Autoregulation of human CYP1A1 gene promoter

Cardnogenesis vol.17 no.3 pp.435-441, 1996

Autoregulation of human CYP1A1 gene promoter activity in HepG2 and MCF-7 cells

Eva Cecilie Bonefeld J0rgensen' and Herman Autrup Department of Environmental and Occupational Medicine. University of Aarhus. Ole Worms Alle" 180. DK-8000 Aarhus C, Denmark 'To whom correspondence should be addressed

Introduction Cytochrome P450 is a superfamily of microsomal hemeproteins which catalyze a variety of biological oxidations of endogenous substrates such as steroids, fatty acids and prostaglandins (1,2). In addition, these monooxygenases play an important role in the metabolism of drugs and other foreign compounds, including activation of procarcinogens (3,4). The P4501A sub-family includes two genes, CYP1A1 and CYP1A2, known to be inducible by xenobiotic compounds such as polycyclic aromatic •Abbreviations: PAH, polycyclic aromatic hydrocarbons; TCDD, 23,7,8tetrachlorodiben7.o-/MJioxin; AHH, aryl hydrocarbon hydroxylase; AHR, aryl hydrocarbon receptor, bHLH, basic helix-loop-helix; XRE, xenobiotic response element; TF, transcription factor; ARNT, AHR nuclear translocator, BTE, basic transcriptional element; NRE, negative regulatory element; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal bovine serum; DMSO, dimethylsulfoxide; CAT, chloramphenicol acetyltransferase; (i-gal. $galactosidase; GAPDH, glyceraldehyde phosphate dehydrogenase; BTEB, BTE binding protein; GRE, glucocorticoid-responsive element. © Oxford University Press

435

Downloaded from http://carcin.oxfordjournals.org/ at Statsbiblioteket on January 2, 2015

Cytochrome CYP1A1 gene expression, induced by polycyclic aromatic hydrocarbons and dioxins, e.g. 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD), is regulated mainly at the level of transcription. Inducible activation of the CYP1A1 promoter is mediated by a ligand-dependent transcription factor dimer complex including the aryl hydrocarbon receptor (AHR) and the AHR nuclear translocator (ARNT) proteins. Additional factors seem to be involved in tissue- and cell-specific modification of the induction process. In the present study HepG2 and MCF7 cell lines were used to examine a possible cell-specific autoregulation of CYP1A1 promotor function. Chimeric CYP1A1-CAT reporter constructs and a human CYP1A1 cDNA expression plasmid were used in transient co-expression experiments. In HepG2 cells co-expression of increasing amounts of CYP1A1 cDNA significantly down-regulated constitutive as well as the TCDD-induced CYP1A1 promotor driven CAT activity. In contrast, co-transfection of MCF-7 cells with a 3-fold molar excess of CYP1A1 cDNA relative to the CYP1A1-CAT reporter construct caused an -2-fold increase in the TCDD-induced CAT activity, whereas no effect was observed on constitutive promotor activity. This autoregulatory mechanism(s) of the human CYP1A1 gene product was independent of specific 5' flanking promotor segments tested. RT-PCR analyses did not indicate any changes in mRNA level of AHR and ARNT in the co-transfection studies. Thus these studies show that the human CYP1A1 gene is exposed to cell-specific autoregulation, probably achieved via different functions of trans-acting factors.

hydrocarbons (PAH*) and dioxins such as 2,3,7,8-tetrachlorodibenzo-/j-dioxin (TCDD). The mechanism of PAH- and dioxin-mediated induction of the CYP1A family has been studied extensively, with the emphasis on CYP1A1. Understanding regulation of the CYP1A1 gene, encoding the enzyme aryl hydrocarbon hydroxylase (AHH), has been of particular concern in chemical carcinogenesis, since P4501A1 is the most active form in activating PAH to highly carcinogenic metabolites (5,6). Large interindividual and tissue variations in CYP1A1 inducibility have been observed in the human population (7-10) and a number of studies have shown a correlation between the level of AHH and susceptibility to chemically induced cancer, especially lung carcinoma (11,12). Moreover, a positive correlation between moderate to high constitutive CYP1A1 expression and poor prognosis in human breast cancer has been reported (13,14). A variety of toxic, teratogenic and carcinogenic responses in PAH- or TCDD-exposed animals are mediated by a cytosolic receptor, the aryl hydrocarbon receptor (AHR) (15,16), a member of the basic helix-loop-helix (bHLH) family of DNA binding proteins (2,17,18). As a ligand-activated transcription factor AHR activates a large number of genes (2,15,19-21), including the drug metabolizing phase I and phase II enzymes. Recently construction and characterization of an AHR-deficient mouse line showed that AHR plays an important role in expression of dioxin-inducible genes and in normal development of the liver and immune system (22). Since CYP1A1 is the most thoroughly studied of the AHRregulated genes, it serves as a useful system for studying the molecular mechanisms through which the receptor acts. Transcriptional activation is initiated upon inducer binding to AHR, dissociation of hsp90 (heat shock protein), receptor activation/transformation, nuclear translocation and interaction with specific positive c/s-acting DNA response elements, termed xenobiotic response elements (XREs) or dioxin response elements. The activated DNA binding transcription factor (TF) complex is a heterodimer of two bHLH proteins, AHR and the AHR nuclear translocator (ARNT) (17,18,23). Tissue- and cell-specific factors seem to influence the induction process, such as protein kinase C-dependent phosphorylation of the TF complex (24,25), AHR level/affinity and CYP1A1 expression (16,26,27). XREs have also been shown to interact with non-receptor molecules involved in induction and/or basal gene activity (28,29). A basic transcriptional element (BTE) in the upstream region was shown to be required for maximal induction of rat CYPJA1 (30). In addition, other m-acting promotor sequences have been identified as mediating negative regulation of mouse cyplal (18) and hamster (31) and human CYP1A1 (32-34) and an autoregulatory effect has been suggested for endogenous mouse CYP1 Al protein (35). A negative regulatory element (NRE) of human CYP1A1, localized between -833 and -558 bp (32), has been demonstrated to down-regulate a heterologous promotor/enhancer in HepG2 cells (33). Recently we reported a cell-specific effect of the NRE on its natural promotor/enhancer (34). In HepG2 the

E.C.BjBrgensen and H.Autrup

NRE influenced relative TCDD-induced promotor activity, whereas no effect was observed in MCF-7 cells. This difference in NRE function was supported by differences in formation of cell-specific protein—NRE complexes. The purpose of the present study was to examine the possible existence of an autoregulatory function of the human CYP1A1 gene and cell specificity in HepG2 and MCF-7 cells. The autoregulatory effect(s) was analyzed by co-transfection of CTP7/*/-reporter constructs and an expression plasmid carrying human CYP1A1 cDNA. The involvement of AHR and ARNT has been assessed by RT-PCR on total RNA from cotransfected cells. Materials and methods

Cells and culture conditions The human hepatoblastoma cell line HepG2, purchased from the American Type Culture Collection (Rockville, MD), was cultured in Dulbecco's modified Eagle's medium (DMEM) with glutamax-1 supplemented with 10% fetal bovine serum (FCS) and 64 Jig/ml garamycin" (passages 77-87). The human breast carcinoma cell line MCF-7 was obtained from the Breast Cancer Task Force Cell Culture Bank (Mason Research Institute, Worcester) and propagated in DMEM without phenol red supplemented with 1% FCS, 64 (ig/ml garamycinR, 2.5 mM glutamine and 6 u.g/1 insulin (passages 298-310). Subcultivation of the cells has previously been described (34). All cell cultures were maintained in a humidified incubator at 37°C in 5% COT/95% air. Plasm ids The reporter plasmids used in this study are shown in Figure I. pRNHIlc, pRNHIoc and pRNH23c have been described (32) and were provided by Dr R.N.Hines (Detroit, MI). The construction of pSdN 1 lc has been previously described (34). The expression plasmid pSVh 1AI, carrying the human CYPIAI cDNA in front of the non-inducible SV40 promotor, was described (36) and provided by Dr W.A.Schmalix (Technische Universitat, MUnchen, Germany). The plasmid constructs were verified by restriction endonuclease digestion and/or DNA sequence analysis (37). Transient expression assays Transfections of HepG2 and MCF-7 cells were done in triplicate in 60 mm diameter dishes as earlier described (34). Briefly, DNA was introduced into the cells employing the lipofecun (Gibco BRL)-mediated transfection protocol using serum-free medium. The HepG2 and MCF-7 cells were transfected 24 and 48 h after cell plating respectively. After transfection the cells were treated with 0.1% dimethylsulfoxide (DMSO) (solvent control) or 10 nM TCDD (induction) for 26 (HepG2) or 21 h (MCF-7). In co-transfection experiments expression plasmid pSVhlAl was introduced at a 0-, I- or 3fold molar excess relative to the chloramphenicol acetyltransferase (CAT) reporter plasmid. To ensure that the cells were not overloaded with DNA the total amount of DNA (15.6 |ig) transfected in each experiment was kept constant by substituting varying amounts of pBR327. pRSV-pgal (0.6 |ig) was used as an internal standard for transfection efficiency. Cell harvest, CAT assays and determination of p"-galactosidase (fi-gal) enzyme content were performed as previously described (34). CAT activity was normalized to transfection efficiency (fi-gal) and protein content. Northern and RNA slot blot Total RNA was extracted from the co-transfected cells using the method of Chomczynski and Sacchi (38). For Northern blots 10 ng total RNA were fractionated on denaturing agarose-formaldhyde gels and transferred to Hybond™-N + nylon membranes, which were baked for 30 min at 80°C and illuminated for 3 min with 312 nm UV light. Filters were pre-hybndized and hybndized in glass bottles according to the manufacturer's instructions in an automatic hybridization oven (HYBAFD). Hybridizations were carried out for 16 h with 2X10 6 c.p.mJml 32P-labeled randomly primed probe. To reprobe

436

RT-PCR Three micrograms of the prepared total RNA from co-transfected cells were reverse transcribed to synthesize cDNA at 37°C for I h using 0.2 p.g random hexamers [d(N)6] in a volume of 33 |il containing 45 mM Tns-HCl, pH 8 3, 68 mM KC1, 9 mM MgCI2, 1.8 mM each dNTP, 50 U RNAguard, 0.001% gelatine. 1.5 mM dithiothreitol, 20 U AMV reverse transcriptase The RNAcDNA duplex was denatured at 90°C for 5 min and placed on ice or stored at -20°C For PCR titration series were performed to obtain the optimal amount of cDNA and molar concentration of each primer set to obtain specific bands within the linear range, ensuring that the level of specific RT-PCR products synthesized was proportional to the respective input of cDNA. In order for the RT-PCR method to be relatively quantitative tube-to-tube variability in cDNA synthesis was corrected by normalizing to the amplified internal standard, glyceraldehyde phosphate dehydrogenase (GAPDH). PCR amplification with specific primers was carried out in 100 |il mixtures consisting of 20 mM Tris-HCI, pH 8.4, 50 mM KCI, O.I mM each dNTP, 1.5 mM MgCI2, 2 5 U Taq polymerase (Gibco), I u.Ci [32P]dATP and 0.01% W-l For each sample two PCR were performed in a reaction mix containing 0.2 nM forward and 0.2 |iM reverse primers for AHR, ARNT or GAPDH and cDNA mixtures equal to 0.2, 0.1 or 0.4 (ig RNA, however, 10 times less cDNA mixture were used for amplification of GAPDH. The PCR comprised 23 cycles for GAPDH and 25 cycles for AHR and ARNT. The reactions were heated to 94°C for 3 min and immediately cycled through a denaturating step at 95°C for 60 s, annealing at 65°C for 30 s and an elongation step at 72°C for 60 s. Following the final cycle a 10 min elongation step was performed. The PCR products were then subjected to 5% polyacrylamide gel electrophoresis and the radioactivity was quantitated by phosphonmager analyses. Oligonucleotides utilized in PCR amplification were as follows: AHR AHR3 ATACTGAAGCAGAGCTGTGC AHR4 AAAGCAGGCGTGCATTAGAC ARNT ANT3 CGGAACAAGATGACAGCCTAC ANT8 ACAGAAAGCCATCTGCTGCC GAPDH GAP I ACATCGCTCAGACACCATGG GAP2 GTAGTTGAGGTCAATGAAGGG The AHR, ARNT and GAPDH primers were predicted to produce RT-PCR products of 180, 225 and 150 bp respectively. The design and position of pnmers encoding the AHR and ARNT products has been previously described (7).

Results A negative autoregulatory feedback loop has been reported for the murine cypla-1 gene in a mouse hepatoma cell line system (35). We wanted to examine whether the expression of human CYP1A1 might be under autoregulatory control, possibly through the NRE promotor segment. Co-transfection studies were carried out with CYP1A1-CAT reporter plasmids (Figure 1) carrying intact 5' regulatory sequences (pRNHIlc) or plasmids where the NRE was substituted by unrelated DNA sequences (pSdNllc) and the expression vector pSVhlAl in 0-, 1- or 3-fold molar excess relative to the CAT reporter plasmid. Transfection of HepG2 with the reporter plasmids alone showed, as earlier observed (34), an ~ 10-fold induction of CAT activity upon TCDD treatment. Figure 2A summarizes the relative CAT activity in DMSO- or TCDD-treated HepG2 cells upon co-transfection. The data for DMSO (1:0) and TCDD (1:0) were set to 100%. An autorepressing effect was clearly observed, since co-transfection with increasing amounts of the non-inducible CYPIA] expression plasmid (pSVhlAl) significantly decreased CAT activity for both the parent plasmid pRNH 11 c and the NRE-substituted construct (pSdNlie). Upon co-transfection with a 3-fold molar excess of pSVhlAl basal CAT activity of pRNHIlc and pSdNllc was decreased 53.1 (P < 0.01) and 65.4% (P < 0.01) respectively and TCDD-


Materials All chemicals were obtained at the highest punty available from Sigma, Boehringer or Pharmacia. Restriction endonuclease and DNA modification enzymes were purchased from either New England Biolabs, Pharmacia or Life Technologies A/S. Mammalian cell culture media, fetal calf serum and lipofectin were from Gibco BRL l-Deoxy(dichloroacetyl-l-[l4C])chloramphenicol (55 Ci/mmol) in 0.25 M Tris-HCI, pH 7.5, [oc-32P]dATP (3000 Ci), Mega prime DNA labeling system RPNI604 and Hybond™-N+ nylon filters were acquired from Amersham Corp. TCDD was obtained from Wellington Laboratories (Lenexa, KS) and oligonucleotides from DNA technology (Science Park, Aarhus, Denmark).

filters 0.5% SDS was heated to boiling temperature and the probe removed by placing the filters in the solution until room temperature was reached. Filters were then pre-hybridized again before rehybridization. For RNA slot blots 5 |ig total RNA were spotted to the Hybond™-N+ membrane using a slot blotting apparatus (Schleicher & Schuell). The relative amount of radioactivity was quantified by phosphonmager analyses (Molecular Dynamics).

Autoregulation of human CYP1AI gene •1140

pRNH11c

-631

-560

-49

I

CAT

pSdN11c

CAT

pRNH16c

CAT

HepG2

s

I o

pRNH23c Fig. 1. Structure of CYPIAI-CAT reporter plasmids. The parent plasmid pRNHl 1c contains -1140 to +2424 bp of CYPIA1 including 5' flanking sequences, exon 1 and intron I in front of the CAT gene (32). To create pSdNHc the NRE fragment (-831 to -560) was substituted by an unrelated fragment, the 275 bp BamH\-Sall fragment from the tcp gene in pBR327 (34). The position of the inserted fragment is shown by dotted lines. To create pRNHloc the distal 5'-portion of pRNHllc, -832 to-1140 (32). carrying three XREs (39), was deleted. To create pRNH23c an internal proximal 5'-portion of pRNHl lc, -49 to -560 (32). carrying one XRE (39). was deleted. E, exon 1.

a

DtlSO

TCOD

IN

11c:hCYP1A1

B

CYP1A1

DMSO

TCOD

IND

SdN11c:hCYP1A1

GAPDH

1 2 3 4 5 6 Fig. 2. Co-transfection studies in HepG2 cells. (A) Percent relative CAT activity and percent fold induction upon co-transfection with pRNHl lc or pSdNI lc and pSVhlAl. pSVhlAl, containing human CYP/A1 cDNA was introduced at a 0-, I- or 3-fold molar excess relative to the CAT reporter plasmid. The data were corrected for transfection efficiency (fi-gal) and normalized to CAT activity at a molar ratio of 1:0 (pRNHl lc or pSdnllc:pSVhlAl), which was set to 100%. %IND, percent fold induction, the ratio of TCDD- to DMSO-induced CAT activity. Values are given as mean ± SD, n = 3-4. (B) RNA slot blot. In each transfection experiment total RNA was isolated from cells transfected in parallel with the CAT analyses. The RNA was analyzed by slot blot assay, probed with a 32 Plabeled 1.52 kp hP,450 5' £coRI fragment (40) and, for use as internal standard, a GAPDH-specific probe. The probes were verified by Northern blot before use. 1 and 2, transfection of pRNHllc alone (1:0) in DMSOand TCDD-treated cells respectively. 3 and 4, co-transfection with pRNHllc and pSVhlAl in the molar ratio 1:3 in DMSO- and TCDDexposed cells respectively. 5 and 6, co-transfection with pSdNI lc and pSVhlAl in the molar ratio 1:3 in DMSO- and TCDD-treated cells respectively. In both cell lines the expression of GAPDH RNA was unaffected by TCDD exposure as determined by both Northern blot and slot blot analyses.

Previously we reported differences in CYP1A1 promotor function in HepG2 and MCF-7 cells (34). To examine whether the autoregulatory mechanism(s) of CYP1A1 depends on cellspecific factors, causing differences between cell lines representing various tissues, MCF-7 cells were co-transfected with pRNHllc or pSdNllc and a 0-, 1- or 3-fold molar excess of pSVhlAl. Similarly to HepG2 cells, TCDD inducibility of the reporter plasmids in MCF-7 was increased ~ 10-fold. Figure 3B shows that co-transfection of MCF-7 with CYP1A1 cDNA does not influence basal CAT activity of either pRNHllc or pSdNllc. In contrast to HepG2 cells, co-expression of a 3fold molar excess of pSVhlAl in TCDD-exposed MCF-7 cells resulted in an induced CAT activity significantly higher than 437


induced activity 26.5 (P < 0.01) and 53.6% (P = 0.01) respectively. pSdNllc-dependent CAT activity was more repressed by CYP1A1 cDNA expression than pRNHllc activity. Since basal CAT activity was more affected than induced expression, increasing amounts of co-transfected CYP1A1 expression plasmid increased the induction ratio compared with transfection with pRNHllc or pSdNllc alone, by 1.6and 1.4-fold (P < 0.01 and P = 0.05) respectively. As a control for the transient SV40 promotor-directed expression of exogenous introduced CYP1A1 cDNA total RNA was isolated from cells transfected in parallel with cells for CAT assays and analyzed by slot blotting (Figure 2B). Relative to the amount of endogenous CYP1A1 RNA from cells transfected with the CAT reporter plasmids alone, co-transfection with non-inducible pSVhlAl in a 3-fold molar excess contributed to an ~5-fold and 3-fold increase in CYP1A1 RNA within cells treated with DMSO or TCDD respectively. In summary, these results indicate that introduction and increased expression of exogenous CYP1A1 cDNA driven by a heterogenous promotor leads to an autoregulatory effect on CYP1A1 promotor-driven CAT activity in a NRE-independent manner. To test whether other specific promotor sequences of the human CYP1A1 gene are involved in this autorepressing activity of the gene product HepG2 cells were co-transfected in parallel with pRNHl lc, pRNHloc or pRNH23c (see Figure 1) and pSVhlAl in a molar ratio of 1:0 or 1:3. The results in Figure 3A illustrate that neither deletion of the distal (pRNHloc) nor the proximal (pRNH23c) portions of 5' flanking sequences of the CYP1A1 gene affect the autoregulatory effect of the exogenous CYP1A1 gene product. A similar pattern of repression of CAT activity upon co-expression with CYP1A1 cDNA was observed for pRNHllc, pRNHloc and pRNH23c. However, compared with pRNHllc and pRNH23c deletion of the distal part of the promotor region (pRNHloc) resulted in a more marked reduction in TCDD-induced CAT activity (31.2, 39.5 and 72.3% respectively). Expression of exogenous CYP1A1 cDNA, analyzed in parallel transfected cells, showed an ~5-fold (DMSO) and 3-fold (TCDD) increase in CYP1A1 RNA upon transfection of a 3-fold molar excess of pSVhlAl (not shown). Thus each of the 5' flanking promotor segments analyzed can independently mediate the negative autoregulatory effect of the CYP1A1 gene product in HepG2 cells.

a

E.C.Bj0rgensen and H.Autrup

B

MCF-7

HepG2

O CO

Q

£

a

a z

2

DUSOTCDD

IND

11c:hCYP1A1

DM8OTC00 IND

DUSOTCOO IND

16c:hCYP1A1

23c:hCYP1A1

DUSO

DMSO

TCOO

TCOO

SdN11c:hCYP1A1

11c:hCYP1A1

upon transfection of pRNHIlc or pSdNllc alone, 1.9- and 1.4-fold (P = 0.01 and P = 0.04) respectively. Since basal CAT activity was unaffected in these co-expression studies, the induction ratio (TCDD:DMSO) was similar to the induced activities. The increment in CYP1A1 RNA upon introduction of a 3-fold molar excess of pSVhlAI in the co-transfection assays in MCF-7, supposed to reflect expression of exogenous CYP1A1 cDNA over endogenous CYP1A1, was in the same range as observed in the studies of HepG2 cells (5- and 3fold for DMSO- and TCDD-treated cells respectively; data not shown). Thus the co-expression studies in MCF-7 cells further indicate that cell-specific factors are involved in autoregulation, causing a down-regulation of expression of the human CYP1A1 gene in HepG2 cells and enhanced xenobiotic induction in MCF-7 cells. The two known components of activated TF complex, AHR and ARNT, may be involved in the autoregulatory mechanism(s) of CYP1A1. It was considered whether the increment in CYP1A1 expression might affect gene expression or RNA stability of AHR and ARNT. In the total RNA isolated from cells transfected in parallel with cells for CAT analyses the steady-state levels of mRNA for AHR and ARNT were analyzed by a relative quantitative RT-PCR method. A linear increase in the radiolabeled PCR product proportional to the input of cDNA was ascertained, as described in Materials and methods. We found, however, no indication of changes in the mRNA levels of AHR and ARNT in the two cell lines upon increased expression of exogenous CYP1A1 cDNA (Figure 4). Discussion A negative autoregulatory feedback loop has been suggested for control of mouse cypla-1 gene expression, extending to other genes in the dioxin-inducible [Ah] battery (35,41). It was shown that stable integration of wild-type cypla-1 cDNA in mutant mouse hepatoma cells, expressing markedly elevated transcripts but defective CYP1A1 enzyme, was sufficient to restore repression of the endogenous cypla-1 gene, as well as inducibility by TCDD. Similar experiments with human CYP1A2 cDNA indicated that each member of the CYP1A family has a transcriptional regulatory function in mouse hepatoma cells that extends beyond these two genes and 438

1

2

3

4 5

6

7

8

9 10 11 12 13 U 15 16 17 18

ARNT . AHR ' GAPDHi

225bp 180bp 150 bp

DMSO + + + + + + + + + TCDO + + + + + + + + + 1:0 + + + + + +

B

1 2 3 4

5 6

7 8

9 10 11 12

ARNT AHR GAPDHi

-«225bp •» i80bp •* 150bp

DMSO + + + + + + TCDD + + + + + + 10 + - + - + - + - + - + _ 13 - + - + - + _ + _ + _ + Fig. 4. Expression of AHR and ARNT mRNA in transfected HepG2 and MCF-7 cells. Total RNA samples were prepared from HepG2 (A) and MCF-7 (B) cells transfected in parallel with cells for CAT analyses (Figures 2 and 3). Relative quantitative RT-PCR analyses of these samples were carried out using AHR, ARNT, and GAPDH primers as described in Materials and methods. Arrows represent the predicted size of each RT-PCR product. (A) RT-PCR products of total RNA from HepG2 transfectants exposed to DMSO or TCDD. Lanes 1-6, AHR; lanes 7-12, ARNT; lanes 13-18, GAPDH. Lanes I, 4, 7, 10, 13 and 16, molar ratio pRNHllc:pSVhlAl, 1:0; lanes 2,5, 8, 11, 14 and 17, molar ratio pRNHllc:pSVhlAI, 1:3; lanes 3, 6, 9, 12, 15 and 18, molar ratio pSdNllc:pSVhlAI, 1:3. (B) RT-PCR of total RNA from MCF-7 transfectants exposed to DMSO or TCDD. Lanes \-4, AHR; lanes 5-8, ARNT; lanes 9-12, GAPDH. Lanes 1, 3, 5, 7, 9 and 11, molar ratio pRNH 11 c:pS Vh 1A1, 1:0; lanes 2, 4, 6, 8, 10 and 12, molar ratio pRNHIlc:pSVhlAl, 1:3.

includes at least two other genes in the [Ah] battery, Nco-1 and Aldh-3 (35). Previously we observed cell-specific differences in human CYP1A1 promotor function; the NRE promotor segment exerted a negative effect on TCDD-induced gene expression in hepatoma HepG2 cells, whereas no effect was observed in breast carcinoma MCF-7 cells (34). The major objective of


Fig. 3. Co-expression studies in HepG2 and MCF-7 cells. (A) Percent relative CAT activity and percent fold induction upon co-transfection of HepG2 with pRNHIlc, pRNHloc or pRNH23c and pSVhlAI. Mean ± SD, n = 3-4. (B) Percent relative CAT activity and percent fold induction upon co-transfection of MCF-7 with pRNHIlc or pSdNllc and pSVhlAI. Mean ± SD, n = 4-5. For both (A) and (B) the experimental design was as described in the legend to Figure 2 and Materials and methods.

Autoregulation of human CYPlAl gene

differences between the two cell lines. The reason for this discrepancy concerning TCDD-dependent maximal indue ibility between previous reported results and this study is unclear, but may be explained by differences in the serum concentration used in the culture media for the two cell lines. Multiple factors may contribute to the autoregulatory effects of the P4501A1 gene product. A BTE sequence containing the GC box consensus is located immediately upstream of the TATA box in the CYPlAl gene, including the human gene (30,44). Yanagida et al. demonstrated that maximal basal transcription, as well as high inducible expression, of the rat CYPlAl gene depends on these BTE sequences (30). Moreover, transient expression analysis in the monkey CV-1 cell line indicated that promoters such as the promotor of the CYPlAl gene, which has a single GC box, are activated by the Spl transcription factor but repressed by a rat BTE binding protein (BTEB) (45). The BTE sequence might be involved in autoregulation of the CYPlAl gene and different expression levels of Spl and/or BTEB between HepG2 and MCF-7 cells could possibly explain the divergent results observed in the two cell lines. However, other mechanisms must be involved, since deletion of BTE in the pRNHloc construct did not abolish down-regulation of CAT activity upon co-expression of CYPlAl cDNA in HepG2 cells. Several glucocorticoidresponsive element (GRE) sequences have been identified in intron 1 of the CYPlAl gene and deletion of most of intron 1 was reported to eliminate the positive synergism of dexamethasone on induced CYPlAl activity without affecting the inducible property of the promotor (32). Transient transfection and DNA binding studies will help to clarify whether the BTE and/or the GRE elements are involved in the autoregulatory mechanisms. The negative autoregulation observed in HepG2 cells might involve down-regulation of AHR DNA binding activity by labile or inducible factors, as suggested for rat and murine hepatoma cell lines (46). The AHR and ARNT proteins are reported to heterodimerize via bHLH motifs (17,43). Interaction or dimerization with cell-specific factors may explain the different effect of CYPlAl cDNA co-expression observed in HepG2 and MCF-7 cells, causing different kinetics of XRE binding activity, i.e. inhibition in the former and enhanced affinity in the latter. Combinatorial interaction of the AHR TF complex with other transduction pathways, such as for example the XRE binding C/EBP factor (28,29), may cause cell-specific modulation of transcriptional regulation of the [Ah] gene battery. Furthermore, basal and TCDD-induced expression of CYPlAl vary greatly in different human breast cancer cell lines (47,48) and an estrogen receptor-dependent, AHR-mediated induction of CYPlAl has been demonstrated (49). In addition, alteration in other proteins participating in dioxin responsiveness, such as hsp90, protein kinase C and phosphatase(s) (43), may increase the diversity of dioxin-induced gene expression. Future transfection analyses using reporter constructs carrying a single XRE in front of a minimal promotor and DNA binding studies will elucidate the dependence on XREs and/or a role for the AHR-ARNT complex in the autoregulatory mechanisms of the CYPlAl gene. It is well known that allelic differences at the Ah locus result in striking differences in susceptibility to chemically induced cancer among different strains of mice (15). Differences in autoregulation causing altered control of CYPlAl expression may also influence an organism's or tissue's susceptibility to dioxin and tumor promotion. A strict control of 439


the current study was to investigate a possible cell-specific difference in autoregulation of the CYPlAl gene in a human system, using HepG2 and MCF-7 cells as models. This report demonstrates that transient co-expression of increasing amounts of the non-inducible CYPlAl expression plasmid pSVhlAl significantly down-regulate basal as well as TCDD-induced activity of the CYPlAl promotor in HepG2 cells. In contrast, co-transfection of MCF-7 cells with a 3-fold molar excess of pSVhlAl caused an ~2-fold increase in induced CYPlAl promotor-driven CAT activity (pRNHllc), whereas it had no effect on basal promotor activity. In summary, these results indicate that trans-acting cell-specific factors are involved in both autoregulation of CYPlAl gene expression and the effect of the NRE on CYPlAl promotor activity. To test if autoregulation of CYPlAl was dependent on specific promotor segments CAT reporter constructs with substitution or deletion of CYPlAl promotor sequences were co-transfected with the pSVhlAl expression plasmid. Deletion of neither the distal (pRHNloc) nor the proximal (pRNH23c) part of the promotor nor substitution of the NRE abolished down-regulation of CAT activity upon co-expression of exogenous CYPlAl cDNA in HepG2 cells, although a more markedly reduced CAT activity was observed with pRNHloc, probably because deletion of this segment, including three XREs, contributes considerably to both basal and induced promotor activity (32,39). CAT activity of the NRE-substituted construct (pSdNllc) was also more affected by co-expression of the pSVhlAl expression plasmid in both cell lines. This may be correlated with the capacity of the NRE to contribute to maximal promotor activity (34). These data show that deletion of neither of the specific 5' promotor segments abolishes the autoregulatory function of the CYPlAl gene product, suggesting that modulation of fra/w-acting factor(s) and/or other DNA sequences are involved in this mechanism. In the mouse hepatoma cells translation of CYPlAl mRNA into a functional enzyme was reported to be an essential requirement for autoregulation (35). TCDD exposure of both cell lines caused an ~6-fold increase in steady-state levels of endogenous CYPlAl mRNA in cells transfected with CAT reporter plasmids only. An additional increase of ~5-fold (DMSO) and 3-fold (TCDD) in Cm^U-specific mRNA was determined upon co-expression with a 3-fold molar excess of pSVhlAl. Since previous studies have reported a linear correlation between CYPlAl mRNA and AHH enzyme activity in both HepG2 and MCF-7 (42), our results support a role for P4501A1 enzyme in an autoregulatory feedback mechanism(s) on the transcriptional activity of the gene that encodes it. Two possible mechanisms have been suggested for P450mediated negative autoregulation (35), i.e. either a substrate of CYP1A involved in stimulation of transcription is inactivated by the enzyme and/or CYP1A enzymes convert the substrate into a metabolite involved in transcriptional repression. Among candidates involved in autoregulation are the AHR and ARNT proteins, known to form an activating TF complex (17,43). Using relative quantitative RT-PCR it was assessed whether differences in gene expression and/or RNA stability of these two factors might explain the effects of the CYPlAl gene product on its promotor activity. However, the results showed no indication of an alteration in mRNA level for AHR or ARNT in either of the co-transfected cell lines. It has been reported that a 5-fold higher level of TCDD is necessary to induce CYPlAl maximally in MCF-7 cells as compared with HepG2 (42), however, we did not observe this

E.C.Bj0rgensen and HAutrup

genes responding to environmental pollution and/or involved in detoxification of toxic chemicals must be of great importance. Understanding cell-specific differences in regulation of CYP1A1 expression will increase insight into the relationships between genetic susceptibility to compounds like TCDD. Currently we are focused on further studies which will elucidate the cell-specific mechanisms of transcriptional regulation of the CYP1A1 gene. Acknowledgements We thank Drs R.N Hines and W.A.Schmalix for plasmids and Birgit Dall and Lisbet Kjeldberg for excellent technical assistance. This project was supported by a grant from the Danish Cancer Society to Herman Autrup.

References

440


1. Nelson,D.R, Kamataki.T., Waxman.DJ. et al. (1993) The P450 superfamily: update on new sequences, gene mapping, accession numbers, early trivial names of enzymes, and nomenclature. DNA Cell Biol., 12, 1-51. 2.Gonzales,F.J. (1989) The molecular biology of cylochrome P450s. Pharmacol. Rev., 40, 243-288. 3.Cooper,C.S., Grover.P.L. and Sims,P. (1983) The metabolism and activation of benzo[a]pyrene. In BndgesJ.W. and Chasseaud.L.F. (eds). Progress in Drug Metabolism. Wiley, New York, NY, vol. 7, pp. 295-396. 4.Jones,C.A., Santella.R.M., Huberman.E., SelkirkJ.K. and Grunberger.D. (1983) Cell specific activation of benzo[a]pyrene by fibroblasts and hepatocytes. Carcinogenesis, 4, 1351-1357. 5. Ioannides.C. and Parker,D.V. (1990) The cytochrome P450 I gene family of microsomal hemoproteins and their role in the metabolic activation of chemicals. Drug Metab. Rev., 22, 1-85. 6.WiIson,N.M., Chnstou.M., Tumer.C.R., Wrighton.S.A. and Jefcoate.C.R. (1984) Binding and metabolism of benzo[a]pyrene and 7,12dimethylbenz[a]anthracene by seven purified forms of cytochrome P-450. Carcinogenesis, 5, 1475-1483. 7.Hayashi,S., WatanabeJ., Nakachi.K., Eguchi.H., Gotoh.O. and Kawajiri.K. (1994) Interindividual differences in expression of human Ah receptor and related P450 genes. Carcinogenesis, 15, 801-806. 8.Fujino,T., Gottlieb.K., Manchester.D.K., Park.S.S., West.D., Gurtoo.H.L., Tarone.R.E. and Gelboin.H.V. (1984) Monoclonal antibody phenotyping of interindividual differences in cytochrome P-450-dependent reactions of single and twin human placenta Cancer Res., 44, 3916-3923. 9.Crofts,F., Taioh.E., Trachman.J., Cosma,G N , Curne.D., Toniolo.P. and Garte.SJ. (1994) Functional significance of different human CYP1A1 genotypes. Carcinogenesis, 15, 2961-2963. lO.Schrenk.D., StUven.T, Gohl.G., Viebahn.R. and Bock,K.W (1995) Induction of CYPIAI and glutathione S-transferase acuvities by 2,3,7,8tetrachlorodibenzo-p-dioxin in human hepatocyte cultures. Carcinogenesis, 16, 943-946. ll.Kouri.R.E., Mckinney.C.E., Slomiany.D.J., Snodgrass.D.R.. Wray.N.P. and McLemore.T.L. (1982) Positive correlation between high aryl hydrocarbon hydroxylase activity and primary lung cancer as analyzed in cryopreserved lymphocytes. Cancer Res., 42, 5030-5037. 12. McLemore.T.L., Adelberg.S., Liu.M.C. et al. (1990) Expression of CYPIA1 gene in patients with lung cancer, evidence for cigarette smoke-induced gene expression in normal lung tissue and for altered gene regulation in primary pulmonary carcinomas. J. Nail Cancer Inst., 82, 1333-1339. 13.Pyykko,K., Tuimala.R., Aalto.L. and Perkio.T. (1991) Is aryl hydrocarbon hydroxylase activity a new prognostic indicator for breast cancer? Br. J. Cancer, 63, 596-600. l4.Murray,G.I., Foster.C.O., Barnes.T.S., Weaver.RJ., Ewen.S.W.B., Melvin, W.T. and Burke.M.D. (1991) Expression of cytochrome P45OIA in breast cancer. Br. J. Cancer, 63, 1021-1023. 15.Ncbert,D.W. (1989) The Ah locus: genetic differences in toxicity, cancer, mutation, and birth defects. CRC Crit. Rev. Toxicol.. 20, 153-174. 16. Landers J.P. and Bunce.NJ. (1991) The Ah receptor and the mechanism of dioxin toxicity. J. Biochem., 276, 273-287. l7.Swanson,H.I. and Bradfield.C.A. (1993) The AH-receptor genetics, structure and function. Pharmacogenetics, 3, 213—230. 18.Nebert,D.W., Puga.A. and Vasiliou.V. (1993) Role of the Ah receptor and the dioxin-inducible [Ah] gene battery in toxicity, cancer, and signal transduction. Ann. NY Acad. Sci., 685. 624-640. 19..Nebert,D.W., Petersen.D.D. and Fomace.AJ.Jr (1990) Cellular responses to oxidative stress: the [Ah] gene battery as a paradigm. Environ Hlth Perspect., 88. 13-25.

2O.Sutter,T.R. and Greenlee.W.F. (1992) Classification of members of the Ah gene battery. Chemosphere, 25, 223-226. 21.Okey,A.B., Riddick.D.S. and Harper,P.A. (1994) The Ah receptor: mediator of the toxicity of 2.3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Toxicol Leu., 70, 1-22. 22. Fernandez-Salguero.P., Pineau.T., Hilbert.D.M. et al. (1995) Immune system impairment and hepatic fibrosis in mice lacking the dioxin-binding Ah receptor. Science, 268, 722-726. 23. WhitelawX., Pongratz.I., Wilhelmsson,A., Gustafsson J.-A.A. and Poellinger, L. (1993) Ligand-dependent recruitment of the Arm coregulator determines DNA recognition by the dioxin receptor. Mol. Cell. Biol., 13, 2504-2514. 24.Carrier,F., Owens.R.A., Nebert.D.W. and Puga.A. (1992) Dioxin-dependent activation of munne cypla-l gene transcription requires kinase Cdependent phosphorylation. Mol. Cell. Biol., 12, 1856-1863. 25. Berghard.A., Gradin.K., Pongratz.l., Whitelaw.M. and Poellinger.L. (1993) Cross-coupling of signal transduction pathways: the dioxin receptor mediates induction of cytochrome P-450IA1 expression via a protein kinase C-dependent mechanism. Mol Cell. Biol., 13, 677-689. 26.Singh,S.S. and Perdew.G.H. (1993) Alteration in the Ah receptor level after staurosporine treatment. Arch. Biochem. Biophys., 305, 170-175 27.Moore,M., Narasimhan.T.R., Steinberg.M.A., Wang.X. and Safe.S. (1993) Potentiation of CYPIAI gene expression in MCF-7 human breast cancer cells cotreated with 2,3,7,8-tetrachlorodibenzo-p-dioxin and I2-0tetradecanoylphorbol-13-acetate. Arch. Biochem Biophys., 305, 483-488. 28. Pimental.R A., Liang.B., Yee.G.K., Wilhelmsson.A., Poellinger.L. and Paulson,K.E. (1993) Dioxin receptor and C/EBP regulate the function of the glutathione S-transferase Ya gene xenobiotic response element. Mol. Cell. Biol., 13, 4365-4373. 29.Carner,F, Chang,C.-Y, DuhJ.-L., Nebert.D W. and Puga,A. (1994) Interaction of the regulatory domains of the murine cyplal gene with two DNA-binding proteins in addition to the Ah receptor and the Ah receptor nuclear translocator (ARNT). Biochem. Pharmacol, 48, 1767-1778. 30.Yanagida,A., SogawaX, Yasumoto.K-I. and Fuji-Kuriyama,Y. (1990) A novel c«-acting DNA element required for a high level of inducible expression of the rat P450c gene. Mol. Cell. Biol., 10, 1470-1475. 3I.Sagami,I., Kikuchi.H., Ikawa,S. and Watanabe.M. (1994) Characterization of hamster CYPIAI gene1 inducible expression and negative regulation. J. Biochem., 116. 801-810. 32.Hines,R.N.. MathisJ.M and Jacob.C.S. (1988) Identification of multiple regulatory elements on the human cytochrome P450IA1 gene. Carcinogenesis, 9, 1599-1605. 33. Boucher.P.D , Ruch.R.J. and Hines.R.N. (1993) Specific nuclear protein binding to a negative regulatory element on the human CYP\A\ gene. J. Biol. Chem., 268, 17384-17391. 34.j0rgensen,E.C.B and Autrup,H. (1995) Effect of a negative regulatory element (NRE) on the human CYPIAI gene expression in breast carcinoma MCF-7 and hepatoma HepG2 cells. FEBS Lett, 365, 101-107. 35.RayChaudhun,B., Nebert,D.W. and Puga.A. (1990) The murine cypla-l gene negatively regulates its own transcription and that of other members of the aromatic hydrocarbon-responsive [Ah] gene battery. Mol. Endocrinol, 4, 1773-1781. 36.Schmalix,W.A., Maser.H., Kiefer.F. et al. (1993) Stable expression of human cytochrome P4501AI cDNA in V79 Chinese hamster cells and metabolic activation of benzo[a]pyrene. Eur. J. Pharmacol.—Environ. Toxicol. Pharmacol Sea., 248, 251-261. 37.Sanger,F, Nicklen.S. and Coulson.A.R. (1977) Cloning in single stranded bacteriophage as an aid to rapid DNA sequencing. J. Mol. Biol., 143, 161-178. 38.Chomczynski,P. and Sacchi.N. (1987) Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162, 156-159. 39.Kubota,M., Sogawa.K., Kaizu.Y. Sawaya,T, WatanabeJ., Kawajin,K., Gotoh.O. and Fujii-Kunyama,Y. (1991) Xenobiotic responsive element in 5'-upstream region of the human P-450c gene. J. Biochem., 110, 232-236. 40.Jaiswal.A.K., Gonzales.F.J. and NeberuD.W. (1985) Human dioxinmducible cytochrome P|-450: complementary DNA and amino acid sequence. Science, 228, 80-83. 41.Gonzales.F.J. and NebertD.W. (1985) Autoregulation plus upstream positive and negative control regions associated with transcriptional activation of the mouse P,450 gene. Nucleic Acids Res., 13, 7269-7288. 42.Cresteil,T, JaiswalAK. and Eisen.HJ. (1987) Transcriptional control of human cytochrome P|-450 gene expression by 2,3,7,8-tetrachlorodibenzop-dioxin in human tissue culture cell lines Arch. Biochem. Biophys., 253, 233-240. 43. WhitlockJ.P. (1993) Mechanistic aspects of dioxin action. Chem. Res. Toxicol., 6, 754-763. 44.Jaisval,A.K., Gonzales.FJ. and Nebert.D.W. (1985) Human P,-450 gene

Autoregulation of human CYP1A1 gene sequence and correlation of mRNA with genetic differences in benzo[a]pyrene metabolism. Nucleic Acids Res.. 13. 4503-4520. 45.Imataka.H.. Sogawa.K.. Yasumoto.K., Kikuchi.Y.. Sasano.K.. Kobayashi.A., Hayamijvl. and Fujn-Kuriyama,Y. (1992) Two regulatory proteins that binds to the basic transcnption element (BTE), a GC box sequence in the promotor region of the rat P-4501AI gene. EMBO 7 . 1 1 . 3663-3671. 46. Reick.M., Robertson.R.W., Pasco.D.S. and FaganJ.B. (1994) Downregulation of nuclear aryl hydrocarbon receptor DNA-binding and transactivation functions: requirement for a labile or inducible factor. Mol. Cell. Biol.. 14, 5653-5660. 47.Pasanen.M.. Stacey.S., Lykkesfeldt.A.. Briand.P.. Hines.R. and Autrup.H. (1988) Induction of cytochrome P-450IAI gene expression in human breast tumour cell lines. Chem.-Biol. Interact., 66. 223-232. 48.ThomsenJ.S.. Nissen.L., Slacey.S.N.. Hines.R.N. and Autrup.H. (1991) Differences in 2.3.7.8-tetrachlorodibenzo-p-dioxin-inducible CYP1AI expression in human breast carcinoma cell lines involve altered transacting factors. Eur. J. Biochem., 197. 577-582. 49.ThomsenJ.S., Wang.X., Hines.R.N. and Safe.S. (1994) Restoration of aryl hydrocarbon (Ah) responsiveness in MDA-MB-231 human breast cancer cells by transient expression of the estrogen receptor. Carcinogenesis. 15. 933-937. Downloaded from http://carcin.oxfordjournals.org/ at Statsbiblioteket on January 2, 2015

Received on August 9. 1995: revised on November 17. 1995; accepted on November 22. 1995

441