The promoters of Drosophila genes encoding DNA replication-related proteins contain transcription regulatory element DRE (5'-TATCGATA) in addition to E2F recognition sites. A specific DRE-binding factor, DNA replication-related element factor or DREF, positively regulates DRE-containing genes. In addition, it has been reported that DREF can bind to a sequence in the hsp70 scs' chromatin boundary element that is also recognized by boundary element-associated factor, and thus DREF may participate in regulating insulator activity. To examine DREF function in vivo, transgenic flies were established in which ectopic expression of DREF was targeted to the eye imaginal discs. Adult flies expressing DREF exhibited a severe rough eye phenotype. Expression of DREF induces ectopic DNA synthesis in the cells behind the morphogenetic furrow that are normally postmitotic, and abolishes photoreceptor specifications of R1, R6, and R7. Furthermore, DREF expression caused apoptosis in the imaginal disc cells in the region where commitment to R1/R6 cells takes place, suggesting that failure of differentiation of R1/R6 photoreceptor cells might cause apoptosis. The DREF-induced rough eye phenotype is suppressed by a half-dose reduction of the E2F gene, one of the genes regulated by DREF, indicating that the DREF overexpression phenotype is useful to screen for modifiers of DREF activity. Among Polycomb/trithorax group genes, it was found that a half-dose reduction of some of the trithorax group genes involved in determining chromatin structure or chromatin remodeling (brahma, moira, and osa) significantly suppresses and that reduction of Distal-less enhances the DREF-induced rough eye phenotype. The results suggest a possibility that DREF activity might be regulated by protein complexes that play a role in modulating chromatin structure. Genetic crosses of transgenic flies expressing DREF to a collection of Drosophila deficiency stocks allowed identification of several genomic regions, deletions of which caused enhancement or suppression of the DREF-induced rough eye phenotype. These deletions should be useful to identify novel targets of DREF and its positive or negative regulators (Hirose, 2001).

The promoters of Drosophila genes related to DNA replication, such as those for the 180-kDa catalytic subunit and 73-kDa subunit polypeptide of DNA polymerase α and proliferating-cell nuclear antigen (PCNA), contain a common 8-bp palindromic sequence (5'-TATCGATA), named the DNA replication-related element (DRE) (2003) in addition to E2F recognition sites. The DRE requirement for promoter activation has been confirmed in both cultured cells and transgenic flies (Yamaguchi, 1995a; Yamaguchi, 1996). Studies using the latter have shown that DRE is required for the function of the PCNA promoter throughout development except in adult females. A specific DRE-binding factor (DREF) was identified consisting of an 80-kDa polypeptide homodimer, and molecular cloning of its cDNA has led to confirmation that DREF is a trans activator for DRE-containing genes (Hirose, 1996; Hirose, 2001 and references therein).

The N-terminal fragment of the DREF polypeptide containing a region responsible for DRE binding and dimer formation acts as dominant negative effector against DREF (1999). Expression of a dominant-negative form of DREF in cells of salivary glands and eye imaginal discs using the GAL4-upstream activation site (UAS) system inhibits endoreplication of larval salivary gland cells and significantly reduces DNA replication in the second mitotic wave, respectively (1999). The results indicate that DREF is required for DNA replication in both the mitotic cell cycle and the endo cell cycle. However, progression of DNA replication requires a number of genes involved in growth signal transduction, cell cycle regulation, DNA replication itself, or transcription regulation. So far, no clues habe been available to determine which of these genes is a critical target of the dominant negative form of DREF (Hirose, 2001).

Screening for the DRE sequence in the Drosophila genome permitted an estimate that more than 10³ Drosophila genes are regulated by DREF (Matsukage, 1995). These genes were classified into five groups by their functional category: (1) DNA replication related, (2) translation related, (3) signal transduction/cell cycle regulation related, (4) transcription factor, and (5) others. The fourth category contains specific transcriptional regulators required for normal development or determination of cell differentiation, such as zeste, caudal, Antennapedia, serendipity-beta, and some RNA-biding proteins. These observations suggest a possibility that DREF may play an important role in cell differentiation other than activation of genes involved in cell proliferation (Hirose, 2001).

Hart haa proposed a novel function of DREF as an antagonist of boundary element-associated factor (BEAF), which is involved in boundary activity of the scs' region of the Drosophila hsp 70 gene (Hart, 1997; Hart, 1999). Staining of polytene chromosomes with anti-DREF and anti-BEAF antibodies revealed that about half of the signals from the two proteins overlapped and that the other half were from only BEAF or only DREF. Furthermore, using a chromatin precipitation method, Hart demonstrated that DREF can bind to the same sequences as BEAF, establishing the chromatin boundary in the scs' special chromatin domain present in Drosophila melanogaster 87A7 hsp 70. From the results, Hart assumed that competition of binding between DREF and BEAF is important for regulation of activity at the chromatin boundary (Hirose, 2001).

To clarify in vivo functions of DREF, transgenic flies expressing DREF in eye imaginal disc cells by using the GAL4-UAS system. If overexpression of DREF in eye imaginal discs confers a specific phenotype in the eyes of adult flies, the flies can be utilized to screen for modifiers of DREF activity or its target genes. Such an approach has been successfully undertaken with dE2F-overexpressing flies. This study analyzed the consequences of ectopic expression of DREF polypeptide in cells of eye imaginal discs. Ectopic expression of DREF in the cells posterior to the MF induced ectopic DNA synthesis and apoptosis in normally postmitotic cells, inhibited differentiation of the photoreceptor cells, and resulted in a severe rough eye phenotype in the adult flies. Furthermore, mutations were identified that enhance or suppress the DREF-induced phenotype. Candidates for DREF interaction include cell cycle regulators and regulators of chromatin-structure, thus giving clues to resolve the molecular mechanisms of how DREF activates many genes related to cell proliferation (Hirose, 2001).

Involvement of the DRE-DREF system in the regulation of a considerable variety of genes has been suggested by the results of DNA database searches. In about 3.5% of the Drosophila genome, 73 copies of 5'-TATCGATA sequences were found to be localized within 0.6 kb upstream regions of 61 genes, including those involved in transcription, translation, growth signal transduction, cell cycle regulation, and transcriptional regulation, in addition to genes related to DNA replication (Matsukage, 1995). Genes related to cell proliferation, such as those for PCNA, DNA polymerase alpha 180- and 73-kDa subunits, CycA, D-raf, and E2F, have been shown to be under DRE-DREF system regulation (Ohno, 1996; Ryu, 1997; Sawado, 1998). In addition, ectopic expression of the truncated form of DREF, possessing dominant negative activity, in salivary glands or eye imaginal discs significantly reduces DNA replication. Evidence suggests that DREF is required for DNA replication in both the endo and mitotic cell cycles. However, it has hitherto not been clear whether the accumulation of DREF is sufficient to induce DNA replication as well as ectopic expression of dE2F to drive normally quiescent cells to enter S phase (Hirose, 2001).

This study demonstrates that ectopic expression of DREF in eye imaginal discs, using the Glass-responsive GMR promoter, induces ectopic DNA synthesis and apoptosis in the cells behind the morphogenetic furrow. The results appear to be very similar to phenomena observed in eye imaginal discs expressing dE2F/dDP. One possible molecular mechanism is direct up-regulation of genes involved in DNA replication such as those encoding PCNA and DNA polymerase alpha. Another possible mechanism is indirect up-regulation of genes related to DNA replication or genes required for the G₁/S transition by means of activation of dE2F expression. Overexpression of DREF in eye imaginal disc cells enhances the promoter activity of dE2F (Sawado, 1998). Therefore, overexpression of DREF may result in E2F accumulation behind the morphogenetic furrow and consequently induce an ectopic S phase. This view is consistent with the observation that reduction of the dE2F gene dose at least partially suppresses the rough eye phenotype induced by DREF. However, it should be noted that there are some differences between the DREF- and dE2F/dDP-induced phenotypes: (1) In eye imaginal discs expressing dE2F/dDP under control of the GMR promoter, most S phases are seen in uncommitted cells, whereas in imaginal discs expressing DREF, nuclei of cells under differentiation process located apically with respect to the imaginal disc appear to incorporate BrdU; (2) it has been reported that coexpression of baculovirus p35 protein, an inhibitor of cell death, strongly enhances the dE2F/dDP phenotype, indicating that the majority of cells ectopically entering S phase as a result of E2F/dDP expression are eliminated by apoptosis, while the DREF phenotype, in contrast, was significantly suppressed by coexpression of p35; and (3) overexpression of dE2F/dDP does not grossly interfere with the initiation of neuronal cell differentiation or the stepwise recruitment of cells into preclusters, whereas expression of DREF perturbs photoreceptor specifications of R1, R6, and R7, whose differentiation normally begins after those of R8, R2, R5, R3, and R4. The molecular mechanisms underlying these differences are unclear, but it is speculated that elevated amounts of DREF not only affect DNA replication-related genes but also disturb the transcription of genes required for normal processes of differentiation, consequently inducing apoptosis (Hirose, 2001).

The present study focused on the use of flies for screening interaction with Polycomb and trithorax group genes. It is known that DREF and BEAF share binding sites with functions as boundary elements in Drosophila (scs' region of hsp70, BE76, and BE28) and also the promoter region of DNA polymerase alpha 180-kDa subunit gene. Polycomb and trithorax group proteins appear to regulate chromatin boundary activity by establishing higher-order domains of chromatin organization required for the assembly of functional insulators at the nuclear matrix, and the present genetic screening showed that a half-dose reduction of the trithorax group genes brm, osa, mor, and E(Pc) suppressed while a half-dose reduction of Dll enhanced the DREF-induced rough eye phenotype (Hirose, 2001).

brm is the Drosophila homologue of the yeast SWI2/SNF2 gene, isolated as a dominant suppressor of Pc mutations. osa shows strong genetic interactions with brm, suggesting that its gene product may cooperate closely with the BRM complex. The product of mor is homologous to yeast Swi3p and has also recently been identified as a BRM complex-associated protein, BAP. BRM, OSA, and MOR are part of a large multiprotein complex containing several other proteins with extensive homology to yeast SWI/SNF or the related chromatin-remodeling complex, and therefore the BRM complex has been predicted to be one of the Drosophila chromatin-remodeling complexes. It has been demonstrated that the BRM complex containing OSA and MOR is an essential coactivator for the trithorax group protein Zeste, a sequence-specific transcription regulator. Several homeotic and other genes carrying the Zeste-binding sequence in their promoter regions thus appear to be candidate target genes (or chromatin regions) for the BRM complex (Hirose, 2001).

Suppression of the DREF-induced rough eye phenotype by a half-dose reduction of some members of BRM complex genes suggests that DREF activity may be positively regulated by the BRM complex. The molecular basis for any interaction is not yet clear, but several possibilities exist. The first is that genetic interaction may result from a direct physical interaction between the two. To test this possibility, biochemical analyses are being undertaken. An alternative possibility is that the BRM complex is a regulator of the expression of some genes critical for induction of the rough eye phenotype by DREF overexpression. This is difficult to test because it is estimated that several hundred genes are under DREF control. Finally, it is possible that the BRM complex generally activates transcription by remodeling chromatin. However, a half-dose reduction of brm, mor, or osa did not change the severity of the rough eye phenotype induced by BEAF32, and mutations in brm, mor or osa enhance the rough eye phenotype induced by overexpression of dE2F/dDP/p35. Therefore, suppression of the DREF-induced rough eye phenotype by reduction of the dosage of brm, mor, or osa may reflect specific interactions between DREF and the BRM complex. Further studies are necessary to clarify the molecular basis and biological meaning of the interaction between DREF and the BRM complex (Hirose, 2001).