Cloning and characterization of the 5′-flanking sequence for the human DNA topoisomerase II beta gene1

Gene 203 (1997) 113–119

Cloning and characterization of the 5∞-flanking sequence for the human DNA topoisomerase II beta gene1 Shu-Wing Ng *, Yan Liu, Lowell E. Schnipper Hematology/Oncology Division, Harvard Institutes of Medicine, Beth Israel Deaconess Medical Centre, East Campus, 330 Brookline Ave, Boston, MA 02215, USA Received 15 April 1997; received in revised form 25 July 1997; accepted 1 August 1997; Received by C.M. Kane

Abstract Mammalian cells express two isoforms of type II DNA topoisomerase which are the intracellular targets of many structurally diverse antineoplastic agents. The levels of topoisomerase II isozymes are important determinants for the sensitivity of cells to the cytotoxicity of drugs that target topoisomerase II. To investigate whether the expression of topoisomerase II isoforms is coordinated and the mechanisms governing their expression in the context of drug resistance, the 5∞-flanking sequence for the gene of human topoisomerase IIb isoform was cloned and characterized. The 5∞-flanking region has a very high GC content and contains no canonical TATA box element. Two separate transcriptional start sites are located to an adenine and a guanine, 193 and 89 nucleotides, respectively, upstream from the ATG translation initiation codon. Except for a small region immediately upstream of the translation initiation codon, there is no obvious sequence homology between the 5∞-flanking sequences of human topoisomerase IIb gene and the previously described a gene. Transient expression assays with different 5∞- and 3∞-deletions of the 5∞-flanking region of the topoisomerase IIb gene have delineated regions important for transcriptional regulation of the gene. Interestingly, sequences within the first intron also contribute to the promoter activity. Gel mobility shift studies demonstrate that protein factors from the nuclear extracts can bind specifically to the downstream elements and may participate in transcriptional regulation. © 1997 Elsevier Science B.V. Keywords: Genomic clone; Promoter; Drug target; Transcriptional regulation

1. Introduction Mammalian Top II is the nuclear enzyme responsible for the regulation of topological configuration of DNA and is essential for a number of critical cellular processes such as DNA replication, transcription, recombination and mitosis (for reviews see Andoh et al., 1993; Hsieh, 1990). It is also a significant structural component of the nuclear matrix scaffold and plays a major role in folding and organization of nuclear chromatin (Laemmli * Corresponding author. Present address: Department of Obstetrics, Gynecology and Reproductive Biology, Laboratory of Gynecologic Oncology, Brigham and Women’s Hospital, BLI-449A, 221 Longwood Avenue, Boston, MA 022115, USA. Tel.: +1 617 278-0072; Fax: +1 617 975-0818. 1 The nucleotide sequence reported in this paper has been submitted to the GenBank_ Data Bank with accession number U65315. Abbreviations: Top II, type II DNA topoisomerase; kb, kilobase(s); bp, base pair(s); nt, nucleotide; RT–PCR, reverse transcription–polymerase chain reaction; mdr, multidrug resistance; ICE, inverted CCAAT element. 0378-1119/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 03 7 8 -1 1 1 9 ( 9 7 ) 0 0 5 00 - 3

et al., 1992). Top II relieves both torsional and interlocking constraints in DNA by passing an intact helix through a transient double-stranded break made in a second DNA segment, and religation of the broken duplex DNA in an ATP-dependent manner (Osheroff, 1986). This enzyme is also the intracellular target of a number of structurally diverse antitumor agents, including anthracyclines, acridines and epipodophylotoxins (Liu, 1989). These Top II-targeting drugs stabilize the Top II enzyme–DNA intermediate complex and result in DNA lesions and cell death via apoptosis (Andoh et al., 1993; Hsieh, 1990). Two Top II isoforms encoded by different genes are identified in eukaryotic cells. The Top IIa form has subunits of 170 kDa, whereas the Top IIb form has subunits of 180 kDa ( Tan et al., 1992). These two isoforms share strong amino acid sequence homology and have similar in vitro activity, but have different patterns of drug sensitivity, thermal stability, and possibly different functions in eukaryotic cells (Drake et al., 1989). The 170 kDa form is localized in the nucleoplasm

114

S.-W. Ng et al. / Gene 203 (1997) 113–119

and its expression is low in the non-cycling cells that are in G0/G1 phases, but high in the proliferating cells and in the G2/M phases of the cell cycle (Hsiang et al., 1988). The 180 kDa form is localized mostly in the nucleolus and its level is independent of the proliferation profile of the cells ( Woessner et al., 1991). The two isozymes also demonstrate differential expression in tissue development; while Top IIa is highly expressed in spleen and thymus, tissues which are composed of rapidly proliferating cells, the expression of Top IIb is more ubiquitous among tissues ( Tsutsui et al., 1993). The elevated expression of Top IIb in neuronal cells of developing rat brain and the association of active RNA synthesis in neuronal nuclei suggest a possible role for Top IIb in transcription in this tissue ( Tsutsui et al., 1993; Watanabe et al., 1994). Resistance to agents that target Top II can be mediated by pleiotropic mechanisms. Alterations of Top II (Danks et al., 1987) and overexpression of the drug transporter P-glycoprotein of the mdr1 gene (Gottesman and Pastan, 1993) are major mechanisms proposed for drug resistance. Mutations in Top II isozyme proteins have been identified in drug-resistant cell lines (Bugg et al., 1991; Hinds et al., 1991; Chan et al., 1993). Reduced expression of the Top II isoforms, that results in fewer molecular targets and hence reduced proteinassociated DNA breaks generated by the Top II-targeting drugs, has been commonly observed in drugresistant cell lines (Cole et al., 1991; Webb et al., 1991; Jaffrezou et al., 1994; Hosking et al., 1994), as well as in tumors that failed chemotherapy (Beck et al., 1995; Davies et al., 1996). In many of these studies, significant mdr1 overexpression was identified only in the most highly resistant sublines and in recurrent relapses of tumors, suggesting that modification of Top II expression may precede mdr1 overexpression and represent a predominant mechanism underlying acquired drug resistance (Jaffrezou et al., 1994; Hosking et al., 1994; Beck et al., 1995). In order to determine whether expression of top IIa and b genes is coordinated and to understand the mechanisms governing expression of Top II isoforms in association with drug resistance and cell differentiation, we have previously identified and characterized the functions of five inverted CCAAT elements that are conserved in the human and Chinese hamster top IIa promoters (Ng et al., 1995). In this report we describe the isolation and partial characterization of the 5∞-flanking sequence for the human top IIb gene. 2. Experimental and discussion 2.1. Cloning of the 5∞-flanking region of the human top IIb gene A 0.8 kb 5∞-end cDNA probe for the human top IIb gene was prepared by the reverse transcription–polymer-

ase chain reaction (RT–PCR) method, using oligonucleotide primers (5∞-CGCTAAGCTTATGGCCAAGTCGGGTG-3∞ and 5∞-CGCTGGATCCCTTCTAGTCATGAG-3∞) according to the published top IIb cDNA sequence (Jenkins et al., 1992) and mRNA isolated from human Burkitt lymphoma Raji cell line. The cDNA probe was used to screen a human lung fibroblast genomic library (Stratagene Ltd.) under stringent conditions. Out of approximately 106 plaques screened, a total of four overlapping genomic clones were isolated. Restriction mapping of the insert DNA isolated from the positive clones with a 23-mer oligonucleotide (5∞-ACCCAGGTCAGTGCC CCGTTGCC-3∞) which is complementary to the top IIb cDNA sequence 46–68 base pairs downstream of the ATG codon has located the putative 5∞-genomic sequence for the human top IIb gene to a 4.2 kb EcoRI fragment of the positive clones. This fragment was subsequently subcloned into the pBluescript KS vector and analyzed. Sequencing of this fragment from one EcoRI end revealed the first 69 bp of the cDNA coding sequence in the first exon, which was followed by a consensus intron donor sequence ( Fig. 1A). The ATG translation initiation codon of the cDNA is located at a NcoI site of the genomic fragment. 2.2. Determination of the transcription start site of the human top IIb gene RNase protection and primer extension assays were performed to locate the transcription start site. An antisense riboprobe was prepared from XhoI linearized pBluescript KS-4.2 EcoRI template with T7 RNA polymerase and [a-32P]CTP. Fig. 2A demonstrates two protected RNA fragments with sizes of about 260 bases and 155 bases, respectively. The 23-mer oligonucleotide mentioned in Section 2.1. was employed in the primer extension experiment ( Fig. 2B). Although several nonspecific extension products were observed, there were two strong extension products which match the sizes of expected extension products as deduced from the RNase protection results. The longest transcript, as deduced from the sequence ladder produced with the same primer, is initiated from the adenine 193 nucleotides upstream from the ATG codon. Another shorter transcript is initiated at the guanine 89 nucleotides upstream from the ATG. Probably due to the difficulty of the reverse transcriptase to transcribe the GC-rich region, the long transcript in the primer extension experiment ( Fig. 2B) is not as prominent as in the RNase protection assay ( Fig. 2A). Downstream from the transcription start sites, the ATG initiation codon deduced from the cDNA sequence (Jenkins et al., 1992) is the first in-frame ATG codon in the genomic sequence and the sequence, GCCATGG, is perfectly matched to the optimal sequence for initiation by eukaryotic ribosomes

S.-W. Ng et al. / Gene 203 (1997) 113–119

Fig. 1. The 5∞-genomic region of the human top IIb gene. (A) The restriction map of the 4.2 kb EcoRI fragment in the pBluescript KS vector (broken lines) and partial 5∞-genomic sequence of the human top IIb gene are presented. The solid box shows the coding region of the first exon of the top IIb gene, while open boxes represent the noncoding upstream and intronic sequences. Some restriction sites presented are not unique: B, BamHI (in the vector); Bg, BglII; E, EcoRI; H, HindIII; P, PstI; X, XhoI. Bold letters in the sequence represent the coding sequence. Arrows indicate the two transcription start sites and the numbering of the sequence is with reference to the first start site. Consensus sequences potential for transcription factor binding are underlined. (B) Alignment of the 5∞-flanking sequences surrounding the ATG codon for the human top IIa and b genes. Identical nucleotides are marked with colons and the ATG codons are underlined.

(A/GCCATGG) as suggested by Kozak (1986). Based on the relative intensities of the ribonuclease protected fragments, the first transcription start site at the adenine 193 is assigned as the predominant start site and designated hereafter as +1. 2.3. Sequence analysis of the 5∞-flanking region of the human top IIb gene The sequence shown in Fig. 1A was analyzed with MacVector and Lasergene Navigator sequence analysis programs. The sequence has a much higher GC content of 77% when compared with the 62% of the 5∞-flanking sequence of the human top IIa gene (Hochhauser et al., 1992). Like the top IIa gene and other house-keeping genes ( Koller et al., 1991), there is no canonical TATA box sequence and there is a high frequency of CpG dinucleotides (57 in total ) in the top IIb upstream region. These are consistent with the observation that Top IIb is expressed ubiquitously among tissues ( Tsutsui

115

Fig. 2. Determination of the transcription start sites of the human top IIb gene. (A) RNase protection assay of: lane 1, 2 mg of Raji poly(A)+RNA; lane 2, 5 mg of yeast RNA. M is the pBR322 MspI digest size marker (New England Biolabs). The arrows with solid lines indicate the two RNase protected fragments and the arrow with broken line indicates the remains of the riboprobe. (B) Primer extension experiment of: lane 1, 2 mg of Raji poly(A)+RNA; lane 2, 5 mg of yeast RNA. The same primer used in the extension reactions was also employed in generating the 5∞-genomic sequence shown on the left. Arrows indicate the extension products that are consistent with the RNase protection results. Methods: RNase protection assays were performed with the Ambion RPA II @ kit. 1.3×105 cpm (about 260 pg) of labeled RNA was hybridized to 2 mg of Raji poly(A)+ RNA or 5 mg of control yeast RNA. The hybridized products were digested at 37°C for 30 min with a mixture of 20 units of RNase T1 and 1 mg of RNase A. For the primer extension experiment, the 23-mer oligonucleotide was labeled by T4 polynucleotide kinase and [c-32P]ATP. Labeled primer (5×105 cpm) was hybridized to 2 mg of Raji poly(A)+ RNA, or 5 mg of control yeast RNA. The annealed primer was extended by Superscript reverse transcriptase (Gibco BRL).

et al., 1993; Watanabe et al., 1994). A slight sequence identity was found in the 5∞-sequence immediately upstream of the ATG codon for the human top IIa and b genes (Fig. 1B). However, there is no further significant homology between the two sequences (data not shown). It is not known if this slight homology represents a region of shared importance for the transcription or translation of the two genes. The fact that no other notable sequence elements are found in common between the two sequences suggests that the expression of top IIa and b genes may not be coordinated. The region upstream of the ATG of the human top IIb sequence has a 78% GC content. A pair of perfect direct repeat sequences of CGCCGCGC are located at positions −50 to −43 and −23 to −16. Another pair of direct repeat sequences of GGAGCGGCGG are located downstream of the second start site at positions +113 to +122 and +147 to +156. The region down-

116

S.-W. Ng et al. / Gene 203 (1997) 113–119

stream of ATG also demonstrates a high GC content of 76%. Interestingly, several consensus sequence elements for transcription factor binding were identified at the 5∞-end of the first intron. There are two perfect GC boxes (GGGCGG) with the potential for Sp1 binding at +285 and +316, respectively. The reverse strand sequence between +267 and +280 (CGGGGACCAGCCAC ) has 10 out of 14 matches to the consensus CTF/NF-I binding site. There is also one perfect AP2 binding site sequence (CCGCGGGC ) on the reverse strand between +305 and +312. 2.4. Delineation of DNA regions important for top IIb promoter activity To localize DNA regions which are important for in vivo promoter activity, constructs with nested 5∞-deletions of the 5∞-flanking sequence were placed upstream of the luciferase coding sequence and transfected into HeLa cells. Luciferase activities were assayed from the cell lysates (Fig. 3A). Decreasing luciferase activity was observed with increasing 5∞-deletions of the 5∞-flanking sequence. An 8-fold decrease in promoter activity was observed when the upstream sequence was deleted from −1000 to −500, indicating that important transactivation element(s) reside in this region. When the upstream sequence was reduced further to −14, there was an enhancement of promoter activity back to about 20% of the wild-type level, suggesting that the region between −500 and −14 may have element(s) which negatively regulate the promoter activity. Interestingly, there was still significant residual promoter activity when the 5∞-flanking region was deleted through the transcription start site (the Top 2b (+56/+326)-LUC construct and the Top2b (+193/+326)-LUC construct). The results of these two constructs suggest that besides the upstream elements, positive element(s) are present downstream of the transcription start site and confer the residual promoter activity. To delineate the 3∞-boundary of the functional promoter, 3∞-deletions of the 5∞-flanking sequence were prepared from the pTop2b (−14/+326)-LUC construct, which has about 20% of wild-type promoter activity. The results of transient expression assays with the new constructs (Fig. 3B) demonstrate that deletion of most of the first intron sequence in the pTop2b (−14/+279)LUC construct significantly eliminated the promoter activity to background level. Complete deletion of the intronic sequence in the pTop2b (−14/+196)-LUC construct gave a further decrease of the already diminished luciferase activity. Potential consensus sequences for the binding of transcription factors have been identified in the first intron. Deletions of the pTop2b (−14/+326)LUC construct to remove the Sp1 and AP2 sequences in the first intron abolished the residual promoter activ-

Fig. 3. Transient expression assays with different deletions of the 5∞-flanking sequence of the human top IIb gene in HeLa cells. (A) Expression assays with 5∞-deletion constructs and (B) expression assays with 3∞-deletion constructs. At least four separate transfections were performed and the luciferase activities were normalized to the activities of the co-transfected b galactosidase-encoding plasmids. The average luciferase activity for each construct was presented relative to the Top2b (−3400/+326)-LUC construct. The solid boxes in the 5∞-flanking regions represent the first exon, while open boxes represent the upstream and intronic sequences. Methods: a 3.7 kb BglII–BamHI fragment (the BamHI site was from the vector) derived from the pBluescript KS-4.2 EcoRI plasmid DNA was cloned in the sense orientation upstream of the luciferase-coding sequence of the pXP2 reporter vector (Nordeen, 1988). The resulting plasmid is designated as pTop2b (−3400/+326)-LUC and it is the construct that contains the longest 5∞-flanking sequence employed in the transfection assays. To prepare 5∞-deletion constructs, a 1.3 kb HindIII–EcoRI fragment with the EcoRI site blunted by Klenow fill-in reaction was isolated from the 4.2 kb EcoRI genomic fragment and cloned into the pXP2 plasmid at the HindIII and blunt-ended XhoI sites. The resulting plasmid is designated as pTop2b (−1000/+326)-LUC. Other 5∞-deletion constructs were obtained by digestion of the pTop2b (−1000/+326)-LUC plasmid with HindIII and one internal restriction enzyme, blunt-ending, and self-ligation. The internal restriction enzymes used for generating pTop2b (−500/+326)-LUC were NotI, XhoI for the pTop2b (−14/+326)-LUC, NruI for the pTop2b (+56/+326)-LUC, and NcoI for the pTop2b (+193/+326)-LUC. To prepare the 3∞-deletion construct pTop2b (−14/+279)-LUC, the pTop2b (−14/+326)-LUC plasmid was digested with BglII, the enzyme for a cloning site 3∞- to the top IIb genomic fragment in the construct, and the internal enzyme SmaI. The resulting vector-containing DNA was blunt-ended by Klenow fill-in and self-ligated. To prepare the construct pTop2b (−14/+196)-LUC, the pTop2b (−14/+326)-LUC plasmid was digested with BglII and the internal enzyme NcoI, and the vectorcontaining DNA was recovered, blunt-ended by Klenow enzyme fill-in and self-ligated. Test constructs (5 mg) were cotransfected with 5 mg of b-galactosidase expression plasmid, pCH110 (Pharmacia), into 3×105 growing HeLa cells by calcium phosphate precipitation method (Gorman et al., 1982). After 48 h, cells were harvested and lysates were prepared. Luciferase assays were performed using the kit from Promega Corporation and the luciferase activity was measured in a Turner Designs Luminometer Model 20. Amounts of lysates employed for the luciferase activity assays were normalized to the b-galactosidase activities as described (Ng et al., 1995).

S.-W. Ng et al. / Gene 203 (1997) 113–119

ity (Fig. 3B). The results of these transfection assays suggest that sequence elements in the first intron can contribute to the promoter activity. The combined results of the transient expression assays also suggest that the sequence between −14 and +326 is the minimal 5∞-flanking sequence for basal promoter activity. 2.5. Analysis of the binding activity to the downstream sequence by gel mobility shift assays Since regulation by downstream elements is new for the topoisomerase genes, specific protein factors that bind to the downstream DNA sequence were investigated. A 141 bp fragment containing the first exon and the 64 bp first intron sequence was radiolabeled and gel mobility shift assays were performed with HeLa nuclear extracts. Protein factors from the nuclear extracts bound to the DNA fragment and migrated as distinct complexes on the native polyacrylamide gel (Fig. 4). All the complexes were competed out when a 20- or 60-fold molar excess of the same unlabeled fragments were included in the binding reaction ( lanes 3 and 4, Fig. 4). The complexes formed are sequence-specific because they

Fig. 4. Gel mobility shift assays of the top IIb downstream sequence with HeLa nuclear extracts. The 141 bp NcoI–BglII fragment containing the 5∞-end of the first intron was labeled for the assays. Lane 1 was incubation without nuclear extracts, while the other lanes were reactions with 2 mg of HeLa nuclear extracts. Competitor DNAs used: lanes 1 and 2, none; lane 3, 20-fold molar excess of the same unlabeled DNA fragment; lane 4, 60-fold molar excess of the same unlabeled DNA fragment; lane 5, 20-fold molar excess of the nonspecific pBluescript DNA; lane 6, 60-fold molar excess of the nonspecific pBluescript DNA. Arrows indicate the specific complexes formed. Methods: the probe was isolated from the pTop2b (−1000/+326)LUC plasmid and was 3∞-end labeled at the NcoI site with Klenow fragment of DNA polymerase I and [a-32P]dCTP. Procedures for the preparation of nuclear extracts and gel mobility shift assays have been described previously (Ng et al., 1995).

117

were not depleted by the addition of a 20-fold molar excess of nonspecific pBluescript competitor DNA ( lane 5 in Fig. 4). The intensities of the upper two complexes diminished when 60-fold molar excess of nonspecific competitor DNA was included ( lane 6 in Fig. 4), suggesting that these two complexes are not very stable. Since consensus AP2 and Sp1 binding sequences were identified in the first intron, the binding of nuclear extract proteins to the downstream sequence was competed by the addition of double-stranded oligonucleotides containing respectively AP2 and Sp1 consensus sequences ( Fig. 5A). Competition by AP2 and Sp1 competitor DNAs resulted in disappearance of most of the bands, especially the lowest specific complex, suggesting that they were formed by AP2 and Sp1 proteins. Some residual protein complexes were still observable

Fig. 5. Characterization of protein binding to the top IIb downstream sequence. (A) Competition gel mobility shift assays with the labeled top IIb downstream sequence. Note that the batch of nuclear extracts employed in this experiment contained more protein than that used in Fig. 4, and additional bands of complexes are more visible. The competitor DNAs used were: lane 1, none; lane 2, 50-fold molar excess of AP2 competitor oligonucleotide; lane 3, 50-fold molar excess of Sp1 competitor oligonucleotide; lane 4, 25-fold molar excess of both AP2 and Sp1 competitor oligonucleotides; lane 5, 50-fold molar excess of the nonspecific DNA. The arrow indicates the lowest migrating complex that disappeared in the competition assays with AP2 and Sp1 competitor oligonucleotides. (B) Gel mobility shift assays with purified AP2 protein. Instead of HeLa nuclear extracts, 0.25 footprinting units of AP2 protein were added in the incubation with the radiolabeled DNA. The competitor DNAs used were: lane 1, none; lane 2, 50-fold molar excess of AP2 competitor oligonucleotide; lane 3, 50-fold molar excess of the nonspecific DNA. Arrows indicate the specific AP2 bound complexes. (C ) Gel mobility shift assays with purified Sp1 protein. 0.25 footprinting units of Sp1 protein were added in the incubation with the radiolabeled DNA. The competitor DNAs used were: lane 1, none; lane 2, 50-fold molar excess of Sp1 competitor oligonucleotide; lane 3, 50-fold molar excess of the nonspecific DNA. Arrows indicate the two specific Sp1 bound complexes. Methods: the conditions of experiments were the same as for Fig. 4. Double-stranded AP2 and Sp1 competitor oligonucleotides and the purified proteins were purchased from Promega.

118

S.-W. Ng et al. / Gene 203 (1997) 113–119

in the presence of both competitor DNAs ( lane 4 of Fig. 5A), which might be formed by other protein factors. To further demonstrate that AP2 and Sp1 bind to the downstream sequence, purified AP2 and Sp1 proteins were incubated with the radiolabeled downstream DNA fragment and gel mobility shift assays were performed (Fig. 5B and C ). Specific binding of both proteins to the DNA fragment was observed. The slow mobility of Sp1 bound complexes in Fig. 5C is also observed in the characterization of other Sp1 binding elements ( Yang et al., 1995), which may be caused by multimerization of the Sp1 protein in the binding (Pascal and Tjian, 1991).

3. Conclusions (1) We have cloned and sequenced the 5∞-flanking region of the human top IIb gene. The derived sequence has a very high GC content and contains no CCAAT and TATA boxes. There is no significant sequence homology between human top IIa and b promoters. The first intron of the top IIb gene contains many potential binding sites for transcription factors. (2) Two transcriptional start sites were located at 193 and 89 nucleotides upstream from the ATG translation initiation codon, respectively. (3) Transient transfection assays with different deletions of the 5∞-flanking sequence have indicated the presence of positive element(s) between −1000 and −500 and probably negative element(s) in the region between −500 and −14. In conjunction with the gel mobility shift assays, the results of transfection assays with 3∞-deletion constructs suggest that specific proteins binding to the downstream element(s) in the first intron may contribute to the top IIb promoter activity. Involvement of intronic elements in the modulation of gene expression has been reported for many other genes (Alder et al., 1992; Lazar et al., 1994; Zastawny and Ling, 1993). One interesting example is the finding that the first intron sequence of the hamster homolog of the drug resistance P-glycoprotein mdr1 gene, pgp1, possesses enhancer activity and is conserved both between family members and across species ( Zastawny and Ling, 1993). It may be of interest to examine if the first intronic sequences of the human top IIb gene is also conserved across species. (4) Gel mobility shift assays showed that there is specific binding of proteins to the downstream elements. Competition of the binding by AP2 and Sp1 oligonucleotides and gel mobility shift assays with purified Sp1 and AP2 proteins provide evidence for the involvement of these two proteins in the complex formation. The transcription factor AP2, unlike the ubiquitously expressed Sp1 factor, is transcribed predominantly in cells of neural and epidermal lineages (Leask et al., 1991). The

fact that top IIb expression is elevated in the developing brain ( Tsutsui et al., 1993; Watanabe et al., 1994) may suggest the involvement of AP2 in regulating top IIb expression in neuronal cells. Further studies are being conducted to characterize possible involvement of AP2 and/or other factors in governing gene expression during cell differentiation and in conferring drug resistance.

Acknowledgement We thank Drs. Dong-Er Zhang and Daniel G. Tenen for the provision of pXP2 vector and Dr. K.B. Tan for the gift of Raji cells. The excellent assistance of Josephine Ng during the preparation of the manuscript is greatly appreciated. This work was supported in part by National Cancer Institute Grant CA70216 (to S.W.N.).

References Alder, H., Yoshinouchi, M., Prystowsky, M.B., Appasamy, P., Baserga, R., 1992. A conserved region in intron 1 negatively regulates the expression of the PCNA gene. Nucleic Acids Res. 20, 1769–1775. Andoh, T., Ikeda, H., Oguro, M. ( Eds.), 1993. Molecular Biology of DNA Topoisomerases and its Application to Chemotherapy. CRC Press, Boca Raton, FL. Beck, J., Handgretinger, R., Dopfer, R., Klingebiel, T., Niethammer, D., Gekeler, V., 1995. Expression of mdr1, mrp, topoisomerase II alpha/beta, and cyclin A in primary or relapsed states of acute lymphoblastic leukaemias. Br. J. Haematol. 89, 356–363. Bugg, B.Y., Danks, M.K., Beck, W.T., Suttle, D.P., 1991. Expression of a mutant DNA topoiosmerase II in CCRF-CEM human leukemic cells selected for resistance to teniposide. Proc. Natl. Acad. Sci. USA 88, 7654–7658. Chan, V.T.W., Ng, S.-W., Eder, J.P., Schnipper, L.E., 1993. Molecular cloning and identification of a point mutation in the topoisomerase II cDNA from an etoposide-resistant Chinese hamster ovary cell line. J. Biol. Chem. 268, 2160–2165. Cole, S.P.C., Chanda, E.R., Dicke, F.P., Gerlach, J.H., Mirski, S.E.L., 1991. Non-P-glycoprotein-mediated multidrug resistance in a small cell lung cancer cell line: evidence for decreased susceptibility to drug-induced DNA damage and reduced levels of topoisomerase II. Cancer Res. 51, 3345–3352. Danks, M.K., Yalowich, J.C., Beck, W.T., 1987. Atyptical multiple drug resistance in a human leukemic cell line selected for resistance to teniposide ( VM-26). Cancer Res. 47, 1297–1301. Davies, S.L., Popert, R., Coptcoat, M., Hickson, I.D., Masters, J.R., 1996. Response to epirubicin in patients with superficial bladder cancer and expression of the topoisomerase II alpha and beta genes. Int. J. Cancer 65, 63–66. Drake, F.H., Hofmann, G.A., Bartus, H.F., Mattern, M.R., Crooke, S.T., Mirabelli, C.K., 1989. Biochemical and pharmacological properties of p170 and p180 forms of topoisomerase II. Biochemistry 28, 8154–8160. Gorman, C.L., Moffat, L.F., Howard, B., 1982. Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2, 1044–1051. Gottesman, M.M., Pastan, I., 1993. Biochemistry of multidrug resistance mediated by the multidrug transporter. Annu. Rev. Biochem. 62, 385–427.

S.-W. Ng et al. / Gene 203 (1997) 113–119 Hinds, M., Deisseroth, K., Mayes, J., Altschuler, E., Jansen, R., Ledley, F.D., Zwelling, L.A., 1991. Identification of a point mutation in the topoisomerase II gene from a human leukemia cell line containing an amsacrine-resistant form of topoisomerase II. Cancer Res. 51, 4729–4731. Hochhauser, D., Stanway, C.A., Harris, A.L., Hickson, I.D., 1992. Cloning and characterization of the 5∞-flanking region of the human topoisomerase IIa gene. J. Biol. Chem. 267, 18961–18965. Hosking, L.K., Whelan, R.D., Shellard, S.A., Davies, S.L., Hickson, I.D., Danks, M.K., Hill, B.T., 1994. Multiple mechanisms of resistance in a series of human testicular teratoma cell lines selected for increasing resistance to etoposide. Int. J. Cancer 57, 259–267. Hsiang, Y.-H., Wu, H.-Y., Liu, L.F., 1988. Proliferation-dependent regulation of DNA topoisomerase II in cultured human cells. Cancer Res. 48, 3230–3235. Hsieh, T.S., 1990. Mechanistic aspects of type II DNA topoisomerases. In: Cozzarelli, N.R., Wang, J.C. ( Eds.), DNA Topology and Its Biological Effects. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 243–263. Jaffrezou, J.P., Chen, G., Duran, G.E., Kuhl, J.S., Sikic, B.I., 1994. Mutation rates and mechanisms of resistance to etoposide determined from fluctuation analysis. J. Natl. Cancer Inst. 86, 1152–1158. Jenkins, J.R., Ayton, P., Jones, T., Davies, S.L., Simmons, D.L., Harris, A.L., Sheer, D., Hickson, I.D., 1992. Isolation of cDNA clones encoding the b isozyme of human DNA topoisomerase II and localization of the gene to chromosome 3p24. Nucleic Acids Res. 20, 5587–5592. Koller, E., Hayman, A.R., Trueb, B., 1991. The promoter of the chicken alpha 2( VI ) collagen gene has features characteristic of house-keeping genes and of proto-oncogenes. Nucleic Acids Res. 19, 485–491. Kozak, M., 1986. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44, 283–292. Laemmli, U.K., Kas, E., Poljak, L., Adachi, Y., 1992. Scaffold-associated regions: cis-acting determinants of chromatin structural loops and functional domains. Curr. Opin. Genet. Dev. 2, 275–285. Lazar, J., Desvergne, B., Zimmerman, E.C., Zimmer, D.B., Magnuson, M.A., Nikodem, V.M., 1994. A role for intronic sequences on expression of thyroid hormone receptor alpha gene. J. Biol. Chem. 269, 20352–20359.

119

Leask, A., Byrne, C., Fuchs, E., 1991. Transcription factor AP2 and its role in epidermal-specific gene expression. Proc. Natl. Acad. Sci. USA 88, 7948–7952. Liu, L.F., 1989. DNA topoisomerase poisons as antitumor drugs. Annu. Rev. Biochem. 58, 351–375. Ng, S.-W., Eder, J.P., Schnipper, L.E., Chan, V.T.W., 1995. Molecular cloning and characterization of the promoter for the Chinese hamster DNA topoisomerase IIa gene. J. Biol. Chem. 270, 25850–25858. Nordeen, S.K., 1988. Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6, 454–458. Osheroff, N., 1986. Eukaryotic topoisomerase II. J. Biol. Chem. 261, 9944–9950. Pascal, E., Tjian, R., 1991. Different activation domains of Sp1 govern formation of multimers and mediate transcriptional synergism. Genes Dev. 5, 1646–1656. Tan, K.B., Dorman, T.E., Falls, K.M., Chung, T.D.Y., Mirabelli, C.K., Crooke, S.T., Mao, J., 1992. Topoisomerase IIa and topoisomerase IIb genes: characterization and mapping to human chromosomes 17 and 3, respectively. Cancer Res. 52, 231–234. Tsutsui, K., Tsutsui, K., Okada, S., Watanabe, M., Shohmori, T., Seki, S., Inoue, Y., 1993. Molecular cloning of partial cDNAs for rat DNA topoisomerase II isoforms and their differential expression in brain development. J. Biol. Chem. 268, 19076–19083. Watanabe, M., Tsutsui, K., Tsutsui, K., Inoue, Y., 1994. Differential expressions of the topoisomerase II alpha and II beta mRNAs in developing rat brain. Neurosci. Res. 19, 51–57. Webb, C.D., Latham, M.D., Lock, R.B., Sullivan, D.M., 1991. Attenuated topoisomerase II content directly correlates with a low level of drug resistance in a Chinese hamster ovary cell line. Cancer Res. 51, 6543–6549. Woessner, R.D., Mattern, M.R., Johnson, C.K., Drake, F.H., 1991. Proliferation- and cell cycle-dependent differences in expression of the 170 kilodalton and 180 kilodalton forms of topoisomerase II in NIH-3T3 cells. Cell Growth Diff. 2, 209–214. Yang, X., Fyodorov, D., Deneris, E.S., 1995. Transcriptional analysis of acetylcholine receptor alpha 3 gene promoter motifs that bind Sp1 and AP2. J. Biol. Chem. 270, 8514–8520. Zastawny, R.L., Ling, V., 1993. Structural and functional analysis of 5∞ flanking and intron 1 sequences of the hamster P-glycoprotein pgp1 and pgp2 genes. Biochim. Biophys. Acta 1173, 303–313.