Mammalian SWI/SNF (SWitch/Sucrose Non-Fermentable) complexes are ATP-dependent chromatin remodelers whose subunits have emerged among the most frequently mutated genes in cancer. Studying SWI/SNF function in cancer cell line models has unveiled vulnerabilities in SWI/SNF-mutant tumors that can lead to the discovery of new therapeutic drugs. However, choosing an appropriate cancer cell line model for SWI/SNF functional studies can be challenging because SWI/SNF subunits are frequently altered in cancer by various mechanisms, including genetic alterations and post-transcriptional mechanisms. In this work, we combined genomic, transcriptomic, and proteomic approaches to study the mutational status and the expression levels of the SWI/SNF subunits in a panel of 38 lung adenocarcinoma (LUAD) cell lines. We found that the SWI/SNF complex was mutated in more than 76% of our LUAD cell lines and there was a high variability in the expression of the different SWI/SNF subunits. These results underline the importance of the SWI/SNF complex as a tumor suppressor in LUAD and the difficulties in defining altered and unaltered cell models for the SWI/SNF complex. These findings will assist researchers in choosing the most suitable cellular models for their studies of SWI/SNF to bring all of its potential to the development of novel therapeutic applications.

Long non-coding RNAs (lncRNAs) are a heterogeneous class of non-coding RNAs whose biological roles are still poorly understood. LncRNAs serve as gene expression regulators, frequently interacting with epigenetic factors to shape the outcomes of crucial biological processes, and playing roles in different pathologies including cancer. Over the last years, growing scientific evidence supports the key role of some lncRNAs in tumor development and proposes them as valuable biomarkers for the clinic. In this study, we aimed to characterize lncRNAs whose expression is altered in tumor samples from patients with lung adenocarcinoma (LUAD) compared to adjacent normal tissue samples. On an RT-qPCR survey of 90 cancer-related lncRNAs, we found one lncRNA, DLG2-AS1, which was consistently downregulated in 70 LUAD patients. To gain insight into its biological function, DLG2-AS1 was cloned and successfully re-expressed in LUAD cancer cell lines. We determined that DLG2-AS1 is not a cis-regulatory element of its overlapping gene DLG2, as their transcription levels were not correlated, nor did DLG2-AS1 restoration modify the expression of DLG2 protein. Furthermore, after generating a receiver operating curve (ROC) and calculating the area under curve (AUC), we found that DLG2-AS1 expression showed high sensitivity and specificity (AUC = 0.726) for the classification of LUAD and normal samples, determining its value as a potential lung cancer biomarker.

Pediatric acute B-cell lymphoblastic leukemia (B-ALL) constitutes a heterogeneous and aggressive neoplasia in which new targeted therapies are required. Long non-coding RNAs have recently emerged as promising disease-specific biomarkers for the clinic. Here, we identified pediatric B-ALL-specific lncRNAs and associated mRNAs by comparing the transcriptomic signatures of tumoral and non-tumoral samples. We identified 48 lncRNAs that were differentially expressed between pediatric B-ALL and healthy bone marrow samples. The most relevant lncRNA/mRNA pair was AL133346.1/CCN2 (previously known as RP11-69I8.3/CTGF), whose expression was positively correlated and increased in B-ALL samples. Their differential expression pattern and their strong correlation were validated in external B-ALL datasets (Therapeutically Applicable Research to Generate Effective Treatments, Cancer Cell Line Encyclopedia). Survival curve analysis demonstrated that patients with “high” expression levels of CCN2 had higher overall survival than those with “low” levels (p = 0.042), and this gene might be an independent prognostic biomarker in pediatric B-ALL. These findings provide one of the first detailed descriptions of lncRNA expression profiles in pediatric B-ALL and indicate that these potential biomarkers could help in the classification of leukemia subtypes and that CCN2 expression could predict the survival outcome of pediatric B-cell acute lymphoblastic leukemia patients.