THAP7-AS1 is upregulated in human BC tissues and enhances BC cell metastatic ability
In our previous study, THAP7-AS1 was identified as a novel lncRNA that promotes GC progression by differential expression profiles of lncRNA (GSE72307). To explore the relevance of THAP7-AS1 in human BC, we investigated THAP7-AS1 expression in a cohort of 44 BC tumors and 8 non-tumor breast tissues. The results showed that THAP7-AS1 expression was significantly higher in the tumor tissues than that of in non-tumor breast tissues, and further increased in lymph node metastatic BC compared with those without metastasis (Figures 1A-B). A receiver operating characteristic (ROC) curve was performed to evaluate whether THAP7-AS1 expression could be used to distinguish patients with or without LNM. The area under the curve (AUC) value for THAP7-AS1, which was constructed to distinguish BC cases with and without LNM, reached up to 0.7346 (Figure 1C). Moreover, analyses of The Cancer Genome Atlas (TCGA) database indicated that THAP7-AS1 was expressed at higher levels in various types of human cancer, such as adrenocortical, AML, esophageal, liver, lung, and pancreatic cancer (Supplementary Figures 1A-F), and THAP7-AS1 overexpression was associated with poor prognosis in human cancers, including lung cancer and GC (Supplementary Figures 1G-H), which further suggests that THAP7-AS1 plays an oncogenic role in progression and development of various human cancers. We further examined THAP7-AS1 expression levels in a panel of human BC cell lines. BC cell lines expressed remarkably higher levels of THAP7-AS1 than that of the nontumorous mammary epithelial cell line MCF10A (Figure 1D).
To further study the function of THAP7-AS1, we used Smart Silencer targeting THAP7-AS1 and control. As shown in Figures 1E-H, PcDNA3.1-THAP7-AS1 -mediated overexpression and the Smart Silencer mediated knockdown were used for exogenously guiding the expression of THAP7-AS1 in MDA-MB-231 and MDA-MB-468 cells. Then, Transwell migration and invasion assays showed that THAP7-AS1 overexpression enhanced cell migration and invasion abilities compared with the control in both BC cells (Figures 1I-J). By contrast, knockdown of THAP7-AS1 effectively suppressed cell migration and invasion (Figures 1K-L). Consistent with these findings, less local infiltration and metastatic colonies in the lungs were observed in xenograft tumor models of nude mice using MDA-MB-231 cells that were stably transfected with LV3-sh-THAP7-AS1 (Figures 1M-O). In addition, the number of metastatic nodules decreased in nude mice subjected to tail vein administration of ectopic knockdown of THAP7-AS1 compared with the control group. This difference was further verified following examination of H and E staining of lung sections (Figure 1P). Taken together, these findings suggest that THAP7-AS1 promotes BC metastasis and progression.
THAP7-AS1 activates the EGFR signaling pathway
To investigate the molecular mechanism of THAP7-AS1 in promoting BC metastasis, first, we investigated whether THAP7-AS1 functions in cis, affecting the expression of nearby gene (THAP7; THAP7-AS1 is an antisense transcript of THAP7). However, the overexpression and knockdown of THAP7-AS1 did not affect the expression of THAP7 (Figures 2A-D), indicating that it did not exert its function its function in cis. Then, RNA subcellular fractionation location and FISH assays demonstrated that THAP7-AS1 is mainly localized in the nucleus (Figures 2E-J), thus suggesting that THAP7-AS1 binds to nuclear nucleus molecules or proteins and transcriptionally regulate gene functions. To explore changes in the expression of genes downstream of THAP7-AS1, we established the global effects of THAP7-AS1- silenced MDA-MB-231 and MDA-MB-468 cells and conducted RNA transcriptome sequencing. THAP7-AS1 knockdown, respectively, affected the expression of 1,349 genes (585 downregulated and 764 upregulated) and 725 genes (318 downregulated and 407 upregulated) in MDA-MB-231 cells and MDA-MB-468 cells (Supplementary Figures 2A-B). KEGG pathway analysis and GSEA revealed that THAP7-AS1 regulates many genes that are associated with cancer-related signaling processes, including the ERBB (EGFR) signaling pathway (Supplementary Figures 2C-D and Figure 2K). Remarkably, western blotting assays showed that THAP7-AS1 overexpression significantly upregulated the expression of some major EGFR signaling targets, including EGFR, RRAS, and ELK1 and phosphorylation of BRAF, CRAF, MAPK, and ERK in both BC cells (Figure 2L). However, the expression of BRAF, CRAF, MAPK and ERK did not show a significant difference between the groups. By contrast, THAP7-AS1 depletion reduced the expression of EGFR and major EGFR targets (Figures 2M-N). These data indicate that THAP7-AS1 activates the EGFR signaling pathway in BC. Next, we explored the expression levels of THAP7-AS1 and EGFR in BC tissues by RT-qPCR. We noticed that THAP7-AS1 expression was positively associated with EGFR expression (r =0.2883, p = 0.0382) (Figure 2O). In summary, THAP7-AS1 initiates EGFR expression to activate the EGFR signaling pathway, leading to tumor progression.
THAP7-AS1 directly binds to SWI/SNF Complex
LncRNAs could exert their functions by interacting with RNA binding proteins that regulate the expression of target genes by various mechanisms (15). Therefore, RNA pull-down assay followed by proteomic analysis of the THAP7-AS1-combined protein complex in MDA-MB-468 cells was performed to search for potential THAP7-AS1-related proteins (Figure 3A). SNF2H, SNF2L, and BAF155, three core subunits of the SWI/SNF complex, were selected for further binding confirmation based on the following criteria: (1) molecular weight of approximately 130 kD; (2) peptide score > 300; (3) subcellular localization in the nucleus; and (4) reportedly relates to tumor progression. RNA pull-down and RNA immunoprecipitation was further confirmed the interaction between the three SWI/SNF components and THAP7-AS1 (Figures 3B-H). Moreover, THAP7-AS1 colocalized with SNF2H (Figure 3I), SNF2L (Figure 3J) and BAF155 (Figure 3K) in the nuclei of BC cells. These findings indicate that THAP7-AS1 is correlated with the SWI/SNF complex in the nuclei of BC cells. THAP7-AS1 overexpression and depletion did not affect the expression levels of SNF2H, SNF2L and BAF155 (Figures 2L-N), indicating that THAP7-AS1 is not involved in the regulation of the SWI/SNF complex at the post-translational level. However, overexpression of THAP7-AS1 promoted the combination between SNF2H and SNF2L and BAF155 (Figure 3L). Additionally, through domain mapping, we observed that segment 1 (–) of THAP7-AS1 interacted with SNF2H, segment 2 (–) combined with SNF2L, whereas segment 3 (–) bound to BAF155 (Figure 3M). These findings suggest that THAP7-AS1 directly binds to the SWI/SNF complex and promotes interactions among the three proteins.
THAP7-AS1 triggers EGFR expression through recruitment of the SWI/SNF complex to activate EGFR signaling
There is mounting evidence that the SWI/SNF complex regulates gene expression at the transcriptional level by binding to the promoter region and remodeling chromatin(16). Hence, we thus explored whether the activation of EGFR signaling by THAP7-AS1 is dependent on the initiation of SNF2H/SNF2L/BAF155 on the EGFR promoter. First, we designed and used primers that were located ~2 kb upstream of the transcription start sites (TSSs) of EGFR (sites EGFR 1-10). ChIP assays with antibodies against SNF2H, SNF2L, and BAF155 or control IgG showed that SNF2H, SNF2L, and BAF155 efficiently immunoprecipitated the promoter regions of EGFR (Figures 4A-B), suggesting that the promoter regions of EGFR could be regulated by the SWI/SNF complex. Then, we investigated the effect of THAP7-AS1 on SNF2H, SNF2L, and BAF155 binding with the promoter region of the EGFR gene. We observed that THAP7-AS1 depletion reduced the binding capacity of SNF2H, SNF2L, and BAF155 with the EGFR promoter (Figures 4C-E). Moreover, the interaction of SNF2H, SNF2L, and BAF155 with the EGFR promoter region can be enhanced by THAP7-AS1 upregulation (Figures 4F-H), indicating that THAP7-AS1 enhanced SNF2H, SNF2L, and BAF155 occupancy of the promoter region of the EGFR. To identify which segment of the EGFR promoter is the core binding site for THAP7-AS1, the promoter activities of a 2,012-bp region upstream of the TSS (pGL3-2,000/0) and three deletion constructs, including pGL3-1,512/0, pGL3-1,012/0, and pGL3-512/0, were determined in THAP7-AS1 deleted and upregulated BC cells. We found (approximately) the -1,012 to -512 bp segment of EGFR promoter as a sufficient binding site for THAP7-AS1 (Figures 4I-L). Our findings indicate that THAP7-AS1 recruits the SWI/SNF complex to the EGFR promoter, which then results in its activation. Remarkably, SNF2H and SNF2L were highly expressed in BC tumors compared with non-tumorous breast tissues (Figures 4M-N), indicating that the SWI/SNF complex may participate in the regulation of BC progression. Additionally, we analyzed RNA-Seq data [from The Cancer Genome Atlas (TCGA)] of the SWI/SNF complex of BC. Figure 4O shows that BAF155 was remarkably upregulated in BC tissues from the TCGA data. Additionally, SNF2H, SNF2L, and BAF155 deletion notably impaired the migration and invasion ability in both BC cells (Supplementary Figures 3A-D). Next, we found that SNF2H, SNF2L, and BAF155 knockdown dramatically decreased the expression of EGFR (Figures 4P-Q) and the key downstream molecules including RRAS and ELK1, and phosphorylation of BRAF, CRAF, MAPK and ERK (Supplementary Figure 3E). However, there was no difference in EGFR expression between the SNF2H, SNF2L, and BAF155 upregulation group and the control (Figures 4R-S). THAP7-AS1 overexpression and knockdown significantly respectively promoted and inhibited the expression of EGFR (Figures 4T-W). The above results indicated that the SNF/SWI complex regulates the EGFR upregulation depending on the presence of THAP7-AS1. Taken together, our data indicate that THAP7-AS1 triggers EGFR expression through the recruitment of the SWI/SNF complex to the -1,012 to -512 bp segment of the EGFR promoter, resulting in the activation of EGFR-ELK1 signaling.
P300 and c-MYC synergistically activates THAP7-AS1 transcription
To explore upstream regulatory mechanisms of THAP7-AS1 overexpression in BC, epigenetic modification and detailed promoter analysis were performed. First, we treated four kinds of BC cell lines with 5-azacytidine (5-AZ, demethylating agent), trichostatin A (TSA; histone deacetylase inhibitor) and 3-deazaneplanocin A (Dznep, histone methyltransferase EZH2 inhibitor). The results revealed that THAP7-AS1 levels were significantly increased in TSA-treated BC cells (Figure 5A). As we known, TSA influences the expression of genes through induction of histone hyperacetylation at the promoter region (17). Our findings indicated that histone acetyltransferase (HAT)/histone deacetylase (HDACs)-mediated histone modification may be one of the epigenetic mechanisms that regulate THAP7-AS1 transcription. Human HATs acetylate lysines, such as PCAF, MORF, HAT11, and P300, leading to a more relaxed, open and transcriptionally active chromatin structure (18). The ALGGEN program suggested that there were seven P300-binding sites in the promoter of THAP7-AS1. P300, which possesses intrinsic acetyltransferase activity, belongs to the HAT family and regulates gene transcription via histone acetylation and chromatin remodeling (19). To explore whether P300 virtually plays a role in THAP7-AS1 gene transcription activation, a luciferase activity assay was performed. P300 was able to promote the transcriptional activity of the THAP7-AS1 promoter in BC cells (Figures 5B-C). Furthermore, the THAP7-AS1 level increased in BC cells transfected with pcDNA3.1-P300 plasmid relative to the control cells (Figures 5D-E), and was close to the levels in TSA-treated cells. Moreover, the expression of P300, which was high in BC tumors compared with non-tumorous breast tissues (Supplementary Figure 4A), was positively correlated with THAP7-AS1 levels in BC (Supplementary Figures 4B-C). Next, we used C646, one of the most representative histone acetyltransferase (HAT)-P300 inhibitors, to evaluate the contribution of P300 to THAP7-AS1 expression in BC cells. C646 treatment resulted in a marked reduction in THAP7-AS1 expression in BC cells (Figures 5F-G). In addition, THAP7-AS1 expression decreased after knocking down P300 (Figures 5H-I), close to the level in C646-treated cells. To determine whether P300 interacts with the endogenous THAP7-AS1 promoter, we performed a ChIP assay. Figures 5J-K show that P300 efficiently immunoprecipitated the promoter region of THAP7-AS1 in BC cells. P300 is responsible for the acetylation of H3K27, and elevated H3K27ac levels are a hallmark of active genes (20, 21). Thus, the level of active histone acetylation marker H3K27ac was determined. Our results suggested that P300-induced H3K27ac expression was enhanced in the THAP7-AS1 promoter region and was involved in the transcriptional regulation of THAP7-AS1 (Figures 5L-M).
HATs add acetyl groups to lysine residues in the core histone proteins and are associated with euchromatin, a decondensed chromatin structure that allows transcription to proceed and increases gene expression (18). To elucidate whether transcription factors regulate THAP7-AS1 expression in BC, the promoter activities of a 2,000-bp region upstream of the TSS (pGL3-2000/0) and deletion constructs, including PGL3-509/0, PGL3-210/0, PGL3-155/0, and PGL3-101/0, were investigated. Our findings showed that the promoter activities of PGL3-210/0 resulted in a 52% decrease in luciferase activity compared to that of pGL3-509/0 (Figures 5N-O), suggesting that the region located between -509 and -210 is the basal promoter of THAP7-AS1. Next, the transcription factor-binding site region from -509 and -210 was analyzed using the ALGGEN program. Five putative binding sites, including that of KLF5, ELK1, c-JUN, CREB, and c-MYC, were determined. To explore their roles in regulating THAP7-AS1 transcription, KLF5, ELK1, c-JUN, CREB, and c-MYC, were, respectively, overexpressed in both BC cells that were transiently transfected with the pGL3-509/0 construct. The results revealed that the promoter activities of pGL3-509/0 remarkably increased in the c-MYC and c-JUN overexpression groups (Figures 5P-Q). Additionally, RT-qPCR assay indicated that c-MYC and c-JUN enhanced the expression of THAP7-AS1(Figures 5R-S). The above results suggest that the c-MYC and c-JUN-binding sites are crucial for THAP7-AS1 transcription. To explore the contribution of the three putative c-JUN-binding sites and the one putative c-MYC-binding site to the regulation of the THAP7-AS1 promoter, we introduced four-point mutations, designated as p (-314 to -320) Luc-c-JUN-1, p (-88 to -94) Luc- c-JUN -2, and p (-15 to -21) Luc- c-JUN -3, and p (-120 to -115) Luc-c-MYC. Luciferase assays showed that the four mutation vectors separately led to a decrease in promoter activity compared with the control in the 293T cells (Figure 5T). Additionally, ChIP assays revealed that positive enrichment of the promoter amplicons of THAP7-AS1 in the c-MYC-binding site and three c-JUN binding sites in BC cells, indicating that c-MYC and c-JUN could directly bind to the promoter of THAP7-AS1 and initiate its transcription (Figures 5U-V and Supplementary Figures 4D-E). To investigate whether c-MYC and c-JUN have a similar synergistic action with P300 on THAP7-AS1 transcription, we co-transfected the PGL3-THAP7-AS1 plasmid and the expression plasmids of P300, c-MYC, and c-JUN into 293T cells, and the results indicated that P300 worked synergistically with c-MYC, rather than c-JUN, to increase promoter activity (Figure 5W). Moreover, P300 and c-MYC synergistically enhanced the expression of THAP7-AS1 in both BC cells (Figure 5X and Supplementary Figure 4F). These results indicated that P300 synergistically acted with c-MYC to interact with the THAP7-AS1 promoter and increase THAP7-AS1 transcription.
THAP7-AS1 promotes M2 macrophage polarization
We have demonstrated that THAP7-AS1 activates the EGFR-ELK1 signaling pathway. A previous investigation suggested that ELK-1 could bind to the promoter of IL-4 and IL-10 and activate their expression (22). To explore the role of THAP7-AS1 in IL-4 and IL-10 expression, RT-qPCR and ELISA assays were performed. The mRNA levels of IL-4 and IL-10 was significantly increased in THAP7-AS1-overexpressed BC cells (Figures 6A-B), and decreased in THAP7-AS1- deleted BC cells (Figures 6C-D), which suggests that IL-4 and IL-10 are positively regulated by THAP7-AS1. We also assessed the cytokine levels in the culture supernatants of THAP7-AS1 ectopically overexpressed and depleted BC cells. Significantly increased IL-4 levels (Figures 6E-F), rather than IL-10 levels (Supplementary Figures 4G-H), were observed in THAP7-AS1 overexpression BC cells, while profoundly decreased IL-4 expression was found in THAP7-AS1-depleted cells (Supplementary Figures 4I-J). IL-4 could polarize TAMs to the M2 phenotype, which further enhances cancer cell migration and metastasis (23). To explore the role of THAP7-AS1-activated EGFR-ELK1 signaling in macrophage polarization, the gene expression levels of a typical M0 marker (CD10B), M1 markers (NOS1), and M2 markers (CD206, CD163, TGFB, IL-4, IL-10, and ARG1) were investigated in THP-1 (human myeloid leukemia mononuclear cells) co-cultured with BC cells. Compared to the control group, the THP-1 co-cultured with THAP7-AS1-overexpressed BC cells showed remarkably higher expression of CD206, CD163, TGFB, IL-4, IL-10 and ARG1 and lower expression of NOS1, illustrating a predominant M2 phenotype (Figure 6G-H). By contrast, THP-1 co-cultured with THAP7-AS1-deleted BC cells showed notably decreased CD206, CD163, TGFB, IL-4, IL-10, and ARG1 and increased NOS1 (Figure 6I and Supplementary Figure 4K), indicating a distinct M1 phenotype. Flow cytometry analysis and western blot assay also indicated that the level of CD206 notably increased after THP-1 was co-cultured with THAP7-AS1-overexpressing BC cells (Figures 6J-L). These findings suggest that THAP7-AS1 increased the differentiation of M2 macrophages and decreased the differentiation of M1 macrophages.
M2 TAMs play a positive role in the enhancement of BC progression.
PMA (phorbol myristate acetate) has been shown to induce THP-1 cells differentiating into M2 TAMs (24). As shown in Figures 6M-N, cell morphology and the flow cytometry analysis suggest that PMA could induce the differentiation of THP-1 cells into M2 TAMs. Cell morphology and migration assays were then performed to explore the biological functions of M2 TAMs on BC cells, which were treated with the culture supernatants of TAMs. The results revealed that the BC cells showed epithelial-mesenchymal transformation and increased cell migration abilities when incubated with culture supernatants from TAMs (Figures 6O-P), which were concordant with the findings of previous reports indicating that TAMs play an active role in promoting the tumor cell invasion in human cancers(25). Next, the expression of CD163 (M2 markers) was examined by IHC in tumor xenografts. Figure 1M shows that there were less CD163+ cells in the THAP7-AS1 knockdown group compared with the control group. Additionally, the gene expression of typical M1 markers (NOS1) and M2 markers (CD206, CD163, TGFB, IL-4, IL-10, and ARG1) was assessed. The expression of CD206, CD163, TGFB, IL-4, IL-10, and ARG1 significantly decreased in the THAP7-AS1 knockdown group (Figure 6Q), which coincided with the above results, indicating that tumor invasion abilities decreased in the THAP7-AS1 knockdown group partly due to the inhibition of M2 macrophage polarization. Therefore, these findings suggest that M2 TAMs play a positive role in the enhancement of BC progression.
M2 TAM upregulates THAP7-AS1 expression via IL-10/CEBP-β-dependent P300 expression in BC
We found that TAM not only promotes breast cancer progression, but also enhances THAP7-AS1 expression (Figures 7A-B). To determine whether TAM could mediate P300/c-MYC/c-JUN-induced THAP7-AS1 expression, we detected the levels of P300, c-MYC, and c-JUN in both BC cells with culture supernatants from TAMs. The results revealed that TAM increased the level of P300, but not c-MYC and c-JUN (Figures 7A-B). To confirm the effect of P300 on mediating TAM induced THAP7-AS1 upregulation, both BC cells were transfected with si-P300. The results showed that TAM-induced THAP7-AS1 upregulation was blocked by knocking down P300 (Figures 7C-D), suggesting that the P300 is critical in mediating TAM-induced THAP7-AS1.ChIP assays indicated that TAM directly enhances P300 and H3K27ac-binding abilities onto the THAP7-AS1 promoter (Figures 7E-H). M2 TAMs abundantly secrete cytokines [e.g., epidermal growth factor (EGF), FGF, IL-6, and IL-10], which are associated with tumor progression in many types of tumors, including breast cancer (26-29). Next, we performed Gene Expression Profiling Interactive Analysis (GEPIA) in BC from TCGA based on RNA-seq data, which demonstrate that EGF, FGF2, and IL-10 expression rather than IL-6 is positively correlated with P300 expression (Supplementary Figures 5A-D). To investigate the effect of EGF, FGF2 and IL-10 on P300 expression, we treated both BC cells with recombinant(r) EGF, rFGF2 and rIL-10. We found that rIL-10, rather than rEGF or rFGF2, increased P300 and THAP7-AS1 expression in both BC cells (Figures 7I-J). ELISA demonstrated that IL-10 concentrations in the TAMs supernatants markedly increased (Figure 7K). In addition, P300 knockdown inhibited rIL-10-induced THAP7-AS1 expression (Figures 7L-M), which suggests that rIL-10 increases THAP7-AS1 expression via P300. To investigate whether IL-10 transcriptionally upregulates P300 expression BC, we generated luciferase constructs containing the fragment -2,000 bp to 0 bp upstream of the TSS of the P300 promoter sequences. Luciferase assays showed higher promoter activities of PGL3-2,000/0 compared to that of pGL3-basic (Figures 7N-O and supplementary Figure 5E). Moreover, rIL-10 increased PGL3-2000/0 promoter activity in 293T cells (Figure 7P). Next, ALGGEN program was used to predict the transcription factor-binding site region. Three putative binding sites, including c-MYB, YY1, and CEBP-β-binding sites, were identified. To explore their roles in regulating P300 transcription, luciferase, RT-qPCR, and ChIP assays were performed. The results indicated that CEBP-β directly binds to its promoter and regulates THAP7-AS1 transcription (Figures 7Q-S and supplementary Figure 5F). Additionally, IL-10 activates distinct JAK2-STAT3 pathways to influence nuclear transcriptional events, resulting in the upregulation of transcription factor C/EBPβ (30-32). We also confirmed that CEBP-β expression was upregulated in TAM- and rIL-10-treated BC cells (Figures 7T-U). Additionally, we found that M2 TAM promoted the expression of the key molecules in EGFR signaling (supplementary Figure 5G). Collectively, M2 TAMs induce THAP7-AS1 upregulation and activate EGFR signaling via IL-10/CEBP-β-dependent P300 expression in BC.