YBX1 promotes laryngeal squamous cell carcinoma progression via activating MAPK/ERK signaling as a target of miR-382-5p

Background. Y-box binding protein-1 (YBX1) inuences the onset and progression of laryngeal squamous cell carcinoma (LSCC) remains unknown. The present study therefore sought to explore the mechanistic role of YBX1 in LSCC. Methods.Analyses of the Gene Expression Omnibus (GEO) database and associated bioinformatics analyses revealed that YBX1 was upregulated in LSCC, and we further conrmed this result using primary LSCC patient samples. We additionally explored the impact of siRNA-mediated YBX1 knockdown on LSCC cell proliferation, migration, and invasion using CCK8, wound healing, and Transwell assays. We then conducted interrogated miRNA databases and conducted subsequent luciferase reporter assays to conrm that miR-382-5p binds to YBX1. Additional studies of the mechanisms downstream of this miR-382-5p/YBX1 axis focused on detecting the expression of mitogen-activated protein kinase (MAPK)/extracellular regulated kinase (ERK) signaling-related genes via qPCR and Western blotting. Results.We detected signicant upregulation of YBX1 in LSCC tumors that was signicantly correlated with advanced TNM stage and poor patient prognosis. Knockdown of YBX1 markedly impaired the proliferative, invasive, and migratory activity of Tu212 cells in vitro. From a mechanistic perspective, miR-382-5p was found to bind to the YBX1 3&rsquo;-untranslated region and to thereby inhibit LSCC progression. We further conrmed that miR-382-5p negatively regulated YBX1 to inhibit proliferation via the MAPK/ ERK signaling axis in LSCC. Conclusion.Together, our results indicated that YBX1 is an important promoter of LSCC progression, and that miR-382-5p can suppress YBX1 expression and inactivate MAPK/ERK signaling. These ndings may thus highlight novel and promising prognostic and therapeutic targets in the context of LSCC.


Introduction
Laryngeal squamous cell carcinoma (LSCC) is a common epithelial tumor affecting the head and neck region [1], with roughly 13,430 newly diagnosed cases in 2016 [2]. Roughly 60% of LSCC patients present with advanced disease that is often not amenable to treatment. Indeed, the overall survival (OS) rate for LSCC has declined slowly from 66-63% in recent decades, underscoring the severe morbidity and mortality of this cancer type [3]. The pathogenesis of LSCC is a complex multi-stage process that remains incompletely understood, and as such further studies of the mechanistic basis for these oncogenic processes have the potential to guide the diagnosis or treatment of this condition to improve patient prognosis.
Y-box binding protein-1 (YBX1) is a cold-shock protein that has previously been linked to dysregulated cell proliferation, drug resistance, and chromatin destabilization in the context of cancer [4]. Elevated YBX1 expression is associated with a poor prognosis in breast, prostate, lung, and ovarian cancers and in melanoma patients [5][6][7][8]. Higher YBX1 expression has also been reported that associated with more advances T stage, poorer tumor differentiation, and cervical or distance metastases in HNSCC patients [9].
The mechanisms whereby YBX1 modulates LSCC cell biology, however, remain to be elucidated.
MicroRNAs (miRNAs) are ~20 nucleotide conserved non-coding RNAs that function as posttranscriptional regulators capable of binding to the 3'-untranslated region (UTR) of target mRNAs to suppress their expression [10,11]. In the context of LSCC, many researchers have identi ed roles for these non-coding RNAs as regulators of cancer onset and progression [12]. MiR-382-5p is a miR-382 family member that is downregulated in many cancers, with such downregulation being linked to tumor progression [13]. In ovarian cancer, miR-382 has been shown to suppress tumor cell migratory and invasive activity via suppressing ROR1 expression and thereby impairing the epithelial-mesenchymal transition [14]. To date, however, no studies have explored miR-382-5p expression pro les or functional roles in the context of LSCC.
As such, in the present study, we explored the potential mechanisms whereby YBX1 and miR-382-5p may in uence LSCC through in vitro and in vivo experiments. We found that YBX1 upregulation was common in LSCC tumor tissues, and that such upregulation was associated with enhanced LSCC tumor cell proliferative, migratory, and invasive activity. We further found that miR-382-5p directly targets YBX1 and suppresses its expression, thereby compromising LSCC tumor cell growth and activity. Restoring YBX1 expression to Tu212 cells overexpressing miR-382-5p was su cient to enhance the proliferation, migration, and invasion of these cells. Together, these ndings indicate that YBX1 plays an oncogenic role in LSCC and that it is negatively regulated by miR-382-5p, which might be a novel potential treatment target for LSCC

Material And Methods
Cell lines, and reagents.
The larynx carcinoma cell line TU212 cells were obtained fromGuangzhou Juyan Biological Technology (Guangzhou, Chin) and cultured in RPMI 1640 medium (Gibco; Grand Island, NY) with 10% fetal bovine serum (FBS; Hyclone, Logan, UT), 100 U penicillin and 100 mg/ml streptomycin (Gibco), hereafter referred to as standard media. The cells were kept at 37°C in a humidi ed atmosphere containing 5% CO2.

Tumor samples
In total, 30 patients diagnosed with LSCC at the Ganzhou Tumor Hospital between 2014 and 2016 were enrolled in this study. Patients were an average of 63.6 years old (range: 28-79), and all patients underwent surgical treatment in our department. In addition, 15 normal paracancerous tissue segments (>1 cm from surgical margins) were collected to serve as control samples. Median patient follow-up was 27.1 months (Range: 2-85.6). Clinicopathological features of them are described in table 1. All patients provided written informed consent, and the Ethical Committee of Ganzhou Cancer Hospital approved this study (No.2017008).The association between gene expression and patient survival was assessed using Kaplan-Meier curves constructed in GraphPad Prism 8.0. GEO data analysis GSE51985 and GSE59102 were obtained from the NCBI Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/gds). The download data format is MINIML,and the box plot is implemented by the R software package ggplot2. All data were downloaded from the public databases hence it was not required to obtain additional ethical approval for our study. Cell culture and transfection Human Tu212 LSCC cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and were grown in DMEM (L110, Shanghai BasalMedia Technologies Co., LTD) containing 10% FBS and penicillin/streptomycin (15070063, Thermo, USA) in a 5% CO 2 incubator at 37°C. Transfections were conducted by plating 5×10 4 cells/well of 6-well plates and allowing them to grow until 60-70% con uent, at which time miR-382-5p (10, 50, or 100 nmol/L) was transfected into cells with Lipofectamine RNAiMAX (GenePharma, Shanghai, China), or plasmid DNA or siRNA duplexes Negative siRNA(NC) sense: 5-UUCUCCGAACGUGUCACGUTT-3; antisense: 5-ACGUGACACGUUCGGAGAATT-3

Migration and invasion assays
For Transwell assays, 8 μm inserts (BD Biosciences) that were or were not coated with 1 mg/mL Matrigel (Invitrogen) were seeded with 1×10 4 or 2×10 4 appropriate cells for invasion and migration assays, respectively. The lower chamber was then lled with DMEM containing 10% FBS, and cells were incubated for 24-48 h after which invasive/migratory cells were xed, stained with crystal violet, and counted under 200× magni cation.

Cell viability assay
Cells were transfected with miR-382-5p mimics, YBX1 overexpression plasmids, or YBX1-speci c siRNA, after which a CCK8 kit (Beyotime, WI, USA) was used based on provided instructions to assess cell viability. Viability was assessed at 0, 24, 48, 72, and 96 h by assessing absorbance (OD) at 450 nm using a plate reader, with survival being calculated as follows: survival rate=(OD experimental -OD blank )/(OD control -OD blank ).

Luciferase reporter assay
A Dual-Luciferase reporter assay was conducted based upon provided instructions (Promega, WI, USA). Brie y, HEK-293T cells were transduced with a lentivirus encoding miR-382-5p or a control construct and were then plated in 96-well plates until 70% con uent. After an additional 12 h, cells were co-transfected with 50 ng of pMIR-YBX1-3'UTR-wt or pMIR-YBX1-3'UTR-mut and 10ng of pMIR-GLO (Gene Pharma). After a further 24 h incubation, the Dual-Luciferase Reporter Assay System was used to quantify re y and Renilla luciferase activity.
qRT-PCR TRIzol (Invitrogen, USA) was used to extract cellular RNA, after which a PrimeScript RT reagent kit (TaKaRa, Japan) was utilized to prepare cDNA, while a PrimeScript miRNA cDNA Synthesis Kit (TaKaRa) was instead used for miRNA analyses. SYBR Premix Ex Taq I was then used for qPCR analyses, and all primers used in this study are compiled in table 2. β-actin and U6 served as normalization controls for mRNA and miRNA analyses, respectively, with the 2-ΔΔ Ct approach being used to normalize gene expression.

Immunohistochemistry (IHC)
Para n-embedded tissue sections were depara nized using xylene, after which they were rehydrated with an ethanol gradient. Endogenous peroxidase activity was then blocked using 3% H 2 O 2 , after which samples were heated in a microwave to facilitate antigen retrieval. Next, 5% BSA was used to block nonspeci c antigen binding for 1 hour at 37°C, after which sections were incubated with primary anti-YBX1 (1:250, Abcam, USA) overnight at 4°C. Sections were then incubated for 1 h with secondary antibodies at 37°C, after which diaminobenzidine was used to detect antibody binding and sections were counterstained using hematoxylin. An Olympus light microscope was then used to capture representative images of tissue sections.

Wound healing assay
Wound healing assays were conducted to evaluate cellular migration. Brie y, cells were plated until 80-85% con uent, at which time a sterile micropipette tip was used to scratch the monolayer. PBS was then used to wash away damaged cells, after which DMEM containing 10% FBS was added and cells were incubated for 24 h. The wound site was then imaged via phase-contrast microscope (Zeiss, Gottingen, Germany), with Image-Pro Plus v4.5.1 being used to quantify wound closure.

YBX1 upregulation correlates with disease progression and poor LSCC patient prognosis
To understand key genes associated with LSCC progression, we evaluated gene expression pro les associated with this cancer type in the publically available GSE59102 and GSE51985 dataset. We speci cally focused on genes that were upregulated in LSCC as they may represent viable therapeutic targets or prognostic biomarkers in this disease type. Of these, we selected YBX1 as a gene that was highly upregulated in LSCC tumor tissues (Fig. 1a). We then validated these previously published results by measuring YBX1 expression in 30 LSCC patient tumor tissue samples and 15 control tissues via IHC and Western blotting. These analyses con rmed that YBX1 was signi cantly overexpressed in LSCC tumors relative to healthy tissues (Fig. 1b-c). We further found that YBX1 expression levels were signi cantly higher in patients with tumors of a more advanced TNM stage (t=17.63, P<0.001, Fig. 1d).
Additional prognostic analyses prognostic analysis also revealed that higher YBX1 expression levels were correlated with decreased overall survival (OS) in LSCC patients (95% 1.130 to 8.585, P=0.031, Fig. 1e). Together, these data indicated that YBX1 is upregulated in LSCC and is associated with tumor progression.

YBX1 promotes LSCC tumor cell migratory and proliferative activity
To understand whether YBX1 plays important roles in LSCC, we next conducted loss-of-function assays wherein we transfected Tu212 cells with siRNA constructs speci cally targeting the YBX1 splice junction. Knockdown e ciency was con rmed via qPCR, with the YBX1-1007 and YBX1-658 siRNA constructs signi cantly reducing YBX1 levels in transfected cells (Fig.2a). Tu212 cells transfected with YBX1-1007 were then analyzed in CCK8 assays, which demonstrated that YBX1 knockdown markedly suppressed LLC cell proliferation at all tested time points (48h, p<0.01; 72h and 96h, p<0.001, Fig.2b). In addition, transwell assays revealed that YBX1 knockdown impaired LSCC cell migratory and invasive activity relative to control cells (Fig.2c). Together, these ndings indicate that YBX1 promotes metastasis in Tu212 cells.

YBX1 is a miR-382-5p target in LSCC cells
We next evaluated the regulatory mechanisms governing YBX1 expression in LSCC cells. Using the miRtarbase, miRWalk, starbase, and TargetScan databases, we identi ed three miRNAs (miR-379-5p, miR-382-5p, and miR-137) with the potential to regulate YBX1 expression (Fig3a). We then elected to focus speci cally on the role of miR-382-5p in this oncogenic context. To that end, we constructed luciferase reporters bearing WT and mutant versions of this putative miRNA binding site in the YBX1 3'-UTR (Fig3b).
We con rmed that YBX1 knockdown was associated with the upregulation of miR-382-5p, whereas overexpressing miR-382-5p markedly suppressed YBX1 expression in Tu212 cells (Fig3c). We then transfected cells with a range of miR-382-5p concentrations and found that YBX1 expression was reduced at higher miRNA concentrations (Fig3d). Luciferase reporter assays further con rmed that the transfection of Tu212 cells bearing the WT but not the MUT YBX reporter plasmid was signi cantly reduced following miR-382-5p transfection (Fig3e).

miR-382-5p suppresses YBX1 and thereby inhibits LSCC tumor growth
We next conducted additional functional assays aimed at exploring the function of miR-382-5p-mediated YBX1 in the context of LSCC. To that end, we transfected Tu212 miR-382-5p cells with a YBX1 overexpression vector (Tu212 miR-330-5p+YBX1 ). We found that overexpressing this miRNA signi cantly decreased the migratory and invasive ability of Tu212 cells, as expected ( Fig.4a-c). Importantly, rescue experiments con rmed that overexpressing YBX1 in these cells restored their proliferative and migratory activity (Fig.4a, b). Using an in vivo subcutaneous tumor implantation model, we further found that tumors derived from Tu212 miR-382-5p+YBX1 cells were signi cantly larger than tumors derived from Tu212 miR-382-5p cells (Fig.4c). Together, these loss-and gain-of-function results con rm that miR-382-5p suppresses YBX1 expression and thereby impairs the growth of LSCC tumors.
miR-382-5p inhibits YBX1 to inhibit LSCC growth via MAPK/ERK signaling Lastly, we explored the underlying signaling mechanisms whereby miR-382-5p-mediated repression of YBX1 may enhance LSCC tumor growth. As YBX1 is a known MAPK-dependent regulator of cellular proliferation and gene expression [19], we sought to speci cally evaluate the role of MAPK signaling in the present experimental context. We found that YBX1 knockdown was associated with the inhibition of K-ras and MEK expression in Tu212 tumor cells (Fig 5a,b). We then evaluated the ability of miR-382-5p to indirectly modulate MEK/ERK signaling via YBX1. Through these analyses, we found that overexpressing miR-382-5p signi cantly impaired YBX1, K-Ras, and MEK levels in Tu212 cells, whereas overexpression of YBX1 was su cient to restore the expression of these three signaling components in Tu212 miR-382-5p cells (Fig5c,d). Together, these ndings provide evidence that miR-382-5p can module MEK/ERK signaling via suppressing YBX1 expression and thereby altering MEK and ERK phosphorylation. Overall, we were therefore able to conclude that miR-382-5p inhibits LSCC tumor progression via suppressing YBX1 expression and thereby inactivating MEK/ERK signaling.

Discussion
LSCC patient death is often the result of tumor metastasis, which is a complex multi-stage process that is modulated by the expression of speci c miRNAs and mRNAs. Herein, we found that YBX1 expression was elevated in patients with LSCC, and that such upregulation was associated with unfavorable clinicopathological ndings and poor patient prognosis. Additional in vitro analyses revealed that YBX1 was able to modulate MAPK/ERK signaling and to thereby promote LSCC cell proliferative, migratory, and invasive activity, whereas miR-382-5p negatively regulated such activity. Overall, these ndings suggested that miR-382-5p/YBX1 is a key signaling regulator in LSCC and may thus represent a novel clinical target for the treatment of this deadly cancer. To the best of our knowledge, no prior studies have demonstrated a role for the miR-382-5p/YBX1 axis in the regulation of LSCC progression.
YBX1 regulates many nucleic acid-related intracellular activities including DNA repair, pre-mRNA transcription and splicing, mRNA packaging, translation, and mRNA stability [20]. YBX1 functions as a transcription factor that can control differentiation, proliferation, drug resistance, and stress responses in many tumor types [21]. From a mechanistic perspective, YBX-1 silencing suppresses MMP1 and β-catenin expression in breast cancer cells, thereby disrupting their invasive activity. Herein, we explored the upstream mechanisms regulating YBX1 expression in LSCC, and we determined that elevated YBX1 expression was correlated with poor prognostic indicators including T and TNM stage. Additionally, survival analyses suggested that YBX1 expression levels are associated with decreased OS in LSCC patients. Together, these data indicate that YBX1 thus plays an oncogenic role in the context of LSCC development and progression.
Many studies have demonstrated that miRNAs are important regulators of key pathological processes linked with LSCC onset and progression, governing tumor cell proliferation and metastatic processes [22]. Using appropriate predictive databases, we ultimately determined that miR-382-5p targets YBX1 [15,16]. MiR-382-5p is generated from the 5′-terminus of the miR-382 pre-miRNA, and has previously been shown to be dysregulated and associated with progression in many cancer types [23]. For example, Du et al [24] determined that miR-382-5p is commonly downregulated in the context of liver cancer associated with hepatitis B infections and that such low expression is often associated with tumor metastasis. In breast cancer, miR-382-5p additionally targets the RERG/Ras/ERK pathway to promote oncogenesis, allowing it to serve as a diagnostic or prognostic biomarker in this cancer type [25]. Prior studies, however, have not clari ed the role of miR-382-5p in LSCC. Our ndings provide clear evidence that miR-382-5p-mediated regulation of YBX1 inhibits LSCC progression. As such, miR-382-5p may be a viable biomarker for the detection and treatment of LSCC.
We additionally studied the mechanisms whereby YBX1 may promote tumor progression in the present study. In prior reports, the RRM2/TK1/TYMS axis has been found to regulate nucleotide metabolism and to govern the proliferation and senescence of tumor cells [26]. YBX1 induces the proliferation of cells via promoting GSK3B/Cyclin D1/Cyclin E1 pathway activation in pancreatic ductal adenocarcinoma [27], and drives renal cell carcinoma metastasis through the G3BP1-SPP1-NF-κB signaling pathway [28]. Furthermore, YBX-1 has been shown to modulate Raf/MEK/ERK signaling so as to promote the progression of prostate cancer [29]. Herein, we determined that miR-382-5p targets YBX1 and thereby promotes the upregulation of p-MEK and p-ERK in LSCC, while negatively impacting k-RAS, MEK, and ERK levels. As such, miR-382-5p/YBX1 may inhibit ERK activation by suppressing RAS/MEK/ERK in LSCC cells, consistent with prior ndings in the context of prostate cancer [29]. MAPK/ERK signaling pathway activity can be targeted with drugs including apatinib and anlotinib [30,31]. As such, our results highlight a potentially novel approach that can guide future efforts to treat LSCC.

Conclusion
In conclusion, we identi ed YBX1 as a prognostic factor associated with the progression of LSCC. We additionally found that miR-382-5p directly targets YBX1 and thereby modulates MAPK/ERK signaling so as to inhibit LSCC progression. These ndings thus suggest that targeting the miR-382-5p/YBX1 axis may be a viable approach to treating LSCC. expression levels in early-stage (I,II, n=14) and advanced-stage (III,IV, n=16) LSCC tissues were measured.

Figure 2
Transient YBX1 knockdown impairs LSCC cell in vitro migration, growth, and invasion. a Western blotting was used to con rm successful siRNA-mediated knockdown of YBX1 in Tu212 cells, with YBX1-1007 exhibiting optimal e cacy (n=3/group; Student's t-tests). b Tu212 cell growth following siRNA transfection was assessed via MTT assay (n=6/group; two-way ANOVA). c Wound healing and Transwell assays were used to assess cell migration and invasion, respectively, with migration and invasion indices being graphed (n=3/group; Student's t-test). Data are means±SD. **P<0.01, ***P<0.001. Figure 4 miR-382-5p inhibits the ability of YBX1 expression to inhibit LSCC growth. a The invasive and migratory activity of Tu212 cells was assessed via Transwell and wound healing assays, respectively, following miR-382-5p transfection and/or YBX1 overexpression (n=3/group). b An MTT assay was used to gauge the proliferation of Tu212 cells following miR-382-5p transfection and/or YBX1 overexpression (n=5/group). c The growth of Tu212 tumors in which miR-382-5p had or had not been expressed was assessed using a murine subcutaneous tumor implantation model (n=10/group), with tumor weights and volumes being compared as indicated. Data are means±SD. **P<0.01, ***P<0.001; Student's t-test.