A role for a replicator dominance mechanism in silencing
1999; Springer Nature; Volume: 18; Issue: 13 Linguagem: Inglês
10.1093/emboj/18.13.3808
ISSN1460-2075
Autores Tópico(s)Genomics and Chromatin Dynamics
ResumoArticle1 July 1999free access A role for a replicator dominance mechanism in silencing Madeleine A.Palacios DeBeer Madeleine A.Palacios DeBeer Department of Biomolecular Chemistry, 587 MSC 1300 University Avenue, University of Wisconsin, Madison, WI, 53706-1532 USA Search for more papers by this author Catherine A. Fox Corresponding Author Catherine A. Fox Department of Biomolecular Chemistry, 587 MSC 1300 University Avenue, University of Wisconsin, Madison, WI, 53706-1532 USA Search for more papers by this author Madeleine A.Palacios DeBeer Madeleine A.Palacios DeBeer Department of Biomolecular Chemistry, 587 MSC 1300 University Avenue, University of Wisconsin, Madison, WI, 53706-1532 USA Search for more papers by this author Catherine A. Fox Corresponding Author Catherine A. Fox Department of Biomolecular Chemistry, 587 MSC 1300 University Avenue, University of Wisconsin, Madison, WI, 53706-1532 USA Search for more papers by this author Author Information Madeleine A.Palacios DeBeer1 and Catherine A. Fox 1 1Department of Biomolecular Chemistry, 587 MSC 1300 University Avenue, University of Wisconsin, Madison, WI, 53706-1532 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1999)18:3808-3819https://doi.org/10.1093/emboj/18.13.3808 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The role of the natural HMR-E silencer in modulating replication initiation and silencing by the origin recognition complex (ORC) was examined. When natural HMR-E was the only silencer controlling HMR, the silencer's ORC-binding site (ACS) was dispensable for replication initiation but essential for silencing, indicating that a non-silencer chromosomal replicator(s) existed in close proximity to the silencer. Further analysis revealed that regions flanking both sides of HMR-E contained replicators. In contrast to replication initiation by the intact silencer, initiation by the non-silencer replicator(s) was abolished in an orc2-1 mutant, indicating that these replicators were extremely sensitive to defects in ORC. Remarkably, the activity of one of the non-silencer replicators correlated with reduced silencing; inactivation of these replicators caused by either the orc2-1 mutation or the deletion of flanking sequences enhanced silencing. These data were consistent with a role for the ORC bound to the HMR-E silencer ACS in suppressing the function of neighboring ORC molecules capable of inhibiting silencing, and indicated that differences in ORC-binding sites within HMR itself had profound effects on ORC function. Moreover, replication initiation by natural HMR-E was inefficient, suggesting that closely spaced replicators within HMR contributed to an inhibition of replication initiation. Introduction One characteristic of eukaryotic DNA replication is that individual chromosomes initiate replication at more than one distinct position, or replication origin, during the S-phase of the cell cycle (Hand, 1978; Fangman and Brewer, 1991; DePamphilis, 1996). Although many individual origins are used during a given S-phase, studies in yeast indicate that many of these origins are not required for efficient replication of the eukaryotic genome (Dershowitz and Newlon, 1993), suggesting that individual origins may play roles in chromosome function which extend beyond their direct contribution to chromosome duplication. Consistent with this view is the observation that individual yeast replication origins display unique characteristics. For example, some yeast origins are efficient, initiating once per cell cycle, whereas others are inefficient, initiating in only a fraction of cell divisions (Newlon et al., 1993; Newlon, 1996). Each origin initiates at a specific time during S-phase, with some origins initiating at the beginning and others at the end, after most of the genome has been replicated (Reynolds et al., 1989; Friedman et al., 1997; Yamashita et al., 1997). Moreover, there exists a small class of specialized origins which is closely associated with elements that control the expression of nearby genes (Loo and Rine, 1995). Identifying features that contribute to origins' distinct characteristics should provide insights into the relationships between chromosome maintenance and expression. At some level, the regulation of individual origin function must involve the origin recognition complex (ORC), the protein complex that binds to the conserved sequence, the autonomously replicating sequence (ARS) consensus sequence (ACS), which is common to yeast replicators. A replicator is defined as a genetic element that controls origin activity (Jacob et al., 1963; Stillman, 1993), and in yeast it appears that replicators and their origins are very close together if not coincident (Bielinsky and Gerbi, 1998). The ACS is an 11-bp AT-rich sequence that is necessary, but not sufficient, for replicator function (Van Houten and Newlon, 1990; Deshpande and Newlon, 1992; Marahrens and Stillman, 1992; Rivier and Rine, 1992; Huang and Kowalski, 1993). The ORC is a six-subunit protein complex identified by its ability to bind to the ACS of yeast replicators in an ATP-dependent manner (Bell and Stillman, 1992). Several independent studies indicate that ORC is the best candidate for the eukaryotic replication initiator. For example, each of the genes encoding the ORC subunits is essential in yeast (Foss et al., 1993; Li and Herskowitz, 1993; Micklem et al., 1993; Bell et al., 1995), and mutations in ORC genes cause initiation defects at individual chromosomal origins (Fox et al., 1995; Liang et al., 1995). In addition, ORC homologs have been identified in a number of eukaryotic organisms including humans (Gavin et al., 1995; Gossen et al., 1995), and in some systems have been shown to be required for DNA replication in vitro (Carpenter et al., 1996; Walter et al., 1998). Thus, it is probable that ORC is required for the fundamental function of replicators: the ability to direct origin unwinding. The distinct features of an individual origin may be influenced by a number of factors, including direct interactions between ORC and the DNA comprising a particular replicator (Lee and Bell, 1997), and/or by interactions between ORC and other proteins that bind near the ACS of a particular replicator. The HMR-E silencer in yeast is a member of the small class of replicators associated with elements that control the transcription of nearby genes (Loo and Rine, 1995). Specifically, HMR-E is a DNA element required for repression of the HMR silent mating-type locus, one of two silent mating-type loci that act as storage cassettes for copies of the yeast mating-type genes. In yeast, mating type is controlled by the genes present at the transcriptionally expressed MAT locus (Herskowitz et al., 1992). Copies of the mating-type genes also reside at HMR and HML, where they are repressed by a mechanism known as silencing. Silencing involves the assembly of a specialized chromatin structure analogous to heterochromatin and requires the action of DNA elements that flank each locus, called the E and I silencers (Loo and Rine, 1995). Several proteins are also required for silencing and these fall into two classes. The first includes the silencer-binding proteins ORC, Rap1p and Abf1p, which bind silencer DNA directly through their binding sites within the silencer itself (Shore, 1994; Dillin and Rine, 1995; Loo and Rine, 1995; Loo et al., 1995b). These silencer-binding proteins recruit the second class of proteins, characterized by the four Sir proteins, to silent mating-type cassettes through protein–protein interactions. In particular, evidence from several studies indicates that ORC functions in the recruitment of the Sir1p to the silencer through direct interactions (Chien et al., 1993; Triolo and Sternglanz, 1996; Fox et al., 1997; Gardner et al., 1999), and, in turn, Sir1p helps recruit the three other Sir proteins to HMR and HML, where they function as structural components of silent chromatin (Hecht et al., 1995, 1996; Grunstein, 1997; Strahl-Bolsinger et al., 1997). Significantly, the E and I silencers at both HMR and HML contain an ACS and can provide for autonomous replication of plasmids, indicating an association between silencer and origin function (Loo and Rine, 1995). In particular, the HMR-E silencer also functions as a replicator in its chromosomal context (Rivier and Rine, 1992). Studies of a simplified version of the HMR-E silencer called the synthetic silencer provide evidence for ORC's role in both the silencing and replication origin functions at HMR (Fox et al., 1995; Dillin and Rine, 1997). Although the synthetic version of HMR-E has been useful in the identification and initial characterization of ORC genes (McNally and Rine, 1991; Foss et al., 1993; Fox et al., 1995; Loo et al., 1995a; Dillin and Rine, 1997), evidence indicates that the natural HMR-E silencer possesses additional features that could provide further insights into the mechanisms modulating ORC function at this chromosomal domain. The initial characterization of ORC genes in silencing and replication at HMR exploited strains harboring an HMR locus that contained the synthetic silencer version of HMR-E, in place of the natural HMR-E silencer and a deletion of the HMR-I silencer element. In this relatively simplified context, mutations in the synthetic silencer's ORC-binding site (ACS), or mutations in individual ORC genes abolish or reduce both replication initiation and silencing at HMR (McNally and Rine, 1991; Rivier and Rine, 1992; Foss et al., 1993; Fox et al., 1995). In contrast, neither mutations in the natural HMR-E silencer's ACS (Brand et al., 1987) nor mutations in individual ORC genes reduce silencing at HMR (Foss et al., 1993; Fox et al., 1995), suggesting that additional elements at HMR substitute for ORC's silencing function. However, it is important to note that two different variables in these previous experiments prevent identification of the individual features of natural HMR-E that contribute to this redundancy. First, early studies indicating that the natural HMR-E silencer's ACS is not necessary for silencing HMR were performed in strains containing the HMR-I element. Although the HMR-I silencer is not sufficient for silencing HMR, it can modulate silencing efficiency at this locus (Abraham et al., 1984; Fox et al., 1995; Rivier et al., 1999) and it acts as a chromosomal origin (Rivier et al., 1999). Thus, the presence of HMR-I makes it difficult to analyze the role of the natural HMR-E silencer's ACS, or the role that the ORC plays in the natural HMR-E element's functions. Secondly, the natural HMR-E silencer itself contains near flanking sequences that were removed in the construction of the synthetic silencer, which may contribute to both silencing and replication initiation at HMR (McNally and Rine, 1991). For example, even in the absence of the HMR-I silencer, the replication and silencing functions of natural HMR-E are not reduced by the same mutations in ORC genes that abolish these functions in the synthetic silencer (Fox et al., 1995). Furthermore, in the absence of HMR-I, HMR-E is a more effective silencer than the synthetic silencer (Fox et al., 1995). Intriguingly, sequences that flank HMR-E contain several near matches to the ACS (Loo and Rine, 1995), and an early study indicates that a second region near HMR-E, but distinct from the silencer ACS, can confer autonomous replication to plasmids (Brand et al., 1987). A simple hypothesis to explain these observations is that the redundancy at natural HMR is due, in part, to sequences flanking the HMR-E silencer, which can bind additional ORC molecules and substitute for the role of the ORC bound to the HMR-E silencer's ACS. A test of this hypothesis requires examination of the functions of the natural HMR-E silencer within a chromosomal HMR locus that lacks the HMR-I silencer. In this report, we initiated an analysis of natural HMR-E's silencer and replication origin functions. To focus on the function of natural HMR-E and its flanking sequences, the HMR-I element was deleted from all strains examined. We addressed three issues relevant to the behavior of natural HMR-E. First, we determined whether the DNA region adjacent to natural HMR-E contributed to ORC's function in replication initiation at HMR. Secondly, we determined whether this DNA region contributed to ORC's role in silencing HMR. Thirdly, we examined replication initiation efficiency within an HMR locus under the control of the natural HMR-E silencer and determined how the silencer's ACS affected this efficiency. Our data provided evidence that ORC bound to the silencer ACS suppressed the function of neighboring non-silencer replicators that could direct efficient initiation from HMR, suggesting a second role for ORC in silencing beyond its previously characterized role in recruiting the Sir1 protein. Results Previous studies that established a requirement for ORC in both silencing and origin function at HMR exploited a simplified version of the HMR-E silencer called the synthetic silencer because it is particularly sensitive to defects in ORC. Specifically, in two-dimensional origin-mapping experiments to measure initiation on the chromosome, initiation by the synthetic silencer replicator is severely reduced in an orc2-1 mutant yeast strain (Fox et al., 1995; Figure 1A, panels 2 and 3). In addition, the mating properties of isogenic MATα strains indicate that silencing at HMR under the control of the synthetic silencer is reduced significantly in an orc2-1 mutant (Foss et al., 1993; Fox et al., 1995; Figure 2). In contrast, natural HMR-E's replicator and silencer functions are unaffected by the same orc2-1 mutation that reduces these functions of the synthetic silencer (Fox et al., 1995; Figures 1B and 2). In fact, in a two-dimensional origin-mapping experiment a slight, but reproducible, increase in the number of replication bubble intermediates, relative to small forks, is observed for the natural HMR-E in an orc2-1 mutant compared with a wild-type ORC strain, suggesting that the initiation frequency of natural HMR-E is enhanced slightly by defects in ORC (Fox et al., 1995; Figure 1B, panels 1 and 2). The mating properties of isogenic MATα strains indicate that silencing controlled by the natural HMR-E silencer is also not reduced in an orc2-1 mutant (Fox et al., 1995; Figure 2). Thus, in contrast to the synthetic silencer, natural HMR-E's silencing function is not reduced by the orc2-1 mutation. Furthermore, in contrast to synthetic silencer replicator and other replicators examined in an orc2-1 mutant (Fox et al., 1995; Liang et al., 1995), initiation by the natural HMR-E replicator is not reduced by a defect in ORC caused by the orc2-1 mutation. Figure 1.Replication initiation by the synthetic silencer, but not natural HMR-E, was sensitive to defects in ORC caused by the orc2-1 mutation. (A) Replication initiation was monitored at the synthetic silencer in two-dimensional origin-mapping experiments. Diagram of the HindIII–BglII HMR fragment containing the synthetic silencer examined in these experiments. The white box represents the synthetic silencer shown in expanded form below the fragment. The gray boxes represent the individual elements of the silencer: an ORC-binding site (ACS), a Rap1p-binding site and an Abf1p-binding site. Panel 1: Representation of how two-dimensional origin mapping gels distinguish replication initiation bubbles from replication forks. A qualitative measure of origin efficiency is reflected by the ratio of replication bubbles to small forks; all other factors being equal, a more efficiently used replication origin will have a larger replication bubble to small forks ratio (Fangman and Brewer, 1991). Panels 2 and 3: The results of two-dimensional origin-mapping experiments of a pair of isogenic strains containing the synthetic silencer at HMR and either wild-type ORC or an orc2-1 mutation (CFY36, CFY285). (B) Replication initiation was monitored at natural HMR-E in two-dimensional origin-mapping experiments. Diagram of the HindIII–BglII HMR fragment examined in these experiments. The thick black portion of the fragment flanking the silencer represents the region present in natural HMR-E but deleted from HMR containing the synthetic silencer. The white box represents the natural silencer which is shown in expanded form below the fragment. The binding sites within the silencer itself are represented by white boxes to indicate that the exact sequences differ from the analogous binding sites in the synthetic silencer represented by gray boxes in (A) (McNally and Rine, 1991). In addition, the line representing the natural silencer is thicker than that used to represent the synthetic silencer in (A) to indicate that the sequences in between the binding sites themselves differ between the natural and synthetic silencer (McNally and Rine, 1991). Panels 1 and 2: The results from two-dimensional origin-mapping experiments of a pair of isogenic strains containing the natural HMR-E origin at HMR and either wild-type ORC or an orc2-1 mutation (CFY37, CFY290). (C) The results from two-dimensional origin-mapping experiments of a set of isogenic strains containing natural HMR-E at HMR and either wild-type ORC or an orc2-1 mutation. The strains used to generate panels 2 and 3 also contain a null mutation in SIR2 (CFY37, CFY393, CFY391). The probe used to detect both the synthetic silencer and natural HMR-E-containing fragments is shown by a line marked with an asterisk below the fragment representations in (A) and (B). The primers used to generate the probe DNA fragment complementary to HMR were CTGGTCCTCACAGTTCGCAG and CAAGAAGTTCCCCTTGAAG. Download figure Download PowerPoint Figure 2.Results of quantitative mating assays performed with the strains used for this study. The data are presented as the log of the mating efficiencies of each strain. MATα strains to be tested were mated with an excess of a mating-type cells (CFY616) and diploids were selected at 27°C. The mating of MATα yeast cells with MATa cells can be selected for by selecting for the growth of diploids on selective media. Defects in silencing HMRa in the MATα strains being tested cause defects in the ability of those MATα cells to mate and form diploids that can grow on the selective media. Mating efficiencies are equal to the number of cells that mated divided by the number of viable cells for each strain, and provide a quantitative measure of silencing at HMRa. All the strains were isogenic MATα, lacked the HMR-I silencer at HMRa, and contained the version of the HMR-E silencer indicated together with either wild-type ORC or the orc2-1 mutation. The synthetic-silencer-containing strains were either ORC2 (gray bar, CFY36) or orc2-1 (black bar, CFY285). The natural HMR-E-containing strains with a wild-type ACS were either ORC2 (gray bar, CFY37) or orc2-1 (black bar, CFY290). The natural HMR-E-containing strains with a mutation in the silencer ACS (acs-) were either ORC2 (gray bar, CFY108) or orc2-1 (black bar, CFY201 or CFY143, left to right on the figure, respectively). The minimal HMR-E silencer-containing strain was ORC2 (ACS, gray bar, CFY3). The minimal HMR-E silencer-containing strains with a mutation in the silencer ACS (acs-) were either ORC2 (gray bar, CFY140) or orc2-1 (black bar, CFY244). Download figure Download PowerPoint One feature of the origin associated with HMR-E that distinguishes it from most other origins is that it resides within a region of DNA, the HMR locus, that is assembled into a repressive, or silenced, form of chromatin. Therefore, to determine whether the transcriptional state at HMR influenced the effects of the orc2-1 mutation on replication initiation by natural HMR-E, we evaluated replication initiation by HMR-E in a set of isogenic strains containing a deletion of SIR2 (Figure 1C). The SIR2 gene encodes the Sir2 protein, one of four Sir proteins essential for silencing HMR (Loo and Rine, 1995). In the absence of SIR2, HMR is assembled into a transcriptionally active form of chromatin. Deletion of SIR2 caused a slight enhancement in replication initiation frequency at HMR-E in a two-dimensional origin-mapping experiment (Figure 1C, compare panel 1 with panel 2), indicating that the silenced state did inhibit initiation within HMR somewhat. However, even in absence of SIR2, the orc2-1 mutant yeast cells initiated replication within HMR more frequently than isogenic ORC2 cells, based on the results of a two-dimensional origin-mapping experiment (Figure 1C, compare panel 2 with panel 3). Thus, the transcriptional state at HMR did not significantly influence the effect of the orc2-1 mutation on replication initiation from HMR-E. Since the orc2-1 mutation causes obvious defects in replication initiation efficiency at several origins (Fox et al., 1995; Liang et al., 1995), the effect of this mutation on initiation by natural HMR-E was exceptional. A simple hypothesis to explain the unusual behavior of the natural HMR-E origin was that sequences adjacent to natural HMR-E could substitute for the role of the ORC-binding site within the defined HMR-E silencer itself. Natural HMR contained non-silencer replicator activity If the DNA region immediately adjacent to natural HMR-E contained an additional ORC-binding site(s), which could substitute for the functions of the silencer ORC-binding site (ACS), then the natural HMR-E silencer's ACS would be dispensable for both replication initiation and silencing at HMR. Therefore, we mutated the ACS within the defined HMR-E silencer and determined the effect of this mutation on replication initiation and silencing at HMR (Figure 3A). Figure 3.The ACS within the natural HMR-E silencer was dispensable for initiation but was required for silencing. (A) The results from two-dimensional origin-mapping experiments of a pair of isogenic strains containing natural HMR-E at HMR with either a wild-type ACS or a mutant ACS (acs-) within the silencer (CFY37, CFY108). The probe used for this experiment was the same as that described in Figure 1. (B) The results from mating experiments with the same strains used in (A). The MATα strains were grown on rich media at 23°C for 24 h and then replica-plated into minimal media containing a lawn of MATa cells at 27°C. Growth of diploid cells on minimal media reflected the extent of mating and thus the degree of silencing at HMRa. Download figure Download PowerPoint Mutation of the ACS in HMR-E [HMR-E(acs-)] caused no reduction in replication initiation efficiency at HMR, as measured by a two-dimensional origin-mapping experiment (Figure 3A). This behavior was in contrast to that of the synthetic silencer, which requires its ACS for replicator function (Rivier and Rine, 1992). Therefore the single exact match to an ACS within HMR, the ACS within the defined HMR-E silencer itself, was not required for chromosomal origin function at an HMR locus controlled by natural HMR-E. Thus, natural HMR contained an additional ORC-binding site(s) in the close vicinity of the HMR-E silencer, as reflected by replication initiation, which occurred independently of the HMR-E silencer ACS, and a pattern of replication intermediates indistinguishable from that formed by wild-type HMR-E. If the additional replicator activity at natural HMR was also providing ORC-dependent silencing activity, then the ACS in natural HMR-E would not be required for silencing at HMR. Therefore, silencing by the mutant silencer was measured by comparing the mating properties of an isogenic pair of MATα strains containing either the natural wild-type HMR-E silencer or the natural HMR-E silencer with a mutant ACS at HMRa (Figures 2 and 3B). Significantly, the mutant HMR-E(acs-) silencer failed to provide for efficient silencing at HMRa. Therefore, the replicator activity that was independent of the silencer ACS was referred to as the non-silencer replicator. Since the ACS in HMR-E is dispensable for silencing in the presence of the HMR-I silencer (Brand et al., 1987), these data provided additional evidence for a role of the HMR-I element in modulating silencing (Abraham et al., 1984; Fox et al., 1995; Rivier et al., 1999). More importantly, the origin activity that remained at HMR in the absence of the silencer ACS indicated that the remaining ORC(s), which bound DNA in the close vicinity of the silencer and provided for replicator function, failed to provide for efficient silencing, even though the other elements of the silencer, the Rap1p- and Abf1p-binding sites, were still present. Reduced ORC function enhanced silencing by the mutant HMR-E silencer Although silencing by the mutant HMR-E(acs-) silencer was reduced significantly, it was not abolished. One possibility was that the ORC that functioned at the non-silencer replicator contributed a small amount of residual silencing activity. If this were true then the orc2-1 allele, which reduces the amount of functional ORC in a cell (Bell et al., 1993), might reduce further the small amount of silencing at an HMRa locus controlled by the mutant HMR-E(acs-) silencer. Therefore, the effect of the orc2-1 mutation on silencing at HMRa was measured by comparing the mating properties of an isogenic set of MATα strains containing the mutant HMR-E(acs-) silencer combined with wild-type ORC or the orc2-1 allele (Figures 2 and 4A). Figure 4.Reduced ORC function enhanced silencing by the mutant HMR-E(acs-) silencer and abolished initiation by the non-silencer replicator(s). (A) The results from patch-mating experiments of isogenic MATα strains containing the mutant natural HMR-E(acs-) silencer at HMRa and either wild-type ORC or an orc2-1 mutation (CFY108, CFY201). The strains were grown on rich media at 23°C for 24 h and then replica-plated to minimal media containing a lawn of MATa cells at 27°C. Growth of diploid cells on minimal media reflected the extent of mating and thus the degree of silencing at HMRa. (B) The results from two-dimensional origin-mapping experiments of a set of isogenic MATα strains containing the natural HMR-E origin at HMR with either a wild-type or mutant ACS and wild-type ORC or an orc2-1 mutation (CFY37, CFY108, CFY201). The probe used for this experiment was the same as that described in Figure 1. Download figure Download PowerPoint Surprisingly, the orc2-1 mutation significantly improved silencing at an HMRa locus controlled by the mutant HMR-E(acs-) silencer (Figures 2 and 4A). The ability of the orc2-1 mutation to enhance silencing by the mutant HMR-E(acs-) silencer was recessive, as are the other phenotypes caused by the orc2-1 mutation (Foss et al., 1993), strongly suggesting that a loss of ORC function was responsible for the enhanced silencing phenotype (M.A.Palacios DeBeer and C.A.Fox, unpublished data). Since the orc2-1 mutation causes defects in replication initiation at chromosomal origins, these data provided evidence that ORC's replication function could inhibit the formation of silent chromatin at HMR. Initiation by non-silencer replicator was abolished by the orc2-1 mutation The above data provided evidence that, in the absence of the silencer ACS, ORC's replication function could inhibit the assembly of silent chromatin at HMR. If this ORC-dependent inhibitory activity was a result of ORC bound in the near vicinity of HMR, then it was possible that ORC function at the non-silencer replicator itself actually inhibited silencing. If this were true, then initiation controlled by the non-silencer replicator might be reduced by the orc2-1 mutation. Therefore, non-silencer replicator activity was evaluated directly in isogenic ORC2 and orc2-1 strains (Figure 4B). Strikingly, by two-dimensional origin-mapping experiments, the orc2-1 mutation caused a drastic reduction in non-silencer replicator activity at HMR. Thus, in the absence of the silencer ACS, the remaining origin activity at HMR, which was at least as robust as the origin activity of the intact silencer as measured by two-dimensional origin mapping gels (Figure 4B, compare panel 1 with panel 2), was extremely sensitive to a reduction in ORC activity caused by the orc2-1 mutation. A region adjacent to HMR-E was required for both non-silencer replicator activity and ORC-dependent inhibition of silencing The data described above were consistent with the view that the orc2-1 mutation enhanced silencing by the mutant silencer by reducing the function of the non-silencer replicator at HMR. If this view were correct, deletion of the non-silencer replicator would also enhance silencing by the mutant HMR-E(acs-) silencer. At first it seems as though the synthetic silencer could be used to address this issue, because an HMR locus controlled by this silencer lacks the non-silencer replicator; mutation of the synthetic silencer ACS abolishes all origin function at HMR, indicating that the synthetic silencer is the only functional replicator in the vicinity of an HMR locus lacking the HMR-I element (Rivier and Rine, 1992; Fox et al., 1995). However, the synthetic silencer also differs in a number of other ways from natural HMR-E (McNally and Rine, 1991). Therefore, to focus on the region adjacent to HMR-E, we constructed a minimal version of HMR-E (minimal HMR-E; Figure 5A). The minimal HMR-E silencer was identical to natural HMR-E except that it lacked the region of DNA surrounding the defined silencer (compare Figure 1B with 5A). Figure 5.The minimal HMR-E silencer required its ACS for efficient initiation but not for silencing. (A) Diagram of the HindIII–BglII H
Referência(s)