatant resuspended (cytosolic in 1 pellet fraction) volume was added was discarded. containing concentration The nuclear 0.4 M KC1 pellet was of HED
and
the
is capable
nucleus after
of binding
binding although
ligand
ligand
but
(24). both
is unable
C of the
to translocate
defective contain
to
in ARNT
function,
and
fitting
was
rocked
at 4 for 30 mm.
glycerol
After
to a final
homogenization x g for
until that immunoblotting the
normal
studies in the
levels
of ARNT
mRNA
(22).
In certain
of the
present
pestle,
of 20% 30 mm further
we utilize a CTCDD-dependent
analysis.
complex
to the XRE-binding
state.
Materials
Chemicals. TCDD Chemical Carcinogen preparation, handling, [SJMethionine and DuPont, respectively. quality available.
and Methods
was obtained from the National Cancer Institute Repository. Safety precautions were taken during and disposal of solutions containing TCDD (17). [32PJdATP were from Amersham and NEN All other chemicals purchased were ofthe highest
antiserum against the mouse
for swelling the cells and supplemented with 20 mM prepared in the absence of g for 120 mm rather than dialyzed overnight against
10% glycerol)
the HEDK nuclear extraction sodium molybdate. Nuclear glycerol and were centrifuged
30 mm. Also, the nuclear
supernatant
mm
tation
at 14,000
and
250 volumes of HEDG (HED containing with 20 mM Na2MoO4 and was then centrifuged for 10 X g at 4. The cytosolic extracts for the immunoprecipiexperiments corresponded to the superna-
immunoblotting
Ah receptor was raised in New Zealand White rabbits using the published aminoterminal peptide sequence (25) coupled to keyhole limpet hemocyanin via an additional cysteine at the amino terminus (Immuno-Dynamics). The antiserum was affinity-purified using the amino-terminal peptide coupled to bovine serum albumin and immobilized on CNBr-activated Sepharose (Pharmacia), according to the manufacturers instructions. The human ARNT cDNA was digested with EcoRI, and the carboxylterminal fragment coding for amino acids 399-777 was isolated and cloned into the GST fusion vector pGEX-1AT (Pharmacia). The resuiting plasmid, pGEX1AT-ARNT-1.2, was then used to transform
polyclonal Escherichw coli JM1O1 to ampicillin resistance. Colonies were isolated
Antisera.
Protein method
Immunoprecipitation
Immunoprecipita-
tion ofcytosolic
of the method
and nuclear
described
in HEDG, supplemented (Sigma), 150 mM NaCl, and 5 mM EGTA, were precleared A-agarose (Boehringer-Mannheim) on a rocking table
Affinity-purified antibodies against Ah receptor
with a modification by Poland et al. (30). In brief, extracts diluted with 20 mM Na2MoO4, 1% Tergitol NP4O
with added
.
extracts
was performed
Protein to the
for 3 hr at 4.
were
and checked for correct orientation of the insert. Large-scale cultures were induced with 1 mM isopropyl f3-D-thiogalactopyranoside at an A value of 0.9 and were grown for an additional 5 hr at 37. The fusion protein was purified from inclusion bodies according to the method of Marston (26) and was used to immunize New Zealand White rabbits.
Immuncomplexes were precipitated with Protein A-agarose for 3 hr at 4, washed five times with the aforementioned buffer, and boiled with SDS-sample buffer (2% SDS, 5% 3-mercaptoethanol, 10% glycerol, 40 mM Tris. HCI, pH 6.8) for 8 mm. Supernatants were subjected to SDS-PAGE using a Miniprotean-Il gel electrophoresis apparatus (Bio-Rad). For
immunoblots, proteins were transferred to nitroceilulose (Hybond-
supernatant,
which
was
then
incubated
overnight
at 4
Bacterial
extracts
of JM1O1
expressing
ECL;
Amersham)
by means
of a semi-dry
blotting
technique.
After
in Ah Receptor prepared absence on half 95 kDa, This (22). from (i.e., of was of immunoblots
513
Hepa-1 (i.e., the a
0.1% serum
and or
Methods. human
preparations,
appropriately diluted in 5% BLOTTO. Antigen-antibody complexes were detected with a goat anti-rabbit antibody coupled to horseradish peroxidase (Pierce Chemicals), using the enhanced chemiluminescence (ECL) detection system (Amersham). Prestained molecular weight markers (Sigma) were used. Because the staining process significantly alters the molecular weight of the marker proteins, so that the nominal molecular weights provided by the manufacturer are incorrect, the molecular weights of the prestained markers were calibrated by cornparing them with a series of nonstained molecular weight markers.
human detected
extracts
size that of 87 detected
(Fig.
kDa
1A, lanes
for human extracts with
in nuclear
of both nuclei)
1A,
The
amount
of the
induced activity protein (data
protein
was
greatly
EMSA. Preparation of the 32P-labeled double-stranded synthetic oligonucleotide, which contained base pairs -999 to -979 of the mouse CYP1AJ gene (encompassing XRE1) and additional nucleotides at both ends, and the EMSA were essentially as described (18). In some experiments EMSA was performed with cytosolic extracts that had been incubated with 10 nM TCDD (added from a lOOx stock solution in dimethylsulfoxide) in vitro for 3 hr at room temperature. During in vitro transformation of the cytosolic extracts of the C- mutant by TCDD, COS-7 cytosolic extracts transfected with pBM5/NEO-M1-1 or pBM5/NEO or in vitro synthesized ARNT protein was added before gel shift analysis, as indicated. Generation of the ARNT protein. Human ARNT was generated from the ARNT expression vector pBM5/NEO-M1-1 (22), either via transfection into COS-7 cells or via in vitro transcription/translation. Transfection of COS-7 cells with pBM5/NEO-M1-1 or pBM5/NEO and labeling with [S]methionine were performed as described previously (18). Cytosolic extracts of COS-7 cells were prepared as described above for EMSA. For in vitro transcription/translation assays, the TNT T7-coupled reticulocyte lysate system containing RNase inhibitor (RNAsin) was programmed with pBM5/NEO-M1-1 according to the manufacturers instructions (Promega). Uv cross-linking. The primer 5-TGAGCTCG-3 was annealed to the oligonucleotide 5-TCCGGCTCTTCTCACGCAACTCCGAGCTC A - 3 , and dBrUTP (Sigma) instead of dTTP was used for secondstrand synthesis with DNA polymerase I (Klenow fragment; New England Biolabs). Thus, three dBrUTPs (shown in bold) were incorporated in the second strand, as follows: 5-TGAGCTCGGAGUUGC GUGAAGAAGAGCCGGA-3. Nuclear extracts (100 g of protein), diluted in 25 mM HEPES, pH 7.5, containing 1 mM DTT, 1 mM EDTA, and 10% glycerol, were incubated with 2.5 sg of poly(dI. dC) (Pharmacia) for 20 mm at room temperature and then incubated with the labeled XRE for another 20 mm. uv cross-linking was performed by irradiation with a Photodyne transilluminator, emitting predominantly at A 302 nm, for 30 mm. Either the incubation mixture was irradiated directly, or it was subjected to nondenaturing PAGE and the portion of the gel containing the receptor-XRE complex was irradiated, and the receptor-XRE complex then electroeluted from the gel overnight in 20 mM HEPES, 20 mM Tris . HC1, 1 mM EDTA, pH 8.0. Irradiated material was boiled in SDS-sample buffer for 3 mm to disrupt proteinprotein complexes and all protein-XRE complexes except those in which protein and DNA were covalently cross-linked. The electroeluted protein-XRE complexes were precipitated with acetone and subjected to denaturing SDS-PAGE (7.5%). To the complexes irradiated in solution, affinity-purified Ah receptor or ARNT antibodies or the corresponding preimmune IgG were added in some cases, as indicated in the text, and incubated overnight at 4. Immuncomplexes were precipitated for 2 hr with Protein A-agarose (which had been previously equilibrated in 10 mM Tris.HC1, pH 7.5, 1 mM EDTA, 0.3 M NaCl, 10 zg/ml bovine serum albumin), washed three times in the same buffer, and then subjected to SDS-PAGE (7.5%). The dried gel was exposed at -80 to Kodak X-OMAT AR X-ray film.
nuclei of the C (Fig. 1A, lanes 3 was detected vT12 the These in is
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
not
shown).
transfected with pBM5/NEO-M1-1, mRNA antibodies (22, are fractions. acid 31).
monospecific amino-terminal
nuclear 19-amino
peptide
of the
mouse
Ah receptor
detected
a single
protein
in
1 C
2 N+
3 C
4 N.i-
56 C N+
180-
12486 635439-
HEPA-1
LS18O
ARNT
1 C
4 C
N-
N+
N-
N+
180124-
8663-
54-
39-
LS18O
Results
Characterization of affinity-purified antibodies to
ARNT antibodies. Protein (1 00 g)from uninduced nuclear (N+), or uninduced nuclear (N-) extracts Ah receptor in Materials
and LS1 80 lulose, and
bodies (B). as indicated
Ah receptor and (C), induced of Hepa-1 , C mutant, cells were subjected to SOS-PAGE, transferred to nitrocelthen probed with ARNT antibodies (A) or Ah receptor antiThe positions of the molecular weight markers (recalibrated in Materials and Methods) are indicated on the left.
of affinity-purified
polyclonal
cytosolic
514
uninduced
95 kDa
Probst
et a!.
cytosols
and 110
from
kDa,
Hepa-1
respectively
cells
and
(Fig.
LS18O
1B,
cells,
lanes
of about
1 and 4).
not iments
detected therefore
in the
form that Ah
(Fig.
2, lane
6).
These They
experalso
ARNT receptor
These Hepa-1
are close to the expected sizes for the Ah receptor in and LS18O cells, i.e., 90 kDa (11, 12) and 110 kDa (32),
Thus, the Ah receptor antibodies specifically de-
respectively.
tect the corresponding protein in cytosols of these strains. The antibodies also uniquely detected the Ah receptor in induced nuclei of LS18O cells (Fig. 1B, lane 6). The receptor was not found in uninduced nuclei of LS18O cells (Fig. 1B, lan-c 5), consistent with the known behavior of the receptor. In induced nuclei of Hepa-1 cells, the antibodies detected a protein of about 1 10 kDa, as well as the approximately 95-kDa Ah receptor. spond The larger cells protein (Fig. 1B, was lane also detected ofthe in uninduced does not The Ah receptor. nuclei correlarger of Hepa-1 2) and therefore
Association
the
of cytosol
cell mutant
the 9-S transformed
c4 is defective
receptor
in ARNT
in vitro. function.
but state
The
In this
it by protein extract
cell line,
be (24). conC
in vivo
synthesized added which was
complex is present, to the XRE-binding and 35S-labeled TCDD incubated ARNT to a cytosolic
In
vitro
comitantly mutant,
with then
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
temperature.
When
tor (Fig. protein also
these
3, lane was observed
extracts
1
).
were
immunoprecipitated
not treated after
with
was TCDD,
to a precursor
or derivative
antibodies,
protein (like the Ah receptor) was not detected if the antibodies were preabsorbed with the Ah receptor peptide used to generate the antibodies or if preimmune serum was used (data not shown), demonstrating that the larger protein represents a species that cross-reacts with the Ah receptor antibodies. It
should be noted that ARNT that ARNT unoccupied and the Ah receptor a component from Hepa-1 of the
immunoprecipitation
antibodies
when
(Fig.
preimmune
3, lane
serum IgG,
A very
was used
faint
(Fig.
signal
3, lane
was
5).
faint
signal
ARNT
was most
protein
likely
with
due to a nonspecific
because demonstrate
interaction
of
of labeled
3 and
no precipitation
cells
migrated 9-S
at indistinguishable Ah antibodies.
whether to the this
rates
is not
protein
4
).
occurred
These
in the
experiments
absence
of antibodies
that
(Fig.
exogenous
3,
Demonstrated
cytosolic
immunoprecipitate
receptor
the
complex.
cytosolic contains the
unable
complex
to
in cytosolic
extracts by of Hepa-1 to Fig.
extracts
addition cells was
with
to
ARNT
ascertain
We cytosols
therefore
complex
another
ARNT.
approach
Induced with
nuclei
antibodies
and uninduced
of Hepaand
Ah receptor,
1 cells resulting
incubated
32P-labeled
double-stranded
XRE-containnondenaturing
oligonucleotide
PAGE.
receptor-XRE This XRE was band was also
The
was
position
not
of the
detected
band
corresponding
in 4 (lane excess (Fig. extracts
to
1,
the
arrow).
Ah
from uninduced cytosolic extracts it was detected in immunoprecipitates extracts, antibodies another formal as expected protein with from the (Fig. nuclear latter 2, lane in extracts ofthis protein, 4). immunoprecipitate, interpretation addition
of unlabeled 4, lane
2).
added not
to the with
It
detected
of Hepa-
Ah receptor,
1 cells
is that
1, with induced nuclear extracts of the C- mutant (22), or with uninduced Hepa-1 cytosol (Fig. 4, lane 3). When a cytosolic extract of Hepa-1 cells was treated with TCDD in vitro, the Ah receptor-XRE complex was produced (Fig. 4, lane 4). However, this C When fected to C NEO-M1-1 cytosol ARNT receptor-dependent taming (Fig. complex mutant extracts with the was was not treated prepared human and the or the parental mixture band 4, lane increasing was generated with from ARNT when TCDD COS-7 cDNA plasmid was
7) but
the Ah receptor, in nuclei.] To exclude the possibility Ah receptor antibodies are unable to immunoprecipitate
S cytosolic complex, we also probed the precipitated
a cytosolic
in vitro
of the
6).
4, lane transpBM5/
cells
that
with in the
extracts antibodies Both
The Ah 2, lane 1)
both with
vector
demonstrating probed
observed not
prepared
HSP9O induced
antibodies. cytosolic
1
Whereas HSP9O was identified in the unform of the receptor (Fig. 2, lane 5), it was
2 3 C 4 N 5 C
from
8).
COS-7
Similarly, 6 N
cells
when
containing
the parental
amounts
plasmid
of a rabbit
(Fig.
4, lane
reticulocyte
Extract
124
86
Fig. 2. ARNT is not a component of the untransformed cytosolic Ah receptor complex. Uninduced cytosolic (C) or induced nuclear (N) extracts from Hepa-1 cells (200 tg of protein each) were immunoprecipitated with affinitypurified Ah receptor antibodies, subjected to SOS-PAGE, transferred to nitrocellulose, and probed with Ah receptor, ARNT, and HSP9O antibodies. The molecular weight markers are indicated on the left. Only the range above 65 kOa is shown, to exclude the bands corresponding to
Role of ARNT
2
TCDD Antibody
#{247}
in Ah Receptor
4 5 6 7
Action
9
10 11
515
12
3
+
4
-
5
#{247}
13
Cell line
Extract
HHH NNC
+ . + + . .
HMMMMMMMMM CCC
CCC
CCCC
#{247}
+ +
0.5
190-
12486-
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
6354-
39-
Fig. 3. Exogenous ARNT can associate with the Ah receptor in TCOOtreated cytosol. Uninduced cytosol from the C- mutant (1 50 zg of protein) was incubated at room temperature with 8 l of rabbit reticulocyte lysate containing in vitro transcribed, translated, and S-labeled ARNT, in the presence (+) or absence of 1 0 n TCOO. After treatment with Ah receptor antibodies corresponding preimmune lgG (P), or no anti(-) (+),
precipitates
were
run on SOS-PAGE
and subjected
to auto-
in vitro
resulted
synthesized in the
ARNT generation
were of
added
to the Fig. 4. Restoration of Ah receptor-dependent XRE binding mutant. EMSA of Hepa-1 and C- mutant cells were performed to the C using 100
increasing
amounts of the protein-XRE complex (Fig. 4, lanes 9-12). The protein-XRE complex was not generated when the reticulocyte
lysate (Fig. binding by addition was programmed 4, lane 13). Thus, of the Ah receptor protein of the of ARNT with the complex Ah parental plasmid XRE to can the pBM5/NEO be restored dBrUTP-
in the
C- cytosol,
to the receptor
TCDD-dependent
of protein for cytosolic extracts and 5 ,g of protein extracts. H, Hepa-1 ; M, C mutant; C, uninduced cytosolic
Lg
for nuclear extract; N, induced nuclear extract. Arrow on the left, position of the Ah receptorXRE complex. Lane 2, a 100-fold excess of unlabeled double-stranded
oligonucleotide was included. Lanes 7 and 8, 1 0 g of protein from cytosolic extracts of COS-7 cells, transfected with pBM5/NEO-M1 -1 or pBM5/NEO, respectively, were added. Lanes 9-12, increasing amounts (0.5-3 I) of rabbit reticulocyte lysate programmed with ARNT were added. Lane 13, 3 l of unprogrammed lysate.
Uv
substituted labeled,
dBrUTPs
cross-linking
a three
were
sequence
irradiated with UV to cross-link protein molecules to the XRE core sequence. They were then boiled in SDS-sample buffer to disrupt protein-protein interactions and noncovalent proteinDNA film interactions, (Fig. 5). We did run not on SDS-PAGE, treat the extracts and exposed with DNase to X-ray before
in vitro transformation of cytosol of the C- mutant, the 95kDa cross-linked complex was generated (Fig. 5, lane 8). The 95-kDa complex was not seen when a cytosolic extract of COSing
7 cells transfected with the parental plasmid pBM5/NEO was
electrophoresis, because the electrophoretic mobility of Ah receptor-dependent protein-XRE complexes is not recognizably affected by such treatment (21). Several cross-linked proteinDNA complexes were observed when an induced nuclear extract from Hepa-1 cells was used (Fig. 5, lane 1 The most prominent
).
of the a functional
95-kDa Ah
band receptor,
is
therefore between
not enhance
protein-DNA
(compare from
complexes
of these
had
apparent
molecular
sizes
of 44, 46,
67,
75, 95,
and
140 kDa. The presence of a 100-fold excess of unlabeled XRE abolished all of the bands (Fig. 5, lane 2). The protein-DNA complex at approximately 95 kDa was not observed when either a nuclear or cytosolic extract of uninduced Hepa-1 cells was used (Fig. 5, lanes 4 and 5, respectively). When uninduced
cytosolic extracts from Hepa1 or Cmutant cells were trans-
respectively).
in either nuclei or cytosols of Hepa-1 Fig. 5, lane 1 with lane and lane 6 with lane 5, The latter complexes were also observed using the C- mutant (compare Fig. 5, lane 7 with lane
4
formed with TCDD in vitro and subsequently the 95-kDa band was observed in Hepa-1 but not in the C- mutant (Fig. 5, lane cytosolic extract of COS-7 cells transiently pBM5/NEO-M1-1 was added simultaneously
7).
UV cross-linked, cells (Fig. 5, lane 6) However, when a transfected with with TCDD dur-
5). Formation of these complexes is therefore not dependent upon TCDD, a functional Ah receptor, or ARNT. These complexes may correspond to constitutive XRE-binding proteins, such as those previously detected in Hepa-1 extracts (33), or to other proteins that bind DNA sequences other than the XRE core sequence that are present in the double-stranded oligonucleotide used for the EMSA. When the Ah receptor-XRE complex from TCDD-treated Hepa-1 cells was cross-linked by UV directly in a nondenatu-
516
ProbstetaL
1 2 3 4 5 6 7
Demonstration bind found complex but linked receptor; however, about extracts investigated Ah tide receptor was carried The (to
that
ARNT
and
the
Ah
receptor
both
Cell line
Extract
MM
-
N
-
N
+ +
-
N
-
N
.
C
-
CCC
-
Competitor
TCOOinvivo TCOO in vitro
the XRE core sequence directly. that, after TCDD activation in from to the the 95 kDa) when the and rat XRE liver cytosol molecular core sequence contained weights, and identified. Ah was of this UV in solution experiments distinct apparent
Elferink et at. (21) vitro, the Ah receptor two proteins, that resolved could by described Hepa-1 in using to XRE the the solution. ARNT dBrUTPoligonucleoprepared of similar be crosssubsequent
+
. .
+
-
+
.
#{247}
ARNT
SDS-PAGE.
One
other we detected
of these
was only
proteins
not one
was
shown
As with
to be the
band nuclear
Ah
(at We and
above,
190-
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
124-
substituted
double-stranded
extracts
86-
from
Hepa-1
cells
disrupt SDS,
that
had
been
cultured
were protein-protein
in the
boiled in
presence
SDS-sample interactions), and
of
TCDD.
buffer
63-
cross-linked and
extracts then
noncovalent
diluted
to 0.4%
with
the
ARNT
Ah
antibodies.
from the with band Hepa-1 In the same
Both
a cross-linked
the ARNT
and
Ah receptor
complex (Fig. 6, lanes after as in the
antibodof about
3 and
extract protein6, lanes of material greater
6,
not
antibodies, at 95 kDa
).
39-
XRE
and
5
were
amounts loaded
6, lanes
were
to obtain
4
uV.tSSIjflkOd
in
in Fig. 6, lanes antibody preparations plexes from the Hepa-1 results indicate that
6.) The
not 95-kDa Ah
corresponding
precipitate
extracts different
region
(Fig.
6, lanes
4 and
complexes,
band
gel
contains
to the to during the
two
ARNT
cross-linked
receptor of the This
Fig. 5. UV cross-linking
of the nuclear
individually with
dBrUTP-substituted XRE. Nuclear (N) extracts were prepared from Hepa-1 cells (H) or C mutant cells (M) grown in the presence (+) or absence (-) of 2 nM TCOO for 90 mm. Cytosolic extracts (C) were incubated with (+) or without (-) 1 0 n TCOO in vitro. The extracts were then incubated with the dBrUTP-containing and 32P-labeled doublestranded oligonucleotide, encompassing XRE1 . Proteins were crosslinked to the oligonucleotide in solution, except for lane 3, where the Ah receptor-XRE complex was cross-linked directly in the gel after nondenatunng gel electrophoresis and then eluted from an appropriate slice of the gel. Lane 7, cytosolic extract of the C mutant was cross-linked in the absence of TCDO. Lane 8, cytosol was incubated with an extract
core
electrophoresis.
is consistent
that
these
two proteins
easily be Ah receptor to the Ah kDa), cells, Ah advantage of 110 LS18O the added
in Hepa-1
resolved by antibodies receptor we also experiments in which (see that it in
cells
are both
about
(see another
90 kDa
Fig. 1). protein
and cannot Because the in addition cells and of has detect the (albeit human only
carried
immunoprecipitation receptor
prepared
presence
from COS-7
of TCOO.
Fig.
possesses
ring
ing
ditional
polyacrylamide
conditions,
bands at
run
certain
under
receptor
another antibodies cross-linked L5180 sistent (see linked 13), LS18O antibody Fig.
proteins
that they
that
can
are sufficiently
be resolved from nuclear (Fig. antibodies of about of the by an approximately
different
SDS-PAGE. 95-kDa extracts 6, lane human 110 ARNT
in size from
protein-XRE prepared
one
the
about
95-kDa were
40
complex observed.
and kDa
of the
also
protein-DNA
complexes
Furthermore,
were
precipitated complex cells with 1). grown the The Ah with known
detected.
These last bands were also seen on occasion when cross-linking was performed in solution, particularly when longer times of irradiation were used (data not shown). The 220-kDa band
could represent a complex in which both XRE-binding com-
TCDD size
10).
receptor complex
precipitated kDa
protein-XRE
ponents of the Ah receptor are cross-linked to the same XRE molecule (see below), as suggested by Elferink et al. (21). The 40-kDa band could represent an additional XRE-binding component of the Ah receptor that is cross-linked only under more
intense irradiation conditions or a degradation product of either
which
cells
is consistent
(32) (see Fig. did preparations
with
1). not
the
The
size
of the
any
Ah
receptor
preimmune protein-XRE
in
corresponding
precipitate
complexes
14).
from
Ah receptor
L5180
bind
nuclear
therefore the
extracts
demonstrate XRE directly.
(Fig.
6, lanes
that both
11 and
ARNT
These the
experiments
ARNT
or the Ah receptor.
and
Role of ARNT
1 23 4 5 6 7 8 9 10 11 12 13 14
Action
517
transfor-
posal mation on
is also put
consistent forward by
with analysis,
a model those
receptor
H
.
H
AAPA
+
H
+ .
H
+ -
H L
+
H L
+ .
H
L
+ -
L5 L5 L5
A
. +
LS L5 LS L5
PA
+
-
Gasiewicz
co-workers investigators
(20).
Based
A
+ -
L
+
L
+ -
L
+ .
physico-chemical
suggested
that then
TCDD associates
causes
form
release
of the
of the
free
and
Ah receptor
that the free (20). cytosolic
from
Ah We
the
receptor
9-
S cytosolic
..
receptor
with
factor and
a cytosolic
form as being co-workers for of the ARNT.
protein
receptor
to generate
the
in the factor
high
studdid of the
DNA-binding cytosolic
identify
190
of Gasiewicz appear
to be required
TCDD-dependent
release
124
4:3-
Ah receptor
ARNT does
triggers
from not
the 9-S complex. This therefore suggests that actively participate in the process whereby
dissociation of the 9-S complex.
86
TCDD
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
63#{149}-
have
can be
observed
found of
that
in the about
after
nucleus 6 S
ligand
(18, 19).
binding
However, the 9-S with the
in
cells as Ah
of Hepa-1
species
54
well
as the
in
et al. (34) and Perdew (35) detected 6-S species in nuclei of Hepa-1 cells
culture. Our When we treated induced do not
treated
with with
notion
TCDD
receptor
3934
-
nuclei support
antibodies,
HSP9O
results
did not
therefore
coimmunoprecipitate
can translocate the difference
the
Ah receptor.
to the between
the
ences Fig. 6. ARNT and the Ah receptor both interact directly with the XRE
result
Wilhelmsson
and
to other
those
may relate conditions
obtained
to differused
by
in
in extraction
core sequence. The dBrUTP-substituted and 32P-labeled double-stranded XRE oligonucleotide was cross-linked in solution with nuclear extracts prepared from Hepa-1 (H) or LS1 80 (LS) cells that had been grown with 2 n or 10 nM TCOD, respectively. The extracts were boiled for 3 mm in SOS-sample buffer and diluted, and then ARNT (A) or Ah receptor (L)
antibodies or lgG fractions from the corresponding preimmune sera (PA and P, respectively) were added, as indicated. The resulting immunoprecipitates, or one third of the supematant from each immunoprecipitation, were then subjected to SOS-PAGE. Lanes 1 and 8, cross-linked material, in amounts equivalent to those in the corresponding supematants, was subjected to SOS-PAGE directly, without immunoprecipitation. Closed arrow, position of the immunoprecipated Ah receptor from
the different experiments, and the aforementioned issue be considered to remain unresolved. We previously demonstrated that the 6-S Ah receptor plex extracted from nuclei of Hepa-1 cells grown with
is a heterodimer Furthermore, receptor ments directly evidence rat Ah contains we consisting showed both that proteins of the the (18). Ah receptor However, whether the XRE of the different and form these ARNT indirectly, XRE-binding
should comTCDD
ARNT. of the experibinds by
piggy-backing
Elferink
et al. (21)
provided that
that receptor
the
form two
Sprague-Dawley proteins
bind
teins kDa, affinity receptor
the
XRE
directly.
The
molecular
sizes
of these
two
pro-
were determined and the smaller ligand, behaves protein as detected the
Discussion
The size of the untransformed cytosolic Ah receptor complex
as (21)
is must
likely,
similarly
receptor,
other
by Elferink
has been estimated to be about 280 kDa sucrose gradient analysis (10) and about
intense UV irradiation conditions TCDD-dependent protein-XRE 220 kDa. The 40-kDa protein could protein or a degradation
cross-linking
for subunits
and
SDS-PAGE
97, 96,
(15).
88,
In the
and 46
latter
kDa
case,
was
evidence
obtained.
represent of either
of about
product
Because the Ah receptor, HSP9O, and ARNT are all about 90 kDa, one possibility is that the unliganded untransformed Ah receptor complex contains one molecule of each of these polypeptides. However, our results show that the immunoprecipitated unliganded receptor complex does not contain ARNT, and they are therefore incompatible with this idea. Furthermore, our analysis of the immunoprecipitated nuclear Ah receptor complex confirms that ofthe this complex does contain the 9-S ARNT.
ARNT
or the Ah receptor
that
was generated
during
processing
of the sample. Elferink et at. (21) observed 220 kDa in their cross-linking experiments this that one linked species the molecule to the contained 220-kDa of the same complex the XRE, protein). on our results, we provide the following Ah Ah the species protein Ah receptor. corresponds receptor that and they one They to identified
a species of about and showed that provided a complex molecule are both evidence in which of the other crossoligonu-
XRE-associated
XRE-containing
double-stranded
Thus,
ARNT
triggers
must
release
associate
with
the
Ah
receptor
only
after
cleotide
cross-linked between the 40-kDa Based
molecule
(21).
It is therefore
that we observed and receptor,
likely
ARNT
that
(and
the
TCDD
Ah receptor
from
cytosolic
represents
220-kDa a complex
also model
receptor complex. Our demonstration that exogenously added ARNT can associate with the Ah receptor after TCDD treatment of cytosol in vitro supports this last proposal. This pro-
perhaps working
518
or other
ciation
of HSP9O
possibly
free 3) The Ah
other
receptor heterodimer
proteins
and release
translocates
of the
with into the
heterodimerizes
16. Fujisawa-Sehara, A., K. Sogawa, M. Yamane, and Y. Fujii-Kuriyama. acterization of xenobiotic responsive elements upstream from the tabolizing cytochrome P450c gene: a similarity to glucocorticoid elements. Nucleic Acids Rex. 15:4179-4191 (1987).
17. Watson, A. J., and 0. Hankinson. Dioxin and Ah receptor-dependent
Chardrug meresponse
protein
nucleus. 4) Both ARNT and the Ah receptor the core sequences of XREs located 5 to the of the CYPJAJ gene. 5) Transcription of the thereby activated.
bind
coding
directly
sequence
to is
CYPJAJ
gene
binding to xenobiotic responsive elements and G-rich DNA studied by in vivo footprinting. J. BWL Chem. 266:6874-6878 (1992). 18. Reyes, H., S. Reisz-Porszasz, and 0. Hankinson. Identification of the Ah receptor nuclear translocator protein (Arnt) as a component of the DNA binding form of the Ah receptor. Science (Washington D. C.) 256:1193-1195
(1992).
In summary, unliganded
that Ah
ARNT receptor
is not complex,
a component and
of the
19. Prokipcak,
R. D., and
A. B. Okey.
we demon20.
nuclear
Henry,
forms
Ah receptor
tetrachlorodibenzo-p-dioxin. of
that both subunits of the nuclear Ah receptor complex directly to the XRE core sequence. Finally, we have demonstrated that addition of in vitro synthesized ARNT protein along with TCDD to cytosolic extracts of the C- mutant results in association ofthe ARNT protein with the Ah receptor
and its participation in binding to the XRE. These last results
strate bind
21.
22.
recently reported by Poellinger results substantiate the working also represent a significant step Ah receptor activity in vitro.
23.
E. C., G. Rucci, the Ah receptor 28:6430-6440 (1989). Elferink, C. J., T. A. Gasciewicz, and J. P. Whitloek. Protein-DNA interactions at a dioxin-responsive enhancer. J. Ball. Chem. 265:20708-20712 (1990). Hoffman, E. C., H. Reyes, F-F. Chu, F. Sander, L. H. Conley, B. A. Brooks, and 0. Hankinson. Cloning of a factor required for the activity of the Ah (dioxin) receptor. Science (Washington D. C.) 252:954-958 (1991). Hankinson, 0., B. A. Brooks, K. I. Weir-Brown, E. C. Hoffman, B. S. Johnson, J. Nanthur, H. Reyes, and A. J. Watson. Genetic and molecular analysis ofthe Ah receptor and ofCYP1A1 gene expression. Biochimie (Paris) 73:61-66 (1991). Legraverend, C., R. R. Hannah, H. J. Eisen, I. S. Owens, D. W. Nebert, and 0. Hankinson. Regulatory gene product of the Ah locus. J. BiOL Chem.
257:6402-6407 (1982).
characterization of the in culture to 2,3,7,8Arch. Biochem. Biophys. 267:811-828 (1988). and T. A. Gasiewicz. Characterization of multiple comparison of species and tissues. Biochemistry
Physicochemical
cells
exposed
Downloaded from molpharm.aspetjournals.org at UNAM (PAF) - Fac. De. Quim Est. Profesionales on April 5, 2013
We thank Dr. Stephen Ulirich murine 84-kDa and 86-kDa heat his help. References
for providing us with the antiserum against shock proteins. We also thank Steve Bacsi
M. I., E. Overvik, M. G. F. Mason, and J.-A. Gustafsson. In vitro of the dioxin receptor to a DNA-binding form by food-borne heterocyclic amines. Carcinogenesis (Land.) 13:1619-1624 (1992). 2. Safe, S. H. Comparative toxicology and mechanism of action of polychlorinated dibenzo-p-dioxins and dibenzofurans. Anna. Rev. PharrnacoL ToxicoL 26:371-399 (1986). 3. Whitlock, J. P. Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzop-dioxin action. Annu. Rev. PharrnacoL ToxicoL 30:251-277 (1990). 4. Gonzalez, F. J. The molecular biology of cytochrome P450s. PharrnacoL Rev.
40:243-288 (1989).
1. Kleman, activation
5. Guengerich, 6.
F. P. Characterization
of human
microsomal
enzymes. Annu. Rev. PharmacoL ToxicoL 29:241-264 Landers, J. P., and N. J. Bunce. The Ah receptor dioxin toxicity. Biochem. J. 276:273-287 (1991).
Bradfield, C. A., E. Glover, and A. Poland. Purification and N-terminal amino acid sequence of the Ab receptor from the C57BL/6J mouse. MoL PharmacoL 39:13-19 (1991); erratum in MoL PharrnacoL 39:435 (1991). 26. Marston, F. A. 0. The purification of eucaryotic polypeptides synthesized in Escherichia coil. Biochem. J. 240:1-12 (1986). 27. Ullrich, S. J., E. A. Robinson, L. W. Law, M. Willingham, and E. Appella. A mouse tumor-specific transplantation antigen is a heat shock-related protein. Proc. NatL Acad. Sci. USA 83:3121-3125 (1986). 28. Hankinson, 0. Selections for and against cells possessing cytochrome P4501A1-dependent aryl hydrocarbon hydroxylase activity. Methods Enzymol. 206:381-400 (1991). 29. Shapiro, D. J., P. A. Sharp, W. W. Wahli, and M. J. Keller. A high-efficiency HeLa cell nuclear transcription extract. DNA 7:47-55 (1988). 30. Poland, A., E. Glover, and C. A. Braduield. Characterization of polyclonal antibodies to the Ah receptor prepared by immunization with a synthetic peptide hapten. MoL Pharrno.coL 39:20-26 (1991).
31. Reyes, H. Croning of a subunit of the Ah (dioxin) receptor. Ph.D. thesis,
25.
7. Whitlock, J. P., and D. R. Galeazzi. 2,3,7,8-Tetrachlorodibenzo-p-dioxin receptors in wild-type and variant mouse hepatoma cells. J. BIOL Chem. 259:980-986 (1984). 8. Denison, M. S., P. A. Harper, and A. B. Okey. Ah receptor for 2,3,7,8tetrachlorodibenzo-p-dioxin: codistribution of unoccupied receptor with cytosolic marker enzymes during fractionation of mouse liver, rat liver and cultured Hepa-lcl cells. Eur. J. Biochem. 155:223-229 (1986). 9. Gudas, J. M., S. Karenlampi, and 0. Hankinson. Intracellular localization of the Ah receptor. J. CelL PhysioL 128:441-448 (1986). 10. Denison, M. S., L. M. Vella, and A. B. Okey. Hepatic Ah receptor for 2,3,7,8tetrachlorodibenzo-p-dioxin: partial stabilization by molybdate. J. BIOL
Chem. 261:3987-3995 (1986).
32.
33.
University of California, Los Angeles (1991). Harper, P. A., R. D. Prokipcak, L. E. Bush, C. L. Golas, and A. B. Okey. Detection and characterization of the Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin in the human colon adenocarcinoma cell line LS18O. Arch. Biochem. Biophys. 290:27-36 (1991). Saatciouglou, F., D. J. Perry, D. S. Pasco, and J. B. Fagan. Multiple DNAbinding factors interact with overlapping specificities at the aryl hydrocarbon response element of the cytochrome P4501A1 gene. MoL CelL BiOL 10:64086416 (1990).
34.
35.
11. Ema, M., N. Watanabe, Y. Chujoh, N. Matsushita, 0. Gotoh, Y. Funae, and Y. Fujii-Kuriyama. cDNA cloning and structure of a mouse putative Ah receptor. Biochem. Biophys. Res. Commun. 184:246-253 (1992). 12. Burbach, C. M., A. Poland, and C. A. Braduield. Cloning of the Ah-receptor cDNA reveals a distinctive ligand-activated transcription factor. Proc. NatL Acad. Sci. USA 89:8185-8189 (1992). 13. Perdew, G. H. Association of the Ah receptor with the 90-tWa heat shock protein. J. BIOL Chern. 263:13802-13805 (1988). 14. Denia, M., S. Cuthill, A-C. Wikstroem, L. Poellinger, and J.-A. Gustafsson. Association of the dioxin receptor with the M, 90,000 heat shock protein. Biochem. Biophys. Res. Commun. 155:801-807 (1988).
36.
Wilhelmsson, A., S. Cuthill, M. Denis, A-C. Wikstroem, J.-A. Gustafsson, and L. Poellinger. The specific DNA binding activity of the dioxin receptor is modulated by the 90 1W heat shock protein. EMBO J. 9:69-76 (1990). Perdew, G. H. Comparison of the nuclear and cytosolic form of the Ah receptor from Hepa lclc7 cells: charge heterogeneity and ATP binding properties. Arch. Biochem. Biophys. 291:284-290 (1991). Whitelaw, M., I. Pongratz, A. Wilhelmsson, J.-A. Gustafsson, and L. Poellinger. Ligand-dependent recruitment of the Arnt coregulator determines DNA recognition by the dioxin receptor. MoL Cell. BalL 13:2504-2514 (1993).
Send reprint requests to: Oliver Hankinson, Laboratory and Molecular Medicine, University of California, Los Avenue, Los Angeles, CA 90024.
Biology Veteran