LDHA Promotes Oral Squamous Cell Carcinoma Progression Through Facilitating Glycolysis and Epithelial–Mesenchymal Transition

Aerobic glycolysis is the main pathway for energy metabolism in cancer cells. It provides energy and biosynthetic substances for tumor progression and metastasis by increasing lactate production. Lactate dehydrogenase A (LDHA) promotes glycolysis process by catalyzing the conversion of pyruvate to lactate. Despite LDHA exhibiting carcinogenesis in various cancers, its role in oral squamous cell carcinoma (OSCC) remains unclear. This study demonstrated that LDHA was over-expressed in both OSCC tissues and cell lines, and was significantly associated with lower overall survival rates in patients with OSCC. Using weighted gene correlation network analysis and gene set enrichment analysis for the gene expression data of patients with OSCC (obtained from The Cancer Genome Atlas database), a close association was identified between epithelial–mesenchymal transition (EMT) and LDHA in promoting OSCC progression. The knockdown of LDHA suppressed EMT, cell proliferation, and migration and invasion of OSCC cells in vitro. Moreover, the silencing of LDHA inhibited tumor growth in vivo. Oxamate, as a competitive LDHA inhibitor, was also suppressed diverse malignant biocharacteristics of OSCC cells. Our findings reveal that LDHA acts as an oncogene to promote malignant progression of OSCC by facilitating glycolysis and EMT, and LDHA may be a potential anticancer therapeutic target.


INTRODUCTION
Oral cancer is the most common location for head and neck cancers. Among all oral cancer subtypes, oral squamous cell carcinoma (OSCC) is the most common (1). Despite the existence of mature diagnostic and comprehensive treatment methods, the 5-year survival rate of patients with OSCC remains relatively poor, and ∼50% of patients with OSCC die within 5 years. The main causes of the poor prognosis of OSCC are early local invasion and distant metastasis (2).
Aerobic glycolysis, i.e., the Warburg effect, constitutes one of its important features during the development of OSCC (3)(4)(5). Cancer cells enhance glucose uptake and lactate production through aerobic glycolysis to meet energy metabolism and biosynthesis needs. Aerobic glycolysis can increase the lactate content of cancer cells and reduce the pH of the tumor microenvironment. As an important carcinogen, lactate is closely associated with tumor growth, immune escape, angiogenesis and epithelial-mesenchymal transition (EMT) (6)(7)(8). In the glycolytic pathway, lactate dehydrogenase A (LDHA) irreversibly catalyzes the conversion of pyruvate to lactate via the oxidative dehydrogenation of nicotinamide adenine dinucleotide (NADH) to NAD+. Several studies have reported that LDHA is highly expressed in various cancers such as pancreatic cancer, gastric, gallbladder, and breast cancers, as well as nasopharyngeal and hepatocellular carcinomas; its expression is positively correlated with the malignant progression of the cancers (9)(10)(11)(12)(13)(14)(15). However, the relationship between LDHA expression and the progression of OSCC, as well as the underlying mechanisms that affect the biological behavior of OSCC cells, are still unclear.
Weighted gene correlation network analysis (WGCNA) uses up to 10,000 of the most variable genes (or in some cases all genes) to identify the target gene set, construct a weighted correlation network, and identify modules comprising highly related genes (16). By using WGCNA, it can identify the gene module that is highly synergistic with LDHA expression in order to explore some of the biological processes that have undergone changes in patients with OSCC. The enrichment of gene expression in biological functions and pathways can be analyzed using gene set enrichment analysis (GSEA). This approach ranks genes according to the degree of differential expression of the two sample types; subsequent testing establishes whether the predetermined gene sets are enriched at the top or bottom of the ranking table.
In this study, we aimed to detect the expression of LDHA in OSCC and explore its effect on the malignant progression of OSCC. We first demonstrated that LDHA was up-regulated in OSCC. Next, we combined gene expression data with GSEA and WGCNA to confirm that LDHA was associated with the EMT gene set and identified that EMT progression could be inhibited by LDHA silencing. Furthermore, we found that LDHA knockdown suppressed OSCC cell growth both in vitro and in vivo. Finally, we used a competitive LDHA inhibitor, oxamate (17), which was found to significantly inhibit glycolysis, proliferation, migration and invasion of OSCC cells. Based on these results, it was revealed that LDHA plays an important role in the progression of OSCC through EMT. Therefore, it is proposed that LDHA may be a potential therapeutic target in OSCC.

OSCC Sample, Cell Lines, and Culture
Eighty-nine OSCC tissues samples, 18 matched adjacent noncancerous normal tissues (ANCT), and two normal oral mucosal epithelial tissues were collected at the Hospital of Stomatology, Sun Yat-sen University. The study was approved by the hospital's ethics committee and was conducted in accordance with the Declaration of Helsinki's guidelines, between September 2017 and December 2018. All patients provided informed consent to participate in the study. The fresh tissues were then divided into two parts after tumor resection, one of which was stored in a cryotube at −80 • C for RNA and protein extraction, and the other was fixed in formalin and embedded in paraffin for immunohistochemical analysis. All staging was determined according to the 7th American Joint Committee on Cancer Staging System.

RNA Isolation and Quantitative Real-Time PCR (qRT-PCR)
Total RNA was extracted from the cells and tissues using RNAzol R RT (RN190, Molecular Research Center, USA) following the manufacturer's instructions. Then, 1 µg of total RNA was reverse-transcribed to complementary DNA using the PrimeScript RT reagent kit (RR036A, Takara Bio, Japan). The qPCR reactions were performed with a SYBR Green Master Mix (11201ES08, Yeasen, China) and were run at 95 • C for

Lactate Production and Glucose Consumption Measurement
HSC3 and SCC15 cells were seeded in 6-well plates at 2.5 × 10 5 cells per well and a well without cells as the control. After the cells were attached, they were switched to a complete medium. The medium was collected at 24 h of incubation to determine lactate production and glucose consumption. The respective concentrations of lactate production and glucose consumption were measured with commercial kits (a lactate assay kit [A019] from Nanjing Jiancheng Biotech, and a glucose assay kit [E1010] from Applygen, China) according to the manufacturer's instructions.

Cell Proliferation and Colony Formation Assay
HSC3 and SCC15 cells were seeded in 96-well plates at a density of 2,000 cells per well. Cell proliferation was detected using a Cell Counting Kit-8 (CCK-8, 40203ES80, Yeasen, China) at 0, 24, 48, 72, and 96 h. In each well, 100 µl of 10% CCK-8 reagent (10 µl CCK-8 and 90 µl serum-free media) was added, and the plates were incubated for 1 h at 37 • C. The absorbance values were measured at 450 nm using a microplate reader (Bio-Rad, USA), and the growth curve was plotted based on absorbance and time.
The colony formation assay was conducted by inoculating 10,000 HSC3 cells or 500 SCC15 cells in 6-well plates and fixing and staining after 7-10 days of culture. The number of HSC3 cell colonies were counted under a microscope at 5× magnification by selecting three random fields. Colony numbers in the SCC15 cells were directly enumerated.

Wound-Healing, Migration, and Invasion Assays
For the wound-healing assay, HSC3 and SCC15 cells were seeded and cultured up to 90% of the coverage area in 6-well plates. The cells were scraped with a sterile pipette tip, washed twice with phosphate-buffered saline (PBS), and treated with FBS-free media for 24 h. Five fields of view were randomly selected under a microscope with 100× magnification. The distances between two cell edges were quantified using Image J and were used to calculate the wound healing rate. A Transwell assay of cell migration and invasion was conducted using 5 × 10 4 HSC3 cells or 1.4 × 10 5 SCC15 cells in 200 µl FBS-free media. They were plated on the upper chambers of the Transwell (for cell migration assay, 8-µm pore size) (Corning, China), and Transwell coated with 0.33 mg/ml Matrigel (for cell invasion assay, 10 mg/ml, 354234, Corning, China), while 600 µl of complete media was added to the lower chamber. After a 24-h incubation period, cells that passed through the chamber filter and attached to the surface of the filter were fixed, stained with hematoxylin, and counted under a 100× microscope in five random fields per chamber according to the number of nuclei.

LDH Activity Assay
HSC3 and SCC15 cells were seeded and cultured in 96-well plates to a density of nearly 80%; oxamate was treated for 24 h. The lactate dehydrogenase (LDH) activity of the cells was determined using the LDH Assay Kit (C0017, Beyotime, China), in accordance with the manufacturer's instructions. Absorbance values were detected using a Bio-Rad microplate reader at 490 nm with a reference wavelength of 600 nm.

Tumor Xenograft
Female NOD/SCID mice (5-6 weeks of age) were randomly distributed into three groups: the shNS group, the shLDHA-1 group, and the shLDHA-2 group (n = 6 in each group). Approximately 5 × 10 7 HSC3 cells per mouse were diluted in 200 µl of PBS and were subcutaneously injected into the inner thigh to establish a subcutaneous transplanted model. The transplanted tumors were measured weekly to observe tumor growth. Tumor volume was calculated using the formula V = length × width × width/2. The mice were euthanized 4 weeks after injection, and the tumor xenografts were harvested, photographed, and fixed for immunohistochemical staining. This animal experiment was approved by Sun Yat-sen University's Animal Experiment Ethics Committee.

The Cancer Genome Atlas (TCGA) Dataset Collection and Data Processing
The head and neck squamous cell carcinoma dataset (566 cases) from TCGA, including preprocessed level 3 RNAseq and clinical data, was downloaded using the UCSC Xena browser on 15 May 2019 (https://xenabrowser.net) (18). Among these cases, clinical samples from the oral cavity (oral tongue, floor of mouth, hard palate, alveolar ridge, oral cavity, and buccle mucosa) comprised 284 OSCC and 30 matched normal oral mucosal epithelial tissues. The gene expression data and survival results of 284 patients with OSCC were used for the survival analysis. Next, the 284 patients with OSCC were divided into high-and low-expression groups based on the median expression levels of LDHA for the GSEA and WGCNA. All the above processes were conducted using R software (R version 3.5.3) and related R packages.

WGCNA
First, the top 6000 of the most variant genes and OSCC samples were examined by respectively, applying the goodSamplesGenes function and the sample network method to remove the offending genes and samples (Z.K < −2.5). Second, the WGCNA package was applied to construct a gene co-expression network. A soft threshold power was selected based on approximate scale-free topological criteria and was used to calculate the adjacency. Next, the adjacency matrix was transformed into a topological overlap matrix, from which the corresponding dissimilarity was calculated. Third, a hierarchical clustering gene tree was generated using the topological overlap matrix, and genes with similar expression profiles were combined into the same gene module. Fourth, the module eigengene was the first principal component of a particular module and represented the overall level of gene expression within that module. The module eigengenes were calculated to identify modules that were significantly associated with the LDHA expression. According to the significance of the gene network, gene significance was defined to quantify the correlation between the genes and the LDHA expression. A quantitative measure of module membership was defined as the correlation between the gene expression profile and the module eigengene and the intramodular connectivity was defined as the connectivity between a gene and other genes within a module. Finally, gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed for the genes in the target modules.

GSEA
GSEA was conducted between the LDHA high-expression and low-expression groups. The association of LDHA expression with oncogenetic biological function was performed using the Hallmark gene sets. A p-value of <0.05 and a false discovery rate (FDR) of <0.05 for a gene set were considered statistically significant.

Statistical Analysis
The mean and the standard deviation of data were analyzed using GraphPad prism 7.0. software. The Shapiro-Wilk test was used to test normal distribution. Statistical differences between the two groups were analyzed using a two-tailed unpaired or paired Student's t-test; for multiple groups, ANOVA with Tukey's multiple comparisons tests were performed. The association between LDHA expression and clinicopathological features in patients with OSCC was conducted using Fisher's exact test. The correlation between genes was analyzed using Pearson's correlation. A p-value of <0.05 was considered statistically significant: * p < 0.05, * * p < 0.01, * * * p < 0.001.

LDHA Is Up-Regulated in OSCC and Associated With Poor Prognosis
To detect the expression level of LDHA in OSCC, immunohistochemical staining was performed for 89 OSCC tumor tissues and 18 ANCTs. Depending on the Histo-score, it was revealed that OSCC tissue demonstrated a higher staining score for LDHA for ANCT (Figures 1A,B). In addition, LDHA with a poor degree of differentiation compared to a good degree of differentiation middle-differentiated and well-differentiated LDHA was also highly expressed ( Figure 1C). Patients with TNM III-IV had similar results compared with patients with TNM I-II ( Figure 1D). Further analysis showed that LDHA expression was associated with the degree of differentiation and T stage of the OSCC. However, gender, age, TNM stage, and lymph-node metastasis did not differ between patients with high and low LDHA expression ( Table 1). RT-qPCR analysis of LDHA mRNA levels in 25 OSCC tumor tissues and 15 ANCTs revealed that LDHA mRNA levels were elevated in the tumor tissues of patients with OSCC ( Figure 1E). Additionally, Western blot analysis indicated that the protein level of LDHA was up-regulated in OSCC tumor tissues (Figure 1F). The results of an analysis of the mRNA levels of LDHA in normal oral mucosal epithelial tissues and OSCC cell lines (SCC1, SCC25, HSC6, SCC15, and HSC3) indicated that the mRNA level of , disease-free survival, and disease-specific survival based on patients with OSCC with high-and low-expression LDHA. Differences between the two groups were compared using a log-rank test. The experiment was repeated three times; error bars indicate standard deviation. *p < 0.05, **p < 0.01, ***p <0.001.
LDHA was also highly expressed in OSCC cells lines compared with that in normal oral mucosal epithelial tissues ( Figure 1G). Next, 284 OSCC samples and 30 matched normal oral mucosal epithelial tissues in the TCGA database were compared; an up-regulation of LDHA mRNA levels in the OSCC samples was evident ( Figure 1H). A survival analysis revealed a negative correlation between the high LDHA expression and overall survival rates; however, there was no statistical difference in the rates for disease-free survival and disease-specific survival ( Figure 1I).

Construction of a Weighted Correlation Network and the Target Module Identification
WGCNA was used to explore potential gene modules related to LDHA expression in the gene expression data of patients with OSCC. Seven abnormal samples were removed by establishing the connection screening threshold, and the remaining 277 samples were clustered with LDHA expression levels (Supplementary Figure 1A). The soft threshold was selected as β = 6 to obtain the neighboring and topology matrices, which caused 6,000 of the most variant genes to conform to the scale-free network (Supplementary Figure 1B). The topological overlap of the gene networks can be visualized using heatmaps (Supplementary Figure 1C). The clustering tree was divided into 11 modules using dynamic shearing (gray represents a gene not assigned to any module; the module's minimum gene number is 30), and the modules were merged according to the coefficient of dissimilarity of <0.2 (Supplementary Figures 1D,E). Supplementary Figures 1F,G illustrate the relationships between LDHA expression and the module eigengenes and the correlation heatmap of each module with LDHA expression. The module most significantly associated with LDHA expression (r = 0.39, p = 1 × 10 −11 ) was the red module. By measuring the correlation between gene significance and module membership, the genes in the red module that were highly correlated with LDHA expression were often the most crucial elements of modules associated with LDHA expression (Supplementary Figure 1H).

GO and KEGG Analysis for the Target Module
To further understand the gene function in the red module, this study applied the R package "clusterProfiler" for GO and KEGG analyses (Supplementary Data Sheet 1). The results of the GO analysis results revealed that the enrichment of genes in the biological process (BP) classification includes the cell-cell adhesion via plasma-membrane adhesion molecules, homophilic cell adhesion via plasma membrane adhesion molecules, hemidesmosome assembly, extracellular matrix (ECM) organization, and regulation of ventricular cardiac muscle cell action potential (Figure 2A). Genes in the red module were mainly enriched in the ECM region in the cellular component classification (Figure 2B). The results of KEGG analysis demonstrated that the genes were mainly enriched during ECM-receptor interaction, small cell lung cancer, focal adhesion ( Figure 2C).

LDHA Regulates EMT Progression in OSCC Cells
The R package "clusterProfiler" was used to perform GSEA in all the differentially expressed genes between the LDHA highexpression and low-expression groups in the OSCC samples (19). Statistically significant differences in the GSEA (p-value < 0.05, FDR < 0.05) were revealed using Hallmark gene sets in 39 enriched gene sets (Supplementary Data Sheet 2) (20). GSEA results showed that the LDHA high-expression group activated more EMT progression compared with the LDHA lowexpression group (Figure 2D). Based on the GSEA and WGCNA results, it is assumed that LDHA may induce EMT to promote the malignant progression of OSCC by affecting cell adhesion and changing the components of ECM. Therefore, a Western blot and RT-qPCR was conducted on human HSC3 and SCC15 cells with stable LDHA knockdown (shLDHA-1 and shLDHA-2) in order to detect EMT related markers [ZO1 (TJP1), E-cadherin (CDH1), N-cadherin (CDH2), Vimentin (VIM), and Slug (SNAI2)] (Figures 2E,F). The result showed that LDHA expression was negatively correlated with the epithelial phenotype and positively correlated with the mesenchymal phenotype (Figures 2G,H).
The same results were also evident in the immunohistochemistry of tumors in the shLDHA group compared with those in the shNS group (Figure 2I). By inducing EMT in OSCC cells treated with 10 ng/ml of TGF-β1 for 48 h, we found that TGF-β 1 induced downregulation of the epithelial phenotype marker and the upregulation of the mesenchymal phenotype marker were rescued by LDHA knockdown (Figures 2J,K). Collectively, these data confirmed that LDHA loss inhibited the EMT progression of OSCC cells.

LDHA Silencing Inhibits Aerobic Glycolysis, Proliferation, Migration, and Invasion of OSCC Cells
This study revealed that LDHA promotes the EMT progression of OSCC and that lactate provides the necessary metabolic and biosynthetic raw materials for the growth of tumor cells (8). We then investigated whether LDHA was essential for the proliferation and metastasis of OSCC cells. The knockdown of LDHA decreased glucose consumption and lactate production in HSC3 and SCC15 cells (Figures 3A,B). Subsequently, we confirmed that the inhibition of LDHA suppressed the proliferation of HSC3 and SCC15 cells using theCCK-8 and colony formation assays (Figures 3C,D). Furthermore, woundhealing, migration, and invasion assays were utilized to detect the effect of LDHA on the migration and invasion of OSCC cells. As shown in Figures 3E,F, the knockdown of LDHA inhibited migration and invasion of HSC3 and SCC15 cells. We further examined the role of LDHA on OSCC tumor formation in vivo through subcutaneous injections of HSC3 cells into the inner thighs of NOD/SCID mice. The results revealed a significantly lower tumor volume and weight in the shLDHA group compared with the shNS group. This suggests that silencing LDHA can effectively inhibit OSCC tumor growth in mice (Figures 3G-I).

Effects of Oxamate on the Biological Function of OSCC Cells
To explore the potential efficacy of the targeted inhibition of LDHA, we used oxamate to inhibit the LDH activity to detect the biological function of OSCC cells ( Figure 4A). As shown in Figures 4B,C, glucose consumption and lactate production were reduced in a dose-dependent manner by treating HSC3 and SCC15 cells with various concentrations of oxamate. Furthermore, oxamate also suppressed the proliferation, migration, and invasion of HSC3 and SCC15 cells in a dose-dependent manner (Figures 4D-G).

DISCUSSION
Elevated aerobic glycolysis remains the predominant feature of cancer metabolism and is deeply dependent on dysregulated metabolic enzymes (21,22). The carcinogenic action of LDHA has been reported in several cancers, including pancreatic and gallbladder cancer, nasopharyngeal carcinoma, and breast cancer, and it promotes the malignant progression of cancer by increasing lactate production, accelerating glucose uptake, and regulating many cancer-associated molecules (9,12,(23)(24)(25). However, the role of LDHA in the initiation and progression of OSCC is not yet fully understood. This study demonstrated that LDHA acted as an oncogene in OSCC to facilitate cell growth and metastasis both in vitro and in vivo. Treatment with oxamate reduced glucose consumption and lactate production in OSCC cells and inhibited cell proliferation, migration, and invasion.
To determine the association between LDHA expression levels and OSCC progression, we performed WGCNA and GSEA using OSCC patients' gene expression data from the TCGA database. The results indicated that LDHA was closely associated with EMT and that silencing LDHA decreased the expression of EMTrelated markers in OSCC cells. The present study revealed that LDHA promoted OSCC cell growth and metastasis by inducing EMT Progression. As the last key enzyme in the glycolysis process, LDH plays an important role in the conversion of pyruvate to lactate, indicating its essential role in tumor metabolism. LDH is a homo-or heterotetramer composed of two subunits: LDHA (M) and LDHB (H); these are encoded by LDHA and LDHB, respectively, and exhibit different isoenzyme compositions in different tissues. The LDHA isoform is mainly expressed in skeletal muscle, which preferentially converts pyruvate into lactate, while the LDHB isoform is chiefly expressed in the heart and brain, where it preferentially converts lactate into pyruvate (26). In LDHA knockout human colon adenocarcinoma (LS174T) cells, the LDHB levels were upregulated under normoxic conditions; however, hypoxic conditions had no effect. However, the LDHB expression did not change in LDHA knockout mouse melanoma (B16-F10) cells, as analyzed using Western blotting (27). Additionally, the acquisition of ectopic fibroblast growth factor receptor 1 increased the stability of LDHA by tyrosine phosphorylation and reduced LDHB expression by supporting its promoter methylation, thereby promoting the conversion of prostate cancer cell metabolism from oxidative phosphorylation to aerobic glycolysis (28). In this study, LDHA was highly expressed in OSCC tissues and cells and was associated with poor prognosis in patients with OSCC. This indicates that LDHA acts as an oncogene in the context of OSCC.
To investigate the molecular mechanism by which LDHA promotes the progression of OSCC, we used WGCNA and GSEA to identify gene sets associated with LDHA expression using OSCC gene expression data provided by TCGA. For WGCNA, we constructed a weighted correlation network to identify the most significant module associated with LDHA expression, which contained 238 genes. According to the results of GO and KEGG enrichment analysis, key module genes were primarily enriched in cell adhesion and ECM function regulation. Recent studies reported that LDHA can promote the progression of renal cell carcinoma and bladder cancer by promoting EMT (29,30). Moreover, lactate, the ultimate metabolite of aerobic glycolysis, led to an acidic tumor microenvironment that facilitated the degradation of ECM and the production of actin filaments (31). The degradation and remodeling of the ECM highly influences and controls both epithelial cell phenotype maintenance and EMT progression (32). Additionally, the mechanical properties and organization of the ECM are essential in regulating EMT (33). Thus, the results of GO and KEGG enrichment analysis revealed that EMT played a key role in the malignant progression of LDHA regulation of OSCC. GSEA indicated that the LDHA high-expression group activated more EMT gene sets compared with the LDHA low-expression group. This further demonstrates that LDHA regulates EMT in OSCC. We also examined the effect of LDHA on the expression of EMT-related markers. These results suggest that LDHA might induce EMT to promote the malignant progression of OSCC by affecting cell adhesion and changing the components of the ECM.
EMT is a key mechanism that promotes tumor migration and invasion (34,35). Moreover, aerobic glycolysis remodels the ECM and facilitates EMT-inducing transcription factor expression by increasing the lactate content surrounding cancer cells and reducing the pH of the tumor microenvironment; these factors indicate that LDHA induces EMT and promotes cancer cell metastasis (8,31,33,36). Additionally, the upregulation of LDHA enhances lactate production by aerobic glycolysis, which provides sufficient biomaterials and energy for the growth and metabolism of cancer cells. In recent years, numerous studies have reported that LDHA knockdown could suppress the growth and metastasis of cancer cells (9,12,25). This evidence suggests that LDHA promotes diverse malignant biocharacteristics. The results of the present study demonstrated that LDHA knockdown reduced glucose consumption and lactate production in OSCC cells and suppressed cell proliferation, migration, and invasion in vitro. Meanwhile, silencing LDHA inhibited the growth of OSCC cells in vitro. Results of comprehensive experiments conducted both in vitro and in vivo, LDHA exerted a cancerpromoting effect on the progression of OSCC by regulating cell glycolysis.
The catalytic substrate of LDHA is competitively inhibited by oxamate, a pyruvate analog, and its effect has been identified in various cell lines (15,(37)(38)(39). A study by Zhao et al. (15) demonstrated that oxamate inhibited the activity of gastric cancer cells in a dose-and time-dependent manner. El-Sisi et al. (37) documented that the combination of oxamate and Taxol reduced levels of proinflammatory mediators, such as TNFα, IL-17, MDA and ATP, in solid Ehrlich carcinomas in mice and facilitated apoptosis and anti-angiogenic effects compared with paclitaxel monotherapy. In the present study, oxamate was shown to inhibit OSCC cell glycolysis, proliferation, and metastasis in a dose-dependent manner. However, owing to limited membrane permeability, the effective dose of oxamate administered to cancer cells in vitro is too high relative to in vivo administration, which limits its potential for clinical application. Currently, effective inhibitors of LDHA are under development, including Gossypol (40), FX11 (41), Quinoline 3-sulfonamides (42), NHI (43), Galloflavin (44). Gossypol has demonstrated clinical efficacy in metastatic adrenal cancer and recurrent adult malignant gliomas (45,46). Developing specific and effective LDHA small molecule inhibitors will be a more appropriate therapeutic strategy for OSCC.
In summary, the present study confirmed the role of LDHA as an oncogene during OSCC progression. Compelling in vitro and in vivo evidence demonstrates that LDHA can promote the proliferation, migration, and invasion of OSCC cells by facilitating glycolysis and EMT progression. These findings provide strategies for silencing or deactivating LDHA in carcinoma tissues and may be a new therapeutic target for OSCC.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/Supplementary Material.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by The ethics committee of the Hospital of Stomatology, Sun Yat-sen University. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by Institutional Animal Care and Use Committee of Sun Yat-sen University, Guangzhou, China.