Retinoic Acid-Induced 2 Contributes to Proficient Homologous Recombination and Maintains Genomic Stability in Breast Cancer

Retinoic Acid-Induced 2 Contributes to Proficient Homologous Recombination and Maintains Genomic Stability in Breast Cancer | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Retinoic Acid-Induced 2 Contributes to Proficient Homologous Recombination and Maintains Genomic Stability in Breast Cancer Lena Boettcher, Sarah Greimeier, Kerstin Borgmann, Shabbir S. Mughal, and 19 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3908810/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Genome instability is a fundamental feature and hallmark of cancer associated with aggressiveness, drug resistance and poor prognosis. RAI2 was initially identified as a novel metastasis suppressor protein specifically associated with the presence of disseminated tumour cells in the bone marrow of breast cancer patients, but its molecular function is largely unknown. Methods We analysed the consequences of RAI2 depletion on gene expression and genomic stability in luminal breast cancer cell lines, performed cytotoxicity profiling using a library of pharmacologically active compounds, and characterized the function of the RAI2 protein in the DNA damage response. We performed in silico validation in different breast cancer datasets. Results Analysis of clinical samples revealed that in primary breast tumours, low RAI2 gene expression is significantly associated with genomically unstable tumours and poor prognosis. RAI2 depletion in breast cancer cell lines resulted in loss of mitotic fidelity characterized by prolonged mitosis with increased chromosome segregation errors and micronuclei formation. Drug screening revealed increased sensitivity of RAI2-depleted breast cancer cells to topoisomerase I and Aurora A inhibitors. We also found that genotoxic stress induces RAI2 protein, which shows affinity for poly-(ADP-ribose) and contributes to efficient DNA repair by homologous recombination. We validated the functional association of RAI2 gene expression with DNA double-strand break repair capacity in clinical samples. Conclusions Our findings support, for the first time, an important functional role of RAI2 in the maintenance of mitotic fidelity and DNA repair associated with early metastatic relapse. The underlying molecular mechanisms could be exploited to improve patient diagnosis and treatment. breast cancer genomic instability premature mitotic entry homologous recombination poly-(ADP-ribose) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Metastasis is a multistep process involving the hematogenous dissemination of cells from the primary tumor to distant sites such as the bone marrow, where the cells can either remain dormant as disseminated tumor cells (DTCs) or begin to proliferate and form overt metastases ( 1 ). Due to its complex phenotype, the acquisition of metastatic competence represents a major selective barrier during cancer progression. It has been proposed that such a macroevolutionary leap leading to metastatic competence of tumor cells can only be achieved by large-scale genomic alterations leading to chromosomal instability (CIN) ( 2 ). CIN is the predominant mechanism leading to genomic instability ( 3 ). CIN can arise through either mitotic or premitotic defects ( 4 ) and is associated with aggressiveness, drug resistance, and poor prognosis ( 5 , 6 ). Recent evidence also suggests that genomic instability may shape the antitumor immune response ( 7 ). Fundamental studies have confirmed that intra-tumor heterogeneity mediated by genome doubling and ongoing dynamic CIN drives disease recurrence in both renal cell and lung cancer. Interestingly, this effect was not associated with mutational burden ( 8 , 9 ). More recently, the evolution of late-stage metastatic melanoma has been shown to be dominated by aneuploidy and whole genome doubling ( 10 ). Previously, RAI2 was identified as a novel metastasis suppressor gene associated with both breast and colorectal cancer metastasis, but its molecular function is still largely unknown ( 11 , 12 ). Low RAI2 gene expression in primary breast cancer is significantly correlated with the mutational status of the TP53 gene ( 11 ), which led us to test whether low RAI2 gene expression might also be associated with genomic instability. Here we provide the first evidence that the RAI2 protein maintains genomic stability by enabling efficient repair of DNA double strand breaks (DSB) by homologous recombination (HR), which contributes to the suppression of metastatic relapse in human breast cancer. In addition, RAI2 depletion appears to have synergistic effects with anticancer agents that induce DNA damage, which may further influence the response to chemotherapeutic treatment or progression from minimal residual disease (MRD) to overt metastasis. Methods A comprehensive description of all materials and methods is provided as Supplementary Information. The description of live cell imaging, cytotoxicity profiling and immunoprecipitation combined with protein analysis by quantitative mass spectrometry proteomics, the rigor adherence table (Table S1 ) and the key resources table (Table S2 ) are provided as Supplementary Information only. Clinical in silico validation We used the Bioconductor package (RRID:SCR_006442) to download TCGA-BRCA transcriptome profiles, copy number segments, and clinical data. Normalized gene expression data for RAI2 and other genes of interest in the METABRIC dataset were obtained from cBioPortal. Normalized RAI2 gene expression was correlated with the expression of genes indicative of CIN (CIN70 score) using the weighted genome instability index (wGII). P-values were calculated by Student's t-test. Multivariable regression analysis was performed to evaluate the association between CIN70 score and RAI2 gene expression, adjusting for other covariates such as oestrogen receptor α (ER) status and PAM50 subtype and TP53 status. For linear regression, RAI2 levels were log-transformed. For multivariable analysis, the Cox regression model was used, including those histopathological factors that were clinically significant in the univariable survival analysis in the multivariable analysis (tumour stage, ER status, HER2 status, grade, and molecular subtype) and presented as hazard ratio with 95% confidence interval. To assess whether RAI2 gene expression correlated with patterns of genomic instability, copy number alteration (burden total number of somatic copy number variations (SCNAs) per tumour type) and point mutations (PM) burden were summed. Spearman correlations were performed to compare mRNA abundance and somatic variant burden on a per-gene basis. P-values were corrected using the Benjamini-Hochberg method to obtain false discovery rates (FDRs). For survival analysis, samples from the METABRIC dataset were divided by the median of RAI2 gene expression and CIN70 score. Differences in five-year overall survival between these groups were determined by Kaplan-Meier analysis using the log-rank test. The Recombination Proficiency Score (RPS) was calculated for each tumour sample using normalized expression values for four signature genes involved in the DNA repair pathway ( RIF1 , PARI, RAD51 , and KU80 ). Cell culture MCF-7 (RRID:CVCL_0031), KPL-1 (RRID:CVCL_2094), MCF-10A (RRID:CVCL_0598), CAMA-1 (RRID:CVCL_1115) and 293T (ATTC #CRL-3216) cells were cultured under standard conditions. Cell line authentication was performed using short tandem repeat profiling to exclude cross-contamination between cell lines. Cell lines were tested monthly for mycoplasma contamination. Plasmids, viral transduction and transfection Plasmid construction for overexpression of wild-type RAI2 protein, as well as viral production and transduction procedures, have been described previously ( 11 ). For knockdown experiments, we used plasmids derived from pLKO.1: non-target shRNA (Sigma-Aldrich #SHC016), shRNA1 (TRCN0000139927), and shRNA2 (TRCN0000441623). We used the pH2B-eYFP plasmid (plasmid #51002) to establish KPL-1 cells with constitutive H2B-eYFP overexpression. Gene expression profiling 500 ng total RNA from KPL-1 breast cancer cells was hybridized to the Illumina HT-12 Array v4 BeadChip (Illumina, San Diego, CA, USA) according to the manufacturer's protocols. Microarrays were scanned using the Illumina iScan scanner (Illumina, San Diego, CA, USA) according to the standard Illumina scanning protocol. Bead-level data were aggregated using BeadStudio and normalized using the quantile method. Differential expression on normalized expression data was determined using the samr package (R version 3.4.2) with an unpaired t-test with a delta of 0.06 and a minimum twofold change in gene expression. Functional annotation of differentially expressed genes was performed using the gene functional classification tool in DAVID Bioinformatics Resources 6.8 (RRID:SCR_001881). Quantitative real-time RT-PCR Analysis (qRT-PCR) RNA was extracted from cultured cells during the exponential growth phase using the Nucleospin RNA Kit (Macherey Nagel, Germany) according to the standard protocol. 1000 ng of RNA from each sample was transcribed using the First Strand cDNA Synthesis Kit (Thermo Scientific, MA, USA) and random hexamers. Human Cell Cycle RT² Profiler™ PCR Arrays and reagents (Eppendorf, Germany) were used for cell cycle focused gene expression analysis. The qRT-PCR reactions were performed in triplicate using the Mastercycler Eppendorf Realplex thermal cycler (Eppendorf, Germany). Data analysis and significance testing were performed using QIAGEN GeneGlobe Data Analysis Centre (RRID:SCR_021211). Western blotting Whole cell extracts from cultured cells were prepared by direct lysis and sonication of cells in 2% SDS sample buffer containing phosphatase and protease inhibitors. Cell extracts were separated on 8–15% denaturing polyacrylamide gels and blotted onto nitrocellulose or PVDF membranes. The following antibodies were used for detection: RAI2 (RRID:AB_2800292), Aurora A (RRID:AB_2665504), Aurora B (RRID:AB_10695307), Cyclin A2 (RRID:AB_627334), Cyclin B1 (RRID: AB_2783553), cyclin B2 (RRID:AB_2072392), survivin (RRID:AB_2063948), γH2AX (RRID:AB_2118009), and HSC-70 (RRID:AB_627761). HRP-conjugated anti-rabbit IgG, (RRID:AB_2099233) and (RRID:AB_330924) or infrared dye-labelled anti-rabbit IgG, (RRID:AB_621843) and anti-mouse IgG (RRID:AB_10956588) were used for detection. Differences between signal intensities in different cell lines and results of three independent experiments were evaluated by two-tailed Student's t-test. Cell cycle analysis Cell cycle profiles were assessed by quantifying DNA content using flow cytometry. Cells were fixed in 4% formaldehyde, treated with RNase and stained with propidium iodide. Flow cytometric analysis was performed using a NovoCyte Quanteon (Agilent, CA, USA) flow cytometer. A minimum of 20,000 cells were collected for analysis. After doublet discrimination, cell cycle profiles were automatically calculated using NovoExpress software (Agilent, CA, USA). The Watson model was used for cell cycle fitting. To determine the mitotic fraction, flow cytometry analysis of P-H3(S10) (RRID:AB_1549592) stained cells was performed using a FACS CantoII (Becton Dickinson, NJ, USA) equipped with FACSDiva software (RRID:SCR_001456). Three independent biological replicates were used to calculate the percentage of each cell cycle phase and the standard deviation. The difference was tested by two-tailed t-test. Immunofluorescence staining Cells were fixed in 4% paraformaldehyde, washed three times with PBS, and permeabilized with 0.2% Triton X-100 in PBS. After incubation with 1% non-fat dry milk (w/v) in PBS for 30 minutes, cells were further incubated with primary antibodies: P-H3(S10) (RRID:AB_1549592), anti-centrosome (RRID:AB_212756), RAI2 (RRID:AB_2800292), CtBP1 (RRID:AB_399429), poly(ADP-ribose) (RRID:AB_785249), γH2AX (RRID:AB_2118009), or 53BP1 (RRID:AB_2921289). Specific antibody binding was visualized with Alexa Fluor 488 goat anti-rabbit IgG (RRID:AB_143165) and Alexa Fluor 546 goat anti-mouse IgG (RRID:AB_2534093). Confocal laser scanning microscopy was performed using a Leica TCS SP5 microscope (Leica, Germany) and Imaris imaging software (RRID:SCR_007370) to identify and measure the number of γH2AX and 53BP1 foci. At least 100 independent events were evaluated for quantification, and differences were tested by two-tailed t-test. Metaphase spread analysis For metaphase spreads, exponentially growing KPL-1 cells were treated overnight with Colcemid (0.02 µg/ml), incubated with 0.0075 M KCl, fixed with methanol/acetic acid (3:1), dropped onto slides, stained with 5% Giemsa and mounted with Entellan before imaging with a Zeiss Axioplan 2 microscope (Zeiss, Germany). 100 metaphases per experiment were counted in three independent experiments. Traffic light reporter assay The BFP-TLR-SceI plasmid (Addgene plasmid #31481) was digested with SceI (New England Biolabs, MA, USA) for 4 hours, run on an agarose gel, and purified with a purification kit (Macherey Nagel, Germany) before being used for transfection. For RAI2 overexpression, HEK293T cells were first transfected with phCMV3-RAI2-HA ( 11 ) using OPTIMEM (Thermo Fisher, MA, USA) and Lipofectamin2000 (Invitrogen, MA, USA) according to standard protocol. After 24 hours, the cells were cotransfected with 500 ng cut BFP-TLR-SceI and GFP donor plasmid (Addgene MA, USA, plasmid #31475). At 48 hours post-transfection, trypsinised cells were quenched with media and mCherry, eGFP and BFP fluorescence signal was analysed by flow cytometry (LSR Fortessa, Becton Dickinson, NJ, USA) using 561 nm, 488 nm and 405 nm laser. The percentage of mCherry (for NHEJ events) and eGFP (for HR events) positive cells in the BFP positive cell fraction was used for analysis. Three independent biological replicates were used to calculate relative DNA repair efficiency. Differences were tested by two-tailed t-test. Results Low RAI2 Expression is a Feature of Genetic Unstable Breast Carcinomas To assess whether RAI2 expression is associated with genomic instability in breast cancer patients, we performed an analysis of large published clinical datasets. First, we calculated the weighted genome integrity index (wGII) as an independent measure of CIN ( 13 ) and divided samples into low and high wGII groups based on the median cut-off. We found that RAI2 gene expression was clinically lower in tumors with a high wGII score (Fig. 1 A). We confirmed clinically differences in RAI2 gene expression in estrogen receptor (ER)-positive and -negative subgroups (Fig. 1 B) as well as in basal, luminal A and B molecular subtypes (Fig. 1 C). We then examined RAI2 gene expression in patient groups stratified according to the CIN70 signature of chromosomal instability derived from gene expression profiles ( 14 , 15 ). Consistent with the results from wGII, we observed clinically lower RAI2 gene expression in samples with a high CIN70 score (Fig. 1 D), ER-positive tumors showed higher RAI2 gene expression compared to ER-negative tumors (Fig. 1 E). However, the overall trend remained consistent in all groups of basal, luminal A and B molecular subtypes (Fig. 1 F). For validation a correlation of RAI2 gene expression with the complete CIN70 score was assessed in the METABRIC breast cancer dataset ( 16 ). Univariable analyses showed a clinically significant association between low RAI2 gene expression and high CIN70 score (Figure S1 A-C). Also, in multivariable linear regression analysis a significant association between the CIN70 score and low RAI2 gene expression was found in the METABRIC data set (p < 0.001, multiplicative factor = 0.999 [CI:0.998–0.999]). Likewise, the basal subtype (p = 0.002, multiplicative factor = 0.949 [CI:0.918–0.982]) and HER2 (p = 0.038, multiplicative factor = 0.938 [CI:0.910–0.967]) showed an influence on RAI2 gene expression compared to luminal A, whereas for ER and TP53 no influence was seen (Figure S1 D). In the cancer genome atlas (TCGA) dataset of the invasive breast carcinoma we investigated a possible correlation of RAI2 gene expression with somatic copy number alterations (SCNA) and point mutations (PM) by a previously published approach ( 17 ). RAI2 was among the top genes (rank = 36 out of 12961) whose expression correlated inversely with aneuploidy (spearman rho = -0.47, FDR = 4.96 e-40) (Fig. 1 G). Taken together these results confirm that compared to other genes in the genome there is a strong correlation between low RAI2 gene expression and different characteristics of genome instability in primary breast tumors. Low RAI2 expression predicts poor clinical outcome in breast cancer patients We then evaluated whether low RAI2 gene expression identifies an aggressive subset of tumors associated with poor clinical outcome in the METABRIC dataset. Five-year overall survival analysis of patients showed that low RAI2 gene expression combined with a high CIN score had a worse prognosis compared to other patient groups (p < 0.001). At five years, 36.2% of early relapse patients with low RAI2 gene expression and high CIN score had died, while the best survival rate with only 9.4% of deaths was seen in patients with high RAI2 gene expression and low CIN score. Patients with a mixed phenotype had intermediate 5-year survival rates of 79.4% and 82.1% (Fig. 1 H). Multivariable analysis showed that the combined RAI2 low /CIN high phenotype was an independent poor prognostic factor (hazard ratio: 1.60 [CI: 1.05–2.42, p = 0.027]). Other independent risk parameters were grade (hazard ratio: 1.46 [CI: 1.08–1.99, p = 0.017]), ER status (hazard ratio: 1.88 [CI: 1.28–2.75, p = 0.001]) and stage (hazard ratio: 2.2 [1.63–2.98, p < 0.001]). Taken together, these results indicate that low RAI2 gene expression is a hallmark of genetically unstable tumors and that these patients have a significantly higher risk of early metastatic relapse. RAI2 depletion in human breast cancer cells causes deregulation of cell cycle-related genes and proteins To analyze whether loss of RAI2 protein function causes genomic instability, we first performed gene expression profiling by microarray analysis of the luminal breast cancer cell line KPL-1 after shRNA mediated RAI2 depletion, which has previously been shown to be a suitable cell line model to study RAI2 protein function ( 11 ). Functional annotation revealed that the significantly deregulated genes are associated with the cell cycle, including deregulation of genes involved in microtubule motor activity, mitosis, and spindle apparatus (Fig. 2 A and Supplementary Table S3 and S4). Therefore, the dysregulation of cell cycle-related genes was analyzed in more detail in additional RAI2-depleted cell lines (KPL-1, CAMA-1, MCF-7 and MCF-10A) using pathway-specific arrays (Cell Cycle RT 2 Profiler PCR Arrays). As shown in Fig. 2 B and Supplementary Table S5, RAI2 depletion leads to decreased gene expression of AURKB, CCNA2, CCNB1, CCNB2, CDC20, CDC6, CDK1, CDK5R1, GTSE1, MAD2L2, MCM2, MCM3, MCM4 and MKI67 in all tested cell lines. The downregulation of selected gene products was validated by Western blot using two independent RAI2-specific shRNA sequences (Fig. 2 C). We found reduced protein expression of Aurora A and B, Cyclin A2 in all three RAI2-depleted cell lines, and Cyclin B2 was significantly reduced only in KPL-1 and CAMA-1 cells. Since all these genes and proteins are known to be periodically regulated within the cell cycle ( 18 ), we tested whether the observed changes in gene expression and protein abundance correlated with changes in cell cycle distributions. In the cell lines tested, we found neither changes in the subpopulations in G1, S or G2 (Figure S2 A) nor consistent changes in the mitotic cell fractions of phospho-histone H3(S10)-positive cells (Figure S2 B). Thus, we conclude that the observed deregulation of a key protein of the G2/M transition cannot be explained by changes in the overall cell cycle distribution, but rather by an effect on mitotic progression. RAI2 maintains mitotic accuracy To test whether RAI2 depletion influences mitotic progression, we analyzed individual phospho-histone H3(S10)-positive KPL-1 and MCF-7 cells and first assessed whether these cells exhibit macroscopic changes. In both cell lines, after RAI2 depletion, we found an increase in cells with micronuclei during prophase (Fig. 3 A) as well as an increased frequency of metaphases with unaligned chromosomes (Fig. 3 B), confirming that RAI2 depletion affects mitotic fidelity. Consistent with this, karyotyping of KPL-1 cells revealed a significantly reduced number of chromosomes in RAI2-depleted cells (Fig. 3 C). To enable the analysis of individual cell divisions in real time, we established KPL-1 cells with stable expression of the eYFP-Histone 2B fusion protein and monitored these cells by live cell imaging after RAI2 depletion (Fig. 3 D). By analyzing individual cell divisions, we found that the total mitotic time was longer in RAI2-depleted cells (Fig. 3 E). In addition, we observed increased de novo micronuclei formation in these cells (Fig. 3 F). To better characterize the observed chromosome segregation aberrations, we co-stained RAI2-depleted KPL-1 and MCF-7 cells with antibodies specific for phosphorylated histone H3 (S10) and centromeres. We found that the frequency of acentric fragments was increased during metaphase and anaphase in both cell lines (Fig. 3 G-H) but did not observe relevant differences in the frequency of lagging chromosomes or an increase in chromosome bridges during metaphase and anaphase (Supplementary Fig. 3B-D). In conclusion, we found that RAI2 depletion impairs mitotic fidelity, probably by causing unrepaired DNA double strand breaks (DSB) that lead to chromosome segregation errors during mitosis. RAI2 depletion is synergistically cytotoxic with topoisomerase I and Aurora A inhibitors To determine how RAI2 depletion could potentially contribute to premature mitotic entry, we performed a cytotoxicity screening with a library of 1280 pharmacologically active compounds. In the first screening phase, we tested and compared all components of the library to inhibit cell growth of control and RAI2-depleted KPL-1 cells. To select the best candidates for further validation, a threshold was set as 80% cell growth of non-target shRNA-expressing cells after 48 hours of incubation. Using these thresholds, we identified 27 compounds that inhibited cell growth in RAI2-depleted KPL-1 cells more efficiently than in control cells (Supplementary Table S6). In the second screening phase, we tested eight of these compounds, which were functionally related to DNA metabolism and/or cell cycle control, at different concentrations in KPL-1 and in MCF-7 cells. We found that Aurora-A inhibitor I, camptothecin (topoisomerase I inhibitor) and idarubicin (topoisomerase II inhibitor) (Fig. 4 A) were more effective in RAI2-depleted cells. Relevant differences were verified in three independent experiments for all components at the concentration of 10e-4.5 M (Fig. 4 B). To validate these findings, we performed confluence assays using the two independent shRNA sequences for RAI2 depletion and a longer incubation time of six days. We were able to validate that cell confluence of KPL-1 RAI2-depleted cells is reduced by treatment with Aurora inhibitor I at a concentration of 300 nM and camptothecin at a concentration of 30 nM. For MCF-7 cells, we found differences for Aurora inhibitor I at a concentration of 100 nM and for camptothecin at a concentration of 3 nM (Fig. 4 C). For idarubicin, no differences between control and RAI2-depleted cells were found in this assay (Fig. 4 C). In conclusion, we have shown that RAI2 depletion sensitizes KPL-1 and MCF-7 breast cancer cells to treatment with topoisomerase I and Aurora A inhibitors, indicating a synergistic effect with RAI2 inactivation. This suggests a potential role for RAI2 in the DNA damage response and/or checkpoint regulation. RAI2 protein is associated with the DNA damage response and poly-ADP ribosylation Based on these results, we analyzed RAI2 in cells treated with camptothecin. This genotoxic treatment resulted in an upregulation of RAI2 gene expression (Fig. 5 A) and protein levels, which correlated with an increase in DNA damage as indicated by increased γH2AX signal (Fig. 5 B). To further investigate the molecular function of the uncharacterized RAI2 protein, we sought to identify interacting proteins. To this end, we overexpressed HA-tagged RAI2 in HEK293T cells and performed co-immune precipitation. Protein labeling was performed by stable isotope labeling of amino acids in cell culture (SILAC) followed by quantitative mass spectrometric bottom-up proteomics for protein identification and quantification. In this model we identified nine possible interaction partners of the RAI2 protein with high confidence (CtBP1, CLTC, HSPA8, NONO, HSPA9, HSPA1A, TNRC6B, UPF1 and AP2B1). In addition, we found eleven probable interactions with low confidence (HSPA5, PARP1, SFPQ, NUDT21, KRT8, DDX24, FXR1, HSPB1, COL17A1, SLC25A11, and FARSA) (Supplementary Table S7, Supplementary Figure S4). The highest SILAC ratio was found for CtBP1, which we previously described as a RAI2-interacting protein ( 11 ) and which has been shown to maintain mitotic fidelity ( 19 , 20 ). Also, the low-confidence interacting protein PARP1 represents a key molecule in response to various forms of DNA damage ( 21 ). Interestingly, 75% of the other identified RAI2-interacting proteins, including PARP1 itself, are known substrates of PARP1 and are thus directly affected by post-translational poly-ADP-ribosylation (Supplementary Table S7) ( 22 – 28 ). Therefore, we hypothesized that RAI2 may have a general affinity poly-(ADP-ribose) itself. To prove these results, we performed a colocalization analysis of the RAI2 protein with either CtBP1 or poly-(ADP-ribose) in untreated and camptothecin-treated KPL-1 and MCF-7 cells. As previously reported, RAI2 protein localizes to distinct nuclear foci ( 11 ). In both untreated cell lines, the majority of these RAI2 foci colocalize with poly-(ADP-ribose) and the degree of colocalization is further increased in the presence of camptothecin (Fig. 5 C). For CtBP1, we found colocalization only for some of the larger RAI2 foci in KPL-1 cells, whereas any colocalization was barely seen in MCF-7 cells and was not affected by genotoxic treatment (Figure S5). Taken together, we confirmed a molecular function of the RAI2 protein within the DNA damage response associated with the presence of poly-(ADP-ribose). Since poly-ADP-ribosylation has multiple roles in different DNA repair pathways ( 21 ), we hypothesized that the previously observed phenotype of loss of mitotic fidelity might be the result of incomplete DNA repair prior to mitosis. RAI2 contributes to efficient homologous recombination Next, we investigated the effect of altered RAI2 expression on DNA DSB repair and the two major DSB DNA repair pathways, HR and non-homologous end joining (NHEJ). First, we used confocal laser scanning microscopy to determine the number of γH2AX foci, which are part of the early cellular response to the induction of DNA double-strand breaks ( 29 ). In parallel, we quantified 53BP1 foci, which promotes DSB repair via NHEJ ( 30 , 31 ). As expected, we found that RAI2 depletion caused an increase in γH2AX foci in KPL-1 and MCF-7, indicating an overall increase in DSB lesions caused by RAI2 depletion (Fig. 6 A). In contrast, we did not observe changes in 53BP1 foci in either cell line (Fig. 6 B). To assess whether the increase in DSBs in RAI2-depleted cells could be caused by defects in DNA synthesis, we analyzed DNA replication by DNA fiber assay in RAI2-depleted KPL-1 and MCF-7 cells. We did not observe differences in DNA replication fork speed in RAI2-depleted cells compared to parental cells, either untreated or after hydroxyurea treatment (Supplementary Figure S6). To investigate potential effects of RAI2 on HR and NHEJ, we used the plasmid-based traffic light reporter (TLR) assay ( 32 ) in HEK293T with transient RAI2 overexpression (Fig. 6 C). We showed that forced RAI2 protein expression resulted in higher HR, whereas no relevant changes in NHEJ were observed in this setting (Fig. 6 D). Taken together, these results demonstrate that RAI2 protein contributes to proficient HR. To validate a possible association of RAI2 gene expression with altered DNA repair capacity in clinical samples, we calculated the recombination proficiency score (RPS), which predicts HR efficiency and sensitivity to DNA-damaging chemotherapy in breast cancer ( 33 ), for all TCGA samples and correlated the RPS score with RAI2 gene expression. We found a positive correlation (Pearson correlation coefficient r = 0.52) between RAI2 expression and the RPS score and most of the individual genes that make up the RPS score (Fig. 6 D-E). Thus, we could confirm that primary breast tumors with putative HR deficiency have low RAI2 gene expression. Discussion HR is a major DNA repair mechanism that plays a critical role in the repair of DSBs. When HR is impaired or defective, several problems can arise that affect the accumulation of unrepaired DNA damage and premature mitotic entry ( 34 ). We discovered three distinct related phenotypes in RAI2-depleted breast cancer cells that are indicative of such premature entry into mitosis ( 13 , 35 ): (i) de novo micronuclei formation, (ii) prolonged mitosis, and (iii) mis-segregation of acentric chromosomes. Since cells with persistent DNA damage and defective HR are more susceptible to genomic instability as a hallmark of carcinogenesis ( 36 ), we conclude that defective DNA damage repair by HR is the underlying mechanism for the observed phenotypes in RAI2-depleted breast cancer cells and demonstrate that low RAI2 gene expression is a hallmark of genomically unstable breast tumors. In addition to affecting primary tumor development, genomic instability is known to contribute to the critical progression from MRD to overt metastasis ( 37 ). Because RAI2 was originally described as a potential metastasis suppressor that inhibits the hematogenous dissemination of estrogen receptor-positive breast cancer to the bone marrow ( 11 ), we conjecture that low RAI2 gene expression in these studies was indicative of genomic instability contributing to the early onset of metastatic progression. Interestingly, although our previous finding showed that RAI2 gene expression is strongly associated with the progression of ER-positive breast cancer ( 11 ), here we found that the functional association of loss of RAI2 gene expression with genomic instability is seen in all molecular breast cancer subtypes. Thus, combined RAI2 and CIN assessment seems to characterize a new class of high-risk patients who may deserve a different form of therapy. Despite the striking correlation of RAI2 gene expression with clinical outcomes, the molecular function of the corresponding RAI2 protein is largely unknown. RAI2 is predicted to be an intrinsically disordered protein, lacking a defined three-dimensional structure (38), and likely prone to tolerate flexible binding partners and promote biomolecular condensate formation in a regulated manner ( 39 , 40 ). In this study, we provide the first evidence that the RAI2 protein, in association with poly-(ADP-ribose), is part of the cellular response to genotoxic stress and contributes to efficient HR. Since poly-(ADP-ribose) has multiple roles in DNA repair and chromatin remodeling ( 21 , 41 ), including early recruitment of HR factors to DSB sites ( 42 ) and regulation of biomolecular condensate formation ( 43 , 44 ), investigating whether and how the RAI2 protein is involved in the regulation of either of these mechanisms is an exciting prospect for the future. Although we were able to validate that RAI2 maintains genomic stability at various levels of evidence, there are potential limitations to this study. Yet, little is known about the regulation of RAI2 gene expression, and an open question is what controls the induction of RAI2 expression after genotoxic exposure? Yan et al. provided initial evidence that hypermethylation of the RAI2 gene is common in colon cancer ( 12 ), and potential binding sites of the AP-2 transcription factor family have been identified in the proximal promoter region of RAI2 ( 45 ). Since in this study we have found by multivariable linear regression analysis that low RAI2 gene expression is not independent of TP53 gene mutation status, we propose that regardless of the active role of RAI2 in response to genotoxic stress, low RAI2 gene expression in tumor tissue may indicate a generalized DNA repair deficiency. However, the combination of correlative and causative effects may explain the strong value of low RAI2 gene expression as a prognostic marker and as a biomarker of genomic instability. Conclusions We provide evidence for an important role of RAI2 in maintaining genomic stability and DNA repair through homologous recombination. The underlying molecular mechanisms could be exploited to improve patient diagnosis and treatment. As the unbiased drug screening approach has revealed a synthetic sick interaction of RAI2 inactivation with topoisomerase I inhibitors, which are approved for the treatment of colorectal ( 46 ), ovarian ( 47 ) and small cell lung cancers ( 48 ), we envision that determining RAI2 gene expression could potentially be used to improve chemotherapeutic outcomes in these cancers or may itself represent a target structure for a new pharmacological approach. Taken together, this study provides the first evidence for RAI2 protein in maintaining genomic stability and enabling efficient DNA repair paving the way for further functional and translational investigations. Abbreviations 53BP1 Tumor suppressor p53-binding protein 1 ADP Adenosine diphosphate BFP Blue fluorescent protein BRCA Breast cancer CIN Chromosomal instability CtBP1 C-terminal-binding protein 1 DNA Deoxyribonucleic acid DSB DNA double strand breaks DTC Disseminated tumor cell ER Oestrogen receptor alpha eYFP Enhanced yellow fluorescent protein FACS Fluorescence-activated cell sorting FDR False discovery rate HER2 Human epidermal growth factor receptor 2 HR Homologous recombination MRD Minimal residual disease NHEJ Non-homologous end joining PAM50 Prediction analysis of microarray 50 PBS Phosphate-buffered saline PM Point mutations PVDF Polyvinylidene fluoride qRT-PCR Quantitative reverse transcription polymerase chain reaction RAI2 Retinoic acid-induced 2 RNA Ribonucleic acid RPS Recombination proficiency score SCNA Somatic copy number variation SDS Sodium dodecyl sulfate shRNA Small hairpin RNA SILAC Stable isotope labeling of amino acids in cell culture TCGA The cancer genome atlas TLR Traffic light reporter TP53 Tumor protein P53 wGII Weighted genome instability index γH2AX Phosphorylated form of H2A histone family member X Declarations Ethics approval and consent to participate. For clinical validation we used two breast cancer data sets provided by the cBIO portal for cancer genomics. TCGA Ethics and Policies was originally published by the National Cancer Institute. Consent for publication. Not applicable. Availability of data and materials. The microarray data are available via ArrayExpress accession with identifier E-MTAB-7071. The mass spectrometry proteomics data are available via ProteomeXchange with identifier PXD047189. Competing Interests. We do not declare any conflicts of interest or competing interests of any author. Funding. This work was supported by Wilhelm Sander-Stiftung [2015.148.2 to SW], by Erich und Gertrud Roggenbuck-Stiftung and by Deutsche Krebshilfe (70113304) to SW, by Deutsche Forschungsgemeinschaft [DFG 218826742 to HW, KP and MW], [SPP 2084: µBONE to KP and HW] and [Bo1868 to KB], by Bundesministerium fur Bildung und Forschung [02NUK032 & 02NUK035B to KB] and by European Research Council [ERC 336045 to JOK]. Author Contributions. Conceptualization: LB, KeB, JOK, HS, KP, HW and SW. Investigation: LB, SG, KeB, KaB, SSM, BE, KB, PS, BZ, AVF, KB, EC, CB, SP, SS, BB, SW. Formal analysis: LB, SG, KeB, KB, SSM, BE, KaB, PS, BZ, AVF, KB, EC, CB, VS, AKO, SP, SS, BB, SW. Funding acquisition: KeB, HW, KP, SW. Writing: LB, KeB, KP, HW, SW Acknowledgments. We are grateful for the skillful technical assistance of Bettina Steinbach, Alexandra Zielinski and Jolanthe Kropidlowski. We thank Ayham Moustafa, Sabrina Köcher, Kai Rothkamm and Matthias Wilmanns for scientific advice. We acknowledge the UKE FACS Sorting core unit for excellent technical support. References Pantel K, Brakenhoff RH. Dissecting the metastatic cascade. Nat Rev Cancer. 2004;4(6):448–56. Gerlinger M, McGranahan N, Dewhurst SM, Burrell RA, Tomlinson I, Swanton C. Cancer: evolution within a lifetime. Annu Rev Genet. 2014;48:215–36. Lengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature. 1998;396(6712):643–9. Gisselsson D. Classification of chromosome segregation errors in cancer. Chromosoma. 2008;117(6):511–9. Gordon DJ, Resio B, Pellman D. Causes and consequences of aneuploidy in cancer. Nat Rev Genet. 2012;13(3):189–203. McGranahan N, Burrell RA, Endesfelder D, Novelli MR, Swanton C. Cancer chromosomal instability: therapeutic and diagnostic challenges. EMBO Rep. 2012;13(6):528–38. Decout A, Katz JD, Venkatraman S, Ablasser A. The cGAS-STING pathway as a therapeutic target in inflammatory diseases. Nat Rev Immunol. 2021;21(9):548–69. Jamal-Hanjani M, Wilson GA, McGranahan N, Birkbak NJ, Watkins TBK, Veeriah S, et al. Tracking the Evolution of Non-Small-Cell Lung Cancer. N Engl J Med. 2017;376(22):2109–21. Turajlic S, Xu H, Litchfield K, Rowan A, Chambers T, Lopez JI, et al. Tracking Cancer Evolution Reveals Constrained Routes to Metastases: TRACERx Renal. Cell. 2018;173(3):581–94 e12. Vergara IA, Mintoff CP, Sandhu S, McIntosh L, Young RJ, Wong SQ, et al. Evolution of late-stage metastatic melanoma is dominated by aneuploidy and whole genome doubling. Nat Commun. 2021;12(1):1434. Werner S, Brors B, Eick J, Marques E, Pogenberg V, Parret A, et al. Suppression of early hematogenous dissemination of human breast cancer cells to bone marrow by retinoic Acid-induced 2. Cancer Discov. 2015;5(5):506–19. Yan W, Wu K, Herman JG, Xu X, Yang Y, Dai G, et al. Retinoic acid-induced 2 (RAI2) is a novel tumor suppressor, and promoter region methylation of RAI2 is a poor prognostic marker in colorectal cancer. Clin Epigenetics. 2018;10:69. Burrell RA, McClelland SE, Endesfelder D, Groth P, Weller MC, Shaikh N, et al. Replication stress links structural and numerical cancer chromosomal instability. Nature. 2013;494(7438):492–6. Birkbak NJ, Eklund AC, Li Q, McClelland SE, Endesfelder D, Tan P, et al. Paradoxical relationship between chromosomal instability and survival outcome in cancer. Cancer Res. 2011;71(10):3447–52. Carter SL, Eklund AC, Kohane IS, Harris LN, Szallasi Z. A signature of chromosomal instability inferred from gene expression profiles predicts clinical outcome in multiple human cancers. Nat Genet. 2006;38(9):1043–8. Pereira B, Chin SF, Rueda OM, Vollan HK, Provenzano E, Bardwell HA, et al. The somatic mutation profiles of 2,433 breast cancers refines their genomic and transcriptomic landscapes. Nat Commun. 2016;7:11479. Buccitelli C, Salgueiro L, Rowald K, Sotillo R, Mardin BR, Korbel JO. Pan-cancer analysis distinguishes transcriptional changes of aneuploidy from proliferation. Genome Res. 2017;27(4):501–11. Whitfield ML, Sherlock G, Saldanha AJ, Murray JI, Ball CA, Alexander KE, et al. Identification of genes periodically expressed in the human cell cycle and their expression in tumors. Mol Biol Cell. 2002;13(6):1977–2000. Bergman LM, Birts CN, Darley M, Gabrielli B, Blaydes JP. CtBPs promote cell survival through the maintenance of mitotic fidelity. Mol Cell Biol. 2009;29(16):4539–51. Birts CN, Bergman LM, Blaydes JP. CtBPs promote mitotic fidelity through their activities in the cell nucleus. Oncogene. 2011;30(11):1272–80. Ray Chaudhuri A, Nussenzweig A. The multifaceted roles of PARP1 in DNA repair and chromatin remodelling. Nat Rev Mol Cell Biol. 2017;18(10):610–21. Bartz R, Seemann J, Zehmer JK, Serrero G, Chapman KD, Anderson RG, et al. Evidence that mono-ADP-ribosylation of CtBP1/BARS regulates lipid storage. Mol Biol Cell. 2007;18(8):3015–25. Di Girolamo M, Silletta MG, De Matteis MA, Braca A, Colanzi A, Pawlak D, et al. Evidence that the 50-kDa substrate of brefeldin A-dependent ADP-ribosylation binds GTP and is modulated by the G-protein beta gamma subunit complex. Proc Natl Acad Sci U S A. 1995;92(15):7065–9. Gagne JP, Pic E, Isabelle M, Krietsch J, Ethier C, Paquet E, et al. Quantitative proteomics profiling of the poly(ADP-ribose)-related response to genotoxic stress. Nucleic acids research. 2012;40(16):7788–805. Jungmichel S, Rosenthal F, Altmeyer M, Lukas J, Hottiger MO, Nielsen ML. Proteome-wide identification of poly(ADP-Ribosyl)ation targets in different genotoxic stress responses. Molecular cell. 2013;52(2):272–85. Vivelo CA, Wat R, Agrawal C, Tee HY, Leung AKL. ADPriboDB: The database of ADP-ribosylated proteins. Nucleic acids research. 2017;45(10):6254. Zhang Y, Wang J, Ding M, Yu Y. Site-specific characterization of the Asp- and Glu-ADP-ribosylated proteome. Nature methods. 2013;10(10):981–4. Zhen Y, Zhang Y, Yu Y. A Cell-Line-Specific Atlas of PARP-Mediated Protein Asp/Glu-ADP-Ribosylation in Breast Cancer. Cell reports. 2017;21(8):2326–37. Zhao H, Huang X, Halicka HD, Darzynkiewicz Z. Detection of Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage. Curr Protoc Cytom. 2019;89(1):e55. Bouwman P, Aly A, Escandell JM, Pieterse M, Bartkova J, van der Gulden H, et al. 53BP1 loss rescues BRCA1 deficiency and is associated with triple-negative and BRCA-mutated breast cancers. Nat Struct Mol Biol. 2010;17(6):688–95. Bunting SF, Callen E, Wong N, Chen HT, Polato F, Gunn A, et al. 53BP1 inhibits homologous recombination in Brca1-deficient cells by blocking resection of DNA breaks. Cell. 2010;141(2):243–54. Certo MT, Ryu BY, Annis JE, Garibov M, Jarjour J, Rawlings DJ, et al. Tracking genome engineering outcome at individual DNA breakpoints. Nature methods. 2011;8(8):671–6. Pitroda SP, Bao R, Andrade J, Weichselbaum RR, Connell PP. Low Recombination Proficiency Score (RPS) Predicts Heightened Sensitivity to DNA-Damaging Chemotherapy in Breast Cancer. Clin Cancer Res. 2017;23(15):4493–500. Wright WD, Shah SS, Heyer WD. Homologous recombination and the repair of DNA double-strand breaks. J Biol Chem. 2018;293(27):10524–35. Hirsch B, Shimamura A, Moreau L, Baldinger S, Hag-alshiekh M, Bostrom B, et al. Association of biallelic BRCA2/FANCD1 mutations with spontaneous chromosomal instability and solid tumors of childhood. Blood. 2004;103(7):2554–9. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646–74. Bakhoum SF, Ngo B, Laughney AM, Cavallo JA, Murphy CJ, Ly P, et al. Chromosomal instability drives metastasis through a cytosolic DNA response. Nature. 2018;553(7689):467–72. Lang A, Goradia N, Wikman H, Werner S, Wilmanns M, Ohlenschlager O. (1)H, (13)C, and (15)N backbone assignments of the C-terminal region of the human retinoic acid-induced protein 2. Biomol NMR Assign. 2020;14(2):271-5. Boeynaems S, Alberti S, Fawzi NL, Mittag T, Polymenidou M, Rousseau F, et al. Protein Phase Separation: A New Phase in Cell Biology. Trends Cell Biol. 2018;28(6):420–35. Wright PE, Dyson HJ. Intrinsically disordered proteins in cellular signalling and regulation. Nat Rev Mol Cell Biol. 2015;16(1):18–29. Huang D, Kraus WL. The expanding universe of PARP1-mediated molecular and therapeutic mechanisms. Molecular cell. 2022;82(12):2315–34. Li M, Yu X. Function of BRCA1 in the DNA damage response is mediated by ADP-ribosylation. Cancer Cell. 2013;23(5):693–704. Altmeyer M, Neelsen KJ, Teloni F, Pozdnyakova I, Pellegrino S, Grofte M, et al. Liquid demixing of intrinsically disordered proteins is seeded by poly(ADP-ribose). Nat Commun. 2015;6:8088. Patel A, Lee HO, Jawerth L, Maharana S, Jahnel M, Hein MY, et al. A Liquid-to-Solid Phase Transition of the ALS Protein FUS Accelerated by Disease Mutation. Cell. 2015;162(5):1066–77. Walpole SM, Hiriyana KT, Nicolaou A, Bingham EL, Durham J, Vaudin M, et al. Identification and characterization of the human homologue (RAI2) of a mouse retinoic acid-induced gene in Xp22. Genomics. 1999;55(3):275–83. Folprecht G, Seymour MT, Saltz L, Douillard JY, Hecker H, Stephens RJ, et al. Irinotecan/fluorouracil combination in first-line therapy of older and younger patients with metastatic colorectal cancer: combined analysis of 2,691 patients in randomized controlled trials. J Clin Oncol. 2008;26(9):1443–51. Lihua P, Chen XY, Wu TX. Topotecan for ovarian cancer. Cochrane Database Syst Rev. 2008;2008(2):CD005589. Horita N, Yamamoto M, Sato T, Tsukahara T, Nagakura H, Tashiro K, et al. Topotecan for Relapsed Small-cell Lung Cancer: Systematic Review and Meta-Analysis of 1347 Patients. Sci Rep. 2015;5:15437. Additional Declarations No competing interests reported. Supplementary Files BCRRAI2maintainsgenomicstabilityuncroppedgelsandblots.pdf EHORAI2maintainsgenomicstabilitysupplementaryinformation.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3908810","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":270589939,"identity":"50a26ad4-747c-485d-8eec-8c1192cb22a6","order_by":0,"name":"Lena Boettcher","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Lena","middleName":"","lastName":"Boettcher","suffix":""},{"id":270589940,"identity":"37441eb8-68a4-4d63-a2ea-9fa8008ac7f4","order_by":1,"name":"Sarah Greimeier","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Greimeier","suffix":""},{"id":270589941,"identity":"d7b988cd-d0b3-456d-a7b3-86f96cfbeaeb","order_by":2,"name":"Kerstin Borgmann","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Kerstin","middleName":"","lastName":"Borgmann","suffix":""},{"id":270589942,"identity":"3cb0753e-b59c-451d-b72f-e1713782a8a3","order_by":3,"name":"Shabbir S. Mughal","email":"","orcid":"","institution":"German Cancer Research Center (DKFZ)","correspondingAuthor":false,"prefix":"","firstName":"Shabbir","middleName":"S.","lastName":"Mughal","suffix":""},{"id":270589943,"identity":"ba43b7d0-478b-42bd-a899-99c36a510ec1","order_by":4,"name":"Bernhard Ellinger","email":"","orcid":"","institution":"Fraunhofer Institute for Translational Medicine and Pharmacology ITMP","correspondingAuthor":false,"prefix":"","firstName":"Bernhard","middleName":"","lastName":"Ellinger","suffix":""},{"id":270589944,"identity":"c4f0c88c-a96f-4e2f-83e6-aa36702c185c","order_by":5,"name":"Kai Bartkowiak","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Bartkowiak","suffix":""},{"id":270589945,"identity":"5dedfe4c-7cae-4471-8d08-f87104f5f7ca","order_by":6,"name":"Bernd Zobiak","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Bernd","middleName":"","lastName":"Zobiak","suffix":""},{"id":270589946,"identity":"c251859e-ffd7-4eb3-9995-3e7a664afd2d","order_by":7,"name":"Antonio V. Failla","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Antonio","middleName":"V.","lastName":"Failla","suffix":""},{"id":270589947,"identity":"da1cdc9c-29d9-4853-914f-34354cb7888a","order_by":8,"name":"Pascal Steffen","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Pascal","middleName":"","lastName":"Steffen","suffix":""},{"id":270589948,"identity":"69cc8e0f-64f3-4caf-a13c-c288ae61972b","order_by":9,"name":"Ellen Claus","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Ellen","middleName":"","lastName":"Claus","suffix":""},{"id":270589949,"identity":"9379b3dc-8549-48f3-8647-f1a984037628","order_by":10,"name":"Katharina Besler","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Katharina","middleName":"","lastName":"Besler","suffix":""},{"id":270589950,"identity":"3a266686-7ef5-409e-a8a1-d35a6c0e90d7","order_by":11,"name":"Christopher Buccitelli","email":"","orcid":"","institution":"Max Delbrück Centre for Molecular Medicine","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Buccitelli","suffix":""},{"id":270589951,"identity":"541adc4f-d5b6-46e3-baeb-4f6011ab39f4","order_by":12,"name":"Violetta Schaaf","email":"","orcid":"","institution":"Ruprecht-Karls-Universität Heidelberg","correspondingAuthor":false,"prefix":"","firstName":"Violetta","middleName":"","lastName":"Schaaf","suffix":""},{"id":270589952,"identity":"2916e973-8be3-4906-a92f-0abef1341b49","order_by":13,"name":"Ann-Kathrin Ozga","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Ann-Kathrin","middleName":"","lastName":"Ozga","suffix":""},{"id":270589953,"identity":"c8b27911-59b8-422a-80c7-95211bd274ab","order_by":14,"name":"Simona Parretta","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Simona","middleName":"","lastName":"Parretta","suffix":""},{"id":270589954,"identity":"3a98f5a1-5ef5-46a8-a3a6-880c45259d84","order_by":15,"name":"Svenja Schneegans","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Svenja","middleName":"","lastName":"Schneegans","suffix":""},{"id":270589955,"identity":"3b8251b9-1993-4538-aa31-3c4902a2f9c7","order_by":16,"name":"Wael Y. Mansour","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Wael","middleName":"Y.","lastName":"Mansour","suffix":""},{"id":270589956,"identity":"628e5585-6e9b-4ef6-b31d-d56c9899ad4c","order_by":17,"name":"Jan O. Korbel","email":"","orcid":"","institution":"European Molecular Biology Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"O.","lastName":"Korbel","suffix":""},{"id":270589957,"identity":"f4fd6d97-692e-4d13-ac10-10bbed23dadc","order_by":18,"name":"Hartmut Schlueter","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Hartmut","middleName":"","lastName":"Schlueter","suffix":""},{"id":270589958,"identity":"4b71fcfe-3be4-4a2a-ab40-84760383a478","order_by":19,"name":"Benedikt Brors","email":"","orcid":"","institution":"German Cancer Research Center (DKFZ)","correspondingAuthor":false,"prefix":"","firstName":"Benedikt","middleName":"","lastName":"Brors","suffix":""},{"id":270589959,"identity":"8be25ff1-67b9-4a68-85b4-c3c8b6cc3caa","order_by":20,"name":"Klaus Pantel","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Klaus","middleName":"","lastName":"Pantel","suffix":""},{"id":270589960,"identity":"c02d9602-2ec4-43e7-9260-c5228da59ca3","order_by":21,"name":"Harriet Wikman","email":"","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":false,"prefix":"","firstName":"Harriet","middleName":"","lastName":"Wikman","suffix":""},{"id":270589961,"identity":"ffb09f2c-40e6-45a9-8132-ed9bd55b3f49","order_by":22,"name":"Stefan Werner","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxElEQVRIiWNgGAWjYFAD9gZmIMlMhEo2GIPnAMlaJBKI1CI/v4FN4uOeO3LmM98YG3xgsJYjqMXgGAOb5Ixnz4xlbucYJ85gSDcmrIWNge02z4HDiTOkc4wP8zAcTmwg6LA2oJY/Bw7Xz5A8Y3z4D8PheoJaGIAOu81w4HCChASPcTIDw+EEIvyS2P6z58Bhwxk8acWGPQbphoQd1nz4sMGPA4flJdgPb5b4UWEtT9AWBgZGZGMNiNAwCkbBKBgFo4AwAAAW1jfdcZJBHgAAAABJRU5ErkJggg==","orcid":"","institution":"University Medical Centre Hamburg-Eppendorf","correspondingAuthor":true,"prefix":"","firstName":"Stefan","middleName":"","lastName":"Werner","suffix":""}],"badges":[],"createdAt":"2024-01-29 11:44:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3908810/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3908810/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":50674661,"identity":"754a2d69-8a5d-4f89-91c6-75a8a735f31a","added_by":"auto","created_at":"2024-02-05 15:11:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":332559,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLow RAI2 gene expression is a feature of chromosomally unstable breast tumors and predicts poor patient outcome. A-C)\u003c/strong\u003e Weighted genomic instability score (wGII). wGII score was calculated for the TCGA-BRCA cohort and patients were divided into two groups based on median wGII score. For each group, normalized gene expression of \u003cem\u003eRAI2\u003c/em\u003e was plotted against wGII score in \u003cstrong\u003eA)\u003c/strong\u003e all tumors, \u003cstrong\u003eB)\u003c/strong\u003ein subgroups of ER and \u003cstrong\u003eC)\u003c/strong\u003e in PAM50 molecular subtypes. P-values were calculated by t-test. D-F) TCGA BRCA patients were divided into low and high CIN70 groups based on median CIN70 signature values. Normalized \u003cem\u003eRAI2\u003c/em\u003egene expression was plotted for each group. CIN70 stratification showing \u003cstrong\u003eD)\u003c/strong\u003eall tumors and \u003cstrong\u003eE)\u003c/strong\u003e in subgroups of ER and \u003cstrong\u003eF)\u003c/strong\u003e in PAM50 molecular subtypes. P-values were calculated by t-test. \u003cstrong\u003eG)\u003c/strong\u003e The Cancer Genome Atlas (TCGA) data including exonic point mutations and somatic copy number alterations (SCNA) were examined to analyze the correlation between RAI2 gene expression and the pattern of genome instability. Copy number burden (total number of SCNAs per tumor type) and PM burden were summed. P-values were corrected using the Benjamini-Hochberg method to obtain FDRs. \u003cstrong\u003eH)\u003c/strong\u003e Overall survival (OS) analysis of RAI2 gene expression stratified by median and CIN score in patients in the METABRIC dataset analyzed by Kaplan-Meier estimation. A two-tailed significance level below 0.05 was considered significant.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/e8120970bc70ad83fb56ff8f.png"},{"id":50674641,"identity":"e764cade-4bc0-47f6-87f0-4489a5f33ac7","added_by":"auto","created_at":"2024-02-05 15:11:23","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":276417,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRAI2 depletion in human breast cancer cells causes deregulation of cell cycle-related genes and proteins. A)\u003c/strong\u003e Pathway analysis results of gene expression profiling showing the five most significantly enriched biological themes (GO terms) of molecular function, cellular component and biological process found in KPL-1 cells expressing RAI2-specific shRNA (dark gray). Plots show the negative decadic logarithm of Benjamini scores (black) and p-values (dark gray). The threshold indicates a negative decadic logarithm cut-off of 0.05. \u003cstrong\u003eB)\u003c/strong\u003eGene expression analysis of cell cycle related genes in RAI2-depleted cell lines using Human Cell Cycle RT² Profiler™ PCR Arrays. Fold change (FC) expression is shown relative to B2M expression and non-target shRNA-transduced control cells. Values shown are the mean of three biological replicates. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test. \u003cstrong\u003eC)\u003c/strong\u003e Western blot analysis and quantification of protein expression signals of RAI2 and previously identified cell cycle markers in whole cell extracts of the indicated cell lines after transduction with two independent RAI2-specific or non-target shRNAs. Detection of HSC70 protein is used as a loading control. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/9618cf7f96933c52bce20a50.png"},{"id":50674532,"identity":"a0107ee7-afa8-4bfb-86e3-3ac68d126987","added_by":"auto","created_at":"2024-02-05 15:11:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":309665,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRAI2 depletion induces mis-segregation of acentric chromosomes and micronuclei formation in luminal breast cancer cells. A)\u003c/strong\u003e Analysis of the frequency of micronuclei during prophase and \u003cstrong\u003eB)\u003c/strong\u003e frequency of incomplete metaphases in RAI2-depleted KPL-1 cells. The images show examples of scored events of immunofluorescence-stained cells with a phospho-histone H3 phosphorylation (S10) antibody. Frequencies are the mean of three independent experiments. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test \u003cstrong\u003eC)\u003c/strong\u003e Metaphase spreading of exponentially growing KPL-1 cells 10 days after transduction with non-target shRNA and RAI2-specific shRNA1. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test. \u003cstrong\u003eD)\u003c/strong\u003e Examples of recorded cell divisions of control and RAI2-depleted KPL-1 cells with stable expression of the eYFP-Histone 2B fusion protein. Time in mitosis is indicated in hours:minutes:seconds and unaligned chromosome/chromosome fragments and micronuclei are indicated by arrows. \u003cstrong\u003eE)\u003c/strong\u003eTotal duration of individual cell divisions assessed from at least 100 cells of each cell line. ** indicate a p-value below the significance threshold of 0.01 calculated by a two-sided t-test. \u003cstrong\u003eF)\u003c/strong\u003e \u003cem\u003eDe novo\u003c/em\u003e micronuclei formation as assessed by live cell imaging of KPL-1 non-target and RAI2-depleted cells. \u003cstrong\u003eG)\u003c/strong\u003e acentric chromosome frequency during metaphase; and \u003cstrong\u003eH)\u003c/strong\u003e anaphase. The images show examples of scored events of immunofluorescence-stained cells for DNA (blue), phospho-histone H3 phosphorylation (S10) (orange) and human centrosomes (green). Values shown are the mean frequency of three independent experiments. Student's t-test was used for significant testing; * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/7f42056df088deade8bff9e4.png"},{"id":50674628,"identity":"27e89765-b5de-440e-8cc3-f5041846b923","added_by":"auto","created_at":"2024-02-05 15:11:18","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":251539,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRAI2-depleted cells show increased sensitivity for topoisomerase I and Aurora-A inhibitors. A)\u003c/strong\u003e Analysis of cell viability of RAI2-depleted KPL-1 and MCF-7 cells in the presence of indicated compounds using the CellTiter-Glo assay. One representative replicate comprising three technical replicates for each of the tested components is shown. \u003cstrong\u003eB)\u003c/strong\u003e Mean values of all replicates at the concentration of 4.5x10e-4.5. \u003cstrong\u003eC)\u003c/strong\u003e Analysis of cell proliferation of RAI2-depleted KPL-1 and MCF-7 cells in the presence of indicated compounds. Data represent the mean of three independent experiments. For significant testing Student’s t-test was applied; * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test; n.s.: non-significant difference.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/73461a290fad61596e28c583.png"},{"id":50674600,"identity":"4460837b-1364-4b76-8cd6-895a8149b3af","added_by":"auto","created_at":"2024-02-05 15:11:15","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":352202,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe RAI2 protein associates with DNA damage response and pADPr. A)\u003c/strong\u003e Quantitative RT-PCR analysis of DMSO- or camptothecin (CPT) treated KPL-1 and MCF-7 cells. RAI2 gene expression is expressed as the average fold change normalized to RPLP0 and DMSO-treated cells. Values are the average of three replicates. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test, ns; not significant. \u003cstrong\u003eB)\u003c/strong\u003eWestern blot analysis and signal quantification of RAI2 and γH2AX in whole cell extracts of DMSO or CPT (100 nM) treated KPL-1 and MCF-7 cells. HSC70 was detected as a loading control. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test. \u003cstrong\u003eC)\u003c/strong\u003e Co-localization of RAI2 (red) with pADPr (green) and DNA (blue) in DMSO or CPT (100 nM) treated KPL-1 and MCF-7 cells. Quantification of fluorescence intensity profiles of overlay images by line scanning of 100 foci images of 2 µm distance each. Scale bars: 10 µm.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/5bcae68e1b4f2486afe8ec8d.png"},{"id":50674612,"identity":"6caf0444-3697-4ecf-a722-a3107d477fa4","added_by":"auto","created_at":"2024-02-05 15:11:17","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":210252,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRAI2 contributes to efficient DNA repair by homologous recombination.\u003c/strong\u003e \u003cstrong\u003eA)\u003c/strong\u003eAssessment of γH2AX foci in RAI2 depleted KPL-1 and MCF-7 cells. Cells were stained with anti γH2AX antibody and analyzed by confocal laser-scanning micospropy. The number of γH2AX foci in at least 100 indivudual cells were determined by image analysis software. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test, ns; not significant. \u003cstrong\u003eB)\u003c/strong\u003e Assessment of 53BP1 foci in RAI2 depleted KPL-1 and MCF-7 cells. Cells were stained with anti 53BP1 antibody and analyzed by confocal laser-scanning micospropy. The number of 53BP1 foci in at least 100 indivudual cells were determined by image analysis software.* indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test, ns; not significant. \u003cstrong\u003eC)\u003c/strong\u003e Traffic Light Reporter Assay analysing relative efficncy of non-homologous end-joining and homologous recombination. Validation of RAI2-overexpressing HEK293T cells by Western blot analysis (left pantel). Assessment of relative HR and NHEJ repair efficiency determined by flow cytometry. HR/NHEJ capacity was normalised to empty vector control. Values the mean of three biological replicates. * indicate a p-value below the significance threshold of 0.05 calculated by a two-sided t-test, ns; not significant. \u003cstrong\u003eD)\u003c/strong\u003e Recombinant Proficiency Score- a gene expression based score was used to quantify the DNA repair efficiency and correlation of RAI2 gene expression in TCGA-BRCA tumor samples. \u003cstrong\u003eE)\u003c/strong\u003e Correlation of RAI2 gene expression with individual genes of HR deficiency signature in breast cancer. The heatmap displays the correlation values (r) computed using Pearson’s correlation. The heatmap legend on the left displays the r values, with a deeper red colour indicating a negative and blue indicating a positive correlation.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/8b81247ba998c5fb07e92788.png"},{"id":63368120,"identity":"6995f1a9-32f7-4462-88bc-db3ea09c86ee","added_by":"auto","created_at":"2024-08-27 11:35:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2566933,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/48f8af3a-63ec-45c3-895d-59f433722306.pdf"},{"id":50674526,"identity":"9cca3e61-3d3e-4da1-8bae-376836a3fd28","added_by":"auto","created_at":"2024-02-05 15:11:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":591592,"visible":true,"origin":"","legend":"","description":"","filename":"BCRRAI2maintainsgenomicstabilityuncroppedgelsandblots.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/bcef11602503e0a4b814f530.pdf"},{"id":50674633,"identity":"cb18427a-ffa3-4c95-a745-9fb60e4af9b5","added_by":"auto","created_at":"2024-02-05 15:11:19","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":884174,"visible":true,"origin":"","legend":"","description":"","filename":"EHORAI2maintainsgenomicstabilitysupplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-3908810/v1/062f38e26ceb36d408378c76.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Retinoic Acid-Induced 2 Contributes to Proficient Homologous Recombination and Maintains Genomic Stability in Breast Cancer","fulltext":[{"header":"Background","content":"\u003cp\u003eMetastasis is a multistep process involving the hematogenous dissemination of cells from the primary tumor to distant sites such as the bone marrow, where the cells can either remain dormant as disseminated tumor cells (DTCs) or begin to proliferate and form overt metastases (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Due to its complex phenotype, the acquisition of metastatic competence represents a major selective barrier during cancer progression. It has been proposed that such a macroevolutionary leap leading to metastatic competence of tumor cells can only be achieved by large-scale genomic alterations leading to chromosomal instability (CIN) (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). CIN is the predominant mechanism leading to genomic instability (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). CIN can arise through either mitotic or premitotic defects (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) and is associated with aggressiveness, drug resistance, and poor prognosis (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Recent evidence also suggests that genomic instability may shape the antitumor immune response (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Fundamental studies have confirmed that intra-tumor heterogeneity mediated by genome doubling and ongoing dynamic CIN drives disease recurrence in both renal cell and lung cancer. Interestingly, this effect was not associated with mutational burden (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). More recently, the evolution of late-stage metastatic melanoma has been shown to be dominated by aneuploidy and whole genome doubling (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePreviously, RAI2 was identified as a novel metastasis suppressor gene associated with both breast and colorectal cancer metastasis, but its molecular function is still largely unknown (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Low \u003cem\u003eRAI2\u003c/em\u003e gene expression in primary breast cancer is significantly correlated with the mutational status of the \u003cem\u003eTP53\u003c/em\u003e gene (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), which led us to test whether low RAI2 gene expression might also be associated with genomic instability. Here we provide the first evidence that the RAI2 protein maintains genomic stability by enabling efficient repair of DNA double strand breaks (DSB) by homologous recombination (HR), which contributes to the suppression of metastatic relapse in human breast cancer. In addition, RAI2 depletion appears to have synergistic effects with anticancer agents that induce DNA damage, which may further influence the response to chemotherapeutic treatment or progression from minimal residual disease (MRD) to overt metastasis.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eA comprehensive description of all materials and methods is provided as Supplementary Information. The description of live cell imaging, cytotoxicity profiling and immunoprecipitation combined with protein analysis by quantitative mass spectrometry proteomics, the rigor adherence table (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) and the key resources table (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e) are provided as Supplementary Information only.\u003c/p\u003e \u003cp\u003e \u003cb\u003eClinical\u003c/b\u003e \u003cb\u003ein silico\u003c/b\u003e \u003cb\u003evalidation\u003c/b\u003e\u003c/p\u003e \u003cp\u003eWe used the Bioconductor package (RRID:SCR_006442) to download TCGA-BRCA transcriptome profiles, copy number segments, and clinical data. Normalized gene expression data for RAI2 and other genes of interest in the METABRIC dataset were obtained from cBioPortal. Normalized RAI2 gene expression was correlated with the expression of genes indicative of CIN (CIN70 score) using the weighted genome instability index (wGII). P-values were calculated by Student's t-test. Multivariable regression analysis was performed to evaluate the association between CIN70 score and RAI2 gene expression, adjusting for other covariates such as oestrogen receptor α (ER) status and PAM50 subtype and TP53 status. For linear regression, RAI2 levels were log-transformed. For multivariable analysis, the Cox regression model was used, including those histopathological factors that were clinically significant in the univariable survival analysis in the multivariable analysis (tumour stage, ER status, HER2 status, grade, and molecular subtype) and presented as hazard ratio with 95% confidence interval. To assess whether RAI2 gene expression correlated with patterns of genomic instability, copy number alteration (burden total number of somatic copy number variations (SCNAs) per tumour type) and point mutations (PM) burden were summed. Spearman correlations were performed to compare mRNA abundance and somatic variant burden on a per-gene basis. P-values were corrected using the Benjamini-Hochberg method to obtain false discovery rates (FDRs). For survival analysis, samples from the METABRIC dataset were divided by the median of RAI2 gene expression and CIN70 score. Differences in five-year overall survival between these groups were determined by Kaplan-Meier analysis using the log-rank test. The Recombination Proficiency Score (RPS) was calculated for each tumour sample using normalized expression values for four signature genes involved in the DNA repair pathway (\u003cem\u003eRIF1\u003c/em\u003e, \u003cem\u003ePARI, RAD51\u003c/em\u003e, and \u003cem\u003eKU80\u003c/em\u003e).\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCell culture\u003c/h2\u003e \u003cp\u003eMCF-7 (RRID:CVCL_0031), KPL-1 (RRID:CVCL_2094), MCF-10A (RRID:CVCL_0598), CAMA-1 (RRID:CVCL_1115) and 293T (ATTC #CRL-3216) cells were cultured under standard conditions. Cell line authentication was performed using short tandem repeat profiling to exclude cross-contamination between cell lines. Cell lines were tested monthly for mycoplasma contamination.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003ePlasmids, viral transduction and transfection\u003c/h2\u003e \u003cp\u003ePlasmid construction for overexpression of wild-type RAI2 protein, as well as viral production and transduction procedures, have been described previously (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). For knockdown experiments, we used plasmids derived from pLKO.1: non-target shRNA (Sigma-Aldrich #SHC016), shRNA1 (TRCN0000139927), and shRNA2 (TRCN0000441623). We used the pH2B-eYFP plasmid (plasmid #51002) to establish KPL-1 cells with constitutive H2B-eYFP overexpression.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eGene expression profiling\u003c/h2\u003e \u003cp\u003e500 ng total RNA from KPL-1 breast cancer cells was hybridized to the Illumina HT-12 Array v4 BeadChip (Illumina, San Diego, CA, USA) according to the manufacturer's protocols. Microarrays were scanned using the Illumina iScan scanner (Illumina, San Diego, CA, USA) according to the standard Illumina scanning protocol. Bead-level data were aggregated using BeadStudio and normalized using the quantile method. Differential expression on normalized expression data was determined using the samr package (R version 3.4.2) with an unpaired t-test with a delta of 0.06 and a minimum twofold change in gene expression. Functional annotation of differentially expressed genes was performed using the gene functional classification tool in DAVID Bioinformatics Resources 6.8 (RRID:SCR_001881).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eQuantitative real-time RT-PCR Analysis (qRT-PCR)\u003c/h2\u003e \u003cp\u003eRNA was extracted from cultured cells during the exponential growth phase using the Nucleospin RNA Kit (Macherey Nagel, Germany) according to the standard protocol. 1000 ng of RNA from each sample was transcribed using the First Strand cDNA Synthesis Kit (Thermo Scientific, MA, USA) and random hexamers. Human Cell Cycle RT\u0026sup2; Profiler\u0026trade; PCR Arrays and reagents (Eppendorf, Germany) were used for cell cycle focused gene expression analysis. The qRT-PCR reactions were performed in triplicate using the Mastercycler Eppendorf Realplex thermal cycler (Eppendorf, Germany). Data analysis and significance testing were performed using QIAGEN GeneGlobe Data Analysis Centre (RRID:SCR_021211).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eWestern blotting\u003c/h2\u003e \u003cp\u003eWhole cell extracts from cultured cells were prepared by direct lysis and sonication of cells in 2% SDS sample buffer containing phosphatase and protease inhibitors. Cell extracts were separated on 8\u0026ndash;15% denaturing polyacrylamide gels and blotted onto nitrocellulose or PVDF membranes. The following antibodies were used for detection: RAI2 (RRID:AB_2800292), Aurora A (RRID:AB_2665504), Aurora B (RRID:AB_10695307), Cyclin A2 (RRID:AB_627334), Cyclin B1 (RRID: AB_2783553), cyclin B2 (RRID:AB_2072392), survivin (RRID:AB_2063948), γH2AX (RRID:AB_2118009), and HSC-70 (RRID:AB_627761). HRP-conjugated anti-rabbit IgG, (RRID:AB_2099233) and (RRID:AB_330924) or infrared dye-labelled anti-rabbit IgG, (RRID:AB_621843) and anti-mouse IgG (RRID:AB_10956588) were used for detection. Differences between signal intensities in different cell lines and results of three independent experiments were evaluated by two-tailed Student's t-test.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCell cycle analysis\u003c/h2\u003e \u003cp\u003eCell cycle profiles were assessed by quantifying DNA content using flow cytometry. Cells were fixed in 4% formaldehyde, treated with RNase and stained with propidium iodide. Flow cytometric analysis was performed using a NovoCyte Quanteon (Agilent, CA, USA) flow cytometer. A minimum of 20,000 cells were collected for analysis. After doublet discrimination, cell cycle profiles were automatically calculated using NovoExpress software (Agilent, CA, USA). The Watson model was used for cell cycle fitting. To determine the mitotic fraction, flow cytometry analysis of P-H3(S10) (RRID:AB_1549592) stained cells was performed using a FACS CantoII (Becton Dickinson, NJ, USA) equipped with FACSDiva software (RRID:SCR_001456). Three independent biological replicates were used to calculate the percentage of each cell cycle phase and the standard deviation. The difference was tested by two-tailed t-test.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eImmunofluorescence staining\u003c/h2\u003e \u003cp\u003eCells were fixed in 4% paraformaldehyde, washed three times with PBS, and permeabilized with 0.2% Triton X-100 in PBS. After incubation with 1% non-fat dry milk (w/v) in PBS for 30 minutes, cells were further incubated with primary antibodies: P-H3(S10) (RRID:AB_1549592), anti-centrosome (RRID:AB_212756), RAI2 (RRID:AB_2800292), CtBP1 (RRID:AB_399429), poly(ADP-ribose) (RRID:AB_785249), γH2AX (RRID:AB_2118009), or 53BP1 (RRID:AB_2921289). Specific antibody binding was visualized with Alexa Fluor 488 goat anti-rabbit IgG (RRID:AB_143165) and Alexa Fluor 546 goat anti-mouse IgG (RRID:AB_2534093). Confocal laser scanning microscopy was performed using a Leica TCS SP5 microscope (Leica, Germany) and Imaris imaging software (RRID:SCR_007370) to identify and measure the number of γH2AX and 53BP1 foci. At least 100 independent events were evaluated for quantification, and differences were tested by two-tailed t-test.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eMetaphase spread analysis\u003c/h2\u003e \u003cp\u003eFor metaphase spreads, exponentially growing KPL-1 cells were treated overnight with Colcemid (0.02 \u0026micro;g/ml), incubated with 0.0075 M KCl, fixed with methanol/acetic acid (3:1), dropped onto slides, stained with 5% Giemsa and mounted with Entellan before imaging with a Zeiss Axioplan 2 microscope (Zeiss, Germany). 100 metaphases per experiment were counted in three independent experiments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eTraffic light reporter assay\u003c/h2\u003e \u003cp\u003eThe BFP-TLR-SceI plasmid (Addgene plasmid #31481) was digested with SceI (New England Biolabs, MA, USA) for 4 hours, run on an agarose gel, and purified with a purification kit (Macherey Nagel, Germany) before being used for transfection. For RAI2 overexpression, HEK293T cells were first transfected with phCMV3-RAI2-HA (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e) using OPTIMEM (Thermo Fisher, MA, USA) and Lipofectamin2000 (Invitrogen, MA, USA) according to standard protocol. After 24 hours, the cells were cotransfected with 500 ng cut BFP-TLR-SceI and GFP donor plasmid (Addgene MA, USA, plasmid #31475). At 48 hours post-transfection, trypsinised cells were quenched with media and mCherry, eGFP and BFP fluorescence signal was analysed by flow cytometry (LSR Fortessa, Becton Dickinson, NJ, USA) using 561 nm, 488 nm and 405 nm laser. The percentage of mCherry (for NHEJ events) and eGFP (for HR events) positive cells in the BFP positive cell fraction was used for analysis. Three independent biological replicates were used to calculate relative DNA repair efficiency. Differences were tested by two-tailed t-test.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eLow RAI2 Expression is a Feature of Genetic Unstable Breast Carcinomas\u003c/h2\u003e \u003cp\u003eTo assess whether RAI2 expression is associated with genomic instability in breast cancer patients, we performed an analysis of large published clinical datasets. First, we calculated the weighted genome integrity index (wGII) as an independent measure of CIN (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e) and divided samples into low and high wGII groups based on the median cut-off. We found that \u003cem\u003eRAI2\u003c/em\u003e gene expression was clinically lower in tumors with a high wGII score (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). We confirmed clinically differences in \u003cem\u003eRAI2\u003c/em\u003e gene expression in estrogen receptor (ER)-positive and -negative subgroups (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB) as well as in basal, luminal A and B molecular subtypes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). We then examined \u003cem\u003eRAI2\u003c/em\u003e gene expression in patient groups stratified according to the CIN70 signature of chromosomal instability derived from gene expression profiles (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Consistent with the results from wGII, we observed clinically lower \u003cem\u003eRAI2\u003c/em\u003e gene expression in samples with a high CIN70 score (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD), ER-positive tumors showed higher \u003cem\u003eRAI2\u003c/em\u003e gene expression compared to ER-negative tumors (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE). However, the overall trend remained consistent in all groups of basal, luminal A and B molecular subtypes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor validation a correlation of \u003cem\u003eRAI2\u003c/em\u003e gene expression with the complete CIN70 score was assessed in the METABRIC breast cancer dataset (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Univariable analyses showed a clinically significant association between low \u003cem\u003eRAI2\u003c/em\u003e gene expression and high CIN70 score (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA-C). Also, in multivariable linear regression analysis a significant association between the CIN70 score and low \u003cem\u003eRAI2\u003c/em\u003e gene expression was found in the METABRIC data set (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, multiplicative factor\u0026thinsp;=\u0026thinsp;0.999 [CI:0.998\u0026ndash;0.999]). Likewise, the basal subtype (p\u0026thinsp;=\u0026thinsp;0.002, multiplicative factor\u0026thinsp;=\u0026thinsp;0.949 [CI:0.918\u0026ndash;0.982]) and HER2 (p\u0026thinsp;=\u0026thinsp;0.038, multiplicative factor\u0026thinsp;=\u0026thinsp;0.938 [CI:0.910\u0026ndash;0.967]) showed an influence on \u003cem\u003eRAI2\u003c/em\u003e gene expression compared to luminal A, whereas for ER and \u003cem\u003eTP53\u003c/em\u003e no influence was seen (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003eIn the cancer genome atlas (TCGA) dataset of the invasive breast carcinoma we investigated a possible correlation of \u003cem\u003eRAI2\u003c/em\u003e gene expression with somatic copy number alterations (SCNA) and point mutations (PM) by a previously published approach (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). \u003cem\u003eRAI2\u003c/em\u003e was among the top genes (rank\u0026thinsp;=\u0026thinsp;36 out of 12961) whose expression correlated inversely with aneuploidy (spearman rho = -0.47, FDR\u0026thinsp;=\u0026thinsp;4.96 e-40) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG). Taken together these results confirm that compared to other genes in the genome there is a strong correlation between low \u003cem\u003eRAI2\u003c/em\u003e gene expression and different characteristics of genome instability in primary breast tumors.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eLow RAI2 expression predicts poor clinical outcome in breast cancer patients\u003c/h2\u003e \u003cp\u003eWe then evaluated whether low \u003cem\u003eRAI2\u003c/em\u003e gene expression identifies an aggressive subset of tumors associated with poor clinical outcome in the METABRIC dataset. Five-year overall survival analysis of patients showed that low \u003cem\u003eRAI2\u003c/em\u003e gene expression combined with a high CIN score had a worse prognosis compared to other patient groups (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). At five years, 36.2% of early relapse patients with low \u003cem\u003eRAI2\u003c/em\u003e gene expression and high CIN score had died, while the best survival rate with only 9.4% of deaths was seen in patients with high \u003cem\u003eRAI2\u003c/em\u003e gene expression and low CIN score. Patients with a mixed phenotype had intermediate 5-year survival rates of 79.4% and 82.1% (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH). Multivariable analysis showed that the combined RAI2\u003csup\u003elow\u003c/sup\u003e/CIN\u003csup\u003ehigh\u003c/sup\u003e phenotype was an independent poor prognostic factor (hazard ratio: 1.60 [CI: 1.05\u0026ndash;2.42, p\u0026thinsp;=\u0026thinsp;0.027]). Other independent risk parameters were grade (hazard ratio: 1.46 [CI: 1.08\u0026ndash;1.99, p\u0026thinsp;=\u0026thinsp;0.017]), ER status (hazard ratio: 1.88 [CI: 1.28\u0026ndash;2.75, p\u0026thinsp;=\u0026thinsp;0.001]) and stage (hazard ratio: 2.2 [1.63\u0026ndash;2.98, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001]). Taken together, these results indicate that low \u003cem\u003eRAI2\u003c/em\u003e gene expression is a hallmark of genetically unstable tumors and that these patients have a significantly higher risk of early metastatic relapse.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eRAI2 depletion in human breast cancer cells causes deregulation of cell cycle-related genes and proteins\u003c/h2\u003e \u003cp\u003eTo analyze whether loss of RAI2 protein function causes genomic instability, we first performed gene expression profiling by microarray analysis of the luminal breast cancer cell line KPL-1 after shRNA mediated RAI2 depletion, which has previously been shown to be a suitable cell line model to study RAI2 protein function (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Functional annotation revealed that the significantly deregulated genes are associated with the cell cycle, including deregulation of genes involved in microtubule motor activity, mitosis, and spindle apparatus (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and Supplementary Table S3 and S4). Therefore, the dysregulation of cell cycle-related genes was analyzed in more detail in additional RAI2-depleted cell lines (KPL-1, CAMA-1, MCF-7 and MCF-10A) using pathway-specific arrays (Cell Cycle RT\u003csup\u003e2\u003c/sup\u003e Profiler PCR Arrays). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB and Supplementary Table S5, RAI2 depletion leads to decreased gene expression of \u003cem\u003eAURKB, CCNA2, CCNB1, CCNB2, CDC20, CDC6, CDK1, CDK5R1, GTSE1, MAD2L2, MCM2, MCM3, MCM4\u003c/em\u003e and \u003cem\u003eMKI67\u003c/em\u003e in all tested cell lines. The downregulation of selected gene products was validated by Western blot using two independent RAI2-specific shRNA sequences (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). We found reduced protein expression of Aurora A and B, Cyclin A2 in all three RAI2-depleted cell lines, and Cyclin B2 was significantly reduced only in KPL-1 and CAMA-1 cells. Since all these genes and proteins are known to be periodically regulated within the cell cycle (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e), we tested whether the observed changes in gene expression and protein abundance correlated with changes in cell cycle distributions. In the cell lines tested, we found neither changes in the subpopulations in G1, S or G2 (Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eA) nor consistent changes in the mitotic cell fractions of phospho-histone H3(S10)-positive cells (Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eB). Thus, we conclude that the observed deregulation of a key protein of the G2/M transition cannot be explained by changes in the overall cell cycle distribution, but rather by an effect on mitotic progression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eRAI2 maintains mitotic accuracy\u003c/h2\u003e \u003cp\u003eTo test whether RAI2 depletion influences mitotic progression, we analyzed individual phospho-histone H3(S10)-positive KPL-1 and MCF-7 cells and first assessed whether these cells exhibit macroscopic changes. In both cell lines, after RAI2 depletion, we found an increase in cells with micronuclei during prophase (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA) as well as an increased frequency of metaphases with unaligned chromosomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), confirming that RAI2 depletion affects mitotic fidelity. Consistent with this, karyotyping of KPL-1 cells revealed a significantly reduced number of chromosomes in RAI2-depleted cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). To enable the analysis of individual cell divisions in real time, we established KPL-1 cells with stable expression of the eYFP-Histone 2B fusion protein and monitored these cells by live cell imaging after RAI2 depletion (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). By analyzing individual cell divisions, we found that the total mitotic time was longer in RAI2-depleted cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). In addition, we observed increased \u003cem\u003ede novo\u003c/em\u003e micronuclei formation in these cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). To better characterize the observed chromosome segregation aberrations, we co-stained RAI2-depleted KPL-1 and MCF-7 cells with antibodies specific for phosphorylated histone H3 (S10) and centromeres. We found that the frequency of acentric fragments was increased during metaphase and anaphase in both cell lines (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG-H) but did not observe relevant differences in the frequency of lagging chromosomes or an increase in chromosome bridges during metaphase and anaphase (Supplementary Fig.\u0026nbsp;3B-D). In conclusion, we found that RAI2 depletion impairs mitotic fidelity, probably by causing unrepaired DNA double strand breaks (DSB) that lead to chromosome segregation errors during mitosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eRAI2 depletion is synergistically cytotoxic with topoisomerase I and Aurora A inhibitors\u003c/h2\u003e \u003cp\u003eTo determine how RAI2 depletion could potentially contribute to premature mitotic entry, we performed a cytotoxicity screening with a library of 1280 pharmacologically active compounds. In the first screening phase, we tested and compared all components of the library to inhibit cell growth of control and RAI2-depleted KPL-1 cells. To select the best candidates for further validation, a threshold was set as \u0026lt;\u0026thinsp;80% cell growth of RAI2-depleted cells and \u0026gt;\u0026thinsp;80% cell growth of non-target shRNA-expressing cells after 48 hours of incubation. Using these thresholds, we identified 27 compounds that inhibited cell growth in RAI2-depleted KPL-1 cells more efficiently than in control cells (Supplementary Table S6). In the second screening phase, we tested eight of these compounds, which were functionally related to DNA metabolism and/or cell cycle control, at different concentrations in KPL-1 and in MCF-7 cells. We found that Aurora-A inhibitor I, camptothecin (topoisomerase I inhibitor) and idarubicin (topoisomerase II inhibitor) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA) were more effective in RAI2-depleted cells. Relevant differences were verified in three independent experiments for all components at the concentration of 10e-4.5 M (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo validate these findings, we performed confluence assays using the two independent shRNA sequences for RAI2 depletion and a longer incubation time of six days. We were able to validate that cell confluence of KPL-1 RAI2-depleted cells is reduced by treatment with Aurora inhibitor I at a concentration of 300 nM and camptothecin at a concentration of 30 nM. For MCF-7 cells, we found differences for Aurora inhibitor I at a concentration of 100 nM and for camptothecin at a concentration of 3 nM (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). For idarubicin, no differences between control and RAI2-depleted cells were found in this assay (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). In conclusion, we have shown that RAI2 depletion sensitizes KPL-1 and MCF-7 breast cancer cells to treatment with topoisomerase I and Aurora A inhibitors, indicating a synergistic effect with RAI2 inactivation. This suggests a potential role for RAI2 in the DNA damage response and/or checkpoint regulation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eRAI2 protein is associated with the DNA damage response and poly-ADP ribosylation\u003c/h2\u003e \u003cp\u003eBased on these results, we analyzed RAI2 in cells treated with camptothecin. This genotoxic treatment resulted in an upregulation of \u003cem\u003eRAI2\u003c/em\u003e gene expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA) and protein levels, which correlated with an increase in DNA damage as indicated by increased γH2AX signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). To further investigate the molecular function of the uncharacterized RAI2 protein, we sought to identify interacting proteins. To this end, we overexpressed HA-tagged RAI2 in HEK293T cells and performed co-immune precipitation. Protein labeling was performed by stable isotope labeling of amino acids in cell culture (SILAC) followed by quantitative mass spectrometric bottom-up proteomics for protein identification and quantification. In this model we identified nine possible interaction partners of the RAI2 protein with high confidence (CtBP1, CLTC, HSPA8, NONO, HSPA9, HSPA1A, TNRC6B, UPF1 and AP2B1). In addition, we found eleven probable interactions with low confidence (HSPA5, PARP1, SFPQ, NUDT21, KRT8, DDX24, FXR1, HSPB1, COL17A1, SLC25A11, and FARSA) (Supplementary Table S7, Supplementary Figure S4). The highest SILAC ratio was found for CtBP1, which we previously described as a RAI2-interacting protein (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e) and which has been shown to maintain mitotic fidelity (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Also, the low-confidence interacting protein PARP1 represents a key molecule in response to various forms of DNA damage (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Interestingly, 75% of the other identified RAI2-interacting proteins, including PARP1 itself, are known substrates of PARP1 and are thus directly affected by post-translational poly-ADP-ribosylation (Supplementary Table S7) (\u003cspan additionalcitationids=\"CR23 CR24 CR25 CR26 CR27\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). Therefore, we hypothesized that RAI2 may have a general affinity poly-(ADP-ribose) itself. To prove these results, we performed a colocalization analysis of the RAI2 protein with either CtBP1 or poly-(ADP-ribose) in untreated and camptothecin-treated KPL-1 and MCF-7 cells. As previously reported, RAI2 protein localizes to distinct nuclear foci (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). In both untreated cell lines, the majority of these RAI2 foci colocalize with poly-(ADP-ribose) and the degree of colocalization is further increased in the presence of camptothecin (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). For CtBP1, we found colocalization only for some of the larger RAI2 foci in KPL-1 cells, whereas any colocalization was barely seen in MCF-7 cells and was not affected by genotoxic treatment (Figure S5). Taken together, we confirmed a molecular function of the RAI2 protein within the DNA damage response associated with the presence of poly-(ADP-ribose). Since poly-ADP-ribosylation has multiple roles in different DNA repair pathways (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e), we hypothesized that the previously observed phenotype of loss of mitotic fidelity might be the result of incomplete DNA repair prior to mitosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eRAI2 contributes to efficient homologous recombination\u003c/h2\u003e \u003cp\u003eNext, we investigated the effect of altered RAI2 expression on DNA DSB repair and the two major DSB DNA repair pathways, HR and non-homologous end joining (NHEJ). First, we used confocal laser scanning microscopy to determine the number of γH2AX foci, which are part of the early cellular response to the induction of DNA double-strand breaks (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). In parallel, we quantified 53BP1 foci, which promotes DSB repair via NHEJ (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). As expected, we found that RAI2 depletion caused an increase in γH2AX foci in KPL-1 and MCF-7, indicating an overall increase in DSB lesions caused by RAI2 depletion (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). In contrast, we did not observe changes in 53BP1 foci in either cell line (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). To assess whether the increase in DSBs in RAI2-depleted cells could be caused by defects in DNA synthesis, we analyzed DNA replication by DNA fiber assay in RAI2-depleted KPL-1 and MCF-7 cells. We did not observe differences in DNA replication fork speed in RAI2-depleted cells compared to parental cells, either untreated or after hydroxyurea treatment (Supplementary Figure S6). To investigate potential effects of RAI2 on HR and NHEJ, we used the plasmid-based traffic light reporter (TLR) assay (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e) in HEK293T with transient RAI2 overexpression (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). We showed that forced RAI2 protein expression resulted in higher HR, whereas no relevant changes in NHEJ were observed in this setting (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). Taken together, these results demonstrate that RAI2 protein contributes to proficient HR. To validate a possible association of \u003cem\u003eRAI2\u003c/em\u003e gene expression with altered DNA repair capacity in clinical samples, we calculated the recombination proficiency score (RPS), which predicts HR efficiency and sensitivity to DNA-damaging chemotherapy in breast cancer (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e), for all TCGA samples and correlated the RPS score with \u003cem\u003eRAI2\u003c/em\u003e gene expression. We found a positive correlation (Pearson correlation coefficient r\u0026thinsp;=\u0026thinsp;0.52) between RAI2 expression and the RPS score and most of the individual genes that make up the RPS score (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD-E). Thus, we could confirm that primary breast tumors with putative HR deficiency have low \u003cem\u003eRAI2\u003c/em\u003e gene expression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eHR is a major DNA repair mechanism that plays a critical role in the repair of DSBs. When HR is impaired or defective, several problems can arise that affect the accumulation of unrepaired DNA damage and premature mitotic entry (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). We discovered three distinct related phenotypes in RAI2-depleted breast cancer cells that are indicative of such premature entry into mitosis (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e): (i) de novo micronuclei formation, (ii) prolonged mitosis, and (iii) mis-segregation of acentric chromosomes. Since cells with persistent DNA damage and defective HR are more susceptible to genomic instability as a hallmark of carcinogenesis (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e), we conclude that defective DNA damage repair by HR is the underlying mechanism for the observed phenotypes in RAI2-depleted breast cancer cells and demonstrate that low \u003cem\u003eRAI2\u003c/em\u003e gene expression is a hallmark of genomically unstable breast tumors.\u003c/p\u003e \u003cp\u003eIn addition to affecting primary tumor development, genomic instability is known to contribute to the critical progression from MRD to overt metastasis (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). Because RAI2 was originally described as a potential metastasis suppressor that inhibits the hematogenous dissemination of estrogen receptor-positive breast cancer to the bone marrow (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), we conjecture that low \u003cem\u003eRAI2\u003c/em\u003e gene expression in these studies was indicative of genomic instability contributing to the early onset of metastatic progression. Interestingly, although our previous finding showed that \u003cem\u003eRAI2\u003c/em\u003e gene expression is strongly associated with the progression of ER-positive breast cancer (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), here we found that the functional association of loss of \u003cem\u003eRAI2\u003c/em\u003e gene expression with genomic instability is seen in all molecular breast cancer subtypes. Thus, combined RAI2 and CIN assessment seems to characterize a new class of high-risk patients who may deserve a different form of therapy.\u003c/p\u003e \u003cp\u003eDespite the striking correlation of \u003cem\u003eRAI2\u003c/em\u003e gene expression with clinical outcomes, the molecular function of the corresponding RAI2 protein is largely unknown. RAI2 is predicted to be an intrinsically disordered protein, lacking a defined three-dimensional structure (38), and likely prone to tolerate flexible binding partners and promote biomolecular condensate formation in a regulated manner (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). In this study, we provide the first evidence that the RAI2 protein, in association with poly-(ADP-ribose), is part of the cellular response to genotoxic stress and contributes to efficient HR. Since poly-(ADP-ribose) has multiple roles in DNA repair and chromatin remodeling (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e), including early recruitment of HR factors to DSB sites (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e) and regulation of biomolecular condensate formation (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e), investigating whether and how the RAI2 protein is involved in the regulation of either of these mechanisms is an exciting prospect for the future.\u003c/p\u003e \u003cp\u003eAlthough we were able to validate that RAI2 maintains genomic stability at various levels of evidence, there are potential limitations to this study. Yet, little is known about the regulation of RAI2 gene expression, and an open question is what controls the induction of RAI2 expression after genotoxic exposure? Yan et al. provided initial evidence that hypermethylation of the \u003cem\u003eRAI2\u003c/em\u003e gene is common in colon cancer (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e), and potential binding sites of the AP-2 transcription factor family have been identified in the proximal promoter region of \u003cem\u003eRAI2\u003c/em\u003e (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e). Since in this study we have found by multivariable linear regression analysis that low \u003cem\u003eRAI2\u003c/em\u003e gene expression is not independent of \u003cem\u003eTP53\u003c/em\u003e gene mutation status, we propose that regardless of the active role of RAI2 in response to genotoxic stress, low \u003cem\u003eRAI2\u003c/em\u003e gene expression in tumor tissue may indicate a generalized DNA repair deficiency. However, the combination of correlative and causative effects may explain the strong value of low \u003cem\u003eRAI2\u003c/em\u003e gene expression as a prognostic marker and as a biomarker of genomic instability.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eWe provide evidence for an important role of RAI2 in maintaining genomic stability and DNA repair through homologous recombination. The underlying molecular mechanisms could be exploited to improve patient diagnosis and treatment. As the unbiased drug screening approach has revealed a synthetic sick interaction of RAI2 inactivation with topoisomerase I inhibitors, which are approved for the treatment of colorectal (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e), ovarian (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e) and small cell lung cancers (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e), we envision that determining \u003cem\u003eRAI2\u003c/em\u003e gene expression could potentially be used to improve chemotherapeutic outcomes in these cancers or may itself represent a target structure for a new pharmacological approach. Taken together, this study provides the first evidence for RAI2 protein in maintaining genomic stability and enabling efficient DNA repair paving the way for further functional and translational investigations.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e53BP1\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Tumor suppressor p53-binding protein 1\u003c/p\u003e\n\u003cp\u003eADP\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Adenosine diphosphate\u003c/p\u003e\n\u003cp\u003eBFP\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Blue fluorescent protein\u003c/p\u003e\n\u003cp\u003eBRCA\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Breast cancer\u003c/p\u003e\n\u003cp\u003eCIN\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Chromosomal instability\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCtBP1 \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;C-terminal-binding protein 1\u003c/p\u003e\n\u003cp\u003eDNA \u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Deoxyribonucleic acid\u003c/p\u003e\n\u003cp\u003eDSB\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;DNA double strand breaks\u003c/p\u003e\n\u003cp\u003eDTC\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Disseminated tumor cell\u003c/p\u003e\n\u003cp\u003eER\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Oestrogen receptor alpha\u003c/p\u003e\n\u003cp\u003eeYFP\u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Enhanced yellow fluorescent protein\u003c/p\u003e\n\u003cp\u003eFACS\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Fluorescence-activated cell sorting\u003c/p\u003e\n\u003cp\u003eFDR\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;False discovery rate\u003c/p\u003e\n\u003cp\u003eHER2\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Human epidermal growth factor receptor 2\u003c/p\u003e\n\u003cp\u003eHR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Homologous recombination\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMRD\u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Minimal residual disease\u003c/p\u003e\n\u003cp\u003eNHEJ\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Non-homologous end joining\u003c/p\u003e\n\u003cp\u003ePAM50\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Prediction analysis of microarray 50\u003c/p\u003e\n\u003cp\u003ePBS\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Phosphate-buffered saline\u003c/p\u003e\n\u003cp\u003ePM\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Point mutations\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePVDF\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Polyvinylidene fluoride\u003c/p\u003e\n\u003cp\u003eqRT-PCR \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Quantitative reverse transcription polymerase chain reaction\u003c/p\u003e\n\u003cp\u003eRAI2\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Retinoic acid-induced 2\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRNA\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Ribonucleic acid\u003c/p\u003e\n\u003cp\u003eRPS\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Recombination proficiency score\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSCNA\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Somatic copy number variation\u003c/p\u003e\n\u003cp\u003eSDS\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Sodium dodecyl sulfate\u003c/p\u003e\n\u003cp\u003eshRNA\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Small hairpin RNA\u003c/p\u003e\n\u003cp\u003eSILAC\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Stable isotope labeling of amino acids in cell culture\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTCGA\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;The cancer genome atlas\u003c/p\u003e\n\u003cp\u003eTLR\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Traffic light reporter\u003c/p\u003e\n\u003cp\u003eTP53\u0026nbsp; \u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Tumor protein P53\u003c/p\u003e\n\u003cp\u003ewGII\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Weighted genome instability index\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026gamma;H2AX \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Phosphorylated form of H2A histone family member X\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate.\u0026nbsp;\u003c/strong\u003eFor clinical validation we used two breast cancer data sets provided by the cBIO portal for cancer genomics. TCGA Ethics and Policies was originally published by the National Cancer Institute.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication.\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials.\u003c/strong\u003e The microarray data are available via ArrayExpress accession with identifier E-MTAB-7071. The mass spectrometry proteomics data are available via ProteomeXchange with identifier PXD047189.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests.\u0026nbsp;\u003c/strong\u003eWe do not declare any conflicts of interest or competing interests of any author.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding.\u0026nbsp;\u003c/strong\u003eThis work was supported by Wilhelm Sander-Stiftung [2015.148.2 to SW], by Erich und Gertrud Roggenbuck-Stiftung and by Deutsche Krebshilfe (70113304) to SW, by Deutsche Forschungsgemeinschaft [DFG 218826742 to HW, KP and MW], [SPP 2084: \u0026micro;BONE to KP and HW] and [Bo1868 to KB], by Bundesministerium fur Bildung und Forschung [02NUK032 \u0026amp; 02NUK035B to KB] and by European Research Council [ERC 336045 to JOK].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions.\u0026nbsp;\u003c/strong\u003eConceptualization: LB, KeB, JOK, HS, KP, HW and SW. Investigation: LB, SG, KeB, KaB, SSM, BE, KB, PS, BZ, AVF, KB, EC, CB, SP, SS, BB, SW. Formal analysis: LB, SG, KeB, KB, SSM, BE, KaB, PS, BZ, AVF, KB, EC, CB, VS, AKO, SP, SS, BB, SW. Funding acquisition: KeB, HW, KP, SW. Writing: LB, KeB, KP, HW, SW\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments.\u0026nbsp;\u003c/strong\u003eWe are grateful for the skillful technical assistance of Bettina Steinbach, Alexandra Zielinski and Jolanthe Kropidlowski. We thank Ayham Moustafa, Sabrina K\u0026ouml;cher, Kai Rothkamm and Matthias Wilmanns for scientific advice. We acknowledge the UKE FACS Sorting core unit for excellent technical support.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ePantel K, Brakenhoff RH. Dissecting the metastatic cascade. Nat Rev Cancer. 2004;4(6):448\u0026ndash;56.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGerlinger M, McGranahan N, Dewhurst SM, Burrell RA, Tomlinson I, Swanton C. Cancer: evolution within a lifetime. Annu Rev Genet. 2014;48:215\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature. 1998;396(6712):643\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGisselsson D. Classification of chromosome segregation errors in cancer. Chromosoma. 2008;117(6):511\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGordon DJ, Resio B, Pellman D. Causes and consequences of aneuploidy in cancer. Nat Rev Genet. 2012;13(3):189\u0026ndash;203.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGranahan N, Burrell RA, Endesfelder D, Novelli MR, Swanton C. Cancer chromosomal instability: therapeutic and diagnostic challenges. EMBO Rep. 2012;13(6):528\u0026ndash;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDecout A, Katz JD, Venkatraman S, Ablasser A. The cGAS-STING pathway as a therapeutic target in inflammatory diseases. Nat Rev Immunol. 2021;21(9):548\u0026ndash;69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJamal-Hanjani M, Wilson GA, McGranahan N, Birkbak NJ, Watkins TBK, Veeriah S, et al. Tracking the Evolution of Non-Small-Cell Lung Cancer. N Engl J Med. 2017;376(22):2109\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurajlic S, Xu H, Litchfield K, Rowan A, Chambers T, Lopez JI, et al. Tracking Cancer Evolution Reveals Constrained Routes to Metastases: TRACERx Renal. Cell. 2018;173(3):581\u0026ndash;94 e12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVergara IA, Mintoff CP, Sandhu S, McIntosh L, Young RJ, Wong SQ, et al. Evolution of late-stage metastatic melanoma is dominated by aneuploidy and whole genome doubling. Nat Commun. 2021;12(1):1434.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWerner S, Brors B, Eick J, Marques E, Pogenberg V, Parret A, et al. Suppression of early hematogenous dissemination of human breast cancer cells to bone marrow by retinoic Acid-induced 2. Cancer Discov. 2015;5(5):506\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan W, Wu K, Herman JG, Xu X, Yang Y, Dai G, et al. Retinoic acid-induced 2 (RAI2) is a novel tumor suppressor, and promoter region methylation of RAI2 is a poor prognostic marker in colorectal cancer. Clin Epigenetics. 2018;10:69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurrell RA, McClelland SE, Endesfelder D, Groth P, Weller MC, Shaikh N, et al. Replication stress links structural and numerical cancer chromosomal instability. Nature. 2013;494(7438):492\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBirkbak NJ, Eklund AC, Li Q, McClelland SE, Endesfelder D, Tan P, et al. Paradoxical relationship between chromosomal instability and survival outcome in cancer. Cancer Res. 2011;71(10):3447\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarter SL, Eklund AC, Kohane IS, Harris LN, Szallasi Z. A signature of chromosomal instability inferred from gene expression profiles predicts clinical outcome in multiple human cancers. Nat Genet. 2006;38(9):1043\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePereira B, Chin SF, Rueda OM, Vollan HK, Provenzano E, Bardwell HA, et al. The somatic mutation profiles of 2,433 breast cancers refines their genomic and transcriptomic landscapes. Nat Commun. 2016;7:11479.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuccitelli C, Salgueiro L, Rowald K, Sotillo R, Mardin BR, Korbel JO. Pan-cancer analysis distinguishes transcriptional changes of aneuploidy from proliferation. Genome Res. 2017;27(4):501\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhitfield ML, Sherlock G, Saldanha AJ, Murray JI, Ball CA, Alexander KE, et al. Identification of genes periodically expressed in the human cell cycle and their expression in tumors. Mol Biol Cell. 2002;13(6):1977\u0026ndash;2000.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBergman LM, Birts CN, Darley M, Gabrielli B, Blaydes JP. CtBPs promote cell survival through the maintenance of mitotic fidelity. Mol Cell Biol. 2009;29(16):4539\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBirts CN, Bergman LM, Blaydes JP. CtBPs promote mitotic fidelity through their activities in the cell nucleus. Oncogene. 2011;30(11):1272\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRay Chaudhuri A, Nussenzweig A. The multifaceted roles of PARP1 in DNA repair and chromatin remodelling. Nat Rev Mol Cell Biol. 2017;18(10):610\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBartz R, Seemann J, Zehmer JK, Serrero G, Chapman KD, Anderson RG, et al. Evidence that mono-ADP-ribosylation of CtBP1/BARS regulates lipid storage. Mol Biol Cell. 2007;18(8):3015\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Girolamo M, Silletta MG, De Matteis MA, Braca A, Colanzi A, Pawlak D, et al. Evidence that the 50-kDa substrate of brefeldin A-dependent ADP-ribosylation binds GTP and is modulated by the G-protein beta gamma subunit complex. Proc Natl Acad Sci U S A. 1995;92(15):7065\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGagne JP, Pic E, Isabelle M, Krietsch J, Ethier C, Paquet E, et al. Quantitative proteomics profiling of the poly(ADP-ribose)-related response to genotoxic stress. Nucleic acids research. 2012;40(16):7788\u0026ndash;805.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJungmichel S, Rosenthal F, Altmeyer M, Lukas J, Hottiger MO, Nielsen ML. Proteome-wide identification of poly(ADP-Ribosyl)ation targets in different genotoxic stress responses. Molecular cell. 2013;52(2):272\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVivelo CA, Wat R, Agrawal C, Tee HY, Leung AKL. ADPriboDB: The database of ADP-ribosylated proteins. Nucleic acids research. 2017;45(10):6254.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Wang J, Ding M, Yu Y. Site-specific characterization of the Asp- and Glu-ADP-ribosylated proteome. Nature methods. 2013;10(10):981\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhen Y, Zhang Y, Yu Y. A Cell-Line-Specific Atlas of PARP-Mediated Protein Asp/Glu-ADP-Ribosylation in Breast Cancer. Cell reports. 2017;21(8):2326\u0026ndash;37.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao H, Huang X, Halicka HD, Darzynkiewicz Z. Detection of Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage. Curr Protoc Cytom. 2019;89(1):e55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBouwman P, Aly A, Escandell JM, Pieterse M, Bartkova J, van der Gulden H, et al. 53BP1 loss rescues BRCA1 deficiency and is associated with triple-negative and BRCA-mutated breast cancers. Nat Struct Mol Biol. 2010;17(6):688\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBunting SF, Callen E, Wong N, Chen HT, Polato F, Gunn A, et al. 53BP1 inhibits homologous recombination in Brca1-deficient cells by blocking resection of DNA breaks. Cell. 2010;141(2):243\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCerto MT, Ryu BY, Annis JE, Garibov M, Jarjour J, Rawlings DJ, et al. Tracking genome engineering outcome at individual DNA breakpoints. Nature methods. 2011;8(8):671\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePitroda SP, Bao R, Andrade J, Weichselbaum RR, Connell PP. Low Recombination Proficiency Score (RPS) Predicts Heightened Sensitivity to DNA-Damaging Chemotherapy in Breast Cancer. Clin Cancer Res. 2017;23(15):4493\u0026ndash;500.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWright WD, Shah SS, Heyer WD. Homologous recombination and the repair of DNA double-strand breaks. J Biol Chem. 2018;293(27):10524\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHirsch B, Shimamura A, Moreau L, Baldinger S, Hag-alshiekh M, Bostrom B, et al. Association of biallelic BRCA2/FANCD1 mutations with spontaneous chromosomal instability and solid tumors of childhood. Blood. 2004;103(7):2554\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBakhoum SF, Ngo B, Laughney AM, Cavallo JA, Murphy CJ, Ly P, et al. Chromosomal instability drives metastasis through a cytosolic DNA response. Nature. 2018;553(7689):467\u0026ndash;72.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLang A, Goradia N, Wikman H, Werner S, Wilmanns M, Ohlenschlager O. (1)H, (13)C, and (15)N backbone assignments of the C-terminal region of the human retinoic acid-induced protein 2. Biomol NMR Assign. 2020;14(2):271-5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoeynaems S, Alberti S, Fawzi NL, Mittag T, Polymenidou M, Rousseau F, et al. Protein Phase Separation: A New Phase in Cell Biology. Trends Cell Biol. 2018;28(6):420\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWright PE, Dyson HJ. Intrinsically disordered proteins in cellular signalling and regulation. Nat Rev Mol Cell Biol. 2015;16(1):18\u0026ndash;29.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang D, Kraus WL. The expanding universe of PARP1-mediated molecular and therapeutic mechanisms. Molecular cell. 2022;82(12):2315\u0026ndash;34.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi M, Yu X. Function of BRCA1 in the DNA damage response is mediated by ADP-ribosylation. Cancer Cell. 2013;23(5):693\u0026ndash;704.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAltmeyer M, Neelsen KJ, Teloni F, Pozdnyakova I, Pellegrino S, Grofte M, et al. Liquid demixing of intrinsically disordered proteins is seeded by poly(ADP-ribose). Nat Commun. 2015;6:8088.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel A, Lee HO, Jawerth L, Maharana S, Jahnel M, Hein MY, et al. A Liquid-to-Solid Phase Transition of the ALS Protein FUS Accelerated by Disease Mutation. Cell. 2015;162(5):1066\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalpole SM, Hiriyana KT, Nicolaou A, Bingham EL, Durham J, Vaudin M, et al. Identification and characterization of the human homologue (RAI2) of a mouse retinoic acid-induced gene in Xp22. Genomics. 1999;55(3):275\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFolprecht G, Seymour MT, Saltz L, Douillard JY, Hecker H, Stephens RJ, et al. Irinotecan/fluorouracil combination in first-line therapy of older and younger patients with metastatic colorectal cancer: combined analysis of 2,691 patients in randomized controlled trials. J Clin Oncol. 2008;26(9):1443\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLihua P, Chen XY, Wu TX. Topotecan for ovarian cancer. Cochrane Database Syst Rev. 2008;2008(2):CD005589.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHorita N, Yamamoto M, Sato T, Tsukahara T, Nagakura H, Tashiro K, et al. Topotecan for Relapsed Small-cell Lung Cancer: Systematic Review and Meta-Analysis of 1347 Patients. Sci Rep. 2015;5:15437.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"breast cancer, genomic instability, premature mitotic entry, homologous recombination, poly-(ADP-ribose)","lastPublishedDoi":"10.21203/rs.3.rs-3908810/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3908810/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eGenome instability is a fundamental feature and hallmark of cancer associated with aggressiveness, drug resistance and poor prognosis. RAI2 was initially identified as a novel metastasis suppressor protein specifically associated with the presence of disseminated tumour cells in the bone marrow of breast cancer patients, but its molecular function is largely unknown.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe analysed the consequences of RAI2 depletion on gene expression and genomic stability in luminal breast cancer cell lines, performed cytotoxicity profiling using a library of pharmacologically active compounds, and characterized the function of the RAI2 protein in the DNA damage response. We performed \u003cem\u003ein silico\u003c/em\u003e validation in different breast cancer datasets.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAnalysis of clinical samples revealed that in primary breast tumours, low \u003cem\u003eRAI2\u003c/em\u003e gene expression is significantly associated with genomically unstable tumours and poor prognosis. RAI2 depletion in breast cancer cell lines resulted in loss of mitotic fidelity characterized by prolonged mitosis with increased chromosome segregation errors and micronuclei formation. Drug screening revealed increased sensitivity of RAI2-depleted breast cancer cells to topoisomerase I and Aurora A inhibitors. We also found that genotoxic stress induces RAI2 protein, which shows affinity for poly-(ADP-ribose) and contributes to efficient DNA repair by homologous recombination. We validated the functional association of \u003cem\u003eRAI2\u003c/em\u003e gene expression with DNA double-strand break repair capacity in clinical samples.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eOur findings support, for the first time, an important functional role of RAI2 in the maintenance of mitotic fidelity and DNA repair associated with early metastatic relapse. The underlying molecular mechanisms could be exploited to improve patient diagnosis and treatment.\u003c/p\u003e","manuscriptTitle":"Retinoic Acid-Induced 2 Contributes to Proficient Homologous Recombination and Maintains Genomic Stability in Breast Cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-05 15:08:37","doi":"10.21203/rs.3.rs-3908810/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8a1d671b-7ba3-4bf5-a293-3ba1acd5cfc8","owner":[],"postedDate":"February 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-08-27T11:35:44+00:00","versionOfRecord":[],"versionCreatedAt":"2024-02-05 15:08:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3908810","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3908810","identity":"rs-3908810","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}