GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2006) 11, 829-844. doi:10.1111/j.1365-2443.2006.00981.x
© 2006 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dimitrova, D. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dimitrova, D. S.

Nuclear transcription is essential for specification of mammalian replication origins

Daniela S. Dimitrova*,a

Department of Biological Sciences, State University of New York at Buffalo, Buffalo, NY 14260, USA


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
I have demonstrated that nuclear transcription modulates the distribution of replication origins along mammalian chromosomes. Chinese Hamster Ovary (CHO) cells were exposed to transcription inhibitors in early G1 phase and replication origin sites in the dihydrofolate reductase (DHFR) gene locus were mapped several hours later. DNA within nuclei prepared from control and transcription-deficient G1-phase cells was replicated with similar efficiencies when introduced into Xenopus egg extracts. Replication initiated in the intergenic region within control late-G1 nuclei, but randomly within transcriptionally repressed nuclei. Random initiation was not a consequence of inability to produce an essential protein(s), since initiation was site-specific within cells exposed to the translation inhibitor cycloheximide during the same interval of G1 phase. Furthermore, in vivo inhibition of transcription within late-G1-phase cells reduced the frequency of usage of pre-established DHFR replication origin sites. Transcription rates in the DHFR domain were very low and did not change throughout G1 phase. This implies that, although ongoing nuclear transcription is required, local expression of the genes in the DHFR locus alone is not sufficient to create a site-specific replication initiation pattern. I conclude that epigenetic factors, including general nuclear transcription, play a role in replication origin selection in mammalian nuclei.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Eukaryotic chromosomes are subdivided into multiple units (replicons) that replicate once per cell cycle. Replicon activation is under rigorous spatio-temporal control to ensure that replication initiates within a defined region (replication origin) at a specific and reproducible time during S phase (Dimitrova & Gilbert 1999a). Establishment of replication origin sites in all eukaryotes starts in late telophase with the assembly of prereplication complexes (pre-RCs) (Diffley et al. 1995; Dimitrova et al. 1999; Rowles et al. 1999). Pre-RC assembly is mediated through a hierarchy of molecular interactions, whereby ORC heterohexameric complex (ORC 1–6) binds to DNA and recruits CDC6, which, together with CDT1, is required for the loading of the MCM heterohexameric complex (MCM 2–7). In all systems studied to date, MCM loading in late mitosis (defined as replication licensing) completes functional pre-RC assembly (Diffley et al. 1995; Rowles et al. 1999; Dimitrova et al. 2002). In S phase, following activation by the CDC7-DBF4 kinase and cyclin-dependent protein kinase(s) (Cdk), pre-RCs are transformed into initiation complexes, giving rise to active replication forks.

It is still an enigma how replication origins are established in higher eukaryotes. There is no specific sequence requirement for origin function, and initiation regions vary in size from a few nucleotides to several tens of kilobases. Paradoxically, pre-RC proteins, which are conserved throughout Eukarya and have been shown to bind DNA in a sequence-specific manner in yeast (e.g. ORC), have evolved in higher eukaryotes to exhibit broad sequence tolerance in their DNA binding properties. Hence, it is unclear how mammalian pre-RC proteins select DNA binding sites. An attractive hypothesis is that epigenetic factors play a major role in constraining mammalian replication origins to defined chromosomal regions (DePamphilis 1999; Dimitrova & Gilbert 1999a; Mechali 2001). It is conceivable that the requirement to coordinate DNA replication with other nuclear activities (e.g. gene expression, DNA repair, maintenance of chromosome integrity) imposes limitations on origin usage and creates selective advantage for utilization of the most favorable out of all potential initiation sites.

DNA replication and nuclear transcription are involved in a complex interplay. There is a correlation between transcription status and replication timing of chromosomal domains (Schubeler et al. 2004). DNA replication can exert both repressive and activating effects on gene expression. Conversely, replication origin activity is stimulated by transcription factors bound nearby origins (Li et al. 1998; Danis et al. 2004) and, remarkably, many of the known replication origins are found in the vicinity of transcription units (DePamphilis 1999). Origin positioning relative to active genes is of the utmost importance, however, as transcription through an initiation locus inactivates the origin (reviewed in Dimitrova & Gilbert 1999a). Given that promoter strength inversely correlates with origin activity (Haase et al. 1994), it is likely that transcription interference targets the replication initiation step. Interestingly, most replication origins identified to date are found in intergenic regions (DePamphilis 1999). Furthermore, a survey of yeast replication origin sequences indicated that they contain functional transcriptional terminators, which operate in an orientation-independent manner (Chen et al. 1996). These facts prompted me to speculate that, during G1 phase, ongoing transcription may destabilize and disassemble pre-RCs within active transcription units, effectively pushing replication origins to the intergenic regions (Dimitrova & Gilbert 1999a). In organisms where non-coding sequences represent a small fraction of the genome (e.g. 30% in yeast; Shabalina et al. 2001), specific sequences that direct ORC binding and terminate local transcription may have evolved to ensure stability of pre-RCs. However, specific origin sequences may prove dispensable in organisms with large genomes [e.g. in Homo sapiens non-coding sequences comprise 95% of the genome (Shabalina et al. 2001)].

The present study was designed to test the hypothesis that transcription interference is a factor responsible for modulating mammalian replication origin positions. The replication origin locus downstream of the DHFR gene in CHOC 400 cells was used as a model system. Nuclei prepared from cells synchronized in G1 phase replicate efficiently when introduced into Xenopus egg extracts. Replication in the DHFR locus initiates in a random manner within early G1-phase nuclei, but in a site-specific manner within late-G1-phase nuclei. This mid-G1-phase transition has been termed the Origin Decision Point (ODP) (Wu & Gilbert 1996). The molecular mechanism of the ODP is yet unknown. Here, I demonstrate that ongoing nuclear transcription is essential for focusing of replication initiation to the intergenic spacer downstream of the DHFR gene at the ODP.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Establishing conditions for reliable detection of nascent transcripts

I began this study by testing whether the ODP transition could occur in the absence of nuclear transcription. Transcription activity was initially monitored by brief in vivo pulse-labeling of nascent RNA with 5-bromouridine (BrU) and subsequent immunostaining with fluorescent anti-BrdU antibodies. To my surprise, this widely implemented and profusely cited technique often produced inconsistent results. No fluorescent signal could be detected in many experiments. In the cases when nuclei were labeled, the distribution of fluorescent transcription foci was variable and followed one of three patterns: (i) nucleolar labeling (Fig. 1A), (ii) nucleoplasmic labeling (Fig. 1B), (iii) nucleolar and nucleoplasmic labeling, often of similar intensities (Fig. 1C). The outcome of the positive experiments was influenced by the fixation technique used, with formaldehyde producing mostly nucleoplasmic labeling, whereas alcohol fixation tended to produce predominantly nucleolar labeling. Similar results were obtained when nascent RNA was labeled with BrUTP within permeabilized cells (as described by Jackson et al. 1993; Wansink et al. 1993), thus eliminating BrU import as an explanation for the poor results. Since under normal conditions only the third fluorescent pattern must be present in all interphase nuclei, it is clear that the performance of this methodology must be regarded with caution. An extensive survey of the literature, as well as personal communications with several established investigators, revealed that problems with visualization of BrU-substituted RNA have been encountered (but seldom reported) by many labs in the eukaryotic transcription field (e.g. Jackson et al. 1993; Wansink et al. 1993; Masson et al. 1996; Hall et al. 2002, among others). Various experimental modifications have been proposed to overcome these difficulties, but none of them worked consistently in my hands. The only successful resolution to this problem was provided by the methodology introduced by Boisvert et al. (2000), which employed 5-fluorouridine (FU) for labeling of nascent RNA and identified an anti-BrdU antibody that binds efficiently to this nucleoside analog. The success rate of the experiments improved dramatically and, importantly, RNA labeling was efficient at micromolar FU concentrations, thus reducing the risk of toxic effects on cell physiology. All interphase nuclei exhibited the expected pattern of numerous fine nucleoplasmic foci of moderate intensity and prominently stained nucleolar regions (Fig. 1D). Since rRNA synthesis accounts for ~50% of the total RNA output in CHOC 400 cells (see below), this immunolabeling procedure reveals accurately the transcription activity in various nuclear compartments.


Figure 1
View larger version (65K):
[in this window]
[in a new window]
  		Figure 1  In vivo labeling of nascent RNA with halogenated nucleosides. Asynchronous cultures of CHOC 400 cells were pulse-labeled for 10–15 min with BrU (A–C) or FU (D), fixed and stained with fluorescent anti-BrdU antibodies. DNA was stained with DAPI.

 
Establishing conditions for in vivo inhibition of RNA polymerase II-directed transcription

Three RNA polymerase II (PolII) inhibitors with different modes of action were used in parallel. {alpha}-amanitin ({alpha}-AM) selectively inhibits extranucleolar transcription (Lindell et al. 1970). Actinomycin D (AMD) intercalates between dGpC dinucleotides in DNA and blocks transcript elongation (Sobell 1985). Nucleolar transcription is particularly sensitive to AMD due to the high GC content of rDNA (Perry & Kelley 1970). 5,6-dichloro-1-ß-D-ribofuranosylbenzimidazole (DRB) inhibits PolII transcription, with little effect on RNA polymerases I and III (Granick 1975; Le Panse et al. 1999). DRB does not affect initiation of transcription (Tamm et al. 1980), but enhances transcriptional pausing (Maderious & Chen-Kiang 1984) through inhibition of the positive transcription elongation factor P-TEFb (Marshall & Price 1992).

First, the inhibitor concentrations required to block nuclear transcription in vivo were defined, as these concentrations can vary between cell types. Asynchronous CHOC 400 cells, pre-incubated for 3 h in the presence of inhibitors, were pulse-labeled with FU and immunostained with fluorescent anti-BrdU antibodies. As shown in Fig. 2A, at least 100 µg/mL of {alpha}-AM were required to suppress nucleoplasmic transcription completely. Nucleolar transcription remained unaffected, since PolI is not sensitive to {alpha}-AM (Kedinger & Simard 1974; Wieland & Faulstich 1991). Such high {alpha}-AM concentration required for in vivo PolII inhibition is attributed to a notoriously slow penetration of the chemical into live cells (Kedinger & Simard 1974). Consistent with this notion, at least 3-h incubation with {alpha}-AM was required to achieve efficient inhibition of PolII transcription (data not shown). Similar to {alpha}-AM, DRB (at ≥ 50 µM) inhibited nucleoplasmic, but not nucleolar transcription (Fig. 2C). By contrast, low concentrations of AMD (0.01–0.05 µg/mL) first inhibited nucleolar transcription, whereas ~100-fold higher concentrations were required to suppress extranucleolar RNA synthesis (Fig. 2B). To exclude the possibility that FU is inadvertently incorporated into DNA, instead of RNA, an aliquot of the cells was treated with aphidicolin (Aph), an inhibitor of replicative DNA polymerases. Aph had no effect on nuclear transcription (Fig. 2E), indicating that FU is faithfully incorporated into nascent RNA under these conditions.


Figure 2
View larger version (78K):
[in this window]
[in a new window]
  		Figure 2  Effect of transcription inhibitors on RNA synthesis in vivo. Cultures of CHOC 400 cells, preincubated for 3 h with (A) {alpha}-AM, (B) AMD, (C) DRB or (E) Aph at the indicated concentrations, were pulse-labeled for 15 min with FU, fixed and stained with fluorescent anti-BrdU antibodies. Untreated control is shown in (D). DNA was stained with DAPI.

 
Transcription rates are reduced when halogenated nucleotides are used as precursors for RNA synthesis (Jackson et al. 1993; Wansink et al. 1993). It is possible that the sensitivity of the immunofluorescent labeling method might not be sufficient to detect low levels of transcription. Therefore, a parallel analysis was performed using a different approach for detection of RNA synthesis. CHOC 400 cells, pretreated in vivo for 3 h with transcription inhibitors, were permeabilized with digitonin to prepare intact nuclei. Nuclei were resuspended in transcription cocktail supplemented with 32P-UTP and nuclear run-on analysis was performed. Since initiation does not occur in vitro, the amount of incorporated 32P-UTP would be proportional to the number of in vivo initiated transcripts that are available for in vitro elongation. In agreement with the immunofluorescent analysis, the nuclear run-on experiments confirmed that at least 100 µg/mL of {alpha}-AM are necessary to inhibit completely PolII transcription in vivo (Fig. 3A), which amounts to ~50% of total transcription. No inhibition, or only partial inhibition was observed at lower {alpha}-AM concentrations, whereas doses > 100 µg/mL did not cause further inhibition (not shown). AMD levels that inhibited nucleolar transcription in the immunofluorescent analysis (0.01–0.05 µg/mL) reduced UTP incorporation by approximately half, whereas global suppression of transcription required at least 2.5 µg/mL AMD. In contrast to the immunocytochemical experiments, in vivo treatment with DRB did not result subsequently in reduced levels of UTP incorporation in the run-on assay (Fig. 3A). This discrepancy can be explained by taking into account that the effect of DRB is instantaneously reversible (Tamm et al. 1976). Since DRB does not block initiation and the stalled RNA polymerase complexes and short nascent transcripts remain associated with the template during nuclear preparation, transcription can resume immediately after removal of the drug (e.g. in the run-on assay following nuclear isolation).


Figure 3
View larger version (61K):
[in this window]
[in a new window]
  		Figure 3  Effect of inhibitors on transcription rates in CHOC 400 nuclei. (A) Run-on assays were performed for 20 min at 21 °C with 2.5 x 105 intact nuclei prepared from exponentially growing CHOC 400 cells that were pretreated in vivo for 3 h with the indicated inhibitors (drug concentrations are as in Fig. 2). Total transcription rates were determined by measuring the amount of 32P-UTP incorporated into acid-insoluble RNA. Results were normalized to the values obtained with control cells (100%). Shown are means ± SD of three experiments. (B) Nuclear run-on products from 106 control or inhibitor-treated CHOC 400 cells were hybridized to probes from the CHO DHFR domain. Similar results were obtained in four independent experiments. (C) Map of the Chinese hamster DHFR domain. Shown is a representative blot from a nuclear run-on transcription assay of asynchronous CHOC 400 cells. The positions of the DNA probes are indicated on the map. The arrows above the three genes indicate the direction of transcription. The positions of the three major replication initiation sites (ori-ß, ori-ß' and ori-{gamma}) are shown in the expended segment of the intergenic region.

 
Next, to verify that the results obtained by analyzing global inhibition of transcription accurately reflect the status of the DHFR gene locus, run-on experiments were performed with nuclei prepared from control or transcriptionally inhibited CHOC 400 cells. RNA was purified and hybridized to a panel of unique DNA probes, which span ~120 kb of the DHFR locus (Fig. 3B). The map of the CHO DHFR domain and positions of probes are shown in Fig. 3C. Nascent transcripts purified from exponentially growing CHOC 400 cells hybridized to DNA sequences within the three genes (rep3, DHFR and 2BE2121) identified in this locus, as well as to the region of probe L, which does not contain a known gene (it is possible that this probe detects a previously unidentified intergenic transcript). RNA synthesis in all transcription units was completely inhibited by 100 µg/mL of {alpha}-AM, consistent with the notion that they are transcribed by PolII (Crouse et al. 1985). The residual {alpha}-AM-insensitive, but AMD-sensitive signal at position H in the rep3 gene is likely due to crossreactivity with polyA– transcripts from another unidentified genomic locus/loci (Mitchell et al. 1986). Unlike the immunofluorescent analysis, where 2.5 µg/mL of AMD seemed sufficient to block extranucleolar transcription, the nuclear run-on analysis showed that transcription in the DHFR locus is only partially inhibited (Fig. 3B). Complete inhibition was observed at ≥ 5 µg/mL of AMD (not shown). Consistent with Fig. 3A, RNA extracted from in vivo-DRB-treated cells hybridized efficiently to the promoter-proximal regions of the genes, indicating that the promoters remain active during DRB arrest. The absence of hybridization to promoter-distal probes (S, P and O) at the inhibitory 100 µM DRB concentration, however, indicates that transcription through the rest of the genes is efficiently inhibited by DRB in vivo (transcript extension under run-on conditions is very limited and the RNA polymerases cannot travel far from their original positions). Finally, the strong signal from probe L at 100 µM DRB suggests that this probe might be located near the promoter of the putative intergenic transcript.

RNA polymerase II-directed transcription is required for the establishment of a specific replication initiation zone within the CHO DHFR gene locus

To determine whether transcription affects the ODP transition, pure metaphase populations of CHOC 400 cells were released into G1 phase. Two hours later, at the pre-ODP stage, transcription inhibitors were added to the medium and the cells were incubated for an additional 4 h. This time is sufficient to achieve maximum transcriptional inhibition (Figs 2 and 3) and, as previously shown (Dimitrova & Gilbert 1999b; Dimitrova et al. 1999), allows untreated cells to proceed to the post-ODP stage. Intact nuclei prepared from control or drug-treated cell populations were introduced into Xenopus egg extracts and the distribution of early replication intermediates in the DHFR domain was evaluated by the early labeled fragment hybridization (ELFH) assay. With {alpha}-AM and DRB, the rates of replication were similar in control and drug-treated nuclei, whereas high concentrations of AMD reduced replication (not shown). Figure 4 shows that concentrations of all three inhibitors that suppressed PolII-mediated transcription (see Figs 2 and 3) also inhibited the ODP transition.


Figure 4
View larger version (61K):
[in this window]
[in a new window]
  		Figure 4  Ongoing nuclear transcription, but not protein synthesis, is required for site-specific initiation of DNA replication. (A–D) The distribution of replication initiation sites at the DHFR locus within CHOC 400 cells treated with the indicated inhibitors between 2 and 6 h postmetaphase was evaluated by the ELFH assay (Gilbert et al. 1995; Dimitrova & Gilbert 1998). (A) {alpha}-AM treatment. (B) AMD treatment. (C) DRB treatment. (D) CHX treatment. Similar results were obtained in four independent experiments. The vertical gray line highlights the positions of probes B-R, which encompass the region of peak initiation activity. The right panel in (D) shows translation levels in cultures of CHOC 400 cells incubated in the absence (0) or presence (100) of 100 µg/mL CHX between 2 and 6 h postmetaphase. Protein synthesis was followed by measuring the amount of 35S-Met incorporated into TCA-insoluble material during the last 30 min of the incubation period. (E) Effect of KT5720 on RNA synthesis in vivo. Cell treatment with the indicated drug concentrations, pulse-labeling with FU and immunofluorescent staining of FU-labeled RNA were performed as in Fig. 2.

 
One possible explanation for this result is that inhibition of transcription arrests the cells at the pre-ODP stage due to inability to produce a protein essential for proper G1 phase progression. I addressed this question by incubating in parallel aliquots of the cells with the translation inhibitor cycloheximide (CHX). As shown in Fig. 4D, CHX treatment had no effect on the distribution of replication origin sites in the DHFR locus. This implies that all proteins required for passage through the ODP are present in CHO cells from the beginning of G1 phase.

The finding that inhibition of PolII-mediated transcription results in random initiation of DNA replication in the CHO DHFR locus contradicts the observations of a recent report where it was concluded that "transcription per se is not required for passage through the ODP" (Keezer & Gilbert 2002). This conclusion was based on the observation that KT5720, a PKA/MAPK inhibitor, did not prevent passage through the ODP at a concentration found to inhibit completely all nuclear transcription. One possible explanation for this discrepancy is that the transcription inhibitors used in the present study have pleiotropic effects (a common concern with chemical inhibitor studies) and inhibit the ODP transition through a mechanism unrelated to transcription. Whereas this possibility cannot be excluded, it seems unlikely at least in the case of {alpha}-AM whose only known cellular target is RNA polymerase (Wieland & Faulstich 1991) and therefore drug doses that do not affect transcription are not likely to influence directly other cellular processes. Another possible explanation for the contradiction between the two studies could be that perhaps KT5720 was wrongfully classified as a transcription inhibitor. This possibility was tested by treating CHOC 400 cells with increasing concentrations of KT5720, followed by evaluation of the in vivo incorporation of FU within the nucleolar and extranucleolar compartments (Fig. 4E). No evidence was found that KT5720 acts as a global inhibitor of nuclear transcription even at ten-fold higher concentration than that tested in the previous report (100 µM vs. 10 µM, respectively). At the same time, treatment of an asynchronous culture with 100 µM KT5720 for 18 h resulted in the arrest of > 50% of the cells in mitosis, verifying that the protein kinase inhibitor was active in vivo. Whereas these results do not exclude that KT5720 might alter the expression of individual genes, they demonstrate that the drug is not a bona fide transcription inhibitor.

Next, I investigated whether the drugs could alter a pre-established specific initiation pattern within post-ODP nuclei. To this end, CHOC 400 cells were released from mitosis for 6 h to allow the cells to reach the post-ODP stage (Fig. 5, hatched circles). Parallel cell cultures were supplemented with {alpha}-AM, AMD, or DRB (Fig. 5) and incubation was continued for another ~2 h. This time represents a compromise between the time required for the drugs to exert their effect (e.g. ~3 h in the case of {alpha}-AM) and the time when the cells begin to enter S phase (~8–9 h post-metaphase, as monitored by BrdU labeling). Compared to control cells, treatments with all three transcription inhibitors reproducibly reduced the preference for ori-ß/{gamma} usage within late-G1 nuclei, but failed to erase the pre-established pattern completely. This effect was not due to altered replication efficiency in late-G1 nuclei, since similar amounts of replication intermediates were generated within pre- and post-ODP-treated nuclei when tested in parallel (not shown). There are two possible explanations for the lack of elimination of origin specificity at the post-ODP stage. First, the shorter duration of the post-ODP-cell treatment may not be sufficient for the drugs to exert their effects in full. Second, perhaps the maintenance of specific replication origin configurations is more resistant to transcriptional inhibition than is their establishment. Regardless of the exact reason, to my knowledge, this represents the first demonstration that interference with an independent nuclear activity can affect the preset replication landscape in the DHFR locus.


Figure 5
View larger version (30K):
[in this window]
[in a new window]
  		Figure 5  Effect of transcription inhibitors on a pre-established replication initiation pattern in the DHFR domain within post-ODP cells. Metaphase populations of CHOC 400 cells were released into G1 phase. Six hours later, after passage through the ODP, the indicated drugs were added to aliquots of the cell cultures and incubation was continued for another ~2 h. Replication origin sites in the DHFR locus were mapped as in Fig. 4. Hatched circles and filled squares represent control untreated cells collected at 6 or 8 h postmetaphase, respectively. Filled circles, 100 µg/mL {alpha}-AM at 8 h; triangles, 10 µg/mL AMD at 8 h; diamonds, 100 µM DRB at 8 h. Similar results were obtained in three independent experiments.

 
Collectively, these results show that inhibition of nuclear transcription affects the distribution of replication origins along mammalian chromosomes. I conclude that transcription, but not protein synthesis, is required for establishment of specific replication origin sites in the hamster DHFR domain.

Transcription in the CHO DHFR gene locus during the cell cycle

In a previously proposed model (Dimitrova & Gilbert 1999a), I suggested that transcription might contribute to focusing of replication initiation to specific chromosomal sites by disrupting some of the pre-RCs assembled on chromatin during late telophase (Dimitrova et al. 2002). In mammalian cells, transcription is repressed during mitosis and gradually resumes on entry into G1 phase (Prescott & Bender 1962). Consistent with this, immunocytochemical and biochemical analysis of synchronized CHOC 400 cell populations showed that nucleoplasmic and nucleolar transcription were repressed during prophase and metaphase, respectively (Fig. 6A). Transcription resumed in telophase (Fig. 6A), increased during the next 2 h to a level that was maintained through most of G1 phase (Fig. 6B), and increased ~2-fold at the G1-S transition. If the maintenance of stable pre-RCs on chromatin were incompatible with passage of the transcription machinery, replication origin sites would be preferentially excluded from active transcription units. My data (Fig. 4) show that potential replication initiation sites become focused to the DHFR origin region during mid-G1 phase and ongoing nuclear transcription is required for this event. This raised the intriguing possibility that the ODP might be a direct consequence of induction of the DHFR and 2BE2121 genes in mid-G1 phase.


Figure 6
View larger version (50K):
[in this window]
[in a new window]
  		Figure 6  Cell cycle regulated transcription in the CHO DHFR gene locus. (A) General transcription is repressed during pro/metaphase and resumes in telophase. CHOC 400 cells were pulse-labeled for 10 min with FU and immunostained with anti-BrdU antibodies to visualize in vivo transcription sites. DNA was stained with DAPI. (B) Transcription rates in synchronized cells. Nuclear run-on assays were performed for 20 min at 21 °C with 2.5 x 105 nuclei prepared from CHOC 400 cells synchronized at the indicated times (M—metaphase; 1 h, 2 h, 4 h or 6 h postmetaphase; G1/S transition; 3 or 8 h into S phase; and exponentially growing cells). The amount of 32P-UTP incorporated into acid-insoluble RNA was measured and results were normalized to the values obtained with the G1/S cells. Shown are means ± SD of ten experiments with different batches of synchronized cells. (C) 32P-labeled nuclear run-on transcription products purified from 106 synchronized CHOC 400 cells were hybridized to DNA probes from the hamster DHFR locus. G1/S represents cells released from metaphase for 14 h in the presence of Aph; S8 represents G1/S-arrested cells that were released from the Aph block for 8 h. Similar results were obtained in ten experiments. (D) Relative transcription rates of the hamster DHFR gene across the cell cycle. Nuclear run-on assays were performed with synchronized populations of CHOC 400 cells as described in (C). Relative c.p.m. hybridized to a DHFR cDNA probe (see Fig. 3C, probe A) or a hamster APRT genomic probe were obtained by phosphorimaging analysis. The values for the DHFR probe were divided by the values for the APRT probe and results were normalized to the highest value among the different cell cycle times. Similar results were obtained in four independent experiments.

 
DHFR gene expression has been studied extensively in both human and rodent cell lines. Unfortunately, most of these studies have been performed either with serum-stimulated quiescent cells and, thus, may not reflect properly gene regulation in normally cycling cells, or have analyzed DHFR mRNA levels, which do not necessarily correlate with actual transcription rates. Importantly, experiments that utilize G0-arrested cells are not suitable for analysis of early/mid-G1-phase events. I therefore directly examined transcription rates in the CHO DHFR domain. Nocodazole-arrested CHOC 400 cells were released from the block and run-on assays were performed with aliquots of cells collected at various times thereafter. Run-on products were hybridized to DNA probes from the rep3, DHFR and 2BE2121 genes. Transcription of all three genes and the putative intergenic unit followed a similar pattern (Fig. 6C). During G1 phase, nascent DHFR and 2BE212 RNA levels are very low and barely detectable, but reproducibly above background levels (i.e. hybridization to a probe from {lambda}-phage DNA). Transcription rates in the DHFR locus rise significantly at the G1/S border (Fig. 6C). To evaluate the degree of DHFR gene induction, I normalized the signal hybridizing to probe A (DHFR cDNA) to the signal hybridizing to a CHO adenine phosphoribosyltransferase (APRT) gene probe. APRT gene expression is not cell cycle-regulated (Mitchell et al. 1986). Relative to APRT, DHFR transcription levels increased ~3–4-fold at the G1/S transition and, later in S phase, dropped back to the G1-phase levels (Fig. 6D).

These data show that transcription rates in the DHFR locus reach a constant level within 1–2 h post-metaphase and do not change further during early and mid-G1 phase. This result does not support the hypothesis that the act of transcription of nearby genes is solely responsible for replication origin specification within a chromosomal domain. It is clear that focusing of initiation to the intergenic spacer between the DHFR and 2BE2121 genes occurs during a cell cycle time when no significant change in the transcription pattern in this region takes place.

Transcription in Xenopus egg extracts has no effect on the distribution of replication origin sites established in CHO nuclei in vivo

Early Xenopus embryos are transcriptionally silent until the midblastula transition (Newport & Kirschner 1982) and naked DNA introduced into eggs or egg extracts is not transcribed (Bendig & Williams 1983). It is believed that the excess amount of histones present in Xenopus egg cytosol are responsible for rapid assembly of DNA into repressive chromatin, which precludes the access of promoter-binding factors to the DNA template and, thus, prevents transcription by all three RNA polymerases. However, if transcription factors are allowed to bind to promoters before the DNA template is introduced into Xenopus egg cytosol, transcriptional repression is alleviated and the potentiated template is transcribed efficiently (Prioleau et al. 1994). Mammalian nuclei contain such preactivated templates and it is possible that they might transcribe efficiently when introduced into egg extracts. The ELFH origin mapping assay entails 30–45-min incubation of CHO nuclei into Xenopus egg extract before radioactive labeling of nascent DNA strands by run-on replication (e.g. Fig. 4). It is formally possible that transcription taking place in CHO nuclei under ELFH conditions might rearrange the in vivo-established replication origin positions.

To test this possibility, intact nuclei prepared from exponentially growing CHOC 400 cells were introduced into Xenopus egg cytosol supplemented with 32P-UTP and the rate of incorporation of radioactive nucleotides into RNA was measured by acid precipitation. As shown in Fig. 7A–C (squares), transcription rates were high for ≥ 2-h incubation period. This differs significantly from the kinetics of run-on transcription with nuclei introduced into transcription cocktail, whereby incorporation of radioactive nucleotides declines within 20–30 min (not shown). Transcription was entirely dependent on the presence of CHO nuclei and transcription rates were not affected by inhibition of DNA replication with Aph (Fig. 7C). Furthermore, RNA synthesis showed similar sensitivity to transcription inhibitors whether nuclei were incubated in Xenopus egg extract or in transcription cocktail (not shown).


Figure 7
View larger version (49K):
[in this window]
[in a new window]
  		Figure 7  Transcription takes place in CHO nuclei incubated in Xenopus egg extracts. (A–C) Transcription rates in Xenopus egg cytosol. 5 x 105 nuclei from asynchronous CHOC 400 cells were introduced into LSS extracts supplemented with 32P-UTP and (A) {alpha}-AM (diamonds, 1 µg/mL; circles, 10 µg/mL; triangles, 100 µg/mL); (B) AMD (diamonds, 0.25 µg/mL; circles, 2.5 µg/mL; triangles, 5 µg/mL); (C) DRB (circles, 100 µM) or Aph (diamonds, 100 µg/mL). Aliquots were removed at the indicated times and the amount of 32P-UTP incorporated into RNA was determined by acid precipitation. Squares represent control reaction without drugs. Triangles in (C) represent mock reaction without addition of nuclei. Similar results were obtained in three independent experiments. (D) Intact nuclei were prepared from CHOC 400 cells synchronized as in Fig. 6C. Nuclear run-on assays were carried out with or without prior incubation in Xenopus egg cytosol as described in Experimental procedures. Purified run-on products from 106 nuclei were hybridized to a DHFR cDNA probe and an APRT genomic probe. Similar results were obtained in four independent experiments. (E) Intact nuclei prepared from CHOC 400 cells synchronized at 2 h (pre-ODP; squares and triangles) or 6 h (post-ODP; circles and diamonds) postmetaphase were introduced into Xenopus egg extracts. The distribution of replication initiation sites in the DHFR locus was evaluated by the ELFH assay as in Fig. 4, with (diamonds and triangles) or without (squares and circles) addition of 100 µg/mL {alpha}-AM to the egg extract.

 
These results indicate that CHO nuclei represent good templates for transcription in Xenopus egg extract. Hence, I tested whether genes in the DHFR locus are transcribed under ELFH conditions. Intact nuclei were prepared from CHOC 400 cells synchronized at various cell cycle times. Nuclear run-on assays were performed with aliquots of nuclei either without or with prior incubation of the nuclei in Xenopus egg extract. Total RNA was extracted and hybridized to probes from the hamster DHFR and APRT genes (Fig. 7D). Consistent with Fig. 6C,D, in vivo expression levels of the DHFR gene were very low during G1 phase ("In Culture"). However, significant induction of DHFR gene transcription was observed after 45-min incubation of nuclei in Xenopus egg cytosol, possibly due to nuclear remodeling known to occur in this system (Kikyo et al. 2000). Similar induction was not observed for the APRT gene, which is transcribed at high rates in vivo and therefore is likely to be already in an open chromatin conformation.

These findings raised the question whether the elevated in vitro DHFR transcription rates influence the replication initiation pattern in this locus obtained through the use of the Xenopus in vitro system. To address this issue, I examined the distribution of replication initiation sites within G1 nuclei after incubation in control or {alpha}-AM-supplemented Xenopus egg cytosol. Figure 7E shows that inhibition of PolII transcription in Xenopus egg extracts has no effect on the replication initiation pattern in the DHFR domain. Furthermore, the experiments described in Fig. 4A–C were repeated by incubating CHO G1 nuclei in {alpha}-AM-supplemented Xenopus egg cytosol and obtained identical results as shown in Fig. 4 (not shown). I conclude that transcription in Xenopus egg extracts during the ELFH origin mapping procedure does not cause redistribution of replication initiation sites pre-established in vivo within CHO nuclei.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
The effect of transcription on replication origin selection in mammalian nuclei has been investigated. The data presented here demonstrate that ongoing nuclear transcription is essential for the establishment of specific replication initiation pattern in the hamster DHFR gene locus. Transition from random to site-specific initiation in CHO nuclei does not require synthesis of novel proteins during early G1 phase, since inhibition of translation with cycloheximide has no effect on origin distribution. Although ongoing general transcription is clearly required, active local transcription is not sufficient to restrict replication initiation sites to the intergenic spacer between the DHFR and 2BE2121 genes. An additional event that takes place in mid-G1 phase contributes to focusing of initiation to the DHFR ori-ß/{gamma} region.

Does transcription indeed play a role in replication origin specification?

The findings of the present study disagree with the conclusions reached in a recently published report (Keezer & Gilbert 2002). Using the same inhibitors ({alpha}-AM, AMD and DRB) and the same experimental system, Keezer and Gilbert concluded that transcription per se is not required for passage through the ODP, and is neither necessary nor sufficient for origin specification.

Two possible reasons could account for this discrepancy. First, these investigators used immunofluorescent labeling of BrU incorporated in vivo during abnormally long labeling times as the sole technique for evaluating RNA synthesis. As shown here, this approach can produce inconsistent results. Contrary to the claims made in the previous study, I found no evidence that the protein kinase inhibitor KT5720 acts as a global transcription inhibitor (Fig. 4E). Furthermore, the low doses of the classic transcription inhibitors administered by Keezer & Gilbert (2002) are unlikely to have the desired effect in vivo, since certain RNA polymerases (e.g. PolI) are insensitive to some of the inhibitors, whereas inhibition of other RNA polymerases (e.g. PolII and III) in live cells requires at least an order of magnitude higher drug concentrations (e.g. Figs 2 and 3 in this report and Lindell et al. 1970; Granick 1975; Tamm et al. 1980; Wieland & Faulstich 1991; Jackson et al. 1998; Custodio et al. 1999; Le Panse et al. 1999; Lee et al. 2004, among many others). Thus, the effects observed in the previous study are unlikely to have resulted from transcriptional inhibition. Second, my extensive experience with this system indicates that erratic results can be generated if the synchronized cells are not in good condition (Dimitrova et al. 2002). Indeed, indications of very poor physiological state of the cells used by Keezer & Gilbert (2002) can be found in their report (clumped cells, unusual chromatin texture with signs of highly condensed chromatin, shrunk nuclei with irregular and fragmented shape, etc., all suggestive of ongoing chromatin/nuclear disintegration).

In conclusion, through the use of optimal cell growth and synchronization conditions, as well as efficient methodologies for measurement of RNA synthesis and origin mapping, I demonstrate that ongoing nuclear transcription is required for the ODP transition in CHOC 400 cells.

Interplay between nuclear transcription and replication: a complex relationship

DNA replication and transcription can occur independent of each other. Transcription proceeds in the absence of DNA replication in quiescent cells or during the G1 and G2 phases in proliferating cells. Conversely, one (mammals) or more (flies and amphibians) early embryonic chromosome cycles take place naturally in transcriptionally silent cells (DePamphilis 1999). The results reported in the present study confirm that transcription is not required for DNA replication. Transcriptionally repressed and transcriptionally active G1 nuclei replicate with similar efficiencies when introduced into Xenopus egg extract. Yet, it is clear that whenever the two processes take place within the same nucleus, control of gene expression and regulation of DNA replication interconnect on multiple levels.

Functional mammalian pre-RCs are assembled in late telophase (Dimitrova et al. 1999, 2002), probably at many more sites than the number of active origins in any given S phase. The decision which of all potential pre-RCs should be utilized in the upcoming S phase is uncoupled from pre-RC assembly (Dimitrova et al. 2002) and is made later, during mid-G1 phase (Wu & Gilbert 1996; Dimitrova et al. 2002). The mechanism of origin choice is unknown. One possibility is that cell cycle-regulated synthesis of a novel gene product may be required after each mitosis to reprogram the spectrum of initiation sites to be utilized. Protein synthesis is not required for passage through the ODP (Fig. 4D; Keezer & Gilbert 2002) which implies that all proteins essential for origin specification are present in CHO cells from the onset of G1 phase. The results of the present study, however, leave open the intriguing possibility that the essential gene product might be an RNA molecule. In the past few years there has been an explosion of reports pointing to the important role of non-coding RNAs in genome structure and function (Storz 2002). One of the most fascinating functions of non-coding RNAs is their emerging role in chromatin organization (Bernstein & Allis 2005). Developmentally regulated intergenic transcription plays a role in chromatin remodeling and establishment of active domains in the ß-globin locus (Gribnau et al. 2000). An RNA component participates in maintenance of pericentric heterochromatin structure and it was found that chromodomains are protein–RNA interaction modules (reviewed in Bernstein & Allis 2005). At least some of the regulatory non-coding RNA genes are transcribed by PolII (Lee et al. 2004; Claverie 2005) and, as shown here, PolII-directed transcription is required for replication origin specification. Establishing whether replication proteins utilize non-coding RNAs as adapters for interactions with other chromatin components would be an exciting area of future research. Perhaps a non-coding RNA species produced during early G1 phase contributes to the stabilization of replication origin complexes at select chromosomal sites. Once formed, such complexes might not be strongly dependent on the sustained production of this RNA molecule(s). The observation that the maintenance of specific replication origins in the hamster DHFR gene locus during late-G1 phase is less sensitive to inhibition of transcription compared to their establishment in early G1 (Fig. 5) is consistent with a similar possibility.

I have previously speculated that the act of transcription might interfere with the stability of pre-RCs assembled within transcription units (Dimitrova & Gilbert 1999a). In mammalian cells, transcription shuts down during mitosis. Detectable RNA synthesis resumes at low levels during telophase in CHOC 400 nuclei (Fig. 6A,B), coincident with the assembly of functional pre-RCs on chromatin (Dimitrova et al. 2002). Within the next hour, global transcription rates reach a steady level that is maintained throughout most of G1 phase (Fig. 6B). Thus, during telophase, there is a window of opportunity to assemble pre-RCs uniformly along eukaryotic chromosomes. In transcriptionally silent cells (e.g. early embryonic cells) each pre-RC can potentially be used to initiate replication. By contrast, in transcriptionally active cells, passage of the transcription machinery might subsequently erase some of the pre-RCs as various genes become activated throughout G1 and S phase, effectively restricting initiation to intergenic regions. This model is consistent with the recent findings that initiation is confined to the intergenic region in the single-copy DHFR domain only within cells in which the DHFR gene promoter is active (Saha et al. 2004). Furthermore, failure to terminate DHFR transcription at the natural terminator results in significant shrinkage of the DHFR intergenic replication initiation zone (Mesner & Hamlin 2005).

My results, however, suggest that the assumption that the mere binding of transcription factors to DNA, or the presence of a functional promoter are sufficient to specify locally a replication origin in mammalian nuclei may be too simplistic. DRB inhibition leaves gene promoters active (associated with transcription factors and engaged RNA polymerase complexes), yet initiation in the DHFR locus is random under these conditions (Fig. 4C). Furthermore, DHFR origin specification takes place during a G1 phase interval when no changes in transcription activity in this locus occur (Fig. 6). It is not known what percentage of DHFR and 2BE2121 gene copies are actively transcribed in CHOC 400 cells. Given that similar intensity of hybridization signals was obtained in the run-on assays with probes from the 500-fold amplified DHFR gene and the single-copy APRT gene (Fig. 7D), it is likely that a very small fraction of the DHFR gene copies are transcriptionally active. Interestingly, there is evidence that in CHOC 400 cells replication origins fire within only ~10% of the DHFR amplicons in a single cell cycle (Pemov et al. 1998). Thus, it remains possible that the active replication origins are found within the transcriptionally active DHFR amplicons, similar to the yeast rDNA replication origins (Muller et al. 2000) and, within these 10% of amplicons, the pre-RCs become excluded from the genes and dispersed throughout the intergenic region. During the pre-ODP stage, this could be masked by random initiation in the overwhelming majority of transcriptionally inactive DHFR amplicons. It is clear, however, that an independent mid-G1-phase event leads to a significant redistribution of potential initiation sites in most, if not all, amplicons to a narrower region in the intergenic spacer, encompassing the major DHFR origin sites ori-ß and ori-{gamma}. The evidence presented here demonstrates that ongoing general nuclear transcription is required for this event. Local gene transcription, however, is not sufficient for origin specification at individual chromosomal loci. Transcription may be just one of several factors that collaborate to determine the replication-relevant features of a chromosomal domain.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Chemicals

AMD, {alpha}-AM, DRB, CHX and KT5720 were obtained from Sigma. Aph was from Calbiochem. Drug concentrations needed to inhibit transcription were found to differ significantly between in vivo and in vitro treatments and are described in detail in the text.

Cell culture, synchronization and labeling

CHOC 400 cells (a cell line, in which the DHFR gene locus has been amplified ~500-fold) were cultured and synchronized as previously described (Dimitrova & Gilbert 1998; Dimitrova et al. 1999). Labeling of nascent DNA with BrdU was performed following published procedures (Dimitrova & Gilbert 1999b; Dimitrova 2006). Labeling of proteins with 35S-Met was performed as described in (Ausubel et al. 2003).

Analysis of DNA replication in Xenopus egg extracts

Activated low speed Xenopus egg extracts (LSS) were prepared as described (Dimitrova & Gilbert 1998). Rates and extent of DNA replication with CHOC 400 nuclei were evaluated by measuring the amount of acid-precipitable [{alpha}-32P]dATP (Dimitrova & Gilbert 1998). Specificity of initiation within the DHFR locus in CHO G1 nuclei was measured by the ELFH assay as described (Gilbert et al. 1995; Dimitrova & Gilbert 1998).

Transcription assays

I. Indirect immunofluorescent labeling of nuclear transcription sites
Cells were incubated with 10–100 µM BrU (Sigma) or 10–100 µM FU (Sigma) for 10–15 min to pulse-label nascent RNA in vivo. Increasing the BrU (or FU) concentration to 1–5 mM (Iborra et al. 1998) did not change the outcome of the experiments. Cells were fixed for 10 min either with 2–4% formaldehyde in PBS, or with cold methanol. Several anti-BrdU antibodies were tested for indirect immunofluorescent labeling. The only combination of reagents found to work consistently was labeling of RNA with FU followed by detection with the mouse B-2531 (Sigma) anti-BrdU antibody, as described in (Boisvert et al. 2000). Appropriate Alexa-conjugated secondary antibodies (Molecular Probes) were used for immunofluorescent detection of the primary antibodies. DNA was stained with 0.1 µg/mL 4',6-diamidino-2-phenylindole (DAPI). Microscopy was performed as described (Dimitrova & Gilbert 1999b; Dimitrova & Berezney 2002).

II. Measurements of transcription activity
Nuclear run-on transcription reactions were performed with 2.5–10 x 105 intact nuclei and 25–100 µCi [{alpha}-32P]UTP (New England Nuclear) at 21 °C following published procedures (Ausubel et al. 2003). Total RNA was purified as per (Chomczynski & Sacchi 1987) and hybridized to a panel of unique DNA probes immobilized on nylon membranes (Hybond N+, Amersham; 1 µg DNA per slot). The probes cover ~120 kb region of the CHO DHFR locus (Gilbert et al. 1995). The genomic positions of some of these probes were wrongfully described in a previous report (Lawlis et al. 1996). The correct positions are: probe T, a 1.5-kb fragment, containing the promoter and exons I and II of the DHFR gene; probe N, a 420-bp fragment from the third intron of the DHFR gene; probe S, a 1.8 kb fragment from the fourth intron of the DHFR gene. Plasmid pH2, carrying the Chinese hamster APRT gene (Kessler & Chasin 1996), was a kind gift of Dr L. Chasin (Columbia University, New York). Relative c.p.m. were obtained by phosphorimaging analysis (Molecular Dynamics).

For run-on assays following incubation in Xenopus egg extract, intact nuclei were first incubated at 10 000 per µL of LSS extract for 45 min at 21 °C in the presence of 100 µg/mL Aph (ELFH conditions). After three washes with cold hypotonic buffer (Dimitrova & Gilbert 1998), nuclei were resuspended in transcription cocktail supplemented with [{alpha}-32P]UTP and elongation of preinitiated transcripts was allowed to proceed for 20 min at 21 °C. Total RNA was purified and hybridized to the same DNA probes as above.

For analysis of transcription rates, nuclei were resuspended in either Xenopus egg extract, or in transcription cocktail supplemented with [{alpha}-32P]UTP and incubated at 21 °C. Aliquots were removed at various times thereafter and the amount of 32P-UTP incorporated into RNA was determined by acid precipitation. Calculations of the amount of nascent RNA synthesized in Xenopus egg extracts were based on 1 mM average UTP concentration (Woodland & Pestell 1972).


   Acknowledgements
 
Dr L. Chasin's gift of plasmid pH2 is gratefully acknowledged. I wish to thank all colleagues from the nuclear organization field who shared with me their experience with visualization of nuclear transcription sites, and especially Drs M. Hendzel and R. Berezney for their problem-solving advice. Financial support was provided by Hendricks Fund for Medical Research postdoctoral fellowship.


   Footnotes
 
Communicated by: Susan M. Gasser

aPresent address: The Babraham Institute, Cambridge, CB2 4AT UK Back

* Correspondence: E-mail: daniela.dimitrova{at}bbsrc.ac.uk


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Ausubel, F.M., Brent, R., Kingston, R.E., et al. (eds) (2003) Current Protocols in Molecular Biology. New York: John Wiley & Sons, Inc.

Bendig, M.M. & Williams, J.G. (1983) Replication and expression of Xenopus laevis globin genes injected into fertilized Xenopus eggs. Proc. Natl. Acad. Sci. USA 80, 6197–6201.[Abstract/Free Full Text]

Bernstein, E. & Allis, C.D. (2005) RNA meets chromatin. Genes Dev. 19, 1635–1655.[Abstract/Free Full Text]

Boisvert, F.M., Hendzel, M.J. & Bazett-Jones, D.P. (2000) Promyelocytic leukemia (PML) nuclear bodies are protein structures that do not accumulate RNA. J. Cell Biol. 148, 283–292.[Abstract/Free Full Text]

Chen, S., Reger, R., Miller, C. & Hyman, L.E. (1996) Transcriptional terminators of RNA polymerase II are associated with yeast replication origins. Nucleic Acids Res. 24, 2885–2893.[Abstract/Free Full Text]

Chomczynski, P. & Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159.[Medline]

Claverie, J.M. (2005) Fewer genes, more noncoding RNA. Science 309, 1529–1530.[Abstract/Free Full Text]

Crouse, G.F., Leys, E.J., McEwan, R.N., Frayne, E.G. & Kellems, R.E. (1985) Analysis of the mouse dhfr promoter region: existence of a divergently transcribed gene. Mol. Cell. Biol. 5, 1847–1858.[Abstract/Free Full Text]

Custodio, N., Carmo-Fonseca, M., Geraghty, F., Pereira, H.S., Grosveld, F. & Antoniou, M. (1999) Inefficient processing impairs release of RNA from the site of transcription. EMBO J. 18, 2855–2866.[CrossRef][Medline]

Danis, E., Brodolin, K., Menut, S., Maiorano, D., Girard-Reydet, C. & Mechali, M. (2004) Specification of a DNA replication origin by a transcription complex. Nat. Cell Biol. 6, 721–730.[CrossRef][Medline]

DePamphilis, M.L. (1999) Replication origins in metazoan chromosomes: fact or fiction? Bioessays 21, 5–16.[CrossRef][Medline]

Diffley, J.F., Cocker, J.H., Dowell, S.J., Harwood, J. & Rowley, A. (1995) Stepwise assembly of initiation complexes at budding yeast replication origins during the cell cycle. J. Cell Sci. Suppl. 19, 67–72.[Medline]

Dimitrova, D.S. (2006) Single labelling of nascent DNA with halogenated thymidine analogues. In: Cell Biology Protocols (eds J.R. Harris, J.M. Graham & D. Rickwood), pp. 6.2. Chichester: John Wiley & Sons Ltd.

Dimitrova, D.S. & Berezney, R. (2002) The spatio-temporal organization of DNA replication sites is identical in primary, immortalized and transformed mammalian cells. J. Cell Sci. 115, 4037–4051.[Abstract/Free Full Text]

Dimitrova, D.S. & Gilbert, D.M. (1998) Regulation of mammalian replication origin usage in Xenopus egg extract. J. Cell Sci. 111, 2989–2998.[Abstract]

Dimitrova, D.S. & Gilbert, D.M. (1999a) DNA replication and nuclear organization: prospects for a soluble in vitro system. Crit. Rev. Eukaryot. Gene Expr. 9, 353–361.[Medline]

Dimitrova, D.S. & Gilbert, D.M. (1999b) The spatial position and replication timing of chromosomal domains are both established in early G1 phase. Mol. Cell 4, 983–993.[CrossRef][Medline]

Dimitrova, D.S., Prokhorova, T.A., Blow, J.J., Todorov, I.T. & Gilbert, D.M. (2002) Mammalian nuclei become licensed for DNA replication during late telophase. J. Cell Sci. 115, 51–59.[Abstract/Free Full Text]

Dimitrova, D.S., Todorov, I.T., Melendy, T. & Gilbert, D.M. (1999) Mcm2, but not RPA, is a component of the mammalian early G1-phase prereplication complex. J. Cell Biol. 146, 709–722.[Abstract/Free Full Text]

Gilbert, D.M., Miyazawa, H. & DePamphilis, M.L. (1995) Site-specific initiation of DNA replication in Xenopus egg extract requires nuclear structure. Mol. Cell. Biol. 15, 2942–2954.[Abstract]

Granick, D. (1975) Nucleolar necklaces in chick embryo fibroblast cells. I. Formation of necklaces by dichlororibobenzimidazole and other adenosine analogues that decrease RNA synthesis and degrade preribosomes. J. Cell Biol. 65, 398–417.[Abstract/Free Full Text]

Gribnau, J., Diderich, K., Pruzina, S., Calzolari, R. & Fraser, P. (2000) Intergenic transcription and developmental remodeling of chromatin subdomains in the human ß-globin locus. Mol. Cell 5, 377–386.[CrossRef][Medline]

Haase, S.B., Heinzel, S.S. & Calos, M.P. (1994) Transcription inhibits the replication of autonomously replicating plasmids in human cells. Mol. Cell. Biol. 14, 2516–2524.[Abstract/Free Full Text]

Hall, L.L., Byron, M., Sakai, K., Carrel, L., Willard, H.F. & Lawrence, J.B. (2002) An ectopic human XIST gene can induce chromosome inactivation in postdifferentiation human HT-1080 cells. Proc. Natl. Acad. Sci. USA 99, 8677–8682.[Abstract/Free Full Text]

Iborra, F.J., Jackson, D.A. & Cook, P.R. (1998) The path of transcripts from extra-nucleolar synthetic sites to nuclear pores: transcripts in transit are concentrated in discrete structures containing SR proteins. J. Cell Sci. 111, 2269–2282.[Abstract]

Jackson, D.A., Hassan, A.B., Errington, R.J. & Cook, P.R. (1993) Visualization of focal sites of transcription within human nuclei. EMBO J. 12, 1059–1065.[Medline]

Jackson, D.A., Iborra, F.J., Manders, E.M. & Cook, P.R. (1998) Numbers and organization of RNA polymerases, nascent transcripts, and transcription units in HeLa nuclei. Mol. Biol. Cell 9, 1523–1536.[Abstract/Free Full Text]

Kedinger, C. & Simard, R. (1974) The action of {alpha}-amanitin on RNA synthesis in Chinese hamster ovary cells. Ultrastructural and biochemical studies. J. Cell Biol. 63, 831–842.[Abstract/Free Full Text]

Keezer, S.M. & Gilbert, D.M. (2002) Sensitivity of the Origin Decision Point to specific inhibitors of cellular signaling and metabolism. Exp. Cell Res. 273, 54–64.[CrossRef][Medline]

Kessler, O. & Chasin, L.A. (1996) Effects of nonsense mutations on nuclear and cytoplasmic adenine phosphoribosyltransferase RNA. Mol. Cell. Biol. 16, 4426–4435.[Abstract]

Kikyo, N., Wade, P.A., Guschin, D., Ge, H. & Wolffe, A.P. (2000) Active remodeling of somatic nuclei in egg cytoplasm by the nucleosomal ATPase ISWI. Science 289, 2360–2362.[Abstract/Free Full Text]

Lawlis, S.J., Keezer, S.M., Wu, J.R. & Gilbert, D.M. (1996) Chromosome architecture can dictate site-specific initiation of DNA replication in Xenopus egg extracts. J. Cell Biol. 135, 1207–1218.[Abstract/Free Full Text]

Le Panse, S., Masson, C., Heliot, L., Chassery, J.M., Junera, H.R. & Hernandez-Verdun, D. (1999) 3-D organization of ribosomal transcription units after DRB inhibition of RNA polymerase II transcription. J. Cell Sci. 112, 2145–2154.[Abstract]

Lee, Y., Kim, M., Han, J., et al. (2004) MicroRNA genes are transcribed by RNA polymerase II. EMBO J. 23, 4051–4060.[CrossRef][Medline]

Li, R., Yu, D.S., Tanaka, M., Zheng, L., Berger, S.L. & Stillman, B. (1998) Activation of chromosomal DNA replication in Saccharomyces cerevisiae by acidic transcriptional activation domains. Mol. Cell. Biol. 18, 1296–1302.[Abstract/Free Full Text]

Lindell, T.J., Weinberg, F., Morris, P.W., Roeder, R.G. & Rutter, W.J. (1970) Specific inhibition of nuclear RNA polymerase II by {alpha}-amanitin. Science 170, 447–449.[Abstract/Free Full Text]

Maderious, A. & Chen-Kiang, S. (1984) Pausing and premature termination of human RNA polymerase II during transcription of adenovirus in vivo and in vitro. Proc. Natl. Acad. Sci. USA 81, 5931–5935.[Abstract/Free Full Text]

Marshall, N.F. & Price, D.H. (1992) Control of formation of two distinct classes of RNA polymerase II elongation complexes. Mol. Cell. Biol. 12, 2078–2090.[Abstract/Free Full Text]

Masson, C., Bouniol, C., Fomproix, N., Szollosi, M.S., Debey, P. & Hernandez-Verdun, D. (1996) Conditions favoring RNA polymerase I transcription in permeabilized cells. Exp. Cell Res. 226, 114–125.[CrossRef][Medline]

Mechali, M. (2001) DNA replication origins: from sequence specificity to epigenetics. Nat. Rev. Genet. 2, 640–645.[Medline]

Mesner, L.D. & Hamlin, J.L. (2005) Specific signals at the 3' end of the DHFR gene define one boundary of the downstream origin of replication. Genes Dev. 19, 1053–1066.[Abstract/Free Full Text]

Mitchell, P.J., Carothers, A.M., Han, J.H., et al. (1986) Multiple transcription start sites, DNase I-hypersensitive sites, and an opposite-strand exon in the 5' region of the CHO dhfr gene. Mol. Cell. Biol. 6, 425–440.[Abstract/Free Full Text]

Muller, M., Lucchini, R. & Sogo, J.M. (2000) Replication of yeast rDNA initiates downstream of transcriptionally active genes. Mol. Cell 5, 767–777.[Medline]

Newport, J. & Kirschner, M. (1982) A major developmental transition in early Xenopus embryos: II. Control of the onset of transcription. Cell 30, 687–696.[CrossRef][Medline]

Pemov, A., Bavykin, S. & Hamlin, J.L. (1998) Attachment to the nuclear matrix mediates specific alterations in chromatin structure. Proc. Natl. Acad. Sci. USA 95, 14757–14762.[Abstract/Free Full Text]

Perry, R.P. & Kelley, D.E. (1970) Inhibition of RNA synthesis by actinomycin D: characteristic dose–response of different RNA species. J. Cell Physiol. 76, 127–139.[CrossRef][Medline]

Prescott, D.M. & Bender, M.A. (1962) Synthesis of RNA and protein during mitosis in mammalian tissue culture cells. Exp. Cell Res. 26, 260–268.[CrossRef][Medline]

Prioleau, M.N., Huet, J., Sentenac, A. & Mechali, M. (1994) Competition between chromatin and transcription complex assembly regulates gene expression during early development. Cell 77, 439–449.[CrossRef][Medline]

Rowles, A., Tada, S. & Blow, J.J. (1999) Changes in association of the Xenopus Origin Recognition Complex with chromatin on licensing of replication origins. J. Cell Sci. 112, 2011–2018.[Abstract]

Saha, S., Shan, Y., Mesner, L.D. & Hamlin, J.L. (2004) The promoter of the Chinese hamster ovary dihydrofolate reductase gene regulates the activity of the local origin and helps define its boundaries. Genes Dev. 18, 397–410.[Abstract/Free Full Text]

Schubeler, D., MacAlpine, D.M., Scalzo, D., et al. (2004) The histone modification pattern of active genes revealed through genome-wide chromatin analysis of a higher eukaryote. Genes Dev. 18, 1263–1271.[Abstract/Free Full Text]

Shabalina, S.A., Ogurtsov, A.Y., Kondrashov, V.A. & Kondrashov, A.S. (2001) Selective constraint in intergenic regions of human and mouse genomes. Trends Genet. 17, 373–376.[CrossRef][Medline]

Sobell, H.M. (1985) Actinomycin and DNA transcription. Proc. Natl. Acad. Sci. USA 82, 5328–5331.[Abstract/Free Full Text]

Storz, G. (2002) An expanding universe of noncoding RNAs. Science 296, 1260–1263.[Abstract/Free Full Text]

Tamm, I., Hand, R. & Caliguiri, L.A. (1976) Action of dichlorobenzimidazole riboside on RNA synthesis in L-929 and HeLa cells. J. Cell Biol. 69, 229–240.[Abstract/Free Full Text]

Tamm, I., Kikuchi, T., Darnell, J.E. Jr & Salditt-Georgieff, M. (1980) Short capped hnRNA precursor chains in HeLa cells: continued synthesis in the presence of 5,6-dichloro-1-ß-D-ribofuranosylbenzimidazole. Biochemistry 19, 2743–2748.[CrossRef][Medline]

Wansink, D.G., Schul, W., van der Kraan, I., van Steensel, B., van Driel, R. & de Jong, L. (1993) Fluorescent labeling of nascent RNA reveals transcription by RNA polymerase II in domains scattered throughout the nucleus. J. Cell Biol. 122, 283–293.[Abstract/Free Full Text]

Wieland, T. & Faulstich, H. (1991) Fifty years of amanitin. Experientia 47, 1186–1193.[CrossRef][Medline]

Woodland, H.R. & Pestell, R.Q. (1972) Determination of the nucleoside triphosphate contents of eggs and oocytes of Xenopus laevis. Biochem. J. 127, 597–605.[Medline]

Wu, J.R. & Gilbert, D.M. (1996) A distinct G1 step required to specify the Chinese hamster DHFR replication origin. Science 271, 1270–1272.[Abs