A novel long noncoding RNA AC125257.1 facilitates colorectal cancer progression by targeting miR-133a-3p/CASC5 axis

2.1 Patients and clinical specimens

CRC tissues (n = 36) and adjacent normal tissues (n = 36) were obtained from CRC patients receiving surgery at The First Affiliated Hospital of Nanchang Medical College. Adjacent normal tissues were collected 5 cm away from the cancerous tissues and confirmed by pathological examination. All participants signed the informed consent prior to the study. The information of patients and the correlation between AC125257.1 expression and clinicopathologic characteristics of CRC patients are presented in Table 1.

1. Ethics approval and consent to participate: The experiments were approved by the Ethics Committee of The First Affiliated Hospital of Nanchang Medical College. The contents of this study are under full compliance with government policy and the Declaration of Helsinki.

2.2 Bioinformatics analysis

Interactive Analysis of Gene Expression Profiles, abbreviated as GEPIA, was used for RNA expression analysis (http://gepia.cancer-pku.cn/). Meanwhile, it provides customizable options, such as analysis of differential expression between tumor and normal patients, similar gene detection, and survival analysis [22]. The normal/tumor differential analysis and the prognostic value of AC125257.1 referred in our paper were all from GEPIA. Potential miRNAs targeting CASC5 (also named KNL1) were searched on the starBase database (https://starbase.sysu.edu.cn/) under the condition of CLIP-Data ≥ 5, Degradome-Data ≥ 0, pan-Cancer ≥ 10, and programNum ≥ 2. Only miR-133a-3p was screened out and subject to further analysis. lncRNAs possibly bound to miR-133a-3p was predicted on the starBase website under the condition of Degradome-Data ≥ 1.

2.3 Cell culture and transfection

Human colonic epithelial cells (HCoEpiCs, BFN608006386) and four other CRC cell lines from BLUEFBIO (Shanghai, China), including SW480 (BFN60800644), HCT116 (BFN607200665), Caco-2 (BFN60800651), and HCT-8 (BFN60800647), were cultured in flasks coated with Dulbecco’s modified eagle medium [10% (v/v) fetal bovine serum + 1% penicillin–streptomycin solution, Hyclone]. The culture environment was set at 37°C with 5% CO₂. For cell transfection, sh-AC125257.1#1/2, sh-NC, miR-133a-3p mimics, NC mimics, and pcDNA3.1/CASC5, were designed by Ribobio Technology (Guangzhou, China). HCT-8 and SW480 were transfected with these vectors using Lipofectamine 2000 (Invitrogen, USA), complied with the product’s protocols.

2.4 Reverse-transcription quantitative polymerase chain reaction (RT-qPCR)

RNAs in CRC cells or tumor tissue were isolated using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA), complied to the product’s protocols. RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) was used for reverse transcription. Applied Biosystems™ 7500 Fast Dx Real-Time PCR Instrument was used for qPCR process. The internal controls were U6 and GAPDH. The temperatures for denaturation, annealing, and elongation were set to 95, 60, and 72°C with 40 cycles. The quantitative calculation was performed using 2^−ΔΔCT method. Promega (Madison, WI, USA) synthesized all these primer sequences and the primers are shown in Table 2.

2.5 RNA pull-down assay

A total of 1 × 10⁷ cells transfected with biotinylated miR-133a-3p mimic (bio-miR-133a-3p) or bio-NC were lysed. Magnetic RNA-Protein Pull-Down Kit was used for cultivation with magnetic beads. GAPDH was used as a negative control. CASC5 or AC125257.1 expression in precipitated RNA complex was measured by qPCR.

2.6 RNA immunoprecipitation (RIP)

The RIP assay was performed using EZ-Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (17-701; Millipore), complied with the product’s protocols. A total of 1 × 10⁷ HCT-8 and SW480 cells were lysed and centrifuged at 12,000g for 10 min. Supernatants were then co-incubated with magnetic beads containing anti-IgG (ab172730, 1:100; Abcam) or anti-Ago2 (ab186733, 1:50; Abcam) and then washed with 1,000 μL of RIP wash buffer. The isolated RNA was analyzed by PCR.

2.7 Cell counting kit-8 (CCK-8) assay

HCT-8 and SW480 were cultivated in 96-well plates (4 × 10³ cells/well). The corresponding cultivation time was 12, 24, and 48 h at 37°C. The cells were mixed with 10 μL of CCK-8 solution (Dojindo) at 37°C. After 2.5 h, the optical density (OD) value at 450 nm was detected using microplate reader.

2.8 Flow cytometry

The flow cytometry was performed using the Annexin V-FITC/propidium iodide (PI) Apoptosis Detection Kit (Yeasen, Shanghai, China). Cold phosphate-buffered saline (PBS) (Solarbio, Beijing, China) was used for washing. After washing twice with cold PBS buffer, the CRC cells were centrifuged at 4°C and 5 min later and made into 1 × 10⁶ cells/mL suspension with 100 L of 1× binding buffer. After 15 min of complete reaction with 10 μL of PI solution and 5 μL of Annexin V-FITC, the mixture was placed on ice for detection by flow cytometry (BD Biosciences, USA).

2.9 Western blot

The protein of CRC cells was extracted and quantitatively analyzed by the bicinchonininc acid method (Solarbio, Beijing, China). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (10%) was used for separation. After being transferred to polyvinylidene fluoride membrane (Micropore Company, USA) and sealed with 5% nonfat milk, the mixture was blocked with the tris buffered saline with Tween-20 buffer. After washing with Tween buffer thrice, primary antibodies were added for incubation at 4°C overnight. Then, it was mixed with goat anti-rabbit IgG secondary antibody (1:1,000, ab7090; Abcam, UK) at room temperature. Two hours later, enhanced chemiluminescence reagents (Bio-Rad, Hercules, CA, USA) were used for protein level detection. Primary antibodies were as follows: anti-GAPDH (1:2,500, ab9485), anti-CASC5 (1:1,000, ab70537)), anti-Bcl-2 (1:1,000, ab32124), anti-Bax (1:2,000, ab32503), anti-CDK4 (1:1,000, ab108357), anti-Cyclin A1 (1:1,000, ab270940), and anti-Cyclin D1 (1:1,000, ab16663).

2.10 Animal experiments

BALB/c male nude mice (5 weeks, 20 ± 2 g) were provided by the Vital River (Beijing, China). Mice were randomly divided into the sh-NC and sh-AC125257.1 groups. The cell suspension of HCT116 cells transfected with sh-NC or sh-AC125257.1 (1 × 10⁶ cells/mL) was subcutaneously injected into right axilla of each mouse. Tumor volume growth was monitored weekly by the following formula: volume = width² × length/2. Mice were sacrificed at day 28, and tumors were obtained and weighed. The animal study has been approved by the Animal Care and Use Committee of The First Affiliated Hospital of Nanchang Medical College.

2.11 Statistics analysis

Data were analyzed using SPSS 19.0 and GraphPad 7.0 and expressed as the mean ± standard deviation. One-way analysis of variance was taken to compare means between three or more groups followed by Dunnett’s post hoc test. Student’s t-test was taken to compare means between two groups. Statistical significance was represented as P < 0.05.

3.1 CASC5 was significantly expressed in CRC tissues and cells

From GEPIA database, high expression of CASC5 was found in CRC patients (either with COAD or with rectum adenocarcinoma) (Figure 1a). The overall survival curve showed that patients with high CASC5 level had worse prognosis than those with low level (Figure 1b). RT-qPCR results indicated that the expression of CASC5 was higher in tumor than in adjacent parts (Figure 1c) and higher in CRC cells than in HCoEpiCs (Figure 1d). These results revealed that CASC5 associated with CRC and has bad prognosis.

Figure 1

(a) The distribution of CASC5 in COAD (275 tumor samples and 349 normal samples) and rectum adenocarcinoma (READ, 92 tumor samples and 318 normal samples). Red column represented tumor samples, and gray column represented normal samples. (b) The overall survival curve of patients with low and high CASC5 TPM based on GEPIA database using Log-rank test. (c) Relative CASC5 expression in adjacent and tumor. (d) Relative CASC5 expression in HCoEpiC and four other CRC cell lines. *P < 0.05, **P < 0.01, ***P < 0.001.

3.2 miR-133a-3p suppressed CASC5 expression through sponging

miRNAs with potential binding site to CASC5 were predicted on the starBase database under the condition of CLIP-Data ≥ 5, Degradome-Data ≥ 0, pan-Cancer ≥ 10, and programNum ≥ 2. The results showed that only miR-133a-3p was screened out. RT-qPCR results showed lowly expressed miR-133a-3p in CRC cells (Figure 2a) and tumor samples (Figure 2b). In SW480 and HCT-8 with miR-133a-3p overexpression, the level of CASC5 expression was significantly downregulated (Figure 2c and d). Western blot results further confirmed this finding (Figure 2e). By using a bio-miR-133a-3p probe, RNA pull-down assay showed that CASC5 has high enrichment in the bio-miR-133a-3p group, demonstrating that CASC5 is targeted by miR-133a-3p (Figure 2f). Moreover, the expression of miR-133a-3p was demonstrated to be negatively correlated with CASC5 expression in CRC patient tissues (Figure 2g).

Figure 2

(a) Relative miR-133a-3p expression in HCoEpiC and four other CRC cell lines. (b) Relative expression of miR-133a-3p in adjacent and tumor. (c) Relative miR-133a-3p expression in SW480 and HCT-8 cell lines transfected with over-miR-133a-3p/miR-NC. (d) Relative CASC5 expression in SW480 and HCT-8 cell lines transfected with over-miR-133a-3p/miR-NC. (e) Western blot results of CASC5 in SW480 and HCT-8 cell lines transfected with over-miR-133a-3p/miR-NC. (f) RNA pull-down results were used to detect the enrichment of CASC5 and GAPDH (negative control) in the complex pulled down by bio-miR-133a-3p or bio-NC. (g) The expression correlation between miR-133a-3p and CASC5 in CRC patient tissues was explored using Pearson correlation analysis. **P < 0.01, ***P < 0.001.

3.3 miR-133a-3p reduced CRC cell proliferation and promoted apoptosis while CASC5 partially reversed these effects

By CCK-8 assays, SW480 and HCT-8 transfected with over-miR-133a-3p showed suppressed proliferation, while those transfected with over-miR-133a-3p + CASC5 partially reserved this effect (Figure 3a). As is shown in flow cytometry results, the apoptosis rate of SW480 and HCT-8 transfected with over-miR-133a-3p was remarkably increased, while group transfected with over-miR-133a-3p + CASC5 showed reversed effect (Figure 3b). Further western blot analysis clarified the protein changes (Figure 3c). To be specific, CASC5, BCL2, and CDK4 levels were decreased by overexpressed miR-133a-3p, and BAX level was increased by miR-133a-3p, suggesting the promotive effects of miR-133a-3p on CRC cell apoptosis, further proving the results in Figure 3b. The levels of Cyclin A1 and Cyclin D1 were downregulated, suggesting that overexpressed miR-133a-3p suppressed CRC progression. These effects could all be restored partially by CASC5.

Figure 3

(a) OD value of 450 nm in SW480 and HCT-8 at 0, 12, 24, and 48 h. (b) Flow cytometry results and apoptosis rate of SW480- and HCT-8 transfected with miR-NC, over-miR-133a-3p, and over-miR-133a-3p + CASC5. (c) Western blot results of CASC5, Bcl-2, Bax, CDK4, Cyclin A1, Cyclin D1, and GAPDH in SW480 and HCT-8 transfected with miR-NC, over-miR-133a-3p, and over-miR-133a-3p + CASC5. **P < 0.01, ***P < 0.001 compared with the miR-NC group; ^### P < 0.001 compared with the over-miR-133a-3p group.

3.4 Knockdown of AC125257.1 promoted the expression of miR-133a-3p and suppressed that of CASC5

lncRNAs possibly bound to miR-133a-3p was predicted on the starBase website under the condition of Degradome-Data ≥ 1. Only AC125257.1 was selected under this condition. AC125257.1 showed abundant expression in CRC cells and tumor by RT-qPCR (Figure 4a and b). RNA pull-down assay proved that AC125257.1 was remarkably enriched by the bio-miR-133a-3p probe, indicating that miR-133a-3p is the target miRNA of AC125257.1 (Figure 4c). As shown in Figure 4d–f, the knockdown of AC125257.1 significantly elevated miR-133a-3p expression and suppressed CASC5 expression, which is consistent with the finding in Figure 2c and d. Western blot results further confirmed this finding (Figure 4g). Furthermore, RIP assay validated that compared to the control group (anti-IgG), AC125257.1 and miR-133a-3p were noticeably enriched in microribonucleoproteins with Ago2, suggesting that AC125257.1 acted as the ceRNA for miR-133a-3p and formed an RNA-induced silencing complex in CRC cells (Figure 4h). The expression of CASC5 and AC125257.1 showed positive correlation in CRC patient tissues, while that of miR-133a-3p and AC125257.1 exhibited negative correlation according to Pearson correlation analysis (Figure 4i–g).

Figure 4

(a) Relative AC125257.1 expression in HCoEpiC and four other CRC cell lines. (b) Relative AC125257.1 expression in adjacent and tumor. (c) RNA pull-down results of AC125257.1 using bio-miR-133a-3p or bio-NC. (d) Relative AC125257.1 expression in SW480 and HCT-8 cell lines transfected with sh-NC/sh-AC125257.1. (e) Relative miR-133a-3p expression in SW480 and HCT-8 cell lines transfected with sh-NC/sh-AC125257.1. (f) Relative CASC5 expression in SW480 and HCT-8 cell lines transfected with sh-NC/sh-AC125257.1. (g) Western blot results of CASC5 and GAPDH in SW480 and HCT-8 cell lines transfected with sh-NC/sh-AC125257.1. (h) AC125257.1 and miR-133a-3p levels detected by RIP. (i) Pearson correlation analysis was conducted to examine the expression correlation between CASC5 and AC125257.1. (j) The expression correlation between AC125257.1 and miR-133a-3p in CRC patient tissues. *P < 0.05, **P < 0.01, ***P < 0.001.

3.5 Knockdown of AC125257.1 reduced CRC cell proliferation and promoted apoptosis while CASC5 partially reversed these effects

By CCK-8 assay, SW480 and HCT-8 transfected with sh-AC125257.1 showed suppressed viability, while those transfected with sh-AC125257.1-1/-2 + CASC5 partially reserved this effect (Figure 5a). As is shown in flow cytometry results, the apoptosis rate of SW480 and HCT-8 transfected with sh-AC125257.1 was remarkably increased, while group transfected with sh-AC125257.1-1/-2 + CASC5 showed reversed effect (Figure 5b). Further western blot analysis clarified the protein changes (Figure 5c). To be specific, CASC5, BCL2, and CDK4 levels were decreased by knockdown of AC125257.1, and BAX level was increased by it, suggesting that CRC cell apoptosis was inhibited by AC125257.1. The expression of Cyclin A1 and Cyclin D1 was downregulated, suggesting that the knockdown of AC125257.1 suppressed CRC progression. These effects could all be reversed partially by CASC5.

Figure 5

(a) OD value of 450 nm in SW480- and HCT-8-transfected with sh-NC, sh-AC125257.1, and sh-AC125257.1 + CACS5 at 0, 12, 24, and 48 h. (b) Flow cytometry results and apoptosis rate of SW480 and HCT-8 transfected with sh-NC, sh-AC125257.1, and sh-AC125257.1 + CACS5. (c) WB results of CASC5, Bcl-2, Bax, CDK4, Cyclin A1, Cyclin D1, and GAPDH in SW480- and HCT-8-transfected with sh-NC, sh-AC125257.1, and sh-AC125257.1 + CACS5. **P < 0.01, ***P < 0.001 compared with the sh-NC group; ^# P < 0.05, ^## P < 0.01 compared with the sh-AC125257.1-1/-2 group.

3.6 Knockdown of AC125257.1 suppressed CRC tumor growth in vivo

Xenograft tumor-bearing mouse models were established to examine the function of AC125257.1 on CRC tumorigenesis. The results indicated that AC125257.1 knockdown significantly decreased the tumor growth rate, and mouse tumor volume and weight was significantly smaller in the sh-AC125257.1-1 group compared with the control group (Figure 6a and b). Moreover, we examined the expression of miR-133a-3p and CASC5 in mouse tumor tissues. As revealed by RT-qPCR analysis, the expression of miR-133a-3p was significantly increased while the CASC5 levels showed a significant decrease in the sh-AC125257.1-1 group compared with the control group (Figure 6c and d).

Figure 6

(a) Tumor volume in each group was measured on days 7, 14, 21, and 28. (b) Mouse tumor weight in the sh-NC and sh-AC125257.1-1 groups. (c) RT-qPCR was used to detect the expression of miR-133a-3p in mouse tumor tissues in each group. (d) The expression of CASC5 was measured using RT-qPCR mouse tumor tissues in each group. **P < 0.01, ***P < 0.001.