Differential targeting of Tetrahymena ORC to ribosomal DNA and non-rDNA replication origins
2009; Springer Nature; Volume: 28; Issue: 3 Linguagem: Inglês
10.1038/emboj.2008.282
ISSN1460-2075
AutoresTaraka Donti, Shibani Datta, Pamela Y. Sandoval, Geoffrey M. Kapler,
Tópico(s)Acute Lymphoblastic Leukemia research
ResumoArticle15 January 2009free access Differential targeting of Tetrahymena ORC to ribosomal DNA and non-rDNA replication origins Taraka R Donti Taraka R Donti Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA Search for more papers by this author Shibani Datta Shibani Datta Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USAPresent address: MedImmune Inc., Gaithersburg, MD 20878, USA Search for more papers by this author Pamela Y Sandoval Pamela Y Sandoval Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USA Search for more papers by this author Geoffrey M Kapler Corresponding Author Geoffrey M Kapler Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USA Search for more papers by this author Taraka R Donti Taraka R Donti Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA Search for more papers by this author Shibani Datta Shibani Datta Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USAPresent address: MedImmune Inc., Gaithersburg, MD 20878, USA Search for more papers by this author Pamela Y Sandoval Pamela Y Sandoval Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USA Search for more papers by this author Geoffrey M Kapler Corresponding Author Geoffrey M Kapler Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USA Search for more papers by this author Author Information Taraka R Donti1,‡, Shibani Datta1,2,‡, Pamela Y Sandoval2 and Geoffrey M Kapler 1,2 1Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, College Station, TX, USA 2Texas A&M University Interdisciplinary Genetics Program, College Station, TX, USA ‡These authors contributed equally to this work *Corresponding author. Department of Molecular and Cellular Medicine, Texas A&M Health Science Center, 454 Reynolds Medical Building, College Station, TX 77843-1114, USA. Tel.: +979 847 8690; Fax: +979 847 9481; E-mail: [email protected] The EMBO Journal (2009)28:223-233https://doi.org/10.1038/emboj.2008.282 Present address: MedImmune Inc., Gaithersburg, MD 20878, USA PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The Tetrahymena thermophila origin recognition complex (ORC) contains an integral RNA subunit, 26T RNA, which confers specificity to the amplified ribosomal DNA (rDNA) origin by base pairing with an essential cis-acting replication determinant—the type I element. Using a plasmid maintenance assay, we identified a 6.7 kb non-rDNA fragment containing two closely associated replicators, ARS1-A (0.8 kb) and ARS1-B (1.2 kb). Both replicators lack type I elements and hence complementarity to 26T RNA, suggesting that ORC is recruited to these sites by an RNA-independent mechanism. Consistent with this prediction, although ORC associated exclusively with origin sequences in the 21 kb rDNA minichromosome, the interaction between ORC and the non-rDNA ARS1 chromosome changed across the cell cycle. In G2 phase, ORC bound to all tested sequences in a 60 kb interval spanning ARS1-A/B. Remarkably, ORC and Mcm6 associated with just the ARS1-A replicator in G1 phase when pre-replicative complexes assemble. We propose that ORC is stochastically deposited onto newly replicated non-rDNA chromosomes and subsequently targeted to preferred initiation sites prior to the next S phase. Introduction The conserved Origin Recognition Complex (ORC) determines the sites for replication initiation in eukaryotic chromosomes and serves as a scaffold for pre-replicative complex (pre-RC) assembly. Although ORC subunits are conserved in eukaryotes, the cis-acting DNA sequence requirements for replicator function are not. Saccharomyces cerevisiae ORC binds in a sequence-specific manner to a short motif present at all origins. ORC binding to autonomously replicating sequence (ARS) elements is required for origin activation, whereas other protein–DNA interactions serve lesser, auxiliary roles (Marahrens and Stillman, 1992; Bolon and Bielinsky, 2006). In contrast, Schizosaccharomyces pombe, Drosophila melanogaster and human ORC bind non-specifically to AT-rich DNA sequences (Kim and Huberman, 1998; Vashee et al, 2003; Remus et al, 2004). S. pombe replicators consist of blocks of degenerate sequence that create multiple ORC-binding sites (Segurado et al, 2003; Dai et al, 2005). In Drosophila and the rat, ORC is tethered to the respective chorion gene (DAFC-66D) and aldolase origins by associating with unrelated sequence-specific DNA-binding proteins (Beall et al, 2002; Minami et al, 2006). The interaction of human ORC with HMGA1a similarly creates functional origins in heterochromatic regions (Thomae et al, 2008). Whether tethering is commonly used to recruit ORC to metazoan origins is unclear. To add to the complexity, metazoan replicators vary in size and local density of initiation sites. Although the 1.2 kb human lamin B2 replicator initiates at a single discrete site (Abdurashidova et al, 1998), many origins fire within the 20 kb segment downstream of the hamster DHFR gene (Hamlin and Dijkwel, 1995). Tetrahymena thermophila (Tt) ORC is unusual in that it contains an integral RNA subunit that uses Watson-Crick base pairing to bind to its cognate DNA target in the ribosomal DNA (rDNA) replication origin (Mohammad et al, 2007). This DNA sequence, the type I element, is required for developmentally programmed amplification and cell cycle-controlled vegetative replication of rDNA minichromosomes (Figure 1A) (reviewed in Tower, 2004). Remarkably, the ORC RNA subunit, 26T RNA, corresponds to the terminal 282 nucleotides (nt) of 26S rRNA (Mohammad et al, 2007). Mutations that perturb RNA pairing with the type I element T-rich strand disrupt rDNA origin recognition and activation. Type I elements are recognized by additional single-stranded binding factors, including TIF1p, which binds to the A-rich strand at the origin and controls the timing of rDNA origin activation (Saha et al, 2001; Morrison et al, 2005). Figure 1.Organization of rDNA and non-rDNA macronuclear chromosomes. (A) Schematic of the palindromic rDNA minichromosome. Expanded view of the 5′ NTS includes positioned nucleosomes (ovals) type I elements (black rectangles), pause site elements (grey rectangles), rRNA promoter (thin arrow) and replication origins (ori) which reside in the 430 bp imperfect duplicated sequences, domains 1 and 2 (thick arrows). The sequence of the type IB element T-rich strand and flanking DNA are shown, including predicted base pair interactions with 26T RNA. (B) Chromosome reorganization and replication during macronuclear development. A portion of the micronuclear chromosome encoding the single copy rRNA gene is shown (CBS: chromosome-breakage sequences). See text for additional details. Download figure Download PowerPoint rDNA and non-rDNA chromosomes are differentially replicated during Tetrahymena development. This property stems from the partitioning of chromosome functions into two distinct nuclei within each cell: the 'germline' micronucleus and the 'somatic' macronucleus (reviewed in Karrer, 2000). The non-transcribed micronucleus contains the chromosomes that are transmitted to progeny during conjugation. As such, it undergoes conventional mitosis and meiosis. The transcribed, amitotic macronucleus confers the phenotype of the cell. During conjugation pronuclei are exchanged and fuse to generate a new diploid micronucleus. Following two rounds of DNA replication and nuclear division, two of the four micronuclei differentiate into macronuclei. At this time, the five monocentric chromosomes are fragmented at chromosome-breakage sequence (CBS) elements and further rearranged, generating ∼280 macronuclear chromosomes (size range: 21–>4000 kb) (Figure 1B). Non-rDNA chromosomes re-replicate to a final copy number of ∼45 C, whereas the single copy rDNA locus is rearranged into a 21 kb minichromosome and amplified ∼5000-fold. Once development is complete, micro- and macronuclear chromosomes replicate once per vegetative cell division. Despite the absence of centromeres, macronuclear chromosomes are maintained at a relatively constant copy number. Here, we describe a plasmid shuttle assay that was used to isolate the first non-rDNA Tetarhymena replicator, and show that it is comprised of discrete cis-acting determinants. We provide the first evidence for cell cycle-regulated changes in the specificity of ORC for chromosomal DNA in any eukaryote, and show that these changes do not occur in the rDNA minichromosome. We propose a model for ORC binding to non-rDNA chromosomes, in which ORC associates non-specifically with newly replicated daughter chromosomes and then re-localizes to preferred initiation sites prior to the next S phase. Results Isolation of non-rDNA Tetrahymena replicators Previous transformation studies showed that the 1.9 kb rDNA 5′ NTS supports autonomous DNA replication in Tetrahymena (Pan et al, 1995; Reischmann et al, 1999). Here, we developed a shuttle assay to isolate functional replicators that are not amplified (Figure 2A). To assure that plasmids replicated in Tetrahymena, DNA was isolated from Tetrahymena transformants and digested with the methylation-sensitive restriction endonuclease, DpnI, prior to re-transforming E. coli. Although plasmid DNA prepared from dam+ E coli is sensitive to DpnI, inefficient DNA methylation in Tetrahymena renders replicated molecules resistant to cleavage. Figure 2.Isolation of non-rDNA replicators. (A) Schematic of the plasmid shuttle assay used to isolate Tetrahymena ARS1. DpnI-sensitive (DpnIs) plasmids containing T. thermophila genomic DNA were co-transformed into Tetrahymena. Plasmid DNA from Tetrahymena transformants was re-introduced into E. coli. ARS-containing plasmids should be DpnI-resistant (DpnIr) following replication in Tetrahymena, and can be recovered by re-transforming E. coli. (B) Ethidium bromide negative stain image of BamHI-digested plasmid DNA from two clones obtained from the plasmid shuttle assay (Vector: plasmid backbone; Insert: Tetrahymena sequences). (C) Southern blot of undigested Tetrahymena genomic DNA from 'en masse' transformations using intact ARS1 or a derivative containing just the internal NheI fragment (see Figure 3A for details on the NheI2 derivative). DNA was isolated from pmr co-transformants propagated for 25 generations. ARS1input plasmid DNA (+) was used as a Southern blotting control. Probe: radiolabelled pCC1FOS (no insert). (D) Southern blot of BamHI-digested DNA from wild type Tetrahymena (WT, CU428) and 'en masse' ARS1 co-transformants propagated for up to 60 generations (gen). Probe: radiolabelled ARS1 insert. Download figure Download PowerPoint An E. coli plasmid library was created from T. thermophila DNA fragments in the 5–7 kb size range. Pools of 10 colonies were screened by PCR to eliminate plasmids that contained the rDNA origin. DNA from 100 rDNA-negative colonies was co-transformed into the developing macronucleus with plasmid AN101, which rearranges into a linear rDNA minichromosome and confers paromomycin resistance. Circular plasmid DNA was isolated from co-transformants grown en masse for ∼25 generations, digested with DpnI, and electroporated into E. coli. Twelve kanamycin-resistant colonies were obtained. Two plasmids contained inserts in the expected size range (Figure 2B), whereas the remaining plasmids had small inserts and/or lacked adjoining vector sequences (data not shown). This was not unexpected, as AT-rich Tetrahymena sequences frequently rearrange in E. coli. The DNA sequences of the two large insert clones were identical, corresponding to a 6674 bp segment within a 407 kb macronuclear chromosome (locus identifier: CH445556.2, DNA sequence identifier: AAGF01003027; http://www.ciliate.org). This plasmid, designated ARS1, was reintroduced into Tetrahymena and autonomous replication was assessed by PCR or Southern blotting. In addition to en masse propagation of co-transformants, clonal lines were grown for 60 generations and assayed for retention of plasmid sequences. Approximately half (9/16) of the pmr clones stably propagated the vector backbone (PCR assay, data not shown), and Southern blot analysis of undigested DNA confirmed that the DNA was extrachromosomal (Figure 2C; ARS1). Further analysis of restriction digested DNA with an ARS1-specific probe revealed that plasmid copy number was relatively constant throughout the duration of the experiment (60 generations), approximating that of the endogenous (45 C) macronuclear chromosome (Figure 2D). Similar to the initial library screen, plasmid DNA from ARS1 Tetrahymena transformants was reintroduced into E. coli. A mixture of full-length clones and deletion derivatives was obtained. As a control, the empty pCC1FOS vector and AN101 were co-transformed. Southern blot analysis failed to detect vector sequences in pmr progeny and kanamycin-resistant E. coli clones were not recovered. We conclude that ARS1 confers autonomous DNA replication in Tetrahymena. ARS1 contains two autonomous replicators Eukaryotic replication origins generally reside in intergenic (IG), AT-rich DNA segments and rarely encroach into genes (Paixão et al, 2004; Wang et al, 2004). Computer annotation predicts that ∼80% of the cloned ARS1 interval encodes two proteins of unknown function (Figure 3A). Less than half of the 600 bp intergenic segment, IG-1, resides in the clone, whereas the entire 1.2 kb IG-2 segment was present. To assess the functional organization of ARS1, restriction enzymes were used to generate deletion derivatives or subclone internal fragments into the smaller pSMART vector (Figure 3A). Plasmids were co-transformed into Tetrahymena. En masse cultured co-transformants were propagated and assayed for the presence of plasmid DNA by Southern blotting. Figure 3.ARS1 deletion mapping. (A) Upper schematic: predicted genes (arrows) and intergenic (IG-1, IG-2) segments in the chromosomal ARS1 interval. The end points of deleted (Δ) or retained sequences in ARS1 plasmid derivatives are indicated below. Asterisks denote fragments that were subcloned into pSMART. The remaining inserts are in the original pCC1FOS vector backbone. (B) Southern blot analysis of ARS1 transformants and five of the seven depicted deletion derivatives. DNA was prepared from 'en masse' co-transformants at defined intervals (generations: gen). BamHI-digested samples were probed with ARS1 fragments containing just those sequences present in the plasmid under examination. C: endogenous chromosomal DNA BamHI fragment; P: plasmid-derived BamHI fragment. Download figure Download PowerPoint Deletion of the internal NsiI or NheI fragments had no effect on plasmid DNA replication or long-term maintenance (Figure 3A and B; ΔNsiI; ΔNheI). To our surprise, plasmids containing either the left (NheIL, 1–966) or right (NheIR, 3798–6774) ends of ARS1 were stably propagated in Tetrahymena. Conversely, internal fragments subcloned into the pSMART vector backbone failed to support autonomous DNA replication (Figure 2C, NheI2; Figure 3B, NsiI2), suggesting that replicator function is sequence-dependent. A pSMART derivative containing just the IG-2 segment (NheIR2) was stably maintained, indicating that this vector backbone does not inhibit DNA replication in Tetrahymena. We conclude that ARS1 contains two autonomous replicators rather than one, hereafter designated ARS1-A and ARS1-B (Supplementary Figure S1). An analogous situation has been documented in the human β-globin locus (Wang et al, 2004). Deletion mapping of ARS1-A Approximately 70% ARS1-A (Figure 3A, nt 300–966) is predicted to encode protein (Tetrahymena Genome Database gene prediction Therm 00579140). Reverse transcription (RT)–PCR verified that this segment is actively transcribed. Two DNA intervals were examined, one of which spans a predicted 40 nt intron (Figure 4A). PCR products of the expected size were detected with cDNA (C) and genomic DNA (G) prepared from a log phase vegetative culture, whereas no product was generated from RNA (R) that was not reverse transcribed. The steady state level of this transcript was relatively constant across the cell cycle (Figure 4B), suggesting that transcription factors are constitutively bound to cis-acting regulatory sequences. Figure 4.Deletion mapping of the ARS1-A replicator. (A) RT–PCR analysis of the ARS1-A interval. The diagram shows the relative positions of forward (F1, F4) and reverse complementary (R2–R6) primers, intron 1, and start (ATG) and stop (TGA) codons in the predicted gene. PCRs were performed on genomic DNA (G), reverse transcribed RNA (cDNA, C) and total RNA (R). A negative image of the ethidium bromide-stained gel is shown. (B) Lower panel: RT–PCR analysis of RNA from cells synchronized by centrifugal elutriation (RT primer R6, PCR primers F4 and R5). Log: RNA from an asynchronous cell culture. Upper panel: flow cytometry profile of elutriated cells at defined culturing intervals (min). (C) Southern blot analysis of NheIL-L8 and NheIL-L6 transformants, and three of the six tested NheIL-L8 deletion derivatives (see Table I). DNA was prepared from pmr 'en masse' ARS1 co-transformants at defined intervals (generations: gen). BamHI-digested DNA was probed with ARS1 fragments containing just those sequences present in the plasmid under examination. Arrow: plasmid-derived ARS1A fragment. Download figure Download PowerPoint To assess whether cis-acting replication determinants reside in protein-coding sequence and examine the overall organization of the ARS1-A replicator, thirteen terminal or internal deletion derivatives were tested for the ability to support autonomous DNA replication (Table I). Deleting the last 166 bp (Nhe1L-L8; nt 1–800) had no effect on plasmid propagation; however, removal of an additional 200 bp led to the failure to support autonomous replication (Nhe1L-L6; nt 1–600) (Figure 4C). The internal deletion mutant Nhe1L Δ600–800 also failed to replicate (data not shown), indicating that the protein-coding interval contains at least one functional DNA replication determinant. Internal deletion derivatives, NheIL Δ200–400 and NheIL Δ400–600, were competent for plasmid DNA replication and long-term plasmid maintenance; however, the Nhe1L-R8 derivative (Δ1–200) was not (data not shown). We conclude that the right and left portions of ARS1-A contain essential replication determinants. To test whether essential intergenic (nt 1–200) and coding region (nt 600–800) sequences were interchangeable, a second copy of the first 200 bp was introduced into NheI-L6. The resulting plasmid, NheIL-L6/+1–200, failed to replicate (data not shown), indicating that the respective sequences serve distinct functions. Overlapping 100 bp deletion derivatives of NheIL-L8 were examined to further localize replication determinants (Table I, last six constructs). All three deletions in the intergenic interval (NheIL-L8 Δ1–100, NheIL-L8 Δ50–150, NheIL-L8 Δ100–200) failed to support autonomous replication (data not shown). However, the three deletions in the protein-coding segment (NheIL-L8 Δ600–700, NheIL-L8 Δ650–750, NheIL-L8 Δ700–800) behaved differently (Figure 4C). NheIL-L8 Δ700–800 was maintained at a constant level, whereas NheIL-L8 Δ650–750 initially replicated, but was lost during vegetative propagation. NheIL-L8 Δ600–700 failed to support autonomous DNA replication, indicating that it contains an essential cis-acting replication determinant. We conclude that ARS1-A contains dispersed essential and non-essential replication determinants. Bioinformatic analysis of ARS1-A and ARS1-B replicators A distinguishing feature of the rDNA replicator is the presence of two reiterated regulatory sequences that participate in origin activation—the type I and PSE element (Figure 1A) (Gallagher and Blackburn, 1998; Reischmann et al, 1999; Saha et al, 2001; TL Morrison and GM Kapler, unpublished results). Type I elements form Watson–Crick base pair interactions with the ORC RNA subunit (26T RNA) which targets ORC to this origin (Mohammad et al, 2007). Biochemical data indicate that 26T RNA is an integral component of all ORC complexes. Significant base pairing potential between 26T RNA and DNA sequences in the ARS1-A or ARS1-B intervals was not detected. rDNA origin- and promoter-proximal type I and PSE elements function as in vivo binding sites for the non-ORC protein, TIF1p (Saha et al, 2001). As both sequences are absent from ARS1-A and ARS1-B, we predicted that TIF1p would not associate with these DNA segments. Chromatin immunoprecipitation (ChIP) analysis with a TIF1 peptide antibody supported this prediction (Supplementary Figure S2). These data and results from experiments described below suggest that 26T RNA does not target ORC to the ARS1-A and ARS1-B origins. To look for shared motifs in ARS1-A and ARS1-B, or similarity with other sequences in the rDNA replication origin, the three replicators were subjected to pairwise analysis using BLAST, CLUSTALW and LALIGN. Short dispersed blocks of sequence identity were detected in each analysis, consisting of AT-rich sequences 7–10 bp in length (Supplementary Figure S3). No DNA motifs were present in all three replicators. Pairwise analysis of ARS1-A or ARS1-B with two arbitrarily chosen intergenic sequences produced matches of similar length. Thus, compelling evidence for sequence conservation was lacking. ORC targeting to rDNA and non-rDNA macronuclear chromosomes The ability of ARS1-A and ARS1-B to support episomal DNA replication suggests that ORC is recruited to these sites in endogenous macronuclear chromosomes. To address this prediction, chromatin ChIP was performed with epitope-tagged Orc1p across a 60 kb segment spanning the endogenous ARS1 chromosomal locus (Figure 5B, schematic). Control reactions on rDNA origin (O), promoter (P) and coding (C) sequences produced the expected results: Orc1p bound the rDNA origin and did not associate with other segments in the rDNA minichromosome (Figure 5A). Identical results were obtained for chromatin from an asynchronous logarithmic vegetative culture or G0/G1 cells synchronized by starvation. Figure 5.ORC targeting to rDNA and non-rDNA replicators. (A) Orc1p ChIP analysis of the rDNA domain 1 origin (O), promoter (P) and coding (C) regions. ChIP was performed on T. thermophila strain TD102 using an antibody directed against the protein A IgG-binding epitope-tag in Orc1p. I: total input DNA; (−): no antibody ChIP control; (+): Orc1p ChIP pellet. (B) Orc1p ChIP analysis of the 60 kb segment spanning ARS1 in the endogenous macronuclear chromosome (strain TD102). PCR products derived from primer sets A1 to A12 are spaced at ∼5 kb intervals. (C, D) Streptavidin (SA) chromatin pull-down analysis of rDNA and ARS1 chromosome intervals. Strain MM201 produces a 26T RNA variant that bears a sequence tag extension (Ext), whereas MM202 expresses an aptamer-tagged (Apt) 26T RNA derivative that binds to streptavidin. I: total input DNA; (−) uncoupled sepharose chromatin pull-down pellet, (+): SA-sepharose chromatin pull-down pellet. (E) ChIP analysis of the ARS1 interval with Tetrahymena Mcm6p antibodies. Download figure Download PowerPoint ChIP analysis of the ARS1-containing chromosome produced very different results (Figure 5B). All examined intervals were enriched in the Orc1p immunoprecipitate for chromatin prepared from the asynchronous vegetative culture. The tested segments consisted of ARS1-A, ARS1-B and seven randomly chosen coding and non-coding sequences spanning 60 kb. In contrast, only the ARS1-A A6 segment was enriched in starvation-induced G0/G1 chromatin immunoprecipitates (Figure 5B). To verify that we were studying in vivo binding of the ORC holocomplex and not just Orc1p, chromatin pull-down assays were performed with 26T RNA derivatives that contained a 5′ sequence extension (26T-Ext) or an aptamer sequence tag (26T-Apt), the later of which confers binding to streptavidin (SA) (Srisawat and Engelke, 2002). As reported earlier (Mohammad et al, 2007), the rDNA origin region was selectively enriched in the SA-sepharose pull-down fraction in cells expressing the aptamer-tagged 26T RNA (Figure 5C). rRNA promoter and coding sequences were not enriched in chromatin pull-downs from log phase and starved cultures (compare 26T-Apt to 26T-Ext controls). In contrast, all nine tested segments in the 60 kb ARS1 interval were enriched in chromatin pull downs in asynchronous log phase cells expressing aptamer-tagged 26T RNA, but not the 26T-Ext variant (Figure 5D and data not shown). Moreover, the aptamer-tagged RNA associated with just the ARS1-A A6 fragment in chromatin prepared from G0/G1 synchronized cells. We conclude that 26T RNA is a component of non-rDNA origin binding ORC complexes. The differential binding of ORC to non-rDNA chromosomes in log phase and starved cells cannot be attributed to the presence or absence of 26T RNA. Starvation not only synchronizes the vegetative cell cycle, it prepares Tetrahymena for conjugation and the associated DNA replication programme. To ask whether Orc1p binding to non-rDNA chromosomes is cell cycle regulated, centrifugal elutriation was used to obtain a highly enriched population of G1 phase cells to examine ORC regulation in an unperturbed cell cycle (Figure 6A; see Supplementary Figures S4 and S6 for 0 min time points). Cell cycle-dependent changes in the abundance of Orc1p were observed, including a precipitous drop in Orc1p levels during S phase (Supplementary Figure S4), analogous to mammalian and Drosophila Orc1p (Mendez et al, 2002; Araki et al, 2003). Figure 6.Cell cycle-regulated ORC binding to the ARS1-A chromosomal locus. A log phase Tetrahymena culture was subjected to centrifugal elutriation. The G1 fraction was further cultured and samples harvested at defined intervals (min) for western blotting of Orc1p (Supplementary Figure S4), flow cytometry and ChIP. (A) Flow cytometry profiles of re-fed cultures (the left line demarcates the G1 propidium iodide (PI) peak, and the right line marks the G2 peak). (B) Orc1p cell cycle ChIP analysis (see Figure 5B for PCR primers locations). I: input; (−): no antibody ChIP pellet; (+): Orc1p ChIP pellet. (C) Orc1p ChIP analysis of the rDNA origin (O), promoter (P) and rRNA coding (C) regions (see Figure 5A schematic). Download figure Download PowerPoint Similar to starvation-synchronized chromatin, Orc1p selectively associated with the ARS1-A origin in elutriated G1/early S phase cells (Figure 6B, 30 min, A6 region; Supplementary Figure S6B, 0 and 30 min). ARS1-A enrichment was lost as cells progressed through S phase, when Orc1p levels decline (Figure 6B, 60 and 90 min). Newly synthesized Orc1p was detected for all nine ARS1 region probes in G2 phase cells, showing no preference for the A6 fragment (Figure 6B, 120 and 150 min). We conclude that ORC binding to the non-rDNA ARS1 chromosomal interval is dynamically regulated across the cell cycle. ORC appears to be stochastically deposited onto the newly replicated chromosome, and re-localize to the ARS1-A region prior to the next S phase. As ARS1-B supported episomal DNA replication, but was not bound by Orc1p when pre-RCs assemble on the endogenous chromosome, this segment does not appear to function as a replicator in its normal chromosomal context. Alternatively, the site for ORC binding might have been distal to the PCR primers used in the initial analysis. To address this concern, immunoprecipitated chromatin from the G1 and G2 phase cells was subjected to PCR with five ARS1-B primer sets spanning a 2.1 kb interval (Supplementary Figure S5A). Products were detected for all primer sets in G2 phase chromatin; however, none of these segments was enriched in G1 preparations (Supplementary Figure S5B). In a comparable analysis of the ARS1-A interval, three consecutive primer sets were amplified in immunoprecipitated G1 phase chromatin. Thus, the ARS1-A locus is the preferred site for ORC binding in the endogenous macronuclear chromosome. Cell cycle ChIP analysis of the rDNA minichromosome revealed origin-specific Orc1p binding only, analogous to log phase and starve vegetative cultures (Figure 6C). Origin binding was detected in G1 and early S phase cells (T=30 min) and disappeared later in S phase (T=60 min), concurrent with the decline in Orc1p protein levels. In contrast to the ARS1 chromosome, subsequent rebinding of Orc1p to rDNA chromatin was restricted to the origin region (T=150 min). The collective results indicate that ORC targeting to rDNA and non-rDNA origins occurs by different mechanisms. The MCM complex is selectively targeted to the ARS1-A region To address whether ARS1-A functions as an initiation site in the endogenous chromosome, we examined the association of Mcm6p, a component of the pre-RC that is recruited by ORC and subsequently moves with the replication fork (reviewed in Bell and Du
Referência(s)