Oncogenic Properties of the Antisense lncRNA COMET in BRAF- and RET-Driven Papillary Thyroid Carcinomas
Abstract
Papillary thyroid carcinomas, a prevalent form of thyroid malignancy, exhibit a remarkable degree of molecular heterogeneity, primarily driven by distinct genetic alterations such as RET rearrangements, as well as specific mutations in the BRAF and RAS genes. These varied genetic drivers instigate differential activation of critical cellular signaling pathways, which in turn leads to the emergence of diverse tumor phenotypes, each associated with unique clinical prognoses and responses to treatment. While comprehensive efforts like those undertaken by The Cancer Genome Atlas Consortium have significantly advanced our understanding by identifying distinct tumor subgroups based on their unique protein-coding gene expression signatures, there has remained a crucial gap in the systematic assessment of other vital regulatory molecules. Specifically, the expression patterns of long noncoding RNAs, or lncRNAs, and their intricate correlations with these well-defined, tumor-driving mutations and genomic rearrangements, had not been thoroughly or systematically investigated at a large scale.
In an effort to bridge this critical knowledge gap and gain a more profound understanding of the molecular landscape of papillary thyroid carcinomas, our research involved a meticulous reanalysis of our extensive RNA-sequencing data. This reanalysis was conducted using a sophisticated de novo discovery approach, allowing for the unbiased identification of novel lncRNAs and the precise delineation of tumor subtype-specific signatures composed of both previously annotated and newly identified lncRNAs. Through this rigorous analytical process, we successfully pinpointed a particularly compelling lncRNA, which we subsequently designated as COMET, an acronym for “Correlated-to-MET.” COMET was characterized as a natural antisense transcript, indicating its potential to regulate the expression of its sense counterpart, in this case, the MET oncogene. A pivotal observation was the remarkably high expression of COMET specifically within carcinomas harboring the BRAF V600E mutation or various RET gene rearrangements, collectively categorized as BRAF-like tumors due to their shared molecular characteristics and often aggressive clinical behavior. Furthermore, we ascertained that elevated COMET expression actively induced the downstream activation of the critical MAPK pathway, a signaling cascade notoriously involved in cell proliferation, survival, and differentiation, and frequently dysregulated in various cancers.
Delving deeper into COMET’s functional role within papillary thyroid carcinomas, we discovered that it was not an isolated entity but rather an integral component of a complex coexpression network. This sophisticated network encompassed several key oncogenes predominantly belonging to the very same MAPK pathway, thereby reinforcing COMET’s direct involvement in this pro-oncogenic signaling axis. Intriguingly, we observed a remarkably strong positive correlation between COMET expression and the expression levels of the MET oncogene, suggesting a profound regulatory relationship. To experimentally validate the functional implications of COMET, we conducted targeted depletion studies. The results were compelling: the suppression of COMET expression led to a significant reduction in the expression of numerous genes within this identified oncogenic network, crucially including the MET oncogene itself. Beyond its molecular regulatory effects, the repression of COMET also yielded dramatic inhibitory effects on the malignant characteristics of tumor cells. Specifically, we observed a substantial inhibition of both the viability and proliferative capacity of tumor cells that harbored either the BRAF V600E somatic mutation or a RET oncogene rearrangement, indicating a direct impact on tumor growth. Moreover, COMET repression profoundly reduced the intrinsic motility and invasive capabilities of these tumor cells, critical features associated with metastatic progression and poor prognosis.
Perhaps one of the most clinically significant findings was the observation that silencing COMET markedly enhanced the sensitivity of tumor cells to vemurafenib. Vemurafenib is a commonly employed targeted therapeutic agent designed to inhibit mutated B-raf, yet resistance to this drug remains a significant clinical challenge. The ability of COMET repression to augment vemurafenib efficacy suggests a novel strategy to circumvent or overcome therapeutic resistance in a subset of thyroid cancers. Collectively, the comprehensive findings emanating from our study robustly suggest that COMET represents a novel and promising therapeutic target. Its modulation could potentially serve as an innovative strategy to improve the effectiveness of existing drug-based cancer therapies, particularly benefiting patients with BRAF-mutated and MET-addicted papillary thyroid carcinomas, where current treatment options may face resistance issues.
Significance
These groundbreaking results significantly illuminate the previously underappreciated oncogenic role of the long noncoding RNA, COMET, within the complex biological landscape of thyroid cancer. By elucidating its integral involvement in driving malignant phenotypes and its direct association with key oncogenic pathways, this research provides a deeper molecular understanding of papillary thyroid carcinoma progression. Furthermore, and of paramount clinical importance, our findings unequivocally indicate COMET as a compelling and potentially transformative new therapeutic target. Its strategic inhibition offers a promising avenue to overcome the pervasive problem of vemurafenib resistance, a formidable challenge often encountered in the treatment of BRAF-mutated and MET-addicted carcinomas, thereby paving the way for more effective and durable treatment outcomes for patients suffering from these aggressive forms of thyroid cancer.
Introduction
The landscape of genomic understanding has undergone a profound transformation in recent years, largely propelled by comprehensive initiatives such as the ENCODE Project. This seminal endeavor meticulously mapped various functional elements across the human genome, ultimately unveiling a remarkably pervasive transcriptional activity that extends over an unexpectedly vast proportion of our genetic material. Historically, many of these extensive genomic stretches were presumptively dismissed as “junk DNA,” devoid of significant biological purpose. However, the groundbreaking findings from the ENCODE Project have fundamentally revised this perspective, demonstrating that a substantial number of these regions are actively transcribed into diverse non-coding RNA molecules, collectively known as ncRNAs. Far from being inert, these ncRNAs are now recognized as teeming with intricate regulatory activities, playing pivotal roles in cellular processes. Among the various classes of non-coding RNAs, a particularly significant group comprises the long noncoding RNAs, or lncRNAs, which are specifically defined as RNA transcripts exceeding 200 base pairs in length and notably lacking any discernible protein-coding potential. This distinction in length and function sets them apart from shorter regulatory RNAs and messenger RNAs, highlighting their unique contributions to gene expression and cellular regulation.
These multifaceted RNA molecules can be systematically categorized into distinct groups, with their classification often based on their precise genomic location in relation to protein-coding genes. A notable category within this classification includes those non-coding RNA molecules that physically overlap with other genes but are transcribed from the complementary, or opposite, DNA strand. These are formally designated as antisense non-coding RNAs, or more commonly referred to as Natural Antisense Transcripts, abbreviated as NATs. A growing body of research has compellingly demonstrated that NATs possess remarkable capabilities to regulate gene expression in a highly localized manner, often referred to as cis-regulation. This involves influencing the expression levels of their corresponding sense gene, either by actively enhancing or effectively inhibiting its transcription at the immediate genomic vicinity. Beyond transcriptional control, NATs can also exert influence over messenger RNA stability, directly affecting the longevity and degradation rates of these crucial templates for protein synthesis. Furthermore, some NATs have been observed to facilitate the recruitment of protein-coding messenger RNAs to active polysomes, thereby directly impacting and promoting their efficient translation into proteins. In addition to their localized effects, NATs have also been implicated in the more distant regulation of gene expression, acting in trans to influence genes located elsewhere in the genome through a variety of intricate and diverse molecular mechanisms.
The specific biological roles of many lncRNAs have now been extensively elucidated and well characterized, particularly their critical involvement in the complex processes of cellular transformation and the progression of various cancers. A prominent and illustrative example of a NAT that is directly implicated in the pathogenesis of cancer is the natural antisense transcript for zinc finger E-box binding homeobox 2, known as ZEB2-AS. This particular lncRNA plays a crucial role in regulating the alternative splicing of its corresponding sense gene, ZEB2, a key transcription factor involved in epithelial-mesenchymal transition. The underlying mechanism involves the physical binding of ZEB2-AS to the 5’ untranslated region (5’UTR) of the ZEB messenger RNA. This strategic binding event specifically impedes the correct execution of alternative splicing, leading to an abnormal retention of introns within the ZEB2 mRNA transcript. The consequence of this intron retention is a significant increase in the translation efficiency of ZEB messenger RNAs, ultimately resulting in an elevated production of the ZEB2 protein. This overexpression, in turn, contributes directly to enhanced cellular proliferation, increased invasiveness, and heightened metastatic potential in various cancer cell types. In other instances, such as with the lncRNA SAMMSON, which is notably associated with melanoma, the experimental knockdown of this specific lncRNA has revealed a potent anti-tumorigenic activity, indicating a critical role in sustaining tumor growth. These findings collectively underscore the immense potential of this class of non-coding RNAs as promising candidate drug targets for the development of innovative cancer therapies. Consequently, the ongoing identification and detailed characterization of cancer-related lncRNAs are paramount for significantly advancing our fundamental understanding of cancer biology, which in turn holds substantial implications for improving both tumor diagnosis and therapeutic strategies.
Thyroid cancer stands as the most prevalent endocrine-related malignancy, accounting for approximately four percent of all diagnosed tumors. Alarmingly, its incidence has witnessed a significant increase, rising threefold over the past three decades. The vast majority of thyroid neoplasms, specifically seventy-five to eighty percent, are classified as differentiated forms, most commonly papillary thyroid carcinomas, or PTCs. These PTCs typically exhibit a shared subset of distinctive genetic alterations that are frequently observed in patient cohorts. These alterations notably include mutually exclusive activating mutations in the BRAF and RAS genes, encompassing H-, K-, and NRAS subtypes. These mutations collectively account for a substantial proportion of cases, representing forty to sixty percent and ten to fifteen percent of cases, respectively. Additionally, RET rearrangements, often referred to as RET/PTC oncogenes, are described in approximately twenty percent of PTC cases. Comprehensive large-scale genomic studies, such as those conducted by The Cancer Genome Atlas Consortium, along with independent analyses, have consistently revealed that PTCs harboring somatic mutations in the BRAF gene, particularly the V600E variant, or those with rearrangements in the RET oncogene, exhibit highly overlapping gene expression patterns. This molecular homogeneity defines a distinct subgroup of patients, often referred to as BRAF-like, which demonstrates a markedly unique transcriptional signature compared to RAS-mutated or highly similar RAS-like tumor samples. The Cancer Genome Atlas also reported that the critical MAPK and PI3K signaling pathways are differentially activated in these specific PTC subgroups. The BRAF-like tumors typically display a robust and sustained activation of the ERK-mediated transcriptional program, a phenomenon partly attributed to the diminished sensitivity of mutant B-Raf to the normal inhibitory feedback mechanisms induced by ERK activation. Conversely, the RAS-like tumors also exhibit a pronounced induction of both MAPK and PI3K/AKT signaling; however, unlike BRAFV600E-mutated tumors, RAS-like PTCs retain functional RAF dimers that remain sensitive to ERK-induced inhibition. These fundamental molecular distinctions manifest as distinct tumor phenotypes, primarily characterized by increased aggressiveness, heightened invasiveness, a higher frequency of disease relapse, and the earlier onset of drug resistance specifically in the BRAF-like PTCs. Despite considerable research efforts dedicated to analyzing these subtype-specific expression patterns and their clinical implications, the non-coding fraction of the PTC genome, particularly its functional contributions, has not yet been systematically or comprehensively explored. This represents a significant gap in our understanding of thyroid cancer biology.
In the present investigative work, our research team employed an innovative de novo discovery approach utilizing extensive RNA-Sequencing datasets derived from papillary thyroid carcinoma samples. The primary objective was to meticulously define the unique BRAF- and RAS-like specific lncRNA expression signatures and, crucially, to identify novel candidate lncRNAs that possess the capacity to modulate, or even interfere with, the expression of known cancer driver genes relevant to thyroid neoplasms. Through this rigorous computational and experimental pipeline, we successfully identified a new natural antisense lncRNA, which we termed COrrelated-to-MET, or COMET. This newly discovered lncRNA exhibited a dramatic and significant upregulation specifically in BRAF-like carcinomas. Furthermore, a highly significant positive correlation was observed between the expression levels of COMET lncRNA and the MET oncogene in these same tumor types, suggesting a potential functional relationship. Our investigations confirmed that this novel lncRNA is transcribed antisense to the MET gene locus, and it predominantly localizes within the cytosolic compartment of the cell, indicating a potential role in post-transcriptional regulation. To ascertain its functional significance, we performed COMET-specific siRNA knockdown experiments in a BRAF-like thyroid cancer cell line. The results unequivocally demonstrated that reducing COMET expression levels led to a significant decrease in the expression of the MET oncogene, as well as other genes intricately associated with the MAPK signaling pathway. Crucially, this reduction in COMET expression also resulted in a profound impairment of cellular proliferation, migration, and invasiveness in the BRAF-like thyroid cancer cells. Thus, our collective data provide compelling confirmation of the oncogene-like behavior of certain lncRNAs within the context of tumors. More specifically, these findings offer novel and substantial evidence for the critical role of the newly identified lncRNA, COMET, which actively sustains tumor cell survival and proliferation, and notably contributes to the MET-driven invasive program characteristic of papillary thyroid carcinomas.
Materials And Methods
Rna-Seq Data Analysis And Identification Of New Lncrnas
The entirety of the computational analysis, encompassing all sequential steps from raw data processing to the final identification of novel lncRNAs, followed a meticulously designed and systematic workflow. For the initial processing of RNA-Sequencing data, advanced bioinformatic tools, specifically Cufflinks (version 2.0.2) and Cuffmerge (version 2.0.2), were systematically employed to perform ab initio transcript reconstruction. This process involves assembling individual RNA-Seq reads into full-length transcript structures without relying on prior gene annotations, allowing for the discovery of novel RNA species. A critical filtering step was subsequently applied to ensure that the analysis focused exclusively on non-coding transcripts; output transcripts that shared sequence identity with known exons of protein-coding genes were carefully filtered out to avoid misclassification. The selection of novel lncRNAs proceeded according to a stringent set of criteria. Firstly, candidates were required to possess a transcript length exceeding 200 base pairs, distinguishing them from shorter non-coding RNAs. Secondly, a lack of prior annotation in the widely utilized GENCODE v19 database was a prerequisite, ensuring the novelty of the identified transcripts. Thirdly, only transcripts exhibiting the presence of more than two exons were considered, suggesting a more complex and potentially functional RNA structure. Finally, the protein-coding potential of each candidate transcript was rigorously evaluated using CPAT (version 1.2.2), with transcripts demonstrating any significant protein-coding capacity being excluded from the lncRNA set.
To systematically identify potential functional relationships between newly discovered lncRNAs and known protein-coding genes, a sophisticated approach was adopted for selecting gene/lncRNA pairs. The precise coordinates of transcription start sites (TSSs), as meticulously defined in the GENCODE v19 annotation, were utilized in conjunction with the ClosestBed function of BEDTools (version 2.17.0A). This powerful tool enabled the efficient identification of the nearest protein-coding gene to each novel lncRNA, establishing a foundational basis for investigating potential cis-regulatory interactions. Crucially, the selection of these pairs was further refined by focusing exclusively on those where both the protein-coding gene and the associated lncRNA exhibited differential expression patterns, suggesting a biologically relevant and co-regulated relationship. Furthermore, to pinpoint lncRNAs potentially involved in oncogenesis, cancer driver genes associated with these newly identified lncRNAs were carefully selected from a curated list of 114 genes previously published by Vogelstein, which are widely recognized for their roles in cancer development. To gain insights into the regulatory chromatin environment of these novel lncRNAs, comprehensive data on peaks of various chromatin markers, including H3K27Ac, H3K4me1, p300 binding, and DNase Hypersensitive Sites, were systematically downloaded from the ENCODE project (2012 data release). BEDTools was then meticulously applied to detect these specific chromatin marks within a 1-kilobase window surrounding the transcription start sites of the newly identified lncRNAs. The presence of these marks is indicative of actively regulated genomic regions, providing further evidence for the functional significance of the discovered lncRNAs. The exact exon/intron structure of the critical COMET lncRNA, initially defined through the high-throughput RNA-Sequencing analysis, underwent a multi-stage validation process to ensure its accuracy and robustness. This validation was first achieved through CAGE (Cap Analysis of Gene Expression) data from different cell lines obtained from the extensive FANTOM5 study, providing independent transcriptional evidence. Subsequently, the structure was experimentally confirmed through a combination of traditional molecular biology techniques, including reverse transcription-polymerase chain reaction (RT-PCR), gene cloning, and direct Sanger sequencing, providing definitive experimental verification of its precise architecture.
Cell Lines
For the purposes of this comprehensive study, human papillary thyroid carcinoma cell lines, specifically TPC-1 and BCPAP, along with the normal thyroid follicular epithelial cell line, Nthy-ori 3-1, were generously provided in 2015 by Professor Alfredo Fusco and Professor Massimo Santoro, respectively. Prior to their utilization in experiments, the crucial mutational status of the tumor cell lines was rigorously confirmed to ensure their appropriate characterization for subtype-specific analyses. This confirmation involved direct Sanger Sequencing of PCR amplicons derived from the BRAF gene protein-coding region for the BCPAP cell line, allowing for precise identification of its BRAF mutation status. For the TPC-1 cell line, which is known for its RET activation, quantitative PCR (qPCR) analysis was employed to assess and confirm the presence of RET oncogene activation. The tumor cell lines, TPC-1 and BCPAP, were routinely cultured and maintained in Dulbecco’s modified Eagle’s medium (DMEM), a standard cell culture medium, which was additionally supplemented with 10% fetal bovine serum (FBS), 2mM glutamine, 100 units/mL penicillin, and 100 units/mL streptomycin to provide essential nutrients and prevent bacterial contamination. In contrast, the normal thyroid follicular epithelial cell line, Nthy-ori 3-1, was cultured in RPMI-1640 medium under identical supplementation conditions. All cell lines employed in this research were meticulously maintained in a controlled environment at 37°C with a 5% CO2 atmosphere, conditions optimal for mammalian cell growth. Furthermore, to ensure the integrity and reliability of experimental results, all cell lines were periodically assayed for the presence of Mycoplasma contamination, a common and often undetected issue in cell culture that can significantly impact experimental outcomes. This assessment was performed using highly sensitive methods including chemiluminescence and specific PCR-based assays, with the latest confirmation test conducted in January 2018. To maintain experimental consistency and minimize potential variations associated with prolonged passaging, all experiments conducted on these cell lines were strictly performed within a defined range, specifically between the 4th and the 16th cell passage.
Rna Fractionation And Fish
To precisely determine the subcellular localization of RNA molecules, especially the novel COMET lncRNA, nuclear and cytoplasmic RNA fractions were meticulously isolated from cellular samples. This critical separation was achieved through the use of the Cytoplasmic and Nuclear RNA purification kit obtained from Norgen Biotek Corp, diligently adhering to the manufacturer’s detailed instructions to ensure optimal and accurate fractionation. To validate the successful and distinct separation of these cellular compartments, specific control genes were employed: GAPDH served as a robust indicator for the cytosolic fraction, while U2 snRNA was utilized as a reliable marker for the nuclear fraction. The primers required for the amplification and detection of these control genes, along with the lncRNA of interest, were carefully selected and documented.
Furthermore, to visually confirm the cellular localization of COMET, Fluorescent In Situ Hybridization (FISH) was performed. For this highly sensitive technique, a custom-designed pool comprising six fluorescently labeled probes, specifically engineered to target the full-length sequence of the COMET lncRNA, was procured from Integrated DNA Technologies. Cells designated for FISH analysis were meticulously grown on glass slides to ensure optimal adherence and morphology. Following growth, the cells were carefully fixed using a 4% Formaldehyde solution to preserve their cellular structure and RNA integrity. Subsequent permeabilization was achieved by incubating the fixed cells in a 0.1% Triton solution, allowing the fluorescent probes to access the intracellular RNA targets. Hybridization of the probes to their target RNA was facilitated by an overnight incubation at 37°C in a specialized hybridization buffer, which comprised a precise mixture of 1g of Dextran, 10% Formamide, and 2X SSC (Saline-Sodium Citrate buffer). To delineate distinct cellular components for comprehensive imaging, DAPI (4′,6-diamidino-2-phenylindole) was employed to specifically stain the cell nucleus, while Actin Green was utilized to visualize the cytoplasm. The resulting fluorescent signals, indicative of COMET lncRNA localization, were expertly visualized using a Zeiss Axiophot upright microscope, which was additionally equipped with a Nikon Coolpix995 camera for high-resolution image acquisition. This dual approach of RNA fractionation and FISH provided compelling and complementary evidence regarding the precise subcellular distribution of the COMET lncRNA.
Rt-Pcr, Qpcr And Rna Interference
For the comprehensive analysis of gene expression, total RNA was meticulously isolated from cellular samples utilizing TRIzol reagent, a highly effective denaturing solution known for its robust RNA extraction capabilities. Following isolation, the purified RNA was subsequently reverse transcribed into complementary DNA (cDNA) using the High Capacity cDNA Reverse Transcription Kit from Invitrogen. This critical step was performed in strict adherence to the manufacturer’s instructions, ensuring optimal cDNA synthesis efficiency and fidelity, which is paramount for accurate downstream quantitative analysis. Quantitative Polymerase Chain Reactions (qPCRs) were then systematically performed on the CFX Connect Real-Time PCR Detection System manufactured by Bio-Rad. The reactions incorporated iTaq Universal SYBR Green Supermix, also from Bio-Rad, which enables real-time monitoring of DNA amplification through fluorescence detection. The methodology employed for qPCR was consistent with previously established protocols, ensuring reproducibility and comparability of results. To accurately normalize gene expression data and account for variations in RNA input and reverse transcription efficiency, the PPIA gene (Peptidylprolyl Isomerase A, also known as Cyclophilin A) was consistently used as a reliable reference gene. Relative gene expression levels were subsequently calculated using the comparative 2-ΔΔCt method, a widely accepted quantitative approach that precisely determines the fold change in target gene expression relative to a control. The specific sequences of the qPCR oligonucleotide pairs, carefully designed for accurate amplification of target genes, were rigorously selected and documented.
To investigate the functional roles of specific genes and the newly identified COMET lncRNA, small interfering RNAs (siRNAs) were utilized to induce targeted gene knockdown in TPC-1 and BCPAP cell lines. These cells were efficiently transfected with the siRNAs using Oligofectamine from Life Technologies, following the manufacturer’s meticulously outlined recommendations for optimal transfection efficiency and minimal cytotoxicity. Custom siRNAs, specifically designed to target the 3’ region of the COMET lncRNA, consisting of two distinct duplex siRNAs with sequences 5’-GGAAGTTTGAGTGACTCAT-3’ and 5’-GCTCAGAAATGACACAATT-3’, were procured from IDT. In addition to COMET, siRNAs targeting other genes of interest were also acquired: two specific siRNAs for MET (Origene #SR302873), two specific siRNAs for BRAF (Origene #300470), and two specific siRNAs for FOSL2 (Origene #SR301649). A non-targeting control siRNA (Origene #SR30004) was also included in all experiments to account for non-specific effects of transfection. The efficiency of gene knockdown for each targeted transcript was rigorously assessed by performing qPCR, ensuring that the desired reduction in mRNA levels was achieved prior to conducting downstream functional assays. This systematic approach guaranteed that any observed phenotypic changes could be directly attributed to the specific gene silencing.
Viability And Apoptosis Assays
To comprehensively assess cellular health and survival, TPC-1 and BCPAP cells were meticulously plated into 96-well white opaque plates, a format optimized for luminescence-based assays. Cell viability was quantitatively determined using the CellTiter-Glo kit from Promega, a robust luminescent assay that measures ATP as an indicator of metabolically active cells. This assessment was performed at multiple time points: 0, 24, 48, 72, and 96 hours following the targeted knockdown of COMET, allowing for a detailed kinetic analysis of its impact on cell viability. To investigate the potential induction of programmed cell death, or apoptosis, the activity of caspases 3 and 7, key executioner enzymes of apoptosis, was measured using the Caspase-Glo 3/7 Assay, also from Promega. This luminescent assay was conducted in both COMET knockdown cells and control cells at 16, 24, and 48 hours after silencing, providing insights into the apoptotic pathway. All luminescence measurements generated from these assays were precisely quantified using a Victor X3 Multimode plate reader manufactured by Perkin Elmer, ensuring accurate and reproducible data acquisition for the comprehensive evaluation of cell viability and apoptosis.
Colony Forming And Proliferation Assays
To evaluate the long-term proliferative and tumorigenic potential of thyroid cancer cells following gene manipulation, specific assays were conducted. Twenty-four hours after the successful knockdown of COMET, a precise number of TPC-1 cells, specifically 150 cells, were carefully plated into six-well plates. These cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) generously supplemented with 10% FBS, providing optimal conditions for colony formation. After an incubation period of eight days, allowing sufficient time for colony development, the resulting cell colonies were meticulously fixed using 20% methanol to preserve their structure, and subsequently stained with a 0.5% crystal violet solution to render them visible. The stained cell clones were then accurately counted and analyzed using ImageJ software, a widely utilized image processing program developed by the National Institutes of Health (NIH), ensuring objective quantification of colony numbers.
For a more direct assessment of cellular proliferation dynamics, COMET knockdown cells and their corresponding control cells were harvested. Following harvesting, a precise concentration of 0.1% CellTrace Violet reagent was added to the cells, strictly adhering to the manufacturer’s protocol. This fluorescent dye is designed to be taken up by cells and then progressively diluted with each cell division, allowing for the tracking of proliferation. Intracellular CellTrace content was subsequently measured using flow cytometry, specifically with a Becton Dickinson FACSCanto system, at three distinct time points: 24, 48, and 72 hours post-labeling. The reduction in fluorescence intensity over time provided a quantitative measure of cell division and proliferation rates, offering a comprehensive understanding of the impact of COMET silencing on cellular growth.
Cell Migration And Invasion Assays
To assess the impact of COMET knockdown on the crucial cellular processes of migration and invasion, which are hallmarks of cancer progression, specialized transwell assays were performed. Twenty-four hours after the successful transfection with siRNAs, cells were carefully harvested and subsequently seeded into the upper chamber of a transwell insert, which featured an 8 μm pore size and was obtained from Corning. This setup allows cells to migrate through the pores towards a chemoattractant in the lower chamber. Following a 24-hour incubation period, the cells that had successfully migrated through the pores to the underside of the membrane were meticulously stained with 0.1% crystal violet for 30 minutes, making them visible for quantification. After staining, the crystal violet dye was solubilized using 10% acetic acid, and its concentration was then precisely measured by determining its absorbance at 595 nm using a spectrophotometer. This photometric measurement provided a quantitative assessment of the number of migrated cells.
The invasion assay, which evaluates a cell’s ability to degrade and move through an extracellular matrix, was conducted using similar transwell chambers, but with a crucial modification: the transwell membranes were pre-coated with a 2% Matrigel solution (Corning). Matrigel mimics the basement membrane and presents a barrier that cells must actively invade. The procedure for counting invading cells after the specified incubation period was identical to that described for the migration assay, ensuring consistency in quantification methods. These assays collectively provided critical insights into the role of COMET in modulating the metastatic potential of thyroid cancer cells.
Physical And Chemical Stimuli
To investigate the transcriptional response of TPC-1 cells to environmental stressors, particularly oxygen deprivation, these cells were subjected to controlled hypoxic conditions. Specifically, TPC-1 cells were placed in a specialized chamber continuously gassed with 1% O2 at 37˚C for a period of 24 hours to induce a state of cellular hypoxia. Concurrently, control cells were maintained under standard normoxic conditions, which involved incubation at 21% O2 and 5% CO2 at 37˚C for an equivalent duration, providing a direct comparison for evaluating hypoxia-induced changes.
To explore the activation of specific signaling pathways, Nthy-ori 3-1 cells were first subjected to a starvation period, being cultured in RPMI-1640 medium devoid of FBS for 24 hours to synchronize cellular responses and deplete growth factor influence. Subsequently, Epidermal Growth Factor (EGF), obtained from Thermo Scientific, was added to the cell culture medium at a concentration of 100 ng/ml. Cells were then exposed to EGF for various indicated time points to track the dynamic cellular response. To confirm the successful induction of the MAPK pathway, Western Blot analysis was meticulously performed to detect changes in the phosphorylation status of ERK (p-Erk), a key downstream component of this pathway. Similarly, to investigate the effects of Hepatocyte Growth Factor (HGF), also from Thermo Scientific, on Nthy-ori-3-1 cells, HGF was administered at a concentration of 100 ng/ml after a 24-hour starvation period, allowing for the study of MET receptor activation.
For experiments involving pharmacological manipulation of signaling pathways, BCPAP cells were treated with a fixed concentration of vemurafenib (VMR), a well-known BRAF inhibitor, acquired from Selleckchem.com. To serve as a crucial experimental control, an equivalent volume of vehicle (DMSO) was added to parallel control cells for the same duration, ensuring that any observed effects were attributable solely to vemurafenib treatment. Following VMR treatment, cell viability was rigorously assessed using a viability assay, and COMET expression levels were quantitatively determined via qPCR, providing a comprehensive evaluation of the drug’s impact on both cellular survival and the lncRNA’s expression.
Western Blot Analysis
To meticulously analyze protein expression and phosphorylation status, TPC-1 and Nthy-ori 3-1 cell lines were carefully prepared by lysing them in a specialized lysis buffer. This buffer contained a precise composition, including 50mM HEPES, 150mM NaCl, 10mM EDTA, 10mM Na4P2O7, 2mM Na3VO4, 100mM NaF, 30mM glycerol, and 10% Triton X-100. Crucially, this lysis buffer was freshly supplemented with a comprehensive cocktail of protease and phosphatase inhibitors, obtained from Thermo Scientific, to effectively prevent protein degradation and dephosphorylation during the extraction process. The total protein concentration in each lysate was accurately quantified using a colorimetric assay, specifically the Bradford protein assay from Bio-rad, ensuring consistency in protein loading for subsequent analysis.
Equal amounts of protein lysate were then subjected to Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE) using either 6% or 10% polyacrylamide gels, chosen based on the molecular weight of the target proteins to ensure optimal separation. Following electrophoretic separation, the resolved proteins were efficiently transferred from the gel onto polyvinylidene difluoride (PVDF) membranes via electrophoresis, enabling immobilization for antibody detection. To minimize non-specific antibody binding, the membranes were thoroughly blocked in Tris-buffered saline–Tween 20 (TBS-T) containing 5% non-fat dried milk for a predetermined duration. Subsequently, the membranes were carefully probed with highly specific primary antibodies targeting the proteins of interest. These included c-Met (dilution 1:10000, incubated for 3 hours at room temperature; Cell Signaling #8198), p-Erk (dilution 1:1000, incubated for 3 hours at room temperature; Cell Signaling #9101), and Hsp90 (dilution 1:2000, incubated for 1 hour at room temperature; Origene #TA500494). After extensive washing with TBS-T to remove unbound primary antibody, the membranes were then incubated for 1 hour at room temperature with appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (Bio-rad #170-6516, #170-6515), which bind to the primary antibodies. Finally, immunoreactive protein bands were visually detected on X-ray film using the enhanced chemiluminescence method from Thermo Scientific, allowing for the sensitive and specific visualization of target proteins.
Statistical Analysis
To ensure the robustness and reliability of the experimental findings, all conducted experiments were independently repeated a minimum of three times. This meticulous approach to replication is fundamental for validating results and minimizing the influence of random variations. For the statistical assessment of differences between distinct experimental groups, the two-tailed Student’s t test was rigorously applied. This widely accepted statistical method is particularly suitable for comparing the means of two independent groups, assuming a normal distribution of data. To determine statistical significance, a p-value of less than 0.05 was established as the threshold. This conventional criterion indicates that the observed difference between groups is unlikely to have occurred by chance, thereby supporting the validity of the experimental conclusions.
Results
Identification Of New Long Non-Coding Rnas Altered In Papillary Thyroid Carcinoma
To comprehensively investigate the specific expression patterns of lncRNAs that distinguish BRAF-like and RAS-like papillary thyroid carcinomas (PTCs), a sophisticated computational workflow was meticulously established for the de novo assembly of transcripts. This extensive analysis was performed on our previously published RNA-Sequencing datasets, which comprised 22 thyroid biopsies, ultimately enabling the definition of a novel and more refined PTC model transcriptome. The subsequent filtering and identification process for new lncRNAs adhered to stringent criteria: only transcripts exceeding 200 nucleotides in length, possessing a coding potential score below 0.364 (as determined by specialized algorithms), and crucially, not overlapping with any previously annotated gene loci, were considered. This rigorous selection process led to the identification of an impressive total of 454 novel lncRNAs. The validity and accuracy of this de novo assembly procedure were further corroborated by analyzing the kernel density plots of transcript abundance. These plots compellingly demonstrated that the newly detected lncRNAs exhibited a distribution pattern strikingly similar to that of already annotated lncRNAs, providing strong evidence for the authenticity of the discovered transcripts.
Remarkably, both previously known and these newly discovered lncRNAs displayed markedly different expression levels, with a statistically significant false discovery rate (FDR) of less than 0.05, when comparing patient samples to control samples. This differential expression was clearly evident through unsupervised hierarchical clustering, which robustly grouped samples based on their lncRNA expression profiles. Similarly, an in-depth analysis of the BRAF- and RAS-like specific expression signatures for protein-coding genes distinctly highlighted the transcriptional differences between these two subgroups, thereby reaffirming that the differential activation of the MAPK and PI3K pathways in these molecularly defined subgroups indeed dictates rather distinct transcriptional programs. Interestingly, the expression profiles of lncRNAs within these two subgroups were also profoundly affected by the presence of specific molecular alterations. Specifically, the expression of 202 lncRNAs was found to be significantly different between BRAF- and RAS-like PTCs, with 77 of these being over-expressed and 125 under-expressed in BRAF-like PTCs. Notably, a substantial proportion, 60 of these differentially expressed lncRNAs, remained unannotated in publicly available repositories, highlighting their novelty and the importance of this discovery.
With the overarching aim of identifying hitherto unannotated lncRNAs that met stringent functional and correlational criteria, a comprehensive selection strategy was implemented. This strategy sought lncRNAs that were: first, aberrantly expressed in PTC; second, differentially expressed between the BRAF- and RAS-like subgroups; third, physically localized in close proximity to known cancer driver genes; and fourth, exhibited a high degree of co-expression (correlation) with these established cancer driver genes. To achieve the third and fourth criteria, messenger RNA (mRNA)/lncRNA pairs were meticulously selected using a “nearest transcription start site” (TSS) approach, which links lncRNAs to the closest protein-coding gene, facilitating the investigation of potential regulatory relationships. Through this systematic and multifaceted screening process, the novel lncRNA, subsequently named COMET (COrrelated-to-MET), was successfully identified. Intriguingly, COMET was found to be positively and highly correlated (with a Pearson’s correlation coefficient of r=0.7) with the MET oncogene, suggesting a significant functional interplay between the two.
Characterization Of Comet: A New Natural Antisense Lncrna
The newly identified COMET lncRNA has been precisely mapped to chromosome 7q31.2, a genomic region of significant interest in cancer research. A critical characteristic of COMET is its transcriptional orientation: it is transcribed from the opposite strand relative to the MET oncogene. This inverse orientation implies a natural antisense relationship, a finding that is particularly intriguing given the known oncogenic role of MET. While COMET partially overlaps with previously annotated GENCODE entries AC006159.3 and AC06159.4, which correspond to RefSeq LINC01510, our detailed de novo reconstruction of transcripts, further validated through rigorous PCR experiments, revealed a more complex scenario. This analysis demonstrated that multiple distinct transcripts arise from this genomic locus, with one of them indeed aligning with LINC01510. More profoundly, the RNA-Sequencing data allowed for the discovery of a novel splicing variant, herein designated as LINC01510_var, which notably exhibits the skipping of exon 2 compared to the already annotated LINC01510. Even more significantly, the RNA-Seq data revealed the expression of longer, previously unannotated isoforms of COMET, with their transcription start sites (TSSs) located within the first intron of the MET gene. These longer, unannotated isoforms provide compelling evidence that COMET functions as a Natural Antisense Transcript (NAT) of the MET oncogene, suggesting a direct regulatory relationship.
To provide robust support for the notion of an actively transcribed locus for COMET, our research utilized two distinct but complementary datasets. Initially, our own comprehensive RNA-Seq data were leveraged to precisely assess the genomic structure of this novel NAT and to accurately map its transcription start site. Subsequently, we incorporated recently released data from the extensive FANTOM5 project, which utilizes CAGE (Cap Analysis of Gene Expression) to provide high-resolution mapping of TSSs, further bolstering our structural and TSS predictions. Beyond transcriptional evidence, we extended our investigation to the epigenetic landscape. We meticulously searched publicly available ENCODE Consortium data for the presence of specific chromatin marks in the genomic region surrounding COMET’s TSS and its putative promoter. This in-depth analysis revealed prominent peaks of both H3K4me1 and H3K27Ac, which are well-established epigenetic marks characterizing actively transcribed promoters and enhancers, thus strongly indicating an active regulatory region. Furthermore, a high enrichment of evolutionarily conserved DNase Hypersensitive Sites (DHSs) was observed around COMET’s TSS. These DHSs, detected in an impressive 83 out of 125 cell lines analyzed by ENCODE, are distinctive indicators of open chromatin, suggesting that the DNA in this region is accessible for transcription factor binding and gene expression. Complementing these broad ENCODE findings, we also integrated DHS data from an independent study by Jin, obtained using Pico-Seq on actual papillary and follicular thyroid cancer samples. Notably, many of the DHSs identified in this external dataset were found to be associated with promoters and enhancers of known cancer-related genes. Remarkably, our analysis could pinpoint a specific DHS, present in more than one thyroid cancer sample, located within the putative COMET promoter region, less than 1.5 kilobases upstream of its transcription start site, further strengthening the evidence for its active regulatory potential in cancer.
Additionally, an examination of the MiTranscriptome database, a valuable resource for lncRNA expression, revealed a preferential expression pattern for COMET specifically within thyroid carcinoma samples, a characteristic that differentiates it from LINC01510. This tissue-specific expression was independently confirmed through semiquantitative RT-PCR analyses performed on a panel of different cell lines and actual tumor samples. Interestingly, some of these samples, despite exhibiting high levels of the MET oncogene, did not express COMET, suggesting a nuanced and potentially independent regulatory mechanism for COMET expression beyond direct MET activation in all contexts. The restricted expression pattern observed for COMET, combined with its very low coding potential score (CP=0.0248), which is significantly lower than that typically observed for protein-coding genes, strongly supports its classification as a novel long non-coding RNA. Finally, an analysis of its evolutionary conservation revealed that COMET is conserved specifically in primate genomes, but not across other vertebrate species, suggesting a relatively recent evolutionary emergence and potentially specialized functional roles within primates.
Comet Lncrna Is Highly Expressed In Braf-Like Carcinomas
Consistent with the general understanding that long non-coding RNAs typically exhibit lower expression levels compared to protein-coding genes, our RNA-Sequencing data revealed that COMET is indeed expressed at a markedly lower abundance than its neighboring MET oncogene. Despite this, the RNA-Seq data also compellingly demonstrated that both COMET and MET are significantly over-expressed in the BRAF-like subgroup of papillary thyroid carcinomas (PTCs). This observation strongly suggests that the pathological hyper-activation of Ret and B-Raf proteins, characteristic of BRAF-like tumors, plays a contributory role in their increased expression levels. To further corroborate this crucial finding, we meticulously stratified a larger, independent cohort of samples from our previous work, which included 50 tumor specimens and 11 healthy control tissues. This stratification was based on the presence of common somatic mutations, specifically BRAFV600E, H-, K-, and N-RAS mutations in codons 12, 13, and 61, as well as RET gene rearrangements, allowing for the classification into BRAF-like (n=32) and RAS-like (n=18) subgroups. In this independent and expanded cohort, we rigorously confirmed through quantitative PCR (qPCR) analysis that both the natural antisense lncRNA COMET and its proximal oncogene MET are indeed significantly over-expressed in BRAF-like PTCs when compared to both RAS-like PTCs and healthy control samples.
In a parallel and equally robust validation effort, we analyzed publicly available exome sequencing data from The Cancer Genome Atlas (TCGA) specifically pertaining to thyroid carcinoma (THCA), an extensive dataset comprising 507 patient samples. We systematically stratified these patients according to their primary driver genetic alterations. Subsequently, we analyzed the corresponding RNA-Seq data from these patients. This comprehensive analysis in a significantly larger and independent cohort of PTCs unequivocally confirmed our initial findings, further solidifying the observation that COMET and MET are consistently overexpressed in BRAF-like tumors. This multi-pronged validation approach, utilizing independent sample sets and large-scale public data, strongly supports the conclusion that COMET lncRNA is highly expressed in BRAF-like carcinomas, suggesting its potential involvement in the distinct molecular pathology of this aggressive PTC subtype.
Comet Is A New Mapk-Induced Cytosolic Lncrna
To definitively validate whether COMET expression is a direct transcriptional event occurring downstream of the constitutive activation of Ret and B-Raf proteins, we systematically measured COMET levels in well-characterized papillary thyroid carcinoma (PTC) cell lines. Our chosen models included BCPAP cells, which harbor the BRAFV600E mutation, and TPC-1 cells, characterized by a RET gene rearrangement. These were compared against Nthy-ori 3-1, an immortalized normal thyroid cell line, serving as a critical non-cancerous control. As anticipated, the expression of COMET, along with that of MET, was found to be significantly elevated in the BRAF-like cell lines (BCPAP and TPC-1) when contrasted with the normal thyroid cells. Notably, the TPC-1 cell line displayed the highest levels of COMET expression among the tested models, making it an ideal in vitro system for the majority of the subsequent phenotypic assays described.
Given that BRAF-like PTCs are notably characterized by a robust and sustained activation of the ERK-mediated transcriptional program, we were prompted to investigate whether COMET expression is specifically induced as a consequence of MAPK pathway activation. To address this, we experimentally stimulated the RAS-RAF-MEK-ERK signaling axis in Nthy-ori 3-1 cells by introducing Epidermal Growth Factor (EGF). The successful induction of the MAPK pathway was promptly confirmed by a rapid and significant increase in phosphorylated ERK (p-Erk) levels, as assessed by Western Blot analysis. This early p-Erk increase was subsequently followed by a significant and measurable increase in COMET lncRNA levels, providing strong evidence that COMET expression is indeed responsive to MAPK activation. Conversely, when we chemically and genetically inhibited the constitutively active MAPK pathway in the BRAF-mutated BCPAP cell line, either by directly blocking the hyper-activated B-Raf protein with vemurafenib or through specific BRAF gene knockdown, we observed a significant decrease in COMET levels. However, it was noted that in cells treated with increasing doses of vemurafenib, COMET expression did not exhibit a clear dose-dependent down-modulation, suggesting that additional downstream factors beyond direct BRAF activity might also contribute to COMET regulation.
To gain a more profound understanding of the transcriptional mechanisms by which this lncRNA is induced downstream of MAPK signaling, we leveraged publicly available ChIP-Seq ENCODE data, specifically focusing on transcription factor binding events around COMET’s transcription start site (TSS) and its predicted promoter region. This detailed analysis revealed an enriched binding of the Fosl1:Fosl2 complex (AP-1 transcription factor complex) in more than one cell line, specifically within a DNase Hypersensitive Site (DHS) located approximately 1.5 kilobases from COMET’s TSS. Intriguingly, FOSL2 itself was found to be over-expressed in BRAF-like tumors in both our own RNA-Seq datasets and the TCGA datasets. Additionally, publicly available GTEx RNA-Seq data further indicated that FOSL2, as a member of the AP-1 complex, is highly expressed in thyroid tissue, lending further support to its potential role in thyroid biology and disease.
To experimentally validate FOSL2 as a potential transcriptional factor directly involved in regulating COMET expression, we first confirmed its induction in a RAS-RAF-MEK-ERK-dependent manner in Nthy-ori 3-1 cells. As expected, EGF-mediated stimulation consistently caused a marked increase in FOSL2 expression levels, establishing a clear link between MAPK activation and FOSL2 induction. Subsequently, siRNA-mediated knockdown of FOSL2, a key component of the AP-1 complex, in the TPC-1 cell line resulted in a significant reduction (up to 50%) of COMET lncRNA levels at 12, 24, and 48 hours post-transfection. These cumulative data strongly suggest that FOSL2 plays a significant role in contributing to COMET expression in papillary thyroid carcinomas, particularly in tumors characterized by marked activation of MAPK signaling, such as the BRAF-like patient cohort and derived tumor cell lines.
Finally, a crucial aspect of lncRNA characterization involves determining its subcellular localization, as this often provides key insights into its functional mechanisms. To this end, we evaluated COMET’s cellular distribution through two complementary methods. RNA fractionation coupled with quantitative PCR revealed that, similar to GAPDH, a well-established control protein-coding gene known to be predominantly cytosolic, COMET exhibited a preferential enrichment in the cytosolic RNA fraction. This finding was further and independently corroborated by Fluorescent In Situ Hybridization (FISH), which visually confirmed the prevalent cytoplasmic localization of this novel natural antisense lncRNA, suggesting a potential role in post-transcriptional regulatory processes within the cytoplasm.
Comet Lncrna Knockdown Impairs The Expression Of Different Oncogenes
To gain deeper insights into the potential functional role of COMET and to ascertain its broader biomedical significance within the complex landscape of papillary thyroid cancer, a comprehensive guilt-by-association analysis was systematically performed. Leveraging the extensive TCGA-based lncRNA atlas known as TANRIC, we meticulously assessed the correlation of COMET expression with the expression profiles of numerous other genes in thyroid carcinoma samples. This correlation analysis was performed using Pearson’s correlation coefficient, a robust statistical measure of linear association. In perfect alignment with our initial discovery, the analysis revealed that the highest positive correlation value was indeed measured with the neighboring oncogene MET, with a highly significant Pearson’s r-value of 0.88. This strong correlation further solidified the hypothesis of a direct functional relationship.
Beyond individual gene correlations, a detailed pathway analysis was subsequently conducted on genes that exhibited a statistically significant correlation with COMET (defined as a False Discovery Rate less than 0.05 and a Pearson’s correlation coefficient greater than or equal to 0.7). This pathway analysis remarkably revealed a significant enrichment for genes associated with the MAPK signaling pathway, a central pathway implicated in thyroid cancer progression. Intriguingly, many of these highly correlated genes were also found to be over-expressed in the BRAF-like cohort of PTC samples within the TCGA dataset, reinforcing the notion that COMET’s regulatory influence might be particularly pronounced in this specific aggressive subtype of thyroid cancer.
To directly address the functional relationship between the newly identified COMET lncRNA and other genes within its co-expression network, we proceeded with a crucial experimental approach: targeted knockdown of COMET in BRAF-like tumor cells, followed by a meticulous measurement of the expression levels of these interconnected genes. Small interfering RNAs (siRNAs), which are highly effective tools for targeting cytosolic transcripts, were specifically designed and utilized to achieve efficient knockdown of the COMET natural antisense transcript. The silencing efficiency of COMET knockdown, using a pool of these specific siRNAs, was rigorously assessed at 24, 48, and 72 hours post-transfection in two distinct BRAF-like cell lines, ensuring a comprehensive evaluation of the knockdown kinetics. Additionally, to further confirm the specificity of COMET targeting and rule out off-target effects, individual antisense oligonucleotides (ASOs) and a pooled combination of ASOs were also employed, providing complementary validation of the silencing effect. However, the siRNAs consistently demonstrated superior silencing efficiency, a finding that aligns well with COMET’s observed cytosolic localization. Consequently, all subsequent knockdown experiments were primarily conducted using the optimized pool of siRNAs. As clearly demonstrated, COMET knockdown led to a significant impairment in the expression levels of most genes within the established co-expression network, with particularly notable reductions observed for AKT3, CREB5, and DUSP5. A substantial and consistent drop in both MET mRNA and protein levels was consistently measured following COMET knockdown, highlighting a profound regulatory impact.
To meticulously avoid any potential confounding effects arising from the siRNAs’ off-target activity on MET expression, a critical control experiment was performed. We assessed the silencing specificity by transfecting HEK293 cells, a cell line known to express MET but crucially not COMET, with the identical COMET-specific pool of siRNAs. Remarkably, no variation in MET oncogene levels was detected in this cell line upon COMET knockdown, unequivocally indicating that the COMET-specific siRNAs do not directly affect MET oncogene levels. This experiment provides strong evidence that the observed reduction in MET in thyroid cancer cells is indeed a specific consequence of COMET knockdown, rather than a non-specific siRNA effect. We further explored the possibility of a reciprocal regulation, investigating whether MET knockdown could, in turn, affect the levels of COMET and other genes belonging to the COMET co-expression network. Intriguingly, upon achieving a significant reduction in MET expression (up to 80%) in TPC-1 cells, we could not measure any significant variation in COMET levels, nor in the expression of other genes previously identified as COMET-correlated. This compelling finding strongly suggests that the reduction in their expression levels observed following COMET knockdown is largely independent of MET’s own expression levels, implying that COMET acts upstream or parallel to MET in this regulatory cascade.
Furthermore, to delve into a putative common regulatory mechanism for COMET lncRNA and its neighboring oncogene MET, we investigated whether known MET-inducing stimuli could also trigger COMET lncRNA expression. Our focus centered on two well-established stimuli for MET induction: hypoxia and Hepatocyte Growth Factor (HGF) treatment. TPC-1 cells cultured under hypoxic conditions exhibited a significant twofold increase in COMET levels when compared to cells grown under normoxic conditions, unequivocally demonstrating that hypoxia-mediated transcriptional responses indeed impact COMET expression. Similarly, acute stimulation of Nthy-ori 3-1 cells with HGF, the specific ligand for the membrane-bound c-Met receptor, also induced a notable increase in COMET lncRNA levels. Importantly, treatment with HGF was also observed to induce FOSL2 expression, further strengthening the hypothesis of FOSL2′s contribution to the regulation of COMET expression. These multifaceted lines of evidence collectively paint a detailed picture of COMET’s expression regulation and its intricate relationship with oncogenic signaling pathways.
Comet Knockdown Inhibits Cell Proliferation And Induces Apoptosis
Given that the targeted knockdown of COMET resulted in a significant reduction in the levels of c-Met, a critical oncogene, as well as a notable decrease in the expression levels of various other MAPK-related oncogenes within thyroid cancer cells, we proceeded to thoroughly investigate the downstream phenotypic consequences of COMET knockdown on the behavior of these tumor cells. To this end, we meticulously measured both the viability and proliferation rates of the TPC-1 cell line following COMET knockdown, assessing these parameters at multiple time points to capture the dynamic effects. Intriguingly, the percentage of viable cells in the COMET-knockdown group significantly decreased over time, showing a reduction of up to approximately 40% at the 72-hour mark when compared to scramble-transfected control cells. This substantial reduction in viability underscores COMET’s crucial role in maintaining thyroid cancer cell survival. Concurrently, COMET knockdown markedly suppressed cellular proliferation, as quantitatively measured by a dedicated cell proliferation assay. This result further reinforces the notion that COMET is essential for the unchecked growth characteristic of these cancer cells.
In a concerted effort to unravel the precise mechanisms by which COMET regulates the growth of thyroid cancer cells, we additionally investigated whether COMET knockdown was capable of inducing apoptosis, the process of programmed cell death. This was assessed by measuring the enzymatic activity of caspases 3 and 7, which are key executioner caspases in the apoptotic pathway. As clearly demonstrated by our findings, COMET knockdown led to a significant increase in the number of apoptotic TPC-1 cells as early as 24 hours after gene silencing. This induction of apoptosis provides a critical mechanistic explanation for the observed decrease in cell viability and proliferation, indicating that COMET actively inhibits programmed cell death in these cancer cells.
Finally, to conclusively determine whether COMET knockdown impacts the intrinsic tumorigenic potential of thyroid tumor cells in an in vitro setting, we conducted a colony-forming assay. This assay evaluates a cell’s capacity to form self-sustaining colonies, a surrogate measure of clonogenic survival and tumorigenicity. The colony-forming ability of COMET-knockdown TPC-1 cells was compared against that of control cells. As unequivocally shown, silencing COMET profoundly affected the oncogenic capacity of the thyroid cancer cells, manifested both in a significant reduction in the total number of colonies formed and a noticeable decrease in the size of the individual colonies. These combined results from viability, proliferation, apoptosis, and colony-forming assays provide compelling evidence that COMET lncRNA plays a critical pro-tumorigenic role in papillary thyroid carcinomas, significantly impacting cell survival, growth, and the ability to form colonies.
COMET Silencing Affects Migration And Invasiveness Of Thyroid Cancer Cells
Considering the profound impact of COMET silencing on the expression levels of the c-Met protein, and acknowledging the well-established pivotal role of c-Met in driving invasive growth programs within various tumors, our subsequent investigations were strategically directed toward exploring the specific effects of COMET knockdown on the migratory and invasive capacities of thyroid cancer cells. The comprehensive data obtained from these experiments compellingly illustrate that the targeted reduction of COMET expression significantly diminishes both the intrinsic motility and the overall invasion capacity of TPC-1 cells, with a notable reduction of up to 40%. This finding strongly implicates COMET in the metastatic potential of these cancer cells. In accordance with these observations, tumor cells subjected to COMET knockdown consistently exhibited a significantly lower expression of vimentin, both at the messenger RNA and protein levels. Vimentin is a widely recognized crucial molecular marker of the epithelial-to-mesenchymal transition (EMT), a fundamental cellular process that endows cancer cells with enhanced migratory and invasive properties. Interestingly, this reduction in vimentin expression occurred without any measurable variation in N-cadherin expression, another key EMT marker, suggesting a specific modulation of certain aspects of the EMT program by COMET. Taken together, these compelling in vitro data provide robust support for an oncogenic role for the newly identified COMET lncRNA within thyroid cancer progression. Furthermore, these findings strongly suggest that this oncogenic property relies, at least in part, on its profound modulatory activity on the expression levels of the MET oncogene and other intricately related oncogenes within the MAPK signaling pathway, thereby influencing critical aspects of tumor aggressiveness.
In Vitro Assessment Of The Therapeutic Potential Of COMET Targeting
The activation of receptor tyrosine kinases (RTKs), and particularly the aberrant activation of the c-Met receptor, has been widely reported as a crucial compensatory mechanism employed by tumor cells to evade the inhibitory effects of vemurafenib, a targeted therapy designed to block mutant B-Raf protein. Given that our experimental data strongly support COMET’s role as a novel regulator of MET, we posed a critical question: might COMET also contribute to the responsiveness of BRAF-mutated tumor cells to vemurafenib therapy? To address this, our initial assessment confirmed that, similarly to TPC-1 cells, COMET silencing exerted a powerful inhibitory effect on the viability of BRAFV600E mutated BCPAP cells, leading to a substantial reduction in viability of up to 50%. This effect was accompanied by a concomitant and significant drop in MET protein levels, further underscoring the functional link. A comparable reduction in cell viability, up to 60%, was also observed when BCPAP cells were treated with vemurafenib alone, highlighting the drug’s efficacy.
Building upon these findings, we then subjected BCPAP cells, which had undergone COMET knockdown, to vemurafenib treatment to evaluate whether this combined therapeutic approach might exert an additive or synergistic anti-cancer effect. The data clearly demonstrated that COMET-knockdown BCPAP cells exhibited a markedly improved sensitivity to vemurafenib when compared to control cells, which included cells treated only with vemurafenib or those transfected with control siRNAs. This compelling observation strongly indicates that COMET represents a promising new therapeutic target, which, when modulated, has the potential to significantly enhance the effectiveness of chemotherapy regimens in tumors harboring the BRAFV600E mutation. These results pave the way for future investigations into COMET as a novel strategy to overcome or mitigate therapeutic resistance in a clinically relevant subset of thyroid cancers.
Discussion
Recent large-scale genomic analyses, notably those conducted by The Cancer Genome Atlas (TCGA) and subsequently corroborated by our independent research group, have unequivocally revealed distinct signaling consequences stemming from genetic alterations in the BRAF, RET, and RAS genes within papillary thyroid carcinomas. A particularly striking finding, with profound and intriguing clinical implications, is the observation that BRAF-like tumors exhibit a remarkably robust and sustained activation of the ERK-mediated transcriptional program. This heightened activity is, at least partially, attributable to the diminished sensitivity of mutant B-Raf proteins to the physiological ERK-induced inhibitory feedback loop, which normally serves to dampen signaling. In stark contrast, RAS-like tumors, while also demonstrating a significant induction of both MAPK and PI3K/AKT signaling, retain a greater sensitivity to ERK-mediated inhibition due to the presence of functional Raf dimers. These fundamental molecular distinctions have underscored the urgent necessity to re-evaluate and refine existing thyroid cancer classification schemes, thereby propelling thyroid cancer therapy towards the era of precision medicine, where treatments are tailored to the specific molecular aberrations of each tumor.
Historically, a substantial proportion of research efforts has been directed towards elucidating the roles of protein-coding genes in tumor etiology. Consequently, this intense focus has inadvertently led to a comparatively poor understanding of the multifaceted roles that long non-coding RNAs (lncRNAs) play in neoplastic transformation and, crucially, in mediating responses to chemotherapy. However, in more recent years, a burgeoning number of scientific studies have decisively shifted their focus towards this critically important class of non-coding RNAs. This paradigm shift is driven by the growing recognition that lncRNAs represent highly attractive and promising therapeutic targets, even within the context of papillary thyroid carcinoma (PTC). One of the most recent and compelling demonstrations of this therapeutic potential is the identification of lncRNA AB074169 (lncAB) as a tumor suppressor in PTC tumorigenesis, highlighting the diverse functions these molecules can exert. In the current study, our investigative efforts were specifically concentrated on the systematic identification of lncRNAs that are directly activated by specific mutated oncogenes prevalent in papillary thyroid carcinoma. By meticulously coupling ab initio transcriptome reconstruction with targeted resequencing of somatic mutations, and integrating this with comprehensive analyses of publicly available TCGA omics data and ENCODE tracks for epigenetic marks, we successfully identified COMET as a novel natural antisense transcript. This lncRNA was found to be highly expressed, particularly in the BRAF-like subtype of PTC, suggesting its specific involvement in this aggressive variant. Importantly, the identification of lncRNA profiles intimately associated with specific somatic alterations in PTCs represents a significant advancement. This knowledge not only enhances our fundamental understanding of BRAF-, RET-, and RAS-mediated tumorigenesis and disease progression but also offers new avenues for diagnostic and prognostic markers. In this regard, a few recent reports have indeed proposed certain PTC-deregulated lncRNAs, notably those enriched in BRAF-mutated patients, as new potential diagnostic and therapeutic targets within this specific tumor subtype. Furthermore, recent research has explored the oncogenic or oncosuppressor potential of BRAF-activated lncRNA (BANCR), yielding varied and sometimes conflicting results in the context of PTC, underscoring the complexity of lncRNA function. Similarly, the lncRNA Orilnc1 was recently discovered in BRAF-mutant melanoma, demonstrating its regulation by the RAS-RAF-MEK-ERK signaling pathway, largely mediated through the AP-1 transcriptional complex, further exemplifying the intricate regulation of lncRNAs by oncogenic pathways.
Our current study meticulously identifies COMET as a novel cytosolic lncRNA, characterized by relatively low expression levels, a characteristic consistent with many other functionally important lncRNAs, such as VELUCT. Notably, despite its extremely low abundance, VELUCT has been shown to induce a profound loss-of-function phenotype, underscoring that low expression does not equate to a lack of biological significance. Our findings indicate that COMET is markedly over-expressed in BRAF-like tumors compared to both RAS-driven tumors and healthy thyroid tissue. Crucially, COMET expression is robustly induced downstream of MAPK pathway activation, and its presence is demonstrably required for the sustained survival and growth of thyroid tumor cells. While our study did not definitively delineate the complete transcriptional regulatory mechanisms governing COMET, compelling evidence points towards FOSL2’s contribution. Specifically, the identification of a consensus binding site for FOSL2 upstream of COMET’s transcription start site, coupled with the observed significant reduction in COMET expression upon FOSL2 knockdown, strongly suggests that FOSL2 plays a pivotal role in COMET regulation. This provides a plausible molecular explanation for COMET’s induction in BRAF-like tumors, which are characterized by constitutive activation of the RAF-MEK-ERK pathway. Additionally, the rapid induction of FOSL2 upon Hepatocyte Growth Factor (HGF) treatment, concomitant with an increase in COMET lncRNA levels, further strengthens the hypothesis of FOSL2’s involvement in COMET regulation. However, it is important to acknowledge that the contribution of other transcription factors and/or co-activators to this complex process cannot be entirely excluded. Indeed, as recently reported by Grossman and colleagues in a large-scale analysis, FOSL2 belongs to a cluster of transcriptional activators that, in comparison to other classes of transcription factors (TFs), exhibit a higher propensity to bind to other co-activators, such as p300 and CREB-binding protein, and frequently interact with each other. This suggests a cooperative role among these factors in activating gene transcription. Furthermore, given that the transcription factors comprising the AP-1 complex are generally essential for maintaining accessible chromatin, this accessibility in turn facilitates the binding of other stimulus-regulated transcription factors. Therefore, it is highly probable that FOSL2 is a key MAPK-induced transcription factor contributing to the intricate regulation of COMET expression.
The identification of a precisely defined network of co-expressed genes, whose expression is profoundly perturbed almost exclusively in tumors driven by specific somatic alterations—namely, the BRAFV600E mutation and rearrangement of the RET oncogene—presents a more intricate and nuanced scenario. Within this context, COMET emerges as a compelling new candidate, a non-protein effector of MAPK signaling, playing a significant role in dictating gene expression patterns in these specific tumor subtypes. Intriguingly, the targeted silencing of COMET in BRAF-like tumor cells proves sufficient to significantly down-modulate the expression of a range of cancer-related genes that are integral components of both the MAPK and PI3K signaling pathways, including crucial players such as AKT3, MAPK1, DUSP5, and CREB5. Even more notably, a particularly profound repressing effect was consistently measured on the MET oncogene. This finding is entirely consistent with a very recent and independent report by Cen and colleagues, published concurrently with our work, which demonstrated that LINC01510, a shorter isoform of the COMET lncRNA, actively promotes the proliferation of colorectal cancer cells, at least in part, by modulating MET oncogene expression. Our study herein has unequivocally demonstrated that repressing COMET expression effectively inhibited the viability and proliferation of tumor cells harboring either the RET oncogene rearrangement or the BRAFV600E somatic mutation. Moreover, in congruence with the critical finding that COMET knockdown significantly impairs both the mRNA and protein levels of the oncogene MET—a key driver of the invasive program in various tumors—we have also compellingly demonstrated that COMET repression dramatically curtails both the motility and invasiveness of thyroid tumor cells.
Our findings possess particular relevance, especially for tumors carrying the BRAFV600E mutation. While patients within this subgroup, which constitutes the most abundant fraction of papillary thyroid carcinoma (PTC) as well as anaplastic thyroid carcinoma (ATC) cases, can often be effectively treated with B-Raf inhibitors that specifically target the constitutively active mutated B-Raf protein, a significant clinical challenge arises from the frequent development of drug resistance. Recent research has highlighted that the activation of the c-Met oncogene represents one of the pivotal mechanisms that BRAF-mutated cells adopt to escape vemurafenib-induced blockade of MAPK signaling. Despite this, clinical application of c-Met inhibitors has yielded only partial benefits, ARS-1323 a limitation often attributed to the low specificity of these drugs and their associated off-target effects. Although MET oncogene overexpression was reported in thyroid carcinomas more than two decades ago, it is only recently that it has been specifically proposed as one of the potential contributors to vemurafenib resistance. In line with this, our detailed analysis of independent cohorts of PTC samples consistently revealed that BRAF-like tumors, in stark contrast to RAS-like tumors and healthy thyroid tissues, exhibit significant overexpression of both the MET oncogene and the newly identified COMET lncRNA. The compelling finding that COMET knockdown is capable of substantially impairing the levels of MET and other MAPK-related oncogenes, coupled with the observation that tumors with repressed COMET demonstrate a markedly enhanced response to vemurafenib exposure, positions this lncRNA as a highly attractive therapeutic target. These promising results warrant immediate further investigation in in vivo experimental models to assess its full therapeutic potential. Our collective results strongly advocate for continued and thorough investigation into the non-coding effectors operating downstream of BRAF signaling and suggest that the targeted repression of COMET may indeed represent a novel and potent therapeutic strategy to overcome tumor resistance to BRAF inhibition, ultimately improving clinical outcomes for thyroid cancer patients.
Acknowledgements
The authors express their sincere gratitude for the invaluable support received from several funding sources and individuals. Project SATIN – POR Campania FESR 2014/2020 is acknowledged for its contribution to V.C.’s work, and the Associazione Italiana per la Ricerca sul Cancro (AIRC IG 2013-14689) is acknowledged for its support to A.C. A special thank you is extended to Professor Alfredo Fusco for his generosity in providing the human papillary thyroid carcinoma cell lines TPC-1 and BCPAP, which were instrumental for this research. Similarly, the authors are grateful to Professor Massimo Santoro for kindly providing the normal thyroid follicular epithelial cell line Nthy-ori 3-1. Finally, the technical assistance and resources provided by the FACS and Integrated Microscopy Facilities of the IGB-CNR are greatly appreciated for their crucial support throughout the study.