Serine-arginine splicing factor 2 promotes oesophageal cancer progression by regulating alternative splicing of interferon regulatory factor 3

Abstract

Objective: Often, alternative splicing is used by cancer cells to produce or increase proteins that promote growth and survival through alternative splicing. Although RNA-binding proteins are known to regulate alternative splicing events associated with tumorigenesis, their role in oesophageal cancer (EC) has rarely been explored.

Methods: We analysed the expression pattern of several relatively well characterized splicing regulators on 183 samples from TCGA cohort of oesophageal cancer; the effectiveness of the knockdown of SRSF2 was subsequently verified by immunoblotting; we measured the ability of cells treated with lenti-sh-SRSF2/lenti-sh2-SRSF2 to invade through an extracellular matrix coating by transwell invasion assay; using RNA-seq data to identify its potential target genes; we performed qRT-PCR to detect the changes of exon 2 usage in lenti-sh-SRSF2 transduced KYSE30 cells to determine the possible effect of SRSF2 on splicing regulation of IRF3; RNA Electrophoretic mobility shift assay (RNA-EMSA) was performed by the incubation of purified SRSF2 protein and biotinylated RNA probes; we performed luciferase assay to confirm the effect of SRSF2 on IFN1 promoter activity.

Results: We found upregulation of SRSF2 is correlated with the development of EC; Knock-down of SRSF2 inhibits EC cell proliferation, migration, and invasion; SRSF2 regulates the splicing pattern of IRF3 in EC cells; SRSF2 interacts with exon 2 of IRF3 to regulate its exclusion; SRSF2 inhibits the transcription of IFN1 in EC cells.

Conclusion: This study identified a novel regulatory axis involved in EC from the various aspects of splicing regulation.

Keywords: IFN1; IRF3; SRSF2; alternative splicing; oesophageal cancer; proliferation.

Figure 1.

SRSF2 is upregulated in EC tissues. (a), EC patients’ information from TCGA database. (b), Relative U2AF1, SF3B1, hnRNP A1, hnRNP A2 and hnRNP U mRNA level in EC and normal tissues from TCGA database. (c), Survival analysis of U2AF1, SF3B1, hnRNP A1, hnRNP A2 and hnRNP U. (d), Relative SRSF2 mRNA level in EC and normal tissues from TCGA database. (e), Statistical analysis of survival probability of SRSF2 high- (n = 11) and low-expressing (n = 52) EC patients using Kaplan – Meier analysis. Statistical analysis is described in the Materials and Methods. * p < 0.05; ** p < 0.01.

Figure 2.

Knock-down of SRSF2 inhibits EC cell proliferation, migration and invasion. (a), the SRSF2 mRNA levels were detected in seven EC cell lines (TE1, TE3, KYSE30, KYSE510, KYSE150, KYSE450 and Eca109) by Qrt-PCR analysis. (b), the SRSF2 protein levels were detected seven EC cell lines (TE1, TE3, KYSE30, KYSE510, KYSE150, KYSE450 and Eca109) by immunoblotting analysis. (c), SRSF2 protein levels were detected in KYSE30 cells after transduced with lenti-control or lenti-sh-SRSF2 by immunoblotting analysis. (d), Cell proliferation assay of KYSE30 cells after transduced with lenti-control or lenti-sh-SRSF2 using CCK−8. (e), Wound healing assays of KYSE30 cells after transduced with lenti-control or lenti-sh-SRSF2. The relative ratio of wound closure per field was shown in the right. (f), Transwell analysis of KYSE30 cells after transduced with lenti-control or lenti-sh-SRSF2. The relative ratio of invasive cells per field is shown in the right. For all Qrt-PCR results, the data are presented as the mean ± SEM, and the error bars represent the standard deviation obtained from three independent experiments. * p < 0.05; ** p < 0.01.

Figure 3.

SRSF2 promotes the exclusion of exons 2 of IRF3. (a), schematic diagrams of IRF3(full-length IRF3), IRF3a (E2 replaced by i2), IRF3b (only E2 excluded) and IRF3c (including E2 and E3 excluded). IRF3 exons are numbered from 1 to 4 and introns are numbered from 1 to 2. Arrows indicate the positions of specific primer sets for different IRF3 splice variants determined by quantitative RT-PCR analysis. (b), Relative expression of IRF3 isoforms in KYSE30 cells transfected with lenti-control or lenti-sh-SRSF2.(c), Western blotting was used to detect the changes of IRF3 and IRF3a protein levels in cells with low SRSF2 knock-out.(d) Western blotting was used to detect the expression of SRSF2 protein in Eca109 cells transfected with pcmv6 empty vector or pCMV6-SRSF2.(e) Relative expression of IRF3 isoforms in Eca109 cells transfected with pCMV6 empty vector or pCMV6-SRSF2.(f) Western blotting was used to detect the expression of IRF3 and IRF3a protein in Eca109 cells transfected with pCMV6 empty vector or pCMV6-SRSF2. For all quantitative RT-PCR results, the data were expressed as mean ± SEM, and the error bars represented the standard deviation from three independent experiments. * p < 0.05; * * p < 0.01.

Figure 4.

SRSF2 interacts with specific sequences in IRF3 exon 2. (a), Sequences of exon 2 of IRF3. The putative binding sites and mutant sites for SRSF2 are indicated in underlined italics. (b), Purified SRSF2 proteins were used for RNA electrophoretic mobility shift assay with the indicated biotinylated RNA probes or clod probes. (c), Schematic diagram showing the minigenes containing IRF3 exon 1 to 3. The putative binding sites and mutant sites for SRSF2 are indicated in underlined italics. The arrows indicate the location of specific primer sets designed for Qrt-PCR analysis of different IRF3 splicing variants. (d), QRT-PCR analysis of the splicing variants of with RNAs extracted from Eca109 cells after transfected with pCMV6 empty vector or pCMV6-SRSF2. For all Qrt-PCR results, the data are presented as the mean ± SEM, and the error bars represent the standard deviation obtained from three independent experiments. * p < 0.05; ** p < 0.01.

Figure 5.

SRSF2 inhibits IFN1 transcription in EC cells. (a), QRT-PCR analysis of SRSF2 mRNA and IFN1 mRNA in KYSE30 cells after transduced with lenti-control or lenti-sh-SRSF2 (the left panel), or in Eca109 cells after transfected with pCMV6 empty vector or pCMV6-SRSF2 (the right panel). (b), the relative luciferase activities in KYSE30 cells after transfected with shRNA-control or shRNA specific to SRSF2 (the left panel), or in Eca109 cells after transfected with pCMV6 empty vector or pCMV6-SRSF2 (the right panel). For all quantitative results, the data are presented as the mean ± SEM, and the error bars represent the standard deviation obtained from three independent experiments. * p < 0.05; ** p < 0.01.