LncSNHG14 promotes nutlin3a resistance by inhibiting ferroptosis via the miR-206 /SLC7A11 axis in osteosarcoma cells

The most prevalent form of primary osseous malignant tumor in adolescents and children is osteosarcoma (OS). A combination of surgery and neoadjuvant/post-surgery chemotherapy is currently the standard therapy. While the chemoresistance associated with OS generally leads to poor efficacy of therapeutic agents, the relevant molecular interaction is still elusive. Here, the lncRNA (long non-coding RNA) SNHG14 was found to be significantly upregulated in the nutlin3a-resistant OS cell line NR-SJSA1 and contributes to treatment resistance by suppressing ferroptosis. In NR-SJSA1 cells, knockdown of LncRNA SNHG14 resulted in a reversal of drug resistance and activation of ferroptosis, which disappeared when ferrostatin-1, a ferroptosis inhibitor, was added. Mechanistically, lncRNA SNHG14 targeted and down-regulated the expression of miR-206, further affecting the common ferroptosis inhibitor SLC7A11, and preventing NR-SJSA1 cells from undergoing ferroptosis. In conclusion, our findings highlight the involvement of lncRNA SNHG14 in ferroptosis and chemotherapy resistance of nutlin3a-resistant NR-SJSA1 cells, thus shedding new insight on how to overcome drug resistance in osteosarcoma cells and improve treatment efficacy.


INTRODUCTION
The most prevalent form of primary malignancy found in adolescents and children is osteosarcoma (OS); it originates from stromal cells and has a high degree of aggressiveness [1,2]. The present standard therapy of OS is combined surgery and neoadjuvant/postoperative chemotherapy, but the efficacy is usually unsatisfactory due to chemotherapy resistance [3]. Therefore, in-depth studies of OS cell drug resistance and the underlying mechanisms are of great significance for OS treatment and prognosis.
Ferroptosis, a recently discovered approach of programmed cell death, differs from cell necrosis, apoptosis, and pyrodeath, as the iron-dependent lipid peroxidation is the main driver [4,5]. It is reported that interferon-gamma (IFNγ) released in response to immunotherapy-activated CD8 + T lymphocytes downregulates the expression level of SLC3A2 (solute carrier family 3 member 2) and SLC7A11. These changes elicit the lipid peroxidation and ferroptosis, which consequently suppresses tumor cell growth [6]. Other studies have also demonstrated that promoting ferroptosis sensitized tumor cells to radiotherapy, while reduced ferroptosis would facilitate the survival of tumor cells during metastasis [7,8]. These papers suggest the involvement of ferroptosis in tumor growth and sensitivity to radiotherapy and immunotherapy.
SLC7A11 is a part of cysteine/glutamate anti-transporter (System Xc-) that promotes cysteine uptake and glutathione biosynthesis [9,10], thus reducing oxidative stress and preventing cellular ferroptosis. Previous evidence suggests that SLC7A11 regulates the drug resistance in malignant cells and acts as a bridge between ferroptosis and chemoresistance. The mRNA level of SLC7A11 is upregulated in cisplatin-resistant bladder tumor cells, but this resistance could be reversed by knocking down SLC7A11, whose expression is negatively correlated with clinical prognosis [11]. It has also been shown that low expression of SLC7A11 increases the susceptibility of drug-resistant neck/head tumor cells and that combining cisplatin with sulfasalazine, a ferroptosis inducer, significantly inhibits drug-resistant tumor cell growth [12]. Besides, the literature suggests that BRAF-mutated melanomas resistant to BRAF inhibitors have remarkably subsided when SLC7A11 is inhibited [13]. Though, how SLC7A11 functions in the drug resistance of OS remains elusive.
ncRNAs (non-coding RNAs) account for around 98 percent of transcripts in the primate genome, and the length of lncRNA (long non-coding RNA) is more than 200 nucleotides [14,15]. A growing body of literature has revealed that lncRNA appears to be associated with the sensitization of OS malignant cells to chemotherapeutic compounds [16][17][18]. It is reported that lncRNA SNHG14 is involved in chemotherapeutic resistance in various tumors. LncRNA SNHG14 activated the Nrf2 signal transduction pathway to promote trastuzumab resistance in breast cancer cells [19]. In non-small cell lung cancer (NSCLC) and large intestine tumor, lncRNA SNHG14 can promote cisplatin resistance [20,21]. However, whether SNHG14 contributes to the chemoresistance of OS cells remains uncertain.
Nutlin3a is an inhibitor that disrupts the interaction of mouse double minute 2 (MDM2) and p53 by occupying the p53 binding site of MDM2, stabilizing the p53 protein and exerting a potent anti-proliferation effect [22,23]. Nutlin3a is currently being tested in clinical trials and has been already widely used in anti-cancer research [24][25][26].
The competing endogenous RNA (ceRNA) theory of lncRNAs and miRNAs is broadly accepted at the moment; [27,28] however, the mechanism study of lncRNA SNHG14 in ferroptosis and nutlin3a-resistant OS cells is lacking. We here reveal that the expression of SNHG14 was enhanced in nutlin3a-resistant OS. The SNHG14 knockdown reversed the drug resistance and altered ferroptosis-related indicators. The reversal effect was blocked when ferrostatin-1, a ferroptosis inhibitor, was introduced into the medium. We further demonstrated that lncRNA SNHG14 inhibited ferroptosis and promoted nutlin3a-resistance in OS cells via the miR-206/SLC7A11 axis. Therefore, lncRNA SNHG14 is a promising target to reduce the chemotherapy resistance in OS.

METHOD Construction of nutlin3a-resistant cell, cell culture and cell transfection
The human OS SJSA1 cells of logarithmic growth phase were seeded in the 10 cm petri dish. When the cells acquired up to 70%~80% confluency, we added nutlin3a at the final concentration of 5 μM. Two hours later, the media containing nutlin3a were discarded, then live cells were transferred into a new culture flask with the nutlin3a-free medium. When the cells reached up to 70~80% confluency, the above steps were repeated with 10 μM nutlin3a (final concentration) until cell mortality was less than 5%. The resulting resistant cell line was named NR-SJSA1.

FISH (fluorescence in situ hybridization)
The FISH probe to detect the expression level of lncRNA SNHG14 by Cyanine3 labeling was designed and produced in Shanghai Gema Pharmaceutical Technology Co., Ltd. The sequences are shown as follows: 5′-ccacactgacgacacatcaa-3′, 5′-cacactgaaaccaggactct-3′, 5′-ggggaggtgcagagaaaaca-3′. Four percent paraformaldehyde was used to fix NR-SJSA1 cells, which were then incubated with FISH probes overnight for hybridization, and washed before fluorescence microscopy imaging.

RT-qPCR (Quantitative real-time PCR)
The TRIzol reagent (Invitrogen, USA) was used to extract total RNA from cells using, and the NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA) was used to check the purity of RNA. The PrimeScript™ RT kit (Takara, Japan) was used to synthesize cDNA samples; as per the manufacturer's instructions, the SYBR Green PCR Mix Kit (Takara, Japan) was used to quantify in the RT-qPCR assay. The RT-qPCR result was quantified with the 2 −ΔΔCT method. Table 1 shows the RT-qPCR primers.

MTT assay
It was used to quantify the cell viability after drug treatment and siRNA transfection. Cells were counted and seeded into ninety-six-well plates.
Following 48 h or 72 h of intervention, ten microlitres of MTT reagent (5 mg/mL) were added in each well. After 240 min of incubation, 150 microlitres of dimethyl sulfoxide (DMSO) were added in each well, and crystals were completely dissolved by shaking the plate on a low-speed shaker for 10 min, then the absorbance at 490 nm (A 490 ) was determined in the microplate reader.

Crystal violet staining
Cell viability after drug treatment and siRNA transfection was evaluated by crystal violet assay. 48 h after treatment, PBS (phosphate-buffered saline) was used to wash cells, which were then fixed by methanol for 15 min, followed by staining with one percent crystal violet (BOSTER, China) for 20 min. Following washing with PBS, cells were dried on filter paper, and pictures were taken by a scanner. Then one millilitre 1% SDS was added in each well, and crystals were completely dissolved by shaking the plate on a low-speed shaker for 15 min. The A 570 was determined using a microplate reader.

ROS and C11-Bodipy fluorescence staining and flow cytometry analysis
For ROS fluorescence staining, the cells were exposed to H2DCFDA dye (5 µM), treated for 20 min and images were acquired with a fluorescence microscope. For C11-Bodipy fluorescence staining, cells were exposed to C11-Bodipy (5 µM), treated for 30 min, stained with DAPI (4′,6-diamidino-2phenylindole) for 300 sec, and then images were taken with a fluorescence microscope. For flow cytometry, H2DCFDA was used to treat cells for 20 min, or cells were incubated with C11-BODIPY for 30 min in a 6-well plate, then they were trypsinized to obtain a single-cell suspension, and quantified on a BD FACSMelody sorting flow cytometer. Data analysis was performed using FlowJo 10.

Evaluation of malondialdehyde (MDA) and iron levels
In terms of the manufacturer's instructions, the final lipid peroxidation product MDA was quantified. The MDA lysis buffer was used to homogenize cells, followed by centrifuging at 13,000 × g for 180 sec, and the A 535 was determined in the microplate reader. following the manufacturer's instructions, the Iron Colorimetric Assay Kit (Applygen, China) was used to determine the concentration of iron ions of lysed cells.

Statistics
Every experiment was performed at least thrice. SPSS Statistics 22.0 (IBM, Armonk, NY, USA) was used to conduct statistical analyses. If the data fit the normal distribution, the statistically significant intergroup differences were determined by One-way ANOVA; the Kruskal-Wallis H test was otherwise applied. The data are shown as the mean ± SD, with p < 0.05 as statistically significant.

LncRNA SNHG14 knockdown sensitized NR-SJSA1 cells to nutlin3a
To investigate the mechanisms of OS cell chemoresistance, we generated nutlin3a-resistant NR-SJSA1 cells from a parent human OS cell line (SJSA1 cells). In RT-qPCR, the lncSNHG14 expression was enhanced in NR-SJSA1 cells compared to its parent SJSA1 cells (Fig. 1A), indicating the involvement of lncRNA SNHG14 in the chemoresistance of NR-SJSA1 cells.
To further study the function of lncRNA SNHG14 in nutlin3aresistance, NR-SJSA1 cells were transfected with si-SNHG14 to knockdown lncSNHG14 (Fig. S1). We observed that lncSNHG14 expression was significantly downregulated in the siSNHG14 group after transfection (Fig. 1B), which confirmed that siSNHG14 successfully silenced lncRNA SNHG14 in NR-SJSA1 cells.
In MTT assay, after lncSNHG14 knockdown, the cell survival rate was significantly decreased. The median inhibitory concentration (IC50) of control group and siSNHG14 group were 41.43 μM and 15.72 μM, respectively (Fig. 1C). The reversal index for drug resistance was 2.64. These findings suggested that inhibiting lncSNHG14 could reverse the drug resistance of NR-SJSA1 cells to nutlin3a. We then chose 9 μM, 12 μM, and 15 μM concentrations of nutlin3a for subsequent crystal violet staining based on the results of the MTT assay. As the concentration of nutlin3a was higher, the viable cells of siSNHG14 group decreased remarkably as compared to the negative control ( Fig. 1D and E), indicating that lncSNHG14 knockdown promotes the nutlin3a sensitivity in NR-SJSA1 cells. Since the crystal violet staining showed a substantial difference when the concentration of nutlin3a was 12 μM, we picked this concentration for subsequent experiments.
LncRNA SNHG14 knockdown promotes ferroptosis in NR-SJSA1 cells We speculated that lncRNA SNHG14 exerted its function via altering nutlin3a resistance by interfering with ferroptosis. To verify our hypothesis, we detected ferroptosis-related molecules SLC7A11, GPX4, PTGS2, Nrf2, Keap1, and HO-1 by both RT-qPCR and Western blot. In siSNHG14 group, the mRNA expressions of ferroptosis inhibitory molecules like SLC7A11 and GPX4 as well as antioxidant molecules like Nrf2 and HO-1 were significantly lower. Conversely, the levels of oxidative stress-inducing molecules like Keap1 and PTGS2 were higher ( Fig. 2A). The Western blot results agreed with those of RT-qPCR ( Fig. 2B and C). These suggest that the inhibition of lncRNA SNHG14 could induce ferroptosis in NR-SJSA1 cells.
To validate the above effects of lncRNA SNHG14, DCFH-DA and C11-BODIPY fluorescent probes were used to quantify the cytoplasmic ROS and lipid ROS, respectively. Compared to the negative control group, the green fluorescence intensity representing cytoplasmic ROS in the siSNHG14 group was enhanced (Fig. 2D). The fluorescence intensity of NR-SJSA1 cells was further quantified by flow cytometry, and the data revealed that siSNHG14 group had a considerably higher level of cytoplasmic ROS (Fig. 2E). The green fluorescence of siSNHG14 group was enhanced (Fig. 2F), while the red fluorescence was weakened. These findings indicate that the lipid ROS level following lncRNA SNHG14 knockdown was increased as compared to the negative control group. The flow cytometry analysis also verified this increased ROS level in siSNHG14-transfected NR-SJSA1 cells (Fig.  2G). In contrast to the negative control, the level of MDA content in NR-SJSA1 cells after siSNHG14 transfection was found to be remarkably increased (Fig. 2H).
It is well-known that iron metabolism is a crucial process in ferroptosis, which involves the Fenton reaction between free and unstable Fe 2+ and H 2 O 2 that generates hydroxyl radicals with strong oxidative capacity, triggering lipid peroxidation and promoting ferroptosis. Notably, the concentration of cellular iron ions is inversely related to the degree of ferroptosis. We observed that as compared to the negative control, the iron ion accumulation in NR-SJSA1 cells was increased after lncRNA SNHG14 knockdown (Fig. 2H). The above results demonstrated that lncRNA SNHG14 silencing could promote ferroptosis in NR-SJSA1 cells.
LncRNA SNHG14 knockdown enhances nutlin3a sensitivity of NR-SJSA1 cells by promoting ferroptosis To investigate whether the increased nutlin3a sensitivity of NR-SJSA1 cells following lncRNA SNHG14 downregulation is due to ferroptosis, we added Ferrostatin-1 (Fer-1) into the medium. Fer-1 inhibits ferroptosis by inhibiting lipid peroxidation. According to the available literature and earlier preliminary experiments, the concentration of Fer-1 in the follow-up experiments of this study was 2 μM, which was not toxic to the cells (Fig. S2). Flow cytometry analysis revealed that siSNHG14-transfected NR-SJSA1 cells had lower mean fluorescence intensity (MFI) of cytoplasmic and lipid ROS after ferroptosis inhibitor Fer-1 addition ( Fig. 3A and B), indicating that Fer-1 could reverse ferroptosis promotion mediated by lncRNA SNHG14 knockdown in NR-SJSA1 cells.
After 48-h treatment with different concentrations of nutlin3a and 2 μM Fer-1, the MTT assay was conducted to assess cellular viability. Nutlin3a concentration-dependently reduced the viability of NR-SJSA1 cells in the siSNHG14 group, but cell viability was considerably increased when Fer-1 was introduced (Fig. 3C). Crystal violet staining showed that lncRNA SNHG14 knockdown NR-SJSA1 cells were more sensitive to nutlin3a than the negative control. Importantly, cells regained their resistance after Fer-1 treatment ( Fig. 3D and E). Collectively, these results established that ferroptosis was closely involved in drug resistance of NR-SJSA1 cells, and that lncRNA SNHG14 promoted drug resistance to nutlin3a by inhibiting ferroptosis.

miR-206 is a target of lncRNA SNHG14
The subcellular localization usually determines the molecular mechanisms by which lncRNA exerts their functions in biological processes. To mechanistically explore how lncRNA SNHG14 participates in ferroptosis and drug resistance, we detected the subcellular localization of lncSNHG14 using RNA-FISH assay, which showed that lncRNA SNHG14 localized predominantly in the cytoplasm (Fig. 4A).
It is evident that competing endogenous RNA (ceRNA) is a common post-transcriptional regulation mechanism of lncRNA, which describes that lncRNA indirectly regulates the expression level of mRNA by competitively binding with miRNA, thereby regulating various biological processes. The ceRNA regulation mainly takes place in the cytoplasm [29]. Given the subcellular localization of lncRNA SNHG14 in the cytoplasm, we screened and identified its target miRNAs. The downstream miRNAs of lncRNA SNHG14 were predicted in three databases, i.e., lncBase, starBase, and miRcode. The overlapping hits from these three databases identified five miRNAs with high reliability (Fig. 4B). The miR-206 expression of NR-SJSA cells was meaningfully enhanced after lncRNA SNHG14 knockdown (Fig. 4C), suggesting the possibility of miR-206 as the negatively regulated target of lncRNA SNHG14.
To validate the interaction between lncRNA SNHG14 and miR-206, we performed the luciferase reporter assay (Fig. 4E). For this, the 293 T cells were treated with miR-206 mimic or its specific inhibitor for detection, and the transfection efficiency was examined by RT-qPCR (Fig. 4D). The relative luciferase activity of SNHG14-WT reporter vector was reduced when overexpressing miR-206, and increased upon miR-206 reduction. Agreeably, no difference was observed in the empty vector group (Fig. 4F). This result suggested that lncRNA SNHG14 could directly bind to miR-206. We also constructed luciferase reporters containing a lncRNA SNHG14 sequence with binding site mutation. After transfecting SNHG14-MUT plasmids into the 293 T cells, miR-206 expression alteration failed to alter the relative luciferase activity of the SNHG14-MUT vector, indicating that the binding site was as per our prediction. These suggest that miR-206 could be the target of lncRNA SNHG14, and it can directly bind to lncRNA SNHG14.

miR-206 induces ferroptosis in NR-SJSA1 cells
To elucidate the importance of miR-206 in ferroptosis, NR-SJSA1 cells were transfected with miR-206 mimic and miR-206 inhibitor, and ferroptosis-related mRNAs and proteins were measured. The mRNA and protein levels of ferroptosis inhibitory molecules like SLC7A11, GPX4, Nrf2, and HO-1 decreased after miR-206 mimic transfection, whereas ferroptosis-promoting molecules like PTGS2 and Keap1 were considerably upregulated, in contrast to the NC mimic group. These changes were reversed in miR-206 inhibitor group as compared to NC inhibitor group ( Fig. 5A-C). Furthermore, the levels of cytoplasmic ROS and lipid ROS were determined. The MFI (mean fluorescence intensity) of ROS in NR-SJSA1 cells was significantly increased after miR-206 mimic transfection ( Fig. 5D and E), while the miR-206 inhibitor treatment decreased ROS levels. The concentration of MDA and iron ions were measured thereafter, and we found that both of them increased in the former group and decreased in the latter ( Fig. 5F and G). According to our findings, miR-206 was shown to be closely associated with ferroptosis.

LncSNHG14 promotes nutlin3a resistance in NR-SJSA1 cells via inhibiting ferroptosis by targeting miR-206
To test whether lncRNA SNHG14 knockdown enhances ferroptosis by interacting with miR-206, we transfected the miR-206 inhibitor in lncRNA SNHG14 knockdown NR-SJSA1 cells and detected the changes in ferroptosis-related indicators. After transfection with miR-206 inhibitor in siSNHG14 cells, the expression of SLC7A11, Nrf2, GPX4, and HO-1 increased, and that of Keap1 and PTGS2 decreased ( Fig. 6A-C). These suggest that miR-206 is an essential interacting molecule of lncRNA SNHG14 in ferroptosis regulation. We then measured cytoplasmic ROS, lipid ROS, MDA, and iron ions in siSNHG14 cells after suppressing miR-206 expression by flow cytometry. As previously observed, the levels of all of these biomarkers were increased in NR-SJSA1 cells when lncRNA SNHG14 was downregulated. However, we saw decreased levels of these molecules in cells exposed to the miR-206 inhibitor (Fig. 6D-G). These results revealed that silencing lncRNA SNHG14 promotes ferroptosis, but dual-inhibition of lncRNA SNHG14 and miR-206 protects NR-SJSA1 cells against ferroptosis. In other words, lncRNA SNHG14 targets miR-206 to inhibit ferroptosis in nutlin3a-resistant cells.
To scrutinize the miR-206 function with regard to the drug resistance regulated by lncRNA SNHG14, we detected the viability of miR-206 downregulated cells in the presence of nutlin3a. Transfection with miR-206 inhibitors significantly enhanced the viability of nutlin3a-resistant cells (Fig. 6H and I), which indicates that the nutlin3a sensitivity induced by lncRNA SNHG14 knockdown can be reversed by inhibiting miR-206. These results showed that lncRNA SNHG14 promotes nutlin3a resistance in NR-SJSA1 cells by binding with miR-206.

SLC7A11 is a target of miR-206
Although we have found that lncRNA SNHG14 enhanced drug resistance by inhibiting ferroptosis via targeting miR-206, the entire mechanism was not fully elucidated. Hence, we predicted the downstream gene of miR-206 using TargetScan, starBase, miRWALK, and DIANA. The intersection of these four databases presented the predicted genes of highest possibility, including SLC7A11 (Fig. 7A). Given that SLC7A11 expression level changed after lncRNA SNHG14 and miR-206 silencing (Figs. 5A-C and 6A-C), SLC7A11 was further investigated in detail. To that end, the NR-SJSA1 cells were transfected with two siRNAs to silence SLC7A11, and the western blot analysis confirmed that the first siRNA (#1) was more efficient in SLC7A11 silencing ( Fig. 7B and C), so we used siSLC7A11 #1 in the following experiments. The DLR gene assay was useful in determining the direct binding between miR-206 and SLC7A11. The PmirGLO, PmirGLO-SLC7A11-WT, PmirGLO-SLC7A11-MUT, and miRNA (negative control /miR-206 mimic/miR-206 inhibitor) were co-transfected in the 293 T cells. As shown in Fig. 7E, in the pmirGLO-SLC7A11-WT group, the relative fluorescence intensity was decreased after miR-206 overexpression, and an increased fluorescence intensity was noted after miR-206 inhibition, suggesting that miR-206 could directly interact with SLC7A11. The cells co-transfected with PmirGLO-SLC7A11-MUT and miRNA (miR-206 mimic/miR-206 inhibitor/negative control) were not dramatically different. We demonstrated that SLC7A11 was a downstream target of miR-206.

miR-206 promotes ferroptosis by downregulating SLC7A11
To test whether miR-206 promotes ferroptosis by downregulating SLC7A11, we transfected NR-SJSA1 cells with siSLC7A11 #1 alone or based on miR-206 suppression. As shown in Fig. 8A-C, silencing of SLC7A11 undermined the expression of GPX4, Nrf2, and HO-1 and enhanced the expression of Keap1 and PTGS2, indicating that SLC7A11 is a critical factor for ferroptosis. Moreover, compared to the miR-206 inhibitor group, GPX4, Nrf2, and HO-1 expressions in the miR-206 inhibitor and siSLC7A11 #1 co-transfection groups were dramatically weakened, whereas expressions of Keap1 and PTGS2 were meaningfully increased, suggesting enhanced cell ferroptosis.
The ferroptosis-related indicators were detected to further confirm the ferroptosis alteration in NR-SJSA1 cells after co-transfection. SLC7A11 knockdown improved the accumulation of cytoplasmic ROS, lipid ROS, MDA, and iron ions, which were reduced by miR-206 silencing, indicating that miR-206 promoted ferroptosis by downregulating SLC7A11 (Fig. 8D-F). Since lncRNA SNHG14 has been proven to target miR-206, we argued that it inhibited ferroptosis via the miR-206/SLC7A11 axis and subsequently improved nutlin3a resistance in NR-SJSA1 cells (Fig. 8G).

DISCUSSION
OS is the most familiar primary malignancy in childhood and adolescence, with a mesenchymal origin, a high degree of malignancy, early distant metastases, and a high mortality rate. The current standard treatment for OS includesa combination of neoadjuvant chemotherapy, surgery, and postoperative chemotherapy, but the chemoresistance usually induces poor efficacy. Therefore, thorough investigations of the underlying mechanisms are critical for OS treatment and prognosis. The carcinogenesis involves intricate genetic and epigenetic alterations, in which lncRNA regulates a wide range of biological processes through epigenetic, transcriptional, and post-transcriptional regulations. Growing evidence indicate that lncRNAs are abnormally expressed in malignant tissues and are closely correlated with carcinogenesis and cancer progression. LncRNA SNHG14, a long non-coding RNA encoded by the gene of human chromosome 15q11.2, exerts diverse functions. It is abundantly expressed in bladder tumor, glioma, hepatocellular carcinoma, among others. It participates in cancer cell proliferation, migration and invasion, and is related to clinical manifestation and outcome. Importantly, SNHG14 is overexpressed in OS tissues and inhibits OS cell apoptosis via the miR-433-3p/FBXO22 axis, thereby promoting OS progression [30]. However, to date no investigation has been conducted on the relationship between lncRNA SNHG14 and chemoresistance in OS cells.
Ferroptosis is a newly discovered type of RCD (regulated cell death), which is elicited by the Fe-dependent lipid peroxidation, and differs from other kinds of RCD in morphology and mechanism. The recent interest in determining the critical role of ferroptosis in malignancy inhibition and sensitivity to immunotherapy and chemotherapy is increasing [6][7][8]. Although limited, several studies suggest that lncRNAs impact on ferroptosis in malignancy.
For instance, lncRNA LINC00336 inhibited ferroptosis in lung tumor via binding with miR-6852, which regulated the expression level of ferroptosis marker cystathionine-β-synthase [31]. We hypothesized that reduced ferroptosis would facilitate the drug resistance of OS, and the dysregulation of lncRNA SNHG14 might regulate the ferroptosis and subsequent chemotherapy sensitivity.
In this study, SNHG14 was abundantly expressed in a nutlin3aresistant human OS NR-SJSA1 cells, and its downregulation reversed nutlin3a resistance. Next, we examined whether ferroptosis is a part of the mechanism of SNHG14 regulated nutlin3a resistance, and our RT-qPCR and western blot analysis supported our speculation. The measurement of ROS level also demonstrated that lncRNA SNHG14 knockdown induced an increased ferroptosis in NR-SJSA1 cells. Furthermore, tour next studies discovered the bridge connecting lncRNA SNHG14 and ferroptosis. Growing evidence indicate that lncRNA acts as a competitive RNA and bind to miRNA when regulating gene expression. Thus, we conducted bioinformatics analysis to screen the downstream miRNAs target of lncRNA SNHG14. We identified miR-206 as a potential candidate. Several studies have shown that lncRNA SNHG14 regulates downstream protein expression in NSCLC and cervical cancer by interacting with miR-206 [32,33]. Using DLR assay, we proved the direct binding between lncRNA SNHG14 and miR-206. SLC7A11, a functional molecule of system Xc, transports extracellular cysteine and intracellular glutamate, thus, regulating the cellular redox homeostasis and ferroptosis. The function of SLC7A11 in tumor cell resistance is spotlighted recently. The melanoma cells with dysfunctional SLC7A11 are more sensitive to immunotherapy and radiotherapy [7]. Nutlin3a-mediated p53 activation could induce ferroptosis by inhibiting SLC7A11 [9]. Bioinformatics analysis showed SLC7A11 as a potential target of miR-206. We confirmed the regulation of SLC7A11 expression by miR-206 and lncRNA SNHG14. However, a substantial divergence was noted between the miR-206 inhibitor and siSLC7A11 co-transfection groups and the siNC group (Fig. 8), suggesting that miR-206 was not the sole upstream molecule of SLC7A11. Collectively, the lncRNA SNHG14/miR-206/SLC7A11 axis was demonstrated to suppress ferroptosis and promote resistance to nutlin3a. The proposed mechanism of lncRNA SNHG14 in OS cell drug resistance is summarized in Fig. 8H. Our findings highlighted a novel potential target that could be explored to reduce the chemoresistance of OS cells in clinical treatment. Further in vivo investigation is needed, and it is also meaningful to identify other regulators for ferroptosis.

CONCLUSION
In conclusion, lncRNA SNHG14 promotes nutlin3a resistance by inhibiting ferroptosis via the miR-206/SLC7A11 axis in osteosarcoma cells. Our studies confirm that ferroptosis is closely associated with drug resistance and shows a valuable target for improving tumor cell chemosensitivity.

DATA AVAILABILITY
The datasets generated during and/or analysed during the current study are not publicly available due [REASON WHY DATA ARE NOT PUBLIC] but are available from the corresponding author on reasonable request.