Full Text

Turn on search term navigation

ARTICLE

Received 7 Mar 2016 | Accepted 31 Aug 2016 | Published 11 Oct 2016

Cancer is a disease associated with genomic instability that often results from oncogene activation. This in turn leads to hyperproliferation and replication stress. However, the molecular mechanisms that underlie oncogene-induced replication stress are still poorly understood. Oncogenes such as HRASV12 promote proliferation by upregulating general transcription factors to stimulate RNA synthesis. Here we investigate whether this increase in transcription underlies oncogene-induced replication stress. We show that in cells overexpressing HRASV12, elevated expression of the general transcription factor TATA-box binding protein (TBP) leads to increased RNA synthesis, which together with R-loop accumulation results in replication fork slowing and DNA damage. Furthermore, overexpression of TBP alone causes the hallmarks of oncogene-induced replication stress, including replication fork slowing, DNA damage and senescence. Consequently, we reveal that increased transcription can be a mechanism of oncogene-induced DNA damage, providing a molecular link between upregulation of the transcription machinery and genomic instability in cancer.

DOI: 10.1038/ncomms13087 OPEN

Increased global transcription activity as a mechanism of replication stress in cancer

Panagiotis Kotsantis1, Lara Marques Silva2, Sarah Irmscher2, Rebecca M. Jones1, Lisa Folkes3, Natalia Gromak2 & Eva Petermann1

1 Institute of Cancer and Genomic Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. 2 Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE, UK. 3 Department of Oncology, CRUK/MRC Oxford Institute for Radiation Oncology, University of Oxford, Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ, UK. Correspondence and requests for materials should be addressed to N.G. (email: mailto:[email protected]

Web End [email protected] ) or to E.P. (email: mailto:[email protected]

Web End [email protected] ).

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087

Cancer is a disease of genomic instability, characterized by high mutation rates and genomic rearrangements that ultimately drive aggressiveness and resistance to therapy1.

One of the mechanisms proposed to cause genomic instability in cancer is replication stress, which occurs when DNA replication fork progression in S phase slows or stalls. This leads to collapse of forks into DNA double-strand breaks (DSBs), as well as incomplete sister chromatid separation in the following mitosis2. Markers of spontaneous replication stress are found in tumour samples and cells expressing active oncogenes, and replication stress promotes chromosomal instability, the most common form of genomic instability in sporadic cancers36. Spontaneous replication stress is therefore increasingly regarded as a central feature of cancer cells and there is much interest in specically targeting this phenotype for cancer therapy7. However, progress in this eld is hindered, because the molecular mechanisms underlying spontaneous replication stress in cells are still largely unknown. This impairs our ability to investigate replication stress in vitro and in vivo, and to identify potential biomarkers or therapeutic targets.

How can mechanisms of spontaneous replication stress be identied? The overexpression of oncogenes such as RAS, MOS, MYC, CDC25A or CYCLIN E is sufcient to induce replication stress in cultured cells4,5,811. These oncogenes all act in the growth factor signalling pathways that stimulate proliferation by promoting cell growth and division. Molecular changes associated with increased proliferation are therefore prime candidates for causing replication stress. Indeed, we and others have reported that CYCLIN E-induced replication stress results from accelerated S-phase entry and increased replication initiation during S phase9,12. So far, however, little attention has been paid to the fact that oncogenes such as RAS and MYC do not only activate the cell cycle machinery but also promote cell growth through the activation of transcription and protein translation1316. Overexpressed c-MYC acts as a universal amplier, stimulating transcription by all three RNA polymerases14,17,18. Oncogenic RAS promotes transcription through the mitogen-activated protein kinase extracellular signal-regulated kinase (ERK), which activates transcription factors such as TIFIA (RRN3), UBTF, TIFIIIB (BRF1) and TATA-box binding protein (TBP)14,15,19, and can also promote transcription through other factors such as TERT20. Moreover, components of the general transcription machinery itself are found mutated, differentially expressed and deregulated across a variety of cancers14,21.

Upregulation of transcription in cancer cells has the potential to be a direct cause of replication stress as interference between transcription and replication leads to replication fork slowing and genomic instability22,23. This can result from direct collisions or torsional stress between the two active protein complexes24. Another important source of replication stress is the collision of the replication machinery with RNADNA hybrids (R-loops) where the nascent RNA has re-annealed with the template25. Deregulated transcription is a common feature of cancers, but its importance for replication stress and consequent genomic instability has not been examined.

Here we use HRASV12 overexpression5,26 to investigate whether the oncogene-induced increase in transcription is a mechanism of endogenous replication stress. We report that increased transcription activity in cells expressing HRASV12 causes replication stress, as high levels of RNA synthesis and transcription intermediates interfere with replication fork progression. Replication stress in these cells depends on R-loop accumulation and on increased expression of the general transcription factor TBP. Importantly, overexpression of TBP alone induces replication stress and genomic instability. Our data

suggest that increased transcription activity is a mechanism contributing to replication stress in cancer.

ResultsHRASV12 increases nascent RNA synthesis and R-loop formation. To investigate whether increased transcription activity contributes to oncogene-induced replication stress, we used immortalized human broblasts that have been stably transfected with pBabe-HRASV12-ERTAM, to express tamoxifen-inducible HRASV12 (BJ-hTert HRASV12ER-TAM)27. HRASV12-ERTAM has been well characterized as a system for inducing oncogenic RAS26,28,29. Addition of 4-hydroxytamoxifen (4OHT) led to HRASV12 accumulation, which activated mitogen-activated protein kinase signalling as evidenced by ERK1/2 phosphorylation (Fig. 1a). Cells proliferated slightly faster for up to 6 days followed by growth arrest, which was previously shown to result from DNA damage-induced apoptosis and senescence4,5 (Supplementary Fig. 1ae). When using immortalized BJ broblasts that were not expressing 4OHT-inducible HRASV12, 4OHT treatment did not affect any of the phenotypes investigated in this study (Supplementary Fig. 1fi).

First, we investigated the effect of oncogenic HRAS on global transcription activity. For this, we quantied nascent RNA synthesis using nuclear incorporation of the modied RNA precursor 5-ethynyluridine (EU) for 1 h (Fig. 1b,c). RNA synthesis increased rapidly after HRASV12 induction and was elevated more than twofold after 48 and 72 h HRASV12 induction (Fig. 1d). RNA synthesis after HRASV12 induction was higher than the highest activity observed in control cells (Fig. 1d), arguing against changes in cell cycle distribution as the sole explanation for the increase. Seventy-two hours of HRASV12 induction was used for most subsequent experiments.

We next used the S9.6 antibody that detects RNA/DNA hybrids30,31, to test whether R-loop formation was increased in HRASV12-overexpressing cells. First, we performed slot blot analysis of isolated genomic DNA32, which revealed a threefold increase in R-loops after 72 h HRASV12 induction (Fig. 1e,f). The S9.6 signal could be removed by treatment with recombinant RNase H, supporting that it was specic to R-loops. We then used S9.6 immunoprecipitation of RNA/DNA hybrids from cells (DNA immunoprecipitation (DIP))33 after 72 h HRASV12 induction, to investigate the distribution of R-loops across RAS target versus control genes (Fig. 2af). We used quantitative PCR (qPCR) to quantify R-loop distribution on DUSP6, SPRY2 and C-FOS, genes that are upregulated by activated RAS (Fig. 2ac and Supplementary Fig. 2a). We observed an increased R-loop formation on the promoter-proximal and selected intron regions of all three genes. DIP analysis of C-FOS showed that in line with previously described R-loop accumulation in actively transcribed genes33, R-loops were signicantly increased over the transcribed regions of the gene (Fig. 2c and also see Supplementary Table 1 for PCR primer sequences). We also quantied R-loop formation on non-RAS target control genes GAPDH1, ACTB (b-ACTIN)

and ACTG (g-ACTIN). We observed no increase in R-loops across any of these genes (Fig. 2df). RNase H treatment conrmed that DIP specically detected R-loops (Fig. 2af). RNase A treatment conrmed that DIP signal was not due to annealing of free RNA species to DNA during sample preparation or to S9.6 antibody recognizing double-stranded RNA (Supplementary Fig. 2be). These data support that stimulation of transcription by HRASV12 results in increased R-loop formation.

HRASV12 causes replication stress and G1 53BP1 foci. HRASV12 overexpression caused replication fork slowing after

2 NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087 ARTICLE

48 h 72 h

4OHT: + + +

HRAS

24 h

kDa 58

EU 1 h EU 1 h EU 1 h

+/4OHT

24 h 48 h 72 h

pERK1 pERK2

ERK1 ERK2

-ACTIN

Control Ras

100

DAPI EU

RNA synthesis (a.u)

17.4 3.1

25.82.9 17.9

0.9

41.7 3.9

22.9 0.2

41.3 1.4

0 con Ras con Ras con Ras

24 h 48 h 72 h

e Genomic DNA

RNase H

1/1

Control

Ras

1/2

1/4

1/8

1/16

S9.6 signal (normalized)

***

Control

Ras

S9.6 (R-loop)

ssDNA

RNase H

0 + +

Ras

Control

Figure 1 | HRASV12 overexpression increases transcription activity. (a) Protein levels of HRAS, pERK1/2, ERK1/2 and b-ACTIN in BJ-HRASV12 cells after RAS induction for the times indicated. (b) EU incorporation (1 h) was used to measure nascent RNA synthesis after RAS induction for the times indicated. (c) Representative images of EU staining (red) after RAS induction for 72 h. (d) Quantication of nuclear EU intensity after RAS induction for the times indicated. N 4 (48 and 72 h), N 5 (24 h). (e) RNA/DNA hybrid slot blot of genomic DNA from BJ-HRASV12 cells after RAS induction for 72 h,

RNase H. S9.6 antibody was used to detect RNA/DNA hybrids (top panel) with single-strand DNA antibody (bottom panel) as a loading control. Serial dilutions of genomic DNA (1/1 4 mg) were probed with S9.6 antibody for standards (left panel). (f) Fold enrichment in RNA/DNA hybrids

compared with control. N 3. Means s.e.m. (bars) are shown. Students t-test, *Po0.05, **Po0.01 and ***Po0.001. Scale bars, 10 mm.

48 h, indicative of replication stress (Fig. 3ac) and induced phosphorylation of replication stress response factors RPA32 (serine 33) and CHK1 (serine 345) (Fig. 3d,e). These modications were already evident at 24 h HRASV12 induction, which may also reect the proposed role for RPA and ATR in the transcription stress response34. As expected, fork slowing was associated with nuclear foci formation of DNA damage markers gH2AX and 53BP1 (Fig. 3f,g). One consequence of replication fork slowing is mitotic entry with under-replicated DNA, which leads to micronuclei formation and appearance of 53BP1 bodies in the following G1 phase35,36. Accordingly, 53BP1 foci increased after 4 days HRASV12 induction and were mostly found in G1 cells (Fig. 3h). Similarly, micronuclei formation peaked at 4 days HRASV12 induction (Fig. 3i). This suggests that HRASV12-induced DNA damage requires mitotic progression, as was previously reported for CYCLIN E10. Four days of HRASV12 induction was used for subsequent investigation of 53BP1 foci formation.

Transcription promotes HRASV12-induced replication stress. To test whether increased transcription causes replication stress in cells harbouring HRASV12, we transiently inhibited RNA synthesis using small molecule inhibitors before measuring replication fork progression. To minimize effects on gene expression, incubations were kept short at 100 min for triptolide and 5,6-dichloro-1-b-D-ribofuranosyl-1H-benzimidazole (DRB)

and 4 h for a-amanitin (Fig. 4a). These treatments inhibited ongoing RNA synthesis (Fig. 4b) but did not affect protein levels of HRASV12 or levels of proteins with short half lives such as CYCLIN B1 and p53 (Supplementary Fig. 3a,b). All three transcription inhibitors increased replication fork speeds specically in the presence of HRASV12 (Fig. 4cf). Although such short incubations with transcription inhibitors may not be sufcient to reverse all effects of transcription, these data suggest that RAS-induced replication stress is promoted by active RNA synthesis. DRB and triptolide rescued replication more effectively than a-amanitin, suggesting that effective inhibition of early

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087

stages of transcription removes more obstacles to replication forks than does inhibiting transcription during elongation (Fig. 4f).

We previously observed that increased CDK activity and new origin ring underlies replication stress in cells overexpressing CYCLIN E12. In contrast, inhibiting new origin ring using CDK inhibitor roscovitine could not rescue HRASV12-induced replication fork slowing (Supplementary Fig. 3ce). Together, our data suggest that replication stress in cells expressing HRASV12 is promoted by transcription but not by CDK activity.

To test whether HRASV12-induced DNA damage was transcription dependent, we rst incubated cells with DRB for 100 min and stained for 53BP1 24 h later. DRB reduced 53BP1 foci formation; however, its impact was limited by the short incubation time (Supplementary Fig. 3f,g). We therefore used gH2AX chromatin immunoprecipitation (ChIP) to test whether transcription and R-loop formation was associated with DNA damage in cells harbouring HRASV12. Indeed, we observed an increase in gH2AX signal over the transcribed promoter-proximal region of the C-FOS gene, correlating with strong induction of R-loops, 72 h after HRASV12 induction (Figs 2c and 4g). This gH2AX induction was replication dependent, as it could be prevented by blocking replication with Aphidicolin (Fig. 4g and Supplementary Fig. 1d). In contrast, we did not detect an increase in replication-dependent gH2AX levels across the intron 1 region of the b-ACTIN gene (Fig. 4h). This suggests that HRASV12 triggers R-loop-associated DNA damage that also depends on replication.

R-loops promote HRASV12-induced replication stress. We next decided to further investigate the role of R-loops in HRASV12-induced replication stress. We used transient transfection to express green uorescent protein (GFP)-tagged human RNaseH1, an enzyme that degrades RNA/DNA hybrids on overexpression37

(Fig. 5a,b). Interestingly, we observed that protein and messenger RNA levels of endogenous RNaseH1 were elevated in cells overexpressing HRASV12, suggesting an increased requirement for R-loop processing activities (Fig. 5b,c). The specicity of RNaseH1 antibody was veried using small interfering RNA (siRNA) depletion of RNaseH1 (Supplementary Fig. 4a). Overexpression of GFP-RNaseH1 reduced R-loop levels in the nucleus, as indicated by S9.6 immunostaining (Fig. 5d,e and also see Supplementary Fig. 5 for validation of immunostaining method). As the expression construct contains the RNaseH1 mitochondrial targeting sequence, mitochondrial R-loops were also reduced (Fig. 5d). RNaseH1 overexpression effectively improved replication fork progression in cells harbouring HRASV12 (Fig. 5f,g). GFP-RNaseH1 also decreased HRASV12 induction of 53BP1 foci (Fig. 5h,i) and DSB induction measured by pulse-eld gel electrophoresis (Supplementary Fig. 4b,c). GFP-RNaseH1 overexpression did not affect hydroxyurea-induced 53BP1 foci formation or the cell cycle prole (Supplementary Fig. 4d,e). These data support that HRASV12-induced replication stress is promoted by the presence of R-loops.

We also tested whether some of the replication stress caused by HRASV12 might be due to ribonucleotide (NTP) depletion as a result of increased RNA synthesis. Supplementing growth medium with ribonucleosides improved fork speeds and reduced 53BP1 foci formation in the presence of HRASV12 (Supplementary Fig. 6ac). However, HPLC quantication showed no reduction in NTP levels or dNTP levels as detectable (Supplementary Fig. 6d,e). This is in line with previous reports demonstrating that nucleoside supplementation can rescue fork progression even in the absence of detectable nucleotide depletion38. Our data thus suggest that transcription contributes to HRASV12-induced replication stress predominantly via direct conicts with transcription machinery and R-loops (Fig. 5j).

DUSP6: R-loop DIP b

SPRY2: R-loop DIP

C-FOS: R-loop DIP

-ACTIN: R-loop DIP

2 3 4

prom in1A in1B

prom in1

Intergenic region

4.0%

1.0%

5.0%

* *

** **

ControlRNase HRasRNase H + Ras

0.8%

3.0%

***

ControlRNase HRasRNase H + Ras

4.0%

R-loops % input

3.0%

0.6%

2.0%

R-loops % input

2.0%

0.4%

1.0%

0.2%

0.0%

prom in1

prom in1A in1B

cFos2 cFos3 cFos4

Intergenic region

d e f

GAPDH1: R-loop DIP

-ACTIN: R-loop DIP

prom in1

7.0%

NS NS

5.0%

ControlRNase HRasRNase H + Ras

6.0%

5.0%

ControlRNase HRasRNase H + Ras

R-loops % input

ControlRNase HRasRNase H + Ras

4.0%

5.0%

R-loops % input

4.0%

3.0%

4.0%

3.0%

2.0%

1.0%

0.0%

prom in1

5'prom in1

Figure 2 | HRASV12 overexpression increases R-loop formation. DIP analysis of R-loop induction on the DUSP6 (a), SPRY2 (b), C-FOS (c), GAPDH (d), g-ACTIN (e) and b-ACTIN (f) genes in BJ-HRASV12 cells after RAS induction for 72 h. Intergenic region upstream of C-FOS gene (c) served as a background control. Values are percentage of input. N 5. Primer positions are shown in upper panels of af (white boxes represent exons). Means

s.e.m. (bars) are shown. Students t-test, *Po0.05, **Po0.01 and ***Po0.001. NS, nonsignicant.

4 NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087 ARTICLE

48 h

24 h 72 h

48 h

72 h

CldU IdU +/4OHT

24 h 48 h 72 h

Control Ras

20 20

4OHT:

24 h + + +

+ + +

kDa

RPA-pS33 30

RPA

GAPDH

Control Ras

* **

% Forks

Median fork speed (kb min1 )

0.8

4OHT: HU kDa

CHK1

GAPDH 30

0.6

CHK1-pS345

pCHK1/CHK1

0.4

0.2

<0.2 <0.4 <0.6 <0.8 <1.0 <1.2 <1.4 <1.6 <1.8 <2.0 <2.2

0.13 0.41 0.07 0.11 0.01 0.52

0.01

Fork speed (kb min1)

0 24 h 48 h 72 h

Control Ras

53BP1

H2AX

DAPI

2AX 53BP1

Days Ras

DAPI (micronuclei)

** *****

53BP1 Cyclin A DAPI

% Positive cells

% Cells with 53BP1 foci

% Cells with micronuclei

1 25 >5

S/G2

0 Con 1d 2d 3d 4d 5d 6d

0 G1 S/G2

0 Con 1 2 3 4 5 6

Days Ras

Figure 3 | HRASV12 causes replication stress and genomic instability. (a) Top: DNA bre labelling was performed in BJ-HRASV12 cells after RAS induction for 24, 48 and 72 h. Bottom: representative images of DNA bres after 72 h RAS induction. (b) Distributions of replication fork speeds after RAS induction. N 3. (c) Median replication fork speeds after RAS induction. N 3. (d) Protein levels of phospho-S33 RPA32, RPA32 and GAPDH after RAS

induction. (e) Protein levels of phospho-S345 CHK1, CHK1 and GAPDH after RAS induction. Hydroxyurea (HU, 2 mM for 24 h) was used as a positive control. (f) Representative images of gH2AX and 53BP1 foci after RAS induction for 96 h. (g) Percentages of cells containing more than 8 gH2AX or 53BP1 foci after RAS induction for 96 h. N 2. Asterisks compare with control. (h) Cell cycle distribution of 53BP1-positive cells after 96 h RAS induction as

determined by co-staining with Cyclin A. Right panel: representative images; left panel: quantication. N 3. (i) Representative images of cells with

micronuclei and percentages of cells containing micronuclei after RAS induction for 16 days. N 2. Asterisks compare with control. Means s.e.m. (bars)

are shown. Students t-test, *Po0.05 and **Po0.01. Scale bars, 10 mm.

HRASV12-induced replication stress depends on TBP. We next turned our attention to transcription factors that may promote RNA synthesis and therefore replication stress downstream of HRASV12. TBP expression is induced by RAS signalling in a number of human, murine and Drosophila cell types19,39,40, and TBP mRNA and protein levels were accordingly increased after HRASV12 induction (Fig. 6ac). We used siRNA depletion to test whether TBP promotes HRASV12-induced replication stress (Fig. 6d,e). TBP depletion decreased nascent RNA synthesis in HRASV12-expressing cells (Fig. 6f). Importantly, TBP depletion also prevented HRASV12-induced fork slowing (Fig. 6g,h) and TBP-depleted cells displayed fewer HRASV12-induced 53BP1 foci (Fig. 6i). Similar results were obtained using a different siRNA sequence targeting TBP (Supplementary Fig. 7ae). We noticed that the alternative TBP siRNA also reduced RNA synthesis and replication fork speeds in control cells, suggesting that generally low RNA synthesis may affect replication (Supplementary Fig. 7c,d). TBP depletion did not prevent hydroxyurea-induced 53BP1 foci (Supplementary Fig. 6f) and did not affect the cell cycle prole of 53BP1 foci-positive cells, suggesting that the reduction in foci was not due to a G2 arrest (Supplementary Fig. 7g).

To test whether TBP acts in the same pathway as ongoing RNA synthesis, we combined TBP siRNA with DRB treatment. Combining both treatments rescued replication fork progression and 53BP1 foci formation to a similar extent as TBP depletion

alone (Fig. 6j,k and Supplementary Fig. 8a). To further test whether ribonucleoside addition was affecting replication stress via a different pathway to transcription41, we combined TBP siRNA and DRB with nucleoside supplementation. Compared with transcription inhibition alone, exogenous nucleosides had no additional effect on either the rescue of replication fork progression or the reduction in 53BP1 foci (Supplementary Fig. 8be).

These data suggest that the main pathway of HRASV12-induced replication stress is via upregulation of transcription, and that TBP is involved in this upregulation of transcription and the replication fork slowing and genomic instability that result from it (Fig. 6l).

TBP overexpression alone causes replication stress. We reasoned that if TBP was a downstream effector in HRASV12-induced replication stress, then overexpression of TBP alone should cause replication stress. We therefore generated BJ-hTert broblasts for doxycycline-inducible overexpression of TBP (BJ-TBPind; Fig. 7a). Inducing TBP overexpression over several days led to a steady increase in RNA synthesis activity as measured by EU incorporation (Fig. 7b,c). Importantly, the increase in transcription activity during TBP overexpression was accompanied by replication fork slowing, consistent with replication stress (Fig. 7d and Supplementary Fig. 9a).

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087

a b

Control + transcription inhibitor

+/4OHT

+/ Transcription inhibitor

1 h

RNA synthesis (a.u.)

Control + DMSORas + DMSO Ras + transcription inhibitor

72 h

14 h

CldU IdU

20 20

0 Triptolide -amanitin DRB

c d

Ras

amanitin

% Forks

<0.2 <0.4 <0.6 <0.8 <1.0 <1.2 <1.4 <1.6 <1.8 <2.0 <2.2

Fork speed (kb min1)

RasRas + triptolide

Fork speed (kb min1)

Control + transcription inhibitor

Control + DMSORas + DMSO Ras + transcription inhibitor

1.2

Median fork speed (kb min1 )

* * **

% Forks

0.8

0.6

0.4

0.2

<0.2 <0.4 <0.6 <0.8 <1.0 <1.2 <1.4 <1.6 <1.8 <2.0 <2.2

Fork speed (kb min1)

RasRas + DRB

0 Triptolide -amanitin DRB

g h

2 3 4 C-FOS

5 prom in1

[afii9826]-actin

p (A)

C-FOS: H2AX ChIP

ConCon + APH RasRas + APH

-actin: H2AX ChIP

3.0

2.0

Control Control + APH RasRas + APH

2.0

H2AX/H2AX

1.0

0.0

cFOS2 cFOS3 cFOS4

5 prom in1

Figure 4 | HRASV12-induced replication stress is promoted by ongoing transcription. (a) Seventy-two hours after RAS induction, BJ-HRASV12 cells were incubated with transcription inhibitors triptolide, a-amanitin, DRB or dimethylsulfoxide (DMSO) for 100 min to 4 h before and during EU or DNA bre labelling. (b) Quantication of nascent RNA synthesis by EU incorporation in cells treated with transcription inhibitors 72 h after RAS induction.

N 2 (con triptolide, con DRB), N 3 (all other samples). (ce) Distributions of replication fork speeds in cells treated with transcription inhibitors

72 h after RAS induction. N 3 (DRB, control a-amanitin, control triptolide), N 5 (other samples). (f) Median replication fork speeds in cells

treated with transcription inhibitors 72 h after RAS induction. (g) ChIP analysis of gH2AX versus total H2AX levels on the promoter of the C-FOS gene 72 h after RAS induction. APH samples were treated with 0.5 mM Aphidicolin for 2 h before ChIP. N 3. (h) ChIP analysis of gH2AX versus total H2AX levels

on the b-ACTIN gene 72 h after RAS induction as in g. N 3. qPCR primer positions are shown in upper panels. Means s.e.m. (bars) are shown. Students

t-test, *Po0.05 and **Po0.01.

Similar results were observed in human MRC5 broblasts over-expressing TBP (Supplementary Fig. 9bd). To test whether TBP-induced fork slowing depended on ongoing transcription, we treated TBP-overexpressing cells with DRB for 100 min, to inhibit nascent RNA synthesis (Fig. 7e). DRB treatment

rescued the TBP-induced fork slowing (Fig. 7f and Supplementary Fig. 9e), supporting that TBP overexpression causes replication stress through nascent RNA synthesis.

We next characterized the effect of TBP overexpression on DNA damage, genomic instability and proliferation. Similar to

6 NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087 ARTICLE

a b

Vector RNH1-GFP

RNH1-GFP transfection

+/4OHT

4OHT: + +

GFP

CldU IdU

HRAS

Tubulin

20 20

53BP1 foci

kDa

48 h

24 h

RNaseH1 30

S9.6 e

Nuclear S9.6 intensity (normailsed)

Control

RAS

3.5

** *

3.5

Fold change versus control

RT-qPCR: RNASEH1

Control

Ras

3.0

Vector RNaseH1-GFP

2.5

1.0

2.0

0.87 0.02

1.30 0.02

1.05 0.03

1.5

1.0

0.5

0.0

Vec RH1 Vec RH1

Control

Ras

Vector RNH1-GFP

Median fork speed (kb min1 )

0.8

% Forks

0.6

0.4

0.2

0 <0.2 <0.4 <0.6 <0.8 <1.0 <1.2 <1.4 <1.6 <1.8 <2.0

Fork speed (kb min1)

Ras + vector

Ras + RNH1-GFP

0 Control Ras

i HRASV12

R-loops

53BP1

Vector RNH1-GFP

Control Ras

RNH1-GFPVector

% Cells with 53BP1 foci

Control Ras

Replication stress

Collisions

Figure 5 | HRASV12 induces replication stress through R-loop formation. (a) After RAS induction for 48 h, BJ-HRASV12 cells were transfected with pCMV6-AC-RNaseH1-GFP or empty vector and processed for DNA bre analysis or 53BP1 foci staining after 24 or 48 h. (b) Protein levels of HRAS, GFP, RNaseH1 and TUBULIN (loading control) 72 h after RAS induction and 24 h after transfection with pCMV6-AC-RNaseH1-GFP. (c) RNASEH1 mRNA quantication by quantitative reverse transcriptasePCR after 72 h RAS induction. RNASEH1 mRNA levels were normalized to GAPDH and control.(d) Representative images of S9.6 immunostaining pCMV6-AC-RNaseH1-GFP 72 h after RAS induction. (e) Quantication of nuclear S9.6 intensity pCMV6-AC-RNaseH1-GFP 72 h after RAS induction. N 3. (f) Distribution of fork speeds pCMV6-AC-RNaseH1-GFP 72 h after RAS induction.

N 3. (g) Median replication fork speeds pCMV6-AC-RNaseH1-GFP 72 h after RAS induction. N 3. (h) Representative images of cells with 53BP1 foci,

pCMV6-AC-RNaseH1-GFP 96 h after RAS induction. (i) Percentage of cells displaying more than 8 53BP1 foci, pCMV6-AC-RNaseH1-GFP 96 h after RAS induction. N 3. (j) Model of how HRASV12 causes replication stress via increasing transcription. Means s.e.m. (bars) are shown. Students t-test,

*Po0.05 and **Po0.01. Scale bars, 10 mm.

HRASV12-expressing cells, cells overexpressing TBP displayed increased formation of micronuclei and 53BP1 foci (Fig. 7g), and 53BP1 foci were mostly found in G1 cells (Fig. 7h). Finally, TBP overexpression eventually led to growth arrest and senescence, similar to that caused by HRASV12 (Fig. 7i and Supplementary Fig. 9f,g). In agreement with this, p53 levels were increased after TBP induction (Fig. 7a and also see Supplementary Fig. 10 for original images of western blottings).

Thus, our data show that TBP alone is able to increase transcription activity and cause replication stress with features that resemble oncogene-induced replication stress.

TBP expression and replication stress in cancer. Finally, we investigated the relationship of TBP and RNASEH1 mRNA expression with oncogenes and replication stress markers in tumour samples using The Cancer Genome Atlas (TCGA) data

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 7

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087

b d

RT-qPCR: TBP

Control Ras

TBPsi #1

Control Ras

+/4OHT

siRNA

transfection

DRB CldU/IdU 40

53BP1 foci1 h

3.0

Fold change versus control

Normalized TBP intensity

2.5

24 h

2.5

48 h 72 h

2.0

24 h kDa

46 30

2.0

4OHT: + + +

nonTsi TBPsi #1

1.5

4OHT: + +

kDa

46 30 58 46 30

1.0

TBP

GAPDH

TBP

HRAS

GAPDH

0.5

0.0

24 h 48 h 72 h

g i

NonTsi

TBPsi #1

NonTsi TBPsi #1

NonTsi

Median fork speed (kb min1 )

% Cells with 53BP1 foci

48 h

** *

RNA synthesis (a.u.)

0.8

% Forks

TBPsi #1 + Ras

0.6

0.4

0.2

Con Ras

<0.2 <0.4 <0.6 <0.8 <1.0 <1.2 <1.4 <1.6 <1.8 <2.0 <2.2

0 Control Ras

Fork speed (kb min1)

NonTsi + Ras

DMSO DRB NonTsiNonTsi + DMSO

TBPsi #1

1.2

TBPsi #1 + DRB

** *

HRASV12

Transcription

Replication stress

Median fork speed (kb min1 )

% Cells with 53BP1 foci

* **

DMSO DRB NonTsi

NonTsi + DMSO

TBP

0.8

TBPsi #1

TBPsi #1 + DRB

0.6

0.4

0.2

Control Ras

Figure 6 | TBP is an effector in HRASV12-induced replication stress. (a) TBP mRNA quantication by quantitative reverse transcriptasePCR in BJ-HRASV12 cells 72 h after RAS induction. TBP mRNA levels were normalized to GAPDH and control. (b) Protein levels of TBP and b-ACTIN after RAS induction for the times indicated. (c) Densitometry quantication of TBP levels based on western blotting as in b after RAS induction for the times indicated. Values were normalized to 72 h control. N 3 (24 and 48 h), N 6 (72 h). (d) Twenty-four hours after RAS induction, cells were transfected with

TBP siRNA (TBPsi #1) or control siRNA (nonTsi). Cells were processed for DNA bre analysis or western blotting 48 h later and for 53BP1 staining 24 h later. (e) Protein levels of TBP, HRAS and GAPDH (loading control) 72 h after RAS induction and 48 h after siRNA transfection. (f) Quantication of nascent RNA synthesis by EU incorporation TBPsi #1 72 h after RAS induction. N 3. (g) Distribution of replication fork speeds TBPsi #1 72 h after RAS

induction. N 3. (h) Median replication fork speeds TBPsi #1 72 h after RAS induction. N 3. (i) Percentages of cells containing more than eight 53BP1

foci, TBPsi #1 96 h after RAS induction. N 3. (j) Median replication fork speeds in cells treated with TBPsi #1 and DRB 72 h after RAS induction,

compared with TBPsi #1 or DRB alone. N 3. (k) Percentages of cells treated with TBPsi #1 and DRB containing more than eight 53BP1 foci after 96 h after

RAS induction, compared with TBPsi #1 or DRB alone. N 3. (l) Model for the role of TBP in HRASV12-induced replication stress. Means s.e.m. (bars) are

shown. Students t-test, *Po0.05, **Po0.01 and ***Po0.001.

sets4250 (Table 1). In a number of cancers, TBP or RNASEH1 expression correlated positively with expression of RAS or MYC oncogenes and the CHEK1 or CHEK2 checkpoint kinases, which are activated by oncogene-induced replication stress4,5. These data support that the relationship between oncogenes, TBP, R-loops and replication stress could be present in cancer tissues.

DiscussionWe report that increased transcription is a new mechanism of oncogene-induced replication stress. Overexpression of oncogenic HRASV12 increases RNA synthesis and R-loop formation, and this directly contributes to replication stress induced by HRASV12. HRASV12-induced replication stress is mediated by the general transcription factor TBP, which is a downstream target of RAS signalling. Accordingly, overexpression of TBP itself causes replication stress and genomic instability. The transcription machinery is frequently deregulated in cancer cells21 and our data

suggest that transcription-associated replication stress could be an important mechanism promoting genomic instability in cancer (Fig. 7j).

Markers of replication stress are observed in early tumours and cancers and other conditions of high proliferation, such as stem cell reprogramming or viral infection51,52. Recent studies have provided rst candidates for the mechanisms causing such endogenous replication stress. So far, all of these have involved deregulation of the cell cycle, including re-replication53 and premature or increased replication initiation, resulting in depletion of nucleotides or replication enzymes9,10,12. In addition, reactive oxygen species play important roles in DNA damage in cancer, but whether they can cause replication stress is still uncertain54,55. Our data support that increased transcription is another pathway causing endogenous replication stress, which may act in parallel or independently of cell cycle deregulation. HRASV12 overexpression did increase the density of active replication origins, which could be reversed by CDK inhibition.

8 NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087 ARTICLE

b c

DAPI

* *

Days doxycycline

22.5 3.8

26.5 4.4

29.3 4.5

27.8 3.4

35.3 3.5

Control

TBP

TUB

Con 1 2 3 4 5 6

kDa

p53 46 58

RNA synthesis (a.u)

TBP

0 Con 1 2 3 4

Days TBP

e f

1.2

DMSO DRB

***

Median fork speed (kb min1 )

1.0

RNA synthesis (a.u.)

Median fork speed (kb min1 )

12.1 0.5

14.1 0.2

0.8

* *

0.6

7.5 0.05

7.7 0.07

0.6

0.4

0.2

0.0

0 Con 1 2 3 4

Con TBP Con TBP

Control TBP

Days TBP

DMSO DRB

53BP1

DAPI (micronuclei)

% Cells with 53BP1 foci

53BP1 foci Micronuclei

% Positive cells

* ** *** *

* ***

** **

0 Con 1 2 3 4 5 6

0 G1 S/G2

Days TBP

-Galactosidase

EGFR Other GF

HRASV12

Control (6 days) TBP (6 days)

% -Gal positive cells

Transcription R-loops

Genomic instability apoptosis/senescence

Replication stress

Other

0 Con 1 2 3 4 5 6

Days TBP

Figure 7 | TBP overexpression causes replication stress and senescence. (a) Protein levels of TBP, p53 and TUBULIN (loading control) in BJ-TBPind cells treated with doxycycline for 16 days, to induce TBP overexpression. No doxycycline was used as a control. (b) Nascent RNA synthesis as measured by EU incorporation after TBP induction for 3 days. Scale bars, 10 mm. (c). Quantication of nuclear EU intensity after TBP induction for 14 days. N 4 (con, day

1, 2 and 4), N 5 (day 3). (d) Median replication fork speeds in after TBP induction. N 2 (day 1, 2 and 4), N 4 (con, day 3). Asterisks compare with

control. (e) Nascent RNA synthesis as measured by EU incorporation in cells treated with DRB or dimethylsulfoxide (DMSO) (control) for 100 min,72 h after TBP induction. N 3. (f) Median replication fork speeds in BJ-TBPind cells treated with DRB 72 h after TBP induction. N 3. (g). Percentage of

cells displaying more than eight 53BP1 foci or micronuclei after TBP induction. Right panel: representative images of cells with 53BP1 foci and micronuclei. Asterisks compare with control. N 27. (h) Cell cycle distribution of 53BP1-positive cells 96 h after TBP induction as determined by co-staining with

Cyclin A. N 2. (i) Representative images and percentages of b-galactosidase staining after TBP induction for 16 days. Scale bars, 100 mm. (j) Model of

how HRASV12 and other growth factor oncogenes such as epidermal growth factor receptor (EGFR) induce replication stress by increasing transcription through TBP and other transcription factors. Additional mechanisms, such as reactive oxygen species, may also contribute to HRASV12-induced DNA damage. Means s.e.m. (bars) are shown. Students t-test, *Po0.05, **Po0.01 and ***Po0.001.

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 9

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087

Table 1 | TCGA data sets showing correlation between mRNA expression of TBP or RNASEH1 with oncogenes and replication

stress markers.

TCGA data set MYC RAS CHEK1 CHEK2 TBP mRNA expression correlation with:

Colorectal adenocarcinoma 0.41 (5%)** 0.54 (1%)** 0.42 (2%)** Glioblastoma multiforme 0.41 (5%)** 0.33 (10%)* 0.47 (1%)**Stomach adenocarcinoma 0.34 (10%)** 0.42 (5%)** 0.45 (1%)** Lung adenocarcinoma 0.39 (1%)** 0.37 (1%)**Lung squamous cell carcinoma 0.32 (5%)* Bladder urothelial carcinoma 0.46 (1%)** 0.34 (5%)*Acute myeloid leukemia 0.31 (5%)* 0.30 (10%)* Skin cutaneous melanoma 0.31 (1%)**Pheochromocytoma and Paraganglioma 0.31 (5%)* 0.36 (5%)*

Liver hepatocellular carcinoma 0.45 (1%)**

RNASEH1 mRNA expression correlation with:Colon adenocarcinoma 0.48 (1%)**Glioblastoma multiforme 0.71 (1%)**Lung adenocarcinoma 0.5 (1%)** 0.48 (1%)** 0.38 (5%)** Lung squamous cell carcinoma 0.31 (5%)* 0.44 (1%)** 0.36 (1%)* Breast invasive carcinoma 0.37 (10%)** 0.57 (1%)** 0.4 (5%)** Pheochromocytoma and Paraganglioma 0.38 (10%)**Prostate adenocarcinoma 0.3 (10%)**Pancreatic adenocarcinoma 0.48 (1%)**Thyroid carcinoma 0.51 (5%)**Brain lower grade glioma 0.37 (10%)**

Blank cells, no correlation; MYC, CMYC/NMYC; RAS, NRAS/HRAS/KRAS; TBP, TATA-box binding protein.

Values: Pearsons coefcient r (0.30.5: positive correlation, 0.51: strong positive correlation, 0.3 0.5: negative correlation, empty cell: no correlation). Values in brackets: top % of all genes;

*Po0.00005 and **Po0.000001, Students t-test.

However, CDK inhibition was unable to relieve HRASV12-induced replication fork slowing, suggesting that increased replication initiation is not a cause but a consequence of HRASV12-induced replication stress (Supplementary Fig. 3d,e). Our data suggest that HRASV12 causes replication stress by a mechanism that is different from oncogenes such as CYCLIN E and CDC25A10,12.

Our data support that HRASV12 overexpression promotes accumulation of R-loops, which can be a major cause for replication fork slowing and DNA breakage23. An elegant study recently reported that loss of tumour suppressors BRCA1 or BRCA2 increases R-loop levels, because these proteins act to prevent R-loop formation56. Our ndings add an important new angle to this observation, showing that oncogenes can induce R-loops, which suggests that increased R-loop levels might also be common in cancer cells that are procient in BRCA1 or BRCA2. Interestingly, we observed increased protein levels of endogenous RNaseH1 in cells overexpressing HRASV12 (Fig. 5b,c). RNaseH1 and RNaseH2 are the main RNase activities counteracting R-loop formation and upregulation of these enzymes may be a response of cancer cells to increased R-loop levels23. Although the biology of increased RNaseH1 expression in response to HRASV12 requires further investigation, it supports the idea that oncogenic replication stress is strongly connected with R-loop metabolism.

One exciting implication of our ndings is the potential to discover new factors involved in promoting replication stress. Components of the transcription machineries are found overexpressed in cancer21,57. TBP is one of several general transcription factors that have been implicated in oncogene-induced cell transformation39,58. The TBP promoter contains binding sites for oncogenic transcription factors and TBP is differentially expressed in cancers (Table 1)59. RAS has been reported to upregulate TBP expression through RAF-MEK and

RALGDS signalling19, and TBP levels are also increased by other growth factor signalling pathways such as epidermal growth factor/epidermal growth factor receptor59,60 (Fig. 7j). We report here that increased TBP levels cause replication stress and markers of genomic instability. TBP expression also correlates with the expression of MYC, RAS and checkpoint kinases in a number of cancers (Table 1). As TBP overexpression did not increase RNA synthesis as strongly as HRASV12 overexpression and TBP depletion did not completely rescue HRASV12-induced replication stress, other transcription factors such as UBTF may also be involved downstream of HRASV12 (Fig. 7j).

A previous report showed that TBP is important for HRASV12 transforming function, and that TBP overexpression alone promotes anchorage-independent growth and tumour growth in vivo49. Our data presented here support the idea that TBP promotes the oncogenic phenotype by increasing transcription activity and replication stress. There is however no evidence that TBP itself is an oncogene, as TBP overexpression promoted tumour growth but not tumourigenesis itself49. Some cancer types show recurrent mutations in TBP that affect the length of the amino-terminal glutamine-rich motif (for example, Q72dup in diffuse large B-cell lymphoma and adrenocortical carcinoma)61. It is yet unknown whether such mutations have pathogenic relevance in cancer and whether they could be oncogenic.

Finally, increased RNA synthesis and elevated levels of R-loops could be correlated with replication stress in cancer. Identifying the molecular mechanisms that directly cause spontaneous replication stress should therefore help to predict and detect replication stress more accurately in cells and tissues. Nascent RNA synthesis can only be measured in live cells but increased expression of transcription factors and RNA polymerase subunits21, as well as increased R-loops and expression of RNases H1 or H2, might be promising candidates for markers of transcription-associated replication stress.

10 NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087 ARTICLE

Methods

Cell lines and reagents. Human BJ-hTert HRASV12ER-TAM (Agami and de Vita labs26,27) BJ-hTert and MRC5 broblasts (ATCC) were authenticated using 8-locus short tandem repeat (STR) proling (LGC Standards). Cells were conrmed to be free of Mycoplasma infection and were grown in DMEM medium (Sigma) with 10% fetal bovine serum (Sigma) supplemented with L-glutamine (Gibco) in a humidied atmosphere containing 5% CO2. For BJ-TBPind cells, human TBP complementary DNA (Origene, SC118124) was inserted into a pInducer20 lentivirus construct to generate TBP-pInducer20 and infected BJ-hTert cells were selected with 500 mg ml 1 G418 (Gibco). HRASV12 was induced with 333 nM 4OHT (Sigma) and TBP expression was induced with 2 mg ml 1 doxycycline (Sigma). 4OHT or doxycycline remained present during all experiments.

DRB (100 mM) was from Sigma, and triptolide (1 mM) and a-amanitin(10 mg ml 1) were from Tocris Bioscience. Ribonucleosides (adenosine, guanosine, uridine and cytidine, 10 mM) were obtained from Sigma. Roscovitine (25 mM) was from Sigma. Hydroxyurea (2 mM) was from Acros Organics, bleomycin(10 mg ml 1) was from Sigma and cycloheximide (100 mg ml 1) was from

Calbiochem.

DNA bre analysis. Cells were pulse labelled with 25 mM CldU and 250 mM IdU for 20 min and harvested. DNA bre spreads were prepared by spotting 2 ml of cells (5 105 cells per ml in PBS) onto microscope slides followed by lysis with 7 ml of

0.5% SDS, 200 mM Tris-HCl pH 7.4 and 50 mM EDTA. Slides were tilted and DNA spreads xed in methanol/acetic acid (3:1). HCl-treated bre spreads were incubated with rat anti-bromodeoxyuridine (detects CldU, BU1/75, AbD Serotec, 1:1,000) and mouse anti-bromodeoxyuridine (detects IdU, B44, Becton Dickinson, 1:500) for 1 h, xed with 4% paraformaldehyde (PFA) to increase staining intensity and incubated with anti-rat IgG AlexaFluor 555 and anti-mouse IgG AlexaFluor 488 (Molecular Probes) for 1.5 h. Images were acquired on an Nikon E600 microscope using a Nikon Plan Apo 60 (1.3 numerical aperture) oil lens,

a Hamamatsu digital camera (C4742-95) and the Volocity acquisition software (Perkin Elmer). Images were analysed using ImageJ (http://rsbweb.nih.gov/ij/

Web End =http://rsbweb.nih.gov/ij/).

In each independent experiment, at least 300 bres were measured per condition.

Immunouorescence. Cells were washed and xed as follows. gH2AX and 53BP1: CSK buffer (10 mM PIPES, 300 mM sucrose, 100 mM NaCl and 3 mM MgCl2) for 1 min, 0.5% Triton X-100 in CSK buffer for 1 min and 4% PFA for 10 min at room temperature; Cyclin A and CldU: 4% PFA for 10 min and 0.3% Triton X-100 in PBS for 5 min at room temperature; S9.6: methanol for 10 min on ice and 0.5% Triton X-100 in PBS for 5 min at room temperature. RNase H (New England Biolabs) was used at 0.05 U ml 1 for 36 h at 37 C. RNase A (Invitrogen) was used at 0.05 ng ml 12 mg ml 1 for 2 h at 37 C. Samples were blocked with 3% BSA/10%

fetal bovine serum. Primary antibodies were mouse anti-phospho-Histone H2AX (Ser139) (JBW301, Millipore 05-636, 1:1,000), rabbit anti-53BP1 (Bethyl A300-272A, 1:30,000), mouse anti-Cyclin A (6E6, Thermo Scientic MS1061, 1:50), rat anti-CldU (BU1/75, AbD Serotec OBT0030G, 1:250) and mouse anti-RNA/DNA hybrid (S9.6, gift from Professor Richard Gibbons, hybridoma supernatant 1:100). Secondary antibodies were anti-mouse IgG AlexaFluor 488 and anti-rabbit IgG AlexaFluor 555 (Molecular Probes). DNA was counterstained with 4,6-diamidino-2-phenylindole (DAPI) and images acquired as above. For quantication of nuclear S9.6 intensity, ImageJ was used to generate nuclear masks based on DAPI staining and mean S9.6 uorescence intensities per pixel were quantied per nucleus.

EU incorporation assay. EU incorporation assays were performed using the Click-iT RNA Alexa Fluor 594 Imaging Kit (Invitrogen) according to the manufacturers instructions. Cells were incubated with 1 mM EU for 1 h, xed with 4% PFA for 15 min at room temperature, permeabilized with 0.5% Triton X-100 for 15 min and Click-iT reaction was performed. DNA was counterstained with DAPI and images were acquired as above. ImageJ was used to generate nuclear masks based on DAPI staining and mean AlexaFluor 594 uorescence intensities per pixel were quantied per nucleus.

siRNA and DNA transfection. siRNAs against TBP (TBPsi #1: sense 50-GAA UCUUGGUUGUAAACUU-30 and TBPsi #2: sense 50-GGAUAAGAGAGCC ACGAAC-30) were purchased from Ambion and Dharmacon. siRNA against RNaseH1 (s48357) was purchased from Ambion. Allstars negative control siRNA was purchased from Qiagen. Cells were transfected with 50 nM siRNA using Dharmafect 1 reagent (GE Dharmacon). For RNaseH1 overexpression, cells were transfected with 2.5 mg pCMV6-AC-RNase H1-GFP (Origene) using TransIT-2020 (Mirus Bio). Empty pcDNA 3.1 ( ) vector was purchased from Invitrogen.

Western blotting. Cell extracts were prepared in UTB buffer (50 mM Tris-HCl pH 7.5, 150 mM b-mercaptoethanol and 8 M urea) and sonicated to release DNA-bound proteins. Primary antibodies used were mouse anti-HRAS(Santa Cruz sc-29, 1:500), mouse anti-TBP (1TBP18, Abcam ab818, 1:2,000), rabbit anti-RNaseH1 (Abcam ab83179, 1:3,000), rabbit anti-phospho-ERK1/2 (Cell Signaling 9101, 1:1,000), rabbit anti-ERK1/2 (Cell Signaling 9102, 1:1,000),

rabbit anti-gH2AX (Bethyl A300-081A, 1:1,000), rabbit anti-phospho-S345 CHK1 (Cell Signaling 2341, 1:1,000), rabbit anti-CHK1 (Cell Signaling 2345S, 1:1,000), rabbit anti-phospho-S33 RPA32 (Bethyl A300-426A, 1:1,000), mouse anti-RPA32 (9H8, Abcam ab2175, 1:2,000), rabbit anti-Turbo-GFP (Evrogen AB513, 1:10,000), mouse anti-p53 (DO-1, gift from Professor David Lane, 1:100), mouse anti-CyclinB1 (V152, Abcam ab72, 1:1,000), mouse anti-aTUBULIN (B512, Sigma

T6074, 1:10,000), rabbit anti-bACTIN (Cell Signaling A967S, 1:5,000) and mouse anti-GAPDH (6C5, Abcam ab8245, 1:10,000).

ChIP analysis. ChIP was performed using 5 mg of anti-gH2AX (Merck Millipore, 07-164) and H2AX (Merck Millipore, 07-627) antibodies as previously described62,63. Where indicated, samples were treated with 0.5 mM Aphidicolin (Sigma) for 2 h before ChIP. The immunoprecipitated (gH2AX or H2AX antibody in immunoprecipitation (IP) reaction), control (beads only) and input DNAs were used as templates for qPCR, containing QuantiTect SYBR Green PCR Master Mix (QIAGEN) and gene-specic primers (see Supplementary Table 1).

RNA/DNA hybrid immunoprecipitation. DIP analysis was performed with RNA/DNA hybrid antibody (0.3 mg ml 1 per IP reaction) puried from S9.6 hybridoma cell lines as previously described33. The immunoprecipitated(S9.6 antibody in IP reaction), control (beads only) and input DNAs were used as templates for qPCR. DIP RNase H-sensitivity analysis was carried out before IP step with the addition of 25 U RNase H (NEB, M0297S). One hundred microlitres of nuclease digestion reaction contained 1 reaction buffer and it was performed

for 3 h at 37 C.

Slot-blot experiments. Slot-blot experiments were carried out as described32. Genomic DNA (1.2 mg) were treated with 2 U of RNase H per mg of DNA (NEB, M0297S) for 2 h at 37 C before loading on the slot blot. Half of the DNA sample was probed with S9.6 antibody (1:1,000) and the other half with anti-ssDNA antibody (MAB3031, Millipore, 1:25,000) as described32. Secondary antibody was goat anti-mouse horseradish peroxidase (1:10,000). Images were acquired with LAS-4000 (Fujilm) and quantied using Image Studio Lite software (Li-COR Biosciences).

Cell proliferation and b-galactosidase assays. For proliferation assays, 1 105

or 2 105 cells were seeded in 12-well plates and incubated with 10 mg ml 1

resazurin for 2 h at the indicated time points. Resorun uorescence at 590 nm was measured using a BMG Labtech PHERAstar FS microplate reader. Cell numbers were determined by multiplying the number of initially seeded cells with (FRn/FR0)

(FRn: uorescence reading at time point, FR0: initial uorescence reading). b-Galactosidase staining was performed using the Senescence b-Galactosidase Staining Kit (Cell Signaling) according to the manufacturers instructions.

Nucleotide quantication. Cells were harvested and nucleotides extracted with 70% ice-cold methanol. Precipitated proteins were removed by centrifugation and supernatants stored at 80 C. Supernatants were dried in a heated vacuum

centrifuge and reconstituted in HPLC starting eluent. Samples were analysed by HPLC (Waters 2695, Watford, UK) with a photodiode array detector (Waters 2996). Separation was achieved using an Ace C18 (3 mm, 3 125 mm, Hichrom,

UK) column maintained at 35 C with eluent A: 10 mM potassium dihydrogen phosphate and 10 mM tetrabutylammonium hydrogen sulfate, 10% methanolpH 6.9; eluent B: 50 mM potassium dihydrogen phosphate, 6 mM tetrabutylammonium hydrogen sulfate and 30% methanol pH 7, using a ow rate of 0.6 ml min 1 and a gradient of 2580% B over 25 min, with a run time of30 min. Nucleotides were identied by comparing with absorbance spectra and retention times of commercially available standards.

Quantitative real-time PCR. Total RNA was harvested using TRIZOL reagent (Invitrogen) followed by DNase I treatment (Roche), 1.5 mg of total RNA was reverse transcribed using SuperScript Reverse Transcriptase III (Invitrogen) with random hexamers (Invitrogen), following manufacturers instructions. The qPCR primers for amplication are listed in Supplementary Table 1. For quantitative real-time PCR, 2 ml of cDNA was analysed using a Rotor-Gene RG-3000 real-time

PCR machine (Corbett Research) with QuantiTect SYBR green (Qiagen). Cycling parameters were 95 C for 15 min, followed by 45 cycles of 94 C for 20 s,58 C62 C for 20 s and 72 C for 20 s. Fluorescence intensities were plotted against the number of cycles by using an algorithm provided by the manufacturer.

Pulsed eld gel electrophoresis. Cells (2 106) per sample were treated as

indicated, harvested and melted into 1.0% InCert-Agarose (Lonza) inserts. Inserts were digested in 0.5 M EDTA-1% N-laurylsarcosyl-proteinase K (1 mg ml 1) at room temperature for 48 h and washed three times in TE buffer. Inserts were loaded onto a separation gel (1% chromosomal-grade agarose, Bio-Rad). Separation was performed using a CHEF DR III (BioRad; 120 eld angle, 240 s switch time, 4 V cm 1, 14 C) for 20 h. Images of ethidium bromide-stained gels were acquired using a Syngene G:BOX gel imaging system. DSBs (chromosome

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 11

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087

fragments 42 Mb) were quantied by densitometry using ImageJ. Intensity of DNA entering the gel was normalized to total DNA and untreated control to obtain nal values.

Statistical analysis. Unless stated otherwise, all values are means 1 s.e.m. of results from independent biological repeats. Scatter plots show pooled data, but numerical values displayed on plots represent the means 1 s.e.m. of the results from independent repeats. Numbers of repeats N are indicated in the gure legends. Statistical tests were performed using the one-tailed Students t-test. Coexpression data were obtained using CBioPortal61,64 and statistical analyses were performed using the Students t-test based on t (r*On 2)/(O r2) where r is

Pearsons coefcient and n is number of samples in analysis.

Data availability. The authors declare that all the data supporting the ndings of this study are available within the article and its Supplementary Information les and from the corresponding authors upon reasonable request.

References

1. McGranahan, N., Burrell, R. A., Endesfelder, D., Novelli, M. R. & Swanton, C. Cancer chromosomal instability: therapeutic and diagnostic challenges. EMBO Rep. 13, 528538 (2012).

2. Kotsantis, P., Jones, R. M., Higgs, M. R. & Petermann, E. Cancer therapy and replication stress: forks on the road to perdition. Adv. Clin. Chem. 69, 91138 (2015).

3. Burrell, R. A. et al. Replication stress links structural and numerical cancer chromosomal instability. Nature 494, 492496 (2013).

4. Bartkova, J. et al. Oncogene-induced senescence is part of the tumorigenesis barrier imposed by DNA damage checkpoints. Nature 444, 633637 (2006).

5. Di Micco, R. et al. Oncogene-induced senescence is a DNA damage response triggered by DNA hyper-replication. Nature 444, 638642 (2006).

6. Dereli-Oz, A., Versini, G. & Halazonetis, T. D. Studies of genomic copy number changes in human cancers reveal signatures of DNA replication stress. Mol. Oncol. 5, 308314 (2011).

7. Dobbelstein, M. & Sorensen, C. S. Exploiting replicative stress to treat cancer. Nat. Rev. Drug Discov. 14, 405423 (2015).

8. Dominguez-Sola, D. et al. Non-transcriptional control of DNA replication by c-Myc. Nature 448, 445451 (2007).

9. Bester, A. C. et al. Nucleotide deciency promotes genomic instability in early stages of cancer development. Cell 145, 435446 (2011).

10. Neelsen, K. J., Zanini, I. M., Herrador, R. & Lopes, M. Oncogenes induce genotoxic stress by mitotic processing of unusual replication intermediates.J. Cell Biol. 200, 699708 (2013).11. Carlos, A. R. et al. ARF triggers senescence in Brca2-decient cells by altering the spectrum of p53 transcriptional targets. Nat. Commun. 4, 2697 (2013).

12. Jones, R. M. et al. Increased replication initiation and conicts with transcription underlie Cyclin E-induced replication stress. Oncogene 32, 37443753 (2013).

13. Beier, R. et al. Induction of cyclin E-cdk2 kinase activity, E2F-dependent transcription and cell growth by Myc are genetically separable events. EMBO J. 19, 58135823 (2000).

14. White, R. J. RNA polymerases I and III, growth control and cancer. Nat. Rev. Mol. Cell Biol. 6, 6978 (2005).

15. Zhao, J., Yuan, X., Frodin, M. & Grummt, I. ERK-dependent phosphorylation of the transcription initiation factor TIF-IA is required for RNA polymerase I transcription and cell growth. Mol. Cell 11, 405413 (2003).

16. Coleman, M. L., Marshall, C. J. & Olson, M. F. RAS and RHO GTPases in G1-phase cell-cycle regulation. Nat. Rev. Mol. Cell Biol. 5, 355366 (2004).

17. Lin, C. Y. et al. Transcriptional amplication in tumor cells with elevated c-Myc. Cell 151, 5667 (2012).

18. Nie, Z. et al. c-Myc is a universal amplier of expressed genes in lymphocytes and embryonic stem cells. Cell 151, 6879 (2012).

19. Johnson, S. A., Mandavia, N., Wang, H. D. & Johnson, D. L. Transcriptional regulation of the TATA-binding protein by Ras cellular signaling. Mol. Cell Biol. 20, 50005009 (2000).

20. Gonzalez, O. G. et al. Telomerase stimulates ribosomal DNA transcription under hyperproliferative conditions. Nat. Commun. 5, 4599 (2014).

21. Bywater, M. J., Pearson, R. B., McArthur, G. A. & Hannan, R. D. Dysregulation of the basal RNA polymerase transcription apparatus in cancer. Nat. Rev. Cancer 13, 299314 (2013).

22. French, S. Consequences of replication fork movement through transcription units in vivo. Science 258, 13621365 (1992).

23. Aguilera, A. & Garcia-Muse, T. R loops: from transcription byproducts to threats to genome stability. Mol. Cell 46, 115124 (2012).

24. Mirkin, E. V. & Mirkin, S. M. Replication fork stalling at natural impediments. Microbiol. Mol. Biol. Rev. 71, 1335 (2007).

25. Gan, W. et al. R-loop-mediated genomic instability is caused by impairment of replication fork progression. Genes Dev. 25, 20412056 (2011).

26. Voorhoeve, P. M. et al. A genetic screen implicates miRNA-372 and miRNA-

373 as oncogenes in testicular germ cell tumors. Cell 124, 11691181 (2006).

27. De Vita, G. et al. Dose-dependent inhibition of thyroid differentiation by RAS oncogenes. Mol. Endocrinol. 19, 7689 (2005).

28. Drost, J. et al. BRD7 is a candidate tumour suppressor gene required for p53 function. Nat. Cell Biol. 12, 380389 (2010).

29. Herranz, H., Hong, X., Hung, N. T., Voorhoeve, P. M. & Cohen, S. M. Oncogenic cooperation between SOCS family proteins and EGFR identied using a Drosophila epithelial transformation model. Genes Dev. 26, 16021611 (2012).

30. Boguslawski, S. J. et al. Characterization of monoclonal antibody to DNA.RNA and its application to immunodetection of hybrids. J. Immunol. Methods 89, 123130 (1986).

31. El Hage, A., French, S. L., Beyer, A. L. & Tollervey, D. Loss of topoisomerase I leads to R-loop-mediated transcriptional blocks during ribosomal RNA synthesis. Genes Dev. 24, 15461558 (2010).

32. Sollier, J. et al. Transcription-coupled nucleotide excision repair factors promote R-loop-induced genome instability. Mol. Cell 56, 777785 (2014).

33. Skourti-Stathaki, K., Proudfoot, N. J. & Gromak, N. Human senataxin resolves RNA/DNA hybrids formed at transcriptional pause sites to promote Xrn2-dependent termination. Mol. Cell 42, 794805 (2011).

34. Derheimer, F. A. et al. RPA and ATR link transcriptional stress to p53. Proc. Natl Acad. Sci. USA 104, 1277812783 (2007).

35. Chan, K. L., Palmai-Pallag, T., Ying, S. & Hickson, I. D. Replication stress induces sister-chromatid bridging at fragile site loci in mitosis. Nat. Cell Biol. 11, 753760 (2009).

36. Lukas, C. et al. 53BP1 nuclear bodies form around DNA lesions generated by mitotic transmission of chromosomes under replication stress. Nat. Cell Biol. 13, 243253 (2011).

37. Groh, M., Luno, M. M., Wade-Martins, R. & Gromak, N. R-loops associated with triplet repeat expansions promote gene silencing in Friedreich ataxia and fragile X syndrome. PLoS Genet. 10, e1004318 (2014).

38. Techer, H. et al. Signaling from Mus81-Eme2-dependent DNA damage elicited by Chk1 deciency modulates replication fork speed and origin usage. Cell Rep. 14, 11141127 (2016).

39. Johnson, S. A. et al. Increased expression of TATA-binding protein, the central transcription factor, can contribute to oncogenesis. Mol. Cell Biol. 23, 30433051 (2003).

40. Wang, H. D., Trivedi, A. & Johnson, D. L. Hepatitis B virus X protein induces RNA polymerase III-dependent gene transcription and increases cellular TATA-binding protein by activating the Ras signaling pathway. Mol. Cell Biol. 17, 68386846 (1997).

41. Aird, K. M. et al. Suppression of nucleotide metabolism underlies the establishment and maintenance of oncogene-induced senescence. Cell Rep. 3, 12521265 (2013).

42. Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature 487, 330337 (2012).

43. Brennan, C. W. et al. The somatic genomic landscape of glioblastoma. Cell 155, 462477 (2013).

44. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of gastric adenocarcinoma. Nature 513, 202209 (2014).

45. Cancer Genome Atlas Research Network. Comprehensive molecular proling of lung adenocarcinoma. Nature 511, 543550 (2014).

46. Cancer Genome Atlas Research Network. Comprehensive genomic characterization of squamous cell lung cancers. Nature 489, 519525 (2012).

47. Cancer Genome Atlas Research Network. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N. Engl. J. Med. 368, 20592074 (2013).

48. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 507, 315322 (2014).

49. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature 490, 6170 (2012).

50. Cancer Genome Atlas Research Network. The molecular taxonomy of primary prostate cancer. Cell 163, 10111025 (2015).

51. Ruiz, S. et al. Limiting replication stress during somatic cell reprogramming reduces genomic instability in induced pluripotent stem cells. Nat. Commun. 6, 8036 (2015).

52. Zimmerman, E. S. et al. Human immunodeciency virus type 1 Vpr induces DNA replication stress in vitro and in vivo. J. Virol. 80, 1040710418 (2006).

53. Davidson, I. F., Li, A. & Blow, J. J. Deregulated replication licensing causes DNA fragmentation consistent with head-to-tail fork collision. Mol. Cell 24, 433443 (2006).

54. Ogrunc, M. et al. Oncogene-induced reactive oxygen species fuel hyperproliferation and DNA damage response activation. Cell Death Differ. 21, 9981012 (2014).

55. Rai, P. et al. Continuous elimination of oxidized nucleotides is necessary to prevent rapid onset of cellular senescence. Proc. Natl Acad. Sci. USA 106, 169174 (2009).

12 NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13087 ARTICLE

56. Bhatia, V. et al. BRCA2 prevents R-loop accumulation and associates with TREX-2 mRNA export factor PCID2. Nature 511, 362365 (2014).

57. Ribeiro, J. R., Lovasco, L. A., Vanderhyden, B. C. & Freiman, R. N. Targeting TBP-associated factors in ovarian cancer. Front. Oncol. 4, 45 (2014).

58. Bywater, M. J. et al. Inhibition of RNA polymerase I as a therapeutic strategy to promote cancer-specic activation of p53. Cancer Cell 22, 5165 (2012).

59. Fromm, J. A., Johnson, S. A. & Johnson, D. L. Epidermal growth factor receptor 1 (EGFR1) and its variant EGFRvIII regulate TATA-binding protein expression through distinct pathways. Mol. Cell Biol. 28, 64836495 (2008).

60. Zhang, Q., Zhong, Q., Evans, A. G., Levy, D. & Zhong, S. Phosphorylation of histone H3 serine 28 modulates RNA polymerase III-dependent transcription. Oncogene 30, 39433952 (2011).

61. Cerami, E. et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2, 401404 (2012).

62. Kadener, S., Fededa, J. P., Rosbash, M. & Kornblihtt, A. R. Regulation of alternative splicing by a transcriptional enhancer through RNA pol II elongation. Proc. Natl Acad. Sci. USA 99, 81858190 (2002).

63. West, S., Gromak, N. & Proudfoot, N. J. Human 50 --4 30 exonuclease Xrn2 promotes transcription termination at co-transcriptional cleavage sites. Nature 432, 522525 (2004).

64. Gao, J. et al. Integrative analysis of complex cancer genomics and clinical proles using the cBioPortal. Sci. Signal. 6, pl1 (2013).

Acknowledgements

P.K. and R.M.J. are supported by a Worldwide Cancer Research grant (13-1048) and a Medical Research Council New Investigator Research Grant (MR/J007595/1) to E.P. N.G. is supported by a Royal Society University Research Fellowship and a Medical Research Council New Investigator Research Grant (MR/J007870/1). L.M.S. is supported by a fellowship from the Portuguese Foundation for Science and Technology. The results shown here are in part based on data generated by the TCGA Research Network: http://cancergenome.nih.gov/

Web End =http://cancergenome.nih.gov/ . We thank Profs Gabriella de Vita, Grant S. Stewart, David Lane, Richard Gibbons and Dr Alexandra Milona for providing materials.

Author contributions

P.K. designed and performed most of the experiments, analysed data and contributed to writing the paper. L.M.S. performed DIP, ChIP, reverse transcriptasequantitative PCR (RTqPCR) and slot-blot experiments. S.I. performed ChIP and DIP experiments. R.M.J performed roscovitine and CldU staining experiments. L.F. designed, performed and analysed nucleotide quantication. N.G. designed, performed and analysed DIP, ChIP, RTqPCR and slot-blot experiments, and contributed to writing the paper.

E.P. conceived and coordinated the project, designed experiments, analysed data and wrote the paper.

Additional information

Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Web End =http://www.nature.com/ http://www.nature.com/naturecommunications

Web End =naturecommunications

Competing nancial interests: The authors declare no competing nancial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

Web End =http://npg.nature.com/ http://npg.nature.com/reprintsandpermissions/

Web End =reprintsandpermissions/

How to cite this article: Kotsantis, P. et al. Increased global transcription activity as a mechanism of replication stress in cancer. Nat. Commun. 7, 13087 doi: 10.1038/ncomms13087 (2016).

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/

r The Author(s) 2016

NATURE COMMUNICATIONS | 7:13087 | DOI: 10.1038/ncomms13087 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 13

Word count: 10392

Show less

Abstract

Translate

Cancer is a disease associated with genomic instability that often results from oncogene activation. This in turn leads to hyperproliferation and replication stress. However, the molecular mechanisms that underlie oncogene-induced replication stress are still poorly understood. Oncogenes such as HRAS^V12 promote proliferation by upregulating general transcription factors to stimulate RNA synthesis. Here we investigate whether this increase in transcription underlies oncogene-induced replication stress. We show that in cells overexpressing HRAS^V12 , elevated expression of the general transcription factor TATA-box binding protein (TBP) leads to increased RNA synthesis, which together with R-loop accumulation results in replication fork slowing and DNA damage. Furthermore, overexpression of TBP alone causes the hallmarks of oncogene-induced replication stress, including replication fork slowing, DNA damage and senescence. Consequently, we reveal that increased transcription can be a mechanism of oncogene-induced DNA damage, providing a molecular link between upregulation of the transcription machinery and genomic instability in cancer.

Details

Title

Increased global transcription activity as a mechanism of replication stress in cancer

Author

Kotsantis, Panagiotis; Silva, Lara Marques; Irmscher, Sarah; Jones, Rebecca M; Folkes, Lisa; Gromak, Natalia; Petermann, Eva

Pages

13087

Publication year

2016

Publication date

Oct 2016

Publisher

Nature Publishing Group

e-ISSN

20411723

Source type

Scholarly Journal

Language of publication

English

DOI

https://doi.org/10.1038/ncomms13087

ProQuest document ID

1827631466

Increased global transcription activity as a mechanism of replication stress in cancer

Jump to:

Full Text

Abstract

Details

Suggested sources