1 Opposing roles for DNA replication initiator proteins ORC1

2 and CDC6 in control of Cyclin E transcription 3 4 Manzar Hossain and Bruce Stillman 5 6 Cold Spring Harbor Laboratory 7 1 Bungtown Road 8 Cold Spring Harbor, NY 11724 USA 9 10 Corresponding author: Bruce Stillman, [email protected] 11 12 Abstract 13 Newly-born cells either continue to proliferate or exit the cell division cycle. This 14 decision involves delaying expression of Cyclin E that promotes DNA replication. 15 ORC1, the Origin Recognition Complex (ORC) large subunit, is inherited into 16 newly-born cells after it binds to condensing during the preceding 17 mitosis. We demonstrate that ORC1 represses Cyclin E gene (CCNE1) 18 transcription, an E2F1 activated gene that is also repressed by the 19 Retinoblastoma (RB) protein. ORC1 binds to RB, the histone methyltransferase 20 SUV39H1 and to its repressive histone H3K9me3 mark. ORC1 cooperates with 21 SUV39H1 and RB protein to repress E2F1-dependent CCNE1 transcription. In 22 contrast, the ORC1-related replication protein CDC6 binds Cyclin E-CDK2 kinase 23 and in a feedback loop removes RB from ORC1, thereby hyper-activating 24 CCNE1 transcription. The opposing effects of ORC1 and CDC6 in controlling the 25 level of Cyclin E ensures genome stability and a mechanism for linking directly 26 DNA replication and cell division commitment. 27 28 Conflicts: The authors represent that there are no conflicts to declare. 29 30

1 31 Impact Statement 32 Two related DNA replication initiation proteins contribute to the decision of 33 whether to enter a new round of the cell division cycle or enter into a period of 34 proliferative quiescence. 35 36 Introduction 37 In addition to its role in the initiation of DNA replication, ORC1, the largest 38 subunit of the Origin Recognition Complex (ORC) controls Cyclin E-dependent 39 duplication of centrosomes and centrioles in cells by acting as an inhibitor of 40 Cyclin E-CDK2 activity (Hemerly et al., 2009, Hossain and Stillman, 2012). The 41 Cyclin E-CDK2 kinase inhibitory activity is compromised by ORC1 Meier-Gorlin 42 Syndrome mutations that also alter the interaction between ORC1 and histone 43 H4K20me2 (Hossain and Stillman, 2012, Kuo et al., 2012, Zhang et al., 2015, 44 Bicknell et al., 2011b, Bicknell et al., 2011a, de Munnik et al., 2012). Unlike the 45 well-characterized yeast complex that is a stable, six subunit complex throughout 46 the cell division cycle, ORC in human cells is a very dynamic complex (Sasaki 47 and Gilbert, 2007, DePamphilis, 2005). ORC1 binds to mitotic chromosomes as 48 cells enter into mitosis (Kara et al., 2015, Okuno et al., 2001), and in human cells 49 it is modified by ubiquitin and then degraded during the G1 to S phase transition 50 (Abdurashidova et al., 2003, Kara et al., 2015, Kreitz et al., 2001, Mendez et al., 51 2002, Ohta et al., 2003, Siddiqui and Stillman, 2007, Tatsumi et al., 2000). The 52 assembly of the full ORC occurs in mid G1 phase of the cell division cycle in 53 preparation for its role in assembly of the pre-replicative complex (pre-RC) at 54 sites across chromosomes (Kara et al., 2015, Siddiqui and Stillman, 2007). 55 56 The ORC1-related protein CDC6 is also required for pre-RC assembly, but it is 57 targeted for proteasome degradation by the SCFCyclin F ubiquitin ligase complex 58 late in the cell cycle and the anaphase-promoting complex/cyclosome (APC/C) in 59 early G1 phase and then stabilized in mid G1 phase by Cyclin E-CDK2 mediated 60 phosphorylation (Mailand and Diffley, 2005, Petersen et al., 2000, Walter et al., 61 2016). This phosphorylation is mediated by the direct interaction between Cyclin

2 62 E and CDC6 and CDC6 and Cyclin E-CDK2 cooperate to promote the initiation of 63 DNA replication (Coverley et al., 2002, Furstenthal et al., 2001, Cook et al., 64 2002). 65 66 As proliferating cells divide, they must make a decision whether to continue to 67 proliferate or enter into proliferative quiescence. This decision is made by a 68 complex regulatory process known as START in yeast and the restriction point in 69 mammalian cells (Johnson and Skotheim, 2013). Key among these regulators 70 are the Cyclin D-CDK4/6 kinases that mono-phosphorylate the Retinoblastoma 71 protein (RB) and contributes to the release of repression of E2F-transcription 72 factors (Narasimha et al., 2014, Ewen et al., 1993, Hinds et al., 1992, Lundberg 73 and Weinberg, 1998, Resnitzky et al., 1994). E2F1-regulated include 74 genes encoding Cyclin E (CCNE1) and CDC6 (Hateboer et al., 1998, Ohtani et 75 al., 1998, Yan et al., 1998, DeGregori et al., 1995). Cyclin E-CDK2 amplifies the 76 phosphorylation of RB, but how this is achieved is not known (Narasimha et al., 77 2014). Here we demonstrate that ORC1 is required for repression of the gene 78 encoding Cyclin E and that CDC6 is involved in relieving the repression in 79 cooperation with Cyclin E-CDK2. We suggest that ORC1 establishes a period in 80 the newly-born cells during which Cyclin E is not expressed, allowing time for the 81 cells to decide whether to proliferate again or enter into replicative quiescence. 82 83 Results 84 ORC1 binds to RB and SUV39H1 and represses CCNE1 gene 85 transcription 86 Apart from its role in DNA replication, human ORC1 controls centriole and 87 centrosome copy number by binding and inhibiting the kinase activity of Cyclin E- 88 CDK2 (Hemerly et al., 2009). That work suggested that ORC1 might also control 89 Cyclin E by regulating its protein level during the G1 phase of the cell division 90 cycle. To determine if this was the case, ORC1 was depleted using siRNA in 91 U2OS cells that had been synchronized in mitosis by nocodazole treatment and 92 released into the next cell cycle. As early as 9hr post release, Cyclin E protein

3 93 level was elevated in ORC1-depleted U2OS cells, compared to control siRNA 94 treated cells (Figure 1A). The expression of CCNE1 mRNA increased at all times 95 following ORC1 depletion in synchronized U2OS cells (Figure 1B) and 96 quantitation of multiple experiments showed significant increases from 6-24 97 hours post release (Figure 1-figure supplement 1A-D). This data suggests that 98 ORC1 inhibits CCNE1 . 99 Transcription of the gene encoding Cyclin E (CCNE1) is known to be 100 regulated by the E2F1 transcription factor and to be repressed by retinoblastoma 101 protein (RB) (Nielsen et al., 2001, Geng et al., 1996, DeGregori et al., 1995), a 102 member of the so-called “pocket protein” family (Dick and Rubin, 2013, Dyson, 103 1998, Giacinti and Giordano, 2006). Since ORC1 binds RB (Mendoza- 104 Maldonado et al., 2010) we explored a role for ORC1 in transcriptional repression 105 of the CCNE1 gene. 106 RB was expressed as a fusion to the Maltose Binding Protein (MBP) and it 107 bound to S35-labelled ORC1 protein (Figure 1-figure supplement 2A). RB binds 108 other binding partners dependent on a canonical LxCxE motif and although D 109 ORC1 has a conserved LPCR /E sequence, it was not required for the interaction 110 between ORC1 and RB (Figure 1-figure supplement 2A and B). Consistent with 111 this finding, two pocket mutants within RB (R661W and N757F) that are defective 112 in binding to LxCxE containing proteins (Chen and Wang, 2000) showed no 113 defect in binding to ORC1 in vitro (Figure 1-figure supplement 2C) or in vivo 114 (Figure 1C). In fact, the mutant RB proteins bound ORC1 better than the wild 115 type, perhaps due to loss of competition between ORC1 and other RB binding 116 proteins or due to conformational changes in RB. When Green Fluorescent 117 Protein (GFP)-RB fusion protein was expressed in cells with ORC1-Flag-tagged 118 protein, the ORC1-Flag protein bound WT and pocket mutant RB, but other ORC 119 subunits did not bind to RB (Figure 1D). We confirmed this observation by 120 showing that immunoprecipitation with anti-ORC2 or anti-ORC3 antibodies failed 121 to precipitate GFP-RB or its mutants, but did bind ORC1-Flag (Figure 1E). Given 122 that the ORC1-RB interaction is independent of other ORC subunits, it suggests

4 123 that ORC1 has additional functions that are separate from its role in DNA 124 replication. 125 The phosphorylation of RB is an important step that relieves repression of 126 E2F target genes (Rubin, 2013). In the presence of purified Cyclin E-CDK2 127 kinase plus ATP RB no longer bound to ORC1 (Figure 1-figure supplement 2A 128 and 2C), suggesting that Cyclin E might feedback and relieve repression of the 129 CCNE1 gene by disrupting the RB-ORC1 interaction. 130 RB interacts with chromatin and histone modifying enzymes to repress 131 E2F1 transcription activity (Dick and Rubin, 2013). Specifically, it has been 132 suggested that RB binds to the SUV39H1 histone methyltransferase that tri- 133 methylates histone H3K9 and then HP1α binds to histone H3K9me3, and that 134 these direct interactions contribute to repression of CCNE1 transcription (Nielsen 135 et al., 2001). It is important to note, however, that others suggested that another 136 protein mediates the RB-SUV39H1 interaction (Vandel et al., 2001). Since ORC1 137 interacts with both RB and HP1α (Prasanth et al., 2010), we tested whether 138 ORC1 also interacts with SUV39H1. Using purified proteins we found that both 139 RB and ORC1 directly bound to SUV39H1, and ORC1 directly bound to CDC6 140 (Saha et al., 1998) (Figure 2A and Figure 2-figure supplement 1). Under these 141 conditions ORC1 bound to HP1α weakly, but the binding increased when higher 142 levels of GST-HP1α were added, consistent with published data (Prasanth et al., 143 2010). Although both RB and ORC1 directly interacted with SUV39H1, ORC1 144 bound SUV39H1 much better than RB (Figure 2-figure supplement 2 A-D). 145 Consistent with the in vitro data, we observed that anti-SUV39H1 antibodies 146 precipitated ORC1 and RB in the RB positive cells (U2OS and MCF7), with 147 ORC1 being the predominant interacting protein in MCF7 cells (Figure 2B). 148 Furthermore, the interaction between SUV39H1 with ORC1 protein did not 149 require RB because anti-ORC1 antibodies co-precipitated SUV39H1 from RB 150 negative SaOS-2 cells (Figure 2C). This observation was confirmed when GFP- 151 RB or GFP-ORC1 were transiently expressed with T7-SUV39H1 in 293 cells; 152 ORC1 more readily bound to SUV39H1 (Figure 2D) whereas higher levels of 153 GFP-RB were required to observe an interaction with SUV39H1 (Figure 2-figure

5 154 supplement 3A). Domain mapping demonstrated that SUV39H1 interacted with 155 ORC1 through its SET domain containing C-terminus, which is required for its 156 histone methyltransferase (HMT) activity. Moreover, SUV39H1 interacted with 157 ORC1 and ORC1 mutants that could not be phosphorylated by Cyclin-CDK, and 158 did not interact in a complex with other ORC subunits, similar to the ORC1-RB 159 interaction (Figure 2-figure supplement 3B-C). These data suggest that the 160 ORC1-SUV39H1 interaction does not require other ORC subunits or CDK 161 phosphorylation of ORC1, and therefore ORC1 plays a role independent of its 162 role in DNA replication. 163 Since SUV39H1 tri-methylates histone H3K9 after its de-acetylation by 164 HDAC1, a prerequisite step for establishing CCNE1 gene repression, we 165 explored the ORC1 protein interactions with different histone H3 modifications 166 (Nicolas et al., 2003, Stewart et al., 2005, Vaute et al., 2002). Purified ORC1 167 bound to unmodified or mono-, di- or tri-methylated H3K9 or H3K14-tri-methyl 168 modifications, while the interaction was abolished when histone H3 was 169 acetylated at the same positions or when H3 serine-10 was phosphorylated, 170 which normally occurs during mitosis (Figure 2E). Thus ORC1 can bind to RB, 171 SUV39H1 and to the repressive H3K9-me3 modification on histone H3, 172 suggesting that it might mediate repression of E2F1-dependent CCNE1 173 transcription. 174 To test if ORC1 has a role in repression of the E2F1-regulated CCNE1 175 gene, the CCNE1 promoter was linked to the luciferase coding region to create a 176 reporter for CCNE1 gene transcription (Geng et al., 1996). Transfection of 177 plasmids expressing E2F1 and its binding partner DP1 activated gene 178 expression, whereas expression of ORC1 or SUV39H1 protein repressed 179 CCNE1 expression in a dose-dependent manner (Figure 2F). In a GAL4-based 180 reporter gene assay, expression of either GAL4-ORC1 or GAL4-SUV39H1 also 181 repressed transcription of the reporter gene in a dose-dependent manner (Figure 182 2-figure supplement 4A). In contrast, neither expression of ORC3 nor ORC4 183 altered E2F1 driven CCNE1 transcription (Figure 2-figure supplement 4B). 184 Moreover, SUV39H1 co-operated with ORC1 to further repress the CCNE1

6 185 promoter, but a mutant of SUV39H1 (H324K) (Li et al., 2002, Rea et al., 2000, 186 Stewart et al., 2005) that has lost its catalytic activity was unable to do so (Figure 187 2G). We therefore conclude that ORC1-SUV39H1 co-operation for transcriptional 188 repression of the CCNE1 gene is mediated through the HMT activity of 189 SUV39H1. 190 Cell cycle dependent association of ORC1, RB, SUV39H1 and CDC6 191 proteins with the CCNE1 gene promoter 192 Having established that ORC1 can repress CCNE1 gene transcription in 193 cooperation with the methyltransferase activity of SUV39H1, we investigated the 194 positioning of ORC1, RB and SUV39H1 proteins within the CCNE1 promoter in 195 vivo. A previous publication (Dellino et al., 2013) reported ORC1 chromatin 196 immuno-precipitation following crosslinking (ChIP) and re-analysis this whole 197 genome Chip-Seq data revealed a weak ORC1 peak (peak height of 7 reads) 198 within the CCNE1 promoter, however a duplicate was not reported. Therefore, to 199 test whether ORC1 associated with the CCNE1 promoter, we modified a 200 chromatin immunoprecipitation (ChIP) method for ORC1 initially using 201 asynchronously growing MCF7 cells and the immunoprecipitated DNA was 202 analyzed by polymerase chain reaction (PCR) using multiple primer pairs across 203 the CCNE1 promoter (Figure 3A). ORC1 bound to the CCNE1 promoter 204 encompassing the region from -280 to +63 base pairs (probes E and F), a region 205 that is known to bind the E2F1 transcription factor (see Gene Expression 206 Omnibus [GEO] transcription factor binding site accession numbers GSM935484 207 and GSM935477) and contains five E2F1 consensus binding sites (red bars, 208 Figure 3A). Our ChIP analysis also showed that RB bound to the same regions 209 bound by ORC1, but not to other probes (Figure 3B). When a similar ChIP 210 analysis was performed using antibodies targeted to SUV39H1, histone 211 H3K9me3 and CDC6, these proteins bound to the same region (-342 to +63) of 212 the CCNE1 promoter (Figure 3C, also see Figure 4 below). 213 We next studied the temporal dynamics of protein binding to the CCNE1 214 promoter during the G1 phase of the cell cycle in synchronized U2OS cells. In 215 this analysis, cells were blocked in mitosis with nocodazole and released into the

7 216 next G1 phase and ChIP analyses for ORC1, CDC6, RB and SUV39H1 were 217 performed using primer pairs B and E (Figures 4A and B). ORC1 and RB bound 218 to the probe E region, but not the probe B region, at 3 hr post mitosis, and then 219 binding was reduced at 6 hr and eliminated by 9 hr (Figure 4A), even though 220 ORC1 protein remained in the cell (Figure 4C). SUV39H1 was detected at 3 hr 221 and 6 hr, but not at 9 hr. Interestingly, CDC6 transiently bound to the promoter 222 only at 6 hr (Figure 4B), precisely the time when Cyclin E proteins levels 223 dramatically increase (Figure 4C). In the following section, we investigated the 224 physiological role of CDC6 binding to CCNE1 promoter. 225 226 CDC6 cooperates with Cyclin E-CDK2 to remove RB from ORC1 and 227 activate CCNE1 gene transcription 228 Cyclin E-CDK2 cooperates with CDC6 to stimulate entry from G1 phase 229 into S-phase of the mammalian cell division cycle (Cook et al., 2002, Coverley et 230 al., 2002, Hateboer et al., 1998). Since CDC6 protein levels increase during late 231 G1 phase (Hateboer et al., 1998, Mendez and Stillman, 2000) and CDC6 binds 232 ORC1 (Figure 2A), we hypothesized that CDC6 may help alleviate ORC1- 233 mediated repression of the CCNE1 gene. CDC6 and Cyclin E-CDK2 were 234 purified and increasing amounts were titrated into a mixture containing purified 235 RB and ORC1 (Figure 5-figure supplement 1) and the interaction between RB 236 and ORC1 was monitored. MBP-ORC1 bound to CDC6 (Figure 5A) and 237 increasing amounts of Cyclin E-CDK2 did not interfere with this interaction 238 (Figure 5B). Moreover, increased binding between CDC6 and ORC1 did not 239 interfere with binding between ORC1 and RB (Figure 5B). Cyclin E-CDK2 240 disrupted the interaction between ORC1 and RB (Figure 5C and Figure 2-figure 241 supplement 2A), but it took relatively high levels of Cyclin E-CDK2. Importantly, 242 the addition of CDC6 along with CyclinE-CDK2 cooperatively disrupted the 243 interaction between ORC1 and RB, even at low concentrations of Cyclin E-CDK2 244 (Figure 5C). As shown before with Xenopus proteins (Furstenthal et al., 2001), 245 human CDC6 bound to Cyclin E-CDK2 in a manner dependent on the CDC6

8 246 Cyclin “Cy” binding motif CDC694RRL96 but not to the CDC694ARA96 “Cy” mutant 247 (Figure 5D). 248 Based upon the biochemical data, we hypothesized that CDC6 and Cyclin 249 E-CDK2 cooperated to relieve ORC1-SUV39H1-RB mediated repression of 250 CCNE1. If this is the case, then CDC6 overexpression should increase CCNE1 251 expression in vivo. To test this hypothesis, we expressed GFP-tagged CDC6 wild 252 type or its “Cy” mutant in nocodazole arrested U2OS cells and released them 253 from the nocodazole block to estimate the level of endogenous Cyclin E protein 254 at different times in G1. Expression of wild type CDC6 but not its “Cy” mutant led 255 to an increase in the endogenous level of Cyclin E protein (Figure 5E). 256 Consistent with these results, overexpression of CDC6 wild type but not its “Cy” 257 mutant further enhanced E2F1-DP1 activated transcription from the CCNE1 258 promoter (Figure 5F). Based on these data, we suggest that CDC6 co-operates 259 with Cyclin E-CDK2 kinase to abolish the interaction between RB and ORC1, 260 contributing to alleviating the repression imposed by RB on the CCNE1 gene, 261 leading to CCNE1 gene transcription. 262 Data presented so far suggests that ORC1 mediated CCNE1 263 transcriptional repression also requires SUV39H1. We therefore transfected into 264 RB+ U2OS cells a CCNE1-luciferase reporter plasmid with E2F1-DP1 and 265 depleted ORC1, SUV39H1 or CDC6 using siRNA (Figure 6A). Depletion of either 266 ORC1 or SUV39H1 with two different siRNAs led to significant increases in 267 CCNE1 promoter activity above its basal, E2F1/DP1-dependent level. In 268 contrast, depletion of CDC6 protein had no effect on the basal, E2F1/DP1- 269 dependent promoter activity (Figure 6A). We further investigated the binding to 270 the CCNE1 promoter of SUV39H1 and the presence of its product, the histone 271 H3K9me3 mark, upon depletion of ORC1 protein in asynchronously growing 272 U2OS cells by ChIP assay. Upon depletion of ORC1 compared to control siRNA 273 the methyltransferase activity of SUV39H1 was drastically reduced on the 274 CCNE1 promoter as evident by a dramatic reduction in the histone H3K9m3 275 mark at the promoter, while the binding of SUV39H1 was only slightly reduced 276 (Figure 6B). The reduced binding of H3K9me3 was most evident within the

9 277 region -280 to -143 of the CCNE1 promoter, where ORC1 binding was centered 278 (Figure 6B, Figure 6 supplement 1). Depletion of ORC1 also resulted in reduced 279 levels of the histone H3K9me3 mark as well as a slight reduction in the level of 280 SUV39H1 protein (Figure 6C). Our results support the hypothesis that ORC1 is 281 involved in transcription repression by recruiting the SUV39H1 protein, which 282 thereby creates the histone H3K9me3 transcriptional repressor mark on CCNE1 283 promoter. 284 To separate the function of ORC1 as a transcription co-repressor from its 285 well-established role in DNA replication, we identified C-terminus truncation 286 mutants (1-700 aa and 1-768 aa) of ORC1 (full length ORC1 is 1-861 aa) that 287 were defective in binding to the other ORC subunits, but still capable of binding 288 RB and SUV39H1 (Figure 7A-C). These ORC1 mutants were fully active in 289 repressing CCNE1 transcription (Figure 7D), demonstrating that the effects of 290 ORC1 on CCNE1 gene repression were not due to an indirect effect of the role of 291 ORC in DNA replication. 292 293 Discussion 294 ORC1 in plants is known to activate transcription in via a plant 295 homeodomain (PHD) that is not found in animal and fungi ORC1 (Sasaki and 296 Gilbert, 2007, de la Paz Sanchez and Gutierrez, 2009). In contrast, ORC1 in the 297 budding yeast S. cerevisiae binds to the Sir1 protein and is involved in repression 298 of transcription of mating type genes (Bell et al., 1993, Triolo and Sternglanz, 299 1996). In Drosophila and animal cells, ORC, including human ORC1, binds the 300 HP1 proteins that are often involved in transcriptional repression, but a direct link 301 between this interaction and transcription repression has not been established 302 (Auth et al., 2006, Badugu et al., 2003, Lidonnici et al., 2004, Pak et al., 1997, 303 Prasanth et al., 2010, Sasaki and Gilbert, 2007). In this report, we have 304 established opposing roles for ORC1 and CDC6 in control of transcription of the 305 Cyclin E gene CCNE1. ORC1, with RB and SUV39H1, bind to the CCNE1 306 promoter adjacent to the E2F1 transcription factor binding sites and repress the 307 E2F1-dependent promoter and this repression is relieved by CDC6 that is bound

10 308! to Cyclin E-CDK2 kinase. Consistent with this model, CDC6 binds to the CCNE1 309! promoter transiently, just at the time during G1 phase when Cyclin E levels 310! increase dramatically. We suggest that low levels of both Cyclin E and CDC6, 311! both E2F1-controlled genes, appear upon activation of Cyclin D-CDK4 312! (Narasimha et al., 2014). The expression of CDC6-Cyclin E-CDK2 provides 313! positive feedback, disrupting the repressive RB-ORC1 interaction. The disruption 314! of RB-ORC1 interaction further weakens ORC1 association with the promoter 315! and results in a dramatic reduction in the repressive histone H3K9me3 mark at 316! the promoter, increasing CCNE1 gene transcription and amplifying the levels of 317! CDC6 and Cyclin E-CDK2 (Figure 8, bottom panel). The opposing expression 318! levels of ORC1 and CDC6 during the cell cycle (Figure 8, top panel) are 319! consistent with this model. The production of higher levels of Cyclin E-CDK2 and 320! CDC6 can then function in pre-RC assembly, as they are known to do (Cook et 321! al., 2002, Coverley et al., 2002, Hateboer et al., 1998). 322! Our work also shows that ORC1 binds to CCNE1 promoter and recruits 323! SUV39H1, forming the repressive H3K9me3 mark that can then bind the HP1 324! protein. Interestingly, HP1 is known to bind both SUV39H1 and ORC1 (Prasanth 325! et al., 2010, Stewart et al., 2005). Thus ORC1 participates in multiple protein- 326! protein interactions to ensure stable repression of CCNE1 and perhaps other 327! E2F1-regulated genes during early G1 phase. 328! In this way ORC1, which binds mitotic chromosomes and is then inherited 329! into the daughter cells (Kara et al., 2015, Okuno et al., 2001), creates an 330! opportunity to repress CCNE1 gene transcription and allow the newborn cells 331! time to integrate information to decide whether to proliferate or exit the cell 332! division cycle. If proliferation and cell division are destined to occur, Cyclin D- 333! CDK4/6 mono-phosphorylates RB and primes it for Cyclin E-dependent 334! activation of E2F regulated genes (Narasimha et al., 2014). We show that low 335! levels of CDC6-Cycin-E-CDK2 can, in a feedback loop, amplify this commitment 336! by antagonizing ORC1-RB interactions (Figure 8). Consistent with the model, we 337! demonstrated that over-expression of CDC6, but not a CDC6 that cannot bind 338! Cyclin E, enhances the levels of endogenous Cyclin E during the very earliest

! 11 339 period of G1 phase. Cell-cycle regulated Cyclin E is essential for maintenance of 340 genome stability since cancer cells that have de-regulated Cyclin E fail to 341 produce sufficient pre-RCs during G1 phase and as a consequence have 342 problems in S phase and accumulate DNA damage (Ekholm-Reed et al., 2004, 343 Jones et al., 2013, Odajima et al., 2010). Of interest is our observation that at the 344 time CDC6 is recruited to the CCNE1 promoter, ORC1 binding to the promoter 345 declines. 346 In human cells, ORC1 shows a dynamic, temporally regulated nuclear 347 localization pattern such that in early G1 phase it is distributed in a punctate 348 pattern throughout the nucleus, but in late G1 phase, ORC1 predominantly binds 349 to regions of the genome that replicate late in the subsequent S phase (Kara et 350 al., 2015). We suggest that the events that occur at the CCNE1 promoter, ORC1 351 binding first and recruitment of SUV39H1 and RB, followed by recruitment of 352 Cyclin E-CDK2 and CDC6, may occur at many ORC1 binding sites in the 353 genome, even those sites that are destined to assemble the entire ORC protein 354 and promote pre-RC formation. SUV39H1 binding to ORC1 may influence its 355 temporally dynamic nuclear localization during G1 phase. Such a scenario may 356 also explain observations that have implicated a role for RB in DNA replication 357 (Bosco et al., 2001, Kennedy et al., 2000, Sterner et al., 1998). The ORC1-CDC6 358 switch might temporally influence pre-RC assembly just like it does for CCNE1 359 gene regulation, a possibility we are investigating. 360 361 Materials and Methods 362 363 Cell culture and cell synchronization, with siRNA treatment or transient 364 transfection of expression plasmids. 365 U2OS, HEK293 and MCF7 cells were obtained from the Cold Spring Harbor 366 Laboratory cell culture collection and cultured in DMEM containing high glucose 367 (Gibco) supplemented with 10% inactivated fetal calf serum and 368 Penicillin/Streptomycin. RB defective SaOS-2 cells were obtained from ATCC 369 (HTB-85) and were cultured in McCoy’s media supplemented with 15%

12 370 inactivated fetal calf serum and Penicillin/Streptomycin. All the cell lines were 371 tested negative for the mycoplasma contamination. To synchronize U2OS cells 372 at G2/M boundary, 100 ng/mL of nocodazole was added to fresh medium for 16 373 hours. After 16 hours of block the cells were washed two times with 1x 374 Phosphate Buffered Saline (PBS) and subsequently, released into the fresh 375 media. For depletion of ORC1 protein in synchronized U2OS cells, the U2OS 376 cells transiently transfected with 100nM of siRNAs (control GFP as well as ORC1 377 siRNA) using Lipofectamine RNAiMax and at the same time were treated with 378 nocodazole, then the cells were incubated and released as described. Plasmids 379 expressing exogenous genes were transfected using 2.5 μg of DNA except 380 where indicated using lipofectamine 2000 transfection reagents (ThermoFisher 381 Scientific). The sequences of the siRNAs used are listed in the Supplementary 382 table 1. 383 384 Plasmid Construction and Mutagenesis 385 Plasmids expressing GFP-RB, 10-4 CCNE1 (Addgene: Cyclin E gene) promoter, 386 E2F1, DP1, HDAC1 and pGL2-GAL4-UAS-Luc were purchased from Addgene. 387 pCMV-LacZ plasmid was purchased from Clontech. Plasmid Flag-SUV39H1 was 388 a gift from Peter Zhou (Dong et al., 2013). ORC1-Flag or its mutant derivatives, 389 ORC3-Flag, ORC4-Flag and T7-SUV39H1 plasmids were generated with a C- 390 terminal Flag tag or N-terminal T7 tag in the pLPC vector (McCurrach et al., 391 1997) (gift from Scott Lowe, Memorial Sloan Kettering Cancer Center) for 392 mammalian expression. Human ORC1 or its mutants, RB or its mutants, CDC6 393 or its mutants and SUV39H1 or its mutants were also cloned into pEGFP-C1 394 (Clontech, Mountain View, CA) for transient overexpression in mammalian cells. 395 GAL4-DBD, GAL4-ORC1 and GAL-SUV39H1 plasmids were made by replacing 396 the GFP insert in pEGFP-C1 constructs with in frame GAL4 DBD sequences. 397 MBP-GFP-ORC1 plasmid is made by cloning GFP-ORC1 insert (amplified from 398 GFP-ORC1 plasmid) in pMALc2E vector, while MBP-RB plasmid is made similar 399 to MBP-ORC1 as described previously (Hossain and Stillman, 2012) for protein 400 expression in bacterial cells. Human RB or its mutants, CDC6 or its mutant,

13 401 SUV39H1 or its mutant, HP1α and HDAC1 were also cloned in pGEX-6P1 vector 402 to produce GST fusion proteins in bacterial cells. The mutant plasmids were 403 generated following the site directed mutagenesis protocol (Quickchange, 404 Stratagene, CA). All of the plasmid constructs were verified by sequencing. The 405 oligonucleotide sequences used to generate the plasmids are listed in the 406 Supplementary table 1. 407 408 Recombinant protein expression and purification 409 Wild type ORC1, GFP-ORC1 and RB were fused to the maltose binding protein 410 (MBP) in frame by cloning in the bacterial expression vector pMALc2E (New 411 England Biolab). To generate GST fusion proteins, wild type RB or its mutants 412 (R661W and N757F), HDAC1, SUV39H1 or its fragments, HP1α or CDC6 were 413 cloned into the bacterial expression vector pGEX6P1 (GE Healthcare Life 414 Sciences, NJ). The MBP fusion recombinant proteins were expressed and 415 purified using amylose beads according to the procedure described previously 416 (Hossain and Stillman, 2012). For GST fusion proteins, transformed E. coli BL21 417 cells with their respective plasmids were induced for 12 hours with 0.3mM of 418 IPTG at 16°C. The induced cells were pelleted, washed, and further lysed with 419 sonication in a lysis buffer A containing 25 mM Tris-HCl at pH 7.5, 150 mM NaCl, 420 0.02% NP-40, 5 mM benzamidine-HCl, 1mM phenylmethylsulfonylfluoride, 421 Protease cocktail inhibitor tablets [Roche], 10% glycerol) plus 100 mg/ml 422 lysozyme. The lysed bacterial cells were centrifuged and the clarified supernatant 423 is incubated with pre-washed Glutathione sepharose beads for 3 hours at 4°C. 424 The bead bound proteins were washed with three column volumes of buffer A 425 plus 0.05% NP-40 + 500 mM NaCl and further with five column volumes of buffer 426 A alone. Fusion protein was eluted in a stepwise manner with buffer A containing 427 20 mM reduced glutathione, pH7.5. Fractions containing purified proteins were 428 pooled, concentrated, and dialyzed, and protein concentration was estimated 429 using a standard Bradford protein assay. 430 431 Pull down assays

14 432 For MBP-RB or MBP-ORC1 pull down assay, the same protocol as described 433 previously was used (Hossain and Stillman, 2012), except that here we have 434 used a different binding buffer with following composition; 25 mM Tris-Cl at pH 435 7.5, 100 mM KCl, 0.1% Nonidet P-40, 0.1 mM EDTA, 5 mM magnesium acetate, 436 1 mM DTT. For MBP-ORC1 pull down, Cyclin E-CDK2 protein was incubated 437 with wild type GST-RB or its mutants in the presence or absence of 1mM ATP. 438 The pull down was further immunoblotted with anti-GST antibody (27-4577-01; 439 GE Healthcare Life Sciences, NJ). 440 The histone peptide pull down assay followed the procedure described previously 441 (Hossain and Stillman, 2012) with the minor modification of binding buffer 442 composition (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 0.05% NP-40). The pull 443 down was silver stained or immunoblotted with anti-MBP antibody (E8038S; New 444 England Biolab). The biotin-labeled histone H3 peptides were purchased from 445 AnaSpec (Fremont, CA) and bound to Streptavidin beads (Sigma-Aldrich, St. 446 Louis, MO) prior to pull down studies. 447 For GST pull down assay, bead bound GST fusion proteins (RB, HDAC1, 448 SUV39H1 or its mutants, HP1α, and GST-CDC6 or its mutant) were incubated 449 with either MBP-ORC1 or MBP-RB or Cyclin E-CDK2 protein. The composition of 450 binding buffer used in the assay was; 25 mM Tris-Cl at pH 7.5, 150 mM KCl, 451 0.15% Nonidet P-40, 0.1 mM EDTA, 5 mM magnesium acetate, 1 mM DTT. The 452 pull down was further immunoblotted with respective anti-GST (27-4577-01; GE 453 Healthcare Life Sciences, NJ) or anti-Cyclin E antibodies (sc-247; Santa Cruz 454 Biotechnology, Dallas, TX). 455 For titration-based binding experiments, increasing molar amounts of GST-CDC6 456 and/or Cyclin E-CDK2 along together with equimolar amounts of MBP-RB 457 (20nM) and MBP-GFP-ORC1 (20nM) were used in different combinations. The 458 reaction mixture was incubated for 4 hours at 4°C followed by 1 hour more 459 incubation with bead coupled anti-GFP antibody (ABP-NAB-GFPA025; Allele 460 Biotech, San Diego, CA) to precipitate MBP-GFP-ORC1. The reaction mixture 461 was incubated and washed with binding buffer with following composition; 25 mM 462 Tris-Cl at pH 7.5, 150 mM KCl, 0.15% Nonidet P-40, 0.1 mM EDTA, 5 mM

15 463 magnesium acetate, 1 mM DTT and 1mM ATP. The pull down was 464 immunoblotted with the following antibodies; anti-ORC1 antibody (pKS1-40), anti- 465 RB antibody (#9309; Cell Signaling, Danvers, MA), anti-GST (27-4577-01; GE 466 Healthcare Life Sciences, NJ) and anti-Cyclin E antibody (sc-247; Santa Cruz 467 Biotechnology, Dallas, TX). 468 469 Immunoprecipitation 470 For expression of proteins, HEK293 cells were transiently transfected with the 471 indicated plasmids with lipofectamine 2000 transfection reagents (ThermoFisher 472 Scientific). Immunoprecipitation from HEK293 cells was performed using the 473 procedure described previously (Hemerly et al., 2009) with slight modification in 474 the protocol. Following expression of proteins, the cells were harvested and 475 washed in PBS and lysed in a buffer containing 20mM Tris-HCl pH7.5, 200mM

476 NaCl, 0.3% NP-40, 5mM MgCl2, 0.1 mM EDTA, 10% Glycerol, 1mM DTT, 1mM

477 CaCl2, 20uM MG132 and protease as well as phosphatase inhibitor tablets 478 (Roche). Benzonase (Sigma-Aldrich, St. Louis, MO) was added to the buffer and 479 the suspension incubated for 30 min on ice with intermittent mixing. The 480 concentration of NaCl and NP-40 was reduced to 100 mM and 0.15%, 481 respectively with dilution buffer after 30 minutes incubation on ice. The extract 482 was centrifuged at 14, 000 rpm for 15 min at 4°C. The proteins were precipitated 483 with specific antibodies as indicated in figure legends using FLAG, GFP, ORC2 484 or ORC3 antibodies. The whole cell extract was first incubated with antibodies for 485 4 hours and subsequently, 2 hours with pre-washed gamma bind G sepharose 486 beads with end-to end shaking at 4°C. The beads were washed 3 times with 487 washing buffer containing 20mM Tris-HCl pH7.5, 100mM NaCl, 0.1% NP-40,

488 5mM MgCl2, 0.1 mM EDTA, 10% Glycerol, 1mM DTT and protease as well as 489 phosphatase inhibitor tablets from Roche. Finally, the washed beads were 490 suspended in Laemmli sample buffer and 8% SDS-PAGE gels were run and 491 immunoblotted. For immunoprecipitation of endogenous ORC1 and SUV39H1 492 proteins from U2OS or MCF7 cells and SaOS-2 cells, we have used monoclonal 493 ORC1 antibody coupled to magnetic beads as described previously (Kara et al.,

16 494 2015) as well as rabbit polyclonal SUV39H1 antibody (A302-128A; Bethyl 495 Laboratories), respectively. 496 For immunoprecipitations, the following antibodies were used: rabbit polyclonal 497 anti-Flag antibody (F7425; Sigma), GFP nAb (ABP-NAB-GFPA025; Allele 498 Biotech, San Diego, CA), polyclonal anti-SUV39H1 antibody (A302-128A; Bethyl 499 Laboratories), mouse monoclonal ORC1 78-1-172 (Kara et al., 2015), rabbit 500 polyclonal anti-ORC2 antibody (CS205-5) (Prasanth et al., 2004), and rabbit 501 polyclonal anti-ORC3 (CS1890) antibody (Siddiqui and Stillman, 2007). For 502 Immunoblots, monoclonal FLAG antibody (F1804; Sigma), polyclonal GFP 503 antibody (G1544; Sigma), monoclonal mouse SUV39H1 antibody (05-615; 504 Millipore), monoclonal mouse T7 antibody (Cold Spring Harbor Laboratory 505 antibody facility), mouse monoclonal anti-ORC1 antibody (pKS1-40) (Hemerly et 506 al., 2009), monoclonal mouse ORC2 antibody (920-2-44) (Siddiqui and Stillman, 507 2007), mouse monoclonal ORC3 antibody (PKS1-16) (Prasanth et al., 2004), 508 goat polyclonal anti-ORC4 antibody (ab9641; Abcam, Cambridge, MA), 509 monoclonal mouse E2F-1 antibody (KH95; Santa Cruz Biotechnology, Dallas, 510 TX) and rabbit GAL4 antibody (sc-577; Santa Cruz Biotechnology, Dallas, TX) 511 were used. 512 513 RNA extraction and qRT-PCR 514 Nocodazole arrested U2OS cells were transfected with 100nM siRNA 515 (Dharmacon Inc., Lafayette, CO) and RNA was prepared at the indicated time 516 points post-drug release using the RNeasy Mini Kit (Qiagen cat. #74104) 517 including on-column DNase digestion (Qiagen cat. # 79254) and eluted in the 518 supplied RNase-free water. The cDNA used for Q-PCR was prepared from 1μg 519 each RNA sample using TaqMan Reverse Transcription Reagents (Applied 520 Biosystems #N808-0234) with random hexamer priming in a GeneAmp PCR 521 system 9700 thermocycler. Each Q-PCR reaction was prepared using 2μL of 1- 522 to-20 diluted cDNA and 13μL LightCycler 480 SYBR Green I Master Mix (Roche 523 #04887352001) and were performed in 384-well plates using the LightCycler 480 524 (Roche) as per manufacturer’s instructions.

17 525 For semi-quantitative RT-PCR, equal amounts of RNA (0.2 μg) were used for 526 RT-PCR using the Qiagen’s One-Step RT PCR kit following manufacturer’s 527 instructions. PCR was performed for 22 cycles and subsequently, run on 1.8% 528 agarose gel. The primer sequences used for RT-PCR analysis are listed in the 529 Supplementary table 1. 530 531 Luciferase reporter assay 532 For the luciferase reporter assay, 0.3x106 U2OS cells were seeded in 6-well 533 plates. Cells were transfected with 0.5 μg CCNE1 promoter luciferase plasmid 534 (p10-4), 50ng E2F1 plasmid, 50 ng of DP1 plasmid and 20 ng LacZ plasmid. 535 Together with above plasmids, the cells were also transfected either with ORC1- 536 Flag or its mutants, ORC3-Flag, ORC4-Flag, T7-SUV39H1, GFP-SUV39H1 or its 537 mutant, GFP-CDC6 or its mutant for overexpression at the indicated amounts in 538 micrograms. The cells were transfected using Lipofectamine 2000 (Life 539 Technologies) for 24 hours. For GAL4 based luciferase assay, the U2OS cells 540 were transfected with 0.5 μg of GAL4-UAS-Luciferase promoter plasmid and 20 541 ng of LacZ plasmid along with GAL4-DBD fused ORC1 or SUV39H1 at the 542 indicated amounts. In each of the experiments the amount of plasmid was kept 543 constant by the addition of empty vector DNA. For siRNA-mediated depletion of 544 proteins and luciferase measurement, 100nM of specific siRNAs targeting ORC1, 545 SUV39H1 or CDC6 were used together with p10-4, E2F1/DP1 and LacZ 546 plasmids. GFP siRNA is used as negative control. β-galactosidase activity was 547 measured as described previously (Smale, 2010), while luciferase activity was 548 measured using luciferase (Promega) luminescent assay kit according to the 549 manufacturer’s instructions. Luciferase activities were normalized to β- 550 galactosidase activities and denoted as Relative Light Units (RLU). Expression or 551 depletion of proteins was confirmed by Western blot. All the experiments were 552 done in triplicates. 553 554 Chromatin Immunoprecipitation 555 MCF7 and U2OS cells were used for chromatin immunoprecipitation. The cells

18 556 were harvested by trypsinization and washed with cold PBS. The cells (3.0x107) 557 were fixed with 1% formaldehyde for 10 minutes at room temperature. The cross- 558 linking was stopped by addition 125 mM of glycine for 5 minutes on ice. The fixed 559 cells were washed with cold PBS and lysed for 10 minutes on ice with buffer

560 containing 10 mM Tris pH 8.0, 10 mM NaCl, 2 mM MgCl2, 0.4% NP40, 1 mM 561 DTT, 10% Glycerol, protease and phosphatase inhibitors. Preliminary 562 experiments demonstrated that ORC1 was most efficiently extracted from the 563 resulting cells by digestion of the cross-linked chromatin with micrococcal 564 nuclease (MNase; Sigma) and extraction with high salt (300-600 mM NaCl). Thus 565 the nuclei were treated with MNase with 1 mM CaCl2 at 37°C (so that most of the 566 extracted DNA ran on an agarose gel as ~1-4 nucleosome length), the reaction 567 was quenched with 2 mM EGTA and washed with buffer containing 10 mM Tris

568 pH 8.0, 10 mM NaCl, 2 mM MgCl2, 1 mM DTT, 10% Glycerol, protease and 569 phosphatase inhibitors. The pelleted residual nuclei were further lysed in buffer 570 containing 20mM Tris pH 8.0, 1 mM EDTA, 0.5 mM EGTA, 300 mM NaCl, 0.5% 571 TritonX-100, 0.05% Sodium Deoxycholate, 0.1% IGE-PAL, 1 mM PMSF and 572 protease/phosphatase inhibitors for 30 min at 4°C with rotation. The lysate was 573 further diluted to bring the salt concentration to 200 mM and sonicated very 574 briefly. The lysate was centrifuged (20,800 X g in an Eppendorf centrifuge) at 4°C 575 to pellet down the debris and supernatant was used for pre-clearing with protein 576 G Dynabeads (Invitrogen) for 2 hours 4°C. The antibodies were bound to protein 577 G Dynabeads for 5 hours at 4°C in PBS containing 0.5% BSA. The antibodies 578 used for chromatin immunoprecipitation were as follows: mouse monoclonal 579 ORC1 [78-1-172; (Kara et al., 2015)], mouse RB (#9309, Cell Signaling), rabbit 580 polyclonal SUV39H1 (A302-128A; Bethyl Laboratories), rabbit polyclonal CDC6 581 (CS1881), rabbit polyclonal histone H3K9me3 (ab8898, Abcam) as well as 582 control mouse and rabbit IgG (Invitrogen). Pre-cleared nuclear extracts was 583 incubated with antibody bound beads overnight with rotation at 4°C. 0.4 ml of the 584 nuclear extract was used for each IP and 0.1 ml was kept aside for “Input” in Q- 585 PCR analysis. Beads were then washed with 3 times with low salt buffer (20mM 586 Tris pH8.0, 200mM NaCl, 0.5% TritonX-100, 0.05% Sodium Deoxycholate, 0.1%

19 587 IGE-PAL, 1 mM PMSF), 3 times high salt buffer (20mM Tris pH8.0, 500mM NaCl, 588 0.5% TritonX-100, 0.05% Sodium Deoxycholate, 0.1% IGE-PAL, 1 mM PMSF), 3 589 times with lithium chloride buffer (20mM Tris pH8.0, 200mM NaCl, 250mM LiCl, 590 0.5% TritonX-100, 0.05% Sodium Deoxycholate, 0.1% IGE-PAL, 1 mM PMSF) 591 and twice with TE buffer (10mM Tris pH8.0, 1 mM EDTA). Chromatin was eluted 592 from beads twice with 100 ul elution buffer (100 mM sodium bicarbonate, 1% 593 SDS) at room temperature. Protein-DNA crosslinks in the IP samples as well as 594 in input samples were reversed overnight by addition of 300 mM NaCl and 2 ug 595 RNase A at 65°C. The ChIP and input samples were then incubated with 60 ug 596 of proteinase K for 2 hours at 42°C. The samples were further extracted with 597 phenol:chloroform:isoamyl alchol [IAA] (once for ChIP and twice for input) and 598 once with chloroform extraction. The ethanol precipitation were done by adding 599 10 ug of glycogen, washed with 70% ethanol, air dried and then re-suspended in 600 50 ul of RNase-DNase free water. The purified DNA was used as template in 601 different PCR amplifications (Applied Biosystems Thermocycler). The sequences 602 of the various primers is listed in Supplementary Table 1. Due to very high GC 603 content of human CCNE1 promoter, the number of PCR cycles was extensively 604 and independently optimized with different DNA polymerases (Supplementary 605 Table 2) for each primer set to maintain linear amplification in all experiments. 606 PCR products were resolved by 2% agarose gel electrophoresis. Images of 607 ethidium bromide-stained DNAs were acquired using an UV trans-illuminator 608 equipped with a digital camera. Intensities of the amplified PCR bands were 609 quantitated by ImageJ software. 610 611 Data analysis. 612 Data are shown as the average ± the standard deviation (SD) of results of at 613 least three independent experiments. In the luciferase assay, the statistical 614 differences between cells overexpressing ORC1 and/or SUV39H1 or CDC6 or no 615 E2F1/DP1 were compared to E2F1/DP1 alone without overexpression, and 616 statistically evaluated by Student’s t test analysis. The lines above the bar graph 617 were used to indicate the statistical differences between overexpressing

20 618 samples. In the luciferase assay with depletion of specific proteins, the statistical 619 differences between siRNAs against the ORC1, SUV39H1 or CDC6 were 620 compared to control GFP siRNA using the Student’s t test. 621 622 Acknowledgements: We thank Patty Wendel for technical assistance and 623 Jaclyn Jansen for commenting on the manuscript. This research was supported 624 by a grant from the US National Cancer Institute (CA13106) and by a NCI 625 Cancer Center Grant (CA45508). 626 627 Author Contributions: M.H. and B.S. designed and analyzed the experiments 628 and wrote the paper. M.H. performed the experiments. 629 630 References 631 632 Abdurashidova, G, Danailov, MB, Ochem, A, Triolo, G, Djeliova, V, Radulescu, 633 S, Vindigni, A, Riva, S & Falaschi, A (2003) Localization of proteins bound 634 to a replication origin of human DNA along the cell cycle. EMBO J, 22, 635 4294-303. 10.1093/emboj/cdg404

636 Auth, T, Kunkel, E & Grummt, F (2006) Interaction between HP1alpha and 637 replication proteins in mammalian cells. Exp Cell Res, 312, 3349-59. 638 10.1016/j.yexcr.2006.07.014

639 Badugu, R, Shareef, MM & Kellum, R (2003) Novel Drosophila heterochromatin 640 protein 1 (HP1)/origin recognition complex-associated protein (HOAP) 641 repeat motif in HP1/HOAP interactions and chromocenter associations. J 642 Biol Chem, 278, 34491-8. 10.1074/jbc.M305262200

643 Bell, SP, Kobayashi, R & Stillman, B (1993) Yeast origin recognition complex 644 functions in transcription silencing and DNA replication. Science, 262, 645 1844-9. 10.1126/science.8266072

646 Bicknell, LS, Walker, S, Klingseisen, A, Stiff, T, Leitch, A, Kerzendorfer, C, 647 Martin, CA, Yeyati, P, Al Sanna, N, Bober, M, Johnson, D, Wise, C, 648 Jackson, AP, O'Driscoll, M & Jeggo, PA (2011a) Mutations in ORC1, 649 encoding the largest subunit of the origin recognition complex, cause 650 microcephalic primordial dwarfism resembling Meier-Gorlin syndrome. Nat 651 Genet, 43, 350-5. 10.1038/ng.776

21 652 Bicknell, LS, Bongers, EM, Leitch, A, Brown, S, Schoots, J, Harley, ME, Aftimos, 653 S, Al-Aama, JY, Bober, M, Brown, PA, van Bokhoven, H, Dean, J, Edrees, 654 AY, Feingold, M, Fryer, A, Hoefsloot, LH, Kau, N, Knoers, NV, Mackenzie, 655 J, Opitz, JM, Sarda, P, Ross, A, Temple, IK, Toutain, A, Wise, CA, Wright, 656 M & Jackson, AP (2011b) Mutations in the pre-replication complex cause 657 Meier-Gorlin syndrome. Nat Genet, 43, 356-9. 10.1038/ng.775

658 Bosco, G, Du, W & Orr-Weaver, TL (2001) DNA replication control through 659 interaction of E2F-RB and the origin recognition complex. Nature Cell 660 Biology, 3, 289-295. Doi 10.1038/35060086

661 Chen, TT & Wang, JY (2000) Establishment of irreversible growth arrest in 662 myogenic differentiation requires the RB LXCXE-binding function. Mol Cell 663 Biol, 20, 5571-80. 10.1128/MCB.20.15.5571-5580

664 Cook, JG, Park, CH, Burke, TW, Leone, G, DeGregori, J, Engel, A & Nevins, JR 665 (2002) Analysis of Cdc6 function in the assembly of mammalian 666 prereplication complexes. Proc Natl Acad Sci U S A, 99, 1347-52. 667 10.1073/pnas.032677499

668 Coverley, D, Laman, H & Laskey, RA (2002) Distinct roles for cyclins E and A 669 during DNA replication complex assembly and activation. Nature Cell 670 Biology, 4, 523-528. 10.1038/ncb813

671 de la Paz Sanchez, M & Gutierrez, C (2009) Arabidopsis ORC1 is a PHD- 672 containing H3K4me3 effector that regulates transcription. Proc Natl Acad 673 Sci U S A, 106, 2065-70. 10.1073/pnas.0811093106

674 de Munnik, SA, Bicknell, LS, Aftimos, S, Al-Aama, JY, van Bever, Y, Bober, MB, 675 Clayton-Smith, J, Edrees, AY, Feingold, M, Fryer, A, van Hagen, JM, 676 Hennekam, RC, Jansweijer, MC, Johnson, D, Kant, SG, Opitz, JM, 677 Ramadevi, AR, Reardon, W, Ross, A, Sarda, P, Schrander-Stumpel, CT, 678 Schoots, J, Temple, IK, Terhal, PA, Toutain, A, Wise, CA, Wright, M, 679 Skidmore, DL, Samuels, ME, Hoefsloot, LH, Knoers, NV, Brunner, HG, 680 Jackson, AP & Bongers, EM (2012) Meier-Gorlin syndrome genotype- 681 phenotype studies: 35 individuals with pre-replication complex gene 682 mutations and 10 without molecular diagnosis. Eur J Hum Genet, 20, 598- 683 606. 10.1038/ejhg.2011.269

684 DeGregori, J, Kowalik, T & Nevins, JR (1995) Cellular targets for activation by 685 the E2F1 transcription factor include DNA synthesis- and G1/S-regulatory 686 genes. Mol Cell Biol, 15, 4215-24.

687 Dellino, GI, Cittaro, D, Piccioni, R, Luzi, L, Banfi, S, Segalla, S, Cesaroni, M, 688 Mendoza-Maldonado, R, Giacca, M & Pelicci, PG (2013) Genome-wide 689 mapping of human DNA-replication origins: levels of transcription at ORC1 690 sites regulate origin selection and replication timing. Genome Res, 23, 1- 691 11. 10.1101/gr.142331.112

22 692 DePamphilis, ML (2005) Cell cycle dependent regulation of the origin recognition 693 complex. Cell Cycle, 4, 70-9.

694 Dick, FA & Rubin, SM (2013) Molecular mechanisms underlying RB protein 695 function. Nat Rev Mol Cell Biol, 14, 297-306. 10.1038/nrm3567

696 Dong, C, Wu, Y, Wang, Y, Wang, C, Kang, T, Rychahou, PG, Chi, YI, Evers, BM 697 & Zhou, BP (2013) Interaction with Suv39H1 is critical for Snail-mediated 698 E-cadherin repression in breast cancer. Oncogene, 32, 1351-62. 699 10.1038/onc.2012.169

700 Dyson, N (1998) The regulation of E2F by pRB-family proteins. Genes Dev, 12, 701 2245-62.

702 Ekholm-Reed, S, Mendez, J, Tedesco, D, Zetterberg, A, Stillman, B & Reed, SI 703 (2004) Deregulation of cyclin E in human cells interferes with 704 prereplication complex assembly. J Cell Biol, 165, 789-800. 705 10.1083/jcb.200404092

706 Ewen, ME, Sluss, HK, Sherr, CJ, Matsushime, H, Kato, J & Livingston, DM 707 (1993) Functional interactions of the with 708 mammalian D-type cyclins. Cell, 73, 487-97.

709 Furstenthal, L, Kaiser, BK, Swanson, C & Jackson, PK (2001) Cyclin E uses 710 Cdc6 as a chromatin-associated receptor required for DNA replication. J 711 Cell Biol, 152, 1267-78.

712 Geng, Y, Eaton, EN, Picon, M, Roberts, JM, Lundberg, AS, Gifford, A, Sardet, C 713 & Weinberg, RA (1996) Regulation of cyclin E transcription by E2Fs and 714 retinoblastoma protein. Oncogene, 12, 1173-80.

715 Giacinti, C & Giordano, A (2006) RB and cell cycle progression. Oncogene, 25, 716 5220-7. 10.1038/sj.onc.1209615

717 Hateboer, G, Wobst, A, Petersen, BO, Le Cam, L, Vigo, E, Sardet, C & Helin, K 718 (1998) Cell cycle-regulated expression of mammalian CDC6 is dependent 719 on E2F. Mol Cell Biol, 18, 6679-97. 10.1128/MCB.18.11.6679

720 Hemerly, AS, Prasanth, SG, Siddiqui, K & Stillman, B (2009) Orc1 controls 721 centriole and centrosome copy number in human cells. Science, 323, 789- 722 793. 10.1126/science.1166745

723 Hinds, PW, Mittnacht, S, Dulic, V, Arnold, A, Reed, SI & Weinberg, RA (1992) 724 Regulation of retinoblastoma protein functions by ectopic expression of 725 human cyclins. Cell, 70, 993-1006. 10.1016/0092-8674(92)90249-C

23 726 Hossain, M & Stillman, B (2012) Meier-Gorlin syndrome mutations disrupt an 727 Orc1 CDK inhibitory domain and cause centrosome reduplication. Genes 728 Dev, 26, 1797-810. 10.1101/gad.197178.112

729 Johnson, A & Skotheim, JM (2013) Start and the restriction point. Curr Opin Cell 730 Biol, 25, 717-23. 10.1016/j.ceb.2013.07.010

731 Jones, RM, Mortusewicz, O, Afzal, I, Lorvellec, M, Garcia, P, Helleday, T & 732 Petermann, E (2013) Increased replication initiation and conflicts with 733 transcription underlie Cyclin E-induced replication stress. Oncogene, 32, 734 3744-53. 10.1038/onc.2012.387

735 Kara, N, Hossain, M, Prasanth, SG & Stillman, B (2015) Orc1 Binding to Mitotic 736 Chromosomes Precedes Spatial Patterning during G1 Phase and 737 Assembly of the Origin Recognition Complex in Human Cells. J Biol 738 Chem, 290, 12355-69. 10.1074/jbc.M114.625012

739 Kennedy, BK, Barbie, DA, Classon, M, Dyson, N & Harlow, E (2000) Nuclear 740 organization of DNA replication in primary mammalian cells. Genes & 741 Development, 14, 2855-2868. Doi 10.1101/Gad.842600

742 Kreitz, S, Ritzi, M, Baack, M & Knippers, R (2001) The human origin recognition 743 complex protein 1 dissociates from chromatin during S phase in HeLa 744 cells. J Biol Chem, 276, 6337-42. 10.1074/jbc.M009473200

745 Kuo, AJ, Song, J, Cheung, P, Ishibe-Murakami, S, Yamazoe, S, Chen, JK, Patel, 746 DJ & Gozani, O (2012) The BAH domain of ORC1 links H4K20me2 to 747 DNA replication licensing and Meier-Gorlin syndrome. Nature, 484, 115-9. 748 10.1038/nature10956

749 Li, J, Lin, Q, Yoon, HG, Huang, ZQ, Strahl, BD, Allis, CD & Wong, J (2002) 750 Involvement of histone methylation and phosphorylation in regulation of 751 transcription by thyroid hormone receptor. Mol Cell Biol, 22, 5688-97. 752 10.1128/MCB.22.16.5688-5697

753 Lidonnici, MR, Rossi, R, Paixao, S, Mendoza-Maldonado, R, Paolinelli, R, 754 Arcangeli, C, Giacca, M, Biamonti, G & Montecucco, A (2004) Subnuclear 755 distribution of the largest subunit of the human origin recognition complex 756 during the cell cycle. J Cell Sci, 117, 5221-31. 10.1242/jcs.01405

757 Lundberg, AS & Weinberg, RA (1998) Functional inactivation of the 758 retinoblastoma protein requires sequential modification by at least two 759 distinct cyclin-cdk complexes. Mol Cell Biol, 18, 753-61. 760 10.1128/MCB.18.2.753

761 Mailand, N & Diffley, JF (2005) CDKs promote DNA replication origin licensing in 762 human cells by protecting Cdc6 from APC/C-dependent proteolysis. Cell, 763 122, 915-26. 10.1016/j.cell.2005.08.013

24 764 McCurrach, ME, Connor, TM, Knudson, CM, Korsmeyer, SJ & Lowe, SW (1997) 765 bax-deficiency promotes drug resistance and oncogenic transformation by 766 attenuating p53-dependent apoptosis. Proc Natl Acad Sci U S A, 94, 767 2345-9.

768 Mendez, J & Stillman, B (2000) Chromatin association of human origin 769 recognition complex, cdc6, and minichromosome maintenance proteins 770 during the cell cycle: assembly of prereplication complexes in late mitosis. 771 Mol Cell Biol, 20, 8602-12.

772 Mendez, J, Zou-Yang, XH, Kim, SY, Hidaka, M, Tansey, WP & Stillman, B (2002) 773 Human origin recognition complex large subunit is degraded by ubiquitin- 774 mediated proteolysis after initiation of DNA replication. Mol Cell, 9, 481-91.

775 Mendoza-Maldonado, R, Paolinelli, R, Galbiati, L, Giadrossi, S & Giacca, M 776 (2010) Interaction of the retinoblastoma protein with Orc1 and its 777 recruitment to human origins of DNA replication. PLoS One, 5, e13720. 778 10.1371/journal.pone.0013720

779 Narasimha, AM, Kaulich, M, Shapiro, GS, Choi, YJ, Sicinski, P & Dowdy, SF 780 (2014) Cyclin D activates the Rb tumor suppressor by mono- 781 phosphorylation. Elife, 310.7554/eLife.02872

782 Nicolas, E, Roumillac, C & Trouche, D (2003) Balance between acetylation and 783 methylation of histone H3 lysine 9 on the E2F-responsive dihydrofolate 784 reductase promoter. Molecular and Cellular Biology, 23, 1614-1622. 785 10.1128/MCB.23.5.1614-1622.2003

786 Nielsen, SJ, Schneider, R, Bauer, UM, Bannister, AJ, Morrison, A, O'Carroll, D, 787 Firestein, R, Cleary, M, Jenuwein, T, Herrera, RE & Kouzarides, T (2001) 788 Rb targets histone H3 methylation and HP1 to promoters. Nature, 412, 789 561-5. 10.1038/35087620

790 Odajima, J, Kalaszczynska, I & Sicinski, P (2010) Cyclins A and E trigger DNA 791 damage. Cell Cycle, 9, 1231-2. 10.1016/S0092-8674(00)80415-8

792 Ohta, S, Tatsumi, Y, Fujita, M, Tsurimoto, T & Obuse, C (2003) The ORC1 cycle 793 in human cells: II. Dynamic changes in the human ORC complex during 794 the cell cycle. J Biol Chem, 278, 41535-40. 10.1074/jbc.M307535200

795 Ohtani, K, Tsujimoto, A, Ikeda, M & Nakamura, M (1998) Regulation of cell 796 growth-dependent expression of mammalian CDC6 gene by the cell cycle 797 transcription factor E2F. Oncogene, 17, 1777-85. 10.1038/sj.onc.1202105

798 Okuno, Y, McNairn, AJ, den Elzen, N, Pines, J & Gilbert, DM (2001) Stability, 799 chromatin association and functional activity of mammalian pre-replication 800 complex proteins during the cell cycle. EMBO J, 20, 4263-77. 801 10.1093/emboj/20.15.4263

25 802 Pak, DT, Pflumm, M, Chesnokov, I, Huang, DW, Kellum, R, Marr, J, 803 Romanowski, P & Botchan, MR (1997) Association of the origin 804 recognition complex with heterochromatin and HP1 in higher eukaryotes. 805 Cell, 91, 311-23. 10.1016/S0092-8674(00)80415-8

806 Petersen, BO, Wagener, C, Marinoni, F, Kramer, ER, Melixetian, M, Lazzerini 807 Denchi, E, Gieffers, C, Matteucci, C, Peters, JM & Helin, K (2000) Cell 808 cycle- and cell growth-regulated proteolysis of mammalian CDC6 is 809 dependent on APC-CDH1. Genes Dev, 14, 2330-43. 10.1101/gad.832500

810 Prasanth, SG, Mendez, J, Prasanth, KV & Stillman, B (2004) Dynamics of pre- 811 replication complex proteins during the cell division cycle. Philosophical 812 Transactions of the Royal Society of London Series B-Biological Sciences, 813 359, 7-16. 10.1098/rstb.2003.1360

814 Prasanth, SG, Shen, Z, Prasanth, KV & Stillman, B (2010) Human origin 815 recognition complex is essential for HP1 binding to chromatin and 816 heterochromatin organization. Proc Natl Acad Sci U S A, 107, 15093-8. 817 10.1073/pnas.1009945107

818 Rea, S, Eisenhaber, F, O'Carroll, D, Strahl, BD, Sun, ZW, Schmid, M, Opravil, S, 819 Mechtler, K, Ponting, CP, Allis, CD & Jenuwein, T (2000) Regulation of 820 chromatin structure by site-specific histone H3 methyltransferases. 821 Nature, 406, 593-9. 10.1038/35020506

822 Resnitzky, D, Gossen, M, Bujard, H & Reed, SI (1994) Acceleration of the G1/S 823 phase transition by expression of cyclins D1 and E with an inducible 824 system. Mol Cell Biol, 14, 1669-79. doi: 10.1128/MCB.14.3.1669

825 Rubin, SM (2013) Deciphering the retinoblastoma protein phosphorylation code. 826 Trends Biochem Sci, 38, 12-9. 10.1016/j.tibs.2012.10.007

827 Saha, P, Chen, J, Thome, KC, Lawlis, SJ, Hou, ZH, Hendricks, M, Parvin, JD & 828 Dutta, A (1998) Human CDC6/Cdc18 associates with Orc1 and cyclin-cdk 829 and is selectively eliminated from the nucleus at the onset of S phase. Mol 830 Cell Biol, 18, 2758-67. doi: 10.1128/MCB.18.5.2758

831 Sasaki, T & Gilbert, DM (2007) The many faces of the origin recognition complex. 832 Curr Opin Cell Biol, 19, 337-43. 10.1016/j.ceb.2007.04.007

833 Siddiqui, K & Stillman, B (2007) ATP-dependent assembly of the human origin 834 recognition complex. J Biol Chem, 282, 32370-83. 835 10.1074/jbc.M705905200

836 Smale, ST (2010) Beta-galactosidase assay. Cold Spring Harb Protoc, 2010, pdb 837 prot5423. 10.1101/pdb.prot5423

26 838 Sterner, JM, Dew-Knight, S, Musahl, C, Kornbluth, S & Horowitz, JM (1998) 839 Negative regulation of DNA replication by the retinoblastoma protein is 840 mediated by its association with MCM7. Molecular and Cellular Biology, 841 18, 2748-2757. doi: 10.1128/MCB.18.5.2748

842 Stewart, MD, Li, JW & Wong, JM (2005) Relationship between histone H3 lysine 843 9 methylation, transcription repression, and heterochromatin protein 1 844 recruitment. Molecular and Cellular Biology, 25, 2525-2538. 845 10.1128/MCB.25.7.2525-2538.2005

846 Tatsumi, Y, Tsurimoto, T, Shirahige, K, Yoshikawa, H & Obuse, C (2000) 847 Association of human origin recognition complex 1 with chromatin DNA 848 and nuclease-resistant nuclear structures. J Biol Chem, 275, 5904-10. doi: 849 10.1074/jbc.275.8.5904

850 Triolo, T & Sternglanz, R (1996) Role of interactions between the origin 851 recognition complex and SIR1 in transcriptional silencing. Nature, 381, 852 251-3. 10.1038/381251a0

853 Vandel, L, Nicolas, E, Vaute, O, Ferreira, R, Ait-Si-Ali, S & Trouche, D (2001) 854 Transcriptional repression by the retinoblastoma protein through the 855 recruitment of a histone methyltransferase. Mol Cell Biol, 21, 6484-94. doi: 856 10.1128/MCB.21.19.6484-6494.2001

857 Vaute, O, Nicolas, E, Vandel, L & Trouche, D (2002) Functional and physical 858 interaction between the histone methyl transferase Suv39H1 and histone 859 deacetylases. Nucleic Acids Research, 30, 475-481. Doi 860 10.1093/Nar/30.2.475

861 Walter, D, Hoffmann, S, Komseli, ES, Rappsilber, J, Gorgoulis, V & Sorensen, 862 CS (2016) SCF(Cyclin F)-dependent degradation of CDC6 suppresses 863 DNA re-replication. Nat Commun, 7, 10530. 10.1038/ncomms10530

864 Yan, Z, DeGregori, J, Shohet, R, Leone, G, Stillman, B, Nevins, JR & Williams, 865 RS (1998) Cdc6 is regulated by E2F and is essential for DNA replication in 866 mammalian cells. Proc Natl Acad Sci U S A, 95, 3603-8.

867 Zhang, W, Sankaran, S, Gozani, O & Song, J (2015) A Meier-Gorlin syndrome 868 mutation impairs the ORC1-nucleosome association. ACS Chem Biol, 10, 869 1176-80. 10.1021/cb5009684 870 871

27 872 Figure legends 873 Figure 1. ORC1 represses Cyclin E gene expression and interacts with RB. 874 (A-B), Nocodazole arrested U2OS cells were transfected with control or ORC1 875 siRNA then released into the next cycle. (A), protein levels were estimated by 876 immunoblotting with antibodies against ORC1, Cyclin E, Cyclin A and α-Tubulin. 877 Low and high indicates different exposures. (B), mRNA levels of ORC1, Cyclin E 878 (CCNE1), Cyclin A (CCNA2) and GAPDH. Quantitation of mRNA levels from 879 multiple experiments is shown in Figure 1-figure supplement 1. (C-E), Interaction 880 between ORC1 and RB or its pocket mutants. GFP, GFP-tagged wild-type or 881 mutant RB were co-transfected in HEK293 cells with either ORC1-Flag or empty 882 vector. Immunoprecipitation with anti-Flag antibody (C) or GFP antibody (D) from 883 cell lysates followed by immunoblotting with the indicated antibodies. (E), Cell 884 lysate from HEK293 cells overexpressing GFP-tagged wild type or mutant RB 885 and ORC1-Flag were immunoprecipitated with normal rabbit serum (NRS) or 886 ORC2 or ORC3 antibodies, immunoblotted with the indicated antibodies. Binding 887 of ORC1 to wild type and RB mutants and the effect of Cyclin E-CDK2 is shown 888 in Figure 1-figure supplement 2. 889 890 Figure 1-figure supplemental 1. ORC1 represses Cyclin E gene expression 891 and interacts with RB. CCNE1 (Cyclin E) gene transcription is up-regulated 892 upon ORC1 depletion. (A-D), Quantitative PCR analysis of ORC1, CCNE1 893 (Cyclin E), CCNA2 (Cyclin A) and GAPDH transcript levels in U2OS cells 894 transfected with either control siRNA targeting GFP (blue bar) or ORC1 siRNA 895 (grey bar). The siRNA treated cells were released after nocodazole arrest and 896 mRNA levels estimated at different time points as indicated in hours. The values 897 shown are average fold change (mean±SEM) from three independent experiment 898 normalized to β-actin transcripts. Statistical analysis was performed using the 899 Student’s t test. *p < 0.01; **p < 0.001; ***p < 0.0001; NS, not significant. 900 901 Figure 1-figure supplemental 2. ORC1 represses Cyclin E gene expression 902 and interacts with RB. Cyclin E-CDK2 phosphorylation controls ORC1 and

28 903 RB interaction. (A), Interaction between purified RB and ORC1 in a MBP pull 904 down assay. MBP-fused wild-type RB was bound to amylose resin and further 905 incubated with in vitro translated, S35-labeled wild type ORC1 or its mutants in 906 the presence or absence of Cyclin E-CDK2 and 1mM ATP. Beads were isolated 907 and bound proteins were separated by gel electrophoresis. MBP was used as a 908 control in the assay. (B), Alignment of ORC1 sequences is shown with conserved 909 LxCxE motif. The conserved residues of LxCxE motif are indicated with different 910 colors. The alignment shows conserved residues in ORC1 from different species 911 in vertebrate and invertebrate classes (Invertebrates: Brugia malayi, 912 Caenorhabditis briggsae, Caenorhabditis elegans, Strongylocentrotus 913 purpuratus, Culex quinquefasciatus, Apis mellifera, Drosophila melanogaster, 914 Aedes aegypti, and Pediculus humanus; Vertebrates: Danio rerio, Xenopus 915 laevis, Xenopus tropicalis, Gallus gallus, Taeniopygia guttata, Mus musculus and 916 Homo sapiens). (C), Interaction between ORC1 and RB in a MBP pull-down 917 assay. GST-fused wild-type RB or its mutant proteins were incubated with wild 918 type MBP-ORC1 in the presence or absence of Cyclin E-CDK2 and/or 1mM ATP. 919 Amylose bead bound proteins were isolated and bound proteins were separated 920 by gel electrophoresis followed by immunoblotting with anti-GST antibody. 921 Recombinant MBP was used as a control in the assay. 922 923 924 Figure 2. ORC1 binds SUV39H1 to control Cyclin E gene transcription. (A), 925 Purified MBP-ORC1 and various GST-fused proteins were mixed and proteins 926 bound in a GST-pull down were detected by immunoblotting with anti-MBP 927 antibodies. The purified proteins are shown in Figure 2-figure supplement 1. (B), 928 U2OS and MCF7 cell lysates were immunoprecipitated with SUV39H1 antibody 929 and immunoblotted with the indicated antibodies. Rabbit IgG served as control 930 antibody. Asterisk indicates the cross-reacting antibody band; arrow indicates the 931 SUV39H1 protein. (C), Immunoprecipitation from RB negative SaOS-2 cell 932 lysates with ORC1 antibody or IgG and immunoblotted with antibodies against 933 ORC1, SUV39H1 or ORC3. (D), HEK293 cells were transiently co-transfected

29 934 with GFP, GFP-ORC1 or GFP-RB plus T7-SUV39H1 plasmids (2.5 μg each). 935 GFP antibody immunoprecipitates were immunoblotted with the indicated 936 antibodies. The interaction between ORC1 and SUV39H1 and between RB and 937 SUV39H1 is shown with purified proteins and quantitated in Figure 2-figure 938 supplement 2. Higher levels of RB are required to demonstrate an interaction 939 with SUV39H1 in vivo and ORC1 interacts with the SET domain of SUV39H1, 940 Figure 2-figure supplement. (E), MBP-ORC1 was incubated with bead-bound 941 histone peptides with or without the indicated modifications and bound MBP- 942 ORC1 was observed by immunoblotting with anti-MBP antibody (lower box) or 943 silver staining (upper box). (F-G), Wild-type CCNE1-luciferase reporter assay in 944 U2OS cells. U2OS cells were transiently co-transfected with 500ng of 10-4 945 CCNE1 promoter, 50ng E2F1, 50 ng DP1 and 20 ng pCMV-LacZ plasmids along 946 with the indicated amounts ORC1 and/or SUV39H1 plasmids. (F), Increasing 947 amounts of ORC1-Flag or T7-SUV39H1 repress Cyclin E gene promoter. 948 Experiments were carried out in triplicate. Expression of proteins was confirmed 949 by Immunoblot; α-Tubulin as loading control. Statistical analysis was performed 950 using the Student’s t test. *p < 0.05; **p < 0.005; ***p < 0.001. (G), ORC1-Flag 951 cooperates with wild type but not mutant SUV39H1 to repress CCNE1 gene 952 expression. The Experiments were carried out in triplicate. Expression of proteins 953 was confirmed by western blots. α-Tubulin served as a control for equal loading 954 of each sample. Statistical analysis was performed using the Student’s t test. *p < 955 0.01; **p < 0.005; ***p < 0.0001. Repression of transcription by ORC1 and 956 SUV39H1 was also demonstrated using an artificial promoter and tethering the 957 proteins via the GAL4 DNA binding domain in Figure 2-figure supplement 4. 958 959 Figure 2-figure supplemental 1. ORC1 binds SUV39H1 to control Cyclin E 960 gene transcription. Bacteria expressed and purified recombinant proteins. 961 Silver stain of purified GST, GST-RB, GST-HDAC1, GST-SUV39H1, GST-HP1α, 962 GST-CDC6 and MBP-ORC1 proteins. MW stands for protein molecular weight 963 marker in kilodalton. 964

30 965 Figure 2-figure supplemental 2. ORC1 binds SUV39H1 to control Cyclin E 966 gene transcription. SUV39H1 interaction with ORC1 and RB. (A), Interaction 967 between purified MBP-ORC1 or MBP-RB with GST-SUV39H1 in a MBP pull 968 down assay. MBP-fused proteins were bound to amylose resin and incubated 969 with GST-SUV39H1. Bound proteins were separated by gel electrophoresis 970 followed by immunoblotting with either anti-GST or anti-SUV39H1 antibodies. 971 Recombinant MBP was used as a control. (B), (C), GST-SUV39H1 protein was 972 bound to the resin and incubated with either MBP-ORC1 or MBP-RB proteins. 973 Bead bound proteins were immunoblotted with anti-MBP antibody. Recombinant 974 GST was used as a control. (D), Concentration dependent interaction with 975 increasing levels of GST-SUV39H1 and either 100 nM of MBP or MBP-ORC1 or 976 MBP-RB followed by pull down with amylose beads and subsequently 977 immunoblotted with anti-GST antibody. Bands were quantified and represented 978 in a graph after background subtraction with MBP control protein. 979 980 Figure 2 figure supplement 3. ORC1 binds SUV39H1 to control Cyclin E 981 gene transcription. ORC1 interaction with SET domain of SUV39H1 does 982 not involve other ORC subunits. (A), HEK293 cells were transiently co- 983 transfected with GFP, GFP-ORC1 or GFP-RB and T7-SUV39H1 expressing 984 plasmids at the indicated amounts in micrograms. The whole cell lysate prepared 985 from HEK293 cells expressing the indicated constructs were immunoprecipitated 986 with GFP antibody followed by immunoblotting with specific antibodies. (B), GFP- 987 tagged wild-type ORC1 or its mutants (A-A: [ORC235ARA237 “Cy” motif mutant] or 988 ORC1S258A,S273A,T375A [CDK: with mutants in CDK target sites]) were co- 989 transfected in HEK293 cells with either Flag-SUV39H1 or its empty vector. 990 Immunoprecipitation with anti-Flag antibody from cell lysates of HEK293 cells 991 overexpressing the indicated constructs followed by immunoblotting with the 992 indicated antibodies. (C), Schematic showing domains of human SUV39H1 993 protein. In the GST-pull down assay, GST-SUV39H1 or its truncation mutant 994 proteins were incubated with MBP-ORC1 protein as indicated and immunoblotted 995 with anti-MBP antibody. GST protein served as negative control.

31 996 997 Figure 2 figure supplement 4. ORC1 binds SUV39H1 to control Cyclin E 998 gene transcription. ORC1, but not ORC3 or ORC4 can repress gene 999 transcription. (A), The U2OS cells were transfected with a Gal4-driven 1000 luciferase reporter as shown in the schematic with increasing amounts of 1001 Gal4DBD-ORC1 or Gal4DBD-SUV39H1 together with pCMV-LacZ plasmids. 1002 Relative luciferase activity was determined and normalized to lacZ activity. 1003 Experiments were carried out in triplicate. The whole cell extract was 1004 immunoblotted with anti-Gal4 antibody for expression of Gal4DBD fusion 1005 plasmids. α-Tubulin served as a loading control. Statistical analysis was 1006 performed using the Student’s t test. **p < 0.01; ***p < 0.005. (B), Wild-type 1007 CCNE1 promoter-luciferase reporter assay in U2OS cells. The U2OS cells were 1008 transiently co-transfected with 500ng of the 10-4 CCNE1 promoter, 50ng E2F1, 1009 50 ng DP1 and 20ng pCMV-LacZ plasmids along with the indicated amounts 1010 ORC3-Flag or ORC4-Flag plasmids. The increasing amounts of ORC3-Flag or 1011 ORC4-Flag do not repress the CCNE1 gene promoter as indicated by relative 1012 light units (RLU) normalized to β-galactosidase activity. Experiments were carried 1013 out in triplicate. Expression of proteins was confirmed by Western blot. α-Tubulin 1014 served as a loading control. 1015 1016 Figure 3. Binding of ORC1, RB, SUV39H1 and CDC6 proteins to the CCNE1 1017 promoter. (A), Schematic of the CCNE1 promoter and the regions amplified with 1018 different primer pairs used for ChIP assay were indicated as follows: A (−1462 to 1019 −1318); B (−683 to −575); C (−456 to −357); D (−342 to −195); E (−280 to −143); 1020 F (−159 to +63); G (+350 to +490); H (+816 to +944). The red bars indicate five 1021 E2F1 consensus sites. The truncated box indicates the first exon of the CCNE1 1022 gene. (B-C), The occupancy of ORC1, RB, SUV39H1 and CDC6 proteins was 1023 analyzed by chromatin immunoprecipitation at the CCNE1 promoter in 1024 asynchronous growing MCF7 cells. ORC1 and RB are mouse antibodies, while 1025 SUV39H1 and CDC6 are rabbit antibodies. In the marker lanes, the two bands

32 1026 are 100 and 200 base pairs. The experiments were done in triplicate and two of 1027 these experiments are shown. 1028 1029 Figure 4. Dynamic association of ORC1, RB, SUV39H1 and CDC6 proteins 1030 to the CCNE1 promoter during the cell cycle. (A-B), Nocodazole arrested 1031 U2OS cells were released for different times (3 hour, 6 hour and 9 hour) and 1032 analyzed for occupancy of ORC1, RB, SUV39H1 and CDC6 proteins at the 1033 CCNE1 promoter by ChIP assay. The primer pairs used to analyze two different 1034 regions of the CCNE1 promoter are indicated. The experiments were done in 1035 triplicate with results similar to those shown. (C), Whole cell protein levels of 1036 nocodazole arrested and released U2OS cells at different time points (as 1037 indicated in hours) by immunoblotting with antibodies against ORC1, RB, 1038 SUV39H1, CDC6, Cyclin E, ORC3 and α-Tubulin. 1039 1040 Figure 5. CDC6 co-operates with Cyclin E-CDK2 to activate E2F1-dependent 1041 CCNE1 gene transcription. (A-C), Equimolar amounts of MBP-GFP-ORC1 and 1042 MBP-RB proteins were incubated with increasing amounts of GST-CDC6 and/or 1043 Cyclin E-CDK2. MBP-GFP-ORC1 protein was immunoprecipitated with GFP 1044 antibody, then immunoblotted with the indicated antibodies. The purified proteins 1045 used in these experiments are shown in Figure 5-figure supplement 1 (A), MBP- 1046 GFP-ORC1 binds GST-CDC6. MBP protein served as control. (B), The binding 1047 of MBP-GFP-ORC1 protein to either GST-CCD6 in the presence of Cyclin E- 1048 CDK2 (left section) or MBP-RB (right section). (C), The binding of MBP-GFP- 1049 ORC1 to MBP-RB in the presence of increasing molar amounts of Cyclin E- 1050 CDK2 (right section) or both GST-CDC6 and Cyclin E-CDK2 (left section). (D), 1051 GST-pull down assay using GST-CDC6 wild type or CDC694ARA96 mutant 1052 (CDC6A-A) with purified Cyclin E-CDK2 protein followed by Immunoblotting with 1053 Cyclin E antibody. GST protein served as control. (E), Nocodazole arrested 1054 U2OS cells were transfected with 500ng of GFP, GFP-CDC6 wild type or 1055 CDC694ARA96 mutant (CDC6A-A) plasmids, then released into the next cell 1056 cycle. At indicated times, whole cell extracts were immunoblotted with specific

33 1057 antibodies against GFP and Cyclin E. α-Tubulin served as loading control. (F), 1058 CCNE1 promoter-luciferase reporter assay in U2OS cells. Cells transiently co- 1059 transfected with 500ng of 10-4 CCNE1 promoter, 50 ng E2F1, 50 ng DP1 and 20 1060 ng pCMV-LacZ plasmids together with increasing amounts GFP-CDC6 WT or 1061 CDC694ARA96 plasmids for 24 hours. Relative luciferase activity was normalized 1062 to co-transfected LacZ control. Experiments in triplicate. Protein expression 1063 determined by immunoblot; α-Tubulin as loading control. Statistical analysis was 1064 performed using the Student’s t test. *p < 0.05; **p < 0.001; ***p < 0.0005. 1065 1066 Figure 5 figure supplement 1. CDC6 co-operates with Cyclin E-CDK2 to 1067 activate E2F1-dependent CCNE1 gene transcription. Purified Proteins. 1068 Coomassie Brilliant Blue stained gel of purified MBP, MBP-GFP-ORC1, MBP-RB 1069 and GST-CDC6 proteins. MW stands for protein molecular weight marker in 1070 kilodalton. 1071 1072 Figure 6. ORC1 depletion decreases association of SUV39H1 and H3K9me3 1073 with the CCNE1 promoter and increases CCNE1 gene transcription. (A), 1074 CCNE1-luciferase reporter assay in U2OS cells. U2OS cells transiently co- 1075 transfected with 500 ng of 10-4 CCNE1 promoter, 50 ng E2F1, 50 ng DP1 and 1076 20 ng pCMV-LacZ. U2OS cells were also transiently transfected for 24 hours with 1077 two different siRNAs targeting either ORC1, SUV39H1 or CDC6. GFP siRNA 1078 was used as a control. Relative luciferase activity was normalized to co- 1079 transfected LacZ control. Depletion of proteins was confirmed by Immunoblot; α- 1080 Tubulin as loading control. Statistical analysis was performed using the Student’s 1081 t test. *p < 0.005; **p < 0.001; ***p < 0.0005. (B), The CCNE1 promoter was 1082 analyzed for SUV39H1 binding and the presence of the H3K9me3 mark by ChIP 1083 assay in U2OS cells treated with either control siRNA or ORC1 siRNA for 48 1084 hours. The experiments were done in triplicate and one experiment is shown. 1085 (C), Immunoblot of protein levels following control and ORC1 siRNA treatment at 1086 different times post nocodazole release. α-Tubulin was used as loading control. 1087

34 1088 Figure 6 figure supplement 1. The ChIP qPCR bands were quantified using 1089 ImageJ software to analyze the extent of binding of SUV39H1 and histone 1090 H3K9me3 to the CCNE1 promoter region (-280 to -143 bp) in ORC1 siRNA 1091 treated U2OS cells compared to control siRNA treated cells. All values were 1092 normalized to the corresponding input DNA. Statistical analysis was performed 1093 using the Student’s t test. The p-values are indicated above the bars. The 1094 reduction in SUV39H1 was not significant due to one out of the three replicates 1095 showing little reduction in ORC1 siRNA treated U2OS cells compared to the 1096 control. Binding of SUV39H1 to the CCNE1 promoter was also low compared to 1097 histone H3K9me3 in asynchronous U2OS cells treated with siRNAs. 1098 1099 Figure 7. ORC1 mutants separate it role as a transcription co-repressor 1100 from its role in DNA replication. (A), HEK293 cells were transfected with 1101 ORC1-Flag or its truncation mutants and GFP-RB as indicated. Whole cell 1102 extracts were immunoprecipitated with anti-Flag antibody and 1103 immunoprecipitates were analyzed by immuoblot with the indicated antibodies. 1104 (B and C), In vivo interaction between SUV39H1 and ORC1 or its truncation 1105 mutants. GFP-tagged wild-type ORC1 (1-861) or its truncation mutants (1-700 1106 and 1-768) plasmids were co-transfected into HEK293 cells with Flag-SUV39H1 1107 plasmid and either GFP-vector or empty vector as a control plasmids. 1108 Immunoprecipitation with anti-GFP antibody (B) or anti-Flag antibody (C) from 1109 cell lysates of HEK293 cells expressing the indicated constructs, followed by 1110 immunoblotting with the indicated antibodies. (D). U2OS cells transiently 1111 transfected for 24 hours with increasing amounts of wild type ORC1-Flag (1-861) 1112 or truncation mutants (1-700 and 1-768). Relative luciferase activity normalized 1113 to co-transfected lacZ control. Experiments were in triplicate. Expression of 1114 proteins determined by Immunoblot; α-Tubulin as loading control. Statistical 1115 analysis was performed using the Student’s t test. *p < 0.005; **p < 0.001; ***p < 1116 0.0005. 1117 1118 Figure 8. Model showing contrasting roles of DNA replication proteins ORC1 and

35 1119 CDC6 in the regulation of CCNE1 transcription and commitment to pre-RC 1120 assembly and cell division. Top panel summarizes the cycle levels of ORC1 and 1121 CDC6. The bottom panel shown the ORC1 associated complexes at different 1122 stages of the cell division cycle. Blue arrow, positive feedback inhibition of 1123 ORC1-RB interaction. 1124

36 Figure 1 D A Control siRNA ORC1 siRNA 0 h 3 h 6 h 9 h 12 h 24 h 0 h 3 h 6 h 9 h 12 h 24 h GFP GFP WT GFP-RB R661W GFP-RB N757F GFP-RB GFP WT GFP-RB R661W GFP-RB N757F GFP-RB ORC1-Flag α-ORC1 + + + + + + + + α-GFP α-Cyclin E (Low) α-Flag

α-Cyclin E (High) α-ORC3 α-ORC4 α-Cyclin A α-E2F1 α-Tubulin Input (2.5%) α-GFP IP (25%)

E NRS α-ORC2 α-ORC3 B Control siRNA ORC1 siRNA 0 h 3 h 6 h 9 h 12 h 24 h 0 h 3 h 6 h 9 h 12 h 24 h ORC1 GFP-RB N757F GFP-RB N757F GFP-RB N757F GFP-RB N757F GFP-RB GFP-RB WT WT GFP-RB R661W GFP-RB WT GFP-RB R661W GFP-RB WT GFP-RB R661W GFP-RB WT GFP-RB R661W GFP-RB CCNE1 + + + + + + + + + + + + ORC1-Flag

CCNA2 α-Flag

GAPDH α-ORC2

α-ORC3

C α-ORC4

α-CDC6

α-GFP Input (2.5%) IP (25%) GFP-RB WT WT GFP-RB WT GFP-RB WT GFP-RB WT GFP-RB GFP-RB R661W GFP-RB R661W GFP-RB GFP-RB N757F GFP-RB N757F GFP-RB + − − − + − − − Empty Vector − + + + − + + + ORC1-Flag

α-Flag

α-GFP

α-ORC3 Input (2.5%) α-Flag IP (25%) Figure 1-figure supplement 1

A B

0.004 0.004 0.0035 *** *** 0.0035 0.003 *** 0.003 NS

0.0025 *** level mRNA 0.0025 mRNA level mRNA NS 0.002 *** 0.002 NS *** *** ORC1 0.0015 GAPDH 0.0015 NS * 0.001 0.001 0.0005 0.0005 Relative Relative 0 0 0 3 6 9 12 24 0 3 6 9 12 24 Hours Hours

C D

** 0.014 0.018 *** *** *** 0.012 0.016 0.014 0.01

mRNA level mRNA level mRNA 0.012 *** *** 0.008 NS 0.01 ** NS NS * NS 0.006 0.008 CCNE1 CCNA2 0.006 0.004 0.004 0.002

Relative Relative 0.002 0 0 0 3 6 9 12 24 0 3 6 9 12 24 Hours Hours Figure 1-figure supplement 2 B A C 200 116 116 45 66 97 (2.5%) (2.5%)

GST RB WT Input (2.5%) (2.5%) GST RB R661W ORC1 Input GST RB N757F ORC A-A + +

GST RB WT MBP ORC1 LxCxE + + − − − +

GST RB R661W + + − − + − LxCxE

GST RB N757F − − MBP − + − −

GST RB WT

− − + − − + (20%) (20%)

GST RB R661W

− − MBP-ORC1 MBP-ORC1 + − + −

GST RB N757F (20%)

− + + + − −

GST RB WT

− + − − − + GST RB R661W

− + H sapiens H sapiens M T G X X D P A D melanogaster A C S C C B − − + −

GST RB N757F MBP-RB

guttata humanus aegypti mellifera tropicalis laevis purpuratus malayi rerio quinquefasciatus elegans briggasea gallus musculus + +

GST RB WT − + − −

+ +

GST RB R661W + − − +

+ +

+ − + −

GST RB N757F

1mM 1mM ATP (30nM) E-CDK2 Cyclin

+ + − − α

-GST Cyclin ORC1 ORC1 A-A ORC1 WT S

35 -ORC1 -ORC1 E-Cdk2 E-Cdk2 LxCxE

Figure 2 B A D C Input (5%) Input (5%) + + − − SaOS-2 Cells SaOS-2 Cells U2OS Cells U2OS Cells (

+ ( RB RB Input Input (5%) − + + − Input +

+

− , ,

GST (20%) down Pull GST p16 p16

+ Mouse IgG + − − Rabbit IgG + GST-RB + − − + − + α-ORC1 α-SUV39H1 ) GFP IP (25%)

) +

α α α GST-HDAC1 − + + − -SUV39H1 -SUV39H1 -ORC3 -ORC1

+ ( MCF7 Cells

RB GST-SUV39H1

+ + − − Input

+ + GST-HP1α , Rabbit IgG p16 T7-SUV39H1 T7-SUV39H1 GFP-RB GFP-ORC1 GFP α α α α α + -E2F1 -E2F1 -ORC3 -SUV39H1 -T7 -GFP α-SUV39H1 GST-CDC6 + ) MBP-ORC1 MBP-ORC1 α ★ α α α α -MBP -SUV39H1 -SUV39H1 -ORC3 -RB -ORC1 IgG band GFP-SUV39H1 WT GFP-SUV39H1 GFP-SUV39H1 GFP-SUV39H1 T7-SUV39H1 ORC1-Flag ( ORC1-Flag ORC1-Flag ( ORC1-Flag E2F1/DP1 E2F1/DP1 F E G E2F1/DP1 E2F1/DP1 α α Mut -Tubulin RLU -Tubulin α α α RLU -GFP -Flag -Flag -Flag -Flag

0.6 0.8 1.2 0.2 0.4 0.6 0.8 1.2 0.2 0.4 α ( ( ug

-T7 ( ug ug ug ug 1 0 1 0 Input (2.5%) ) ) ) ) ) Beads *** *** − − − − − − −

H3 Unmodified Peptide Pull down (25%) (25%) down Pull Peptide − − +

− − − + H3K9me1

0.5 H3K9me2 ** − + 1.0 ** ** − − +

H3K9me3 1.0 − + * 0.5 H3S10 ** − − + ** 2.0

*** − + H3K9me3 S10

0.5 − − + *

0.5 H3K14ac **

− +

H3K9ac 0.5 1.0 *** 1.0 − + ** − + H3K14me3

0.5 1.0 2.0 ** − + ** − + α

-MBP

Figure 2-figure supplement 1

α MW MW GST GST-RB GST-HDAC1 GST-SUV39H1 GST-HP1 GST-CDC6 MBP-ORC1 200 116 97 66

45

30 Figure 2-figure supplement 2

A B C Input(5%) MBP MBP-ORC1 MBP-RB + + + + GST-SUV39H1 GST GST SUV39H1 GST GST SUV39H1 GST GST Input(5%) GST GST α-GST Input(5%) + + + MBP-Orc1 + + + MBP-RB α-SUV39H1 α-MBP α-MBP Pull Down (25%) (25%) (25%)

D MBP-ORC1 MBP-RB MBP (100 nM)

40

nM nM nM nM 35 nM nM GST- INPUT 14.5 29 43.4 58 86.9 115.8 SUV39H1 30 units) 25 α-GST 20 arbitary 15 MBP-ORC1 (100 nM) MBP-RB (100 nM)

10

nM nM Binding( nM nM nM nM nM nM 5

nM nM nM nM GST- 29 43.4 58 29 43.4 58 14.5 86.9 115.8 14.5 86.9 115.8 SUV39H1 0 0 50 100 150 α-GST GST-SUV39H1 [nM] Figure 2-figure supplement 3

A B

+ − − + − − GFP (2.5 ug) GFP-ORC1 A-A A-A GFP-ORC1 A-A GFP-ORC1 A-A GFP-ORC1 A-A GFP-ORC1 GFP-ORC1 WT WT GFP-ORC1 CDKGFP-ORC1 WT GFP-ORC1 CDKGFP-ORC1 GFP-ORC1WT CDKGFP-ORC1 WT GFP-ORC1 CDKGFP-ORC1 − + − − + − GFP-ORC1 (2.5 ug) + + + − − − + + + − − − Empty Vector − − + − − + GFP-RB (5 ug) − − − + + + − − − + + + Flag-SUV39H1 + + + + + + Flag-SUV39H1 (2.5 ug) α-Flag α-GFP

α-Flag α-GFP

α-ORC3 α-ORC3

α-E2F1 α-ORC4

Input (5%) GFP IP (25%) Input (2.5%) α-Flag IP (25%)

C Chromo Pre-SET SET Post-SET

1 45 90 140 235 365 412 Input(2.5%) GST GST SUV39H1 GST 1-412 SUV39H1 GST 1-240 SUV39H1 GST 132-412 SUV39H1 GST 220-412 + + + + + + MBP-ORC1

α-MBP (25%) Figure 2-figure supplement 4

A 6X GAL4 UAS Luciferase

1.2 1 2 3 4 5 1 200 116 0.8 97 0.6 66 RLU ** ** 0.4 45 ** *** 0.2 30 0 GAL4 (ug) 0.2 − − − − 20 α-GAL4 GAL4-ORC1 (ug) − 0.1 0.2 − − 14 GAL4-SUV39H1 (ug) − − − 0.1 0.2 α-Tubulin 1 2 3 4 5

B 1.2 1 0.8 0.6 RLU 0.4 0.2 0 E2F1/DP1 − + + + + + ORC3-Flag (ug) − − 0.5 1.0 − − ORC4-Flag (ug) − − − − 0.5 1.0

α-Flag

α-Tubulin Figure 3 +604 (ATG) A CCNE1 -1500 +1000 A B C D F G H > < > < > < > < > < > < > < E > < B

-RB -RB -ORC1 -ORC1 Input Input Input Marker α Marker α IgG α IgG α

A (−1462 to −1318)

B (−683 to −575)

C (−456 to −357)

D (−342 to −195)

E (−280 to −143)

F (−159 to +63)

G (+350 to +490)

H (+816 to +944)

Replicate 1 Replicate 2

C

-SUV39H1 -CDC6 -H3K9me3 -SUV39H1 -CDC6 -H3K9me3 Marker Input IgG α α α Marker Input IgG α α α

A (−1462 to −1318)

B (−683 to −575)

C (−456 to −357)

D (−342 to −195)

E (−280 to −143)

F (−159 to +63)

G (+350 to +490)

H (+816 to +944)

Replicate 1 Replicate 2 Figure 4

A B

-SUV39H1 -CDC6 -SUV39H1 -CDC6 -RB -RB -ORC1 -ORC1 Marker Input IgG α α Marker Input IgG α α Marker α Marker IgG α Input Input IgG α Input α

3 hour 3 hour

6 hour 6 hour

9 hour 9 hour B (−683 to −575) E (−280 to −143) B (−683 to −575) E (−280 to −143)

C 3h 6h 9h

α-ORC1

α-RB

α-SUV39H1

α-CDC6

α-Cyclin E

α-ORC3

α-Tubulin Figure 5 A Input D (2.5%) α-GFP IP MBP MBP-GFP-ORC1

20 20 20 20 20 20 20 20 20 20 20 20 MBP or MBP-GFP-ORC1 (nM) 20 20 20 20 20 20 0 0 0 0 0 0 MBP-RB (nM)

0 10 20 30 40 50 0 10 20 30 40 50 GST-CDC6 (nM) Input(5%) GST CDC6 WT GST A-A CDC6 GST /2 orGC6 /2 cE MBP-RB MBP-GFP-ORC1 0 5 10 15 20 25 0 0 0 0 0 0 CycE-CDK2 (nM) + + + + Cyclin E-CDK2 α-ORC1 α-GST α-Cyclin E α-RB α-Cyclin E (20%) B E 0 h 3 h 6 h 9 h Input (2.5%)

α-GFP IP 20 20 20 20 20 20 20 20 20 20 20 20 MBP-GFP-ORC1 (nM) GFP-CDC6 WT GFP-CDC6WT A-A GFP-CDC6 GFP-CDC6WT A-A GFP-CDC6 GFP-CDC6WT A-A GFP-CDC6 GFP-CDC6WT A-A GFP-CDC6 GFP GFP GFP GFP GFP 0 0 0 0 0 0 20 20 20 20 20 20 MBP-RB (nM) 0 10 20 30 40 50 0 10 20 30 40 50 GST-CDC6 (nM) /2 orGC6 /2 α-GFP cE MBP-RB MBP-GFP-ORC1 0 5 10 15 20 25 0 0 0 0 0 0 CycE-CDK2 (nM) α-ORC1 α-GST α-Cyclin E α-RB α-Tubulin α-Cyclin E C (20%) Input (2.5%) F 2 ** α-GFP IP 1.75 20 20 20 20 20 20 20 20 20 20 20 20 MBP-GFP- ORC1 (nM) 1.5 *** *** 20 20 20 20 20 20 20 20 20 20 20 20 MBP-RB (nM) 1.25 * * * 0 10 20 30 40 50 0 0 0 0 0 0 GST-CDC6 (nM) 1 RLU /2 orGC6 /2 0 5 10 15 20 25 0 5 10 15 20 25 0.75 cE MBP-RB MBP-GFP-ORC1 CycE-CDK2 (nM) 0.5 α-ORC1 0.25 *** α-GST 0 E2F1/DP1 − + + + + + + + α-RB GFP-CDC6 WT (ug) − − 0.3 0.4 0.5 − − − α-Cyclin E GFP-CDC6 A-A (ug) − − − − − 0.3 0.4 0.5 (20%) α-GFP

α-Tubulin Figure 5-figure supplement 1 MW MW MBP MBP MBP-GFP-ORC1 MBP-RB GST-CDC6

200

116 97 66

45 Figure 6 A 3 ** 2.5 *** 2 ** 1.5 RLU * * 1 * 0.5 *** 0 E2F1/DP1 − + + + + + + + α-ORC1

α-SUV39H1

α-CDC6

α-RB

α-Tubulin

GFP #1 #2 #1 #2 #1 #2 siRNA ORC1 SUV39H1 CDC6 siRNA siRNA siRNA B C Control siRNA ORC1 siRNA

siRNA siRNA

-SUV39H1 -SUV39H1 -H3K9me3 -H3K9me3 Marker Input α Marker α IgG α Input IgG α Control ORC1

A (−1462 to −1318) α-ORC1

α-SUV39H1 B (−683 to −575) α-H3K9me3 C (−456 to −357) α-Cyclin E D (−342 to −195) α-CDC6

E (−280 to −143) α-ORC3

F (−159 to +63) α-Tubulin

G (+350 to +490)

H (+816 to +944) Figure 6-figure supplement 1

Control siRNA Orc1 siRNA P=0.0005 5 4.5 4 units) 3.5 3 arbitary 2.5 2 1.5

BandIntensity( 1 P=0.4 P=0.2 0.5 0 IgG SUV39H1 H3K9me3

Antibodies Figure 7 A B + − − − + − − − Empty Vector + − − − + − − − GFP − + − − − + − − ORC1-Flag 1-861 − + − − − + − − GFP-ORC1 1-861 − − + − − − + − ORC1-Flag 1-700 − − + − − − + − GFP-ORC1 1-700 − − − + − − − + ORC1-Flag 1-768 − − − + − − − + GFP-ORC1 1-768 + + + + + + + + GFP-RB + + + + + + + + Flag-SUV39H1

α-Flag α-GFP

α-Flag α-GFP α-ORC3 α-ORC3 α-ORC4 α-ORC4 Input (5%) α-GFP IP (25%) Input (5%) α-Flag IP (33%) C + + + − − − + + + − − − Empty Vector − − − + + + − − − + + + Flag-SUV39H1 + − − + − − + − − + − − GFP-ORC1 1-861 − + − − + − − + − − + − GFP-ORC1 1-700 − − + − − + − − + − − + GFP-ORC1 1-768

α-GFP

α-ORC3 D α-ORC4 1.2 α-Flag 1 0.8 Input (5%) α-Flag IP (25%) 0.6 RLU ** * *** 0.4 *** *** ** 0.2 *** 0 E2F1/DP1 − + + + + + + + ORC1-Flag 1-861 (ug) − − 1.0 2.0 − − − − ORC1-Flag 1-700 (ug) − − − − 1.0 2.0 − − ORC1-Flag 1-768 (ug) − − − − − − 1.0 2.0

α-Flag

α-Tubulin Figure 8

M G1 S G2 M Human ORC1

Cyclin E-CDK2 Cyclin A-CDK2

Human CDC6

Window of Switching

SUV39H1 SUV39H1 CDC6 SUV39H1 CyclinE-CDK2 ORC1 RB Cyclin D-CDK4 RB HP1 HP1 Pre-RC me3 me3 ORC1 H3K9 ORC1 E2F1 E2F1 H3K9 assembly Cyclin E