Patient samples. A total of 67 human ocular melanoma tissues, 14 human nevi tissues and 5 adjacent normal tissues were collected for immunohistochemistry (IHC) from Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine from 2007 to 2017. Patients’ clinical and demographic features are detailed in Supplementary Table 6. The study was performed in accordance with World Medical Association Declaration of Helsinki and was approved by the Ethics Committee of Shanghai Jiao Tong University.
Immunohistochemistry (IHC). IHC staining of tissue slides was performed by Servicebio (Wuhan, China). Tissues were deparaffinized and rehydrated through an alcohol series, followed by antigen retrieval with sodium citrate buffer. Then tissue sections were blocked with 3% bovine serum albumin (BSA) 30 min at room temperature and then incubated with anti-A3B (abcam, ab184990, 1:100) antibody at 4°C overnight. Finally, the tissues were covered with horseradish peroxidase (HRP) labeled secondary antibody and incubated at room temperature for 50 minutes. All immunostained slides were scanned on 3D Histech Quant Center (3D Histech, Hungary), and computerized image analysis was performed by Halo (Indica labs, USA). Immunostaining for A3B was analyzed in adjacent normal tissues, nevi tissues and melanoma tissues using percentage of positive cells and histochemistry score (H-score). H-score was determined based on the proportion of positive cells and the intensity of nuclear staining as previously described 48. H-score=∑(pi×i)= (percentage of cells of weak intensity ×1) + (percentage of cells of moderate intensity ×2)+ (percentage of cells of strong intensity ×3). In the formula, pi represents the percentage of positive cells in the slide; i represents the intensity of A3B staining.
Cell culture. The adult retinal pigment epithelium cell line ARPE-19 was obtained from the Cell Bank/Stem Cell Bank (Chinese Academy of Sciences). Human cutaneous melanocyte cell line PIG1 was obtained from the Department of Ophthalmology, Peking University Third Hospital. The human UVM cell lines OMM1, OMM2.3, MEL285 and MEL290 and conjunctiva melanoma (CoM) cell lines CRMM1, CRMM2, and CM2005.1 were kind gifts from Prof. Martine J. Jager (Leiden University Medical Center, Leiden, The Netherlands). The human UVM cell line MUM2B and 92.1 were kindly supplied by Prof. John F. Marshall (Tumor Biology Laboratory, Cancer Research UK Clinical Center, John Vane Science Centre, London, UK). HEK293T human embryonic kidney cell line was purchased from the American Type Culture Collection (Manassas, VA, USA). The cell lines used in this study were authenticated by short tandem repeat (STR) profiling. Human HEK293T, A375 and A2058 cells was cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco). Other cells were cultured in RPMI1640 medium (Gibco). All mediums are supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin/streptomycin (P/S; Thermo Fisher Scientific). All cells were cultured in a 37℃ humidified incubator containing 5% CO2.
RNA extraction and quantification. RNA purification was performed using EZ-press RNA purification kit (EZBioscience, B0004DP) according to the manufacturer’s guidelines. Then, cDNA synthesis was achieved using the PrimeScript RT Master Mix (Takara, Japan). Finally, real-time quantitative PCR (RT-qPCR) was carried out using the SYBR Green qPCR Master Mix (Applied Biosystems). mRNA expression values were calculated using the ΔΔCt method and GAPDH gene as a control. A detailed list with primers used in the present study is provided in Supplementary Table 7.
Western blotting. Total cell lysates were prepared in RIPA lysis buffer (Beyotime, P0013B). Protein samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore). After blocking with 5% non-fat milk for 1 h at room temperature, the membranes were incubated with anti-p-RPA32 (BETHYL, A300-245A, 1:2,000), anti-RPA70 (CST, 2267S, 1:1,000), anti-A3B (Abcam, ab184990, 1:1,000), anti-GFP (Invitrogen, A-11122, 1:1000) or anti-V5 (Abcam, ab15828, 1:2000) antibodies overnight at 4 ℃ and then with the appropriate secondary antibodies conjugated to a fluorescent tag (Invitrogen) for 1 h at room temperature. Anti-β-Actin (Proteintech, 66009-1-Ig, 1:20000) antibody served as the loading control. The immunoblots were recorded with the Odyssey infrared imaging system (LI-COR Biosciences, USA).
Co-immunoprecipitation (co-IP). Cells were pelleted and washed with PBS, and then lysates were prepared in 500 µL lysis buffer containing 120 mM NaCl, 20 mM Tris-Cl, 2 mM EDTA, 1% NP40 and 5% Glycerol supplemented with 1× protease inhibitor cocktail (Roche). Anti-S9.6 antibody (Kerafast, ENH001, 1:100) or normal mouse IgG (Santa cruz, sc-2025, 1:10000) was incubated with the cell lysates overnight at 4 ℃, after which 30 µL protein A magnetic beads (CST, #73778) were added and incubated for 2 additional hours. Then, the magnetic beads were washed three times with lysis buffer. For IP-MS analysis, 100 µL glycine solution was used for each tube to elute the protein complexes from the beads, while 1 × SDS loading buffer (NCM, China) was used per sample for SDS-PAGE analysis.
Transfection and virus packaging. Three shRNA sequences targeting A3B were cloned into the pGIPZ-TurboGFP-puro vector. The A3B targeting sequences were: 5’-TAAAGTTGAAAGTGAATGTGTT-3’(shA3B_1); 5’-TTAAAGTTGAAAGTGAATGTGG-3’ (shA3B_2). The full length of human A3B (isoform NM_004900.5) and RNASEH1_EGFP (isoform NM_002936.3) were cloned into the pHAGE-puro vector respectively for overexpression of A3B and RNASEH1. PolyJet DNA In Vitro Transfection Reagent (SignaGen) was used for plasmid transfection following the manufacturer’s instructions. After lentiviral packaging with HEK293T cells, cells were infected with lentiviruses and selected by incubation with 2 µg/mL puromycin for 3 days. For transfection of RNASEH1 D210N mutant construct, cells were transfected with ppyCAG_RNASEH1_D210N_V5 plasmid (Addgene, 111904) and selected with 300 µg/mL Hygromycin B (Selleck, S2908) for 3 days.
Colony formation. A volume of 2 mL of complete medium containing 1,000 cells was seeded in each well of a 12-well plate and incubated for about 2 weeks. For drug sensitivity colony formation assay, cells were treated with the indicated drug concentrations every 48 h. Once colonies formed in DMSO control conditions, plates were fixed with 4% paraformaldehyde, stained with crystal violet and the number of colonies was counted. Colony formation efficiency indicated by area of colonies treated relative to area of control colonies was used to measure survival between wells. RSR inhibitors and chemotherapy drugs for drug sensitivity assay were purchased from Selleck: VE-821 (S8007) targeting ATR, VE-822 (S7102) highly selective and potent derivatives of VE-821, Rabusertib (S2626) targeting Chk1, Cisplatin (S1166), Carboplatin (S1215), Dacarbazine (S1221).
Cell viability. To determine cell viability, MEL290 cells were seeded in 96-well plates at a density of 1000 cells per well and treated with 2.5 nM BAY-1895344 (Selleck, S8666) or 2 µM VE-822. After incubation with 10 µL CCK-8 reagent (Dojindo Laboratories, Japan) per well, the absorbance was measured at a wavelength of 450 nm at the indicated time points. The data were recorded and analyzed.
Immunofluorescence (IF). S9.6 IF was performed essentially as described49. Briefly, cells were fixed with 100% ice-cold methanol, blocked with 2% BSA for 1 h at room temperature and incubated with S9.6 (Kerafast, ENH001; 1:200) overnight at 4°C. Then coverslips were washed three times in phosphate buffered saline (PBS) and incubated with Alexa Fluor 594 secondary antibody (Invitrogen, #A21203; 1:1000) for 1 h at room temperature. Finally, cells were stained with DAPI and mounted in ProLong Gold AntiFade reagent (Invitrogen). For DNA replication stress assessment by p-RPA32 (S4/S8) immunostaining, cells adhering to a glass slide were fixed with 4% paraformaldehyde (Biosharp, 70071800) for 15min, permeabilized with 0.1% Triton X-100 for 15min and then blocked with 2% BSA solution for 1 h at room temperature. After incubation with primary antibody against p-RPA32 (S4/S8) (BETHYL, A300-245A, 1:2000) overnight, the cells were washed three times with PBS and subsequently incubated with Alexa Fluor 594 secondary antibody (Invitrogen, A11012, 1:1000) for 1 h. The coverslips were then mounted with ProLong Gold mounting medium with DAPI (Invitrogen, P36931) and observed under an inverted fluorescence microscope (Nikon). Images were acquired with Leica TCS SP8 confocal microscope (Leica, Mannheim, Germany). Nuclear intensity was quantified using FIJI (ImageJ) image processing package. The nuclear mean gray value for S9.6 was measured for each condition. Percentage of cells with more than 10 p-RPA32 foci was quantified. All results presented were obtained from three independent biological replicates; at least 100 cells were measured per replicate.
RNA-seq. RNA sequencing was performed by Novogene (Beijing, China). Total RNA was harvested from MEL290 cells overexpressing A3B and/or RNASEH1 and 92.1 cells transfected with A3B knockdown or empty vector using Trizol. The integrity of the RNA was assessed by Agilent 2100 bioanalyzer (Thermo Fisher Scientific, MA, USA). Approximately 1 µg mRNA from each sample was used for RNA sequencing (Illumina HiSeq PE150 platform). Short reads were aligned to the reference human genome GRCh38 using STAR50. Gene expression in terms of Fragments Per Kilobase of transcript per Million mapped reads (FPKM) of genes overlapping with gene annotations in Ensembl release 75 was calculated by Cufflinks51. Differential gene expression was determined using the Cufflinks Cuffdiff package51. Gene Set Enrichment Analysis (GSEA) was performed to interpret the function of regulated genes using the cancer hallmark gene sets, canonical pathways gene sets and GO biological process gene sets.
ChIP-seq. ChIP assay was performed as previously described, with minor modifications52. Briefly, OMM2.3 and 92.1 (1 × 107 cells) were cross-linked for 10 min at room temperature with 1% formaldehyde-containing medium. The reaction was then halted using a 1× glycine solution. Then cells were pelleted before resuspension in sonication buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 0.5 mM EGTA and 1 × protease inhibitor cocktail). The cellular chromatin was then sonicated to lengths between 100 and 600 bp. 10% chromatin for each ChIP reaction was kept as input DNA. The rest chromatin was then incubated with anti-A3B antibody (abcam, ab184990, 1:50) or normal rabbit IgG (CST, 2729, 1:500) at 4℃ overnight. 30 µL magnetic protein A/G beads (CST) were then added for another 1 h of incubation at 4℃ before going through a series of salt washes. Chromatin was then eluted from the magnetic beads in the elution buffer at 65°C for 15 minutes while vortexing. The supernatant was removed and treated with RNase A followed by Proteinase K. ChIP DNA was then purified using MinElute PCR Purification Kit (QIAGEN, 28006). The ChIP-seq libraries were constructed by Novogene (Beijing, China). Pair-end sequencing of sample was performed on Illumina platform (Illumina, CA, USA). Library quality was assessed on the Agilent Bioanalyzer 2100 system. Raw data of fastq format was processed using FastQC 0.11.9 software53. Clean reads were aligned to the reference genome GRCh38 using Bowtie2 2.2.554 to generate bam files. After removing PCR duplicates with Sambamba 0.8.2 software55, we used MACS2 2.2.7.1 to identify regions of IP enrichment over background56. A p-value threshold of 0.05 was used for all data sets. Deeptools 3.5.1 was used to transfer deduplicated bam file to BigWig format and to generate heatmaps and density plot Figs. 57. ChIPseeker package and ChIPpeakAnno package were used for peak annotation and visualization58, 59.
CUT&Tag. CUT&Tag assay was performed using Hyperactive In-Situ ChIP Library Prep Kit for Illumina (Vazyme, TD901-TD902) following the manufacturer’s protocol. Briefly, 5 × 105 OMM2.3 cells transfected with V5 tagged RNASEH1_D210N were collected. After binding to concanavalin A–coated magnetic beads (ConA beads), bead-bound cells were incubated with anti-V5 antibody (Abcam, ab15828, 5ug/1:10) or normal rabbit IgG (CST, 2729, 0.5ug/1:100) for 2h at room temperature. After brief wash with dig-wash buffer, cells were then incubated with goat anti rabbit secondary antibody (Abcam, ab6702, 1:100) for 1h at room temperature. When antibody binding procedures were finished, the bead-bound cells were then mixed with hyperactive pA-Tn5 transposon and tagmentated with tagmentation buffer. Tagmentated DNA was then extracted and amplified to form the sequencing-ready libraries. After the PCR reaction, libraries were purified with the DNA clean beads (Vazyme, N411) and library quality was assessed on the Agilent Bioanalyzer 2100 system. The clustering of the index-coded samples was performed on a cBot Cluster Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumina). The library preparations were sequenced on Illumina Novaseq platform at Novogene (Beijing, China) and 150bp paired-end reads were generated. Raw data was firstly processed using FastQC 0.11.9. Clean data was obtained by removing adapters using Cutadapt 4.1 with default parameters60. Paired-end reads were aligned to genome GRCh38 using Bowtie2 2.2.5 as follows: bowtie2 --local --very-sensitive --no-mixed --no-discordant --phred33 -I 10 -X 700. After removing the duplicated reads with Picard 2.27.3 (https://broadinstitute.github.io/picard/), we calculated the read coverage and depth with Samtools 1.15.161. To ensure that our data were of high quality and reproducibility, we filtered data with minQualityScore༞2 using Samtools 1.15.1. Then we filtered and kept the mapped read pairs. Bedtools was used to convert files into bed file format and calculate reads coverage at the genome level. Peak calling was performed with SEACR 1.362. EdgeR package was used to get differential peaks63. BigWig file and reads distributions across genes were acquired using Deeptools 3.5.1. Peak annotation was performed with ChIPseeker and ChIPpeakAnno package as previously described. Gene ontology enrichment analysis was implemented by Metascape (www.metascape.org). In Venn diagrams, numbers represent genes co-occurring between conditions.
TCGA data analysis. Two TCGA data sets (PRAD and UVM) were used to test the correlation between the selected genes and patient survival. Based on the expression levels of A3B and RNASEH1 and their median expression values, respectively, patients were further discretized as four groups: “A3B High and RNASEH1 High”, “A3B High and RNASEH1 Low”, “A3B Low and RNASEH1 High” and “A3B Low and RNASEH1 Low”. The statistical analysis was performed by the R package ‘‘survival’’ and survival curves were fitted by the survfit function. RSR signature score was calculated based on the median expression levels of the RSR signature gene sets (Supplementary Table 8)64.
Statistical analysis. All statistical analyses were performed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA), employing unpaired t test. Descriptive values are presented as mean ± standard error unless stated elsewhere. Differences between two groups were analyzed by unpaired Student’s t-test while differences among multiple groups were analyzed by one-way analysis of variance (ANOVA) with post-hoc intergroup comparisons with Turkey’s test. Results were considered statistically significant when p < 0.05. Pearson correlation analysis was performed on GEPIA (http://gepia.cancer-pku.cn/).