SLCO4A1-AS1 expression is downregulated in highly metastatic lung cancer cells and NSCLC patients

We identified candidate lncRNAs that may be associated with metastatic capacity by comparing the gene expression profiles of highly metastatic PC9/gef and low metastatic PC9 cells from the GEO database (GSE60189) and identified 79 differentially expressed lncRNAs (Fig. 1A; Additional file 3: Table S3). Among these, 22 lncRNAs were upregulated and 57 were downregulated in highly metastatic PC9/gef cells compared to those in PC9 cells. Nine lncRNAs were selected for RT-qPCR analysis (Fig. 1B). Among them, four upregulated (DLEU2, LINC00467, lnc-LYZL1-15, and lnc-PDPK1-3) and four downregulated lncRNAs (PAX8-AS1, MIR31HG, ZBTB-AS1, and SLCO4A1-AS1) were consistent with microarray findings, with the exception of MALAT1 (Fig. 1B). Notably, SLCO4A1-AS1 exhibit the most significant downregulation in PC9/gef cells. Furthermore, we examined SLCO4A1-AS1 expression levels in NSCLC tissues and normal samples using the Gene Expression Profiling Interactive Analysis (GEPIA) database [28]. SLCO4A1-AS1 transcript levels were significantly lower in both lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) tissues. Additionally, the expression of SLCO4A1-AS1 was consistently downregulated in both breast invasive carcinoma (BRCA) and liver hepatocellular carcinoma (LIHC) tissues (Fig. 1C). Moreover, we extended our analysis to the Taiwan Lung Adenocarcinoma Patient Cohort (dbGaP Study Accession: phs001954.v1.p1), comprising of 90 pairs of lung tumors and adjacent normal tissues [29]. In this cohort, the expression level of SLCO4A1-AS1 was significant lower in tumor tissues than in the corresponding normal tissues (Fig. 1D). Notably, lung cancer patients with tumors with higher SLCO4A1-AS1 expression had significantly improved overall survival and extended time to recurrence based on Kaplan–Meier plotter analysis (Fig. 1E; https://kmplot.com/analysis/; GSE31210) [30]. Thus, the aberrant downregulation of SLCO4A1-AS1 in lung tumors correlated with clinical outcomes and prognosis.

Fig. 1

SLCO4A1-AS1 is significantly downregulated in highly metastatic lung cancer cells and NSCLC patients. A Heat map of lncRNA expression in PC9 and PC9/gef cells from the GEO database (GSE60189). B The differential expression fold change of lncRNAs is listed from the database (left) and validated by RT-qPCR (right). C SLCO4A1-AS1 expression in LUAD, LUSC, BRCA, and LIHC tissues from Gene Expression Profiling Interactive Analysis (GEPIA) database. D RNA-seq data revealed SLCO4A1-AS1 transcript expression in 90 pairs of lung tumors and corresponding adjacent normal tissues in Taiwan Lung Adenocarcinoma Patient Cohort (dbGaP Study Accession: phs001954.v1.p1). The data was analyzed using a paired Wilcoxon test (p = 0.038). E The correlation between SLCO4A1-AS1 expression and overall survival (log rank test, p = 0.033), and first progression (log rank test, p = 0.034) of lung cancer patients from GSE31210 in Kaplan–Meier plotter database website. LUAD lung adenocarcinoma, LUSC lung squamous cell carcinoma, BRCA breast invasive carcinoma, LIHC liver hepatocellular carcinoma

Biological process enrichment analysis of SLCO4A1-AS1

We established H1299-mock and H1299-SLCO4A1-AS1 cells to compare the differentially expressed gene (DEG) profiles of H1299-mock and H1299-SLCO4A1-AS1 cells by performing RNA-seq (Fig. 2A; Additional file 8: Fig. S1A). Using Gene Ontology (GO) enrichment analysis, we categorized 520 DEGs into 15 GO biological processes, including extracellular matrix organization, cell migration and mobility, cell junction organization, and cell differentiation and morphogenesis (Fig. 2B). These results indicated that SLCO4A1-AS1 plays a role in regulating cell adhesion and motility.

Fig. 2

Enrichment analysis for biological process of SLCO4A1-AS1. A Volcano plot comparing differential gene expression in H1299-mock vs. H1299-SLCO4A1-AS1 cells. The cutoff was set as |log2(fold change)| > 1.5 and adjusted p value < 0.05. B Enrichment map of the top 520 SLCO4A1-AS1-related genes. Nodes represent Gene Ontology (GO) terms, node colors and size indicate the enrichment significance and number of GO-associated genes. Edges indicate substantial overlap between GO-related genes

SLCO4A1-AS1 plays a tumor suppressor role in lung cancer cells

We further explored the role of SLCO4A1-AS1 in lung cancer by overexpressing it in lung cancer cell lines (PC9/gef, CL1-5, and H1299) using a plasmid encoding the full-length sequence of the SLCO4A1-AS1 transcript (NR_024470). RT-qPCR confirmed the elevated expression of SLCO4A1-AS1 (Additional file 8: Fig. S1A). We assessed the effects of SLCO4A1-AS1 on various aspects of cancer cell behavior, including proliferation, stemness, sphere formation, drug resistance, invasion, and migration. Sphere-forming assays indicated that SLCO4A1-AS1-overexpressing cells exhibited reduced sphere size and number compared to control cells (Additional file 8: Fig. S1B, C). Increased expression and activity of aldehyde dehydrogenase 1A1 (ALDH1A1) have been reported as robust cancer stem cell markers [31]. SLCO4A1-AS1-overexpression decreased the population of ALDH+ cells and ALDH1A1 mRNA levels compared to control cells (Additional file 8: Fig. S1D, E). Additionally, we used Hoechst 33342 dye and verapamil to characterize the side population, which has been shown to be enriched for cancer stem-like cells. We found a significant decrease in the side population fraction in PC9/gef-SLCO4A1-AS1 cells compared to PC9/gef-mock cells (0.38% vs. 0.20%) and H1299-SLCO4A1-AS1 cells compared to H1299-mock cells (5.80% vs. 4.08%) (Additional file 8: Fig. S1F), which supports a negative correlation between SLCO4A1-AS1 and stem-like characteristics. However, SLCO4A1-AS1-overexpression did not alter cell proliferation or EGFR-TKI/chemotherapeutic agent sensitivity (Additional file 8: Fig. S2A, B). SLCO4A1-AS1 exerted a suppressive effect on cancer stemness but not on EGFR TKI/chemotherapeutic agent sensitivity.

Cell migration and invasion play a significant role in cancer metastasis. Therefore, we evaluated the role of SLCO4A1-AS1 in cell invasion and migration. Overexpression of SLCO4A1-AS1 significantly attenuated cell migration and invasion in the lung cancer cell lines (p < 0.05; Fig. 3A; Additional file 8: Fig. S3A). Furthermore, we investigated whether SLCO4A1-AS1 inhibited cell migration and invasion via cytoskeletal remodeling by evaluating the expression of focal adhesion kinase (FAK) and paxillin, which are present beneath the plasma membrane at adhesion sites. Immunofluorescence staining revealed that the fluorescent signals of phospho-FAK and phospho-paxillin around the periphery of cell membranes were decreased in SLCO4A1-AS1 expressing cells (Fig. 3B; Additional file 8: Fig. S3B). Additionally, SLCO4A1-AS1 decreased F-actin (filopodia) formation (Fig. 3B; Additional file 8: Fig. S3B), suggesting that SLCO4A1-AS1 inhibited cell migration and invasion by remodeling the cytoskeleton.

Fig. 3

SLCO4A1-AS1 suppresses cell motility. A The effect of SLCO4A1-AS1 on migration and invasion of lung cancer cells was performed using migration and invasion assays, respectively. Quantification of migratory and invasive cell numbers are shown (*p < 0.05, **p < 0.01). B Immunofluorescence (IF) staining for p-FAK (red), p-paxillin (red), rhodamine-phalloidin for F-actin (red), and nuclei (DAPI, blue) in the control and SLCO4A1-AS1-overexpressing lung cancer cell lines, indicated by white arrows and scale bar = 20 μm

To prevent unforeseeable changes in cancer cell behavior caused by the random genomic integration of SLCO4A1-AS1, we employed transient transfection of SLCO4A1-AS1 to assess its effect on migration and invasion in PC9/gef and H1299 cells. The results showed that transient SLCO4A1-AS1 expression significantly reduced cell migration and invasion in PC9/gef and H1299 cell lines (Additional file 8: Fig. S4A–D).

Furthermore, we utilized siRNA to silence SLCO4A1-AS to assess its impact on the migration and invasion capabilities of PC9/gef and H1299 cells. We found that knockdown of SLCO4A1-AS1 significantly enhanced the migration and invasion ability of PC9/gef and H1299 cells (Additional file 8: Fig. S5A–D). However, knockdown of SLCO4A1-AS1 did not affect SLCO4A1 mRNA and protein expression (Additional file 8: Fig. S5E, F). Additionally, PC9/gef cell growth was unaffected by SLCO4A1-AS1 knockdown compared with control cells (Additional file 8: Fig. S5G).

Furthermore, an intriguing question arose regarding whether the parental gene of SLCO4A1-AS1, namely SLCO4A1, affects cell migration and invasion. To investigate this, we used the siRNA system to knockdown SLCO4A1 in PC9/gef cells and evaluated its impact on their migration and invasion abilities (Additional file 8: Fig. S6A). Our results indicate that silencing of SLCO4A1 did not affect the migration and invasion abilities, nor did it influence the expression of SLCO4A1-AS1 in SLCO4A1 knockdown PC9/gef cells (Additional file 8: Fig. S6B, C).

We assessed the effect of SLCO4A1-AS1 on tumor metastasis in vivo by injecting SLCO4A1-AS1-overexpressing and control cells (H1299-SLCO4A1-AS1 and H1299-mock) into the tail vein of NOD/SCID mice. After 7 weeks, the mice were sacrificed and the lungs were removed for further examination. We observed that SLCO4A1-AS1-overexpressing cells markedly reduced lung metastatic nodules and pulmonary metastatic lesions (Fig. 4A, B). Histological analyses of the lungs confirmed that overexpression of SLCO4A1-AS1suppressed lung cancer metastasis in vivo (Fig. 4C, D).

Fig. 4

Overexpression of SLCO4A1-AS1 inhibited lung metastasis in NOD/SCID mice. A The gross appearance of pulmonary nodules dissected from NOD/SCID mice. B The number of lung metastatic nodules were calculated (n = 5 for each group; *p < 0.05). C SLCO4A1-AS1 expression from representative lung metastatic nodes. The data of RT-qPCR are presented as mean ± SD (**p < 0.01). D Representative H&E staining images of lung tissues sections and scale = 1 mm

Moreover, we evaluated the distant organ metastatic potential between mock-expressing and SLCO4A1-AS1-overexpressing cancer cells. H1299-mock and H1299-SLCO4A1-AS1 cells were injected into mice via tail vein, with 15 mice in each group. After 8–10 weeks, 14 H1299-mock and 11 H1299-SLCO4A1-AS1 cell-injected mice developed lung nodules. Additionally, two H1299-mock cell-injected mice developed malignant ascites as well as liver and kidney metastasis, whereas no liver and kidney metastasis were observed in the H1299-SLCO4A1-AS1 cell-injected mice (Additional file 4: Table S4; Additional file 8: Fig. S7A). These results provide additional evidence that SLCO4A1-AS1 effectively reduces the metastatic ability of lung cancer cells in vivo.

SLCO4A1-AS1 interacts with TOX4

LncRNAs typically exert their functions through physical interactions with various cellular molecules. We performed a protein pull-down assay using biotinylated SLCO4A1-AS1 on three lung cancer cell lines that overexpressed SLCO4A1-AS1 (PC9/gef-SLCO4A1-AS1, CL1-5-SLCO4A1-AS1, and H1299-SLCO4A1-AS1) to identify associated proteins (Fig. 5A). These proteins were analyzed through SDS-PAGE and silver staining (Fig. 5B). Following band screening, we identified 407 candidate target proteins for SLCO4A1-AS1 across the three cell lines using mass spectrometry data intersection (Additional file 5: Table S5). Subsequently, we assessed the target proteins related to cell mobility and found that TOX4 (also known as MIG7) is associated with the invasion ability in glioma cells [32].

Fig. 5

SLCO4A1-AS1 binds to TOX4. A Overview of the in vitro RNA pull-down assay and identification of lncRNA SLCO4A1-AS1 associated cellular proteins. B Silver staining of biotinylated SLCO4A1-AS1 associated proteins. The SLCO4A1-AS1-specific bands (arrows) were excised and analyzed by mass spectrometry (MS), which were identified as TOX4. C Western blotting for proteins from SLCO4A1-AS1 pull-down assays. D Western blotting of TOX4 in samples pulled down by full-length (fragment A, including Exon1-4) or truncated SLCO4A1-AS1 (fragment B–D). Exon mapping showed that SLCO4A1-AS1 exon 3 and exon 4 interacted with TOX4 and was indispensable. E The PC9 cell lysates were immunoprecipitated with control rabbit IgG or anti-TOX4 antibody, and complexes were analyzed for enrichment of SLCO4A1-AS1 by RT-qPCR. The data are presented as mean ± SD (***p < 0.001). Specific immunoprecipitation of TOX4 was confirmed by western blotting (Inset). F TOX4 enriched SLCO4A1-AS1 was confirmed in SLCO4A1-AS1-expressing H1299 cell lysates. RT-qPCR data are presented as mean ± SD (***p < 0.001; ns, not statistically significant). Specific immunoprecipitation of TOX4 was confirmed by western blotting (Inset). G Western blot quantification of cytosolic and nuclear protein levels of TOX4 in H1299-TOX4 and CL1-0-TOX4 cells. H Immunofluorescence staining showed that TOX4 was localized in the nucleus. Scale bar = 20 μm. I RNA fluorescence in situ hybridization (RNA-FISH) and immunofluorescence staining showed that SLCO4A1-AS1 co-localized with TOX4 in SLCO4A1-AS1-expressing H1299-TOX4 cell line. Scale bar = 20 μm

The interaction was confirmed by subjecting the pull-down extracts to SDS-PAGE and probing with the TOX4 antibody in H1299-SLCO4A1-AS1 and CL1-5-SLCO4A1-AS1 cells (Fig. 5C). SLCO4A1-AS1 did not affect TOX4 protein expression in SLCO4A1-AS1-overexpressing H1299 or A549 cells (Additional file 8: Fig. S8A). To identify the TOX4-interacting region of SLCO4A1-AS1, we constructed serial plasmids encoding different exon deletion fragments of SLCO4A1-AS1 for in vitro transcription (IVT) to obtain the corresponding transcripts (fragments A–E; Fig. 5D, left). These transcripts were biotinylated and used in pull-down assays. We found that transcripts A, B, and C interacted with TOX4, whereas transcripts D and E inhibited the interaction between SLCO4A1-AS1 and TOX4. This suggests that exons 3 and 4 of SLCO4A1-AS1 are pivotal for their interaction with TOX4 (Fig. 5D, right). To confirm these findings, we performed immunoprecipitation of the TOX4 complex from cell lysates using either an isotype control IgG or an anti-TOX4 antibody and detected the SLCO4A1-AS1 transcript using RT-qPCR. SLCO4A1-AS1 was enriched in anti-TOX4 immunoprecipitation compared with anti-IgG immunoprecipitation (Fig. 5E, F). Furthermore, we assessed the cellular localization of the TOX4 protein using H1299-TOX4 and CL1-0-TOX4 cell fraction lysates and immunofluorescence staining of H1299-TOX4 and A549-TOX4 cells. We found that TOX4 was predominantly localized in the nuclei of lung cancer cells (Fig. 5G, H). RNA-FISH and immunofluorescence confirmed the nuclear co-localization of SLCO4A1-AS1 and TOX4 (Fig. 5I). The co-transfection efficiency was confirmed by RT-qPCR and western blotting (Additional file 8: Fig. S8B, C). Collectively, these results indicated a direct association between TOX4 and SLCO4A1-AS1.

SLCO4A1-AS1-overexpression reverted TOX4-induced cancer cell migration and invasion

We investigated whether TOX4 modulates cell migration and invasion by overexpressing TOX4 in H1299 and A549 cells (Fig. 6A) and silencing TOX4 in H1299 and CL1-5 cells (Additional file 8: Fig. S9A). TOX4 knockdown significantly reduced cell migration and invasion in both H1299 and CL1-5 cells (Additional file 8: Fig. S9B). In contrast, TOX4-overexpression augmented lung cancer cell migration and invasion (Fig. 6B). Phospho-FAK, phospho-paxillin, and F-actin levels increased in TOX4-overexpressing H1299 cells (Additional file 8: Fig. S10A). Furthermore, the overexpression of SLCO4A1-AS1 counteracted the TOX4-induced elevation of phospho-FAK, phospho-paxillin, and F-actin levels, thereby affecting the cytoskeletal structure (Additional file 8: Fig. S11A) and mitigating TOX4-induced migration and invasion (Fig. 6C). Moreover, survival analysis using the Kaplan–Meier plotter revealed that patients with lung adenocarcinoma who had higher TOX4-expressing tumors had shorter overall survival and recurrence than patients with lower TOX4-expressing tumors (Fig. 6D).

Fig. 6

TOX4 facilitates lung cancer cell migration and invasion. A Stable overexpression of TOX4 in H1299 and A549 was evaluated by western blotting. B The effect of TOX4 overexpression on H1299 and A549 cells migration and invasion was evaluated. Quantification of migratory and invasive cell numbers are shown (*p < 0.05, **p < 0.01, ***p < 0.001). C SLCO4A1-AS1 overexpression rescued the TOX4-evoked increase in cells migration and invasion. Quantification of migratory and invasive cell numbers are shown (*p < 0.05, **p < 0.01, ***p < 0.001). D The correlation between TOX4 expression and overall survival (log rank test, p < 0.001), and first progression (log rank test, p = 0.0026) of lung cancer patients from Kaplan Meier plotter database website. E The gross appearance of pulmonary nodules dissected from NOD/SCID mice. F Number of metastatic nodules in the lung tissue (n = 5 for each group; *p < 0.05). G Representative H&E staining images of lung tissues sections and scale = 1 mm. H Immunohistochemistry staining to assess TOX4 expression in metastatic lung node. Scale = 1 mm and 250 μm

In the mouse tail vein metastasis model, TOX4-overexpression significantly increased the formation of metastatic nodules in the lungs (Fig. 6E, F). Histological analysis of the lungs confirmed lung metastases in different groups (Fig. 6G). Immunohistochemistry of the resected lungs validated the role of TOX4 in the promotion of lung cancer metastasis in vivo (Fig. 6H). These results suggest that TOX4 is essential for lung cancer cell migration and invasion, and is crucial for the promotion of lung cancer progression.

SLCO4A-AS1 interrupted the binding of TOX4 to NTSR1 promoter and suppressed the expression of NTSR1

We explored the molecular mechanisms governed by SLCO4A1-AS1 and TOX4 by analyzing differential gene expression between two paired cell lines (H1299-mock vs. H1299-SLCO4A1-AS1 and H1299-mock vs. H1299-TOX4) using RNA-seq. We focused on protein-coding genes selected from the RNA-seq data and found that 54 coding genes were upregulated and 353 were downregulated in H1299-TOX4 cells (p < 0.05; fold change > 1.5). Additionally, 72 coding genes were upregulated and 136 were downregulated in H1299-SLCO4A1-AS1 cells (Fig. 7A). Furthermore, eight candidate target proteins were simultaneously regulated in TOX4- and SLCO4A1-AS-overexpressing cells (Fig. 7A, B; Additional file 6: Table S6 and Additional file 7: Table S7). Notably, NTSR1 (neurotensin receptor 1) was the most promising target within the SLCO4A1-AS1/TOX4 axis. NTSR1 was upregulated by 2.5-fold in TOX4-overexpressing cells and downregulated 0.18-fold in SLCO4A1-AS1-overexpressing cells compared to the corresponding control cells (Fig. 7B). Further validation revealed that TOX4-overexpression significantly increased NTSR1 mRNA and protein expression in H1299 and A549 cells (Fig. 7C; Additional file 8: Fig. S12A), whereas NTSR1 expression significantly decreased in SLCO4A1-AS1-expressing cells (Fig. 7D; Additional file 8: Fig. S12B). To clarify whether different transcripts of SLCO4A1-AS1 exhibited differential effects on the downstream NTSR1, we established three different SLCO4A1-AS1 transcript-expressing PC9/gef cells. Among them, the full-length transcript SLCO4A1-AS1:4, encompassing all four exons, exhibited a more pronounced suppression of NTSR1. In contrast, transcriptsSLCO4A1-AS1:1 and 1:2, lacking exon 2 and 3 for TOX4 recruitment, showed diminished regulatory capacity (Additional file 8: Fig. S12C, D). Consistent with the data from cultured cells, NTSR1 was downregulated in SLCO4A1-AS1 overexpressing lung tumors and upregulated in TOX4-overexpressing lung tumor tissues from a mouse tail vein metastasis model (Fig. 7E). Given that TOX4 is a DNA-binding protein, we investigated whether TOX4 is a potential transcriptional regulator of NTSR1. First, we determined whether TOX4 binds to the NTSR1 promoter through ChIP assay in H1299-TOX4 and A549-TOX4 cells. The region spanning from − 10 to + 90 relative to the transcription start site (TTS, + 1) of the NTSR1 promoter exhibited significant enrichment upon TOX4 immunoprecipitation (Fig. 7F). These results indicated that TOX4 binds to the NTSR1 promoter and regulates NTSR1 protein expression. Furthermore, overexpression of SLCO4A1-AS1 interrupted the interaction between TOX4 and the NTSR1 promoter (Fig. 7G), indicating that TOX4 is a transcriptional regulator of NTSR1 and that the binding of TOX4 to the NTSR1 promoter can be blocked by SLCO4A1-AS1.

Fig. 7

Identification of downstream target for the SLCO4A1-AS1/TOX4 axis. A Screening of candidate genes using intersection of two pairs of RNA-seq data (H1299-mock vs. H1299-TOX4 and H1299-mock vs. H1299-SLCO4A1-AS1). The eight protein coding candidate genes were upregulated in TOX4-overexpressing cells and downregulated in SLCO4A1-AS1-overexpressing cells. B The fold change of candidate genes was validated in the TOX4-overexpressing cells and SLCO4A1-AS1-overexpressing H1299 cells by RT-qPCR assay. CNTSR1 mRNA and protein expression was detected in TOX4-expressing H1299 cells by RT-qPCR and western blotting. RT-qPCR data are presented as mean ± SD (***p < 0.001). DNTSR1 mRNA and protein expression was detected in SLCO4A1-AS1-expressing H1299 cells by RT-qPCR and western blotting. RT-qPCR data are presented as mean ± SD (***p < 0.001). E Immunohistochemistry staining was performed to evaluated NTSR1 expression in mouse tail vein lung metastasis model and scale = 50 μm. F Chromatin immunoprecipitation (ChIP) was performed with an antibody against TOX4 or IgG in H1299-TOX4 and A549-TOX4 cells. The enriched NTSR1 promoter was purified and quantified by RT-qPCR using primers targeting the NTSR1 promoter region (inset). The data of RT-qPCR are presented as mean ± SD (***p < 0.001). G ChIP was performed to assess the effect of SLCO4A1-AS1 overexpression on the interaction of TOX4 and NTSR1 promoter. The NTSR1 promoter was quantified by RT-qPCR using primers targeting the NTSR1 promoter region. The data of RT-qPCR are presented as mean ± SD (***p < 0.001)

We also conducted NTSR1 promoter luciferase assays to determine whether TOX4 activated the transcription of NTSR1. We used cell pairs with differential expression levels of TOX4 (H1299-TOX4 and H1299-mock; PC9 and PC9/gef). We observed a higher expression level of endogenous TOX4 and NTSR1 promoter luciferase activity in PC9/gef cells compared to PC9 cells (Additional file 8: Fig. S12E). Furthermore, overexpression of TOX4 in H1299 cells resulted in a significant increase in luciferase activity in TOX4-overexpressing H1299 compared to corresponding control H1299-mock cells (Additional file 8: Fig. S12F). Additionally, overexpression of SLCO4A1-AS1 reversed TOX4-promoted transcriptional activity of the NTSR1 promoter in PC9/gef cells (Additional file 8: Fig. S12G). Therefore, we suggest that TOX4 may activate the transcription of NTSR1.

NTSR1 is essential in SLCO4A1-AS1/TOX4-mediated migration and invasion

We verified the role of NTSR1 in TOX4-mediated cell movement by silencing NTSR1 in H1299-TOX4 and A549-TOX4 cells and examined their migratory and invasive abilities. The loss-of-function assay showed that NTSR1 knockdown significantly reduced the migration and invasion of H1299-TOX4 and A549-TOX4 cells (Fig. 8A, B; Additional file 8: Fig. S12H, I). Conversely, NTSR1-overexpression in H1299 and A549 cells increased phosphorylation of FAK and paxillin, and remodeled the F-actin structure (Fig. 8C and D; Additional file 8: Fig. S12J, K). According to the Kaplan–Meier plotter database, higher NTSR1 expression was associated with shorter overall survival in patients with lung adenocarcinoma (Fig. 8E). The evidence showed that NTSR1 is a novel and convergent downstream target of SLCO4A1-AS1/TOX4 and is involved in regulating SLCO4A1-AS1/TOX4-mediated migration and invasion in lung cancer cells. In summary, SLCO4A1-AS1 exerts its inhibitory influence on NTSR1 expression through its interaction with TOX4, thereby exerting a modulatory control over the migratory and invasive behaviors of lung cancer cells (Fig. 8F).

Fig. 8

SLCO4A1-AS1/TOX4-mediate migration and invasion through NTSR1. A H1299-TOX4 cells were transfected with NTSR1 small interfering RNAs (siNTSR1) or scramble siRNA (siCTL). The effect of siRNAs was evaluated by western blotting. B The effect of NTSR1 knockdown on H1299-TOX4 cell migration and invasion was measured. Quantification of migratory and invasive cell numbers are shown (*p < 0.05). C Stable overexpression of NTSR1 in H1299 was evaluated by western blotting. D IF staining for p-FAK (red), p-paxillin (red), rhodamine-phalloidin for F-actin (red), and nuclei (DAPI, blue) in the control and H1299-NTSR1 cells, indicated by white arrows and scale bar = 20 μm. E The correlation between NTSR1 expression and overall survival of lung adenocarcinoma patients from Kaplan Meier plotter database website (log rank test, p = 0.028). F Hypothetical model for the tumor suppressor roles of SLCO4A1-AS1 in lung cancer