Background

Gastric cancer (GC) is the third most common cause of cancer-related deaths [1, 2]. Although treatments for GC have been greatly improved, the survival remains poor due to the inability to diagnose this cancer in early stage [3, 4]. The most common route of GC metastasis is lymph node metastasis, followed by peritoneal dissemination metastasis and liver metastasis [5, 6]. Approximately one-third of GC patients are diagnosed at an advanced stage with metastasis, and 4–14% have metastatic disease to the liver [7, 9B), suggesting ILF2/ILF3 complex may play a critical role in modulating dsRNA stability. Interestingly, a recent publication by Mahale and colleagues provided very similar evidence for the role of antisense RNA (IER3-AS1) and sense RNA (IER3 )[35]. Based on RNA-seq analysis, they found that activation of FGF2/FGFR signaling greatly enhanced the mRNA levels of IER3 and IER3-AS1. IER3 and IER3-AS1 formed RNA duplex and regulated their mRNA stability each other by interacting with HNRNPK protein [35].

Fig. 9
figure 9

The working model of the SNAI2-ELF3-AS1 feedback loop. A ELF3-AS1 and ELF3 shared promoters and were transcriptionally repressed by SNAI1/2. B The molecular mechanisms underlying how ELF3-AS1 regulates ELF3 mRNA stability. C The molecular mechanisms of the biological role of ELF3-AS1 in GC. D The working model of SNAI2-ELF3-AS1 feedback loop in driving GC progression

Full size image

The biological effects of transcriptional repression of ELF3-AS1 by SNAI2 remains unknown. SNAI2 is a rapid-turnover protein [36]. Recently, Kang et al. had reported that SNAI2 protein turnover was regulated by the ubiquitin-proteasome system (UPS) [12, 13]. However, in theory, the regulation of SNAI2 protein turnover should be not only at the (post-) translational level, but also at the (post-) transcriptional level. Herein, our finding strongly indicated that the SNAI2-repressed lncRNA ELF3-AS1 played an essential role in maintaining SNAI2 mRNA stability. Knockdown of ELF3-AS1 results in decreased expression levels of miRNAs targeting SNAI2, upregulation of SNAI2 mRNA and protein, and activation of downstream signaling of SNAI2. Additionally, the overall survival analysis based on the TCGA data showed that ELF3-AS1 and SNAI2 possessed opposite prognoses in pan-cancer (Figure S5). These findings highlighted that SNAI2 achieves self-overexpression by transcriptionally repressing ELF3-AS1. Once SNAI2 is overexpressed, it can transcriptionally repress ELF3-AS1 expression, thereby maintaining self-overexpression state in tumor metastasis. On the other hand, the downregulation of lncRNA ELF3-AS1 promoted GC cell proliferation by accelerating the G1/S transition and increasing histone-coding gene expression (Fig 9C).

Conclusions

In summary, a novel double-negative feedback loop between SNAI2 and lncRNA ELF3-AS1 was identified in GC. The SNAI2-ELF3-AS1 feedback loop drives GC metastasis by continuously activating SNAI2 signaling and regulating ELF3 expression at transcriptional and post-transcriptional levels (Fig 9D). In GC, SNAI2 was overexpressed, resulting in decreased expression level of ELF3 and ELF3-AS1. In turn, ELF3-AS1 downregulation further drives tumor progression by continuously activating SNAI2 signaling and promoting cell proliferation, thereby leading to a poor prognosis in GC (Fig 9D).