- Email: [email protected]
Detection and characterization of the Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin in the human colon adenocarcinoma cell line LS180
ARCHIVES
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 290, No. 1, October, pp. 27-36, 1991
Detection and Characterization of the Ah Receptor for 2,3,7,8-Tetrachlorodibenzo-p-dioxin in the Human Colon Adenocarcinoma Cell Line LSI 80 Patricia
A. Harper,
Rebecca D. Prokipcak,
Leslie
E. Bush,
Cheryl
L. Golas, and Allan
B. Okey’
Division of Clinical Pharmacology and Toxicology, Research Institute, The Hospital for Sick Children and Departments of Pediatrics and Pharmacology, University of Toronto, Toronto, Ontario. Canada M5S IA8
Received March 12, 1991, and in revised form June 1, 1991
The Ah (aromatic hydrocarbon) receptor mediates induction of aryl hydrocarbon hydroxylase (AHH; an enzyme activity associated with cytochrome P450IAl) by polycyclic aromatic hydrocarbon carcinogens such as 3methylcholanthrene (MC) and benzo[a]pyrene (BP) and the halogenated toxin 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD). Until recently the AhR seemed to be present only at very low levels in human cells and tissue. With a modified assay (the presence of sodium molybdate and a reduction in the amount of charcoal used to adsorb “excess” ligand) we found that cytosol from LS180 cells contains a high concentration of AhR (400-500 fmol/mg cytosolic protein) when detected by [3H]TCDD or [3H]MC. Cytosolic receptor also was detected with [3H]BP but at a level that was 35% of that detected with [3H]TCDD or [3H]MC. These levels are similar to those found in mouse Hepa-l hepatoma cells in which AhR has been extensively characterized. The apparent binding affinity (&) of the cytosolic receptor for [3H]TCDD and for [3H]MC was about 5 nM. As with Hepa-1, the human LS180 cytosolic AhR sedimented at about 9 S on sucrose gradients when detected with [3H]TCDD, [3H]BP or [3H]MC. The nuclear-associated ligand * receptor complex recovered from cells incubated in culture with [3H]TCDD sedimented at about 6.2 S. The 9.8 S cytosolic form corresponds to a multimeric protein of a relative molecular mass (M,) of about 285,000 whereas the 6.2 S nuclear receptor corresponds to a multimeric protein of M,. 175,000. The smallest specific ligand-binding subunit (detected by sodium dodecyl sulfate-polyacrylamide electrophoresis under denaturing conditions of receptor photoaffinity labeled with [3H]TCDD) was about M, 110,000. AHH activity was induced in cells exposed in culture to TCDD or benz[a]anthracene (BA). The EC&,
’ To whom correspondence should be addressed at the Department of Pharmacology. Fax (416) 978-6395. 0003.9861/91 $3.00 Copyright GJ 1991 by Academic Press, All rights of reproduction in any form
was 4 X lo-” M for TCDD and 1.5 X lo-’ M for BA. For both inducers the ECSO in LSl80 cells was shifted about one log unit to the right as compared to the ECeO for AHH induction in mouse Hepa-l cells. The lower sensitivity of the LS180 cells to induction of AHH activity by TCDD or BA is consistent with the lower affinity of TCDD and MC for binding to human AhR. The ligand-binding properties, physicochemical properties, and mode of action of the AhR in this human cell line are therefore very similar to those of the extensively characterized AhR in 8 1991 Academic press, I~C. rodent cells and tissues.
The Ah (aromatic hydrocarbon)’ receptor mediates induction of aryl hydrocarbon hydroxylase (AHH: P450IAl) by polycyclic aromatic hydrocarbons and has been extensively characterized in rodent cells and tissues (l-7). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and related halogenated and nonhalogenated hydrocarbons bind to the AhR in the cytoplasm. The receptor - ligand complex is then “transformed” to a nuclear-associated protein which binds to specific regulatory regions of DNA and increases synthesis of P450IAl mRNA, among other responses (8-15). Recently the AhR has been described in several human cell lines and tissues (16-23). With improved techniques for detection and quantitation, the human AhR appears to function in a manner similar to the AhR in rodent cells and tissues. However, the human AhR has appeared to * Abbreviations used: Ah, aromatic hydrocarbon; AHH, aryl hydrocarbon hydroxylase; AhR, aromatic hydrocarbon receptor; BA, benz[a]anthracene; BP, benzo[a]pyrene; BSA, bovine serum albumin; CAT, catalase; DB[a,h]A, dibenz[a,h]anthracene; DMSO, dimethyl sulfoxide; MC, 3-methylcholanthrene; MEM, minimal essential medium; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; TCDF, 2,3,7,8-tetrachlorodibenzofuran. 27
Inc. reserved.
28
HARPER
be much less stable than the rodent receptor, making it difficult to do a full characterization of the human AhR’s physicochemical properties. In this paper we present our observations on human AhR detected in the colon carcinoma cell line LS180. With suitable assay conditions an abundant specific binding protein is readily identifiable. The AhR detected in LS180 cells is sufficiently stable to allow more extensive characterization by both gel permeation chromatography and SDS-PAGE
MATERIALS
electrophoresis AND
of photoaffinity-labeled
AhR.
METHODS
Chemicals. [$H]TCDD (35 Ci/mmol) and nonradioactive TCDF were generous gifts from Dr. S. Safe (Texas A & M University). [3H]Triamcinolone acetonide (33 Ci/mmol) and [3H]benzo[a]pyrene (96 Ci/mmol) were obtained from DuPont Canada-NEN Research products (Lachine, Quebec). [3H]MC (generally labeled, 37 Ci/mmol) was from Amersham Corp. (Oakville, Ontario). TCDD and TCDF are extremely toxic substances and should be handled with care, as described by Poland and Glover (24). Sephacryl S 300 was from Pharmacia. All SDS-polyacrylamide gel supplies were from Bio-Rad. Sucrose (density gradient grade) was from Beckman Instruments (Toronto, Ontario); 4-(-hydroxyethyl)-l-piperazineethanesulfonic acid was from CalbiochemBehring Corp. (La Jolla, CA); nonradioactive BA, BP, DB[a,h]A, and MC were obtained from Aldrich Chemical Co. (Milwaukee, WI); molybdate (sodium salt), dithiothreitol, dexamethasone, bovine serum albumin, and catalase were obtained from Sigma Chemical Co. (St Louis, MO); dimethyl sulfoxide, glycerol, charcoal (Norit A), ethylenediaminetetraacetic acid, and all other routine chemicals were obtained from Fisher Scientific Co. (Toronto, Ontario). Buffers. The main buffer for cytosol preparation was HEGDM (25 mM 4-(-hydroxyethyl)-l-piperazineethanesulfonic acid, 3 mM ethylenediaminetetraacetic acid, 2 mM dithiothreitol, 10% glycerol (v/v), and 20 mM sodium molybdate). HEGDMK buffer is HEGDM with the addition of 0.5 M potassium chloride. HEDM is a hypotonic buffer consisting of HEDGM but without the glycerol; HEPGMD is HEDM plus 20% v/v glycerol. For preparation of some nuclear extracts cells were homogenized in MDH buffer (3 mM magnesium chloride, 1 mM dithiothreitol, and 25 mM Hepes) and nuclei were extracted with HDK (25 mM Hepes, 1 mM dithiothreitol, and 0.5 M NaCl). All buffers were adjusted to pH 7.4 at room temperature. Hepa-l is a derivative of a transplanted Cell lines and cell culture. hepatoma, originating in a C57L/J mouse (8, 25). The subclone, Hepalclc9, used in the present study was isolated and kindly provided by Dr. M. Dufresne (University of Windsor, Windsor, Ontario). Throughout the remainder of the paper the term Hepa-l refers to the subclone Hepalclc9 cells. The LS180 cell line, derived from an adenocarcinoma of the colon, was obtained from the American Type Culture Collection (Rockville, Maryland). Each cell line was grown as a monolayer culture in a-minimum essential medium ((u-MEM) supplemented with 10% fetal calf serum. The cells were routinely harvested by trypsinization when the density on the culture dish reached 1.5 to 3 X lo5 cells per centimeter squared for Hepa-l cells or when cells appeared confluent in the case of the LS180 cells and then washed in Dulbecco’s phosphate-buffered saline before homogenization. Cell fractionation. Prewashed Hepa-l cells were resuspended in the hypotonic buffer HEDM at about 2 to 4 X 10s cells per milliliter. As LS180 cells in suspension are extremely self-adherent, cell number could not be determined; therefore, the cells from 10 petri dishes (10 cm diameter) were resuspended in 1 ml of HEDM buffer. Each cell line was homogenized with a Polytron fitted with a PT 7 generator (Brinkmann Instruments, Rexdale, Ontario) until >90% of the cells were fractured but the nuclei were still intact (as determined by phase contrast mi-
ET AL. croscopy). The homogenate was immediately diluted 1:l with HE2GDM buffer and centrifuged at 105,OOg for 1 h. The resulting supernatant fraction (cytosol) was stored in l- to 2-ml aliquots in liquid nitrogen until analysis. Protein concentrations were determined by the method of Bradford (26) using BSA as the standard. Preparation of nuclear extract. Nuclear extracts were prepared from cells exposed to [3H]TCDD in culture; cytosols also were obtained from these cells. The cell homogenate was centrifuged for 4 min at 10,OOOg in a Beckman microfuge and the resulting pellet was washed two to three times with HEGDM buffer, incubated for 1 h with HEGDK buffer at 4°C to extract the nuclear receptor, and then centrifuged at 105,OOOg for 1 h. The resulting supernatant fraction is termed the nuclear extract. Nuclear extracts for gel permeation chromatography and photoaffinity labeling were prepared essentially as described by Denison et al. (11). Briefly, cells were incubated with 2 nM [aH]TCDD in a-MEM for 2 h at 37“C, washed, and then incubated for 15 min in 10 mM Hepes, pH 7.5. The cells were scraped off the dish, collected, and homogenized in MDH buffer using a Dounce homogenizer with a close-fitting pestle. The homogenate was centrifuged at 1OOOgfor 5 min and the resulting pellet was washed two to three times with MDH buffer containing 0.1 M NaCl and incubated for 45 min with HDK buffer at 4°C to extract the nuclear receptor; glycerol was added to a final concentration of 10% and the mixture was then centrifuged at 105,OOOgfor 1 h. The resulting supernatant fraction is termed the nuclear extract. Preparation of cytosol for analysis by velocity sedimentation and Sephacryl S-300 anulysis. The procedure used was essentially that of Denison et al. (27). Briefly, cytosols to be analyzed for AhR were incubated with [3H]TCDD (or other [3H]ligand as indicated in the text and figure legends), in the absence or presence of a loo-fold molar excess of the competitor TCDF (unless noted otherwise in the text). Samples analyzed for glucocorticoid receptor were incubated with [aH]triamcinolone acetonide with or without the competitor dexamethasone. All incubations were for 2 h at 4°C. After incubation unbound radioligand was removed by adding the cytosols to a pellet of dextran-coated charcoal; 0.1 mg charcoal per milligram cytosolic protein was used for samples to be analyzed on sucrose gradients, or 0.5 mg/ml for samples to be analyzed on Sephacryl S-300. The charcoal-to-cytosol ratio was increased for samples to be analyzed on Sephacryl S-300 to reduce the free [3H]TCDD in the void volume; further increase in the charcoal-to-cytosol ratio resulted in the loss of [sH]TCDD. AhR. Cytosol was incubated with dextran-coated charcoal for 15 min at 4°C; charcoal was removed by centrifugation at 4000g. Nuclear extracts obtained from cells labeled in culture with [3H]TCDD were not treated with dextran-coated charcoal; these samples were applied directly to sucrose gradients or to Sephacryl S-300 columns. Samples were analyzed Velocity sedimentation on sucrosegradients. by density gradient centrifugation using a vertical tube rotor technique as described by Tsui and Okey (28) and Harper et al. (17). [“‘ClFormaldehyde-labeled BSA (4.4 S) and [‘4C]formaldehyde-labeled catalase (11.3 S) were included in each gradient as internal sedimentation markers. Marker proteins were labeled with [‘*C]formaldehyde as previously described (29). Monolayer cells Induction and assay of aryl hydrocarbon hydroxylase. (subconfluent) were treated with various concentrations of [3H]TCDD or BA dissolved in dimethyl sulfoxide. The final concentration of DMSO was 1% in the culture medium in all samples. This concentration of DMSO was not toxic and did not increase the AHH activity above the basal noninduced level. Radiolabeled TCDD was used because of the extremely low concentrations of TCDD which were sufficient to induce AHH activity; the concentrations were verified by determining the radioactivity in each dilution and, where possible, in the incubating media. After 24 h cells were harvested by trypsinization and washed two to three times with phosphate-buffered saline. AHH activity in whole cells was determined by the method of Nebert and Gelboin (30) with cell protein levels of 0.75 to 1.0 mg per assay flask incubated for 30 min at 37’C. There was no fluorescence contribution at any concentration of [3H]TCDD, BA, or DB[a,h]A, since parallel induced samples assayed
Ah RECEPTOR
IN HUMAN
in the absence of the substrate BP gave values that were essentially zero. One unit of AHH activity is defined as the amount of enzyme catalyzing in I min at 37°C the formation of hydroxylatedproduct causing fluorescence equivalent to that of 1 pmol of 3-hydroxybenzo[a]pyrene (recrystallized) standard. Activity is expressed as pmol per minute per milligram cell protein. Protein concentrations for each sample were determined by the method of Bradford (26). Gel permeation chromatography Gel permeation chromatography. was done essentially as described by Denison et al. (29) and Prokipcak and Okey (31). Sephacryl S-300 was equilibrated in HEGDM or HEGD buffer containing 0.5 M NaCl and poured into 1.6 X loo-cm glass columns. Standards used to calibrate the gel permeation columns and their stokes radii were thyroglobulin (8.5 nm), ferritin (6.1 nm), catalase (5.2 nm), bovine serum albumin (3.6 nm), ovalbumin (2.7 nm), and cytochrome c (1.8 nm). The Stokes radius for catalase was obtained from a published table (32), all others were calculated as described by Denison et al. (29). The applied sample size was about 12 mg of cytosolic protein or about 5 mg nuclear extract protein. Distribution coefficients (Ko), Stokes radius (R,), sedimentation coefficient (S xO,w), relative molecular mass (M,), frictional ratio (f/s), and axial ratio (a/b) of a hydrodynamically equivalent prolate ellipsoid were calculated as described by Denison et al. (27). For each sample the measured S 20,wand R, were used to calculate M,, f/f,,, and a/b. These values, derived separately for each individual sample, were then used to calculate the mean and standard deviation for the group of samples within a replicate series. Photoafinity labeling with [3H]TCDD and separation by SDS-PAGE Cytosol and nuclear exof the AhR from cytosol and nuclear extracts. tracts from LS180 cells were photoaffinity labeled with [3H]TCDD essentially as described by Prokipcak and Okey (33). Briefly [aH]TCDDlabeled cytosol prepared in HEDGM buffer (1 to 1.5 ml at -3 mg protein/ ml; 300-400 fmol specific binding/mg protein) and nuclear extracts prepared in HDK buffer (1 to 1.5 ml at 1 to 2 mg protein; 1006150 fmol specific binding/mg protein) were irradiated on ice for 15-20 min, 11 cm from an 85-W lamp (Johns Scientific) or for 2 min, 2 cm from a 450-W lamp (Hanovia Ace Glass). After irradiation proteins were precipitated in cold acetone (-20’(Z), collected by centrifugation, then dissolved directly in sample buffer. After boiling for 2 to 4 min samples and standard molecular mass markers were analyzed by electrophoresis on discontinuous polyacrylamide gels by the method of Laemmli (34). i4C-labeled standards (Amersham) included myosin (200 kDa), phosphorylase b (doublet of 100 kDa and 92.5 kDa), bovine serum albumin (66 kDa), ovalbumin (43 kDa), and carbonic anhydrase (30 kDa). Fluorography. Fixed gels were impregnated with Enlightening (DuPont), dried, and placed in contact with Kodak X-OMAT film for 6 weeks to 5 months before developing (33). RESULTS
LS180
29
CELLS 20
1
TCDD + -o-
-o-.-
I
24 nM [JH]MC +2pMTCDF
I
30 Oi
“.
19 nhl[3H]TCDD +2pMTCDF
‘v’
.
‘v’
.
.
-&
17 nM [3H]BP
f
+2pMTCDF
.
25
FIG. 1. Binding of [3H]TCDD, [3H]MC, and [aH]BP to the AhR. Cytosol from LS180 cells prepared in HEDGM buffer was incubated with 20 nM [3H]TCDD, 20 nM [3H]MC, or 20 nM [aH]BP in the absence or presence of a loo-fold molar excess of the competitor TCDF. Samples were analyzed by velocity sedimentation on sucrose gradients. [i4C]formaldehyde-labeled BSA (4.4 S) and [‘%]formaldehyde-labeled catalase (11.3 S) were added to the gradients as internal sedimentation markers (arrowheads).
Cytosolic Receptor The AhR can be detected by sucrose density gradient analysis of cytosolic samples that have been incubated with [3H]TCDD. Specific binding of the ligand to the AhR is determined by the comparison of profiles from samples incubated in the presence of a loo-fold excess of a specific competitor such as TCDF with binding in the absence of competitor. Characteristically, a specific peak of radioactivity is detectable in the 9 S region of the sucrose gradient (approximately at fraction numbers 10-15). Several ligands known to be agonists for the AhR were tested for their ability to bind to the LS180 AhR. As shown in Fig. 1 incubation of LS180 cytosol with [3H]TCDD, [“H]MC, or [3H]BP resulted in a specific binding peaks in the 9 S region of the sucrose gradient for each ligand;
each of these peaks was eliminated by the presence of loo-fold excess TCDF in the incubation. When rodent cytosols are incubated with [3H]MC and analyzed by a sucrose density gradient there is a specific binding peak in the 4 to 5 S region of the gradient in addition to the 9 S peak. In LS180 cells specific binding was detected with [3H]MC only in the 9 S region of the gradient. When the same cytosol preparation was analyzed using all three ligands the concentration of specific binding sites for [3H]TCDD (355 fmol/mgprotein) and [3H]MC (376 fmol/ mg protein) were very similar, whereas the concentration of binding sites detected with [3H]BP was lower (121 fmol/ mg protein). The amount of AhR detected by [3H]TCDD in LS180 cytosol (300 to 400 fmol/mg protein) was com-
30
HARPER
0
10
20
40
30
ET AL.
50
0
20
10
FREE [3H]TCDD (nM)
30
0.8 -
B map 611 fmol/mg
250
B map 548 fmol/mg K,= 4.6 nM r = 0.98
0.0 liLL 0
500
[3H]TCDD BOUND (fmollmg
50
0.8
0.6
0
40
FREE [3H]MC (nM)
protein)
250
500
[3H]MC BOUND (fmollmg
protein)
15
B map 549 fmol/mg
5-
K,= 5.2 nM r= 1.0
-1
1
3
5
[3H]TCDD UNBOUND (fmoUmg protein, thousands)
7
I 1. -1
1
I 3
.
I 5
.
I 7
’
[3H]MC UNBOUND (fmoVmg protein, thousands)
FIG. 2. Saturation analysis and determination of apparent affinity of binding of [“HITCDD and [3H]MC in cytosols from LS180 cells. Cytosol aliquots (about 4 mg protein/ml) prepared in HEDGM buffer were incubated at 4°C for 2 h with [3H]TCDD (left) or [3H]MC (right) at concentrations ranging from 2 to 40 nM. Specific binding in the 9 S peak was determined for each sample by sucrose density gradient analysis. The Scatchard plots (middle) and Woolf plots (bottom) were derived from data shown in the saturation plots (top). Binding parameters shown for the Scatchard and Woolf plots were calculated by least squares linear regression.
Ah RECEPTOR
IN HUMAN TABLE
Determination
of the Apparent
Affinity
31
LS180 CELLS
I
of [ 3H]TCDD
and [ 3H]MC
for Cytosolic
[3H]MCb
[sH]TCDD”
K+ (nM) B,,,
(fmol/mg
protein)
AhR
Woolf
Scatchard
Woolf
Scatchard
5.2 + 0.83' 554 f108
5.6 f 0.5 551 f108
5.3 +- 3.3 458 f191
5.8 zk 2.6 455 +170
Note. Scatchard and Woolf plots were derived from saturation binding data as shown in Fig. 2. Binding parameters and Woolf plots were calculated by least squares linear regression. u Number of replicates = 5. * Number of replicates = 3. ’ Mean + S.D.
parable to that detected in the rodent cell line Hepa-l (300 to 600 fmol/mg protein). Incubating LS180 cytosol in the presence of [3H]TCDD and loo-fold excess of benz[a]anthracene, benzo[a]pyrene, 3-methylcholanthrene, or dibenz[a,h]anthracene inhibited [3H]TCDD binding to the AhR, whereas phenobarbital, dexamethasone, and diethylstilbestrol had no effect on binding of [3H]TCDD in the 9 S region (data not shown). When LS180 cytosol was incubated with [3H]triamcinolone acetonide (a glucocorticoid analogue), very little specific binding to the glucocorticoid receptor was observed (about 15 fmol/mg cytosolic protein) as compared with [3H]TCDD binding to the AhR of 350 fmol/mg cytosolic protein in the same cytosol (data not shown). In previous studies on human tissues it has been necessary to add sodium molybdate to the homogenizing buffers during cytosol preparation in order to stabilize AhR. A comparison of cytosols prepared from LS180 cells in the presence and absence of sodium molybdate showed no such dependence for the initial detection of receptor. The amount of receptor detected in cytosols with or without molybdate was essentially the same. Although the presence of molybdate did not increase the amount of receptor detected, sodium molybdate did stabilize the receptor during long-term manipulation (>12 h, data not shown) and therefore was routinely added unless otherwise noted. Estimation of Binding Affinity of [3Hj TCDD and [3H]MC for AhR in LS180 Cell Cytosol The apparent Kd for the interaction of [3H]TCDD or [3H]MC with the AhR was estimated by incubating cytosols with a wide range of radioligand concentrations (1 to 40 nM), then determining the specific binding in the 9 S region by sucrose gradient analysis for each sample. As illustrated in Fig. 2, about 20 nM TCDD or MC was necessary to saturate the receptor. Binding data were further analyzed by Scatchard and Woolf plots (Fig. 2 middle and bottom). The kinetic parameters Kd and B,,,
shown for the Scatchard
computed from both plots are shown in Table I. The presence of sodium molybdate in the homogenizing buffer had no effect on the apparent Kd and B,,, (data not shown). Identification
of a Nuclear Receptor- Ligand Complex
The binding of a ligand such as TCDD to the cytosolic receptor results in transformation of the ligand . receptor complex into a nuclear binding protein. This nuclear-associated receptor complex can be extracted from cell nuclei by high salt concentrations and identified by sucrose gradient analysis with no further incubation with [3H]TCDD. Figure 3 shows the sucrose gradient profiles obtained from nuclear and cytosolic fractions of LS180 cells incubated in suspension culture with [3H]TCDD prior to cell fractionation in HEGDM. The nuclear-associated ligand . receptor complex has a lower sedimentation value of 6.2 S (this work and Ref. (31)) as compared to the cytosolic receptor complex which sediments at about 9 S (Fig. 3, top). The 6.2 S binding peak is specific since nuclear extracts from cells incubated with [3H]TCDD in the presence of a loo-fold excess of the competitor TCDF in the medium show no binding in this region of the sucrose gradient (Fig. 3, bottom). AHH Induction
by [3H]TCDD
and BA
Aryl hydrocarbon hydroxylase is one of the best characterized enzymatic responses associated with AhR-mediated cytochrome P450IAl induction. If the AhR is fully functional, exposure of cells to an AhR agonist in culture should result in the induction of AHH activity. LS180 and Hepa-l cells were incubated with various concentrations of [3H]TCDD or BA for 24 h then harvested and the AHH activity in each sample was determined. Since TCDD is active at very low concentrations, [3H]TCDD was used so that the inducer concentration in the culture medium could be accurately determined. The concentration dependency for AHH induction by TCDD and BA is shown in Fig. 4. The absolute AHH
32
HARPER 4
1
BSA v
3
ET AL.
Velocity Sedimentation and Gel Permeation Chromatography on Sephacryl S-300 of Cytosolic and Nuclear Forms of the AhR
CAT
V -0
NUCLEAR
-D
CYTOSOL
2
0.8 -
0.6 -
The physicochemical properties of cytosolic AhR were analyzed in greater detail by performing replicate analyses on several cytosol preparations. Sodium molybdate (20 mM final concentration) was added to stabilize the ligand . AhR complex since inclusion of 0.1 M NaCl in the column buffers in the absence of sodium molybdate results in a mixture of 9 S and 6 S forms of the receptor (Harper et al., in preparation). Also, the absence of sodium molybdate greatly reduces the ligand-binding capability of the receptor during the time necessary for the analysis. When cytosolic LS180 AhR was separated on a Sephacryl S-300 column equilibrated in HEGDM buffer containing 0.1 M NaCl, a specific [3H]TCDD binding peak eluted with a Stokes radius of about 7 nm (Table I). The first peak corresponds to the void volume of the column and was nonspecific. The shoulder is the specific AhR . TCDD peak as determined by comparison of samples incubated with or without competitor (TCDF, Fig. 5, top).
o-2-
-O-
2 nM PH]TCDD
-.-
+l@TCDF
OI’...I....I,...I...“I.“..’ TOP 5 10
FRACTION
15
20
25
I I
NUMBER
FIG. 3. Demonstration of cytosolic receptor and nuclear AhR- ligand complex from cells incubated in culture with [3H]TCDD. LS180 cells were incubated in culture for 2 h at 37°C with 2 nM [3H]TCDD. Cytosol prepared in HEGDM and a 0.5 M KC1 extract of whole nuclei were analyzed by velocity sedimentation analysis on sucrose gradients. Top, cytosol and nuclear extract from the same cells incubated in culture with [3H]TCDD. Bottom, nuclear extract from cells incubated with 2 nM [sH]TCDD in the presence or absence of 1 pM TCDF.
activity after induction by TCDD was 3.1 pmol/min/mg cell protein for LS180 cells compared with 7.0 pmol/min/ mg cell protein for Hepa-l cells. In addition, the absolute AHH activity after induction by BA was 2.3 pmol/min/ mg cell protein for LS180 cells compared with 5.8 pmol/ min/mg cell protein for Hepa-l cells. The most notable major difference between the human LS180 cells and rodent Hepa-l cells was in the EC& value (i.e., the concentration at which the agonist induces the half-maximal response). In Hepa-l cells the EC& for AHH induction by TCDD was 1.2 X 10-l’ M, whereas in LS180 cells the EC& was 4 X 10P1’ M, approximately a 1 log unit shift to the right (Fig. 4). A similar shift to the right of the ECsO is also seen for AHH induction by BA (3 X 10e7 M as compared to 1.5 X 10m5M, Fig. 4) and with DB[a,h]A (2 X 10m6 M as compared to 1 X lOPa M, data not shown).
UNTREATED
LOG,,
-13
-12
-11
-10
-9
LSI 80
Hepa-l
rn
-8
-6
-7
INDUCER CONCENTRATION
-5
-4
(M)
FIG. 4. Dose-response curves for AHH induction by TCDD and BA. LS180 or Hepa-l cells were incubated for 24 h with the concentrations of [3H]TCDD or BA indicated on the abscissa. AHH activity was determined by the method of Nebert and Gelboin (30). The maximum activities attained for induction in LS180 cells were 3.07 for TCDD and 2.25 pmol for BA, units are 30H-BP/min/mg whole cell protein. The maximum activities attained for induction in Hepa-l cells were 7.0 for TCDD and 5.8 for BA.
Ah RECEPTOR
IN HUMAN
LS180 CELLS
33
300 in the presence of 20 mM sodium molybdate and 100 NaCl sedimented at about 9 S when reanalyzed by sedimentation analysis, whereas the 7-nm nuclear form sedimented as a single peak at about 6 S on sucrose gradients containing 0.5 M NaCl (data not shown). Since sequential analysis confirms that the 9.8 S form of the cytosolic receptor corresponds to the 7.1-nm form detected by gel permeation chromatography, and that the 6.2 S form of the nuclear receptor detected by sucrose gradient analysis corresponds to the 6.8-nm form, the Stokes radii and sedimentation coefficients were used to calculate relative molecular mass (M,), frictional ratios, and axial ratios for both cytosolic and nuclear forms of the human AhR. Several determinations were carried out on cytosols and nuclear extracts from different cell preparations. The molecular parameters obtained for LS180 cytosolic and nuclear receptor are summarized in Table II. The molecular mass for the nuclear receptor (176,000) is significantly different from the molecular mass of the cytosolic receptor (M, = 285,000). mM
FRACTION NUMBER FIG. 5. Analysis of LS180 AhR by gel permeation chromatography: Comparison of cytosolic and nuclear fractions. Top, LS180 cytosol was incubated with 20 nM [3H]TCDD in the absence (0) or presence(0) of a loo-fold excess of nonradioactive TCDF. After treatment with dextrancoated charcoal the samples were applied to Sephacryl S-300 columns equilibrated with EDGDM buffer containing 100 mM NaCl, then eluted as described under Materials and Methods. Bottom, LS180 nuclear extracts labeled with [3H]TCDD (0) or [3H]TCDD in the presence of a loo-fold excess of nonradiolabeled TCDF (0) were prepared in HDK buffer as described under Materials and Methods. Aliquots were applied to Sephacryl S-300 columns that had been equilibrated in HEGD buffer containing 500 mM NaCl, then eluted as described under Materials and Methods. Arrows at the top represent elution positions of the following standard proteins (from left to right): thyroglobulin (8.8 nm), ferritin (6.1 nm), catalase (5.2 nm), bovine serum albumin (3.6 nm), ovalbumin (2.7 nm), and cytochrome c (1.8 nm).
Separation of [3H]TCDD-labeled nuclear extract on a Sephacryl S-300 column equilibrated in HEGD containing 0.5 M NaCl resulted in the elution of a single symmetrical [3H]TCDD-binding peak with a Stokes radius of about 7 nm (Fig. 5, bottom). In this instance the radioactivity in the void volume was much reduced since no additional [3H]TCDD had been added to the sample. To confirm that the forms of the receptor observed by gel permeation chromatography correspond to the forms observed in sucrose gradients, specific fractions eluted from Sephacryl S-300 were transferred back onto sucrose gradients for reanalysis of the sedimentation coefficients. The 7-nm cytosolic form which eluted from Sephacryl S-
Photoafinity Labeling of the Ligand-Binding Subunit Cytosolic and nuclear extracts, photoaffinity labeled with [3H]TCDD, were analyzed by SDS-PAGE. The M, for nuclear AhR appears to be the same as that of the cytosolic receptor under denaturing conditions. A single radiolabeled peak was observed at about M, 110,000, as compared to radiolabeled standard proteins run in adjacent lanes. Comparison of [3H]TCDD-labeled cytosolic AhR and [3H]TCDD-labeled nuclear AhR analyzed in adjacent lanes, show a similar degree of migration (Fig. 6). Further, a 1:l mix of photoaffinity-labeled cytosol and photoaffinity-labeled nuclear extract analyzed by SDSPAGE resulted in the detection of a single band of radioactivity. Photoaffinity labeling of cytosolic AhR with [3H]TCDD in the presence of increasing concentrations of TCDF resulted in a concentration-dependent decrease in the radiolabeled signal (data not shown). Similarly, photoaffinity labeling of nuclear AhR by [3H]TCDD was reduced in nuclear extracts from cells treated in culture with [3H]TCDD in the presence of a loo-fold excess nonradiolabeled TCDD. Using 7.5% SDS-PAGE followed by fluorography the M, for the ligand-binding subunit of AhR from mouse Hepa-l cells was compared to the M, for the ligand-binding subunit of AhR from human LS180 cells. A M, of 90,000 to 95,000 for the cytosolic ligand-binding subunit from Hepa-l cells has been previously reported (35, 36). From our data, AhR from Hepa-l cells has a M, of 94,200 + 1900 (n = 6), distinctly different from 110,300 + 3200 (n = 7) observed for AhR from LS180 cells (Fig. 7). Thus, the LS180 cytosolic and nuclear multimeric forms of AhR contain a common ligand-binding subunit of M, N 110,000 which is distinctly larger than the ligandbinding subunit observed for AhR from Hepa-l cell cytoso1s.
34
HARPER TABLE
ET AL. II
Molecular Characteristics of Ah Receptor from Human Cell Line LS180
Cytosolic AhR (low ionic strength) Nuclear AhR (high ionic strength)
f/f0
Axial ratio (prolate ellipsoid)
6,600 (4)
1.62 f 0.03 (5)
11.6 + 0.6 (5)
176,200 f 11,000 (7)
1.88 + 0.05 (7)
17.3 f 1.1 (7)
%WV,
R. (nm)
9.8 + 0.23 (6)
7.1 It 0.15 (5)
285,500 f
6.2 f 0.39 (7)
6.9 f 0.15 (7)
Note. Sedimentation coefficients were determined Sephacryl S-300 columns as described under Materials indicate the number of replicate determinations.
M,
by velocity sedimentation on sucrose gradients and Stokes radii by chromatography on and Methods. Values are expressed as means f standard deviation. Numbers in parentheses
DISCUSSION The LS180 cell line has proven to be a good model system for investigating the events which occur during AhR-mediated cytochrome P450IAl induction in human cells. In many respects the mechanism of AhR-mediated induction is similar to that already described for the rodent receptor. The cytosolic AhR sediments as a 9.8 S complex on sucrose gradients in low ionic strength. A 6.2 S receptor. ligand complex can be identified in nuclear extracts from cells exposed to [3H]TCDD in culture. Exposure of LS180 cells to AhR agonists such as TCDD, DB[u,h]A, or BA results in the induction of AHH activity. It is interesting that in vitro [3H]TCDD and [3H]MC have similar binding affinities (apparent) and detect similar numbers of binding sites, yet the nonhalogenated hydrocarbons are markedly less effective as inducers of AHH activity in culture or in uiuo. The reason for this difference is unknown. Thus, this human cell line has a complete regulatory mechanism for AHH induction. Unlike other human cell lines previously studied, the amount of detectable receptor approaches that found in the rodent cell line, Hepa-1. Because of the relatively high concentration of Ah receptor in the LS180 cells we were able to characterize the sedimentation coefficients and stokes radii of both cytosolic and nuclear human Ah receptor to obtain a reliable estimate of their relative molecular mass by hydrodynamic analysis. Also, the receptor is sufficiently stable to allow photoaffinity labeling with [3H]TCDD of the ligandbinding subunit of either cytosolic or nuclear receptor and further analysis by SDS-PAGE. The human receptor has previously been characterized primarily by density gradient centrifugation. Various investigators using cells from several different tissues have reported a range of values. Harris et al. (19) reported a value of 9.7 to 10.4 S for cytosolic receptor from human breast cell lines, similar to that previously reported for the skin cell line A431 (17). More recently there have been several descriptions of receptor from the human liver cell lines HepG2 and HEPB3 (18, 22) where cytosolic receptor had a sedimentation value of 8 to 9 S. Our anal-
ysis of the sedimentation coefficient for AhR from the colon cell line LS180 is in the same range as previously reported values for other human cells and tissues. We find a sedimentation value of 9.8 S for cytosolic receptor as compared to 6.2 S for nuclear receptor. These results are similar to those reported by Prokipcak and Okey (31) for AhR from Hepa-l cells. There have been no previous estimates of the Stokes radii for either cytosolic or nuclear multimeric forms of human AhR. Receptor from human tissues or cell lines has proven to be difficult to work with since the levels are relatively low and the receptor is unstable over prolonged manipulation in vitro. Our data indicate that the LS180 cell line has abundant AhR with a B,,, of about 600 fmol/mg cytosolic protein for TCDD or MC, a level comparable to that detected in the widely studied mouse Hepa-l cell line. Additionally, AhR in LS180 cells is sufficiently stable in the presence of 20 mM sodium molybdate to withstand the longer times necessary for S-300 gel permeation chromatography or photoaffinity labeling with [3H]TCDD. In the present work we calculated Stokes radii of about 7 nm for both cytosolic and nuclear receptors. Sequential analysis from the S-300 column back to sucrose density gradient, however, confirmed that the nuclear receptor differs from the cytosolic receptor despite the similarities in Stokes radii. These results are similar to that previously reported for rodent receptor (27, 31). Recently it has been possible to identify the smallest ligand-binding subunit of the AhR. Poland and coworkers (35, 36), using 2-azido-3-[1251]iodo-7,8-dibromodibenzo-pdioxin to photoaffinity label the cytosolic AhR, reported a M, for the ligand-binding subunit as determined by SDS-PAGE of 95-110 kDa (the size depends on the species). In particular the M, for human AhR (from HeLa cells and a lymphoblast cell line) was found to be 104,000 as compared to 95,000 for AhR from Hepa-l cells. Similarly, Landers et al. (37) using [3H]TCDD reported a molecular mass of about 91,000 for mouse Hepa nuclear receptor, and Wang et al. (38) using [3H]TCDD or [125117-iodo-2,3-dibromodibenzo-p-dioxin have reported a mo-
Ah RECEPTOR
IN HUMAN
LS180
35
CELLS
lecular mass of about 97,000 for AhR in Hepa-l and 110,000 for AhR in HepG2 (human hepatoma cell line). The data presented here are in good agreement with these previous reports. Using [3H]TCDD to photoaffinity label the AhR, we find that the M, for the cytosolic ligandbinding subunit of the AhR from human is 110,000 and from Hepa-l is 94,000. In addition, we have been able to label and identify nuclear receptor from human LS180 cells. Our data indicate that the ligand-binding subunits from multimeric nuclear and cytosolic receptors have an identical molecular mass as determined by SDS-PAGE. These results are similar to those recently reported for Hepa- 1 by Prokipcak and Okey (33) who showed that the mass of the ligand-binding subunits from cytosolic and nuclear Hepa-l AhR is 94,200. From the data presented here and elsewhere it would appear that although the human AhR is more fragile to
100 92.5
69
46
w-+ LS180 (human)
Hepa-l (mouse)
FIG. 7. Comparison of photoaffinity-labeled Hepa-l cytosolic LS180 cytosolic AhRs. Cytosol from Hepa-l cells or LS180 cells photoaffinity-labeled with [3H]TCDD in the absence or presence loo-fold excess of TCDF. Samples were analyzed by SDS-PAGE fluorography as described under Materials and Methods.
69
46
and were of a and
experimental manipulation, its mode of action and basic physicochemical properties are very similar to those that have been well described for rodent AhR. In this regard the LS180 cell line should prove to be a useful experimental system for examining human AhR since this cell line has abundant, functional AhR which mediates the induction of cytochrome P450IAl and is sufficiently stable to withstand extensive experimental manipulation. ACKNOWLEDGMENTS This work was supported by a grant from the National Cancer Institute of Canada to A.B.0 and P.A.H. We thank Dr. Gerald Batist, Montreal General Hospital, for suggesting the use of LS180 cells.
FIG. 6. Comparison of M, for cytosolic and nuclear forms of AhR from LS180 cells. LS180 cytosol was labeled with 10 nM [3H]TCDD in vitro for 2 h, at 4°C. Following dextran-coated charcoal treatment the sample was brought to a final concentration of 0.5 M NaCl. Nuclear extracts were prepared from cells incubated in culture with 2 nM [3H]TCDD. Labeled cytosol and nuclear extract were irradiated and analyzed as described under Materials and Methods. Lanes 1, left and right: molecular weight markers; an aliquot of [3H]TCDD photoaffinitylabeled cytosolic protein; a 1:l mix of [3H]TCDD photoaffinity-labeled cytosolic proteins and [sH]TCDD photoaffinity-labeled nuclear proteins; an aliquot of [aH]TCDD photoaffinity-labeled nuclear protein.
REFERENCES 1. Poland, A., Glover, E., and Kende, A. S. (1976) J. Biol. Chem. 251, 4936-4946. 2. Okey, A. B., Bondy, G. P., Mason, M. E., Kahl, G. F., Eisen, H. J., Guenthner, T. M., and Nebert, D. W. (1979) J. Biol. Chem. 256, 11,636-11,648. 3. Poland, A., and Knutson, J. C. (1982) Annu. Reu. Pharmol. 22,517-554.
Toricol.
36
HARPER
4. Safe, S. H. (1986) Annu. Reu. Pharmacol. ToricoZ. 26, 371-399. 5. Hannah, R. R., Lund, J., Poellinger, L., Gillner, M., and Gustafsson, J.-A. (1986) Eur. J. Biochem. 156, 237-242.
6. Whitlock,
J. P., Jr. (1986) Annu. Reu. Pharmacol.
Tozicol. 26,333-
369. 7. Whitlock,
J. P., Jr. (1987) Pharmacol.
Reu. 39, 1477161.
8. Okey, A. B., Bondy, G. P., Mason, M. E., Kahl, G. F., Eisen, H. J., Guenthner, T. M., and Nebert, D. W. (1980) J. Biol. Chem. 255, 11,415-11,422. 9. Gasiewicz, T., and Bauman, P. A. (1987) J. Biol. Chem. 262,21162120. 10. Denison, M. S., Fisher, J. M., and Whitlock, Chem. 263, 17,221-17,224. 11. Denison, M. S., Fisher, J. M., and Whitlock, Natl. Acad. Sci. USA 85, 2528-2535.
J. P., Jr. (1988) J. Biol. J. P., Jr. (1988) Proc.
12. Fujisawa-Sehara, A., Yamane, M., and Fujii-Kuriyama, Proc. Natl. Acad. Sci. USA 65, 5859-5863.
Y. (1988)
252. J. P., Jr. (1990) Annu. Reu. Pharmacol.
Toxicol. 30,251-
277. 15. Okey, A. B. (1990) Phurmacol. Ther. 45, 241-298. 16. Manchester, D. K., Golas, C. L., Roberts, E. A., and Okey, A. B. (1987) Cancer Res. 47,4861-4868. 17. Harper, P. A., Golas, C. L., and Okey, A. B. (1988) Cancer Res. 48, 2388-2395. 18. Labruzzo, P., Yu, X. F., and Dufresne, M. J. (1989) Biochem. Pharmacol. 38, 2339-2348. 19. Harris, M., Piskorska-Pliszczynska, J., Zacharewski, M., and Safe, S. (1989) Cancer Res 49,4531-4535. 20. Gillner, M., Haldosen, L.-A., Gustafsson, J.-A. (1989) Toricol. Lett. 47, 41-51.
21. Cook, J. C., and Greenlee, W. F. (1989) Mol. Pharmacol. 35, 713719. 22. Roberts, E. A., Johnson, K. C., Harper, P. A., and Okey, A. B. (1990) Arch. Biophys. Biochem. 276,442-450. 23. Waithe, W. I., Michaud, M., Harper, P. A., Okey, A. B., and Anderson, A. (1990) Biochem. Pharmucol. 41,85-92. 24. Poland, A., and Glover, E. (1975) Mol. Pharmucol. 11,389-398. 25. Bernhard, H. P., Darlington, G. J., and Ruddle, F. H. (1974) Deu. Biol. 35, 83-96. 26. Bradford, M. M. (1976) Anal. Biochem. 72,248-254. 27. Denison, M. S., Vella, L. M., and Okey, A. B. (1986) J. Biol. Ckm.
261,3987-3995. 28. Tsui, H. W., and Okey, A. B. (1981) Can. J. Physiol. Pharmacol. 59,927-931. 29. Denison, M. S., Vella, L. M., and Okey, A. B. (1986) J. Biol. Chem. 261, 10,189-10,195. 30. Nebert, D. W., and Gelboin, H. V. (1968) J. Biol. Chem. 243,6242-
6247.
13. Nebert, D. W., and Jones, J. E. (1989) Znt. J. Biochem. 21, 24314. Whitlock,
ET AL.
T., Romkes,
S. A., and Gustafsson,
31. Prokipcak,
R. D., and Okey, A. B. (1988) Arch. Biochem. Biophys.
267,811-826. 32. Mantle, T. J. (1978) in Techniques
in Protein and Enzyme Biochemistry (Kornberg, H. L., Metcalfe, J. C., Northcote, D. H., Rogson, C. J., and Tipton, K. F., Eds.), pp. 1-17, Elsevier-North Holland Biomedical, Amsterdam 33. Prokipcak, R. D., and Okey, A. B. (1990) Biochem. Biophys. Res. Commun. 172,698-704. 34. Laemmli, U. K. (1970) Nature 227, 680-685. 35. Poland, A., and Glover, E. (1987) Biochem. Biophys. Res. Commun.
146,1439-1449. 36. Poland, A., and Glover, E. (1990) Mol. Pharmucol. 38, 306-312. 37. Landers, J. P., Piskorska-Pliszczynska, J., Zacharewski, T., Bunce, N. J., and Safe, S. (1989) J. Biol. Chem. 264, 18,463-18,471. 38. Wang, X., Narasimhan, T. R., Morrison, V., and Safe, S. (1991) Arch. Biochem. Biophys. 287, 186-194.
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 290, No. 1, October, pp. 27-36, 1991
Detection and Characterization of the Ah Receptor for 2,3,7,8-Tetrachlorodibenzo-p-dioxin in the Human Colon Adenocarcinoma Cell Line LSI 80 Patricia
A. Harper,
Rebecca D. Prokipcak,
Leslie
E. Bush,
Cheryl
L. Golas, and Allan
B. Okey’
Division of Clinical Pharmacology and Toxicology, Research Institute, The Hospital for Sick Children and Departments of Pediatrics and Pharmacology, University of Toronto, Toronto, Ontario. Canada M5S IA8
Received March 12, 1991, and in revised form June 1, 1991
The Ah (aromatic hydrocarbon) receptor mediates induction of aryl hydrocarbon hydroxylase (AHH; an enzyme activity associated with cytochrome P450IAl) by polycyclic aromatic hydrocarbon carcinogens such as 3methylcholanthrene (MC) and benzo[a]pyrene (BP) and the halogenated toxin 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD). Until recently the AhR seemed to be present only at very low levels in human cells and tissue. With a modified assay (the presence of sodium molybdate and a reduction in the amount of charcoal used to adsorb “excess” ligand) we found that cytosol from LS180 cells contains a high concentration of AhR (400-500 fmol/mg cytosolic protein) when detected by [3H]TCDD or [3H]MC. Cytosolic receptor also was detected with [3H]BP but at a level that was 35% of that detected with [3H]TCDD or [3H]MC. These levels are similar to those found in mouse Hepa-l hepatoma cells in which AhR has been extensively characterized. The apparent binding affinity (&) of the cytosolic receptor for [3H]TCDD and for [3H]MC was about 5 nM. As with Hepa-1, the human LS180 cytosolic AhR sedimented at about 9 S on sucrose gradients when detected with [3H]TCDD, [3H]BP or [3H]MC. The nuclear-associated ligand * receptor complex recovered from cells incubated in culture with [3H]TCDD sedimented at about 6.2 S. The 9.8 S cytosolic form corresponds to a multimeric protein of a relative molecular mass (M,) of about 285,000 whereas the 6.2 S nuclear receptor corresponds to a multimeric protein of M,. 175,000. The smallest specific ligand-binding subunit (detected by sodium dodecyl sulfate-polyacrylamide electrophoresis under denaturing conditions of receptor photoaffinity labeled with [3H]TCDD) was about M, 110,000. AHH activity was induced in cells exposed in culture to TCDD or benz[a]anthracene (BA). The EC&,
’ To whom correspondence should be addressed at the Department of Pharmacology. Fax (416) 978-6395. 0003.9861/91 $3.00 Copyright GJ 1991 by Academic Press, All rights of reproduction in any form
was 4 X lo-” M for TCDD and 1.5 X lo-’ M for BA. For both inducers the ECSO in LSl80 cells was shifted about one log unit to the right as compared to the ECeO for AHH induction in mouse Hepa-l cells. The lower sensitivity of the LS180 cells to induction of AHH activity by TCDD or BA is consistent with the lower affinity of TCDD and MC for binding to human AhR. The ligand-binding properties, physicochemical properties, and mode of action of the AhR in this human cell line are therefore very similar to those of the extensively characterized AhR in 8 1991 Academic press, I~C. rodent cells and tissues.
The Ah (aromatic hydrocarbon)’ receptor mediates induction of aryl hydrocarbon hydroxylase (AHH: P450IAl) by polycyclic aromatic hydrocarbons and has been extensively characterized in rodent cells and tissues (l-7). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and related halogenated and nonhalogenated hydrocarbons bind to the AhR in the cytoplasm. The receptor - ligand complex is then “transformed” to a nuclear-associated protein which binds to specific regulatory regions of DNA and increases synthesis of P450IAl mRNA, among other responses (8-15). Recently the AhR has been described in several human cell lines and tissues (16-23). With improved techniques for detection and quantitation, the human AhR appears to function in a manner similar to the AhR in rodent cells and tissues. However, the human AhR has appeared to * Abbreviations used: Ah, aromatic hydrocarbon; AHH, aryl hydrocarbon hydroxylase; AhR, aromatic hydrocarbon receptor; BA, benz[a]anthracene; BP, benzo[a]pyrene; BSA, bovine serum albumin; CAT, catalase; DB[a,h]A, dibenz[a,h]anthracene; DMSO, dimethyl sulfoxide; MC, 3-methylcholanthrene; MEM, minimal essential medium; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; TCDF, 2,3,7,8-tetrachlorodibenzofuran. 27
Inc. reserved.
28
HARPER
be much less stable than the rodent receptor, making it difficult to do a full characterization of the human AhR’s physicochemical properties. In this paper we present our observations on human AhR detected in the colon carcinoma cell line LS180. With suitable assay conditions an abundant specific binding protein is readily identifiable. The AhR detected in LS180 cells is sufficiently stable to allow more extensive characterization by both gel permeation chromatography and SDS-PAGE
MATERIALS
electrophoresis AND
of photoaffinity-labeled
AhR.
METHODS
Chemicals. [$H]TCDD (35 Ci/mmol) and nonradioactive TCDF were generous gifts from Dr. S. Safe (Texas A & M University). [3H]Triamcinolone acetonide (33 Ci/mmol) and [3H]benzo[a]pyrene (96 Ci/mmol) were obtained from DuPont Canada-NEN Research products (Lachine, Quebec). [3H]MC (generally labeled, 37 Ci/mmol) was from Amersham Corp. (Oakville, Ontario). TCDD and TCDF are extremely toxic substances and should be handled with care, as described by Poland and Glover (24). Sephacryl S 300 was from Pharmacia. All SDS-polyacrylamide gel supplies were from Bio-Rad. Sucrose (density gradient grade) was from Beckman Instruments (Toronto, Ontario); 4-(-hydroxyethyl)-l-piperazineethanesulfonic acid was from CalbiochemBehring Corp. (La Jolla, CA); nonradioactive BA, BP, DB[a,h]A, and MC were obtained from Aldrich Chemical Co. (Milwaukee, WI); molybdate (sodium salt), dithiothreitol, dexamethasone, bovine serum albumin, and catalase were obtained from Sigma Chemical Co. (St Louis, MO); dimethyl sulfoxide, glycerol, charcoal (Norit A), ethylenediaminetetraacetic acid, and all other routine chemicals were obtained from Fisher Scientific Co. (Toronto, Ontario). Buffers. The main buffer for cytosol preparation was HEGDM (25 mM 4-(-hydroxyethyl)-l-piperazineethanesulfonic acid, 3 mM ethylenediaminetetraacetic acid, 2 mM dithiothreitol, 10% glycerol (v/v), and 20 mM sodium molybdate). HEGDMK buffer is HEGDM with the addition of 0.5 M potassium chloride. HEDM is a hypotonic buffer consisting of HEDGM but without the glycerol; HEPGMD is HEDM plus 20% v/v glycerol. For preparation of some nuclear extracts cells were homogenized in MDH buffer (3 mM magnesium chloride, 1 mM dithiothreitol, and 25 mM Hepes) and nuclei were extracted with HDK (25 mM Hepes, 1 mM dithiothreitol, and 0.5 M NaCl). All buffers were adjusted to pH 7.4 at room temperature. Hepa-l is a derivative of a transplanted Cell lines and cell culture. hepatoma, originating in a C57L/J mouse (8, 25). The subclone, Hepalclc9, used in the present study was isolated and kindly provided by Dr. M. Dufresne (University of Windsor, Windsor, Ontario). Throughout the remainder of the paper the term Hepa-l refers to the subclone Hepalclc9 cells. The LS180 cell line, derived from an adenocarcinoma of the colon, was obtained from the American Type Culture Collection (Rockville, Maryland). Each cell line was grown as a monolayer culture in a-minimum essential medium ((u-MEM) supplemented with 10% fetal calf serum. The cells were routinely harvested by trypsinization when the density on the culture dish reached 1.5 to 3 X lo5 cells per centimeter squared for Hepa-l cells or when cells appeared confluent in the case of the LS180 cells and then washed in Dulbecco’s phosphate-buffered saline before homogenization. Cell fractionation. Prewashed Hepa-l cells were resuspended in the hypotonic buffer HEDM at about 2 to 4 X 10s cells per milliliter. As LS180 cells in suspension are extremely self-adherent, cell number could not be determined; therefore, the cells from 10 petri dishes (10 cm diameter) were resuspended in 1 ml of HEDM buffer. Each cell line was homogenized with a Polytron fitted with a PT 7 generator (Brinkmann Instruments, Rexdale, Ontario) until >90% of the cells were fractured but the nuclei were still intact (as determined by phase contrast mi-
ET AL. croscopy). The homogenate was immediately diluted 1:l with HE2GDM buffer and centrifuged at 105,OOg for 1 h. The resulting supernatant fraction (cytosol) was stored in l- to 2-ml aliquots in liquid nitrogen until analysis. Protein concentrations were determined by the method of Bradford (26) using BSA as the standard. Preparation of nuclear extract. Nuclear extracts were prepared from cells exposed to [3H]TCDD in culture; cytosols also were obtained from these cells. The cell homogenate was centrifuged for 4 min at 10,OOOg in a Beckman microfuge and the resulting pellet was washed two to three times with HEGDM buffer, incubated for 1 h with HEGDK buffer at 4°C to extract the nuclear receptor, and then centrifuged at 105,OOOg for 1 h. The resulting supernatant fraction is termed the nuclear extract. Nuclear extracts for gel permeation chromatography and photoaffinity labeling were prepared essentially as described by Denison et al. (11). Briefly, cells were incubated with 2 nM [aH]TCDD in a-MEM for 2 h at 37“C, washed, and then incubated for 15 min in 10 mM Hepes, pH 7.5. The cells were scraped off the dish, collected, and homogenized in MDH buffer using a Dounce homogenizer with a close-fitting pestle. The homogenate was centrifuged at 1OOOgfor 5 min and the resulting pellet was washed two to three times with MDH buffer containing 0.1 M NaCl and incubated for 45 min with HDK buffer at 4°C to extract the nuclear receptor; glycerol was added to a final concentration of 10% and the mixture was then centrifuged at 105,OOOgfor 1 h. The resulting supernatant fraction is termed the nuclear extract. Preparation of cytosol for analysis by velocity sedimentation and Sephacryl S-300 anulysis. The procedure used was essentially that of Denison et al. (27). Briefly, cytosols to be analyzed for AhR were incubated with [3H]TCDD (or other [3H]ligand as indicated in the text and figure legends), in the absence or presence of a loo-fold molar excess of the competitor TCDF (unless noted otherwise in the text). Samples analyzed for glucocorticoid receptor were incubated with [aH]triamcinolone acetonide with or without the competitor dexamethasone. All incubations were for 2 h at 4°C. After incubation unbound radioligand was removed by adding the cytosols to a pellet of dextran-coated charcoal; 0.1 mg charcoal per milligram cytosolic protein was used for samples to be analyzed on sucrose gradients, or 0.5 mg/ml for samples to be analyzed on Sephacryl S-300. The charcoal-to-cytosol ratio was increased for samples to be analyzed on Sephacryl S-300 to reduce the free [3H]TCDD in the void volume; further increase in the charcoal-to-cytosol ratio resulted in the loss of [sH]TCDD. AhR. Cytosol was incubated with dextran-coated charcoal for 15 min at 4°C; charcoal was removed by centrifugation at 4000g. Nuclear extracts obtained from cells labeled in culture with [3H]TCDD were not treated with dextran-coated charcoal; these samples were applied directly to sucrose gradients or to Sephacryl S-300 columns. Samples were analyzed Velocity sedimentation on sucrosegradients. by density gradient centrifugation using a vertical tube rotor technique as described by Tsui and Okey (28) and Harper et al. (17). [“‘ClFormaldehyde-labeled BSA (4.4 S) and [‘4C]formaldehyde-labeled catalase (11.3 S) were included in each gradient as internal sedimentation markers. Marker proteins were labeled with [‘*C]formaldehyde as previously described (29). Monolayer cells Induction and assay of aryl hydrocarbon hydroxylase. (subconfluent) were treated with various concentrations of [3H]TCDD or BA dissolved in dimethyl sulfoxide. The final concentration of DMSO was 1% in the culture medium in all samples. This concentration of DMSO was not toxic and did not increase the AHH activity above the basal noninduced level. Radiolabeled TCDD was used because of the extremely low concentrations of TCDD which were sufficient to induce AHH activity; the concentrations were verified by determining the radioactivity in each dilution and, where possible, in the incubating media. After 24 h cells were harvested by trypsinization and washed two to three times with phosphate-buffered saline. AHH activity in whole cells was determined by the method of Nebert and Gelboin (30) with cell protein levels of 0.75 to 1.0 mg per assay flask incubated for 30 min at 37’C. There was no fluorescence contribution at any concentration of [3H]TCDD, BA, or DB[a,h]A, since parallel induced samples assayed
Ah RECEPTOR
IN HUMAN
in the absence of the substrate BP gave values that were essentially zero. One unit of AHH activity is defined as the amount of enzyme catalyzing in I min at 37°C the formation of hydroxylatedproduct causing fluorescence equivalent to that of 1 pmol of 3-hydroxybenzo[a]pyrene (recrystallized) standard. Activity is expressed as pmol per minute per milligram cell protein. Protein concentrations for each sample were determined by the method of Bradford (26). Gel permeation chromatography Gel permeation chromatography. was done essentially as described by Denison et al. (29) and Prokipcak and Okey (31). Sephacryl S-300 was equilibrated in HEGDM or HEGD buffer containing 0.5 M NaCl and poured into 1.6 X loo-cm glass columns. Standards used to calibrate the gel permeation columns and their stokes radii were thyroglobulin (8.5 nm), ferritin (6.1 nm), catalase (5.2 nm), bovine serum albumin (3.6 nm), ovalbumin (2.7 nm), and cytochrome c (1.8 nm). The Stokes radius for catalase was obtained from a published table (32), all others were calculated as described by Denison et al. (29). The applied sample size was about 12 mg of cytosolic protein or about 5 mg nuclear extract protein. Distribution coefficients (Ko), Stokes radius (R,), sedimentation coefficient (S xO,w), relative molecular mass (M,), frictional ratio (f/s), and axial ratio (a/b) of a hydrodynamically equivalent prolate ellipsoid were calculated as described by Denison et al. (27). For each sample the measured S 20,wand R, were used to calculate M,, f/f,,, and a/b. These values, derived separately for each individual sample, were then used to calculate the mean and standard deviation for the group of samples within a replicate series. Photoafinity labeling with [3H]TCDD and separation by SDS-PAGE Cytosol and nuclear exof the AhR from cytosol and nuclear extracts. tracts from LS180 cells were photoaffinity labeled with [3H]TCDD essentially as described by Prokipcak and Okey (33). Briefly [aH]TCDDlabeled cytosol prepared in HEDGM buffer (1 to 1.5 ml at -3 mg protein/ ml; 300-400 fmol specific binding/mg protein) and nuclear extracts prepared in HDK buffer (1 to 1.5 ml at 1 to 2 mg protein; 1006150 fmol specific binding/mg protein) were irradiated on ice for 15-20 min, 11 cm from an 85-W lamp (Johns Scientific) or for 2 min, 2 cm from a 450-W lamp (Hanovia Ace Glass). After irradiation proteins were precipitated in cold acetone (-20’(Z), collected by centrifugation, then dissolved directly in sample buffer. After boiling for 2 to 4 min samples and standard molecular mass markers were analyzed by electrophoresis on discontinuous polyacrylamide gels by the method of Laemmli (34). i4C-labeled standards (Amersham) included myosin (200 kDa), phosphorylase b (doublet of 100 kDa and 92.5 kDa), bovine serum albumin (66 kDa), ovalbumin (43 kDa), and carbonic anhydrase (30 kDa). Fluorography. Fixed gels were impregnated with Enlightening (DuPont), dried, and placed in contact with Kodak X-OMAT film for 6 weeks to 5 months before developing (33). RESULTS
LS180
29
CELLS 20
1
TCDD + -o-
-o-.-
I
24 nM [JH]MC +2pMTCDF
I
30 Oi
“.
19 nhl[3H]TCDD +2pMTCDF
‘v’
.
‘v’
.
.
-&
17 nM [3H]BP
f
+2pMTCDF
.
25
FIG. 1. Binding of [3H]TCDD, [3H]MC, and [aH]BP to the AhR. Cytosol from LS180 cells prepared in HEDGM buffer was incubated with 20 nM [3H]TCDD, 20 nM [3H]MC, or 20 nM [aH]BP in the absence or presence of a loo-fold molar excess of the competitor TCDF. Samples were analyzed by velocity sedimentation on sucrose gradients. [i4C]formaldehyde-labeled BSA (4.4 S) and [‘%]formaldehyde-labeled catalase (11.3 S) were added to the gradients as internal sedimentation markers (arrowheads).
Cytosolic Receptor The AhR can be detected by sucrose density gradient analysis of cytosolic samples that have been incubated with [3H]TCDD. Specific binding of the ligand to the AhR is determined by the comparison of profiles from samples incubated in the presence of a loo-fold excess of a specific competitor such as TCDF with binding in the absence of competitor. Characteristically, a specific peak of radioactivity is detectable in the 9 S region of the sucrose gradient (approximately at fraction numbers 10-15). Several ligands known to be agonists for the AhR were tested for their ability to bind to the LS180 AhR. As shown in Fig. 1 incubation of LS180 cytosol with [3H]TCDD, [“H]MC, or [3H]BP resulted in a specific binding peaks in the 9 S region of the sucrose gradient for each ligand;
each of these peaks was eliminated by the presence of loo-fold excess TCDF in the incubation. When rodent cytosols are incubated with [3H]MC and analyzed by a sucrose density gradient there is a specific binding peak in the 4 to 5 S region of the gradient in addition to the 9 S peak. In LS180 cells specific binding was detected with [3H]MC only in the 9 S region of the gradient. When the same cytosol preparation was analyzed using all three ligands the concentration of specific binding sites for [3H]TCDD (355 fmol/mgprotein) and [3H]MC (376 fmol/ mg protein) were very similar, whereas the concentration of binding sites detected with [3H]BP was lower (121 fmol/ mg protein). The amount of AhR detected by [3H]TCDD in LS180 cytosol (300 to 400 fmol/mg protein) was com-
30
HARPER
0
10
20
40
30
ET AL.
50
0
20
10
FREE [3H]TCDD (nM)
30
0.8 -
B map 611 fmol/mg
250
B map 548 fmol/mg K,= 4.6 nM r = 0.98
0.0 liLL 0
500
[3H]TCDD BOUND (fmollmg
50
0.8
0.6
0
40
FREE [3H]MC (nM)
protein)
250
500
[3H]MC BOUND (fmollmg
protein)
15
B map 549 fmol/mg
5-
K,= 5.2 nM r= 1.0
-1
1
3
5
[3H]TCDD UNBOUND (fmoUmg protein, thousands)
7
I 1. -1
1
I 3
.
I 5
.
I 7
’
[3H]MC UNBOUND (fmoVmg protein, thousands)
FIG. 2. Saturation analysis and determination of apparent affinity of binding of [“HITCDD and [3H]MC in cytosols from LS180 cells. Cytosol aliquots (about 4 mg protein/ml) prepared in HEDGM buffer were incubated at 4°C for 2 h with [3H]TCDD (left) or [3H]MC (right) at concentrations ranging from 2 to 40 nM. Specific binding in the 9 S peak was determined for each sample by sucrose density gradient analysis. The Scatchard plots (middle) and Woolf plots (bottom) were derived from data shown in the saturation plots (top). Binding parameters shown for the Scatchard and Woolf plots were calculated by least squares linear regression.
Ah RECEPTOR
IN HUMAN TABLE
Determination
of the Apparent
Affinity
31
LS180 CELLS
I
of [ 3H]TCDD
and [ 3H]MC
for Cytosolic
[3H]MCb
[sH]TCDD”
K+ (nM) B,,,
(fmol/mg
protein)
AhR
Woolf
Scatchard
Woolf
Scatchard
5.2 + 0.83' 554 f108
5.6 f 0.5 551 f108
5.3 +- 3.3 458 f191
5.8 zk 2.6 455 +170
Note. Scatchard and Woolf plots were derived from saturation binding data as shown in Fig. 2. Binding parameters and Woolf plots were calculated by least squares linear regression. u Number of replicates = 5. * Number of replicates = 3. ’ Mean + S.D.
parable to that detected in the rodent cell line Hepa-l (300 to 600 fmol/mg protein). Incubating LS180 cytosol in the presence of [3H]TCDD and loo-fold excess of benz[a]anthracene, benzo[a]pyrene, 3-methylcholanthrene, or dibenz[a,h]anthracene inhibited [3H]TCDD binding to the AhR, whereas phenobarbital, dexamethasone, and diethylstilbestrol had no effect on binding of [3H]TCDD in the 9 S region (data not shown). When LS180 cytosol was incubated with [3H]triamcinolone acetonide (a glucocorticoid analogue), very little specific binding to the glucocorticoid receptor was observed (about 15 fmol/mg cytosolic protein) as compared with [3H]TCDD binding to the AhR of 350 fmol/mg cytosolic protein in the same cytosol (data not shown). In previous studies on human tissues it has been necessary to add sodium molybdate to the homogenizing buffers during cytosol preparation in order to stabilize AhR. A comparison of cytosols prepared from LS180 cells in the presence and absence of sodium molybdate showed no such dependence for the initial detection of receptor. The amount of receptor detected in cytosols with or without molybdate was essentially the same. Although the presence of molybdate did not increase the amount of receptor detected, sodium molybdate did stabilize the receptor during long-term manipulation (>12 h, data not shown) and therefore was routinely added unless otherwise noted. Estimation of Binding Affinity of [3Hj TCDD and [3H]MC for AhR in LS180 Cell Cytosol The apparent Kd for the interaction of [3H]TCDD or [3H]MC with the AhR was estimated by incubating cytosols with a wide range of radioligand concentrations (1 to 40 nM), then determining the specific binding in the 9 S region by sucrose gradient analysis for each sample. As illustrated in Fig. 2, about 20 nM TCDD or MC was necessary to saturate the receptor. Binding data were further analyzed by Scatchard and Woolf plots (Fig. 2 middle and bottom). The kinetic parameters Kd and B,,,
shown for the Scatchard
computed from both plots are shown in Table I. The presence of sodium molybdate in the homogenizing buffer had no effect on the apparent Kd and B,,, (data not shown). Identification
of a Nuclear Receptor- Ligand Complex
The binding of a ligand such as TCDD to the cytosolic receptor results in transformation of the ligand . receptor complex into a nuclear binding protein. This nuclear-associated receptor complex can be extracted from cell nuclei by high salt concentrations and identified by sucrose gradient analysis with no further incubation with [3H]TCDD. Figure 3 shows the sucrose gradient profiles obtained from nuclear and cytosolic fractions of LS180 cells incubated in suspension culture with [3H]TCDD prior to cell fractionation in HEGDM. The nuclear-associated ligand . receptor complex has a lower sedimentation value of 6.2 S (this work and Ref. (31)) as compared to the cytosolic receptor complex which sediments at about 9 S (Fig. 3, top). The 6.2 S binding peak is specific since nuclear extracts from cells incubated with [3H]TCDD in the presence of a loo-fold excess of the competitor TCDF in the medium show no binding in this region of the sucrose gradient (Fig. 3, bottom). AHH Induction
by [3H]TCDD
and BA
Aryl hydrocarbon hydroxylase is one of the best characterized enzymatic responses associated with AhR-mediated cytochrome P450IAl induction. If the AhR is fully functional, exposure of cells to an AhR agonist in culture should result in the induction of AHH activity. LS180 and Hepa-l cells were incubated with various concentrations of [3H]TCDD or BA for 24 h then harvested and the AHH activity in each sample was determined. Since TCDD is active at very low concentrations, [3H]TCDD was used so that the inducer concentration in the culture medium could be accurately determined. The concentration dependency for AHH induction by TCDD and BA is shown in Fig. 4. The absolute AHH
32
HARPER 4
1
BSA v
3
ET AL.
Velocity Sedimentation and Gel Permeation Chromatography on Sephacryl S-300 of Cytosolic and Nuclear Forms of the AhR
CAT
V -0
NUCLEAR
-D
CYTOSOL
2
0.8 -
0.6 -
The physicochemical properties of cytosolic AhR were analyzed in greater detail by performing replicate analyses on several cytosol preparations. Sodium molybdate (20 mM final concentration) was added to stabilize the ligand . AhR complex since inclusion of 0.1 M NaCl in the column buffers in the absence of sodium molybdate results in a mixture of 9 S and 6 S forms of the receptor (Harper et al., in preparation). Also, the absence of sodium molybdate greatly reduces the ligand-binding capability of the receptor during the time necessary for the analysis. When cytosolic LS180 AhR was separated on a Sephacryl S-300 column equilibrated in HEGDM buffer containing 0.1 M NaCl, a specific [3H]TCDD binding peak eluted with a Stokes radius of about 7 nm (Table I). The first peak corresponds to the void volume of the column and was nonspecific. The shoulder is the specific AhR . TCDD peak as determined by comparison of samples incubated with or without competitor (TCDF, Fig. 5, top).
o-2-
-O-
2 nM PH]TCDD
-.-
+l@TCDF
OI’...I....I,...I...“I.“..’ TOP 5 10
FRACTION
15
20
25
I I
NUMBER
FIG. 3. Demonstration of cytosolic receptor and nuclear AhR- ligand complex from cells incubated in culture with [3H]TCDD. LS180 cells were incubated in culture for 2 h at 37°C with 2 nM [3H]TCDD. Cytosol prepared in HEGDM and a 0.5 M KC1 extract of whole nuclei were analyzed by velocity sedimentation analysis on sucrose gradients. Top, cytosol and nuclear extract from the same cells incubated in culture with [3H]TCDD. Bottom, nuclear extract from cells incubated with 2 nM [sH]TCDD in the presence or absence of 1 pM TCDF.
activity after induction by TCDD was 3.1 pmol/min/mg cell protein for LS180 cells compared with 7.0 pmol/min/ mg cell protein for Hepa-l cells. In addition, the absolute AHH activity after induction by BA was 2.3 pmol/min/ mg cell protein for LS180 cells compared with 5.8 pmol/ min/mg cell protein for Hepa-l cells. The most notable major difference between the human LS180 cells and rodent Hepa-l cells was in the EC& value (i.e., the concentration at which the agonist induces the half-maximal response). In Hepa-l cells the EC& for AHH induction by TCDD was 1.2 X 10-l’ M, whereas in LS180 cells the EC& was 4 X 10P1’ M, approximately a 1 log unit shift to the right (Fig. 4). A similar shift to the right of the ECsO is also seen for AHH induction by BA (3 X 10e7 M as compared to 1.5 X 10m5M, Fig. 4) and with DB[a,h]A (2 X 10m6 M as compared to 1 X lOPa M, data not shown).
UNTREATED
LOG,,
-13
-12
-11
-10
-9
LSI 80
Hepa-l
rn
-8
-6
-7
INDUCER CONCENTRATION
-5
-4
(M)
FIG. 4. Dose-response curves for AHH induction by TCDD and BA. LS180 or Hepa-l cells were incubated for 24 h with the concentrations of [3H]TCDD or BA indicated on the abscissa. AHH activity was determined by the method of Nebert and Gelboin (30). The maximum activities attained for induction in LS180 cells were 3.07 for TCDD and 2.25 pmol for BA, units are 30H-BP/min/mg whole cell protein. The maximum activities attained for induction in Hepa-l cells were 7.0 for TCDD and 5.8 for BA.
Ah RECEPTOR
IN HUMAN
LS180 CELLS
33
300 in the presence of 20 mM sodium molybdate and 100 NaCl sedimented at about 9 S when reanalyzed by sedimentation analysis, whereas the 7-nm nuclear form sedimented as a single peak at about 6 S on sucrose gradients containing 0.5 M NaCl (data not shown). Since sequential analysis confirms that the 9.8 S form of the cytosolic receptor corresponds to the 7.1-nm form detected by gel permeation chromatography, and that the 6.2 S form of the nuclear receptor detected by sucrose gradient analysis corresponds to the 6.8-nm form, the Stokes radii and sedimentation coefficients were used to calculate relative molecular mass (M,), frictional ratios, and axial ratios for both cytosolic and nuclear forms of the human AhR. Several determinations were carried out on cytosols and nuclear extracts from different cell preparations. The molecular parameters obtained for LS180 cytosolic and nuclear receptor are summarized in Table II. The molecular mass for the nuclear receptor (176,000) is significantly different from the molecular mass of the cytosolic receptor (M, = 285,000). mM
FRACTION NUMBER FIG. 5. Analysis of LS180 AhR by gel permeation chromatography: Comparison of cytosolic and nuclear fractions. Top, LS180 cytosol was incubated with 20 nM [3H]TCDD in the absence (0) or presence(0) of a loo-fold excess of nonradioactive TCDF. After treatment with dextrancoated charcoal the samples were applied to Sephacryl S-300 columns equilibrated with EDGDM buffer containing 100 mM NaCl, then eluted as described under Materials and Methods. Bottom, LS180 nuclear extracts labeled with [3H]TCDD (0) or [3H]TCDD in the presence of a loo-fold excess of nonradiolabeled TCDF (0) were prepared in HDK buffer as described under Materials and Methods. Aliquots were applied to Sephacryl S-300 columns that had been equilibrated in HEGD buffer containing 500 mM NaCl, then eluted as described under Materials and Methods. Arrows at the top represent elution positions of the following standard proteins (from left to right): thyroglobulin (8.8 nm), ferritin (6.1 nm), catalase (5.2 nm), bovine serum albumin (3.6 nm), ovalbumin (2.7 nm), and cytochrome c (1.8 nm).
Separation of [3H]TCDD-labeled nuclear extract on a Sephacryl S-300 column equilibrated in HEGD containing 0.5 M NaCl resulted in the elution of a single symmetrical [3H]TCDD-binding peak with a Stokes radius of about 7 nm (Fig. 5, bottom). In this instance the radioactivity in the void volume was much reduced since no additional [3H]TCDD had been added to the sample. To confirm that the forms of the receptor observed by gel permeation chromatography correspond to the forms observed in sucrose gradients, specific fractions eluted from Sephacryl S-300 were transferred back onto sucrose gradients for reanalysis of the sedimentation coefficients. The 7-nm cytosolic form which eluted from Sephacryl S-
Photoafinity Labeling of the Ligand-Binding Subunit Cytosolic and nuclear extracts, photoaffinity labeled with [3H]TCDD, were analyzed by SDS-PAGE. The M, for nuclear AhR appears to be the same as that of the cytosolic receptor under denaturing conditions. A single radiolabeled peak was observed at about M, 110,000, as compared to radiolabeled standard proteins run in adjacent lanes. Comparison of [3H]TCDD-labeled cytosolic AhR and [3H]TCDD-labeled nuclear AhR analyzed in adjacent lanes, show a similar degree of migration (Fig. 6). Further, a 1:l mix of photoaffinity-labeled cytosol and photoaffinity-labeled nuclear extract analyzed by SDSPAGE resulted in the detection of a single band of radioactivity. Photoaffinity labeling of cytosolic AhR with [3H]TCDD in the presence of increasing concentrations of TCDF resulted in a concentration-dependent decrease in the radiolabeled signal (data not shown). Similarly, photoaffinity labeling of nuclear AhR by [3H]TCDD was reduced in nuclear extracts from cells treated in culture with [3H]TCDD in the presence of a loo-fold excess nonradiolabeled TCDD. Using 7.5% SDS-PAGE followed by fluorography the M, for the ligand-binding subunit of AhR from mouse Hepa-l cells was compared to the M, for the ligand-binding subunit of AhR from human LS180 cells. A M, of 90,000 to 95,000 for the cytosolic ligand-binding subunit from Hepa-l cells has been previously reported (35, 36). From our data, AhR from Hepa-l cells has a M, of 94,200 + 1900 (n = 6), distinctly different from 110,300 + 3200 (n = 7) observed for AhR from LS180 cells (Fig. 7). Thus, the LS180 cytosolic and nuclear multimeric forms of AhR contain a common ligand-binding subunit of M, N 110,000 which is distinctly larger than the ligandbinding subunit observed for AhR from Hepa-l cell cytoso1s.
34
HARPER TABLE
ET AL. II
Molecular Characteristics of Ah Receptor from Human Cell Line LS180
Cytosolic AhR (low ionic strength) Nuclear AhR (high ionic strength)
f/f0
Axial ratio (prolate ellipsoid)
6,600 (4)
1.62 f 0.03 (5)
11.6 + 0.6 (5)
176,200 f 11,000 (7)
1.88 + 0.05 (7)
17.3 f 1.1 (7)
%WV,
R. (nm)
9.8 + 0.23 (6)
7.1 It 0.15 (5)
285,500 f
6.2 f 0.39 (7)
6.9 f 0.15 (7)
Note. Sedimentation coefficients were determined Sephacryl S-300 columns as described under Materials indicate the number of replicate determinations.
M,
by velocity sedimentation on sucrose gradients and Stokes radii by chromatography on and Methods. Values are expressed as means f standard deviation. Numbers in parentheses
DISCUSSION The LS180 cell line has proven to be a good model system for investigating the events which occur during AhR-mediated cytochrome P450IAl induction in human cells. In many respects the mechanism of AhR-mediated induction is similar to that already described for the rodent receptor. The cytosolic AhR sediments as a 9.8 S complex on sucrose gradients in low ionic strength. A 6.2 S receptor. ligand complex can be identified in nuclear extracts from cells exposed to [3H]TCDD in culture. Exposure of LS180 cells to AhR agonists such as TCDD, DB[u,h]A, or BA results in the induction of AHH activity. It is interesting that in vitro [3H]TCDD and [3H]MC have similar binding affinities (apparent) and detect similar numbers of binding sites, yet the nonhalogenated hydrocarbons are markedly less effective as inducers of AHH activity in culture or in uiuo. The reason for this difference is unknown. Thus, this human cell line has a complete regulatory mechanism for AHH induction. Unlike other human cell lines previously studied, the amount of detectable receptor approaches that found in the rodent cell line, Hepa-1. Because of the relatively high concentration of Ah receptor in the LS180 cells we were able to characterize the sedimentation coefficients and stokes radii of both cytosolic and nuclear human Ah receptor to obtain a reliable estimate of their relative molecular mass by hydrodynamic analysis. Also, the receptor is sufficiently stable to allow photoaffinity labeling with [3H]TCDD of the ligandbinding subunit of either cytosolic or nuclear receptor and further analysis by SDS-PAGE. The human receptor has previously been characterized primarily by density gradient centrifugation. Various investigators using cells from several different tissues have reported a range of values. Harris et al. (19) reported a value of 9.7 to 10.4 S for cytosolic receptor from human breast cell lines, similar to that previously reported for the skin cell line A431 (17). More recently there have been several descriptions of receptor from the human liver cell lines HepG2 and HEPB3 (18, 22) where cytosolic receptor had a sedimentation value of 8 to 9 S. Our anal-
ysis of the sedimentation coefficient for AhR from the colon cell line LS180 is in the same range as previously reported values for other human cells and tissues. We find a sedimentation value of 9.8 S for cytosolic receptor as compared to 6.2 S for nuclear receptor. These results are similar to those reported by Prokipcak and Okey (31) for AhR from Hepa-l cells. There have been no previous estimates of the Stokes radii for either cytosolic or nuclear multimeric forms of human AhR. Receptor from human tissues or cell lines has proven to be difficult to work with since the levels are relatively low and the receptor is unstable over prolonged manipulation in vitro. Our data indicate that the LS180 cell line has abundant AhR with a B,,, of about 600 fmol/mg cytosolic protein for TCDD or MC, a level comparable to that detected in the widely studied mouse Hepa-l cell line. Additionally, AhR in LS180 cells is sufficiently stable in the presence of 20 mM sodium molybdate to withstand the longer times necessary for S-300 gel permeation chromatography or photoaffinity labeling with [3H]TCDD. In the present work we calculated Stokes radii of about 7 nm for both cytosolic and nuclear receptors. Sequential analysis from the S-300 column back to sucrose density gradient, however, confirmed that the nuclear receptor differs from the cytosolic receptor despite the similarities in Stokes radii. These results are similar to that previously reported for rodent receptor (27, 31). Recently it has been possible to identify the smallest ligand-binding subunit of the AhR. Poland and coworkers (35, 36), using 2-azido-3-[1251]iodo-7,8-dibromodibenzo-pdioxin to photoaffinity label the cytosolic AhR, reported a M, for the ligand-binding subunit as determined by SDS-PAGE of 95-110 kDa (the size depends on the species). In particular the M, for human AhR (from HeLa cells and a lymphoblast cell line) was found to be 104,000 as compared to 95,000 for AhR from Hepa-l cells. Similarly, Landers et al. (37) using [3H]TCDD reported a molecular mass of about 91,000 for mouse Hepa nuclear receptor, and Wang et al. (38) using [3H]TCDD or [125117-iodo-2,3-dibromodibenzo-p-dioxin have reported a mo-
Ah RECEPTOR
IN HUMAN
LS180
35
CELLS
lecular mass of about 97,000 for AhR in Hepa-l and 110,000 for AhR in HepG2 (human hepatoma cell line). The data presented here are in good agreement with these previous reports. Using [3H]TCDD to photoaffinity label the AhR, we find that the M, for the cytosolic ligandbinding subunit of the AhR from human is 110,000 and from Hepa-l is 94,000. In addition, we have been able to label and identify nuclear receptor from human LS180 cells. Our data indicate that the ligand-binding subunits from multimeric nuclear and cytosolic receptors have an identical molecular mass as determined by SDS-PAGE. These results are similar to those recently reported for Hepa- 1 by Prokipcak and Okey (33) who showed that the mass of the ligand-binding subunits from cytosolic and nuclear Hepa-l AhR is 94,200. From the data presented here and elsewhere it would appear that although the human AhR is more fragile to
100 92.5
69
46
w-+ LS180 (human)
Hepa-l (mouse)
FIG. 7. Comparison of photoaffinity-labeled Hepa-l cytosolic LS180 cytosolic AhRs. Cytosol from Hepa-l cells or LS180 cells photoaffinity-labeled with [3H]TCDD in the absence or presence loo-fold excess of TCDF. Samples were analyzed by SDS-PAGE fluorography as described under Materials and Methods.
69
46
and were of a and
experimental manipulation, its mode of action and basic physicochemical properties are very similar to those that have been well described for rodent AhR. In this regard the LS180 cell line should prove to be a useful experimental system for examining human AhR since this cell line has abundant, functional AhR which mediates the induction of cytochrome P450IAl and is sufficiently stable to withstand extensive experimental manipulation. ACKNOWLEDGMENTS This work was supported by a grant from the National Cancer Institute of Canada to A.B.0 and P.A.H. We thank Dr. Gerald Batist, Montreal General Hospital, for suggesting the use of LS180 cells.
FIG. 6. Comparison of M, for cytosolic and nuclear forms of AhR from LS180 cells. LS180 cytosol was labeled with 10 nM [3H]TCDD in vitro for 2 h, at 4°C. Following dextran-coated charcoal treatment the sample was brought to a final concentration of 0.5 M NaCl. Nuclear extracts were prepared from cells incubated in culture with 2 nM [3H]TCDD. Labeled cytosol and nuclear extract were irradiated and analyzed as described under Materials and Methods. Lanes 1, left and right: molecular weight markers; an aliquot of [3H]TCDD photoaffinitylabeled cytosolic protein; a 1:l mix of [3H]TCDD photoaffinity-labeled cytosolic proteins and [sH]TCDD photoaffinity-labeled nuclear proteins; an aliquot of [aH]TCDD photoaffinity-labeled nuclear protein.
REFERENCES 1. Poland, A., Glover, E., and Kende, A. S. (1976) J. Biol. Chem. 251, 4936-4946. 2. Okey, A. B., Bondy, G. P., Mason, M. E., Kahl, G. F., Eisen, H. J., Guenthner, T. M., and Nebert, D. W. (1979) J. Biol. Chem. 256, 11,636-11,648. 3. Poland, A., and Knutson, J. C. (1982) Annu. Reu. Pharmol. 22,517-554.
Toricol.
36
HARPER
4. Safe, S. H. (1986) Annu. Reu. Pharmacol. ToricoZ. 26, 371-399. 5. Hannah, R. R., Lund, J., Poellinger, L., Gillner, M., and Gustafsson, J.-A. (1986) Eur. J. Biochem. 156, 237-242.
6. Whitlock,
J. P., Jr. (1986) Annu. Reu. Pharmacol.
Tozicol. 26,333-
369. 7. Whitlock,
J. P., Jr. (1987) Pharmacol.
Reu. 39, 1477161.
8. Okey, A. B., Bondy, G. P., Mason, M. E., Kahl, G. F., Eisen, H. J., Guenthner, T. M., and Nebert, D. W. (1980) J. Biol. Chem. 255, 11,415-11,422. 9. Gasiewicz, T., and Bauman, P. A. (1987) J. Biol. Chem. 262,21162120. 10. Denison, M. S., Fisher, J. M., and Whitlock, Chem. 263, 17,221-17,224. 11. Denison, M. S., Fisher, J. M., and Whitlock, Natl. Acad. Sci. USA 85, 2528-2535.
J. P., Jr. (1988) J. Biol. J. P., Jr. (1988) Proc.
12. Fujisawa-Sehara, A., Yamane, M., and Fujii-Kuriyama, Proc. Natl. Acad. Sci. USA 65, 5859-5863.
Y. (1988)
252. J. P., Jr. (1990) Annu. Reu. Pharmacol.
Toxicol. 30,251-
277. 15. Okey, A. B. (1990) Phurmacol. Ther. 45, 241-298. 16. Manchester, D. K., Golas, C. L., Roberts, E. A., and Okey, A. B. (1987) Cancer Res. 47,4861-4868. 17. Harper, P. A., Golas, C. L., and Okey, A. B. (1988) Cancer Res. 48, 2388-2395. 18. Labruzzo, P., Yu, X. F., and Dufresne, M. J. (1989) Biochem. Pharmacol. 38, 2339-2348. 19. Harris, M., Piskorska-Pliszczynska, J., Zacharewski, M., and Safe, S. (1989) Cancer Res 49,4531-4535. 20. Gillner, M., Haldosen, L.-A., Gustafsson, J.-A. (1989) Toricol. Lett. 47, 41-51.
21. Cook, J. C., and Greenlee, W. F. (1989) Mol. Pharmacol. 35, 713719. 22. Roberts, E. A., Johnson, K. C., Harper, P. A., and Okey, A. B. (1990) Arch. Biophys. Biochem. 276,442-450. 23. Waithe, W. I., Michaud, M., Harper, P. A., Okey, A. B., and Anderson, A. (1990) Biochem. Pharmucol. 41,85-92. 24. Poland, A., and Glover, E. (1975) Mol. Pharmucol. 11,389-398. 25. Bernhard, H. P., Darlington, G. J., and Ruddle, F. H. (1974) Deu. Biol. 35, 83-96. 26. Bradford, M. M. (1976) Anal. Biochem. 72,248-254. 27. Denison, M. S., Vella, L. M., and Okey, A. B. (1986) J. Biol. Ckm.
261,3987-3995. 28. Tsui, H. W., and Okey, A. B. (1981) Can. J. Physiol. Pharmacol. 59,927-931. 29. Denison, M. S., Vella, L. M., and Okey, A. B. (1986) J. Biol. Chem. 261, 10,189-10,195. 30. Nebert, D. W., and Gelboin, H. V. (1968) J. Biol. Chem. 243,6242-
6247.
13. Nebert, D. W., and Jones, J. E. (1989) Znt. J. Biochem. 21, 24314. Whitlock,
ET AL.
T., Romkes,
S. A., and Gustafsson,
31. Prokipcak,
R. D., and Okey, A. B. (1988) Arch. Biochem. Biophys.
267,811-826. 32. Mantle, T. J. (1978) in Techniques
in Protein and Enzyme Biochemistry (Kornberg, H. L., Metcalfe, J. C., Northcote, D. H., Rogson, C. J., and Tipton, K. F., Eds.), pp. 1-17, Elsevier-North Holland Biomedical, Amsterdam 33. Prokipcak, R. D., and Okey, A. B. (1990) Biochem. Biophys. Res. Commun. 172,698-704. 34. Laemmli, U. K. (1970) Nature 227, 680-685. 35. Poland, A., and Glover, E. (1987) Biochem. Biophys. Res. Commun.
146,1439-1449. 36. Poland, A., and Glover, E. (1990) Mol. Pharmucol. 38, 306-312. 37. Landers, J. P., Piskorska-Pliszczynska, J., Zacharewski, T., Bunce, N. J., and Safe, S. (1989) J. Biol. Chem. 264, 18,463-18,471. 38. Wang, X., Narasimhan, T. R., Morrison, V., and Safe, S. (1991) Arch. Biochem. Biophys. 287, 186-194.