The Jun/miR-22/HuR regulatory axis contributes to tumourigenesis in colorectal cancer

Abstract

Background: Colorectal cancer (CRC) is a severe health problem worldwide. Clarifying the mechanisms for the deregulation of oncogenes and tumour suppressors in CRC is vital for its diagnosis, treatment, prognosis and prevention. Hu antigen R (HuR), which is highly upregulated in CRC, functions as a pivotal oncogene to promote CRC progression. However, the underlying cause of its dysregulation is poorly understood.

Methods: In CRC tissue sample pairs, HuR protein levels were measured by Western blot and immunohistochemical (IHC) staining, respectively. HuR mRNA levels were also monitored by qRT-PCR. Combining meta-analysis and microRNA (miRNA) target prediction software, we predicted miRNAs that targeted HuR. Pull-down assay, Western blot and luciferase assay were utilized to demonstrate the direct binding of miR-22 on HuR's 3'-UTR. The biological effects of HuR and miR-22 were investigated both in vitro by CCK-8, EdU and Transwell assays and in vivo by a xenograft mice model. JASPAR and SABiosciences were used to predict transcriptional factors that could affect miR-22. Luciferase assay was used to explore the validity of putative Jun binding sites for miR-22 regulation. ChIP assay was performed to test the Jun's occupancy on the C17orf91 promoter.

Results: We observed a significant upregulation of HuR in CRC tissue pairs and confirmed the oncogenic function of HuR both in vitro and in vivo. We found that an important tumour-suppressive miRNA, miR-22, was significantly downregulated in CRC tissues and inversely correlated with HuR in both CRC tissues and CRC cell lines. We demonstrated that miR-22 directly bound to the 3'-UTR of HuR and led to inhibition of HuR protein, which repressed CRC proliferation and migration in vitro and decelerated CRC xenografted tumour growth in vivo. Furthermore, we found that the onco-transcription factor Jun could inhibit the transcription of miR-22.

Conclusions: Our findings highlight the critical roles of the Jun/miR-22/HuR regulatory axis in CRC progression and may provide attractive potential targets for CRC prevention and treatment.

Keywords: Colorectal cancer; HuR; Jun; Migration; Proliferation; miR-22.

Fig. 1

HuR protein but not mRNA is significantly upregulated in CRC tissues. a and b Western blot analysis of HuR levels in CRC tissue pairs. c IHC staining of HuR in CRC tissue pairs. d qRT-PCR analysis of HuR mRNA levels in CRC tissue pairs. e Pearson’s correlation scatter plot of the fold changes of HuR protein and mRNA levels in CRC tissue pairs. ***P < 0.001

Fig. 2

HuR is a potential target gene of miR-22 and miR-129. a A Venn diagram was used to search for potential miRNAs that could target HuR in CRC. b Levels of candidate miRNAs in CRC tissue pairs. c Efficiency and specificity of the HuR probe. d Pull-down assays showed that among the downregulated 13 candidate miRNAs, 7 could bind to HuR mRNA. e and f Schematic descriptions of the hypothetical duplexes formed by miR-22 or miR-129 with the 3’-UTR of HuR. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 3

miR-22 and miR-129 can inhibit HuR by binding to its 3’-UTR. a Western blot analysis of HuR levels in normal colon mucosal epithelial cell line NCM460 and 7 CRC cell lines. b-e qRT-PCR analysis of miR-22 and miR-129 levels and the correlation between miRNA and HuR levels in the aforementioned 8 cell lines. f-i qRT-PCR analysis of miR-22 and miR-129 levels and the correlation between fold changes of miRNA and HuR levels in CRC tissue pairs. j Western blot analysis of HuR levels in 3 CRC cell lines after treatment with miR-22/miR-129 mimic or inhibitor. k Relative luciferase activities in SW480 treated with a miR-22/miR-129 mimic or inhibitor. **P < 0.01; ***P < 0.001

Fig. 4

miR-22 inhibits SW480 proliferation and migration in vitro by targeting HuR. a, c and d miR-22 inhibits SW480 proliferation. a: CCK-8 assay; c and d: EdU assay. b, e and f Recovery experiments indicated that the suppression of SW480 proliferation by miR-22 was due to its inhibitory effect on HuR. b: CCK-8 assay; e and f: EdU assay. g and h Transwell assays revealed that miR-22 could inhibit SW480 migration. g and i Recovery experiments indicated that the suppression of SW480 migration by miR-22 was due to its inhibitory effect on HuR. **P < 0.01; ***P < 0.001

Fig. 5

miR-22 suppresses CRC tumour growth in vivo by targeting HuR. a-c miR-22 slowed down CRC xenografted tumour growth. a: Photos of CRC tumours; b: Tumour volume curves; c: Tumour weights. d qRT-PCR analysis of miR-22 levels in CRC xenografted tumours. e Western blot analysis of HuR levels in CRC xenografted tumours. f and g HE staining and IHC staining for HuR and Ki-67 in xenografted tumours. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 6

miR-22 is inhibited by Jun at the transcriptional level. a Schematic descriptions of the genomic location of miR-22 and Jun’s putative binding sites in the promoter region of miR-22 host gene C17orf91. b-d The influences of Jun on the levels of mature miR-22, pri-miR-22 and C17orf91, respectively. e Luciferase activities of different miR-22 promoter reporter constructs, co-transfected with si-Jun or a negative control. f and g ChIP assay for Jun’s occupancy on the C17orf91 promoter. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 7

The Jun/miR-22/HuR axis exists in SW480 and CRC tissues. a Western blot analysis of HuR levels after altering Jun expression or activity in SW480. b and c Western blot analysis of HuR levels in CRC tissue pairs. d Pearson’s correlation scatter plot of the fold changes of Jun protein and miR-22 levels in CRC tissue pairs. e Pearson’s correlation scatter plot of the fold changes of Jun and HuR protein levels in CRC tissue pairs. ***P < 0.001

Fig. 8

Working model of the Jun/miR-22/HuR axis in CRC