LncRNA MALAT1 was highly expressed in CRC tissues and cell lines
We compared MALAT1 expression levels between normal and cancer tissues via the Cancer Genome Atlas (TCGA) database. As it showed, MALAT1 was more elevated in tumor tissues compared to normal tissues in CRC (Fig. 1A). We divided CRC patients in TCGA database into MALAT1_low and MALAT1_high groups on the basis of the median MALAT1 expression value. The overall survival was shorter in the MALAT1_high group than that in the MALAT1_low group (Fig. 1B). To investigate the involvement of MALAT1 in CRC progression, we detected MALAT1 expression in both primary tumor and metastatic tumor tissues, and discovered that MALAT1 expression was significantly increased in the metastatic tumors (Fig. 1C). Real-time PCR assay was used to investigate the expression of MALAT1 in resected CRC tissues and corresponding adjacent normal tissues (n = 57). The results showed that MALAT1 expression was higher in CRC tissues comparing with that of non-tumor tissues (Fig. 1D–E). The clinicopathological characteristics of the registered patients were shown in Table 1. Then, we detected the expression of MALAT1 in a normal intestinal epithelial cell line, HIEC, and CRC cell lines including HCT116, SW480, HCT8, HT29, RKO, LOVO, and colo205. The results showed that HCT116 cell possessed the highest expression of MALAT1 (Fig. 1F). Therefore, HCT116 was chosen for subsequent designs to study the role of MALAT1 in CRC radiosensitivity.
Table 1
The statistic of clinicopathological characteristics from 57 CRC patients
Clinical parameters
|
Numbers
|
Cases
|
57
|
Age
|
|
< 65
|
32
|
≥ 65
|
25
|
Tumor size
|
|
Small size (< 5 cm)
|
27
|
Large size (≥ 5 cm)
|
30
|
Gender
|
|
Male
|
37
|
Female
|
20
|
Invasion levels
|
|
Mucosa
|
5
|
Submucosa
|
4
|
Muscle
|
10
|
Serosa
|
37
|
TNM stage
|
|
Stage 1–2
|
36
|
Stage 3–4
|
21
|
Lymph node metastasis
|
|
Positive
|
28
|
Negative
|
29
|
Grade of tumors
|
|
Low grade
|
15
|
Intermediate grade
|
34
|
High grade
|
8
|
Vascular invasion
|
|
Positive
|
24
|
Negative
|
33
|
Perineural invasion
|
|
Positive
|
19
|
Negative
|
38
|
Downregulated expression of MALAT1 inhibited the proliferation of HCT116
Lentivirus-mediated shRNA targeting MALAT1 was used to knock down MALAT1 expression, and puromycin (3 ug/mL) was used to select stably transfected cells. The transfected efficiency was verified by real-time PCR (Fig. 2A). To confirm the short-term influence of MALAT1 silencing-related growth inhibition in human CRC HCT116, stably transfected cells were seeded and incubated for 24 h, 48 h, 72 h, and 96 h, and cell viability was measured by CCK8 assays. As shown in Fig. 2B, MALAT1 silencing inhibited the proliferation of HCT116. A long-term effect of proliferative inhibition was determined by colony formation assays, which implied that MALAT1 silencing remarkable decreased the number of colonies (Fig. 2C–D). These results suggested that MALAT1 silencing inhibited CRC cell proliferation.
MALAT1 silencing inhibited migration in CRC cells
Wound healing migration assays were performed to estimate the effect of MALAT1 on migration. The ability of cell migration was obviously suppressed due to MALAT1 silencing, and the inhibitory effect became more distinct after 4 Gy X-ray treatment (Fig. 3A–B). We tested the expression of metastasis-related proteins through western blotting assays. The results showed that the expression level of matrix metalloproteinase 2 (MMP2) and Vimentin were declined and E-cadherin was increased in HCT116 cells after transfecting with shRNA target MALAT1 (Fig. 3C), and such changes were more distinct in combination with irradiation. Taken together, we surmised that the downregulation of MALAT1 in HCT116 cells could significantly impair the migration of HCT116 cells.
MALAT1 silencing enhanced the radiosensitivity of HCT116
The plate clone formation assay was considered as the gold standard in evaluating radiosensitivity. Cells stably transfected with shRNA targeted MALAT1 and NC were seeded into 6-well plates in different densities, then exposed to 0, 2, 4, 6, and 8 Gy X-rays respectively and incubated for another 10 days. A multitarget single-hit model was used to explore the effect of MALAT1 silencing on radiosensitivity of HCT116 cells. After MALAT1 silencing, the radiosensitivity of HCT116 was improved as shown in Fig. 4A. D0 values were used to appraise the radiosensitivity, and higher D0 values indicated lower sensitivity to radiation. The results showed that MALAT1 silencing could remarkably decline the levels of colony formation after irradiation, and the downregulation of MALAT1 (D0 sh1 = 0.8122, D0 sh3 = 0.8010) significantly reduced the colony survival fraction in a dose-dependent manner, compared with the NC group (D0 = 0.8882) (Fig. 4B and Table 2). According to the parameters of radiosensitization showed in Table 2, downregulating MALAT1 in HCT116 exhibited higher sensitivity enhancement ratios (SER) (SER sh1 = 1.0936, SER sh3 = 1.1089). Thus, the results showed that silencing MALAT1 could increase the radiosensitivity of CRC cells.
Knockdown of MALAT1 induced cell cycle arrest
In addition to the results from prior experiments, we used TCGA data to conduct a generalized analysis of the G2/M checkpoint and found that the G2/M checkpoint was significantly activated in CRC (Fig. 4C). Although there was no correlation between the G2/M checkpoint and MALAT1 (Fig. 4D–E), the following results confirm that downregulation of MALAT1 combined with 4 Gy X-ray irradiation could lead to more G2/M phase arrest in HCT116 cells (Fig. 4F–G). These data indicated that MALAT1 silencing combined with X-rays could increase the radiosensitivity of CRC cells via activating the G2/M checkpoint.
Table 2
The D0, Dq, N and SER values in NC, sh-MALAT1 and sh-MALAT3 transfected HCT116 cells. The SER value was artificial using the multi-target single hit model.
Group
|
D0
|
Dq
|
N
|
SER
|
NC
|
0.882
|
1.727
|
6.991
|
|
Sh-MALAT1-1
|
0.8122
|
1.0412
|
3.604
|
1.0936
|
Sh-MALAT1-3
|
0.801
|
1.1862
|
4.397
|
1.1089
|
MALAT1 silencing induced DNA double-strand breaks in CRC cells
Thereafter, a generalized analysis of the DNA_REPAIR found that DNA_REPAIR was significantly activated in CRC (Fig. 5A–B). In addition, there was a significant positive correlation between DNA_REPAIR and MALAT1 expression (Fig. 5C). To examine whether MALAT1 silencing improved the formation of nuclear γ-H2AX foci induced by irradiation at 0, 0.5, 4, and 12 h, western blot analysis was conducted (Fig. 5D). The results confirmed that MALAT1 silencing could significantly increase the formation of γ-H2AX foci after 4 Gy X-ray irradiation, especially at 0.5 h, which was further confirmed by immunofluorescence assay (Fig. 5E–F). These results suggested that MALAT1 silencing could induce higher expression of γ-H2AX and more γ-H2AX foci formation, regardless of whether it is combined with irradiation.
MALAT1 silencing induced apoptosis in CRC cells
Severe DNA damage may result in apoptosis, so we further investigated the relationship between apoptosis and the expression of MALAT1. As shown in Fig. 6A–B, apoptosis was significantly inhibited in CRC. Moreover, the expression level of MALAT1 was negatively relevant to apoptosis (Fig. 6C). Then, we verified the impact of MALAT1 on CRC cells using flow cytometry analysis. The results demonstrated that MALAT1 silencing increased apoptosis, and it was more obvious after combining it with radiotherapy (Fig. 6D). As showed in Fig. 6E, the apoptotic ratios of shMALAT1-1 (6.77 ± 0.31%) and shMALAT1-3 (6.71 ± 0.26%) were significantly higher than that of shNC (3.33 ± 0.17%). After 4 Gy X-ray irradiation, the apoptotic ratios came to 20.15 ± 1.38% in shMALAT1-1, 24.16 ± 1.85% in shMALAT1-3 and 6.32 ± 0.63% in shNC. However, we detected the expression of apoptosis-related proteins. The expression of BAX was increased, and Bcl-2 was decreased after cells were treated with shRNA targeted MALAT1 compared to NC, especially when it was combined with radiotherapy (Fig. 6F). All of these results suggested that MALAT1 silencing could increase apoptosis in CRC cells.
Preliminary proteomic analysis indicated the underlying mechanisms of MALAT1 in CRC
To investigate the difference of protein levels between HCT116 siMALAT1 and HCT116 siNC, we extracted total proteins of stably transfected cells and then conducted iTRAQ proteomic analysis. The amounts of the total identified and quantified proteins as well as differentially expressed proteins are separately summarized in Supplemental Table S2-3. Twenty-seven proteins (quantitative ratio over 1.3) were found to be upregulated, whereas 46 proteins (quantitative ratio less than 0.77) were downregulated. Volcano plot displayed the top 10 up- and downregulated proteins in HCT116 siMALAT1 compared with HCT116 siNC (Fig. 7A and Table S4-5); the green ball represented downregulated proteins and the red ball represented upregulated ones. To further analyze their functions, the differentially-abundant proteins (DAPs) were categorized into the major Gene Ontology (GO) categories—biological process (BP), cellular component (CC), and molecular function (MF)—based on their GO annotations. The mainly enriched proteins in BP were related with the following terms: protein maturation by protein folding, regulation of protein depolymerization, regulation of protein complex disassembly, ensheathment of neurons, and protein maturation (Fig. 7B). In the CC category, the predominantly enriched terms were elongin complex, Flemming body, F-actin capping protein complex, dynactin complex, and microtubule associated complex (Fig. 7C). The DAPs in the MF category were connected with the following terms: complement component C1q binding, protein complex binding, macromolecular complex binding, complement binding, and opsonin binding (Fig. 7D). The detailed results of GO enrichment analysis are given in Supplemental Table S6-8. To further investigate the discrepancy between the cellular pathways of HCT116 siMALAT1 and HCT116 siNC, KEGG enrichment analysis was performed via DAVID. The results indicated that the whole DAPs mapped to 34 KEGG pathways. The pathways of the DAPs based on the KEGG database are shown in Fig. 7E. The maximal number of DAPs were enriched in the glycosaminoglycan degradation, human immunodeficiency virus 1 infection, Ubiquitin-mediated proteolysis, glycosphingolipid biosynthesis-globo and isoglobo series, and lysosome pathways, detailed results of which can be found in Supplemental Table S9. These signaling pathways may play an crucial role in MALAT1 silencing HCT116, and more research ought to be carried out on them.