Fig 2: Pol3 depletion leads to RPA accumulation on chromatin and checkpoint activation.A. Western blot against Rfa1 on soluble and chromatin-bound fractions of log-phase POL3-AID cells treated with the indicated IAA concentration for 2h. B. Western blots as in (A), but using cells collected during G1 arrest, or in mid- or late S-phase as indicated. C. Western blots to detect Rad53 phosphorylation in asynchronous cultures of POL3-AID cells grown in YPD supplemented with IAA. D. Serial dilution spot tests of POL3-AID or POL3-AID mec1Δ sml1Δ cells exposed to 1 mM IAA for 0, 2 or 4h of growth during logarithmic phase, followed by plating on YPD without IAA. E. DNA content measured by flow cytometry for POL3-AID or POL3-AID mec1Δ sml1Δ cells released from ⍺-factor-mediated G1 arrest at 30°C. As in Fig 1C, individual cultures were treated with the indicated concentration of IAA for 2h during the initial arrest, and subsequently released into media containing the same concentration of IAA.

Fig 3: The chimera Rfa1-MN provides a genetic system to study the repair of DSBs at replication forks.(A) ChEC analysis of RFA1-MN cells arrested in G1 with α-factor and released into S phase for 30 minutes. Total DNA from cells permeabilized and treated with 2 mM CaCl2 for different times is shown, as well as the FACS profiles. A scheme with the rational of the ChEC approach is shown on the left. (B) 2D/ChEC analysis of replication intermediates of RFA1-MN cells synchronised in G1 with α-factor and released into S phase for 30 minutes. Total DNA from cells permeabilized and treated with Ca2+ for different times was digested with specific restriction enzymes and analysed by 2D electrophoresis. A schematic representation of the migration pattern of replication intermediates is shown on the right. (C) Rfa1 expression in wild-type and RFA1-MN cells from exponentially growing cultures as determined by western blot analysis. (D) MMS and HU sensitivity of RFA1-MN cells. Wild-type and rad52∆ cells were included as controls. (E) DSB sensitivity of RFA1-MN cells transformed with pGAL-HO and grown in glucose (GAL1p repression) and galactose-containing medium (GAL1p activation). An HO-induced DSB at the MAT locus can be repaired by NHEJ or, preferentially, by HR with the HMR or HML donor. The analysis was performed in wild-type and ku70∆ background (defective in NHEJ). (F) Rad53 activation in wild-type and RFA1-MN cells as determined by western blot analysis of exponentially growing cultures either in the absence or presence of 0.005% MMS for 1 hour. (G) Cell cycle progression of wild-type and RFA1-MN cells synchronised in G1 with α-factor and released into S phase for different times as determined by FACS analysis. (H) RFA1-MN rad52∆ lethality as determined by tetrad analysis. (I) Effect of restricting Rad52 expression to G2/M in wild type (G2::cRAD52) and RFA1-MN cells (RFA1-MN G2::cRAD52). (J) HO-induced DSB repair in cells defective in HR (rad52∆) and/or NHEJ (ku70∆). Cells were transformed with pGAL-HO and grown in glucose (GAL1p repression) and galactose-containing medium (GAL1p activation). (K) Proposed model for the essential role of HR in Rfa1-MN expressing cells. (L) Effect of the pif1∆, pol32∆, yen1∆, mus81∆ and mus81∆ yen1∆ mutations in the growth of RFA1-MN cells in the absence and presence of different CaCl2 concentrations. At high concentration, CaCl2 can form crystals that did not affect the reproducibility of the assay. (D-E, I-J, L) Cell growth was determined by spotting 10-fold serial dilutions of the same number of mid-log growing cells onto the indicated mediums. All the analyses were repeated at least twice with similar results.

Fig 4: Checkpoint, replication fork stability factors and cohesins facilitate the repair of DSBs at forks.(A–F) Effect of the indicated mutations in the growth of RFA1-MN cells as determined by spotting 10-fold serial dilutions of the same number of mid-log growing cells onto SMM medium without or with the indicated concentrations of CaCl2. Mutants scored in the SGA screening are shown in bold. (G) Checkpoint activation of the indicated strains transformed with plasmid pMDL5 (expressing RAD52 under control of the GAL1 promoter) in galactose or after 9 hours in glucose, as determined by western blot against Rad53. The analyses were repeated at least twice with similar results.

Fig 5: Proposed mechanisms for the repair of a DSB at forks.After fork breakage by Rfa1-MN, HR might be operating at a deDSB generated either between Okazaki fragments (A) or by collapse of a converging fork with the gap left at the non-broken strand (B), or at a seDSB generated after cleavage of the fork junction, more likely at the lagging strand (C). In response to seDSBs, the core HR machinery (with the help of fork-associated HR factors) would promote the invasion of the sister chromatid, generating a D-loop structure that primes a conservative, error-prone replication by a migrating bubble. This BIR-like restart mechanism would be facilitated by cohesins, checkpoint activation, Dun1-mediated increase in dNTPs, and replisome components that would be retained at the proximity of the D-loop for the stability of the migrating D-loop or, alternatively, the conversion of this structure into a canonical fork upon the activity of the Mus81 nuclease. The formation of the D-loop structure would prevent checkpoint activation and inhibition of late replication origins. The activation of these origins would also prevent BIR-associated genetic instability by fork merging with the D-loop structure and subsequent Mus81/Yen1-dependent HJ resolution. In line with this later mechanism, regulation of G1 length by Sic1, Cdh1 and Whi5 would facilitate the rescue of broken replication forks by ensuring a sufficient number of active origins, especially in response to massive fork breakage or fork breakage at specific regions like the end of chromosomes or common fragile sites.