Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Feb 21;43(1):52.
doi: 10.1186/s13046-024-02971-7.

A tumor suppressor protein encoded by circKEAP1 inhibits osteosarcoma cell stemness and metastasis by promoting vimentin proteasome degradation and activating anti-tumor immunity

Ying Zhang #  1   2 , Zhaoyong Liu #  3   4 , Zhigang Zhong #  4   5 , Yanchen Ji  6 , Huancheng Guo  3 , Weidong Wang  7 , Chuangzhen Chen  6
Affiliations

A tumor suppressor protein encoded by circKEAP1 inhibits osteosarcoma cell stemness and metastasis by promoting vimentin proteasome degradation and activating anti-tumor immunity

Ying Zhang et al. J Exp Clin Cancer Res. .

Abstract

Background: Osteosarcoma (OS) is one of most commonly diagnosed bone cancer. Circular RNAs (circRNAs) are a class of highly stable non-coding RNA, the majority of which have not been characterized functionally. The underlying function and molecular mechanisms of circRNAs in OS have not been fully demonstrated.

Method: Microarray analysis was performed to identify circRNAs that are differentially-expressed between OS and corresponding normal tissues. The biological function of circKEAP1 was confirmed in vitro and in vivo. Mass spectrometry and western blot assays were used to identify the circKEAP1-encoded protein KEAP1-259aa. The molecular mechanism of circKEAP1 was investigated by RNA sequencing and RNA immunoprecipitation analyses.

Results: Here, we identified a tumor suppressor circKEAP1, originating from the back-splicing of exon2 of the KEAP1 gene. Clinically, circKEAP1 is downregulated in OS tumors and associated with better survival in cancer patients. N6-methyladenosine (m6A) at a specific adenosine leads to low expression of circKEAP1. Further analysis revealed that circKEAP1 contained a 777 nt long ORF and encoded a truncated protein KEAP1-259aa that reduces cell proliferation, invasion and tumorsphere formation of OS cells. Mechanistically, KEAP1-259aa bound to vimentin in the cytoplasm to promote vimentin proteasome degradation by interacting with the E3 ligase ARIH1. Moreover, circKEAP1 interacted with RIG-I to activate anti-tumor immunity via the IFN-γ pathway.

Conclusion: Taken together, our findings characterize a tumor suppressor circKEAP1 as a key tumor suppressor regulating of OS cell stemness, proliferation and migration, providing potential therapeutic targets for treatment of OS.

Keywords: ARIH1; Osteosarcoma; Vimentin; circKEAP1.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Fig. 1
Fig. 1
Validation and expression of circKEAP1 in OS tissues. (A). A circRNA microarray was used for 3 pairs of OS tissues and case-matched normal tissues. Circular heatmap showing the differentally-expressed circRNAs. (B). Expression of circKEAP1 was detected by qRT-PCR in OS (n = 26) and normal tissues (n = 11) from our own tissue bank. (C). The expression of circKEAP1 in normal (n = 3) and OS tissues (n = 3) was also found in GSE140256 from GEO database. (D). CircKEAP1 expression was detected by qRT-PCR in various human OS cell lines (HOS, Saos-2, MG63 and U2OS) and osteoblast cells (hFOB1.19). **, *** indicates significant differences compared with the hFOB1.19 at a p value < 0.01, < 0.001, respectively. (E). Schematic illustration showing the formation and head-to-tail splicing site of circKEAP1 via the circularization of exon 2 in KEAP1. The presence of circKEAP1 was validated by Sanger sequencing. (F). RT-PCR validated the existence of circKEAP1 in HOS and U2OS cell lines. CircKEAP1 was amplified by divergent and convergent primers in complementary DNA (cDNA) and genomic DNA (gDNA). (G). The expression of circKEAP1 and KEAP1 mRNA in both HOS and U2OS cell lines was detected by RT-PCR and qRT-PCR with/without RNase R (20 U/µl) treatment. (H). Expression of KEAP1 and circKEAP1 was detected in the cytoplasm and nuclei of OS cells. (I). FISH images demonstrating circKEAP1is mainly located in the cytoplasm of OS cells (scale bars, 20 μm). (J). Expression of circKEAP1 was detected by ISH in OS tissues (n = 72) and normal tissues (n = 6) (scale bars, 100 μm, 50 μm). (K). Overall survival analysis based on circKEAP1 expression in OS. (L). ROC curve was shown based on the expression of circKEAP1 in OS tissues. Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the normal group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 2
Fig. 2
Tumor suppressive functions of circKEAP1 in vivo and in vitro. (A). OS cells were transfected with circKEAP1, two circKEAP1 shRNAs or control vectors. Colony formation of OS cells was determined in the indicated transfection groups. (B). Caspase 3 and PARP levels were determined by western blotting in the indicated transfection groups. (C). Apoptosis of the indicated OS cells was detected by flow cytometry. (D). Cell migration of the indicated OS cells was determined by wound healing assay. (E). Cell migration of indicated OS cells was determined by transwell assay. (F). The indicated U2OS cells (NC and circKEAP1) were injected into nude mice via tail vein and the lungs were harvested after 35 days. Representative image and H&E staining of mouse lung metastasis is shown (scale bars, 200 μm). (G). The indicated U2OS cells (NC and circKEAP1) were injected subcutaneously into nude mice (n = 5 per group) and tumor volume was measured every 7 days. After 28 days, tumors were collected and weighed. (H). Tumor tissues stained with Ki67 antibody and IHC score are shown (scale bars, 50 μm). Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the control group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 3
Fig. 3
CircKEAP1 encodes a novel 259-amino acid protein denoted KEAP1-259aa. (A). Putative ORF in circKEAP1 and the sequences of the putative ORF are shown. (B). Putative IRES activity in circKEAP1 was tested by dual luciferase activity assay. (C). U2OS cells were transfected with control vector, circKEAP1 ATG-mutated vector, circKEAP1 wild-type vector and linear KEAP1-259aa vector. CircKEAP1 expression was detected by qRT-PCR. (D). Total proteins from circKEAP1 or control plasmid-transfected U2OS cells were separated via SDS-PAGE. The differential gel bands between 25 kDa and 35 kDa were cut and subjected to LC-MS/MS. The identified KEAP1-259aa amino acids are shown. (E). KEAP1 and KEAP1-259aa expression were detected in eight paired normal and case-matched OS samples. (F). MG63 and U2OS cells were transfected with control or two circKEAP1 shRNAs. KEAP1 and KEAP1-259aa levels were characterized by western blotting. (G). KEAP1 and KEAP1-259aa expression were detected in control-, circKEAP1-, circKEAP1 IRSE mut-, and circKEAP1 ATG mut-transfected HOS and U2OS cells by western blotting. (H). Flag-tagged KEAP1-259aa was transfected into HOS and U2OS cells. Immunofluorescence staining using anti-Flag was performed to show KEAP1-259aa cellular localization (scale bars, 20 μm). Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the empty vector group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 4
Fig. 4
KEAP1-259aa exerts the biological function in OS cells. (A). OS cells were transfected with empty vector, circKEAP1 vector, ATG-mutated circKEAP1 vector and IRES-mutated circKEAP1 vector. Colony formation was performed in the indicated transfection groups. (B). Cell proliferation of the indicated cells was characterized by CCK-8 assay. (C). Apoptosis of the indicated cells was detected by flow cytometry. (D). Cell apoptosis of the indicated cells was characterized by western blotting assay. (E). Cell migration of the indicated cells was characterized by wound healing assay. (F). Cell migration of the indicated cells was characterized by transwell assay. (G). Tumor sphere assay was performed to detect stemness of the indicated cells (scale bar, 50 μm). (H). Flow cytometry assay was used to examine percentage of CD133 + cells in the indicated cells. (I). SOX2, ABCG2, Oct4 and Nanog expression in cells was characterized by western blotting. (J). The indicated cells (1 = NC, 2 = KEAP1-259aa, 3 = IRSE mut, 4 = ATG mut) were injected subcutaneously into nude mice (n = 4 per group) and tumor volume was measured every 7 days. After 28 days, tumors were collected and weighed. (K). The indicated cells were injected into nude mice (n = 4 per group) via tail vein and the lungs were harvested after 35 days. Representative image and H&E staining of mouse lung metastases is shown (scale bars, 200 μm). Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the control group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 5
Fig. 5
KEAP1-259aa interacts with ARIH1 to promote vimentin degradation. (A). Total immunoprecipitated proteins from Flag-KEAP1-259aa cells were separated via SDS-PAGE. Vimentin and ARIH1 were identified by LC/LC-MS. (B). Interaction of GFP-vimentin, Myc-ARIH1 and Flag-KEAP1-259aa were determined by co-IP assay. (C). Flag-tagged KEAP1-259aa and GFP-vimentin were co-transfected into MG63 cells and immunofluorescence staining was performed using anti-Flag and anti-GFP antibodies (scale bar, 20 μm). (D). Expression of vimentin in OS cells transfected with the indicated plasmid or siRNAs following the treatment with CHX (50 µg/ml). (E). Expression of vimentin was detected in OS cells transfected with the indicated plasmid or siRNAs following the treatment with MG132 (10 µM). (F). Ubiquitin of vimentin, in MG132 (10 µM)-treated cells, was detected by IP using GFP antibodies in the indicated cells followed by immunoblotting using HA antibodies. (G). Wild-type HA-Ub, K11R-, K48R- or K63R-mutant HA-Ub plasmids were transfected into OS cells. Wild-type HA-Ub, K11 HA-Ub plasmids were transfected into OS cells. Ubiquitin of vimentin was detected by IP assay in the treatment of MG132 (10 µM). (H). Wild-type GFP-vimentin, or vimentin S39A, S56A or S73A mutant plasmids were transfected into OS cells. The binding of vimentin and KEAP1-259aa was detected by co-IP assay. (I). Wild-type GFP-Vimentin, or vimentin S39A, S56A or S73A mutant plasmids were transfected into OS cells. Ubiquitin of vimentin was detected by IP assay in the treatment of MG132 (10 µM). Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the control group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 6
Fig. 6
CircKEAP1 suppresses OS malignancy through decreasing vimentin levels. (A). OS cells were transfected with circKEAP1 shRNA and control vectors followed by treatment with WFA (5 µM). Control vectors, vimentin or both circKEAP1 and vimentin were transfected into OS cells. Caspase-3 and PARP expression were determined by western blotting. (B). Proliferation of the indicated OS cells was measured by CCK8 assay. (C). Apoptosis of the indicated OS cells was detected by flow cytometry. (D). Migration of the indicated OS cells was quantified by transwell assay. (E). Migration of the indicated OS cells was quantified by wound healing assay. (F). Colony formation ability of OS cells was determined. (G). The indicated cells were injected subcutaneously into nude mice (n = 6 per group) and tumor volume was measured every 7 days. After 28 days, tumors were collected and weighed. Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the control group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 7
Fig. 7
CircKEAP1 expression is modulated by m6A methylation. (A). MeRIP assays for m6A-modified circKEAP1 in HOS and U2OS cell lines, as well as cells over-expressing circKEAP1 with or without the m6A site mutated. (B). RNA pulldown assay was performed using a circKEAP1 or empty control probe and YTHDF2, METTL3, and FTO were detected by western blotting. (C). RIP assay and PCR analysis for circKEAP1 in U2OS cells using FTO, METTL3, YTHDF1/2 antibodies. (D). Expression of circKEAP1 in HOS and U2OS cells transfected with the indicated siRNAs under actinomycin D (5 µg/ml) treatment for 0, 6 and 12 h by qPCR. (E). Expression of KEAP1 was detected by western blotting in HOS and U2OS cells transfected with the indicated siRNAs. (F). QPCR analysis of circKEAP1 in HOS and U2OS cells transfected with control, si-METTL3 or si-FTO. (G). MeRIP assays for m6A-modified circKEAP1 in U2OS cells transfected with control, si-METTL3 or si-FTO. Error bars represent three independent experiments. *, **, *** indicates significant differences compared with 0 h or control group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 8
Fig. 8
CircKEAP1 provokes cellular immune responses via RIG-I. (A). RNA sequencing was conducted in control and circKEAP1 overexpression cells. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) pathway analyses were performed to find the most differentially-changed signaling pathways. (B). GSEA analysis for circKEAP1 compared to the control group. (C). Fold change of the indicated mRNAs in circKEAP1 knockdown and control OS cells was measured by qRT-PCR. (D). Expression levels of the indicated mRNAs in cells following RIG-I knockdown and/or circKEAP1-overexpression in U2OS and HOS cell lines. (E). Protein levels in HOS and U2OS cells with circKEAP1 overexpression and/or RIG-I knockdown were detected by western blotting. (F). Levels of CXCL10, CCL5 and IFNγ in the supernatants of cells was measured by ELISA following RIG-I knockdown or/and circKEAP1 overexpression in U2OS and HOS cell lines. (G). Expression of inflammatory cytokines in the supernatants of cells was characterized using a human inflammation antibody array. (H). Protein expression of STAT1 and P65 was detected by western blotting in the nuclear and cytoplasmic fractions of the indicated cells. (I). Cellular localization of STAT1 and P65 was detected by immunofluorescence staining (scale bars, 20 μm). (J). Biotin pulldown assay was performed using circKEAP1 probe and the anti-Flag was detected in U2OS and HOS cells. (K). FISH-IF staining was performed to show the cellular localization of circKEAP1 and RIG-I (scale bars, 20 μm). Error bars represent three independent experiments. *, **, *** indicates significant differences compared with the control or indicated group at a p value < 0.05, < 0.01, < 0.001, respectively
Fig. 9
Fig. 9
Schematic diagram of the project. In OS, circKEAP1 inhibits cell proliferation, migration and stemness by encoding a novel protein KEAP1-259aa, which mediated ubiquitination and stability of vimentin as well as triggers an immune defense response via the RIG-I pathway
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Lin Z, Xie X, Lu S, Liu T. Noncoding RNAs in osteosarcoma: implications for drug resistance. Cancer Lett. 2021;504:91–103. doi: 10.1016/j.canlet.2021.02.007. - DOI - PubMed
    1. Yan GN, Lv YF, Guo QN. Advances in osteosarcoma stem cell research and opportunities for novel therapeutic targets. Cancer Lett. 2016;370(2):268–74. doi: 10.1016/j.canlet.2015.11.003. - DOI - PubMed
    1. Zhou WY, Cai ZR, Liu J, Wang DS, Ju HQ, Xu RH. Circular RNA: metabolism, functions and interactions with proteins. Mol Cancer. 2020;19(1):172. doi: 10.1186/s12943-020-01286-3. - DOI - PMC - PubMed
    1. Soghli N, Qujeq D, Yousefi T, Soghli N. The regulatory functions of circular RNAs in osteosarcoma. Genomics. 2020;112(4):2845–56. doi: 10.1016/j.ygeno.2020.03.024. - DOI - PubMed
    1. Li Z, Li X, Xu D, Chen X, Li S, Zhang L, Chan MTV, Wu WKK. An update on the roles of circular RNAs in osteosarcoma. Cell Prolif. 2021;54(1):e12936. doi: 10.1111/cpr.12936. - DOI - PMC - PubMed

MeSH terms