JBC Ideal method for primary cell transfection

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M208296200 on September 12, 2002

J. Biol. Chem., Vol. 277, Issue 46, 43873-43880, November 15, 2002
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
277/46/43873    most recent
M208296200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Whetstine, J. R.
Right arrow Articles by Matherly, L. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Whetstine, J. R.
Right arrow Articles by Matherly, L. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The Human Reduced Folate Carrier Gene Is Regulated by the AP2 and Sp1 Transcription Factor Families and a Functional 61-Base Pair Polymorphism*

Johnathan R. WhetstineDagger , Teah L. Witt§, and Larry H. MatherlyDagger §

From the Dagger  Department of Pharmacology, and the § Experimental and Clinical Therapeutics Program, Barbara Ann Karmanos Cancer Institute, Wayne State University School of Medicine, Detroit, Michigan 48201

Received for publication, August 13, 2002, and in revised form, September 11, 2002

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Recently, our laboratory reported an intricate regulation of the human reduced folate carrier (hRFC) gene, involving multiple promoters and noncoding exons. We localized promoter activity to a 452-bp GC-rich region upstream of noncoding exon A, including a 47-bp basal promoter with a CRE/AP-1-like consensus element that bound the bZip family of DNA-binding proteins (e.g. CREB-1 and c-Jun). We now report that three nearly identical tandem repeats (49-61 bp) in the hRFC-A upstream region are involved in regulating promoter activity. By in vitro binding assays, multiple transcription factors (e.g. AP2 and Sp1/Sp3) bound this region. When AP2 was cotransfected with the hRFC-A reporter construct into HT1080 cells, promoter activity increased 3-fold. In Drosophila SL2 cells, Sp1 transactivated promoter A and showed synergism with CREB-1. However, c-Jun was antagonistic to the effects of Sp1. A sequence variant in the hRFC-A repeated region was identified, involving an exact duplication of a 61-bp sequence. This variant had an allelic frequency of 78% in 72 genomic DNAs and resulted in a 63% increase in promoter activity. These results identify important regions in the hRFC-A promoter and critical roles for AP2 and Sp1, in combination with the bZip transcription factors. Moreover, they document a functionally novel polymorphism that increases promoter activity and may contribute to interpatient variations in hRFC expression and effects on tissue folate homeostasis and antitumor response to antifolates.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Reduced folates are essential cofactors involved in one-carbon transfer reactions that lead to the synthesis of nucleotides and amino acids required for cell proliferation. Since mammals lack the capacity for de novo synthesis of reduced folates, these derivatives must be obtained from the diet (1).

The primary route for the transport of reduced folates into mammalian cells involves the ubiquitously expressed reduced folate carrier (RFC)1 (2-5). RFC is also the major cellular uptake system for antifolates used for cancer therapy, including methotrexate (Mtx) and Tomudex (2-5). High levels of transport by RFC are critical to the antitumor effects of antifolate inhibitors, and impaired transport results in decreased antitumor activity and a drug-resistant phenotype (6-10). Moreover, impaired Mtx transport results from sustained exposures of human and murine tumor cells to Mtx in vitro (6-10) and has been reported to develop in murine tumor cells in vivo during Mtx chemotherapy (11).

In studies of childhood acute lymphoblastic leukemia and osteosarcoma, diseases for which Mtx remains a cornerstone of modern therapies (12, 13), an important role for human RFC (hRFC) was, likewise, implied. For instance, we previously reported an 88-fold range of hRFC expression in B-precursor acute lymphoblastic leukemia lymphoblasts and a proportional loss of Mtx transport capacity with changes in hRFC (14). Similarly, Guo et al. (13) showed that low levels of hRFC gene expression in osteosarcomas accompanied decreased Mtx transport and a poor prognosis. These studies suggest that relative levels of hRFC gene expression can potentially have a major impact on clinical outcome following chemotherapy with Mtx.

The differential expression of hRFC in these patient populations could be a result of altered promoter usage or transcriptional activity but could also be associated with interindividual variations in the regulatory regions of the hRFC gene. A better understanding of the molecular mechanisms that regulate hRFC gene expression and function in malignant and normal cells is critical, since this could foster improvements in the clinical use of antifolates for cancer therapy. Moreover, this could shed light on the intra- and extracellular signals that influence patterns of hRFC expression in normal human tissues and contribute to various pathophysiologic conditions associated with folate deficiency (e.g. cardiovascular disease (15), fetal abnormalities (16), neurological disorders (17), and cancer (18)).

Recent studies suggest a remarkably complex regulation of hRFC the gene expression in tissues and tumors (19). We found that hRFC gene is ubiquitously but differentially expressed in human tissues and is regulated by seven noncoding exons (designated A1, A2, A, B, C, D, and E) and at least three promoters (A, B, and C) (19). Several of the noncoding exons are capable of alternative splicing. Altogether, there are as many as 18 unique hRFC transcripts with distinct 5'-UTRs linked to a common open reading frame (19). The exact function of each 5'-UTR has yet to be determined, however, an effect on hRFC mRNA stabilities, intracellular targeting, and/or translation efficiencies can be envisaged (20-28).

We recently began to identify and characterize critical transcription factor families involved in regulating the hRFC-B and -A promoters. For the basal promoter B, transactivation involved binding of Sp1 and Sp3 to a GC-box element. For hRFC-A, regulation involved binding of different members of the bZip superfamily, including c-Jun, CREB-1, and ATF-1, to a CRE/AP1-like element in the minimal promoter (29). Most recently, we reported that the frequent use of promoter B in malignant tissues was associated with its regulation by multiple transcription factors, including oncogenic proteins such as c-Myc and Ikaros, and by histone acetylation (30).2 Accordingly, cell- and tissue-specific usage of alternate hRFC promoters/noncoding exons could serve to regulate relative expression and function of hRFC in response to differences in folate levels within tissues, intracellular distributions of transcription factors, or epigenetic events. It is this complexity of controls that probably ensures adequate levels of hRFC transcripts (and protein) in response to metabolic requirements for folates, or other tissue- or cell-specific signals.

In this report, we expand our initial analysis of hRFC-A, a major promoter in liver, bone marrow, and immortalized cell lines (19), as well as in primary malignant cells.2 Our results demonstrate that AP2 and Sp1 binding to a series of unique tandem nucleotide repeats in the hRFC-A upstream region is critical to promoter transactivation. Moreover, we document the presence of a high frequency polymorphism in the repeated region, involving the insertion of an additional 61 nucleotides that influence hRFC-A transcriptional activity. This is the first report of a functional polymorphism in the hRFC gene that may contribute to interpatient variations in hRFC gene expression and response to antifolate cancer chemotherapeutics.

    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Chemicals and Reagents-- [gamma -32P]ATP (3000 Ci/mmol) was purchased from PerkinElmer Life Sciences. Synthetic oligonucleotides were purchased from Genosys Biotechnologies, Inc. (The Woodlands, TX). Lipofectin® was purchased from Invitrogen. Restriction and modifying enzymes, reporter gene vectors (pGL3-Basic, pGL3-Pro, pRLSV40), and other molecular biologicals were obtained from Promega (Madison, WI). The pHSV-AP2alpha , pHSV-AP2beta , and pHSV-AP2gamma constructs were provided by Michael A. Tainsky (Karmanos Cancer Institute). The Mzf-1 expression construct (in pCDNA3) was provided by Yubin Ge (Karmanos Cancer Institute). The following Drosophila SL2 expression vectors were used in cotransfections: pPacO and pPacSp1 (Robert Tjian, University of California, Berkley); pPac CREB (Timothy Osborne, University of California, Irvine); and pPac Jun (John Noti, Guthrie Foundation).

hRFC-Luciferase Reporter Constructs-- The full-length hRFC-A/-903 promoter construct (positions -903 to -277) was previously described by Zhang et al. (31). The promoter A construct with the additional 61-nucleotide repeat inserted between positions -715 and -714 (hRFC/Repeat 1 +) was previously described by Whetstine and Matherly (29).

Specific DNA binding sites in promoter A were mutated by both standard PCR and "Soeing" PCR (32), using the GC-Rich Kit (Roche Molecular Biochemicals) with Tgo polymerase (Roche Molecular Biochemicals). The hRFC-A/-903 construct was used as a template. The hRFC-A/-903 mE construct, containing the mutated E-box, was prepared by a standard PCR using the -903mE1-2 (XhoI) (5'-ACTGCCTCGAGGGTACCGGTGGGGAACGGGGCCAATGGGCCGATTGTCGGGGGCTGCGGGGTGTCTCGGGGCCCT-3'; mutant positions noted by underlines) and RFCO5 (HindIII) (positions -278 to -319, 5'-CTAGCTAAGCTTGCTCCAAGGGGAAGTTGCACCTACAAAGCTTCTGGCACAGG-3') primers. Standard PCR was also used to make the hRFC-A/-903 mAP2(1) and hRFC-A/-903 mMzf-1(1) mutant constructs using the -903 mAP2(1) (XhoI) (5'-ACTGCCTCGAGGGTACCGGTGGGGAACGGGGCCACGGGGCCGCGTGTCGGGGGCTGCAAAGTGTCTCGGGGCCCT-3'; mutant positions noted by underlines) and -903 mMzf-1(1) (XhoI) (5'-ACTGCCTCGAGGGTACCGGTGTTTAACGGGGCCACGGGGCCGCGTGTCGGGGGCTGCGGGGTGTCTCGGGGCCCT-3'; mutant positions noted by underlines) primers with the RFCO5 (HindIII) primer. Standard PCR conditions were 95 °C for 15 s and 72 °C for 120 s for 40 cycles.

"Soeing" PCR was used to mutate specific elements in promoter A using the primers shown in Table I. Aliquots of the primary PCR products were mixed and amplified with the ProA (XhoI) and RFCO5 (HindIII) primers for the secondary PCR amplifications. For primary PCR amplifications, conditions were 95 °C for 15 s and 72 °C for 120 s for 40 cycles, and the secondary PCR amplifications used 95 °C for 15 s and 72 °C for 120 s for 40 cycles. The amplicons were isolated from 2% LE-agarose, digested with XhoI and HindIII, and subcloned into pGL3-Basic in the sense orientation. Mutations were verified by automated DNA sequencing.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Primers used to create mutant promoter A plasmids
Mutant constructs were prepared by Soeing PCR using homologous mutant sense and antisense primers. For each mutant, two amplifications were performed (ProA(XhoI) + antisense primer and sense primer + RFCO5 (HindIII). The primary reactions were then combined and used in a secondary amplification reaction with ProA (XhoI) and RFCO5.

Cell Culture, Transient Transfections, and Reporter Gene Assays-- The HT1080 human fibrosarcoma cell line (American Type Culture Collection) was cultured, transiently transfected with hRFC-A promoter constructs, and assayed for luciferase activity exactly as previously described (29). Cotransfections generally included up to 200 ng of AP2 (alpha , beta , and gamma ) and/or Mzf-1 expression constructs. Firefly luciferase activities were normalized with beta -galactosidase activity (pRSV-beta -gal; Promega).

Drosophila SL2 cells (5 × 105 cells/12.5 cm2 flask) were transfected with hRFC-A promoter constructs (1 µg) and 200 ng of total expression vector (10 ng of pPacSp1 with either 190 ng of pPac CREB, pPac Jun, or pPacO) as previously described (29). After 48 h, the cells were lysed, and luciferase activities were assayed with the Single Luciferase reporter assay system (Promega). Relative luciferase activities were normalized to protein concentrations of the cell lysates. For all transfections, three or more experiments were performed in duplicate.

DNase I Footprinting-- The Pro A oligonucleotide probe spanning positions -940 to -570 of the hRFC-A promoter was PCR-amplified with P430 (5'-TCCGGGAGCCCCAGGGCAGCCGCCCCGCCG-3') and Del7 (5'-CTAGCTAAGCTTCTGCCAGGCGCCCGGGCCACCGCGACCCCCGCGGAGACC-3') primers from a previously described genomic clone (31). The amplicon was subcloned into pGEM-T Easy vector and then isolated by digestion with EcoRI. The Pro A probe (300 ng) was end-labeled with [gamma -32P]ATP as previously described (1-3 × 106 cpm) (29), digested with HindIII (for sense strand) or KpnI (for antisense strand), and incubated with or without 120 µg of HT1080 nuclear extract (33) for 30 min. The binding reaction mixtures were treated with 50 µl of magnesium/calcium solution (Core Fooprinting Kit; Promega) for 1 min, followed by RQ1 DNase I for 4 min and, finally, 90 µl of stop buffer. Reactions were treated with Proteinase K (20 µg) for 45 min at 42 °C, phenol/chloroform-extracted, and precipitated overnight with carrier tRNA. The DNase I-treated reactions were resuspended in 5 µl and loaded onto an 8% denaturing sequencing gel. Product sizes were determined from the migration of a sequencing reaction prepared with Sequenase 7.0 and sequencing primers to the 5'- and 3'-ends of this fragment. The images were visualized on an Amersham Biosciences PhosphorImager.

Gel Mobility Shift Assays-- Gel shift assays were performed exactly as previously described (29). 12 µg of HT1080 nuclear extract was used in each binding reaction. The [gamma -32P]ATP end-labeled probes (0.01 pmol, corresponding to 5 × 106 to 7 × 106 cpm) used in the binding reactions were as follows: hRFC-A/-903, 5'-GGTACCGGTGGGGAACGGGGCCACGGGGCCGCGTGTCGGGGGCTGCGGGGTGTCTCGGGGCCCT-3'; hRFC-A/R1, 5'-CTGGGGTCCGCGGGGCCCTGGGGAGGGTGCGGGGCGTGGGCCGGGGTCTGCGGTCTGCAGC-3'; and hRFC-A/R2, 5'-CTGGGGTCTGGGGGGCCCTGGGGAGGGTGCGGGGCGTGGCCGGGGTCTGCGGTCTGCAGC-3' (consensus sequences are underlined). The competition assays were performed with a 150-fold molar excess of unlabeled wild-type oligonucleotides, mutant oligonucleotides (hRFC-A/R1 mAP2(2), 5'-CTGGGGTCCGCAAAGCCCTGGGGAGGGTGCGGGGCGTGGGCCGGGGTCTGCGGTCTGCAGC-3'; hRFC-A/R1 mMzf-1(2), 5'-CTGGGGTCCGCGGGGCCCTTTTTAGGGTGCGGGGCGTGGGCCGGGGTCTGCGGTCTGCAGC-3'; hRFC-A/R1 mAP2(2)/mMzf-1(2), 5'-CTGGGGTCCGCAAAGCCCTTTTTAGGGTGCGGGGCGTGGGCCGGGGTCTGCGGTCTGCAGC-3'; mutations in consensus sequences are in italic type), or oligonucleotides containing conserved DNA binding sites for specific transcription factors (E-box, AP2, and GC-box (Santa Cruz Biotechnologies, Inc., Santa Cruz, CA); Mzf-1 1-4 and 5-13 (34)). DNA-protein complexes were supershifted with USF-1 (Santa Cruz Biotechnology), Sp1 (Geneka Biotechnologies), and AP2 (Santa Cruz Biotechnology) antisera. The gels were dried and visualized by autoradiography.

Polymorphism Analysis-- Genomic DNAs (gDNAs) from 72 normal (i.e. nondisease) individuals were provided by M. Norris and M. Haber (Children's Cancer Institute Australia for Medical Research, Randwick, Australia). The gDNAs (50 ng) were amplified with the P430 and RFCO5 primers to the hRFC-A promoter region at 95 °C for 15 s, 63 °C for 45 s, and 72 °C for 120 s for 40 cycles. The amplicons were visualized on 2.5% LE-agarose gels stained with ethidium bromide and isolated, and the DNA sequence was confirmed by automated sequencing. The allelic frequency of the 61-bp insertion polymorphism (see below) was calculated according to the Hardy-Weinberg equilibrium.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

In Vitro Binding of USF-1, AP2, and Mzf-1 to the hRFC-A Promoter

Our previous studies of the regulation of the hRFC gene confirmed that promoter activity was associated with a 452-bp GC-rich region immediately upstream from exon A that was devoid of a TATA-box but contained a number of putative transcription elements (31). The basal hRFC-A promoter was localized to within 47 bp (positions -501 to -455; see Fig. 1) and included a CRE/AP1-like consensus sequence capable of binding the bZIP family of transcription factors, including CREB-1, ATF-1, and c-Jun (29, 31). However, the contributions of individual transcription factors capable of binding cis elements upstream of this minimal region were not established.


View larger version (45K):
[in this window]
[in a new window]
  		Fig. 1.   Nucleotide sequence for the hRFC-A promoter. The upstream nucleotide sequence (-903 to -606) with tandem repeated regions in promoter A is shown in the sense direction, and the numbering corresponds to the promoter A fragment in the hRFC-A/-903 reporter construct used in transient transfections. The basal promoter region (-501 to -376) for hRFC-A is also indicated (29). The potential transcription factor binding sites are boxed and labeled with the name of the transcription factor family type. The tandem repeats are indicated with R1 (single underline), R2 (double underline), and R3 (broken line). The italicized letters correspond to the 5'-end of exon A as previously described (19, 31).

A characteristic feature of the original promoter A sequence involves the presence of three nearly identical tandem repeats of 49-61 bp and spanning 170 bp of upstream sequence from position -775 to -606 (Fig. 1). Potential binding sites identified in this region include E-box (-884 to -867), Mzf-1 (site 1, -897 to -889; site 2, -765 to -752), AP2 (site 1, -864 to -853; site 2, -771 to -760), and GC-box (site 1, -689 to -674; site 2, -629 to -613) elements (Fig. 1). In transient transfections of hRFC-A deletion clones, striking variations in promoter activity were observed upon deleting this region (31). The most significant changes occurred between positions -903 to -809 and between -721 and -558 (31).

In vitro binding assays were used to identify the key proteins that interact with the first 298 bp of the hRFC-A promoter between positions -903 and -606. This region was analyzed by DNase I footprinting with HT1080 nuclear extracts, and gel shift assays were then used to identify the probable transcription factors capable of binding to the hRFC-A upstream sequence (Fig. 2).


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 2.   Cis elements within promoter A bind multiple transcription factors in vitro. DNase I footprinting (positions -940 to -570) and gel shift assays were performed with HT1080 nuclear extracts as described under "Materials and Methods." Panel A, the DNase I footprinted region from -883 to -852 is shown in the left panel. Lanes 1 and 3, probe alone; lane 2, probe and nuclear extract. The protected regions correspond to potential AP2 and E-box sites. In the center and right panels, the hRFC-A/-903 double-stranded oligonucleotide (containing Mzf-1(1), E-box, and AP2(1) elements) was used as a probe for gel shift assays with HT1080 nuclear extracts. Lane 4, probe alone (without nuclear extract); lanes 5 and 10, probe plus nuclear extract; lane 6, 150-fold molar excess of the unlabeled wild type hRFC-A/-903 oligonucleotide (self); lanes 7, 8, and 11, 150-fold molar excess of the unlabeled E-box consensus sequence (lane 7), Mzf-1 consensus sequences (Mzf1 1-4; lane 8), or AP2 consensus sequence (lane 11); lanes 9 and 12, antiserum to USF-1 or AP2 beta , respectively, was added to the binding reactions. Panel B, the left panel shows the DNase I footprinted region from position -777 to -735, corresponding to Repeat 1 (R1). Lanes 1 and 3, probe alone; lane 2, probe and nuclear extract. The protected regions correspond to potential Mzf-1 (Mzf-1(2)) and AP2 (AP2(2)) sites. The center and right panels show gel shift results with a 32P-end-labeled hRFC-A/R1 oligonucleotide incubated with nuclear extract. Lane 4, probe alone; Lanes 5 and 11, hRFC-A/R1 plus nuclear extract; Lane 6, 150-fold molar excess of unlabeled wild type hRFC-A/R1 oligonucleotide (self); lanes 7 and 8, unlabeled Mzf-1 consensus sequences (Mzf1 1-4 and Mzf1 5-13, respectively); lane 9, AP2 consensus sequence; lane 10, AP2 beta  antiserum was added to the binding reactions; lanes 11-13, 32P-end-labeled mutant R1 mAP2(2) (lane 11), R1mMzf-1(2) (lane 12), or R1 mAP2(2)/mMzf-1(2) (lane 13) oligonucleotides were incubated with nuclear extract. C, the DNase I footprinted region from -676 to -711 is shown in the left panel, corresponding to Repeat 2 (R2). Lanes 1 and 3, probe alone; lane 2, probe and nuclear extract. The protected region corresponds to a potential GC/GT-box (GC-box(1)). In the right panel, the gel shifts show double-stranded oligonucleotide hRFC-A/R2 incubated without (lane 4) or with (lane 5) nuclear extract. The complexes were competed with the unlabeled hRFC-A/R2 oligonucleotide (lane 6) and a GC-box (lane 7) consensus sequence. In lanes 8 and 9, antibodies were added to the nuclear extract including that for Sp1 (lane 8) and Sp3 (lane 9). For A-C, the complexes for each gel shift probe are labeled with letters and arrowheads, and the supershifted complexes are marked with arrows. The nonspecific band is noted with a cross. Gel-shifted complexes were resolved on a 5% nondenaturing gel, and DNase I footprints were resolved on a denaturing 8% gel.

-903 to -840-- DNase I footprint analysis identified two protected regions from position -883 to -867 (a putative E-box) and from -865 to -852 (a putative AP2 site; AP2(1)) (Fig. 2A, compare lanes 1 and 3 with lane 2). Protein binding to a putative Mzf-1 element (positions -897 to -889) was also localized to the most 5'-end of the footprint probe (data not shown).

The proteins that bound the -903 to -840 region in HT1080 nuclear extracts were identified with the hRFC-A/-903 double-stranded oligonucleotide probe in gel shift assays. Three DNA-protein complexes were detected (labeled A-C in Fig. 2A, lanes 5 and 10) and effectively competed by excess unlabeled hRFC-A/-903 probe (lane 6). Complex A was competed with an E-box consensus oligonucleotide (lane 7) and was supershifted with an antibody to an E-box-binding protein (USF-1; lane 9) (35, 36). An AP2 consensus oligonucleotide competed with complex B (lane 11), and an AP2 beta  antibody resulted in a decreased DNA/protein signal and a supershifted complex (lane 12, arrow). Although complex C was effectively competed with a Mzf-1 consensus sequence (Mzf-1 1-4) (lane 8), the identity of the protein in this complex could not be verified due to the lack of a commercially available Mzf-1 antibody. When each of these elements was individually mutated in the hRFC-A/-903 oligonucleotide and the 32P-mutant probes used directly in gel shift assays, protein binding was abolished (data not shown). Collectively, these data confirm binding of USF1, AP2, and a Mzf-1-like protein to the -903 to -840 region of the hRFC-A promoter.

-775 to -715-- Footprint analysis of the -775 to -715 region identified protected regions spanning positions -775 to -761 and -758 to -742 (Fig. 2B, compare lanes 1 and 3 with lane 2). These sites correspond to candidate AP2 (-771 to -760; AP2(2)) and Mzf-1 (-765 to -752; Mzf-1(2)) elements (37) that overlap in the first repeated region (designated R1 in Fig. 1). When an end-labeled double-stranded hRFC-A/R1 oligonucleotide probe (positions -775 to -715) was incubated with a HT1080 nuclear extract, a series of DNA-protein complexes was detected (complexes A-D in Fig. 2B, lanes 5 and 11) and were competed by a molar excess of unlabeled probe (lane 6, self). Mzf-1 competitor oligonucleotides (Mzf-1 1-4 and 5-13) added to the binding reactions in molar excess resulted in a slightly diminished signal for all complexes (lanes 7 and 8, respectively). An AP2 consensus oligonucleotide potently competed for complexes A and C while perturbing complex D (lane 9). Similarly, an AP2 antibody (AP2 beta  antiserum) perturbed complexes A, C, and D and resulted in a supershifted complex (lane 10; noted with an arrow).

The identities of the DNA-protein complexes were further assessed on gel shifts with end-labeled hRFC-A/R1 oligonucleotides with mutations in the potential AP2(2) and/or Mzf-1(2) sites (lanes 12-14). Thus, mutation of the AP2(2) element (hRFC-A/R1 mAP2(2)) resulted in decreased levels of all four complexes (lane 12). When the Mzf-1 site was mutated (hRFC-A/R1 mMzf-1(2)), the intensities of complexes A, B, and D were all decreased, whereas complex C increased in intensity (lane 13). When both the Mzf-1 and AP2 elements were mutated (hRFC-A/R1 mAP2(2)/mMzf-1(2)), there was a complete abolition of protein binding (lane 14). These results demonstrate the interactive binding of AP2 and Mzf-1-like proteins to the AP2(2) and Mzf-1(2) elements in the hRFC-A Repeat 1 region.

-714 to -655 and -654 to -606-- A single protected region was identified by DNase I footprinting of the hRFC-A/Repeat 2 antisense sequence (R2 in Fig. 1; positions -714 to -655). This protected sequence (positions -677 to -693; Fig. 2C) corresponds to a putative GC/GT-box (GC-box(1), positions -689 to -674) (38, 39) and is identical to the protected GC-box(2), (positions -629 to -613) in the hRFC-A/Repeat 3 sequence (R3 in Fig. 1; positions -654 to -606) (data not shown). The hRFC-A/R2 oligonucleotide probe was used in gel shift assays to identify the DNA-protein complexes between positions -714 and -665. The hRFC-A/R3 oligonucleotide, including the identical 48 bp in Repeat 3, was also synthesized and used in gel shift assays to identify the complexes between positions -654 and -606.

For both probes, three specific complexes (designated A-C) were detected (Fig. 2C, lane 5, shows data for the hRFC-A/R2 probe; the hRFC-A/R3 probe results in identical complexes when compared with hRFC-A/R2, so the data are only shown for the hRFC-A/R3 probe). Since the DNA-protein complexes were completely competed by a GC-box consensus oligonucleotide and were supershifted with Sp1 and Sp3 antibodies (lanes 7-9, respectively; supershifts are noted with arrows), all complexes clearly involved members of the Sp family of transcription factors. When the GC-box elements were mutated, the competitions seen with the wild type oligonucleotides were abolished (data not shown).

Mutagenesis of Transcription Factor Binding Elements in the hRFC-A Upstream Region

To further assess the importance of the major binding sites identified in the hRFC-A upstream region by in vitro binding assays, the individual elements were mutated. The mutant hRFC-A promoter constructs in pGL3 Basic vector were transiently transfected into HT1080 cells, and luciferase activities were compared with that of the wild type hRFC-A/-903 construct. When the Mzf-1(1), E-box, Mzf-1(2), and GC-box(1) sites (Fig. 1) were individually mutated, insignificant changes in promoter A activity were observed (<10%; p > 0.05). However, the effects of other individual mutations were significant (p < 0.05) and ranged from a 23 ± 5% (for AP2(1)) or 25 ± 3% (for AP2(2)) increase in promoter activity to a 37 ± 6% decrease in activity (GC-Box(2)). Interestingly, when both the AP2(2) and Mzf-1(2) elements were mutated, a striking 37 ± 5% decrease in promoter activity was observed, consistent with our gel shift findings (Fig. 2B) that imply a functional interaction between AP2 and Mzf-1-like factors bound to these sites. Collectively, our mutation results suggest that the AP2(1) and AP2(2), Mzf-1(2), and GC-box(2) binding sites are important for promoter A activity.

Cotransfections of Promoter A with AP2 and Mzf-1 Expression Constructs

AP2 and Mzf-1 expression constructs were cotransfected into HT1080 cells with the hRFC-A/-903 promoter construct to assess their effects on promoter activity. Whereas Mzf-1, alone, was transcriptionally inert in this assay, AP2 (alpha , beta , and gamma ) markedly stimulated (~3-fold) luciferase activities (Fig. 3). When Mzf-1 was expressed in combination with the AP2 isoforms, there was no further increase in luciferase activity from that with AP2 alone (data not shown). These data further support the notion of a critical role for the family of AP2 proteins in the regulation of hRFC-A transcription.


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 3.   Effects of co-expressed AP2 and Mzf-1 on hRFC-A promoter activity. HT1080 cells were cotransfected with the hRFC-A/-903 promoter construct and 200 ng of Mzf-1 or 100 ng of the AP2 expression vectors. Luciferase activities were assayed with the Single Luciferase Kit and normalized to beta -galactosidase activities. Relative luciferase activities for cotransfections of HSV-AP2alpha , HSV-AP2beta , HSV-AP2gamma , and pCDNA3-Mzf-1 constructs with the hRFC-A/-903 reporter construct were compared with that for a cotransfection of hRFC-A/-903 and empty cytomegalovirus vector. The error bars represent S.E.

Sp1 Regulates Promoter A Activity in Drosophila SL2 Cells

Sp1 primarily acts as a transactivating factor (38, 39). Sp3 exists as three isoforms, generated by different translation starts (40), all of which can act (to varying degrees) as activators or repressors of transcription, depending on the cell and promoter context (38-44). Since the GC-box(2) element appears to be critical for promoter A activity (see above), Sp-null Drosophila SL2 cells (45-47) were used to directly assess the impact of exogenous Sp transcription factors on hRFC-A transactivation. Although the hRFC-A basal promoter does not respond to Sp1 cotransfection (data not shown) (29), the full-length hRFC-A/-903 construct was stimulated ~46-fold over pPacO (Fig. 4A). The long isoform of Sp3 (but not the short Sp3 isoforms) was also able to activate promoter A, although it was less effective than Sp1 (~6-fold; data not shown). When GC-box(2) was mutated (Repeat 3 mGC(2)), the net stimulation by Sp1 was decreased by 55%; however, there was no significant effect of mutating GC-box(1) (Repeat 3 mGC(1)) (Fig. 4B). Interestingly, the AP2(2)/Mzf-1(2) double mutant resulted in a 34% decrease in the Sp1 stimulation (Fig. 4B). Overall, these data further confirm that the GC-box(2) element is functional and that Sp1 binding to this site has a direct and profound effect on hRFC-A promoter transactivation. Moreover, they suggest that the extent of Sp1 stimulation is positively affected by the binding of AP2 and Mzf-1-like proteins to the upstream AP2(2)/Mzf-1(2) elements.


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 4.   Effects of Sp1, CREB-1, and/or c-Jun on promoter A activity in Drosophila SL2 (D. SL2) cells. A, Drosophila SL2 cells were cotransfected with 1 µg of the hRFC-A promoter constructs in pGL3-Basic and pPacSp1 (10 ng), pPac CREB (190 ng), pPac Jun (190 ng), and/or empty pPacO vector. Transfections were supplemented with pPacO vector so that the total plasmid DNA was 200 ng. Luciferase activities were expressed relative to pPacO alone. B, the hRFC-A/-903 or the mutant constructs, Repeat 1 mAP2(2)/mMzf-1(2), Repeat 2 mGC(1), and Repeat 3 mGC(2), were cotransfected with pPacSp1 (10 ng) or pPacO alone. Luciferase activities for the mutant constructs were compared with that for the wild type hRFC-A/-903 reporter construct. Luciferase activities were assayed after 48 h as described under "Materials and Methods." The error bars represent S.E.

Based on our earlier findings that the bZip DNA binding proteins regulate hRFC-A basal promoter activity (29), expression vectors for c-Jun (48) and CREB-1 (49) were cotransfected into Drosophila SL2 cells with the hRFC-A/-903 construct in the presence or absence of pPacSp1. Whereas CREB-1 alone was transcriptionally inert, in combination with Sp1, a dramatic 5-fold increase in promoter activity was observed over that with Sp1 alone (~210-fold over pPacO; Fig. 4A). In contrast, c-Jun resulted in a modest 3-fold stimulation of promoter activity over that with pPacO and suppressed the stimulation by Sp1 by 70% (Fig. 4A). The Sp3 proteins were incapable of either stimulating or repressing CREB-1 or c-Jun (data not shown), indicating that the interactions between Sp1 and the bZip proteins and their effects on the hRFC-A promoter are specific. Thus, binding of different members of the bZip family of transcription factors, presumably to the hRFC-A basal promoter (29), can exert potent and opposing effects on the magnitude of hRFC-A transactivation, accompanying Sp1 binding to the upstream GC-box(2) element.

Identification of a Polymorphism in the hRFC-A Promoter

Our original description of the hRFC upstream region was based on DNA sequence from a genomic clone isolated from a human peripheral blood leukocyte genomic library (31). While subsequently amplifying across the hRFC-A tandem repeats (positions -775 to -606; Fig. 1) from gDNAs prepared from cell lines and peripheral blood samples, a distinct doublet (722 and 661 bp) was detected (Fig. 5A). Both amplicons were isolated and sequenced. The smaller fragment was identical to our original published sequence for promoter A (GenBankTM accession number AF046920) (31). The 722-bp product was, likewise, identical with the exception of an additional 61 bp that is identical to the R1 sequence (including the overlapping AP2(2) and Mzf-1(2) elements) inserted between positions -715 and -714 (Figs. 1 and 5B).


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 5.   Identification and characterization of a 61-bp polymorphism in the hRFC-A promoter. A 61-bp polymorphism was identified in human gDNA samples by PCR amplification and resulted in increased promoter A activity. A, promoter A sequence was amplified from gDNAs from nondiseased patients with P430 and RFC05 primers as described under "Materials and Methods." The amplicons were separated on a 2% agarose gel and visualized with ethidium bromide. The amplicons, including the 61-bp repeat (722 bp, R1+), are easily resolved from the R1- amplicon (661 bp). The data shown include three homozygous R1+, one homozygous R1-, and two heterozygous R1+/R1- genotypes. B, a schematic representation of the polymorphic region in promoter A is shown. The putative transcription factor binding sites in hRFC-A, as described under "Results," are also indicated. C, a comparison of relative luciferase activities for HT1080 cells transiently transfected with the hRFC-A/Repeat 1 + or hRFC-A/Repeat 1 - reporter constructs. Data are reported as relative firefly luciferase activities, normalized to Renilla luciferase activities. The error bars represent S.E.

To evaluate whether this insertion is a mutation or a polymorphism, gDNAs from 72 normal (nondisease) patients were amplified and analyzed on agarose gels as seen in Fig. 5A. Surprisingly, 43 (63%) of the specimens were homozygous for the 722-bp amplicon ("Repeat 1 +"), and 21 (30%) were heterozygous ("Repeat 1 +/-"). Only five individuals (7%) were homozygous for the 661-bp amplicon ("Repeat 1 -"). Based on the Hardy-Weinberg equilibrium, the allelic frequencies for Repeat 1 + and Repeat 1 - genotypes are 78 and 22%, respectively. Thus, the additional 61-bp repeat sequence in the hRFC-A promoter is a previously unrecognized polymorphism.

To establish the effect of the insertion polymorphism on promoter A activity, full-length hRFC-A promoter constructs with and without the 61-bp insertion (designated hRFC-A Repeat 1 + and Repeat 1 -, respectively) were subcloned into pGL3 Basic and transfected into HT1080 cells. The hRFC-A Repeat 1 + construct resulted in a highly significant 63 ± 12% increase in luciferase activity compared with the activity of the Repeat 1 - construct (p < 0.05) (Fig. 5C). Since the 61-bp insertion generates additional AP2(2) and Mzf-1(2) binding sites (Fig. 5B), cotransfections were performed with the hRFC-A Repeat 1 + construct and AP2 expression vectors (alpha , beta , and gamma ) for comparison with the results of the AP2 cotransfections with the Repeat 1 - construct (Fig. 3). A 3-fold stimulation of luciferase activity was observed for all the AP2 isoforms with the Repeat 1 + promoter A construct (data not shown). Thus, although the presence of the additional 61-bp R1 repeat clearly augments promoter A activity above an elevated base-line level in transient transfections, the relative stimulation by AP2 is identical to that for the Repeat 1 - construct.

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

We previously localized the minimal hRFC-A promoter to within 47 bp, including a CRE/AP-1-like element that bound the bZip family of transcription factors (e.g. CREB-1, ATF-1, and c-Jun) (29). In this report, we characterize the role of the hRFC-A upstream region, which includes a novel series of tandem nucleotide repeats, and describe a polymorphic 61-bp insertion in the repeated region.

Using DNase I footprinting, protected regions corresponding to transcription factor binding sites were identified in the first 298 bp of promoter A. In vitro binding of AP2-, USF-, Mzf-1-, and Sp-related proteins to the candidate sites in the -903 to -606 region was confirmed by gel shift assays. The AP2 sites were of particular interest, since mutations of either the AP2(1) (positions -864 to -853) or AP2(2) (positions -771 to -760) elements significantly altered promoter A activity in transient transfections, and cotransfections with the AP2 isoforms (alpha , beta , and gamma ) resulted in a 3-fold stimulation for the hRFC-A/-903 promoter construct. These data strongly suggest a critical transactivating role for the AP2 isoforms.

AP2 designates a family of genes (alpha , beta , and gamma ) that encodes a series of 48-52-kDa proteins capable of recognizing the consensus sequence 5'-GCCNNNGGC-3' (50). The AP2 genes each exhibit cell- and tissue-specific expression patterns that determine their effects on transcription of various genes involved in normal tissue development and differentiation (50). Reports that AP2 transactivation can be influenced by the binding of other transcription factors (e.g. Sp1, Ikaros, and bZip proteins) (51-54) are of particular interest, given the close proximity of the binding sites for Mzf-1 (i.e. Mzf-1(2)) and AP2 (i.e. AP2(2)) in the hRFC-A sequence. In our study, mutations of the overlapping AP2 and Mzf-1 elements in the hRFC-A Repeat 1 region resulted in a variety of effects on promoter activity, ranging from a lack of effect (Mzf-1(2)) or a stimulatory response (AP2(2)) to a potent suppression of promoter activity when both elements were mutated together. Interactive binding between AP2 and Mzf-1 to the hRFC-A R1 region (positions -775 to -715) was also suggested by the results of gel shift assays. Due to the absence of Mzf-1 antibody, the identity of the factor(s) bound to Mzf-1(2) could not be unequivocally confirmed in vitro. Therefore, it is possible that a protein other than Mzf-1 may be involved in regulating promoter A.

Although Sp1 had no effect on transcription from the 47-bp hRFC-A minimal hRFC-A promoter (29), this factor markedly stimulated activity for the full-length hRFC-A construct and synergized with CREB-1. Interestingly, these effects on promoter activity were specific to Sp1, since the long isoform of Sp3 was less effective and failed to synergize with CREB-1, and the short Sp3 isoforms were completely inert. However, c-Jun was notably antagonistic to the transactivating effects of Sp1. These results suggest that the relative ratios of the bZip transcription factors can potentially regulate the hRFC-A promoter in different cell or tissue types, in direct support of our earlier finding of disparate effects of bZip proteins (e.g. CREB-1, ATF-1, and c-Jun) on the hRFC-A basal promoter (29). Furthermore, the response of the full-length promoter to the bZip DNA-binding proteins would be profoundly influenced by the differential expression of the Sp family of factors, thus providing an intricate mechanism for regulating cell- or tissue-specific transcription of hRFC-A.

There is a burgeoning interest in the identification of functional polymorphisms that could influence the activity or expression of folate-dependent gene products. For instance, a high frequency "low function" polymorphic variant of 5,10-methylene tetrahydrofolate reductase (C677T) has been associated with elevated serum levels of homocysteine, increased risk of cardiovascular disease and neural tube defects, and a lower risk of colon cancer (55). For human thymidylate synthase, double or triple tandem repeats of a 28-bp cis-acting enhancer element has been associated with increased levels of gene expression (56). Our laboratory recently reported that the only established polymorphism for the hRFC gene (a G to A at position 80 in the hRFC coding sequence) results in a modified carrier with an Arg to His substitution at position 27 with virtually identical transport properties for reduced folates and antifolates to the wild type hRFC protein (57).

In this report, we identify the first functional polymorphism for hRFC. We originally reported that the hRFC-A promoter contained three tandem repeated sequences (31); however, we now show that up to 78% of the hRFC alleles in a small cohort of patients have an additional R1 insertion between nucleotides -715 and -714. The Repeat 1 + genotype corresponds to the hRFC-A promoter sequence originally identified by Williams and Flintoff (GenBankTM accession number AF077610) (58). Since AP2 appears to have a critical role in promoter A activity, the polymorphic repeat could potentially influence the level of promoter transactivation. Indeed, in HT1080 cells, the 61-bp nucleotide repeat resulted in a statistically significant increase in promoter activity (63%) over the hRFC-A/Repeat 1 - reporter construct and was stimulated to an identical extent in cotransfections with AP2 expression vectors over this elevated base-line value. These data strongly argue that the cellular composition of transcription factors and the number of tandem repeats in the hRFC-A promoter can profoundly alter patterns of hRFC gene expression in cells and tissues.

Since 5'-UTRs can influence transcript stability and translation efficiency (20-28), the enhanced promoter activity for the Repeat 1 + genotype could further increase the fraction of hRFC transcripts with exon A sequence, which in turn could increase total levels of hRFC protein due to post-transcriptional effects. For instance, Wong et al. (59) reported dramatically increased amounts of hRFC protein to transcripts in Chinese hamster ovary cells transfected with the hRFC cDNA containing the exon A 5'-UTR sequence when compared with the hRFC cDNA with the exon B 5'-UTR sequence, raising the possibility of such post-transcriptional controls. Thus, the joint effects of variations in the levels of the bZip, AP2, and Sp families of proteins and the presence or absence of the functional hRFC-A polymorphism could effectively modulate transcription from promoter A. Combined with the possibility of post-transcriptional controls, this could result in a wide range of hRFC protein and relative uptakes for reduced folates and antifolates.

In conclusion, our studies document the critical role of a novel series of tandem repeats in the hRFC-A upstream region, including functionally important AP2 and Sp1 elements, and the occurrence of a high frequency 61-bp polymorphism involving one of the repeat sequences that alters promoter activity. The importance of our findings draws from the critical role of hRFC in folate homeostasis and cancer chemotherapy with antifolate drugs and the possibility that the homozygosity of the Repeat 1 + genotype and/or increased chromosome 21 ploidy could substantially impact net hRFC levels and transport activities in tumor cells and normal tissues. Accordingly, increased transcriptional activity from the polymorphic hRFC-A promoter could contribute to untoward toxicities in patients receiving antifolate chemotherapy or explain patterns of elevated hRFC expression and increased therapeutic responses. Based on the central role of hRFC in folate accumulation in mammalian cells and tissues (2-5) and the potential contributions of folate deficiencies to chromosomal instability (18), cardiovascular disease (15), and fetal abnormalities (16), population studies of this hRFC-A promoter variant are clearly warranted and may have far reaching significance to human health and disease.

    ACKNOWLEDGEMENTS

We thank Drs. Michael A. Tainsky and Yubin Ge for providing expression constructs for AP2 and Mzf-1, respectively; Dr. Guntram Suske for pPacSp3 and pPacUSp3; Dr. Robert Tjian for pPacO and pPacSp1; Drs. John Noti and Timothy Osborne for pPac Jun and pPac CREB, respectively; and Drs. Murray Norris and Michelle Haber for the gDNAs from nondiseased patients.

    FOOTNOTES

* This work was supported by NCI, National Institutes of Health, Grant CA53535.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Experimental and Clinical Therapeutics Program, Karmanos Cancer Institute, 110 E. Warren Ave., Detroit, MI 48201. Tel.: 313-833-0715 (ext. 2407); Fax: 313-832-7294; E-mail: matherly@karmanos.org.

Published, JBC Papers in Press, September 12, 2002, DOI 10.1074/jbc.M208296200

2 J. R. Whetstine, R. M. Flatley, and L. H. Matherly, submitted for publication.

    ABBREVIATIONS

The abbreviations used are: RFC, reduced folate carrier; Mtx, methotrexate; hRFC, human RFC; 5'-UTR, 5'-untranslated region; gDNA, genomic DNA; CRE, cAMP-response element; CREB, CRE-binding protein.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1. Stokstad, E. L. R. (1990) in Folic Acid Metabolism in Health and Disease (Picciano, M. F. , Stokstad, E. L. R. , and Gregory, J. F., eds) , pp. 1-21, Wiley-Liss, New York
2. Goldman, I. D., and Matherly, L. H. (1985) Pharmacol. Ther. 28, 77-100[CrossRef][Medline] [Order article via Infotrieve]
3. Sirotnak, F. M. (1985) Cancer Res. 45, 3992-4000[Free Full Text]
4. Jansen, G. (1999) in Anticancer Development Guide: Antifolate Drugs in Cancer Therapy (Jackman, A. L., ed) , pp. 293-321, Humana Press Inc., Totowa, NJ
5. Matherly, L. H. (2001) Prog. Nucleic Acids Res. Mol. Biol. 67, 131-162[Medline] [Order article via Infotrieve]
6. Schuetz, J. D., Matherly, L. H., Westin, E. H., and Goldman, I. D. (1988) J. Biol. Chem. 263, 9840-9847[Abstract/Free Full Text]
7. Wong, S. C., McQuade, R., Proefke, S. A., and Matherly, L. H. (1997) Biochem. Pharmacol. 53, 199-206[CrossRef][Medline] [Order article via Infotrieve]
8. Jansen, G., Mauritz, R., Drori, S., Sprecher, H., Kathman, I., Bunni, M., Priest, D. G., Noordhuis, P., Schornagel, J. H., Pinedo, H. M., Peters, G. J., and Assaraf, Y. G. (1998) J. Biol. Chem. 273, 30189-30198[Abstract/Free Full Text]
9. Gong, M., Yess, J., Connolly, T., Ivy, S. P., Ohnuma, T., Cowanm, K. H., and Moscow, J. A. (1997) Blood 89, 2494-2499[Abstract/Free Full Text]
10. Wong, S. C., Zhang, L., Witt, T. L., Proefke, S. A., Bhushan, A., and Matherly, L. H. (1999) J. Biol. Chem. 274, 10388-10394[Abstract/Free Full Text]
11. Sirotnak, F. M., Moccio, D. M., Kelleher, L. E., and Goutas, L. (1981) Cancer Res. 41, 4447-4452[Abstract/Free Full Text]
12. Matherly, L. H., and Taub, J. W. (1996) Leukemia Lymphoma 21, 359-368[Medline] [Order article via Infotrieve]
13. Guo, W., Healey, J. H., Meyeers, P. A., Ladanyai, M., Huvos, A. G., Bertino, J. R., and Gorlick, R. (1999) Clin. Can. Res. 5, 621-627[Abstract/Free Full Text]
14. Zhang, L., Taub, J. W., Williamson, M., Wong, S. C., Hukku, B., Pullen, J., Ravindranath, Y., and Matherly, L. H. (1998) Clin. Can. Res. 4, 2169-2177[Abstract]
15. Refsum, H., Ueland, P., Nygard, O., and Vollset, S. E. (1998) Annu. Rev. Med. 49, 31-62[CrossRef][Medline] [Order article via Infotrieve]
16. Butterworth, C. E., Jr., and Bendich, A. (1996) Annu. Rev. Nutr. 16, 73-97[CrossRef][Medline] [Order article via Infotrieve]
17. Serot, J. M., Christmann, D., Dubost, T., Bene, M. C., and Faure, G. C. (2001) J. Neural Transm. 108, 93-99[CrossRef][Medline] [Order article via Infotrieve]
18. Choi, S-W., and Mason, J. B. (2000) J. Nutrition 130, 129-132[Abstract/Free Full Text]
19. Whetstine, J. R., Flatley, R. M., and Matherly, L. H. (2002) Biochem. J. 367, 629-640[CrossRef][Medline] [Order article via Infotrieve]
20. Roberts, S. J., Chung, K-N., Nachmanoff, K., and Elwood, P. C. (1997) Biochem. J. 326, 439-447[Medline] [Order article via Infotrieve]
21. Elwood, P. C., Nachmanoff, K., Saikawa, Y., Page, S. T., Pacheco, P., Roberts, S., and Chung, K-N. (1997) Biochemistry 36, 1467-1478[CrossRef][Medline] [Order article via Infotrieve]
22. Chen, L., Qi, H., Korneberg, J., Garrow, T. A., Choi, Y-J., and Shane, B. (1996) J. Biol. Chem. 271, 13077-13087[Abstract/Free Full Text]
23. Roy, K., Mitsugi, K., and Sirotnak, F. M. (1996) J. Biol. Chem. 271, 23820-23827[Abstract/Free Full Text]
24. Turner, F. B., Andreassi II, J. L., Ferguson, J., Titus, S., Tse, A., Taylor, S. M., and Moran, R. G. (1999) Cancer Res. 59, 6074-6079[Abstract/Free Full Text]
25. Turner, F. B., Taylor, S. M., and Moran, R. G. (2000) J. Biol. Chem. 275, 35960-35968[Abstract/Free Full Text]
26. Kocarek, T. A., Zangar, R. C., and Novak, R. F. (2000) Arch. Biochem. Biophys. 376, 180-190[CrossRef][Medline] [Order article via Infotrieve]
27. Fiaschi, T., Chiarugi, P., Veggi, D., Raugei, G., and Ramponi, G. (2000) FEBS Lett. 473, 42-46[CrossRef][Medline] [Order article via Infotrieve]
28. Fournier, B., Trunong-Bolduc, Q. C., Zhang, X., and Hooper, D. C. (2001) J. Bacteriol. 183, 2367-2371[Abstract/Free Full Text]
29. Whetstine, J. R., and Matherly, L. H. (2001) J. Biol. Chem. 276, 6350-6358[Abstract/Free Full Text]
30. Whetstine, J. R., Witt, T., and Matherly, L. H. (2002) Proc. Am. Assoc. Cancer Res. 43, 101
31. Zhang, L., Wong, S. C., and Matherly, L. H. (1998) Biochem. J. 332, 773-780[Medline] [Order article via Infotrieve]
32. Horton, R. M., Cai, Z. L., Ho, S. N., and Pease, L. R. (1990) BioTechniques 8, 528-535[Medline] [Order article via Infotrieve]
33. Ausuble, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1998) Current Protocols in Molecular Biology , pp. 12.1.1-12.1.9, John Wiley and Sons, Inc., New York
34. Morris, J. F., Hromas, R., and Rauscher, F. J. (1994) Mol. Cell. Biol. 14, 1786-1795[Abstract/Free Full Text]
35. Gregor, P. D., Sawadogo, M., and Roeder, R. G. (1990) Genes Dev. 4, 1730-1740[Abstract/Free Full Text]
36. Pognonec, P., and Roeder, R. G. (1991) Mol. Cell. Biol. 11, 5125-5136[Abstract/Free Full Text]
37. Quandt, K., Frech, K., Karas, H., Wingender, E., and Werner, T. (1995) Nucleic Acids Res. 23, 4878-4884[Abstract/Free Full Text]
38. Philipsen, S., and Suske, G. (1999) Nucleic Acids Res. 27, 2991-3000[Abstract/Free Full Text]
39. Suske, G. (1999) Gene (Amst.) 238, 291-300[CrossRef][Medline] [Order article via Infotrieve]
40. Kennett, S. B., Udvadia, A. J., and Horowitz, J. M. (1997) Nucleic Acids Res. 25, 3110-3117[Abstract/Free Full Text]
41. Hagen, G., Müller, S., Beato, M., and Suske, G. (1992) Nucleic Acis Res. 20, 5519-5525[Abstract/Free Full Text]
42. Hagen, G., Müller, S., Beato, M., and Suske, G. (1994) EMBO J. 13, 3843-3851[Medline] [Order article via Infotrieve]
43. Ihn, H., and Trojanowska, M. (1997) Nucleic Acids Res. 25, 3712-3717[Abstract/Free Full Text]
44. Chen, S. J., Artlett, C. M., Jimenez, S. A., and Varga, J. (1998) Gene (Amst.) 215, 101-110[CrossRef][Medline] [Order article via Infotrieve]
45. Courey, A. J., and Tjian, R. (1988) Cell 55, 887-898[CrossRef][Medline] [Order article via Infotrieve]
46. Courey, A. J., Holtzman, D. A., Jackson, S. P., and Tjian, R. (1989) Cell 59, 827-836[CrossRef][Medline] [Order article via Infotrieve]
47. Pascal, E., and Tjian, R. (1991) Genes Dev. 5, 1646-1656[Abstract/Free Full Text]
48. Noti, J. D. (1997) J. Biol. Chem. 272, 24038-24045[Abstract/Free Full Text]
49. Dooley, K. A., Bennett, M. K., and Osborne, T. F. (1999) J. Biol. Chem. 274, 5285-5291[Abstract/Free Full Text]
50. Hilger-Eversheim, K., Moser, M., Schorle, H., and Buettner, R. (2000) Gene (Amst.) 260, 1-12[CrossRef][Medline] [Order article via Infotrieve]
51. Xu, Y., Porntadavity, S., and St. Clair, D. K. (2002) Biochem. J. 362, 401-412[CrossRef][Medline] [Order article via Infotrieve]
52. Ito, T., Nomura, S., Okada, M., Katsumata, Y., Kikkawa, F., Rogi, T., Tsujimoto, M., and Mizutani, S. (2002) Biochem. Biophys. Res. Commun. 290, 1048-1053[CrossRef][Medline] [Order article via Infotrieve]