Long Non-coding RNA LINC01787 Drives Breast Cancer Progression via Disrupting miR-125b Generation

Breast cancer is still the most common and leading cause of cancer-related deaths in women worldwide. Long noncoding RNAs (lncRNAs) and microRNAs (miRNAs) have shown key regulator roles in various cancers. Previous reports have identified miR-125b as a critical tumor suppressor in breast cancer. However, the role of lncRNAs in breast cancer is far from well-characterized. In this study, we identified a novel lncRNA LINC01787, which specifically binds pre-miR-125b, inhibits the binding between DICER and pre-miR-125b, represses the processing of pre-miR-125b by DICER, and therefore induces pre-miR-125b accumulation and represses mature miR-125b generation. Functional assays showed that LINC01787 promotes breast cancer cell proliferation and migration and breast cancer xenograft growth in vivo, which is abolished by the mutation of pre-miR-125b binding sites on LINC01787 or overexpression of miR-125b. Furthermore, LINC01787 is up-regulated in breast cancer tissues and is associated with advanced stages and poor survival. The expression of LINC01787 is inversely associated with that of miR-125b in breast cancer tissues. In conclusion, our findings identified a novel up-regulated and oncogenic lncRNA LINC01787 in breast cancer, which binds pre-miR-125b and represses mature miR-125b generation. Our data suggests LINC01787 as a potential prognostic biomarker and therapeutic target for breast cancer.


INTRODUCTION
According to global cancer statistics in 2018, breast cancer remains the most commonly diagnosed cancer and the leading cause of cancer death in women worldwide (1). Breast cancer accounts for 24.2% of the total incidence of cancers in women and 15.0% of the total deaths of cancer in women, with 2,088,849 estimated new cases and 358,989 estimated deaths in 2018 worldwide (1). Despite advances in surgery, chemotherapy, radiotherapy, endocrine therapy, molecular-targeted therapy, and immunotherapy, breast cancer remains the cause of a vast number of deaths (2)(3)(4)(5). Currently, the outcome of breast cancer patients is still far from satisfactory, which is largely due to an unclear understanding of the molecular mechanisms underlying the initiation and progression of breast cancer (6)(7)(8)(9).
Whole-genome and whole-transcriptome sequencings have found that most human genomes are transcribed, and while only about 2% of human genomes encode proteins, which suggests that most of human transcriptomes are non-coding RNAs (10). Among these great number of non-coding RNAs, microRNA (miRNA) and long non-coding RNA (lncRNA) are two classes of regulatory RNAs, that play important roles in various physiological and pathological processes (11)(12)(13)(14). miRNAs are small non-coding RNAs with 19-25 nucleotides in length (15,16). miRNAs regulate gene expression mainly via binding the 3 ′ -untranslated region (3 ′ -UTR) of target mRNAs, inducing target mRNAs degradation and/or repressing target mRNAs translation, and lastly inhibiting target genes expression (17)(18)(19). Therefore, miRNAs may have oncogenic or tumor-suppressive roles via targeting tumor suppressors or oncogenes (20)(21)(22). In our previous study, we have found that miR-125b exerts tumor suppressive roles in breast cancer via targeting KIAA1522 (23). The tumor suppressive roles of miR-125b in breast cancer were also verified in other reports with several new miR-125b targets being identified, including ETS and SNAI1 (24,25).
lncRNAs are long non-coding RNAs with more than 200 nucleotides that are long and lack an extended open reading frame (26). Whole-transcriptome sequencings have found that the human transcriptome contains more than 58,000 lncRNAs, while the number of protein-coding genes is only about 21,000 (10). Although the expression of lncRNAs is relatively lower than that of protein-coding genes, the expression of lncRNAs are more disease and tissue-specific (27). Aberrant expression of lncRNAs are frequently observed in various cancers, including breast cancer (28). Like miRNAs, lncRNAs play oncogenic or tumor-suppressive roles in cancers (29)(30)(31). lncRNAs may modulate the proliferation, apoptosis, cell cycle, migration, invasion, growth, metastasis, drug-resistance, angiogenesis, and so on of cancer cells (32)(33)(34)(35). Compared with miRNAs, the mechanisms of action of lncRNAs are relatively more complex and varied (36). lncRNAs can modulate protein-coding genes ′ expression at a transcriptional level, post-transcriptional level, translational level, and post-translational modification (37)(38)(39). Furthermore, lncRNAs can modulate miRNAs and further regulate the expression of key miRNAs targets (40). Routinely, lncRNAs can sponge miRNAs and relieve the repressing roles of miRNAs on their targets (41). Furthermore, lncRNAs may regulate the generation of miRNAs via binding pre-miRNAs (42). miRNA-coding genes first transcribe into pri-miRNAs, which are further processed by Drosha to generate pre-miRNAs. Then, the pre-miRNAs are translocated to the cytoplasm and processed by DICER to generate mature miRNAs. We have found that miR-125b is a critical tumor suppressor in breast cancer (23). We further hypothesized that lncRNAs, which can regulate miR-125b, may also have important roles in breast cancer.
First, we predicted the lncRNAs which could specifically bind miR-125b using starBase (http://starbase.sysu.edu.cn/). However, none of these predicted lncRNAs show a significant correlation with the survival of breast cancer patients in The Cancer Genome Atlas (TCGA) data. Next, we searched for lncRNAs that may modulate miR-125b generation. The lncRNAs that could bind pre-miR-125b were searched for using Blast (https:// blast.ncbi.nlm.nih.gov/Blast.cgi). Only lncRNA LINC01787 was predicted to have a strong binding potential with pre-miR-125b. Furthermore, analyzing the TCGA data found that a high expression of LINC01787 is correlated with poor survival of breast cancer patients with a hazard ratio of 1.52. In this study, we further investigated the expression, clinical association, roles, and functional mechanisms of LINC01787 in breast cancer.

Cell Culture
Human breast cancer cell lines MDA-MB-231 and MCF-7 were acquired from the American Type Culture Collection (Manassas, VA, USA) and passaged in our lab. The cells were cultured in DMEM medium (Invitrogen, Thermo Fisher Scientific, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Invitrogen, Thermo Fisher Scientific) and 100 U/ml penicillin and 100 mg/ml streptomycin (Invitrogen, Thermo Fisher Scientific). The cells were incubated in a humidified incubator containing 5% CO 2 at 37 • C.

Isolation of Cytoplasmic and Nuclear RNA
Cytoplasmic and nuclear RNA were extracted from indicated breast cancer cells with the Cytoplasmic & Nuclear RNA Purification Kit (Norgen, Belmont, CA) following the provided protocol. The RNA present in the cytoplasm and nucleus was measured by qRT-PCR as above.

Cell Proliferation Assay
A cell counting kit-8 (CCK-8) assay and a 5-ethynyl-2'deoxyuridine (EdU) incorporation assay were undertaken to analyze cell proliferation. For the CCK-8 assay, indicated breast cancer cells were seeded 3,000 cells per well into 96-well plates and incubated for 0-3 days. At an indicated time, the CCK-8 reagent (Beyotime) was added to the plates and the cells were further incubated for 2 h. The optical density at 450 nm was detected to calculate cell proliferation. EdU incorporation assay was undertaken using the EdU Kit (RiboBio, Guangzhou, China) following the provided protocol. The results were acquired using Zeiss photomicroscope (Carl Zeiss, Oberkochen, Germany) and analyzed using Image-Pro plus 6.0 software.

Cell Migration Assay
A Transwell migration assay was undertaken to analyze cell migration. Indicated breast cancer cells re-suspended in fetal bovine serum free DMEM were seeded to the upper chamber of a transwell insert (Millipore). DMEM with 10% fetal bovine serum was added to the lower chamber. After incubation for 48 h, the cells remaining on the upper chamber were wiped off with a cotton swab. The cells migrated into the lower surface were fixed using 4% paraformaldehyde, stained using 0.1% crystal violet, and photographed using Zeiss photomicroscope.

Animal Experiment
Four-six week old female athymic BALB/c nude mice were acquired from Shanghai Lingchang Biological Technology Ltd (Shanghai, China). 3 × 10 6 indicated breast cancer cells were subcutaneously implanted into the mice. Subcutaneous xenograft volume was measured weekly using a caliper and calculated following the formula V = 0.5 × L × S 2 (L, longest diameter; S, shortest diameter). At the 28th day after implantation, the mice were sacrificed, and subcutaneous xenografts were resected and weighed. The subcutaneous xenografts were fixed in formalin and made into paraffin-embedded sections. The paraffinembedded sections were stained with primary antibodies against PCNA (ab29, 1:6,000, Abcam), KIAA1522 (ab122203, 1:150, Abcam), ETS1 (ab220361, 1:500, Abcam), or SNAI1 (ab53519, 1:1,000, Abcam) following the routine immunohistochemistry (IHC) method. The sections were also used to perform terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay with the in situ Cell Death Detection Kit (Roach) following the provided protocol. The animal experiments were reviewed and approved by the Ethics Committee of Xinxiang Medical University (Xinxiang, China).

Clinical Tissues
Eighty-nine pairs of breast cancer tissues and normal adjacent tissues were obtained from breast cancer patients who received surgery at Xinxiang Medical University (Xinxiang, China). All tissues were diagnosed by pathological examination and stored at −80 • C until use. The use of clinical tissues was reviewed and approved by the Ethics Committee of Xinxiang Medical University (Xinxiang, China). Written informed consents were acquired from all patients.

Statistical Analysis
Statistical analyses were undertaken using GraphPad Prism v6.0 (GraphPad Software, La Jolla, CA, USA). For comparison, oneway ANOVA followed by Dunnett's multiple comparisons test, Kruskal-Wallis test followed by Dunn's multiple comparisons test, Wilcoxon signed rank test, Spearman correlation analysis, Log-rank test, and Pearson chi-square test were performed as indicated. Difference was considered as significant when P < 0.05.

LINC01787 Binds pre-miR-125b and Represses miR-125b Generation
The interaction between LINC01787 and pre-miR-125b-1 and pre-miR-125b-2 was predicted by Blast (https://blast.ncbi. nlm.nih.gov/Blast.cgi) (Figures 1A,B). The nearly completely consistent sequences between LINC01787 and stem regions of both pre-miR-125b implied that LINC01787 may bind another stranded of stem regions of both pre-miR-125b (Figures 1A,B). Moreover, both pre-miR-125b and LINC01787 were mainly located in the cytoplasm in breast cancer cells (Figures S1A,B), which support the potential interaction between LINC01787 and pre-miR-125b. To investigate whether LINC01787 could bind pre-miR-125b, RNA pull-down assays were undertaken using in vitro-transcribed biotin-labeled LINC01787. The results showed that both pre-miR-125b was significantly enriched in the pull-down material acquired by biotin-labeled LINC01787 (Figure 1C, Figure S1C). Mutation of the pre-miR-125b binding sites on LINC01787 abolished the specific enrichment of pre-miR-125b ( Figure 1C, Figure S1C). After transient overexpression of wild-type (wt) or pre-miR-125b binding sites-mutated LINC01787 in MDA-MB-231 cells via transfection of LINC01787-wt or LINC01787-mut overexpression plasmids (Figure S1D), LINC01787 antisense biotinylated probes was used to pulldown LINC01787 and interacted RNAs. The results revealed that both pre-miR-125b-1 and pre-miR-125b-2 were significantly pulled down by LINC01787-wt, which was abolished by the mutation of pre-miR-125b binding sites (Figures 1D,E). Pre-miRNAs are bound and processed by DICER to generate mature miRNAs. Therefore, FIGURE 1 | LINC01787 binds pre-miR-125b and represses the binding between pre-miR-125b and DICER. (A,B) Schematic diagram of the predicted interaction between LINC01787 and pre-miR-125b. (C) RNA pull-down assays using in vitro transcribed biotin-labeled LINC01787 was performed to detect the binding between LINC01787 and pre-miR-125b. (D,E) After transient overexpression of wild type LINC01787 (LINC01787-wt) or pre-miR-125b binding sites mutated LINC01787 (LINC01787-mut) in MDA-MB-231 cells, RNA pull-down assays using LINC01787 antisense biotinylated probes were performed to detect the binding between LINC01787 and pre-miR-125b-1 (D) or pre-miR-125b-2 (E). (F,G) After transient overexpression of LINC01787-wt or LINC01787-mut in MDA-MB-231 cells, RIP assays were performed to detect the binding between DICER and pre-miR-125b-1 (F) or pre-miR-125b-2 (G). (H,I) After transient transfection of LINC01787 specific shRNAs into MDA-MB-231 cells, RIP assays were performed to detect the binding between DICER and pre-miR-125b-1 (H) or pre-miR-125b-2 (I). Results are shown as mean ± SD of 3 independent experiments. **P < 0.01, ***P < 0.001 by one-way ANOVA followed by Dunnett's multiple comparisons test, compared with beads, control, or shCtl group.
we further explored whether the direct binding between LINC01787 and pre-miR-125b disturb the binding and process of pre-miR-125b by DICER. After transient overexpression of LINC01787-wt or LINC01787-mut in MDA-MB-231 cells, RIP assays were performed to detect the binding between pre-miR-125b and DICER. The results revealed that DICER could bind both pre-miR-125b (Figures 1F,G). LINC01787 overexpression markedly inhibited the binding between DICER and both pre-miR-125b, which was abolished by the mutation of pre-miR-125b binding sites on LINC01787 (Figures 1F,G). Conversely, LINC01787 knockdown markedly promoted the binding between DICER and both pre-miR-125b (Figures 1H,I,  Figure S1E). The previously reported binding between pre-miR-483 and DICER was used to verify RIP efficiencies, which was not regulated by LINC01787 (Figures S1F,G). The exact number of molecules of LINC01787, pre-miR-125b-1, and pre-miR-125b-2 were quantified in MDA-MB-231 cells. The results showed that the number of molecules of LINC01787 was comparable with that of pre-miR-125b-1 and pre-miR-125b-2 in breast cancer cells (Figure S1H), which support the effects of LINC01787 on pre-miR-125b via direct interaction. Next, we measured pre-miR-125b and mature miR-125b expression levels in MDA-MB-231 cells after transient overexpression or knockdown of LINC01787. The results revealed that LINC01787 overexpression promoted the accumulation of both pre-miR-125b and reduced mature miR-125b level, which were abolished by the mutation of pre-miR-125b binding sites on LINC01787 (Figures 2A,B). LINC01787 knockdown reduced both pre-miR-125b levels and increased mature miR-125b level (Figures 2C,D). Therefore, these data suggested that LINC01787 binds pre-miR-125b, represses the binding and cleavage of pre-miR-125b by DICER, and therefore inhibits the generation of mature miR-125b.

LINC01787 Overexpression Promotes Breast Cancer Cell Proliferation and Migration in a miR-125b Dependent Manner
Due to LINC01787 repressing miR-125b generation and miR-125b has tumor suppressive roles in breast cancer, we next investigated the roles of LINC01787 in breast cancer. Wild-type (wt) or pre-miR-125b binding sites mutated LINC01787 stably overexpressed MDA-MB-231 and MCF7 cells were constructed via stable transfection of LINC01787-wt or LINC01787mut overexpression plasmids (Figures 3A,B). CCK-8 assays showed that LINC01787 overexpression significantly accelerated breast cancer cell proliferation, which was abolished by the mutation of pre-miR-125b binding sites (Figures 3C,D). EdU incorporation assays also revealed that LINC01787 overexpression promoted breast cancer cell proliferation, which was abolished by the mutation of pre-miR-125b binding sites ( Figure 3E). Transwell assays showed that LINC01787 overexpression markedly increased migration ability of breast cancer cells, which was abolished by the mutation of pre-miR-125b binding sites ( Figure 3F). These data suggested that LINC01787 overexpression promotes breast cancer cell proliferation and migration in a miR-125b dependent manner.

LINC01787 Knockdown Inhibits Breast Cancer Cell Proliferation and Migration
To completely elucidate the roles of LINC01787 in breast cancer, LINC01787 stably depleted MDA-MB-231 and MCF7 cells were constructed via stable transfection of two independent LINC01787 specific shRNAs (Figures 4A,B). CCK-8 assays showed that LINC01787 knockdown significantly repressed breast cancer cell proliferation (Figures 4C,D). EdU incorporation assays also revealed that LINC01787 knockdown reduced breast cancer cell proliferation ( Figure 4E). Transwell assays showed that LINC01787 knockdown markedly decreased migration ability of breast cancer cells ( Figure 4F). These data suggested that LINC01787 knockdown inhibits breast cancer cell proliferation and migration.

LINC01787 Promotes Breast Cancer Xenograft Growth in a miR-125b Dependent Manner
Next, we investigated the effects of LINC01787 in breast cancer in vivo. LINC01787-wt or LINC01787-mut stably overexpressed MDA-MB-231 cells were subcutaneously implanted into nude mice. The results showed that LINC01787 overexpression significantly promoted MDA-MB-231 xenograft growth in vivo, which was abolished by the mutation of pre-miR-125b bind sites (Figures 5A,B). LINC01787 overexpression efficiencies and the repressive roles of LINC01787 on miR-125b were further confirmed in the xenograft (Figure 5C). Proliferation marker PCNA IHC staining showed that the xenograft formed by LINC01787 overexpressed MDA-MB-231 cells had significantly more PCNA positive cells than that formed by control cells (Figure 5D). The increasing of PCNA positive cells was abolished by the mutation of pre-miR-125b bind sites ( Figure 5D). TUNEL assays showed that the xenograft formed by LINC01787 overexpressed MDA-MB-231 cells had significantly less apoptotic cells than that formed by control cells, which was also abolished by the mutation of pre-miR-125b bind sites ( Figure 5E). These data suggested that LINC01787 promotes breast cancer xenograft growth in vivo in a miR-125b dependent manner.

miR-125b Overexpression Reverses the Oncogenic Roles of LINC01787 in Breast Cancer
To further elucidate whether the oncogenic roles of LINC01787 in breast cancer are dependent on the negative regulation of miR-125b, we stably overexpressed miR-125b in LINC01787 stably overexpressed MDA-MB-231 cells (Figure 6A). CCK-8 assays showed that enhanced expression of miR-125b reversed the accelerated cell proliferation caused by LINC01787 overexpression (Figure 6B). EdU incorporation assays also revealed that enhanced expression of miR-125b reversed    Results are shown as mean ± SD of 5 mice in each group. *P < 0.05, **P < 0.01, ns, not significant by Kruskal-Wallis test followed by Dunn's multiple comparisons test, compared with control group.
the pro-proliferative roles of LINC01787 in breast cancer ( Figure 6C). Transwell assays showed that enhanced expression of miR-125b reversed the increased migration ability of breast cancer cells caused by LINC01787 overexpression (Figure 6D). These data suggested that miR-125b overexpression reverses the roles of LINC01787 in promoting breast cancer cell proliferation and migration. In addition, miR-125b and LINC01787 stably overexpressed and control MDA-MB-231 cells were subcutaneously implanted into nude mice. The results showed that enhanced expression of miR-125b reversed the accelerated xenograft growth caused by LINC01787 overexpression (Figures 6E,F). LINC01787 and miR-125b overexpression efficiencies were confirmed in the xenografts ( Figure S2A). IHC detection of miR-125b targets KIAA1522, ETS1, and SNAI1 in the xenografts revealed that LINC01787 overexpression up-regulated the protein levels of KIAA1522, ETS1, and SNAI1, which were abolished by the overexpression of miR-125b (Figures S2B-D). Proliferation marker PCNA IHC staining showed that enhanced expression of miR-125b reversed the increasing of PCNA positive cells caused by LINC01787 overexpression (Figure 6G). TUNEL assays showed that enhanced expression of miR-125b reversed the reduction of apoptotic cells caused by LINC01787 overexpression (Figure 6H). Collectively, these data suggested that miR-125b overexpression reverses the oncogenic roles of LINC01787 in breast cancer.

LINC01787 Is Up-Regulated in Breast Cancer and Associated With Poor Prognosis
To investigate the clinical significance of LINC01787 in breast cancer, we collected 89 pairs of breast cancer tissues and normal adjacent tissues. The expression of LINC01787 in these tissues was measured by qRT-PCR. The results showed that LINC01787 was significantly up-regulated in breast cancer tissues compared with normal adjacent tissues ( Figure 7A). The expression of miR-125b was measured in the same breast cancer tissues. The results showed that miR-125b expression level was inversely associated with LINC01787 expression level in breast cancer tissues ( Figure 7B). Analyzing the correlation between LINC01787 expression level and clinicopathologic characteristics of these 89 breast cancer patients revealed that high expression level of LINC01787 was positively correlated with tumor size, lymph node metastasis, and advanced clinical stage ( Table 1). No correlation was observed between LINC01787 expression level and age, ER status, PR status, or CerbB2 status ( Table 1). The correlation between LINC01787 expression level and prognosis of these 89 breast cancer patients showed that breast cancer patients with high LINC01787 expression level had worse prognosis that those with low LINC01787 expression level ( Figure 7C). The correlation between LINC01787 expression levels and prognosis of breast cancer patients was further analyzed by The Kaplan Meier plotter (http://kmplot.com/ analysis/), which includes 1089 breast cancer cases. The result also showed that high expression of LINC01787 was associated with poor prognosis of breast cancer patients ( Figure S3). Consistent with previous report (24), breast cancer patients with low miR-125 expression level also had worse prognosis that those with high miR-125 expression level in our cohort ( Figure 7D). These data support the negative regulation of miR-125b by LINC01787 in vivo. These data also demonstrated that LINC01787 is up-regulated in breast cancer and associated with poor prognosis of breast cancer patients.

DISCUSSION
More and more studies are focusing on the important roles and clinical significance of lncRNAs in cancers (43). Due to their key roles in cancers, enhancing tumor suppressive lncRNAs or targeting oncogenic lncRNAs represent novel therapeutic approaches for various cancers. The aberrant expressions and important roles of lncRNAs in breast cancer have begun to be revealed (44)(45)(46). lncRNA CYTOR (LINC00152) was revealed to be up-regulated in breast cancer and associated with poor survival of breast cancer patients (44). CYTOR was also shown to be required for breast cancer cell proliferation, migration, and cytoskeletal organization (44). lncRNA DSCAM-AS1 was found to promote breast cancer progression and tamoxifen resistance (45). LncRNA-Hh was revealed to enhance breast cancer stem cells generation (46). Although several lncRNAs have been studied in breast cancer, the roles of lncRNAs in breast cancer are far from being understood. In this study, we identified a novel lncRNA LINC01787, which locates at chromosome 1p21.3 and has six exons containing 1994 nucleotides in length. The expression and roles of LINC01787 in cancers are still unknown. In this study, we found that LINC01787 is up-regulated in breast cancer and associated with larger tumor size, lymph node metastasis, advanced clinical stage and poor prognosis of breast cancer patients. Functional assays showed that overexpression of LINC01787 enhances the proliferation and migration of breast cancer cells, and conversely knockdown of LINC01787 represses the proliferation and migration of breast cancer cells. Furthermore, we also found that LINC01787 promotes breast cancer xenograft growth in vivo. Thus, we identified that LINC01787 plays key oncogenic roles in breast cancer. Our data also suggested that LINC01787 is a potential therapeutic target for breast cancer.
The molecular mechanisms underlying the roles of lncRNAs are complex. The most frequently reported mechanism is to bind proteins and change the location, modification, expression, and function of the interacted proteins (47). Another common mechanism of lncRNAs is to competitively bind common miRNAs and relieve the repressive roles of miRNAs on their targets (48). These lncRNAs are classified as competing endogenous RNA (ceRNA), such as the binding of miR-200s by lncRNA-ATB, the binding of miR-133 and miR-135 by linc-MD1, and the binding of miR-145 and miR-181a by lincRNA-RoR (49)(50)(51). In this study, we identified a relative novel mechanism of lncRNA. We found that LINC01787 could specifically bind pre-miR-125b. Mature miR-125b, also known as miR-125b-5p, was generated from two genomic sites. MIR125B1 is located at chromosome 11q24.1, which transcribes pre-miR-125b-1. MIR125B2 is located at chromosome 21q21.1, which transcribes pre-miR-125b-2. Both pre-miR-125b-1 and pre-miR-125b-2 are bound and processed by DICER to generate mature miR-125b. In this study, using RNA pull-down assay with in vitro-transcribed LINC01787 and ChIRP assay with LINC01787 antisense biotinylated probes, we found that LINC01787 specifically binds both pre-miR-125b. Although both these assays could not exclude the potential that the interaction between LINC01787 and pre-miR-125b might be mediated by other factors, such as proteins, the findings that the mutation of pre-miR-125b binding sites on LINC01787 abolished the interaction between LINC01787 and pre-miR-125b are in favor of a direct interaction between LINC01787 and pre-miR-125b. Further, we found that via binding pre-miR-125b, LINC01787 represses the binding between pre-miR-125b and DICER, represses the processing of pre-miR-125b by DICER, and therefore inducing the accumulation of pre-miR-125b and repressing mature miR-125b generation. Through inhibiting mature miR-125 generation, LINC01787 up-regulates the expression of miR-125b targets including KIAA1522, ETS1, and SNAI1. Therefore, we provided new evidence about the regulation of miRNA by lncRNA via modulating miRNAs generation.
The negative regulation of miR-125b by LINC01787 was further verified by functional rescue assays and expression association in clinical tissues. The mutation of the pre-miR-125b binding sites on LINC01787 abolished the roles of LINC01787 in promoting breast cancer cell proliferation, migration, and in vivo xenograft growth. Furthermore, overexpression of miR-125b also abolished the roles of LINC01787 in promoting breast cancer cell proliferation, migration, and in vivo xenograft growth. Thus, these functional rescue assays suggested that the oncogenic roles of LINC01787 are dependent on the negative modulation of miR-125b. The expression correlation of LINC01787 and miR-125b was analyzed in 89 breast cancer tissues. Our data revealed that the expression of miR-125b was significantly inversely correlated with that of LINC01787 in breast cancer tissues. Conversely to LINC01787, low expression of miR-125b predicts poor survival of breast cancer patients. The inverse expression association between LINC01787 and miR-125b supports the negative regulation of miR-125b by LINC01787 in vivo.
In conclusion, our results identified LINC01787 as a critical oncogenic lncRNA in breast cancer, promoting breast cancer cell proliferation, migration, and in vivo xenograft growth via binding pre-miR-125b and repressing mature miR-125b generation. Furthermore, we identified LINC01787 as an up-regulated and poor survival-associated lncRNA in breast cancer. Our results thus suggests that LINC01787 is a potential prognostic biomarker and therapeutic target for breast cancer.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of Xinxiang Medical University (Xinxiang, China). The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by the Ethics Committee of Xinxiang Medical University (Xinxiang, China).