
© Non-coding RNA Investigation. All rights reserved. Non-coding RNA Investig 2021;5:2 | http://dx.doi.org/10.21037/ncri-2020-007 Non-alcoholic steatohepatitis (NASH) is the progressive state of non-alcoholic fatty liver disease (NAFLD). Instead of hepatitis, NAFLD has become the major concern among liver diseases with the change of lifestyle and chronic alcoholism as well as over nutrition (1). There will be 15– 20% of NAFLD patients becomes NASH and nearly 20% of NASH patients then further develop into cirrhosis, which permanently irreversible and liver transplantation is needed. Despite the increasing prevalence, currently, there are limited approved therapeutic drugs available for NASH (2). Recently, Zhao et al. from Sun Yat-sen University provided a novel circular RNA (circRNA) based therapy for NASH (3). Specifically, they showed that mitochondrial steatohepatitis-associated circRNA ATP5B Regulator (SCAR) could inhibit mitochondrial ROS (mROS) output and fibroblast activation via blocking mitochondrial permeability transition pore opening. Notably, PGC1-α, a key transcriptional co-activator modulating mitochondrial biogenesis (4), were predicted as the regulator to the consensus site which is responsible for SCAR transcription in fibroblast. Interestingly, the authors clearly revealed the mechanism between PGC1-α and SCAR and further clarified a CHOP-PGC1-αSCAR axis under the ER Stress induced by lipid overload. Moreover, the authors successfully deliver SCAR to mitochondrial and overexpression of circRNA could attenuate the high fat diet-induced cirrhosis and insulin resistance in vivo. To be mentioned, the function and mechanism of circRNA are still under the center of the storm of controversy. It is no doubt that this study showed the new evidence that circRNAs could serve as the putative therapeutic target in liver disease. Compared with the viral based gene therapy, there is more and more fundamental evidence highlight that RNA molecular could be the future for next generation gene therapy (5). Viral approaches including lentivirus and AAV may lead to genome integration, insertional mutagenesis and carcinogenesis. Thus, non-genome integration platforms such as in vitro transcribed mRNA or circRNA, modified RNA and even nanoparticle mediated overexpression plasmid delivery could enhance the safety concern. In addition, compare to linear RNA, special structure of circRNA results the high RNase resistance and longer expression lifespan (6), suggesting the high application value. Multiple mechanisms have been investigated for circRNA’s effect in NASH (7,8). One of the approved mechanism is the circRNA-miRNA-mRNA axes (9,10). Guo et al. reported that both circRNA_046366 and circRNA_046367 could enhance the expression of miR-34a and indirectly remove the inhibitor of transcribed peroxisome proliferatoractivated receptor α mRNA abundance (11,12). A similar mechanism didn’t apply for circRNA SCAR when Zhao et al. tries to find out its sponged microRNAs. The SCAR also did not have any effect on its host gene’s transcripts. Instead, they find the circRNA-protein directly binding pathway. The double-stranded stem-loop structure of Editorial Commentary
© Non-coding RNA Investigation. All rights reserved. Non-coding RNA Investig 2020;4:6 | http://dx.doi.org/10.21037/ncri.2020.03.04 At present, lung cancer remains the first cause of cancerrelated death worldwide, and non-small cell lung cancer (NSCLC) accounts for 80% of cases. Despite advances in the treatment of NSCLC, patient prognosis remains dismal, with overall survival rates ranging from 55% in local disease to 4% in metastatic disease. This scenario, combined with the absence of proper screening tools, highlights the urgency of developing new diagnostic biomarkers that can identify lung cancer patients at early-stage where surgery with curative intent is the preferred treatment. Although early diagnosis is related to increased survival, in the first five years after surgery, up to 40% of resected patients relapse (1). Therefore, it also urges the development of new biomarkers to identify lung cancer patients with a high risk of relapse after surgery in order to improve their clinical management and survival. In this line, the study of long non-coding RNAs (lncRNAs) has emerged as one of the most promising fields in the development of new biomarkers for all cancer types, including NSCLC (2-4). LncRNAs display an extremely specific expression pattern, and their capacity to discriminate between tumor and normal tissue is superior to coding genes or pseudogenes, making them an ideal source of biomarkers (5). The recently published manuscript by Acha-Sagredo et al. reports their study of deregulated lncRNAs associated with NSCLC pathogenesis (6) and provides some additional information to the state of the art (2-4). This was twophase study (Figure 1). An initial discovery data set based on microarray analysis of 44 paired normal and NSCLC tissue was used for the identification of dysregulated lncRNAs in NSCLC. This discovery phase was followed by qPCR validation in 38 additional specimens, which was cross-validated using available TCGA data [n=850: lung adenocarcinoma (LUAD) =585; squamous cell carcinoma (LUSC) =504]. During the discovery phase, a total of 1122 downregulated and 771 upregulated lncRNAs [including long intergenic non-coding RNAs (lincRNAs) and antisense transcripts] were identified. The validation phase confirmed the upregulation of six lncRNAs and the downregulation of six other lncRNAs in tumor tissue compared to normal tissue. Some of the upregulated lncRNAs identified have previously been described as oncogenes in NSCLC. For example, LINC00673 and PCAT6 have been linked to several key processes in cancer progression: LINC00673 to epithelial-to-mesenchymal transition (EMT) and cellular senescence (7,8); and PCAT6 to the recruiting of chromatinmodifying complexes (9). Consistent with these results, some of the downregulated lncRNAs identified in the study by Acha-Sagredo et al. (6) have previously been described by Salavaty et al. using TCGA LUAD data (10). For instance, they explored and confirmed the role of the lncRNAs LANCL1-AS1, FENDRR, LINC00968 and ADAMTS9-AS2 as tumor suppressors in NSCLC adenocarcinoma. Salavaty et al. analyzed the TCGA data using two different webEditorial Commentary
© Non-coding RNA Investigation. All rights reserved. Non-coding RNA Investig 2020 | http://dx.doi.org/10.21037/ncri.2020.03.05 Breast cancer is a complex disease. In addition to lifestyle and environmental factors, genetic factors play a pivotal role in predisposition to breast cancer. Germline mutations in breast cancer predisposition genes BRCA1, BRCA2, PALB2, CHEK2 and ATM confer moderate to high (i.e., >2fold) risks to develop breast cancer and cluster within breast cancer families. The proteins encoded by these five genes are all involved in the repair of DNA double-strand breaks via homologous recombination repair (1,2). In addition to these five breast cancer risk genes, more than 170 breast cancer risk variants have been identified that confer a low (i.e., <1.3-fold) risk to develop breast cancer (3-7). Although the breast cancer risk a single variant confers is very low, these variants are common in the population and act multiplicatively. Therefore, when combined in a polygenic risk score (PRS), individuals in the top centile of this PRS have a moderate lifetime overall breast cancer risk of 32.6% (8). This risk is higher than the lifetime risk conferred by, for example, the c.1100delC mutation in the moderate-risk gene CHEK2 (9). The 172 low-risk variants were identified through genome-wide association studies (GWASs) in which tens of thousands of breast cancer cases and controls are genotyped for more than half a million SNPs in the human genome. Subsequent imputation of the acquired genotyping data using a human genome reference panel even allows the assessment of more than 10 million SNPs (5). The majority of GWAS-identified variants, however, are not the causal variants conferring the breast cancer risk. Usually, they are tags for the causal variants that lie in close linkage disequilibrium (LD) with the GWAS-identified variants. Therefore, GWASs need to be followed up by fine-mapping of the region and in silico and experimental functional studies to identify the breast cancer risk-causing variants (10,11). Identifying the causal variants is important for two reasons. First, the causal variant is likely to confer a higher breast cancer risk than the GWAS-identified variant. Therefore, including all casual variants and their corresponding risk in the PRS will lead to more accurate breast cancer risk stratification of individuals in the population. Moreover, the PRS also modifies the breast cancer risk of individuals from breast cancer families either carrying a BRCA1, BRCA2 or CHEK2 germline mutation or not (12,13). For a 161-SNP PRS this caused, for example, a change in screening recommendation for 11.5% to 19.8% of the familial non-BRCA1/2 mutation carriers, depending on the breast cancer screening guidelines used (14). Second, identification of the causal variants and the normal cellular functions they disturb will help the research community to understand the biological mechanisms driving breast cancer predisposition. This knowledge could eventually be used to develop novel preventive and therapeutic strategies for breast cancer. Editorial Commentary