Silencing of long non-coding RNA CASC15 enhances radiosensitivity of esophageal squamous cell carcinoma through microRNA-140-5p-dependent inhibition of MMP7

LncRNA CASC15 is highly-expressed and associated with a poor prognosis in ESCC

Initially, the expression profiles of lncRNA and mRNA in esophageal mucosa samples obtained from normal individuals and ESCC patients were analyzed using the microarray data. The ESCC-related gene expression dataset GSE45168 in the Gene Expression Omnibus (GEO) database revealed a total of 941 differentially expressed lncRNAs. Amongst them, 5 lncRNAs exhibited significant differences in the expression between the ESCC and normal tissues (Figure 1A), with lncRNA CASC15 being markedly elevated in ESCC tissues (Figure 1B). Meanwhile, the expression profile of lncRNA CASC15 was verified using The Cancer Genome Atlas (TCGA) database, the results of which revealed that the expression of lncRNA CASC15 was evidently augmented in ESCC tissues (Figure 1C), which was consistent with the results obtained from the GEO database. Due to the existence of numerous spliced transcripts in lncRNA CASC15, we designed various forms of primers to detect lncRNA CASC15 expression. The primer #1 exhibiting the highest detection efficiency (Figure 1D), was thus selected for subsequent detection of the lncRNA CASC15 expression patterns. In order to further validate the predicted results, the expression patterns of lncRNA CASC15 in 79 ESCC tissue specimens (including 6 cases in stage I, 31 in stage II, 27 in stage III, and 15 in stage IV) and corresponding adjacent normal tissues were detected using reverse transcription quantitative polymerase chain reaction (RT-qPCR) (Figure 1E), which revealed that lncRNA CASC15 was up-regulated in ESCC tissues, especially in an advanced TNM stage, relative to adjacent normal tissues (p < 0.05). Additionally, the expression of lncRNA CASC15 in the five ESCC cell lines (EC109, TE-1, KYSE70, KYSE270, and KYSE30) and the human esophageal epithelial cell (HEEC) cell line was measured using RT-qPCR. As shown in Figure 1F, compared with the HEEC cell line, all ESCC cell lines exhibited significantly elevated expression profiles of lncRNA CASC15 (p < 0.05), with KYSE270 and EC109 cells presenting with the highest lncRNA CASC15 expression. Therefore, the KYSE270 and EC109 cell lines were selected for further experiments. Subsequently, a Kaplan-Meier survival curve was plotted to analyze the correlation between the expression of lncRNA CASC15 and the overall survival rate of ESCC patients (Figure 1G), which revealed a negative correlation between the expression of lncRNA CASC15 and the survival rate of ESCC patients. In addition, as shown in Table 1, the expression of lncRNA CASC15 was irrelevant to parameters such as age, sex and tumor size of ESCC patients, yet being closely associated with the TNM stage, lymph node metastasis (LNM) and differentiation. Conjointly, the aforementioned findings suggested that lncRNA CASC15 was expressed at a higher level in ESCC, and further associated with a poor prognosis of ESCC.

Figure 1. LncRNA CASC15 is overexpressed with poor prognostic value in ESCC. (A) Analysis of five aberrant lncRNAs in ESCC-related gene expression dataset GSE45168 in GEO database. (B) The expression of five aberrant lncRNAs in GSE45168 (n = 5. the left box plot represents the expression of lncRNAs in the normal sample, and the right box plot represents the expression of lncRNAs in the ESCC samples). (C) The expression of lncRNA CASC15 in ESCC and adjacent normal tissues analyzed in the TCGA database. (D) The expression of lncRNA CASC15 in EC109 cells detected by different primers. (E) The expression of lncRNA CASC15 in ESCC and adjacent normal tissues (n = 79 including 6 cases in stage I, 31 in stage II, 27 in stage III, and 15 in stage IV) determined by RT-qPCR (*p < 0.05 vs. the adjacent normal tissues). (F) The expression of lncRNA CASC15 in ESCC cell lines and HEEC cells determined by RT-qPCR (*p < 0.05 vs. the HEEC cell line). (G) Kaplan-Meier survival curve to analyze the correlation between lncRNA CASC15 expression and overall survival rate of 79 ESCC patients (log-rank test for the difference significance). The measurement data in panel (D–F) were presented as mean ± standard deviation and compared with one-way analysis of variance, followed by Tukey’s post hoc test.

Silencing lncRNA CASC15 inhibits proliferation and invasion of ESCC cells

Since lncRNA CASC15 was extensively expressed in ESCC cells, we chose to silence lncRNA CASC15 using a short hairpin RNA (shRNA) in KYSE270 and EC109 cells to evaluate its effects on ESCC cells. The silencing efficiency was detected by means of RT-qPCR (Figure 2A), which revealed that the expression of lncRNA CASC15 in cells transfected with sh-CASC15 was significantly lower compared to that observed in cells transfected with sh-negative control (NC) (p < 0.05), indicating that lncRNA CASC15 was efficiently knocked down. Subsequently, Cell Counting Kit-8 (CCK-8) and Transwell assays were employed to evaluate the viability and invasion of KYSE270 and EC109 cells after lncRNA CASC15 silencing (Figure 2B and 2C). In comparison to the cells transfected with sh-NC, cells transfected with sh-CASC15 exhibited remarkably decreased viability and invasive abilities (p < 0.05). Moreover, Western blot analysis was applied to determine the expression patterns of proliferation-related factor Ki-67 and invasion-related factor MMP2 (Figure 2D and 2E), both of which were evidently reduced upon silencing of lncRNA CASC15 (p < 0.05). Thus, these findings suggested that the proliferation and invasion of ESCC cells were decreased after silencing lncRNA CASC15.

Figure 2. Silencing lncRNA CASC15 reduces the proliferation and invasion of ESCC cells. (A) RT-qPCR to detect the silencing efficiency of lncRNA CASC15. (B) CCK-8 assay to assess the viability of KYSE270 and EC109 cells in response to silencing lncRNA CASC15. (C) Transwell assay to evaluate the invasion of KYSE270 and EC109 cells after silencing lncRNA CASC15 (× 200). (D–E) Western blot analysis to determine the protein level of Ki-67 (a proliferation-related factor) and MMP2 (an invasion-related factor) in KYSE270 and EC109 cells. *p < 0.05 vs. the ESCC cells transfected with sh-NC. The measurement data were presented as the mean ± standard deviation. Data in panel (A, C, E) were analyzed by unpaired t test and those in panel (B) by one-way analysis of variance, followed by Tukey’s post hoc test. the experiment was repeated 3 times. sh-NC, cells transfected with negative control plasmids of short hairpin RNA; sh-CASC15, cells transfected with short hairpin RNA against lncRNA CASC15.

LncRNA CASC15 up-regulates MMP7 expression in ESCC by binding to miR-140-5p

Fluorescence in situ hybridization (FISH) was performed in order to identify the subcellular localization of lncRNA CASC15 in EC109 cells. As shown in Figure 3A, lncRNA CASC15 was primarily expressed in the cytoplasm of EC109 cells, thereby suggesting the potential function of lncRNA CASC15 as a regulator of EC109 cellular behaviors. Therefore, the downstream miRNAs regulated by lncRNA CASC15 were predicted from the RNA22 database and RAID database, respectively. Moreover, based on the GEO database, we analyzed the ESCC-related miRNA expression dataset GSE97049 and identified 45 poorly expressed miRNAs in ESCC (Figure 3B). Interestingly, only one intersected miRNA (miR-140-5p) was profound, which was expressed at a low level in ESCC and regulated by lncRNA CASC15 (Figure 3C), signifying the potential regulatory properties of lncRNA CASC15 in ESCC via possible regulation of miR-140-5p. As shown in Figure 3D, RT-qPCR further confirmed that the expression of miR-140-5p was down-regulated in both ESCC tissues and cells (p < 0.05). Next, we analyzed the correlation between the lncRNA CASC15 and miR-140-5p expression in ESCC tissues (Figure 3E) and cells (Figure 3F). Noticeably, a negative correlation was apparent between the expression of lncRNA CASC15 and miR-140-5p. Additionally, in order to further clarify the downstream regulation mechanism of miR-140-5p, the downstream target genes of miR-140-5p were predicted using the RNA22 database (Figure 3G), and the putative target genes were intersected with the up-regulated genes in GSE45168 dataset. Finally, a sum of 145 potential genes were obtained, which were presented in Figure 3H. Protein-protein interaction correlation analysis exhibited that the MMP7 gene was at the core of these 145 genes (Figure 3I). Remarkably, a close association between MMP7 and the radiosensitivity of tumors was evident [27, 28]. Thus, it was speculated that the MMP7 gene mediated by miR-140-5p may be involved in the regulation of the radiosensitivity of ESCC. To further ascertain this speculation, the expression patterns of MMP7 and other candidate downstream target genes KCNA2 and S100A14 in ESCC tissues and cells were measured by means of RT-qPCR and Western blot analyses (Figure 3J). The results illustrated that MMP7 was solely up-regulated at mRNA and protein levels in both ESCC tissues and cells (p < 0.05), and no significant differences were observed in the expressions of KCNA2 and S100A14 (p > 0.05). Thus, MMP7 was chosen as the downstream target of miR-140-5p and lncRNA CASC15 for subsequent research. Next, the effect of miR-140-5p on lncRNA CASC15 and MMP7 was assessed. As displayed in Figure 3K, miR-140-5p could negatively regulate the expression of lncRNA CASC15 and MMP7. The correlation between the expression of lncRNA CASC15 and MMP7 in ESCC was analyzed, with the results illustrating a positive correlation in Figure 3L. Subsequently, the putative binding sites between lncRNA CASC15 and miR-140-5p, as well as between miR-140-5p and MMP7 were predicted with the help of a biological prediction website (Figure 3M). Dual-luciferase reporter assay was performed to verify the regulatory relationship between lncRNA CASC15 and miR-140-5p as well as the binding relationship between miR-140-5p and MMP7 using recombinant wild type (wt) CASC15/MMP7 and mutant type (mut) CASC15/MMP7 plasmids (Figure 3N–O). In comparison to the EC109 cells transfected with NC mimic, the luciferase signals of wt-CASC15 and wt-MMP7 were noted to be decreased in the EC109 cells transfected with miR-140-5p mimic (p < 0.05), but there were no significant differences in the luciferase activities of mut-CASC15 and mut-MMP7 between the EC109 cells transfected with NC mimic and miR-140-5p mimic (p > 0.05). The binding relationship between lncRNA CASC15 and miR-140-5p or between lncRNA CASC15 and MMP7 was detected using the RNA-binding protein immunoprecipitation (RIP) experiment (Figure 3P). In comparison to the anti-IgG control, cells treated with anti-Ago2 exhibited significantly increased enrichment of miR-140-5p, lncRNA CASC15, and MMP7 (p < 0.05). Thereafter, to further verify the aforementioned finding, RIP was performed to assess the amount of MMP7 mRNA interacting with miR-140-5p in response to over-expression or knockdown of CASC15. The results revealed that the amount of MMP7 mRNA interacting with miR-140-5p was reduced after lncRNA CASC15 over-expression, yet opposite trends were noted upon lncRNA CASC15 silencing (p < 0.05) (Figure 3Q). Moreover, RT-qPCR (Figure 3R) and dual-luciferase reporter assay (Figure 3S) revealed that mRNA expression and luciferase activity of MMP7 were elevated after lncRNA CASC15 over-expression, which was rescued by miR-140-5p elevation (p < 0.05). With these results as basis, we speculated that lncRNA CASC15 may bind to miR-140-5p in ESCC, thereby up-regulating the expression of MMP7.

Figure 3. LncRNA CASC15 up-regulates MMP7 expression in ESCC by binding to miR-140-5p. (A) Subcellular localization of lncRNA CASC15 in EC109 cells (× 400). (B) A heat map of down-regulated miRNAs in GSE97049 dataset (the X label indicates the sample number, while the Y label indicates the miRNA name, and each small square indicates the expression of one miRNA in one sample). (C) The intersection of predicted miRNAs regulated by lncRNA CASC15 from RNA 22 database, RAID database and GSE97049 dataset (three circles represent the miRNAs in the RNA 22 database, RAID database and GSE97049 dataset respectively, and the middle part indicates the intersection). (D) RT-qPCR to detect the expression of miR-140-5p in ESCC tissues and cells (*p < 0.05 vs. the adjacent normal tissues or HEEC cells). (E) The correlation analysis of lncRNA CASC15 and miR-140-5p expression in ESCC tissues. (F) The correlation between lncRNA CASC15 and miR-140-5p expression in EC109 cells (*p < 0.05 vs. EC109 cells transfected with oe-NC, #p < 0.05 vs. EC109 cells transfected with sh-NC). (G) A heat map of differentially expressed genes in GSE45168 dataset (the X axis indicates the sample number, while the Y axis indicates the miRNA name, and each small square indicates the expression of one miRNA in one sample). (H) The intersection of predicted downstream target genes of miR-140-5p from RNA22 database and differentially expressed genes obtained from GSE45168 dataset (two circles indicate up-regulated genes in the predicted results from RNA22 database and GSE45168 dataset, and the middle part indicates their interaction). (I) The correlation analysis of the predicted downstream target genes of miR-140-5p (each circle refers to one gene, and the ligature between two circles represents interaction between two genes). (J) RT-qPCR and Western blot analyses to detect the expression of MMP7 and other candidate target genes (KCNA2 and S100A14) at mRNA and protein levels in ESCC tissues and cells (*p < 0.05 vs. the adjacent normal tissues or HEEC cells). (K) RT-qPCR to detect the expression of lncRNA CASC15 and MMP7 measured after upregulating or downregulating miR-140-5p (*p < 0.05 vs. EC109 cells transfected with NC mimic, #p < 0.05 vs. EC109 cells treated with NC inhibitor). (L) The correlation analysis between lncRNA CASC15 and MMP7 expression in ESCC. (M) A bioinformatic prediction website to predict the putative binding sites between miR-140-5p and lncRNA CASC15 or MMP7. (N) Dual-luciferase reporter assay to examine the luciferase activity after the co-transfection of miR-140-5p mimic with wt-CASC15 and mut-CASC15. (O) Dual-luciferase reporter assay to test the luciferase activity after the co-transfection of miR-140-5p mimic with wt-MMP7 and mut-MMP7. (*p < 0.05 vs. EC109 cells transfected with NC mimic). (P) RIP to assess the enrichment of lncRNA CASC15, miR-140-5p, and MMP7 (*p < 0.05 vs. normal mouse IgG). (Q) RIP to study the effect of lncRNA CASC15 on the binding of miR-140-5p and MMP7 (*p < 0.05 vs. the treatment of oe-NC anti-Ago2; #p < 0.05 vs. the treatment of sh-NC anti-Ago2). (R) RT-qPCR to detect the mRNA expression of MMP7 in ESCC cells (*p < 0.05 vs. the treatment of oe-NC and NC mimic; #p < 0.05 vs. the treatment of oe-CASC15 and NC mimic). (S) Dual-luciferase reporter assay to test the luciferase activity after the co-transfection of oe-CASC15 + miR-140-5p mimic with wt-MMP7 and mut-MMP7 (*p < 0.05 vs. EC109 cells transfected with oe-NC + NC mimic; #p < 0.05 vs. EC109 cells transfected with oe-CASC15 + NC mimic). The measurement data were presented as the mean ± standard deviation. Data in panel (D) (left) and (J) (left) were analyzed by paired t test and those in panel (D) (right) and (F, J) (right) and (K, N, O, P, Q, R, S) by unpaired t test; the experiment was repeated 3 times. NC mimic, EC109 cells transfected with negative control plasmids of mimic; miR-140-5p mimic, EC109 cells transfected with miR-140-5p mimic; oe-NC, EC109 cells transfected with negative control plasmids of overexpression vector; sh-NC, EC109 cells transfected with negative control plasmids of short hairpin RNA; wt-CASC15, EC109 cells transfected with luciferase reporter plasmids containing wild-type lncRNA CASC15; mut-CASC15, EC109 cells transfected with luciferase reporter plasmids containing mutant lncRNA CASC15; wt-MMP7, EC109 cells transfected with luciferase reporter plasmids containing wild-type MMP7; mut-MMP7, EC109 cells transfected with luciferase reporter plasmids containing mutant MMP7.

LncRNA CASC15 inhibits the radiosensitivity of ESCC by down-regulating miR140-5p and up-regulating MMP7

In order to investigate the regulatory role of lncRNA CASC15 in response of radiosensitivity against ESCC, the expression patterns of lncRNA CASC15, miR-140-5p, and MMP7 in the radiotherapy-sensitive and radiotherapy-resistant ESCC patients were determined by means of RT-qPCR (Figure 4A) and Western blot analyses (Figure 4B). Results unraveled that the expressions of lncRNA CASC15 and mRNA and protein expression of MMP7 were higher, while those of miR-140-5p were lower in radiotherapy-resistant patients compared to radiotherapy-sensitive patients (p < 0.05). Meanwhile, the expression of lncRNA CASC15 was further determined in the ESCC parental cell line TE-1 and radiation-resistant cell line TE-1R (Figure 4C). It was found that lncRNA CASC15 was expressed at a high level in the TE-1R cell line relative to the TE-1 line (p < 0.05). The aforementioned findings suggested that the lncRNA CASC15/miR-140-5p/MMP7 axis affected the radiosensitivity against ESCC.

Figure 4. LncRNA CASC15 inhibits the sensitivity of ESCC to radiotherapy by decreasing miR-140-5p and upregulating MMP7. (A) RT-qPCR to detect the expression of lncRNA CASC15, miR-140-5p, and MMP7 in the radio-sensitive and radio-resistant ESCC patients (*p < 0.05 vs. the sensitive group). (B) Western blot analysis to measure the protein expression of MMP7 in the radio-sensitive and radio-resistant ESCC patients (*p < 0.05 vs. the sensitive group). (C) RT-qPCR to determine the expression of lncRNA CASC15 in ESCC naive cell line TE-1 and radiation resistant cell line TE-1R (*p < 0.05 vs. ESCC parental cell line TE-1). (D) RT-qPCR to test the transfection efficiency in EC109 cells. (E) CCK-8 assay to evaluate cell viability of EC109 cells. (F) Flow cytometry to assess cell apoptosis of EC109 cells. (G) Western blot analysis to determine the protein level of Fas, Ki-67 and cleaved Caspase 3 in EC109 cells. (H) Clone formation ability of EC109 cells after X-ray radiation. In panel (C–G), *p < 0.05 vs. EC109 cells transfected with 4 Gy + NC mimic. #p < 0.05 vs. EC109 cells transfected with 4 Gy + sh-NC. &p < 0.05 vs. EC109 cells transfected with 4 Gy + oe-NC + sh-NC; @p < 0.05 vs. EC109 cells transfected with 4 Gy + oe-CASC15 + sh-NC. The measurement data were presented as the mean ± standard deviation. Data in panel (A, B) were analyzed by unpaired t test, and data in panel (C, E, F) were analyzed by one-way analysis of variance, followed by Tukey’s post hoc test. Data in panel D were analyzed by repeated-measures analysis of variance, followed by Tukey’s post hoc test, and data in panel (E) were analyzed by two-way analysis of variance, followed by Tukey’s post hoc test; the experiment was repeated 3 times. 4 Gy + NC mimic, EC109 cells transfected with negative control of mimic and treated with 4 Gy irradiation; 4 Gy + miR-140-5p mimic, EC109 cells transfected with miR-140-5p mimic and treated with 4 Gy irradiation; 4 Gy + sh-NC, EC109 cells transfected with negative control plasmids of short hairpin RNA and treated with 4 Gy irradiation; 4 Gy + sh-MMP7, EC109 cells transfected with short hairpin RNA against matrix metallopeptidase 7 and treated with 4 Gy irradiation; 4 Gy + oe-NC + sh-NC, EC109 cells transfected with negative control plasmids of overexpression vector and short hairpin RNA and treated with 4 Gy irradiation; 4 Gy + oe-CASC15 + sh-NC, EC109 cells transfected with plasmids overexpressing lncRNA CASC15 and negative control of short hairpin RNA and treated with 4 Gy irradiation; 4 Gy + oe-CASC15 + sh-MMP7, EC109 cells transfected with plasmids overexpressing lncRNA CASC15 and short hairpin RNA against matrix metallopeptidase 7 and treated with 4 Gy irradiation.

Furthermore, to investigate the role of this axis in the radiosensitivity of ESCC cells, KYSE270 and EC109 cells were transfected with different plasmids (including NC mimic, miR-140-5p mimic, sh-NC, sh-MMP7, sh-CASC15, oe-NC, oe-CASC15) and irradiated with 4 Gy X-ray. The transfection efficiency was detected by means of RT-qPCR (Figures 4D and 5A), which demonstrated that the transfection efficiency reached the required threshold for further experiments. Next, cell viability was measured by conducting a CCK-8 assay (Figure 4E and 5B). The results revealed that compared to the matched controls, over-expression of lncRNA CASC15 in KYSE270 and EC109 cells at 4 Gy significantly enhanced cell viability (p < 0.05), which was reversed by knocking down MMP7 (p < 0.05). Besides, decreased cell viability was evident following the treatment of single miR-140-5p mimic or sh-MMP7 at 4 Gy (p < 0.05). Cell apoptosis was then determined by conducting flow cytometry (Figure 4F and 5C), the results of which suggested that over-expression of lncRNA CASC15 in KYSE270 and EC109 cells irradiated with 4 Gy remarkably decreased cell apoptosis (p < 0.05), which could be overturned by silencing MMP7 (p < 0.05). The treatment of single miR-140-5p mimic or sh-MMP7 at 4 Gy exclusively induced apoptosis (p < 0.05). Western blot analysis was conducted to determine the expression patterns of proliferation-related factor Fas and Ki-67 and apoptosis-related factor cleaved Caspase 3. As shown in Figure 4G and 5D, KYSE270 and EC109 cells transfected with oe-CASC15 at 4 Gy showed increased expression of Ki-67 and dropped expression of Fas and cleaved caspase-3, while co-transfection of sh-MMP7 inhibited the effects of oe-CASC15 on Fas, Ki-67 and cleaved Caspase-3 levels (p < 0.05). In addition, a single transfection of the miR-140-5p mimic or sh-MMP7 weakened the expression patterns of Ki-67 and augmented the expression profiles of Fas and cleaved Caspase-3. Moreover, KYSE270 and EC109 cells were treated with 0, 2, 4, 6, and 8 Gy X-ray respectively, and then clone formation experiments were conducted (Figure 4H and 5E). Results demonstrated that lncRNA CASC15 could potentially enhance the clone forming ability of KYSE270 and ESCC cells at variable doses of X-ray (p < 0.05), which was prevented after down-regulation of MMP7 (p < 0.05). Moreover, the clone forming ability was also diminished after miR-140-5p elevation and MMP7 inhibition (p < 0.05). Thus, the aforementioned results indicated that lncRNA CASC15 inhibited the sensitivity of ESCC to radiotherapy by down-regulating miR-140-5p and up-regulating MMP7.

Figure 5. LncRNA CASC15 impairs the radiosensitivity of ESCC by decreasing miR-140-5p and elevating MMP7. (A) RT-qPCR to test the transfection efficiency in KYSE270 cells. (B) CCK-8 assay to assess cell viability of KYSE270 cells. (C) Flow cytometry to evaluate cell apoptosis of KYSE270 cells. (D) Western blot analysis to determine the protein level of Fas, Ki-67 and cleaved Caspase 3 in KYSE270 cells. E, Clone formation ability of KYSE270 cells after X-ray radiation. *p < 0.05 vs. KYSE270 cells transfected with 4 Gy + NC mimic; #p < 0.05 vs. KYSE270 cells transfected with 4 Gy + sh-NC; &p < 0.05 vs. KYSE270 cells transfected with 4 Gy + oe-NC + sh-NC; @p < 0.05 vs. KYSE270 cells transfected with 4 Gy + oe-CASC15 + sh-NC. The measurement data were presented as the mean ± standard deviation. Data in panel (A, C, D) were analyzed by one-way analysis of variance, followed by Tukey’s post hoc test. Data in panel B were analyzed by repeated-measures analysis of variance, followed by Tukey’s post hoc test, and data in panel C were analyzed by two-way analysis of variance, followed by Tukey’s post hoc test; the experiment was repeated 3 times. 4 Gy + NC mimic, KYSE270 cells transfected with negative control of mimic and treated with 4 Gy irradiation; 4 Gy + miR-140-5p mimic, KYSE270 cells transfected with miR-140-5p mimic and treated with 4 Gy irradiation; 4 Gy + sh-NC, KYSE270 cells transfected with negative control plasmids of short hairpin RNA and treated with 4 Gy irradiation; 4 Gy + sh-MMP7, KYSE270 cells transfected with short hairpin RNA against matrix metallopeptidase 7 and treated with 4 Gy irradiation; 4 Gy + oe-NC + sh-NC, KYSE270 cells transfected with negative control plasmids of overexpression vector and short hairpin RNA and treated with 4 Gy irradiation; 4 Gy + oe-CASC15 + sh-NC, KYSE270 cells transfected with plasmids overexpressing lncRNA CASC15 and negative control of short hairpin RNA and treated with 4 Gy irradiation; 4 Gy + oe-CASC15 + sh-MMP7, KYSE270 cells transfected with plasmids overexpressing lncRNA CASC15 and short hairpin RNA against matrix metallopeptidase 7 and treated with 4 Gy irradiation.

Silencing lncRNA CASC15 increases the radiosensitivity of ESCC in vivo

In an attempt to assess the effect of lncRNA CASC15 on the radiosensitivity of ESCC in vivo, nude mice were injected with KYSE270 and EC109 cells infected with a lentivirus containing sh-CASC15. Upon observing the mean volume of tumors approaching 300 mm³, the mice were irradiated once a week by 4 Gy X-ray. The volume of tumors was measured every 3 days after irradiation, and the nude mice were euthanized 21 days later with the tumors collected. As detected by RT-qPCR, the lncRNA CASC15 was efficiently impeded by lentivirus-packaged shRNA, after which the expression of miR-140-5p was increased while that of MMP7 was reduced in tumors (Figure 6A) (p < 0.05). Moreover, the volume and weight of tumors were found to be lower after treatment with 4 Gy and sh-CASC15 relative to treatment with 4 Gy and sh-NC (Figure 6B and 3C) (p < 0.05). Then, the expression patterns of Ki-67 and cleaved Caspase 3 and MMP7 in tumors were measured by means of Western blot and immunohistochemistry (IHC) analyses (Figure 6D and 6E), which revealed that treatment with 4 Gy and sh-CASC15 led to a reduction in the level of Ki-67, and an elevation in the level of cleaved Caspase 3 and MMP7 in contrast to the treatment with 4 Gy and sh-NC (p < 0.05). These results demonstrated that silencing lncRNA CASC15 could effectively increase the radiosensitivity of ESCC in vivo.

Figure 6. Silencing lncRNA CASC15 increases the radiosensitivity of ESCC in vivo. (A) RT-qPCR to detect the silencing efficiency of lncRNA CASC15 and the expression of miR-140-5p and MMP7 in tumors. (B) The volume of tumors following X-ray radiation and lncRNA CASC15 silencing. (C) The weight of tumors following X-ray radiation and lncRNA CASC15 silencing. (D) The levels of MMP7, Ki-67 and cleaved Caspase 3 in tumors in response to X-ray radiation and lncRNA CASC15 silencing determined by Western blot analysis. (E) The levels of MMP7, Ki-67 and cleaved Caspase 3 in tumors in response to X-ray radiation and lncRNA CASC15 silencing detected by IHC analysis (× 200). *p < 0.05 vs. the 4 Gy + sh-NC group. n = 12. The measurement data were presented as the mean ± standard deviation. Data in panel (A, C–E) were analyzed by unpaired t test and panel B by repeated-measures analysis of variance, followed by Tukey’s post hoc test; the experiment was repeated 3 times. 4 Gy + sh-CASC15, 4 Gy irradiated mice with EC109 cells infected with lentivirus containing shRNA targeting CASC15, 4 Gy + sh-NC, 4 Gy irradiated mice with EC109 cells infected with lentivirus containing negative control shRNA.

Ethical statement

The current study protocols were approved by the Ethics Committee of Affiliated Cancer Hospital of Zhengzhou University (Henan Cancer Hospital). Signed informed consents were obtained from all participants prior to enrollment. All animal experiments were in accordance with the guidelines issued in the Guide for the Care and Use of Laboratory Animal by National Institutes of Health, and great efforts were made to minimize the suffering of the included animals.

Bioinformatics prediction

Microarray data, including data related to cancerous and normal tissues from ESCC patients, were retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). Subsequently, the differentially expressed lncRNAs were identified using the Limma package of R language, with the cut-off value set at p < 0.05 and |log2 (fold change)| > 2. The most significant differentially expressed lncRNAs in ESCC-related microarray were verified in the TCGA. After the lncRNAs were screened, their downstream miRNAs were predicted using the RNA22 (https://cm.jefferson.edu/rna22/Precomputed/OptionController?species=HomoSapiens) and RAID database (http://www.rna-society.org/raid/search.html). Moreover, the target genes of the candidate miRNA were further predicted using the RNA22 database (https://cm.jefferson.edu/rna22/Precomputed/). Subsequently, the protein interaction network was obtained from the String database (https://string-db.org/).

Study subjects

ESCC tissues and adjacent normal tissues were obtained from ESCC patients who underwent esophagectomy at the Affiliated Cancer Hospital of Zhengzhou University (Henan Cancer Hospital) between February 2016 and July 2017, which were preserved in liquid nitrogen. None of the included patients underwent chemotherapy or radiotherapy or any other comprehensive anti-tumor therapies prior to tissue collection. Patients were staged according to the criteria of the American Joint Commission for Cancer (AJCC).

Furthermore, 79 additional patients clinically diagnosed with ESCC at the Affiliated Cancer Hospital of Zhengzhou University (Henan Cancer Hospital) between April 2016 and June 2018 were selected and assigned into the sensitive group and the resistant group after instilling a two-month regimen of radiotherapy. The clinical criteria employed to categorize patients into the sensitive group and the resistant group were as follows: the position of the patients was fixed by the body membrane or neck and shoulder membrane, and breast computed tomography (CT) was applied to enhance the scan orientation with series scanning through 5 mm of thickness. In addition, a three-dimensional treatment planning system was transmitted through the network system. The target region, which was defined based on the 50^th and 62^nd reporting principles of the International Commission Radiological Units (ICRU), was plotted in reference to the breast CT and esophagus barium meal X ray images. Then, the target regions of the primary tumor (GTV-T) and metastatic lymph nodes (GTV-N), clinical target volume (CTV, GTV-T’s external expansion of 0.6 - 0.8 cm at horizontal level; manually modified after external expansion of 3 - 5 cm toward head or foot), planning target volume (PTV, CTV’s external expansion of 0.5 cm) and organ at risk were plotted. Next, 6 MV-X ray was employed for conducting three-dimensional conformal radiation therapy (3DCRT) or intensity modulated radiation therapy (IMRT) (1.8 - 2.2 Gy/time; 5 times/week; total dose, 57 - 67 Gy; median dose, 60.2 Gy). Four weeks after being subjected to the radiotherapy, esophageal barium meal examination was conducted. Patients with complete or partial remission were regarded as the sensitive group, and those with non-remission as the resistant group. Those patients comprised of 64 males and 15 females, with a calculated mean age of (50.44 ± 9.15) years. No significant differences were evident in terms of age and sex between the sensitive and resistant groups. The enrolled ESCC patients were identified with the following clinical manifestations: 1) not underwent radiotherapy before surgery; 2) normal functions of other organs such as the heart, lung, liver, and kidney; 3) no distant metastasis or other malignant tumors. Relevant clinical data are shown in Table 1.

Cell treatment

Normal HEEC cell line along with five ESCC cell lines (EC109, TE-1, KYSE70, KYSE270, and KYSE30) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were cultured in high glucose Dulbecco’s modified eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Gibco, Carlsbad, CA, USA) at 37° C under saturated humidity containing 95% air and 5% CO₂, which were subcultured upon reaching 90% confluence. RT-qPCR was then conducted to screen the cell line with the highest expression of lncRNA CASC15, among which the KYSE270 and EC109 cell lines exhibiting the highest expression of lncRNA CASC15 were chosen for the follow-up experiments. The TE-1R cell line resistant to radiation was screened following the principles of a former study [57, 58].

Then, the cells were plated in 6-well plates and transfected with the following plasmids using lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the provided instructions, including shRNA against lncRNA CASC15 (sh-CASC15), miR-140-5p mimic, shRNA against MMP7 (sh-MMP7), plasmid over-expressing lncRNA CASC15 (oe-CASC15), wt-CASC15, mut-CASC15, wt-MMP7, mut-MMP7 and relative NC plasmids alone or in combination. The aforementioned plasmids were purchased from Dharmacon, Inc. (Lafayette, CO, USA). After 48 h of transfection, the cells were subjected to subsequent experiments.

RNA isolation and quantification

Total RNA content was extracted from tissues and KYSE270 and EC109 cells using an RNeasy Mini Kit (Qiagen, Valencia, CA, USA). Next, the total RNA was reverse transcribed into complementary DNA (cDNA) (20 μL) using the Primescript^TM RT reagent Kit (RR047A, Takara Bio Inc., Otsu, Shiga, Japan). Real time qPCR was performed with the help of a SYBR® Premix Ex TaqTM II reagent kit (RR420A, Takara Bio Inc., Otsu, Shiga, Japan) in an ABI7500 real-time quantitative PCR system (Applied Biosystems, Foster City, CA, USA). The primers used are listed in Table 2, which were synthetized by Shanghai Sangon Biotechnology Co., Ltd. (Shanghai, China). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or U6 was regarded as endogenous controls, and relative quantification (2^-∆∆Ct method) was adopted to calculate the fold changes [59].

FISH

FISH technique was adopted to identify the subcellular localization of lncRNA CASC15 in the ESCC cells. EC109 cells were hybridized with several nucleotide probes against lncRNA CASC15 following the manufacturer’s instructions for the RiboTM lncRNA FISH Probe Mix (Red) (C10920, Guangzhou RiboBio Co., Ltd., Guangzhou, China). In brief, the cells were grown in 24-well plates until reaching 60 - 70% confluence. After fixation with 4% paraformaldehyde, the cells were permeabilized in phosphate-buffered saline (PBS) containing 0.5% Triton X-100 and blocked using 200 uL pre-hybridization solution at 37° C for 30 min. Cells were then incubated with the nucleotide probe against lncRNA CASC15 (Wuhan Genecreate Bioengineering Co., Ltd., Hubei, China) overnight at 37° C in conditions avoiding exposure to light. Next, the cells were stained using 4’,6-diamidino-2-phenylindole (DAPI) and observed under a fluorescence microscope (Olympus Optical Co., Ltd., Tokyo, Japan) with 5 randomly selected fields from each sample.

RIP assay

RIP was conducted in the normal EC109 cells or those transfected with the plasmids of sh-CASC15 or oe-CASC15. EC109 cells were then lysed using 100 mL RIP lysis buffer (N653, Haoran Bio Technologies Co., Ltd., Shanghai, China) over ice for 5 min. Next, the cell lysate was added for incubation with 50 μL magnetic beads and 0.5 mL RIP wash buffer (EHJ-BVIS08102, Xiamen Huijia Biotechnology Co., Ltd., Fujian, China). After discarding the supernatant, the magnetic beads were incubated with 5 μg Argonaute 2 (AGO2) antibody (P10502500, Otwo Biotech Co., Ltd., Shenzhen, China) at room temperature for 30 min, with the normal mouse immunoglobulin G (IgG) serving as control. Subsequently, the antibody-coated magnetic beads were resuspended in 900 μL RIP Immunoprecipitation Buffer (P10403138, Otwo Biotech Co., Ltd., Shenzhen, China). After purification with 150 μL proteinase K buffer, RNA context was extracted from the magnetic beads using the Trizol reagent, followed by RT-qPCR detection. The experiment was performed in triplicate to obtain a mean value.

Western blot analysis

Total protein was extracted from the tissues and cells using radio-immunoprecipitation assay (RIPA) lysis buffer containing phenylmethane sulfonyl fluoride (PMSF). After supernatant collection, a bicinchoninic acid (BCA) kit was employed to determine the protein concentration. A total of 50 μg of protein was separated by conducting sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene fluoride membrane. Next, a membrane blockade was conducted using 5% skimmed milk at room temperature for 1 h and subjected to incubation with the diluted rabbit anti-human primary antibodies purchased from Abcam (Cambridge, UK) at 4° C overnight: GAPDH (as an internal control, ab9485, dilution ratio of 1 : 2,500), Fas (ab133619, dilution ratio of 1 : 5000), Ki-67 (ab92742, dilution ratio of 1 : 5,000), MMP7 (ab5706, dilution ratio of 1: 1000), and cleaved Caspase 3 (ab2302, dilution ratio of 1 : 1,000). Then, the membrane was incubated with the horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG secondary antibody (ab97051, dilution ratio of 1 : 2000, Abcam company, Shanghai, China) for 1 h at 37° C. Subsequently, enhanced chemiluminescence (ECL) was adopted to visualize the bands using a Bio-Rad Gel imager (Bio-Rad, Richmond, CA, USA). The gray value was measured using the Quantity One version 4.6.2 software, and the gray value ratio was calculated relative to the internal reference. The experiment was performed in triplicate in each group to obtain the mean value.

Dual-luciferase reporter assay

Luciferase reporter plasmids containing wt-CASC15, mut-CASC15, wt-MMP7, as well as mut-MMP7 were designed and constructed by GenePharma Co., Ltd., (Shanghai, China). The NC mimics and miR-140-5p mimics were respectively co-transfected with wt-CASC15, mut-CASC15, wt-MMP7 or mut-MMP7 plasmids and the renilla plasmid into EC109 cells. 48 h after transfection, a Dual-Luciferase Reporter Assay kit (D0010, Beijing Solarbio Science and Technology Co., Ltd. (Beijing, China) was employed to detect the firefly and renilla luciferase activities. The luminance was measured using a GLomax 20/20 Luminometer (E5311, Shaanxi Zhongmei Biotechnology Co., Ltd., Xi’an, China). The experiment was performed in triplicate in each group to obtain the mean value.

CCK-8 assay for cell viability

Cells were inoculated in 96-well plates at a density of 2 × 10³ cells/well. After 24 h of transfection, the cells were irradiated with X-ray at 4 Gy. Cells in each well were incubated with 10 μL of CCK-8 solution at 37° C at 0, 24, 48, 72, and 96 h after irradiation, respectively. The absorbance of the samples was measured at an excitation wavelength of 450 nm using an enzyme labeling instrument (Bio-Rad, Hercules, CA, USA). The experiment was performed in triplicate in each group to obtain the mean value. The cell viability curve was then plotted based on the absorbance ratio of the experimental group relative to the blank control group.

Transwell assay for cell invasion

The upper chamber of the transwell filter chamber (40111ES08, Yeasen Company, Shanghai, China) was coated with diluted Matrigel (1 : 8 dilution in pre-cold serum-free DMEM) and incubated at 37° C for 4 - 5 h. The transfected cells were diluted using 100 μL serum-free medium. Cell suspension (1 × 10⁶ cells/mL) was then inoculated into the upper chamber of the transwell filter (in triplicate) and cultured for 24 h at 37° C with 5% CO₂ in air (addition of 500 μL DMEM medium containing 20% FBS to the lower chamber). The invasive cells on the inferior chamber of the filter were fixed with 5% glutaraldehyde at 4° C, stained with 0.1% crystal violet, and observed under an inverted fluorescence microscope (TE2000, Nikon, Tokyo, Japan). Five fields in each transwell filter were selected randomly and photographed. The experiment was performed in triplicate to obtain the mean value of the number of the invasive cells in each group.

Flow cytometry for cell apoptosis

Cells were inoculated into 6-well plates at a density of 1 × 10⁶ cells/well. After a 24 h period post-transfection, the cells were irradiated with X-ray at 4 Gy. Cells were collected 24 h after irradiation and cell apoptosis was detected using Annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI) cell apoptosis kit (Invitrogen, Carlsbad, California, USA) with a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA). The experiment was performed in triplicate in each group to obtain the mean value.

Colony formation assay

A total of 1 × 10³ cells at the logarithmic growth phase were inoculated in 6-well plates supplemented with complete medium and cultured overnight. Cells were then exposed to 0, 2, 4, 6, and 8 Gy X-ray radiations at a dose of 3.5 Gy/min. After incubation at 37° C with 5% CO₂ in air for 14 d, the cells were fixed using methanol for 5 min and stained with crystal violet for 3 min. Colonies encompassing more than 50 cells were counted and documented under a microscope. Survival fraction (SF) = [the number of colonies with more than 50 cells / (the number of inoculated cells × plating efficiency)] × 100%, and the plating efficiency = the number of colonies / the number of inoculated cells were calculated. The experiment was performed in triplicate in each group to obtain the mean value.

Tumor xenografts in nude mice

A total of 24 female BALA/C nude mice (aged 4 weeks; weighing: 18 - 25 g) were obtained from the Animal Experimental Center of Zhengzhou University. The mice were then assigned into the sh-NC and sh-CASC15 groups and raised under specific pathogen-free conditions. Lentivirus containing shRNA against lncRNA CASC15 (LV-sh-CASC15) or scramble shRNA (LV-sh-NC) were purchased from GenePharma Co., Ltd., (Shanghai, China). A total of 1 × 10⁶ EC109 cells soundly infected with lentivirus were subcutaneously injected into the right abdomen of the nude mice. Upon observing the mean volume of transplanted tumors as about 300 mm³, the mice were irradiated once a week using X-ray at 4 Gy. The volume of the transplanted tumors was measured regularly every three days after being subjected to irradiation. The volume of the transplanted tumor = 1/2 × tumor length × tumor width². Twenty-one days after the first irradiation, the nude mice were euthanized by carbon dioxide asphyxia and the tumors were excised for subsequent weighing and IHC analysis.

IHC analysis

Paraffin-embedded tissue slices were dewaxed using the conventional manner and subsequently underwent antigen retrieval under high pressure at 95° C in 0.01 M citric acid buffer (pH = 6.0) for 20 min. The slices were then immersed in 3% H₂O₂ for 15 min to terminate the peroxidase activity. After conducting a blockade using 3% bovine serum albumin (BSA), the slices were incubated with the rabbit anti-human Ki-67 (ab92742, dilution ratio of 1 : 500, Abcam, Cambridge, UK) or rabbit anti-human cleaved Caspase 3 (ab2302, dilution ratio of 1 : 200, Abcam, Cambridge, UK) at 37° C for 2 h. After an additional PBS rinse, the slices were incubated with the HRP-labeled goat anti-rabbit IgG secondary antibody (Abcam, Cambridge, UK) in a wet box for 30 min at 37° C. The slices were then counter-stained with hematoxylin (Shanghai Fusheng Industrial Co., Ltd., Shanghai, China), washed under running water, and mounted with 10% glycerol/PBS. The experiment was performed in triplicate and the results were evaluated and scored by two individuals independently.

Statistical analysis

All experimental data were processed and analyzed using the Statistic Package for Social Science (SPSS) 22.0 statistical software (IBM Corp. Armonk, NY, USA). Measurement data are presented as mean ± standard deviation. Differences between two groups were analyzed by unpaired t test or paired t test. Comparisons among multiple groups were analyzed using one-way analysis of variance (ANOVA), followed by Tukey’s post-hoc test. Tumor volumes between the groups were compared using repeated-measures ANOVA. Pearson correlation analysis was adopted to verify the correlations between the expression of lncRNA CASC15 and miR-140-5p or MMP7 in ESCC tissues, and Kaplan-Meier analysis with the log-rank test was performed to assess the survival time of patients with aberrant lncRNA CASC15 expression. A value of p < 0.05 was considered to be statistically significant.