DNA replication-related elements cooperate to enhance promoter activity of the drosophila DNA polymerase alpha 73-kDa subunit gene.

An analysis was carried out on the promoter region of the Drosophila DNA polymerase alpha 73-kDa subunit gene and the factor(s) activating the promoter. Transcription initiation sites were newly identified in the region downstream of the previously determined sites. Full promoter activity resided within the region from -285 to +129 base pairs with respect to the newly determined major site. Within this region, we found three sequences identical or similar to the DNA replication-related element (DRE), 5'-TATCGATA, which is known as a common promoter-activating element for the Drosophila DNA polymerase alpha 180-kDa subunit gene and the proliferating cell nuclear antigen gene. These sites were located at positions -77 to -70 (DREalpha-I), -44 to -37 (DREalpha-II), and +3 to +10 (DREalpha-III). Footprinting analysis using the recombinant DRE-binding factor (DREF) or Kc cell nuclear extract demonstrated that DREF can bind to all three DRE-related sites. Introduction of mutation in even one of the three DRE-related sequences caused extensive reductions of the promoter activity and also the DREF-binding activity of the promoter-containing fragment. The results indicate that the three DREF-binding sites cooperate to enhance promoter activity of the DNA polymerase alpha 73-kDa subunit gene.

An analysis was carried out on the promoter region of the Drosophila DNA polymerase ␣ 73-kDa subunit gene and the factor(s) activating the promoter. Transcription initiation sites were newly identified in the region downstream of the previously determined sites. Full promoter activity resided within the region from ؊285 to ؉129 base pairs with respect to the newly determined major site. Within this region, we found three sequences identical or similar to the DNA replication-related element (DRE), 5-TATCGATA, which is known as a common promoter-activating element for the Drosophila DNA polymerase ␣ 180-kDa subunit gene and the proliferating cell nuclear antigen gene. These sites were located at positions ؊77 to ؊70 (DRE␣-I), ؊44 to ؊37 (DRE␣-II), and ؉3 to ؉10 (DRE␣-III). Footprinting analysis using the recombinant DRE-binding factor (DREF) or Kc cell nuclear extract demonstrated that DREF can bind to all three DRE-related sites. Introduction of mutation in even one of the three DRE-related sequences caused extensive reductions of the promoter activity and also the DREF-binding activity of the promotercontaining fragment. The results indicate that the three DREF-binding sites cooperate to enhance promoter activity of the DNA polymerase ␣ 73-kDa subunit gene.
Five distinct species of DNA polymerases (␣, ␤, ␥, ␦, and ⑀) have been isolated from eukaryotes and characterized (1,2). Of these, three are thought to be involved in chromosomal DNA replication including the DNA polymerase ␣-primase, which has been implicated in this process by many lines of evidence (1)(2)(3)(4)(5). The DNA polymerase ␣-primase consists of four subunits with molecular masses of 165-182, 68 -86, 54 -60, and 46 -50 kDa (2). The largest polypeptide is known to be the DNA polymerase catalytic subunit (6), and the two smallest subunits are responsible for the primase activity (7). The function of the second largest subunit of 68 -86 kDa is not clear yet. cDNA clones for the four subunits of the mouse DNA polymerase ␣-primase were isolated, and their expressions were examined in mouse cells during the cell cycle (8). When cells at the quiescent state are stimulated to proliferate, levels of mRNAs for all four subunits of the enzyme increase almost simultaneously prior to DNA synthesis (8), and therefore, transcription of these genes is likely regulated by a common mechanism.
In budding yeast, promoter regions of many DNA replica-tion-related genes contain a common nucleotide sequence (5Ј-ACGCGT) named MCB (MluI cell cycle box) (9), and the specific transcription factor MBF (MCB-binding factor) is required for the transcription of these genes at the G 1 -S boundary (10,11). In mammalian cells, the transcription factor E2F binds to the E2F-recognition site (5Ј-TTTCGCGC) and positively regulates transcription of a group of genes whose products are required for cell proliferation (12,13) such as DNA polymerase ␣, dihydrofolate reductase, thymidine kinase, c-Myc, c-Myb, Cdc2, proliferating cell nuclear antigen (PCNA), 1 cyclin D, and cyclin E (14 -17).
We have isolated Drosophila genes for DNA polymerase ␣ 180-kDa subunit (18) and PCNA (19). Promoter regions of these genes contain a common 8-base pair (bp) palindromic sequence (5Ј-TATCGATA), named DNA replication-related element (DRE) (20). Three DREs are present in the DNA polymerase ␣ 180-kDa subunit gene at nucleotide positions Ϫ217, Ϫ86, and Ϫ30 with respect to the transcription initiation site and one DRE in the PCNA gene at a position Ϫ100 (20). The requirement for DREs for the activities of promoters of these genes has been confirmed both in cultured cells (20) and in transgenic flies (21). Furthermore, we found a specific DREbinding factor (DREF) consisting of an 80-kDa polypeptide homodimer (20). It is therefore of interest to determine whether the DRE/DREF transcriptional regulatory system functions in the transcription of genes for other DNA replication enzymes.
A cDNA and the genomic regions for the Drosophila DNA polymerase ␣ 73-kDa subunit have been cloned, and their nucleotide sequences have been determined (22). This sequence apparently contains one DRE sequence and close to that two additional DRE-related sequences. However, all of these are located around the first ATG codon in the transcribed region.
In the work presented here, we have carried out a more detailed analysis of the transcription initiation sites in the 73-kDa subunit gene and have determined new sites downstream of those previously identified. The DRE-related sequences are located at Ϫ76, Ϫ44, and ϩ3 with respect to the most prominent new transcription initiation site. We have therefore examined the role of the DRE-related sequences in the promoter activity. The obtained results suggest that these sites cooperate to activate the promoter of the DNA polymerase ␣ 73-kDa subunit gene.

EXPERIMENTAL PROCEDURES
Cell Culture-Kc cells derived from Drosophila melanogaster embryos were grown at 25°C in M(3)BF medium (23) supplemented with * This work was supported in part by a grant-in-aid from the Ministry of Education, Science, and Culture of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Plasmid Constructions-All nucleotide positions of the 73-kDa subunit gene in the following part of this paper were expressed with respect to one of the major transcription initiation sites, which were determined in the present study. Plasmid A contains all of the transcribed region and at least 5 kilobases of the upstream region of the Drosophila DNA polymerase ␣ 73-kDa subunit gene (22). Plasmid B contains about 1.2 kilobases of the upstream region from the position ϩ7 of the 73-kDa subunit gene in the vector pBluescript (22). The plasmid pDhsp70-L contains firefly luciferase cDNA combined with the Drosophila hsp70 gene promoter was used for cotransfection with CAT plasmids as an internal control (21).
To construct the plasmid that contained the upstream region from the position Ϫ302 to the position ϩ129 of the 73-kDa subunit gene, PCR was performed using plasmid A DNA as a template and a combination of primers B-1 and B-2. The PCR product was blunt-ended using T4 phage DNA polymerase, digested with XhoI, and then replaced with the gene fragment between EcoRV and XhoI sites of the plasmid B to create the plasmid pDPOLA73BLU, which contains the gene region from about Ϫ1200 to ϩ129.
To construct the plasmid used for the CAT transient expression assay, pDPOLA73BLU was digested with KpnI and SacI, and then the DNA fragment that contained the 73-kDa gene fragment was isolated. Then, this DNA fragment was placed between KpnI and SacI sites of the plasmid pSKCAT (24). The resultant plasmid was named as pDPOLA73CAT.
To construct the plasmid p-302DPOLA73CAT, pDPOLA73CAT was digested with XhoI to remove an upstream DNA fragment between two XhoI sites. The remaining part was then self-ligated using T4 DNA ligase.
To construct the plasmid p-302DPOLA73CATmutIII containing a mutation in the DRE sequence (DRE␣-III), p-302DPOLA73CAT was digested at the center of the DRE sequence with ClaI and then bluntended using T4 DNA polymerase, followed by self-ligation using T4 DNA ligase. By this treatment, two base pairs, GC, were inserted at the center of the DRE sequence. To construct the plasmids that contained mutations in the DRE-related sequences (DRE␣-I and DRE␣-II), a first PCR was performed using pDPOLA73CAT as a template with the following combinations of primers: 1) B-1 and 73␣Imc, 2) B-2 and 73␣Im, 3) B-1 and 73␣IImc, and 4) B-2 and 73␣IIm. The resultant products were named as Imc, Im, IImc, and IIm, respectively. A second PCR was performed using Imc and Im as templates with primers B-1 and B-2 or IImc and IIm as templates with primers B-1 and B-2. Under these conditions, only the DNA fragment produced depending on hybridization of two templates was amplified, and the hybridized region contains the mutated DRE-related sequence. These products were digested with BamHI and XhoI and then used to replace the region carrying the wild type sequence between BamHI and XhoI sites of p-302DPOLA73CAT to create the plasmids p-302DPOLA73CATmutI and p-302DPOLA73CATmutII.
To construct the plasmid p-302DPOLA73CATmutI II, first PCR was performed as described above using p-302DPOLA73CATmutI as a template with primers 73␣IImc and 73␣IIm, then followed with the second PCR using primers B-1 and B-2. To construct the plasmids with mutation in the DRE␣-III in addition to mutation(s) in DRE␣-I, DRE␣-II, or both, the plasmid carrying the mutation(s) in site I, site II, or both were digested with ClaI and then blunt-ended using T4 DNA polymerase, followed by self-ligation using T4 DNA ligase.
Primer Extension-Total cellular RNA was extracted from Drosophila embryos at 0 -2 h and 2-4 h old or from Kc cells by the method as described (25). A 38-mer primer (Primer 2) that was complementary to the region downstream of the second ATG codon (see Fig. 1) was chemically synthesized. The 5Ј-end of the primer was labeled with 32 P and hybridized with 50 g of total RNA for 16 h at 45°C in a solution containing 75% formamide, 62.5 mM PIPES (pH 6.4), 0.5 M NaCl. The RNA and primer were ethanol-precipitated and redissolved in a solution for the primer extension. The primer was extended for 90 min at 45°C in a 25 l reaction mixture containing 50 mM Tris-HCl (pH 8.3), 3 mM MgCl 2 , 10 mM dithiothreitol, 2.5 mM each of dNTP, 1.5 units of RNase inhibitor (Takara), 400 units of reverse transcriptase (Superscript II, Life Technologies, Inc.). After incubation, the reaction was stopped by adding 1 l of 0.5 M EDTA, and 1 l of RNase A (1 mg/ml) was added to digest RNA. The digestion was carried out for 30 min at 37°C. Glycogen (1 g) was added to the reaction as a carrier, and then the sample was extracted with phenol-chloroform. The reaction product was ethanol precipitated and redissolved in 4 l of a sequencing dye mixture. The sample was analyzed by a gel electrophoresis under denaturing conditions, followed by autoradiography. 35 S-Labeled DNA fragments, produced in the dideoxy sequencing reaction with the plasmid pDPOLA73BLU as a template using Primer 2 as a primer, were run in parallel, allowing precise mapping of the cap site.
Gel Mobility Shift Analysis-The expression plasmid for glutathione S-transferase (GST)-DREF(16 -608) fusion protein was constructed, and the fusion protein was expressed in Escherichia coli as described (26). The specific DRE-binding activity of DREF protein resides within 16 -105-amino acid residues (26). The E. coli cells were collected and suspended in a solution containing 25 mM Hepes, pH 7.9, 1 mM EDTA, 0.02% 2-mercaptoethanol, 10% glycerol, 0.1% Tween 80, 0.2 M KCl. The suspension was sonicated and then centrifuged. The supernatant was collected and used for the gel mobility shift assay. Kc cell nuclear extracts were prepared as described (20) and used for the gel mobility shift analysis. The probes containing a part of the Drosophila DNA polymerase ␣ 73-kDa subunit gene were end-labeled with 32 P. Gel mobility shift analysis was performed as described (20) with minor modifications. 32  sonicated calf thymus DNA (average size, 0.2 kilobase) on ice for 5 min. When necessary, unlabeled DNA fragments were added as competitors at this step. Then, the E. coli lysate containing GST-DREF(16 -608) fusion protein or Kc cell nuclear extract was added, and a reaction mixture was incubated for 15 min on ice. DNA-protein complexes were electrophoretically resolved on a 4% polyacrylamide gel in 50 mM Tris borate, pH 8.3, 1 mM EDTA containing 2.5% glycerol at 25°C. The gel was dried and autoradiographed.
DNA Transfection and CAT Assay-Kc cells (2 ϫ 10 6 /dish) were grown in 60-mm plastic dishes for 24 h and cotransfected with 10 g of the reporter plasmid DNA and 50 ng of pDhsp70-L DNA by a calcium phosphate coprecipitation method as described (27). Cells were harvested 48 h after DNA transfection. Cell extracts for determination of CAT activities were prepared as described (28). Radioactivities of spots corresponding to acetylated [ 14 C]chloramphenicols were quantified with the imaging analyzer BAS2000 (Fuji Film). The luciferase assay was carried out by means of a PicaGene assay kit (Toyo Inc.) as described previously (29). All assays were performed within the range of linear relation of the activities to incubation time and protein amounts. CAT activities were normalized to luciferase activities.
DNase I Footprinting Analysis-DNase I footprinting analysis was performed essentially as described (20). The DNA fragment obtained from digestion of pDPOLA73BLU with BamHI and XhoI was labeled at 5Ј-end of the upper or lower strand (1 ng, 1 ϫ 10 4 cpm) and added to 30 l of a reaction mixture containing 25 mM Hepes, pH 6.7, 40 mM KCl, 5 mM MgCl 2 , 0.1 mM EDTA, 1 g of sonicated calf thymus DNA, 1 g of poly(dI-dC). E. coli lysate containing GST-DREF (16 -242) or Kc cell nuclear extract were added last, and the binding reaction was performed for 15 min on ice. DNase I (2 l, 100 units/l) that was freshly diluted in 10 mM Hepes, pH 7.6, 5 mM CaCl 2 , 10 mM MgCl 2 , and 100 g/ml bovine serum albumin was added to each reaction. After 1 min of digestion at 25°C, reactions were terminated by adding 100 l of a solution containing 40 mM EDTA, 0.4% SDS, 0.3 M NaCl, 40 g/ml E. coli tRNA, and 100 g/ml proteinase K. The samples were incubated for 30 min at 45°C and then extracted with phenol-chloroform. The reaction products were precipitated with ethanol and then loaded on a 6% polyacrylamide/8 M urea sequencing gel in parallel with products of Maxam-Gilbert sequencing reactions using the same DNAs as probes for sequencing controls. After electrophoresis, gels were dried and autoradiographed.

Determination of Transcription Initiation Sites of the DNA
Polymerase ␣ 73-kDa Subunit Gene-cDNA and the gene for the Drosophila DNA polymerase ␣ 73-kDa subunit were isolated, and their nucleotide sequences were determined (22). We found a sequence identical to DRE (5Ј-TATCGATA) and two sequences similar to DRE around the previously identified ATG translation initiation codon (ATG1 in Fig. 1). These locations are several hundred base pairs downstream from the previously determined multiple transcription initiation sites (22). Because these three sites were found to be bound by DREF as described below, we named these sites as DRE␣-I, DRE␣-II, and DRE␣-III. In our previous studies on the genes for DNA polymerase ␣ 180-kDa subunit and PCNA, DREs are localized  Fig. 1) complementary to the DNA polymerase ␣ 73-kDa subunit mRNA was hybridized with total RNA isolated from Drosophila embryos, Kc cells, or E. coli rRNA. The primer was extended using reverse transcriptase as described under "Experimental Procedures." To align the extended products with the genomic DNA sequence, a parallel dideoxy sequencing reaction was carried out by using the same 38-mer primer (lanes A, C, G, and T). Shorter exposure of the autoradiogram around the region of the major transcription initiation site is shown on the right. The numbers at the left indicate the nucleotide positions from the major transcription initiation site, which was defined as ϩ1. in the adjacent upstream regions of their transcription initiation sites (20). We therefore thought that other transcription initiation sites might be present downstream of those previously identified.
We searched this by a primer extension experiment using the new primer (Primer 2 shown in Fig. 1), and several new transcription initiation sites of the 73-kDa subunit gene were identified in addition to those previously determined using Primer 1 (22) (Fig. 1 and 2). The most prominent of the newly determined transcription initiation sites was now defined as the nucleotide position ϩ1 and was mapped 3 bp downstream of the first ATG codon as suggested previously (22) (Fig. 1). Both previously and newly identified sites seem to be frequently used in early embryos. The signal of newly identified site was especially prominent with RNA extracted from Kc cells (Fig. 2), and those corresponding to previously determined ones were rather weak. A TFIID target sequence, 5Ј-TTATTG (30), was found 12 bp upstream of the major site of the new transcription initiation sites but not around the previously determined sites.
In the region downstream of the newly determined transcription initiation site, the ATG codon (ATG2 in Fig. 1) was found in the position of ϩ130, and its location is 132 bp downstream from the previously determined first ATG (ATG1 in Fig. 1). Thus, the polypeptide coded by the new sequence is 44 amino acids shorter than that coded by the previously determined sequence. Homology searches indicate that amino acid sequence of the N-terminal end of the newly suggested sequence corresponds to that of the mammalian 73-kDa subunit sequence, and therefore, the previously determined coding frame has an extra sequence of 44 amino acid residues that is absent in the mammalian homolog.
Determination of the Promoter Region of the DNA Polymerase ␣ 73-kDa Subunit Gene-The fragment spanning positions about Ϫ1200 to ϩ129, which contains both previously and newly determined transcription initiation sites, was placed adjacent upstream of the CAT gene (pDPOLA73CAT). Deletions were made unidirectionally from its 5Ј-end, and then the plasmids carrying various deletions were transfected into Drosophila Kc cells. A deletion from position Ϫ1200 to position Ϫ508 did not show any significant change in CAT expression (data not shown). Further deletions to position Ϫ285 also did not affect CAT activity significantly (Fig. 3). About 60% reduction of CAT expression was observed with a deletion from Ϫ285 to Ϫ266. Because this region contains a sequence similar to the E2F-binding site, 5Ј-TTTCGCGG, the transcription factor E2F might play a role for activation of the promoter of this gene as reported with the PCNA gene (31). Further deletions resulted in progressive reduction of the CAT expression level. Therefore,  (lanes a, c, e, and g) or without (lanes b, d, f, and h) -i), or the PCNA gene promoter fragment carrying a 6-base deletion in the DRE sequence (lanes j-l) were added to the reaction as competitors. The amount of competitors were as follows; 2 (lanes d, g, and j), 10 (lanes e, h, and k), and 20 ng (lanes f, i, and l). C, the 32 P-labeled fragment C was used as a probe and incubated with (lanes b-k) or without (lane a) Kc cell nuclear extract. Excess amount of the unlabeled fragment C (lanes c-e), the PCNA gene promoter fragment containing DRE (lanes f-h), or the PCNA gene promoter fragment carrying a 6-base deletion at the DRE sequence (lanes i-k) were added to the reaction as competitors. The amounts of the competitors were as follows; 2 (lanes c, f, and i), 10 (lanes d, g, and j), and 20 ng (lanes e, h, and k). D, the 32  the region containing DRE and DRE-related sequences are required for high promoter activity.
A deletion construct from the 3Ј-end was also made. The plasmid carrying the region from Ϫ302 to ϩ129 showed high CAT expression. However, the plasmid carrying the region from Ϫ302 to ϩ12 showed no detectable level of CAT expression, although it contained the DRE, the DRE-related sequences, and the major transcription initiation site (Fig. 3). These results indicate that the region from ϩ12 to ϩ129 contains an element(s) essential for the promoter activity. Taken together, it is concluded that the promoter region of the DNA polymerase ␣ 73-kDa subunit gene is localized between positions Ϫ285 and ϩ129.
Determination of the DREF-binding Sites in the DNA Polymerase ␣ 73-kDa Subunit Gene-To examine whether DRE and its related sequences can be recognized by DREF, the DREbinding factor identified previously (20), we carried out a gel mobility shift assay using the GST-DREF (16 -608) fusion protein (26) and Kc cell nuclear extract, in which four DNA fragments from various regions of the gene (Fig. 1) were used as probes. The shifted band was observed with the 430-bp fragment C but not with fragments A, B, and D (Fig. 4A). The shifted band was competed by adding an excess amount of the same fragment (Fig. 4B, lanes d-f) and the 151-bp DNA fragment containing the region from positions Ϫ168 to Ϫ26 of the Drosophila PCNA gene promoter, which contained a DRE sequence (Fig. 4B, lanes g-i). In contrast, the PCNA gene fragment carrying the 6-base deletion in the DRE sequence did not compete (Fig. 4B, lanes j-l). Similar nucleotide sequence-specific complex formation was observed with the fragment C and the Kc cell nuclear extract (Fig. 4C). Furthermore, the addition of the anti-DREF monoclonal antibody (monoclonal antibody 4) (26) to the binding reaction resulted in super-shift of the DNAprotein complex (Fig. 4D, lanes d-f). These results indicate that DREF can specifically bind to the fragment C-containing DRE and its related sequences of the DNA polymerase ␣ 73-kDa subunit gene.
DNase I footprinting analysis was performed to determine the exact DREF-binding site(s) in the DNA fragment C. 5Ј-end of either the upper or lower strand of the fragment spanning from position Ϫ302 to ϩ129 was 32 P-labeled and was used for the analysis. As shown in Fig. 5A, GST-DREF (16 -242) fusion protein protected three regions of the upper strand corresponding to positions from Ϫ78 to Ϫ62 (␣-I), from Ϫ55 to Ϫ34 (␣-II), and from Ϫ4 to ϩ18 (␣-III). Similar regions were protected when the lower strand was used as a probe (Fig. 5B). When the Kc cell nuclear extract was added in the reaction (Fig. 5C), three regions of about 22 bp corresponding to positions from Ϫ78 to Ϫ62, from Ϫ55 to Ϫ34, and from Ϫ4 to ϩ18 were also protected. The region ␣-III contains the DRE sequence, whereas the regions ␣-I and ␣-II contain the DRE-related sequences that matches 5 bp out of the 8-bp DRE sequence almost at the center of each protected region. The results indicate that DREF can bind to the DRE-related sequences as well as the DRE sequence.
Effects of Mutations in DRE and Its Related Sequences on Promoter Activity of the DNA Polymerase ␣ 73-kDa Subunit Gene-To examine roles of DREF-binding sequences for promoter activity of the 73-kDa subunit gene, we constructed CAT expression plasmids having mutations in the DRE and its related sequences (Fig. 6A), and CAT transient expression assays in Kc cells were performed.
Mutation of either of three DREF-binding sequences resulted in extensive reduction (75-95%) of the CAT expression (Fig. 6B). Mutations in any two DRE-related sequences completely abolished the CAT expression. The results indicate that all of the three DREF-binding sequences are required for the high promoter activity.
The above evidence suggests that the three sites cooperate to enhance the promoter activity. To gain further insight into the molecular mechanism, we examined the effects of the mutations on the DREF binding to the DNA fragments by a gel mobility shift assays. When GST-DREF fusion protein was incubated with the 32 P-labeled fragment C without competitor fragments, the shifted bands were detected (Fig. 7A, lane c). When a fragment containing intact DRE and DRE-related sequences was added to the reaction as a competitor, the shifted band was decreased extensively (Fig. 7A, lanes d-g). In contrast, when any of mutant fragments with mut I, mut II, or mut III was added to the reactions, shifted bands were decreased to only a limited extent. (Fig. 7A, lanes h-k, l-o, and p-s, and quantified results in Fig. 7B). Therefore, the presence of all three DREF-binding sequences in intact forms is required for formation of the strong DNA-protein complex, and an extensive decrease of the promoter activity by mutation in any of three DREF-binding sequences might be due to loss of affinity of DREF to the three DRE sequences. Thus, three DREF-binding sequences appear to cooperate to conduct high promoter activity.

DISCUSSION
In the present study, we have mapped new transcription initiation sites for the Drosophila DNA polymerase ␣ 73-kDa subunit gene, which are downstream of those previously reported. The sites in both upstream and downstream regions seem to be utilized in Drosophila early embryos, whereas the newly mapped sites are prominent in the cultured Kc cells. The newly mapped major transcription initiation site was located 3 bp downstream of the first ATG codon that was previously predicted as a translation initiation site (22). These results suggest that the second ATG codon located at 132 bp downstream of the first ATG codon functions as a translation initiation site in the mRNA, which is synthesized from the newly mapped transcription initiation sites. Thus, the previously predicted open reading frame contains 44 additional amino acid residues to the N terminus of the polypeptide started from the second ATG. Comparison of amino acid sequences of these predicted amino acid sequences with that of the mammalian 73-kDa subunit (8) revealed that the N-terminal of mammalian homolog corresponds to that predicted by the second ATG. However, this does not rule out the possibility that Drosophila embryos also contain a 73-kDa subunit with the extra N-terminal sequence. Biological significance of the possible heterogeneity of the DNA polymerase ␣ 73-kDa subunit remains to be clarified.
Previously, we reported that an 8-bp palindromic sequence of DRE and not neighboring sequences are responsible for activating promoters of the DNA polymerase ␣ 180-kDa subunit and PCNA genes in both cultured cell and transgenic fly systems (20,21). We also reported that the 2-base substitution within the 8-bp sequence of DRE abolished the binding to DREF (20). We found one DRE sequence and two DRE-related sequences in an adjacent region to the newly mapped transcription initiation sites. Although the nucleotide sequence of DRE is identical to that reported previously, two DRE-related sequences match only 5 bp out of the 8-bp DRE sequence. However, gel shift analyses have shown that all these three sites are essential for formation of the stable DNA-protein complex. Furthermore, mutation in any one of these three DREF-binding sites resulted in extensive reduction of the promoter activity. Therefore, they probably cooperate to enhance the promoter activity.
Recently, a Drosophila homolog of the mammalian E2F1 was isolated (32)(33)(34), and E2F-binding sites were found in promoter regions of the DNA polymerase ␣ 180-kDa subunit (32) and PCNA (31) genes. These sites appeared to function in both cultured Drosophila cells and living flies (31). A deletion of one of two E2F-recognition sequences in the DNA polymerase ␣ 73-kDa subunit gene promoter remarkably reduced the promoter activity. Our observations therefore suggest that the 73-kDa subunit gene is also regulated by E2F like the 180-kDa subunit and PCNA genes, although further analysis is necessary to clarify this point. In addition, our results suggest that the region between positions ϩ12 and ϩ129 is also important for the promoter activity. However, the precise sequences responsible for the control have yet to be identified.
Organizations of transcriptional regulatory elements of Drosophila genes for the DNA polymerase ␣ 73-kDa subunit, 180-kDa subunit, and PCNA are summarized in Fig. 8. DRE(s) and the E2F-binding sites are commonly observed among these genes. A similar organization of DRE and the E2F-binding site most likely represents a common regulatory mechanism for the expression of these three DNA replication-related genes.