Upregulation of long intergenic noncoding RNA LINC00092 indicates favorable survival in lung adenocarcinoma via tumor growth inhibition

Background Long intergenic noncoding RNA 00092 (LINC00092) is a recently identified novel RNA that acts on cancer-associated fibroblasts (CAFs) to drive the progression of ovarian cancer. Because CAFs also play a vital role in lung cancer, we hypothesized that LINC00092 is associated with lung cancer.Methods The expression level of LINC00092 was examined in 93 cases of non-small cell lung cancer (NSCLC) tumor tissues and adjacent normal lung tissues, and its clinical effect on prognosis was evaluated using the Kaplan-Meier method, log-rank test and Cox regression analysis based on The Cancer Genome Atlas (TCGA) data. The effect of LINC00092 on tumor growth was further assessed in vitro .Results LINC00092 was significantly downregulated in 76.3% (71/93) of lung cancer tissues compared to that in their normal counterparts ( P < 0.001), and high expression of LINC00092 led to a better prognosis with increased survival time (1632 days vs. 1171 days; P = 0.087) and decreased mortality (hazard rate, HR = 0.73, 95%CI = 0.51-1.05) than low expression in lung adenocarcinoma (LUAD) but not in lung squamous cell carcinoma (LSCC) patients. Additionally, upregulation of LINC00092 inhibited LUAD cell proliferation and tumorigenic ability in vitro .Conclusions LINC00092 is an indicator of favorable LUAD prognosis. Targeted molecular therapy directed at LINC00092 upregulation may be a valuable strategy to fight LUAD.


Abstract
Background Long intergenic noncoding RNA 00092 (LINC00092) is a recently identified novel RNA that acts on cancer-associated fibroblasts (CAFs) to drive the progression of ovarian cancer. Because CAFs also play a vital role in lung cancer, we hypothesized that LINC00092 is associated with lung cancer.Methods The expression level of LINC00092 was examined in 93 cases of non-small cell lung cancer (NSCLC) tumor tissues and adjacent normal lung tissues, and its clinical effect on prognosis Background Lung cancer has ranked as the leading cause of cancer-related death for several years. According to data from the American Cancer Society, 158,080 lung cancer deaths were estimated in America in 2016, accounting for 26.5% of all cancer deaths (Siegel, Miller et al., 2016). Data from the National Central Cancer Registry of China also showed that the lung cancer mortality was 610.2/10 5 in 2015, ranking first among all cancer deaths (Chen, Zheng et al., 2016). The 5-year survival rate of lung cancer patients is still low, with values lower than 20%. However, advancements in treatment strategies for lung cancer have improved in lung cancer survival rates ranging from 12% in 1975to 1977, 13% in 1987-1989to 18% in 2004-2010(Siegel, Miller et al., 2015. The effect is small but significant. This increase in the last two decades is mainly due to targeted molecular therapy, such as tyrosine kinase inhibitors targeting the epidermal growth factor receptor used in the treatment of lung cancer, which has resulted in considerable improvements in overall survival, indicating promising prospects in the fight against lung cancer (Chen, Kronenberger et al., 2012;Chee, Robinson et al., 2017). Nevertheless, the downside is that patient responses to targeted anticancer agents are discrepant and variable, which may be a major reason limiting improvement in lung cancer survival.
Thus, testing for predictive biomarkers for targeted therapy is becoming an important part of the diagnostic workup. We still need to exert great efforts to identify effective prognostic biomarkers.
Long intergenic noncoding RNAs (lincRNAs) are defined as intergenic non-protein coding transcripts longer than 200 nucleotides. Thus far, many lincRNAs have been identified as possible oncogenes or tumor suppressors that drive or inhibit tumorigenesis and cancer progression. Further, some lincRNAs, such as LINC00673 , LINC00511 (Sun, Li et al., 2016), LINC00672 (Li, Li et al., 2017), HULC (Panzitt, Tschernatsch et al., 2007) and CASC15 (Russell, Penikis et al., 2015), have been suggested to be possible diagnostic and prognostic biomarkers or therapeutic targets owing to their significant correlation with clinical features in cancer. However, there is still little knowledge on the role of lincRNAs in lung cancer.
LINC00092 (NR_024129.1) is a newly identified lincRNA that is upregulated in ovarian cancer and correlated with poor prognosis in patients (Zhao, Ji et al., 2017).
Mechanistic analyses have shown that LINC00092 can induce cancer-associated fibroblasts (CAFs) to promote metastasis of cancer cells (Zhao, Ji et al., 2017). CAFs also play a definite role in lung cancer development (Micke and Ostman, 2004). We thus hypothesized that LINC00092 is associated with the progression of lung cancer.
In the current study, we evaluated the expression levels of LINC00092 in 132 pairs

LINC00092 overexpression vector construction and cell transfection
The whole transcript of LINC00092 was cloned into the overexpression vector pcDNA3.1. The empty vector pcDNA3.1 was used as a control. The vectors were transiently transfected into two LUAD cell lines, A549 and PC-9. The cells were placed in a CO 2 -containing incubator (SANYO Electric Co., Ltd., Japan) with constant 90% humidity and 5% CO 2 .

Tumor growth assay in vitro
Cell proliferation was tested using the Cell Counting Kit-8 (CCK8) assay according to the manufacturer's protocols. LUAD cells were seeded in a 96-well plate, and cell viability was measured after 24, 48, 72 and 96 h. The clonogenic ability of LUAD cells was assessed via plate colony assays. Cells were plated in a 6-well plate (100 cells/well) and cultured for two weeks. Then, the visible colonies were counted after staining. The clonogenic ability of LUAD cells was also analyzed using soft agar assays. Cell suspensions were mixed with 0.35% soft agar in medium containing 10% FBS and layered in triplicate on 0.75% solidified agar in the same growth medium (3 × 10 3 cells/well). After three weeks of culture, colonies were stained, photographed and counted.

Transwell migration and invasion assays
Cell migration and invasion were respectively tested using Corning Transwell insert chambers and BD BioCoat Matrigel Invasion Chambers in accordance with the manufacturer's protocols. Generally, 2 × 10 4 or 2 × 10 5 cells in serum-free RPMI 1640 medium were cultured in the upper chamber for migration and invasion, respectively, cell medium with 10% FBS was added to the lower compartment. After 24 h (for migration) or 48 h (for invasion), cells in the upper chamber were carefully scraped off using a cotton swab, and cells in the lower surfaces of the membrane were counted.

Statistical analysis
Differences in gene expression between NSCLC tumor tissues and their normal counterparts were tested by paired t tests. Comparisons between groups were performed via Student's t test or one-way ANOVA. Survival curves were constructed using the Kaplan-Meier method. The effect of LINC00092 on survival was evaluated using the log-rank test, while the strength of association between LINC00092 expression and survival was calculated by Cox models adjusted for age, sex, smoking status, histological type and clinical stage. All tests were two-sided and performed using Stata 12.0 software, and a P value less than 0.05 was considered statistically significant.

LINC00092 is downregulated in NSCLC tissues
A total of 98 samples were collected to detect LINC00092 expression, and 93 cases among them were finally included in the study using the detection of expression levels of LINC00092 in both cancerous and normal tissues. Unlike its upregulation in ovarian cancer (Zhao, Ji et al., 2017), we found that LINC00092 was significantly downregulated in 76.3% (71/93) of lung cancer tissues compared to that in their normal counterparts (P < 0.001; Fig. 1a). The expression levels of LINC00092 were significantly higher in lung cancer tissues than in adjacent normal lung tissues (mean ± standard deviation: 0.208 ± 0.327 vs. 0.435 ± 0.255; P < 0.001; Fig. 1b).

LINC00092 may be correlated with metastasis in the total cohort
We assessed the correlation between LINC00092 expression levels in cancer tissues and clinicopathological parameters. As shown in Table1, the results revealed that LINC00092 expression levels were likely to be correlated with the presence or absence of metastasis (M), as a lower level was observed in patients with positive metastasis (0.144 ± 0.272) than in patients with negative metastasis (0.258 ± 0.359). The difference showed a clear tendency to being significant (P = 0.096).
However, no significant or slightly significant (P < 0.1) associations were observed between LINC00092 expression and primary tumor dissociation (T; P = 0.357), extent of lymph node involvement (N; P = 0.191) or clinical stage (P = 0.311). Upregulated LINC00092 inhibits LUAD cell growth showed weaker growth ability in plate colony and soft agar assays (Fig. 2a, b).

Upregulation of LINC00092 does not impair migration and invasion in
A previous study reported a beneficial role of LINC00092 on ovarian cancer cell migration and invasion (Zhao, Ji et al., 2017); thus, using Transwell assays, we investigated whether LINC00092 affects the migration and invasion of LUAD cells.
The assay revealed that upregulation of LINC00092 did not significantly affect either migratory or invasive ability of LUAD cells (Fig. 3).
Gene coexpression network with regard to LINC00092 To evaluate possible coding genes related to LINC00092, we constructed a coexpression network based on TCGA data using Weighted Correlation Network analysis (WGCNA) in R. The analysis revealed that LINC00092 expression displays significant, positive and substantial correlations with 11 coding genes (Fig. 4a), addition, TCGA data indicated that all of the above coding genes were downregulated in LUAD (Fig. 4c).

Discussion
Multiple studies have discovered a limited number of lincRNAs as diagnostic or prognostic biomarkers for several cancers, yet efforts are still needed to identify more lincRNAs with effects on cancer. Here, we demonstrated that LINC00092 suppresses growth of lung tumors and contributes to favorable survival in LUAD patients, which is controversial to its role in ovarian cancer according to a previously published study (Zhao, Ji et al., 2017). We found that LINC00092 is significantly downregulated in NSCLC tumorous tissues compared to that in normal   Gene coexpression network with regard to LINC00092. a. WGCNA showed a series of coding g

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download. Figure S1.tif