A lncRNA–miRNA axis regulates the balance between proliferative and invasive states in melanoma

A lncRNA–miRNA axis regulates the balance between proliferative and invasive states in melanoma | 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 A lncRNA–miRNA axis regulates the balance between proliferative and invasive states in melanoma Carlo Presutti, Lucrezia De Santis, Alessandro De Santis, Vito Antonio Amico, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9504736/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Melanoma is a highly aggressive malignancy characterized by significant phenotypic plasticity, allowing tumour cells to transit between different phenotypic states. This process, known as phenotype switching, drives metastasis and therapeutic resistance. While the primary transcriptional markers defining these states are well-established, the non-coding regulatory networks that stabilize these phenotypes remain incompletely understood. In this study, we identify the long non-coding RNAs (lncRNAs) LINC00504 and LINC00520 as key components of a multilayered regulatory circuit associated with the regulation of the proliferative/invasive melanoma subtypes. Functional assays reveal that experimental silencing of LINC00504 or LINC00520 in proliferative cells is sufficient to enhance the migratory and invasive capacity of melanoma cells, effectively mimicking the phenotype switch toward the invasive state. Mechanistically, these lincRNAs function as competing endogenous RNAs (ceRNAs) in the cytoplasm, where they sequester specific microRNAs to prevent the repression of mRNA targets, including MITF . Furthermore, we establish that MITF directly binds to the regulatory elements of both lincRNAs to drive their transcription, forming a positive feedback loop that reinforces the proliferative transcriptional program. Disruption of this MITF-lncRNA-miRNA axis facilitates the transition toward an invasive phenotype. Collectively, our findings highlight a novel integrated regulatory circuit that maintains melanoma cell state stability and suggest that targeting these non-coding components may provide a strategy to overcome phenotypic plasticity and therapeutic adaptability in melanoma. Biological sciences/Cancer/Tumour heterogeneity Biological sciences/Molecular biology/Epigenetics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 7 Figure 8 Introduction Melanoma is an aggressive skin cancer responsible for more than 70% of all skin cancer-related deaths, mostly due to its strong metastatic potential 1 , 2 . It is considered one of the most metastatic and therapy resistant malignancies largely because of its pronounced heterogeneity 3 , which arises from both genetic alterations and non-genetic mechanisms. Approximately 50% of cutaneous melanoma harbour BRAFV600E mutation 4 , which promotes tumour growth, survival, and invasiveness 5 . However, melanoma progression is not exclusively driven by genetic events, as epigenetic regulation, metabolic reprogramming, and developmental programs also play a crucial role in shaping tumour behaviour and therapeutic response 3 . Melanoma arises from malignant transformation and uncontrolled proliferation of melanocytes, the melanin-producing cells located in the basal layer of the epidermis. Melanocytes originate from neural crest cells that, during embryonic development, originate melanoblasts, an embryonic, highly migratory, and multipotent cell population capable of generating diverse lineages 6 . This neuroectodermal origin confers intrinsic plasticity, which is retained during tumorigenesis and enables melanoma cells to reversibly transition between differentiated and dedifferentiated states, a process commonly referred to as phenotype switching. Phenotype switching contributes to intratumoral heterogeneity and represents a mechanism underlying tumour progression and therapeutic resistance 7 . Transcriptomic analyses have identified two predominant and dynamically interchangeable cellular states, a “proliferative” (or melanocytic) and a “invasive” (or mesenchymal-like), which can coexist in the same tumour 8 . These states are defined by distinct transcriptional programs controlled by specific master regulators. The proliferative phenotype is characterized by high expression of Melanocyte Inducing Transcription Factor ( MITF ), which drives differentiation and cell growth 9 , whereas the invasive phenotype is associated with high expression of the receptor tyrosine kinase AXL and genes linked to migration, extracellular matrix remodelling, and stress responses 10 – 12 . Importantly, the invasive state is associated with metastatic dissemination and cross-resistance to both targeted therapies and immunotherapies 13 . However, emerging evidence indicates that melanoma cell states are not limited to two subtypes, but there are different intermediate phenotypes. These transitional populations may display distinct functional properties and contribute differentially to tumour growth, dissemination, and drug resistance 14 . The dynamic interconversion between these states is largely driven by epigenetic mechanisms, including DNA methylation, histone modifications, and non-coding RNAs (ncRNAs), which enable rapid and reversible adaptation to environmental and therapeutic pressures 15 . Among ncRNAs, long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) have emerged as key regulators of gene expression networks in cancer 16 . LncRNAs are transcripts longer than 200 nucleotides that regulate gene expression at epigenetic, transcriptional, and post-transcriptional levels, depending on their cellular localization 17 . Several lncRNAs have been implicated in tumor progression, where they can function as oncogenes or tumor suppressors. LINC00504 is not characterized in melanoma, but its role has been detected in other tumours, such as non-small cell lung, ovarian, colon, and breast cancers 18 , 19 . It has an oncogenic role in favouring tumour progression. Patients with high expression of LINC00504 are associated with poor prognosis and it may be a new biomarker for breast cancer 20 . LINC00520 is widely expressed in various tissues, and it is upregulated in 11 cancers including lung cancer 21 , non-small cell lung cancer (NSCLC) 22 , breast cancer (BC) 23 , and melanoma (MM) 24 . Notably, its role in melanoma appears to be context dependent. In primary melanoma, higher LINC00520 expression has been associated with better survival, whereas in metastatic samples its upregulation correlates with poor prognosis 25 . These observations suggest that LINC00520 may play distinct roles depending on disease stage and cellular state, potentially reflecting the underlying phenotypic plasticity of melanoma cells. Mechanistically, both lncRNAs have been reported to function as competing endogenous RNAs (ceRNAs), acting as molecular sponges that modulate miRNA availability in different cancer system, such as breast cancer 20 , ovarian cancer 26 , hepatocellular carcinoma 27 and colorectal cancer 28 . microRNAs (miRNAs) are small non-coding RNA with a length of ∼18–22 nucleotides that post-transcriptionally repress gene expression by binding to complementary sequences in target mRNAs 29 . They regulate a broad spectrum of cellular processes, and their dysregulation contributes to cancer initiation, progression, and therapy response 30 . MiRNAs are aberrantly expressed in many tumours depending on cancer type, stage, and other clinical variables. Depending on the cellular context, miRNAs may act as oncogenes or tumour suppressors 31 . Current therapeutic strategies for melanoma include surgical resection for localized disease and systemic treatments for advanced stages, such as targeted therapy, immunotherapy, and chemotherapy 32 . Despite significant clinical improvements, therapeutic resistance remains a major challenge. Recently, treatment with the combination of Romidepsin and Interferon-α2b (RI) has been shown to induce growth arrest and cell death in melanoma cells 33 . Transcriptomic analysis of primary melanoma cells revealed distinct clusters corresponding to the proliferative and invasive states 34 and RI treatment was found to promote a shift from a proliferative toward a more invasive transcriptional phenotype. Results Identification of LINC00504 and LINC00520 in Proliferative cell state The re-analysis of the differentially expressed genes (DEGs) from the GSE221386 dataset indicated that two lincRNAs, LINC00504 and LINC00520 , are strongly downregulated after the treatment with Romidepsin and Interferon-α2b (RI) in the proliferative cells (Fig. 1 A). This data was also validated using Real-time quantitative polymerase chain reaction (RT-qPCR) (Fig. 1 B) in the same primary melanoma cells, confirming a significant reduction in the expression of both lincRNAs in these samples. Given that RI treatment has been shown to trigger a transcriptional shift from a proliferative toward an invasive-like phenotype, the observed downregulation of these lincRNAs suggests they could potentially be associated with this phenotypic remodelling. To further investigate this association in a clinical context, their expression levels were investigated in melanoma patients from The Cancer Genome Atlas (TCGA) cohort. After stratifying melanoma patient samples into proliferative or invasive subtypes based on their transcriptional signatures, it was observed that LINC00504 and LINC00520 are highly expressed in proliferative tumour compared to the invasive one (Fig. 1 C). Furthermore, a correlation analysis within the TCGA dataset was performed to evaluate the potential coordinated regulation of these transcripts. A strong and statistically significant correlation was found between the expression levels of LINC00504 and LINC00520 (Fig. 1 D). To further explore the clinical implications of these findings, we evaluated the prognostic relevance of both lincRNAs using TCGA survival data through the GEPIA platform. We observed that higher expression of these transcripts is associated with poorer patient prognosis (Supplementary Fig. S1 ). Selection of a Proliferative Cellular Model and Subcellular Localization Analysis Given that LINC00504 and LINC00520 were found to be highly expressed in the proliferative state, it was essential to identify an appropriate in vitro model that closely mimics this phenotypic state in melanoma cells. To this aim, different immortalized melanoma cells (A375, CHL1, SKmel28, and 624-mel) were screened for the expression of established proliferative markers, such as MITF and MLANA , and the two lincRNAs. This characterization aimed to identify a cell line exhibiting a proliferative signature suitable for subsequent functional studies. Analysis of mRNA levels revealed that the 624-mel cell line exhibited the highest expression of both proliferative markers and lincRNAs (Fig. 2 A-B), thereby serving as the model for further investigation. Following the selection of the cellular model, the subcellular localization of LINC00504 and LINC00520 was investigated in 624-mel cells, as the spatial distribution of lncRNAs is often indicative of their biological function. Two complementary approaches were employed to ensure technical robustness: biochemical cellular fractionation followed by RT-qPCR, and RNA Fluorescence In Situ Hybridization (RNA-FISH). The results from both methods demonstrated that LINC00504 is predominantly localized within the cytoplasm, whereas LINC00520 exhibits a dual distribution in both the nucleus and the cytoplasm (Fig. 2 C-D). The distinct cellular distribution of these two lncRNAs suggest they may act through different molecular mechanisms based on their cellular localization. Identification of the lncRNA–miRNA–mRNA Regulatory Network To explore the possible biological function of LINC00504 and LINC00520 as miRNA sponge in melanoma, particularly regarding the cytoplasmatic fraction, computational analysis was performed to identify potential miRNA interactors. Using the prediction algorithms miRcode and RNA22, a set of candidate miRNAs with putative binding sites for these lincRNAs was identified. Among the predicted candidates, 6 miRNAs (miR-23a-3p, miR-24-3p, miR-27a-3p, miR-27b-3p, miR-31-5p, and miR-125b-5p) were selected for further investigation based on their binding affinity, as illustrated in the heatmap in Fig. 3 A. To evaluate the clinical relevance of these findings, miRNAs’ expression was analysed in the TCGA-SKCM cohort, using the previously described stratification in proliferative and invasive subtypes. The miRNAs showed significantly higher expression levels in the invasive subtype compared to the proliferative one (Fig. 3 B). This pattern is in direct contrast to the expression of LINC00504 and LINC00520 (Fig. 1 C), which are enriched in proliferative samples. The inverse correlation in patient data supports the hypothesis that these lincRNAs and miRNAs may be part of a coordinated regulatory network involved in melanoma phenotype switching. To further delineate the downstream effects of this regulatory axis, a second computational screening was performed to identify potential mRNA targets for the previously selected miRNAs. By integrating data from multiple bioinformatic software, including Mienturnet, TargetScan, and miRTarBase, we identified a specific set of genes likely regulated by these miRNAs. Notably, these targets include master regulators of the melanocyte lineage, such as MITF and MLANA . The complexity of this regulatory network is summarized in Fig. 3 C, which presents both a schematic representation of the proposed regulatory network and a summary table of the identified miRNA–mRNA interactions. To explore the clinical relevance of this network, the expression of these target mRNAs was evaluated across the TCGA-SKCM cohort. In agreement with the proposed model, many of the identified mRNA targets displayed significantly higher expression levels in the proliferative subtype compared to the invasive one (Fig. 3 D). Notably, their expression profiles show a direct correlation with the levels of the two lincRNAs (Fig. 1 C) and a consistent inverse correlation with their targeting miRNAs (Fig. 3 B). This highly coordinated expression pattern across patient samples strongly suggests the existence of a functional lincRNA–miRNA–mRNA axis that characterizes the proliferative phenotype in melanoma. Experimental Validation of lincRNA–miRNA Physical Interaction To provide evidence of a direct interaction between the lncRNAs and the candidate miRNAs, we performed an RNA pull-down assay followed by RT-qPCR. To this end, we utilized the same biotinylated oligonucleotide probes previously employed for RNA-FISH (Fig. 2 D), which were designed to specifically hybridize with LINC00504 and LINC00520 . As shown in Fig. 4 A, the pull-down efficiency was confirmed by the significant and specific enrichment of both lincRNAs. We analysed the co-enriched small RNA fraction to detect the presence of the predicted miRNA interactors. The RT-qPCR analysis revealed that a subset of the candidate miRNAs was significantly pulled down alongside the lincRNAs, suggesting a direct physical association within the cellular environment. Specifically, we observed a robust enrichment for miR-23a-3p , miR-24-3p , miR-31-5p , and miR-125b-5p (Fig. 4 B). Conversely, although miR-27a-3p and miR-27b-3p did not show a statistically significant enrichment under these specific experimental conditions, they were nonetheless retained for further functional investigations. Overall, these results substantiate the hypothesis that LINC00504 and LINC00520 act as molecular platforms for miRNA binding. LincRNA Silencing Promotes an Invasive-like Phenotypic Shift As mentioned above, treatment with RI induces a phenotype switch in melanoma cells, driving them from a proliferative to an invasive state. Under the same conditions, we observed a significant downregulation of both LINC00504 and LINC00520 . To investigate whether their reduced expression contributes to this transition, we acted directly on the lincRNAs by silencing them via RNA interference (RNAi). The silencing efficiency was confirmed by RT-qPCR, which demonstrated a substantial reduction in the levels of both transcripts (Fig. 5 A). We then evaluated if the direct depletion of these lincRNAs was sufficient to trigger the functional hallmarks of the invasive phenotype, studying cell migratory and invasive properties. Transwell migration assays were performed using uncoated inserts (Fig. 5 B-C), while invasion assays were conducted using gel-coated chambers (Fig. 5 D-E). Migration and invasion assays revealed that silencing either LINC00504 or LINC00520 resulted in a significant increase in both migratory and invasive capacities. Notably, this enhanced cellular motility occurred without any significant impact on cell proliferation (data not shown), suggesting that the downregulation of LINC00504 and LINC00520 is not merely a consequence of the phenotype switch, but an event that contributes to the acquisition of a more invasive profile. To understand the mechanisms behind this functional change, we next examined the molecular consequences of lincRNA silencing on the previously identified regulatory network. Molecular Disruption of the ceRNA Network Following lincRNA Silencing To investigate the molecular mechanisms underlying the observed phenotypic shift, we examined how the silencing of LINC00504 and LINC00520 impacts the downstream components of the predicted regulatory network. Following the RNAi-mediated knockdown of LINC00504 and LINC00520 , which was first validated via RT-qPCR (Fig. 5 A), we investigated the resulting changes in the expression levels of the previously identified miRNAs and their mRNA targets. We observed that the candidate miRNAs were significantly upregulated following the silencing of LINC00520 (Fig. 6 A). Under the same experimental conditions, we registered an efficient reduction in the expression of the mRNA targets, as well as a decrease in the levels of established proliferative markers (Fig. 6 B). However, some mRNA targets show a reduction in their expression levels only after the silencing of one of the lincRNAs (Fig. 6 C). To further validate these findings at the protein level, we analysed the expression of two specific targets. We selected Nuclear Receptor Binding SET Domain Protein 2 ( NSD2 ) (Fig. 6 D), which in our previous work was described as a direct target of miR-23a, miR-24, and miR-31 within the same proliferative subtype, and MITF (Fig. 6 E), as it is the most established proliferative marker in melanoma. In line with the transcriptomic data, we observed a consistent reduction in protein expression levels following the silencing of both lincRNAs. MITF Directly Regulates the Expression of LINC00504 and LINC00520 To investigate the transcriptional regulation of LINC00504 and LINC00520 , we performed an integrative analysis of transcriptomic and epigenomic data 25 , 35 (GSE137776, GSE61965, GSE94488, ENCODE ENCSR008SDL, GSE167496, GSE61966, GSE163646, GSE283855). First, we assessed the impact of MITF perturbation on lncRNA expression levels across multiple independent RNA-seq datasets. As shown in Fig. 7 A, both lncRNAs were significantly downregulated following MITF silencing (via siRNA, shRNA, and CRISPR-KO) and markedly upregulated upon MITF overexpression, suggesting a positive regulatory relationship. This observation was further supported by expression correlation analysis in large-scale melanoma cohorts (Fig. 7 B). We found a strong positive correlation (Spearman 𝜌) between MITF and both lncRNAs in cutaneous melanoma cell lines (CCLE) and a consistent correlation in patient tumours (TCGA-SKCM). To determine whether this regulation is direct, we analysed the genomic landscape surrounding both loci using ATAC-seq and DNase-seq data (Fig. 7 C-D). We initially screened a ± 200 kb window to identify Candidate Regulatory Elements (CREs) characterized by high chromatin accessibility and active enhancer marks (H3K27ac). Notably, publicly available ChIP-seq datasets showed variable MITF occupancy across these loci. The integration of these epigenomic tracks highlighted that the regions immediately upstream of the Transcription Start Site (TSS) also exhibited strong signatures of open chromatin and regulatory activity. Given that the proximal promoter is fundamental for the initiation of transcription, we focused our experimental validation on these proximal regulatory regions (~ 2 kb upstream of the TSS). This approach allowed us to prioritize these upstream sequences for experimental validation, testing whether MITF physically occupies these proximal regions to directly modulate the core promoter activity and functional expression of both lincRNAs. This coordinated expression reflects the broader transcriptional program governed by MITF ; indeed, functional enrichment analysis of MITF-target genes confirms a dual role in activating pigmentation and differentiation pathways while simultaneously repressing genes associated with cell adhesion and mesenchymal development (Supplementary Fig. S2 ). Validation of the MITF-lincRNA Regulatory Axis To validate the direct binding of MITF to the regulatory elements of LINC00504 and LINC00520 , we performed Chromatin Immunoprecipitation (ChIP) assays in melanoma cells. The immunoprecipitated DNA fractions were analysed via both semi-quantitative PCR and quantitative real-time PCR (qPCR). Agarose gel electrophoresis of the PCR products showed a clear enrichment for the target genomic regions in the anti-MITF fractions compared to the Rabbit IgG negative control (Fig. 8 A). These results were further supported by ChIP-qPCR analysis (Fig. 8 B), which demonstrated a significant fold-enrichment of both lincRNAs, confirming that MITF physically occupies these loci. To assess the functional impact of MITF on the expression of these transcripts, we performed gain-of-function experiments. We generated a MITF overexpression construct by cloning the MITF transcript (ENST00000394351.9) into a pIRESneo-FLAG/HA vector, replacing the EYFP sequence. This plasmid was transfected into the A375 melanoma cell line, which was selected for its relatively low basal expression of MITF, providing an ideal model to observe transcriptional induction. Following transfection, Western blot analysis confirmed the successful overexpression of the MITF protein (Fig. 8 C). Subsequent RT-qPCR analysis revealed a significant increase in the steady-state mRNA levels of both LINC00504 and LINC00520 ( Fig. 8 D ) . Taken together, these data demonstrate that MITF not only binds to the regulatory regions of these lincRNAs but also actively drives their transcriptional upregulation, establishing a direct regulatory link. Discussion We identify LINC00504 and LINC00520 as components of a regulatory network associated with the proliferative state of melanoma cells. Both lncRNAs are highly expressed in proliferative melanoma subtype and localize to both nuclear and cytoplasmic compartments, suggesting multifunctional roles. Our findings support a model in which these transcripts contribute to the maintenance of the proliferative phenotype through coordinated transcriptional and post-transcriptional mechanisms. At the cytoplasmic level, LINC00504 and LINC00520 act as competing endogenous RNAs, sequestering specific miRNAs and limiting their ability to repress downstream targets. This sponging activity correlates with increased expression of shared mRNA targets, including MITF , a master regulator of the melanocytic lineage and a key determinant of the proliferative state. The positive association between lncRNA levels and target mRNA expression, together with the inverse relationship observed with miRNA abundance, supports the existence of a functional lncRNA–miRNA–mRNA regulatory axis in proliferative melanoma cells. Importantly, our data extend this post-transcriptional model by identifying MITF as an upstream transcriptional regulator of LINC00504 and LINC00520 . MITF binding to their promoter regions suggests the existence of a positive feedback loop in which high MITF activity promotes lncRNAs transcription, while the lncRNAs sustain MITF expression indirectly through miRNA sequestration. Disruption of this network, either during phenotype switching toward the invasive state or following experimental silencing of the lncRNAs, leads to reduced lncRNAs expression, release of miRNAs from sequestration, and subsequent repression of target mRNAs, including MITF . The resulting decrease in MITF protein levels likely contributes to the collapse of the proliferative program and prevents reactivation of the lncRNA-mediated circuit. These observations are consistent with the dynamic and reversible nature of melanoma cell states, where regulatory networks rather than fixed genetic alterations can govern phenotypic transitions. Melanoma plasticity, rooted in the neural crest origin of melanocytes, allows tumour cells to oscillate between transcriptional states in response to intrinsic and extrinsic stimuli. Our findings suggest that the absence of LINC00504 and LINC00520 contribute to the acquisition of a more invasive cell state, while their presence stabilize the proliferative state within this plastic framework. Given that phenotype switching is closely linked to therapeutic resistance, regulatory axes involving non-coding RNAs may represent critical modulators of melanoma adaptability. Overall, this work highlights a multilayered regulatory circuit in which MITF and lncRNAs cooperate to sustain the proliferative phenotype through coordinated transcriptional and post-transcriptional control. Targeting components of this network may offer new opportunities to interfere with melanoma cell state stability and overcome resistance mechanisms. Materials and Methods Bioinformatic Analysis Publicly available transcriptomic datasets were analysed to evaluate the expression profiles of long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and target messenger RNAs (mRNAs) in melanoma. RNA-sequencing (TPM) and miRNA-sequencing (RPM) data from the TCGA-SKCM cohort were retrieved and processed. Samples were stratified into "Proliferative" (Melanocytic and Transitory subtypes) and "Invasive" (Neural crest-like and Undifferentiated subtypes) classes based on established molecular signatures to assess expression differences across melanoma progression states. Additionally, differential expression analysis was performed on the GSE221386 dataset, with results visualized via volcano plots generated using the ggplot2 R package. To characterize the regulatory network of the selected miRNAs, we utilized the scanMiR R/Bioconductor package. miRNA seed matches were predicted across the sequences of LINC00504 , LINC00520 . Binding site affinity was quantified as an aggregated binding affinity, which integrates site type and context to estimate the regulatory impact of each miRNA on its targets. Expression levels across TCGA subtypes were compared using the Wilcoxon rank-sum test, with p-values visualized using the ggpubr package. Heatmaps representing the predicted repression scores for both lncRNAs and coding genes were generated using ggplot2 and pheatmap, employing color gradients to represent binding potency. To investigate the transcriptional control of LINC00504 and LINC00520 by MITF , we performed an integrative analysis of transcriptomic and epigenomic data. The impact of MITF on lincRNA expression was assessed across multiple independent RNA-seq datasets involving MITF perturbation (siRNA, shRNA, CRISPR-KO, and overexpression). Correlation of expression (Spearman 𝜌) between MITF and the lincRNAs was calculated using the CCLE (Cancer Cell Line Encyclopedia) and TCGA-SKCM cohorts. To identify direct binding sites, we analysed the genomic landscape surrounding the lincRNA loci (± 200 kb from the TSS) using ATAC-seq, and DNase-seq data retrieved from public data. Candidate Regulatory Elements (CREs) were identified based on high chromatin accessibility and active enhancer marks (H3K27ac). All primary data are public. Accessions, cell lines, and source: GSE137776, GSE61965, GSE94488, ENCODE ENCSR008SDL, GSE167496, GSE61966, GSE163646, GSE283855. All computational analyses were conducted in the R statistical environment (v4.x). Cell culture In this study were used human melanoma cell lines. The human melanoma cell lines A375, CHL1, and 624-mel were kindly provided by Antonio Filippini (University of Rome, Sapienza), while SKmel28 was a gift from Lucia Gabriele (Istituto Superiore di Sanità, Rome). A375, CHL1, and 624-mel were cultured in high-glucose DMEM with sodium pyruvate (Corning, Catalog Number: 15-013-CV) with 1 mmol/L L-glutamine (Gibco, Catalog number: 25030081), 100 U/mL penicillin, and 100 µg/mL streptomycin (Corning, Catalog number: 30-002-CI), and 10% Fetal Bovine Serum (Corning, Catalog number: 35-015-CV). SKmel28 was cultured in RPMI 1640 medium (Corning, Catalog Number: 15-040-CV) supplemented with 1 mmol/L L-glutamine, 100 U/mL penicillin and 100 µg/mL streptomycin, and 10% Fetal Bovine Serum. Cells were incubated in the humidified incubator at 37°C in a 5% CO 2 humidified atmosphere, and periodically checked for Mycoplasma. Quantitative RT-PCR RNA extraction was performed using QIAzol reagent and the miRNeasy Mini Kit (QIAGEN, Catalog number 217004) according to the manufacturer’s instructions. Reverse transcription was conducted using SensiFAST cDNA Synthesis Kit (Bioline, Catalog number: BIO-65054) and SuperScript™ IV Reverse Transcriptase (Invitrogen, Catalog number: 18090050) for mRNA; for miRNA were used miRCURY LNA RT Kit (QIAGEN, Catalog number: 339340). The Real Time PCR was performed with SensiFAST SYBR Hi-ROX Kit (Bioline, Catalog number: BIO-92020) for mRNA and miRCURY LNA SYBR Green PCR Kit (QIAGEN: Catalog number: 339346) for miRNA. The reactions were conducted by using the StepOnePlus System (Applied Biosystems) according to the set reaction conditions. The HPRT gene and miR191-5p were selected as the reference gene for normalization of mRNA and miRNA analysis, respectively. Western blot Cells were lysed with RIPA Lysis and Extraction Buffer (Thermo Scientific™, Catalog number: 89901) added with protease inhibitors cOmplete™ EDTA-free Protease Inhibitor Cocktail (Merck, Catalog number: 11873580001) and phosphatase inhibitors PhosSTOP™ (Merck, Catalog number: 4906845001). The concentration of extracted protein was measured by Pierce™ Detergent Compatible Bradford Assay Kit (Thermo Scientific™, Catalog number: 23246). The total protein was separated in 8% and 10% PAGE prepared from Acrylamide/Bis-acrylamide, 40% solution (19:1) (Merck, Catalog number: A9926-5X100ML). Proteins were transferred to nitrocellulose membranes (Amersham Protran WB membrane, Merck; Catalog number: GE10600007). Antibodies against NSD2 (Cell Signaling Technology, D4Z8Q, 1:7 000), HPRT (Prodotti Gianni srl, EPR5299, 1:10 000), MITF (Cell Signaling, D3B4T Rabbit mAb #97800), and HA (Invitrogen, Mouse, Catalog number: 26183, 1:5 000) were incubated at 4°C overnight. Secondary HRP peroxidase-conjugated goat anti-rabbit Ab (Thermo Fisher Scientific Cat#31460, RRID: AB_228341, 1:10 000) and goat anti-mouse (Thermo Fisher Scientific Cat# 31430, RRID: AB_228307, 1:10 000) were used. The bands were detected using Clarity Western ECL Substrate (BIORAD Catalog number: BRD1705061) and ChemiDoc MP Imaging System (BIORAD). PageRuler™ Prestained Protein Ladder (Thermo Scientific, Catalog number 26616) was used as marker. Cell fractionation To perform cellular fractionation was used the hypotonic buffer lysis A (HLB): Tris (20 mM, pH 8.0), NaCl (10 mM), MgCl2 (3 mM), EDTA (0.2 mM), NP40 (0.1%), and fresh DTT (1 mM), cOmplete™ EDTA-free Protease Inhibitor Cocktail (1x) (Merck, Catalog Number: 11873580001), and RNaseOUT™ Recombinant Ribonuclease Inhibitor (1 U/µl) (Invitrogen, Catalog Number: 10777019). Cultured cells were resuspended in PBS and centrifugated to resuspend the pellet in HLB. With following centrifuge and wash, nucleus and cytoplasm can be separated in the pellet and supernatant respectively. Briefly, the pellet was resuspended in HLB, incubate in ice for 5 minutes and centrifugated 400xg 5 minutes at 4°C. Repeat this steps and collect the supernatant after the second centrifugation because it contains the cytoplasm fraction. Wash twice the pellet with HLB and collect the pellet after the second centrifugation; it is the nucleus fraction. The fractionation is followed by RNA extraction, retro transcription and qPCR. RNA FISH RNA FISH was conducted as previously described 36 . Were used 16 and 20 probes with 5’ biotin complementary with LINC00504 and LINC00520 respectively. Probes were designed using Stellaris tool and ordered from Bio-Fab research Srl. For the lncRNA visualization, probes were incubated with the coverslip overnight at 37°C in a humid box. The next day Streptavidin, Alexa Fluor™ 488 Conjugateanti-biotin (Invitrogen™, Catalog number: S32354) were incubated 1h in a humid box. After the staining with DAPI, the signals were acquired with Nikon Eclipse 50i (100x magnification, 385nm and 475nm filters). RNA pulldown assay RNA pulldown was conducted as previously described 37 . RNA pull-down experiments were performed using 5′-biotinylated antisense DNA probes complementary to the target lncRNAs, also used in the RNA FISH. A 5′-biotinylated probe against lacZ, a non-human bacterial gene, was employed as a negative control to monitor non-specific binding and validate probe specificity. Cells were lysed using Lysis buffer (LB): Tris-HCl pH 7.5 50 mM, NaCl 150 mM, MgCl2 3 mM, NP40 0.5% (IGEPAL® CA-630, Sigma-Aldrich, Catalog number: I8896), EDTA 2 mM, and fresh DTT 1 mM, 1× PIC (cOmplete™ Protease Inhibitor Cocktail, Roche, Catalog number: 11697498001), and RNaseOUT™ Recombinant Ribonuclease Inhibitor (0.2 U/µl) (Invitrogen, Catalog number: 10777019). After the lysis, collect at least 1 mg of total extract protein for each pulldown (PD) condition and 0,1 mg for the input (10% of PD). Subsequently, sample were diluted with Hybridization buffer (HB): Tris-HCl pH 7.5 50 mM, NaCl 150 mM, MgCl2 3 mM, NP40 0.5%, EDTA 2 mM, Dextran Sulfate Salt DSS (2,5%) (Merck, Catalog Number: D8906-5G), and fresh DTT 1 mM, 1× PIC, and RNase Inhibitors (0.2 U/µl). Add the probes and incubate rotating for at least 4h at + 4°C. Then, add Streptavidin MagneSphere® Paramagnetic Particles (Promega, Catalog number: Z5481) pre-washed with HB and incubate 1h at room temperature. After 4 wash using magnetic rack, add QIAzol and detach the beads by vortex the samples. Proceed with RNA extraction, retro transcription using SuperScript™ IV Reverse Transcriptase and qPCR. RNA interference To silence LINC00520 were used the FlexiTube GeneSolution GS645687 from QIAGEN, while for LINC00504 were constructed siRNA custom made from QIAGEN. The negative control is Negative Control siRNA (20 nmol) (QIAGEN, Catalog number: 1027310). After the seeding of the cells 24h before transfection, cells were treated with Lipofectamine™ RNAiMAX Transfection Reagent (Invitrogen, Catalog number: 13778150) and siRNA. 48h after the transfection, the RNA was extracted following the protocol, retro transcribed with SensiFAST cDNA Synthesis Kit and analysed with qPCR. Cell invasion and migration assay Transwell assay was used to detect the invasiveness of melanoma cell. 48h after the transfection with siRNA against the lincRNAs, cells were counted and were placed on Geltrex-coated TC-inserts (Geltrex Gibco A14132-02; TC-inserts Sarstedt 83.3932.800). In the transwell, 100,000 cells were seeded in DMEM 2% FBS. The culture medium with 20% fetal bovine serum was used as chemical attractant and it was added to the lower chamber of the well. After 8h, the invasive cells were fixed and stained with crystal violet (Sigma-Aldrich, Catalog Number: S-C0775-25G). The number of invasive cells was counted in ten random fields under 20x magnification, and the mean for each condition was determined. For the migration assay, transwell inserts were used following the same protocol, but without the Geltrex coating. Migrated cells were fixed and coloured with crystal violet, and they were counted in ten random fields under 20x magnification. Plasmids and transfections The MITF overexpressing plasmid was constructed by inserting the full length of MITF ENST00000394351.9 isoform into pIRESneo-FLAG/HA EYFP (AddGene, Plasmid #10825) replacing the EYFP . pIRESneo-FLAG/HA EYFP construct was used as control. pIRESneo-FLAG/HA MITF and pIRESneo-FLAG/HA EYFP were transfected into cells using Lipofectamine™ 3000 Transfection Reagent (Invitrogen™, Catalog number: L3000001). 24 hours after transfection, MITF overexpression was validated using Western Blot. Moreover, total RNA was extracted to analyse changes in expression levels of LINC00504 and LINC00520 . Chromatin Immuno Precipitation Chromatin Immuno Precipitation was performed using MAGnify™ Chromatin Immunoprecipitation System (Applied Biosystems™, Catalog number: 492024) following the protocol of the kit. After 624-mel cell lysis, the sonication was performed using Bioruptor® Pico (Diagenode, B01080010) with 12 cycles 30” ON/30” OFF. We used antibody against MITF (Cell Signaling, D3B4T Rabbit mAb #97800), H3K4Me3 (Cell signaling, C42D8 Rabbit mAb #9751), and IgG Rabbit (provided with the kit). The DNA immunoprecipitated was analysed using PCR and qPCR with primers against the promoter region of LINC00504 and LINC00520 . Declarations Conflicts of Interest The authors declare no conflicts of interest. Author Contributions Conceptualization, LDS and CP. Study design, LDS, LG and CP. Experiments, LDS, VAA, and ST. Project administration, funding acquisition and resources, CP. Bioinformatics analysis, ADS. Supervision, CP. Writing of original draft, LDS and CP. Editing, LDS, AF, LG and CP. All authors had access to the study data and approved the final submission. Acknowledgements This work was supported by the PNRR Project CN3-National Center for Gene Therapy and Drugs based on RNA Technology (Spoke 3). Primary melanoma cells were kindly provided by Doctor Stefania D’Atri, Molecular Oncology Laboratory, Istituto Dermopatico Dell’Immacolata IDI-IRCCS, Rome, Italy. We thank Sapienza University of Rome for providing the facilities and resources that made this study possible. We are also grateful to our colleagues for their insightful feedback and suggestions during the development of this project. Data Availability Statement Full length western blots are available as Supplementary File. Data sharing is not applicable to this article as no new data were created or analyzed in this study. References World Cancer Research Fund. World Cancer Research Fund https://www.wcrf.org/ . Li, F. Z., Dhillon, A. S., Anderson, R. L., McArthur, G. & Ferrao, P. T. Phenotype Switching in Melanoma: Implications for Progression and Therapy. Front. Oncol. 5, 126186 (2015). Rambow, F., Marine, J.-C. & Goding, C. R. Melanoma plasticity and phenotypic diversity: therapeutic barriers and opportunities. Genes & Development 33, 1295 (2019). Pillai, M. & Jolly, M. K. Systems-level network modeling deciphers the master regulators of phenotypic plasticity and heterogeneity in melanoma. iScience 24, 103111 (2021). Smalley, K. S. M. & Herlyn, M. Loitering with Intent: New Evidence for the Role of BRAF Mutations in the Proliferation of Melanocytic Lesions. J Invest Dermatol 123, xvi–xvii (2004). J, T., L, R., K, P., C, G., Km, S., J, L., et al. Multi-stage Differentiation Defines Melanoma Subtypes with Differential Vulnerability to Drug-Induced Iron-Dependent Oxidative Stress. Cancer cell 33, (2018). Huang, F., Santinon, F., González, R. E. F. & Rincón, S. V. del. Melanoma Plasticity: Promoter of Metastasis and Resistance to Therapy. Frontiers in Oncology 11, (2021). Tirosh, I., Izar, B., Prakadan, S. M., Marc H Wadsworth, I. I., Treacy, D., Trombetta, J. J., et al. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq. Science (New York, N.Y.) 352, 189 (2016). MLANA/MART1 and SILV/PMEL17/GP100 Are Transcriptionally Regulated by MITF in Melanocytes and Melanoma. The American Journal of Pathology 163, 333–343 (2003). Hoek, K. S., Schlegel, N. C., Brafford, P., Sucker, A., Ugurel, S., Kumar, R., et al. Metastatic potential of melanomas defined by specific gene expression profiles with no BRAF signature. Pigment Cell Research 19, 290–302 (2006). I, A. & C, W. Phenotype plasticity as enabler of melanoma progression and therapy resistance. Nature reviews. Cancer 19, (2019). M, S., M, C., G, C., G, N., F, A., G, T., et al. Human cutaneous melanomas lacking MITF and melanocyte differentiation antigens express a functional Axl receptor kinase. The Journal of investigative dermatology 131, (2011). J, L., J, K., M, R., T, B., M, R., M, C., et al. Melanomas resist T-cell therapy through inflammation-induced reversible dedifferentiation. Nature 490, (2012). Karras, P., Bordeu, I., Pozniak, J., Nowosad, A., Pazzi, C., Van Raemdonck, N., et al. A cellular hierarchy in melanoma uncouples growth and metastasis. Nature 610, 190–198 (2022). N, S., S, G., E, D. S., N, V. & G, B. Epithelial-to-Mesenchymal Transition: Epigenetic Reprogramming Driving Cellular Plasticity. Trends in genetics: TIG 33, (2017). Schmitt, A. M. & Chang, H. Y. Long Noncoding RNAs in Cancer Pathways. Cancer cell 29, 452 (2016). Akhade, V. S., Pal, D. & Kanduri, C. Long Noncoding RNA: Genome Organization and Mechanism of Action. Long Non Coding RNA Biology 47–74 (2017) doi: 10.1007/978-981-10-5203-3_2 . Feng, J., Ma, J., Liu, S., Wang, J. & Chen, Y. A noncoding RNA LINC00504 interacts with c-Myc to regulate tumor metabolism in colon cancer. Journal of Cellular Biochemistry 120, 14725–14734 (2019). Feng, J., Li, Y., Zhu, L., Zhao, Q., Li, D., Li, Y., et al. STAT1 mediated long non-coding RNA LINC00504 influences radio-sensitivity of breast cancer via binding to TAF15 and stabilizing CPEB2 expression. Cancer Biology & Therapy 22, 630 (2021). Hou, T., Ye, L. & Wu, S. Knockdown of LINC00504 Inhibits the Proliferation and Invasion of Breast Cancer via the Downregulation of miR-140-5p. Onco Targets Ther 14, 3991–4003 (2021). Xia, G., Li, X., Chen, F. & Shao, Z. LncRNA LINC00520 Predicts Poor Prognosis and Promotes Progression of Lung Cancer by Inhibiting MiR-3175 Expression. Cancer Management and Research 12, 5741 (2020). Jf, W., Zn, X., Hj, S., Z, B. & Yh, Q. SP1-induced overexpression of LINC00520 facilitates non-small cell lung cancer progression through miR-577/CCNE2 pathway and predicts poor prognosis. Human cell 34, (2021). Guo, Q., Xu, L., Peng, R., Ma, Y., Wang, Y., Chong, F., et al. Characterization of lncRNA LINC00520 and functional polymorphisms associated with breast cancer susceptibility in Chinese Han population. Cancer Medicine 9, 2252 (2020). Ding, Y., Li, M., Tayier, T., Zhang, M., Chen, L. & Feng, S. Bioinformatics analysis of lncRNA–associated ceRNA network in melanoma. Journal of Cancer 12, 2921 (2021). Haller, A., Gambi, G., D’Agostino, M., Davidson, G., Lallement, A., Mengus, G., et al. Interaction of lncRNA LENT with DHX36 regulates translation and suppresses autophagy in melanoma. Cell Death & Disease 17, 121 (2025). Y, L., X, H., Y, C. & D, C. Long non-coding RNA LINC00504 regulates the Warburg effect in ovarian cancer through inhibition of miR-1244. Molecular and cellular biochemistry 464, (2020). Q, L., W, W., T, Y., D, L., Y, H., G, B., et al. LINC00520 up-regulates SOX5 to promote cell proliferation and invasion by miR-4516 in human hepatocellular carcinoma. Biological chemistry 403, (2022). H, L., G, Z., C, L. & X, S. LINC00520 promotes colorectal cancer progression through miRNA-195-3p / NAT2 axis. Cellular and molecular biology (Noisy-le-Grand, France) 70, (2024). Pd, V., Pj, L., A, F. & O, R. Modulation of miRNA function by natural and synthetic RNA-binding proteins in cancer. Cellular and molecular life sciences: CMLS 76, (2019). Krol, J., Loedige, I. & Filipowicz, W. The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genet 11, 597–610 (2010). microRNAs as oncogenes and tumor suppressors. Developmental Biology 302, 1–12 (2007). Domingues, B., Lopes, J. M., Soares, P. & Pópulo, H. Melanoma treatment in review. ImmunoTargets and Therapy 7, 35 (2018). Fragale, A., Stellacci, E., Romagnoli, G., Licursi, V., Parlato, S., Canini, I., et al. Reversing vemurafenib-resistance in primary melanoma cells by combined romidepsin and type I IFN treatment through blocking of tumorigenic signals and induction of immunogenic effects. International Journal of Cancer 153, 1080–1095 (2023). A, D. S., L, D. S., F, R., S, G., V, L., Va, A., et al. NSD2 and miRNAs as Key Regulators of Melanoma Response to Romidepsin and Interferon-α2b Treatment. Cancer medicine 14, (2025). Dilshat, R., Fock, V., Kenny, C., Gerritsen, I., Lasseur, R. M. J., Travnickova, J., et al. MITF reprograms the extracellular matrix and focal adhesion in melanoma. eLife 10, e63093 (2021). T, S., J, M. & M, B. Visualization of Nuclear and Cytoplasmic Long Noncoding RNAs at Single-Cell Level by RNA-FISH. Methods in molecular biology (Clifton, N.J.) 2157, (2021). F, D., E, D., P, L. & M, B. Advances in endogenous RNA pull-down: A straightforward dextran sulfate-based method enhancing RNA recovery. Frontiers in molecular biosciences 9, (2022). Additional Declarations There is no duality of interest Supplementary Files SupplementaryFigures.pdf originalwesternblotandgel.pdf table.xlsx Cite Share Download PDF Status: Under Review 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-9504736","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":631754809,"identity":"a8654ec3-c5e4-4867-84ed-f6252ba0466f","order_by":0,"name":"Carlo Presutti","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIiWNgGAWjYHACxgMItoENAwMzmCXBwIZHD0wLY8MBgzSYFgsGNjx6kLQwHIYJVjDgsoa//eyDAx8Y7uTz858xf/yh4Hzidnbeh49u1Egw8Mk3YNUicSbd4OAMhmeWM2fkGAIddjtxZzO7sXHOMQmcDjNgSGM4zMNw2MDgBg9Ey4bDbGzSOWx4tPA/g2ixP38GpOUcVMs/PFokYLYwgB12AKIltw23FokbzxgOzjB4ZiBxI61wxhmDZGOgFmbj3D4JHja2BOwh1p/G+OBDxR0D/v7DGz5U/LGT3XD+GOPjnG91cvLNB7BbA3EeFkkePOpBAJ95o2AUjIJRMOIBANKHWhNNzq3tAAAAAElFTkSuQmCC","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Carlo","middleName":"","lastName":"Presutti","suffix":""},{"id":631754810,"identity":"259a78e6-3124-4571-9f7f-9d2a82ab5624","order_by":1,"name":"Lucrezia De Santis","email":"","orcid":"https://orcid.org/0009-0003-5022-5494","institution":"University of Rome, Sapienza","correspondingAuthor":false,"prefix":"","firstName":"Lucrezia","middleName":"","lastName":"De Santis","suffix":""},{"id":631754811,"identity":"a9454702-efd0-46e3-be9a-344249e12a49","order_by":2,"name":"Alessandro De Santis","email":"","orcid":"","institution":"University of Rome, Sapienza","correspondingAuthor":false,"prefix":"","firstName":"Alessandro","middleName":"","lastName":"De Santis","suffix":""},{"id":631754812,"identity":"825f7ff2-a13b-4e42-8ec7-58c84293c161","order_by":3,"name":"Vito Antonio Amico","email":"","orcid":"","institution":"University of Rome, Sapienza","correspondingAuthor":false,"prefix":"","firstName":"Vito","middleName":"Antonio","lastName":"Amico","suffix":""},{"id":631754813,"identity":"6eb24de7-cdea-4de1-ae1f-e072957766ac","order_by":4,"name":"Sofia Testarmata","email":"","orcid":"","institution":"University of Rome, Sapienza","correspondingAuthor":false,"prefix":"","firstName":"Sofia","middleName":"","lastName":"Testarmata","suffix":""},{"id":631754814,"identity":"8714ed8a-b85d-4d55-934f-1037c72e522a","order_by":5,"name":"Alessandra Fragale","email":"","orcid":"","institution":"Istituto Superiore di Sanità","correspondingAuthor":false,"prefix":"","firstName":"Alessandra","middleName":"","lastName":"Fragale","suffix":""},{"id":631754815,"identity":"474737bd-b1b1-415c-acdf-5b067224f8c2","order_by":6,"name":"Lucia Gabriele","email":"","orcid":"https://orcid.org/0000-0002-1483-866X","institution":"Istituto Superiore di Sanità","correspondingAuthor":false,"prefix":"","firstName":"Lucia","middleName":"","lastName":"Gabriele","suffix":""}],"badges":[],"createdAt":"2026-04-23 09:20:54","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9504736/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9504736/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109255010,"identity":"7ec93f6e-3ebb-4897-88ed-3dda0c69557e","added_by":"auto","created_at":"2026-05-14 09:55:31","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1689387,"visible":true,"origin":"","legend":"\u003cp\u003eLINC00504 and LINC00520 expression in melanoma patients. \u003cstrong\u003e(A) \u003c/strong\u003eVolcano plot showing modulated genes in proliferative primary melanoma cells before and after treatment with Romidepsin and Interferon a2b, displaying each gene's −log10 (FDR) and log2 fold change with the chosen covariate. Upregulated genes (FDR ≤ 0.1 and log2 fold change \u0026gt; 0.7) are highlighted in red. Downregulated genes (FDR ≤ 0.1 and log2 fold change \u0026lt; 0.7) are indicated in green. Vertical dashed lines indicate thresholds for up-regulated and down-regulated genes, and the horizontal dashed line represents the FDR threshold. Data sourced from the GSE221386 dataset. \u003cstrong\u003e(B)\u003c/strong\u003e Real-time quantitative polymerase chain reaction (RT-qPCR) analysis of LINC00504 and LINC00520 following treatment with RI in the proliferative cell lines. The y-axis represents the relative fold change compared to the control condition. Each colour represents a different treatment condition. Mean and standard error (SE) are depicted in the bar chart. N=4 (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001). \u003cstrong\u003e(C)\u003c/strong\u003e Analysis of the expression of LINC00504 and LINC00520 in TCGA melanoma dataset. In yellow the invasive subtype and in black the proliferative one. The y-axis shows the log10 Transcripts Per Million (TPM) (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001).\u003cstrong\u003e (D)\u003c/strong\u003eExpression correlation between LINC00504 and LINC00520 using all TCGA melanoma samples.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/2602310861875ca07b912a7b.png"},{"id":109254976,"identity":"3aec942c-28ac-4c33-b9c5-bd4723d7b7fd","added_by":"auto","created_at":"2026-05-14 09:55:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2819044,"visible":true,"origin":"","legend":"\u003cp\u003eLINC00504 and LINC00520 expression levels in different melanoma cell lines (A375, CHL1, SKmel28, and 624-mel) and their cellular localization. \u003cstrong\u003e(A) \u003c/strong\u003eRT-qPCR analysis of proliferative markers in four melanoma cell lines (A375, CHL1, SKmel28, and 624-mel). The y-axis represents the 2-∆Ct where ∆Ct = Ct (gene of interest) – Ct (housekeeping gene) using HPRT as the housekeeping gene. Mean and standard error (SE) are depicted in the bar chart. N=3 (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001). \u003cstrong\u003e(B)\u003c/strong\u003e RT-qPCR analysis of LINC00504 and LINC00520 in four melanoma cell lines (A375, CHL1, SKmel28, and 624-mel). The y-axis represents the 2-∆Ct where ∆Ct = Ct (gene of interest) – Ct (housekeeping gene) using HPRT as the housekeeping gene. Mean and standard error (SE) are depicted in the bar chart. N=3 (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001). \u003cstrong\u003e(C)\u003c/strong\u003e RT-qPCR analysis of LINC00504 and LINC00520 using nuclear and cytoplasm RNA fractions after cellular fractionation in 624-mel cell line. The nuclear marker is Xist, while the cytoplasmatic are HPRT and Actin. N=3 \u003cstrong\u003e(D)\u003c/strong\u003e RNA FISH using biotinylated probes against LINC00504 and LINC00520 (Scale bar: 10 µm).\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/a9c9786bdc4bcb894603f9ba.png"},{"id":109254973,"identity":"a8d16f5d-3132-42dc-8498-ab455c19171f","added_by":"auto","created_at":"2026-05-14 09:55:25","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1652318,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of the expression levels of miRNAs and mRNA targets and identification of regulatory network. \u003cstrong\u003e(A)\u003c/strong\u003e The heatmap displays aggregated scores calculated via scanMiR, representing the predicted binding activity between the selected miRNAs and the two lincRNAs (\u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e). Darker colour intensity indicates a higher predicted probability of physical association between the lincRNA transcripts and the candidate miRNAs. \u003cstrong\u003e(B) \u003c/strong\u003eAnalysis of the expression of miRNAs in TCGA melanoma dataset. In yellow the invasive subtype and in black the proliferative one. The y-axis shows the log10 Transcripts Per Million (TPM) (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001).\u003cstrong\u003e (C)\u003c/strong\u003e Schematic representation of lncRNA-miRNA-mRNA interactions \u003cstrong\u003e(D)\u003c/strong\u003e Analysis of the expression of mRNA targets in TCGA melanoma dataset. In yellow the invasive subtype and in black the proliferative one. The y-axis shows the log10 Transcripts Per Million (TPM) (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001).\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/aaac8a3350a7a6a267d05b46.png"},{"id":109254954,"identity":"c8794094-cbb4-4e8b-8fa5-78a127fe9348","added_by":"auto","created_at":"2026-05-14 09:55:25","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":596720,"visible":true,"origin":"","legend":"\u003cp\u003eEnrichment of lincRNAs and their miRNAs interactor. \u003cstrong\u003e(A)\u003c/strong\u003e LINC00504 and LINC00520 enrichment after RNA pulldown assay. The y-axis shows the enrichment of the lncRNA in terms of % of input. The negative control is HPRT which doesn’t result enrich after the pulldown. Pulldown using probes specific to LacZ don’t show enrichment of lincRNAs. Mean and standard error (SE) are shown, N=4. \u003cstrong\u003e(B)\u003c/strong\u003e miRNA enrichment after lincRNAs RNA pulldown. The y-axis shows the enrichment of the lncRNA in terms of % of input. Mean and standard error (SE) are shown, N=4.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/414d031fba4818337ee7f140.png"},{"id":109254988,"identity":"dfaaab6a-b7fe-4d99-b2d0-30e594a2d3c1","added_by":"auto","created_at":"2026-05-14 09:55:27","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":3544851,"visible":true,"origin":"","legend":"\u003cp\u003ePhenotypic effects after lincRNAs silencing obtained using specific siRNAs. \u003cstrong\u003e(A)\u003c/strong\u003e RT-qPCR analysis of LINC00504 and LINC00520 expression levels after RNA interference. The y-axis represents the relative fold change compared to the control condition, using HPRT as the housekeeping gene. Mean and standard error (SE) are shown, N=4. (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001). \u003cstrong\u003e(B)\u003c/strong\u003eRepresentative images of migration assays (20x magnification). Images are representative of 10 random fields per experimental condition. \u003cstrong\u003e(C)\u003c/strong\u003eQuantification of the number of migrated cells, calculated as the average count of 10 fields per condition using ImageJ software. Mean and standard error (SE) are shown, N=3. \u003cstrong\u003e(D)\u003c/strong\u003e Representative images of invasion assays (20x magnification) and \u003cstrong\u003e(E)\u003c/strong\u003e respective quantification, conducted as described above, of the invasive capacity of melanoma cells after lincRNAs silencing. Mean and standard error (SE) are shown, N=3. (p-values: ns \u0026gt; 0.05, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001).\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/69741c0b393eb049c63de541.png"},{"id":109255050,"identity":"55c1da83-e83d-4632-bf9c-c5911ddd35a4","added_by":"auto","created_at":"2026-05-14 09:55:44","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1845106,"visible":true,"origin":"","legend":"\u003cp\u003eIntegrative summary of MITF regulation of LINC00504 and LINC00520. \u003cstrong\u003e(A)\u003c/strong\u003e Forest plots of expression changes following MITF perturbation, where blue bars indicate loss-of-function (siRNA, shRNA, CRISPR-KO) and red bars indicate gain-of-function (overexpression), with MITF itself used as a positive control. \u003cstrong\u003e(B)\u003c/strong\u003e Positive correlation (Spearman 𝜌) between MITF and lincRNAs expression levels in TCGA-SKCM patient tumours and CCLE cutaneous melanoma cell lines. \u003cstrong\u003e(C-D)\u003c/strong\u003e The genome-browser views provide a detailed map of the LINC00504 and LINC00520 loci, showing regions of open chromatin (DNase-seq and ATAC-seq), and active enhancer marks (H3K27ac) across multiple melanoma cell lines.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/2ae3e3adc5c892a20ebdf42a.png"},{"id":109254980,"identity":"d70de7a2-7468-4947-8fa4-3b974cfc4331","added_by":"auto","created_at":"2026-05-14 09:55:26","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":939365,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of MITF direct binding and transcriptional regulation of LINC00504 and LINC00520. \u003cstrong\u003e(A)\u003c/strong\u003e Chromatin Immunoprecipitation (ChIP) analysis followed by agarose gel electrophoresis shows specific enrichment bands for LINC00504 and LINC00520 promoter regions after immunoprecipitation with anti-MITF antibody compared to the IgG control. \u003cstrong\u003e(B)\u003c/strong\u003e Quantitative PCR (qPCR) analysis of the ChIP-purified DNA confirms a significant fold-enrichment of MITF at the candidate regulatory elements of LINC00504 and LINC00520, with data expressed as a percentage of the input. \u003cstrong\u003e(C)\u003c/strong\u003e Western blot analysis confirms the efficient overexpression (OE) of MITF protein levels in melanoma cells, serving as a functional validation for the subsequent expression assays. \u003cstrong\u003e(D)\u003c/strong\u003eQuantitative real-time PCR (RT-qPCR) reveals a significant increase in the expression levels of both LINC00504 and LINC00520 following MITF overexpression, demonstrating that MITF not only binds to their regulatory regions but actively drives their transcription. Mean and standard error (SE) are shown, N=3 (p-values: * ≤ 0.05, ** ≤ 0.01).\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/5935e6fede56022d55f1719a.png"},{"id":109296532,"identity":"3859c478-0748-4b0c-9b42-cae44f6c81fc","added_by":"auto","created_at":"2026-05-15 08:47:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":11176645,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/9ff2bb9f-4dc8-48fb-bfa6-14b8f1e152dd.pdf"},{"id":109255017,"identity":"a9d0c4ea-579f-4d21-9e5b-8b0266620c56","added_by":"auto","created_at":"2026-05-14 09:55:37","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":479150,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigures.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/e868536e1d90379cda00dc71.pdf"},{"id":109255042,"identity":"2093f385-b694-43d8-ac3a-97d3f1e94c5d","added_by":"auto","created_at":"2026-05-14 09:55:43","extension":"pdf","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":429880,"visible":true,"origin":"","legend":"","description":"","filename":"originalwesternblotandgel.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/e1b044e2d60610492711f335.pdf"},{"id":109254974,"identity":"15655ecb-98c3-4b75-8dd4-fb415c6550c9","added_by":"auto","created_at":"2026-05-14 09:55:25","extension":"xlsx","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":11116,"visible":true,"origin":"","legend":"","description":"","filename":"table.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9504736/v1/2d7968a6885fc57c4ead8ab0.xlsx"}],"financialInterests":"There is no duality of interest","formattedTitle":"A lncRNA–miRNA axis regulates the balance between proliferative and invasive states in melanoma","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMelanoma is an aggressive skin cancer responsible for more than 70% of all skin cancer-related deaths, mostly due to its strong metastatic potential\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. It is considered one of the most metastatic and therapy resistant malignancies largely because of its pronounced heterogeneity\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e, which arises from both genetic alterations and non-genetic mechanisms. Approximately 50% of cutaneous melanoma harbour \u003cem\u003eBRAFV600E\u003c/em\u003e mutation\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, which promotes tumour growth, survival, and invasiveness\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. However, melanoma progression is not exclusively driven by genetic events, as epigenetic regulation, metabolic reprogramming, and developmental programs also play a crucial role in shaping tumour behaviour and therapeutic response\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMelanoma arises from malignant transformation and uncontrolled proliferation of melanocytes, the melanin-producing cells located in the basal layer of the epidermis. Melanocytes originate from neural crest cells that, during embryonic development, originate melanoblasts, an embryonic, highly migratory, and multipotent cell population capable of generating diverse lineages\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. This neuroectodermal origin confers intrinsic plasticity, which is retained during tumorigenesis and enables melanoma cells to reversibly transition between differentiated and dedifferentiated states, a process commonly referred to as phenotype switching.\u003c/p\u003e \u003cp\u003ePhenotype switching contributes to intratumoral heterogeneity and represents a mechanism underlying tumour progression and therapeutic resistance\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Transcriptomic analyses have identified two predominant and dynamically interchangeable cellular states, a \u0026ldquo;proliferative\u0026rdquo; (or melanocytic) and a \u0026ldquo;invasive\u0026rdquo; (or mesenchymal-like), which can coexist in the same tumour\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. These states are defined by distinct transcriptional programs controlled by specific master regulators. The proliferative phenotype is characterized by high expression of Melanocyte Inducing Transcription Factor (\u003cem\u003eMITF\u003c/em\u003e), which drives differentiation and cell growth\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e, whereas the invasive phenotype is associated with high expression of the receptor tyrosine kinase \u003cem\u003eAXL\u003c/em\u003e and genes linked to migration, extracellular matrix remodelling, and stress responses\u003csup\u003e\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Importantly, the invasive state is associated with metastatic dissemination and cross-resistance to both targeted therapies and immunotherapies\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, emerging evidence indicates that melanoma cell states are not limited to two subtypes, but there are different intermediate phenotypes. These transitional populations may display distinct functional properties and contribute differentially to tumour growth, dissemination, and drug resistance\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. The dynamic interconversion between these states is largely driven by epigenetic mechanisms, including DNA methylation, histone modifications, and non-coding RNAs (ncRNAs), which enable rapid and reversible adaptation to environmental and therapeutic pressures\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAmong ncRNAs, long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) have emerged as key regulators of gene expression networks in cancer\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. LncRNAs are transcripts longer than 200 nucleotides that regulate gene expression at epigenetic, transcriptional, and post-transcriptional levels, depending on their cellular localization\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Several lncRNAs have been implicated in tumor progression, where they can function as oncogenes or tumor suppressors. \u003cem\u003eLINC00504\u003c/em\u003e is not characterized in melanoma, but its role has been detected in other tumours, such as non-small cell lung, ovarian, colon, and breast cancers\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. It has an oncogenic role in favouring tumour progression. Patients with high expression of \u003cem\u003eLINC00504\u003c/em\u003e are associated with poor prognosis and it may be a new biomarker for breast cancer\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eLINC00520\u003c/em\u003e is widely expressed in various tissues, and it is upregulated in 11 cancers including lung cancer\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, non-small cell lung cancer (NSCLC)\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, breast cancer (BC)\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, and melanoma (MM)\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Notably, its role in melanoma appears to be context dependent. In primary melanoma, higher \u003cem\u003eLINC00520\u003c/em\u003e expression has been associated with better survival, whereas in metastatic samples its upregulation correlates with poor prognosis\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. These observations suggest that \u003cem\u003eLINC00520\u003c/em\u003e may play distinct roles depending on disease stage and cellular state, potentially reflecting the underlying phenotypic plasticity of melanoma cells. Mechanistically, both lncRNAs have been reported to function as competing endogenous RNAs (ceRNAs), acting as molecular sponges that modulate miRNA availability in different cancer system, such as breast cancer\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, ovarian cancer\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e, hepatocellular carcinoma\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e and colorectal cancer\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003emicroRNAs (miRNAs) are small non-coding RNA with a length of \u0026sim;18\u0026ndash;22 nucleotides that post-transcriptionally repress gene expression by binding to complementary sequences in target mRNAs\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. They regulate a broad spectrum of cellular processes, and their dysregulation contributes to cancer initiation, progression, and therapy response\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. MiRNAs are aberrantly expressed in many tumours depending on cancer type, stage, and other clinical variables. Depending on the cellular context, miRNAs may act as oncogenes or tumour suppressors\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCurrent therapeutic strategies for melanoma include surgical resection for localized disease and systemic treatments for advanced stages, such as targeted therapy, immunotherapy, and chemotherapy\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Despite significant clinical improvements, therapeutic resistance remains a major challenge.\u003c/p\u003e \u003cp\u003eRecently, treatment with the combination of Romidepsin and Interferon-α2b (RI) has been shown to induce growth arrest and cell death in melanoma cells\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Transcriptomic analysis of primary melanoma cells revealed distinct clusters corresponding to the proliferative and invasive states\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e and RI treatment was found to promote a shift from a proliferative toward a more invasive transcriptional phenotype.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eIdentification of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e in Proliferative cell state\u003c/p\u003e \u003cp\u003eThe re-analysis of the differentially expressed genes (DEGs) from the GSE221386 dataset indicated that two lincRNAs, \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e, are strongly downregulated after the treatment with Romidepsin and Interferon-α2b (RI) in the proliferative cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). This data was also validated using Real-time quantitative polymerase chain reaction (RT-qPCR) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB) in the same primary melanoma cells, confirming a significant reduction in the expression of both lincRNAs in these samples. Given that RI treatment has been shown to trigger a transcriptional shift from a proliferative toward an invasive-like phenotype, the observed downregulation of these lincRNAs suggests they could potentially be associated with this phenotypic remodelling.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further investigate this association in a clinical context, their expression levels were investigated in melanoma patients from The Cancer Genome Atlas (TCGA) cohort. After stratifying melanoma patient samples into proliferative or invasive subtypes based on their transcriptional signatures, it was observed that \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e are highly expressed in proliferative tumour compared to the invasive one (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). Furthermore, a correlation analysis within the TCGA dataset was performed to evaluate the potential coordinated regulation of these transcripts. A strong and statistically significant correlation was found between the expression levels of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). To further explore the clinical implications of these findings, we evaluated the prognostic relevance of both lincRNAs using TCGA survival data through the GEPIA platform. We observed that higher expression of these transcripts is associated with poorer patient prognosis (Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSelection of a Proliferative Cellular Model and Subcellular Localization Analysis\u003c/p\u003e \u003cp\u003eGiven that \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e were found to be highly expressed in the proliferative state, it was essential to identify an appropriate \u003cem\u003ein vitro\u003c/em\u003e model that closely mimics this phenotypic state in melanoma cells. To this aim, different immortalized melanoma cells (A375, CHL1, SKmel28, and 624-mel) were screened for the expression of established proliferative markers, such as \u003cem\u003eMITF\u003c/em\u003e and \u003cem\u003eMLANA\u003c/em\u003e, and the two lincRNAs. This characterization aimed to identify a cell line exhibiting a proliferative signature suitable for subsequent functional studies. Analysis of mRNA levels revealed that the 624-mel cell line exhibited the highest expression of both proliferative markers and lincRNAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA-B), thereby serving as the model for further investigation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFollowing the selection of the cellular model, the subcellular localization of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e was investigated in 624-mel cells, as the spatial distribution of lncRNAs is often indicative of their biological function. Two complementary approaches were employed to ensure technical robustness: biochemical cellular fractionation followed by RT-qPCR, and RNA Fluorescence In Situ Hybridization (RNA-FISH). The results from both methods demonstrated that \u003cem\u003eLINC00504\u003c/em\u003e is predominantly localized within the cytoplasm, whereas \u003cem\u003eLINC00520\u003c/em\u003e exhibits a dual distribution in both the nucleus and the cytoplasm (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC-D). The distinct cellular distribution of these two lncRNAs suggest they may act through different molecular mechanisms based on their cellular localization.\u003c/p\u003e \u003cp\u003eIdentification of the lncRNA\u0026ndash;miRNA\u0026ndash;mRNA Regulatory Network\u003c/p\u003e \u003cp\u003eTo explore the possible biological function of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e as miRNA sponge in melanoma, particularly regarding the cytoplasmatic fraction, computational analysis was performed to identify potential miRNA interactors. Using the prediction algorithms miRcode and RNA22, a set of candidate miRNAs with putative binding sites for these lincRNAs was identified. Among the predicted candidates, 6 miRNAs (miR-23a-3p, miR-24-3p, miR-27a-3p, miR-27b-3p, miR-31-5p, and miR-125b-5p) were selected for further investigation based on their binding affinity, as illustrated in the heatmap in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA. To evaluate the clinical relevance of these findings, miRNAs\u0026rsquo; expression was analysed in the TCGA-SKCM cohort, using the previously described stratification in proliferative and invasive subtypes. The miRNAs showed significantly higher expression levels in the invasive subtype compared to the proliferative one (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). This pattern is in direct contrast to the expression of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC), which are enriched in proliferative samples. The inverse correlation in patient data supports the hypothesis that these lincRNAs and miRNAs may be part of a coordinated regulatory network involved in melanoma phenotype switching.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further delineate the downstream effects of this regulatory axis, a second computational screening was performed to identify potential mRNA targets for the previously selected miRNAs. By integrating data from multiple bioinformatic software, including Mienturnet, TargetScan, and miRTarBase, we identified a specific set of genes likely regulated by these miRNAs. Notably, these targets include master regulators of the melanocyte lineage, such as \u003cem\u003eMITF\u003c/em\u003e and \u003cem\u003eMLANA\u003c/em\u003e. The complexity of this regulatory network is summarized in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, which presents both a schematic representation of the proposed regulatory network and a summary table of the identified miRNA\u0026ndash;mRNA interactions. To explore the clinical relevance of this network, the expression of these target mRNAs was evaluated across the TCGA-SKCM cohort. In agreement with the proposed model, many of the identified mRNA targets displayed significantly higher expression levels in the proliferative subtype compared to the invasive one (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Notably, their expression profiles show a direct correlation with the levels of the two lincRNAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC) and a consistent inverse correlation with their targeting miRNAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). This highly coordinated expression pattern across patient samples strongly suggests the existence of a functional lincRNA\u0026ndash;miRNA\u0026ndash;mRNA axis that characterizes the proliferative phenotype in melanoma.\u003c/p\u003e \u003cp\u003eExperimental Validation of lincRNA\u0026ndash;miRNA Physical Interaction\u003c/p\u003e \u003cp\u003eTo provide evidence of a direct interaction between the lncRNAs and the candidate miRNAs, we performed an RNA pull-down assay followed by RT-qPCR. To this end, we utilized the same biotinylated oligonucleotide probes previously employed for RNA-FISH (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), which were designed to specifically hybridize with \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, the pull-down efficiency was confirmed by the significant and specific enrichment of both lincRNAs. We analysed the co-enriched small RNA fraction to detect the presence of the predicted miRNA interactors. The RT-qPCR analysis revealed that a subset of the candidate miRNAs was significantly pulled down alongside the lincRNAs, suggesting a direct physical association within the cellular environment. Specifically, we observed a robust enrichment for \u003cem\u003emiR-23a-3p\u003c/em\u003e, \u003cem\u003emiR-24-3p\u003c/em\u003e, \u003cem\u003emiR-31-5p\u003c/em\u003e, and \u003cem\u003emiR-125b-5p\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Conversely, although \u003cem\u003emiR-27a-3p\u003c/em\u003e and \u003cem\u003emiR-27b-3p\u003c/em\u003e did not show a statistically significant enrichment under these specific experimental conditions, they were nonetheless retained for further functional investigations. Overall, these results substantiate the hypothesis that \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e act as molecular platforms for miRNA binding.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eLincRNA Silencing Promotes an Invasive-like Phenotypic Shift\u003c/p\u003e \u003cp\u003eAs mentioned above, treatment with RI induces a phenotype switch in melanoma cells, driving them from a proliferative to an invasive state. Under the same conditions, we observed a significant downregulation of both \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e. To investigate whether their reduced expression contributes to this transition, we acted directly on the lincRNAs by silencing them via RNA interference (RNAi). The silencing efficiency was confirmed by RT-qPCR, which demonstrated a substantial reduction in the levels of both transcripts (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). We then evaluated if the direct depletion of these lincRNAs was sufficient to trigger the functional hallmarks of the invasive phenotype, studying cell migratory and invasive properties. Transwell migration assays were performed using uncoated inserts (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB-C), while invasion assays were conducted using gel-coated chambers (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD-E). Migration and invasion assays revealed that silencing either \u003cem\u003eLINC00504\u003c/em\u003e or \u003cem\u003eLINC00520\u003c/em\u003e resulted in a significant increase in both migratory and invasive capacities. Notably, this enhanced cellular motility occurred without any significant impact on cell proliferation (data not shown), suggesting that the downregulation of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e is not merely a consequence of the phenotype switch, but an event that contributes to the acquisition of a more invasive profile. To understand the mechanisms behind this functional change, we next examined the molecular consequences of lincRNA silencing on the previously identified regulatory network.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eMolecular Disruption of the ceRNA Network Following lincRNA Silencing\u003c/p\u003e \u003cp\u003eTo investigate the molecular mechanisms underlying the observed phenotypic shift, we examined how the silencing of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e impacts the downstream components of the predicted regulatory network. Following the RNAi-mediated knockdown of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e, which was first validated via RT-qPCR (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA), we investigated the resulting changes in the expression levels of the previously identified miRNAs and their mRNA targets. We observed that the candidate miRNAs were significantly upregulated following the silencing of \u003cem\u003eLINC00520\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Under the same experimental conditions, we registered an efficient reduction in the expression of the mRNA targets, as well as a decrease in the levels of established proliferative markers (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). However, some mRNA targets show a reduction in their expression levels only after the silencing of one of the lincRNAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). To further validate these findings at the protein level, we analysed the expression of two specific targets. We selected Nuclear Receptor Binding SET Domain Protein 2 (\u003cem\u003eNSD2\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD), which in our previous work was described as a direct target of miR-23a, miR-24, and miR-31 within the same proliferative subtype, and \u003cem\u003eMITF\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE), as it is the most established proliferative marker in melanoma. In line with the transcriptomic data, we observed a consistent reduction in protein expression levels following the silencing of both lincRNAs.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eMITF Directly Regulates the Expression of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e\u003c/p\u003e \u003cp\u003eTo investigate the transcriptional regulation of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e, we performed an integrative analysis of transcriptomic and epigenomic data\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e (GSE137776, GSE61965, GSE94488, ENCODE ENCSR008SDL, GSE167496, GSE61966, GSE163646, GSE283855). First, we assessed the impact of MITF perturbation on lncRNA expression levels across multiple independent RNA-seq datasets. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, both lncRNAs were significantly downregulated following \u003cem\u003eMITF\u003c/em\u003e silencing (via siRNA, shRNA, and CRISPR-KO) and markedly upregulated upon \u003cem\u003eMITF\u003c/em\u003e overexpression, suggesting a positive regulatory relationship. This observation was further supported by expression correlation analysis in large-scale melanoma cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). We found a strong positive correlation (Spearman \u0026#120588;) between \u003cem\u003eMITF\u003c/em\u003e and both lncRNAs in cutaneous melanoma cell lines (CCLE) and a consistent correlation in patient tumours (TCGA-SKCM).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo determine whether this regulation is direct, we analysed the genomic landscape surrounding both loci using ATAC-seq and DNase-seq data (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC-D). We initially screened a\u0026thinsp;\u0026plusmn;\u0026thinsp;200 kb window to identify Candidate Regulatory Elements (CREs) characterized by high chromatin accessibility and active enhancer marks (H3K27ac). Notably, publicly available ChIP-seq datasets showed variable \u003cem\u003eMITF\u003c/em\u003e occupancy across these loci. The integration of these epigenomic tracks highlighted that the regions immediately upstream of the Transcription Start Site (TSS) also exhibited strong signatures of open chromatin and regulatory activity. Given that the proximal promoter is fundamental for the initiation of transcription, we focused our experimental validation on these proximal regulatory regions (~\u0026thinsp;2 kb upstream of the TSS). This approach allowed us to prioritize these upstream sequences for experimental validation, testing whether \u003cem\u003eMITF\u003c/em\u003e physically occupies these proximal regions to directly modulate the core promoter activity and functional expression of both lincRNAs. This coordinated expression reflects the broader transcriptional program governed by \u003cem\u003eMITF\u003c/em\u003e; indeed, functional enrichment analysis of MITF-target genes confirms a dual role in activating pigmentation and differentiation pathways while simultaneously repressing genes associated with cell adhesion and mesenchymal development (Supplementary Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eValidation of the MITF-lincRNA Regulatory Axis\u003c/p\u003e \u003cp\u003eTo validate the direct binding of \u003cem\u003eMITF\u003c/em\u003e to the regulatory elements of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e, we performed Chromatin Immunoprecipitation (ChIP) assays in melanoma cells. The immunoprecipitated DNA fractions were analysed via both semi-quantitative PCR and quantitative real-time PCR (qPCR). Agarose gel electrophoresis of the PCR products showed a clear enrichment for the target genomic regions in the anti-MITF fractions compared to the Rabbit IgG negative control (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). These results were further supported by ChIP-qPCR analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB), which demonstrated a significant fold-enrichment of both lincRNAs, confirming that MITF physically occupies these loci.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo assess the functional impact of MITF on the expression of these transcripts, we performed gain-of-function experiments. We generated a MITF overexpression construct by cloning the MITF transcript (ENST00000394351.9) into a pIRESneo-FLAG/HA vector, replacing the EYFP sequence. This plasmid was transfected into the A375 melanoma cell line, which was selected for its relatively low basal expression of MITF, providing an ideal model to observe transcriptional induction. Following transfection, Western blot analysis confirmed the successful overexpression of the MITF protein (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eC). Subsequent RT-qPCR analysis revealed a significant increase in the steady-state mRNA levels of both \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520 (\u003c/em\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eD\u003cem\u003e)\u003c/em\u003e. Taken together, these data demonstrate that MITF not only binds to the regulatory regions of these lincRNAs but also actively drives their transcriptional upregulation, establishing a direct regulatory link.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe identify \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e as components of a regulatory network associated with the proliferative state of melanoma cells. Both lncRNAs are highly expressed in proliferative melanoma subtype and localize to both nuclear and cytoplasmic compartments, suggesting multifunctional roles. Our findings support a model in which these transcripts contribute to the maintenance of the proliferative phenotype through coordinated transcriptional and post-transcriptional mechanisms.\u003c/p\u003e \u003cp\u003eAt the cytoplasmic level, \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e act as competing endogenous RNAs, sequestering specific miRNAs and limiting their ability to repress downstream targets. This sponging activity correlates with increased expression of shared mRNA targets, including \u003cem\u003eMITF\u003c/em\u003e, a master regulator of the melanocytic lineage and a key determinant of the proliferative state. The positive association between lncRNA levels and target mRNA expression, together with the inverse relationship observed with miRNA abundance, supports the existence of a functional lncRNA\u0026ndash;miRNA\u0026ndash;mRNA regulatory axis in proliferative melanoma cells.\u003c/p\u003e \u003cp\u003eImportantly, our data extend this post-transcriptional model by identifying \u003cem\u003eMITF\u003c/em\u003e as an upstream transcriptional regulator of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e. \u003cem\u003eMITF\u003c/em\u003e binding to their promoter regions suggests the existence of a positive feedback loop in which high \u003cem\u003eMITF\u003c/em\u003e activity promotes lncRNAs transcription, while the lncRNAs sustain \u003cem\u003eMITF\u003c/em\u003e expression indirectly through miRNA sequestration.\u003c/p\u003e \u003cp\u003eDisruption of this network, either during phenotype switching toward the invasive state or following experimental silencing of the lncRNAs, leads to reduced lncRNAs expression, release of miRNAs from sequestration, and subsequent repression of target mRNAs, including \u003cem\u003eMITF\u003c/em\u003e. The resulting decrease in \u003cem\u003eMITF\u003c/em\u003e protein levels likely contributes to the collapse of the proliferative program and prevents reactivation of the lncRNA-mediated circuit. These observations are consistent with the dynamic and reversible nature of melanoma cell states, where regulatory networks rather than fixed genetic alterations can govern phenotypic transitions.\u003c/p\u003e \u003cp\u003eMelanoma plasticity, rooted in the neural crest origin of melanocytes, allows tumour cells to oscillate between transcriptional states in response to intrinsic and extrinsic stimuli. Our findings suggest that the absence of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e contribute to the acquisition of a more invasive cell state, while their presence stabilize the proliferative state within this plastic framework. Given that phenotype switching is closely linked to therapeutic resistance, regulatory axes involving non-coding RNAs may represent critical modulators of melanoma adaptability.\u003c/p\u003e \u003cp\u003eOverall, this work highlights a multilayered regulatory circuit in which \u003cem\u003eMITF\u003c/em\u003e and lncRNAs cooperate to sustain the proliferative phenotype through coordinated transcriptional and post-transcriptional control. Targeting components of this network may offer new opportunities to interfere with melanoma cell state stability and overcome resistance mechanisms.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003eBioinformatic Analysis\u003c/p\u003e \u003cp\u003ePublicly available transcriptomic datasets were analysed to evaluate the expression profiles of long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and target messenger RNAs (mRNAs) in melanoma. RNA-sequencing (TPM) and miRNA-sequencing (RPM) data from the TCGA-SKCM cohort were retrieved and processed. Samples were stratified into \"Proliferative\" (Melanocytic and Transitory subtypes) and \"Invasive\" (Neural crest-like and Undifferentiated subtypes) classes based on established molecular signatures to assess expression differences across melanoma progression states. Additionally, differential expression analysis was performed on the GSE221386 dataset, with results visualized via volcano plots generated using the ggplot2 R package.\u003c/p\u003e \u003cp\u003eTo characterize the regulatory network of the selected miRNAs, we utilized the scanMiR R/Bioconductor package. miRNA seed matches were predicted across the sequences of \u003cem\u003eLINC00504\u003c/em\u003e, \u003cem\u003eLINC00520\u003c/em\u003e. Binding site affinity was quantified as an aggregated binding affinity, which integrates site type and context to estimate the regulatory impact of each miRNA on its targets.\u003c/p\u003e \u003cp\u003eExpression levels across TCGA subtypes were compared using the Wilcoxon rank-sum test, with p-values visualized using the ggpubr package. Heatmaps representing the predicted repression scores for both lncRNAs and coding genes were generated using ggplot2 and pheatmap, employing color gradients to represent binding potency.\u003c/p\u003e \u003cp\u003eTo investigate the transcriptional control of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e by \u003cem\u003eMITF\u003c/em\u003e, we performed an integrative analysis of transcriptomic and epigenomic data. The impact of \u003cem\u003eMITF\u003c/em\u003e on lincRNA expression was assessed across multiple independent RNA-seq datasets involving MITF perturbation (siRNA, shRNA, CRISPR-KO, and overexpression). Correlation of expression (Spearman \u0026#120588;) between \u003cem\u003eMITF\u003c/em\u003e and the lincRNAs was calculated using the CCLE (Cancer Cell Line Encyclopedia) and TCGA-SKCM cohorts. To identify direct binding sites, we analysed the genomic landscape surrounding the lincRNA loci (\u0026plusmn;\u0026thinsp;200 kb from the TSS) using ATAC-seq, and DNase-seq data retrieved from public data. Candidate Regulatory Elements (CREs) were identified based on high chromatin accessibility and active enhancer marks (H3K27ac). All primary data are public. Accessions, cell lines, and source: GSE137776, GSE61965, GSE94488, ENCODE ENCSR008SDL, GSE167496, GSE61966, GSE163646, GSE283855.\u003c/p\u003e \u003cp\u003eAll computational analyses were conducted in the R statistical environment (v4.x).\u003c/p\u003e \u003cp\u003eCell culture\u003c/p\u003e \u003cp\u003eIn this study were used human melanoma cell lines. The human melanoma cell lines A375, CHL1, and 624-mel were kindly provided by Antonio Filippini (University of Rome, Sapienza), while SKmel28 was a gift from Lucia Gabriele (Istituto Superiore di Sanit\u0026agrave;, Rome). A375, CHL1, and 624-mel were cultured in high-glucose DMEM with sodium pyruvate (Corning, Catalog Number: 15-013-CV) with 1 mmol/L L-glutamine (Gibco, Catalog number: 25030081), 100 U/mL penicillin, and 100 \u0026micro;g/mL streptomycin (Corning, Catalog number: 30-002-CI), and 10% Fetal Bovine Serum (Corning, Catalog number: 35-015-CV). SKmel28 was cultured in RPMI 1640 medium (Corning, Catalog Number: 15-040-CV) supplemented with 1 mmol/L L-glutamine, 100 U/mL penicillin and 100 \u0026micro;g/mL streptomycin, and 10% Fetal Bovine Serum. Cells were incubated in the humidified incubator at 37\u0026deg;C in a 5% CO\u003csub\u003e2\u003c/sub\u003e humidified atmosphere, and periodically checked for Mycoplasma.\u003c/p\u003e \u003cp\u003eQuantitative RT-PCR\u003c/p\u003e \u003cp\u003eRNA extraction was performed using QIAzol reagent and the miRNeasy Mini Kit (QIAGEN, Catalog number 217004) according to the manufacturer\u0026rsquo;s instructions. Reverse transcription was conducted using SensiFAST cDNA Synthesis Kit (Bioline, Catalog number: BIO-65054) and SuperScript\u0026trade; IV Reverse Transcriptase (Invitrogen, Catalog number: 18090050) for mRNA; for miRNA were used miRCURY LNA RT Kit (QIAGEN, Catalog number: 339340). The Real Time PCR was performed with SensiFAST SYBR Hi-ROX Kit (Bioline, Catalog number: BIO-92020) for mRNA and miRCURY LNA SYBR Green PCR Kit (QIAGEN: Catalog number: 339346) for miRNA. The reactions were conducted by using the StepOnePlus System (Applied Biosystems) according to the set reaction conditions. The HPRT gene and miR191-5p were selected as the reference gene for normalization of mRNA and miRNA analysis, respectively.\u003c/p\u003e \u003cp\u003eWestern blot\u003c/p\u003e \u003cp\u003eCells were lysed with RIPA Lysis and Extraction Buffer (Thermo Scientific\u0026trade;, Catalog number: 89901) added with protease inhibitors cOmplete\u0026trade; EDTA-free Protease Inhibitor Cocktail (Merck, Catalog number: 11873580001) and phosphatase inhibitors PhosSTOP\u0026trade; (Merck, Catalog number: 4906845001). The concentration of extracted protein was measured by Pierce\u0026trade; Detergent Compatible Bradford Assay Kit (Thermo Scientific\u0026trade;, Catalog number: 23246). The total protein was separated in 8% and 10% PAGE prepared from Acrylamide/Bis-acrylamide, 40% solution (19:1) (Merck, Catalog number: A9926-5X100ML). Proteins were transferred to nitrocellulose membranes (Amersham Protran WB membrane, Merck; Catalog number: GE10600007). Antibodies against NSD2 (Cell Signaling Technology, D4Z8Q, 1:7 000), HPRT (Prodotti Gianni srl, EPR5299, 1:10 000), MITF (Cell Signaling, D3B4T Rabbit mAb #97800), and HA (Invitrogen, Mouse, Catalog number: 26183, 1:5 000) were incubated at 4\u0026deg;C overnight. Secondary HRP peroxidase-conjugated goat anti-rabbit Ab (Thermo Fisher Scientific Cat#31460, RRID: AB_228341, 1:10 000) and goat anti-mouse (Thermo Fisher Scientific Cat# 31430, RRID: AB_228307, 1:10 000) were used. The bands were detected using Clarity Western ECL Substrate (BIORAD Catalog number: BRD1705061) and ChemiDoc MP Imaging System (BIORAD). PageRuler\u0026trade; Prestained Protein Ladder (Thermo Scientific, Catalog number 26616) was used as marker.\u003c/p\u003e \u003cp\u003eCell fractionation\u003c/p\u003e \u003cp\u003eTo perform cellular fractionation was used the hypotonic buffer lysis A (HLB): Tris (20 mM, pH 8.0), NaCl (10 mM), MgCl2 (3 mM), EDTA (0.2 mM), NP40 (0.1%), and fresh DTT (1 mM), cOmplete\u0026trade; EDTA-free Protease Inhibitor Cocktail (1x) (Merck, Catalog Number: 11873580001), and RNaseOUT\u0026trade; Recombinant Ribonuclease Inhibitor (1 U/\u0026micro;l) (Invitrogen, Catalog Number: 10777019). Cultured cells were resuspended in PBS and centrifugated to resuspend the pellet in HLB. With following centrifuge and wash, nucleus and cytoplasm can be separated in the pellet and supernatant respectively. Briefly, the pellet was resuspended in HLB, incubate in ice for 5 minutes and centrifugated 400xg 5 minutes at 4\u0026deg;C. Repeat this steps and collect the supernatant after the second centrifugation because it contains the cytoplasm fraction. Wash twice the pellet with HLB and collect the pellet after the second centrifugation; it is the nucleus fraction. The fractionation is followed by RNA extraction, retro transcription and qPCR.\u003c/p\u003e\n\u003ch3\u003eRNA FISH\u003c/h3\u003e\n\u003cp\u003eRNA FISH was conducted as previously described\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Were used 16 and 20 probes with 5\u0026rsquo; biotin complementary with \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e respectively. Probes were designed using Stellaris tool and ordered from Bio-Fab research Srl. For the lncRNA visualization, probes were incubated with the coverslip overnight at 37\u0026deg;C in a humid box. The next day Streptavidin, Alexa Fluor\u0026trade; 488 Conjugateanti-biotin (Invitrogen\u0026trade;, Catalog number: S32354) were incubated 1h in a humid box. After the staining with DAPI, the signals were acquired with Nikon Eclipse 50i (100x magnification, 385nm and 475nm filters).\u003c/p\u003e \u003cp\u003eRNA pulldown assay\u003c/p\u003e \u003cp\u003eRNA pulldown was conducted as previously described\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. RNA pull-down experiments were performed using 5\u0026prime;-biotinylated antisense DNA probes complementary to the target lncRNAs, also used in the RNA FISH. A 5\u0026prime;-biotinylated probe against lacZ, a non-human bacterial gene, was employed as a negative control to monitor non-specific binding and validate probe specificity. Cells were lysed using Lysis buffer (LB): Tris-HCl pH 7.5 50 mM, NaCl 150 mM, MgCl2 3 mM, NP40 0.5% (IGEPAL\u0026reg; CA-630, Sigma-Aldrich, Catalog number: I8896), EDTA 2 mM, and fresh DTT 1 mM, 1\u0026times; PIC (cOmplete\u0026trade; Protease Inhibitor Cocktail, Roche, Catalog number: 11697498001), and RNaseOUT\u0026trade; Recombinant Ribonuclease Inhibitor (0.2 U/\u0026micro;l) (Invitrogen, Catalog number: 10777019). After the lysis, collect at least 1 mg of total extract protein for each pulldown (PD) condition and 0,1 mg for the input (10% of PD). Subsequently, sample were diluted with Hybridization buffer (HB): Tris-HCl pH 7.5 50 mM, NaCl 150 mM, MgCl2 3 mM, NP40 0.5%, EDTA 2 mM, Dextran Sulfate Salt DSS (2,5%) (Merck, Catalog Number: D8906-5G), and fresh DTT 1 mM, 1\u0026times; PIC, and RNase Inhibitors (0.2 U/\u0026micro;l). Add the probes and incubate rotating for at least 4h at +\u0026thinsp;4\u0026deg;C. Then, add Streptavidin MagneSphere\u0026reg; Paramagnetic Particles (Promega, Catalog number: Z5481) pre-washed with HB and incubate 1h at room temperature. After 4 wash using magnetic rack, add QIAzol and detach the beads by vortex the samples. Proceed with RNA extraction, retro transcription using SuperScript\u0026trade; IV Reverse Transcriptase and qPCR.\u003c/p\u003e \u003cp\u003eRNA interference\u003c/p\u003e \u003cp\u003eTo silence \u003cem\u003eLINC00520\u003c/em\u003e were used the FlexiTube GeneSolution GS645687 from QIAGEN, while for \u003cem\u003eLINC00504\u003c/em\u003e were constructed siRNA custom made from QIAGEN. The negative control is Negative Control siRNA (20 nmol) (QIAGEN, Catalog number: 1027310). After the seeding of the cells 24h before transfection, cells were treated with Lipofectamine\u0026trade; RNAiMAX Transfection Reagent (Invitrogen, Catalog number: 13778150) and siRNA. 48h after the transfection, the RNA was extracted following the protocol, retro transcribed with SensiFAST cDNA Synthesis Kit and analysed with qPCR.\u003c/p\u003e \u003cp\u003eCell invasion and migration assay\u003c/p\u003e \u003cp\u003eTranswell assay was used to detect the invasiveness of melanoma cell. 48h after the transfection with siRNA against the lincRNAs, cells were counted and were placed on Geltrex-coated TC-inserts (Geltrex Gibco A14132-02; TC-inserts Sarstedt 83.3932.800). In the transwell, 100,000 cells were seeded in DMEM 2% FBS. The culture medium with 20% fetal bovine serum was used as chemical attractant and it was added to the lower chamber of the well. After 8h, the invasive cells were fixed and stained with crystal violet (Sigma-Aldrich, Catalog Number: S-C0775-25G). The number of invasive cells was counted in ten random fields under 20x magnification, and the mean for each condition was determined. For the migration assay, transwell inserts were used following the same protocol, but without the Geltrex coating. Migrated cells were fixed and coloured with crystal violet, and they were counted in ten random fields under 20x magnification.\u003c/p\u003e \u003cp\u003ePlasmids and transfections\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eMITF\u003c/em\u003e overexpressing plasmid was constructed by inserting the full length of \u003cem\u003eMITF\u003c/em\u003e ENST00000394351.9 isoform into pIRESneo-FLAG/HA EYFP (AddGene, Plasmid #10825) replacing the \u003cem\u003eEYFP\u003c/em\u003e. pIRESneo-FLAG/HA EYFP construct was used as control. pIRESneo-FLAG/HA MITF and pIRESneo-FLAG/HA EYFP were transfected into cells using Lipofectamine\u0026trade; 3000 Transfection Reagent (Invitrogen\u0026trade;, Catalog number: L3000001). 24 hours after transfection, \u003cem\u003eMITF\u003c/em\u003e overexpression was validated using Western Blot. Moreover, total RNA was extracted to analyse changes in expression levels of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eChromatin Immuno Precipitation\u003c/p\u003e \u003cp\u003eChromatin Immuno Precipitation was performed using MAGnify\u0026trade; Chromatin Immunoprecipitation System (Applied Biosystems\u0026trade;, Catalog number: 492024) following the protocol of the kit. After 624-mel cell lysis, the sonication was performed using Bioruptor\u0026reg; Pico (Diagenode, B01080010) with 12 cycles 30\u0026rdquo; ON/30\u0026rdquo; OFF. We used antibody against MITF (Cell Signaling, D3B4T Rabbit mAb #97800), H3K4Me3 (Cell signaling, C42D8 Rabbit mAb #9751), and IgG Rabbit (provided with the kit). The DNA immunoprecipitated was analysed using PCR and qPCR with primers against the promoter region of \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflicts of Interest\u003c/h2\u003e \u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e \u003cp\u003eConceptualization, LDS and CP. Study design, LDS, LG and CP. Experiments, LDS, VAA, and ST. Project administration, funding acquisition and resources, CP. Bioinformatics analysis, ADS. Supervision, CP. Writing of original draft, LDS and CP. Editing, LDS, AF, LG and CP. All authors had access to the study data and approved the final submission.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was supported by the PNRR Project CN3-National Center for Gene Therapy and Drugs based on RNA Technology (Spoke 3). Primary melanoma cells were kindly provided by Doctor Stefania D\u0026rsquo;Atri, Molecular Oncology Laboratory, Istituto Dermopatico Dell\u0026rsquo;Immacolata IDI-IRCCS, Rome, Italy. We thank Sapienza University of Rome for providing the facilities and resources that made this study possible. We are also grateful to our colleagues for their insightful feedback and suggestions during the development of this project.\u003c/p\u003e\u003ch2\u003eData Availability Statement\u003c/h2\u003e \u003cp\u003eFull length western blots are available as Supplementary File.\u003c/p\u003e \u003cp\u003eData sharing is not applicable to this article as no new data were created or analyzed in this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWorld Cancer Research Fund. World Cancer Research Fund \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.wcrf.org/\u003c/span\u003e\u003cspan address=\"https://www.wcrf.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, F. Z., Dhillon, A. S., Anderson, R. L., McArthur, G. \u0026amp; Ferrao, P. T. Phenotype Switching in Melanoma: Implications for Progression and Therapy. Front. Oncol. 5, 126186 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRambow, F., Marine, J.-C. \u0026amp; Goding, C. R. Melanoma plasticity and phenotypic diversity: therapeutic barriers and opportunities. Genes \u0026amp; Development 33, 1295 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePillai, M. \u0026amp; Jolly, M. K. Systems-level network modeling deciphers the master regulators of phenotypic plasticity and heterogeneity in melanoma. iScience 24, 103111 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmalley, K. S. M. \u0026amp; Herlyn, M. Loitering with Intent: New Evidence for the Role of BRAF Mutations in the Proliferation of Melanocytic Lesions. J Invest Dermatol 123, xvi\u0026ndash;xvii (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ, T., L, R., K, P., C, G., Km, S., J, L., et al. Multi-stage Differentiation Defines Melanoma Subtypes with Differential Vulnerability to Drug-Induced Iron-Dependent Oxidative Stress. Cancer cell 33, (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, F., Santinon, F., Gonz\u0026aacute;lez, R. E. F. \u0026amp; Rinc\u0026oacute;n, S. V. del. Melanoma Plasticity: Promoter of Metastasis and Resistance to Therapy. Frontiers in Oncology 11, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTirosh, I., Izar, B., Prakadan, S. M., Marc H Wadsworth, I. I., Treacy, D., Trombetta, J. J., et al. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq. Science (New York, N.Y.) 352, 189 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMLANA/MART1 and SILV/PMEL17/GP100 Are Transcriptionally Regulated by MITF in Melanocytes and Melanoma. The American Journal of Pathology 163, 333\u0026ndash;343 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoek, K. S., Schlegel, N. C., Brafford, P., Sucker, A., Ugurel, S., Kumar, R., et al. Metastatic potential of melanomas defined by specific gene expression profiles with no BRAF signature. Pigment Cell Research 19, 290\u0026ndash;302 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eI, A. \u0026amp; C, W. Phenotype plasticity as enabler of melanoma progression and therapy resistance. Nature reviews. Cancer 19, (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM, S., M, C., G, C., G, N., F, A., G, T., et al. Human cutaneous melanomas lacking MITF and melanocyte differentiation antigens express a functional Axl receptor kinase. The Journal of investigative dermatology 131, (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ, L., J, K., M, R., T, B., M, R., M, C., et al. Melanomas resist T-cell therapy through inflammation-induced reversible dedifferentiation. Nature 490, (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarras, P., Bordeu, I., Pozniak, J., Nowosad, A., Pazzi, C., Van Raemdonck, N., et al. A cellular hierarchy in melanoma uncouples growth and metastasis. Nature 610, 190\u0026ndash;198 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eN, S., S, G., E, D. S., N, V. \u0026amp; G, B. Epithelial-to-Mesenchymal Transition: Epigenetic Reprogramming Driving Cellular Plasticity. Trends in genetics: TIG 33, (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmitt, A. M. \u0026amp; Chang, H. Y. Long Noncoding RNAs in Cancer Pathways. Cancer cell 29, 452 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAkhade, V. S., Pal, D. \u0026amp; Kanduri, C. Long Noncoding RNA: Genome Organization and Mechanism of Action. Long Non Coding RNA Biology 47\u0026ndash;74 (2017) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-981-10-5203-3_2\u003c/span\u003e\u003cspan address=\"10.1007/978-981-10-5203-3_2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeng, J., Ma, J., Liu, S., Wang, J. \u0026amp; Chen, Y. A noncoding RNA LINC00504 interacts with c-Myc to regulate tumor metabolism in colon cancer. Journal of Cellular Biochemistry 120, 14725\u0026ndash;14734 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeng, J., Li, Y., Zhu, L., Zhao, Q., Li, D., Li, Y., et al. STAT1 mediated long non-coding RNA LINC00504 influences radio-sensitivity of breast cancer via binding to TAF15 and stabilizing CPEB2 expression. Cancer Biology \u0026amp; Therapy 22, 630 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHou, T., Ye, L. \u0026amp; Wu, S. Knockdown of LINC00504 Inhibits the Proliferation and Invasion of Breast Cancer via the Downregulation of miR-140-5p. Onco Targets Ther 14, 3991\u0026ndash;4003 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXia, G., Li, X., Chen, F. \u0026amp; Shao, Z. LncRNA LINC00520 Predicts Poor Prognosis and Promotes Progression of Lung Cancer by Inhibiting MiR-3175 Expression. Cancer Management and Research 12, 5741 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJf, W., Zn, X., Hj, S., Z, B. \u0026amp; Yh, Q. SP1-induced overexpression of LINC00520 facilitates non-small cell lung cancer progression through miR-577/CCNE2 pathway and predicts poor prognosis. Human cell 34, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo, Q., Xu, L., Peng, R., Ma, Y., Wang, Y., Chong, F., et al. Characterization of lncRNA LINC00520 and functional polymorphisms associated with breast cancer susceptibility in Chinese Han population. Cancer Medicine 9, 2252 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing, Y., Li, M., Tayier, T., Zhang, M., Chen, L. \u0026amp; Feng, S. Bioinformatics analysis of lncRNA\u0026ndash;associated ceRNA network in melanoma. Journal of Cancer 12, 2921 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaller, A., Gambi, G., D\u0026rsquo;Agostino, M., Davidson, G., Lallement, A., Mengus, G., et al. Interaction of lncRNA LENT with DHX36 regulates translation and suppresses autophagy in melanoma. Cell Death \u0026amp; Disease 17, 121 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eY, L., X, H., Y, C. \u0026amp; D, C. Long non-coding RNA LINC00504 regulates the Warburg effect in ovarian cancer through inhibition of miR-1244. Molecular and cellular biochemistry 464, (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQ, L., W, W., T, Y., D, L., Y, H., G, B., et al. LINC00520 up-regulates SOX5 to promote cell proliferation and invasion by miR-4516 in human hepatocellular carcinoma. Biological chemistry 403, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH, L., G, Z., C, L. \u0026amp; X, S. LINC00520 promotes colorectal cancer progression through miRNA-195-3p / NAT2 axis. Cellular and molecular biology (Noisy-le-Grand, France) 70, (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePd, V., Pj, L., A, F. \u0026amp; O, R. Modulation of miRNA function by natural and synthetic RNA-binding proteins in cancer. Cellular and molecular life sciences: CMLS 76, (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrol, J., Loedige, I. \u0026amp; Filipowicz, W. The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genet 11, 597\u0026ndash;610 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003emicroRNAs as oncogenes and tumor suppressors. Developmental Biology 302, 1\u0026ndash;12 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDomingues, B., Lopes, J. M., Soares, P. \u0026amp; P\u0026oacute;pulo, H. Melanoma treatment in review. ImmunoTargets and Therapy 7, 35 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFragale, A., Stellacci, E., Romagnoli, G., Licursi, V., Parlato, S., Canini, I., et al. Reversing vemurafenib-resistance in primary melanoma cells by combined romidepsin and type I IFN treatment through blocking of tumorigenic signals and induction of immunogenic effects. International Journal of Cancer 153, 1080\u0026ndash;1095 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eA, D. S., L, D. S., F, R., S, G., V, L., Va, A., et al. NSD2 and miRNAs as Key Regulators of Melanoma Response to Romidepsin and Interferon-α2b Treatment. Cancer medicine 14, (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDilshat, R., Fock, V., Kenny, C., Gerritsen, I., Lasseur, R. M. J., Travnickova, J., et al. MITF reprograms the extracellular matrix and focal adhesion in melanoma. eLife 10, e63093 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eT, S., J, M. \u0026amp; M, B. Visualization of Nuclear and Cytoplasmic Long Noncoding RNAs at Single-Cell Level by RNA-FISH. Methods in molecular biology (Clifton, N.J.) 2157, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eF, D., E, D., P, L. \u0026amp; M, B. Advances in endogenous RNA pull-down: A straightforward dextran sulfate-based method enhancing RNA recovery. Frontiers in molecular biosciences 9, (2022).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"cell-death-and-disease","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cddis","sideBox":"Learn more about [Cell Death \u0026 Disease](http://www.nature.com/cddis/)","snPcode":"41419","submissionUrl":"https://mts-cddis.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Disease","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9504736/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9504736/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMelanoma is a highly aggressive malignancy characterized by significant phenotypic plasticity, allowing tumour cells to transit between different phenotypic states. This process, known as phenotype switching, drives metastasis and therapeutic resistance. While the primary transcriptional markers defining these states are well-established, the non-coding regulatory networks that stabilize these phenotypes remain incompletely understood. In this study, we identify the long non-coding RNAs (lncRNAs) \u003cem\u003eLINC00504\u003c/em\u003e and \u003cem\u003eLINC00520\u003c/em\u003e as key components of a multilayered regulatory circuit associated with the regulation of the proliferative/invasive melanoma subtypes. Functional assays reveal that experimental silencing of \u003cem\u003eLINC00504\u003c/em\u003e or \u003cem\u003eLINC00520\u003c/em\u003e in proliferative cells is sufficient to enhance the migratory and invasive capacity of melanoma cells, effectively mimicking the phenotype switch toward the invasive state. Mechanistically, these lincRNAs function as competing endogenous RNAs (ceRNAs) in the cytoplasm, where they sequester specific microRNAs to prevent the repression of mRNA targets, including \u003cem\u003eMITF\u003c/em\u003e. Furthermore, we establish that \u003cem\u003eMITF\u003c/em\u003e directly binds to the regulatory elements of both lincRNAs to drive their transcription, forming a positive feedback loop that reinforces the proliferative transcriptional program. Disruption of this MITF-lncRNA-miRNA axis facilitates the transition toward an invasive phenotype. Collectively, our findings highlight a novel integrated regulatory circuit that maintains melanoma cell state stability and suggest that targeting these non-coding components may provide a strategy to overcome phenotypic plasticity and therapeutic adaptability in melanoma.\u003c/p\u003e","manuscriptTitle":"A lncRNA–miRNA axis regulates the balance between proliferative and invasive states in melanoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-14 09:54:08","doi":"10.21203/rs.3.rs-9504736/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-disease","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cddis","sideBox":"Learn more about [Cell Death \u0026 Disease](http://www.nature.com/cddis/)","snPcode":"41419","submissionUrl":"https://mts-cddis.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Disease","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5e16acfc-9d36-4335-8389-17c14d8bf6a5","owner":[],"postedDate":"May 14th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67244062,"name":"Biological sciences/Cancer/Tumour heterogeneity"},{"id":67244063,"name":"Biological sciences/Molecular biology/Epigenetics"}],"tags":[],"updatedAt":"2026-05-14T09:54:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-14 09:54:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9504736","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9504736","identity":"rs-9504736","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}