OGDHL closely associates with tumor microenvironment and can serve as a prognostic biomarker for papillary thyroid cancer

Abstract Background Papillary thyroid cancer (PTC) is the most common type of thyroid cancer. However, due to the lack of reliable prognostic biomarkers for PTC, overtreatment has been on the rise. Therefore, our research aims to identify new and promising prognostic biomarkers and provide fresh perspectives for clinical decision making. Methods The RNA‐seq data and clinical data of PTC samples were obtained from The Cancer Genome Atlas data portal. GSE64912 and GSE83520 datasets were downloaded through the GEOquery R package. The difference in the expression of oxoglutarate dehydrogenase like (OGDHL) between PTC and normal tissues was explored by the Wilcoxon test. Kaplan–Meier (KM) and Cox regression analyses were used to further explore the prognostic value of OGDHL. The tumor microenvironments of PTC patients were explored based on ssGSEA and Tumor Immune Estimation Resource online database. Gene Set Enrichment Analysis (GSEA) was performed to explore the biological processes associated with OGDHL. Results The expression level of OGDHL in PTC was significantly altered compared to that in normal tissues (p < 0.05). Various biological processes associated with OGDHL were also explored through GSEA. KM analysis suggested that the low‐OGDHL group had a better overall survival [OS, p = 3.49e‐03, hazard ratio (HR) = 4.567]. The receiver operating characteristic curve also indicated the favorable prognostic potential of OGDHL. Moreover, OGDHL was proved to be an independent prognostic indicator in Cox analysis (p = 1.33e‐02, HR = 0.152). In the analysis of the tumor microenvironment, the low‐OGDHL group showed a lower immune score and stromal score, while tumor purity was higher. The expression of OGDHL was also closely correlated with the infiltration of immune cells. Conclusion Our study elucidated the influence of OGDHL on the prognosis of PTC and demonstrated its potential as a novel biomarker, which would provide new insights into the prognosis monitoring and clinical decision making in PTC patients.

prognostic biomarkers and provide fresh perspectives for clinical decision making. Methods: The RNA-seq data and clinical data of PTC samples were obtained from The Cancer Genome Atlas data portal. GSE64912 and GSE83520 datasets were downloaded through the GEOquery R package. The difference in the expression of oxoglutarate dehydrogenase like (OGDHL) between PTC and normal tissues was explored by the Wilcoxon test. Kaplan-Meier (KM) and Cox regression analyses were used to further explore the prognostic value of OGDHL.
The tumor microenvironments of PTC patients were explored based on ssGSEA and Tumor Immune Estimation Resource online database. Gene Set Enrichment Analysis (GSEA) was performed to explore the biological processes associated with OGDHL. Results: The expression level of OGDHL in PTC was significantly altered compared to that in normal tissues (p < 0.05). Various biological processes associated with OGDHL were also explored through GSEA. KM analysis suggested that the low-OGDHL group had a better overall survival [OS, p = 3.49e-03, hazard ratio (HR) = 4.567]. The receiver operating characteristic curve also indicated the favorable prognostic potential of OGDHL. Moreover, OGDHL was proved to be an independent prognostic indicator in Cox analysis (p = 1.33e-02, HR = 0.152).
In the analysis of the tumor microenvironment, the low-OGDHL group showed a lower immune score and stromal score, while tumor purity was higher. The expression of OGDHL was also closely correlated with the infiltration of immune cells.

| INTRODUCTION
Papillary thyroid cancer (PTC) is the most common type of thyroid cancer (TC), accounting for approximately 80% of all reported cases of TCs. 1,2 In recent decades, the development of thyroid cancer diagnostic technology has increased the detection rate of TC rapidly, and the incidence of TC is predicted to replace colon cancer as the fourth leading cancer by 2030. 3 As the most differentiated TC, the vast majority of PTC patients are at low risk, and do not require immediate and aggressive intervention. 4 However, due to lack of reliable biomarkers for determining the degree of malignancy of PTC, overtreatment of PTC has been increasing in recent years. 5 Therefore, it has become essential to find new and reliable prognostic biomarkers for PTC to provide better clinical decisions for patients and reduce overtreatment.
Oxoglutarate dehydrogenase like (OGDHL), a subunit of the oxoglutarate dehydrogenase complex (OGDHC), is not only the main rate-limiting component of OGDHC for the degradation of glucose and glutamate 6,7 but is also involved in the tricarboxylic acid cycle, and in the development and progression of many types of cancers. [8][9][10][11][12] Sen et al. reported that forced expression of OGDHL in cervical cancer can downregulate the protein kinase B (AKT) signaling cascade and reduce the phosphorylation of nuclear factor kappa lightchain enhancer of activated B cells (NF-κB) mediated by caspase 3. These activities increase the production of reactive oxygen species (ROS) and lead to cell apoptosis, which, in turn, can inhibit the proliferation and metastasis of cervical cancer cells. 8 A relevant study also showed that the miR-214/ TWIST1 (microRNA 214/ Twist-related protein 1) negative feedback loop mediated by OGDHL can inhibit the growth and metastasis of pancreatic cancer. 9 Shen et al. demonstrated that in hepatocellular carcinoma (HCC), promoter hypermethylation and DNA copy deletion of OGDHL were associated with reduced OGDHL expression, and silencing of OGDHL can further lead to the occurrence and development of HCC by regulating the glutamine metabolism pathway. 10 Similarly, in breast cancer and sporadic colorectal cancer, abnormal hypermethylation of the OGDHL gene promoter also promotes cancer progression. 11,12 These studies have revealed that OGDHL is closely associated with the development and progression of various cancers. However, the relationship between OGDHL and PTC still remains poorly understood and needs further investigation.
In the present study, we explore the possible mechanisms closely related to the tumor microenvironment in an attempt to elucidate and comprehensively analyze the relationship between the expression of OGDHL and the prognosis of PTC. We hope to provide a reliable base for prognostic prediction and clinical decision-making optimization in patients with PTC.

between PTC and normal tissues
The fragments per kilobase of exon model per million mapped reads (FPKM) RNA-seq data and clinical data of 493 PTC samples and 58 non-tumor samples were obtained from The Cancer Genome Atlas (TCGA) data portal (https://cance rgeno me.nih.gov/). The Ensembl IDs were converted into gene symbols based on the Genome Reference Consortium human (GRCh38.98) genome file acquired from the Ensembl database (http://asia.ensem bl.org/index.html). The expression file for OGDHL was then extracted for further analysis. The Wilcoxon test was performed to ascertain the differential expression between PTC and normal tissues. Additionally, the expression profiles of GSE64912 (with 18 primary PTC samples and 4 non-tumor samples) and GSE83520 (with 12 primary PTC samples and 12 non-tumor samples) were also downloaded from Gene Expression Omnibus (GEO. http:// www.ncbi.nlm.nih.gov/geo/) and the results were verified using the Wilcoxon test or paired Wilcoxon test through the GEOquery R package.

COX analysis
The prognostic value of OGDHL was estimated through survival analysis using a log2 (FPKM value +1) data format. We used Kaplan-Meier (KM) analysis to confirm statistical significance, and the cutoff value was determined from its median value. Univariate and multivariate Cox analyses were Conclusion: Our study elucidated the influence of OGDHL on the prognosis of PTC and demonstrated its potential as a novel biomarker, which would provide new insights into the prognosis monitoring and clinical decision making in PTC patients.

K E Y W O R D S
OGDHL, papillary thyroid cancer, ssGSEA, tumor microenvironment performed to compare the survival impact of OGDHL with other clinical characteristics. The receiver operating characteristic (ROC) curves within 3-years and 5-years were established to further analyze the prognostic potential of OGDHL in PTC using the timeROC R package.

| OGDHL mRNA level and tumor microenvironment
The cell infiltration score of the tumor microenvironment between high and low-OGDHL groups was further analyzed in order to explore the biological mechanism of OGDHL in PTC. The tumor microenvironment is generally composed of a variety of cell types, including immune cells, mesenchymal cells, endothelial cells, inflammatory mediators, and extracellular matrix (ECM) molecules. 13 In this study, the ESTIMATE algorithm was used to determine the quantitative cell score of the tumor microenvironment for each PTC patient. Furthermore, the difference between the high and low-OGDHL groups was further compared using the Wilcoxon test.
Additionally, the tumor immune cell infiltrating abundance of each PTC patient was calculated using the Tumor Immune Estimation Resource online database (https://cistrome. shinyapps.io /timer) to further explore the immune-related mechanisms between OGDHL expression and the prognosis of PTC patients. 14 Pearson correlation analysis was used to further evaluate the relationship between the infiltrating abundance of each component and OGDHL expression. A p value < 0.05 and /R/ >0.15 indicated that a correlation existed between OGDHL expression and immune cell infiltration.

| Gene Set Enrichment Analysis (GSEA)
GSEA was used to explore the signaling pathways and biological processes which were differentially activated between the high and low-OGDHL groups. We also identified an ordered list of genes using the edgeR R package and executed GSEA on the gene with adjusted p < 0.05. The number of random sample permutations was set at 1000, while the minimum size of the gene set was stipulated at 350.

| Statistical analysis
All statistical analyses were performed using the R software (version 3.6.1) (http://www.r-proje ct.org/) and the corresponding R packages. The KM analysis and Cox regression analyses were completed using the survival R package. The Area Under Curve (AUC) of the ROC curve was calculated using the timeROC package for R. The c5.bp.v6.2.entrez. gmt file from the Molecular Signatures Database (MSigDB, http://softw are.broad insti tute.org/gsea/index.jsp) was downloaded for the GSEA and was completed by using the clus-terProfiler R package.

between PTC and normal samples
OGDHL expression in the TCGA RNA-seq, GSE83520, and GSE64912 datasets showed significant statistical differences (Figure 1). OGDHL was lower in PTC samples than in the normal samples (p < 0.05, Figure 1A-1C).

Cox analysis
We observed that patients with high-OGDHL expression were correlated with worse overall survival in the KM analysis [p =3.49e-03, hazard ratio (HR) = 4.567, 95% confidence interval (CI) = 1.677-12.435] compared to patients with low-OGDHL expression (Figure 2A). In the meantime, the time-dependent ROC curve analysis indicated that OGDHL had a promising prognostic potential for predicting overall survival in PTC (area under the curve, AUC 3-year = 0.759, AUC 5-year = 0.833) ( Figure 2B). In addition, comparison with other clinical features by univariate and multivariate Cox regression analyses also revealed that OGDHL was independently associated with overall survival (OS) (HR = 6.576, p = 1.33e-02) ( Table 1).

| High-and low-OGDHL groups displayed different tumor microenvironments
The results from the ESTIMATE algorithm showed that the immune score, stromal score, and tumor purity of the tumor microenvironment were statistically different between the high and low-OGDHL groups. The low-OGDHL group had a lower immune score, lower stromal score, but a higher tumor purity than the high-OGDHL group ( Figure 3A-3C, p < 0.05). These observations indicated that a difference in OGDHL expression in PTC patients results in a difference in the respective tumor microenvironments.
In addition, we further analyzed the relationship between the immune cell components of the tumor immune microenvironment and the expression of OGDHL. Pearson correlation analysis showed that immune cell infiltration in the tumor microenvironment was associated with expression of OGDHL ( Figure 4A, p < 0.05). The B cells, CD4+ cells, neutrophils, macrophages, and dendritic cells were all negatively correlated with OGDHL according to our criterion ( Figure 4B-F, p < 0.05).

| The different immune-related biological processes associated with OGDHL
GSEA revealed that various immune-associated biological processes were enriched, including immune response, immune system process, regulation of immune system response, and immune system development ( Figure 5A-5D, S1).

| DISCUSSION
OGDHL, as the main rate-limiting component of 2-oxoglutarate dehydrogenase complex (OGDHC) for glucose and glutamic acid degradation, plays an important role in the development of a wide variety of cancers 7 and tumors by influencing cell cycle arrest, 9 cell apoptosis, 8,15 and energy metabolism. 10 Studies have shown that methylation of the OGDHL promoter contributes to the development of many cancers such as breast cancer 11 and colon cancer. 16 Our analysis also revealed significant changes in OGDHL expression in PTC samples, thereby justifying our pursuit in elucidating the role of OGDHL in PTC.
It is worth noting that as per the KM analysis, the low-OG-DHL group had a better OS than the high-OGDHL group, while the differential expression analysis showed a decreased level of OGDHL expression in PTC samples than that in the normal tissues. This observation was rather remarkable as it suggested that OGDHL may not only play different roles in the occurrence and progression of PTC but may also be closely related to the unique molecular mechanisms and functions of OGDHL affecting PTC. In addition, OGDHL was also determined as an independent prognostic marker of PTC by Cox analysis and ROC curve analysis. Therefore, it is evident that OGDHL has a novel and promising potential to be employed as a reliable prognostic biomarker for PTC patients. We performed GSEA to explore the biological mechanism of OGDHL acting on PTC and demonstrated that diverse biological processes endowed with different OGDHL expression levels were closely related to tumor immunity, including immune response, regulation of the immune system, immune system processes and overall immune system development. The immune system is known to coordinate with other systems for maintaining the internal stability and physiological balance of the body, 17,18 but in-depth studies on tumor antigens have revealed that the immune response-a physiological process employed by the immune system to eliminate antigens-plays an increasingly important role in the development of tumors. [19][20][21] The regulation of the immune system also acts on many aspects of the immune response process such as maintaining the immune responses at an appropriate level and avoiding immune response disorder, which, in turn, are essential for sustaining the stability of the body's internal environment and influencing the development of cancers. 18,22 Existing studies have demonstrated the increasing importance of immunity in PTC, and various immunotherapies have also been applied in PTC. [23][24][25] These observations are consistent with our analysis that the expression of OGDHL in PTC is closely related to biological immune processes. Therefore, we believe that our preliminary analysis can broaden our understanding of the influence of OGDHL on PTC immunology and provide a perspective for further exploration.
Tumor development is influenced not only by its characteristics but also by the unique tumor microenvironment. Increasing evidence indicates that the tumor microenvironment plays an important role in predicting tumor development and prognosis. [26][27][28] Our findings from the ESTIMATE algorithm also indicates that groups with different OGDHL expression levels show different tumor microenvironments,

F I G U R E 3
The relationships between the expression of OGDHL and immune score, stromal score, and tumor purity of tumor microenvironment. The low-OGDHL group demonstrated (A) lower immune score, (B) lower stromal score, and (C) higher tumor purity than the high-OGDHL group.
thereby further supporting the potential of OGDHL as a PTC prognostic marker closely related to the PTC tumor microenvironment.
Another important aspect of the tumor microenvironment is immune cell infiltration, which deeply impacts the outcomes of PTC. 29 Some existing studies have shown that neutrophils are involved in promoting tumor proliferation, metastasis, and tumor angiogenesis, 30,31 while others suggest that neutrophils plays an anti-tumor role through the blocking of TGF-β or the activation of cytokines. 32,33 In our study, Pearson correlation analysis also showed that neutrophil infiltration can be beneficial to the prognosis of PTC. Ugolini et al. proved that the infiltration of dendritic cells was associated with a decrease in the invasion rate and the prolongation of disease-free survival of PTC, 34 which is consistent with our analysis. In addition, B cells have been observed to accumulate in primary thyroid tumors, producing antibodies and presenting antigens to T cells, thereby enhancing the immune response, and although there was no evidence collocating B-cell infiltration with disease severity, the positive influence of B cells in eliminating cancer was confirmed. [35][36][37] M1 macrophages, similar to B cells, reportedly promote the cytotoxic natural killing ability and T-cell responses by presenting antigens, which play a role in cancer inhibition. 38 Bastman et al. found that CD8+T cells infiltrated at different statuses of thyroid cancer, such as locally invasive differentiated, anaplastic, and distant metastases of thyroid cancer, which can indirectly protect the patients. 29 In addition, CD4+T cells often differentiate into helper T cells, producing cytokines, and cooperating with CD8+T cells to kill cancer cells yy. 37 Therefore, the interactions between OGDHL and tumor microenvironment immune cell infiltration can be a potential mechanism for correlating OGDHL expression with prognosis in PTC. These observations can be further explored for a holistic understanding of the nuances of PTC microenvironment immune cell infiltration.

| CONCLUSION
This study not only comprehensively discusses the effect of OGDHL on the prognosis of PTC but is also, to our knowledge, the first to elaborate on the relationship between OGDHL and PTC. However, there are certain limitations that need to be considered in our study. First, due to lack of in vitro experiments, the specific mechanism of OGDHL in PTC still needs to be clinically verified. Second, the transcriptome data used in this study were all from core samples F I G U R E 5 Parts of Gene Set Enrichment Analysis (GSEA) results indicating that various immune-associated biological processes are also associated with OGDHL, including (A) immune response, (B) regulation of immune system, (C) immune system processes, and (D) immune system development.
of tumor tissue, so the influence of the microenvironment in different tumor regions on tumor progression could not be considered. Therefore, a well-designed clinical trial is required to further validate our observations, although the mode of action of OGDHL in PTC has been effectively elucidated through various methods, which would provide new insights for prognosis monitoring and clinical decision making in PTC patients.