RNA-seq data and clinical information for 594 LUAD (including 107 paired samples) and 551 LUSC (including 49 paired samples) were downloaded from the TCGA database (https://cancergenome.nih.gov/) (USA) in order to evaluate the characteristics of LINC01572 expression and its prognostic value.

The Lncipedia portal (https://lncipedia.org/) is a public database used to obtain lncRNA sequences and annotations [23]. The LncLocator portal (http://www.csbio.sjtu.edu.cn/bioinf/lncLocator/) (China) is an online tool used to predict the subcellular localization of lncRNA [24]. Both these portals were used in the present study of LINC01572.

The differential expression of LINC01572 between tumor and normal samples in LUAD and LUSC was analyzed by R based on the TCGA data. lnCAR (https://lncar.renlab.org/) (China), an online database for lncRNA analysis, was also used to evaluate LINC01572 expression levels in LUAD and LUSC from the GSE75037 and GSE88862 databases. The associations of LINC01572 expression with different clinical variables were also analyzed by R (3.6.3, R Foundation for Statistical Computing, Vienna, Austria) (R is the language and operating environment for statistical analysis, plotting, and operation, and it is an excellent tool for statistical calculations and statistical mapping).

The pROC package in R was used to perform diagnostic ROC analysis with clinical data obtained from a published study [25]. The results were then visualized using the ggplot2 package (3.3.6, Springer-Verlag New York, 2016).

Pairs of cancerous and adjacent tissue (n = 20) were obtained from patients diagnosed with NSCLC who underwent resection at the Affiliated Nanjing Hospital, Nanjing Medical University, between 2015 and 2018. The tissues were stored at –80 °C until analysis. The use of human tissues was in accordance with guidelines from the Declaration of Helsinki.

Mice were obtained from Hangzhou Ziyuan Laboratory Animal Technology Co., Ltd. The animal ethics committee of Nanjing Medical University approved the use of mice for experiments. A human bronchial epithelial cell line (16HBE) and several tumor cell lines (H1299, A549, SKMES1, H460, and SBC3) were purchased from the Cell Bank, Chinese Academy of Sciences. We confirm that mycoplasma testing has been done for the cell lines used. In our study, STR genotyping was used for cell cross-contamination and characterization. These were used to confirm abnormal expression of LINC01572 and to examine cell phenotypes. All cell lines were cultured at 37 °C in an atmosphere containing 5% CO${}_2$ and 95% air in RPMI-1640 medium (HyClone Cytiva; Logan, UT, USA) supplemented with 1% penicillin/streptomycin and 10% FBS (Cytiva, Logan, UT, USA).

To achieve knockdown of LINC01572 expression in A549 and H1299 cells, two small interfering RNAs (siLINC01572-1, siLINC01572-2) that were specific against four LINC01572 transcript variants (NR_159372.1, NR_126330.2, NR_159371.1, NR_159370.1) were designed and synthesized by GenePharma, together with the negative control (siNC). The cells were seeded at an initial concentration of 1 $\times$ 10${}^5$ cells/well in 6-well plates and grown at 37 °C until they reached 80% confluence. Transfection was then carried out using Lipofectamine® 3000 (Invitrogen; Thermo Fisher Scientific, Inc, Waltham, MA, USA) with either the siRNAs or siNC as recommended by the manufacturer. The knockdown efficiency of the siRNAs was evaluated using RT-qPCR technique.

A stable, low-expression plasmid of LINC01572 was designed with the si-LINC01572 and vector (pGPU6) sequences (shRNA, GenePharma, China), the information of siRNAs was listed in Table 1.

The relative expression of LINC01572 in NSCLC tissues and cells was evaluated by RT-qPCR. Total RNA in the NSCLC tissues and cell lines was extracted using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). cDNA for analysis by RT-qPCR was produced from 2 µg of total RNA using the BestarTM qPCR RT kit (cat. no. 2220; DBI Bioscience, Shanghai, China) as recommended by the manufacturer. qPCR was conducted using BestarTM qPCR MasterMix (cat. no. 2043; DBI Bioscience, China) on an ABI7500 system using the following primers: LINC01572 forward (F), 5${}^{\prime }$-AATTTGTGGCGCTTGGTGTC-3${}^{\prime }$ and reverse (R), 5${}^{\prime }$-CTGAGATCCCTGGTTCCACTG-3${}^{\prime }$; GAPDH F, 5${}^{\prime }$-ACAACTTTGGTATCGTGGAAGG-3${}^{\prime }$ and R, 5${}^{\prime }$-GCCATCACGCCACAGTTTC-3${}^{\prime }$. LINC01572 expression was normalized to that of GAPDH. Pre-denaturation was at 95 °C for 5 minutes, followed by 30 cycles of denaturation at 95 °C, annealing at 60 °C for 30 seconds, and extension at 72 °C for 30 seconds. The reactants were finally kept at 72 °C for 7 minutes.

A CCK-8 kit (CK04; Dojindo, Kumamoto, Japan) was used according to the manufacturer’s instructions. Transfected A549 and H1299 cells were grown in 96-well culture plates and then incubated for 0, 24, 48, and 72 h with 100 µL of CCK-8 solution. The cells were then incubated with CCK-8 solution for an additional 4 hours, after which the Optical Density (OD) at 450 nm was evaluated using a spectrophotometer. Experiments were conducted in triplicate.

Colony formation assays were performed with A549 and H1299 cells to assess the impact of LINC01572 knockdown on NSCLC cell proliferation. Following transfection with shLINC01572-1 or shNC for 48 h, cells were seeded at an initial density of 250 cells/well in 6-well plates and incubated for 2 weeks at 37 °C with 5% CO${}_2$. Colonies were then fixed with 70% methanol for 1 h and stained with Giemsa solution for 20 minutes. The number of visible colonies was counted manually under a microscope, with triplicate wells evaluated for each group.

The effect of LINC01572 knockdown on the cell cycle was evaluated by flow cytometry. After 48 h transfection with shLINC01572-1, shLINC01572-2 or shNC, the cells were trypsinized, centrifuged at 2000 rpm for 10 minutes, and washed three times with ice-cold PBS. They were then fixed with 70% ethanol at –20 °C overnight, and the cell cycle distribution analyzed using a cell cycle kit (BestBio, China) and a Beckman Coulter FACSCalibur flow cytometer (Beckman Coulter, Inc., USA).

Transwell assays were carried out using chambers with 8 µm pores (Corning, NY, USA) and Matrigel matrix. A549 and H1299 cells transfected with shLINC01572-1, shLINC01572-2 or shNC were collected after 48 h growth and resuspended in serum-free culture medium at a concentration of 2 $\times$ 10${}^5$ cells/mL. A 200 µL cell suspension was added to the upper chamber, and 500 µL of 10% FBS-supplemented culture medium to the lower chamber. After culturing for 24 h, migrating cells were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet. Cell numbers were counted under a microscope, and this was repeated three times for each of the shLINC01572-1, shLINC01572-2 and shNC groups.

The potential mechanisms for LINC01572 involvement in the development of LUAD were investigated using the GSEA method. LUAD samples were divided into “high” and “low” groups based on the median expression level of LINC01572. Subsequently, the two groups were compared using GSEA (Broad Institute, Cambridge, Mass. USA) software and the fold-change value was used to rank differences. Enrichment analysis was conducted based on the Hallmark gene set, which was previously identified using biological knowledge. The normalized enrichment score (NES) was used as the primary statistic for assessing gene set enrichment results, while the false discovery rate (FDR) was used to estimate the probability of false positive findings in a given gene set with a particular NES value. Enrichment significance was considered to be present when the $|$NES$|$ value was $>$1 and the FDR was 0.25.

Male BALB/c mice aged 8 weeks were obtained from the Shanghai SLAC Laboratory Animal Company Ltd., Shanghai, China. These were used to evaluate the impact of LINC01572 knockdown on NSCLC progression by establishing in vivo tumor growth assays. The department of Laboratory Animal Science, Fudan University, approved all animal experiments. A549 cells that were stably transfected with either shLINC01572-1 or shNC were suspended in culture medium at a concentration of 5 $\times$ 10${}^6$ cells/mL. A 200 µL cell suspension was then subcutaneously injected into the right flank of nude mice. Tumor growth was monitored on days 2, 4, 6, 8, 10, 12, 14, 16, 18 and 20 following injection. The volume of tumors was measured using the formula: length $\times$ width${}^2$$\times$ 0.5. Euthanasia was carried out by exposing the mice to carbon dioxide. The percentage of chamber volume replaced by carbon dioxide flow was approximately 50% vol/min.

The immune, stromal and ESTIMATE scores are valuable and proven indicators for predicting cancer prognosis [26, 27]. The R software package ESTIMATE (https://bioinformatics.mdanderson.org/estimate/index.html) (Houston, TX, USA) was used to calculate these scores based on TGGA LUAD data. The relationship between LINC01572 and immune, stromal and ESTIMATE scores was explored using the psych R package (version 2.1.6). Pearson correlation analysis was used to assess the associations between these factors.

The GSVA package (1.34.0) (Barcelona, Catalonia, Spain) [28] was used to analyze immune cell infiltration in the tumor microenvironment (TME) of LUAD patients by using expression data from TCGA and immunocyte markers from previous studies [29]. Correlations between LINC01572 and co-stimulator and co-inhibitor molecules were also investigated using markers obtained from previous studies [30]. The ggplot2 package provides visualization of significant results only.

R packages were used to construct and analyze survival curves. Cox regression analysis was performed using the “survival” package, and the resulting data was visualized with the “survminer” package. Appropriate statistical tests including the chi-squared test, Student’s t-test, Wilcoxon test, and Kruskal–Wallis tests were used to assess the significance of differences between groups. Statistical significance was confirmed with a p-value 0.05.

The expression of LINC01572 in NSCLC was investigated using the GSE75037 and GSE88862 datasets. LINC01572 expression was higher in tumor samples than in matched normal tissues in both LUAD (Fig. 1A) and LUSC (Fig. 1D). In addition, the TCGA data showed significant upregulation of LINC01572 in LUAD (Fig. 1B) and LUSC (Fig. 1E). Analysis of LINC01572 expression in the 20 pairs of human tumor-adjacent normal tissue of NSCLC further confirmed the above findings (Fig. 1C).

Fig. 1.

Abnormal expression of LINC01572. (A,D) LINC01572 was overexpressed in tumor samples from GSE75037 (LUAD) and GSE88862 (LUSC), respectively. (B,E) Upregulation of LINC01572 was also observed in LUAD and LUSC from the TCGA database, respectively. (C) LINC01572 expression in tumor and adjacent normal tissue from 20 paired LUAD samples. (F) Prediction of the subcellular localization of LINC01572. ***, p 0.001. LUAD, Lung adenocarcinoma; LUSC, Lung squamous cell carcinoma.

The subcellular localization of LINC01572 was predicted by LncLocator based on the characteristics of the LINC01572 RNA sequence. This showed that LINC01572 was located in the cytoplasm with a score of 0.6353 (Fig. 1F).

The above findings suggest that LINC01572 is likely to promote the initiation and development of NSCLC, although further investigation is needed.

Correlation analyses were performed between the LINC01572 expression level and various clinicopathological features in order to assess the clinical value of LINC01572 in LUAD. These showed clear associations between LINC01572 expression and male gender, smoking, T4 stage, and pathological stage IV (Fig. 2A–F; p 0.05 for each).

Fig. 2.

Correlation analysis between clinical characteristics and LINC01572 expression. Significantly higher LINC01572 expression was detected in males (A), smokers (B), patients with metastasis (C), black/African-American patients (D), T4 stage (E), and pathological stage IV (F). The AUC for diagnostic ROC was calculated (G) and Kaplan-Meier survival analysis was conducted using TCGA LUAD data (H). *, p 0.05; **, p 0.01. LINC, Large intergenic non-coding; AUC, Areas under ROC curves; ROC, Receiver operating characteristic.

Analysis of the diagnostic value of LINC01572 expression in LUAD revealed the area under the curve (AUC) was 0.958 (95% CI: 0.942–0.973) (Fig. 2G), thereby indicating strong potential as a diagnostic biomarker. Using median expression as the cutoff value, survival analysis showed that LUAD patients with higher expression of LINC01572 had significantly shorter overall survival (OS) (HR: 1.381; 95% CI: 1.033–1.8481) (Fig. 2H).

In summary, LINC01572 appears to be a promising and novel biomarker for the diagnosis of LUAD, as well as having prognostic value and being a potential therapeutic target. We next investigated the underlying biological mechanisms of LINC01572 in LUAD.

The results of GSEA showed that G2M checkpoint and DNA repair signaling pathways were significantly activated in patients with high LINC01572 expression (Fig. 3A,B). This finding suggests that enhanced tumor cell proliferation and resistance to DNA damage are the main mechanisms by which LINC01572 promotes tumor development.

Fig. 3.

Analysis of signaling pathway enrichment and of the immune microenvironment. GSEA results showed that G2M checkpoint (A) and DNA repair (B) signaling pathways were significantly activated in samples with high LINC01572 expression. ESTIMATE analysis indicated that LINC01572 expression was negatively correlated with local immune cell infiltration (C–E). Lymphocyte infiltration analysis further indicated that fewer cytotoxic and DC cells were present in samples with high LINC01572 expression (F). ns, no significance. *, p 0.05; **, p 0.01; ***, p 0.001. aDC, antigen Dendritic Cells; DC, Dendritic Cells; NK cells, Natural killer cells; Th 1, helper T lymphocyte1.

The immune response plays an essential role in both tumor progression and elimination. Many different molecules can influence the tumor microenvironment (TME), leading to considerable diversity. Tumor cells can escape or fade depending on whether an immune suppressive TME or an immune-activated TME is dominant.

According to ESTIMATE results, LINC01572 expression was negatively correlated to stromal (Fig. 3C) and immune (Fig. 3D) scores. This result indicates that upregulation of LINC01572 leads to an immune suppressive TME. The relationship between LINC01572 expression and specific immune cell infiltration was further analyzed in the TCGA LUAD cohort using the GSVA method. This showed that high LINC01572 expression was associated with lower levels of infiltrating T cells (p 0.05), CD8+ T cells (p 0.001), cytotoxic T cells (p 0.001) and DC (p 0.001) (Fig. 3E). These results confirm those of the ESTIMATE analysis and suggest that LINC01572 could impair the immune response, leading to tumor progression and shorter OS.

Based on the current analysis, LINC01572 was found to be overexpressed in LUAD, as well as being a prognostic factor. Next, in vitro and in vivo experiments were conducted to further investigate the role of LINC01572 in tumor progression.

The LINC01572 expression level in NSCLC cell lines was first evaluated under baseline conditions. As shown in Fig. 4A, tumor cells have significantly higher expression of LINC01572 compared to 16HBE, which is a benign human bronchial epithelial cell line. Among the tumor cell lines investigated here, H1299 and A549 exhibited relatively high overexpression of LINC01572 and were thus selected for subsequent experiments.

Fig. 4.

Abnormal LINC01572 expression, knockdown efficacy, and colony forming assay results. qPCR analysis of LINC01572 expression in cell lines (A), and LINC01572 knockdown efficacy in A549 and H1299 cells (B–D). 96 h CCK-8 assays (E,F) and 2-week colony formation assays (G–J) were also performed using A549 and H1299 cells. *, p 0.05; **, p 0.01; ***, p 0.001.

Two small interfering RNAs, siLINC01572-1 and siLINC01572-2, were designed to inhibit the expression of LINC01572 in A549 and H1299 cells. The knockdown efficiency was determined using RT-qPCR. LINC01572 expression was found to be significantly downregulated in the intervention groups (Fig. 4B,C). In addition, the two corresponding shRNAs also caused stable and significant reduction of LINC01572 expression (Fig. 4D).

The effect of LINC01572 knockdown on cell proliferation was evaluated using CCK-8 assays. Clear reductions in cell proliferation activity were observed in the LINC01572 knockdown groups compared with the shNC group (Fig. 4E,F). Colony formation assay showed that the number of colonies following transfection with shLINC01572-1 or shLINC01572-2 was significantly less than with shNC for A549 (Fig. 4G,H; p 0.05) and H1299 (Fig. 4I,J; p 0.05) cells.

Flow cytometry was conducted to evaluate the effect of LINC01572 knockdown on the cell cycle. Compared to the shNC group, a significant increase was observed in the number of cells in the G1 phase and a corresponding decrease in the number of cells in the S and G2 phases. Knockdown of LINC01572 resulted in significant changes in the cell cycle distribution in A549 (Fig. 5A,B, p 0.05) and H1299 (Fig. 5C,D, p 0.05) cells.

Fig. 5.

Assessment of cell cycle distribution and cell migration ability. The proportion of cells in G1 phase was increased, and the proportion in G2 and M phases correspondingly decreased following knockdown of LINC01572 in A549 (A,B) and H1299 (C,D) cells. Impaired migration ability was observed in the shLINC01572-1 groups compared to the blank and shNC groups in A549 and H1299 cells (10$\times$) (E). Cellular quantitative changes in migration ability in A549 cells and H1299 cells (F,G). **, p 0.01; ***, p 0.001.

In addition, the transwell assay showed that the migratory ability of A549 and H1299 cells was significantly reduced in the LINC01572 knockdown groups compared to both the blank control and shNC-treated groups (Fig. 5E–G; p 0.05).

To investigate the effects of LINC01572 knockdown on tumor growth in vivo, A549 cells with stable knockdown of LINC01572 were subcutaneously injected into the right flank of nude mice (Fig. 6A). Tumor growth was reduced compared to tumors formed after the injection of control cells (Fig. 6B). The final tumor volume and weight were significantly smaller in the LINC01572 knockdown group compared to the shNC group (Fig. 6C,D; p 0.05). These results confirm that LINC01572 knockdown significantly inhibited the progression of LUAD in vivo.

Fig. 6.

In vivo tumor growth experiments. (A) Representative photographs of subcutaneous tumors formed after injection of A549 cells with LINC01572 knockdown. (B) Tumor growth in mice over time. (C) Harvested transplant tumors from LINC01572 knockdown and control mice groups. (D) Weight of the harvested transplant tumors. *, p 0.05; **, p 0.01.