circHIPK3 prevents cardiac senescence by acting as a scaffold to recruit ubiquitin ligase to degrade HuR

Abstract

Rational: Senescence is a major aging process that contributes to the development of cardiovascular diseases, but the underlying molecular mechanisms remain largely unknown. One reason is due to the lack of suitable animal models. We aimed to generate a cardiomyocyte (CM)-specific senescent animal model, uncover the underlying mechanisms, and develop new therapies for aging associated cardiac dysfunction. Methods: The gain/loss of circHIPK3 approach was used to explore the role of circHIPK3 in cardiomyocyte (CM) senescence. To investigate the mechanisms of circHIPK3 function in cardiac senescence, we generated CM-specific tamoxifen-induced circHIPK3 knockout (CKO) mice. We also applied various analyses including PCR, Western blot, nuclear and cytoplasmic protein extraction, immunofluorescence, echocardiography, RNA immunoprecipitation assay, RNA-pulldown assay, and co-immunoprecipitation. Results: Our novel CKO mice exhibited worse cardiac function, decreased circHIPK3 expression and telomere length shortening in the heart. The level of the senescence-inducer p21 in the hearts of CKO mice was significantly increased and survival was poor compared with control mice. In vitro, the level of p21 in CMs was significantly decreased by circHIPK3 overexpression, but increased by circHIPK3 silencing. We showed that circHIPK3 was a scaffold for p21 mRNA-binding protein HuR and E3 ubiquitin ligase β-TrCP. circHIPK3 silencing weakened the interaction between HuR and β-TrCP, reduced HuR ubiquitination, and enhanced the interaction between HuR and p21 mRNA. Moreover, we found that mice injected with human umbilical cord mesenchymal stem cell-derived exosomes (UMSC-Exos) showed increased circHIPK3 levels, decreased levels of p21, longer telomere length, and good cardiac function. However, these beneficial effects exerted by UMSC-Exos were inhibited by silencing circHIPK3. Conclusions: We successfully generated CM-specific CKO mice for aging research. Our results showed that deletion of circHIPK3 led to exaggerated CM senescence and decreased cardiac function. As a scaffold, circHIPK3 enhanced the binding of E3 ubiquitin ligase β-TrCP and HuR in the cytoplasm, leading to the ubiquitination and degradation of HuR and reduced p21 activity. In addition, UMSC-Exos exerted an anti-senescence and cardio-protective effect by delivering circHIPK3. These findings pave the way to the development of new therapies for aging associated cardiac dysfunction.

Keywords: RNA-binding protein; aging; circHIPK3; exosome; senescence; ubiquitin ligase.

Figure 1

circHIPK3 inhibits cardiomyocyte senescence. (A) Quantification of circHIPK3 expression in different mouse tissues. n = 4. (B) qRT-PCR analysis of circHIPK3 in the hearts of mice of different ages. n = 4. (C) qRT-PCR analysis of the levels of circHIPK3 and HIPK3 in UMSC treated with RNase R. n = 3. (D) qRT-PCR analysis of circHIPK3 level in H9C2 cardiomyocyte 40 h after siRNA transfection. n = 6. (E) qRT-PCR analysis of circHIPK3 level in H9C2 cardiomyocyte 40 h after transfection of overexpression plasmid. n = 5. (F) The proliferation of primary cardiomyocytes was detected by EdU incorporation after transfection with circHIPK3 siRNA or overexpression plasmid for 40 h. n = 5. (G) Primary cardiomyocytes transfected with circHIPK3 siRNA or control were subject to β-gal staining, n = 5. (H-J) The mRNA and protein levels of p16 and p21 were detected by qRT-PCR or Western blot in H9C2 after transfection with circHIPK3 siRNA or overexpression plasmid for 40 h. n = 4-6. A and B, one-way ANOVA test. D-J two-tailed Student's t test.

Figure 2

Specific deletion of circHIPK3 in cardiomyocytes induces cardiac senescence. (A) SINEs were deleted using CRISPR/Cas9 systems. gRNAs were designed to delete the SINEs sequence. (B) qRT-PCR analysis of circHIPK3 expression in cells after SINEs deletion. n = 4. (C) Schematic illustration of the procedure to generate cardiomyocyte-specific knockout circHIPK3 mice. 8-week-old mice were subjected to intraperitoneal injection of tamoxifen and the mice were used for subsequent experiment. (D) Cardiac function was analyzed by echocardiography in circHIPK3 inducible knockout (CKO) mice after tamoxifen induction and littermate control mice without tamoxifen induction. n = 6. (E-F) qRT-PCR analysis of circHIPK3 and HIPK3 mRNA expressions in the hearts of circHIPK3 CKO and control mice. NS, not significant. (G) Telomere length was detected using the telomere length assay. n = 6. (H) qRT-PCR analysis of p16 and p21 mRNAs expression. n = 6. (I) Western blot analysis of p16 and p21 expression in circHIPK3 CKO and control mice. n = 4. B, one-way ANOVA test. D-I, two-tailed Student's t test.

Figure 3

circHIPK3 regulates HuR protein expression. (A) Cytoplasm distribution of HuR in control and circHIPK3 knockout mice. n = 4. (B) qRT-PCR analysis of HuR mRNA expression in control and CKO mouse heart. n = 5, NS, not significant. (C) Western blot analysis of HuR protein level in control and CKO mouse heart. n = 4. (D) Localization of HuR by immunofluorescence in H9C2 cells transfected with circHIPK3 siRNA. (E-F) qRT-PCR and Western blot analysis of HuR mRNA protein levels in H9C2 cells with circHIPK3 silencing or overexpression. n = 4. Data were analyzed by two-tailed Student's t test.

Figure 4

circHIPK3 promotes HuR degradation by recruiting E3 ubiquitin ligase. (A) The interaction network of HuR and β-TrCP protein was analyzed by the STRING database. (B) Western blot analysis of HuR in H9C2 cells transfected with circHIPK3 siRNA cells with or without MG132 treatment for 6 h. n = 5. (C-D) H9C2 cells were transfected with circHIPK3 overexpression plasmid for 40 h. RNA-pull down assay (C) and RIP (D) experiments were performed to demonstrate the interaction between circHIPK3 and HuR. n = 4. (E) RIP experiment showed the interaction between circHIPK3 and the ubiquitin E3 ligase β-TrCP. n = 4. (F) The cells were transfected with circHIPK3 siRNA for 40 h. Co-IP experiment showed that the interaction between HuR and β-TrCP was attenuated and the binding of HuR and ubiquitin was decreased by circHIPK3 siRNA. n = 4. (G) H9C2 cells were transfected with circHIPK3 siRNA for 40 h. RIP experiment showed that the interaction between HuR and p21 mRNA was enhanced by circHIPK3 silence. n = 4. (H) The half-life of p21 mRNA in NC siRNA and HuR siRNA groups was assessed 40 h after transfection by treating the cells with actinomycin D (2 mg/ml); mRNA half-life was calculated by qRT-PCR. n = 3. (I) Western blot analysis of p21 protein in H9C2 transfected with NC siRNA, circHIPK3 siRNA and circHIPK3 siRNA+HuR siRNA. n = 4. B, two-way ANOVA test. D, E, G, two-tailed Student's t test. I, one-way ANOVA test.

Figure 5

Exosomes prevent cardiomyocyte senescence by releasing circHIPK3. (A) The expression of circRNAs in exosome was determined by qRT-PCR analysis. n = 4. (B) PKH26-labeled Exo uptake was detected by fluorescence microscopy. n = 6. (C) qRT-PCR analysis of the expression of circHIPK3 in H9C2 cells treated with exosome for 24 h. n = 7. (D) circHIPK3 siRNA transfection in UMSC resulted in decreased expression of circHIPK3 in exosome. n = 4. (E) The proliferative rate of primary cardiomyocytes treated with NC-Exo or si-Exo (exosomes isolated from UMSC transfected with circHIPK3 siRNA). n = 5. (F) β-gal staining for primary cardiomyocytes treated with NC-Exo or si-Exo. n = 5. (G-H) qRT-PCR and Western blot analysis of p16, p21 mRNAs and proteins in H9C2 cells treated with NC-Exo or si-Exo for 24 h. n = 5. (I) EdU staining of proliferating primary cardiomyocytes treated with NC siRNA, circHIPK3 siRNA, and circHIPK3 siRNA+Exo. n = 4. (J) β-gal staining for primary cardiomyocytes treated with NC siRNA, circHIPK3 siRNA and circHIPK3 siRNA+Exo. n = 5. (K-M) The mRNA and protein levels of p16 and p21 were detected by qRT-PCR and Western blot in H9C2 cells treated with NC siRNA, circHIPK3 siRNA, and circHIPK3 siRNA+Exo group. n = 4. A, one-way ANOVA test. C-H, two-tailed Student's t test. I-M, one-way ANOVA test.

Figure 6

Exosomes prevent cardiac senescence in circHIPK3 knockout mice. (A) PKH26-labeled exosomes were injected via the tail vein. After 30 min, the hearts were harvested and the labeled exosomes in hearts were observed by fluorescence microscopy. (B) Schematic illustration of experiments performed in panels C-H. 8-week-old mice were subjected to intraperitoneal injection of tamoxifen and the mice were injected with exosome via the tail vein into CKO mice three times a week. By day 20, the mice were subjected to subsequent experiment. (C) Cardiac function analyzed by echocardiography in circHIPK3 CKO mice injected with PBS, NC-Exo and si-Exo. n = 5. (D) qRT-PCR analysis of circHIPK3 in circHIPK3 CKO mice injected with PBS, NC-Exo and si-Exo. n = 4. (E-G) qRT-PCR analysis of telomere length assay, p16 and p21 mRNAs in circHIPK3 CKO mice injected with PBS, NC-Exo and si-Exo. n = 4. (H) Western blot of p16 and p21 proteins in circHIPK3 CKO mice injected with PBS, NC-Exo and si-Exo. n = 4. Data were analyzed by one-way ANOVA test.

Figure 7

Proposed mechanisms of circHIPK3/HuR/p21 regulation in cardiac premature senescence. HuR in the cytoplasm binds to AU-rich elements in the 3'UTR of p21 mRNA, promoting the expression of p21 protein. Exosomal circHIPK3 binds to HuR and β-TrCP to promote the degradation of HuR via the ubiquitin-proteasome pathway. Increased circHIPK3 protects heart from premature senescence through destabilizing HuR protein and decreasing p21 expression.