Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 44 (6), 2846-58

Foxo3 Circular RNA Retards Cell Cycle Progression via Forming Ternary Complexes With p21 and CDK2

Affiliations

Foxo3 Circular RNA Retards Cell Cycle Progression via Forming Ternary Complexes With p21 and CDK2

William W Du et al. Nucleic Acids Res.

Abstract

Most RNAs generated by the human genome have no protein-coding ability and are termed non-coding RNAs. Among these include circular RNAs, which include exonic circular RNAs (circRNA), mainly found in the cytoplasm, and intronic RNAs (ciRNA), predominantly detected in the nucleus. The biological functions of circular RNAs remain largely unknown, although ciRNAs have been reported to promote gene transcription, while circRNAs may function as microRNA sponges. We demonstrate that the circular RNA circ-Foxo3 was highly expressed in non-cancer cells and were associated with cell cycle progression. Silencing endogenous circ-Foxo3 promoted cell proliferation. Ectopic expression of circ-Foxo3 repressed cell cycle progression by binding to the cell cycle proteins cyclin-dependent kinase 2 (also known as cell division protein kinase 2 or CDK2) and cyclin-dependent kinase inhibitor 1 (or p21), resulting in the formation of a ternary complex. Normally, CDK2 interacts with cyclin A and cyclin E to facilitate cell cycle entry, while p21works to inhibit these interactions and arrest cell cycle progression. The formation of this circ-Foxo3-p21-CDK2 ternary complex arrested the function of CDK2 and blocked cell cycle progression.

Figures

Figure 1.
Figure 1.
The effect of circ-Foxo3 on cell proliferation. (A) Left, Real-time PCR showed that circ-Foxo3 was highly expressed in the non-cancer cell lines of mouse embryo fibroblast (MEF), mouse cardiac fibroblast (MCF) and NIH3T3, as compared with the cancer cell lines 67NR, 66C14, 4T07, 4T1 and B16. Right, The levels of Foxo3 linear mRNA were not correlated with circ-Foxo3. Asterisks indicate significant differences. **P < 0.001, Error bars, SD (n = 4). (B and C) Different cell lines as indicated (B) or NIH3T3 fibroblasts (C) were seeded at the cell density of 1 × 105 cells/well on 6-well dishes in 10% FBS/DMEM medium until 50, 80, 100% or over-confluence based on the coverage of the surface of the tissue culture plates, followed by determination of circ-Foxo3 levels and cell cycle distribution. Increased cell densities expressed higher levels of circ-Foxo3 (B) and more cells were detected in the G1 phase (C). (D) NIH3T3 fibroblasts were incubated in basal medium with EGF (0, 2, 10 and 50 ng/ml, left) or AG1478 (0, 0.5, 1.5 and 5 μM, right) for 24 h. Real-time PCR showed circ-Foxo3 expression decreased after EGF treatment but increased after AG1478 treatment. **P < 0.001, Error bars, SD (n = 4). (E) Left, A siRNA was designed to specifically target circ-Foxo3. Right, Cells were transfected with siRNA targeting circ-Foxo3 or a control oligo. RNAs isolated were subject to real-time PCR to confirm downregulation of circ-Foxo3 in the siRNA-transfected cells. (F) MCF cells transfected without (wild-type) or with circ-Foxo3 siRNA or 2 oligos with random sequences were cultured in DMEM supplemented with 5% FBS for up to 5 days. The siRNA-transfected cells grew fast compared to the controls. **P < 0.001, Error bars, SD (n = 4). (G) B16 cells transfected without or with circ-Foxo3 siRNA or the control oligos were cultured in DMEM with 2.5% FBS for up to 5 days. Cell proliferation assays showed that circ-Foxo3 siRNA-transfected cells grew fast compared to the controls. **P < 0.001, Error bars, SD (n = 4). (H) Silencing circ-Foxo3 decreased the number of cells in G1 phase, but increased the number of cells in S and G2 phase. **P < 0.01. Error bars, SD (n = 4).
Figure 2.
Figure 2.
circ-Foxo3 repressed cell cycle entry. (A) NIH3T3 cells were cultured in serum-free medium for 48 h, then changed with medium containing 10% FBS for cell cycle synchronization. Cells were collected each hour for up-to 24 h and subject to flow cytometry for cell cycle analysis. Error bars, SD (n = 3). (B) Circ-Foxo3 levels were determined by real-time PCR. Inset, the correlation between circ-Foxo3 levels and percentage of cells in G1 phase was analyzed by Graph Prism 4. It showed that circ-Foxo3 levels were highly correlated with the percentages of cells in G1 phase. R squares, 0.778; P < 0.001, n = 25. (C) Levels of circ-Foxo3 were determined by real-time PCR in NIH3T3 cells transfected with GFP vector, Foxo3, mock control, circ-Amotl1 and circ-Foxo, or in over confluence (OC) cells. **P < 0.01. Error bars, SD (n = 4). (D) NIH3T3 fibroblasts transfected without or with circ-Foxo3, the control vector, or a GFP plasmid were cultured in DMEM with 5% FBS. Cell proliferation assays showed that circ-Foxo3 expressing cells grew slowly compared to the controls. **P < 0.01. Error bars, SD (n = 4). (E) The cells were also cultured in DMEM supplemented with 5% FBS for 2 days, and then processed to flow cytometry. Expression of circ-Foxo3 significantly increased the number of cells in the G1 phase, and decreased the number of cells in the S and G2 phases. **P < 0.01. Error bars, SD (n = 4). (F) B16 cells were transfected as above and cultured in DMEM with 5% FBS. Cell proliferation assays showed that circ-Foxo3 expressing cells grew slowly compared to the controls. **P < 0.01. Error bars, SD (n = 4). (G) The cells were also subject to flow cytometry. Expression of circ-Foxo3 significantly increased the number of cells in G1 phase and decreased the number of cells in S and G2 phases. **P < 0.01. Error bars, SD (n = 4).
Figure 3.
Figure 3.
circ-Foxo3 interacted with CDK2 and P21. (A) Cell lysis prepared from NIH3T3 cells were subject to immuno-precipitation (IP) with antibodies against rabbit IgG, mouse IgG, cyclin A, cyclin B, cyclin C, cyclin D1, cyclin E, CDK2, CDK4, CDK6, p16, p18, p21, p27 and p57, followed by real-time PCR with primers specific for linear Foxo3 mRNA or circ-Foxo3. Anti-CDK2, CDK6, p16, p21and p27 antibodies pulled-down circ-Foxo3, but not linear Foxo3 mRNA. It was especially obvious that precipitating CDK2 or P21 pulled-down circ-Foxo3. **P < 0.01. Error bars, SD (n = 4). (B) Cell lysates prepared from NIH3T3 cells transfected with circ-Foxo3 or mock control were subject to immunoprecipitation with anti-rabbit IgG, mouse IgG, cyclin A, cyclin B, cyclin C, cyclin D, cyclin E, CDK2, CDK4, CDK6, p16, p18, p21, p27 and p57 antibodies. Western blot showed that immunoprecipitation pulled down similar amount of proteins in both control and circ-Foxo3-transfected cells. (C) The immuno-precipitated mixtures were also subject to real-time PCR with primers specific for circ-Foxo3. Anti-CDK2, CDK6, p16, p21and p27 antibodies pulled-down more circ-Foxo3 from NIH3T3 cells transfected with circ-Foxo3 than from mock control. *P < 0.01. Error bars, SD (n = 4). (D) NIH3T3 cells were subject to flow cytometry to sort cells in G1, S and G2 phases, followed by immunoprecipitation with anti-rabbit IgG, mouse IgG, p21 and Cdk2 antibodies and real-time PCR with primers specific for circ-Foxo3. Antibodies against p21 and Cdk2 precipitated significantly more circ-Foxo3 in G1 phase than in G2 an S phases. **P < 0.01. Error bars, SD (n = 4). (E) Cell lysates prepared were subject to immunoprecipitation with anti-rabbit IgG, mouse IgG, p21 and Cdk2 antibodies, followed by real-time PCR with primers specific for circ-Foxo3, circ-DNAJA1, circ-MRPL47, circ-NDUF53, circ-RPS5 and circ-RPF5. Antibodies against p21 and Cdk2 pulled-down circ-Foxo3, but not the other circular RNAs. **P < 0.01. Error bars, SD (n = 4).
Figure 4.
Figure 4.
circ-Foxo3 enhanced the interaction between p21 and CDK2. (A) Lysates prepared from NIH3T3 cells transfected with circ-Foxo3 or mock control were subject to IP with anti-CDK2 antibody, followed by Western blotting. CDK2 precipitation pulled down more p21, and less cyclin A and cyclin E in the circ-Foxo3-transfected cells than the control. (B) Left, the lysates were subject to IP with anti-cyclin A antibody, followed by Western blotting. Cyclin A precipitation pulled down less CDK2 in the circ-Foxo3-transfected cells than in the control. Right, the lysates were subject to IP with anti-cyclin E antibody, followed by western blotting. Cyclin E precipitation pulled-down less CDK2 in the circ-Foxo3-transfected cells than in the control. (C) Circ-Foxo3 transfection did not change expression of cyclin A and cyclin E. (D) Cell lysates prepared from NIH3T3 cells transfected with GFP vector, Foxo3, circular RNA vector, circ-Amotl1 and circ-Foxo3, or over-confluence culture were subject to western blot probed with antibodies against CDK2, p21 and β-action. The lysates were also subject to immunoprecipitation with antibody against p21 or CDK2. Anti-p21 antibody pulled-down more Cdk2 and anti-CDK2 antibody pulled-down more p21 in the cells transfected with circ-Foxo3 or overgrown. (E) Lysates prepared from NIH3T3 cells transfected with circ-Foxo3 and a vector, or over-grown cells were mixed with biotinylated probes against circ-Foxo3 or an oligo. Western blotting showed that levels of CDK2 and p21 were not affected (upper). However, the circ-Foxo3 probe pulled down more CDK2 and p21 in the cells transfected with circ-Foxo3 or over-grown relative to the controls (lower). (F) Lysates prepared from NIH3T3 cells sorted into G1, S or G2 phase were subject to Western blotting. The levels of p21 and CDK2 were similar in cells of different phases (upper). Anti-p21 and anti-CDK2 antibodies pulled down more CDK2 and p21, respectively, in the G1 phase cells. (G) The lysates prepared from circ-Foxo3- and mock-transfected cells were subject to native gradient gel electrophoresis followed by western blotting probed with antibodies against p21 or CDK2. (H) Diagram of our hypothesis showing the effect of circ-Foxo3 on cell cycle progression. Circ-Foxo3 retards cell cycle entry via enhancing interaction between p21 and CDK2, which repressed CDK2–cyclin complex formation.
Figure 5.
Figure 5.
Silencing circ-Foxo3 enhanced cell cycle entry and decreased the interaction between p21 and CDK2. (A) Lysates prepared from subconfluence NIH3T3 cells transfected with circ-Foxo3 siRNA or a control oligo were subject to western blotting. Silencing circ-Foxo3 did not change expression of p21 and CDK2. (B) The lysates were subject to IP with anti-p21 antibody followed by western blotting. p21 precipitation pulled-down less CDK2 in the circ-Foxo3 siRNA-transfected cells relative to the control. (C) The lysates were subject to IP with anti-CDK2 antibody, followed by western blotting. CDK2 precipitation pulled-down less p21, but more cyclin A and cyclin E in the siRNA-transfected cells compared to the control. (D) Silencing circ-Foxo3 did not change expression of cyclin A and cyclin E. (E) Cyclin A and cyclin E precipitations pulled down more CDK2 in the circ-Foxo3 siRNA-transfected cells. (F) The lysates were subject to native gradient gel electrophoresis followed by western blotting probed with antibodies against p21 or CDK2. (G) Cell lysates prepared from p21 siRNA- and a control oligo-transfected NIH3T3 cells were subject to immunoprecipitation with anti-rabbit IgG, mouse IgG, p21 and Cdk2 antibodies, followed by real-time PCR. Anti-p21 antibody pulled-down less circ-Foxo3 in the p21 siRNA-transfected cells. **P < 0.01. Error bars, SD (n = 4). (H) The lysates were hybridized with circ-Foxo3 probe or a control oligo for RNA pull-down assays. Real-time PCR showed that the circ-Foxo3 probe pulled down same levels of circ-Foxo3 in the cells transfected with p21 siRNA. Error bars, SD (n = 4). (I) Silencing p21 did not affect CDK2 expression. However, anti-p21 antibody pulled-down less Cdk2 and p21 in the p21 siRNA-transfected cells. Anti-CDK2 antibody pulled-down less p21, but the same amount of Cdk2 in the p21 siRNA-transfected cells. (J) Silencing CDK2 did not affected circ-Foxo3 levels. (K) Anti-CDK2 antibody pulled-down less circ-Foxo3 in the Cdk2 siRNA-transfected cells. **P < 0.01. Error bars, SD (n = 4). (L) Silencing CDK2 did not affect p21 expression. However, anti-CDK2 antibody pulled-down less Cdk2 and p21 in the CDk2 siRNA-transfected cells. Anti-p21 antibody pulled-down less CDK2, but the same amount of p21 in the CDK2 silencing cells.
Figure 6.
Figure 6.
Pulling down circ-Foxo3 precipitated both p21 and CDK2. (A) Lysates prepared from NIH3T3 cells transfected with circ-Foxo3 or a control vector, were subject to RNA pull-down assays. Real-time PCR showed that the circ-Foxo3 probe pulled down high levels of circ-Foxo3 in the cells transfected with circ-Foxo3. **P < 0.01. Error bars, SD (n = 4). (B) Left: the lysates were mixed with biotinylated probes against circ-Foxo3 or a control oligo and subject to western blot with antibodies against CDK2, p21 and β-action. Right: the pull-down mixture was subject to Western blotting. Pulling down circ-Foxo3 also pulled down CDK2 and p21 in cells transfected with circ-Foxo3. (C) Lysates prepared from NIH3T3 cells transfected with circ-Foxo3 siRNA or a control oligo, were subject to RNA pull down assays. Real-time PCR showed that the probe pulled down less circ-Foxo3 in the cells transfected with circ-Foxo3 siRNA. **P < 0.01. Error bars, SD (n = 4). (D) Left, transfection with circ-Foxo3 siRNA did not affect expression of CDK2 and p21. Right, Pulling down circ-Foxo3 precipitated less CDK2 and p21 in the cells transfected with circ-Foxo3 siRNA than those transfected with the control oligo. (E) Silencing p21 had little effect on circ-Foxo3 pulling down CDK2. (F) Silencing Cdk2 had little effect on circ-Foxo3 pulling down p21.
Figure 7.
Figure 7.
CDK2 and p21 protected the circ-Foxo3 binding site. (A and B) The lysates prepared from circ-Foxo3- and mock-transfected cells were incubated with different concentrations of RNAse A followed by native gradient gel electrophoresis. The gels were subject to Western blotting probed with antibodies against p21 (A) or CDK2 (B). (C) Cell lysates from NIH3T3 cells transfected with circ-Foxo3 and mock control were treated with or without RNAse A (0.1 mg/ml), and subject to western blotting. While treatment with RNAse A did not affect levels of CDK2 and p21 (left), anti-p21 antibody was able to pull-down Cdk2 and anti-CDK2 antibody was able to pull down p21 in the circ-Foxo3-transfected cells with or without RNAse A treatment. (D) The lysates prepared from NIH3T3 cells grown to 50% conference and over-confluence were incubated with different concentrations of RNAse A followed by native gradient gel electrophoresis. The gels were subject to western blot probed with antibody against p21 and then against Cdk2.

Similar articles

See all similar articles

Cited by 283 PubMed Central articles

See all "Cited by" articles

References

    1. Jeck W.R., Sharpless N.E. Detecting and characterizing circular RNAs. Nat. Biotechnol. 2014;32:453–461. - PMC - PubMed
    1. AbouHaidar M.G., Venkataraman S., Golshani A., Liu B., Ahmad T. Novel coding, translation, and gene expression of a replicating covalently closed circular RNA of 220 nt. Proc. Natl. Acad. Sci. U.S.A. 2014;111:14542–14547. - PMC - PubMed
    1. Haimovich G., Medina D.A., Causse S.Z., Garber M., Millan-Zambrano G., Barkai O., Chavez S., Perez-Ortin J.E., Darzacq X., Choder M. Gene expression is circular: factors for mRNA degradation also foster mRNA synthesis. Cell. 2013;153:1000–1011. - PubMed
    1. Wilusz J.E., Sharp P.A. Molecular biology. A circuitous route to noncoding RNA. Science. 2013;340:440–441. - PMC - PubMed
    1. Ivanov A., Memczak S., Wyler E., Torti F., Porath H.T., Orejuela M.R., Piechotta M., Levanon E.Y., Landthaler M., Dieterich C., et al. Analysis of intron sequences reveals hallmarks of circular RNA biogenesis in animals. Cell Rep. 2015;10:170–177. - PubMed

Publication types

MeSH terms

Feedback