![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Department of Cell Biology and Kaplan Cancer Center, New York University School of Medicine, New York, New York (V.H.B.) and Department of Medicine, Section of Medical Toxicology, University of Colorado Health Sciences Center, Denver, Colorado (L.C.Q.)
(Received September 25, 2003; accepted February 23, 2004)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). Members of the mammalian CYP1A family (CYP1A1 and CYP1A2) metabolize aromatic hydrocarbons (AHs)1. In this role they have been considered agents of toxicity and carcinogenesis. CYP1A2 is a major contributor to the metabolism of heterocyclic amines and aflatoxin B1 to carcinogenic compounds (McManus et al., 1990), but it also metabolizes commonly consumed agents, such as caffeine and acetaminophen (Buters et al., 1996; Peters et al., 1999). In addition, CYP1A family members metabolize steroids, retinoids, and arachidonic acid (Rifkind et al., 1995; Raner et al., 1996; Lee et al., 2003). These endogenous substrates participate in development and differentiation, maintenance of homeostasis, hormone response, and modification of signal transduction processes. Induction of CYP1A1 and CYP1A2 by AHs or other exogenous agents can, potentially, alter levels of these physiologically active substances, resulting in environmentally or diet-induced dysfunction and disease.
Current understanding of the mechanism of CYP1A1 induction by AHs has been summarized in a recent review (Okino and Whitlock, 2000). Briefly, AH ligands bind to a cytoplasmic aromatic hydrocarbon receptor (AHR), which is associated with a chaperone complex containing heat shock protein 90, AHR-interacting protein, and other cytoplasmic factors. The liganded AHR enters the nucleus where, released from its chaperone complex, it dimerizes with its partner, ARNT (Heid et al., 2000). The AHR/ARNT heterodimer interacts with response elements on the DNA variously called dioxin response elements (DREs), xenobiotic response elements (XREs), or aromatic hydrocarbon response elements (AHREs). The binding of AHR/ARNT at one or more of the multiple AHREs identified in the CYP1A1 gene ultimately results in increased transcription.
It is often assumed that the regulation of the CYP1A2 gene is identical to that of CYP1A1 because it is responsive to AHs. However, AHR/ARNT binding sites have been reported in CYP1A2 genes of only two species and characterized only for the human gene (Quattrochi and Tukey, 1989; Strom et al., 1992; Quattrochi et al., 1994). The two putative AHR/ARNT binding sites previously identified in the human gene are located much further 5' (-2495 and -2000) (Quattrochi et al., 1994) than the AHREs in CYP1A1; only one is functional in both binding and induction of reporter activity. The lower number of AHREs in the CYP1A2 gene may place more restrictions on its response to AHs. Detailed analysis of CYP1A2 regulation has been limited by the fact that CYP1A2 mRNA is expressed in only a few cell lines (MCF-7, LS-180, OMC-3, and HepG2) and CYP1A2 protein is detectable only at low levels, if at all (Li et al., 1998; Iwanari et al., 2002). In contrast, CYP1A1 is expressed in many cells lines at easily detectable levels.
In this study, we sought to increase knowledge of the CYP1A2 gene by analyzing the 5'-flanking region of the CYP1A2 gene in the guinea pig, a species known to be highly sensitive to dioxin. Two putative binding sites for AHR/ARNT (-830 and -575) were identified in the proximal -1 kb of the guinea pig CYP1A2 gene. We characterized these sites in a homologous cell line, guinea pig adenocarcinoma cells (GPC16), in which we demonstrated, for the first time, AH-induced expression of both CYP1A1 and CYP1A2. Both sites were capable of binding nuclear proteins prepared from GPC16 cells in an AH-induced manner, but only one (-830) was capable of mediating AH-induced reporter gene activity. Concatamers of oligonucleotides corresponding to this site were able to replicate the fold induction in reporter gene activity, and mutation of the site suppressed AH-induced activity, further supporting its involvement in AH-mediated induction of the guinea pig CYP1A2. However, the level of AH-induced reporter gene activity was considerably less than that reported for CYP1A1, indicating that other elements modulate CYP1A2 response to AH.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of Genomic DNA and Genomic DNA Clones. Guinea pig genomic DNA was prepared according to published protocols (Davis et al., 1986). Using the Universal GenomeWalker Kit (BD Biosciences Clontech, Palo Alto, CA), five pools or "libraries" of uncloned, adaptor-ligated genomic DNA fragments were constructed. Each library was amplified by PCR using the GenomeWalker adaptor primers and a gene-specific primer (5'-GGCAGGTGCAGGAGGTTAGCAGGATTG-3') designed against the 5' end of published cDNA sequence for guinea pig CYP1A2 available in GenBank (Accession numbers: D50457
[GenBank]
, U23501
[GenBank]
) (nt 3-29, D50457
[GenBank]
). The longest PCR products (1.2 kb) were obtained from the library created using the restriction enzyme StuI. These were subcloned into pCR2.1 using the Original TA Cloning Kit (Invitrogen, Carlsbad, CA). The distance between the putative first intron and the end of the known cDNA sequence, judged by comparison with the rat CYP1A2 gene (Sogawa et al., 1985), did not allow for creation of a true nested primer for genome walking. Therefore, the veracity of the clones was confirmed by PCR. The StuI library served as the template, and primers corresponded to the 5' end of the cDNA, inset from that used for genome walking (nt 8-31, U23501
[GenBank]
), and sequence at the 3' end of the 1.2-kb clone (nt -255 to -273, 5'-CTTAGCCTGCTCTGACCTTGAGAG-3'). PCR products were subcloned using the TOPO TA Cloning Kit for Sequencing (Invitrogen).
Sequence Analysis. Initial sequence of two 1.2-kb insert-bearing clones was obtained using the adaptor and gene-specific primers. Full-length sequence for each was obtained by "walking," using oligonucleotides corresponding to the sequence as it became known. Sequence for the subcloned PCR products was obtained using oligonucleotides corresponding to M13 priming sites flanking the plasmid cloning site.
Sequencing was performed by core facilities at New York University School of Medicine and University of Colorado Health Sciences Center. GCG Wisconsin programs (Accelrys, San Diego, CA) available at New York University were used for sequence analysis and in comparing guinea pig CYP1A2 sequence with other CYP1A2 gene sequences available in GenBank (Accession numbers: human, U02993
[GenBank]
, AF253322
[GenBank]
; rat, K03241
[GenBank]
; mouse, AB042202
[GenBank]
, M10022
[GenBank]
, X00479
[GenBank]
; hamster, D10914
[GenBank]
; rabbit, X05686
[GenBank]
) and with information available in published papers [human (Ikeya et al., 1989; Quattrochi and Tukey, 1989; Quattrochi et al., 1994), rabbit (Strom et al., 1992), rat (Sogawa et al., 1985), mouse (Gonzalez et al., 1985; Ikeya et al., 1989; Uchida et al., 2002)]. Potential binding sites for transcription factors were identified using MatInspector v2.2 (Quandt et al., 1995) available through TRANSFAC, the transcription factor database (Heinemeyer et al., 1998).
Potential start sites of transcription were estimated using a Primer Extension System (Promega, Madison, WI), following instructions provided by the manufacturer. 32P-labeled antisense oligonucleotide primer, corresponding to the proximal primer used in genome walking, was hybridized with total RNA prepared from liver tissue of 3MC-treated guinea pigs and extended using reverse transcriptase.
Reporter Constructs. The enhancer region encompassing both putative AHR/ARNT binding sites (-459 to -936) was sequentially subcloned through pBluescript into the reporter vector pSEAP2-Promoter (BD Biosciences Clontech), taking advantage of available restriction enzyme sites. The pSEAP2 vector contains the reporter gene, secreted alkaline phosphatase (SEAP), driven by an SV40 promoter. Fragments containing only one of the putative AHR/ARNT binding sites were obtained by cleaving the 
500-bp fragment with PvuII. These were subcloned directly into the Nru I site of the pSEAP2-Promoter. Concatamers of the more distal site (-830) were prepared by blunt ligation of ds oligonucleotides corresponding to its sequence (see below) and inserted into the Nru site of pSEAP2-Promoter. The set used in this study contained two copies of the site (5'-GGCCGTGGCGTGGGCTCTTGGCCGTGGCGTGGGCTCTT-3'). To facilitate site-directed mutational analysis in the context of the full-length 1.2-kb promoter region, the entire segment was subcloned into pGL3 reporter vectors (Promega), which contain the reporter gene luciferase. The 1.2-kb segment minus a 3' fragment encompassing the TATA box (-93 to -1204) was subcloned into the pGL3 promoter vector containing the SV40 promoter, while the entire 1.2-kb segment (-1 to -1204) was subcloned into the pGL3 basic vector, which lacks a promoter. Mutagenesis, performed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) and oligonucleotides mutated in the core binding sequence of the AHRE (5'-CGT/CG/T-3'), was confirmed by sequencing.
Cell Culture, Drug Treatment, RT-PCR, Northern and Western Blot Analysis, and Transfection. The guinea pig colon cell line GPC-16 (ATCC CCL 242) was purchased from the American Type Culture Collection (Rockville, MD). The cells were plated at 1 x 105 in 35-mm wells for RNA analysis and transfection experiments and at 5 x 106 in 150-mm dishes for preparation of microsomes and nuclear extracts. They were grown in Eagle's minimum essential medium (MEM) containing phenol red as an indicator of pH, sodium pyruvate (110 µg/ml) and Earle's balanced salt solution (90%) with fetal bovine serum (10%) in the presence of antibiotics (100 U/ml penicillin G, 100 µg/ml streptomycin) in an atmosphere of 5% CO2 at 37°C. The cells were treated with TCDD or 3MC, at concentrations given below, dissolved in dimethylsulfoxide (DMSO). The final concentration of DMSO was 0.1%. Controls received 0.1% DMSO only.
To characterize mRNA expression, cells were treated 36 to 48 h after plating with TCDD (1 and 10 nM) or 3MC (0.1 and 1 µM) for 0 to 48 h. RNA was prepared using RNA-Stat-60 (Tel-Test Inc., Friendswood, TX). RNA integrity was assessed by ethidium bromide staining of samples separated on formaldehyde-containing agarose gels. RT-PCR analyses for levels of CYP1A1, CYP1A2, AHR, and ARNT mRNA were performed using guinea pig-specific primers (Table 1) and were normalized by comparison with RT-PCR products for ß-actin, as previously described (Black et al., 1998b). Primers for AHR and ARNT corresponded to unpublished cDNA sequence for their periodicity/ARNT/simple-minded domains obtained in our laboratory in collaboration with Drs. Mark Hahn and colleagues, Woods Hole Oceanographic Institute (Woods Hole, MA) (Black et al., 1998a). The correspondence of the AHR primers to recently published sequence for guinea pig AHR (Korkalainen et al., 2001) is given in Table 1. On the basis of the mRNA analyses, further experiments were performed using TCDD (1 nM) or 3MC (0.1 µM).
|
Microsomes were prepared after 18 and 36 h of treatment and analyzed for levels of CYP1A1 and CYP1A2 immunoreactive protein, as previously described (Black et al., 1998b). Two primary antibodies were used: a monoclonal antibody, raised against scup CYP1A, that recognizes CYP1A1 of many vertebrate species and shows little cross-reactivity with CYP1A2 (Mab 1-12-13, received from Dr. John Stegeman, Woods Hole Oceanographic Institute); and a polyclonal antibody, raised against hamster CYP1A2, that has some cross-reactivity with CYP1A1 (Puracyp, La Jolla, CA). Reactions with peroxidase-labeled secondary antibodies were visualized using diaminobenzidine or enhanced chemiluminescence (Amersham Biosciences Inc., Piscataway, NJ).
Transfections were performed 36 to 48 h after plating, at 50 to 65% confluency, using LipofectAMINE (Invitrogen). DNA of the reporter vector, containing the reporter constructs described above (1 µg per 35-mm dish) and LipofectAMINE (3 µg per 35-mm dish) were prepared, combined, and plated, according to the manufacturer's instructions. Control transfections were carried out in a similar fashion. Reporter vectors lacking the CYP1A2 inserts served as a negative control. As a positive control, some cells were transfected with control vectors, provided by the manufacturer. To assess transfection efficiency, some cultures were cotransfected; pSEAP was paired with a vector-containing gene sequence for ß-galactosidase (pSV-ß-Galactosidase Control Vector; Promega) and pGL3 with a vector bearing Renilla luciferase (phRG-TK, 0.5 µg per 35-mm dish; Promega). The LipofectAMINE-containing medium was left on for 3 to 6 h and then replaced with Eagle's MEM. The following day this medium was replaced with Eagle's MEM containing TCDD (1 nM) or 3MC (0.1 µM) in DMSO (0.1% final concentration) or DMSO (0.1%) alone in duplicate or triplicate cultures for each condition. Reporter gene products and protein were assayed in cell lysates at the end of the treatment period.
Transfections were also performed using a cell line derived from a human hepatocarcinoma (HepG2) and isolated rat hepatocytes. HepG2 cells were grown under conditions described previously (Quattrochi and Tukey, 1989; Quattrochi et al., 1998). Rat hepatocytes were isolated and maintained as primary cultures, according to published protocols (Li et al., 1991). The culture medium for hepatocytes included vitamin C (50 µg/ml) and dexamethasone (10-7 M); both are important for maintaining optimal expression of CYP1A2 (Silver et al., 1990; Mori et al., 1997), as well as insulin (0.25 U/ml). HepG2 cells and isolated rat hepatocytes were transfected using LipofectAMINE and treated with TCDD (1 nM) and 3MC (0.1 µM), as described above.
Assays of reporter molecules were performed in 96-well plates, according to the manufacturer's instructions (pSEAP2; BD Biosciences Clontech; and Luciferase or Dual-Glo Luciferase Assay Systems and ß-Galactosidase Enzyme Assay System with Reporter Lysis Buffer; Promega). Luminescence was read in a microtiter plate luminometer (Microlite ML3000; Dynatech Laboratories, Chantilly, VA). A standard curve was run with each SEAP assay. Values for experimental cultures were on the linear portion of the curve within the range of 10 x 10-13 to 10 x 10-14 mol SEAP. Luciferase activity was determined to be in the linear portion of the instrument's detection range. Protein was determined using Pierce Micro BCA protein assay reagent (Pierce Chemical, Rockford, IL). Activities of both primary reporter molecules were expressed as relative light units (RLU) normalized to cotransfected reporter molecules and/or microgram of cell protein. Average values of duplicate or triplicate cultures were determined for each experimental condition. Three to five experiments were performed for each condition tested. Activity (RLU/µg protein) varied between experiments, but induction expressed as the fold increase of reporter molecule activity in treated cultures versus cultures exposed to DMSO alone was not significantly different for primary vectors bearing the same reporter construct, normalized to cotransfected controls or to protein alone. Therefore, values shown for fold induction represent pooled data from cotransfected and noncotransfected cultures. Activity in cultures transfected with noninsert-bearing plasmid was generally 80% of the level seen in cultures transfected with insert-bearing plasmid (pSEAP2, 0.81 ± 0.34; pGL3, 0.84 ± 0.07) and showed no significant increase in activity in the presence of TCDD or 3MC.
Statistical analysis was performed using InStat (GraphPad Software Inc., San Diego, CA). RLU per microgram protein values were analyzed by analysis of variance for repeated measures and Dunnett's post test for multiple comparisons. Analysis of variance followed by Bonferroni's multiple comparison post test was used to compare fold induction values from all experiments. Values are expressed as the mean ± standard deviation. Significance was set as P < 0.05.
Preparation of Nuclear Extracts and Electrophoretic Mobility Shift Assay (EMSA). GPC16 cells were treated as described above 36 to 48 h after plating and harvested 1.5 to 3 h later. Nuclear extracts were prepared by standard methods (Ausubel et al., 1990), and protein content was determined by the Bradford assay, as previously described (Black et al., 1998). Sense and antisense oligonucleotides corresponding to the mouse XRE (DRE3) (5'-GAGCTCGGAGTTGCGTGAGAAGAGCC-3') (Denison et al., 1988) and guinea pig CYP1A2 putative AHR/ARNT binding sites (R/NT1: -830, 5'-GGCCGTGGCGTGGGCTCTT-3'; R/NT2: -575, 5'-TCTCACTGCGTGGCAGCCA-3') were annealed to form 19 bp ds oligonucleotide probes and gel-purified before use, according to published protocols (Ausubel et al., 1990). Oligonucleotides corresponding to the more distal site (-830), but mutated in the core binding sequence (5'-CGT/CG/T-3'), were prepared in a similar manner. These probes were end-labeled with 
32P-ATP, purified on Sepharose columns (TE SELECT-D G-25; Eppendorf-5 Prime, Inc., Boulder, CO) and quantitated by fluorescence spectroscopy. EMSA was performed essentially as described by Denison et al. (Bank et al., 1995). The final reaction mixture contained 10 µg of protein, 25 mM Hepes, 1 mM EDTA, 1 mM dithiothreitol, and 10% glycerol (v/v), with 1 to 3 µg of poly(dI:dC) and 180 to 200 mM KCl, depending on the probe used. Unlabeled competitors were included in some reactions, as indicated in the figure legends, at 25- to 100-fold the concentration of the labeled probes. An oligonucleotide corresponding to the consensus sequence for the NF-
B binding site (5'-AGTTGAGGGGACTTTCCCAGGC-3'; Promega) was selected for use as a nonspecific probe.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
58 bp) is consistent with that reported for CYP1A2 in other species [human, 62 or 64 bp (Ikeya et al., 1989; Quattrochi et al., 1994, respectively), rabbit, 59 bp (Strom et al., 1992; X05686
[GenBank]
); rat and mouse, 60 bp (Gonzalez et al., 1985; Sogawa et al., 1985)]. Further comparison of the guinea pig 5'-flanking sequence with sequence reported for other CYP1A2 genes revealed two stretches of highly conserved sequence (80-90%) within the proximal 150 bp. One (-15 to -40) encompasses the TATA box. The other (-50 to -150) includes highly conserved core sequences for several transcription factor binding sites: NF-1 and GATA, as well as CAAT and E boxes. Further 5', the sequences diverged from one another considerably. Consistent with previous reports for CYP1A2, no sequence was found that corresponded to the basic transcription element found in CYP1A1 and other genes involved in xenobiotic metabolism (Ikeya et al., 1989; Fujii-Kuriyama et al., 1992).
|
Identification of AHRE Sites and Comparison with Those of Other CYP1A2 Genes. Computer analysis of more distal sequence identified two putative AHR/ARNT binding sites centered around -830 and -575 in the guinea pig gene. For the sake of simplicity, we called these putative binding sites AHR/ARNT 1 (R/NT1) and AHR/ARNT 2 (R/NT2), respectively. They possessed the canonical five-base pair sequence (GCGTG) characteristic of the AHR/ARNT binding sequence (Denison et al., 1988) and had high overall identity (R/NT1, 89.5%; R/NT2, 78.9%) with the 19-bp consensus sequences defined for the mouse DRE (Yao and Denison, 1992) (Fig. 2). The four-base pair core (CGTG) is essential for binding (Denison et al., 1988; Quandt et al., 1995), but sequence flanking this core can affect binding. R/NT1 had substitutions of two nucleotides in the binding consensus sequence (15, A/T to C and 19, C/G to T) and one in the functional consensus sequence (15, A/T to C). R/NT2 had substitutions of one base pair in the binding consensus sequence (19, C/G to A) and two in the functional binding consensus sequence (5, C/G to A and 6, T to C).
|
Characterization of CYP1A2 mRNA and Protein Expression in GPC16 Cells. GPC16 cells have been used in studies of AHR activation (Aarts et al., 1995; Heath-Pagliuso et al., 1998), making them a reasonable choice for further characterization of the putative AHRE sites in guinea pig CYP1A2. However, to our knowledge, GPC16 cells had not been characterized for expression of the target genes, CYP1A1 or CYP1A2. RT-PCR and Northern blot analyses (Fig. 3, A and B) demonstrated that mRNA CYP1A1 and CYP1A2 mRNA were induced in GPC16 cells following AH treatment, whereas mRNA for the two transcription factors involved in mediating effects of AHs, AHR, and ARNT were expressed at similar levels in control and treated cells (Fig. 3A). Comparison of the levels of induction of CYP1A produced by 1 versus 10 nM TCDD and 0.1 versus 1 µM 3MC showed that induction was somewhat diminished at the higher doses. Therefore, the lower doses were chosen for further experiments. A time course revealed that at these doses the levels of CYP1A1 and CYP1A2 mRNA increased by 3 h, remained at the induced level until 18 h, then became more variable and gradually decreased, returning to control levels in some cultures by 48 h (Fig. 3C). Levels of AHR and ARNT mRNA did not change over most of this period but did show a slight decrease at 48 h (Fig. 3C). Western blot analysis of microsomes prepared at 18 and 36 h after treatment revealed immunodetectable protein for CYP1A1 and CYP1A2 at both time points (Fig. 3D).
|
Characterization of AHRE Sites by EMSA and Transient Transfection Analysis. Binding of nuclear proteins to the two putative AHR/ARNT binding sites was tested by EMSA analyses. Nuclear extracts were prepared from cultures treated with TCDD or 3MC. Similar results were obtained using both AHs. For simplicity, results for TCDD are consistently shown. AH induced the binding of ds oligonucleotides corresponding to the mouse XRE (DRE3) to nuclear proteins from GPC16 cells (Fig. 4A). This binding was completed by unlabeled XRE (Fig. 4A), as well as by unlabeled ds oligonucleotides corresponding to R/NT1and R/NT2 (Fig. 4B), to a greater extent than by the nonspecific probe, ds oligonucleotide corresponding to the consensus sequence for NF-B (Fig. 4, A and B). Competition by unlabeled XRE was more effective than that by the unlabeled R/NT probes, probably reflecting differences in the consensus sequence flanking the core binding sequences, noted above. Yao and Denison (1992) showed that substitutions in positions 5 and 6 of mouse DRE3, seen in R/NT2, have no effect on TCDD-inducible complex formation. However, substitutions in nucleotides 15 and 19 of the DRE oligomers, seen in both R/NT sites, significantly decreased AH-induced nuclear protein binding affinity of DRE oligomers.
|
Binding of labeled ds oligonucleotides corresponding to R/NT1 and R/NT2 to nuclear proteins prepared from GPC16 cells was also AH-induced and specific (Fig. 4, B and C). The binding of R/NT1 and R/NT2 was greater to nuclear proteins from cultures treated with TCDD than to nuclear proteins from control cultures treated with the vehicle alone. The induced binding of both labeled R/NT probes was completed by unlabeled homologous oligonucleotide, as well as by unlabeled oligonucleotide corresponding to the other R/NT site (Fig. 4B) and to the mouse XRE (DRE3) (Fig. 4C), but not by the unlabeled nonspecific probe, NF-B (Fig. 4, B and C). However, the DNA-protein complex formed by R/NT2 was consistently less intense than that formed by R/NT1, and unlabeled R/NT2 was a less effective competitor of R/NT1 than of R/NT2. Oligonucleotides corresponding to R/NT1 in which the core binding site had been mutated did not bind to nuclear proteins from control or treated cultures, nor did they compete for binding (Fig. 4D). In some cases, two induced complexes could be distinguished (Fig. 4, C and D), consistent with the previous report of two distinct protein complexes in assays using guinea pig hepatic cytosol and mouse XRE (DRE3) (Bank et al., 1995). Both of the complexes reported by Bank et al. contained AHR and bound to the DRE with the same specificity; differences in subunit composition were suggested as an explanation for the different sizes of the protein-DNA complexes. These remain to be elucidated.
The functionality of the two AHR/ARNT sites was tested by transient transfection of host cells with reporter constructs containing these sites and analysis of reporter gene activity after treatment with TCDD or 3MC (Fig. 5). GPC16 cells, HepG2 cells, and primary cultures of isolated rat hepatocytes were transfected with pSEAP2-Promoter vector containing a segment encompassing both AHR/ARNT binding sites (-936 to -459). Treatment with TCDD increased SEAP activity and resulted in a 2- to 3-fold induction of the SEAP reporter molecule in all three cell culture systems (Fig. 5, A and B). Treatment of GPC16 cells transfected with a reporter construct containing R/NT1, but not R/NT2 (-936 to -633), also resulted in an approximately 2-fold induction of SEAP activity. However, cells transfected with a reporter construct encompassing R/NT2, but not R/NT1 (-633 to -459), showed no detectable, inducible SEAP activity (data not shown). These results suggest that the inducible reporter gene activity of the 477-bp fragment (-936 to -459) is mediated by R/NT1, not R/NT2.
|
To test if the R/NT1 site alone could mediate inducible reporter gene activity, host cells were transiently transfected with reporter constructs containing a concatamer of ds oligonucleotides corresponding to its sequence. TCDD treatment resulted in fold induction of SEAP activity that was not significantly different from that mediated by the reporter construct encompassing R/NT1 (-936 to -633) or the larger construct encompassing both AHR/ARNT binding sites (-936 to -459) (Fig. 5C). This supports the concept that the 19-bp site at -830 is functional, i.e., capable of mediating response to aromatic hydrocarbons, but it does not exclude the possibility that adjacent elements modulate this activity.
To test the functionality of R/NT1 in the context of the full-length 1.2-kb 5' flanking region, GPC16 cells were transfected with reporter constructs in the pGL3 vector containing the wild-type promoter or one in which R/NT1 had been mutated in the core binding sequence. TCDD treatment resulted in a fold induction of luciferase reporter gene activity, whether driven by a heterologous SV40 promoter or by the homologous promoter, that was not significantly different from that seen with the R/NT1 containing fragment (-936 to -633) or concatamers of R/NT1 in the pSEAP-2 promoter vector (Fig. 5D). Mutation of the core binding sequence at -830 suppressed this AH-induced reporter gene activity. These results indicate that the homologous promoter is functional and further support the concept that the 19-bp site at -830 is a functional AHRE. We have chosen to call the apparently functional site AHRE1.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
), including the human colon carcinoma cell line LS-180 (Li et al., 1998). Because its extrahepatic expression has only recently been recognized, there have been few analyses of CYP1A2 gene regulation in nonhepatic cells. In this article, we report initial cloning of the 5'-flanking region of the guinea pig CYP1A2 and demonstrate CYP1A2 expression in a guinea pig colon adenocarcinoma cell line, GPC16, for the first time. We identified two putative AHR/ARNT binding sites in the first -1.2 kb of the guinea pig CYP1A2 and used the homologous GPC16 cells to characterize one of these as a functional AHRE. This is the only AHRE to be characterized in CYP1A2, other than that previously characterized in the human gene (Quattrochi et al., 1994), and is one of few analyses of CYP1A2 gene regulation in extrahepatic cells.
Because CYP1A2 metabolizes and is induced by AHs, it has been assumed that CYP1A2 genes are regulated in a manner similar to CYP1A1. However, CYP1A2 genes do not contain the well characterized basic transcription element just proximal to the TATA box that is found in the promoter of many cytochromes P450, including CYP1A1, and several other genes involved in xenobiotic metabolism (Ikeya et al., 1989). Furthermore, although CYP1A1 genes contain multiple AHREs in the initial -1 kb, CYP1A2 genes appear to have fewer functional AHREs, diversely placed. In the human CYP1A2 gene, most putative binding sites identified are considerably more distal (-2 to -3.7 kb and beyond) (Quattrochi and Tukey, 1989; Quattrochi et al., 1994; Corchero et al., 2001) and, thus far, only one site (-2500) has been shown to be functional (Quattrochi et al., 1994). No AHREs have been identified in the initial -4 kb of the mouse Cyp1a2, suggesting more distal location in this species as well (Uchida et al., 2002; GenBank accession number AB042202
[GenBank]
). However, in the guinea pig CYP1A2 gene, as well as in the rabbit CYP1A2 gene (Strom et al., 1992), putative AHR/ARNT binding sites are located within the first -1 kb of the 5'-flanking region and a computer search performed as a part of our analysis revealed putative AHR/ARNT binding sites in the rat at -925 bp and in the hamster at -445 and -1645 bp. Although the possibility of more distal sites in guinea pig CYP1A2 gene, as well as in those of other species for which distal sequence is not yet available, cannot be excluded, our data indicate that one proximal AHRE site in the guinea pig CYP1A2 gene is functional.
It is important to note, however, that although the CYP1A2 promoter is functional, the level of AH-induced CYP1A2 reporter gene activity is lower than that reported for CYP1A1. In HepG2 cells, instead of 50-fold levels of CYP1A1 reporter gene activity induced by TCDD (1 nM) in transient transfection assays (Postlind et al., 1993), the human CYP1A2 shows responses of 1.8- to 6.5-fold (Aitchison et al., 2000; Postlind et al., 1993, respectively). The 2- to 3-fold levels of TCDD (1 nM)-induced reporter gene activity shown here for guinea pig CYP1A2 in three cell culture systems are within the range of levels reported for human CYP1A2 in HepG2 cells. The reasons for the consistently lower level of AH-responsiveness of CYP1A2 are not clear at this time, but low response suggests that other factors are involved in achieving the levels of CYP1A2 expression seen in vivo after AH exposure. These could include: 1) elements that mediate and/or modify AH-response located within, proximal or distal to regions containing the functional AHREs demonstrated thus far, such as more distal AHREs that act synergistically and/or sites for other transcription factors that suppress or enhance AH response; 2) interaction of AHR and/or ARNT with receptors or cofactors involved in other signaling pathways; or 3) post-transcriptional mechanisms altering mRNA stability.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Abbreviations used are: AH, aromatic hydrocarbon; AHR, AH receptor; ARNT, AH nuclear translocator; DRE, dioxin response element; XRE, xenobiotic response element; AHRE, AH response element; kb, kilobase; RT-PCR, reverse transcriptase-polymerase chain reaction; nt, nucleotide; 3MC, 3-methylcholanthrene; SEAP, secreted alkaline phosphatase; MEM, minimal essential medium; DMSO, dimethylsulfoxide; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; RLU, relative light units; EMSA, electrophoretic mobility shift assay; bp, base pair; NF, nuclear factor. 
Address correspondence to: Dr. Virginia H. Black, Department of Cell Biology, New York University School of Medicine, 550 First Avenue, New York, NY 10016. E-mail: blackv01{at}med.nyu.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P Guzelian, L Quattrochi, N Karch, L Aylward, and R Kaley Does dioxin exert toxic effects in humans at or near current background body levels?: an evidence-based conclusion Human and Experimental Toxicology, February 1, 2006; 25(2): 99 - 105. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
