USP50 suppresses alternative RecQ helicase use and deleterious DNA2 activity during replication

Abstract

Mammalian DNA replication relies on various DNA helicase and nuclease activities to ensure accurate genetic duplication, but how different helicase and nuclease activities are properly directed remains unclear. Here, we identify the ubiquitin-specific protease, USP50, as a chromatin-associated protein required to promote ongoing replication, fork restart, telomere maintenance, cellular survival following hydroxyurea or pyridostatin treatment, and suppression of DNA breaks near GC-rich sequences. We find that USP50 supports proper WRN-FEN1 localisation at or near stalled replication forks. Nascent DNA in cells lacking USP50 shows increased association of the DNA2 nuclease and RECQL4 and RECQL5 helicases and replication defects in cells lacking USP50, or FEN1 are driven by these proteins. Consequently, suppression of DNA2 or RECQL4/5 improves USP50-depleted cell resistance to agents inducing replicative stress and restores telomere stability. These data define an unexpected regulatory protein that promotes the balance of helicase and nuclease use at ongoing and stalled replication forks.

Fig. 1. USP50 Ile-141 promotes Ub-binding and chromatin recruitment after HU treatment.

Where included, graphs indicate the mean ± SEM, exact P values are shown, and number of biological repeats is listed (n). All statistical analysis in this figure was performed using a two-tailed unpaired t test. Source data are provided with this paper. A USP50:Ub interaction predicted by Alphafold2. Electrostatic densities of invariant USP50 residues are shown (yellow). In the inset Isoleucine-44 of Ub and Isoleucine-141 of USP50 are shown as electrostatic density. B Immunoprecipitation of FLAG epitopes from HeLa cells expressing FLAG-USP50 or I141R-FLAG-USP50 and Myc-Ub, probed for FLAG and Myc. Mean Myc-Ub normalised to both Myc-Ub and FLAG-USP50 expression in the whole-cell lysate. n = 3. C Immunoblot of whole-cell lysate and the chromatin fraction of FLAG-USP50 or I141R-FLAG-USP50, Tubulin and H2B from cells untreated or treated with 5 mM HU for 3 hours. Graph shows mean FLAG-USP50 in the chromatin fraction relative to FLAG-USP50 in the untreated sample. n = 3. D FLAG-USP50 expressing cells were treated with VCPi (CB-5083) for 3 hours, or MG132 for 4 hours before lysis. Immunoblots show whole-cell lysate and chromatin fraction. FLAG-USP50 expression for NTC vs 10 µM MG132 was quantified and shown for both the whole-cell extract (top right graph) and chromatin-enriched fraction (bottom right). n = 2. E Mean proximity ligation assay (PLA) foci of FLAG-USP50 or I141R-FLAG-USP50 with Biotin-EdU. DNA was either labelled with EdU for 24 hours (bottom left) or 15 mins followed by 3 hours 5 mM HU treatment (bottom right). Representative images (top) of the 24 hours EdU incubation with 10 µm scale bar. n = 3, >150 cells per condition. F Mean PLA foci of FLAG-USP50 or I141R-FLAG-USP50 with Biotin-EdU after 5 mins EdU treatment followed by 3 hours 5 mM HU, with, or without, expression of Myc-Ub. Control conditions were pre-treatment with 5 μM MG132 1 hour before EdU treatment (pMG132) or 100 μM thymidine added for 5 mins after EdU treatment, before HU treatment (Thy chase). n = 3, >196 cells per condition.

Fig. 2. USP50 promotes replication in unperturbed and stressed conditions.

Where included, graphs indicate the mean ± SEM, exact P values are shown, and number of biological repeats is listed (n). Other than 2H which was two-way ANOVA, all statistical analysis in this figure was performed using a two-tailed unpaired t test. Source data are provided with this paper. A Mean% of first-label terminations after non-targeting control (NTC) siRNA (−) or shUSP50 (+) and complemented with FLAG-USP50 (left), I141R-FLAG-USP50 (right) or uninduced (−). n = 3, >195 fibres per condition. B Mean ratio of second-label tracts either side of first label after treatment as in A. n = 3, >35 first-label origins per condition. C Mean% of stalled forks after treatment as in A. n = 3, >240 fibres per condition. D Immunoblot of RPA, pRPA and vinculin following 3 hours 5 mM HU. Quantification shows mean pRPA from n = 5. E Native BrdU tracts length after treatment as in A and with HU. n = 3, >1400 tracks per condition. F 53BP1 foci numbers after treatment as in A. n = 3, >150 cells per condition. G 53BP1 foci numbers after treatment as in A. n = 2, >100 cells per condition. H Colony survival after treatment as in A and 16 hours HU. n = 4. I Colony survival after siNTC or shUSP50 and 24 hours Pyridostatin. n = 2. J 53BP1 foci numbers after siNTC (−) or shUSP50 (+), with or without 24 hours 100 µM Pyridostatin. n = 3, 250 cells per condition. K GC content of break-adjacent heximeric sequences enriched or reduced in USP50:siNTC treated cells from overlapping sequences between n = 2 biological repeats. Occ-p: p value of the occurrence difference. Statistical test: hypergeometric for over/under-representation. All significant sequences and p values shown in Supplementary Fig. 4B.

Fig. 3. USP50 promotes WRN-FEN1 localisation at stalled forks.

Where included, graphs indicate the mean ± SEM, exact P values are shown, and number of biological repeats is listed (n). All statistical analysis in this figure was performed using a two-tailed unpaired t test. Source data are provided with this paper. A Mean% of stalled forks after shUSP50 and WRN siRNA. n = 3, >200 fibres per condition. B Immunoblot of indicated proteins in whole-cell lysate and chromatin fraction after shUSP50 and 3 hours 5 mM HU (representative of two). C Mean proximity ligation assay (PLA) foci of endogenous WRN with HUS1 after siNTC, siWRN, siUSP50 and 3 hours 5 mM HU. n = 3, >145 cells per condition. D Mean% first-label terminations after siRNA to USP50 and GFP-WT-WRN (WT), GFP-E84A-WRN (EA) or GFP-K577M-WRN (KM) expression. n = 3, >200 fibres per condition. E Mean ratio of second-label tracts on either side of first labels after treatment as in D. n = 3, >34 first-label origins per condition. F Mean 53BP1 foci after USP50 siRNA and GFP or GFP-WRN expression. n = 3, >130 cells per condition. G Immunoprecipitation with anti-WRN or control IgG (G) after siNTC, shUSP50 and 3 hours 5 mM HU, probed for WRN or FEN1. Numbers below: ratio of FEN1:WRN relative to siNTC (performed once). H Mean PLA foci of endogenous WRN with FEN1 after siNTC, siFEN1, shUSP50, FLAG-USP50 or I141R-FLAG-USP50 expression and 3 hour 5 mM HU. n = 3, >150 cells per condition. I Immunoblot of indicated proteins input or purified by iPOND after shUSP50 and 3 hours 5 mM HU (representative of three). J Mean% stalled forks after shUSP50 and siRNA to FEN1. n = 3, >200 fibres per condition. K Mean ratio of second-label tracts, either side of first labels after shUSP50 and WT Myc-FEN1 (WT) or E359K-Myc-FEN1 (EK) expression. n = 3, >70 first-label origins measured. L Mean% of stalled forks after treatment as in K. n = 3, >200 fibres per condition. M Mean 53BP1 foci after treatment as in K. n = 3, 150 cells per condition.

Fig. 4. Replication defects in USP50 deficient cells are driven by DNA2.

Where included, graphs indicate the mean ± SEM, exact P values are shown, and number of biological repeats is listed (n). All statistical analysis in this figure was performed using a two-tailed unpaired t test. Source data are provided with this paper. A Mean DNA2 foci after shUSP50, 3 hours 5 mM HU, wash and recovery for 20 mins. Representative images (left) with 10 µm scale bar shown. n = 3, >140 cells per condition. B PLA foci of DNA2 and Biotin-EdU after siNTC (−), DNA2 siRNA, or shUSP50 (+) in cells pulsed with EdU for 15 mins and treated with 3 hours 5 mM HU, or HU-treated and recovered for 20 mins “released”. n = 3, >150 cells per condition. C Mean% of stalled forks after shUSP50, with or without DNA2 siRNA. n = 3, >350 fibres per condition. D Mean% of stalled forks treated as in C. n = 3, >200 fibres per condition. E Immunoblot of pRPA (S4/8) and Vinculin after shUSP50, with and without DNA2 siRNA and 3 hours 5 mM HU (performed twice). F Mean% of stalled forks after shUSP50 with and C5 DNA2i (20 µM). n = 3, >200 fibres per condition. G 53BP1 foci after shUSP50 with and without DNA2 siRNA and 3 hours 5 mM HU for 3 hours washed and recovered for 20 mins. n = 3, >90 cells per condition. H Mean% of chromatids with lagging-strand telomere loss after shUSP50 and DNA2 siRNA. n = 3, scored chromatid numbers per experiment: NTC: 1482, 956, 898; shUSP50: 1446, 652, 970; siDNA2: 1238, 1236, 714; shUSP50/siDNA2: 170, 1350, 286. I Colony survival after USP50 siRNA and 16 hours 1.25 mM HU treatment with or without 20 µM DNA2i. n = 6 for siNTC and siUSP40, n = 4 for conditions including DNA2i treatment.

Fig. 5. Replication defects in USP50 deficient cells are driven by RECQL4/5.

Where included, graphs indicate the mean ± SEM, exact P values are shown, and number of biological repeats is listed (n). All statistical analysis in this figure was performed using a two-tailed unpaired t test. Source data are provided with this paper. A Mean% of stalled forks after shUSP50 with RECQL4 siRNA. n = 3, >200 fibres per condition. B Mean% of stalled forks after shUSP50 with RECQL5 siRNA. siNTC and shUSP50 conditions are shared with 5 A. n = 3, >200 fibres per condition. C Mean% of first-label terminations after shUSP50, with or without RECQL4 or RECQL5 siRNA. n = 3, >200 fibres per condition. D Native BrdU tracts length after shUSP50 and siRNA targeting RECQL4 and RECQL5 or DNA2 and 3 hours 5 mM HU. siNTC and shUSP50 conditions are shared with 2E. n = 3, >1400 tracks per condition. E Mean 53BP1 foci after shUSP50 and RECQL4 and RECQL5 siRNA and 3 hours 5 mM HU followed by 20 mins recovery. n = 3, 150 cells per condition. F Mean% of stalled forks after FEN1 siRNA with and without siRNA targeting RECQL4 and RECQL5. n = 3, >400 fibres per condition. G Mean RECQL4 foci after shUSP50 and 3 hours 5 mM HU, followed by 20 mins recovery. Representative images left, scale bar is 10 µm. n = 3, 150 cells per condition. H Mean PLA foci of RECQL4 with Biotin-EdU after RECQL4 siRNA, or shUSP50 in cells pulsed with EdU for 15 mins and 3 hours 5 mM HU, or HU treated, washed and allowed to recover for 20 mins (HU release). n = 3, >150 cells per condition. I Mean PLA foci of RECQL4 with Biotin-EdU after RECQL5 siRNA, or shUSP50 in cells pulsed with EdU for 15 mins and 3 hours 5 mM HU, or HU treated, washed and allowed to recover for 20 mins (HU release). n = 4, >200 cells per condition. J Colony survival after shUSP50 and siRNAs to DNA2 or RECQL4 and RECQL5 and 24 hours 25 μM Pyridostatin. n = 7 for siNTC and shUSP50, n = 3 for all other conditions.

Fig. 6. Model of the influence of USP50 on WRN-FEN1, RECQL4/5, and DNA2 during replication.

A USP50 promotes the recruitment of WRN-FEN1 to stalled replication structures. Helicase and nuclease competent WRN (dark blue), and GEN and WRN interaction-competent FEN1 (light blue) can rescue the lack of USP50 to promote ongoing replication and suppress DSB foci. B Without USP50, increased RECQL4 and RECQL5 (dark blue) and DNA2 (light blue) result in extended resection and MUS81-dependent DNA breaks. DNA DSBs are more common near CG-rich sequences.