Background
Osteosarcoma is a rare bone-forming tumor that primarily affects adolescents [
1‐
3]. Although it accounts for approximately 5% of childhood and adolescent cancers, it accounts for the majority of childhood cancer mortality. This is especially true for cases after the 1970s, and its survival rate has not improved as much as other childhood cancers [
4]. A few low-grade osteosarcomas can be treated via surgery alone [
5], but this discussion focuses on the more common and aggressive high-grade osteosarcomas [
6]. The disease is thought to be comprised of solid and drug-resistant tumors, and until the 1970s, it was primarily treated by surgery. Even when the lesion can be fully controlled locally, 80–90% of patients develop fatal lung metastases within one year of diagnosis. Indicative of the malignant nature of osteosarcomas, diagnosis is made assuming undetectable micrometastases, and systemic treatment is performed outside of surgery.
Recent advances in oncology research and related technologies have made it possible to identify the minor components of tumorigenic cells in cancers such as osteosarcoma; these cells are termed CSCs. CSCs derived from osteosarcomas have the characteristics of self-renewal, pluripotency, tumorigenicity, multidrug resistance and metastasis. The presence of CSCs is responsible for treatment failure, including recurrence, metastasis, and chemotherapy resistance [
7,
8]. Stemness hallmarks, including the upregulation of CD133, ABCG2, CXCR4, CD117, Oct4, Nanog, Sox2 and ALDH1A1 expression in CSCs, are considered specific markers [
9,
10]. CSCs are widely regulated by regular physiological regulatory pathways, such as the TGF-β, Notch, Hedgehog, and P42/44 MAPK signaling pathways. Moreover, increasing evidence has confirmed that other factors, including BMP-2, TNF-α, SOX2 and OCT3/4, may be involved in the maintenance of CSCs [
8]. Despite many efforts to characterize CSCs, new therapeutics to eradicate CSCs have not been established due to our limited understanding of CSC biology. Additionally, the presence of CSCs was reported to be a key factor of chemoresistance; however, the exact mechanisms responsible for this effect are unknown. Therefore, further in-depth study of osteosarcoma characteristics, CSC regulatory mechanisms and related signaling pathways will help to identify new molecular targets for therapeutic interventions of osteosarcoma.
LRPPRC belongs to a large family of RNA-binding proteins that are characterized by a canonical 35 residue repeat motif [
11]. LRPPRC is reported to be essential for maintaining mitochondrial homeostasis by protecting the mitochondrial membrane potential against reactive oxygen species formation by interacting directly with Beclin 1 and Bcl-2 [
12,
13]. In recent years, the role of LRPPRC in tumor progression has been investigated, and it has been reported that LRPPRC expression is markedly decreased in diethylnitrosamine (DEN)-induced mouse HCC tissues. Depletion of LRPPRC increased the incidence of DEN-induced HCC [
14]. In contrast, in human lung adenocarcinoma tissues, LRPPRC was found to be downregulated compared to adjacent nontumor tissues [
15]. LRPPRC knockdown or depletion in human bladder and lung cancer cells inhibits malignancies both in vitro and in vivo [
15,
16]. Liu and colleagues also reported that in human hepatocellular carcinoma tissues, LRPPRC is significantly upregulated and is correlated with worse survival for HCC patients [
17]. These findings raised the question of the exact role of LRPPRC in osteosarcoma.
The forkhead box transcription factor FOXM1 is involved in a variety of cellular processes, such as proliferation, cell cycle progression, cell differentiation, DNA damage repair, tissue homeostasis, angiogenesis, apoptosis, and REDOX signaling [
18]. In addition to being beneficial to the normal function of cells, FOXM1 is involved in several disease scenarios, including cancer. It has been assigned an oncogenic status based on multiple lines of evidence of its role in tumor development and progression. FOXM1 is highly expressed in several cancers and is also associated with poor prognosis. Notably, LRPPRC was found to directly regulate FOXM1 and thus exert a protective effect against oxidative stress via mitochondrial mtRNA metabolism and the circANKHD1/FOXM1 axis [
18]. FOXM1 was also reported to be an inducer of chemoresistance, and several strategies have been developed to inhibit FOXM1 to induce chemosensitivity [
19]. These results indicate that LRPPRC might regulate malignancies in osteosarcoma by interacting with FOXM1, which is worth further investigation.
In this study, we characterized LRPPRC as a potentially key oncogenic factor in osteosarcoma and derived a subpopulation of cells termed CSCs. Our data showed that the knockdown of LRPPRC markedly decreased malignant behaviors in osteosarcoma and osteosarcoma-derived CSCs. LRPPRC knockdown decreased mitochondrial homeostasis and led to mitochondrial dysfunction. Importantly, we showed for the first time that LRPPRC modulates stemness in CSCs derived from osteosarcoma cells by regulating FOXM1 and oxidative stress. Taken together, our results demonstrate the oncogenic role of LRPPRC in osteosarcomas and indicate that it is a promising therapeutic target for osteosarcomas.
Methods
Cell cultures
The human osteosarcoma cells SaOS2, MG63, HOS, 143B and U2OS, and a human osteoblast cell hFOB1.19 were purchased from Type Culture Collection of the Shanghai Institute for Biological Sciences and cultured in DMEM medium containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin at 37 °C in a 5% CO2 incubator. Every two days, medium was refreshed and passaged when confluence reached to 90%.
Enrichment of CSCs in vitro
Enrichment of CSCs from osteosarcoma cells was performed by culturing cells in serum-free medium. Briefly, cells were suspended in serum-free DMEM/F12 medium supplemented with 20ng/ml EGF, 20ng/ml bFGF, and 1X B27supplement. Every three days, medium was half-replaced with addition of EGF, bFGF, and 1X B27supplement. The single cell suspension was plated in 6-well ultralow attachment plates at a density of 5,000–10,000 cells per well and cultured for 10–14 d at 37 °C and 5% CO2. Every five days, spheres were imaged using under a X71 (U-RFL-T) fluorescence microscope (Olympus, Melville, NY).
Western blot analysis
Total protein were harvested and lysed using Animal Tissue/cells/bacteria total protein isolation kit (DocSense, Chengdu, China) from cells. Then, quantification of protein lysates was measured with a protein BCA assay Kit (Bio-Rad) and fractionated by 10% SDS-PAGE and transferred to a nitrocellulose membrane. The blots were cut according to the predicted size of target protein prior to hybridization with antibodies. Then, membranes were blocked with 5% nonfat milk in TBST (50 mmol/L Tris–HCl (pH 7.6), 150 mmol/L NaCl, 0.3% Tween 20) for 1 h at room temperature and then incubated with primary antibody diluted in blocking buffer at 4 °C overnight: Anti-LRPPRC/GP130 antibody (ab259927, 1:1000); Anti-Nanog antibody (ab109250, 1:1000); Anti-Oct4 antibody (ab200834, 1:1000); Anti-SOX2 antibody (ab92494, 1:1500); Anti-FOXM1 antibody (ab207298, 1:2000); Anti-Peroxiredoxin 3/PRDX3 antibody (ab128953, 1:2000); Anti-SOD2/MnSOD antibody (ab68155, 1:1000); Anti-Catalase antibody (ab209211, 1:1000); Anti-beta Actin antibody (ab8226, 1:4000). After three washes with TBST, the membranes were incubated with horseradish peroxidase-coupled Goat Anti-Rabbit IgG H&L secondary antibody (1:50,000; abcam, ab6721) for 1 h at room temperature avoiding from light. Detection was performed with enhanced chemiluminescence according to the manufacturer’s instructions (ECL kit, Life Technologies).
Cell apoptosis assay
Cell apoptosis was measured by performing flow cytometry with the Annexin V-FITC Apoptosis Detection Kit (Sigma-Aldrich, USA) according to the manufacturer’s instructions. Briefly, cells were harvested after treatment, then washed for three times using chilled PBS. Cells were pelleted and resuspended in 200 µL binding buffer, and incubated with 5 µL Annexin V-FITC and 10 µL PI for 20 min at room temperature. Then, 800 µL binding buffer was added to adjust cell concentration to 1 × 106 cells /mL. Subsequently, the number of stained cells was assessed with a 3 laser Navios flow cytometers, Beckman Coulter, Brea, CA, USA).
Cells were cultured in 6-well plates (Corning, New York, USA) at 2000 cells per well in 2 ml medium. After 14 days, colonies were fixed using 4% paraformaldehyde at room temperature for 10 min. Then, fixed cells were washed for three times using PBS and stained with 0.25% cystal violet for 30 min at room temperature. After three washed with PBS, colonies were imaged under a X71 (U-RFL-T) fluorescence microscope (Olympus, Melville, NY).
Cell invasion assay
For measuring invasion ability, cells were suspended a plated into upper chamber (8-µm pore size; Corning Inc., Corning, NY. USA). Transwell membrane was coated with a basement membrane Matrigel (37 µg/filter). Cells were seeded into the upper inserts 1.5 × 104 cells per insert in 150 µl of serum-free RPMI-1640. Outer wells were filled with 600 µl of RPMI-1640 containing 10% FBS as a chemoattractant. Cells were incubated at 37 °C with 5% CO2 for 48 h. The non-invading cells were gently removed and the invading cells on the under surface of the filter were fixed and stained with 0.25% crystal violet (Sigma–Aldrich). Four independent experiments were performed.
Cell transfection and lentivirus production
To produce lentivirus, HEK293T cells were co-transfected with lentivirus expression vector containing target sequence and packing plasmids (pMD2.G and psPAX2, Addgene MA, USA) by calcium phosphate precipitation, then refreshed with culture medium 16 h after transfection. Thirty-six hours later, the lentivirus supernatant was harvested and stored in aliquots at − 80 °C until use. Target cells were infected with lentiviral supernatant supplemented with 8 µg/mL polybrene (Millipore, Billerica, MA).
Cell cycle analysis
Cells were collected and washed for three times using chilled PBS. 1ml of 75% chilled ethanol was added to suspend 1 × 106 cells and stored in 4℃ for 24 h. After removing supernatant, cells were stained using 0.05 mg/mL propidium iodide (Sigma-Aldrich, USA.) for 15 min at room temperature avoiding from light. Then, cells were analyzed using a 3 laser Navios flow cytometers, Beckman Coulter, Brea, CA, USA).
Quantitative ROS measurement
The cells were labeled with the fluorescent DCFH-DA probe to determine the intracellular ROS generation and measured with flow cytometer. The cells were incubated with DCFH-DA at 37℃ for 30 min and harvested. After that, ROS accumulation was performed followed by ROS detection using a GloMax 96 Microplate Luminometer (Promega).
JC-1 staining (determination of mitochondrial membrane potential)
A JC-1 staining assay kit was used to determinate the alteration of mitochondrial membrane potential in the cells according to the instructions. Briefly, the cells were washed with PBS and stained with 20ug/ml JC-1 at 37℃ for 30 min in the dark. After washing twice with staining buffer, the cells were detected by a 3 laser Navios flow cytometers, Beckman Coulter, Brea, CA, USA).
ATP measurement
The cells were plated in 96-well plates at a density of 5 × 104 cells per well. ATP-dependent luciferase luminescence was measured in cells incubated using the CellTiterGlo kit according to manufacturer’s instructions and analyzed using luciferase chemiluminescence Multiskan spectrum microplate reader (Thermo Electron Corporation, Waltham, MA, USA).
Data analysis and statistics
All data are presented as mean ± SD. Statistical differences between different groups were analyzed by one way ANOVA post hoc Dunnett T-test using Prism 6 (GraphPad Software, San Diego, CA). Value of P < 0.05 or P<0.01 was considered statistically significant.
Discussion
Mitochondria are indispensable in tumorigenesis because many aspects of mitochondrial functions, such as metabolism, oxidative stress regulation, cell death susceptibility and signal transduction, are involved in tumorigenesis [
21]. Many reports have demonstrated that the LRPPRC protein and other family members critically regulate mitochondrial homeostasis and function [
11]. Nevertheless, the biological function and potential molecular mechanisms by which LRPPRC is involved in osteosarcoma progression are still unclear. In this study, the regulatory roles of LRPPRC in osteosarcoma progression by regulating mitochondrial homeostasis and function were revealed. It is the first time to evaluate the roles of LRPPRC on stemness of osteosarcoma cells, via regulating mitochondrial homeostasis potentially by regulating its downstream target genes. These findings suggest LRPPRC as a potential therapeutic target in osteosarcomas.
Accumulating studies have reported that the expression of LRPPRC is upregulated in several types of cancers, including lung adenocarcinoma [
25], esophageal squamous cell carcinoma [
22], gastric cancer [
23], colon cancer [
24], mammary and endometrial adenocarcinoma [
25], urothelial carcinoma of the bladder (UCB) [
26], and prostate cancer [
27]. This demonstrates that
LRPPRC functions as an oncogene. According to Wei et al. [
18], LRPPRC is a potential oncogene in UCB that regulates redox homeostasis via the circANKHD1/FOXM1 axis to enhance UCB tumorigenesis. Therefore, we hypothesized that LRPPRC may play a critical role in the development and progression of osteosarcomas. LRPPRC was found to be upregulated in several kinds of cancers, compared to adjacent tissues, however, in osteosarcoma, LRPPRC was found presenting same level compared to adjacent tissues. To reveal the role of LRPPRC, in vitro experiments in osteosarcomas were performed. These studies showed that LRPPRC knockdown inhibited malignancies, including proliferation, invasion, colony formation and tumor formation, in soft agar. Moreover, LRPPRC knockdown induced chemosensitivity and promoted apoptosis. Our results further demonstrate the role of LRPPRC as an oncogene that regulates osteosarcoma progression.
Despite great improvements in treatment strategies, progress in the prognosis of osteosarcoma patients has leveled off in recent decades. According to the CSC hypothesis, most current treatments target the majority of tumor masses, which consist of rapidly differentiating cells, rather than the relatively slow-growing migrating and resistant CSCs. The presence of CSCs with progenitor cell properties is the cause of treatment failures, such as tumor recurrence and metastasis. The characterization of CSCs has been of great interest due to the discovery and development of CSC-related therapies and the identification of key molecules involved in controlling the unique properties of the CSC population [
8]. In this study, we also evaluated the role of LRPPRC in regulating this subpopulation of osteosarcoma cells. CSCs derived from MG63 and U2OS cells express LRPPRC at relatively high levels compared to those of PCs. LRPPRC knockdown inhibited sphere formation in CSCs, indicating its role in maintaining stemness. Moreover, we also found that in LRPPRC-deficient osteosarcoma cells, the sphere formation capacity was markedly decreased. These results demonstrate that LRPPRC is a key regulator of the stemness of osteosarcomas. Considering that the existence of CSCs is critical for tumorigenesis and recurrence, LRPPRC might be a promising therapeutic target to avoid tumorigenesis and recurrence in osteosarcoma. However, we failed to evaluate the effects of LRPPRC on cell behaviors in an animal model, which is a limitation of this study.
Altered mitochondrial metabolism in cancer cells is known as the Warburg effect. This effect distinguishes tumor cells from normal cells [
28] and is considered a possible anticancer therapeutic target. We documented the altered mitochondrial status of osteosarcoma cells and osteosarcoma-derived CSCs after LRPPRC knockdown. This resulted in the loss of the mitochondrial membrane potential and lowered mitochondrial mass and ATP synthesis. This is consistent with our previous findings [
29] showing altered mitochondrial metabolism in pancreatic cancer. Notably, the mitochondrial membrane potential of LRPPRC-deficient osteosarcoma cells and osteosarcoma-derived CSCs remained unchanged under normoxic conditions and dramatically decreased under hypoxic conditions. Thus, we can also speculate that under normoxic conditions, LRPPRC might not be necessary to scavenge low levels of generated ROS and is critical for scavenging accumulated levels of ROS under oxidative stress conditions.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.