Bioinformatics Modeling Research Articles

E2F1-induced autocrine IL-6 inflammatory loop mediates cancer-immune crosstalk that predicts T cell phenotype switching and therapeutic responsiveness.

Melanoma is a metastatic, drug-refractory cancer with the ability to evade immunosurveillance. Cancer immune evasion involves interaction between tumor intrinsic properties and the microenvironment. The transcription factor E2F1 is a key driver of tumor evolution and metastasis. To explore E2F1's role in immune regulation in presence of aggressive melanoma cells, we established a coculture system and utilized transcriptome and cytokine arrays combined with bioinformatics and structural modeling. We identified an E2F1-dependent gene regulatory network with IL6 as a central hub. E2F1-induced IL-6 secretion unleashes an autocrine inflammatory feedback loop driving invasiveness and epithelial-to-mesenchymal transition. IL-6-activated STAT3 physically interacts with E2F1 and cooperatively enhances IL-6 expression by binding to an E2F1-STAT3-responsive promoter element. The E2F1-STAT3/IL-6 axis strongly modulates the immune niche and generates a crosstalk with CD4+ cells resulting in transcriptional changes of immunoregulatory genes in melanoma and immune cells that is indicative of an inflammatory and immunosuppressive environment. Clinical data from TCGA demonstrated that elevated E2F1, STAT3, and IL-6 correlate with infiltration of Th2, while simultaneously blocking Th1 in primary and metastatic melanomas. Strikingly, E2F1 depletion reduces the secretion of typical type-2 cytokines thereby launching a Th2-to-Th1 phenotype shift towards an antitumor immune response. The impact of activated E2F1-STAT3/IL-6 axis on melanoma-immune cell communication and its prognostic/therapeutic value was validated by mathematical modeling. This study addresses important molecular aspects of the tumor-associated microenvironment in modulating immune responses, and will contribute significantly to the improvement of future cancer therapies.

Transcriptome-Optimized Hydrogel Design of a Stem Cell Niche for Enhanced Tendon Regeneration.

Bioactive hydrogels have emerged as promising artificial niches for enhancing stem cell-mediated tendon repair. However, a substantial knowledge gap remains regarding the optimal combination of niche features for targeted cellular responses, which often leads to lengthy development cycles and uncontrolled healing outcomes. To address this critical gap, an innovative, data-driven materiomics strategy is developed. This approach is based on in-house RNA-seq data that integrates bioinformatics and mathematical modeling, which is a significant departure from traditional trial-and-error methods. It aims to provide both mechanistic insights and quantitative assessments and predictions of the tenogenic effects of adipose-derived stem cells induced by systematically modulated features of a tendon-mimetic hydrogel (TenoGel). The knowledge generated has enabled a rational approach for TenoGel design, addressing key considerations, such as tendon extracellular matrix concentration, uniaxial tensile loading, and in vitro pre-conditioning duration. Remarkably, our optimized TenoGel demonstrated robust tenogenesis in vitro and facilitated tendon regeneration while preventing undesired ectopic ossification in a rat tendon injury model. These findings shed light on the importance of tailoring hydrogel features for efficient tendon repair. They also highlight the tremendous potential of the innovative materiomics strategy as a powerful predictive and assessment tool in biomaterial development for regenerative medicine.

A Novel Urine DNA Predictor for Noninvasive Early Diagnosis and Monitoring Minimal Residual Disease of Upper Tract Urothelial Carcinoma.

For early detection and postoperative monitoring of upper tract urothelial carcinoma (UTUC), the traditional detection method was limited to its invasiveness and insufficient sensitivity. We aim to use urine tumour DNA (utDNA) for detecting minimal residual disease (MRD), early diagnosis and perioperative monitoring in UTUC. We previously established a utDNA multidimensional bioinformatic valuation model, named utLIFE, using low-coverage whole-genome sequencing and targeted deep sequencing. This prospective cohort enrolled 93 patients diagnosed with UTUC without metastasis. We collected morning urine samples on the day of surgery and the discharge day after the operation for utLIFE testing. In addition, we also enrolled 80 healthy controls to further validate the specificity of the utLIFE model in the study. The utLIFE of preoperative samples could discriminate UTUC with high specificity (96.25%, 77/80), and high sensitivity (96.77%, 90/93) regardless of stage and grade. The sensitivity of utLIFE was significantly higher than urine cytology (p < 0.001) and fluorescence insitu hybridisation (FISH) (p < 0.001) (N = 19), especially in early-stage and low-grade UTUC. Postoperative utLIFE scores were significantly decreased compared with those of preoperative samples (79 vs. 36, p < 0.001), indicating its association with tumour burden. For special pathology types, utLIFE performed less well in sensitivity and perioperative alteration. In conclusion, we established a bioinformatic utDNA valuation model, utLIFE, which was validated to be a rapid and noninvasive approach with high sensitivity for early detection and MRD monitoring for UTUC.

Syn-COM: A Multi-Level Predictive Synergy Framework for Innovative Drug Combinations.

Drug prediction and treatment using bioinformatics and large-scale modeling have emerged as pivotal research areas. This study proposes a novel multi-level collaboration framework named Syn-COM for feature extraction and data integration of diseases and drugs. The framework aims to explore optimal drug combinations and interactions by integrating molecular virtuality, similarity clustering, overlap area, and network distance. It uniquely combines the characteristics of Chinese herbal medicine with clinical experience and innovatively assesses drug interaction and correlation through a synergy matrix. Gouty arthritis (GA) was used as a case study to validate the framework's reliability, leading to the identification of an effective drug combination for GA treatment, comprising Tamaricis Cacumen (Si = 0.73), Cuscutae Semen (Si = 0.68), Artemisiae Annuae Herba (Si = 0.62), Schizonepetae Herba (Si = 0.73), Gleditsiae Spina (Si = 0.89), Prunellae Spica (Si = 0.75), and Achyranthis Bidentatae Radix (Si = 0.62). The efficacy of the identified drug combination was confirmed through animal experiments and traditional Chinese medicine (TCM) component analysis. Results demonstrated significant reductions in the blood inflammatory factors IL1A, IL6, and uric acid, as well as downregulation of TGFB1, PTGS2, and MMP3 expression (p < 0.05), along with improvements in ankle joint swelling in GA mice. This drug combination notably enhances therapeutic outcomes in GA by targeting key genes, underscoring the potential of integrating traditional medicine with modern bioinformatics for effective disease treatment.

Bioinformatics-driven untargeted metabolomic profiling for clinical screening of methamphetamine abuse.

Amphetamine-type stimulants are very common, and their usage is becoming a very big social problem all over the world. Thousands of addicts encounter several health problems including mental, metabolic, behavioral and neurological disorders. In addition to these, there are several reports about the elevated risk of tendency on committing criminal cases by addicted persons. Hence, methamphetamine addiction is not only an individual health problem but also a social problem. In our study, we aimed to investigate the pathogenesis of chronic usage of methamphetamine via untargeted metabolomics approach. 38 plasma samples were carefully collected and extracted for untargeted metabolomics assay. A liquid-liquid extraction was performed to get as much metabolite as possible from the samples. After the extraction procedure, samples were transferred into vials and they were evaluated via time of flight mass spectrometry instrument. Significantly, altered metabolites were identified by the fold analysis and Welch's test between the groups. 42 different compounds were annotated regarding to data-dependent acquisition method. Pathway analysis were also performed to understand the hazardous effect of methamphetamine on human body. It has been reported that drug exposure may affect several metabolic pathways for amino acids, fats, energy metabolism and vitamins. An alternative bioinformatic model was also developed and validated in order to predict the chronic methamphetamine drug users in any criminal cases. This generated model passes the ROC curve analysis and permutation test and classify the controls and drug users correctly by evaluating the metabolic alterations between the groups.

Mathematical Modeling in Bioinformatics: Application of an Alignment-Free Method Combined with Principal Component Analysis

Mathematical Modeling in Bioinformatics: Application of an Alignment-Free Method Combined with Principal Component Analysis

USP53 Affects the Proliferation and Apoptosis of Breast Cancer Cells by Regulating the Ubiquitination Level of ZMYND11

Breast cancer is the most common female malignancy worldwide. Ubiquitin-specific peptidase 53 (USP53) has been shown to exert cancer-suppressing functions in several solid tumors, but its role and the underlying mechanism in breast cancer has not been clearly elucidated. Therefore, we have carried out a series of detailed studies on this matter at the levels of bioinformatics, clinical tissue, cell function and animal model. We found that USP53 expression was downregulated in breast cancer specimens and was negatively correlated with the clinical stages. Gain- and loss-of-function experiments demonstrated USP53 inhibited proliferation, clonogenesis, cell cycle and xenograft growth, as well as induced apoptosis and mitochondrial damage of breast cancer cells. Co-immunoprecipitation data suggested that USP53 interacted with zinc finger MYND-type containing 11 (ZMYND11), and catalyzed its deubiquitination and stabilization. The 33–50 amino acid Cys-box domain was key for USP53 enzyme activity, but not essential for its binding with ZMYND11. The rescue experiments revealed that the anti-tumor role of USP53 in breast cancer cells was at least partially mediated by ZMYND11. Both USP53 and ZMYND11 were prognostic protective factors for breast cancer. USP53-ZMYND11 axis may be a good potential biomarker or therapeutic target for breast cancer, which can provide novel insights into the diagnosis, treatment and prognosis.

The Effects of Khat Chewing among Djiboutians: Dental Chemical Studies, Gingival Histopathological Analyses and Bioinformatics Approaches.

This study examined the effects of khat chewing on oral gingival conditions by adopting a targeted process which combined physicochemical analyses of the teeth, histopathological examinations of the gums, and bioinformatics modeling. The physicochemical evaluation of teeth in khat consumers compared to non-consumers was carried out using specific analytical techniques; hence, the results of this initial investigation revealed significant erosion of the tooth enamel due to khat chewing, as well as an alteration of the essential chemical composition of the teeth. Additionally, the histopathological analyses complemented preliminary studies by showing severe inflammation of the gums and oral mucosa in khat users. The understanding of these studies was enriched by bioinformatics analysis, where modeling was carried out via computational methods. This analytical phase examined molecular docking mechanisms, including the interaction between cathinone, the main alkaloid of khat, and the protein receptors involved in the protection of gingival tissues against infections. In summary, this multidisciplinary research provided an in-depth view of the oral health issues related to khat chewing, combining experimental studies with bioinformatics perspectives.

Heavy metals and elderly kidney health: A multidimensional study through Enviro-target Mendelian Randomization

Chronic Kidney Disease (CKD), closely linked to environmental factors, poses a significant public health challenge. This study, based on 529 triple-repeated measures from key national environmental pollution area and multiple gene-related public databases, employs various epidemiological and bioinformatics models to assess the impact of combined heavy metal exposure (Chromium [Cr], Cadmium [Cd], and Lead [Pb]) on early renal injury and CKD in the elderly. Introducing the novel Enviro-Target Mendelian Randomization method, our research explores the causal relationship between metals and CKD. The findings indicate a positive correlation between increased levels of metal and renal injury, with combined exposure caused renal damage more significantly than individual exposure. The study reveals that metals primarily influence CKD development through oxidative stress and metal ion resistance pathways, focusing on three related genes (SOD2, MPO, NQO1) and a transcription factor (NFE2L2). Metals were found to regulate oxidative stress levels in the body by increasing the expression of SOD2, MPO, NQO1, and decreasing NFE2L2, leading to CKD onset. Our research establishes a new causal inference framework linking environmental pollutants-pathways-genes-CKD, assessing the impact and mechanisms of metal exposure on CKD. Future studies with more extensive in vitro evidence and larger population are needed to validate.

Novel bioinformatic analyses of somatic cell contamination in sperm samples

The assessment of epigenetic profiles in sperm is sensitive to somatic cell contamination, which can influence methylation signals at gene promoters. This contamination is particularly problematic in the assessment of DNA methylation in samples with low sperm counts, where fractional amounts of somatic cell DNA can lead to significant shifts in measured methylation state. In this study, a new method of detecting possible somatic cell contamination is proposed through two multi-region bioinformatic models: a traditional differential methylation analysis and a machine learning logistic regression model. These models were trained on publicly available sperm (n = 489) and blood (n = 1029) DNA methylation array data and tested on a contamination set, wherein the sperm of four donors with normal sperm counts were run on a 450k methylation array with four permutations each, including pure blood, half blood and half sperm by DNA concentration, half blood and half sperm by cell count, and pure sperm (n = 16). The DMR and logistic regression model classified the contamination testing set with 100% and 94% accuracy, respectively. These new methods of detecting the effects of somatic cell contamination allow for more accurate differentiation between epigenetic profiles that contain a biological somatic-like shift and those that have somatic-like signatures because of contamination.

PdPANA: phagemid display as peptide array for neutralizing antibodies, an engineered in silico vaccine candidate against COVID-19

The COVID-19 pandemic has tested the technical, scientific, and industrial resources of all countries worldwide. Faced with the absence of pharmacological strategies against the disease, an effective plan for vaccinating against SARS-CoV-2 has been essential. Due to the lack of production means and necessary infrastructure, only a few nations could adequately confront this pathogen with a production, storage, and distribution scheme in place. This disease has become endemic in many countries, especially in those that are developing, thus necessitating solutions tailored to their reality. In this paper, we propose an in silico method to guide the design towards a thermally stable, universal, efficient, and safe COVID-19 vaccine candidate against SARS-CoV-2 using bioinformatics, immunoinformatics, and molecular modeling approaches for the selection of antigens with higher immunogenic potential, incorporating them into the surface of the M13 phage. Our work focused on using phagemid display as peptide array for neutralizing antibodies (PdPANA). This alternative approach might be useful during the vaccine development process, since it could bring improvements in terms of cost-effectiveness in production, durability, and ease of distribution of the vaccine under less stringent thermal conditions compared to existing methods. Our results suggest that in the heavily glycosylated region of SARS-CoV-2 Spike protein (aa 344–583), from its inter-glycosylated regions, useful antigenic peptides can be obtained to be used in M13 phagemid display system. PdPANA, our proposed method might be useful to overcome the classic shortcoming posed by the phage-display technique (i.e., the time-consuming task of in vitro screening through great sized libraries with non-useful recombinant proteins) and obtain the most ideal recombinant proteins for vaccine design purposes.

Cloning, Expression, and Bioinformatics Modeling of Human Papillomavirus Type 52 L1/L2 Chimeric Protein in Escherichia coli BL21 (DE3)

Human papillomavirus (HPV) L1 major capsid protein generates a highly immunogenic virus like particles (VLPs), which have been used as the main component of its prophylactic vaccine. However, the neutralizing antibodies against L1 VLPs are mostly type specific and may not be effective to prevent infection from different strains of HPV. On the other hand, HPV L2 minor capsid protein has low antigenic variation, thus can induce cross-neutralization. This study aims to obtain HPV 52 L1/L2 chimeric protein, which is designed based on HPV type 52 as one of the most circulated high-risk types in Indonesia, to develop a broad-spectrum HPV vaccine. Substitution of HPV 52 H4 helix L1 region with an HPV 52 L2 epitope was carried out using overlap extension PCR. HPV 52 L1/L2 chimeric gene was constructed into pET-SUMO expression vector and expressed in Escherichia coli BL21 (DE3). Bioinformatics modeling suggested that L2 epitope was located inside of the loop region in monomer form, and on the contrary, it was located outside of the pentamer surface. Furthermore, B cell and T cell epitopes predictions were conducted using Immune Epitope Database (IEDB) analysis. The B cell epitopes prediction revealed eleven potential epitopes, whereas the T cell epitopes prediction showed seven potential epitopes for each MHC class I and MHC class II. This study showed that HPV 52 L1/L2 chimeric protein has the potential to induce cross-neutralizing antibodies and can be developed as a promising candidate for a new HPV vaccine.

Investigating the pathogenesis of vitreous in postmortem COVID patients via untargeted metabolomics based bioinformatics model

SARS-CoV-2 virus has become a worldwide pandemic causing millions of death. This severe disaster lead to a immense panic and stress all over the world. Several studies were dedicated to understand its mechanism, pathogenesis and spreading characteristics. By this way, scientists try to develop different therapy and diagnose strategies. For these reasons, several metabolomics, proteomics and genomics studies were also carried out to improve knowledge in this newly identified virus. In this study, we are aimed to explain the pathogenesis of SARS-CoV-2 exposure on postmortem COVID (+) patients via untargeted metabolomics analysis. To carry out this study, a Data Independent Acquisition SWATH method is optimized and performed. Vitreous samples were analyzed in both MS1 and MS2 ESI(+) mode. An orthogonal Partial Least Square Discriminant Analysis were performed for classification. It was observed that lipid metabolism, several amino acids and oxidative stress biomarkers were strongly affected due to high inflammation and possible cytokine storm.

Deep neural network model for enhancing disease prediction using auto encoder based broad learning

Bioinformatics and Healthcare Integration Disease prediction models have been revolutionized by Big Data. These models, which make use of extensive medical data, predict illnesses before symptoms appear. Deep neural networks are well-known for their ability to increase accuracy by extending the network's depth and modifying weights through gradient descent. Traditional approaches, however, are hindered by issues such as gradient instability and delayed training. As a substitute, the Broad Learning (BL) system is introduced, which avoids gradient descent in favor of quick reconstruction by incremental learning. However, BL has trouble extracting complicated features from medical data, which makes it perform poorly in cases involving complex healthcare. We suggest ABL, which combines the effectiveness of BL with the noise reduction of Denoising Auto Encoder (AE), to address this. Robust feature extraction is an area in which the hybrid model shines, especially in intricate medical environments. Accuracy of up to 98.50 % is achieved by remarkable results from validation using a variety of datasets. The ability of ABL to quickly adapt through incremental learning suggests that it may be used to forecast diseases in complicated healthcare contexts with agility and accuracy.

Integrating bioinformatics and experimental models to investigate the mechanism of the chelidonine-induced mitotic catastrophe via the AKT/FOXO3/FOXM1 axis in breast cancer cells.

Breast cancer (BC) is currently the most frequent and lethal cancer among women, and therefore, identification of novel biomarkers and potential anticancer agents for BC is crucial. Chelidonine is one of the main active ingredients of Chelidonium majus, which has been applied in Chinese medicine prescriptions to treat cancer. This paper aimed to evaluate the ability of chelidonine to trigger mitotic catastrophe in BC cells and to clarify its mechanism through the AKT/FOXO3/FOXM1 pathway. Bioinformatics analysis revealed that forkhead box O3 (FOXO3) was downregulated in different subtypes of BC. Factors such as age, stage, Scarff-Bloom-Richardson (SBR) grade, diverse BC subclasses, and triple-negative status were inversely correlated to FOXO3 levels in BC patients compared with healthy controls. Notably, patients exhibiting higher FOXO3 expression levels demonstrated better overall survival (OS) and relapse-free survival (RFS). Moreover, FOXM1 levels were negatively correlated with both OS and RFS in BC patients. These results revealed that FOXO3 might be considered a predictive biomarker for the prognosis of BC. By utilizing Gene Set Enrichment Analysis (GSEA), we delved into the main Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathways of FOXO3, and the results suggested that FOXO3 was mainly involved in cancer-related pathways and the cell cycle. Thereafter, MTT and flow cytometry (FCM) analysis indicated that chelidonine inhibited BC cell line proliferation and induced M phase arrest. It was found that chelidonine treatment induced MCF-7 cell apoptosis, significantly reduced the expression of survivin and promoted the expression of p53 and caspase-9. Further morphological observation illustrated depolymerization of the actin skeleton and shortening of actin filaments in BC cells, leading to the typical characteristics of mitotic catastrophe, such as abnormal mitosis and multinucleated cells. Western blot analysis demonstrated that chelidonine inhibited the expression of p-AKT to promote the expression of FOXO3 protein and weaken the expression levels of FOXM1 and polo-like kinase 1 (PLK1). Taken together, our present work proved that FOXO3 might be considered a potential therapeutic target for BC. Chelidonine emerges as a promising agent to treat BC by inducing M phase arrest of BC cells and hindering the AKT/FOXO3/FOXM1 axis, thereby inducing mitotic catastrophe in BC.

Mogrol-mediated enhancement of radiotherapy sensitivity in non-small cell lung cancer: a mechanistic study.

This study investigated mogrol's impact on non-small cell lung cancer (NSCLC) radiosensitivity and underlying mechanisms, using various methods including assays, bioinformatics, and xenograft models. CCK-8, clonogenic, flow cytometry, TUNEL, and Western blot assays evaluated mogrol and radiation effects on NSCLC viability and apoptosis. Ubiquitin-specific protease 22 (USP22) expression in NSCLC patient tissues was determined by RT-qPCR and Western blot. A xenograft model validated mogrol's effects on tumor growth. Bioinformatics identified four ubiquitin-specific proteases, including USP22, in NSCLC. Kaplan-Meier analysis confirmed USP22's value in lung cancer survival. Human Protein Atlas (HPA) database analysis indicated higher USP22 expression in lung cancer tissues. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis implicated ERK1/2 in NSCLC progression, and molecular docking showed stability between mogrol and ERK1/2. Further in vivo and in vitro experiments have demonstrated that mogrol enhances the inhibitory effect of radiation on NSCLC cell viability and clonogenic capacity. Cell viability and clonogenic capacity are reduced by >50%, and an increase in cellular apoptosis is observed, with apoptotic levels reaching 10%. USP22 expression was significantly elevated in NSCLC tissues, particularly in radiotherapy-resistant patients. Mogrol downregulated USP22 expression by inhibiting the ERK/CREB pathway, lowering COX2 expression. Mogrol also enhanced radiation's inhibition of tumor growth in mice. Mogrol enhances NSCLC radiosensitivity by downregulating USP22 via the ERK/CREB pathway, leading to reduced COX2 expression.NEW & NOTEWORTHY Mogrol enhances non-small cell lung cancer (NSCLC) cell sensitivity to radiotherapy by downregulating USP22 through the ERK/CREB pathway, reducing COX2 expression. These findings highlight mogrol's potential as an adjunct to improve NSCLC radiotherapy and open avenues for further research and clinical applications.

Data-driven modeling of core gene regulatory network underlying leukemogenesis in IDH mutant AML

Acute myeloid leukemia (AML) is characterized by uncontrolled proliferation of poorly differentiated myeloid cells, with a heterogenous mutational landscape. Mutations in IDH1 and IDH2 are found in 20% of the AML cases. Although much effort has been made to identify genes associated with leukemogenesis, the regulatory mechanism of AML state transition is still not fully understood. To alleviate this issue, here we develop a new computational approach that integrates genomic data from diverse sources, including gene expression and ATAC-seq datasets, curated gene regulatory interaction databases, and mathematical modeling to establish models of context-specific core gene regulatory networks (GRNs) for a mechanistic understanding of tumorigenesis of AML with IDH mutations. The approach adopts a new optimization procedure to identify the top network according to its accuracy in capturing gene expression states and its flexibility to allow sufficient control of state transitions. From GRN modeling, we identify key regulators associated with the function of IDH mutations, such as DNA methyltransferase DNMT1, and network destabilizers, such as E2F1. The constructed core regulatory network and outcomes of in-silico network perturbations are supported by survival data from AML patients. We expect that the combined bioinformatics and systems-biology modeling approach will be generally applicable to elucidate the gene regulation of disease progression.

Developmental and Housekeeping Genes: Two Types of Genetic Organization in the Drosophila Genome.

We developed a procedure for locating genes on Drosophila melanogaster polytene chromosomes and described three types of chromosome structures (gray bands, black bands, and interbands), which differed markedly in morphological and genetic properties. This was reached through the use of our original methods of molecular and genetic analysis, electron microscopy, and bioinformatics data processing. Analysis of the genome-wide distribution of these properties led us to a bioinformatics model of the Drosophila genome organization, in which the genome was divided into two groups of genes. One was constituted by 65, in which the genome was divided into two groups, 62 genes that are expressed in most cell types during life cycle and perform basic cellular functions (the so-called "housekeeping genes"). The other one was made up of 3162 genes that are expressed only at particular stages of development ("developmental genes"). These two groups of genes are so different that we may state that the genome has two types of genetic organization. Different are the timings of their expression, chromatin packaging levels, the composition of activating and deactivating proteins, the sizes of these genes, the lengths of their introns, the organization of the promoter regions of the genes, the locations of origin recognition complexes (ORCs), and DNA replication timings.

Abstract 6814: E2F1-induced autocrine IL-6 inflammatory loop mediates cancer-immune cell crosstalk that regulates metastatic properties in melanoma

Abstract Melanoma is a cancer that is metastatic, drug-refractory and fatal if managed late or inadequately. The transcription factor E2F1 is a key player in melanoma pathology, promoting metastatic traits like EMT, neoangiogenesis, chemoresistance and evasion of immune surveillance. Particularly, high levels of E2F1 trigger tumor progression by developing protein-protein interactions (PPI) with coregulators that enhance its potential to activate a prometastatic gene regulatory network (GRN). We aimed to investigate the extent to which E2F1-expressing melanoma cells influence the immune response. Methods: We combined transcriptomic approaches with bioinformatics and structural modeling to search for target genes involved in E2F1-activated immunomodulatory and prometastatic GRNs in aggressive melanoma. Immunoprecipitation and ChIP were performed to verify both PPI and protein-DNA interactions. Functional analyses, including qRT-PCR, immunoblotting and luciferase assays, were performed to verify expression and regulation of downstream targets. Additionally, we established a melanoma-immune cell coculture system with patient-derived CD4+ and CD8+ T cells. Cytokine arrays and ELISA were used for measurement of cytokine secretion to analyze the E2F1-induced tumor-immuno-crosstalk. Results: We identified an E2F1-dependent GRN with IL-6 as a central factor mediating EMT and immune regulation. Mechanistically, E2F1-induced activation and secretion of IL-6 leads to an autocrine inflammatory forward-feedback loop, which in turn further promotes invasiveness and EMT marker expression in melanoma cells. Interestingly, IL6 activated STAT3 associates with E2F1 in a structure-dependent manner, forming a complex that enhances IL-6 expression by binding to a composite E2F1:STAT3-responsive promoter element. Moreover, we observed that the E2F1-IL6 axis dispenses immunomodulatory effects on surrounding CD4+ and CD8+ T cells, further stimulating IL-6 release, but also leading to secretion of IL-10 into the TME. Although IL-10 has an immunosuppressive function, it also showed anti-metastatic properties in melanoma cells. Thus, in contrast to IL-6, IL-10 decreased cell motility and reduced the expression of Vim, Slug, and Snai1, while simultaneous knockdown of E2F1 increased these effects. The findings are supported by clinical data sets showing that melanoma patients who have high levels of E2F1 and IL-6 and low IL-10 exhibit poor survival. Conclusion: Our results unveil a novel gene regulatory program by which E2F1-associated IL-6 release induces the formation of an E2F1:STAT3 complex, resulting in a forward-feedback loop that promotes cancer invasion and regulates T cells response through intercellular cytokine signaling in the tumor microenvironment. Citation Format: Prabir Dhar, Krishna P. Singh, Shailendra K. Gupta, Dhanya Peringot, Alf Spitschak, Brigitte M. Pützer. E2F1-induced autocrine IL-6 inflammatory loop mediates cancer-immune cell crosstalk that regulates metastatic properties in melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6814.

Abstract 3911: A FGFR-p53 developmental signaling axis drives progression of salivary cancer

Abstract Mucoepidermoid carcinoma (MEC) is the most frequently occurring salivary gland (SG) malignancy. Treatments for high grade (HG) MEC are lacking due to a limited understanding of its diverse histological and clinical features. Consequently, patients with non-resectable HG MEC are often treated with adjuvant radiotherapy and suffer from increased rates of recurrence and metastasis with poor 5-year survival rates (30-50%) as compared to low grade (LG) MEC (76-95%). We previously demonstrated a role for IGF-1 signaling, an important mediator for SG development and branching morphogenesis, in MEC pathogenesis (Musicant et al, 2021). Here, we analyzed transcriptome profiles shared between fetal and adult normal parotid (PAR) tissues with MEC tumors to investigate programs involved in MEC progression. Using this information, we developed the first genetically engineered mouse model (GEMM) of advanced HG MEC via targeted oncogene expression in SG ductal cells. MEC samples from 21 PAR tumors (8 LG/13 HG), 8 age-matched normal, and 3 fetal PAR glands aged between 22-24 weeks of gestation (Saitou et al, 2020) were retrospectively collected from UCSF or UNC Surgical Pathology. RNAseq was performed on FFPE samples using Illumina HiSeq. Differential gene expression, gene set enrichment analyses (GSEA), and alternative splicing analyses were compared between fetal/normal/MEC according to tumor grade and CRTC1 (CREB-regulated transcription coactivator 1)/MAML2 (Mastermind-like 2) (C1/M2) status. Based on these analyses, we developed a GEMM combining C1/M2 expression with TP53 loss. GEMM tumors were subjected to RNAseq and reviewed by two independent pathologists for diagnostic confirmation of HG MEC.GSEA revealed significant overlap between MEC tumors and fetal SG for genes governing proliferation and developmental pathways associated with branching morphogenesis. Interestingly, upregulation of the negatively regulated p53 target FOXM1 along with a mutant-p53 gene signature (Troester et al, 2006) revealed that p53 dysregulation in HG tumors phenocopies that seen in fetal SG. Cre-inducible expression of C1/M2 alone was insufficient to induce MEC-like lesions when targeted to distinct salivary cell types (ductal Krt14-CreER, acinar Mist1-CreER, serous demilune Dcpp1-CreER). However, p53 alteration (Tp53KO) with ductal C1/M2 expression drove MEC formation with histological and molecular features of human HG-MEC. Notably, transcriptome profiles implicate an FGFR-p53 developmental signaling axis in MEC progression. Our autochthonous GEMM recapitulates advanced HG-MEC in vivo providing the field with a powerful platform for identifying pathways for developing targeted therapies. Taken together, the integrated bioinformatics and mouse modeling of this study provide a critical step toward elucidating salivary MEC progression and suggest a role for FGFR targeted therapies in the treatment of non-resectable HG MEC. Citation Format: Julia M. Billington, Adele M. Musicant, Jessica L. Cote, Jennifer Modliszewski, Yi-Hsuan Tsai, Luane J. Landuau, John Powers, Renee Betancourt, Radhika Sekhri, Ricardo J. Padilla, Juan C. Hernandez-Prera, David N. Hayes, Trevor G. Hackman, Omer Gokcumen, Sarah M. Knox, Jimena Guidice, Antonio L. Amelio. A FGFR-p53 developmental signaling axis drives progression of salivary cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3911.