Mann A et al. (2022), POLθ prevents MRE11-NBS1-CtIP-dependent fork br...

XB-ART-59465

Mol Cell 2022 Nov 17;8222:4218-4231.e8. doi: 10.1016/j.molcel.2022.09.013.

Show Gene links Show Anatomy links

POLθ prevents MRE11-NBS1-CtIP-dependent fork breakage in the absence of BRCA2/RAD51 by filling lagging-strand gaps.

Mann A , Ramirez-Otero MA , De Antoni A , Hanthi YW , Sannino V , Baldi G , Falbo L , Schrempf A , Bernardo S , Loizou J , Costanzo V .

???displayArticle.abstract???
POLθ promotes repair of DNA double-strand breaks (DSBs) resulting from collapsed forks in homologous recombination (HR) defective tumors. Inactivation of POLθ results in synthetic lethality with the loss of HR genes BRCA1/2, which induces under-replicated DNA accumulation. However, it is unclear whether POLθ-dependent DNA replication prevents HR-deficiency-associated lethality. Here, we isolated Xenopus laevis POLθ and showed that it processes stalled Okazaki fragments, directly visualized by electron microscopy, thereby suppressing ssDNA gaps accumulating on lagging strands in the absence of RAD51 and preventing fork reversal. Inhibition of POLθ DNA polymerase activity leaves fork gaps unprotected, enabling their cleavage by the MRE11-NBS1-CtIP endonuclease, which produces broken forks with asymmetric single-ended DSBs, hampering BRCA2-defective cell survival. These results reveal a POLθ-dependent genome protection function preventing stalled forks rupture and highlight possible resistance mechanisms to POLθ inhibitors.

???displayArticle.pubmedLink??? 36400008
???displayArticle.link??? Mol Cell

Species referenced: Xenopus laevis
Genes referenced: brca2 gmnn h2bc21 mbp mre11 nbn rad51 rasl12 rbbp8 smarcal1
GO keywords: DNA double-strand break processing [+]

???attribute.lit??? ???displayArticles.show???

	Graphical Abstract.
	Figure 1. Xenopus POLθ is recruited to stalled replication forks (A) Human and Xenopus POLθ. (B) Representative western blot (WB) of chromatin-binding time course of mock and POLθ-depleted extracts treated with DMSO or APH (1.5 mM). Chromatin isolation time points following nuclei addition to egg extracts are indicated. Chromatin-bound RPA quantification was determined by relative optical density (ROD) compared with histone H2B. (C) POLθ chromatin-binding ratio with histone H2B. Columns indicate mean ± SD; n = 3; unpaired t test; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. (D) Representative WB of the indicated proteins from DMSO or APH-treated extracts supplemented with buffer or geminin. ROD for the POLθ band compared with H2B is indicated. (E) Representative iPOND experiment showing nascent chromatin isolated from extracts treated with DMSO or APH 60 min after nuclei addition in the presence or absence of biotin-dUTP added 10 min earlier.
	Figure 2. Replicative polymerases inhibition induces stalled OKFs detected by EM (A and B) (A) Representative EM images of RIs isolated from an extract treated with DMSO-containing buffer or (B) 1.5 mM aphidicolin (APH) added 60 min after nuclei addition to extracts and incubated for additional 45 min. Letters indicate daughter (D) and parental (P) strands. Outlined windows show the magnified section highlighted in dotted rectangle. Schematic drawings represent dsDNA (continuous line) and ssDNA (dotted line) thickness, respectively. Bar lengths in nucleotides (nt) are indicated. Red arrows highlight ssDNA. (C and D) Representative replication forks isolated following 20 μM APH (M-APH) added to the extract as in (B) showing OKFs separated by ssDNA. (E) Scheme showing asymmetric ssDNA gap distribution. (F) Gap quantification of DNA molecules represented in (A), (B), (C), and (D). Each dot represents the gap length for each RI isolated from extracts treated as indicated. Horizontal axis numbers indicate gap position from fork junction as in (E) for the same RI containing 1, 2, 3, or 4 gaps. Measurements were conducted on 300 RIs pooled from three independent experiments (n = 300); 100 RIs were counted for each treatment; horizontal bars indicate mean ± SD; unpaired t test; ∗∗p < 0.01;∗∗∗∗p < 0.0001. Percentage of RIs with APH- and M-APH-induced-gaps is indicated in parentheses. (G) OKF length detected on RIs isolated from extracts treated with M-APH as in (C) and (D). Measurements were conducted on RIs pooled from independent experiments (n = 100). Horizontal bars indicate mean ± SD.
	Figure 3. POLθ polymerase inhibition impacts on replication fork ssDNA accumulation (A) Chemical structure of POLθi. (B) Representative denaturing gel showing POLθ-FL-mediated 3′-extension of a fluorescent DNA oligo in the presence of the indicated concentrations of POLθi. Percentage of DNA elongation inhibition is shown for each lane. (C) Representative autoradiography showing incorporation of α-32PdCTP in sperm nuclei incubated for the indicated times. ODs are indicated. (D) Representative EM image of a replication fork isolated from an extract treated with POLθi (5 μM) for 60 min following sperm nuclei addition. Red arrow indicates ssDNA. (E) Quantification of total ssDNA gap length for each replication fork represented in (D), (G), and (H) isolated from extracts treated with POLθi, M-APH, or both added 60 min after sperm nuclei addition and incubated for 45 min before DNA isolation. RIs were pooled from three independent experiments (n = 280); 70 RIs were counted for each treatment; horizontal bars indicate mean ± SD; unpaired t test; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. (F) Quantification of total ssDNA gap length for each fork isolated from mock or POLθ-depleted extracts. RIs were pooled from three independent experiments (n = 150); 75 RIs were counted for each treatment; horizontal bars indicate mean ± SD; unpaired t test; ∗∗∗∗p < 0.0001. (G and H) Representative EM images of replication forks isolated from an extract treated with POLθi and M-APH added 60 min after sperm nuclei addition and incubated for an additional 45 min before DNA isolation.
	Figure 4. POLθ polymerase and helicase process stalled OKFs (A) Quantification of fork junction ssDNA gap length for each replication fork isolated from extracts supplemented with buffer or POLθ-FL (25 nM) and treated with APH. Measurements were conducted on 150 RIs pooled from three independent experiments (n = 150); 75 RIs were counted for each treatment; horizontal bars indicate mean ± SD; unpaired t test; ∗∗p < 0.01. (B and C) (B) Representative EM images of replication forks isolated from extracts supplemented buffer or (C) POLθ-FL and APH. (D) Quantification of total ssDNA length for each replication fork represented in (E–G) isolated from extracts supplemented with buffer, POLθ-FL, POLθ-POL, or POLθ-HEL and treated with M-APH. RIs were pooled from three independent experiments (n = 280); 70 RIs were counted for each condition; horizontal bars indicate mean ± SD; unpaired t test; ∗p < 0.05; ∗∗∗∗p < 0.0001. (E–G) (E) EM images of replication forks isolated from extracts supplemented with POLθ-FL, (F) POLθ-POL (25 nM), or (G) POLθ-HEL treated with M-APH. Blue arrows indicate DNA flaps. See also Figures S3–S5.
	Figure 5. POLθ and RAD51 bound to lagging-strand DNA protect stalled forks from MRE11-dependent cleavage (A) Quantification of fork junction ssDNA gap length for each RI isolated from extracts that were treated as indicated. RIs pooled from three independent experiments (n = 300); 100 RIs were counted for each treatment; unpaired t test; ∗∗∗∗p < 0.0001. (B and C) (B) Representative EM images of replication forks isolated from extracts supplemented with GST-BRC4 (BRC4) or (C) GST-BRC4 and POLθ-FL. (D) Representative WB showing the indicated proteins in extract or bound to chromatin isolated 60 min after addition of sperm nuclei to egg extracts supplemented with buffer or recombinant hPOLα complex at the indicated concentrations. (E) Representative WB showing chromatin-binding time course of the indicated proteins in extracts that were treated with GST or GST-BRC4. Chromatin was isolated at the indicated times after sperm nuclei addition to interphase egg extracts. Rectangles delimitate relevant lanes derived from the same gel. (F) Quantification of broken forks isolated from extracts supplemented with POLθi (5 μM), GST-BRC4 (0.4 mg/mL), 100 μM Mirin, or 100 μM PFM01 as indicated. RIs were pooled from independent experiments (n = 420); 70 RIs were counted for each treatment; columns indicate mean ± SD; unpaired t test; ∗∗∗p < 0.001; ns, non-significant. (G–I) (G) Representative EM images of replication forks isolated from extracts supplemented with GST-BRC4 and buffer, (H) GST-BRC4 and POLθi, or (I) GST-BRC4, POLθi, and PFM01. Numbers in white show the approximate length in nt for each DNA segment. Quantification for the representative images is shown in (F).
	Figure 6. Inhibition of POLθ polymerase induces ssDNA gaps and MRE11-dependent DSB formation limiting BRCA2-defective cell survival (A and B) (A) Schematic of the IdU/CldU pulse-labeling protocol followed by S1 nuclease treatment (top). Dot plot of IdU tract lengths (μm) in DLD1 cells or (B) DLD1 BRCA2−/− cells that were treated with DMSO (UN), 2 μM POLθi or 2 μM POLθi and 50 μM Mirin per experimental condition during the IdU pulse labeling. Horizontal bar indicates the mean. Kruskal-Wallis test; ∗p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 (middle). Representative immunofluorescence images of labeled DNA fibers (bottom). (C) Representative EM images of broken replication forks isolated from DLD1 BRCA2−/− cells treated with POLθi for 16 h. Numbers in white show the approximate length in nt for each DNA segment. Red arrows mark ssDNA gaps. Blue arrow marks DNA flaps. (D) Representative confocal microscopy immunofluorescence of EdU-labeled DLD1 and DLD1 BRCA2−/− cells showing γH2AX foci (red) following siRNAs against the indicated targets and exposure to POLθi for 24 h. EdU-positive cells in green. DAPI staining in blue. Bars indicate 10 nm. (E) Representative WB of the indicated proteins in DLD1 cells following siRNA-mediated knockdown. (F) Dot plot of γH2AX foci number in EdU-labeled DLD1 BRCA2−/− cell nuclei as shown in (D); unpaired t test; ∗∗p < 0.01; ∗∗∗∗p < 0.0001. (G) Viability of DLD1 BRCA2−/− cells treated with increasing amounts of POLθi and the indicated siRNAs. Cell viability was assessed after 5 days, and it is expressed as a percentage of POLθi-treated viable cells relative to the untreated controls. Values represent the means ± SEMs of three independent biological replicates (n = 3); unpaired t test; ∗∗∗∗p < 0.0001. See also Figure S7.
	Figure 7. Proposed model for RAD51- and POLθ-mediated fork protection from MRE11 endonuclease (A) BRCA2/RAD51 dynamically associates to replication forks, protecting ssDNA emerging from parental dsDNA unwinding and facilitating POLα/POLδ-mediated lagging-strand DNA synthesis. (B) OKF stalling and ssDNA accumulation stimulate POLθ and BRCA2/RAD51 recruitment onto chromatin to process OKFs. POLθ polymerase extends OKFs, whereas POLθ helicase displaces 5′-flaps containing OKFs. (C) Without BRCA2/RAD51, ssDNA accumulates attracting POLθ, which fills inter-OKF gaps. (D) In the absence of POLθ, BRCA2/RAD51 binding to chromatin is sufficient to protect ssDNA. (E) In the absence of both POLθ and BRCA2/RAD51, the MRE11-NBS1/NBN-CtIP endonuclease gains access to gapped forks. (F) The MRE11-NBS1/NBN-CtIP endonuclease cleaves DNA producing broken forks.
	Figure S1. Isolation and characterization of Xenopus POLq. Related to Figures 1 and 2. A) Agarose gel showing PCR amplification of Xenopus POLq-FL cDNA. B) Multiple alignments of the protein sequences of X. laevis, H. sapiens and M. musculus POLq polymerase and helicase domains with amino acids coloured according to conservation and chemical nature (using Clustal X colour scheme). Protein sequences alignment was done using Clustal Omega and Jalview online tools. C) Coomassie stained gels with 2 µg of the recombinant proteins: 6H-MBP-POLq full length (POLq-FL), 6H-POLq polymerase-domain (POLq-POL) and POLq helicase domain (POLqHEL). D) Representative WB showing POLq protein in 0.5 µl egg extract (extract) and 20 ng of recombinant 6H-MBP-POLq-FL (POLq-FL). E) Representative WB showing the binding of the indicated proteins to replicating chromatin. Detection of POLq-FL was obtained using anti-Histidine antibody.
	Figure S2. Characterization of Xenopus POLq activity and sensitivity to APH. Related to Figure 3. A) Representative gel images showing time course of 5’-labeled fluorescent DNA oligo extension reaction, in the presence of POLq-FL (left) or POLq-POL (right) proteins (50 nM, final concentration). The not-extended primer fraction was quantified calculating the intensity of remaining substrate normalized to total DNA present in the reaction without POLq. Rectangle delimitate relevant lanes derived from the same gel. B) Representative gel image showing 5’-labeled fluorescent DNA oligo extension reaction in the presence of POLq-FL and the indicated APH concentrations. C) Representative gel images showing 5’-labeled fluorescent DNA oligo extension reaction using the indicated templates 1 or 2 in the presence of no enzyme, POLq-FL or T4 Polymerase (50 nM). The fractions of the primers that were halted in their extension (halted extension) or fully extended (full extension) were quantified by calculating the intensity of the respective products normalized to total DNA present in the reaction without enzyme. The not-extended primer fraction was quantified calculating the intensity of remaining substrate normalized to total DNA present in the reaction without enzyme. Rectangles delimitate relevant lanes derived from the same gel.
	Figure S3. Denaturing EM of DNA replication intermediates. Related to Figure 4. A) Representative EM image of a denatured replication bubble isolated from an interphase extract treated with DMSO containing buffer (Buffer). B) Representative EM image of denatured replication fork isolated from an extract treated with 1.5 mM APH added 60 minutes after sperm nuclei addition and incubated for an additional 45 minutes. Arrows indicate ssDNA at fork junction. Quantification of APH-induced RIs is reported in Figure 2F. C) Representative EM image of a reversed fork isolated from an extract treated as in (B). Quantification of reversed forks is reported in Figure S5I. D) Representative EM image of a denatured reversed fork isolated from an extract treated as in (B). Outlined windows show magnified section of the image highlighted in the dotted rectangle. Schematic drawings represent dsDNA (thick line) and ssDNA (thin line). Bar lengths expressed in nucleotides (nt) are indicated.
	Figure S4. Characterization of chromatin and RIs structures induced by APH in the absence of nuclease and fork remodeling activities. Related to Figure 4. A) Quantification of fork junction ssDNA gap length for each RIs isolated from extracts that were untreated (CTRL) or supplemented with APH and Buffer, Mirin (100 µM) or DNA2i (100 µM). Measurements were conducted on 280 RIs pooled from independent experiments (n=280); 70 RIs were counted for each treatment; horizontal bars indicate mean ± SD; unpaired t-test; ***p<0.0001; ns non-significant. B) Representative WB of the indicated proteins bound to chromatin isolated at the indicated times from the addition of sperm nuclei to mock or SMARCAL1-depleted extracts supplemented with buffer or APH. NS indicates no sperm nuclei addition. C) Graph showing percentage of reversed forks relative to total number of RIs isolated from mock or SMARCAL1-depleted extracts supplemented with DMSO or APH 1.5 mM. Percentage was calculated on three independent experiments (n=3); 70 RIs were counted for each treatment; columns indicate mean ± SD; unpaired t-test; p<0.05, ***p<0.001. D-G) Representative EM images of replication forks isolated from extracts treated with 20 µM APH (M-APH) added 60 minutes after sperm nuclei addition and incubated for additional 45 minutes containing at least two stalled OKFs. Red arrows indicate ssDNA between stalled OKFs. Blue arrows indicate flaps. Quantification for RIs represented by these images is reported in Figure 2F.
	Figure S5. Effects of POLqi and POLq-FL on DNA replication in unchallenged and stressful conditions. Related to Figure 4. A) Quantification of total ssDNA gap length for each replication fork represented in Supp Fig 5B-D and 5F-H isolated from extracts treated with Buffer (CTRL), POLqi 5 µM, POLai (CD437,100 µM), POLai and POLqi or POLq-FL (25 nM), PARPi (Olaparib, 100 µM), or POLai plus POLqi or POLq-FL. Chemical inhibitors were added 60 minutes after sperm nuclei addition and incubated for 45 minutes before DNA isolation. POLq-FL protein was added to egg extract with sperm nuclei. RIs were pooled from independent experiments (n=560); 70 RIs were counted for each treatment; horizontal bars indicate mean ± SD; unpaired t-test; **p<0.0001. Representative EM images of a replication fork isolated from extracts treated with (B) POLai, (C) POLai and POLqi or (D) POLq-FL. Red arrows indicate ssDNA between stalled OKFs. Blue arrows indicate flaps. E) Representative WB of the indicated proteins bound to chromatin isolated from extracts treated binding treated with DMSO or POLai (CD437,100 µM). Samples were taken at the indicated times after the addition of sperm nuclei to egg extract. Extract alone and sample with no sperm addition (NS) were also included. Representative EM images of a replication fork isolated from extracts treated with (F) PARPi, (G) PARPi and POLqi or (H) POLq-FL as indicated. I) Percentage of reversed forks represented in Figure S3C relative to total number of RIs for each of the indicated treatment condition. RIs were isolated from extracts treated with DMSO containing Buffer or APH in the absence or presence of POLq-FL (25 nM). Scoring was performed on 210 RIs pooled from independent experiments (n=210); 70 RIs were counted for each treatment; statistical analysis was conducted by unpaired t-test; p<0.01, ***p < 0.001.
	Figure S6. Effects of RAD51, POLq and MRE11 inhibition on RIs structure. Related to Figure 5. A) Representative EM image of a replication fork isolated from an extract supplemented with GST-BRC4 in the presence of APH. Red arrow indicates ssDNA. B) Quantification of chromatin binding intensity ratio of the band corresponding to RAD51 relative to histone H2B in three independent experiments (n=3) one of which is shown in Figure 5D. C) Quantification of chromatin binding intensity measured as optical density of the band corresponding to POLq of a representative experiment shown in Figure 5E. D) Scheme showing the formation of RIs with asymmetric broken arms. Normal Y-shaped RIs have two symmetric arms of similar length (within a ±10% error range) corresponding to the replicated strands. These are produced by PvuII-mediated restriction digestion of the replication bubbles following genomic isolation for EM analysis (Left). Asymmetric RIs lack symmetric branches due to spontaneous rupture or endogenous cleavage of one of the replicated strands, thus producing single-ended forks (Right). Broken forks are scored by measuring the strand length in RIs that contain three branches. RIs are classified as broken asymmetric forks when the length of each of the three branches differs from the others at least more than 10%. E) Representative EM image of converging replication forks isolated from an extract supplemented with GST-BRC4 and POLqi. F) Representative EM image of converging replication forks isolated from an extract supplemented with GST-BRC4, POLqi and PFM01. G) Graph showing percentage of broken converging forks relative to the experiment shown in Fig 5F. 70 RIs were counted for each treatment; statistical analysis was conducted by unpaired t-test; p<0.01; *p < 0.001.
	Figure S7. POLqi effects on DLD1 and DLD1 BRCA2-/- cells. Related to Figure 6. A) Quantification of cell viability showing survival of DLD1 and DLD1 BRCA2-/- cells in the presence of increasing concentrations of POLqi (left) from three independent experiments (n=3). Error bars indicate SD. B) Representative immunofluorescence images of labeled DNA fibers isolated from DLD1 and DLD1 BRCA2-/- cells that were treated with DMSO (UN) or incubated with POLqi. C) Graph showing fork rate in DLD1 and DLD1 BRCA2-/- cells treated as indicated. POLqi was present for the whole labeling time. Bars indicate interquartile values. Horizontal lines indicate median ± SD; unpaired t-test; p<0.05; p<0.001; *p<0.0001; ns non-significant. D) Dot plot of CldU tract lengths (μm) in DLD1 BRCA2-/- cells that were untreated or treated with 2 µM POLqi, or 2 µM POLqi and 25 µM PFM01, as indicated. Horizontal bar indicates the mean. p-values were calculated with Kruskal-Wallis test; p<0.05, *p<0.001, p<0.0001. E) Representative confocal microscopy immunofluorescence showing RPA (red) and EdUlabeled (green) in DLD1 and DLD1 BRCA2-/- cells treated with DMSO or POLqi as indicated. DAPI staining in blue. F) Dot plot of RPA32 immunofluorescence integrated intensity in EdU-labeled DLD1 BRCA2/- cell nuclei treated as indicated in (E). Horizontal bar indicates the mean. Statistical analysis was conducted by unpaired t test; p<0.0001. G) Quantification of fork arrest shown as IdU/CldU ratio in DLD1 and DLD1 BRCA2-/- cells treated as indicated. Bars indicate interquartile values; unpaired t-test; p<0.01; ns nonsignificant. H) Representative confocal microscopy immunofluorescence showing gH2AX (red) and EdUlabeled (green) in DLD1 and DLD1 BRCA2-/- cells treated with 2 µM POLqi or DMSO as indicated for 24 h. DAPI staining in blue. Bars indicate 10 µm. I) On the left a dot plot shows total gH2AX foci number in EdU-labelled DLD1 and DLD1 BRCA2-/- cell nuclei treated as indicated. Horizontal bar indicates the median. Statistical analysis was conducted by unpaired t-test; **p<0.0001; *p<0.001. On the right a bar graph shows the percentage of EdU-labelled DLD1 and DLD1 BRCA2-/- cells with more than 10 gH2AX foci in two independent experiments; columns indicate mean ± SD. J) Percentage of broken forks represented in Figure 6C relative to total number of RIs isolated from DLD1 and DLD1 BRCA2-/- cells treated for 16 h with 2 µM POLqi. Scoring was performed on 120 RIs pooled from two independent experiments; 60 RIs were counted for each condition; columns indicate mean ± SD. K) Representative images showing gH2AX immunofluorescence (red) in EdU-labeled DLD1 and DLD1 BRCA2-/- cells treated with DMSO, POLqi or POLqi and PFM01. DAPI staining in blue. Bars indicate 10 µm.