LncRNA-SARCC sensitizes osteosarcoma to cisplatin through the miR-143-mediated glycolysis inhibition by targeting Hexokinase 2

Abstract

Chemotherapy is one of the primary treatments used against cancer. Cisplatin is a conventional chemotherapy drug used to treat osteosarcoma; however, due to the development of cisplatin resistance, advantageous therapeutic outcomes and prognosis of osteosarcoma remain low. Thus, investigation of the specific targeted therapies to circumvent the anti-chemoresistance of osteosarcoma depends on understanding the molecular mechanisms underlying cisplatin resistance. Tumor cells display an increased utilization of glycolysis rather than oxidative phosphorylation. This phenomenon is called the “Warburg effect,” which presents a survival advantage for tumor cells, leading to chemoresistance. To date, the molecular mechanism underlying osteosarcoma cisplatin resistance remains to be fully elucidated. In this study, we reported the significant down-regulation of the long noncoding RNA-Suppressing Androgen Receptor in Renal Cell Carcinoma (lncRNA-SARCC) in the cells of osteosarcoma and in the specimens from osteosarcoma patients. Moreover, we observed a negative correlation between the lncRNA-SARCC and cisplatin resistance in the osteosarcoma tissues. Overexpression of the lncRNA-SARCC sensitizes osteosarcoma cells to cisplatin. From microarray analysis, we screened several miRNAs, which are significantly regulated by the lncRNA-SARCC in osteosarcoma cells, and revealed that lncRNA-SARCC promoted microRNA-43 (miR-143) expression in osteosarcoma. Interestingly, miR-143 showed the same expression pattern with the lncRNA-SARCC in osteosarcoma patient specimens. By establishing a cisplatin-resistant cell line from Sarcoma Osteogenic-2 (Saos-2), we found the cisplatin-resistant cells with down-regulated expressions of the lncRNA-SARCC and miR-143, but with a higher glycolysis rate compared to that in parental cells. We identified the glycolysis key enzyme, Hexokinase 2 (HK2), as a direct target for miR-143 in osteosarcoma. Restoration of the HK2 expression in the lncRNA-SARCC-overexpressing osteosarcoma cells reversed cisplatin resistance, suggesting that lncRNA-SARCC-mediated cisplatin sensitivity may be via glycolysis in the miR-143-inhibited osteosarcoma cells. Finally, results from both in vitro and in vivo xenograft models demonstrated that the lncRNA-SARCC was an effective therapeutic agent for overcoming cisplatin resistance in osteosarcoma. Our findings suggest an essential axis of the lncRNA-SARCC-miR-143-HK2 in regulation of osteosarcoma chemosensitivity, presenting the lncRNA-SARCC as a new therapeutic target against cisplatin-resistant osteosarcoma.

Keywords

lncRNA-SARCC microRNA-143 cisplatin resistance osteosarcoma the Warburg effect

1. Introduction

Osteosarcoma (OS) is the most common type of cancer that arises in the bones, generally found at the epiphysis of long bones. OS often presents as malignant bone tumors, which occurs predominantly in children and young adults, resulting in a poor prognosis and a high mortality rate [1]. Traditional therapeutic strategies against osteosarcoma are surgery, radiation, and adjuvant chemotherapy [2, 3].

Cisplatin is a widely used platinum-based antitumor agent used against solid tumors, including osteosarcoma, and acts by inducing DNA damage that inhibits tumor cell division and promotes apoptosis of the cancer cells [4]. Although advanced chemotherapy has significantly improved the prognosis for patients with osteosarcoma, a large number of cases subsequently develop chemoresistance, which remains a leading cause of death [5]. Thus, there is an urgent need to identify the molecular mechanisms responsible for the acquired cisplatin resistance in osteosarcoma.

Recent studies have revealed that the endogenous non-coding RNAs including microRNAs ( $\sim$ 22 nt) and the long non-coding RNA [lncRNA ( $\sim$ 200 nt)] are dysregulated in osteosarcoma, and play essential roles as oncogenes or tumor-suppressive genes during oncogenesis, tumor development, and chemoresistance of OS [6, 7, 8], depending on their specific mRNA targets. Currently, multiple lncRNAs have been reported to participate in the progression of osteosarcoma. Among them, the lncRNA-Hepatic Nuclear Factor Alpha-Antisense RNA 1 (HNF1A-AS1) may promote the progression of OS by regulating the Wingless Int-1 (Wnt)/ $\beta$ -catenin pathway [9]. Moreover, the lncRNA-Urothelial Cancer-Associated 1 (UCA1) has been reported to contribute to osteosarcoma tumorigenesis and progression, suggesting that lncRNA-UCA1 may be a novel OS prognostic marker [10]. Interestingly, recent studies have demonstrated that the lncRNA-SARCC suppresses the proliferation of renal cancer cells by modulating the androgen receptor [11, 12], suggesting the lncRNA-SARCC may have tumor-suppressive roles. However, the precise functions of the lncRNA-SARCC in osteosarcoma remain under investigation.

Tumor cells, which have higher nutritional demands owing to their rapid proliferation and display of distinct metabolic features compared to that in normal cells [13]. Tumor cells demonstrate a high rate of anaerobic glycolysis, a pathway that catalyzes the breakdown of glucose to allow energy to be harnessed in the form of ATP even in the presence of abundant oxygen [13, 14]. This process is known as the “Warburg effect,” which has been widely accepted as a new hallmark of cancer [13, 14]. Furthermore, the dysregulated glucose metabolism of cancer cells has been linked to chemoresistance [15], indicating that targeting the glycolytic pathway of cancer cells contributes to overcoming chemoresistance.

In the present study, we examined the differentially expressed lncRNA-SARCC and miR-143 in human osteosarcoma cells and specimens from patients. The biological functions of the lncRNA-SARCC and miR143 in OS cell lines and tissues were investigated particularly for their roles in cell growth and glycolysis. Additionally, investigation was performed to understand the molecular mechanisms of the lncRNA-SARCC-mediated cisplatin sensitization, using in vitro and in vivo models. The potential direct target of the lncRNA-SARCC and miR-143 was identified and verified. In summary, we propose that the lncRNA-SARCC-miR-143 axis may be a promising therapeutic target against cisplatin-resistant osteosarcoma.

2. Materials and methods

2.1 Cell culture and selection of the cisplatin-resistant cells

Human osteosarcoma cell lines [HOS, Sarcoma Osteogenic-2 (SaoS-2), MG63, 143B, and U2OS] and normal osteoblast cell lines, hFOB 1.19, were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cells were cultured in the Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen, CA, USA) supplemented with 10% fetal bovine serum (FBS) (Bio-Rad, Hercules, CA, USA) at 37 ${}^{\circ}$ C in a 5% CO ${}_{2}$ incubator. For the establishment of the cisplatin-resistant OS cells, SaoS-2 cells were treated for two months with gradually elevated concentrations of cisplatin (Sigma-Aldrich, Shanghai, China) from 2 $\mu$ M to 10 $\mu$ M. Survived cells were pooled for downstream experiments.

2.2 Osteosarcoma patient samples

Human OS specimens and their adjacent normal tissues were obtained from 40 patients including 20 cisplatin-sensitive and 20 cisplatin-resistant patients at the Department of Our institute, from August 2016 to December 2018. Tissues were frozen immediately in liquid nitrogen and stored at $-$ 80 ${}^{\circ}$ C until further use. This study was approved by the Institutional Review Board of Our institute. Informed consent was obtained from patients and all tissue specimens were evaluated pathologically. No patients received other cancer chemo- or radio-therapy prior to tissue resection.

2.3 Reagents and antibodies

Rabbit monoclonal antibodies against HK2 (#2867), LDHA (#3582), PKM2 (#4053) and $\beta$ -actin (#4970) were purchased from Cell Signaling Technology (Danvers, MA, USA). Cisplatin and 2-Deoxy-D-glucose (2-DG) were purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.4 Cell transfection

Lentivirus vectors, plasmid DNA, and miRNAs were transfected using the Lipofectamine ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}{\textregistered}}$ 2000 reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. The recombinant lentivirus encoding the lncRNA-SARCC and its shRNA were constructed from Hanbio Technology Co., Ltd. (Shanghai, China). The miRNA control and miR-143 precursor were synthesized by GenePharma Co., Ltd. (Shanghai, China). Hexokinase 2 (HK2)-overexpression vector and control vector were obtained from OriGene ${}^{\text{TM}}$ Technologies (Rockville, MD, USA). MiRNAs were transfected at 50 nM for 48 h. Plasmid DNA was transfected at 2 ug for 48 h.

2.5 RNA isolation and qRT-PCR

Total RNA from the OS patient tissues and cells were extracted with TRIzol ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}{\textregistered}}$ reagent (Thermo Fisher Scientific Inc., Waltham, MA, USA) followed by cDNA synthesis using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the manufacturer’s instructions. The quality and concentration of RNA samples were assessed by a Nanodrop spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA). For mRNA detection, 1 ug of total RNA was used for reverse transcription. Real-time PCR was performed using a SYBR ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}{\textregistered}}$ Green Real-Time PCR Master Mix (Thermo Fisher Scientific Inc., Waltham, MA, USA) and the MyCycler ${}^{\text{TM}}$ Thermal Cycler 580BR-3001 System (Bio-Rad, Hercules, CA, USA). Specific primer sequences were as follows: lncRNA-SARCC, F: 5’-AAA ACCATGATGTTGCCGCT-3’, R: 5’-AGCAGATTTCA CAGGCTCCT-3’; HK2, F: 5’-TGATCGCCTGCTTAT TCACGG-3’, R: 5’-AACCGCCTAGAAATCTCCAGA -3’; PKM2, F: 5’-TCGAGGAACTCCGCCGCCTG- 3’, R: 5’-CCACGGCACCCACGGCGGCA-3’; LDHA, F: 5’-TGTCTCCAGCAAAGACTACTGT-3’, R: 5’-G ACTGTACTTGACAATGTTGGGA-3’; GAPDH, F: 5’-GGAGCGAGATCCCTCCAAAAT-3’, R: 5’-GGCT GTTGTCATACTTCTCATGG-3’. For detection of the miR-143 expressions, 100 ng of the total RNA was used for cDNA synthesis using the TaqMan MicroRNA Reverse Transcription kit (Thermo Fisher Scientific Inc., Waltham, MA, USA). The primer used for the miR-143 detection was: 5’-GCGTGAGATGAAGCACTGTAGC TC-3’. The GAPDH gene or the U6 gene was used as the internal control for mRNA or miRNA, respectively. The PCR conditions were 25 cycles at 95 ${}^{\circ}$ C for 30 s, 56 ${}^{\circ}$ C for 30 s, and 72 ${}^{\circ}$ C for 1 min. The relative expressions of mRNA and miRNAs were calculated using the 2 ${}^{-\Delta\Delta\text{Ct}}$ method. All experiments were performed in triplicate and repeated three times.

2.6 Cell proliferation and viability assays

Osteosarcoma cells were seeded into 24-well plates at 3 $\times$ 10 ${}^{3}$ cells per well and cultured overnight. Then cells were cultured for 48 h, 72 h, or 96 h. Cell number was calculated using a Trypan blue direct cell counting method. Cell viability was examined with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were seeded into 96-well plates at a density of 5 $\times$ 10 ${}^{4}$ cells per well and cultured overnight. After cells were treated with different concentrations of cisplatin for 48 h, 250 $\mu$ l of 5 mg/ml MTT was added into each well and incubated for 4 h in an incubator containing 5% CO ${}_{2}$ at 37 ${}^{\circ}$ C. Media were removed and 150 $\mu$ l of DMSO was added. Plates were covered with tinfoil and placed on an orbital shaker for 15 min. The absorbance output read at 570 nm using a SpectraMax ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}{\textregistered}}$ M2 plate reader (Molecular Devices, LLC., San Jose, CA, USA). Experiments were performed in triplicate and repeated three times.

2.7 Measurements of cellular glycolysis rate

The glucose uptake was measured using a Glucose Uptake Colorimetric Assay Kit (#MAK083, Sigma-Aldrich, Shanghai, China) according to the manufacturer’s instructions. The lactate product was measured using a Lactate Assay Kit (#MAK064, Sigma-Aldrich, Shanghai, China) according to the manufacturer’s instructions. The OD value of each sample was normalized by total cell numbers in each treatment. Experiments were performed in triplicate and repeated three times.

2.8 Luciferase assay

The luciferase assay was performed according to a previously described method [16]. Briefly, cells were seeded into 24-well plates and transfected with the pGL3-reporter luciferase vector containing the WT 3’UTR of HK2 or the mutant 3’UTR of HK2 by the Lipofectamine method according to the manufacturer’s instructions. Control miRNAs or miR-143 precursor were co-transfected with the luciferase vector to examine the luciferase activity. After 48 h post-transfection, luciferase activity was measured using a Dual-luciferase reporter assay system (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The ratio of Firefly-to-Renilla luciferase activity was calculated for the relative luciferase activity. Experiments were performed in triplicate and repeated three times.

2.9 Mice xenograft experiment

A total of forty 4-week-old female nude mice (BALB/c) were purchased from the Beijing Weitong Lihua Experimental Animal Technology Co. Ltd (Beijing, China) and housed under traditional conditions. Mice had free access to food and water and exposed to the natural light-dark cycle. In total, 50 $\mu$ l of the Saos-2 cisplatin-resistant cells with 2 $\times$ 10 ${}^{8}$ cells/mL concentration was injected subcutaneously into the flanks of mice. When the tumor size reached 100 mm ${}^{3}$ , mice were separated into four groups, and each group was randomly assigned ten nude mice. The four groups were intraperitoneally injected with normal saline, cisplatin alone, the lncRNA-SARCC alone, or cisplatin plus lncRNA-SARCC twice a week at a dosage of 15 mg/kg. The survival rates of the mice were calculated each day. After 8 weeks of treatments, tumors from the dead mice or the surviving mice were removed, then fixed with 4% paraformaldehyde for downstream experiments. Animal experiments were approved by the Animal Ethics committee from the Institutional Review Board of the Renmin Hospital of Wuhan University and were in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC).

2.10 Immunohistochemistry

The staining of HK2 in the human OS tissues, normal adjacent tissues, and mice tumors were performed according to the methods described by previous reports [12]. Samples were de-paraffinized with xylene, then dehydrated with a graded series of alcohol. Samples on the slides were induced for antigen retrieval by heat treatment in 0.1 M citrate buffer. After blocking with IHC blocking buffer, slides were incubated with primary antibodies followed by secondary antibodies after washing with phosphate-buffered saline (PBS). Experiments were repeated three times.

2.11 Western blot

Cells were directly lysed in a RIPA lysis buffer (Beyotime, Shanghai, China) with a 1X protease inhibitor cocktail (Sigma, Shanghai, China) then placed on ice for 15 min, followed by centrifugation at 12,000 RPM for 15 min at 4 ${}^{\circ}$ C. Proteins in the supernatants were collected and quantified by the Bradford method. Equal amounts of protein for each sample were loaded into 10% SDS-PAGE. After separation by SDS-PAGE, proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (Life Technologies, Carlsbad, CA, USA). Membranes were blocked with Tris-buffered saline, 0.1% Tween 20 (TBST) buffer containing 5% BSA for 1 h at room temperature, and incubated overnight with primary antibodies at 4 ${}^{\circ}$ C. After washing with TBST, the blots were incubated with horseradish peroxidase-conjugated anti-rabbit secondary antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for 2 h at room temperature. The protein bands were detected by the chemiluminescence method using the Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific Inc., Waltham, MA, USA). Experiments were repeated three times.

2.12 Statistical analysis

Experiments were performed in triplicate and repeated three times. Data have been presented as means $\pm$ standard deviation (SD). Differences between two samples were analyzed by the Student’s $t$ -test using GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA) for significance. Comparisons of more than two groups were performed using one-way ANOVA analysis. $p<$ 0.05 was considered statistically significant.

Figure 1.

Downregulation of lncRNA-SARCC in human osteosarcoma. (A) Human OS ( $n=$ 25) and their corresponding non-cancerous tissues ( $n=$ 25) were subjected to qRT-PCR analysis to compare the expressions of lncRNA-SARCC. (B) Expressions of lncRNA-SARCC in five human osteosarcoma cell lines and one normal osteoblast cell line were detected by qRT-PCR. GAPDH was used as an internal control for lncRNA detection. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01; ${}^{***}$ , $p<$ 0.001.

3. Results

3.1 LncRNA-SARCC is dow-nregulated in osteosarcoma and negatively correlated with cisplatin resistance

Recent studies have demonstrated that the lncRNA-SARCC suppresses renal cell carcinoma (RCC) progression through modification of the androgen receptor [11, 12], suggesting LncRNA-SARCC may play a tumor-suppressive role. To elucidate the role of the lncRNA-SARCC in human osteosarcoma, we compared the expression levels of the lncRNA-SARCC in specimens from 25 OS patients and their corresponding levels in normal tissues. As hypothesized, qRT-PCR analysis showed that the lncRNA-SARCC was significantly down-regulated in OS tissues compared to the levels in the normal bone tissues (Fig. 1A). Furthermore, we evaluated the expression of the lncRNA-SARCC in five human osteosarcoma cell lines, HOS, Saos-2, MG63, 143B, and U2OS, as well as in hFOB1.19, a normal osteoblast cell line by real-time quantitative PCR. The expression of the lncRNA-SARCC was significantly less in the OS cells compared with that in the hFOB1.19 osteoblasts (Fig. 1B). To further elucidate the role of the lncRNA-SARCC in chemoresistance of human osteosarcoma, we evaluated the association between its expression and resistance to cisplatin using specimens from 20 cases of cisplatin-sensitive and 20 cases of cisplatin-resistant osteosarcoma. As shown in Fig. 2A, the results of qRT-PCR showed that the expression of the lncRNA-SARCC was significantly down-regulated in the cisplatin-resistant OS tissues compared with those of the cisplatin-sensitive tissues. To investigate the molecular mechanisms involved in the lncRNA-SARCC-mediated cisplatin resistance, we established a cisplatin-resistant in vitro model using the Saos-2 cell line. To verify cisplatin resistance, a cell viability assay was performed by treating the Saos-2 parental and the resistant cells with different concentrations of cisplatin. Parental Saos-2 cells showed a gradual reduction in cell viabilities with cisplatin treatments (Fig. 2B). However, the IC ${}_{50}$ of the Saos-2 cisplatin resistant-cells was significantly higher compared with the IC ${}_{50}$ of parental cells (Fig. 2B). Consistent with the above-mentioned clinical results of the OS patients, the cisplatin-resistant OS cells displayed a significant down-regulated lncRNA-SARCC expression compared with the expression in the parental cells (Fig. 2C). To investigate the direct functions of the lncRNA-SARCC in cisplatin sensitivity, we transfected lncRNA-SARCC into the Saos-2 and the U2OS cells (Fig. 2D). Cell viability assays were performed to examine the sensitivity of OS cells without or with the lncRNA-SARCC overexpression. Results from Fig. 2E and F demonstrated that osteosarcoma cells with exogenous lncRNA-SARCC were more sensitive to cisplatin. In summary, these data demonstrate a negative correlation between the lncRNA-SARCC expression and cisplatin resistance in osteosarcoma, suggesting lncRNA-SARCC may be selected as a new therapeutic target against chemoresistance in osteosarcoma.

3.2 LncRNA-SARCC positively regulates miR-143 in osteosarcoma

Next, we investigated the molecular mechanisms involved in the lncRNA-SARCC-modulated cisplatin resistance in osteosarcoma. Accumulating evidence revealed that microRNAs play important roles in biological processes of various human-associated tumors [17]. To identify potential miRNAs involved in chemoresistance, we performed microRNA microarrays to screen differentially expressed miRNAs from the control cells and the lncRNA-SARCC-overexpressed osteosarcoma cells. Results from Fig. 3A illustrated that among these miRNAs, miR-143 was significantly up-regulated in the lncRNA-SARCC-overexpressed SaoS-2 and the lncRNA-SARCC-overexpressed U2OS cells, suggesting miR-143 may be located downstream regulator of the lncRNA-SARCC and may regulate cisplatin sensi-

Figure 2.

Negative correlation between lncRNA-SARCC and cisplatin resistance in osteosarcoma. (A) Expressions of lncRNA-SARCC were detected from cisplatin-sensitive ( $n=$ 20) and cisplatin-resistant ( $n=$ 20) osteosarcoma patient specimens by qRT-PCR. U6 was used as an internal control. (B) Establishment and identification of the SaoS-2 cisplatin-resistant cells. The SaoS-2 parental cells were treated with gradually elevated concentrations of cisplatin for 2 months. The surviving cells were pooled for downstream assay. (C) Expressions of the lncRNA-SARCC were detected from the SaoS-2 parental cells and cisplatin-resistant cells by qRT-PCR. (D) SaoS-2 and U2OS cells were transfected with control vector or the lncRNA-SARCC-overexpression vector for 48 h. Expressions of the lncRNA-SARCC were detected by qRT-PCR. GAPDH was used as an internal control for lncRNA detection. (E) SaoS-2 and (F) U2OS cells without or with lncRNA-SARCC overexpression were treated with cisplatin at the indicated concentrations for 48 h, the cell viability was determined by MTT assay. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01; ${}^{***}$ , $p<$ 0.001.

Figure 3.

LncRNA-SARCC positively regulates miR-143. (A) Selective microRNA microarray results demonstrated miR-143 was up-regulated by the lncRNA-SARCC in the SaoS-2 and the U2OS cells. (B) Five human osteosarcoma cells were transfected with control vector and the lncRNA-SARCC for 40 h, followed by miR-143 detection by qRT-PCR. (C) Positive correlation between lncRNA-SARCC and miR-143 expressions in human OS patient specimen. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01; ${}^{***}$ , $p<$ 0.001.

Figure 4.

MiR-143 is negatively correlated with cisplatin resistance. (A) miR-143 expressions were detected by qRT-PCR and compared in the human OS ( $n=$ 25) and their corresponding non-cancerous tissues ( $n=$ 25). (B) miR-143 expressions were detected by qRT-PCR and compared in the cisplatin-sensitive and the cisplatin-resistant osteosarcoma patient specimens. (C) miR-143 expressions were detected by qRT-PCR in the SaoS-2 parental cisplatin-resistant cells. U6 was used as an internal control for miR-143. (D) SaoS-2 and U2OS cells were transfected with control miRNAs or miR-143 precursor for 48 h. Expressions of miR-143 were measured by qRT-PCR. (E) SaoS-2 and (F) U2OS cells were transfected with control miRNAs or miR-143 precursor for 48 h. Cells were then subjected to elevated cisplatin treatments for 48 h, followed by cell viability assay. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01; ${}^{***}$ , $p<$ 0.001.

Figure 5.

LncRNA-SARCC suppresses growth and glycolysis of osteosarcoma cells through miR-143. (A) The glucose uptake and lactate product assays were performed in the SaoS-2 parental and the cisplatin-resistant cells. (B) The SaoS-2 and U2OS cells were transfected with control, the lncRNA-SARCC vector, miR-143 precursor or the lncRNA-SARCC plus miR-143 for 48 h. Cells were seeded in 12-well plate for 48 h, followed by cell growth measurement by the MTT assay. (C) The SaoS-2 and U2OS cells were transfected with control, the lncRNA-SARCC vector, miR-143 precursor or the lncRNA-SARCC plus miR-143 for 48 h, the glucose uptake and (D) lactate product were detected. (E) SaoS-2 cells were transfected with the above vectors for 48 h. The protein and mRNA expressions of HK2, PKM2 and LDHA were detected by Western blotting and (F) qRT-PCR assay. $\beta$ -actin and GAPDH were used as internal controls for Western blotting and qRT-PCR, respectively. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01; ${}^{***}$ , $p<$ 0.001.

Figure 6.

MiR-143 directly targets HK2, leading to sensitization of osteosarcoma to cisplatin. (A) Predicted miR-143 binding sites on the 3’-UTR of HK2 mRNA by Targetscan.org. (B) Conserved miR-143 binding sites on 3’-UTR of HK2 mRNA in multiple species analyzed by Targetscan.org. (C) The SaoS-2 (upper) and U2OS (lower) cells were transfected with control miRNAs and miR-143 precursor for 48 h, expressions of HK2 were detected by Western blotting. $\beta$ -actin was used as an internal control. (D) Luciferase reporter vector with WT or binding sites mutant 3’-UTR was co-transfected with control miRNA or miR-143 precursor into SaoS-2 (upper) and U2OS (lower) cells. The luciferase activity was detected according to the manufacturer’s instructions for the reagent. (E) Negative correlation between miR-143 and HK2 mRNA expressions was observed in osteosarcoma patient specimen. (F) IHC staining of HK2 in the osteosarcoma tissue specimens and corresponding normal tissue specimens. (G) The Saos-2 (upper) and U2OS (lower) cells were transfected with control or lncRNA-SARCC-overexpression plasmid and control siRNA or lncRNA-SARCC-siRNA for 48 h, and the expressions of HK2 were detected by Western blotting. (H) The SaoS-2 (upper) and U2OS (lower) cells were transfected with negative control, the lncRNA-SARCC inhibitor for 48 h. Cells were then treated with cisplatin alone or cisplatin plus 2DG for 48 h, followed by a cell viability assay. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01; ${}^{***}$ , $p<$ 0.001.

Figure 7.

LncRNA-SARCC sensitizes osteosarcoma cells to cisplatin through miR-143-mediated glycolysis inhibition in vitro and in vivo. (A) The SaoS-2 cisplatin-resistant cells were transfected with negative control, the lncRNA-SARCC or the lncRNA-SARCC plus HK2 for 48 h. Cell were then treated with cisplatin at 0 uM, 2.5 uM, 5 uM, 10 uM, 20 uM, or 40 uM for 48 h, followed by a cell viability assay. (B) Four groups mice were treated with control, cisplatin alone, the lncRNA-SARCC alone or the lncRNA-SARCC plus cisplatin. Tumor sizes were measured every 5 d. (C) Tumors were dissected and visualized. (D) Survival curves of the four groups mice with treatments of control, cisplatin alone, the lncRNA-SARCC alone or the lncRNA-SARCC plus cisplatin. (E) Tumors were dissected from the above four groups mice. Total RNAs were isolated and the expressions of miR-143. U6 was used as an internal control for miR-143 detection. (F) IHC staining of HK2 protein from xenograft tumors of the above-mentioned treated mice. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.0; ${}^{***}$ , $p<$ 0.001.

tivity. We performed qRT-PCR assays in multiple OS cells, which were transfected with the control or the lncRNA-SARCC overexpression vector. As shown in Fig. 3B, the expression of miR-143 was significantly elevated in the lncRNA-SARCC-overexpressed osteosarcoma cells. To further confirm that lncRNA-SARCC increased the miR-143 expression, we assessed the correlation between the lncRNA-SARCC and miR-143 expressions in osteosarcoma patient specimens. As hypothesized, miR-143-overexpressing OS tissues accompanied a relative increase in the lncRNA-SARCC expression (Fig. 3C). Collectively, the above results from in vitro cell lines and the OS specimens demonstrated that the lncRNA-SARCC up-regulated the miR-143 expression in osteosarcoma.

3.3 Overexpression of miR-143 sensitizes osteosarcoma cells to cisplatin

We explored the roles of miR-143 in cisplatin resistance. Recent studies revealed that miR-143 was down-regulated in multiple cancers and contributed to anti-cancer therapy [18]. As hypothesized, we detected a down-regulated expression of miR-143 in osteosarcoma patient tissues compared with the expression in the corresponding normal bone tissues (Fig. 4A). Importantly, miR-143 expression was significantly low in the cisplatin-resistant OS patient specimens compared with that from the cisplatin-sensitive OS patient specimens (Fig. 4B). Consistent with these results, miR-143 expression was significantly down-regulated in the SaoS-2 cisplatin-resistant cells (Fig. 4C). To assess the direct function of miR-143 in cisplatin resistance, we transfected miR-143 into the SaoS-2, and the U2OS cells (Fig. 4D) then detected the cisplatin sensitivity. Cell viability assays showed that overexpression of miR-143 sensitized osteosarcoma cells to cisplatin. The IC ${}_{50}$ decreased to 2.25 uM from 8.65 uM and to 2.85 uM from 9.33 uM, respectively (Fig. 4E and F). In summary, the above results demonstrated that overexpression of miR-143 was able to sensitize OS cells to cisplatin.

3.4 LncRNA-SARCC suppresses growth and cellular glycolysis of osteosarcoma through upregulating miR-143

It has been well-documented that malignant cells demand accelerated glucose and energy metabolism through a higher rate of anaerobic glycolysis, even in the presence of sufficient oxygen. This phenomenon was known as the “Warburg effect” [13, 14]. In addition, the Warburg effect has been strongly correlated to chemoresistance; for example, cancer cells with a higher glycolytic rate have been frequently associated with chemoresistance [15]. As hypothesized, the cisplatin-resistant OS cells showed an increased glycolytic rate (Fig. 5A). To determine the effect of the lncRNA-SARCC-mediated miR-143 up-regulation on the cellular glucose metabolism of OS cells, we transfected a control vector, lncRNA-SARCC overexpression vector, miR-143 precursor or lncRNA-SARCC overexpression vector, and miR-143 inhibitor into the SaoS-2 cells. Cell proliferation assay demonstrated overexpression of the lncRNA-SARCC significantly suppressed cell growth, glucose uptake, and lactate production of the OS cells (Fig. 5B–D), suggesting the the lncRNA-SARCC-modulated cisplatin sensitization may be achieved by glycolytic inhibition. Similar results were observed in the miR-143-overexpressed SaoS-2 cells (Fig. 5B–D). Importantly, OS cells with co-transfection of the lncRNA-SARCC and the miR-143 inhibitor showed recovery in both cellular growth and glycolytic capacity (Fig. 5B–D), indicating the miR-143 was a downstream effector of the lncRNA-SARCC, which led to the suppression of the growth of OS cells and suppression of glycolysis through positively-regulating miR-143. To further confirm the aforementioned observations, we examined the protein and mRNA expressions of the glycolysis rate-limiting enzymes, Hexokinase 2 (HK2), Pyruvate kinase M2 (PKM2), and Lactate dehydrogenase A (LDHA). As hypothesized, overexpression of the lncRNA-SARCC and miR-143 significantly inhibited protein and mRNA expressions of these enzymes (Fig. 5E and F). Meanwhile, inhibition of the lncRNA-SARCC-regulated miR-143 significantly restored the cell growth and glycolytic capacity of the SaoS-2 cells (Fig. 5E and F). These results revealed that the lncRNA-SARCC-mediated miR-143 up-regulation directly contributed to the glycolytic suppression in OS cells.

3.5 The lncRNA-SARCC-miR-143-mediated osteosarcoma glycolysis inhibition is through direct targeting HK2

Empirical evidence has shown that microRNAs bind to the 3’-untranslated region (3’-UTR) of target mRNAs to down-regulate their expressions. Next, we explored the TargetScan database in search of potential targets of miR-143. Among the candidates, we found in the 3’-UTR of the human HK2, an essential glycolytic enzyme containing a putative miR-143 seeding sequence binding region (2291 nt $-$ 2298 nt) (Fig. 6A). Importantly, this binding site is conserved in multiple species, including mouse and rat (Fig. 6B), suggesting that miR-143-HK2 interaction may play a vital role during development and evolution. To examine whether HK2 was indeed the target of miR-143, we transfected the miR-143 precursor or control miR into the SaoS-2 and the U2OS cell lines. Overexpression of miR-143 significantly decreased the protein expression of HK2 compared to that in the control group (Fig. 6C). To verify whether miR-143 could directly bind to the 3’-UTR of HK2 mRNA, a luciferase reporter assay was performed with co-transfection of control miRNA, miR-143, and WT-HK2 3’-UTR or mutant HK2 3’-UTR into SaoS-2 and U2OS cells. Luciferase assay showed that miR-143 significantly decreased the luciferase activity of HK2 wild type 3 $\prime$ -UTR compared with the binding site of the mutant group (Fig. 6D). Furthermore, the Spearman’s correlation analysis was performed and showed that the expression of miR-143 were negatively correlated with HK2 mRNA expressions in the OS patient specimens (Fig. 6E). IHC staining of HK2 demonstrated that HK2 was up-regulated in the OS tissues compared with HK2 levels in the corresponding normal tissues (Fig. 6F). Moreover, consistent results showed that the upstream molecule of miR-143 and the lncRNA-SARCC negatively regulated the HK2 protein expression in the OS cell lines (Fig. 6G). Taken together, these data illustrated that HK2, an oncoprotein, was a direct target gene of miR-143 and could be inhibited by lncRNA-SARCC in osteosarcoma cells.

To test whether the lncRNA-SARCC-mediated glycolytic inhibition led to cisplatin sensitization, we transfected the Saos-2 (Fig. 6F) and the U2OS (Fig. 6G) cells with the control vector, the lncRNA-SARCC inhibitor, or the lncRNA-SARCC inhibitor plus the glycolytic inhibitor, 2DG. Consistent results demonstrated that inhibition of the lncRNA-SARCC significantly decreased the cisplatin sensitivity of the OS cells. At the same time, co-treatment with the glycolytic inhibitor was able to reverse the cisplatin resistance caused by lncRNA-SARCC inhibition.

3.6 Sensitization of cisplatin-resistant osteosarcoma cells by lncRNA-SARCC-miR-143-glycolysis axis in vitro and in vivo

Finally, we investigated whether inhibition of cellular glycolysis of the osteosarcoma cells by lncRNA-miR-143 could sensitize the cisplatin-resistant cells. The SaoS-2 cisplatin-resistant cells were transfected with the control vector, the lncRNA-SARCC-overexpression vector, or the lncRNA-SARCC-plus-HK2-overexpres- sion vectors. Results from Fig. 7A demonstrated that the exogenous lncRNA-SARCC significantly sensitized the cisplatin-resistant SaoS-2 cells. As hypothesized, restoration of HK2 in the lncRNA-SARCC-overexpressing cells significantly recovered the cisplatin sensitivity (Fig. 7A). These results further indicated that HK2 was a major target of the LncRNA-SARCC-miR-143 axis and was a response to cisplatin resistance of the osteosarcoma cells.

Figure S1.

HK2 mRNA was detected from xenograft tumors from mice with the indicated treatments by qRT-PCR. GAPDH was used an internal control for mRNA detection. Data have been shown as mean $\pm$ S.D. ( $n=$ 3). ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.0; ${}^{***}$ , $p<$ 0.001.

To verify the above in vitro results, we studied the functions of the lncRNA-SARCC using an in vivo xenografted osteosarcoma model. We investigated whether the combination of lncRNA-SARCC with cisplatin could significantly increase the survival rate. The SaoS-2 cisplatin-resistant cells were subcutaneously injected into the mammary fat pads of nude mice to establish xenograft tumors. After one week, mice were intraperitoneally injected weekly for two months with control PBS, cisplatin alone, the lncRNA-SARCC lentivirus alone, or cisplatin plus lncRNA-SARCC. We observed limited tumor sizes in mice from the lncRNA-SARCC-treated group compared to the tumor sizes in the control or the cisplatin-treated alone group (Fig. 7B and C). As hypothesized, the combination of cisplatin plus lncRNA-SARCC dramatically reduced the size of tumors in the xenograft mice (Fig. 7B and C). Moreover, mice died with the control or lone treatment with cisplatin (Fig. 7D). The lncRNA-SARCC treatment slightly increased the survival rate of the xenografted mice, but the combination of cisplatin plus lncRNA-SARCC dramatically showed a prolonged survival rate (Fig. 7D). Two months after injection, survival mice were sacrificed, and tumor tissues were obtained for analysis of the miR-143 and the HK2 mRNA expressions. Consistent with the in vitro results, the lncRNA-SARCC treatment significantly decreased the miR-143 and HK2 mRNA expressions (Fig. 7E and F, and Fig. S1). Furthermore, the combination of cisplatin plus lncRNA-SARCC exerted a synergistically inhibitory effect on miR-143 and HK2 expression (Fig. 7E and F, and Fig. S1), suggesting that lncRNA-SARCC may be an effective therapeutic agent against cisplatin resistance in osteosarcoma.

4. Discussion

LncRNAs are non-protein-coding transcripts, which have recently gained widespread attention in the diverse process of tumorigenesis, particularly for regulatory noncoding RNAs [19]. Emerging evidence has shown that lncRNAs play essential roles in tumorigenesis and chemoresistance of osteosarcoma [20]. A recent study has revealed that the lncRNA-SARCC was able to suppress the progression of renal cell carcinoma (RCC) through regulation of the androgen receptor (AR) [11, 12], suggesting that lncRNA-SARCC may have tumor-suppressive roles, which contribute to therapeutic potential. However, little is known about the function of the lncRNA-SARCC in osteosarcoma. In the present study, we revealed that lncRNA-SARCC was significantly down-regulated in the human osteosarcoma tissues and cells compared with that in the surrounding non-tumor cells and normal osteoblast cells. In addition, we observed a negative correlation between the lncRNA-SARCC expression and cisplatin resistance in the osteosarcoma tissues and cells, which was consistent with the tumor-suppressive roles of the lncRNA-SARCC in RCC [12].

Cancer cells prefer the utilization of glycolysis more than the utilization of the mitochondrial oxidative phosphorylation (OXPHOS) for their energy requirements, also known as the “Warburg effect” [13, 14], which has been extensively studied and is recognized as a new hallmark of cancer. In addition, the metabolic characteristics of cancer cells provide an advanced environment for developing drug resistance [15]. Multiple studies have reported that increased cellular glycolysis contributes to chemoresistance of cancers, such as resistance against 5-Fu [21], doxorubicin [22], and cisplatin [23]. Meanwhile, the strong correlation between the lncRNAs and the metabolic reprogramming of cancer cells has been described [24]. Glucose Transporter 1 (GLUT1), a glycolysis transporter, has been reported as a downstream effector of the lncRNA-antisense non-coding RNA in the INK4 locus (ANRIL), which upregulates GLUT1 and LDHA expressions in nasopharyngeal cancer, resulting in an increased glucose uptake [25]. In addition, studies have demonstrated that lncRNA-UCA1 is overexpressed and promotes glycolysis in the colon cancer cells by up-regulating HK2 [26]. In this study, we have described a negative glycolytic regulator, the lncRNA-SARCC, which suppresses glycolysis of the osteosarcoma cells via inhibition of HK2, suggesting that the lncRNA-SARCC may be be selected as a new anti-glycolytic agent for osteosarcoma therapy.

Furthermore, we observed that the pyruvate kinase isoenzyme M2 (PKM2) and LDHA were also regulated by lncRNA-SARCC and miR-143, respectively. It may be possible that they are indirectly regulated by SARCC and miR-143 in addition to HK2, which is directly targeted by miR-143. Since the HK2-catalyzed reaction (conversion of glucose to glucose-6-phosphate) is in the upstream of the glycolysis pathway, and PKM2 and LDHA-catalyzed reactions are in the relative downstream, the higher level of Glu-6-P in the cytosol of cancer cells will activate other pathways to synthesize (or stabilize) more PKM2 proteins and LDHA proteins. Therefore, lncRNA-SARCC and miR-143 also regulated the expression of PKM2 and LDHA through indirect mechanisms.

In our future work, a more comprehensive in vivo mice model will be established to investigate the precise molecular mechanisms for the lncRNA-SARCC-mediated chemosensitivity.

Although the lncRNA-mediated pathways have been revealed to be responsible in chemoresistance of OS, the detailed molecular mechanisms remain be fully elucidated. In this study, we demonstrated that the cisplatin-resistant OS cells displayed an up-regulated glycolytic capacity, which was found to be consistent with the results of previous reports [27]. The lncRNA-SARCC, and miR-143 expression were significantly down-regulated in the cisplatin-resistant OS cells. Moreover, we revealed that the lncRNA-SARCC up-regulates the expression of miR-143, which was able to directly inhibit the rate-limiting glycolytic enzyme, HK2 from both in vitro and in vivo xenograft model. Importantly, restoration of HK2 expression in the lncRNA-SARCC-overexpressing OS cells recovered cisplatin resistance, suggesting that specific lncRNA-SARCC-mediated cisplatin sensitivity may be via miR-143-inhibited osteosarcoma cell glycolysis. Despite the previous studies that have reported the down-regulation of the lncRNA-SARCC in renal cancer and the miR-143 expression by the lncRNA-SARCC, our results are the first to report the tumor-suppressive roles of the lncRNA-SARCC in osteosarcoma and a correlation between lncRNA-SARCC and cisplatin resistance. A previous study has shown that lncRNA-SARCC up-regulated miR-143 through the androgen receptor [12]. In OS, other mechanisms may also be involved in the lncRNA-SARCC-mediated cisplatin sensitivities. However, this study did not evaluate the direct mechanisms for upregulation of miR-143 by lncRNA-SARCC in osteosarcoma. In addition, the lncRNA-SARCC also regulated other microRNAs from our array data. The alternative microRNAs or signaling pathways, responsible for the lncRNA-SARCC-promoted cisplatin sensitization, were not investigated in this study.

In conclusion, the above-mentioned in vivo and in vitro results revealed that lncRNA-SARCC was negatively associated with cisplatin resistance in osteosarcoma, and overexpression of the lncRNA-SARCC significantly enhanced the chemosensitivity of the osteosarcoma cells. We described that the lncRNA-SARCC positively regulates miR-143 expression, which was directly able to bind to the 3’-UTR of HK2, resulting in a glycolytic inhibition, affecting the chemosensitivity of osteosarcoma. These findings suggest a potentially important axis of lncRNA-SARCC-miR-143-HK2 in the regulation of the osteosarcoma chemosensitivity and presentation of the lncRNA-SARCC may serve as a new therapeutic target against cisplatin-resistant osteosarcoma.

References

Morrow

J.J.

and Khanna

, Osteosarcoma genetics and epigenetics: Emerging biology and candidate therapies, Critical Reviews in Oncogenesis 20 (2015), 173–197.

Lamplot

J.D.

et al., The current and future therapies for human osteosarcoma, Current Cancer Therapy Reviews (2013), 55–77.

Xie

and Guo

, Anti-angiogenesis target therapy for advanced osteosarcoma, Oncology Reports 38(2) (2017), 625–636.

Dasari

and Tchounwou

P.B.

, Cisplatin in cancer therapy: Molecular mechanisms of action, European Journal of Pharmacology 740 (2014), 364–378.

Amable

, Cisplatin resistance and opportunities for precision medicine, Pharmacological Research 106 (2016), 27–36.

Lou

et al., Long non-coding RNA KDM5B anti-sense RNA 1 enhances tumor progression in non-small cell lung cancer, Journal of Clinical Laboratory Analysis (2019), e22897.

Kumar

M.M.

and Goyal

, LncRNA as a therapeutic target for angiogenesis, Current Topics in Medicinal Chemistry 17 (2017), 1750–1757.

et al., LncRNA NEAT1 promotes autophagy via regulating miR-204/ATG3 and enhanced cell resistance to sorafenib in hepatocellular carcinoma, Journal of Cellular Physiology (2019), 3402–3413.

Zhang

et al., Long noncoding RNA HNF1A-AS1 indicates a poor prognosis of colorectal cancer and promotes carcinogenesis via activation of the Wnt/β-catenin signaling pathway, Biomedicine & Pharmacotherapy (2017), 877–883.

10.

Xiao

and Huang

, HIF-1α-induced upregulation of lncRNA UCA1 promotes cell growth in osteosarcoma by inactivating the PTEN/AKT signaling pathway, Oncology Reports 39(3) (2018), 1072–1080.

11.

Zhai

et al., Differential regulation of LncRNA-SARCC suppresses VHL-mutant RCC cell proliferation yet promotes VHL-normal RCC cell proliferation via modulating androgen receptor/HIF-2α/C-MYC axis under hypoxia, Oncogene (2017), 4525.

12.

Zhai

et al., LncRNA-SARCC suppresses renal cell carcinoma (RCC) progression via altering the androgen receptor (AR)/miRNA-143-3p signals, Cell Death and Differentiation (2017), 1502–1517.

13.

Lebelo

M.T.

Joubert

A.M.

and Visagie

M.H.

, Warburg effect and its role in tumourigenesis, Archives of Pharmacal Research 42(10) (2019), 833–847.

14.

X.D.

et al., Warburg effect or reverse Warburg effect? A review of cancer metabolism, Oncology Research and Treatment (2015), 117–122.

15.

Bhattacharya

Mohd Omar

M.F.

and Soong

, The Warburg effect and drug resistance, British Journal of Pharmacology 173(6) (2016), 970–979.

16.

et al., HOTAIR regulates HK2 expression by binding endogenous miR-125 and miR-143 in oesophageal squamous cell carcinoma progression, Oncotarget (2017), 86410–86422.

17.

Rupaimoole

and Slack

F.J.

, MicroRNA therapeutics: Towards a new era for the management of cancer and other diseases, Nature Reviews. Drug Discovery 16 (2017), 203–222.

18.

Farooqi

A.A.

et al., MicroRNA-143 as a new weapon against cancer: Overview of the mechanistic insights and long non-coding RNA mediated regulation of miRNA-143 in different cancers, Cellular and Molecular Biology (2019), 1–5.

19.

Bhan

Soleimani

and Mandal

S.S.

, Long noncoding RNA and cancer: A new paradigm, Cancer Research 77(15) (2017), 3965–3981.

20.

Schmitt

A.M.

and Chang

H.Y.

, Long noncoding RNAs in cancer pathways, Cancer Cell 29 (2016), 452–463.

21.

Chen

et al., Baicalein reverses hypoxia-induced 5-FU resistance in gastric cancer AGS cells through suppression of glycolysis and the PTEN/Akt/HIF-1α signaling pathway, Oncology Reports (2015), 457–463.

22.

et al., Targeting cellular metabolism chemosensitizes the doxorubicin-resistant human breast adenocarcinoma cells, BioMed Research International (2015), 453986.

23.

et al., Overcoming cisplatin resistance of ovarian cancer cells by targeting HIF-1-regulated cancer metabolism, Cancer Letters (2016), 36–44.

24.

Sun

et al., Emerging roles of long non-coding RNAs in tumor metabolism, Journal of Hematology & Oncology (2018), 106.

25.

Zou

Z.W.

et al., LncRNA ANRIL is up-regulated in nasopharyngeal carcinoma and promotes the cancer progression via increasing proliferation, reprograming cell glucose metabolism and inducing side-population stemlike cancer cells, Oncotarget (2016), 61741–61754.

26.

Mathupala

S.P.

Y.H.

and Pedersen

P.L.

, Hexokinase-2 bound to mitochondria: cancer’s stygian link to the “Warburg Effect” and a pivotal target for effective therapy, Seminars in Cancer Biology 19(1) (2009), 17–24.

27.

Song

Y.D.

et al., miR-214 modulates cisplatin sensitivity of osteosarcoma cells through regulation of anaerobic glycolysis, Cellular and Molecular Biology (2017), 75–79.