Timely lagging strand maturation relies on Ubp10-mediated PCNA dissociation from replicating chromatin

Synthesis and maturation of Okazaki Fragments is an incessant and highly efficient metabolic process completing the synthesis of the lagging strands at replication forks during S phase. Accurate Okazaki fragment maturation (OFM) is crucial to maintain genome integrity and, therefore, cell survival in all living organisms. In eukaryotes, OFM involves the consecutive action of DNA polymerase Pol ∂, 5’ Flap endonuclease Fen1 and DNA ligase I, and constitutes the best example of a sequential process coordinated by the sliding clamp PCNA. For OFM to occur efficiently, cooperation of these enzymes with PCNA must be highly regulated. Here, we present evidence of a role for the PCNA-deubiquitylase Ubp10 in the maturation of Okazaki fragments in the budding yeast Saccharomyces cerevisiae. We show that Ubp10 associates with lagging-strand DNA synthesis machineries on replicating chromatin to ensure timely ligation of Okazaki fragments by promoting an Elg1ATAD5-independent PCNA unloading mechanism. This document was written without the use of AI. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 2

The POL30 gene of Saccharomyces cerevisiae encodes the sliding clamp PCNA (Proliferating Cell Nuclear Antigen), a conserved ring-shaped protein with essential roles in DNA metabolism as a crucial component of replication and repair machineries. PCNA forms an homotrimer that encircles DNA, where it interacts with a staggering number of proteins involved in every step required for DNA replication or repair. Thus, whereas it is devoid of enzymatic activity itself, PCNA exerts its function by recruiting and, in many instances, also activating numerous interactors1,2. DNA polymerases require free 3’-OH groups to initiate DNA synthesis, therefore, they can only synthesize DNA in a 5’-3’ direction. As a result of this and the antiparallel nature of double stranded DNA, replication of one of the strands is discontinuous through the generation of Okazaki Fragments (OF). This strand is called the lagging strand (in contrast to the leading one). PCNA plays a crucial role in the synthesis and maturation of these OFs. Thus, PCNA is loaded on dsDNA at primer-template junctions to recruit Pol d and enhance its processivity. When the Pol d-PCNA complex collides with the 5’-end of the preceding OF, it displaces a short flap that is cleaved off by the structure-specific flap endonuclease-1 Fen1, Rad27 in S. cerevisiae, upon binding to PCNA. This process generates a nick in the nascent DNA that is sealed by DNA ligase I, Cdc9 in S. cerevisiae, which is also recruited and catalytically activated through its interaction with PCNA3. All three subunits of DNA Pol d, Fen1/Rad27, and DNA ligase I/Cdc9 harbor PIP (PCNA-interacting peptide)-boxes, through which they interact in a coordinated manner with the Inter-Domain Connecting Loop (IDCL) of PCNA, a major interaction site in the sliding clamp4–6. Although the mechanisms and factors involved in lagging strand maturation have been extensively studied key molecular details of OF maturation remain poorly understood. To ensure the successful completion of DNA replication, particularly of the processive synthesis of the lagging strands, continuous recycling of chromatin-bound PCNA is required. The sliding clamp is loaded when required and actively unloaded when no longer needed in order to suppress illegitimate enzymatic reactions2,7,8. RFC (Replication factor C) and RFC-like complexes (RLCs) mediate the loading and unloading of PCNA (reviewed in9). There is a general consensus regarding Rfc1-RFC complex role as the main loader of PCNA on replicating chromatin10,11. One additional complex, Ctf18-RLC, also acts as a PCNA loader, although Ctf18 cannot substitute Rfc1. Moreover, Rfc1 and Ctf18 show a strand preference during replication, with a (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 3 lagging strand bias for Rfc1 and a slight leading strand preference for Ctf1812–14. Both Rfc1 and Ctf18 also exhibit some ability to unload PCNA in vitro. Human Rfc1-RFC, but not yeast Rfc1, is able to unload PCNA in vitro in an ATP-dependent manner15,16 and yeast Ctf18-RLC promotes an in vitro PCNA unloading mechanism, that also requires ATP hydrolysis, in the presence of ssDNA coated with RPA17. Despite these in vitro results, PCNA unloading activities for Rfc1- or Ctf18- complexes in vivo have not been reported. Elg1-RLC complex (ATAD5-RLC in mammals) is considered as the major PCNA unloader when the role of PCNA in DNA replication is completed8,18–22. However, given that ELG1 is a non-essential gene for cell division23–25 and that PCNA accumulated in Elg1-depleted cells ends up being removed from chromatin before M phase18,20, it is likely that additional PCNA unloaders are required during chromosome replication, at least in the absence of Elg1. Therefore, it remains unclear whether Elg1-RLC is the only in vivo PCNA unloader22,26. PCNA functions are regulated by different post-translational modifications such as SUMOylation, ubiquitylation, phosphorylation or acetylation, which confer PCNA the necessary plasticity to interact with its different binding partners1,27,28. In the face of DNA lesions, PCNA is ubiquitylated to mediate damage-tolerance mechanisms that allow circumventing DNA lesions and prevent replication fork stalling29,30. PCNA is mono-ubiquitylated at K164 by the evolutionary conserved RAD6/RAD18 (E2/E3) ubiquitin ligase complex to switch its affinity from replicative polymerases to damage- tolerant translesion synthesis (TLS) DNA polymerases, which, although mutagenic, are capable to bypass damaged bases27,31,32. Furthermore, polyubiquitylation of the same residue by the Rad5/Mms2/Ubc13 PCNA-ubiquitin ligase complex leads to template switching (TS), the DNA damage tolerance (DDT) error-free pathway, to overcome the potentially lethal effects of replication fork stalling27,30. We and others reported that the precise regulation of these processes not only depends on writer enzymes, PCNAK164- Ubiquitin ligase complexes, but also on the erasers of these modifications. Thus, both TLS and TS pathways are limited by PCNA-deubiquitylation processes to minimize their deleterious cellular side effects. In mammals, deubiquitylating enzymes Usp1, Usp7, and Usp10 revert PCNA ubiquitylation caused in response to DNA damage28,33– 35. Knockdown of USP1 induces aberrant PCNA monoubiquitylation, enhanced recruitment of error-prone TLS polymerases and increased mutagenesis levels in human cells33,35. In the case of the budding yeast S. cerevisiae, the PCNA-DUBs Ubp10 and Ubp1236,37 limit the extent of DDT processes during the progression of (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 4 exogenously unperturbed S phase by reverting K164-ubiquitylation of the sliding clamp at replication forks37. Ubp10 ubiquitin-protease has also a role deubiquitylating histone H2BK123 working at genomics sites distinct of the SAGA-related histone H2BK123 DUB Ubp8 sites38–40. Moreover, Ubp10 activity is also involved in the regulation of RNA polymerase I stability41. Remarkably, despite the key roles of this PCNA-DUB, abrogation of the UBP10 gene is viable, even though ubp10 mutated cells have both growth and cell cycle progression defects36,37,41,42. Suppression of either growth or cell cycle defects can be accomplished by mutation of specific targets. Thus, multiple deletion of the TLS polymerases (REV1, REV3 and RAD30) rescues the cell cycle delay in S phase progression caused by the abrogation of Ubp1037, indicating that the role of Ubp10 in supporting normal replication rates through PCNA-K164 deubiquitylation is dependent, at least in part, on the TLS pathway. A number of active roles in the regulation of key cellular mechanisms has been described for ubiquitin-signalling writers, but not so many for erasers. Along this line of thought, it is assumed that pivotal regulatory steps rest on ubiquitin writers while redundant erasers are considered to act automatically or spontaneously after the post- translational modification is added to a given substrate. For these reasons ubiquitin proteases are in general considered to play a minor role, if any, in regulatory controls. In this context, it is assumed that PCNA ubiquitylation is counteracted by constitutive deubiquitylation mediated by redundant PCNA-DUBs43,44. However, recent work with yeast models concerning PCNA-DUBs role in DNA replication shows that something is amiss with this scenario both in fission and budding yeast37,43. Regarding this matter, Ubp10 has a remarkable slow S phase phenotype that we were very interested to understand in full. PCNAK164 ubiquitylation has also been linked to the OFM process. Thus, loss of PCNA ubiquitylation seems to cause inefficient gap-filling which interferes with efficient OF ligation in fission yeast and human cells44,45. Moreover, PCNAK164 ubiquitylation suppresses replication stress resulting from Fen1/Rad27-defective flap processing during OFM46, and the PCNAK164R mutant shows inefficient OF processing in an in vitro DNA replication reconstitution assay using yeast proteins47. Here, we present evidence of the role of Ubp10 in the regulation of OFM. Mass spectrometry analysis of the Ubp10 interactome showed that this PCNA-DUB interacts with all major components of the OF metabolism. Based on this observation and on the cell cycle delay in S phase, characteristic of cells deficient in Ubp10, we focused on deciphering the potential link between Ubp10 and OFM processes. Ablation of Ubp10 leads to accumulation of (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 5 unligated OFs and a markedly increase of chromatin-bound PCNA during S phase. POL30 mutants that conform unstable PCNA homotrimers on chromatin, particularly pol30R14E and pol30D150E alleles, counteract these ubp10∆-associated replication defects, as well as cell cycle delay. In addition, abrogation of ubp10 is strongly additive to elg1 depletion, resulting in substantial increase of the PCNA bound to chromatin during replication. These data indicate that timely dissociation of PCNA during lagging strand synthesis requires Ubp10 in an Elg1/ATAD5 independent manner and is important for OF ligation. All this evidence reveals an important regulatory role for Ubp10 at the lagging strand synthesis.

PCNA-DUB Ubp10 physically interacts with core components of the lagging- strand synthesis machinery. We first analyzed the interactome of the H2B- and PCNA-ubiquitin protease Ubp10 to reveal partners of this DUB during DNA replication by LC-MS-MS analysis. Mass spectrometry analysis of the proteome associated with Ubp10 in unperturbed S phase cells retrieved all major components of the lagging strand synthesis machinery (such as PCNA, Pola/primase, RFC replication clamp loader, DNA polymerase ∂) and Okazaki fragment metabolism (Fen1 flap endonuclease, Cdc9 ligase, RNase H2) (Figure 1). A complementary proteomic analysis of DNA ligase Cdc9 also revealed Ubp10 as one of the DNA ligase interactors (to be published elsewhere). Relevant interactions were confirmed by direct co-immunoprecipitation (ChIP-CoIP) analyses in epitope-tagged backgrounds, as shown for Pol3, catalytic subunit of Pold, (Supplementary Figure 1A), and Cdc9 (Supplementary Figure 1B). Ubp10-PCNA and Ubp10-Fen1 ChIP-CoIP interactions have been described previously36,37. This evidence suggests the existence of a functional link between Pold, Fen1, Cdc9, Ubp10 and PCNA, and, therefore, supports the hypothesis that Ubp10 works on the lagging strand during OFM. Likely significant, our Ubp10-GFP-trap proteomic approach did not detect RLC cofactor Elg1. In contrast, all Rfc1-RFC subunits were identified as Ubp10 unperturbed S phase interactors. Of interest, FACT subunits (Spt16 and Pob3) and RNA pol I subunits (Rpa190, 34, 43, 49 and 135), comprising known functional interactors of Ubp10 Spt16 and Rpa190 41,48, were identified with high scores in our proteomic analyses validating our experimental approach (Supplementary Figure 1C). (All these detected Ubp10 interactions somehow endorse our LC-MS-MS analysis). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 6 The PCNA-DUB Ubp10 acts downstream of the Fen1Rad27 FLAP-endonuclease during DNA replication. Our Ubp10-GFP-Trap proteomic approach shows that Ubp10 interacts physically with major components of the lagging strand synthesis machinery during S phase. Therefore, looking for a functional support of the proteomic evidence, a molecular analysis was designed to reveal a potential role of Ubp10 in the metabolism of OFs. Initially, we focused in discerning the potential interplay of FEN1 and UBP10 in the OFM pathway. We had recently shown that Ubp10 and the Fen1 Flap-endonuclease physically interact in early S phase, as detected by co-immunoprecipitation37. To further understand the functional relevance of this observation, we studied genetic interactions occurring in strains ablated for both replication proteins. We first monitored bulk DNA replication in synchronized cell cultures and observed that depletion of Fen1 suppressed the replication defect characteristic of ubp10 mutants, with ubp10 fen1 cells showing replication dynamics virtually indistinguishable from that of fen1 single mutants (Figure 2A). We then tested fork transitions upon replication stress induction by dNTP pool depletion, in particular by examining the accumulation of anomalous small Ys in ubp10∆ cells37. We found that replication intermediates accumulating in ubp10 fen1 double mutants are very similar to those of fen1 single mutant cells and clearly differ from those of ubp10∆ cells, lacking the characteristic small Y accumulation (Figure 2B), indicating that FEN1 deletion prevents anomalous nascent strand transitions at stalled forks caused by Ubp10 absence. The functional nature of the genetic interaction was further confirmed by testing thermosensitivity and resistance to chronic exposure to HU, where we found that fen1∆ ubp10∆ cells phenocopy single fen1∆ mutants (Figure 2C), indicating that Ubp10 and Fen1 support viability through a single genetically-related pathway and that Fen1 underlies ubp10∆ sensitivity to replication stress. Finally, chromatin fractionation assays in cells synchronously traversing S phase failed to show differences in Fen1 protein accumulation on chromatin in wild-type and ubp10 defective cells (Figure 2D), suggesting that Ubp10 does not markedly influence Fen1 chromatin association during OFM. Taken together, these analyses suggested that absence of the Fen1 endonuclease, required for Okazaki fragment flap-processing, bypasses a yet undefined replication related Ubp10-dependent function. Ubp10 promotes Okazaki fragment ligation. In order to explore the role of the Ubp10 in lagging strand synthesis, we analysed chromatin binding of key OFM proteins during unperturbed S phase upon Ubp10 depletion. We show a control assay displaying a chromatin fractionation experiment in (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 7 full (Supplementary Figure 2), which allows analyzing the partitioning of proteins of interest into chromatin-free and chromatin-bound fractions. For clarity, we only show chromatin-bound fractions in further fractionation assays (including Figure 2D), unless otherwise stated. PCNA is an abundant protein, detectable throughout the cell cycle, that temporarily associates to chromatin during in S phase, in discrete but detectable amounts, in cycles of loading and unloading as the process of DNA synthesis requires1,8,26. Indeed, we observed that PCNA is readily detectable in WCE and chromatin-free fractions, and to a lesser extent on chromatin-bound fractions of cells synchronously undergoing genome replication (Supplementary Figure 2). Therefore, during replication, a major fraction of PCNA is detectable in soluble forms while a small subset associates to chromatin with a quite reproducible periodicity. In S. cerevisiae, CDC9 encodes the DNA ligase I, an essential enzyme that seals Okazaki fragments during DNA replication49,50. Cdc9 has a human homolog, LIG1, also regulated by PCNA during the sealing of nicked DNA at lagging strands51. Remarkably, human LIG1 complements yeast cdc9 temperature-sensitive mutants at the restrictive temperature52. Having observed that in budding yeast PCNA-DUB Ubp10 physically interacts with Cdc9 during S phase, we were interested in understanding a possible functional interaction among them. We studied genetic interactions of UBP10 with a conditional allele of the DNA ligase I by testing the viability of cdc9ts and ubp10D double mutant cells. In our study we used a cdc9-7 allele (from the National BioResource Project, NBRP Japan) a W303 derivative strain that we characterized and sequenced to find that it is synonymous to the cdc9-1 allele53. We found that depletion of Ubp10 aggravates dramatically the thermosensitivity of strains carrying this CDC9 temperature-sensitive mutant allele (Figure 3A). We further observed - by chromatin fractionation assays - that abrogation of UBP10 function in the cdc9-7 ts mutant lead to the accumulation of PCNA on chromatin for a longer period of time than in cdc9-7 controls, which also accumulated PCNA on chromatin when compared to wild-type cells during S phase even under permissive conditions (25ºC) (Figure 3B). These observations indicate that the cdc9-7 allele is - to some extend - defective in nick ligation at 25ºC and clearly suggest that all its defects are aggravated by Ubp10 ablation. However, cdc9-7 mutation did not rescue the characteristic slow bulk DNA replication phenotype of ubp10D mutant (Supplementary Figure 3A), suggesting that OF ligation is unrelated to the ubp10∆ replication defect. We also noticed that deletion of UBP10 in cdc9-7 led to an accumulation of unusual replication intermediates by 2D-gel analysis of HU-treated (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 8 cells equivalent to those observed in ubp10D single mutants (Supplementary Figure 3B). Okazaki fragments can be detected in vivo upon DNA ligase I CDC9 depletion or using conditional alleles of the ligase, which result in the accumulation of nicked DNA49,50. To test a potential accumulation of Okazaki fragments in cdc9-7 ubp10D cells and, particularly, after having observed that cdc9-7 cells grow poorly at 29ºC when combined with deletion of UBP10, we set up cultures of exponentially growing cells incubated at 25ºC to then shifted them to 29ºC. Okazaki fragments were end-labelled in samples taken at one-hour intervals, separated by denaturing agarose gel electrophoresis, transferred to a nitrocellulose membrane, and visualized with a Phosphor Imager. We detected a transitory accumulation of Okazaki fragments in cdc9-7 cells upon sifting from 25º to 29ºC degrees cultures of cells growing exponentially. After end-labelling of DNA and denaturing electrophoresis, we observed the characteristic banding pattern of short and heterogenous nicked DNA that results from defects in the ligation at lagging strands50 (Figure 3C). Remarkably, we also found that cdc9-7 ubp10D cells accumulated OFs more abundantly than cdc9-7 control cells (Figure 3C). We did not detect this characteristic banding pattern of Okazaki fragments in nick ligation competent wild-type cells or single ubp10D mutants tested. In essence, all this evidence indicates that depletion of Ubp10 leads to a strong lagging-strand replication defect phenotype. However, it is important to understand whether this strong phenotype is the cause or the consequence of the slow progression through S phase that characterizes the genetic depletion of Ubp10. Cells with a defective Flap-endonuclease Fen1/Rad27 accumulate unprocessed Okazaki fragments with poorly ligatable ends due to the accumulation of flaps with different sizes54 (and our unpublished observations). Furthermore, impairment of PCNA unloader Elg1 function leads to the accretion of extended Okazaki fragments likely reflecting defective post-replicative nucleosome reposition55. However, in vivo depletion of Cdc9 DNA ligase I causes the accumulation of nicks that are, therefore, in vitro ligatable with no obvious defects in coupling with chromatin assembly50,54. To characterize the molecular nature of the Okazaki fragments accumulating in cdc9-7 ubp10D double mutants, we examined their size by gel electrophoresis and tested the extent to which the isolated DNA fragments are competent for ligation after purification of total DNA (Figure 3D). We observed that cdc9-7 ubp10D cells accumulate Okazaki (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 9 fragments of normal length (Figure 3C). In addition, these are in vitro ligated by T4 DNA (Figure 3D), with an efficiency equivalent to ligase deficient controls50,54. These

indicate that OFs accumulating upon Ubp10 ablation have DNA ends equivalent to those accumulated in cdc9ts mutants and suggest that OF accumulation is due to defects in the last steps of lagging strand maturation. The evidence shown so far suggests a role for Ubp10 in the timely maturation of Okazaki fragments, either in the regulation of nick ligation or in promoting chromatin disassociation of PCNA. Ubp10 is required for timely ligase association to replicating chromatin. PCNA accumulates on replicating chromatin in the absence of Cdc9-mediated Okazaki fragment ligation21. We reasoned that the PCNA accretion observed in Cdc9 Ubp10 doubly depleted cells might be the consequence of defects either in the DNA ligase activity or in the chromatin abundance of Cdc9. Therefore, we next tested a heterologous DNA ligase that can complement Cdc9 depletion56, and found that ubp10D cells overexpressing Chlorella virus DNA ligase exhibit a delay in S phase progression similar to that of controls (Supplementary Figure 4A), strengthening the

that ubp10D deficiency is not related with defects in DNA ligase activity. We then monitored Cdc9 chromatin association in synchronously replicating cells and found that Ubp10 depleted cells show transiently reduced ligase levels compared to wild type cells (Figure 4), a decrease particularly marked at early time points after G1 release, coinciding with the slow progression of bulk DNA replication and in an inverse correlation with PCNA abundance (Supplementary Figure 4C). Therefore, the accumulation or permanence of PCNA on chromatin in ubp10D mutants might be the consequence of a slow recruitment of DNA ligase Cdc9 on chromatin. To further understand whether the poor accretion of Cdc9 on chromatin was the cause or a consequence of the slow S phase phenotype of Ubp10 depleted cells, we tested ubp10D cells overexpressing CDC9, by means of a duplicated galactose-inducible allele, and found that high levels of chromatin-bound Cdc9 did not rescue the characteristic cell cycle defect, lengthy S phase, of ubp10 mutants nor rescued the accumulation of PCNA (Supplementary Figures 4B and 4C). PCNA disassembly-prone mutants pol30R14E and pol30D150E revert ubp10∆- associated replication defects. We reasoned that the excessive abundance of PCNA on chromatin in Ubp10 depleted cells might be related to defective unloading of the sliding clamp. If ubp10∆ phenotypes (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 10 are related with the excessive retention of PCNA on chromatin, two predictions can be made. Firstly, cells depleted for Ubp10 should accumulate PCNA on chromatin during S phase. Secondly, reduction the amount of PCNA bound to chromatin would rescue ubp10 replication defects. Regarding this second prediction, there are a number of POL30 alleles that conform unstable PCNA homotrimers57. In particular, pol30R14E and pol30D150E alleles are described as PCNA trimer-disassembly-prone mutants due to their instability when bound to chromatin21,57. Hence, if the reason underlaying ubp10 defects is related to excessive PCNA on chromatin, the potential suppression by these disassembly-prone PCNA mutants might be easier to observe in a cdc9-7 ubp10∆

due to the greater thermosensitivity as compared to single cdc9-7 cells. In fact, the growth defect of cdc9-7 ubp10∆ is suppressed by pol30R14E and pol30D150E mutant alleles, while they did not rescue cdc9-7 thermosensitivity (Figure 5A). Furthermore, these two-point mutant alleles of PCNA also rescue abnormal replication intermediates accumulated in cdc9-7 ubp10∆ cells arrested in HU (Figure 5B). In fact, these POL30 alleles rescue all tested Ubp10-depletion associated S phase defects, including those related with bulk genomic DNA replication in a CDC9 wild-type-like strain (Figure 5C), implying that the slow S phase observed in Ubp10 depleted cells is a consequence of PCNA accumulation on replicating chromatin. Defective replication intermediate accumulation in HU-treated UBP10 deleted cells is also efficiently suppressed by either pol30R14E or pol30D150E (Figure 5D). This evidence indicates that the replication phenotypes caused by Ubp10 ablation are a consequence of a defective PCNA unloading mechanism during S phase. UBP10 mutants accumulate non-canonical small Y-shaped replication intermediates upon HU-induced fork stalling37. The accumulation of these non-canonical small Ys is characterized by a decrease in large Ys and is abated by mutating RAD5237. We re- examined the accumulation of these small Y-shaped molecules and found that is suppressed in pol30R14E and pol30D150E genetic backgrounds (Figures 5B, 5D). These

strongly suggest that small Y-shaped molecules formed as a direct consequence of PCNA accumulation on replicating chromatin (and, also, that these abnormal structures are generated as cells try to repair through a Rad52-dependent TS-like mechanism). Disassembly-prone-mutant pol30D150E suppresses the increased chromatin association of PCNA as well as the OF accumulation in UBP10 defective cells. As mentioned, if ubp10 phenotypes are the consequence of a defective unloading of PCNA, cells depleted for Ubp10 should accumulate the sliding clamp on chromatin (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 11 during S phase, indeed we have observed a chromatin-bound accretion in ubp10∆ cells when testing Cdc9 levels (Figure 4A and Supplementary Figure 4C). We next re- evaluated how much and for how long the sliding clamp PCNA is bound to chromatin through a synchronized, otherwise unperturbed, S phase in wild-type and ubp10∆ cells. Exponentially growing cultures of either wild-type or ubp10∆ strains were synchronized in G1 with a-factor pheromone and release in fresh media. Samples were taken at 0, 40, 55 and 70 minutes after the release and processed for chromatin fractionation assays and DNA content analysis. As expected from previous results (Figure 5C), we found that the slow S phase progression in ubp10∆ cells correlates with an increased PCNA accumulation on chromatin (Figure 6A and 6B). In parallel, we also evaluated the disassembly-prone pol30D150E point-mutant ability to suppress this chromatin-bound PCNA accretion phenotype and found that this PCNA mutant rescued chromatin retention of PCNA of ubp10∆ (Figure 6A), as well as the bulk DNA replication defect (Figure 5C and 6B). Having shown that PCNA accumulation on replicating chromatin underlies not only the slow S phase progression but also the formation of anomalous small Ys in ubp10 defective cells, we next tested whether the PCNA disassembly-prone pol30D150E allele would be able to rescue the accumulation of unligated Okazaki fragments that can be evidenced in the ubp10 mutant using a cdc9-7 background. It has been shown that pol30R14E and pol30D150E PCNA trimer-disassembly-prone mutants alleviate the Okazaki fragment length extension problem described in elg1D cells55, in agreement with PCNA unloading being a key event in maturation of the lagging strand and nucleosome deposition. We predicted that pol30D150E would mitigate the OF accretion of cdc9-7 ubp10∆ cells. To test this hypothesis, exponentially growing cdc9-7, cdc9-7 ubp10∆, pol30D150E and pol30D150E ubp10∆ cells incubated at 25ºC were shifted to 29ºC, and samples were taken at one-hour intervals and processed for OF analysis. We observed that the pol30D150E point mutant suppresses ubp10∆ Okazaki fragments maturation timing defect (Figure 6C), implying that the main problem caused by Ubp10 depletion during lagging strand replication is PCNA retention on chromatin. Ubp10 as a key regulator of a PCNA unloading mechanism distinct from Elg1- RLC Thus far, our results argue that Ubp10 promotes timely PCNA unloading during lagging strand replication. One reasonable hypothesis is that Ubp10 promotes PCNA deubiquitylation to enhance Elg1-mediated PCNA unloading at the final steps of Okazaki fragment maturation. Fully aware of in vitro evidence showing that human (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 12 ATAD5Elg1 is able to unload both ubiquitylated and deubiquitylated PCNA forms with similar efficiency58, we reasoned that in vivo, during unperturbed DNA replication, chromatin PCNA unloading would be enhanced by deubiquitylated forms of the sliding clamp in order to proceed smoothly and timely through lagging strand synthesis. Yeast Elg1ATAD5 is an evolutionary conserved homolog of the replication factor C (RFC) subunit Rfc123,25. It has been shown that while Rfc1-RFC loads PCNA on replicating chromatin, the Elg1-RLC complex has a role in PCNA unloading in a molecular event preceded by Okazaki fragment ligation18,21. Elg1 forms an alternative RFC hetero- pentameric complex with all RFC2-5 subunits of RFC. Significantly, this alternative Elg1-RLC complex is important but not essential for DNA replication24. Elg1 interacts physically with PCNA and Fen1Rad27, has a role in PCNA unloading during Okazaki fragment maturation and is, therefore, important for efficient S phase progression, and likely has a role in proper nucleosome assembly18,25,55. We have observed that, though Elg1 displays many phenotypes related to chromosome stability, depletion of ELG1 in budding yeast do not cause major replication delays as assayed in synchronous S phase (Supplementary Figure 5A). In fact, bulk DNA replication timing in ELG1 mutants is equivalent to wild-type replication as tested by FACS DNA content analysis (Supplementary Figure 5A). Moreover, deletion of ELG1 is viable while deletion of other RFC components is not (in particular RFC1)24,26 (Supplementary Figure 5B). However, in 10-fold dilution assays we detected that elg1∆ strains show a poor growth rate particularly at high temperatures (Supplementary Figure 5B). A defect exacerbated when single elg1∆ mutation is combined with the deletion of Ubp10 (ubp10∆ elg1∆) at any tested temperature, as compared to single mutants or wild type cells (Supplementary Figure 5B). This semi- lethality is indicative of a genetic interaction suggestive of a role for both factors in a common event, likely PCNA unloading. To test chromatin-bound PCNA levels throughout S phase in wild-type, double mutant elg1∆ ubp10∆, and single mutants elg1∆ and ubp10∆, mid-log phase cultures of the indicated strains were synchronized in G1 with a-factor and released in fresh media. As in previous experiments, samples were taken at indicated time points and processed for chromatin pellet assays and DNA content analysis (Figure 7). We initially expected elg1∆ ubp10∆ double mutants to behave like elg1∆ singles in accordance with the hypothesis that Ubp10 might regulate PCNA unloading through Elg1. Unexpectedly, we found that ubp10D and elg1D, when combined, are additive (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 13 regarding PCNA accumulation on replicating chromatin (Figure 7A and 7B), strongly suggesting the existence of two distinct pathways of PCNA unloading. These results support the existence of a chromatin disassociation of PCNA mechanism regulated by Ubp10 separable from the Elg1ATAD5-dependent PCNA unloading. Even though the proteomic analysis presented in this work revealed no interaction with the RFC cofactor Elg1, we examined whether the ablation of Ubp10 altered the chromatin binding pattern of Elg1 during a synchronized S phase. By chromatin fractionation assays, we found that depletion of the PCNA-DUB does not alter Elg1 interaction with chromatin (Supplementary Figure 5C). Therefore, we concluded that UBP10 mutants do not deregulate Elg1 interaction with chromatin during S phase. Significantly, cells depleted for both UBP10 and ELG1 do not exacerbate the ubp10∆ deletion slow S phase phenotype, though they accumulate PCNA abundantly, far more than individual mutants (Figure 7A and 7B). The fact that depletion of UBP10 alone has a replication progression defect underpins the importance of the PCNA-DUB Ubp10- dependent mechanism of PCNA unloading during lagging strand synthesis.

Okazaki fragment maturation is a complex, yet well understood, process in the synthesis of the lagging strand during DNA replication. Here, we unveil a role of the ubiquitin protease Ubp10 in the latest steps of maturation of Okazaki fragments in the model yeast S. cerevisiae. In yeast, the Ubp10 enzyme have been functionally related to RNA polymerase I through the stabilization of the Rpa190 subunit41. Ubp10 also cooperates with the FACT complex in the maturation of nucleosomes48. Furthermore, Ubp10 is involved in the reversal of histone H2BK123 ubiquitylation38–40. Significantly for this report, a key role of this desubiquitylase is to counteract futile bypass events at replication forks acting as a PCNAK164-DUB37. PCNA deubiquitylation is a requirement conserved throughout evolution as evidence for ScUbp10, SpUbp16 and HsUsp1 shows36,43. With this background in mind, the aim of this work was to understand the functional meaning of the interaction of the PCNA-DUB Ubp10 with proteins involved in Okazaki fragment synthesis and maturation, physical interaction described earlier for the Flap endonuclease Fen1Rad27,37. In this study we have presented ample evidence suggesting that this DUB regulates the dissociation of the sliding clamp PCNA from chromatin and that, by doing so, ensures proper maturation of the lagging strand. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 14 One recent observation, Fen1Rad27-Ubp10 binding, led us to the study of Ubp10’s S phase proteome. The analysis confirmed previous data regarding Ubp10 biology as the RNA polymerase I complex subunits Rpa190, Rpa34, Rpa43, Rpa49 and Rpa135 were among proteins trap with Ubp10. Spt16 and Pob3 FACT subunits were also found to bind Ubp10-GFP. Our approach confirmed Fen1-Ubp10 and PCNA-Ubp10 interactions and identified major components of synthesis and maturation of the lagging strand as feasible interactors of the PCNA-DUB. The proteomic studies were made in crosslinked protein samples from cells synchronized in S phase. We confirmed each observed interaction by individually testing Ubp10 ability to form a complex with each OFM complex component of interest in tagged strains during S phase. A relevant point for this work is understanding the nature of the cell cycle defect of Ubp10 depleted cells, a defect we believed is poorly understood36,37,41,42. Growth and cell cycle defects are separable37,42. In our studies, we did not observe a G1 delay defect36,37. Further, the timing of entry into S phase is close to that of the wild type, with equivalent timing in ARSs activation37. ubp10∆ characterization indicates that the defective cell cycle is a consequence of the slowdown in DNA replication progression. We also found here that this defect is based on the accretion of chromatin-bound PCNA during S phase and that, consistent with this evidence, it is efficiently suppressed by PCNA-disassembly-prone pol30R14E and pol30D150E mutant alleles (see below). Fission yeast cells increase the amount of chromatin-associated PCNA when the K164 of this sliding clamp is ubiquitylated44. Based on their observations, Daigaku and coworkers proposed that in S. pombe an increase in Ub-PCNAK164 works to expand the time for PCNA-Pol d binding to chromatin to allow the completion of OFs. Our findings in S. cerevisae are consistent with a scenario where ubiquitylation of PCNA is a DNA retention signal for the sliding clamp at lagging strands in unperturbed replication. In a cause-and-effect link (further discussed below), the increase in chromatin-bound PCNA likely slows down progression through S phase in budding yeast. This is consistent with previous evidence in S. pombe cells where depletion of PCNA-DUBs leads to a cell cycle delay phenotype suppressed by abrogation of the PCNA-ubiquitin-ligase Rhp1843. Of particular interest for our work were both the study of the cell cycle defect of Ubp10 depleted cells and the analysis of the genetic interactions that arise from individual mutants in the OFM pathway when combined with ubp10∆. One of these analyses has (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 15 been made in a Cdc9 defective background. In this analysis, not only we found a semi- synthetic lethality among cdc9-7 and ubp10∆ but also created a tool and found a temperature for the OF accumulation tests. Indeed, cdc9-7 ubp10∆ results suggest that ablation of the PCNA-DUB Ubp10 leads to a strong lagging-strand replication defect phenotype, possibly owing to defects in Okazaki fragment ligation. On the other hand, we also show that ubp10∆ phenotype is unrelated to Cdc9 presence or activity. Therefore, we surmise that ubp10∆ cells have a genuine defect in the maturation of Okazaki fragments related to PCNA unloading. We have performed a series of experiments to measure the length and molecular nature of the abundant OF accumulated in UBP10 mutant cells and, in summary, all observations indicate that only abundance is affected. Regarding the length and relative accumulation of chromatin repeats (that reflect nucleosome repeats), the detected OF in cdc9-7 ubp10D cells are normal. Moreover, we show here that Ubp10 accumulated OF are in vitro ligatable to the same extent as controls (cdc9-7 ubp10D versus cdc9-7), and controls are in accordance with published data54 again indicating that OFs are conventional nicked DNA. Since it has been demonstrated that lagging-strand synthesis in budding yeast is coupled with chromatin assembly on newly synthesized DNA50 we deduce as well that chromatin assembly is not affected in Ubp10 depleted cells. Smith and Whitehouse have shown that rad9D and tof1D checkpoint mutants when combined with a cdc9td-degron allele accumulate abundant but normal length Okazaki fragments when Cdc9td is proteolyzed50. This observation is particularly strong, in terms of OF abundance, for tof1D mutants and is relevant for our work given the similarity with ubp10D data presented here. There is not such a strong accumulation in cdc9td rad9D case50. Tof1, named after topoisomerase I-interacting factor, is pertinent to this work because is a subunit of the Csm3-Mrc1-Tof1 replication pausing-mediator complex functionally associated with DNA replication forks59–62. On the other hand, the N-terminus of FACT-subunit Spt16 interacts with Tof1 to ensure chromatin replication in vitro63, supporting the hypothesis that FACT is recruited to replication forks by the Tof1-fork replication complex for parental nucleosomes removal. Given that the FACT complex interacts with Ubp10 likely to integrate (transcription and) DNA replication with nucleosome assembly48, a complex Tof1-FACT-Ubp10 connection emerges likely involved in robust DNA replication progression. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 16 As mentioned before, an important conclusion here is that ubp10∆ S phase defects do not involve Cdc9 function. However, we observed that ubp10∆ cell cycle deficiency was assumably a consequence of increased residence time of PCNA during S phase and was directly related with a defective unloading of the sliding clamp in Ubp10 depleted cells. Accordingly, disassembly-prone mutant alleles of POL30, pol30R14E and pol30D150E, rescue ubp10∆ mutant defects. A corollary of these two remarkable observations is that during unperturbed S phase PCNA unloading may occur simultaneously with or even before Cdc9-mediated nick ligation and this may be in contradiction with evidence published regarding Elg1 in the PCNA unloading subject21. However, this is an outstanding preliminary observation; therefore, further work would be required to understand this conundrum. Besides the slowdown in replication progression, the ablation of Ubp10 is characterized by the accumulation of non-canonical replication intermediates in HU- treated cells detected as small Ys by 2D-gel analysis. These intermediates are observed upon Ubp10-DUB depletion, suggesting that they are normally suppressed by PCNA deubiquitylation37. Canonical small Ys reflect passive replication by forks arising outside the probed fragment. We suggest that accumulation of small Ys may reflect also pathological features of lagging strand-associated replication fork defects (nick DNA that may generate breakage structures). Special consideration should be given to replication intermediates when testing drug-treated cells because fork progression is limited in HU, as elongation of DNA synthesis from ARSs is slow64, thus, all replication intermediates detected under our experimental conditions by 2D gel analysis belong to the closest origin of replication for any given restriction fragment tested. Consistent with the idea that non-canonical/small Ys might be pathological replication structures, both fen1rad27 and cdc9-7 mutants accumulate small Ys (comparable to top1 top2 double mutants65) compatible with increased accumulation of nicked DNA, therefore, fragile molecules. The suppression of the accumulation of these non-canonical molecules in HU-treated cells by POL30 disassembly-prone mutant alleles, pol30R14E and pol30D150E, combined with the suppression of the slow bulk DNA replication, show that ubp10∆ mutation have an effect throughout the entire yeast genome. All this evidence indicates that this Ubp10 deubiquitylase plays a significant genome-wide role during the S phase of every unperturbed cell cycle. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 17 PCNA trimer-disassembly-prone mutants pol30R14E and pol30D150E alleviate the Okazaki fragment length extension problem described in elg1D cells55. Further, combining pol30R14E with elg1D largely rescued the elevated mutation rate of the single elg1 mutant57. Although UBP10 mutant cells accumulate OFs, they do not show an elevated mutation rate, in fact their mutation rate is similar to that of wild-type cells36,43. Both, pol30D150E and pol30R14E, PCNA mutant alleles are excellent extragenic suppressors of UBP10 deletion. It has been described that due to their disassembly-prone nature of the homo-trimeric PCNA ring, both POL30 alleles accumulate low levels of PCNA on replicating chromatin. However, based in our own observations, pol30R14E and pol30D150E point mutant alleles differ between them in the total cellular amount of PCNA levels. In whole cell extracts, PCNAD150E levels mimics those of wild-type cells, while PCNAR14E shows significantly reduced levels as compared to PCNAwt. Thus, PCNAD150E behaves like a real prone-disassembly PCNA mutant. Nevertheless, PCNAR14E mirror PCNAD150E low levels of PCNA bound to chromatin and, therefore, it is useful and valuable in our analysis. In summary, disassembly-prone mutants pol30R14E and pol30D150E rescue the chromatin retention of PCNA phenotype of ubp10D (as shown in Figure 6A), as well as cell cycle delay (Figure 5C and 6B) and defective replication intermediate accumulation upon exposure to HU (Figure 5D). Remarkably, the pol30D150E disassembly-prone mutant also rescues Okazaki fragments accumulation observed in DNA ligase I (cdc9ts) when combined with ubp10D (Figure 6C ). In other words, all major S phase phenotypes associated with defective Ubp10 (ubp10D) are relieved by reversion of accretion of PCNA by two different trimer instability PCNA/POL30 mutants (pol30R14E and pol30D150E)55,57. Together, all this evidence indicates that a slow PCNA unloading underlies every replication defect in Ubp10 depleted cells. Another relevant information here is that Elg1-depleted cells transiently accumulate PCNA on chromatin (Figure 7). This transient nature of PCNA retention on chromatin is, on one hand, consistent with the viability elg1∆ deleted strains23–25. However, on the other hand, it may also mean that in the absence of Elg1-RLC complex PCNA is steadily unloaded in vivo, as we show here (Figure 7). Although it does not come as a total surprise, this unanticipated observation is coherent with the fact that Elg1 is not essential for DNA replication23–25. Nonetheless, as reported, Elg1 may be still important for efficient S phase progression18. However, by testing bulk DNA replication in unperturbed conditions at 25ºC, we observed no S phase delays in Elg1-depleted cells (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 18 (in similar conditions where we detected a transient accretion of chromatin-bound PCNA). One strong possibility is that under these experimental circumstances an alternative complex to Elg1-RLC1 is unloading PCNA. The transient accumulation of chromatin-bound PCNA in elg1∆ cells is, therefore, evocative of the existence of an alternative PCNA unloading mechanism during replication. We show here that only when elg1∆ is combined with ubp10∆ PCNA remain bound onto chromatin. Interestingly, we detected a robust accumulation and a very slow-paced PCNA unloading in elg1 ubp10 double-mutant cells in synchronized S phase, suggesting the existence of an Ubp10-regulated PCNA unloading mechanism divergent from that of Elg1-RLC. Undeniably, all the evidence suggests that a Ubp10- dependent mechanism underlies the timely removal of PCNA from replicating chromatin during unperturbed S phase. Eventually, PCNA is unloaded in Elg1 and Ubp10 doubly-depleted cells, as chromatin-bound PCNA in G1 synchronized cells remains very low. Perhaps most significantly, S phase progression is slow in this elg1 ubp10 double mutant. The fact that depletion of Ubp10 has a similar slow S phase phenotype may be indicative of a default mechanism of PCNA unloading regulated by this PCNA-DUB in unperturbed cell cycle. Kubota et al., predicted Elg1-RLC alternative PCNA unloaders back in 201326. In fact, they suggested three models of PCNA unloading based on published evidence at the time and still valid today. A first model where Elg1-RLC would be the main unloader; a second model where Rfc1-RFC would work as a genome-wide unloader and Elg1-RLC would be the unloader of PCNA at specific genome localizations with emphasis in difficult to replicate sites; and a third model where Elg1-RLC complex would unload SUMOylated PCNA and Rfc1-RFC would unload unmodified PCNA rings to directly recycle them at lagging strands. Our findings are consistent with the last two models yet better support the third one where two different PCNA unloaders complexes would effectively recycle PCNA during the synthesis of lagging strands. We suggest that the most likely alternative PCNA unloader is Rfc1-RFC, and that it may unload PCNA preferentially when deubiquitylated. Rfc1 is an essential subunit of the RFC complex that interacts with Ubp10 and PCNA. Ubp10 would eventually regulate the precise timing of PCNA unloading by Rfc1-RFC after Ub-PCNAK164 deubiquitylation. Our working model predicts that PCNA is ubiquitylated at K164 at lagging strands, we hypothesize that such event would be a consequence of the collision of the PCNA-Pol delta complex with the preceding (5’ end) Okazaki Fragment. Prior to Ubp10 action and (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 19 subsequent PCNA unloading, PCNA ubiquitylation would enhance Pol delta release by a collision-release mechanism already described66. A strong case can be made here that yeast cells coordinate the last steps of synthesis and maturation of Okazaki fragments through a Ubp10-regulated PCNA unloading mechanism distinct from that of Elg1ATAD5-RLC (see model in Figure 7D). Our findings are consistent with the hypothesis that S. cerevisiae cells ubiquitylate and deubiquitylate PCNA during S phase in a dynamic, yet ordered, manner to ensure normal DNA replication, such that PCNA ubiquitylation is orderly followed by its deubiquitylation to strengthen PCNA unloading at lagging strands in a genome-wide scale, regulating and ensuring the time frame for Okazaki fragments maturation to generate a continuous lagging double-stranded DNA.

Yeast strains, growth conditions and media All the budding yeast used in this study originate from a MATa W303 RAD5 bar1::LEU2 strain36 and are listed in the Resources Table (Supplementary Information). Budding yeast strains were grown in YPAD medium (1% yeast extract, 2% peptone supplemented with 50 μg/ml adenine) containing 2% glucose. For block-and-release experiments, cells were grown in YPAD with 2% glucose at 25°C and synchronized in G1 with α-factor pheromone (40 ng/ml, 2.5 hours). Cells were then collected by centrifugation (3000 rpm, 3 min) and released into fresh media (supplemented with 50 µg/ml of Pronase) in the absence or in the presence of HU (0.2 M, FORMEDIUM). Overexpression experiments with cells grown in YPAD medium with 2% raffinose at 25°C were conducted by adding to the medium 2.5% galactose (to induce) or 2% glucose (to repress). General experimental procedures General experimental procedures of yeast Molecular and Cellular Biology were used as described previously60,67,68. Generation of tagged alleles and specific gene deletions was performed as described69. Transformation was performed by lithium acetate protocol and transformants were selected by growing in selective mediums. Different selection markers were used (KANMX6, HphMX4, NatMX6, URA3, TRP1, HIS3), as indicated in Resources Table. Constructs were confirmed by PCR and/or sequencing. The presence of tagged proteins was further confirmed by immunoblot. Moreover, strains with tagged alleles were carefully checked for growth rate and sensitivity to HU. No differences with untagged controls were found. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 20 Flow Cytometry Analysis For flow cytometry analyses, 10 7 cells were collected by centrifugation, washed once with water, fixed in 70% ethanol, and processed as described previously68,70. Cells were prepared using a modification of the method, by using SYTOX Green (Molecular PROBES) for DNA staining71,72. The DNA content of individual cells was measured using a Becton Dickinson Accuri C6 plus FACScan. HU sensitivity Assays Stationary cells were counted and serially diluted in YPAD media. Ten -fold dilutions of equal numbers of cells were plated onto YPAD (2% glucose) media (always supplemented with 50 μg/ml adenine), or YPAD containing HU, incubated at the indicated temperatures for 24, 48, 72 or 120 hours and then scanned. Identification of Ubp10 interactors by mass spectrometry Ubp10-GFP expressing- and wild type control cells were synchronized with a-factor and released into fresh YPAD. 30 min, 40 min, 50 min and 60 min time point samples were fixed with formaldehyde and harvested. Cell pellets were resuspended in Lysis Buffer (50 mM Hepes pH 7.5, 140 mM NaCl, 1 mM EDTA, 1% Tritón -X100, 0.1% Na - deoxycholate) and broken using glass beads in a fast -prep. Chromatin extracts were disrupted by sonication, cleared by centrifugation, and incubated for 4 hours at 4ºC with agarose-conjugated GFP-Trap™ beads (Chromotek). Beads were washed once with lysis buffer, once with Wash Buffer (10 mM Tris pH 8, 250 mM LiCl, 1 mM EDTA, 0.5% NP-40, 0.5% Na-deoxycholate) and once with TE. 30 min, 40 min, 50 min and 60 min samples were pulled and resuspended in Laemmli Buffer, and resolved by SDS-PAGE. Regions of interest were excised and digested as previously described73. Samples were digested using trypsin and analyzed using mass spectrometry at the Proteomic Biotechnology Unit of the Cancer Research Center (Salamanca, Spain) . The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE74 partner repository with the dataset identifier PXD048249. Total protein extracts for Immunoblotting Total protein extracts were prepared following cell fixation using trichloroacetic acid (TCA) and resolved by SDS -polyacrylamide gel electrophoresis before transfer to nitrocellulose membranes. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 21 Fractioning and Immunoblotting For chromatin-enriched fractions around 6 x 10 7 exponentially growing cells were harvested by centrifugation and resuspended in 1 ml of Buffer 1 (containing 150 mM Tris pH 8.8, 10 mM dithiothreitol (DTT), and 0.1% sodium azide), and incubated at room temperature for 10 minutes. Cells were pelleted, was hed with 1 ml of Buffer 2 (50 mM KH2PO4/K2HPO4 pH 7.4, 0.6 M Sorbitol, and 10 mM DTT), resuspended in 200 μl of Buffer 2 supplemented with 40 μg Zymolyase -100T and incubated at 37 °C for 10 minutes with intermittent mixing. The resulting spheroplasts were washed with 1 ml of ice-cold Buffer 3 (50 mM HEPES pH 7.5, 100 mM KCl, 2.5 mM MgCl, and 0.4 M Sorbitol), followed by resuspension and a 5 -minute incubation in 100 μl of EBX buffer (50 mM HEPES pH 7.5, 100 mM KCl, 2.5 mM MgCl, 0.25% Triton100, 1 mM phenylmethylsulfonyl fluoride (PMSF), Protease inhibitor tablets (EDTA -free, Roche), Leupeptin 1 μg/ml, Pepstatin 2.5 μg/ml, and RNAse 10 μg/ml), with occasional mixing. Aliquots of 30 μl of these disrupted cell suspensions were collected as whole cell extract samples (WCE). Remaining volume was layered onto 70 μl of cold EBX-S buffer (EBX buffer supplemented with 30% Sucrose) and subjected to centrifugation at 12000 rpm for 10 minutes at 4 °C. Aliquots of 30 μl of the resulting supernatant layer (Chromatin - free fraction) were also collected. After discarding supernatant, chromatin pellets were washed with 200 μl of EBX-S buffer, resuspended in 70μl of EBX buffer supplemented with 0.5 μl of Benzonase, and incubated on ice for 15 minutes (Chromatin fraction). SDS- PAGE loading buffer was added to each fraction. The different protein extracts were separated by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. Antibodies used for detection are listed in the Resources Table and were visualized using ECL reagents (Amersham Pharmacia Biotech) and film. The levels of proteins bound to chromatin were quantified using Quantity One 1-D Analysis Software (BioRad) and normalized with their corresponding Histone H2B values. All data in the bar graphs are presented as means SD in triplicate. A two-way analysis of variance (ANOVA) test was used to determine the statistical significance. Protein interaction analysis Cells expressing tagged or untagged (control) proteins were fixed with 1% formaldehyde and harvested. Chromatin extracts were prepared in a Lysis Buffer containing 50 mM Hepes pH 7.5, 140 mM NaCl, 1 mM EDTA, 1% Tritón-X100, 0.1% Na-deoxycholate, and supplemented with Antiproteolytic Cocktail using glass beads. Extracts were cleared by centrifugation; soluble protein fractions were discarded, and chromatin pellets were (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 22 sheared by sonication. Chromatin extracts were clarified and tagged proteins were enriched by immunoprecipitation with specific Tag antibodies previously bound to Protein G Dynabeads (5 hours at 4°C). Then, antibody-bound Protein G Dynabeads and controls were extensively washed with lysis buffer, and elution was carried out in SDS -PAGE loading buffer. Immunoprecipitates were resolved by SDS -PAGE gels, transferred to nitrocellulose membranes and analysed with specific-HRP conjugated antibodies. Two-dimensional DNA gels (2D-gel analysis) DNA samples for neutral-neutral two-dimensional gel electrophoresis were prepared and analyzed as described previously 60,75. DNA was cut with the NcoI restriction enzyme, transferred to Hybond-XL (GE Healthcare) nitrocellulose membrane, and hybridized to radiolabeled probes spanning the ARS305 and ARS306 origins of DNA replication. For each origin of replication tested, the specific probe corresponds to the following coordinates (retrieved from SGD): ARS305 (39073-40557, Chr III) and ARS306 (73001- 73958, Chr III). Images were acquired using a Molecular Imager FX (BioRad) and different replication-associated DNA molecules were quantified using Quantity One 4.6 software (BioRad). Acknowledgments We are grateful to members of 08 research group at the IBMCC for helpful discussions. We would like to particularly thank Professor Anne Donaldson and Takashi Kubota PhD (University of Aberdeen) for pol30 mutant and ChVLig1 strains. We are also grateful to the National BioResource Project, NBRP Japan for the cdc9-7 strain. This work was supported by the Spanish Ministry of Science (grants PID2019 -109616GB-100 to A.B. and M.P.S. and PID2020 -116003GB-100 to R.B.) and Junta de Castilla y León (grant SA103P20 to A.B). J.Z. was supported by a Predoctoral Fellowship from the Junta de Castilla y León (JCyL). S.M. was supported by a University of Salamanca Postdoctoral Fellowship and a MSCA Postdoctoral Fellowship (grant nº 101106007). E.A. was supported by a JCyL Postdoctoral Fellowship. A.B. and M.P.S. Institution is supp orted by the “Programa de Apoyo a Planes Estratégicos de Investigaci όn de Excelencia” cofunded by the Junta de Castilla y Leόn and the European Regional Development Fund (CLC-2017-01). Author Contributions Conceptualization, A.B., with substantial inputs from M.P.S., J.Z., S.M. and RB; Investigation, J.Z., S.M., E.A., M.A. and M.P.S.; Supervision, A.B. and M.P.S.; Formal (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 23 Analysis, A.B., M.P.S., S.M. and J.Z.; Writing & Editing, A.B. and M.P.S. with the help of R.B. and S.M.; Funding Acquisition, A.B., M.P.S. and R.B. Data availability The datasets supporting the current study are available from the corresponding authors on request.

1. Moldovan, G.-L., Pfander, B. & Jentsch, S. PCNA, the maestro of the replication fork. Cell 129, 665–679 (2007). 2. Bell, S. P. & Labib, K. Chromosome Duplication in Saccharomyces cerevisiae. Genetics 203, 1027–1067 (2016). 3. Zheng, L. & Shen, B. Okazaki fragment maturation: nucleases take centre stage. J. Mol. Cell Biol. 3, 23–30 (2011). 4. Acharya, N., Klassen, R., Johnson, R. E., Prakash, L. & Prakash, S. PCNA binding domains in all three subunits of yeast DNA polymerase {Delta} modulate its function in DNA replication. Proc. Natl. Acad. Sci. USA 108, 17927-17934 (2011). 5. Vijayakumar, S. et al. The C-terminal domain of yeast PCNA is required for physical and functional interactions with Cdc9 DNA ligase. Nucleic Acids Res. 35, 1624–1637 (2007). 6. Gary, R. et al. A Novel Role in DNA Metabolism for the Binding of Fen1/Rad27 to PCNA and Implications for Genetic Risk. Mol. Cell. Biol. 19, 5373–5382 (1999). 7. Shibahara, K. & Stillman, B. Replication-Dependent Marking of DNA by PCNA Facilitates CAF-1-Coupled Inheritance of Chromatin. Cell 96, 575–585 (1999). 8. Ulrich, H. D. New Insights into Replication Clamp Unloading. J. Mol. Biol. 425, 4727– 4732 (2013). 9. Arbel, M., Liefshitz, B. & Kupiec, M. How yeast cells deal with stalled replication forks. Curr Genet 66, 911–915 (2020). 10. Cullmann, G., Fien, K., Kobayashi, R. & Stillman, B. Characterization of the Five Replication Factor C Genes of Saccharomyces cerevisiae. Mol. Cell. Biol. 15, 4661– 4671 (1995). 11. Yao, N. et al. Replication Factor C Clamp Loader Subunit Arrangement within the Circular Pentamer and Its Attachment Points to Proliferating Cell Nuclear Antigen*. J. Biol. Chem. 278, 50744–50753 (2003). 12. Bermudez, V. P. et al. The alternative Ctf18-Dcc1-Ctf8-replication factor C complex required for sister chromatid cohesion loads proliferating cell nuclear antigen onto DNA. Proc. Natl. Acad. Sci. 100, 10237–10242 (2003). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 24 13. Lengronne, A. et al. Establishment of sister chromatid cohesion at the S. cerevisiae replication fork. Mol Cell 23, 787–799 (2006). 14. Liu, H. W. et al. Division of Labor between PCNA Loaders in DNA Replication and Sister Chromatid Cohesion Establishment. Mol Cell 78, 725-738.e4 (2020). 15. Cai, J. et al. Reconstitution of human replication factor C from its five subunits in baculovirus-infected insect cells. Proc. Natl. Acad. Sci. USA 93, 12896–12901 (1996). 16. Yao, N. et al. Clamp loading, unloading and intrinsic stability of the PCNA, β and gp45 sliding clamps of human, E. coli and T4 replicases. Genes Cells 1, 101–113 (1996). 17. Bylund, G. O. & Burgers, P. M. J. Replication protein A-directed unloading of PCNA by the Ctf18 cohesion establishment complex. Mol. Cell. Biol. 25, 5445–5455 (2005). 18. Kubota, T., Nishimura, K., Kanemaki, M. T. & Donaldson, A. D. The Elg1 replication factor C-like complex functions in PCNA unloading during DNA replication. Mol. Cell 50, 273–280 (2013). 19. Lee, K., Fu, H., Aladjem, M. I. & Myung, K. ATAD5 regulates the lifespan of DNA replication factories by modulating PCNA level on the chromatin. J. Cell Biol. (2012) doi:10.1083/jcb.201206084. 20. Shiomi, Y. & Nishitani, H. Alternative replication factor C protein, Elg1, maintains chromosome stability by regulating PCNA levels on chromatin. Genes Cells 18, 946– 959 (2013). 21. Kubota, T., Katou, Y., Nakato, R., Shirahige, K. & Donaldson, A. D. Replication- Coupled PCNA Unloading by the Elg1 Complex Occurs Genome-wide and Requires Okazaki Fragment Ligation. Cell Reports 12, 774-787 (2015). 22. Shiomi, Y. & Nishitani, H. Control of Genome Integrity by RFC Complexes; Conductors of PCNA Loading onto and Unloading from Chromatin during DNA Replication. Genes 8, 52 (2017). 23. Ben-Aroya, S., Koren, A., Liefshitz, B., Steinlauf, R. & Kupiec, M. ELG1, a yeast gene required for genome stability, forms a complex related to replication factor C. Proc. Natl. Acad. Sci. USA 100, 9906–9911 (2003). 24. Bellaoui, M. et al. Elg1 forms an alternative RFC complex important for DNA replication and genome integrity. EMBO J. 22, 4304–4313 (2003). 25. Kanellis, P., Agyei, R. & Durocher, D. Elg1 forms an alternative PCNA-interacting RFC complex required to maintain genome stability. Current Biol. 13, 1583–1595 (2003). 26. Kubota, T., Myung, K. & Donaldson, A. D. Is PCNA unloading the central function of the Elg1/ATAD5 replication factor C-like complex? Cell Cycle 12, 2570–2579 (2013). 27. Hoege, C., Pfander, B., Moldovan, G.-L., Pyrowolakis, G. & Jentsch, S. RAD6- dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature 419, 135–141 (2002). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 25 28. Mailand, N., Gibbs-Seymour, I. & Bekker-Jensen, S. Regulation of PCNA-protein interactions for genome stability. Nat. Rev. Mol. Cell Biol. 14, 269–282 (2013). 29. Chang, D. J. & Cimprich, K. A. DNA damage tolerance: when it’s OK to make mistakes. Nat. Chem. Biol. 5, 82–90 (2009). 30. Friedberg, E. C. Suffering in silence: the tolerance of DNA damage. Nat. Rev. Mol.Cell Biol. 6, 943–953 (2005). 31. Hedglin, M. & Benkovic, S. J. Regulation of Rad6/Rad18 Activity During DNA Damage Tolerance. Annual Rev. Biophysics 44, 207–228 (2015). 32. Stelter, P. & Ulrich, H. D. Control of spontaneous and damage-induced mutagenesis by SUMO and ubiquitin conjugation. Nature 425, 188–191 (2003). 33. Huang, T. T. et al. Regulation of monoubiquitinated PCNA by DUB autocleavage. Nat. Cell Biol. 8, 339–347 (2006). 34. Kashiwaba, S. et al. USP7 Is a Suppressor of PCNA Ubiquitination and Oxidative- Stress-Induced Mutagenesis in Human Cells. Cell Reports 13, 2072-2080 (2015). 35. Lim, K. S. et al. USP1 Is Required for Replication Fork Protection in BRCA1- Deficient Tumors. Mol. Cell 72, 925-941.e4 (2018). 36. Gallego-Sánchez, A., Andrés, S., Conde, F., San-Segundo, P. A. & Bueno, A. Reversal of PCNA ubiquitylation by Ubp10 in Saccharomyces cerevisiae. PLoS Genet. 8, e1002826 (2012). 37. Álvarez, V. et al. PCNA Deubiquitylases Control DNA Damage Bypass at Replication Forks. Cell Reports 29, 1323-1335. (2019). 38. Emre, N. C. T. et al. Maintenance of low histone ubiquitylation by Ubp10 correlates with telomere-proximal Sir2 association and gene silencing. Mol. Cell 17, 585–594 (2005). 39. Gardner, R. G., Nelson, Z. W. & Gottschling, D. E. Ubp10/Dot4p regulates the persistence of ubiquitinated histone H2B: distinct roles in telomeric silencing and general chromatin. Mol. Cell. Biol. 25, 6123–6139 (2005). 40. Schulze, J. M. et al. Splitting the task: Ubp8 and Ubp10 deubiquitinate different cellular pools of H2BK123. Genes & Dev. 25, 2242–2247 (2011). 41. Richardson, L. A. et al. A Conserved Deubiquitinating Enzyme Controls Cell Growth by Regulating RNA Polymerase I Stability. Cell Reports 2, 372-385 (2012). 42. Mapa, C. E., Arsenault, H. E., Conti, M. M., Poti, K. E. & Benanti, J. A. A balance of deubiquitinating enzymes controls cell cycle entry. Mol. Biol. Cell 29, 2821–2834 (2018). 43. Álvarez, V. et al. Orderly progression through S-phase requires dynamic ubiquitylation and deubiquitylation of PCNA. Sci. Rep. 6, 25513 (2016). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 26 44. Daigaku, Y. et al. PCNA ubiquitylation ensures timely completion of unperturbed DNA replication in fission yeast. PLoS Genet. 13, e1006789 (2017). 45. Thakar, T. et al. Ubiquitinated-PCNA protects replication forks from DNA2- mediated degradation by regulating Okazaki fragment maturation and chromatin assembly. Nat. Commun. 11, 2147–14 (2020). 46. Becker, J. R. et al. Genetic Interactions Implicating Postreplicative Repair in Okazaki Fragment Processing. PLoS Genet. 11, e1005659 (2015). 47. Guilliam, T. A. & Yeeles, J. T. P. Reconstitution of translesion synthesis reveals a mechanism of eukaryotic DNA replication restart. Nat. Struct. & Mol. Biol. 27, 450–460 (2020). 48. Nune, M. et al. FACT and Ubp10 collaborate to modulate H2B deubiquitination and nucleosome dynamics. eLIFE 8, e40988 (2019). 49. Johnston, L. H. & Nasmyth, K. A. Saccharomyces cerevisiae cell cycle mutant cdc9 is defective in DNA ligase. Nature 274, 891–893 (1978). 50. Smith, D. J. & Whitehouse, I. Intrinsic coupling of lagging-strand synthesis to chromatin assembly. Nature 483, 434–438 (2012). 51. Blair, K. et al. Mechanism of human Lig1 regulation by PCNA in Okazaki fragment sealing. Nat. Commun. 13, 7833 (2022). 52. Kachroo, A. H. et al. Evolution. Systematic humanization of yeast genes reveals conserved functions and genetic modularity. Science 348, 921–925 (2015). 53. Unternährer, S. & Hinnen, A. Temperature sensitivity of the cdc9-1 allele of Saccharomyces cerevisiae DNA ligase is dependent on specific combinations of amino acids in the primary structure of the expressed protein. Mol. Gen. Genet. MGG 232, 332–334 (1992). 54. Kahli, M., Osmundson, J. S., Yeung, R. & Smith, D. J. Processing of eukaryotic Okazaki fragments by redundant nucleases can be uncoupled from ongoing DNA replication in vivo. Nucleic Acids Res. 47, 1814–1822 (2019). 55. Gali, V. K. et al. Identification of Elg1 interaction partners and effects on post- replication chromatin re-formation. Plos Genet. 14, e1007783 (2018). 56. Sriskanda, V., Schwer, B., Ho, C. K. & Shuman, S. Mutational analysis of Escherichia coli DNA ligase identifies amino acids required for nick-ligation in vitro and for in vivo complementation of the growth of yeast cells deleted for CDC9 and LIG4. Nucleic Acids Res. 27, 3953–3963 (1999). 57. Devakumar, L. J. P. S., Gaubitz, C., Lundblad, V., Kelch, B. A. & Kubota, T. Effective mismatch repair depends on timely control of PCNA retention on DNA by the Elg1 complex. Nucleic Acids Res. 47, 6826–6841 (2019). 58. Kang, M.-S. et al. Regulation of PCNA cycling on replicating DNA by RFC and RFC-like complexes. Nat Commun. 10, 2420 (2019). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 27 59. Katou, Y. et al. S-phase checkpoint proteins Tof1 and Mrc1 form a stable replication-pausing complex. Nature 424, 1078–1083 (2003). 60. Calzada, A., Hodgson, B., Kanemaki, M., Bueno, A. & Labib, K. Molecular anatomy and regulation of a stable replisome at a paused eukaryotic DNA replication fork. Genes & Dev. 19, 1905–1919 (2005). 61. Nedelcheva, M. N. et al. Uncoupling of Unwinding from DNA Synthesis Implies Regulation of MCM Helicase by Tof1/Mrc1/Csm3 Checkpoint Complex. J. Mol. Biol. 347, 509–521 (2005). 62. Bando, M. et al. Csm3, tof1, and mrc1 form a heterotrimeric mediator complex that associates with DNA replication forks. J. Biol. Chem. 284, 34355–34365 (2009). 63. Safaric, B. et al. The fork protection complex recruits FACT to reorganize nucleosomes during replication. Nucleic Acids Res. 50, gkac005 (2022). 64. Poli, J. et al. dNTP pools determine fork progression and origin usage under replication stress. .EMBO J. 31, 883–894 (2011). 65. Bermejo, R. et al. Top1- and Top2-mediated topological transitions at replication forks ensure fork progression and stability and prevent DNA damage checkpoint activation. Genes & Dev. 21, 1921–1936 (2007). 66. Langston, L. D. & O’donnell, M. DNA polymerase delta is highly processive with proliferating cell nuclear antigen and undergoes collision release upon completing DNA. J. Biol. Chem. 283, 29522–29531 (2008). 67. Cordon-Preciado, V., Ufano, S. & Bueno, A. Limiting amounts of budding yeast Rad53 S-phase checkpoint activity results in increased resistance to DNA alkylation damage. Nucleic Acids Res. 34, 5852–5862 (2006). 68. Sánchez, M., Calzada, A. & Bueno, A. The Cdc6 protein is ubiquitinated in vivo for proteolysis in Saccharomyces cerevisiae. J. Biol. Chem. 274, 9092–9097 (1999). 69. Longtine, M. S. et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14, 953–961 (1998). 70. Calzada, A., Sacristán, M., Sánchez, E. & Bueno, A. Cdc6 cooperates with Sic1 and Hct1 to inactivate mitotic cyclin-dependent kinases. Nature 412, 355–358 (2001). 71. Haase, S. B. & Reed, S. I. Improved flow cytometric analysis of the budding yeast cell cycle. Cell Cycle 1, 132–136 (2002). 72. Haase, S. B. Cell cycle analysis of budding yeast using SYTOX Green. Current protocols in cytometry Chapter 7, Unit 7.23-7.23.4 (2004). 73. Shevchenko, A., Wilm, M., Vorm, O. & Mann, M. Mass Spectrometric Sequencing of Proteins from Silver-Stained Polyacrylamide Gels. Anal. Chem. 68, 850–858 (1996). 74. Perez-Riverol, Y. et al. The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Res. 50, 543–552 (2021). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 28 75. Lopes, M. et al. The DNA replication checkpoint response stabilizes stalled replication forks. Nature 412, 557–561 (2001). FIGURE LEGENDS Figure1. PCNA-DUB Ubp10 interacts with core components of lagging-strand synthesis machinery. PCNA-DUB Ubp10 interacts with all major lagging-strand synthesis complex components during unperturbed S phase as detected by LC-MS-MS analysis. Ubp10- GFP was immunoprecipitated from yeast cells pre-synchronized in G1 with a-factor and released 60 minutes in fresh rich media (S phase). Immunoprecipitation was carried out by using a GFP-Trap Agarose beads as described in Methods and using an untagged strain as a negative control. Full IPs were subject to LC-MS-MS analysis. The list shows Ubp10-GFP interacting proteins identified by LC-MS-MS. Potential Ubp10-interactors were identified among proteins with at least 3 unique peptides (only selected interactors are shown). Some selected proteins were chosen for confirmative CoIP analysis as shown in supplementary figure 1 for Pol3 (catalytic subunit of Pol∂) and Cdc9 (DNA ligase I) together with LC-MS-MS results obtained with other proteins of interest regarding Ubp10 biology. Figure 2. Flap-endonuclease FEN1/RAD27 is epistatic to PCNA-DUB UBP10 in the Okazaki fragment maturation pathway. A. S phase progression analysis of cells depleted for Fen1 and Ubp10. Wild-type (wt), ubp10∆, fen1∆ and fen1∆ ubp10∆ exponentially growing were synchronized in G1 with a-factor and release in fresh (rich) media. DNA content by FACS analysis of samples taken at indicated intervals is shown. Red arrows indicate approximated duration of bulk DNA replication for each strain. Note that fen1∆ ubp10∆ cells behave like single fen1∆ mutants. B. Cells described in A were synchronized with a-Factor and released in the presence of 200 mM HU at 25ºC. Samples were taken 60 minutes of treatment and processed for 2D- gel analysis of replication intermediates. Membranes were hybridized consecutively with probes spanning ARS305 and ARS306 replication origins. Note that fen1∆ ubp10∆ cells accumulated X-shaped replication intermediates comparably to fen1∆ single mutant cells and in clear contrast to abnormal small Ys intermediates (red arrows) observed in ubp10∆ cells (described in Álvarez et al., 2019). C. Ten-fold dilutions of the strains indicated in A incubated in YPAD at 25ºC or 35ºC in the absence (unperturbed) or in the chronic presence of HU (25mM or 50 mM as indicated). Data presented in A, (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 29 B and C indicates that FEN1 is epistatic to UBP10. D. S phase chromatin association of Fen1 in wild-type and ubp10∆ cells expressing Fen1-Flag tagged protein. Exponentially growing cultures of wild-type and ubp10∆ cells were synchronized with a-factor and released in fresh media to test S phase chromatin association of Flap- endonuclease Fen1/Rad27. Samples were taken at indicated intervals; chromatin- enriched fractions were prepared and electrophoresed in SDS-PAGE gels. Blots were incubated with a-Flag (to detect Fen1-Flag) or a-H2B antibodies. Blots from a representative experiment are shown. Data in the graph represent the average of three biological replicates (and is expressed as means ±SD in triplicate) (p = 0.3250, two- way ANOVA test). This evidence suggests that chromatin association of Fen1 is not affected by depletion of Ubp10. Figure 3. Abrogation of Ubp10 leads to accumulation of unligated Okazaki fragments. Depletion of Ubp10 increases the thermosensitivity and PCNA accumulation defects of the cdc9-7 ts allele of DNA ligase I. A. Ten-fold dilutions of equal number of cells of the indicated strains were spotted in Petri dishes and incubated either at 25ºC, 28ºC, 30ºC, or 35ºC for 60 hours. B. S phase chromatin association of PCNA in wild-type, cdc9-7, and cdc9-7 ubp10∆ cells. Exponentially growing cultures of wild-type, cdc9-7, and cdc9-7 ubp10∆ cells were synchronized in G1 with a-factor pheromone and released in fresh media to test S phase chromatin association of PCNA. Samples were taken at indicated intervals; chromatin-enriched fractions were prepared and electrophoresed in SDS-PAGE gels. Blots were incubated with a-PCNA or a-H2B antibodies. A blot from a representative experiment is shown. Data in the graph represent the average of three biological replicates (and is expressed as means ±SD in triplicate) (wild-type vs cdc9-7 p = 0.011; cdc9-7 vs cdc9-7 ubp10∆ p = 0.4372, two-way ANOVA test). C. UBP10 mutants show a strong detectable lagging-strand replication phenotype that leads to the accumulation of unligated Okazaki fragments (OF) . Exponentially growing cultures of cdc9-7 and cdc9-7 ubp10∆ cells incubated at 25ºC were shifted to 29ºC. Aliquot samples were taken at the indicated intervals. Purified total genomic DNA was labelled with exonuclease-deficient DNA polymerase I (Klenow) fragment and a-32P-dCTP, separated by agarose denaturing electrophoresis, and visualized using a Phosphor Imager. A representative experiment of two biological replicates is shown. Note that cdc9-7 ts cells transiently accumulate OFs and that depletion of Ubp10 results in further Okazaki fragments accumulation both in quantity and duration. D. In vitro analysis of ubp10D cumulative OFs reveals ligatable nick DNA. OFs were obtained (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 30 from cdc9-7 and cdc9-7 ubp10∆ cells as in C. Where indicated (+) genomic DNA was treated in vitro with T4 DNA ligase before labelling with Klenow and a-32P-dCTP for quantifying the proportion of DNA fragments (OF) ready for nick ligation. As shown, DNA samples from cdc9-7 ubp10∆ cells were as efficiently ligated as cdc9-7 controls. A representative experiment of three replicates is shown. Values of means of these three replicates ± SD are plotted. Figure 4. Steady reduction of chromatin associated Cdc9 during replication in UBP10 defective cells. A. S phase chromatin association of Cdc9 in wild-type and ubp10∆ cells. Exponentially growing cultures of wild-type and ubp10∆ cells were synchronized with a-factor and released in fresh YPAD media to test S phase chromatin association of DNA ligase I Cdc9. Samples were taken at indicated intervals; chromatin-enriched fractions were prepared and electrophoresed in SDS-PAGE gels. Blots were incubated with a-Ha (to detect Cdc9-Ha), a -PCNA or a-H2B antibodies. A blot from a representative experiment is shown. Data in the graph represent the average of three biological replicates (and is expressed as means ±SD in triplicate) (p = 0.0002, two-way ANOVA test). B. Whole cell extract (WCE) aliquots from A were blotted to test Cdc9-Ha protein amounts. Again, a blot from a representative experiment is shown. Data in the graph represent the average of three biological replicates (and is expressed as means ±SD in triplicate) (p = 0.1408, two-way ANOVA test). Figure 5. PCNA disassembly-prone mutants pol30R14E and pol30D150E revert ubp10∆-associated replication defects. A. pol30R14E and pol30D150E alleles rescue ubp10∆-associated defects in cdc9-7 ubp10∆. Ten-fold dilutions of the cdc9-7 indicated strains incubated in YPAD at different temperatures for 60 hours. Note that the increased lethality of cdc9-7 ubp10∆ is suppressed by pol30 mutant alleles that, in turn, do not rescue cdc9-7 thermosensitivity. B. 2D-gel analysis of cells synchronized in early S phase with the ribonucleotide reductase inhibitor HU at 29ºC. Indicated cdc9-7 strains were grown to exponential phase at 25ºC, synchronized in G1 with a-factor, released in fresh media with 200 mM HU at 29ºC for 60 additional minutes. The membrane was hybridized to a probe spanning ARS305 early replication origin. Open red arrow points small Ys intermediates. Under these conditions, cdc9-7 ubp10∆ defects were suppressed by PCNA disassembly-prone mutants. Histogram plots of small/large Y-shaped replication intermediates ratios in cdc9-7, cdc9-7 ubp10∆, cdc9-7 ubp10∆ pol30R14E and cdc9-7 ubp10∆ pol30D150E mutants are shown. C. DNA (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 31 replication progression defects in ubp10∆ cells are abrogated by pol30R14E and pol30D150E alleles. DNA content analysis of wild type, ubp10∆, pol30R14E, ubp10∆ pol30R14E, pol30D150E and ubp10∆ pol30D150E strains. Cells of the indicated strains (all cdc9td with CDC9 ON) were synchronized with a-factor and released in fresh YPAD at 25ºC. The progression of the bulk genome replication was monitored at the indicated time points. Open red arrows indicate approximate S phase duration in every strain where pol30R14E and pol30D150E suppression of the replication defect of ubp10∆ cells can be observed. D. 2D-gel analysis of replication intermediates in cells synchronized in early S phase with HU. Indicated strains grown to exponential phase at 25ºC were pre-synchronized in G1 with a-factor, released in fresh media with 200 mM HU an incubated at the same temperature for one additional hour. Membranes were hybridized to a probe spanning ARS305 early replication origin. Open red arrows indicate small Ys intermediates in blots. Histogram plots of small/large Y-shaped replication intermediates ratios in wild type and ubp10∆, pol30R14E, ubp10∆ pol30R14E, pol30D150E and ubp10∆ pol30D150E mutants are shown. Figure 6. PCNA retention on replicating chromatin underlies Okazaki fragments accumulation in Ubp10 abrogated cells. A. PCNA trimer-disassembly-prone mutant pol30D150E suppresses PCNA retention on chromatin phenotype of Ubp10 depleted cells. Exponentially growing cultures of wild-type, ubp10∆, and pol30D150E ubp10∆ cells were synchronized in G1 with a-factor and released in fresh media to test S phase chromatin association of PCNA. Samples were taken at indicated intervals; chromatin- enriched fractions were prepared and electrophoresed in SDS-PAGE gels. Blots were incubated with a-PCNA or a-H2B antibodies. A blot from a representative experiment is shown. Data in the graph represent the average of three biological replicates (and is expressed as means ±SD in triplicate) (wild-type vs ubp10∆ p <0.0001; ubp10∆ vs pol30D150E ubp10∆ p <0.0001, two-way ANOVA test). B. Slow S phase progression in PCNA-DUB UBP10 defective cells is a direct consequence of PCNA accumulation on replicating chromatin. DNA content analysis of wild type, ubp10∆, and pol30D150E ubp10∆ cells at the indicated time points (aliquot samples of the experiment in A). C. pol30D150E rescues ubp10∆ Okazaki fragments maturation timing defects. Exponentially growing cultures of cdc9-7, cdc9-7 ubp10∆, pol30D150E and pol30D150E ubp10∆ cells incubated at 25ºC were shifted to 29ºC. Aliquot samples were taken at the indicated one-hour intervals. Purified total genomic DNA was labelled with exonuclease-deficient DNA polymerase I, Klenow, fragment and a-32P-dCTP, separated by agarose (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 32 denaturing electrophoresis, and visualized using a Phosphor Imager. A representative experiment of two biological replicates is shown. Figure 7. Ubp10 is a key regulator of a PCNA unloading mechanism distinct from Elg1-RLC. A. Abrogation of Ubp10 is strongly additive to Elg1 depletion resulting in substantial increase of PCNA bound to chromatin during replication. S phase chromatin association of PCNA in wild-type, elg1∆, ubp10∆ and elg1∆ ubp10∆ cells. Exponentially growing cultures of the indicated strains were synchronized in G1 with a- factor and released in fresh media to test S phase chromatin association of PCNA and histone H2B. Samples were taken at indicated intervals; chromatin-enriched fractions were prepared and electrophoresed in SDS-PAGE gels. Blots were incubated with a- PCNA or a-H2B antibodies. Blots from representative experiments are shown. Data in the graphs represent the average of three biological replicates (and are expressed as means ±SD in triplicate) (p < 0.0001, two-way ANOVA test). Note that depletion of Elg1

in transient accumulation of PCNA on chromatin during S phase. B. S phase chromatin association of PCNA and histone H2B in ubp10∆, elg1∆ and elg1∆ ubp10∆ cells. Samples were processed as in A to compare the transient accumulation of PCNA on replicating chromatin in ubp10∆ and elg1∆ single mutants in the same blot (p < 0.0001, two-way ANOVA test). C. S phase progression analysis of wild-type, elg1∆, ubp10∆ and elg1∆ ubp10∆ cells. Progression of bulk genome replication was monitored at the indicated time points by FACS analysis. D. Proposed model for PCNA-DUB Ubp10 during maturation of Okazaki fragments regulating an Elg1- independent PCNA unloading mechanism (see text for details). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint DNA primase large subunit PRI2 21 4 DNA primase small subunit PRI1 12 5 DNA polymerase α subunit POL12 32 9 DNA polymerase α POL1 26 6 DNA pol α−binding protein CTF4 39 8 DNA polymerase δ POL3 32 9 DNA polymerase δ subunit POL31 23 5 PCNA POL30 76 10 Replication factor C subunit 1 RFC1 38 10 Replication factor C subunit 2 RFC2 111 13 Replication factor C subunit 3 RFC3 118 12 Replication factor C subunit 4 RFC4 85 10 Replication factor C subunit 5 RFC5 70 13 DNA mismatch repair protein MSH2 34 10 DNA mismatch repair protein MLH1 12 3 Ribonuclease H2 subunit A RNH201 22 6 Ribonuclease H2 subunit B RNH202 54 9 Replication factor A protein 1 RFA1 98 12 Replication factor A protein 3 RFA3 61 3 Replication factor A protein 2 RFA2 38 5 Ubiquitin hydrolase 10 UBP10 323 35 FLAP endonuclease I FEN1/RAD27 34 8 DNA ligase I CDC9 22 7 Gene names Selected identified Interactors Score Unique peptides S-phase Ubp10 Proteome (Ubp10-GFP-trap) fresh media 0’ 60’ α−FAsyn Figure 1 Zamarreño et al., 2024 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint Figure 2 Zamarreño et al., 2024 A B α−Factor 10’ Asyn 40’ 50’ 60’ 70’ Δ release in fresh YPAD at 25ºC 80’ 90’ 20’ 30’ 1C 2C 1C 2C 1C 2C fen1Δ ubp10Δ 1C 2C wt ubp10Δ fen1Δ 200mM HU 60’αFAsyn 25ºC Fen1-Flag H2B Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type ubp10Δ 0 0.4 0.8 0.2 1.0 Fen1 levels arbitrary units Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type ubp10Δ 0.6 YPAD (25 mM HU) ubp10∆ wt fen1∆ fen1/ubp10∆ YPAD (50 mM HU) 25ºC 35ºC 25ºC YPAD (unperturbed) ubp10∆ wt fen1∆ fen1/ubp10∆ C D ARS306 wild-type 10Δ fen1Δ fen1Δ 10Δ ARS305 wild-type 10Δ fen1Δ fen1Δ 10Δ (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint Figure 3 Zamarreño et al., 2024 B PCNA H2B 28 15 Fα 40´ 55´ 70´40 Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type cdc9-7 cdc9-7 ubp10 Δ KDa Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type cdc9-7 cdc9-7 ubp10 Δ 0 2 4 6 8 PCNA levels arbitrary units fresh media 0’ 70’ α−FAsyn 55’ 40’ C 29ºC 0h Asyn 25ºC 1h 3h 2h 0 1 2 3 0 1 2 3 100 200 300 400 500 cdc9-7 cdc9-7 ubp10Δ size (nt) D in vitro DNA ligase w/o T4 DNA ligase with T4 DNA ligase 100 0 Relative % 50 25 75 control ubp10Δ 44% 47% 100 200 300 400 500 size (nt) 1 2 + + 1: cdc9-7 2: cdc9-7 ubp10Δ A 28ºC 30ºC 35ºC ubp10Δ cdc9-7 cdc9-7 ubp10Δ wild-type 25ºC wild-type wild-type wild-type cdc9-7 cdc9-7 cdc9-7 cdc9-7 ubp10Δ cdc9-7 ubp10Δ cdc9-7 ubp10Δ ubp10Δ ubp10Δ ubp10Δ (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint Figure 4 Zamarreño et al., 2024 H2B Cdc9-Ha Pgk1 Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type ubp10Δ 0 0.75 0.5 1 Cdc9 levels arbitrary units 0.25 Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type ubp10Δ 15 100 130 KDa B A 0 1,5 1 2 Cdc9 levels arbitrary units 0,5 Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type ubp10Δ H2B15 Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ wild-type ubp10Δ Cdc9-Ha100 130 KDa PCNA28 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint Figure 5 Zamarreño et al., 2024 Release in fresh YPAD at 25ºC αF 10´ 20´ 30´ 40´ 50´ 60´ 70´ 80´ 90´ Asyn wild-type ubp10Δ pol30R14E pol30D150E10∆ pol30R14E 10∆ pol30D150E 1C 2C 1C 2C 1C 2C 1C 2C 1C 2C 1C 2C ubp10Δ control pol30R14E pol30D150E ubp10∆ pol30R14E ubp10∆ pol30D150E 28ºC 30ºC25ºC ubp10Δ control pol30R14E pol30D150E ubp10∆ pol30R14E 34ºC 35ºC 36ºC ubp10∆ pol30D150E A 200mM HU 60’αFAsyn 29ºC ba c d ARS305 a: cdc9-7 b: cdc9-7 ubp10∆ c: cdc9-7 ubp10∆ pol30 R14E d: cdc9-7 ubp10∆ pol30D150E 0.5 0 1,0 1,5 0.33 1.20 0.24 0.21 a b c d Small Ys/Large Ys Ratio (arbitrary units) B C 0.5 0 1,0 1,5 0.54 1.23 0.42 0.46 0.58 1.16 0.490.53 Small Ys/Large Ys Ratio (arbitrary units) 3: pol30R14E 5: pol30D150E 4: ubp10∆ pol30R14E 6: ubp10∆ pol30D150E 1: wild-type 2: ubp10Δ 1 2 3 64 1 2 5 ARS305 R14E 10∆ R14E wt 10Δ D150E 10∆ D150E wt 10Δ TEST pol30R14E TEST pol30D150E 200mM HU 60’αFAsyn 25ºC D (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint 100 200 300 400 500 size (nt) cdc9-7 Time (hours) 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 pol30D150E ubp10∆ pol30D150Econtrol ubp10Δ 29ºC Asyn 25ºC 0h 1h 2h 3h Time course ubp10Δwild-type ubp10Δ pol30D150E Release in fresh YPAD αF 40´ 55´ 70´ 1C2C 1C 2C 1C 2CB C Figure 6 Zamarreño et al., 2024 A Time course fresh media 0’ 70’ α−FAsyn 55’ 40’ 0 2 4 1 3 αF 40´ 55´ 70´ αF 40´ 55´ 70´ αF 40´ 55´ 70´ wild-type ubp10Δ PCNA levels arbitrary units pol30D150E ubp10∆ PCNA H2B 28 15 40 wild-type ubp10Δ KDa pol30D150E ubp10∆ αF 40´ 55´ 70´αF 40´ 55´ 70´αF 40´ 55´ 70´ (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint Figure 7 Zamarreño et al., 2024 αF 40´ 55´ 70´ ubp10Δwild-type ubp10Δ elg1∆ 1C2C 1C2C 1C 2C Release in YPAD 25ºC elg1Δ 1C2C ubp10Δ elg1∆ubp10∆ elg1Δ Fα 40´ 55´ 70´ Fα40´ 55´ 70´ Fα40´ 55´ 70´ 25 15 PCNA H2B A B C D lagging strandPCNA PCNA PCNA Ubp10 Ubp10Ubp10 leading strand δ Cdc9Fen1 δ δ PCNA-loading PCNA-unloading lagging strand PCNA-DUB Rfc1 Rfc1-RFC Elg1 Elg1-RLC unchallenged OFM X-RFC ubp10Δ elg1∆wild-type elg1Δ Fα 40´ 55´ 70´ Fα40´ 55´ 70´ Fα40´ 55´ 70´ 25 KDa 15 wild-type Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ elg1∆ ubp10Δ elg1∆ 0 2 4 6 8 PCNA levels arbitrary units 10 12 14 wild-type ubp10Δ Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ Fα 40´ 55´ 70´ ubp10Δ elg1∆ 0 2 4 6 8 10 ubp10Δ elg1∆wild-type ubp10Δ Fα 40´ 55´ 70´ Fα40´ 55´ 70´ Fα40´ 55´ 70´ PCNA H2B 25 15 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 5, 2024. ; https://doi.org/10.1101/2024.01.05.574312doi: bioRxiv preprint