CircECE1 activates energy metabolism in osteosarcoma by stabilizing c-Myc

Abstract

Background: Osteosarcoma (OS) is the most common malignant bone tumor and has a poor prognosis. The potential involvement of circular RNAs (circRNAs) in OS progression remains unexplored. Here, we report that CircECE1, a circular RNA derived from human ECE1, plays a critical role in energy metabolism in OS.

Methods: The RIP chip sequence assay was performed to confirm CircECE1, through overexpression or knockdown of CircECE1 to verify its function in 143B and U2OS. RNA immunoprecipitation and immunoprecipitation were used to verify CircECE1's regulation of protein c-Myc and co- immunoprecipitation was used to verified the competitive binding relationship between CircECE1 and SPOP. The influence of CircECE1 on energy metabolism was evaluated by seahorse experiment, western blot, and immunohistochemistry.

Results: We found that CircECE1 is highly expressed in OS tissues and cells and that CircECE1 knockdown suppresses tumor proliferation and metastasis both in vitro and in vivo. Further, CircECE1 significantly promotes glucose metabolism in OS cells in vitro and in vivo. Mechanistically, CircECE1 interacts with c-Myc to prevent speckle-type POZ-mediated c-Myc ubiquitination and degradation. C-Myc inhibits thioredoxin binding protein (TXNIP) transcription and subsequently activates the Warburg effect.

Conclusions: CircECE1 regulates the Warburg effect through the c-Myc/TXNIP axis. CircECE1 mediated signal transduction plays a important role in OS process and energy metabolism. These findings may identify novel targets for OS molecular therapy.

Keywords: C-Myc; CircECE1; Glucose metabolism; Osteosarcoma; TXNIP.

Fig. 1

CircECE1 validation and expression in osteosarcoma tissue and cells. a. RIP-seq results indicate that CircRNAs interact with c-Myc. The red arrow indicates CircECE1. b. Total RNAs were isolated from the specimens of patients with OS and from chondroma tissues for use in real-time PCR. Tumor samples exhibited significantly higher levels of CircECE1 compared to those in the chondroma tissue. Data represent the mean ± SD (n = 10). c. CircECE1 expression was higher in human OS pulmonary metastasis than in primary OS tissues. Representative images are provided (400× magnification). d. CircECE1 expression in hFOB1.19 and osteosarcoma (OS) cell lines (OS-732, 143B, HOS, SJSA-1, and MG-63) was evaluated by qRT-PCR. Data represent the mean ± standard deviation (SD) (n = 3). * P < 0.05. e. Schematic illustration reveals the ECE1 exon 2–4 circularization that forms CircECE1 (red arrow). The presence of CircECE1 was validated by RT–PCR and subsequent Sanger sequencing. The red arrow represents “head-to-tail” CircECE1 splicing sites. f. The presence of CircECE1 was validated in U2OS and 143B osteosarcoma cell lines by RT–PCR. Divergent primers amplified circECE1 in cDNA but not in genomic DNA. GAPDH was used as a negative control. g. The expression of CircECE1 and ECE1 mRNA in 143B cells treated with or without RNase R was detected by real-time PCR. The relative levels of CircECE1 and ECE1 mRNA were normalized to the value measured in the mock treatment. Data represent the mean ± SD (n = 3). * P < 0.05. h. RNA fluorescence in situ hybridization (FISH) revealed that CircECE1 was predominantly localized within the cytoplasm. CircECE1 probes were labeled with Alexa Fluor 488, and nuclei were stained with DAPI. Scale bar = 50 μm

Fig. 2

CircECE1 increased cell survival and colony formation and promoted the expression of c-Myc targets. a. Silencing or overexpression of CircECE1 in osteosarcoma cells decreased or increased, respectively, the ability of cell proliferation compared to that of the control. **P < 0.01. Data represent the mean ± SD (n = 6). b. CircECE1 functions in tumor cell proliferation as detected by EdU assay. Nuclei were stained with DAPI, and a combined reaction involving EdU and DAPI indicated the cells in S phase. c. CircECE1 overexpression promotes cell growth as determined by colony formation assay (details are shown in the insets). Error bars represent the mean ± SD of three independent experiments. * P < 0.05. d. Cellular transformation was induced by CircECE1 overexpression. Vector or CircECE1-overexpressing OS stable cells were cultured in soft agar for 20 days. Colonies were stained with crystal violet, photographed, and quantified using ImageJ (details are shown in inserts). e. CircECE1 downregulation in osteosarcoma cells increased cell apoptosis (n = 3). f. CircECE1 overexpression in osteosarcoma cells decreased cell anoikis (n = 3). g. Cell migration abilities of U2OS and 143B cells transfected with CircECE1 or vector were evaluated by transwell migration assays. Data represent the mean ± SD (n = 3). * P < 0.05. Scale bar, 50 μm. h. Protein lysates from vector and CircECE1 or shCircECE1-transfected osteosarcoma cells were subject to western blotting. Cell lysates were analyzed using c-Myc, HIF-1α, Cdc25a, Mdm2, ELK1, JUN, and JUNB antibodies. i. The shCircECE1 or CircECE1 and the vector transfected osteosarcoma cells were maintained at 80% confluence and were used for RNA extraction and measurement of c-Myc and c-Myc targets. The CircECE1 cells expressed significantly higher levels of c-Myc targets compared to those of the control

Fig. 3

CircECE1 interacts with c-Myc to prevent SPOP-mediated degradation. a. Identification of the regions of the c-Myc protein that interact with CircECE1. The fragments of the c-Myc protein are illustrated (left); the interaction of c-Myc protein regions with CircECE1 in 293 T cells was confirmed using a RIP assay (right). The interaction of CircECE1 with c-Myc and SPOP was verified using a RIP assay. b-c. Schematic diagram of CircECE1 WT and MUT (left); RNA immunoprecipitation experiments were performed using anti-c-Myc antibodies in 143B cells by qRT-PCR (b) and RT-PCR(c), and specific primers were used to detect CircECE1 or GAPDH; n = 3. d-e. The effect of CircECE1 WT/MUT on the expression of c-Myc protein levels in OS cells as assessed by WB (d) and IF (e). f. The effect of CircECE1 WT/MUT on the expression of c-Myc mRNA levels in OS cells as assessed by qRT-PCR. g. The effect of CHX treatment on the change in the c-Myc protein level mediated by CircECE1 WT and MUT overexpression as detected by western blotting. h. The effect of Bortezomib treatment on the change in the c-Myc protein level mediated by CircECE1 knockdown as detected by western blotting. i. The ubiquitination levels of c-Myc in CircECE1 WT or MUT cells as measured by IP experiments. Bortezomib (250 nM) and NEM (5 μm)(Protease inhibitor) or PR-619 (100 nm)(Non-selective deubiquitinating enzyme inhibitor) were added for 6 h. j. The ubiquitination levels of c-Myc in CircECE1 knockdown cells as measured by IP experiments. Bortezomib (250 nM) and NEM (5 μm) or PR-619 (100 nm) were added for 6 h. k. The effect of SPOP knockdown on the change in the c-Myc protein level mediated by CircECE1 knockdown as detected by western blotting. l. Sequence alignment of c-Myc with the SPOP binding motif (SBC) in known SPOP substrates. m. The interaction of CircECE1 WT and MUT with c-Myc in OS cells was verified using a RIP assay. n. The interaction of c-Myc with SPOP was verified using a Co-IP assay. o. Role of SBC1 and SBC2 in the direct interaction of SPOP with c-Myc. 293 T cells were transfected with expression vectors for SPOP WT and FLAG- c-Myc or with c-Myc that was mutated within the SBC (ΔSBC) 1 or 2 for 24 h. Bortezomib (250 nM) was then added for 6 h. Total cell lysates were prepared, and IP experiment was conducted using anti-FLAG or anti-SPOP antibodies. p. 293 T cells were transfected as indicated with HA-SPOP WT and Flag-c-Myc or Flag-c-Myc ΔSBC1 for 24 h. Immunoblot analyses were conducted to detect for Flag-tagged c-Myc, HA-tagged SPOP, and β-actin

Fig. 4

CircECE1 is a positive regulator of glucose metabolism in osteosarcoma. a. Schematic representation of glucose metabolism in cancer cells. b-c. OS cells infected with either control shRNA or CircECE1 shRNA were subjected to various analyses to measure the expression levels of genes involved in glucose metabolism by real-time PCR (b) or WB (c). OS cells infected with either control vector or CircECE1 overexpression vector were subjected to various analyses to measure the expression levels of genes involved in glucose metabolism by real-time PCR (b) or WB (c). d. CircECE1 overexpression increased ATP production and CircECE1 knockdown decreased ATP production. e. Mitochondrial potential increased in the presence of CircECE1 overexpression and decreased in the presence of CircECE1 knockdown. f. CircECE1 induced lactate production as indicated by the color of the medium. g. CircECE1 downregulation inhibited glucose uptake and lactate production in osteosarcoma cells. h. Extracellular acidification rate (ECAR), an indicator of glycolysis, was reduced in response to CircECE1 knockdown. Oxygen consumption rate (OCR), which reflects mitochondrial respiration, was decreased in CircECE1 inhibited osteosarcoma cells

Fig. 5

CircECE1 regulates glucose metabolism via c-Myc. a-b. After knocking down c-Myc, over-expression of circECE1, the protein and RNA of genes related to glucose metabolism were assessed by real-time PCR (a) and WB (b). c. CircECE1 MUT decreased ATP production induced by CircECE1 WT in U2OS and 143B cells. d. Mitochondrial potential was decreased after CircECE1 MUT overexpression compared to that of CircECE1 WT. e. Glucose uptake and lactate production was lower in the CircECE1 MUT-overexpressing OS cells. f. CircECE1 MUT rescued the upregulated the glycolysis rate induced by CircECE1 WT, and this was supported by the ECAR. CircECE1 MUT rescued the upregulated OCR induced by CircECE1 WT. g. CircECE1 WT function in tumor cell proliferation as detected by CCK-8 assay. h. CircECE1 WT function in tumor cell proliferation as detected by EdU assay. Nuclei were stained with DAPI. The combined reaction between EdU and DAPI indicated cell that were in S phase. i. CircECE1 overexpression promoted cell growth based on the results of a colony formation assay (details are shown in the insets). Error bars represent the mean ± SD of three independent experiments. * P < 0.05. j. Cell migration abilities of U2OS and 143B cells transfected with CircECE1 WT and MUT or vector were evaluated using transwell migration assays. Data represent the mean ± SD (n = 3). * P < 0.05. Scale bar = 50 μm. k. CircECE1 MUT overexpression in osteosarcoma cells increased cell anoikis (n = 3) compared to that observed in response to CircECE1 WT

Fig. 6

TXNIP negatively regulates glucose metabolism and proliferation in osteosarcoma. a. Overexpression of CircECE1 in OS cells, and heat map detailing the RNA-seq. b. Knockdown of c-Myc or CircECE1 increased the levels of TXNIP mRNA. c. Luciferase assay showed that c-Myc binds to the TXNIP promoter. d. c-Myc occupies the E-box of the TXNIP promoter region as measured by ChIP assay. e. Knockdown of c-Myc increased TXNIP protein expression. f. Effect of TXNIP overexpression or knockdown on the expression levels of genes involved in glucose metabolism in OS cells as assessed by real-time PCR. g. TXNIP inhibited cell proliferation as measured by a CCK-8 proliferation kit. h. TXNIP inhibited cell proliferation as measured by a colony formation assay. i. Cell lysates were collected, and immunoblotting was performed using the indicated antibodies. j. TXNIP rescued the expression levels of genes involved in glucose metabolism in OS cells by real-time PCR, and this was induced by CircECE1 overexpression. k-m. ATP and lactate production was lower in the TXNIP and CircECE1 co-expressing OS cells. n. TXNIP and CircECE1 co-expression lowered the glycolysis rate as indicated by the ECAR and OCR. o. TXNIP and CircECE1 function in tumor cell proliferation as detected by CCK-8 assay. p. TXNIP and CircECE1 function in tumor cell proliferation as detected by EdU assay. Nuclei were stained with DAPI. A combined reaction between EdU and DAPI identified the cells in S phase. q. Cell migration abilities of U2OS and 143B cells transfected with CircECE1 and TXNIP, CircECE1 alone, or vector were evaluated using transwell migration assays. Data represent the mean ± SD (n = 3). * P < 0.05. Scale bar = 50 μm. r. TXNIP and CircECE1 overexpression in osteosarcoma cells increased cell anoikis (n = 3) compared to that observed in response to CircECE1 alone

Fig. 7

CircECE1 positively regulates glucose metabolism in a xenograft model. a. 143B cells stably expressing CircECE1 WT, MUT, or empty vector were subcutaneously injected into nude mice. Nude mice were respectively injected with an equal amount of 5 × 10⁶ stable control cells or cells transfected with CircECE1 WT or MUT subcutaneously. After 30 days, tumors were dissected and photographed. b. Tumor weight was calculated on the day mice were euthanized. Data represent the mean ± SEM (n = 6 each group). c. Tumor volumes (ab2/2) were recorded every six days beginning on the day after mice were injected with the stable OS cells. Data represent the mean ± SEM (n = 6 each group). d. Western blotting was used to assess protein expression levels. Expression of the rate-limiting enzymes GLUT1, GLUT4, MCT4, PDK1, and PDK4 was decreased in tumors formed by CircECE1 overexpression in osteosarcoma cells. e. RT-qPCR revealed that the mRNA levels of the rate-limiting enzymes GLUT1, GLUT4, MCT4, PDK1, and PDK4 were decreased in tumors formed by CircECE1 overexpression in osteosarcoma cells. f. H&E staining and immunohistochemistry (IHC) revealed the structure of OS in mice and the relative protein levels of GLUT1, GLUT4, MCT4, PDK1, and PDK4 in tumors of different groups. Scale bars = 100 μm. g. Lung metastasis in mice injected with various stable 143B cells in the tail vein was detected using an in vivo bioluminescence imaging system. Representative images and a histogram are shown (n = 9 each group). h. Fish revealed the expression level of CircECE1 and Immunohistochemistry (IHC) revealed the relative protein levels of GLUT1, GLUT4, MCT4, PDK1, and PDK4 in chondroma and osteosarcoma. Scale bars = 100 μm. I. Schematic illustration of the CircECE1/c-Myc/TXNIP axis