mTOR inhibition enhances the antitumor efficacy of RAF dimer-MEK blockade by inhibiting the ATF4-MTHFD2 pathway

mTOR inhibition enhances the antitumor efficacy of RAF dimer-MEK blockade by inhibiting the ATF4-MTHFD2 pathway | 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 Article mTOR inhibition enhances the antitumor efficacy of RAF dimer-MEK blockade by inhibiting the ATF4-MTHFD2 pathway Sonia Del Rincon, Feiyang Cai, Fan Huang, Christophe Goncalves, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6857127/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 RAF monomer inhibitors are clinically approved for the treatment of BRAF V600 -mutant melanoma in combination with a MEK inhibitor, but ineffective in other melanoma subtypes. Moreover, RAF dimer inhibitors, such as belvarafenib, when combined with MEK inhibitors (cobimetinib) have promising but limited efficacy in non- BRAF -mutant melanomas. Here, we report that the mTOR inhibitor sapanisertib improves the efficacy of combined belvarafenib and cobimetinib therapy in NRAS , NF1 , and KIT -mutant melanomas. Mechanistically, sapanisertib combined with belvarafenib and cobimetinib suppressed ATF4 expression and its target gene MTHFD2 while inducing DNA damage, revealing a previously underappreciated role of the ATF4-MTHFD2 axis in DNA damage repair and drug response. Human and murine models resistant to combined belvarafenib and cobimetinib exhibited elevated levels of ATF4 and MTHFD2 and were sensitive to sapanisertib. This study provides novel treatment opportunities for patients with non- BRAF -mutant melanomas, or those who relapse following belvarafenib and cobimetinib combination therapy. Health sciences/Oncology/Cancer/Skin cancer/Melanoma Biological sciences/Cancer/Skin cancer/Melanoma mTOR belvarafenib cobimetinib sapanisertib melanoma targeted therapy ATF4 MTHFD2 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Melanoma is one of the most aggressive skin cancers, being highly metastatic 1 . Cutaneous melanoma can be classified into four genomic subtypes based on the mutation status of BRAF , NRAS , NF1 , and a fourth subgroup termed triple wild-type 2 . The triple wild-type melanoma is often driven by less frequently detected hotspot mutations, such as in KIT , a receptor tyrosine kinase 2 . Nevertheless, the RAS-RAF-MEK-ERK pathway is an essential signal transduction cascade in all melanoma subtypes 3 , 4 . FDA-approved BRAF V600 inhibitors, vemurafenib, dabrafenib, and encorafenib 4 , 5 , are clinically combined with MEK inhibitors and cause rapid regression of BRAF -mutant melanomas, which account for around 50% melanoma cases. However, these drugs are unable to inhibit both protomers of the RAF dimer and paradoxically activate MAPK signaling in non- BRAF V600 mutant melanoma 6 – 9 . Currently, there is no FDA-approved targeted therapy for NRAS -mutant, NF1 -mutant, and triple wild-type melanomas, making them classically “hard-to-treat” subtypes. In particular, NRAS -mutant melanomas exhibit greater aggressiveness and are associated with a poorer prognosis in comparison to BRAF -mutant melanomas 10 . Hence, it is crucial to expand novel and effective therapies for patients with hard-to-treat melanoma. A novel RAF dimer inhibitor belvarafenib has shown promising pre-clinical results in NRAS -mutant melanoma 11 . However, belvarafenib resistant cell clones exist in melanoma, sustaining MAPK signaling in the presence of belvarafenib treatment. These resistant cells remain dependent on MAPK signaling and are sensitive to downstream MAPK inhibitors (MAPKi), such as MEK inhibitors (e.g., cobimetinib) 11 . As a result, combined belvarafenib and cobimetinib was trialed in NRAS -mutant melanoma patients (NCT03284502), with an overall response rate (ORR) of 38.5% 12 . Another pan-RAF and MEK inhibitor combination therapy (naporafenib + trametinib) is also being investigated in a randomized phase 3 trial for NRAS -mutant melanoma patients (NCT0634067), with an ORR of 46.7% 13 . Despite encouraging data on the combination of belvarafenib and cobimetinib in melanoma, resistance remains a major challenge, as observed in nearly all patients treated with targeted therapy 12 , 14 . In addition to MAPK signaling, driver mutations in melanoma often simultaneously activate the PI3K-AKT-mTOR pathway, which is an essential signaling cascade driving MAPK inhibitor resistance in melanoma 15 – 18 . In BRAF -mutant non-melanoma skin cancer with primary or secondary resistance to BRAF inhibitors, there is also a strong correlation between PI3K/mTOR signaling pathway and BRAF inhibitor resistance 19 . Importantly, most studies in melanoma examining the role of mTOR in the context of MAPK pathway-targeted therapies have focused largely on the BRAF -mutant tumors. As a result, the contribution of mTOR signaling to therapeutic response and resistance in other melanoma subtypes, such as NRAS -mutant or wild-type melanomas, remains largely unexplored, representing a significant gap in our understanding of subtype-specific vulnerabilities. mTOR activation has been linked to resistance to Wee1 inhibitors in lung cancer patients through DNA damage repair pathway 20 , suggesting that mTOR mediates a broader mechanism beyond the well-characterized crosstalk between MAPK and mTOR pathways. Furthermore, to cope with stresses from rapid cell division and anti-tumor therapy, cancer cells often activate the integrated stress response (ISR) pathway, governed by activating transcription factor 4 (ATF4) 21 . Mechanistically, mTOR can control ATF4 translation 22 and plays an important role in resistance to BRAF-targeted therapy 23 . Given the essential roles of the mTOR pathway and ISR signaling in therapy tolerance and resistance, we sought to investigate whether blocking mTOR would augment therapeutic responses to combined belvarafenib plus cobimetinib therapy. We chose to use sapanisertib (a.k.a INK128), an ATP-competitive catalytic inhibitor of mTOR 24 that has been evaluated in clinical trials to treat a variety of solid tumors (NCT02412722, NCT02197572), showing well-tolerated adverse effects 25 , 26 . Unlike first-generation mTOR inhibitors 27 , sapanisertib targets both mTORC1 and mTORC2, blocking feedback loops that reactivate AKT, thereby expanding its potential in therapy-resistant tumors and provides a rationale for its use in combination with other anti-cancer agents. Indeed, we found that sapanisertib potentiated the anti-tumor effects of belvarafenib plus cobimetinib both in vitro and in vivo . By interrogating a list of ATF4 target genes that may underpin the response to belvarafenib + cobimetinib + sapanisertib (referred hitherto as triple therapy), MTHFD2 emerged as a top candidate. MTHFD2 is an enzyme in the folate cycle, which functions as a mediator of purine synthesis and DNA repair 28 . Hence, we hypothesized that MTHFD2 facilitates DNA repair in belvarafenib plus cobimetinib-treated cells, thus promoting their therapy tolerance. Finally, we explored the therapeutic benefit of the triple therapy to unravel how mTOR inhibition rewires MTHFD2-mediated purine synthesis and DNA repair in belvarafenib plus cobimetinib therapy in hard-to-treat melanomas. Accordingly, we demonstrate how incorporating mTOR inhibition is an effective strategy in targeting melanomas that are resistant to belvarafenib plus cobimetinib. Results Inhibiting mTOR augments the efficacy of combined belvarafenib and cobimetinib in melanoma cells. Oncogenic mutations of NRAS , NF1 and KIT in melanoma drive hyperactivated RAS signaling, which contributes to activated MEK/ERK and PI3K/mTOR signaling 29 . Given the emerging role of the mTOR pathway in melanoma response and resistance to MAPK inhibition 15 , we hypothesized that targeting the mTOR pathway could enhance the antitumor activity of belvarafenib plus cobimetinib combination therapy in non-BRAF mutant melanoma cells. We thus determined the half-maximal inhibitory concentration (IC 50 ) of belvarafenib and cobimetinib in the context of combination therapy using a panel of human melanoma cell lines harboring different genetic mutations ( NRAS Q 61 R -mutant BLM cells, NRAS Q 61 K -mutant WM3406 and WM3623 cells, NF1 Q1336* -mutant MeWo cells, c-KIT D820Y -mutant HBL cells and NRAS G 12 V -mutant YUGOE cells 30 ) (Supplementary Fig. 1a). Next, we treated melanoma cell lines with belvarafenib (25 nM), cobimetinib (50 nM), and the mTOR inhibitor sapanisertib (hereafter referred to as INK128 24 , 25 nM) and monitored their impact on colony formation capacity (Fig. 1 a,b and Supplementary Fig. 1b,c). As expected, belvarafenib plus cobimetinib significantly decreased colony formation, but some cells persisted, surviving the 10-day treatment (Fig. 1 a and Supplementary Fig. 1b). Notably, there were fewer colonies in all four melanoma cell lines treated with the triple therapy of INK128 plus belvarafenib plus cobimetinib (Fig. 1 a and Supplementary Fig. 1b). Using western blot, all cell lines were tested to confirm on-target engagement at the chosen drug concentrations (Fig. 1 a,b). We further showed that the triple therapy induced the highest level of apoptosis at 48 hours. Interestingly, INK128 did not induce apoptosis on its own, consistent with its anti-tumoral effects being mostly cytostatic, rather than cytolytic 24 , 31 , but significantly increased belvarafenib plus cobimetinib induced apoptosis in all the melanoma subtypes tested (Fig. 1 c and Supplementary Fig. 1c). The triple therapy induces apoptosis of melanoma cells through the inhibition of ATF4 and induction of DNA damage. To investigate potential mechanisms through which the triple therapy most effectively eliminated melanoma cells, we performed RNA sequencing analysis on the NRAS -mutant WM3406 cell line (Supplementary Fig. 2a,b). Gene set enrichment analysis (GSEA) on differentially expressed genes (DEGs) in cells treated with the triple therapy versus the vehicle-treated cells identified the “cellular response to stress” pathway, centered on the integrated stress response (ISR) signaling, as being downregulated in the triple therapy-treated cells (Fig. 2 a,b and Supplementary Fig. 2c-e). The activation of the ISR, marked by increased expression of its major effector ATF4, plays a crucial role in supporting persister cell survival during BRAF-targeted therapy 32 . Generally, ISR promotes the phosphorylation of eIF2α at serine 51, inhibiting translation initiation and leading to upregulation of ATF4 33 . Western blot analysis showed that the triple therapy resulted in an inhibition of ATF4 protein expression in WM3406 and MeWo cells (Fig. 2 c), without any consistent changes in the phosphorylation status of eIF2α (Fig. 2 c), which is also supported by a study showing mTORC1 controls ATF4 independently of changes in eIF2α phosphorylation 22 . Similar phenotypes were observed in a murine NRAS mutant melanoma cell line termed MaNRAS1007 34,35 (Fig. 2 d and Supplementary Fig. 2g). Melanoma shows intratumor heterogeneity in the activity of MAPK signaling 36 , which we hypothesize, potentially regulates ATF4 level. Thus, we next reanalyzed a spatially resolved, unsupervised transcriptomics dataset generated from tumors induced in an NRAS -mutant melanoma mouse model (Supplementary Fig. 3), from which MaNRAS1007 cell line was derived 37 . Using previously defined pathway-specific gene signatures 38 , 39 , we noted that the tumor sample with the highest overall ATF4 expression also had the highest MAPK- and mTOR-pathway activity (Supplementary Fig. 3a-f), consistent with prior findings that both the MAPK and mTOR pathways drive ATF4 expression in melanoma 32 , 40 . Next, we applied our pathway activity analysis on the spatial transcriptomics data to assess the gene expression profile in each spatially localized environment (region of interest, ROI) (Supplementary Fig. 3a-f). As expected, the activities of MAPK and mTOR pathways were positively correlated (Supplementary Fig. 3g), supporting both pathways being simultaneously activated through common driver mutations (i.e., NRAS ) in melanoma 41 . Interestingly, ROIs with high MAPK activity (pink and green, top 20 percentile) showed highest ATF4 expression, regardless of their mTOR activity. On the contrary, in MAPK lo regions, high mTOR activity (top 20 percentile) is associated with retained ATF4 expression compared with MAPK hi ROIs, whereas MAPK lo mTOR lo regions had the lowest level of ATF4 (Supplementary Fig. 3g). Similarly, compared with ROIs with high MAPK signaling activity, those with low MAPK signaling activity showed stronger correlation between ISR and mTOR pathway activity (Supplementary Fig. 3h). Together, these data suggest that while both the MAPK and mTOR pathways facilitate ATF4 expression 32 , 40 , elevated mTOR activity may compensate for MAPK suppression by maintaining ATF4 levels and driving ISR signaling. Given the importance of the ATF4-governed ISR in melanoma drug tolerance and resistance 23 , we hypothesized that ATF4 may be essential for the survival of melanoma persister cells during combined belvarafenib plus cobimetinib therapy, which ultimately lead to the development of resistance 42 . Although a 24-hour combination treatment with belvarafenib plus cobimetinib repressed ATF4 expression (Fig. 2 c,d), when WM3406 and MeWo cells were treated with belvarafenib plus cobimetinib for 10 days, at the timepoint when only persister cells remained (Fig. 1 a, 2 e), ATF4 expression was no longer impacted by the therapy (Fig. 2 f). Importantly, INK128 was still able to efficiently downregulate ATF4 (Fig. 2 f), consistent with the spatial transcriptomic data suggesting that dual blockade of MAPK and mTOR most efficiently decreases ATF4 expression (Supplementary Fig. 3h). We next formally tested whether ATF4 is required for the survival of tumor cells that persist following belvarafenib plus cobimetinib therapy. ATF4 knockdown using short interfering RNA (siRNA) did not compromise cell survival relative to those treated with non-targeting siRNA ( siCtrl ) (Fig. 2 g,h). However, ATF4 -depleted cells were more sensitive to belvarafenib plus cobimetinib induced apoptosis, as shown by Annexin V-PI staining (Fig. 2 g) and western blotting for cleaved PARP (Fig. 2 h). Together, these data revealed an ATF4-dependent mechanism of tolerance in the MAPKi-induced melanoma persister cells. We suggest then that blockade of mTOR enhances the efficacy of MAPKi by synergistically potentiating and sustaining ATF4 suppression. To better understand the role of ATF4 in modulating the response of melanoma cells to the triple therapy, we further analyzed our transcriptomic data and identified the “DNA damage response” and “double-strand break repair” pathways, as being downregulated in the triple therapy-treated cells (Fig. 2 a and Supplementary Fig. 2e). Given that impaired DNA double-strand break repair can lead to excessive DNA damage and increased apoptosis 43 , we next examined markers of DNA damage, with a primary focus of γH2AX, a well-established read-out of DNA double-strand breaks (DSBs) 44 . Belvarafenib plus cobimetinib consistently induced γH2AX, with INK128 further enhancing this effect (Fig. 2 i,k). Importantly, knockdown of ATF4 also resulted in enhanced γH2AX formation in cells treated with combined belvarafenib plus cobimetinib (Fig. 2 j), phenocopying the effect observed with the mTOR inhibitor (Fig. 2 i,k). Altogether, our data suggest that mTOR inhibition potentiates belvarafenib plus cobimetinib-induced DNA damage through a mechanism downstream of ATF4. The triple therapy suppresses MTHFD2 downstream of ATF4. As an important transcription factor of ISR signaling, ATF4 target genes have been previously documented, wherein each target is evaluated and classified based on high, medium, or low confidence 33 . We interrogated our RNA sequencing data for the expression of 37 high confidence ATF4 target genes and found that several were downregulated by the triple therapy (Fig. 3 a and Supplementary Table 1). We next explored the clinical relevance of these 37 ATF4 target genes by correlating their expression with ATF4 expression, using human melanoma data from The Cancer Genome Atlas (TCGA) (Fig. 3 b). Among the candidate genes of interest, methylenetetrahydrofolate dehydrogenase 2 (MTHFD2) was particularly noteworthy. MTHFD2 is an enzyme that functions in mitochondrial one-carbon metabolism in the tetrahydrofolate (THF) cycle, which is essential for nucleotide synthesis and helps provide building blocks for subsequent DNA repair 28 , making it an attractive candidate for subsequent evaluation. MTHFD2 was repressed by the triple therapy in our RNAseq data and its expression correlated with ATF4 expression in the TCGA dataset (Fig. 3 a,c). MTHFD2 mRNA and protein were decreased in the triple therapy-treated cells (Fig. 3 d). Further supporting ATF4 as a regulator of MTHFD2 expression, knockdown of ATF4 decreased MTHFD2 protein level, while MTHFD1, which is independent of ATF4, remained unaffected (Fig. 3 e). Silencing of MTHFD2 in belvarafenib and cobimetinib combination treated melanoma cells phenocopies the triple therapy by inhibiting purine biosynthesis and increasing DNA damage. Given the essential role of MTHFD2 in nucleotide synthesis and DNA repair, we next sought to explore the link between the ATF4-MTHFD2 axis and the triple therapy-induced DNA damage. Similarly to the data obtained in ATF4 -silenced melanoma cells, MTHFD2 knockdown sensitized melanoma cells to belvarafenib plus cobimetinib treatment with a significantly increased level of DNA damage (Fig. 4 a-c). THF cycle produces one-carbon formyl groups for various cellular processes, including de novo purine synthesis. MTHFD2 in THF cycle can oxidize CH2-THF to 10-formyl THF, which participates in multiple steps in purine synthesis 45 – 47 . Gotoh et al. found that the amount of purine nucleotides is greatly reduced in MTHFD2 knockdown lung cancer cells 48 . Additionally, Zhou et al. showed that impairing nucleotide biosynthesis results in an insufficient supply of metabolites for DNA repair 49 . We thus measured whether an exogenous supply of nucleosides would rescue the phenotypes observed in MTHFD2 -depleted melanoma cells treated with belvarafenib plus cobimetinib. As expected, nucleoside supplementation, which bypasses the need for MTHFD2 in nucleotide biosynthesis (Fig. 4 d), protected MTHFD2 -depleted cells from the apoptosis induced by belvarafenib plus cobimetinib (Fig. 4 a), and significantly reduced γH2AX levels in these conditions (Fig. 4 b,c). Similarly, the addition of nucleosides also attenuated the DNA damage observed in belvarafenib plus cobimetinib-treated cells and triple therapy-treated cells (Fig. 4 e), resulting in an overall reduction of apoptosis following these treatments (Fig. 4 f). Together, these data suggest that the triple therapy induces DNA damage and apoptosis in melanoma cells via blocking the ATF4-MTHFD2 axis-mediated DNA repair pathways. The triple therapy decreases tumor outgrowth and improves survival of NRAS -mutant melanoma bearing mice in vivo . To evaluate the triple therapy in vivo , we used the NRAS Q 61 K -mutant MaNRAS1007 murine melanoma model 34 . Tumor bearing mice were randomized into 4 treatment arms: vehicle, INK128 (0.5 mg/kg/daily), belvarafenib (5 mg/kg/daily) plus cobimetinib (5 mg/kg/daily), and the triple therapy (Fig. 5 a). We have tailored the doses of each drug to ensure on-target engagement and optimize therapeutic efficacy, while minimizing potential toxicity. Melanoma growth was substantially inhibited in the INK128 monotherapy, and as expected, the tumors responded to belvarafenib plus cobimetinib 11 (Fig. 5 b). However, the tumors gradually progressed at approximately day 40 of the belvarafenib plus cobimetinib treatment, suggesting the emergence of drug resistance (Fig. 5 b). Notably, tumor control was significantly extended in the mice administered the triple therapy, compared with those that received INK128 monotherapy or the belvarafenib plus cobimetinib dual therapy. Importantly, the triple therapy cohort showed significantly improved overall survival (OS), compared with any other treatment groups (Fig. 5 c). Body weight was monitored throughout treatment, and no significant weight loss was observed, indicating good tolerability of the therapy (Fig. 5 d). These data suggest that mTOR inhibition may eliminate a substantial proportion of melanoma cells that resist combined belvarafenib plus cobimetinib (Fig. 5 b and Supplementary Fig. 4). The in vivo on-target activity of the therapies was confirmed by immunohistochemistry (IHC) staining after 14 days on treatment. As expected, p-ERK levels were inhibited in with the tumors from the belvarafenib plus cobimetinib cohort, and p-S6 levels were repressed in mice treated with INK128 (Fig. 5 e). There were fewer Ki67 positive stained cells in the melanomas treated with the triple therapy (Fig. 5 e), suggesting an inhibition of melanoma proliferation. Finally, we asked whether ATF4-MTHFD2 pathway would be inhibited by the triple therapy in vivo . We found that nuclear ATF4 expression and the expression of MTHFD2 were discernibly lower in the triple therapy group, suggesting impaired ATF4-MTHFD2 signaling in vivo (Fig. 5 e). mTOR inhibition can overcome the resistance to the combination therapy of belvarafenib and cobimetinib in NRAS -mutant melanoma. Finally, we tested whether blocking mTOR would be of clinical benefit once therapeutic resistance was acquired using cell lines derived from pre-clinical mouse models (Fig. 6 a). For this, mice bearing MaNRAS1007 murine melanomas or WM3406 xenografts were treated with belvarafenib plus cobimetinib; after 35 days, the tumors began to progress, indicating acquired resistance. To confirm resistance to belvarafenib plus cobimetinib, we dissociated the parental and dual drug-resistant melanomas (i.e. derived from MaNRAS1007 or WM3406 melanomas) and tested them in vitro . As expected, the resistant cell lines did not respond to belvarafenib and cobimetinib (Fig. 6 b,d). However, treatment with INK128 significantly induced apoptosis in these resistant cell lines (Fig. 6 c,e), supporting its potential as a salvage therapy. Therefore, we next treated these belvarafenib plus cobimetinib-resistant melanomas with INK128, while keeping the mice on belvarafenib plus cobimetinib. The addition of INK128 inhibited tumor outgrowth for both the MaNRAS1007 melanoma and WM3406 xenograft, which were resistant to belvarafenib plus cobimetinib (Fig. 6 f,h). Consistently, melanomas that were resistant to belvarafenib plus cobimetinib expressed higher ATF4 and MTHFD2. INK128 reduced the levels of ATF4 and MTHFD2 in these belvarafenib plus cobimetinib resistant tumors (Fig. 6 g,i). Taken together, our results provide a rationale to inhibit mTOR activity to overcome acquired resistance to combined belvarafenib plus cobimetinib. Discussion In this present study, we showed that mTOR inhibition can augment therapeutic responses to combined belvarafenib and cobimetinib in cell culture models and pre-clinical mouse models of melanoma. Importantly, inhibiting mTOR demonstrated potent antitumor effects in NRAS mutant melanomas with acquired resistance to belvarafenib and cobimetinib (Fig. 6). While our study supports the established crosstalk between the MAPK and mTOR pathways in regulating ATF4 activity 15,22,50,51 , it also uncovers a novel therapeutic mechanism: combining RAF dimer inhibitors with MEK and mTOR inhibitors effectively suppresses the ATF4-MTHFD2 axis, disrupting purine synthesis and impairing DNA repair. We provide evidence that this multi-targeted approach offers a powerful strategy to overcome resistance and enhance antitumor efficacy in MAPK-driven cancers. Our findings are aligned with prior work demonstrating that combined belvarafenib and cobimetinib therapy exhibits clinical activity in patients with NRAS -mutant melanomas (NCT03284502), with an overall response rate of 38.5% 11,12 . INK128 has also reached clinical testing and is well tolerated (NCT02412722, NCT02197572), thus providing an opportunity to assess whether mTOR inhibition improves the efficacy of combined belvarafenib and cobimetinib. Notably, in NRAS -mutant melanomas treated with belvarafenib plus cobimetinib, acquired resistance can develop, which we have shown is repressed by the blockade of mTOR (Fig. 6). Therefore, INK128 may be tested first in patients who have progressed on combination belvarafenib plus cobimetinib therapy. ATF4 protein is tightly regulated during the ISR, which specifically enhances the translation of ATF4 and other mRNAs containing upstream open reading frames (uORFs) 33 . Oncogenic MAPK signaling induces the expression of ATF4, which plays a central role in stress response and cell survival 32 . In NRAS and NF1 -mutant melanomas, we demonstrate that ATF4 expression is partially repressed following the blockade of MAPK signaling by belvarafenib plus cobimetinib (Fig. 2c), adding an important perspective to an earlier study 40 . Moreover, we showed that inhibiting mTOR in the context of belvarafenib plus cobimetinib further suppressed ATF4 (Fig. 2c), which is consistent with prior work showing that the mTOR pathway is required for the MAPK-mediated ATF4 induction 32 . ATF4 has been shown to have a multifaceted role in cancer, exhibiting both pro-survival and pro-apoptotic effects. Our work showed that both drug persistent melanoma cells in vitro and resistant melanomas in vivo maintain ATF4 expression, which can be impaired upon inhibiting mTOR to overcome therapeutic resistance (Fig. 2f and Fig. 6c,f). These results indicate the involvement of the ATF4 pathway in belvarafenib plus cobimetinib persister cells, consistent with a pro-survival role. Indeed, activation of the ATF4 pathway represents an evolutionarily conserved general stress response that may function in adaptation to MAPK inhibitors 23 . However, Niessner et al. reported that encorafenib plus binimetinib, targeting MAPK signaling, increased the expression of ATF4 in NRAS -mutant melanoma, which played a pro-apoptotic role due to a set of ATF4 target genes mediated by endoplasmic reticulum (ER) stress 52 . BRAF inhibitor vemurafenib or dabrafenib can also induce ATF4 expression transiently (within 4 hours) in BRAF -mutant melanoma 32,53 . Given the importance of ATF4 in drug persistence and resistance, the widespread role of ATF4 target genes in mitochondria function, autophagy, pro-apoptotic pathway and tRNA biosynthesis should not be neglected 33 . Indeed, our study shows that the expression of a subset of ATF4 target genes decrease after the treatment of belvarafenib plus cobimetinib (Fig. 3a). In support of our findings, ATF4 expression and MTHFD2 expression are positively correlated in TCGA SKCM (Skin Cutaneous Melanoma) cohort (Fig. 3c) and there is strong evidence from human and murine studies indicating that ATF4 regulates MTHFD2 transcription 54-57 . Herein, our data in melanoma models support that MTHFD2 is a target of ATF4. Moreover, activated mTOR leads to upregulated MTHFD2 and higher activity of purine synthesis, while knockdown of ATF4 decreases MTHFD2, which further indicates that, as a metabolic effector of mTOR 50 , ATF4 is required for the induction MTHFD2-mediated purine synthesis 58 . Although previous studies have shown that targeting mTOR or ATF4 downregulates MTHFD2 22,50 , our study provides the first evidence that MTHFD2 regulates purine synthesis and DNA damage repair, thereby modulating the response to MAPK-targeted therapy in non- BRAF -mutant melanoma. Our finding of the upregulation of ATF4 and MTHFD2 in melanomas that are resistant to combined belvarafenib and cobimetinib therapy raises the question of how ATF4 and MTHFD2 impact drug resistance. MTHFD2 catalyzes the oxidation of CH2-THF to CHO-THF, a critical reaction required for several key steps in purine synthesis, while producing free nucleotides as building blocks for subsequent DNA synthesis and repair 59,60 . A previous study showed that overexpressing MTHFD2 results in MAPK signaling activation in bladder cancer 61 . Reactivated MAPK signaling is a hallmark of targeted therapy resistance in melanoma cells, indicating a potential link between MTHFD2 and drug resistance. In support of these speculations, MTHFD2 is found to be critical for lung adenocarcinoma that is resistant to EGFR inhibitor, likely through increasing purine nucleotide 48 . MTHFD2 knockdown reduces folate-mediated one-carbon metabolism and purine synthesis in lung cancer cells, increasing the sensitivity to EGFR inhibitor gefitinib 48 . In addition, it is widely accepted that impairing purine synthesis causes DNA repair deficiency, which contributes to the efficacy of radiation therapy in glioblastoma 49 . Here, we show that inhibiting mTOR increases DNA damage in combined belvarafenib and cobimetinib therapy (Fig. 2k), and nucleoside supplementation decreases γH2AX (Fig. 4e), indicating the reduction of the accumulation of DNA damage. Knocking down MTHFD2 promotes DNA damage induced by belvarafenib plus cobimetinib and makes melanoma cells more prone to response to belvarafenib and cobimetinib combination therapy, while the supplementation of nucleoside reduces DNA damage and prevents these MTHFD2 -depleted cells from being sensitized to belvarafenib plus cobimetinib (Fig. a-c). These results are consistent with recent work identifying that the supplementation of nucleoside promotes DNA repair in glioblastoma, where inhibiting GTP synthesis impairs DNA repair and sensitizes resistant glioblastoma cells to radiotherapy 49,62 . Combined belvarafenib plus cobimetinib therapy has demonstrated efficacy in patients with Class 2 and 3 BRAF -mutant or NRAS -mutant melanoma 11,12,63 . In our preclinical models, co-targeting MAPK and mTOR pathway induced apoptosis and suppressed the growth of NRAS -mutant, NF1 -mutant, and KIT -mutant melanoma (Fig. 1). These data suggested that our proposed triple therapy could potentially benefit patients with not only NRAS- , but also NF1- , and KIT- mutant melanoma. Additionally, while BRAF V600 -mutant melanomas are sensitive to combined BRAF inhibitor and MEK inhibitor therapy, acquired resistance invariably occurs, often resulting from the transactivation of MAPK signaling induced by RAF dimerization 7 . Since belvarafenib in our triple therapy is a RAF dimer inhibitor, triple therapy could potentially benefit patients with drug resistant class 1 BRAF -mutant melanomas. In summary, we provide evidence that pharmacologic inhibition of mTOR shows efficacy in preclinical melanoma models when combined with belvarafenib and cobimetinib, via a mechanism involving the blockade of ATF4-MTHFD2 axis-regulated purine synthesis, which results in DNA damage and apoptosis. Methods Mice Male C57BL/6N mice (6-8 weeks old) were purchased from Charles River Laboratories. Female nonobese diabetic (NOD)/severe combined immunodeficiency (SCID) mice (6-10 weeks old) were kindly gifted by Dr. Moulay Alaoui-Jamali. All mice were randomized before injection. For melanoma cell inoculation, MaNRAS1007 cells were injected to male C57BL/6N mice at 1,000,000 cells/mouse. WM3406 cells were injected to female NOD/SCID mice at 1,000,000 cells/mouse. All melanoma cell lines were freshly prepared in PBS and subcutaneously injected to the right flank of mice. Once palpable tumors were formed, tumors were measured in length (L) and width (W). Tumor volumes (V) were calculated based on the formula V=3.1416/6*L*W 2 . For belvarafenib (5 mg/kg, MedChem Express, HY-109080), cobimetinib (5mg/kg, ChemieTek CT-G0973) and INK128 (0.5 mg/kg, MedChem Express, HY-13328) treatment, these drugs were dissolved in DMSO and subsequently diluted in PEG400 (Sigma-Aldrich, P3265), TWEEN® 20 (Sigma-Aldrich, P1379) and PBS. Mice were administered every day by gavage, starting when the tumor volume reached approximately 100 mm 3 . For tumor growth curve, mice were sacrificed when the tumor volume reached approximately 1500 mm 3 . For IHC in Fig. 6, all mice were sacrificed on Day 20 after treatment initiation. The IHC on relapsed tumors were performed when the tumor volume reached 1000 mm 3 . Cells and reagents Sources, culture conditions and treatment timelines of human melanoma cell lines are listed in Supplementary Table 2. All the drug treatments began the next day after cell seeding. To inhibit the purine synthesis, melanoma cells were treated with MTHFD inhibitor LY345899 (10 μM, MedChem Express, HY-101943), GARFT inhibitor lometrexol (10 μM, MedChem Express, HY-14521), DHODH inhibitor brequinar (10 μM, MedChem Express, HY-108325). For nucleoside supplementation, nucleosides (100X, Sigma-Aldrich, ES-008) with 0.73 g/L cytidine, 0.85 g/L guanosine, 0.73 g/L uridine, 0.8 g/L adenosine and 0.24 g/L thymidine was added 1:100 in cell culture media. Cells were collected for apoptosis assay and Western blot after 48h treatment. Colony formation assay Melanoma cells were seeded into 6-well plates at 2,000 cells per well. The next day, belvarafenib, cobimetinib and INK128 (see Supplementary Table 2), or DMSO control were added to the indicated wells. Cells were subsequently cultured for 10 days, during which media was changed and drugs were freshly added every two days. At the end of the assay, cells were fixed with 4% formaldehyde/PBS, stained with 0.5% of crystal violet (Sigma-Aldrich, HT90132) diluted in 70% EtOH and photographed. Colonies were manually quantified using FIJI software (https://github.com/fiji/fiji) to assess cell survival and proliferation. Western blotting Immunoblots were performed as previously described 64 . Briefly, cells were lysed with RIPA buffer (150mmol/L Tris-HCl, pH=7, 150mmol/L NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitors (Roche). Equal amounts of protein samples were loaded, separated on 10% or 12% SDS-PAGE gels, transferred to nitrocellulose membranes, and probed with corresponding antibodies. Detailed antibody information is listed in Supplementary Table 3. Flow cytometry-based assays Melanoma cells were seeded into 6-well plates at 100,000 cells per well. Following indicated treatments, cells were trypsinized, centrifuged at 240g for 5 min, and washed twice in PBS. For apoptosis detection of non-fixed cells, Alexa Fluor™ 647-Annexin V (Invitrogen™, A23204) and Propidium Iodide (PI) Staining Solution (BD Biosciences, 556463) were diluted in 1×binding buffer (BD Biosciences, 556454), and subsequently mixed with cells following the manufacturer’s instructions. All flow cytometry experiments were conducted on the FACSCanto (BD Biosciences). RNA sequencing analyses Sequencing libraries were generated using NEBNext® UltraTM RNA Library Prep Kit for Illumina® (NEB, USA) following manufacturer’s recommendations. The library was sequenced using the Illumina NovaSeq 6000 sequencing platform to generate raw reads. Then, nf-core/rnaseq pipeline v3.8.1 65 was used to perform quality control, trimming and alignment on raw paired-end fastq reads, followed by reference genome-guided transcriptome assembly and gene expression quantification. Differentially expressed genes (DEGs) were identified by DESeq2 66 with cut-off values of a log2|fold-change| > 1 and a p-adjust < 0.05. ClusterProfiler 67 was used to perform functional enrichment analysis and the potential genes in the identified modules were analyzed based on gene ontology (GO) categories. RNA interference siRNAs were transfected into cells using Lipofectamine™ RNAiMAX Transfection Reagent (Invitrogen, 13778) following the manufacturer’s instructions. Media were changed the next day after cells were incubated with siRNAs for 18 hours. All cells were harvested between 48h-96h after siRNA transfection. All siRNA sequences are listed in Supplementary Table 4. Immunofluorescence microscopy WM3406 and MeWo cells were grown on glass coverslips and fixed with 4% (w/v) paraformaldehyde (PFA) in PBS for 15 min at room temperature. PFA-fixed cells were permeabilized with 0.2% (v/v) Triton X-100 in PBS for 20 min at room temperature. Cells were then incubated with blocking buffer (5% bovine serum albumin, and 0.5% Triton X-100 in PBS) for 1 hour at room temperature and then incubated with the γH2AX primary antibodies overnight at 4 °C. After three washes with PBS, cells were incubated for 1 hour at room temperature with the secondary antibodies (Invitrogen, A21206). Cells were then washed with PBS, followed by incubation with 1 μg/ml DAPI nuclear stain for 15 minutes. Cells were washed and mounted to a glass slide using ProLong gold mounting media (Invitrogen, P36930). Slides were stored in the dark until fluorescence images were taken. Stacking and coloring of images was performed using FIJI software. Each quantification was done on at least three biological replicates and at least 3 ROIs per biological replicate. Quantitative real-time PCR Cultured cells were pelleted, and RNA was prepared using the E.Z.N.A. total RNA isolation kit (OMEGA Bio-Tek). RNA concentrations were then quantified using a NanoDrop spectrophotometer (ThermoFisher Scientific) and cDNA was prepared from 1mg of total RNA using iScript cDNA Synthesis Kit (Bio-Rad). Target genes were quantified using the Applied Biosystems 7500 Fast Real-Time PCR System with SYBR Green real-time PCR master mix (Applied Biosystems). Two housekeeping genes were used for each assay. Primers used for qPCR are listed in Supplementary Table 5. Immunohistochemistry (IHC) Staining of mouse samples was performed as previously described 64 . Briefly, formalin-fixed, paraffin-embedded tumor sections were stained with indicated antibodies (Supplementary Table 3), followed by a standard magenta red detection protocol 68 (Agilent Technologies, GV92511-2). Hematoxylin-counterstained slides were mounted with coverslips. Slides were scanned on AxioScan (ZEISS) and positive staining quantified using QuPath v0.5.1. Access and re-analysis of previously published datasets Processed count matrices of spatial transcriptomics (Visium) from NRAS Q61K/° ;Ink4a −/− melanoma mouse model were downloaded from the website of Dr. Jean-Christophe Marine lab (https://marinelab.sites.vib.be/en) 37 . In brief, spots were retained when nFeature_Spatial > 1,000 and percent.mt < 5 and expression data were normalized using SCTransform (Seurat, v.5.0.2) 69 . To determine the MAPK pathway activity, we used the MPAS score based on the MAPK signaling characteristic gene signatures as previously described 38 . To determine the mTOR pathway activity, we used Gene Set Variation Analysis (GSVA) 70 to perform functional enrichment analysis based on HALLMARK categories. The ISR score was also calculated based on the Z-score of stress response regulators 33 . Tumor dissociation and culture MaNRAS1007 and WM3406 melanomas in control group or belvarafenib plus cobimetinib resistant group were resected. Tumors were minced and digested in collagenase A to obtain single-cell suspension. MaNRAS1007 cells were cultured in Ham’s F12 containing 1% FBS, while WM3406 cells were cultured in RPMI containing 1% FBS and 1 x GlutaMAX. Statistics In vitro data are presented as mean ± SD. In vivo data are presented as mean ± SEM. Prism software (GraphPad) was used to determine statistical significance of differences. Figure legends specify the statistical analysis used. P values are indicated in the figures and p < 0.05 were considered significant. Study approval Animal experiments were conducted according to the regulations established by the Canadian Council of Animal Care, and protocols approved by McGill University Animal Care and Use Committee (#2015-7672). Declarations Acknowledgments This research was funded by the Canadian Institutes of Health Research (CIHR) (grants PJT-180379 to AANR, PJT-162260, PJT-178194 to SVDR and grant PJT-183934 to WHM). FC and JM were endowed by Fonds de recherche du Québec - Santé doctoral scholarship. FJ, SVDR and WHM are supported by “Wallonie-Bruxelles International” and “Relations Internationales et Francophonie, Québec” to encourage active collaboration. This research as well as biobanking of biological material and data was made possible through a collaboration with the Réseau de recherche sur le cancer (RRCancer) financially supported by the the Oncopole, the FRQ cancer division, which receives funding from Merck Canada Inc., GSK, Pfizer and the Ministère de l'Économie, de l'Innovation et de l'Énergie du Québec. The RRCancer is affiliated to the Canadian Tumor Repository Network (CTRNet). We thank Madelyn Abraham for technical supports on DNA damage detection. We thank David Papadopoli for critical reading of the manuscript. We thank Christian Young, Mathew Duguay, and Darleen Element for experimental advice and technical supports. Author contributions: FC, FH, SVDR and WHM contributed conceptualization and investigation. FC, CG, HS, NG, EG, LMH, JM, JS, KB, AO, SVDR and WHM contributed methodology. FC, FH, CG, AO, SVDR and WHM contributed visualization. AANR, SVDR and WHM contributed funding acquisition. FC, SVDR and WHM contributed project administration and writing the original draft. SVDR and WHM contributed supervision. FC, FH, JM, FJ, AANR, AO, SVDR and WHM contributed review & Editing. Competing interests: The authors have declared that no conflict of interest exists. References Prabhu, S. A., Moussa, O., Miller, W. H., Jr. & Del Rincon, S. V. The MNK1/2-eIF4E Axis as a Potential Therapeutic Target in Melanoma. Int J Mol Sci 21 (2020). Cancer Genome Atlas, N. Genomic Classification of Cutaneous Melanoma. Cell 161, 1681-1696 (2015). Heidorn, S. J. et al. Kinase-dead BRAF and oncogenic RAS cooperate to drive tumor progression through CRAF. Cell 140, 209-221 (2010). Karoulia, Z., Gavathiotis, E. & Poulikakos, P. I. New perspectives for targeting RAF kinase in human cancer. Nat Rev Cancer 17, 676-691 (2017). Ascierto, P. A. et al. Contribution of MEK Inhibition to BRAF/MEK Inhibitor Combination Treatment of BRAF-Mutant Melanoma: Part 2 of the Randomized, Open-Label, Phase III COLUMBUS Trial. J Clin Oncol 41, 4621-4631 (2023). Hatzivassiliou, G. et al. RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth. Nature 464, 431-435 (2010). Poulikakos, P. I., Zhang, C., Bollag, G., Shokat, K. M. & Rosen, N. RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF. Nature 464, 427-430 (2010). Peng, S. B. et al. Inhibition of RAF Isoforms and Active Dimers by LY3009120 Leads to Anti-tumor Activities in RAS or BRAF Mutant Cancers. Cancer Cell 28, 384-398 (2015). Dankner, M. et al. Dual MAPK Inhibition Is an Effective Therapeutic Strategy for a Subset of Class II BRAF Mutant Melanomas. Clin Cancer Res 24, 6483-6494 (2018). Randic, T., Kozar, I., Margue, C., Utikal, J. & Kreis, S. NRAS mutant melanoma: Towards better therapies. Cancer Treat Rev 99, 102238 (2021). Yen, I. et al. ARAF mutations confer resistance to the RAF inhibitor belvarafenib in melanoma. Nature 594, 418-423 (2021). Shin, S. J. et al. A phase Ib trial of belvarafenib in combination with cobimetinib in patients with advanced solid tumors: Interim results of dose-escalation and patients with NRAS-mutant melanoma of dose-expansion. Journal of Clinical Oncology 39, 3007-3007 (2021). de Braud, F. et al. Initial Evidence for the Efficacy of Naporafenib in Combination With Trametinib in NRAS-Mutant Melanoma: Results From the Expansion Arm of a Phase Ib, Open-Label Study. J Clin Oncol 41, 2651-2660 (2023). Bivona, T. G. & Doebele, R. C. A framework for understanding and targeting residual disease in oncogene-driven solid cancers. Nat Med 22, 472-478 (2016). Wang, B. et al. Targeting mTOR signaling overcomes acquired resistance to combined BRAF and MEK inhibition in BRAF-mutant melanoma. Oncogene 40, 5590-5599 (2021). Shi, H. et al. A novel AKT1 mutant amplifies an adaptive melanoma response to BRAF inhibition. Cancer Discov 4, 69-79 (2014). Villanueva, J. et al. Acquired resistance to BRAF inhibitors mediated by a RAF kinase switch in melanoma can be overcome by cotargeting MEK and IGF-1R/PI3K. Cancer Cell 18, 683-695 (2010). Obenauf, A. C. et al. Therapy-induced tumour secretomes promote resistance and tumour progression. Nature 520, 368-372 (2015). Sen, S., Meric-Bernstam, F., Hong, D. S., Hess, K. R. & Subbiah, V. Co-occurring Genomic Alterations and Association With Progression-Free Survival in BRAFV600-Mutated Nonmelanoma Tumors. J Natl Cancer Inst 109 (2017). Ye, L. et al. Transmembrane-4 L-six family member-1 (TM4SF1) promotes non-small cell lung cancer proliferation, invasion and chemo-resistance through regulating the DDR1/Akt/ERK-mTOR axis. Respir Res 20, 106 (2019). Pakos-Zebrucka, K. et al. The integrated stress response. EMBO Rep 17, 1374-1395 (2016). Park, Y., Reyna-Neyra, A., Philippe, L. & Thoreen, C. C. mTORC1 Balances Cellular Amino Acid Supply with Demand for Protein Synthesis through Post-transcriptional Control of ATF4. Cell Rep 19, 1083-1090 (2017). Yang, C., Tian, C., Hoffman, T. E., Jacobsen, N. K. & Spencer, S. L. Melanoma subpopulations that rapidly escape MAPK pathway inhibition incur DNA damage and rely on stress signalling. Nat Commun 12, 1747 (2021). Hsieh, A. C. et al. The translational landscape of mTOR signalling steers cancer initiation and metastasis. Nature 485, 55-61 (2012). Moore, K. N. et al. Phase I study of the investigational oral mTORC1/2 inhibitor sapanisertib (TAK-228): tolerability and food effects of a milled formulation in patients with advanced solid tumours. ESMO Open 3, e000291 (2018). Patel, C. et al. Phase 1 Study to Evaluate the Effect of the Investigational Anticancer Agent Sapanisertib on the QTc Interval in Patients With Advanced Solid Tumors. Clin Pharmacol Drug Dev 9, 876-888 (2020). Wan, X., Harkavy, B., Shen, N., Grohar, P. & Helman, L. J. Rapamycin induces feedback activation of Akt signaling through an IGF-1R-dependent mechanism. Oncogene 26, 1932-1940 (2007). Yue, L. et al. Mthfd2 Modulates Mitochondrial Function and DNA Repair to Maintain the Pluripotency of Mouse Stem Cells. Stem Cell Reports 15, 529-545 (2020). Perurena, N., Situ, L. & Cichowski, K. Combinatorial strategies to target RAS-driven cancers. Nat Rev Cancer 24, 316-337 (2024). Krauthammer, M. et al. Exome sequencing identifies recurrent mutations in NF1 and RASopathy genes in sun-exposed melanomas. Nat Genet 47, 996-1002 (2015). Janes, M. R. et al. Efficacy of the investigational mTOR kinase inhibitor MLN0128/INK128 in models of B-cell acute lymphoblastic leukemia. Leukemia 27, 586-594 (2013). Nagasawa, I. et al. Disrupting ATF4 Expression Mechanisms Provides an Effective Strategy for BRAF-Targeted Melanoma Therapy. iScience 23, 101028 (2020). Neill, G. & Masson, G. R. A stay of execution: ATF4 regulation and potential outcomes for the integrated stress response. Front Mol Neurosci 16, 1112253 (2023). Petit, V. et al. C57BL/6 congenic mouse NRAS(Q61K) melanoma cell lines are highly sensitive to the combination of Mek and Akt inhibitors in vitro and in vivo. Pigment Cell Melanoma Res 32, 829-841 (2019). Ackermann, J. et al. Metastasizing melanoma formation caused by expression of activated N-RasQ61K on an INK4a-deficient background. Cancer Res 65, 4005-4011 (2005). Arozarena, I. & Wellbrock, C. Phenotype plasticity as enabler of melanoma progression and therapy resistance. Nat Rev Cancer 19, 377-391 (2019). Karras, P. et al. A cellular hierarchy in melanoma uncouples growth and metastasis. Nature 610, 190-198 (2022). Wagle, M. C. et al. A transcriptional MAPK Pathway Activity Score (MPAS) is a clinically relevant biomarker in multiple cancer types. NPJ Precis Oncol 2, 7 (2018). Sigaud, R. et al. MAPK inhibitor sensitivity scores predict sensitivity driven by the immune infiltration in pediatric low-grade gliomas. Nat Commun 14, 4533 (2023). Pathria, G. et al. Translational reprogramming marks adaptation to asparagine restriction in cancer. Nat Cell Biol 21, 1590-1603 (2019). Simanshu, D. K., Nissley, D. V. & McCormick, F. RAS Proteins and Their Regulators in Human Disease. Cell 170, 17-33 (2017). Marin-Bejar, O. et al. Evolutionary predictability of genetic versus nongenetic resistance to anticancer drugs in melanoma. Cancer Cell 39, 1135-1149 e1138 (2021). Roos, W. P. & Kaina, B. DNA damage-induced cell death: from specific DNA lesions to the DNA damage response and apoptosis. Cancer Lett 332, 237-248 (2013). Rogakou, E. P., Pilch, D. R., Orr, A. H., Ivanova, V. S. & Bonner, W. M. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem 273, 5858-5868 (1998). Fan, J. et al. Quantitative flux analysis reveals folate-dependent NADPH production. Nature 510, 298-302 (2014). Locasale, J. W. Serine, glycine and one-carbon units: cancer metabolism in full circle. Nat Rev Cancer 13, 572-583 (2013). Tibbetts, A. S. & Appling, D. R. Compartmentalization of Mammalian folate-mediated one-carbon metabolism. Annu Rev Nutr 30, 57-81 (2010). Nishimura, T. et al. Cancer stem-like properties and gefitinib resistance are dependent on purine synthetic metabolism mediated by the mitochondrial enzyme MTHFD2. Oncogene 38, 2464-2481 (2019). Zhou, W. et al. Purine metabolism regulates DNA repair and therapy resistance in glioblastoma. Nat Commun 11, 3811 (2020). Torrence, M. E. et al. The mTORC1-mediated activation of ATF4 promotes protein and glutathione synthesis downstream of growth signals. Elife 10 (2021). Gao, Y., Feng, L., Zhang, L., Geng, J. & Zhang, E. ATF4/MYC Regulates MTHFD2 to Promote NSCLC Progression by Mediating Redox Homeostasis. Dis Markers 2022, 7527996 (2022). Niessner, H. et al. BRAF Inhibitors Amplify the Proapoptotic Activity of MEK Inhibitors by Inducing ER Stress in NRAS-Mutant Melanoma. Clin Cancer Res 23, 6203-6214 (2017). Nagasawa, I., Kunimasa, K., Tsukahara, S. & Tomida, A. BRAF-mutated cells activate GCN2-mediated integrated stress response as a cytoprotective mechanism in response to vemurafenib. Biochem Biophys Res Commun 482, 1491-1497 (2017). Wang, S. et al. ATF4 Gene Network Mediates Cellular Response to the Anticancer PAD Inhibitor YW3-56 in Triple-Negative Breast Cancer Cells. Mol Cancer Ther 14, 877-888 (2015). Ord, T., Ord, D., Kaikkonen, M. U. & Ord, T. Pharmacological or TRIB3-Mediated Suppression of ATF4 Transcriptional Activity Promotes Hepatoma Cell Resistance to Proteasome Inhibitor Bortezomib. Cancers (Basel) 13 (2021). Han, J. et al. ER-stress-induced transcriptional regulation increases protein synthesis leading to cell death. Nat Cell Biol 15, 481-490 (2013). Freundt, J. K. et al. The Transcription Factor ATF4 Promotes Expression of Cell Stress Genes and Cardiomyocyte Death in a Cellular Model of Atrial Fibrillation. Biomed Res Int 2018, 3694362 (2018). Ben-Sahra, I., Hoxhaj, G., Ricoult, S. J. H., Asara, J. M. & Manning, B. D. mTORC1 induces purine synthesis through control of the mitochondrial tetrahydrofolate cycle. Science 351, 728-733 (2016). Zhu, Z. & Leung, G. K. K. More Than a Metabolic Enzyme: MTHFD2 as a Novel Target for Anticancer Therapy? Front Oncol 10, 658 (2020). Pardo-Lorente, N. & Sdelci, S. MTHFD2 in healthy and cancer cells: Canonical and non-canonical functions. npj Metabolic Health and Disease 2, 3 (2024). Zhu, L. et al. MTHFD2 is a potential oncogene for its strong association with poor prognosis and high level of immune infiltrates in urothelial carcinomas of bladder. BMC Cancer 22, 556 (2022). Ramos, L., Henriksson, M., Helleday, T. & Green, A. C. Targeting MTHFD2 to Exploit Cancer-Specific Metabolism and the DNA Damage Response. Cancer Res 84, 9-16 (2024). Kim, K. H. et al. Exploring Molecular Genetic Alterations and RAF Fusions in Melanoma: A Belvarafenib Expanded Access Program in Patients with RAS/RAF-Mutant Melanoma. Oncologist 29, e811-e821 (2024). Huang, F. et al. Inhibiting the MNK1/2-eIF4E axis impairs melanoma phenotype switching and potentiates antitumor immune responses. J Clin Invest 131 (2021). Ewels, P. A. et al. The nf-core framework for community-curated bioinformatics pipelines. Nat Biotechnol 38, 276-278 (2020). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550 (2014). Yu, G., Wang, L. G., Han, Y. & He, Q. Y. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics 16, 284-287 (2012). Petersen, K. H., Lohse, J. & Ramsgaard, L. Automated sequential chromogenic IHC double staining with two HRP substrates. PLoS One 13, e0207867 (2018). Hao, Y. et al. Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat Biotechnol 42, 293-304 (2024). Hanzelmann, S., Castelo, R. & Guinney, J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14, 7 (2013). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryfilesCaietalncomms.docx Supplementary info nrreportingsummaryFLAT.pdf Reporting Summary 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-6857127","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":484907518,"identity":"5b66cb63-c5de-446a-a577-5b98dd9fec06","order_by":0,"name":"Sonia Del Rincon","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABO0lEQVRIie2RvWrDMBCAzwTsRSSrSoL1ChIBU0ihryIv9eJARk9NJnlx9rxEhy6ZZQT1IgrdPHRwKWTKkGwpZGjkn8FpU+hWir9FOuk+7k4C6Oj4o0gA3Gx60AdrYQK3PClmlxXUKHatjMsrTi+XQtVilAq/FL9RSLxM5Ye4JrfO8k0eYOLaQyWKQ/QaPMTZi6lCBouWQvUzT5cCswRlNE0gGNt9P2aJ3kzXOpwZha1kW8EhlUhjjvAdSATKF8gSQ0uo6TpH3Cgc2gpZhTQ91kp6BDVvlMBrFNJWIA+pQlGlqFMVbtcK93JHlgo9a0xrqkbRaRb9BGpEA2aqXCVasdMs5pExezxrLE7G+y29J04sevttNCED5GzwIVLEy7L33S66Ie5ZY23av4BMiH/K/4JT/Cq9o6Oj47/yCVpldz/zzUGZAAAAAElFTkSuQmCC","orcid":"","institution":"McGill University","correspondingAuthor":true,"prefix":"","firstName":"Sonia","middleName":"Del","lastName":"Rincon","suffix":""},{"id":484907519,"identity":"7ac03847-c88a-4042-abc0-b274260fae61","order_by":1,"name":"Feiyang Cai","email":"","orcid":"https://orcid.org/0000-0002-9977-4190","institution":"Division of Clinical and Translational Research, McGill University","correspondingAuthor":false,"prefix":"","firstName":"Feiyang","middleName":"","lastName":"Cai","suffix":""},{"id":484907520,"identity":"a8a0df6d-5da1-47fa-9e3c-32c3967d0e06","order_by":2,"name":"Fan Huang","email":"","orcid":"","institution":"Department of Experimental Medicine, Faculty of Medicine, McGill University","correspondingAuthor":false,"prefix":"","firstName":"Fan","middleName":"","lastName":"Huang","suffix":""},{"id":484907521,"identity":"208d4e1f-3d2b-4d78-a238-68cfcbcdb215","order_by":3,"name":"Christophe Goncalves","email":"","orcid":"","institution":"Lady Davis Institute","correspondingAuthor":false,"prefix":"","firstName":"Christophe","middleName":"","lastName":"Goncalves","suffix":""},{"id":484907522,"identity":"d6c70f51-d279-4c9f-810d-0703875205ec","order_by":4,"name":"Harinee Srikannan","email":"","orcid":"","institution":"Lady Davis Institute, McGill University","correspondingAuthor":false,"prefix":"","firstName":"Harinee","middleName":"","lastName":"Srikannan","suffix":""},{"id":484907523,"identity":"f95f1fc5-249b-4251-ae3c-370226ef6d49","order_by":5,"name":"Natascha Gagnon","email":"","orcid":"","institution":"Lady Davis Institute","correspondingAuthor":false,"prefix":"","firstName":"Natascha","middleName":"","lastName":"Gagnon","suffix":""},{"id":484907524,"identity":"4750247f-a86d-4ada-b9e6-c036f122da18","order_by":6,"name":"Elizabeth Guettler","email":"","orcid":"https://orcid.org/0000-0002-0176-9407","institution":"Lady Davis Institute, McGill University","correspondingAuthor":false,"prefix":"","firstName":"Elizabeth","middleName":"","lastName":"Guettler","suffix":""},{"id":484907525,"identity":"77eb2dd9-8097-46c2-9042-e38ec0e362d5","order_by":7,"name":"Léa Mukeba-Harchies","email":"","orcid":"","institution":"University of Mons","correspondingAuthor":false,"prefix":"","firstName":"Léa","middleName":"","lastName":"Mukeba-Harchies","suffix":""},{"id":484907526,"identity":"6b66767f-f71c-4604-b0f8-8e95dcb6b3a2","order_by":8,"name":"Jennifer Maxwell","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Jennifer","middleName":"","lastName":"Maxwell","suffix":""},{"id":484907527,"identity":"6055d55a-337e-4f9a-8c06-859d72e1e352","order_by":9,"name":"Jie Su","email":"","orcid":"","institution":"Lady Davis Institute, McGill University","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Su","suffix":""},{"id":484907528,"identity":"74250372-a487-45c0-9b4e-d6f6caf54aaf","order_by":10,"name":"Krikor Bijian","email":"","orcid":"","institution":"Lady Davis Institute, McGill University","correspondingAuthor":false,"prefix":"","firstName":"Krikor","middleName":"","lastName":"Bijian","suffix":""},{"id":484907529,"identity":"fc9fbe7e-2b23-4a63-ac37-e7619d5c699e","order_by":11,"name":"Fabrice Journe","email":"","orcid":"","institution":"UMONS","correspondingAuthor":false,"prefix":"","firstName":"Fabrice","middleName":"","lastName":"Journe","suffix":""},{"id":484907530,"identity":"a4ef89f8-3fae-4735-aca2-472306ded647","order_by":12,"name":"April Rose","email":"","orcid":"https://orcid.org/0000-0002-9845-4603","institution":"Jewish General Hospital","correspondingAuthor":false,"prefix":"","firstName":"April","middleName":"","lastName":"Rose","suffix":""},{"id":484907531,"identity":"24505555-9eb2-4dd0-85ba-1ee810e8f1c8","order_by":13,"name":"Alexandre Orthwein","email":"","orcid":"https://orcid.org/0000-0002-7350-3413","institution":"Emory University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Alexandre","middleName":"","lastName":"Orthwein","suffix":""},{"id":484907532,"identity":"3122401f-f387-46ae-836c-17b207b990da","order_by":14,"name":"Wilson Miller","email":"","orcid":"","institution":"Segal Cancer Centre and Lady Davis Institute for Medical Research","correspondingAuthor":false,"prefix":"","firstName":"Wilson","middleName":"","lastName":"Miller","suffix":""}],"badges":[],"createdAt":"2025-06-09 19:40:43","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6857127/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6857127/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87035102,"identity":"f854c76c-3fcc-43d9-98bf-f1af04c2a675","added_by":"auto","created_at":"2025-07-18 13:12:16","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1107017,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe inhibition of mTOR augments the anti-cancer effects of belvarafenib and cobimetinib combination therapy in melanoma cells.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eTop: Colony number of human melanoma cells following belvarafenib (25 nM), cobimetinib (50 nM) and INK128 (25 nM) treatment for 10 days. Bottom: Western blot analysis of the indicated proteins in human melanoma cells following the treatment of belvarafenib, cobimetinib and INK128 for 2 h (n=3). One-way ANOVA. \u003cstrong\u003eb\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eSchematic diagram showing the MAPK signaling and PI3K/AKT/mTOR signaling in melanoma cells.\u003cstrong\u003e c\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eTop: Percentage apoptotic cells as measured by the sum of PI/Annexin V double-positive and Annexin V-positive staining. Bottom: Western blot analysis of the indicated proteins in human melanoma cells. One-way ANOVA. All data represent mean ± SD.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/aa2087c08dd5e313330bf6f9.png"},{"id":87035106,"identity":"89d5ab44-562f-4b7a-9281-a00d5520042f","added_by":"auto","created_at":"2025-07-18 13:12:16","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1703348,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe triple therapy induces apoptosis of melanoma cells through inhibition of ATF4 and induction of DNA damage.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Dotplot depicting Gene Ontology (GO) enrichment of differential expressed genes (DEGs) in triple therapy treated cells, compared with DMSO treatment. Gene sets that were potentially related to the apoptosis phenotype were colored red. Fisher test. \u003cstrong\u003eb\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eGSEA for these DEGs in cellular response to stress. \u003cstrong\u003ec\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eTop: Relative quantification of ATF4 protein following indicated treatment in WM3406 cells (left) and MeWo cells (right) (n=3). One-way ANOVA. Bottom: Western blot analysis of the indicated proteins in WM3406 (left) and MeWo (right) cells following the treatment of INK128, belvarafenib plus cobimetinib, or the triple therapy. \u003cstrong\u003ed\u003c/strong\u003e, Left: Western blot analysis of ATF4 in mouse NRAS-mutant MaNRAS1007 melanoma cells following indicated treatment. Right: Relative quantification of ATF4 protein following indicated treatment in MaNRAS1007 cells (n=3). One-way ANOVA. \u003cstrong\u003ee\u003c/strong\u003e, Schematic of experimental design for \u003cstrong\u003ef\u003c/strong\u003e. \u003cstrong\u003ef\u003c/strong\u003e, Left: Western blot analysis of ATF4 in WM3406 (top) and MeWo (bottom) cells following belvarafenib plus cobimetinib treatment for 10 days and INK128 treatment for 2 days. Right: Relative quantification of ATF4 protein following indicated treatment in WM3406 cells and MeWo cells (n=3). One-way ANOVA. \u003cstrong\u003eg\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003ePercent apoptosis detected in DMSO- or belvarafenib plus cobimetinib-treated WM3406 (left) or MeWo (right) cells. ATF4 was knocked down prior to belvarafenib plus cobimetinib treatment (n=3). Two-way ANOVA. \u003cstrong\u003eh\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eWestern blot analysis of the indicated proteins in WM3406 (left) and MeWo (right) cells with ATF4 knockdown or belvarafenib plus cobimetinib treatment. \u003cstrong\u003ei\u003c/strong\u003e, Western blot analysis of γH2AX (phosphorylated H2AX at Ser 139) in WM3406 (left) and MeWo (right) cells following indicated treatment, or \u003cstrong\u003ej\u003c/strong\u003e with ATF4 knockdown and belvarafenib plus cobimetinib treatment. \u003cstrong\u003ek\u003c/strong\u003e, (Left) Representative images of WM3406 (top) and MeWo (bottom) cells following indicated treatment for 18 h and processed for γH2AX (green) immunofluorescence. Nuclei were counterstained with DAPI (scale bars: 50 μm). (Right) Quantification of γH2AX foci per cell (left) and γH2AX foci intensity (right) in WM3406 (top) and MeWo (bottom) cells following indicated treatment (n=3). 40 cells per condition were randomly quantified. One-way ANOVA. All data represent mean ± SD.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/8190bacd92ed2c52f74b348c.png"},{"id":87035108,"identity":"03296b1d-6da8-4f39-a6d4-65f04986ae5a","added_by":"auto","created_at":"2025-07-18 13:12:16","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":891986,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe triple therapy suppresses MTHFD2 downstream of ATF4.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eHeatmap showing the expression of ATF4 target genes in drug treated cells relative to DMSO treated cells. \u003cstrong\u003eb\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eThe plot showing the correlation between the expression of ATF4 targets genes and that of \u003cem\u003eATF4\u003c/em\u003eas assessed by Pearson correlation. Red dots indicate correlation coefficient larger than 0.3. \u003cstrong\u003ec\u003c/strong\u003e, The scatter plot showing the correlation between the expression of \u003cem\u003eMTHFD2\u003c/em\u003eand \u003cem\u003eATF4\u003c/em\u003e. Pearson correlation. \u003cstrong\u003ed\u003c/strong\u003e, Top: Fold change of \u003cem\u003eMTHFD2\u003c/em\u003e transcripts following indicated treatment in WM3406 cells (left) and MeWo cells (right) (n=3). RPLP0 was used as a reference gene. One-way ANOVA. Bottom: Western blot analysis of the indicated proteins in WM3406 (left) and MeWo (right) cells following the treatment of INK128, belvarafenib plus cobimetinib, or the triple therapy. \u003cstrong\u003ee\u003c/strong\u003e, Western blot analysis of the indicated proteins in WM3406 (left) and MeWo (right) cells with ATF4 knockdown or belvarafenib plus cobimetinib treatment. All data represent mean ± SD.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/470d743d325b8f6c4970a2ee.png"},{"id":87036287,"identity":"f6370956-0c46-4db3-afe5-ea7fdafcb231","added_by":"auto","created_at":"2025-07-18 13:20:16","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2329008,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe silencing of MTHFD2 sensitizes melanoma cells to belvarafenib and cobimetinib combination therapy by inhibiting purine biosynthesis and increasing DNA damage.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003ePercent apoptosis detected in DMSO- or belvarafenib plus cobimetinib- or belvarafenib plus cobimetinib plus nucleosides-treated WM3406 (left) or MeWo (middle) cells. MTHFD2 was knocked down prior to belvarafenib plus cobimetinib treatment (n=3). Two-way ANOVA. Right:\u003cstrong\u003e \u003c/strong\u003eWestern blot analysis of MTHFD2 in WM3406 (top) and MeWo (bottom) cells with MTHFD2 knockdown. \u003cstrong\u003eb\u003c/strong\u003e, (Left) Representative images of WM3406 (left) and MeWo (right) cells following indicated treatment for 18 h and processed for γH2AX (green) immunofluorescence. Nuclei were counterstained with DAPI (scale bars: 25 μm). \u003cstrong\u003ec\u003c/strong\u003e, Quantification of γH2AX foci per cell (top) and γH2AX foci intensity (bottom) in WM3406 (left) and MeWo (right) cells following indicated treatment (n=3). 40 cells per condition were randomly quantified. Two-way ANOVA. \u003cstrong\u003ed\u003c/strong\u003e, Schematic diagram showing the one-carbon metabolism and nucleotide biosynthesis in melanoma cells. \u003cstrong\u003ee\u003c/strong\u003e, Western blot analysis of γH2AX and \u003cstrong\u003ef\u003c/strong\u003e percent apoptosis detected in WM3406 (left) or MeWo (right) cells following belvarafenib, cobimetinib, INK128 treatment with or without nucleosides supplementation (n=3). Two-way ANOVA. All data represent mean ± SD.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/2bb2b3298365dcb25c2a6bbc.png"},{"id":87035110,"identity":"1240bcad-2b46-4f6e-a007-ee4929b442ba","added_by":"auto","created_at":"2025-07-18 13:12:16","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2703981,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe triple therapy is effective to treat NRAS-mutant melanoma \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ein vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Schematic of experimental design (gavage of drugs) panel \u003cstrong\u003eb-d\u003c/strong\u003e. \u003cstrong\u003eb\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eTumor growth curve comparing MaNRAS1007 melanomas grown on C57BL/6N host mice with indicated treatments.\u003cstrong\u003e c\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eKaplan-Meier curves showing overall survival (OS) of mice bearing MaNRAS1007 melanomas with indicated treatments. Log-rank test. \u003cstrong\u003ed\u003c/strong\u003e, Relative weight change (% initial body weight prior to treatment) in mice receiving indicated treatments. \u003cstrong\u003ee\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eRepresentative images (left) and percentages (right) of p-ERK, p-S6, Ki67, ATF4 and MTHFD2-positive melanoma cells in primary melanoma sections with indicated treatment. The positive signals (pink) were indicated by magenta red and tissues were counterstained with Hematoxylin (purple). Three melanomas from each treatment arm were stained for p-ERK, p-S6 and Ki67, to confirm the on-target effect of the drugs (day 14; n = 3; scale bars: 100 μm). The percentages of ATF4-positive cell nuclei and MTHFD2-positive cells were quantified (day 20; n = 5; scale bars: 100 μm). All data represent mean ± SEM.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/43cf7bf6893642a0d043fe48.png"},{"id":87035109,"identity":"abe275f8-f9b1-439b-b3ab-7ab2bbe9502f","added_by":"auto","created_at":"2025-07-18 13:12:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":2516267,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTargeting mTOR activity overcomes the resistance of belvarafenib and cobimetinib combination therapy in NRAS-mutant melanoma.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Schematic of experimental design for panel \u003cstrong\u003eb-i\u003c/strong\u003e. \u003cstrong\u003eb\u003c/strong\u003e, Left: IC\u003csub\u003e50\u003c/sub\u003e curve for belvarafenib inhibition of MaNRAS1007 tumor dissociated parental (IC\u003csub\u003e50\u003c/sub\u003e = 27.20 nM) and drug resistant (IC\u003csub\u003e50\u003c/sub\u003e = 320.1 nM) cells with the treatment of 10 nM of cobimetinib (n=4). Right: IC\u003csub\u003e50\u003c/sub\u003e curve for cobimetinib inhibition of MaNRAS1007 tumor dissociated parental (IC\u003csub\u003e50\u003c/sub\u003e = 26.19 nM) and drug resistant (IC\u003csub\u003e50\u003c/sub\u003e = 504.4 nM) cells with the treatment of 25 nM of belvarafenib (n=4). \u003cstrong\u003ec\u003c/strong\u003e, Percent apoptosis in belvarafenib plus cobimetinib resistant MaNRAS1007 cells following INK128 treatment (n=3). Two-sided unpaired \u003cem\u003et\u003c/em\u003e test. \u003cstrong\u003ed\u003c/strong\u003e, Left: IC\u003csub\u003e50\u003c/sub\u003e curve for belvarafenib inhibition of WM3406 tumor dissociated parental (IC\u003csub\u003e50\u003c/sub\u003e = 23.88 nM) and drug resistant (IC\u003csub\u003e50\u003c/sub\u003e = 239.5) cells with the treatment of 50 nM of cobimetinib (n=4). Right: IC\u003csub\u003e50\u003c/sub\u003e curve for cobimetinib inhibition of WM3406 tumor dissociated parental (IC\u003csub\u003e50\u003c/sub\u003e = 29.69 nM) and drug resistant (IC\u003csub\u003e50\u003c/sub\u003e = 316.7 nM) cells with the treatment of 25 nM of belvarafenib (n=4). \u003cstrong\u003ee\u003c/strong\u003e, Percent apoptosis in belvarafenib plus cobimetinib resistant WM3406 cells following INK128 treatment (n=3). Two-sided unpaired \u003cem\u003et\u003c/em\u003e test. \u003cstrong\u003ef\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eTumor growth curve comparing MaNRAS1007 melanomas grown on C57BL/6N host mice with or without belvarafenib plus cobimetinib treatments. After melanomas got resistant to belvarafenib plus cobimetinib treatments, mice were randomly split to treat with belvarafenib plus cobimetinib treatments (orange) or the triple therapy (blue). Two-way ANOVA.\u003cstrong\u003e g\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eRepresentative images (left) and percentages (right) of ATF4 and MTHFD2-positive melanoma cells in primary melanoma sections with indicated treatment (n = 5; scale bars: 100 μm). One-way ANOVA.\u003cstrong\u003e h\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eTumor growth curve comparing WM3406 melanomas grown on NOD/SCID host mice with or without belvarafenib plus cobimetinib treatments. After melanomas got resistant to belvarafenib plus cobimetinib treatments, mice were randomly split to treat with belvarafenib plus cobimetinib treatments (orange) or the triple therapy (blue). Two-way ANOVA. \u003cstrong\u003ei\u003c/strong\u003e, Representative images (left) and percentages (right) of ATF4 and MTHFD2-positive melanoma cells in primary melanoma sections with indicated treatment (n = 4; scale bars: 100 μm). One-way ANOVA. In all the IHC data, the positive signals (pink) were indicated by magenta red and tissues were counterstained with Hematoxylin (purple). Panels \u003cstrong\u003eb-e\u003c/strong\u003e represent mean ± SD. Panels \u003cstrong\u003ef-i\u003c/strong\u003e represent mean ± SEM.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/e9a3d46fc48ad8bc4be3b05e.png"},{"id":88638217,"identity":"ed34c939-52f8-4c56-810f-b682473c6980","added_by":"auto","created_at":"2025-08-08 15:28:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":12500248,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/6c9d31aa-086e-4b1d-89b8-1bfd284111a1.pdf"},{"id":87035104,"identity":"d5f24943-1652-4677-8553-3eb78a0084b0","added_by":"auto","created_at":"2025-07-18 13:12:16","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4253371,"visible":true,"origin":"","legend":"Supplementary info","description":"","filename":"SupplementaryfilesCaietalncomms.docx","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/145ec98bcdf5ef3ba5914de0.docx"},{"id":87036288,"identity":"e00c6770-9a64-4cbe-b584-1ebd6e82eafd","added_by":"auto","created_at":"2025-07-18 13:20:16","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2665652,"visible":true,"origin":"","legend":"Reporting Summary","description":"","filename":"nrreportingsummaryFLAT.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6857127/v1/59647d445af5dcb7c7c641d8.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"mTOR inhibition enhances the antitumor efficacy of RAF dimer-MEK blockade by inhibiting the ATF4-MTHFD2 pathway","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMelanoma is one of the most aggressive skin cancers, being highly metastatic \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Cutaneous melanoma can be classified into four genomic subtypes based on the mutation status of \u003cem\u003eBRAF\u003c/em\u003e, \u003cem\u003eNRAS\u003c/em\u003e, \u003cem\u003eNF1\u003c/em\u003e, and a fourth subgroup termed triple wild-type \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. The triple wild-type melanoma is often driven by less frequently detected hotspot mutations, such as in \u003cem\u003eKIT\u003c/em\u003e, a receptor tyrosine kinase \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Nevertheless, the RAS-RAF-MEK-ERK pathway is an essential signal transduction cascade in all melanoma subtypes \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. FDA-approved \u003cem\u003eBRAF V600\u003c/em\u003e inhibitors, vemurafenib, dabrafenib, and encorafenib \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e, are clinically combined with MEK inhibitors and cause rapid regression of \u003cem\u003eBRAF\u003c/em\u003e-mutant melanomas, which account for around 50% melanoma cases. However, these drugs are unable to inhibit both protomers of the RAF dimer and paradoxically activate MAPK signaling in non-\u003cem\u003eBRAF V600\u003c/em\u003e mutant melanoma \u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Currently, there is no FDA-approved targeted therapy for \u003cem\u003eNRAS\u003c/em\u003e-mutant, \u003cem\u003eNF1\u003c/em\u003e-mutant, and triple wild-type melanomas, making them classically \u0026ldquo;hard-to-treat\u0026rdquo; subtypes. In particular, \u003cem\u003eNRAS\u003c/em\u003e-mutant melanomas exhibit greater aggressiveness and are associated with a poorer prognosis in comparison to \u003cem\u003eBRAF\u003c/em\u003e-mutant melanomas \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Hence, it is crucial to expand novel and effective therapies for patients with hard-to-treat melanoma.\u003c/p\u003e\u003cp\u003eA novel RAF dimer inhibitor belvarafenib has shown promising pre-clinical results in \u003cem\u003eNRAS\u003c/em\u003e-mutant melanoma \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. However, belvarafenib resistant cell clones exist in melanoma, sustaining MAPK signaling in the presence of belvarafenib treatment. These resistant cells remain dependent on MAPK signaling and are sensitive to downstream MAPK inhibitors (MAPKi), such as MEK inhibitors (e.g., cobimetinib) \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. As a result, combined belvarafenib and cobimetinib was trialed in \u003cem\u003eNRAS\u003c/em\u003e-mutant melanoma patients (NCT03284502), with an overall response rate (ORR) of 38.5% \u003csup\u003e12\u003c/sup\u003e. Another pan-RAF and MEK inhibitor combination therapy (naporafenib\u0026thinsp;+\u0026thinsp;trametinib) is also being investigated in a randomized phase 3 trial for \u003cem\u003eNRAS\u003c/em\u003e-mutant melanoma patients (NCT0634067), with an ORR of 46.7% \u003csup\u003e13\u003c/sup\u003e. Despite encouraging data on the combination of belvarafenib and cobimetinib in melanoma, resistance remains a major challenge, as observed in nearly all patients treated with targeted therapy \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. In addition to MAPK signaling, driver mutations in melanoma often simultaneously activate the PI3K-AKT-mTOR pathway, which is an essential signaling cascade driving MAPK inhibitor resistance in melanoma \u003csup\u003e\u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. In \u003cem\u003eBRAF\u003c/em\u003e-mutant non-melanoma skin cancer with primary or secondary resistance to BRAF inhibitors, there is also a strong correlation between PI3K/mTOR signaling pathway and BRAF inhibitor resistance \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Importantly, most studies in melanoma examining the role of mTOR in the context of MAPK pathway-targeted therapies have focused largely on the \u003cem\u003eBRAF\u003c/em\u003e-mutant tumors. As a result, the contribution of mTOR signaling to therapeutic response and resistance in other melanoma subtypes, such as \u003cem\u003eNRAS\u003c/em\u003e-mutant or wild-type melanomas, remains largely unexplored, representing a significant gap in our understanding of subtype-specific vulnerabilities. mTOR activation has been linked to resistance to Wee1 inhibitors in lung cancer patients through DNA damage repair pathway \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, suggesting that mTOR mediates a broader mechanism beyond the well-characterized crosstalk between MAPK and mTOR pathways. Furthermore, to cope with stresses from rapid cell division and anti-tumor therapy, cancer cells often activate the integrated stress response (ISR) pathway, governed by activating transcription factor 4 (ATF4) \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Mechanistically, mTOR can control ATF4 translation \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e and plays an important role in resistance to BRAF-targeted therapy \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eGiven the essential roles of the mTOR pathway and ISR signaling in therapy tolerance and resistance, we sought to investigate whether blocking mTOR would augment therapeutic responses to combined belvarafenib plus cobimetinib therapy. We chose to use sapanisertib (a.k.a INK128), an ATP-competitive catalytic inhibitor of mTOR \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e that has been evaluated in clinical trials to treat a variety of solid tumors (NCT02412722, NCT02197572), showing well-tolerated adverse effects \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Unlike first-generation mTOR inhibitors \u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, sapanisertib targets both mTORC1 and mTORC2, blocking feedback loops that reactivate AKT, thereby expanding its potential in therapy-resistant tumors and provides a rationale for its use in combination with other anti-cancer agents. Indeed, we found that sapanisertib potentiated the anti-tumor effects of belvarafenib plus cobimetinib both \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e. By interrogating a list of ATF4 target genes that may underpin the response to belvarafenib\u0026thinsp;+\u0026thinsp;cobimetinib\u0026thinsp;+\u0026thinsp;sapanisertib (referred hitherto as triple therapy), \u003cem\u003eMTHFD2\u003c/em\u003e emerged as a top candidate. MTHFD2 is an enzyme in the folate cycle, which functions as a mediator of purine synthesis and DNA repair \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Hence, we hypothesized that MTHFD2 facilitates DNA repair in belvarafenib plus cobimetinib-treated cells, thus promoting their therapy tolerance. Finally, we explored the therapeutic benefit of the triple therapy to unravel how mTOR inhibition rewires MTHFD2-mediated purine synthesis and DNA repair in belvarafenib plus cobimetinib therapy in hard-to-treat melanomas. Accordingly, we demonstrate how incorporating mTOR inhibition is an effective strategy in targeting melanomas that are resistant to belvarafenib plus cobimetinib.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eInhibiting mTOR augments the efficacy of combined belvarafenib and cobimetinib in melanoma cells.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOncogenic mutations of \u003cem\u003eNRAS\u003c/em\u003e, \u003cem\u003eNF1\u003c/em\u003e and \u003cem\u003eKIT\u003c/em\u003e in melanoma drive hyperactivated RAS signaling, which contributes to activated MEK/ERK and PI3K/mTOR signaling \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Given the emerging role of the mTOR pathway in melanoma response and resistance to MAPK inhibition \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, we hypothesized that targeting the mTOR pathway could enhance the antitumor activity of belvarafenib plus cobimetinib combination therapy in non-BRAF mutant melanoma cells. We thus determined the half-maximal inhibitory concentration (IC\u003csub\u003e50\u003c/sub\u003e) of belvarafenib and cobimetinib in the context of combination therapy using a panel of human melanoma cell lines harboring different genetic mutations (\u003cem\u003eNRAS\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003eR\u003c/em\u003e\u003c/sup\u003e-mutant BLM cells, \u003cem\u003eNRAS\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003eK\u003c/em\u003e\u003c/sup\u003e-mutant WM3406 and WM3623 cells, \u003cem\u003eNF1\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ1336*\u003c/em\u003e\u003c/sup\u003e-mutant MeWo cells, \u003cem\u003ec-KIT\u003c/em\u003e\u003csup\u003e\u003cem\u003eD820Y\u003c/em\u003e\u003c/sup\u003e-mutant HBL cells and \u003cem\u003eNRAS\u003c/em\u003e\u003csup\u003e\u003cem\u003eG\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003eV\u003c/em\u003e\u003c/sup\u003e-mutant YUGOE cells \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e) (Supplementary Fig.\u0026nbsp;1a). Next, we treated melanoma cell lines with belvarafenib (25 nM), cobimetinib (50 nM), and the mTOR inhibitor sapanisertib (hereafter referred to as INK128 \u003csup\u003e24\u003c/sup\u003e, 25 nM) and monitored their impact on colony formation capacity (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea,b and Supplementary Fig.\u0026nbsp;1b,c). As expected, belvarafenib plus cobimetinib significantly decreased colony formation, but some cells persisted, surviving the 10-day treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea and Supplementary Fig.\u0026nbsp;1b). Notably, there were fewer colonies in all four melanoma cell lines treated with the triple therapy of INK128 plus belvarafenib plus cobimetinib (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea and Supplementary Fig.\u0026nbsp;1b). Using western blot, all cell lines were tested to confirm on-target engagement at the chosen drug concentrations (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea,b). We further showed that the triple therapy induced the highest level of apoptosis at 48 hours. Interestingly, INK128 did not induce apoptosis on its own, consistent with its anti-tumoral effects being mostly cytostatic, rather than cytolytic \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, but significantly increased belvarafenib plus cobimetinib induced apoptosis in all the melanoma subtypes tested (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec and Supplementary Fig.\u0026nbsp;1c).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eThe triple therapy induces apoptosis of melanoma cells through the inhibition of ATF4 and induction of DNA damage.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo investigate potential mechanisms through which the triple therapy most effectively eliminated melanoma cells, we performed RNA sequencing analysis on the \u003cem\u003eNRAS\u003c/em\u003e-mutant WM3406 cell line (Supplementary Fig.\u0026nbsp;2a,b). Gene set enrichment analysis (GSEA) on differentially expressed genes (DEGs) in cells treated with the triple therapy versus the vehicle-treated cells identified the \u0026ldquo;cellular response to stress\u0026rdquo; pathway, centered on the integrated stress response (ISR) signaling, as being downregulated in the triple therapy-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea,b and Supplementary Fig.\u0026nbsp;2c-e). The activation of the ISR, marked by increased expression of its major effector ATF4, plays a crucial role in supporting persister cell survival during BRAF-targeted therapy \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Generally, ISR promotes the phosphorylation of eIF2α at serine 51, inhibiting translation initiation and leading to upregulation of ATF4 \u003csup\u003e33\u003c/sup\u003e. Western blot analysis showed that the triple therapy resulted in an inhibition of ATF4 protein expression in WM3406 and MeWo cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec), without any consistent changes in the phosphorylation status of eIF2α (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec), which is also supported by a study showing mTORC1 controls ATF4 independently of changes in eIF2α phosphorylation \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Similar phenotypes were observed in a murine \u003cem\u003eNRAS\u003c/em\u003e mutant melanoma cell line termed MaNRAS1007 \u003csup\u003e34,35\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed and Supplementary Fig.\u0026nbsp;2g).\u003c/p\u003e\u003cp\u003eMelanoma shows intratumor heterogeneity in the activity of MAPK signaling \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, which we hypothesize, potentially regulates ATF4 level. Thus, we next reanalyzed a spatially resolved, unsupervised transcriptomics dataset generated from tumors induced in an \u003cem\u003eNRAS\u003c/em\u003e-mutant melanoma mouse model (Supplementary Fig.\u0026nbsp;3), from which MaNRAS1007 cell line was derived \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Using previously defined pathway-specific gene signatures \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, we noted that the tumor sample with the highest overall \u003cem\u003eATF4\u003c/em\u003e expression also had the highest MAPK- and mTOR-pathway activity (Supplementary Fig.\u0026nbsp;3a-f), consistent with prior findings that both the MAPK and mTOR pathways drive ATF4 expression in melanoma \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Next, we applied our pathway activity analysis on the spatial transcriptomics data to assess the gene expression profile in each spatially localized environment (region of interest, ROI) (Supplementary Fig.\u0026nbsp;3a-f). As expected, the activities of MAPK and mTOR pathways were positively correlated (Supplementary Fig.\u0026nbsp;3g), supporting both pathways being simultaneously activated through common driver mutations (i.e., \u003cem\u003eNRAS\u003c/em\u003e) in melanoma \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Interestingly, ROIs with high MAPK activity (pink and green, top 20 percentile) showed highest ATF4 expression, regardless of their mTOR activity. On the contrary, in MAPK\u003csup\u003elo\u003c/sup\u003e regions, high mTOR activity (top 20 percentile) is associated with retained ATF4 expression compared with MAPK\u003csup\u003ehi\u003c/sup\u003e ROIs, whereas MAPK\u003csup\u003elo\u003c/sup\u003e mTOR\u003csup\u003elo\u003c/sup\u003e regions had the lowest level of \u003cem\u003eATF4\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3g). Similarly, compared with ROIs with high MAPK signaling activity, those with low MAPK signaling activity showed stronger correlation between ISR and mTOR pathway activity (Supplementary Fig.\u0026nbsp;3h). Together, these data suggest that while both the MAPK and mTOR pathways facilitate ATF4 expression \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e, elevated mTOR activity may compensate for MAPK suppression by maintaining \u003cem\u003eATF4\u003c/em\u003e levels and driving ISR signaling.\u003c/p\u003e\u003cp\u003eGiven the importance of the ATF4-governed ISR in melanoma drug tolerance and resistance \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, we hypothesized that ATF4 may be essential for the survival of melanoma persister cells during combined belvarafenib plus cobimetinib therapy, which ultimately lead to the development of resistance \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. Although a 24-hour combination treatment with belvarafenib plus cobimetinib repressed ATF4 expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec,d), when WM3406 and MeWo cells were treated with belvarafenib plus cobimetinib for 10 days, at the timepoint when only persister cells remained (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea, \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee), ATF4 expression was no longer impacted by the therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef). Importantly, INK128 was still able to efficiently downregulate ATF4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef), consistent with the spatial transcriptomic data suggesting that dual blockade of MAPK and mTOR most efficiently decreases ATF4 expression (Supplementary Fig.\u0026nbsp;3h). We next formally tested whether ATF4 is required for the survival of tumor cells that persist following belvarafenib plus cobimetinib therapy. ATF4 knockdown using short interfering RNA (siRNA) did not compromise cell survival relative to those treated with non-targeting siRNA (\u003cem\u003esiCtrl\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg,h). However, \u003cem\u003eATF4\u003c/em\u003e-depleted cells were more sensitive to belvarafenib plus cobimetinib induced apoptosis, as shown by Annexin V-PI staining (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg) and western blotting for cleaved PARP (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eh). Together, these data revealed an ATF4-dependent mechanism of tolerance in the MAPKi-induced melanoma persister cells. We suggest then that blockade of mTOR enhances the efficacy of MAPKi by synergistically potentiating and sustaining ATF4 suppression.\u003c/p\u003e\u003cp\u003eTo better understand the role of ATF4 in modulating the response of melanoma cells to the triple therapy, we further analyzed our transcriptomic data and identified the \u0026ldquo;DNA damage response\u0026rdquo; and \u0026ldquo;double-strand break repair\u0026rdquo; pathways, as being downregulated in the triple therapy-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea and Supplementary Fig.\u0026nbsp;2e). Given that impaired DNA double-strand break repair can lead to excessive DNA damage and increased apoptosis \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e, we next examined markers of DNA damage, with a primary focus of γH2AX, a well-established read-out of DNA double-strand breaks (DSBs) \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. Belvarafenib plus cobimetinib consistently induced γH2AX, with INK128 further enhancing this effect (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei,k). Importantly, knockdown of \u003cem\u003eATF4\u003c/em\u003e also resulted in enhanced γH2AX formation in cells treated with combined belvarafenib plus cobimetinib (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ej), phenocopying the effect observed with the mTOR inhibitor (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei,k). Altogether, our data suggest that mTOR inhibition potentiates belvarafenib plus cobimetinib-induced DNA damage through a mechanism downstream of ATF4.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eThe triple therapy suppresses MTHFD2 downstream of ATF4.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAs an important transcription factor of ISR signaling, ATF4 target genes have been previously documented, wherein each target is evaluated and classified based on high, medium, or low confidence \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. We interrogated our RNA sequencing data for the expression of 37 high confidence ATF4 target genes and found that several were downregulated by the triple therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea and Supplementary Table\u0026nbsp;1). We next explored the clinical relevance of these 37 ATF4 target genes by correlating their expression with \u003cem\u003eATF4\u003c/em\u003e expression, using human melanoma data from The Cancer Genome Atlas (TCGA) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). Among the candidate genes of interest, \u003cem\u003emethylenetetrahydrofolate dehydrogenase 2 (MTHFD2)\u003c/em\u003e was particularly noteworthy. MTHFD2 is an enzyme that functions in mitochondrial one-carbon metabolism in the tetrahydrofolate (THF) cycle, which is essential for nucleotide synthesis and helps provide building blocks for subsequent DNA repair \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e, making it an attractive candidate for subsequent evaluation. \u003cem\u003eMTHFD2\u003c/em\u003e was repressed by the triple therapy in our RNAseq data and its expression correlated with ATF4 expression in the TCGA dataset (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea,c). MTHFD2 mRNA and protein were decreased in the triple therapy-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed). Further supporting ATF4 as a regulator of MTHFD2 expression, knockdown of ATF4 decreased MTHFD2 protein level, while MTHFD1, which is independent of ATF4, remained unaffected (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eSilencing of MTHFD2 in belvarafenib and cobimetinib combination treated melanoma cells phenocopies the triple therapy by inhibiting purine biosynthesis and increasing DNA damage.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eGiven the essential role of MTHFD2 in nucleotide synthesis and DNA repair, we next sought to explore the link between the ATF4-MTHFD2 axis and the triple therapy-induced DNA damage. Similarly to the data obtained in \u003cem\u003eATF4\u003c/em\u003e-silenced melanoma cells, \u003cem\u003eMTHFD2\u003c/em\u003e knockdown sensitized melanoma cells to belvarafenib plus cobimetinib treatment with a significantly increased level of DNA damage (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea-c).\u003c/p\u003e\u003cp\u003eTHF cycle produces one-carbon formyl groups for various cellular processes, including \u003cem\u003ede novo\u003c/em\u003e purine synthesis. MTHFD2 in THF cycle can oxidize CH2-THF to 10-formyl THF, which participates in multiple steps in purine synthesis \u003csup\u003e\u003cspan additionalcitationids=\"CR46\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Gotoh \u003cem\u003eet al.\u003c/em\u003e found that the amount of purine nucleotides is greatly reduced in \u003cem\u003eMTHFD2\u003c/em\u003e knockdown lung cancer cells \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. Additionally, Zhou \u003cem\u003eet al.\u003c/em\u003e showed that impairing nucleotide biosynthesis results in an insufficient supply of metabolites for DNA repair \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. We thus measured whether an exogenous supply of nucleosides would rescue the phenotypes observed in \u003cem\u003eMTHFD2\u003c/em\u003e-depleted melanoma cells treated with belvarafenib plus cobimetinib. As expected, nucleoside supplementation, which bypasses the need for MTHFD2 in nucleotide biosynthesis (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed), protected \u003cem\u003eMTHFD2\u003c/em\u003e-depleted cells from the apoptosis induced by belvarafenib plus cobimetinib (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea), and significantly reduced γH2AX levels in these conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb,c). Similarly, the addition of nucleosides also attenuated the DNA damage observed in belvarafenib plus cobimetinib-treated cells and triple therapy-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee), resulting in an overall reduction of apoptosis following these treatments (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ef). Together, these data suggest that the triple therapy induces DNA damage and apoptosis in melanoma cells via blocking the ATF4-MTHFD2 axis-mediated DNA repair pathways.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eThe triple therapy decreases tumor outgrowth and improves survival of\u003c/b\u003e \u003cb\u003eNRAS\u003c/b\u003e\u003cb\u003e-mutant melanoma bearing mice\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eTo evaluate the triple therapy \u003cem\u003ein vivo\u003c/em\u003e, we used the \u003cem\u003eNRAS\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003eK\u003c/em\u003e\u003c/sup\u003e-mutant MaNRAS1007 murine melanoma model \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Tumor bearing mice were randomized into 4 treatment arms: vehicle, INK128 (0.5 mg/kg/daily), belvarafenib (5 mg/kg/daily) plus cobimetinib (5 mg/kg/daily), and the triple therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). We have tailored the doses of each drug to ensure on-target engagement and optimize therapeutic efficacy, while minimizing potential toxicity. Melanoma growth was substantially inhibited in the INK128 monotherapy, and as expected, the tumors responded to belvarafenib plus cobimetinib\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). However, the tumors gradually progressed at approximately day 40 of the belvarafenib plus cobimetinib treatment, suggesting the emergence of drug resistance (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). Notably, tumor control was significantly extended in the mice administered the triple therapy, compared with those that received INK128 monotherapy or the belvarafenib plus cobimetinib dual therapy. Importantly, the triple therapy cohort showed significantly improved overall survival (OS), compared with any other treatment groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec). Body weight was monitored throughout treatment, and no significant weight loss was observed, indicating good tolerability of the therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ed). These data suggest that mTOR inhibition may eliminate a substantial proportion of melanoma cells that resist combined belvarafenib plus cobimetinib (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb and Supplementary Fig.\u0026nbsp;4).\u003c/p\u003e\u003cp\u003eThe \u003cem\u003ein vivo\u003c/em\u003e on-target activity of the therapies was confirmed by immunohistochemistry (IHC) staining after 14 days on treatment. As expected, p-ERK levels were inhibited in with the tumors from the belvarafenib plus cobimetinib cohort, and p-S6 levels were repressed in mice treated with INK128 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ee). There were fewer Ki67 positive stained cells in the melanomas treated with the triple therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ee), suggesting an inhibition of melanoma proliferation. Finally, we asked whether ATF4-MTHFD2 pathway would be inhibited by the triple therapy \u003cem\u003ein vivo\u003c/em\u003e. We found that nuclear ATF4 expression and the expression of MTHFD2 were discernibly lower in the triple therapy group, suggesting impaired ATF4-MTHFD2 signaling \u003cem\u003ein vivo\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ee).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003emTOR inhibition can overcome the resistance to the combination therapy of belvarafenib and cobimetinib in\u003c/b\u003e \u003cb\u003eNRAS\u003c/b\u003e\u003cb\u003e-mutant melanoma.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFinally, we tested whether blocking mTOR would be of clinical benefit once therapeutic resistance was acquired using cell lines derived from pre-clinical mouse models (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). For this, mice bearing MaNRAS1007 murine melanomas or WM3406 xenografts were treated with belvarafenib plus cobimetinib; after 35 days, the tumors began to progress, indicating acquired resistance. To confirm resistance to belvarafenib plus cobimetinib, we dissociated the parental and dual drug-resistant melanomas (i.e. derived from MaNRAS1007 or WM3406 melanomas) and tested them \u003cem\u003ein vitro\u003c/em\u003e. As expected, the resistant cell lines did not respond to belvarafenib and cobimetinib (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb,d). However, treatment with INK128 significantly induced apoptosis in these resistant cell lines (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec,e), supporting its potential as a salvage therapy. Therefore, we next treated these belvarafenib plus cobimetinib-resistant melanomas with INK128, while keeping the mice on belvarafenib plus cobimetinib. The addition of INK128 inhibited tumor outgrowth for both the MaNRAS1007 melanoma and WM3406 xenograft, which were resistant to belvarafenib plus cobimetinib (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ef,h). Consistently, melanomas that were resistant to belvarafenib plus cobimetinib expressed higher ATF4 and MTHFD2. INK128 reduced the levels of ATF4 and MTHFD2 in these belvarafenib plus cobimetinib resistant tumors (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eg,i). Taken together, our results provide a rationale to inhibit mTOR activity to overcome acquired resistance to combined belvarafenib plus cobimetinib.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this present study, we showed that mTOR inhibition can augment therapeutic responses to combined belvarafenib and cobimetinib in cell culture models and pre-clinical mouse models of melanoma. Importantly, inhibiting mTOR demonstrated potent antitumor effects in \u003cem\u003eNRAS\u003c/em\u003e mutant melanomas with acquired resistance to belvarafenib and cobimetinib (Fig. 6). While our study supports the established crosstalk between the MAPK and mTOR pathways in regulating ATF4 activity\u003csup\u003e15,22,50,51\u003c/sup\u003e, it also uncovers a novel therapeutic mechanism: combining RAF dimer inhibitors with MEK and mTOR inhibitors effectively suppresses the ATF4-MTHFD2 axis, disrupting purine synthesis and impairing DNA repair. We provide evidence that this multi-targeted approach offers a powerful strategy to overcome resistance and enhance antitumor efficacy in MAPK-driven cancers.\u003c/p\u003e\n\u003cp\u003eOur findings are aligned with prior work demonstrating that combined belvarafenib and cobimetinib therapy exhibits clinical activity in patients with \u003cem\u003eNRAS\u003c/em\u003e-mutant melanomas (NCT03284502), with an overall response rate of 38.5% \u003csup\u003e11,12\u003c/sup\u003e. INK128 has also reached clinical testing and is well tolerated (NCT02412722, NCT02197572), thus providing an opportunity to assess whether mTOR inhibition improves the efficacy of combined belvarafenib and cobimetinib. Notably, in \u003cem\u003eNRAS\u003c/em\u003e-mutant melanomas treated with belvarafenib plus cobimetinib, acquired resistance can develop, which we have shown is repressed by the blockade of mTOR (Fig. 6). Therefore, INK128 may be tested first in patients who have progressed on combination belvarafenib plus cobimetinib therapy.\u003c/p\u003e\n\u003cp\u003eATF4 protein is tightly regulated during the ISR, which specifically enhances the translation of ATF4 and other mRNAs containing upstream open reading frames (uORFs) \u003csup\u003e33\u003c/sup\u003e. Oncogenic MAPK signaling induces the expression of ATF4, which plays a central role in stress response and cell survival \u003csup\u003e32\u003c/sup\u003e. In \u003cem\u003eNRAS\u003c/em\u003e and \u003cem\u003eNF1\u003c/em\u003e-mutant melanomas, we demonstrate that ATF4 expression is partially repressed following the blockade of MAPK signaling by belvarafenib plus cobimetinib (Fig. 2c), adding an important perspective to an earlier study \u003csup\u003e40\u003c/sup\u003e. Moreover, we showed that inhibiting mTOR in the context of belvarafenib plus cobimetinib further suppressed ATF4 (Fig. 2c), which is consistent with prior work showing that the mTOR pathway is required for the MAPK-mediated ATF4 induction \u003csup\u003e32\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eATF4 has been shown to have a multifaceted role in cancer, exhibiting both pro-survival and pro-apoptotic effects. Our work showed that both drug persistent melanoma cells \u003cem\u003ein vitro\u003c/em\u003e and resistant melanomas \u003cem\u003ein vivo\u003c/em\u003e maintain ATF4 expression, which can be impaired upon inhibiting mTOR to overcome therapeutic resistance (Fig. 2f and Fig. 6c,f). These results indicate the involvement of the ATF4 pathway in belvarafenib plus cobimetinib persister cells, consistent with a pro-survival role. Indeed, activation of the ATF4 pathway represents an evolutionarily conserved general stress response that may function in adaptation to MAPK inhibitors \u003csup\u003e23\u003c/sup\u003e. However, Niessner \u003cem\u003eet al.\u003c/em\u003e reported that encorafenib plus binimetinib, targeting MAPK signaling, increased the expression of ATF4 in \u003cem\u003eNRAS\u003c/em\u003e-mutant melanoma, which played a pro-apoptotic role due to a set of ATF4 target genes mediated by endoplasmic reticulum (ER) stress \u003csup\u003e52\u003c/sup\u003e. BRAF inhibitor vemurafenib or dabrafenib can also induce ATF4 expression transiently (within 4 hours) in \u003cem\u003eBRAF\u003c/em\u003e-mutant melanoma \u003csup\u003e32,53\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eGiven the importance of ATF4 in drug persistence and resistance, the widespread role of ATF4 target genes in mitochondria function, autophagy, pro-apoptotic pathway and tRNA biosynthesis should not be neglected \u003csup\u003e33\u003c/sup\u003e. Indeed, our study shows that the expression of a subset of ATF4 target genes decrease after the treatment of belvarafenib plus cobimetinib (Fig. 3a). In support of our findings, ATF4 expression and MTHFD2 expression are positively correlated in TCGA SKCM (Skin Cutaneous Melanoma) cohort (Fig. 3c) and there is strong evidence from human and murine studies indicating that ATF4 regulates MTHFD2 transcription \u003csup\u003e54-57\u003c/sup\u003e. Herein, our data in melanoma models support that MTHFD2 is a target of ATF4. Moreover, activated mTOR leads to upregulated MTHFD2 and higher activity of purine synthesis, while knockdown of ATF4 decreases MTHFD2, which further indicates that, as a metabolic effector of mTOR \u003csup\u003e50\u003c/sup\u003e, ATF4 is required for the induction MTHFD2-mediated purine synthesis \u003csup\u003e58\u003c/sup\u003e. Although previous studies have shown that targeting mTOR or ATF4 downregulates MTHFD2 \u003csup\u003e22,50\u003c/sup\u003e, our study provides the first evidence that MTHFD2 regulates purine synthesis and DNA damage repair, thereby modulating the response to MAPK-targeted therapy in non-\u003cem\u003eBRAF\u003c/em\u003e-mutant melanoma.\u003c/p\u003e\n\u003cp\u003eOur finding of the upregulation of ATF4 and MTHFD2 in melanomas that are resistant to combined belvarafenib and cobimetinib therapy raises the question of how ATF4 and MTHFD2 impact drug resistance. MTHFD2 catalyzes the oxidation of CH2-THF to CHO-THF, a critical reaction required for several key steps in purine synthesis, while producing free nucleotides as building blocks for subsequent DNA synthesis and repair \u003csup\u003e59,60\u003c/sup\u003e. A previous study showed that overexpressing MTHFD2 results in MAPK signaling activation in bladder cancer \u003csup\u003e61\u003c/sup\u003e. Reactivated MAPK signaling is a hallmark of targeted therapy resistance in melanoma cells, indicating a potential link between MTHFD2 and drug resistance. In support of these speculations, MTHFD2 is found to be critical for lung adenocarcinoma that is resistant to EGFR inhibitor, likely through increasing purine nucleotide \u003csup\u003e48\u003c/sup\u003e. MTHFD2 knockdown reduces folate-mediated one-carbon metabolism and purine synthesis in lung cancer cells, increasing the sensitivity to EGFR inhibitor gefitinib \u003csup\u003e48\u003c/sup\u003e. In addition, it is widely accepted that impairing purine synthesis causes DNA repair deficiency, which contributes to the efficacy of radiation therapy in glioblastoma \u003csup\u003e49\u003c/sup\u003e. Here, we show that inhibiting mTOR increases DNA damage in combined belvarafenib and cobimetinib therapy (Fig. 2k), and nucleoside supplementation decreases \u0026gamma;H2AX (Fig. 4e), indicating the reduction of the accumulation of DNA damage. Knocking down \u003cem\u003eMTHFD2\u003c/em\u003e promotes DNA damage induced by belvarafenib plus cobimetinib and makes melanoma cells more prone to response to belvarafenib and cobimetinib combination therapy, while the supplementation of nucleoside reduces DNA damage and prevents these \u003cem\u003eMTHFD2\u003c/em\u003e-depleted cells from being sensitized to belvarafenib plus cobimetinib (Fig. a-c). These results are consistent with recent work identifying that the supplementation of nucleoside promotes DNA repair in glioblastoma, where inhibiting GTP synthesis impairs DNA repair and sensitizes resistant glioblastoma cells to radiotherapy \u003csup\u003e49,62\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eCombined belvarafenib plus cobimetinib therapy has demonstrated efficacy in patients with Class 2 and 3 \u003cem\u003eBRAF\u003c/em\u003e-mutant or \u003cem\u003eNRAS\u003c/em\u003e-mutant melanoma \u003csup\u003e11,12,63\u003c/sup\u003e. In our preclinical models, co-targeting MAPK and mTOR pathway induced apoptosis and suppressed the growth of \u003cem\u003eNRAS\u003c/em\u003e-mutant, \u003cem\u003eNF1\u003c/em\u003e-mutant, and \u003cem\u003eKIT\u003c/em\u003e-mutant melanoma (Fig. 1). These data suggested that our proposed triple therapy could potentially benefit patients with not only \u003cem\u003eNRAS-\u003c/em\u003e, but also \u003cem\u003eNF1-\u003c/em\u003e, and \u003cem\u003eKIT-\u003c/em\u003emutant melanoma. Additionally, while BRAF\u003csup\u003eV600\u003c/sup\u003e-mutant melanomas are sensitive to combined BRAF inhibitor and MEK inhibitor therapy, acquired resistance invariably occurs, often resulting from the transactivation of MAPK signaling induced by RAF dimerization \u003csup\u003e7\u003c/sup\u003e. Since belvarafenib in our triple therapy is a RAF dimer inhibitor, triple therapy could potentially benefit patients with drug resistant class 1 \u003cem\u003eBRAF\u003c/em\u003e-mutant melanomas.\u003c/p\u003e\n\u003cp\u003eIn summary, we provide evidence that pharmacologic inhibition of mTOR shows efficacy in preclinical melanoma models when combined with belvarafenib and cobimetinib, via a mechanism involving the blockade of ATF4-MTHFD2 axis-regulated purine synthesis, which results in DNA damage and apoptosis.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eMice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMale C57BL/6N mice (6-8 weeks old) were purchased from Charles River Laboratories. Female nonobese diabetic (NOD)/severe combined immunodeficiency (SCID) mice (6-10 weeks old) were kindly gifted by Dr. Moulay Alaoui-Jamali. All mice were randomized before injection. For melanoma cell inoculation, MaNRAS1007 cells were injected to male C57BL/6N mice at 1,000,000 cells/mouse. WM3406 cells were injected to female NOD/SCID mice at 1,000,000 cells/mouse. All melanoma cell lines were freshly prepared in PBS and subcutaneously injected to the right flank of mice. Once palpable tumors were formed, tumors were measured in length (L) and width (W). Tumor volumes (V) were calculated based on the formula V=3.1416/6*L*W\u003csup\u003e2\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFor belvarafenib (5 mg/kg, MedChem Express, HY-109080), cobimetinib (5mg/kg, ChemieTek CT-G0973) and INK128 (0.5 mg/kg, MedChem Express, HY-13328) treatment, these drugs were dissolved in DMSO and subsequently diluted in PEG400 (Sigma-Aldrich, P3265), TWEEN\u0026reg; 20 (Sigma-Aldrich, P1379) and PBS. Mice were administered every day by gavage, starting when the tumor volume reached approximately 100 mm\u003csup\u003e3\u003c/sup\u003e. For tumor growth curve, mice were sacrificed when the tumor volume reached approximately 1500 mm\u003csup\u003e3\u003c/sup\u003e. For IHC in Fig. 6, all mice were sacrificed on Day 20 after treatment initiation. The IHC on relapsed tumors were performed when the tumor volume reached 1000 mm\u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCells and reagents\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSources, culture conditions and treatment timelines of human melanoma cell lines are listed in Supplementary Table 2. All the drug treatments began the next day after cell seeding.\u003c/p\u003e\n\u003cp\u003eTo inhibit the purine synthesis, melanoma cells were treated with MTHFD inhibitor LY345899 (10 \u0026mu;M, MedChem Express, HY-101943), GARFT inhibitor lometrexol (10 \u0026mu;M,\u0026nbsp;MedChem Express, HY-14521), DHODH inhibitor brequinar (10 \u0026mu;M,\u0026nbsp;MedChem Express, HY-108325). For nucleoside supplementation, nucleosides (100X, Sigma-Aldrich, ES-008) with 0.73 g/L cytidine, 0.85 g/L guanosine, 0.73 g/L uridine, 0.8 g/L adenosine and 0.24 g/L thymidine was added 1:100 in cell culture media. Cells were collected for apoptosis assay and Western blot after 48h treatment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eColony formation assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMelanoma cells were seeded into 6-well plates at 2,000 cells per well. The next day, belvarafenib, cobimetinib and INK128 (see Supplementary Table 2), or DMSO control were added to the indicated wells. Cells were subsequently cultured for 10 days, during which media was changed and drugs were freshly added every two days. At the end of the assay, cells were fixed with 4% formaldehyde/PBS, stained with 0.5% of crystal violet (Sigma-Aldrich, HT90132) diluted in 70% EtOH and photographed. Colonies were manually quantified using FIJI software (https://github.com/fiji/fiji) to assess cell survival and proliferation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWestern blotting\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImmunoblots were performed as previously described \u003csup\u003e64\u003c/sup\u003e. Briefly, cells were lysed with RIPA buffer (150mmol/L Tris-HCl, pH=7, 150mmol/L NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitors (Roche). Equal amounts of protein samples were loaded, separated on 10% or 12% SDS-PAGE gels, transferred to nitrocellulose membranes, and probed with corresponding antibodies. Detailed antibody information is listed in Supplementary Table 3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFlow cytometry-based assays\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMelanoma cells were seeded into 6-well plates at 100,000 cells per well. Following indicated treatments, cells were trypsinized, centrifuged at 240g for 5\u0026thinsp;min, and washed twice in PBS. For apoptosis detection of non-fixed cells, Alexa Fluor\u0026trade; 647-Annexin V (Invitrogen\u0026trade;, A23204) and Propidium Iodide (PI) Staining Solution (BD Biosciences, 556463) were diluted in 1\u0026times;binding buffer (BD Biosciences, 556454), and subsequently mixed with cells following the manufacturer\u0026rsquo;s instructions. All flow cytometry experiments were conducted on the FACSCanto (BD Biosciences).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA sequencing analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSequencing libraries were generated using NEBNext\u0026reg; UltraTM RNA Library Prep Kit for Illumina\u0026reg; (NEB, USA) following manufacturer\u0026rsquo;s recommendations. The library was sequenced using the Illumina NovaSeq 6000 sequencing platform to generate raw reads. Then, nf-core/rnaseq pipeline v3.8.1 \u003csup\u003e65\u003c/sup\u003e was used to perform quality control, trimming and alignment on raw paired-end fastq reads, followed by reference genome-guided transcriptome assembly and gene expression quantification. Differentially expressed genes (DEGs) were identified by DESeq2 \u003csup\u003e66\u003c/sup\u003e with cut-off values of a log2|fold-change| \u0026gt; 1 and a p-adjust \u0026lt; 0.05. ClusterProfiler \u003csup\u003e67\u003c/sup\u003e was used to perform functional enrichment analysis and the potential genes in the identified modules were analyzed based on gene ontology (GO) categories.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA interference\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003esiRNAs were transfected into cells using Lipofectamine\u0026trade; RNAiMAX Transfection Reagent (Invitrogen, 13778) following the manufacturer\u0026rsquo;s instructions. Media were changed the next day after cells were incubated with siRNAs for 18 hours. All cells were harvested between 48h-96h after siRNA transfection. All siRNA sequences are listed in Supplementary Table 4.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunofluorescence microscopy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWM3406 and MeWo cells were grown on glass coverslips and fixed with 4% (w/v) paraformaldehyde (PFA) in PBS for 15 min at room temperature. PFA-fixed cells were permeabilized with 0.2% (v/v) Triton X-100 in PBS for 20 min at room temperature. Cells were then incubated with blocking buffer (5% bovine serum albumin, and 0.5% Triton X-100 in PBS) for 1 hour at room temperature and then incubated with the \u0026gamma;H2AX primary antibodies overnight at 4 \u0026deg;C. After three washes with PBS, cells were incubated for 1 hour at room temperature with the secondary antibodies (Invitrogen, A21206). Cells were then washed with PBS, followed by incubation with 1 \u0026mu;g/ml DAPI nuclear stain for 15 minutes. Cells were washed and mounted to a glass slide using ProLong gold mounting media (Invitrogen, P36930). Slides were stored in the dark until fluorescence images were taken. Stacking and coloring of images was performed using FIJI software. Each quantification was done on at least three biological replicates and at least 3 ROIs per biological replicate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative real-time PCR\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCultured cells were pelleted, and RNA was prepared using the E.Z.N.A. total RNA isolation kit (OMEGA Bio-Tek). RNA concentrations were then quantified using a NanoDrop spectrophotometer (ThermoFisher Scientific) and cDNA was prepared from 1mg of total RNA using iScript cDNA Synthesis Kit (Bio-Rad). Target genes were quantified using the Applied Biosystems 7500 Fast Real-Time PCR System with SYBR Green real-time PCR master mix (Applied Biosystems). Two housekeeping genes were used for each assay. Primers used for qPCR are listed in Supplementary Table 5.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunohistochemistry (IHC)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStaining of mouse samples was performed as previously described \u003csup\u003e64\u003c/sup\u003e. Briefly, formalin-fixed, paraffin-embedded tumor sections were stained with indicated antibodies (Supplementary Table 3), followed by a standard magenta red detection protocol \u003csup\u003e68\u003c/sup\u003e (Agilent Technologies, GV92511-2). Hematoxylin-counterstained slides were mounted with coverslips. Slides were scanned on AxioScan (ZEISS) and positive staining quantified using QuPath v0.5.1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAccess and re-analysis of previously published datasets\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eProcessed count matrices of spatial transcriptomics (Visium) from NRAS\u003csup\u003eQ61K/\u0026deg;\u003c/sup\u003e;Ink4a\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e melanoma mouse model were downloaded from the website of Dr. Jean-Christophe Marine lab (https://marinelab.sites.vib.be/en) \u003csup\u003e37\u003c/sup\u003e. In brief, spots were retained when nFeature_Spatial\u0026thinsp;\u0026gt;\u0026thinsp;1,000 and percent.mt\u0026thinsp;\u0026lt;\u0026thinsp;5 and expression data were normalized using SCTransform (Seurat, v.5.0.2) \u003csup\u003e69\u003c/sup\u003e. To determine the MAPK pathway activity, we used the MPAS score based on the MAPK signaling characteristic gene signatures as previously described \u003csup\u003e38\u003c/sup\u003e. To determine the mTOR pathway activity, we used Gene Set Variation Analysis (GSVA) \u003csup\u003e70\u003c/sup\u003e to perform functional enrichment analysis based on HALLMARK categories. The ISR score was also calculated based on the Z-score of stress response regulators \u003csup\u003e33\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTumor dissociation and culture\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMaNRAS1007 and WM3406 melanomas in control group or belvarafenib plus cobimetinib resistant group were resected. Tumors were minced and digested in collagenase A to obtain single-cell suspension. MaNRAS1007 cells were cultured in Ham\u0026rsquo;s F12 containing 1% FBS, while WM3406 cells were cultured in RPMI containing 1% FBS and 1 x GlutaMAX.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn vitro data are presented as mean \u0026plusmn; SD. In vivo data are presented as mean \u0026plusmn; SEM. Prism software (GraphPad) was used to determine statistical significance of differences. Figure legends specify the statistical analysis used. P values are indicated in the figures and p \u0026lt; 0.05 were considered significant.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnimal experiments were conducted according to the regulations established by the Canadian Council of Animal Care, and protocols approved by McGill University Animal Care and Use Committee (#2015-7672).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by the Canadian Institutes of Health Research (CIHR) (grants PJT-180379 to AANR, PJT-162260, PJT-178194 to SVDR and grant PJT-183934 to WHM). FC and JM were endowed by Fonds de recherche du Qu\u0026eacute;bec - Sant\u0026eacute; doctoral scholarship. FJ, SVDR and WHM are supported by \u0026ldquo;Wallonie-Bruxelles International\u0026rdquo; and \u0026ldquo;Relations Internationales et Francophonie, Qu\u0026eacute;bec\u0026rdquo; to encourage active collaboration. This research as well as biobanking of biological material and data was made possible through a collaboration with the R\u0026eacute;seau de recherche sur le cancer (RRCancer) financially supported by the the Oncopole, the FRQ cancer division, which receives funding from Merck Canada Inc., GSK, Pfizer and the Minist\u0026egrave;re de l\u0026apos;\u0026Eacute;conomie, de l\u0026apos;Innovation et de l\u0026apos;\u0026Eacute;nergie du Qu\u0026eacute;bec. The RRCancer is affiliated to the Canadian Tumor Repository Network (CTRNet). We thank Madelyn Abraham for technical supports on DNA damage detection. We thank David Papadopoli for critical reading of the manuscript. We thank Christian Young, Mathew Duguay, and Darleen Element for experimental advice and technical supports.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u0026nbsp;\u003c/strong\u003eFC, FH, SVDR and WHM contributed conceptualization and investigation. FC, CG, HS, NG, EG, LMH, JM, JS, KB, AO, SVDR and WHM contributed methodology. FC, FH, CG, AO, SVDR and WHM contributed visualization. AANR, SVDR and WHM contributed funding acquisition. FC, SVDR and WHM contributed project administration and writing the original draft. SVDR and WHM contributed supervision. FC, FH, JM, FJ, AANR, AO, SVDR and WHM contributed review \u0026amp; Editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u003c/strong\u003e The authors have declared that no conflict of interest exists.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003ePrabhu, S. A., Moussa, O., Miller, W. H., Jr. \u0026amp; Del Rincon, S. V. The MNK1/2-eIF4E Axis as a Potential Therapeutic Target in Melanoma. \u003cem\u003eInt J Mol Sci\u003c/em\u003e 21 (2020).\u003c/li\u003e\n \u003cli\u003eCancer Genome Atlas, N. Genomic Classification of Cutaneous Melanoma. \u003cem\u003eCell\u003c/em\u003e 161, 1681-1696 (2015).\u003c/li\u003e\n \u003cli\u003eHeidorn, S. J.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Kinase-dead BRAF and oncogenic RAS cooperate to drive tumor progression through CRAF. \u003cem\u003eCell\u003c/em\u003e 140, 209-221 (2010).\u003c/li\u003e\n \u003cli\u003eKaroulia, Z., Gavathiotis, E. \u0026amp; Poulikakos, P. I. New perspectives for targeting RAF kinase in human cancer. \u003cem\u003eNat Rev Cancer\u003c/em\u003e 17, 676-691 (2017).\u003c/li\u003e\n \u003cli\u003eAscierto, P. A.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Contribution of MEK Inhibition to BRAF/MEK Inhibitor Combination Treatment of BRAF-Mutant Melanoma: Part 2 of the Randomized, Open-Label, Phase III COLUMBUS Trial. \u003cem\u003eJ Clin Oncol\u003c/em\u003e 41, 4621-4631 (2023).\u003c/li\u003e\n \u003cli\u003eHatzivassiliou, G.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth. \u003cem\u003eNature\u003c/em\u003e 464, 431-435 (2010).\u003c/li\u003e\n \u003cli\u003ePoulikakos, P. I., Zhang, C., Bollag, G., Shokat, K. M. \u0026amp; Rosen, N. RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF. \u003cem\u003eNature\u003c/em\u003e 464, 427-430 (2010).\u003c/li\u003e\n \u003cli\u003ePeng, S. B.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Inhibition of RAF Isoforms and Active Dimers by LY3009120 Leads to Anti-tumor Activities in RAS or BRAF Mutant Cancers. \u003cem\u003eCancer Cell\u003c/em\u003e 28, 384-398 (2015).\u003c/li\u003e\n \u003cli\u003eDankner, M.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Dual MAPK Inhibition Is an Effective Therapeutic Strategy for a Subset of Class II BRAF Mutant Melanomas. \u003cem\u003eClin Cancer Res\u003c/em\u003e 24, 6483-6494 (2018).\u003c/li\u003e\n \u003cli\u003eRandic, T., Kozar, I., Margue, C., Utikal, J. \u0026amp; Kreis, S. NRAS mutant melanoma: Towards better therapies. \u003cem\u003eCancer Treat Rev\u003c/em\u003e 99, 102238 (2021).\u003c/li\u003e\n \u003cli\u003eYen, I.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e ARAF mutations confer resistance to the RAF inhibitor belvarafenib in melanoma. \u003cem\u003eNature\u003c/em\u003e 594, 418-423 (2021).\u003c/li\u003e\n \u003cli\u003eShin, S. J.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e A phase Ib trial of belvarafenib in combination with cobimetinib in patients with advanced solid tumors: Interim results of dose-escalation and patients with NRAS-mutant melanoma of dose-expansion. \u003cem\u003eJournal of Clinical Oncology\u003c/em\u003e 39, 3007-3007 (2021).\u003c/li\u003e\n \u003cli\u003ede Braud, F.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Initial Evidence for the Efficacy of Naporafenib in Combination With Trametinib in NRAS-Mutant Melanoma: Results From the Expansion Arm of a Phase Ib, Open-Label Study. \u003cem\u003eJ Clin Oncol\u003c/em\u003e 41, 2651-2660 (2023).\u003c/li\u003e\n \u003cli\u003eBivona, T. G. \u0026amp; Doebele, R. C. A framework for understanding and targeting residual disease in oncogene-driven solid cancers. \u003cem\u003eNat Med\u003c/em\u003e 22, 472-478 (2016).\u003c/li\u003e\n \u003cli\u003eWang, B.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Targeting mTOR signaling overcomes acquired resistance to combined BRAF and MEK inhibition in BRAF-mutant melanoma. \u003cem\u003eOncogene\u003c/em\u003e 40, 5590-5599 (2021).\u003c/li\u003e\n \u003cli\u003eShi, H.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e A novel AKT1 mutant amplifies an adaptive melanoma response to BRAF inhibition. \u003cem\u003eCancer Discov\u003c/em\u003e 4, 69-79 (2014).\u003c/li\u003e\n \u003cli\u003eVillanueva, J.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Acquired resistance to BRAF inhibitors mediated by a RAF kinase switch in melanoma can be overcome by cotargeting MEK and IGF-1R/PI3K. \u003cem\u003eCancer Cell\u003c/em\u003e 18, 683-695 (2010).\u003c/li\u003e\n \u003cli\u003eObenauf, A. C.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Therapy-induced tumour secretomes promote resistance and tumour progression. \u003cem\u003eNature\u003c/em\u003e 520, 368-372 (2015).\u003c/li\u003e\n \u003cli\u003eSen, S., Meric-Bernstam, F., Hong, D. S., Hess, K. R. \u0026amp; Subbiah, V. Co-occurring Genomic Alterations and Association With Progression-Free Survival in BRAFV600-Mutated Nonmelanoma Tumors. \u003cem\u003eJ Natl Cancer Inst\u003c/em\u003e 109 (2017).\u003c/li\u003e\n \u003cli\u003eYe, L.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Transmembrane-4 L-six family member-1 (TM4SF1) promotes non-small cell lung cancer proliferation, invasion and chemo-resistance through regulating the DDR1/Akt/ERK-mTOR axis. \u003cem\u003eRespir Res\u003c/em\u003e 20, 106 (2019).\u003c/li\u003e\n \u003cli\u003ePakos-Zebrucka, K.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e The integrated stress response. \u003cem\u003eEMBO Rep\u003c/em\u003e 17, 1374-1395 (2016).\u003c/li\u003e\n \u003cli\u003ePark, Y., Reyna-Neyra, A., Philippe, L. \u0026amp; Thoreen, C. C. mTORC1 Balances Cellular Amino Acid Supply with Demand for Protein Synthesis through Post-transcriptional Control of ATF4. \u003cem\u003eCell Rep\u003c/em\u003e 19, 1083-1090 (2017).\u003c/li\u003e\n \u003cli\u003eYang, C., Tian, C., Hoffman, T. E., Jacobsen, N. K. \u0026amp; Spencer, S. L. Melanoma subpopulations that rapidly escape MAPK pathway inhibition incur DNA damage and rely on stress signalling. \u003cem\u003eNat Commun\u003c/em\u003e 12, 1747 (2021).\u003c/li\u003e\n \u003cli\u003eHsieh, A. C.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e The translational landscape of mTOR signalling steers cancer initiation and metastasis. \u003cem\u003eNature\u003c/em\u003e 485, 55-61 (2012).\u003c/li\u003e\n \u003cli\u003eMoore, K. N.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Phase I study of the investigational oral mTORC1/2 inhibitor sapanisertib (TAK-228): tolerability and food effects of a milled formulation in patients with advanced solid tumours. \u003cem\u003eESMO Open\u003c/em\u003e 3, e000291 (2018).\u003c/li\u003e\n \u003cli\u003ePatel, C.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Phase 1 Study to Evaluate the Effect of the Investigational Anticancer Agent Sapanisertib on the QTc Interval in Patients With Advanced Solid Tumors. \u003cem\u003eClin Pharmacol Drug Dev\u003c/em\u003e 9, 876-888 (2020).\u003c/li\u003e\n \u003cli\u003eWan, X., Harkavy, B., Shen, N., Grohar, P. \u0026amp; Helman, L. J. Rapamycin induces feedback activation of Akt signaling through an IGF-1R-dependent mechanism. \u003cem\u003eOncogene\u003c/em\u003e 26, 1932-1940 (2007).\u003c/li\u003e\n \u003cli\u003eYue, L.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Mthfd2 Modulates Mitochondrial Function and DNA Repair to Maintain the Pluripotency of Mouse Stem Cells. \u003cem\u003eStem Cell Reports\u003c/em\u003e 15, 529-545 (2020).\u003c/li\u003e\n \u003cli\u003ePerurena, N., Situ, L. \u0026amp; Cichowski, K. Combinatorial strategies to target RAS-driven cancers. \u003cem\u003eNat Rev Cancer\u003c/em\u003e 24, 316-337 (2024).\u003c/li\u003e\n \u003cli\u003eKrauthammer, M.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Exome sequencing identifies recurrent mutations in NF1 and RASopathy genes in sun-exposed melanomas. \u003cem\u003eNat Genet\u003c/em\u003e 47, 996-1002 (2015).\u003c/li\u003e\n \u003cli\u003eJanes, M. R.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Efficacy of the investigational mTOR kinase inhibitor MLN0128/INK128 in models of B-cell acute lymphoblastic leukemia. \u003cem\u003eLeukemia\u003c/em\u003e 27, 586-594 (2013).\u003c/li\u003e\n \u003cli\u003eNagasawa, I.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Disrupting ATF4 Expression Mechanisms Provides an Effective Strategy for BRAF-Targeted Melanoma Therapy. \u003cem\u003eiScience\u003c/em\u003e 23, 101028 (2020).\u003c/li\u003e\n \u003cli\u003eNeill, G. \u0026amp; Masson, G. R. A stay of execution: ATF4 regulation and potential outcomes for the integrated stress response. \u003cem\u003eFront Mol Neurosci\u003c/em\u003e 16, 1112253 (2023).\u003c/li\u003e\n \u003cli\u003ePetit, V.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e C57BL/6 congenic mouse NRAS(Q61K) melanoma cell lines are highly sensitive to the combination of Mek and Akt inhibitors in vitro and in vivo. \u003cem\u003ePigment Cell Melanoma Res\u003c/em\u003e 32, 829-841 (2019).\u003c/li\u003e\n \u003cli\u003eAckermann, J.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Metastasizing melanoma formation caused by expression of activated N-RasQ61K on an INK4a-deficient background. \u003cem\u003eCancer Res\u003c/em\u003e 65, 4005-4011 (2005).\u003c/li\u003e\n \u003cli\u003eArozarena, I. \u0026amp; Wellbrock, C. Phenotype plasticity as enabler of melanoma progression and therapy resistance. \u003cem\u003eNat Rev Cancer\u003c/em\u003e 19, 377-391 (2019).\u003c/li\u003e\n \u003cli\u003eKarras, P.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e A cellular hierarchy in melanoma uncouples growth and metastasis. \u003cem\u003eNature\u003c/em\u003e 610, 190-198 (2022).\u003c/li\u003e\n \u003cli\u003eWagle, M. C.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e A transcriptional MAPK Pathway Activity Score (MPAS) is a clinically relevant biomarker in multiple cancer types. \u003cem\u003eNPJ Precis Oncol\u003c/em\u003e 2, 7 (2018).\u003c/li\u003e\n \u003cli\u003eSigaud, R.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e MAPK inhibitor sensitivity scores predict sensitivity driven by the immune infiltration in pediatric low-grade gliomas. \u003cem\u003eNat Commun\u003c/em\u003e 14, 4533 (2023).\u003c/li\u003e\n \u003cli\u003ePathria, G.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Translational reprogramming marks adaptation to asparagine restriction in cancer. \u003cem\u003eNat Cell Biol\u003c/em\u003e 21, 1590-1603 (2019).\u003c/li\u003e\n \u003cli\u003eSimanshu, D. K., Nissley, D. V. \u0026amp; McCormick, F. RAS Proteins and Their Regulators in Human Disease. \u003cem\u003eCell\u003c/em\u003e 170, 17-33 (2017).\u003c/li\u003e\n \u003cli\u003eMarin-Bejar, O.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Evolutionary predictability of genetic versus nongenetic resistance to anticancer drugs in melanoma. \u003cem\u003eCancer Cell\u003c/em\u003e 39, 1135-1149 e1138 (2021).\u003c/li\u003e\n \u003cli\u003eRoos, W. P. \u0026amp; Kaina, B. DNA damage-induced cell death: from specific DNA lesions to the DNA damage response and apoptosis. \u003cem\u003eCancer Lett\u003c/em\u003e 332, 237-248 (2013).\u003c/li\u003e\n \u003cli\u003eRogakou, E. P., Pilch, D. R., Orr, A. H., Ivanova, V. S. \u0026amp; Bonner, W. M. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. \u003cem\u003eJ Biol Chem\u003c/em\u003e 273, 5858-5868 (1998).\u003c/li\u003e\n \u003cli\u003eFan, J.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Quantitative flux analysis reveals folate-dependent NADPH production. \u003cem\u003eNature\u003c/em\u003e 510, 298-302 (2014).\u003c/li\u003e\n \u003cli\u003eLocasale, J. W. Serine, glycine and one-carbon units: cancer metabolism in full circle. \u003cem\u003eNat Rev Cancer\u003c/em\u003e 13, 572-583 (2013).\u003c/li\u003e\n \u003cli\u003eTibbetts, A. S. \u0026amp; Appling, D. R. Compartmentalization of Mammalian folate-mediated one-carbon metabolism. \u003cem\u003eAnnu Rev Nutr\u003c/em\u003e 30, 57-81 (2010).\u003c/li\u003e\n \u003cli\u003eNishimura, T.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Cancer stem-like properties and gefitinib resistance are dependent on purine synthetic metabolism mediated by the mitochondrial enzyme MTHFD2. \u003cem\u003eOncogene\u003c/em\u003e 38, 2464-2481 (2019).\u003c/li\u003e\n \u003cli\u003eZhou, W.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Purine metabolism regulates DNA repair and therapy resistance in glioblastoma. \u003cem\u003eNat Commun\u003c/em\u003e 11, 3811 (2020).\u003c/li\u003e\n \u003cli\u003eTorrence, M. E.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e The mTORC1-mediated activation of ATF4 promotes protein and glutathione synthesis downstream of growth signals. \u003cem\u003eElife\u003c/em\u003e 10 (2021).\u003c/li\u003e\n \u003cli\u003eGao, Y., Feng, L., Zhang, L., Geng, J. \u0026amp; Zhang, E. ATF4/MYC Regulates MTHFD2 to Promote NSCLC Progression by Mediating Redox Homeostasis. \u003cem\u003eDis Markers\u003c/em\u003e 2022, 7527996 (2022).\u003c/li\u003e\n \u003cli\u003eNiessner, H.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e BRAF Inhibitors Amplify the Proapoptotic Activity of MEK Inhibitors by Inducing ER Stress in NRAS-Mutant Melanoma. \u003cem\u003eClin Cancer Res\u003c/em\u003e 23, 6203-6214 (2017).\u003c/li\u003e\n \u003cli\u003eNagasawa, I., Kunimasa, K., Tsukahara, S. \u0026amp; Tomida, A. BRAF-mutated cells activate GCN2-mediated integrated stress response as a cytoprotective mechanism in response to vemurafenib. \u003cem\u003eBiochem Biophys Res Commun\u003c/em\u003e 482, 1491-1497 (2017).\u003c/li\u003e\n \u003cli\u003eWang, S.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e ATF4 Gene Network Mediates Cellular Response to the Anticancer PAD Inhibitor YW3-56 in Triple-Negative Breast Cancer Cells. \u003cem\u003eMol Cancer Ther\u003c/em\u003e 14, 877-888 (2015).\u003c/li\u003e\n \u003cli\u003eOrd, T., Ord, D., Kaikkonen, M. U. \u0026amp; Ord, T. Pharmacological or TRIB3-Mediated Suppression of ATF4 Transcriptional Activity Promotes Hepatoma Cell Resistance to Proteasome Inhibitor Bortezomib. \u003cem\u003eCancers (Basel)\u003c/em\u003e 13 (2021).\u003c/li\u003e\n \u003cli\u003eHan, J.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e ER-stress-induced transcriptional regulation increases protein synthesis leading to cell death. \u003cem\u003eNat Cell Biol\u003c/em\u003e 15, 481-490 (2013).\u003c/li\u003e\n \u003cli\u003eFreundt, J. K.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e The Transcription Factor ATF4 Promotes Expression of Cell Stress Genes and Cardiomyocyte Death in a Cellular Model of Atrial Fibrillation. \u003cem\u003eBiomed Res Int\u003c/em\u003e 2018, 3694362 (2018).\u003c/li\u003e\n \u003cli\u003eBen-Sahra, I., Hoxhaj, G., Ricoult, S. J. H., Asara, J. M. \u0026amp; Manning, B. D. mTORC1 induces purine synthesis through control of the mitochondrial tetrahydrofolate cycle. \u003cem\u003eScience\u003c/em\u003e 351, 728-733 (2016).\u003c/li\u003e\n \u003cli\u003eZhu, Z. \u0026amp; Leung, G. K. K. More Than a Metabolic Enzyme: MTHFD2 as a Novel Target for Anticancer Therapy? \u003cem\u003eFront Oncol\u003c/em\u003e 10, 658 (2020).\u003c/li\u003e\n \u003cli\u003ePardo-Lorente, N. \u0026amp; Sdelci, S. MTHFD2 in healthy and cancer cells: Canonical and non-canonical functions. \u003cem\u003enpj Metabolic Health and Disease\u003c/em\u003e 2, 3 (2024).\u003c/li\u003e\n \u003cli\u003eZhu, L.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e MTHFD2 is a potential oncogene for its strong association with poor prognosis and high level of immune infiltrates in urothelial carcinomas of bladder. \u003cem\u003eBMC Cancer\u003c/em\u003e 22, 556 (2022).\u003c/li\u003e\n \u003cli\u003eRamos, L., Henriksson, M., Helleday, T. \u0026amp; Green, A. C. Targeting MTHFD2 to Exploit Cancer-Specific Metabolism and the DNA Damage Response. \u003cem\u003eCancer Res\u003c/em\u003e 84, 9-16 (2024).\u003c/li\u003e\n \u003cli\u003eKim, K. H.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Exploring Molecular Genetic Alterations and RAF Fusions in Melanoma: A Belvarafenib Expanded Access Program in Patients with RAS/RAF-Mutant Melanoma. \u003cem\u003eOncologist\u003c/em\u003e 29, e811-e821 (2024).\u003c/li\u003e\n \u003cli\u003eHuang, F.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Inhibiting the MNK1/2-eIF4E axis impairs melanoma phenotype switching and potentiates antitumor immune responses. \u003cem\u003eJ Clin Invest\u003c/em\u003e 131 (2021).\u003c/li\u003e\n \u003cli\u003eEwels, P. A.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e The nf-core framework for community-curated bioinformatics pipelines. \u003cem\u003eNat Biotechnol\u003c/em\u003e 38, 276-278 (2020).\u003c/li\u003e\n \u003cli\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol\u003c/em\u003e 15, 550 (2014).\u003c/li\u003e\n \u003cli\u003eYu, G., Wang, L. G., Han, Y. \u0026amp; He, Q. Y. clusterProfiler: an R package for comparing biological themes among gene clusters. \u003cem\u003eOmics\u003c/em\u003e 16, 284-287 (2012).\u003c/li\u003e\n \u003cli\u003ePetersen, K. H., Lohse, J. \u0026amp; Ramsgaard, L. Automated sequential chromogenic IHC double staining with two HRP substrates. \u003cem\u003ePLoS One\u003c/em\u003e 13, e0207867 (2018).\u003c/li\u003e\n \u003cli\u003eHao, Y.\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e Dictionary learning for integrative, multimodal and scalable single-cell analysis. \u003cem\u003eNat Biotechnol\u003c/em\u003e 42, 293-304 (2024).\u003c/li\u003e\n \u003cli\u003eHanzelmann, S., Castelo, R. \u0026amp; Guinney, J. GSVA: gene set variation analysis for microarray and RNA-seq data. \u003cem\u003eBMC Bioinformatics\u003c/em\u003e 14, 7 (2013).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"mTOR, belvarafenib, cobimetinib, sapanisertib, melanoma, targeted therapy, ATF4, MTHFD2","lastPublishedDoi":"10.21203/rs.3.rs-6857127/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6857127/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRAF monomer inhibitors are clinically approved for the treatment of BRAF\u003csup\u003eV600\u003c/sup\u003e-mutant melanoma in combination with a MEK inhibitor, but ineffective in other melanoma subtypes. Moreover, RAF dimer inhibitors, such as belvarafenib, when combined with MEK inhibitors (cobimetinib) have promising but limited efficacy in non-\u003cem\u003eBRAF\u003c/em\u003e-mutant melanomas. Here, we report that the mTOR inhibitor sapanisertib improves the efficacy of combined belvarafenib and cobimetinib therapy in \u003cem\u003eNRAS\u003c/em\u003e,\u003cem\u003e NF1\u003c/em\u003e,\u003cem\u003e \u003c/em\u003eand\u003cem\u003e KIT\u003c/em\u003e-mutant melanomas. Mechanistically, sapanisertib combined with belvarafenib and cobimetinib suppressed ATF4 expression and its target gene MTHFD2 while inducing DNA damage, revealing a previously underappreciated role of the ATF4-MTHFD2 axis in DNA damage repair and drug response. Human and murine models resistant to combined belvarafenib and cobimetinib exhibited elevated levels of ATF4 and MTHFD2 and were sensitive to sapanisertib. This study provides novel treatment opportunities for patients with non-\u003cem\u003eBRAF\u003c/em\u003e-mutant melanomas, or those who relapse following belvarafenib and cobimetinib combination therapy.\u003c/p\u003e","manuscriptTitle":"mTOR inhibition enhances the antitumor efficacy of RAF dimer-MEK blockade by inhibiting the ATF4-MTHFD2 pathway","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-18 13:12:11","doi":"10.21203/rs.3.rs-6857127/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":"0c036c49-4d4c-41fb-a4bc-620c006f5ecf","owner":[],"postedDate":"July 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":51471548,"name":"Health sciences/Oncology/Cancer/Skin cancer/Melanoma"},{"id":51471549,"name":"Biological sciences/Cancer/Skin cancer/Melanoma"}],"tags":[],"updatedAt":"2025-12-03T10:01:46+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-18 13:12:11","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6857127","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6857127","identity":"rs-6857127","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}