DNA binding properties of dioxin receptors in wild-type and mutant mouse hepatoma cells

Analysis of the aryl hydrocarbon receptor (AhR) signal transduction pathway

Analysis of the Ah Receptor Signal Transduction Pathway (Michael S. Denison, Jane M. Rohers, S. Renee Rushing. Carol L. Jones, Selwyna C. Tetangico, and Sharon Heath-Pagliuso, University of California, Davis, California).The protocols in this unit will allow researchers to detect the Ah receptor and characterize its functional activities (i.e., ligand binding, transformation and DNA binding, and gene expression) in their biological test system and to use these methods to detect chemical and biochemical events that affect this signaling system.

Oxygen tension regulates the expression of a group of procollagen hydroxylases

In this study, we have characterized the influence of hypoxia on the expression of hydroxylases crucially involved in collagen fiber formation, such as prolyl-4-hydroxylases (Ph4) and procollagen lysyl-hydroxylases (PLOD). Using the rat vascular smooth muscle cell line A7r5, we found that an hypoxic atmosphere caused a characteristic time-dependent five- to 12-fold up-regulation of the mRNAs of the two P4h alpha-subunits [alphaI (P4ha1) and alphaII (P4ha2)] and of two lysylhydroxylases (PLOD1 and PLOD2). These effects of hypoxia were mimicked by the iron-chelator deferoxamine (100 micro m) and by cobaltous chloride (100 micro m). The hypoxic induction of these genes was also seen in the mouse juxtaglomerular As4.1 cell line and mouse hepatoma cell line Hepa1 but was almost absent in the mutant cell line Hepa1C4, which is defective for the hypoxia-inducible transcription factor 1 (HIF-1). In addition, the enzyme expression was induced by hypoxia in mouse embryonic fibroblasts but not in embryonic fibroblasts lacking the HIF-1alpha subunit. These findings indicate that hypoxia stimulates the gene expression of a cluster of hydroxylases that are indispensible for collagen fiber formation. Strong indirect evidence, moreover, suggests that the expression of these enzymes during hypoxia is coordinated by HIF-1.

The cellular oxygen tension regulates expression of the endoplasmic oxidoreductase ERO1-Lalpha

The formation of disulfide bonds in the endoplasmic reticulum requires protein disulfide isomerase (PDI) and endoplasmic reticulum oxidoreductin 1 (ERO1) that reoxidizes PDI. We report here that the expression of the rat, mouse and human homologues of ERO1-Like protein alpha but not of the isoform ERO1-Lbeta are stimulated by hypoxia in rats vivo and in rat, mouse and human cell cultures. The temporal pattern of hypoxic ERO1-Lalpha induction is very similar to that of genes triggered by the hypoxia inducible transcription factor (HIF-1) and is characteristically mimicked by cobalt and by deferoxamine, but is absent in cells with a defective aryl hydrocarbon receptor translocator (ARNT, HIF-1beta). We speculate from these findings that the expression of ERO1-Lalpha is probably regulated via the HIF-pathway and thus belongs to the family of classic oxygen regulated genes. Activation of the unfolded protein response (UPR) by tunicamycin, on the other hand, strongly induced ERO1-Lbeta and more moderately ERO1-Lalpha expression. The expression of the two ERO1-L isoforms therefore appears to be differently regulated, in the way that ERO1-Lalpha expression is mainly controlled by the cellular oxygen tension, whilst ERO1-Lbeta is triggered mainly by UPR. The physiological meaning of the oxygen regulation of ERO1-Lalpha expression likely is to maintain the transfer rate of oxidizing equivalents to PDI in situations of an altered cellular redox state induced by changes of the cellular oxygen tension.

Comparison of expression of aldehyde dehydrogenase 3 and CYP1A1 in dominant and recessive aryl hydrocarbon hydroxylase-deficient mutant mouse hepatoma cells

The mouse hepatoma cell line Hepa-1 is inducible by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) for both CYP1A1 (aryl hydrocarbon hydroxylase, AHH) and class 3 aldehyde dehydrogenase (ALDH3) enzymes. To test the hypothesis of a common regulatory mechanism, several AHH deficient mutants of Hepa-1 were studied for their ALDH3 activities and specific mRNA levels before and after TCDD treatment. The recessive (with respect to the wild-type Hepa-1) mutants have defects in Cypla-1 structural gene (mutant c1) or in the Ah (aryl hydrocarbon) receptor (mutants c2 and c6 with decreased levels of Ah receptor; mutant c4 defective in the DNA binding of the Ah receptor). The results with these mutants suggested that Ah receptor nuclear translocator protein, ARNT, is needed for ALDH3 expression. Two dominant mutants, one of which is characterized by preventing the binding of the Ah receptor complex to DNA, were also studied. Surprisingly, these mutants possessed elevated levels of ALDH3 mRNA and enzyme activities which were also inducible by TCDD. The binding of Ah receptor-ligand complex to DNA was thus not needed for the expression of ALDH3. A dominant repressor for Cypla-1 gene transcription did not prevent the derepression or induction of ALDH3. The results thus suggest that Aldh-3 gene is regulated by a mechanism independent of the Ah receptor.

Mechanism of benzo[a]pyrene-induced Cyp1a-1 gene expression in mouse Hepa 1c1c7 cells: role of the nuclear 6 s and 4 s proteins

Treatment of wild-type (wt) aryl hydrocarbon (Ah)-responsive mouse Hepa 1c1c7 cells with benzo[a]pyrene (B[a]P) caused a concentration-dependent induction of ethoxyresorufin O-deethylase (EROD) activity. In contrast, B[a]P was inactive as an inducer in Ah nonresponsive class 1 and class 2 mutant cell lines. In parallel experiments, the nuclear fractions from wt cells treated with 10(-7) M [3H]B[a]P contained both the 4 s carcinogen binding protein and the 6 s (Ah receptor) complexes, whereas only the 4 s complex was present in the nuclear fraction of the class 2 mutant cells. The results obtained from cotreatment of wt Hepa 1c1c7 cells with 10(-6) or 10(-7) M B[a]P and 5 x 10(-7) or 10(-7) M 6-methyl-1,3,8-trichlorodibenzofuran (MCDF) showed that MCDF inhibited the induction of EROD activity and Cyp1a-1 mRNA levels by B[a]P. Moreover, using 10(-7) M [3H]B[a]P and unlabeled MCDF, it was shown that MCDF not only inhibited the induction response but also caused a concentration-dependent decrease in levels of the nuclear 6 s complex but not the 4 s complex. In contrast, in situ competition studies with unlabeled 10(-6) M benzo[ghi]-perylene (B[ghi]P) resulted in the elimination of the nuclear [3H]B[a]P 4 s complex (but not the 6 s complex); however, the EROD activity and Cyp1a-1 mRNA levels in cells treated with 10(-7) M B[a]P in the presence or absence of 10(-6) M B[ghi]P were not significantly different. These results indicate that the 4 s binding protein is not required for the induction of Cyp1a-1 gene expression in Hepa 1c1c7 cells and suggest that B[a]P and 2,3,7,8-tetrachlorodibenzo-p-dioxin induce Cyp1a-1 gene expression via a common mechanism which involves binding to the Ah receptor.

Comparison of the nuclear and cytosolic forms of the Ah receptor from Hepa 1c1c7 cells: charge heterogeneity and ATP binding properties

2-[125I]iodo-7,8-dibromo-p-dioxin ([125I]Br2DpD) and 2-[125I]iodo-3-azido-7,8-dibromo-p-dioxin ([125I]N3Br2-DpD) are both capable of binding to the Ah receptor (AhR) with a high degree of specificity in cultured Hepa 1c1c7 cells. After incubation with either [125I]N3Br2DpD or [125I]Br2DpD Hepa 1c1c7 cytosolic and high salt nuclear extracts were analyzed by sucrose density gradient analysis with the following results: (i) With both radioligands an approximately 9 S form of the AhR was observed in cytosolic extracts. (ii) Nuclear extracts labeled with [125I]N3Br2DpD revealed both approximately 6 S and approximately 9 S forms of the AhR. (iii) In contrast, analysis of nuclear extracts labeled with [125I]Br2DpD revealed only an approximately 6 S form of the AhR. The approximately 9 S [125I]N3Br2DpD-labeled AhR was preferentially extracted with 100 mM KCl from a nuclear fraction and mixed with monoclonal antibody 8D3, an anti-90-kDa heat shock protein antibody. Monoclonal antibody 8D3 was able to bind to the approximately 9 S nuclear form of the AhR and caused the receptor to sediment as a heavier complex on sucrose density gradients. This would indicate that the AhR can reside in the nucleus bound to 90-kDa heat shock protein. The [125I]N3Br2DpD-labeled approximately 6 S peak fractions were collected and subjected to denaturing two-dimensional gel electrophoresis. A comparison of [125I]N3Br2DpD-labeled cytosolic (9 S) AhR preparations with the nuclear (6 S) AhR by 2-D gel electrophoresis was performed. The cytosolic form of the AhR was present in the apparent pI range of 5.2-5.7; the nuclear form focused between 5.5 and 6.2. The [125I]N3Br2DpD-labeled nuclear extracts were incubated with ATP-agarose and 43% of the photoaffinity-labeled AhR bound to the affinity gel. In contrast, approximately threefold lower binding of [125I]N3Br2DpD-labeled receptor was obtained when GTP-, AMP-, or ADP-agarose was used. Only 2% of the [125I]N3Br2DpD-labeled cytosolic AhR was able to bind to ATP-agarose. These results suggest that after the AhR translocates into the nucleus the following biochemical changes occur: (i) The sedimentation value for the AhR changes from an approximately 9 S to an approximately 6 S species. (ii) The AhR attains the ability to bind with specificity to ATP. (iii) The AhR undergoes a shift to a more basic pI.

DNA binding properties of the Ah receptor in wild-type and variant mouse hepatoma cells

Mouse hepatoma Hepa 1c1c7 cells and nonresponsive mutants have been extensively used as models for investigating the molecular mechanism of induction of CYP1A1 gene transcription by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Incubation of cytosolic [3H]TCDD-aryl hydrocarbon (Ah) receptor from wild-type Hepa 1c1c7 cells for 16 h at 4 degrees C in 0.4 M KCl resulted in the formation of transformed liganded receptor which exhibited increased binding affinity on DNA-Sepharose columns. The elution properties of the peak with the highest DNA binding affinity were similar to the elution profiles of the nuclear receptor complex isolated from wild-type cells. TAOc1BPrcl (class I) nonresponsive mutant cells were characterized by relatively low levels of the cytosolic and nuclear Ah receptor complex. The BPrcl (class II) variant cell line contained levels of cytosolic receptor which were comparable to those observed in the wild-type cells; however, significantly reduced levels of nuclear receptor complex were observed in the class II variant cell line. Incubation of the nuclear or transformed liganded cytosolic Ah receptor from wild-type cells with a consensus 32P-labeled dioxin responsive element (DRE) in a gel shift assay gave a retarded band associated with the receptor-DRE complex. Incubation of the cytosolic receptor complex from the class I and II mutant cells for 16 h at 4 degrees C in 0.4 M KCl or for 2 h at 20 degrees C did not yield complexes with increasing binding affinities on DNA-Sepharose columns. Moreover, incubation of these complexes with 32P-labeled DRE did not give a retarded band in a gel shift assay. However, coincubation of the liganded class II mutant cytosol with cytosol from class I cells resulted in transformation of the liganded receptor and this was confirmed in both the DNA-Sepharose and gel retardation assays. These results suggest that the failure of class II mutant cells to respond to TCDD is due to a defect in the factors responsible for transformation of the cytosolic receptor complex.

Liver cells contain constitutive DNase I-hypersensitive sites at the xenobiotic response elements 1 and 2 (XRE1 and -2) of the rat cytochrome P-450IA1 gene and a constitutive, nuclear XRE-binding factor that is distinct from the dioxin receptor

Dioxin stimulates transcription from the cytochrome P-450IA1 promoter by interaction with the intracellular dioxin receptor. Upon binding of ligand, the receptor is converted to a form which specifically interacts in vitro with two dioxin-responsive positive control elements located in close proximity to each other about 1 kb upstream of the rat cytochrome P-450IA1 gene transcription start point. In rat liver, the cytochrome P-450IA1 gene is marked at the chromatin level by two DNase I-hypersensitive sites that map to the location of the response elements and exist prior to induction of transcription by the dioxin receptor ligand beta-naphthoflavone. In addition, a DNase I-hypersensitive site is detected near the transcription initiation site and is altered in nuclease sensitivity by induction. The presence of the constitutive DNase I-hypersensitive sites at the dioxin response elements correlates with the presence of a constitutive, labile factor which specifically recognizes these elements in vitro. This factor appears to be distinct from the dioxin receptor, which is observed only in nuclear extract from treated cells. In conclusion, these data suggest that a certain protein-DNA architecture may be maintained at the response elements at different stages of gene expression.

Liver cells contain constitutive DNase I-hypersensitive sites at the xenobiotic response elements 1 and 2 (XRE1 and -2) of the rat cytochrome P-450IA1 gene and a constitutive, nuclear XRE-binding factor that is distinct from the dioxin receptor

Dioxin stimulates transcription from the cytochrome P-450IA1 promoter by interaction with the intracellular dioxin receptor. Upon binding of ligand, the receptor is converted to a form which specifically interacts in vitro with two dioxin-responsive positive control elements located in close proximity to each other about 1 kb upstream of the rat cytochrome P-450IA1 gene transcription start point. In rat liver, the cytochrome P-450IA1 gene is marked at the chromatin level by two DNase I-hypersensitive sites that map to the location of the response elements and exist prior to induction of transcription by the dioxin receptor ligand beta-naphthoflavone. In addition, a DNase I-hypersensitive site is detected near the transcription initiation site and is altered in nuclease sensitivity by induction. The presence of the constitutive DNase I-hypersensitive sites at the dioxin response elements correlates with the presence of a constitutive, labile factor which specifically recognizes these elements in vitro. This factor appears to be distinct from the dioxin receptor, which is observed only in nuclear extract from treated cells. In conclusion, these data suggest that a certain protein-DNA architecture may be maintained at the response elements at different stages of gene expression.

Role of the ligand in intracellular receptor function: receptor affinity determines activation in vitro of the latent dioxin receptor to a DNA-binding form

To reconstitute the molecular mechanisms underlying the cellular response to soluble receptor ligands, we have exploited a cell-free system that exhibits signal- (dioxin-)induced activation of the latent cytosolic dioxin receptor to an active DNA-binding species. The DNA-binding properties of the in vitro-activated form were qualitatively indistinguishable from those of in vivo-activated nuclear receptor extracted from dioxin-treated cells. In vitro activation of the receptor by dioxin was dose dependent and was mimicked by other dioxin receptor ligands in a manner that followed the rank order of their relative affinities for the receptor in vitro and their relative potencies to induce target gene transcription in vivo. Thus, in addition to triggering the initial release of inhibition of DNA binding and presumably allowing nuclear translocation, the ligand appears to play a crucial role in the direct control of the level of functional activity of a given ligand-receptor complex.

Mechanism of action of 2,3,7,8-tetrachlorodibenzo-p-dioxin antagonists: characterization of 6-[125I]methyl-8-iodo-1,3-dichlorodibenzofuran-Ah receptor complexes

6-Methyl-8-iodo-1,3,-dichlorodibenzofuran (I-MCDF) and its radiolabeled analog [125I]MCDF have been synthesized and used to investigate the mechanism of action of 1,3,6,8-substituted dibenzofurans as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) antagonists. Like 6-methyl-1,3,8-trichlorodibenzofuran (MCDF), I-MCDF partially antagonized the induction by TCDD of microsomal aryl hydrocarbon hydroxylase (AHH) and ethoxyresorufin O-deethylase (EROD) activities in rat hepatoma H-4-II E cells and male Long-Evans rat liver. Incubation of rat liver cytosol with [125I]MCDF followed by velocity sedimentation analysis on sucrose gradients gave a specifically bound peak which sedimented at 9.6 S. This radioactive peak was displaced by coincubation with a 200-fold excess of unlabeled I-MCDF, 6-methyl-1,3,8-trichlorodibenzofuran (MCDF), 2,3,7,8-tetrachlorodibenzofuran (TCDF), and benzo [a]pyrene. Based on the velocity sedimentation results and the elution profile from a Sephacryl S-300 gel permeation column, the Stokes radius and apparent molecular weights of the cytosolic [125I]MCDF-Ah receptor complex were 6.5 nm and 259,200, respectively. In addition, the nuclear [125I]MCDF-receptor complex eluted at a salt concentration of 0.29 M KCl from a DNA-Sepharose column. Velocity sediment analysis of the nuclear [125I]MCDF-Ah receptor complex from rat hepatoma H-4-II E cells gave a specifically bound peak at 5.6 +/- 0.8 S. All of these properties were similar to those observed using [3H]TCDD as the radioligand. In addition, there were several ligand-dependent differences observed in the properties of the I-MCDF and TCDD receptor complexes; for example, the [125I]MCDF rat cytosolic receptor complex was unstable in high salt buffer and was poorly transformed into a form with increased binding affinity on DNA-Sepharose columns; Scatchard plot analysis of the saturation binding of [3H]TCDD and [125I]MCDF with rat hepatic cytosol gave KD values of 1.07 and 0.13 nM and Bmax values of 137 and 2.05 fmol/mg protein, respectively. The nuclear extract from rat hepatoma H-4-II E cells treated with I-MCDF or TCDD interacted with a dioxin-responsive element in a gel retardation assay. These results suggest that the mechanism of antagonism may be associated with competition of the antagonist receptor complex for nuclear binding sites.

Role of the ligand in intracellular receptor function: receptor affinity determines activation in vitro of the latent dioxin receptor to a DNA-binding form

To reconstitute the molecular mechanisms underlying the cellular response to soluble receptor ligands, we have exploited a cell-free system that exhibits signal- (dioxin-)induced activation of the latent cytosolic dioxin receptor to an active DNA-binding species. The DNA-binding properties of the in vitro-activated form were qualitatively indistinguishable from those of in vivo-activated nuclear receptor extracted from dioxin-treated cells. In vitro activation of the receptor by dioxin was dose dependent and was mimicked by other dioxin receptor ligands in a manner that followed the rank order of their relative affinities for the receptor in vitro and their relative potencies to induce target gene transcription in vivo. Thus, in addition to triggering the initial release of inhibition of DNA binding and presumably allowing nuclear translocation, the ligand appears to play a crucial role in the direct control of the level of functional activity of a given ligand-receptor complex.

Regulation of polyamine biosynthesis by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)

We have examined the effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on tissue polyamine concentrations in CD1 mice. Two days after a low dose treatment with TCDD, polyamine content of the liver and thymus of treated mice showed a 49-82% decrease, but that of spleen was not affected. Based on this finding, we examined the role of alterations in polyamine levels in the toxicity of TCDD. We administered the polyamine biosynthetic inhibitor difluoromethylornithine (DFMO) to animals treated with a toxic dose of TCDD. DFMO dramatically increased the toxicity of TCDD as measured by mortality, ascites and changes in organ weights. In addition, administration of the polyamine putrescine was able to reduce the toxicity of TCDD. These results suggest that a decrease in polyamine concentrations in critical organs may play an important role in the toxic effects of TCDD.

Analysis of photoaffinity-labeled aryl hydrocarbon receptor heterogeneity by two-dimensional gel electrophoresis

The level of charge heterogeneity in the aryl hydrocarbon receptor (AhR) was examined by high-resolution denaturing two-dimensional (2D) gel electrophoresis. Hepa 1c1c7 cell cytosolic fraction was photoaffinity-labeled with 2-azido-3-[125I]iodo-7,8-dibromodibenzo-p-dioxin and applied to isoelectric focusing (IEF) tube gels. After optimization of focusing conditions a broad peak of radioactivity was detected in the apparent pI range of 5.2-5.7. IEF tube gels were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by visualization of the radiolabeled AhR by autoradiography; three distinct isoforms were detected. The same 2D electrophoretic isoform pattern was obtained when the AhR from Hepa 1c1c7 was photoaffinity-labeled in cell culture. BPrCl cells, a mutant line derived from Hepa 1c1c7 cells, contain an AhR that is unable to bind to DNA. Photoaffinity-labeled BPrCl cytosolic fractions were subjected to 2D gel electrophoretic analysis resulting in essentially the same molecular weight and isoform pattern as seen in Hepa 1c1c7 cytosol. This result would suggest that if a mutation is present in the BPrCl AhR it has not caused a significant change in its IEF pattern, although a small shift in the pI values was observed. Two-dimensional gel electrophoresis of photoaffinity-labeled cytosolic fractions from HeLa cells, the rat liver tumor cell line McA-RH7777, and buffalo rat thymus revealed three isoforms, essentially the same isoform pattern as in Hepa 1c1c7 cells. This would indicate that despite the considerable molecular weight polymorphism between species the level of charge heterogeneity is highly conserved.

Specific protein-DNA interactions at a xenobiotic-responsive element: copurification of dioxin receptor and DNA-binding activity

Upon binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin (called dioxin or TCDD), the dioxin receptor exhibits increased affinity for the cell nucleus in vivo and for DNA in vitro. To define the recognition sequence of the dioxin receptor and its relationship with that of the glucocorticoid receptor, oligonucleotides derived from dioxin-responsive elements of the rat cytochrome P-450c gene were tested for their ability to form specific protein-DNA complexes in a gel retardation assay. We found that a previously defined sequence motif that is similar to the glucocorticoid-responsive element and exhibits strong enhancer activity in response to dioxin receptor ligands bound a dioxin-inducible factor with high specificity but was not recognized by the DNA-binding domain of the glucocorticoid receptor. Binding to this element was only observed in nuclear extracts of wild-type mouse hepatoma cells in a time- and dose-dependent manner and not in nuclear extracts from a nonresponsive mutant cell line deficient in DNA binding of the dioxin receptor. The specific DNA-binding activity in wild-type nuclear extracts comigrated in a Superose size-exclusion column and cosedimented on sucrose gradients with the in vivo labeled dioxin receptor. These experiments strongly suggest that the dioxin receptor is a sequence-specific DNA-binding protein and is not only biochemically but also functionally similar to the steroid receptor family.

Ligand-dependent interaction of the dioxin receptor with target DNA

Wild type and nuclear transfer deficient mouse hepatoma cell lines were used to study the specific DNA binding of a dioxin inducible factor. This factor interacts with XRE only after dioxin treatment and is absent in receptor mutant containing cells even after treatment. Thus, evidence is provided to substantiate the claim that the dioxin receptor is involved in the specific DNA interaction with dioxin response enhancer elements. It is also shown that the molybdate stabilised dioxin-receptor interacts with hsp90 suggesting that, in similarity to the glucocorticoid receptor, the dioxin receptor is kept in a non-transformed state in the absence of ligand.