Drosophila ORC specifically binds to ACE3, an origin of DNA replication control element

Abstract

In the yeast Saccharomyces cerevisiae, sequence-specific DNA binding by the origin recognition complex (ORC) is responsible for selecting origins of DNA replication. In metazoans, origin selection is poorly understood and it is unknown whether specific DNA binding by metazoan ORC controls replication. To address this problem, we used in vivo and in vitro approaches to demonstrate that Drosophila ORC (DmORC) binds to replication elements that direct repeated initiation of replication to amplify the Drosophila chorion gene loci in the follicle cells of egg chambers. Using immunolocalization, we observe that ACE3, a 440-bp chorion element that contains information sufficient to drive amplification, directs DmORC localization in follicle cells. Similarly, in vivo cross-linking and chromatin immunoprecipitation assays demonstrate association of DmORC with both ACE3 and two other amplification control elements, AER-d and ACE1. To demonstrate that the in vivo localization of DmORC is related to its DNA-binding properties, we find that purified DmORC binds to ACE3 and AER-d in vitro, and like its S. cerevisiae counterpart, this binding is dependent on ATP. Our findings suggest that sequence-specific DNA binding by ORC regulates initiation of metazoan DNA replication. Furthermore, adaptation of this experimental approach will allow for the identification of additional metazoan ORC DNA-binding sites and potentially origins of replication.

Figure 1

A construct containing nine repeats of ACE3 localizes DmORC. (A) Diagram of P-element constructs (Orr-Weaver et al. 1989; Carminati et al. 1992). Construct A₄₈O₂₈ has 7.7 kb of chorion DNA. Construct A₅₄O₁₈ is the same as A₄₈O₂₈ except that 319 bp of ACE3 is deleted. Construct M9 has nine repeats of the 440-bp ACE3 fragment. (B–D) DmORC protein was visualized by immunofluorescence microscopy in the follicle cells of stage 10A egg chambers of fly lines transformed with P-element constructs containing third chromosome chorion DNA. DNA was visualized by staining with DAPI. The merged red (DmORC) and blue (DNA) channels are shown. (B) Localization of DmORC in flies transformed with the A₄₈O₂₈ construct. Three foci were observed per nucleus. (C) Localization of DmORC in flies transformed with construct A₅₄O₁₈. Two foci were observed per nucleus. (D) Localization of DmORC in flies transformed with the M9 construct. Three foci of were observed per nucleus. (E) Western blot performed using anti-DmORC2 serum. Samples on the blot are 80 ng of recombinant DmORC2 protein (lane 1), extract from 0- to 12-hr embryos (lane 2), and nuclear extract from Schneider line 2 cells (lane 3).

Figure 2

High-resolution microscopy of DmORC foci in follicle cells. Immunofluorescence microscopy was used to visualize DmORC (red), DNA (stained with DAPI, in blue), and lamin (green) in a small, early stage 10B egg chamber (this staging is defined in Royzman et al. 1999). The doughnut-like structures are consistently observed in early 10B egg chambers, and are not readily visible in stage 10A egg chambers (data not shown).

Figure 3

Association of DmORC with chorion loci elements in vivo. Chromatin-containing extracts were prepared from formaldehyde-treated stage 10 egg chambers and were immunoprecipitated with either anti-DmORC2 serum (lane 3) or preimmune serum (lane 4). DNA was amplified using PCR primers specific to ACE3 (A), ACE1 (B), or AER-d (C), and primers specific to actin (A–C). Quantitation of the PCR products indicates the following enrichments of ACE elements compared to actin DNA in the precipitation: 28-fold enrichment for ACE3 compared to actin; 10-fold enrichment of ACE1 compared to actin; and 14-fold enrichment of AER-d compared to actin. These primers were also used to amplify dilutions of DNA isolated from the extracts before immunoprecipitation (Input DNA, lanes 1,2). The amount of input DNA used in these reactions was the equivalent of 0.5% (lane 1) and 0.125% (lane 2) of the total DNA present before immunoprecipitation.

Figure 4

Association of DmORC with the M9 P-element construct in vivo. Chromatin immunoprecipitation analysis of stage 10 egg chambers isolated from the M9-2 transformant line. (A) Diagram of the P-element insertion site of the transformant line M9-2. The diagram shows the location of the PCR primers used for this experiment, primers sets A, B, C, ry, and M9. (B) PCR analysis was performed on DNA isolated from immunoprecipitations with preimmune serum (lanes 4,9,14,19) or anti-DmORC2 serum (lanes 5,10,15,20). PCR analysis was also performed on input DNA equivalent to 2% (lanes 1,6,11,16), 0.5% (lanes 2,7,12,17), and 0.125% (lanes 3,8,13,18) of the total DNA present before immunoprecipitation. (C) PCR analysis was performed on DNA isolated from the wild-type flies after immunoprecipitations with preimune serum (lane 22) or anti-DmORC2 serum (lane 23). The wild-type flies lack the M9 construct. In the M9-2 transformant line, the M9 construct is present at one copy per genome equivalent whereas the A, B, and C. primer sets recognize loci that are present in two copies per genome equivalent. The rosy sequence recognized by the ry primer set is present at three copies per genome equivalent in the M9-transformant line. Consequently, PCR reactions performed with input DNA from the M9-2 transformant line always yield less product for the M9 DNA compared to the other sequences.

Figure 5

ATP-dependent binding of DmORC to ACE3 in vitro. (A) DmORC was purified from embryo extracts. The peak fraction (2 μl) from the final purification step was separated on a 10% polyacrylamide gel and subsequently silver stained (lane 1). The mobility of each DmORC subunit is located to the left of lane 1. (B) Electrophoretic mobility-shift assays were performed with radiolabeled ACE3 DNA. Binding reactions contained, as indicated, DmORC protein (lanes 3–6), ATPγS (lane 4–6), 50 ng affinity purified anti-DmORC2 antibody (lane 5), 50 ng anti-Drosophila RNA polymearse II (PolII) antibody (lane 6). The ATP-dependent species is indicated by the arrow.

Figure 6

DmORC binds specifically to ACE3 and AER-d in vitro. DmORC electrophoretic mobility-shift assays were performed with labeled DNA fragments. The DNA fragments, as diagramed at the bottom of the figure, were ACE3 DNA (lanes 1–4), DNA that flanks ACE3 (lanes 5–8), AER-d DNA (lanes 13–16), or DNA that flanks AER-d (lanes 9–12). Binding reactions contained DmORC (lanes 2,3,6,7,10,11,14,15) and ATPγS (lanes 3,4,7,8,11,12,15,16). The ATP-dependent species are indicated by the arrow.

Figure 7

DmORC interacts most strongly with the middle-third of ACE-3. (A) Diagram of ACE3 and flanking DNA (RF fragment). Different parts of the ACE3 (fragments a,b, and c) were subcloned into the StuI site (indicated by a verticle arrow) located in the center of the RF fragment. These subclones were used to make the DNA probes (B), for an electrophoretic mobility-shift assay (C). Binding reactions contained probe RF (lanes 1–4) probe RF + a (lanes 5–8), probe RF + b (lanes 9–12), or probe RF + c (lanes 13–16). Binding reactions also contained DmORC (lanes 2,3,6,7,10,11,14,15) or ATPγS (lanes 1,2,5,6,9,10,13,14).