Non‐enzymatic function of WRN RECQL helicase regulates removal of topoisomerase‐I‐DNA covalent complexes and triggers NF‐κB signaling in cancer

Abstract Mutation in Werner (WRN) RECQL helicase is associated with premature aging syndrome (Werner syndrome, WS) and predisposition to multiple cancers. In patients with solid cancers, deficiency of the WRN RECQL helicase is paradoxically associated with enhanced overall survival in response to treatment with TOP1 inhibitors, which stabilize pathological TOP1‐DNA‐covalent‐complexes (TOP1cc) on the genome. However, the underlying mechanism of WRN in development of chemoresistance to TOP1 inhibitors is not yet explored. Our whole‐genome transcriptomic analysis for ~25,000 genes showed robust activation of NF‐κB‐dependent prosurvival genes in response to TOP1cc. CRISPR‐Cas9 knockout, shRNA silencing, and under‐expression of WRN confer high‐sensitivity of multiple cancers to TOP1 inhibitor. We demonstrated that WRN orchestrates TOP1cc repair through proteasome‐dependent and proteasome‐independent process, unleashing robust ssDNA generation. This in turn ensues signal transduction for CHK1 mediated NF‐κB‐activation through IκBα‐degradation and nuclear localization of p65 protein. Intriguingly, our site‐directed mutagenesis and rescue experiments revealed that neither RECQL‐helicase nor DNA‐exonuclease enzyme activity of WRN (WRNE84A, WRNK577M, and WRNE84A‐K577M) were required for TOP1cc removal, ssDNA generation and signaling for NF‐κB activation. In correlation with patient data and above results, the TOP1 inhibitor‐based targeted therapy showed that WRN‐deficient melanoma tumors were highly sensitive to TOP1 inhibition in preclinical in vivo mouse model. Collectively, our findings identify hitherto unknown non‐enzymatic role of WRN RECQL helicase in pathological mechanisms underlying TOP1cc processing and subsequent NF‐κB‐activation, offering a potential targeted therapy for WRN‐deficient cancer patients.


| INTRODUC TI ON
Camptothecin (CPT) derivatives, for example, topotecan and irinotecan specifically inhibit Topoisomerase 1 (TOP1) and are clinically approved chemotherapeutics for a wide array of cancers including ovarian, colorectal and lung cancers .
TOP1 introduces transient single-strand breaks (SSBs) through formation of TOP1-DNA covalent complexes (TOP1cc) and re-joins DNA strands to allow the removal of negative as well as positive supercoiling of DNA during active replication and transcription. TOP1 inhibitors stabilize TOP1cc, leading to replication runoff and/ or stalled/reversed/collapsed forks mediated DNA double-strand breaks (DSBs) (Murai et al., 2019;. To this end, DSB formation is primarily dependent on persistence of TOP1cc, which are lethal if not repaired .
However, these studies have been carried out at acute micromolar concentrations of CPT, which are far higher than physiologically achievable concentrations. Understanding various molecular players involved in the removal of TOP1cc and CPT-resistance at nanomolar concentrations may be helpful in targeting cancers with better therapeutic outcomes.
Mutation in Werner (WRN) RECQL helicase is associated with premature aging syndrome (Werner syndrome, WS) and predisposition to multiple cancers. WS patients exhibit heightened incidence of neoplasia, for example, soft tissue sarcoma, osteosarcoma, malignant melanoma, meningioma, thyroid cancer, breast cancer, and leukemias (Lauper et al., 2013;Sugimoto et al., 2011). Imperatively, it is reported that WRN expression is epigenetically downregulated in multiple cancers in patients. We and others have previously demonstrated the instrumental role of WRN in DSB repair process and its therapeutic implications in response to radiation and chemotherapy (Gupta et al., 2021;Shamanna, Lu, de Freitas, et al., 2016). It has also been shown that WS patient cells or WRN-depleted cancer cells are hypersensitive to TOP1 inhibitors due to defective S-phase checkpoint and repair (Cheng et al., 2008;Patro et al., 2011). Survival of colorectal cancer patients with low expression of WRN was significantly increased after TOP1 inhibitor treatment (Agrelo et al., 2006). In the

METABRIC (Molecular Taxonomy of Breast Cancer International
Consortium) cohort comprising 1977 breast cancers,  have shown recently that aggressiveness and adverse prognostic outcome in breast cancer patients were correlated with altered TOP1 and WRN expression in the tumor . Although WRN is known to remove TOP1cc through proteasome-mediated degradation at very high (micromolar) concentrations of topotecan (Christmann et al., 2008), the precise mechanism of WRN-mediated removal of TOP1cc and the downstream effects at physiologically relevant concentrations (nanomolar) of CPT is not yet known.
Several studies have also shown that TOP1 inhibition-mediated DNA damage triggers NF-κB, a transcription factor, to induce prosurvival signaling, leading to resistance in different cancers (Martin et al., 2011). In response to DSB inducing genotoxic agents, PARP1 and ATM are activated, leading to their poly ADP-ribosylation (PARylation) and the assembly of ATM and NEMO (NF-κB essential modifier) or IKKγ, PIASy complex (Stilmann et al., 2009). Further, NEMO is SUMOylated and phosphorylated by PIASy and ATM, respectively, to trigger its translocation to cytoplasm and activation of IκB kinase (IKK) complex (IKKα, IKKβ, and NEMO). This in turn stimulates phosphorylation, ubiquitination, and subsequent degradation of inhibitor of NF-κB(IκBα) to release NF-κB (p65/p50) heterodimer (Stilmann et al., 2009;Yang et al., 2011). NF-κB (p65/p50) now freely enters the nucleus and stimulates gene expression for cell survival (Hayden & Ghosh, 2008). Since WRN-depleted cancer cells or WS patient-derived cells are highly sensitive to TOP1 inhibitors, we envisaged that WRN might be orchestrating TOP1cc removal and subsequent NF-κB activation to offer therapeutic resistance in WRN-proficient cells. Here, by using multiple approaches, we show that WRN RECQL helicase, independent of its helicase and exonuclease activity, regulates TOP1cc removal, leading to accumulation of single-stranded DNA (ssDNA) and activation of NF-κB to trigger robust resistance to physiologically relevant concentrations of CPT in vitro and in vivo.

| WRN triggers intrinsic resistance via NF-κB activation in response to TOP1cc
Previously, we have shown that WRN expression protects osteosarcoma cells from the toxic effects of the TOP1ccs at micromolar concentration of CPT (Patro et al., 2011). In order to unravel the hitherto unknown role of WRN in CPT-resistance in response F I G U R E 1 WRN regulates activation of NF-κB gene expression and induces intrinsic chemoresistance in response to TOP1cc. (a) Microarray expression profile of overall differential gene expression (DGE) with cutoff 1.5-fold in samples treated with 1 μM (5 h) and 50 nM (5-16 h) of CPT. (b) Heatmap of NF-κB related DGE in response to CPT (50 nM) at different time points. (c-e) WRN-proficient and WRN-deficient U2-OS, B16-F10, and COLO-205 cancer cells were treated with indicated concentration of CPT for 2 days and clonogenic survival of cells in drug-free medium were assessed. (f) WRN-WT and WRN-KO U2-OS cells were treated with CPT, as mentioned above, in the absence or presence of NF-κB inhibitor (Ro 106-9920, 1 µM) and clonogenic survival was assessed. (g) Schematic representation of luciferase expression through NF-κB promoter (h) WRN-WT and WRN-KO cells expressing NF-κB driven luciferase reporter were treated with CPT (50 nM) for indicated time periods, and NF-κB activation was assessed in terms of fold increase in luciferase activity. (i,j) WRN-WT and WRN-KO cells were treated with CPT (50 nM), for indicated time periods, and NF-κB activation was assessed in term of nuclear translocation of p65 (Bar: 10 μm; zoomed images: 5 μm). Quantification of p65 intensity in the nucleus is shown in (j). All the values indicated are mean ± SEM (n = 3-5). *p < 0.05 and **p < 0.01 with respect to respective untreated control cells. ns, not significant to TOP1cc, we initially performed whole-genome transcriptomic analysis of mRNAs in U2-OS cells treated with nanomolar concentration of CPT. Upon analysis, we observed a drastic change in the landscape of differential gene expression (DGE) at nanomolar concentration in time-dependent manner and also with higher concentration of CPT (i.e., micromolar) ( Figure 1a). The DGE at nanomolar concentration of CPT was unexpectedly different from the DGE at micromolar concentration of CPT ( Figure 1a, Table   S1). Further, to annotate the pathways activated at nanomolar concentration of CPT treatment, gene ontology (GO) based pathway classification was performed using PANTHER (Protein ANalysis THrough Evolutionary Relationships) classification system (Mi et al., 2019), which revealed that NF-κB-regulated genes were upregulated in response to nanomolar concentration of CPT (Figure 1b, Figures S1 and S2). In order to evaluate the role of WRN in cancer resistance to TOP1cc, WRN expression was abolished in U2-OS osteosarcoma by CRISPR-Cas9 double nickase system (WRN-KO vs. WRN-WT cells expressing control CRISPR-Cas9 double nickase vector; Figure S3a). Our results revealed that WRN-deficient (WRN-KO) cells showed higher sensitivity to CPT in clonogenic assay, which was rescued by ectopic expression of WRN in WRN-KO cells (Figure 1c). We also depleted WRN in B16-F10 melanoma cells by shRNA expressing lentiviral system or overexpressed WRN in COLO-205 colon carcinoma cells ( Figure S3b,c), which are known have low WRN expression due to epigenetic silencing (Agrelo et al., 2006). In both B16-F10 melanoma and COLO205 colon carcinoma cells, WRN deficiency was strongly associated with enhanced sensitivity to nanomolar concentrations of CPT (Figure 1d,e). In order to unravel whether WRN expression might be linked to NF-κB activation and de novo resistance to CPT treatment, we assessed the CPT sensitivity of WRN-proficient cells (WRN-WT U2-OS) in the presence of an NF-κB-specific pharmacological inhibitor (Ro 106-9920). We found that NF-κB inhibition significantly enhanced sensitivity of WRN-WT cells in response to CPT while it had no or marginal effect on WRN-KO cells, indicating a key role of WRN in NF-κB activation mediated CPT-resistance ( Figure 1f). We further employed a luciferase-based NF-κB reporter assay (Figure 1g), in order to assess the role of WRN in activation of NF-κB at nanomolar concentration of CPT. In accordance with our microarray results, we found a significant time-dependent enhancement of luciferase (NF-κB) activity in WRN-WT cells, even at nanomolar concentrations of CPT, while no or marginal increase was observed in WRN-KO cells ( Figure 1h). Consequently, IκBα degradation, a critical requirement for NF-κB activation, was rapidly and significantly enhanced in WRN-WT cells while it was significantly abrogated in WRN-KO cells in response to CPT treatment ( Figure S3d,e).
Moreover, nuclear translocation of p65 (NF-κB), from cytoplasm, was observed with CPT treatment in a time-dependent manner in WRN-WT cells, which was significantly ameliorated in WRN-KO cells (Figure 1i,j). Altogether, these results showed a novel role of WRN in triggering intrinsic/de novo resistance in cancers toward CPT to TOP1 inhibition through regulation of NF-κB activation.

| WRN regulates NF-κB activation via CHK1 and PARP1
Although extensive studies are available for TOP1cc formation, DNA repair and NF-κB activation in response to micromolar concentrations of TOP1 inhibitors, only few studies describe the remodeling of replication forks in response to clinically relevant nanomolar concentrations of TOP1 inhibitors (Berti et al., 2013;Iannascoli et al., 2015;Ray Chaudhuri et al., 2012). However, the precise role of TOP1cc dynamics and NF-κB activation at such low concentrations of TOP1 inhibitor is not known so far.
To identify the WRN regulated pathway that might activate NF- with our previous studies (Gupta et al., 2021;Patro et al., 2011), showing that WRN-deficient cells are defective in the activation of ATM and ATR-CHK1 signaling in response to genotoxic agents ( Figure 2a). Previously, we and others have shown that WRN undergoes degradation in response to ionizing radiation (Gupta et al., 2021) and high concentration of CPT (10 μM) . Interestingly, WRN mRNA level was enhanced while WRN protein level remained unaffected in response to physiologically relevant nanomolar concentration of CPT ( Figure S4a,b). Therefore, the above observed effects were not related to perturbed WRN protein level in WT cells. We also observed that EdU (5-ethynyl 2'-deoxyuridine) incorporation was similar in WRN-WT and WRN-KO cells ( Figure S5b), suggesting the compromised DDR is also not related to differential S-phase distribution in these cells.
Further, we sought to know whether ATM and NEMO mediated activation of canonical NF-κB pathway, a primary mechanism in various genotoxic stimuli (Stilmann et al., 2009), might also be associated with cellular response to nanomolar concentrations of CPT. In this regard, IκBα degradation was severely abrogated in NEMO silenced

| WRN regulates TOP1cc removal through TDP1 and CTIP
TOP1 inhibitors induce TOP1cc mediated SSBs and replication runoff for generation of DSBs . In order to identify the molecular damage responsible for WRN-mediated NF-κB activation, IκBα degradation in response to CPT was assessed in the presence of aphidicolin (APH). APH is known to block replication runoff and enhance stalled replication forks, through its ability to inhibit DNA polymerase (Seiler et al., 2007;Su et al., 2014). In this regard, our results showed that IκBα degradation was grossly unperturbed by presence of APH in response to CPT in WRN-WT cells ( Figure S7a Further, we focused on the role WRN in TOP1cc removal and NF-κB activation. First, we assessed TOP1cc formation in live cancer cells by employing a previously reported confocal microscopybased FRAP (Fluorescence Recovery After Photobleaching) assay (Das et al., 2016). In this assay, ectopically expressing GFP-tagged human TOP1 (EGFP-TOP1) was photobleached at a small region of interest (ROI) in the cell nucleus and kinetics of replacement of the photobleached EGFP-TOP1 in ROI by fluorescent EGFP-TOP1 proteins was assessed in the absence or presence of CPT ( Figure 3a). Figure 3a Figure   S8, the kinetics of TOP1cc formation, in response to nanomolar concentration of CPT, was significantly delayed in both WRN-WT and KO quiescent cells. Interestingly, TOP1cc removal was not observed up to 8 h in both the cells. This result suggests TOP1cc formation and removal follows a WRN-independent process in non-proliferating but transcriptionally active quiescent (G 0 ) cells. Further, to evaluate whether the role of WRN is U2-OS cell line specific, HCT116 colorectal cancer cells were also employed. Note, WRN expression is severely downregulated in HCT116 due to epigenetically inactivation of WRN promoter (Agrelo et al., 2006) . In accordance with this report, we too observed severe downregulation of WRN expression in HCT116 as compared to U2-OS cells ( Figure S9a conditions. It is also known that WRN interacts/influences some of these proteins (Franchitto & Pichierri, 2004;Murfuni et al., 2013).

As shown in
Further, we sought to know whether WRN-induced removal of TOP1cc is mediated through some of these proteins in response to low dose of CPT. To this end, our results showed that TDP1 silencing led to delay in TOP1cc removal while MRE11 silencing or inhibition of exonuclease activity MRE11 by mirin have marginal/no impact on the TOP1cc removal (Figure S10a-c). Interestingly, depletion of CTIP, by lentivirus (Gupta et al., 2021), led to significant inhibition of TOP1cc removal in response to CPT ( Figure S11a,b). Together, our results suggested that WRN-mediated TOP1cc removal might be influenced through TDP1 and CTIP proteins.

| Neither helicase nor exonuclease activity of WRN is essential for TOP1cc removal
WRN has exonuclease and helicase activities and multiple proteininteracting domains and a C-terminal motif for binding DNA (Bohr, 2008). We sought to know whether enzymatic activities of WRN are essential for TOP1cc removal at nanomolar concentration of CPT ( Figure 4a). For this, FLAG-tagged full length WRN (WRN WT ) was mutated at 84 and 577 amino acid positions to create exonuclease (WRN E84A ) and helicase (WRN K577M ) defective WRN (Sharma et al., 2003), respectively, by using site-directed mutagenesis ( Figure 4b). Besides, a double mutant of WRN (WRN E84A−K577M ), F I G U R E 3 WRN is a key regulator of TOP1cc removal and NF-κB activation. (a-c) WRN-WT and WRN-KO cells expressing EGFP-TOP1 were pretreated with CPT (100 nM) for 6 h. Fluorescence of EGFP-TOP1, at a small ROI in the nucleus, was bleached with laser irradiation, and fluorescence recovery was assessed by confocal microscopy at indicated time points (Bar: 10 μm). Quantification of fluorescence recovery is shown in B and C. Error bars represent mean ± SE (n = 10). (d) Schematic representation for stabilized TOP1cc in the presence of CPT. (e-h) WRN-WT and WRN-KO cells were treated with CPT (50 nM) in the absence or presence of MG132 (10 μM) or PARPi (100 nM) for indicated time periods, and TOP1cc was analyzed by RADAR based slot blot assay. DNA was probed as a loading control, in the same samples by DNAspecific antibody. Quantification of dot blot assay is shown in F-H. (i) WRN-WT and WRN-KO cells were treated with CPT (50 nM) in the absence of presence of APH (500 nM) for indicated time periods, and TOP1cc level was analyzed by RADAR based slot blot assay. All the values indicated are mean ± SEM (n = 5) **p < 0.01 with respect to vehicle treatment in the respective cell types.

| TOP1cc removal by WRN induces ssDNA to activate CHK1 for NF-κB activation
We have previously shown that WRN induces phosphorylation of RPA2 (ssDNA formation) and CHK1 at both nanomolar ( Figure 2a) and micromolar concentration of CPT (Patro et al., 2011). We also showed a key role of CHK1 in NF-κB activation and intrinsic resistance to CPT treatment (Figure 2f and S2a,b). Next, we sought to know whether enzymatic activity of WRN is essential for ssDNA formation and CHK1 activation in response to CPT treatment. In correlation with defective TOP1cc removal, we found that both RPA2 (ssDNA) and CHK1 phosphorylation were also severely impaired in WRN-KO cells, as compared to WRN-WT cells in response to CPT (Figures 2a and 5a).  (Figure 6a), which was assessed by immunofluorescence microscopy (Gupta et al., 2021;Patro et al., 2011). Our results revealed that CPT treatment induced ssDNA formation and a corresponding enhancement of p65 nuclear translocation in WRN-WT cells (Figure 6b,c).
In WRN-WT cells, 48.6 ± 3.5% of nuclei were ssDNA +ve and p65 +ve whereas in WRN-KO cells, 12.4 ± 3.4% nuclei were ssDNA +ve and p65 +ve , suggesting a pivotal role of ssDNA in NF-κB activation (Figure 6b,c). This further validated our previous results pertaining to the regulatory role of WRN in these two interlinked processes (Figures 1h,i and 2a). Imperatively, ectopic expression of WT, exonuclease dead, helicase dead or double mutant forms of WRN almost completely restored ssDNA formation and p65 nuclear translocation in WRN-KO cells (Figure 6b,c). These results further validated a nonenzymatic role of WRN in ssDNA formation and NF-κB activation.
Altogether, these results showed that non-enzymatic WRN plays a key role TOP1cc removal, ssDNA formation and the activation of CHK1 and NF-κB to mediate therapeutic resistance to TOP1 inhibitors in vitro. Finally, we assessed therapeutic potential of TOP1 inhibitor for targeting WRN-deficient melanoma tumor, which otherwise is known to be resistant to other therapeutic modalities (Figure 6d,e) (Gupta et al., 2021). In this regard, our results of animal studies revealed that CPT treatment (1 mg/kg body weight) significantly reduced WRN-depleted melanoma, as compared to WRN-proficient melanoma, indicating translational potential of TOP1 inhibitor therapy for cancer patients with WRN deficiency. Besides, we have also analyzed patient survival from TCGA database of colorectal adenocarcinoma (available at c-Bioportal) (Cerami et al., 2012;Gao et al., 2013) and found WRN-deficient cancer patient has more survival rate in contrast to increased WRN expressing cancer patients ( Figure 6f). To this end, WRN deficiency in cancer patients may be targeted with TOP1 inhibitor therapy for better clinical outcomes.

| DISCUSS ION
NF-κB plays a crucial role in chemoresistance to TOP1 inhibitors in various preclinical and clinical settings (Rasmi et al., 2020;Tomicic & Kaina, 2013). Currently, combination therapy of TOP1 inhibitors and NF-κB inhibitors is undergoing different stages of clinical trials (Rasmi et al., 2020;Tomicic & Kaina, 2013). Although activation of NF-κB in response to DSB inducing genotoxic stresses is known, the molecular mechanism for NF-κB activation in response to physiologically relevant concentrations of CPT, which do not primarily cause DSBs (Volcic et al., 2012), remains elusive. To the best of our knowledge, the present study provides the first evidence that WRN RECQL protein is a key factor for activation of NF-κB pathway, leading to intrinsic resistance to nanomolar concentrations of CPT. We F I G U R E 6 Role of WRN in ssDNA generation and nuclear translocation of p65 and its therapeutic implications. (a) Schematic representation for WRN-mediated removal of TOP1cc, generation of ssDNA and CHK1 phosphorylation and NF-κB activation. (b, c) WRN-WT and WRN-KO cells were treated with BrdU for 36 h for uniform labeling of BrdU on both the DNA strands. Further cells were exposed to CPT for indicated time periods for TOP1cc removal mediated ssDNA generation and nuclear translocation of NF-κB (p65). BrdU exposed in ssDNA (red) and p65 (green) was assessed in non-denaturing native condition by immunofluorescence microscopy (Bar: 5 μm). % Nuclei with more than threshold fluorescence were considered positive for p65 and/or BrdU and plotted in C. All the values indicated are mean ± SEM (n = 3). *p < 0.05 with respect to respective untreated cell types. (d,e) Mice bearing WRN-WT and WRN-KO melanoma tumors were treated with vehicle or CPT (1 mg/kg, once on 1st, 3rd, and 5th day per week for 4 weeks). Tumor volume was measured once every alternate day. After 30 days, mice were sacrificed and tumors were removed and analyzed. Data represent the mean ± SD, n = 6 per group. *p < 0.01 w.r.t corresponding vehicle-treated tumor. (f) Survival of patients with WRN high and low expression in cancer using data available at cBioportal. The EXP < −1 denotes mRNA expression is <1 standard deviations (SD) below the mean, and EXP > 1 denotes mRNA expression is >1 SD above the mean. have demonstrated that NF-κB-related genes are differentially expressed in a time-dependent manner in response to nanomolar concentrations of CPT (Figure 1b, Table S1). Luciferase-based reporter assay for NF-κB activation revealed significantly higher NF-κB acti-

vation in WRN-WT cells compared to WRN-KO cells in response to
CPT (Figure 1g,h). In agreement with this result, IκBα degradation and subsequent translocation of NF-κB (p65) to nucleus were significantly impeded in WRN-KO cells, as compared to WRN-WT cells in response to TOP1cc ( Figure S3d,e and Figure 1i,j). Imperatively, WRN-deficient osteosarcoma, melanoma, and colon carcinoma were highly sensitive to nanomolar concentration of CPT (Figure 1c-e).
We also demonstrated that inhibition of NF-κB, with Ro106-9920, enhanced the chemosensitivity of WRN-WT cells while the effect was marginal in WRN-KO cancer cells (Figure 1f). Taken together, our data establish a key role of WRN in the activation of NF-κB to induce intrinsic chemoresistance to CPT.
Replication plays an essential role in regulating sensitivity/resistance to TOP1 inhibitors, suggesting an underlying involvement of replication-associated-DNA-lesions in this process. Recently, it has been shown that replication forks undergo fork reversal in response to nanomolar concentrations of CPT (Ray Chaudhuri et al., 2012).

Deficiency of RECQL1 enhances the level of reversed forks in re-
sponse to CPT (Berti et al., 2013). Besides, WRN is known to protect nascent replicating DNA strands from MRE11 mediated degradation at micromolar but not at nanomolar concentrations of CPT (Berti et al., 2013;Palermo et al., 2016;Su et al., 2014). At micromolar concentrations of CPT, CDK1 causes phosphorylation of WRN to induce DNA2-dependent extensive resection of replication-associated DSBs (collapsed replication forks). In contrast, DNA2-mediated limited processing of nascent strands was observed in WT cells while exonuclease defective WRN causes extensive degradation of nascent strand in response to nanomolar concentrations of CPT (Iannascoli et al., 2015). It is plausible that some of these replicationassociated-DNA-lesions (stalled replication forks, degraded forks, reverse/regressed forks, or collapsed forks) may be responsible for activation of NF-κB pathway in response to nanomolar concentrations of CPT. Intriguingly, our results showed that treatment with APH, a replication inhibitor (Seiler et al., 2007;Su et al., 2014), had no or marginal effects on TOP1cc formation and NF-κB activation Accumulation of TOP1cc without repair for longer periods is known to enhance the sensitivity of proliferating cancer cells.
Strategies to modify chemical structures of TOP1 inhibitors to generate stabilized and persistent TOP1cc showed enhanced sensitivity of cancer cells in vitro and in vivo (Crawley et al., 2015;Kundu et al., 2019). Persistent TOP1cc is also associated with enhanced and persistent DSBs, resulting in hypersensitivity of proliferating cancer cells (Kundu et al., 2019;Marzi et al., 2018). Faster removal of TOP1cc is considered as one of the reasons for chemoresistance to TOP1 inhibitors. Human cancer cells use both proteasome-dependent and proteasome-independent endonucleolytic pathways (TDP1, PARP1, XPF-ERCC1, MRE11, CTIP, and MUS81-EM1) to remove TOP1cc (Tomicic & Kaina, 2013). In this regard, we showed that a rapid induction of TOP1cc and its subsequent removal in WRN-proficient cancer at nanomolar concentration of CPT, while this was abrogated in WRN-KO cells, suggesting a key role of WRN in TOP1cc removal/ repair (Figure 3a,e). Interestingly, inhibition of proteasome or PARP1 inhibitor partially or completely abrogated TOP1cc removal in WRN-WT cells, respectively (Figure 3e). This suggested that WRN regulates TOP1cc removal through both proteasome-dependent and independent pathways. Interestingly, TDP1 depletion delayed TOP1cc removal while CTIP depletion significantly affected TOP1cc removal, suggesting a role of WRN in influencing/recruiting TDP1 and CTIP to the site of TOP1cc (Figures S10 and S11). Imperatively, impaired helicase and/or exonuclease activities or deletion of exonuclease domain in WRN did not impede TOP1cc removal ability of WRN (Figure 4 and Figure S12), clearly suggesting a non-enzymatic role of WRN in the process of TOP1cc removal.
Since our results revealed that replication may not play a major role in NF-κB activation, we focused on exploring the role of TOP1cc removal/repair in this process. It is plausible that WRN-mediated TOP1cc removal and subsequent SSB repair may involve the generation of ssDNA, which may elicit signaling to trigger NF-κB activation.
Notably, SSB repair involves DNA-end resection by APE2 for generation of ssDNA to induce ATR-CHK1 DDR pathway (Lin et al., 2018;Willis et al., 2013). In WRN-WT cells, we observed RPA2 phosphorylation and ATR-CHK1 phosphorylation in a time-dependent manner in response to nanomolar concentrations of CPT. Phosphorylation RPA2 and CHK1 was severely defective in WRN-KO cells (Figures   2a and 5a), showing a key role of WRN in ssDNA generation and CHK1 activation. Interestingly, helicase and exonuclease activities of WRN are not essential for these processes (Figure 5a and Figure   S15). Further, the ability of non-enzymatic WRN to resolve TOP1cc removal and ssDNA generation was significantly correlated with NF-κB activation mediated p65 nuclear translocation at individual cell level and at cell population level (NF-κB luciferase assay) (Figure 6a-c and Figure S14). This was defective in WRN-KO cells in response to CPT treatment. Besides, our results revealed that WRN-induced CHK1 activation is essential for NF-κB activation and intrinsic resistance to CPT (Figure 2f, Figures S6a and S15). In summary, our results pinpoint WRN as a critical regulator of TOP1cc removal and ssDNA generation and activation of CHK1 signaling for NF-κB mediated chemoresistance to TOP1 inhibitors (Figure 6g).
In conclusion, the present study establishes the significance of WRN-mediated removal of TOP1cc and NF-κB activation in chemoresistance of cancers to physiologically relevant concentration of CPT. Efficient removal of TOP1cc also requires PARP1 and CHK1 activation are key requirement for NF-κB activation in response to CPT. Aggressiveness and adverse prognostic outcome in breast cancer patients was correlated with altered TOP1 and WRN expression in the tumor Shamanna, Lu, de Freitas, et al., 2016). Therefore, our current investigation also provides insights to establish a foundation to devise new synthetic lethal strategies by combining CHK1 or PARP1 inhibitors with TOP1 inhibitors to counteract WRN and NF-κB mediated intrinsic tumor resistance in clinical settings. Besides, WRN expression is downregulated in multiple cancers in patients due to epigenetic silencing. Thus, identifying WRN deficiency in individual cancer patient may help in achieving better and predictable prognosis through TOP1 inhibitor-based personalized/targeted therapy. Our investigation, which demonstrates the association between impaired removal of TOP1cc and defective NF-κB activation and WRN deficiency, explains enhanced survival of colorectal cancer patients with WRN deficiency in response TOP1 inhibitor treatment (Agrelo et al., 2006).
All the cell lines used were certified tested and authenticated by DNA profiling for polymorphic short tandem repeat markers.

| Clonogenic assay
Cells were seeded (500/well) in 6-well plates and were grown in the presence of vehicle or different concentrations of CPT for 24 h and then allowed to grow in CPT-free media for 10-14 days. For combination treatment with different inhibitors, the treatment with inhibitor was given 0.5 h before CPT treatment and then removed after 24 h and allowed to grow in complete medium for 10-14 days.
Subsequently, cells were washed in PBS, fixed in methanol, and stained with crystal violet (0.5% in PBS). Colonies of ≥30 cells were manually counted, and the survival curves were derived from the colony numbers normalized to respective control. COLO-205 cells were also treated as mentioned above. After 3 days of recovery, viable cells were counted by staining with trypan blue (0.4% in PBS).

| Generation of knockout and knockdown cells
U2-OS cells were transfected with WRN/CRISPR-Cas9 and control CRISPR-Cas9 double nickase plasmid system from SCBT according to the manufacturer's protocol. Cells were selected with puromycin for stable knockout (KO) of WRN. To generate stable WRN knockdown cells, the B16-F10 cells were exposed to lentiviral particles encoding either scrambled or WRN-specific shRNAs

| Microarray data analysis
The RNA isolation, labeling, and hybridization were conducted by a commercial Affymetrix array service (M/S Vimta Lab Services, India). The manufacturer's protocol was followed for the determination of gene expression data using Affymetrix Human GeneCHIP 1.0 ST arrays. Briefly, this method includes first and second strand cDNA synthesis, double-stranded cDNA purification, cDNA synthesis, biotin-labeled cDNA quantification, and cDNA fragmentation followed by subsequent hybridization. Following hybridization and washing, the Affymetrix arrays were scanned with an Affymetrix GeneChip. Image generation and feature extraction were performed using the Affymetrix Software. The data from arrays that passed the manufacturer's quality specifications were used for further analysis.
The p value cutoff for differentially expressed genes was set at 0.05.
In analysis, for differential gene analysis cutoff set for fold change is set to be log2.

| Slot blot assay for trapped TOP1cc
TOP1cc (TOP1-DNA covalent complex) formation was assessed by RADAR (rapid approach to DNA adduct recovery) assay, as reported previously (Kiianitsa & Maizels, 2013 where MFI represents mean fluorescence intensity. Recovery graphs were plotted using percentage recovery data.

| Immunofluorescence assay
Immunofluorescence assay was carried out as per the reported protocol (Gupta et al., 2021) with minor modifications. For measuring ssDNA and p65, cells were grown on coverslips in presence of BrdU (10 μM) for 36 h, chased for 3 h in BrdU free medium, and treated with CPT for indicated time period. Subsequently, cells were fixed with 4% PFA, washed twice in PBS, and fixed with chilled methanol for overnight. For immunostaining, cells were permeabilized in Triton X-100 (0.5% in PBS) on ice for 10 min, washed three times with PBS, and subjected to indirect immunofluorescence using primary antibody against p65 and BrdU and suitable secondary antibody tagged with Alexa-595 and Alexa-488. Slides were mounted in Prolong Gold anti-fade reagent with DAPI. Images were captured using an LSM780 Meta laser scanning confocal microscope (Zeiss, Germany).

| Animal studies
Six-week-old male C57BL6 mice were procured from BARC animal house facility, Mumbai, India, in accordance with ethical clearance from the BARC Animal Ethics Committee. Melanoma tumor development, housing and routine care of animals were carried out as per the standard animal maintenance guidelines and our previous report (Gupta et al., 2021). For melanoma tumor induction, B16-F10 WRN-WT and WRN-KD cells (1 × 10 5 cells/0.2 ml DMEM/mouse) were injected subcutaneously in the right flank of the mice. Tumor-bearing mice were randomly grouped (6 mice/group), and tumors were allowed to grow to an average volume of 100 mm 3 before treatment. Palpable tumors were noticed approximately after 7 days of tumor induction. Mice bearing palpable melanoma tumors were treated with vehicle or CPT (1 mg/kg in three doses on 1st, 3rd, and 5th day per week) through i.p.
injection. During treatment, tumor volumes were measured by caliper once every alternate day and calculated according to the formula (L × W 2 )/2 (L, length; W, width). Upon completion of the experiments, the mice were sacrificed after an overdose of thiopental, the tumors were excised and their weights measured. The animals were dissected, and macroscopic analysis was also carried out to visualize major morphological changes in the organs.

| Statistical analysis
The data are presented as the mean ± SEM or mean ± SD. Comparisons between two groups were performed using an unpaired Student's t-test in GraphPad software. Comparisons between two groups were performed using one-way analysis of variance (ANOVA) and two-way ANOVA for multiple comparisons in GraphPad software. A value of p < 0.05 was considered statistically significant.

ACK N OWLED G M ENTS
We acknowledge BARC animal house for providing all the fa-

CO N FLI C T O F I NTE R E S T
None declared.

AUTH O R CO NTR I B UTI O N S
BSP and PG involved in conceptualization. PG, AGM, and BSP involved in investigation, methodology, data analysis, writing original draft and revised manuscript. BSP involved in supervision.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data are available in the main text or the supplementary materials. Additional data related to this paper may be requested from the authors.