Over 40% of global GC cases occur in China, with the majority of these patients being diagnosed when the disease is in a later stage, at which time curative surgical tumor resection is no longer an option [1, 2]. This delayed diagnosis, coupled with the high rates of GC tumor chemoresistance, results in a poor prognosis for the majority of these GC patients. As such, it is vital that novel diagnostic biomarkers and therapeutic targets associated with this cancer type be identified in order to provide a foundation for the development of novel personalized treatment strategies capable of markedly improving patient outcomes.
HOXA-AS3 is a lncRNA that was recently discovered and that was found to play a role in EZH2-dependent H3K27 trimethylation and consequent regulation of the ability of human bone marrow mesenchymal stem cells to undergo differentiation into osteoblasts or adipocytes [24]. In oncogenic contexts, HOXA-AS3 has also been shown to promote the proliferation, survival, and proliferation of glioma cells [8], while knocking down this lncRNA in lung adenocarcinoma was sufficient to impair the proliferative and invasive activity of these cells [10]. HOXA-AS3 has also been identified as a novel target in NSCLC patients owing to its ability to downregulate HOXA3 and to thereby modulate tumor cell resistance to cisplatin treatment [11]. However, no prior studies have explored the role of this lncRNA in GC. Herein, we determined that GC patient tumor tissues and cell lines exhibit HOXA-AS3 upregulation that was correlated with tumor size, H. pylori infection status, T stage, and N stage. Consistent with findings in glioma and HCC patients [8, 9], this elevated HOXA-AS3 expression was an independent predictor of poor GC patient prognosis. When we knocked down this lncRNA in GC cells, we found that this was sufficient to compromise their growth and metastatic potential in vitro and in vivo. Overall, these data suggested that HOXA-AS3 functions to promote GC development and progression, making it a potentially viable biomarker for the evaluation of GC patient disease status and prognosis.
NF-κB signaling is a central mediator of GC progression [25]. HOXA-AS3 is able to colocalize with NF-κB in the promoter region of certain genes, and can directly interact with this transcription factor in order to enhance its activation by regulating IκBα expression and p65 subunit K310 acetylation status [17]. As such, we speculated that HOX-AS3 may drive GC progression via the activation of NF-κB. Our experiments confirmed this hypothesis, as knocking down HOXA-AS3 was sufficient to suppress NF-κB, whereas activating NF-κB reversed the negative impacts of HOXA-AS3 knockdown on GC cell proliferative and migratory activity.
There is clear evidence that HOXA-AS3 can function as a ceRNA in the context of HCC, glioma, and atherosclerosis [12, 13, 21]. In HCC, this lncRNA controls the miR-29c/BMP1 axis and thereby promotes tumor cell proliferation, EMT, and metastasis via activating MEK/ERK signaling [9]. As such, we hypothesized that HOXA-AS3 may similarly control NF-κB signaling activity by functioning as a ceRNA in GC. To that end, we identified putative miRNA targets of HOXA-AS3 in GC via the use of a predictive bioinformatics program, leading to the identification of miR-29c and miR-29a-3p as complementary binding targets. Given that miR‐29a-3p had previously been shown to be expressed at low levels in GC and to be correlated with poor GC patient prognosis [26, 27], we examined this miRNA target in-depth and confirmed that it was able to directly bind to HOXA‐AS3, which negatively regulated its expression and function. We further identified LTβR as a downstream miR‐29a-3p target, such that HOXA‐AS3 was able to indirectly control the expression of this gene by modulating miR‐29a-3p activity. Prior studies have demonstrated that LTβR can induce non-classical NF-κB signaling in the context of H. pylori infection, intestinal metaplasia, and atypical hyperplasia, the exact mechanistic role of this gene in GC remains to be studied in detail[23, 28]. Even so, we were able to demonstrate through the use of luciferase reporter assays that the HOXA-AS3 knockdown-dependent inhibition of NF-κB activation was reversed when GC cells overexpressed LTβR or were transfected with a miR-29a-3p inhibitor, and Western blotting further confirmed that these manipulations were sufficient to reverse the HOXA-AS3 knockdown-dependent inhibition of IKKβ, IκBα phosphorylation, and nuclear P65 expression in GC cells. As such, our data provide clear evidence that the HOXA-AS3/miR-29a-3p/LTβR signaling axis plays a key role in the regulation of GC development owing to its ability to modulate NF-κB signaling.