HOX transcript antisense RNA (HOTAIR) is an oncogenic long non-coding RNA frequently overexpressed in cancer. HOTAIR can enhance the malignant behavior of tumors by sponging microRNAs with tumor suppressor functions. Vasculogenic mimicry is a hypoxia-activated process in which tumor cells form three-dimensional (3D) channel-like networks, resembling endothelial blood vessels, to obtain nutrients. However, the role of HOTAIR in vasculogenic mimicry and the underlying mechanisms are unknown in human cancers. In the current study, we investigated the relevance of HOTAIR in hypoxia-induced vasculogenic mimicry in metastatic MDA-MB-231 and invasive Hs-578t triple negative breast cancer cells. Analysis of The Cancer Genome Atlas (TCGA) database using cBioPortal confirmed that HOTAIR was upregulated in clinical breast tumors relative to normal mammary tissues. Our quantitative RT-PCR assays showed a significant increase in HOTAIR levels after 48 h hypoxia relative to normoxia in breast cancer cell lines. Remarkably, knockdown of HOTAIR significantly abolished the hypoxia-induced vasculogenic mimicry which was accompanied by a reduction in the number of 3D channel-like networks and branch points. Likewise, HOTAIR silencing leads to reduced cell migration abilities of cancer cells. Bioinformatic analysis predicted that HOTAIR has a potential binding site for tumor suppressor miR-204. Luciferase reporter assays confirmed that HOTAIR is a competitive endogenous sponge of miR-204. Congruently, forced inhibition of HOTAIR in cells resulted in augmented miR-204 levels in breast cancer cells. Further bioinformatic analysis suggested that miR-204 can bind to the 3′ untranslated region of focal adhesion kinase 1 (FAK) transcript involved in cell migration. Western blot and luciferase reporter assays confirmed that FAK is a novel target of miR-204. Finally, silencing of HOTAIR resulted in low levels of cytoplasmic FAK protein and alterations in the organization of cellular cytoskeleton and focal adhesions. In summary, our results showed, for the first time, that HOTAIR mitigates cell migration and vasculogenic mimicry by targeting the miR-204/FAK axis in triple negative breast cancer cells.

Organotypic three-dimensional (3D) cell cultures more accurately mimic the characteristics of solid tumors in vivo in comparison with traditional two-dimensional (2D) monolayer cell models. Currently, studies on the regulation of long non-coding RNAs (lncRNAs) have not been explored in breast cancer cells cultured in 3D microenvironments. In the present research, we studied the expression and potential roles of lncRNAs in estrogen receptor-positive luminal B subtype BT-474 breast cancer cells grown over extracellular matrix proteins-enriched 3D cultures. Global expression profiling using DNA microarrays identifies 290 upregulated and 183 downregulated lncRNAs in 3D cultures relative to 2D condition. Using a co-expression analysis approach of lncRNAs and mRNAs pairs expressed in the same experimental conditions, we identify hundreds of regulatory axes modulating genes involved in cancer hallmarks, such as responses to estrogens, cell proliferation, hypoxia, apical junctions, and resistance to endocrine therapy. In addition, we identified 102 lncRNAs/mRNA correlations in 3D cultures, which were similar to those reported in TCGA datasets obtained from luminal B breast cancer patients. Interestingly, we also found a set of mRNAs transcripts co-expressed with LINC00847 and CTD-2566J3.1 lncRNAs, which were predictors of pathologic complete response and overall survival. Finally, both LINC00847 and CTD -2566J3.1 were co-expressed with essential genes for cancer genetic dependencies, such as FOXA1 y GINS2. Our experimental and predictive findings show that co-expressed lncRNAs/mRNAs pairs exhibit a high degree of similarity with those found in luminal B breast cancer patients, suggesting that they could be adequate pre-clinical tools to identify not only biomarkers related to endocrine therapy response and PCR, but to understand the biological behavior of cancer cells in 3D microenvironments.

The 3D organotypic cultures, which depend on the growth of cells over the extracellular matrix (ECM) used as a scaffold, can better mimic several characteristics of solid cancers that influence tumor biology and the response to drug therapies. Most of our current knowledge on cancer is derived from studies in 2D cultures, which lack the ECM-mediated microenvironment. Moreover, the role of miRNAs that is critical for fine-tuning of gene expression is poorly understood in 3D cultures. The aim of this study was to compare the miRNA expression profiles of breast cancer cells grown in 2D and 3D conditions. On an on-top 3D cell culture model using a basement membrane matrix enriched with laminin, collagen IV, entactin, and heparin-sulfate proteoglycans, the basal B (Hs578T) and luminal (T47D) breast cancer cells formed 3D spheroid-like stellate and rounded mass structures, respectively. Morphological changes in 3D cultures were observed as cell stretching, cell–cell, and cell–ECM interactions associated with a loss of polarity and reorganization on bulk structures. Interestingly, we found prolongations of the cytoplasmic membrane of Hs578T cells similar to tunneled nanotubes contacting between neighboring cells, suggesting the existence of cellular intercommunication processes and the possibility of fusion between spheroids. Expression profiling data revealed that 354 miRNAs were differentially expressed in 3D relative to 2D cultures in Hs578T cells. Downregulated miRNAs may contribute to a positive regulation of genes involved in hypoxia, catabolic processes, and focal adhesion, whereas overexpressed miRNAs modulate genes involved in negative regulation of the cell cycle. Target genes of the top ten modulated miRNAs were selected to construct miRNA/mRNA coregulation networks. Around 502 interactions were identified for downregulated miRNAs, including miR-935/HIF1A and miR-5189-3p/AKT that could contribute to cell migration and the response to hypoxia. Furthermore, the expression levels of miR-935 and its target HIF1A correlated with the expression found in clinical tumors and predicted poor outcomes. On the other hand, 416 interactions were identified for overexpressed miRNAs, including miR-6780b-5p/ANKRD45 and miR-7641/CDK4 that may result in cell proliferation inhibition and cell cycle arrest in quiescent layers of 3D cultures. In conclusion, 3D cultures could represent a suitable model that better resembles the miRNA transcriptional programs operating in tumors, with implications not only in the understanding of basic cancer biology in 3D microenvironments, but also in the identification of novel biomarkers of disease and potential targets for personalized therapies in cancer.

Prostate cancer (PCa) is a prevalent malignancy in men globally. Current diagnostic methods like PSA testing have limitations, leading to overdiagnosis and unnecessary treatment. Castration-resistant prostate cancer (CRPC) emerges in some patients receiving androgen deprivation therapy (ADT). This study explores the potential of circulating microRNA-107 (miR-107) in liquid biopsies as a prognosis tool to differentiate CRPC from non-castration-resistant PCa (NCRPC). We designed a case-control study to evaluate circulating miR-107 in serum as a potential prognosis biomarker. We analyzed miR-107 expression in liquid biopsies and found significantly higher levels (p < 0.005) in CRPC patients, compared to NCRPC. Notably, miR-107 expression was statistically higher in the advanced stage (clinical stage IV), compared to stages I–III. Furthermore, CRPC patients exhibited significantly higher miR-107 levels (p < 0.05), compared to NCRPC. These findings suggest that miR-107 holds promise as a non-invasive diagnostic biomarker for identifying potential CRPC patients.

A growing body of research on the transcriptome and cancer genome has demonstrated that many gynecological tumor-specific gene mutations are located in cis-regulatory elements. Through chromosomal looping, cis-regulatory elements interact which each other to control gene expression by bringing distant regulatory elements, such as enhancers and insulators, into close proximity with promoters. It is well known that chromatin connections may be disrupted in cancer cells, promoting transcriptional dysregulation and the expression of abnormal tumor suppressor genes and oncogenes. In this review, we examine the roles of alterations in 3D chromatin interactions. This includes changes in CTCF protein function, cancer-risk single nucleotide polymorphisms, viral integration, and hormonal response as part of the mechanisms that lead to the acquisition of enhancers or super-enhancers. The translocation of existing enhancers, as well as enhancer loss or acquisition of insulator elements that interact with gene promoters, is also revised. Remarkably, similar processes that modify 3D chromatin contacts in gene promoters may also influence the expression of non-coding RNAs, such as long non-coding RNAs (lncRNAs) and microRNAs (miRNAs), which have emerged as key regulators of gene expression in a variety of cancers, including gynecological malignancies.

Extracellular matrix (ECM) stiffening is a defining biophysical feature of solid tumors that reshape gene regulation through mechanotransduction. Increased collagen crosslinking and stromal remodeling enhance integrin engagement, focal-adhesion signaling and force transmission to the nucleus, where key hubs such as lysyl oxidase (LOX), focal adhesion kinase (FAK) and the Hippo co-activators YAP1 and TAZ (WWTR1) promote proliferation, invasion, stemness and therapy resistance. Here, we synthesize evidence that quantitative changes in matrix stiffness remodel the miRNome and lncRNome in both tumor and stromal compartments, including extracellular vesicle cargo that reprograms metastatic niches. To address heterogeneity in experimental support, we classify mechanosensitive ncRNAs into studies directly validated by stiffness manipulation (e.g., tunable hydrogels/AFM) versus indirect associations based on mechanosensitive signaling, and we summarize physiological versus pathophysiological stiffness ranges across tissues discussed. We further review competing endogenous RNA (ceRNA) networks converging on mechanotransduction nodes and ECM remodeling enzymes, and discuss translational opportunities and challenges, including targeting mechanosensitive ncRNAs, combining ncRNA modulation with anti-stiffening strategies, delivery barriers in dense tumors, and the potential of circulating/exosomal ncRNAs as biomarkers. Overall, integrating ECM mechanics with ncRNA regulatory circuits provides a framework to identify feed-forward loops sustaining aggressive phenotypes in rigid microenvironments and highlights priorities for validation in physiologically relevant models.