Genes associate with Immunity and Amino Acid Metabolism in Lung adenocarcinoma (LUAD): A bioinformatic analysis

doi:10.21203/rs.3.rs-3344958/v1

Download PDF

Article

Genes associate with Immunity and Amino Acid Metabolism in Lung adenocarcinoma (LUAD): A bioinformatic analysis

https://doi.org/10.21203/rs.3.rs-3344958/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Lung adenocarcinoma (LUAD) represents the most prevalent subtype of primary lung cancer. Amino acids play a vital role as essential nutrients for both tumor cells and immune cells. Both tumor cells and immune cells exhibit specific and distinctive amino arequirements. Many tumors overexpress enzymes that degrade amino acids, which provide energy and metabolites for anabolic processes and also act as a mechanism for immune evasion of cancells. Thus, an in-depth exploration of the relationship between immunity and amino acid metabolism in LUAD is crucial. The identification of stable and reliable tumor markers can facilitate patient screening for poor prognosis, leading to more aggressive treatment approaches.

Methods

This study utilized 539 LUAD samples and 59 normal samples obtained from The Cancer Genome Atlas (TCGA) database. Differentially expressed genes ( between LUAD and normal tissue were identified through analysis of processed expression profile data. The study focused on genes associated with immune response and amino acid metabolism among the differentially expressed genes. Subsequently, potential mechanisms, biological characteristics, and pathways related to LUAD were investigated in the cancer and normal groups using Gene Ontology/Kyoto Encyclopedia of Genes and Genomes (GO/KEGG), Gene Set Enrichment Analysis (GSEA), and Gene Set Variation Analysis (GSVA). A prognostic model was then established through LASSO-COX analysis, considering risk scores and prognostic factors to identify markers influencing the occurrence and prognosis of LUAD.

Results

Differential expression analysis identified 377 genes at the intersection of up-regulated differentially expressed genes and Amino Acid Metabolism-related genes (AAMGs). Protein-protein interaction (PPI) analysis on these 377 genes, associated with immunity and amino acid metabolism, yielded 17 hub genes selected based on top 30 scores from five algorithms. A LASSO regression analysis-based prognosis model was constructed to evaluate the prognostic value of these 17 hub genes using the TCGA-LUAD dataset. Validation with a combined dataset confirmed four genes, polo-like kinase(PLK1), Ribonucleotide Reductase Subunit M2 (RRM2), Thyroid Hormone Receptor Interactor 13 (TRIP13), and Hyaluronan-Mediated Motility Receptor (HHMR), as consistent results in the TCGA-LUAD dataset. The accuracy of the model was further verified through ROC curve analysis and the COX model. Additionally, immunohistochemical analysis of PLK1 expression in LUAD tumor tissue and normal thyroid tissue from the HPA database, using antibody HPA053229, showed higher PLK1 expression levels in LUAD tumor tissue.

Conclusion

LUAD development is strongly associated with immunity and amino acid metabolism. Four genes, namely, PLK1, RRM2, TRIP13, and HMMR, hold prognostic value for lung adenocarcinoma. High expression of PLK1 in LUAD may contribute to tumorigenesis by regulating the cell cycle and could serve as a prognostic indicator for clinical outcomes.

Biological sciences/Cancer

Biological sciences/Immunology

Lung adenocarcinoma

immunity

amino acid metabolism

GEO

TCGA

DEGs

hub genes

PLK1

RRM2

TRIP13

HHMR

Lung cancer, a prevalent malignant tumor, poses a significant threat to human health. The latest statistical report on the global cancer burden, released by the International Agency for Research on Cancer of the World Health Organization (WHO/IARC), reveals that lung cancer ranks second in terms of incidence and first in mortality^[1–2]. Among the different types of lung cancer, lung adenocarcinoma (LUAD), classified as non-small cell lung cancer (NSCLC), stands as the most common subtype of primary lung cancer, accounting for 40% of all lung cancer cases and displaying an upward trend^[3]. Consequently, the identification of stable and reliable tumor markers becomes crucial to screen patients for poor prognosis and to offer more aggressive treatment options.

Amino acids play a vital role as essential nutrients for both tumor cells and immune cells. Both tumor cells and immune cells exhibit specific and distinctive amino acid requirements^[4–5]. T cells, which are central to the immune system, play a pivotal role in inducing apoptosis of tumor cells and inhibiting tumor occurrence and development through various mechanisms^[6–7]. The activation, differentiation, and function of T cells heavily depend on amino acid transport and metabolism. Furthermore, modern studies have shown that malignant tumor cells undergo rapid growth, characterized by exuberant tissue metabolism, abnormal cell growth, and accelerated anabolism and catabolism. Many tumors overexpress enzymes that degrade amino acids, which provide energy and metabolites for anabolic processes and also act as a mechanism for immune evasion of cancer cells^[8]. Studies have shown that alanine-serine-transporter 2 (ASCT2), an essential amino acid transporter for life, can regulate the differentiation of CD4 + T cell clusters, thus promoting anti-tumor immune response and achieving the goal of inhibiting tumor growth^[9]. Hence, a comprehensive exploration of the interplay between immunity, amino acid metabolism, and LUAD is necessary to decipher the conditions under which specific genes promote or inhibit tumor survival.

In this paper, we analyze the expression changes of genes related to Immunity and Amino Acid Metabolism in LUAD by downloading and sorting the LUAD expression profile data and clinical information of patients from The Cancer Genome Atlas (TCGA) database. Our objective is to explore their correlation with diagnosis and prognosis and ultimately identify key genes such as PLK1, RRM2, TRIP13, and HHMR. Polo-like kinase (PLK) belongs to the serine/threonine protein kinase polo family and includes five subtypes, namely Polo-like kinase 1–5. PLK1, a known cell cycle regulator, governs cell mitosis, cytokinesis, DNA damage response, and development^[10]. It is often associated with lower survival rates and frequently observed in various human cancers due to its overexpression.

Data Acquisitions

The expression matrix of the Lung adenocarcinoma (LUAD) dataset (TCGA-LUAD) was downloaded from The Cancer Genome Atlas (TCGA) through the R package TCGAbiolinks^[11]. This dataset comprised 539 lung adenocarcinoma samples (cancer group, designated as LUAD) and 59 adjacent samples (normal group, designated as normal). All samples included in this study had matched clinical information and were standardized into Fragments Per Kilobase per Million (FPKM) data format. The corresponding clinical data were obtained from the UCSC Xena database^[12](http://genome.ucsc.edu). The Tpm data and count sequencing data of the TCGA-LUAD dataset were standardized using the R-package limma^[13].

Additionally, we acquired LUAD-related datasets GSE118370, which included small cell lung cancer (SCLC), and non-small cell lung cancer (NSCLC) GSE40275 from the GEO database^[15] using the R-package GEOquery^[14]. The GSE118370 dataset consisted of microarray gene-expression profiles of 6 lung adenocarcinoma patients and 6 matched normal tissues adjacent to cancer, derived from Homo Sapiens and the data platform GPL570 [Hg-U133 _ Plus _ 2] Affymetrix Human Genome U133 Plus 2.0 Array. On the other hand, the GSE40275 dataset comprised microarray gene expression profile data of 12 lung adenocarcinoma patients and 12 perfectly matched normal tissue samples adjacent to cancer, also from Homo Sapiens, with the data platform GPL15974 Human Exon 1.0 ST Array [CDF: Brain Array Version 9.0.1, HSEX 10stv2 _ hs _ refseq].

GeneCards database^[16] was utilized to gather Immune-related genes (IRGs) and Amino Acid Metabolism-related genes (AAMGs) using "immune" and "amino acid metabolism" as search keywords, respectively. Additionally, the MSigDB (Molecular Signatures Database) ^[17] database was employed to conduct searches with keywords "Immune" and "amino acid metabolisms," resulting in the download of all relevant genomes. These datasets were merged, ultimately providing 3531 genes related to both immune and amino acid metabolisms (I&AA MGS).

Analysis of Differentially Expressed Genes(DEGs) Related to Immunity and Acid Metabolism Between the Normal and LUAD Groups

For data analysis, the R language was utilized with preinstalled packages such as GEOquery and limma. The datasets were downloaded from GEO, GeneCards, and MSigDB databases. To explore the potential mechanisms, related biological characteristics, and pathways of differential genes in the cancer group and normal group of LUAD, we first standardized the TCGA-LUAD, GSE118370, and GSE40275 datasets using the R-package limma. Subsequently, the TCGA-LUAD dataset was divided into the cancer group and normal group to analyze the differences in the processed expression profile data. Differentially expressed genes (DEGs) between different groups of the TCGA-LUAD dataset were identified, and the results of the differential analysis were visually represented using a volcano plot created with the ggplot2 R package. We selected the differentially expressed genes with |logFC| > 2 and P.adj < 0.05. The obtained differentially expressed genes were then merged with I&AAMGs (genes related to immune and amino acid metabolism) and visualized using a Venn diagram to identify iaamrdegs (immune and amino acid metabolism related differentially expressed genes) in LUAD. The expression levels of these genes in the TCGA-LUAD dataset were visualized using the R-package pheatmap.

Protein-protein interaction (PPI)

Protein-protein interaction (PPI) involves the association of individual proteins through their interactions. To explore PPI networks, we utilized the STRING database ^[18], which enables the search for known protein interactions and the prediction of interactions between proteins. In this study, we constructed a PPI network based on differentially expressed genes with a minimum required interaction score of 0.400. The closely related local regions within the PPI network may represent molecular complexes with specific biological functions. We employed the Markov Clustering (MCL) algorithm, a graph-based clustering method, to initially cluster the PPI network, setting the expansion parameter to 3. Subsequently, we used the MCC (Maximal Clique Centrality) algorithm to determine the importance of each node based on the maximum clique theory. Additionally, we calculated the importance of each node using the Degree algorithm, which is based on the number of nodes adjacent to each node. The EPC (Edge Percolated Component) algorithm identified the maximum connected subgraph in the network by removing edges, thereby determining the importance of each node. Moreover, the MNC (Maximum Neighborhood Component) algorithm measured the importance of each node based on the maximum number of neighborhoods to which each node belongs. Lastly, the Dnnc (density of maximum network neighborhood connectivity) algorithm assessed the importance of each node by calculating the maximum neighborhood density to which each node belongs. Using these five algorithms, we assigned scores to IAAAMRDEGs related to other nodes in the PPI network. We ranked the IAAAMRDEGs based on these scores and selected the top 30 genes as hub genes (mRNAs) associated with LUAD disease.

Gene Ontology (GO) Enrichment Analysis

For large-scale function enrichment research, we conducted Gene Ontology (GO) ^[19] analysis, which encompasses biological process (BP), molecular function (MF), and cellular component (CC). Additionally, we utilized the Kyoto Encyclopedia of Genes and Genomes (KEGG) ^[20], a comprehensive database containing information on genomes, biological pathways, diseases, and drugs. To perform GO annotation analysis on IAAAMRDEGs, we employed the R-package clusterProfiler^[21]. The selection criteria for entries were set at P < 0.05 and FDR value (q.value) < 0.2. We applied the Benjamini-Hochberg (BH) method to correct the P values.

Gene Set Enrichment Analysis (GSEA)

Gene Set Enrichment Analysis (GSEA) ^[22] evaluates the distribution trend of genes in a predefined gene set within a gene list sorted by phenotypic correlation to determine its contribution to the phenotype. In our study, we utilized the R-package clusterProfiler to perform GSEA on all genes in the cancer group and normal group of the TCGA-LUAD dataset. The GSEA enrichment analysis parameters were set as follows: 123 seeds, 1000 permutations, a minimum of 10 genes per gene set, and a maximum of 500 genes. The Benjamini-Hochberg (BH) method was used to correct the P values. We utilized the "c2.cp.v7.2.symbols" gene set from the Molecular Signatures Database (MSigDB) and considered significant enrichment at a threshold of P < 0.05 and FDR value (q.value) < 0.25.

Gene Set Variation Analysis (GSVA)

Gene Set Variation Analysis (GSVA) ^[23] is a nonparametric unsupervised analysis method used to evaluate gene set enrichment results of microarray nuclear transcriptomes. It transforms the expression matrix of genes between different samples into the expression matrix of gene sets between samples, thus determining whether different pathways are enriched among different samples. For this analysis, we obtained the "h.all.v7.4.symbols.gmt" gene set from the MSigDB database. GSVA analysis was performed on the TCGA-LUAD dataset at the gene expression level, and the differences in functional enrichment between the two groups were calculated. Significant enrichment was determined at a threshold of P < 0.05.

Least Absolute Shrinkage and Selection Operator (LASSO)

To develop the prognosis model of hub genes for LUAD, we employed 10-fold cross-validation with 123 seeds for regression using the Least Absolute Shrinkage and Selection Operator (LASSO) method ^[24]. We ran 1000 cycles to prevent overfitting and improve the model's generalization ability. LASSO regression is often used for constructing prognosis models. By increasing the penalty term (the absolute value of lambda × slope) on top of linear regression, the model's overfitting is reduced and its generalization ability is enhanced. The risk score was determined through the LASSO regression prognosis model, and the grouping and survival outcomes of each cancer group sample were visualized on the risk factor map.

Cox regression analysis

To investigate the clinical prognostic value of the screened hub genes in LUAD, we included the expression of IAAAMRDEGs from the TCGA-LUAD dataset in a multivariate Cox regression analysis. The significance threshold for the P value was set at 0.1. Based on the results of the multivariate Cox regression analysis, we constructed a forest plot to visualize the outcomes. Furthermore, a nomogram was established to predict the 1-year, 3-year, and 5-year survival of LUAD patients. A nomogram is a graphical representation that depicts the functional relationship between multiple independent variables through intersecting line segments in a plane rectangular coordinate system. It scores and characterizes the variables in the multivariate regression model using a specific scale and calculates the total score to predict the probability of events. The accuracy and resolution of the nomogram were evaluated using calibration curves. Calibration curves assess the prediction performance of the model by plotting the fit of actual probabilities against the probabilities predicted by the model under different circumstances. The "rms" R package was employed to construct the nomograms and calibration curves. Additionally, the decision curve analysis (DCA) was used to evaluate the prediction effect of 1-year, 3-year, and 5-year survival outcomes in LUAD patients. DCA is a straightforward method for evaluating clinical prediction models, diagnostic tests, and molecular markers. The R-package ggDCA^[25] was utilized to draw the DCA curves.

Statistical Analyses

The specific technical roadmap of our study was shown in the figure (Fig. 1). All data processing and analyses were conducted using R software (Version 4.1.2). Continuous variables were presented as mean ± standard deviation. The Wilcoxon rank-sum test was used for comparing continuous variables between two groups, while the statistical significance of normally distributed variables was estimated using the independent Student t-test. The Kruskal-Wallis test was applied for comparing three or more groups. For categorical variables, the Chi-square test or Fisher's exact test was used to analyze the statistical significance between two groups. LASSO regression analysis was performed using the R-package glmnet^[26], and receiver operating characteristic (ROC) curves were generated using the R-package pROC^[27]. Unless otherwise specified, all results were calculated using Spearman correlation analysis, and P values were two-tailed, with statistical significance set at P < 0.05.

Analysis of Differentially Expressed Genes(DEGs) Related to Immunity-Acid Metabolism Between the Normal and LUAD Groups

From the TCGA-LUAD dataset, a total of 18,436 DEGs were identified, with 1,666 genes highly expressed and 564 genes lowly expressed (Fig. 2A). We further cross-referenced the up-regulated DEGs with I&AAMGs, resulting in 377 genes that were related to both immune and amino acid metabolism (IAAAMRDEGs) (Table 1). The expression patterns of these 377 IAAAMRDEGs in the TCGA-LUAD dataset were depicted in a Venn diagram (Fig. 2B) and represented as a heat diagram (Fig. 2C).

PPI Interaction Network Analysis of Immune-amino Acid Metabolism Related Genes

To analyze the interactions between proteins encoded by IAAAMRDEGs and predict the PPI network, we constructed a network using the STRING database with an interaction threshold set at 0.400. A total of 31 closely related IAAAMRDEGs were selected. We utilized five algorithms, namely MCC, DMNC, MNC, EPC, and Degree, to mine scores related to other PPI network nodes for the IAAAMRDEGs. The top 30 node genes scored by each algorithm were identified as hub genes (mRNAs) associated with LUAD disease and visualized using Cytoscape (Fig. 3).

GO/KEGG enrichment analysis of genes related to immune-amino acid metabolism

For functional enrichment analysis, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses of the 17 hub genes using the "ClusterProfiler" package^[21]. After converting gene IDs, we analyzed the hub genes through GO and KEGG (Table 2). The GO annotations of the 17 hub genes included Cellular Component (CC), Biological Process (BP), and Molecular Function (MF), providing insights into the functional enrichment of DEGs. KEGG analysis revealed associations between the 17 hub genes and signaling pathways. Specifically, the 17 hub genes were mainly linked to Progesterone-mediated oocyte maturation (hsa04914), Cell cycle (hsa04110), and Oocyte meiosis (hsa04114) (Fig. 4A, B). The results of GOKEGG enrichment analysis were visualized using a chord diagram, combining logFC with circle diagrams (Fig. 4C, D). The circle diagram highlighted that the mitotic cell cycle phase transition (GO: 0044772) showed significantly up-regulated expression, and Progesterone-mediated oocyte maturation (HSA04914) was the KEGG pathway with significantly up-regulated expression.

GSEA of immune-amino acid metabolism related genes in high and low risk groups

To determine the influence of gene expression levels on the difference between the LUAD cancer group and normal group, we performed GSEA enrichment analysis on all genes expressed in the TCGA-LUAD dataset. The criteria for significant enrichment were set at P < 0.05 and FDR value (q.value) < 0.25. The results revealed significant enrichment in several biological processes, cell components, and molecular functions. Specifically, the genes in the TCGA-LUAD dataset were significantly enriched in pre-notch expression and processing (Fig. 5B), signaling by NOTCH (Fig. 5C), TCF dependent signaling in response to Wnt (Fig. 5D), 4249 hedgehog signaling pathway (Fig. 5E), and other pathways (Table 3). Mountain and path maps were generated to visualize these enrichments in the TCGA-LUAD dataset (Fig. 5A, Fig. 5B-E).

GSVA of immune-amino acid metabolism related genes in high and low risk groups

To explore the differences in hallmark gene sets between the LUAD cancer group (group: Tumor) and the Normal group (group: normal), we analyzed the expression of all genes in the TCGA-LUAD dataset using GSVA. The GSVA analysis showed significant differences in 46 hallmark gene sets between the LUAD cancer group and the normal group (P < 0.05, Fig. 6A, Table 4). Among these, we selected 45 hallmark gene sets that demonstrated significant differences (P < 0.001) or played important roles in tumor occurrence and development. A group comparison chart was created (Fig. 6B) to visually display the results. The genes in the TCGA-LUAD dataset were notably enriched in hallmark pathways such as adipogenesis, allograft rejection, androgen response, angiogenesis, and apical junction, among others.

Screening Prognostic Related Genes by Lasso Regression Analysis

To assess the prognostic value of the 17 Hub genes in the TCGA-LUAD dataset, we performed LASSO regression analysis to construct a prognostic model (Fig. 7A-B), which included 4 Hub genes. Excluding normal samples, the cancer group was divided into a high-risk group (grouping: High Risk) and a low-risk group (grouping: Low Risk) based on the median of the Risk Score obtained from the LASSO model. The high-risk grouping results of the LASSO regression were visualized and presented in a risk factor map (Fig. 7C). To validate the prognosis model of LASSO, we conducted a statistical analysis of the clinical information of LUAD patients obtained from the TCGA-LUAD dataset (Table 5). We then merged the GSE118370 and GSE40275 datasets to create a Combined Dataset for verification. Using the variable coefficients from the LASSO model, we constructed a Risk Score for the cancer group samples in the combined dataset. Similar to the TCGA-LUAD dataset, the cancer group in the combined dataset was divided into a high-risk group (grouping: High Risk) and a low-risk group (grouping: Low Risk). The expression levels of the four Hub genes in both datasets were grouped according to high and low risks, and a comparison chart was generated (Fig. 7D-E) to show the statistical significance of the expression level differences (P < 0.05 is considered statistically significant). From the chart, we can observe consistent results for four genes (PLK1, RRM2, TRIP13, and HHMR) in the merged dataset compared to the TCGA-LUAD dataset.

Correlation between the expression of key genes and the occurrence of LUAD

To investigate the correlation between the expression of the four key genes (PLK1, RRM2, TRIP13, and HHMR) and the occurrence of LUAD, we plotted ROC curves for these genes in the TCGA-LUAD dataset, using clinical status (Tumor vs Normal) as the outcome variable (Fig. 8A-D). The AUC values for PLK1, RRM2, TRIP13, and HHMR were obtained from the ROC curves. Additionally, we examined the clinical correlations of these four Hub genes in subgroups based on clinical T-stage, N-stage, M-stage, and overall survival (OS) (Fig. 8E-T). The results indicated that the expression of PLK1 was associated with the differences between T1 and T2 in clinical T-stage (P < 0.0001), N0 and N1 in N stage (P < 0.05), M0 and M1 in clinical M stage (P < 0.01), and the difference between survival and death in OS events (P < 0.001). The expression of RRM2 was associated with the differences between T1 and T2 in clinical T-stage (P < 0.001), N0 and N1 in N stage (P < 0.001), M0 and M1 in clinical M stage (P < 0.05), and the difference between survival and death in OS events (P < 0.001). The expression of TRIP13 was associated with the differences between T1 and T2 in clinical T-stage (P < 0.001), N0 and N1 in N stage (P < 0.001), and the difference between survival and death in OS events (P < 0.001). The expression of HHMR was associated with the differences between T1 and T2 in clinical T stage (P < 0.001), N0 and N1 in N stage (P < 0.001), M0 and M1 in clinical M stage (P < 0.05), and the difference between survival and death in OS events (P < 0.001).

Multivariate Cox regression analysis of clinical correlation between expression of key genes and prognosis

To further validate the prognostic model established by LASSO regression, we conducted multivariate Cox regression analysis to analyze the correlation between the expression of the four key genes (PLK1, RRM2, TRIP13, HHMR) and prognosis. The multivariate Cox regression model was constructed (Table 6), and a forest plot was generated to display the results (Fig. 9A). Then we performed nomogram analysis to assess the predictive power of the multivariate Cox regression model and created a nomogram (Fig. 9B). Additionally, we calibrated the nomogram for 1-year (Fig. 9C), 3-year (Fig. 9D), and 5-year (Fig. 9E) survival predictions and drew the calibration curves (Fig. 9C-E). From the calibration curves, it can be observed that the blue line representing three years is closest to the gray ideal line, indicating that the prediction effect of the model in the third year is better than that in the first year and fifth year. Subsequently, we used decision curve analysis (DCA) to evaluate the clinical effectiveness of the constructed LASSO-Cox regression prognosis model for one year (Fig. 9F), three years (Fig. 9G), and five years (Fig. 9H). The results are shown in Figs. 9E-H, where the X-axis range of the blue line representing the model is higher than that of the all-positive red line and all-negative gray line, suggesting that the model performs better in five years but less so in one year and three years.

PLK1 has a higher expression level in LUAD tumor tissues

Using the key gene PLK1 from the multivariate Cox model for further analysis, we analyzed the expression of PLK1 in LUAD tumor tissue and normal lung tissue by immunohistochemistry in the HPA database, using the HPA053229 antibody. The results revealed that the expression level of PLK1 in LUAD tumor tissue (Fig. 10B) was higher compared to normal lung tissue (Fig. 10A).

Lung adenocarcinoma (LUAD) represents a common form of lung cancer, accounting for approximately 40% of all lung cancer cases^[28]. Despite advancements in early detection and treatment options, the survival rate for LUAD remains low, underscoring the need for further research into the molecular mechanisms underpinning this disease. A key area of interest is the potential contribution of alterations in amino acid metabolism and immune signaling pathways, particularly involving the protein PLK1, to LUAD pathogenesis. Exploring differentially expressed genes associated with immune-amino acid metabolism in LUAD could enhance early-stage prediction, facilitate the evaluation of LUAD patient conditions, and aid in determining their prognosis, thus offering valuable insights for subsequent research and clinical practice.

In this study, we initially identified differentially expressed genes between LUAD samples and normal lung tissue, specifically screening for genes associated with immune-amino acid metabolism. Subsequently, we elucidated the potential mechanisms of action and relevant biological characteristics and pathways in both the cancer and normal LUAD groups through the utilization of GO/KEGG, GSEA, and GSVA analyses. Following this comprehensive exploration, we employed LASSO-COX analysis, risk scoring, and prognostic modeling to identify PLK1, RRM2, TRIP13, and HHMR as genes significantly associated with LUAD and immune-amino acid metabolism. Additionally, we performed immunohistochemical analysis of PLK1 expression in LUAD tumor tissue and normal lung tissue using the HPA database, revealing higher expression levels of PLK1 in LUAD compared to normal tissues. This finding aligns with previous studies that have also reported elevated PLK1 protein expression in various cancer types, such as osteosarcoma^[29], colon cancer^[30–31], and lung squamous cell carcinoma^[32].

Our analysis confirms the significant overexpression of PLK1 in LUAD in TCGA data, suggesting its potential as a diagnostic marker. Specifically, in LUAD, the expression of PLK1 serves as a promising diagnostic marker, with an impressive AUC exceeding 0.9. Moreover, our findings indicate a correlation between PLK1 and the clinical stage of LUAD, particularly the T stage, providing further support for the notion that PLK1 expression might be associated with the degree of malignancy in LUAD. PLK1, a serine/threonine kinase, plays a critical role in cell cycle progression and mitosis. Its frequent overexpression in cancer cells is associated with enhanced cell proliferation, cell cycle arrest, and resistance to apoptosis^[33]. Consistent with previous research, our study aligns with the evidence linking high PLK1 expression to poor prognosis in LUAD^[34–35]. Recent studies suggest that PLK1 may also play a role in immune signaling pathways in cancer cells. For instance, it has been shown that PLK1-mediated phosphorylation of vimentin accelerates the transfer of cytotoxic T cells and immune escape in LUAD^[36].

Moving on to ribonucleotide reductase subunit M2 (RRM2), this enzyme holds crucial importance in DNA synthesis and repair and has been associated with various cancers, including lung adenocarcinoma. In lung adenocarcinoma cells, RRM2 exhibits upregulation, and its expression levels have been linked to poor prognosis^[37]. One potential mechanism through which RRM2 contributes to the development and progression of lung adenocarcinoma involves its impact on immune system function. Multiple studies suggest that RRM2 has been found to affect lung cancer progression and tumor immune cell infiltration. For example, RRM2 inhibition has been found to be effective in promoting M1 macrophage polarization and inhibiting M2 macrophage polarization in vitro and in vivo^[38]. Additionally, in bladder cancer, RRM2 is positively correlated with immune checkpoints and cytotoxic T lymphocytes^[39].

Thyroid Hormone Receptor Interactor 13 (TRIP13) encodes an AAA + ATPase that plays crucial roles in various cellular processes, including meiotic spindle assembly and DNA repair. Recent investigations have revealed frequent amplification and overexpression of TRIP13 in diverse cancer types, including LUAD. Wei Li et al. (2016) ^[40] observed upregulation of TRIP13 in LUAD tissues compared to adjacent normal tissues, and its expression was associated with poor prognosis in LUAD patients. Furthermore, Mechanistically, TRIP13 was shown to regulate the expression of multiple genes involved in cell cycle progression and apoptosis, the knockdown of TRIP13 can arrest lung cancer cells in G2/M phase and regulate the expression levels of genes related to cell cycle checkpoints^[41].

Hyaluronan Mediated Motility Receptor (HMMR) is a gene encoding a protein that plays a pivotal role in tumor immune evasion and amino acid metabolism. The expression level of the HMMR gene in LUAD is closely associated with tumor occurrence and prognosis. Studies have found that high HMMR gene expression is linked to poor prognosis in LUAD patients and increased tumor recurrence^[42]. A large number of studies have confirmed that HMMR is overexpressed in non-small cell lung cancer^[43]. Abnormal tumor microenvironment is closely related to the occurrence of tumors, and immune cells are an important part of tumor microenvironment. Studies have identified a connection between high HMMR gene expression and tumor immune in liver cancer. Specifically, HMMR induces a macrophage-related immune response likely by activating subgroup M2, and its high expression predicts increased infiltration levels of B cells, CD8 + T cells, CD4 + T cells, macrophages, neutrophils, and dendritic cells^[44].

The results of GO/KEGG analysis revealed that the mitotic cell cycle phase transition was a significantly up-regulated biological process in lung adenocarcinoma, which is consistent with previous studies. Cell division processes play a vital role in cancer development. Error-free chromosomal segregation and cell proliferation during mitosis are central events in the life cycle. Proto-onco genes and tumor suppressor genes are directly or indirectly involved in the regulation of cell cycle, or it may be the result of genetic damage of genes encoding cyclins^[45]. Additionally, the development of lung adenocarcinoma is also associated with the inactivation of tumor suppressor genes like p53 and Rb, as well as mutations in oncogenes such as EGFR and KRAS, all of which are closely related to cell cycle regulation^[46–47]. As we all know, steroid hormone receptors, including progesterone receptors, play an important role in the development of hormone-targeted tissue cancers, such as breast cancer and endometrial cancer^[48]. While there is currently no direct evidence linking LUAD to progesterone-mediated oocyte maturation, some studies have suggested that progesterone might play a role in the occurrence and progression of LUAD. Researchers have found that The combination of estrogen and progesterone in NSCLC cells synergistically promotes the expression of vascular endothelial growth factor (VEGF) by increasing the proliferation of endothelial cells from adjacent vessels^[49]. Moreover, other studies have shown that the high expression of the progesterone receptor in patients with LUAD is not related with malignancy and prognosis of lung adenocarcinoma^[50].

In our study, GSEA analysis revealed significantly enriched pathways in LUAD, including pre-notch expression and processing, signaling by NOTCH, TCF dependent signaling in response to WNT, and 4249 hedgehog signaling pathway. Notch, Wnt, and Hedgehog are all developmental signaling pathways. Previous research has reported that developmental signaling pathways, including Notch, WNT, and Hedgehog, are frequently altered in cancer stem cells (CSCs) to sustain the survival of CSCs and interact with other oncogenic signaling pathways such as MAPK, NF-κB, PI3K, and EGFR^[51]. Although CSCs constitute only a small subset, accounting for less than 1% of the tumor microenvironment, they possess the characteristics of self-renewal and contribute to tumor initiation, metastasis, spread, and resistance.

The GSVA enrichment analysis of the TCGA-LUAD dataset revealed significant enrichment in several hallmark pathways, including adipogenesis, allograft rejection, androgen response, angiogenesis, and apical junction pathways. As LUAD is a malignant tumor, its development is influenced by the tumor microenvironment, which includes factors such as fat cells, immune cells, and blood vessels. Adipogenesis, the process of fat formation in adipose tissue, has been shown to be related to the occurrence and progression of lung cancer^[52]. Angiogenesis is the formation of new blood vessels. It plays a critical role in the growth and metastasis of lung cancer and other malignancies. The density of blood vessels in normal lung tissue is lower than that in lung adenocarcinoma, suggesting that angiogenesis may accelerate the growth and spread of lung adenocarcinoma. Factors such as VEGF, which promote angiogenesis, have been implicated in promoting the growth and spread of lung cancer^[53]. As for androgen, modern research showed that androgen receptor exist in normal human lung, non-small cell, and small cell lung cancer tissue^[54]. In both male and female patients with lung cancer, androgen levels and their response may influence tumor growth and progression^[55]. Androgen response in hepatocarcinoma has attracted the attention of researchers, who have found that androgens can promote the growth and metastasis of hepatocellular carcinoma and are associated with the prognosis of the disease^[56]. The apical junction is a crucial structure that interconnects cells and maintains the integrity and stability of tissues. Abnormal expression of proteins involved in cell junctions in LUAD may lead to structural damage and increased proliferation of tumor cells. Understanding the regulation of epithelial cell junctions may have implications for preventing the metastasis of lung adenocarcinoma and improving therapeutic strategies.

LUAD is a complex disease characterized by dysregulation in various biological processes, including amino acid metabolism and immune signaling pathways. The prognosis model was constructed by LASSO, and the dataset TCGA-LUAD was verified with the combined datasets (GSE118370, GSE40275). The results showed that there were four identical genes, namely, PLK1, RRM2, TRIP13 and HHMR. We drew the ROC curves of four hub genes, and the AUCs of PLK1, RRM2, TRIP13 and HHMR were all greater than 0.9, indicating that the genes related to immunity and amino acid metabolism selected by us were accurate in predicting the presence of LUAD. In addition, multivariate Cox regression analysis revealed that the expression levels of four hub genes were correlated with the prognosis of LUAD. PLK1, RRM2, TRIP13 and HMMR as the pivotal protein, may serve as a crucial link between these different processes by regulating both amino acid metabolism and immune cell function. Further research is essential to comprehensively understand how PLK1 contributes to LUAD development, and to develop targeted therapies that can exploit these pathways for potential therapeutic advantages. Although this study has enhanced our understanding of the interplay between immunity, amino acid metabolism, and LUAD, it does have some limitations. One major limitation is the lack of in vitro and in vivo validation for our findings. Additionally, due to the design constraints of this study, there might be other important signaling pathways related to LUAD that were not explored, and further investigation is needed to uncover these potential pathways.

Ethical Approval: Not applicable.

Competing interests: All authors declare no competing interests.

Authors’ contributions: The authors confirm contribution to the paper as follows: Conceptualization, methodology: Y.Z. and Q.L.; data curation and analysis: Y.Z. and Y.W.; writing—original draft preparation: Y.Z. and R.Z. writing—review & editing: Y.Z., R.Z., Y.W. and Q.L. All authors have read and agreed to the published version of the manuscript.

Funding:

Special Fund for Basic Research Business Expenses of Central Higher Education Institutions Subsidies Key Research Projects of Beijing University of Chinese Medicine(2020-JYB-ZDGG-127)

Beijing University of Chinese Medicine started the fund project for new teachers(2023-JYB-XJSJJ050)

Availability of data and materials: The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

LungCancer2020:Epidemiology,Etiology,andPrevention.ClinChestMed.2020Mar;41(1):1-24.
Noone AM, Howlader N, Krapcho M, et al. SEER Cancer Statistics Review, 19752015, National Cancer Institute. Bethesda, MD, Available online: https://seer.cancer.gov/csr/1975_2015/, based on November 2017 SEER data submission, posted to the SEER website, April 2018.
HUTCHINSON B D,SHROFF G S,TRUONG M T,et al.Spectrum of lung adenocarcinoma[J].Semin Ultrasound CTMR,2019,40(3):255-264.
LUENGO A,GUI D Y,VANDER HEIDEN M G.Targeting metabolism for cancer therapy[J].Cell Chem Biol,2017,24(9):1161-1180.
FAUBERT B,SOLMONSON A,DEBERARDINIS R J.Metabolic reprogramming and cancer progression[J].Science,2020,368(6487):eaaw5473.
NAKAYA M,XIAO Y,ZHOU X,et al.Inflammatory T cell responses rely on amino acid transporter ASCT2 facilitation of glutamine uptake and m TORC1 kinase activation[J].Immunity,2014,40(5):692-705.
SISKA P J,RATHMELL J C.T cell metabolic fitness in antitumor immunity[J].Trends Immunol,2015,36(4):257-264.
Ananieva E. Targeting amino acid metabolism in cancer growth and anti-tumor immune response. World J Biol Chem. 2015;6(4):281-289. doi:10.4331/wjbc.v6.i4.281
Nakaya M, Xiao Y, Zhou X, et al. Inflammatory T cell responses rely on amino acid transporter ASCT2 facilitation of glutamine uptake and mTORC1 kinase activation. Immunity. 2014;40(5):692-705. doi:10.1016/j.immuni.2014.04.007
ILIAKI S,BEYAERT R,AFONINA I S.Polo-like kinase 1(PLK1) signaling in cancer and beyond[J].Biochem Pharmacol,2021,193:114747.
Goldman, M.J., et al., Visualizing and interpreting cancer genomics data via the Xena platform. Nature biotechnology, 2020. 38(6): p. 675-678.
Goldman, M.J., et al., Visualizing and interpreting cancer genomics data via the Xena platform. Nat Biotechnol, 2020. 38(6): p. 675-678.
Ritchie, M.E., et al., limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic acids research, 2015. 43(7): p. e47-e47.
Davis, S. and P.S. Meltzer, GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics, 2007. 23(14): p. 1846-7.
Barrett, T., et al., NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res, 2013. 41(Database issue): p. D991-5.
Stelzer, G., et al., The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr Protoc Bioinformatics, 2016. 54: p. 1.30.1-1.30.33.
Liberzon, A., et al., The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst, 2015. 1(6): p. 417-425.
Szklarczyk, D., et al., STRING v11: protein protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic acids research, 2019. 47(D1): p. D607-D613.
Gene Ontology, C., Gene Ontology Consortium: going forward. Nucleic Acids Res, 2015. 43(Database issue): p. D1049-56.
Kanehisa, M. and S.J.N.a.r. Goto, KEGG: kyoto encyclopedia of genes and genomes. 2000. 28(1): p. 27-30.
Yu, G., et al., clusterProfiler: an R package for comparing biological themes among gene clusters. Omics, 2012. 16(5): p. 284-7.
Subramanian, A., et al., Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A, 2005. 102(43): p. 15545-50.
Hänzelmann, S., R. Castelo, and J. Guinney, GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics, 2013. 14: p. 7.
Tibshirani, R.J.S.i.m., The lasso method for variable selection in the Cox model. 1997. 16(4): p. 385-395.
Tataranni, T. and C. Piccoli, Dichloroacetate (DCA) and Cancer: An Overview towards Clinical Applications. Oxid Med Cell Longev, 2019. 2019: p. 8201079.
Engebretsen, S. and J. Bohlin, Statistical predictions with glmnet. Clin Epigenetics, 2019. 11(1): p. 123.
Mandrekar, J.N., Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol, 2010. 5(9): p. 1315-6.
Sun, X., Mao, Y., Zang, X., Zhang, J., Tang, G., Guo, F., ... & Li, A. (2019). Optimization of comprehensive therapy for patients with mixed-type lung adenocarcinoma: a multicenter retrospective study. Translational lung cancer research, 8(1), 31-39.
Chou YS,Yen CC,Chen WM,et al.Cytotoxic mechanism of PLK1 inhibitor GSK461364 against osteosarcoma:Mitotic arrest,apoptosis,cellular senescence,and synergistic effect with paclitaxel[J].Int J Oncol,2016,48(3):1187-1194.
Dasgupta N,Thakur BK,Ta A,et al.Polo-like kinase 1 expression is suppressed by CCAAT/enhancer-binding protein alpha to mediate colon carcinoma cell differentiation and apoptosis[J].Biochim Biophys Acta Gen Subj,2017,1861(7):1777-1787.
Fernández-Aceňero MJ,Cortés D,Gómez del Pulgar T,et al.PLK-1 expression is associated with histopathological response to neoadjuvant therapy of hepatic metastasis of colorectal carcinoma[J].Pathol Oncol Res,2016,22(2):377-383.
Li H,Wang H,Sun Z,et al.The clinical and prognostic value of polo-like kinase 1 in lung squamous cell carcinoma patients:immunohistochemical analysis[J].Biosci Rep,2017[Epub ahead of print].
Strebhardt K. Multifaceted polo-like kinases: drug targets and antitargets for cancer therapy. Nat Rev Drug Discov. 2010;9(8):643-660. doi:10.1038/nrd3184
Wang ZX, Xue D, Liu ZL, et al. Overexpression of polo-like kinase 1 and its clinical significance in human non-small cell lung cancer. Int J Biochem Cell Biol. 2012;44(1):200-210. doi:10.1016/j.biocel.2011.10.017
Li H, Wang H, Sun Z, Guo Q, Shi H, Jia Y. The clinical and prognostic value of polo-like kinase 1 in lung squamous cell carcinoma patients: immunohistochemical analysis. Biosci Rep. 2017;37(4):BSR20170852. doi:10.1042/BSR20170852
Jang HR, Shin SB, Kim CH, et al. PLK1/vimentin signaling facilitates immune escape by recruiting Smad2/3 to PD-L1 promoter in metastatic lung adenocarcinoma [published correction appears in Cell Death Differ. 2021 Aug 17;:]. Cell Death Differ. 2021;28(9):2745-2764. doi:10.1038/s41418-021-00781-4
Ma C, Luo H, Cao J, Gao C, Fa X, Wang G. Independent prognostic implications of RRM2 in lung adenocarcinoma. J Cancer. 2020;11(23):7009-7022. Published 2020 Oct 17. doi:10.7150/jca.47895
Tang B, Xu W, Wang Y, et al. Identification of critical ferroptosis regulators in lung adenocarcinoma that RRM2 facilitates tumor immune infiltration by inhibiting ferroptotic death. Clin Immunol. 2021;232:108872. doi:10.1016/j.clim.2021.108872
Zhou Z, Song Q, Yang Y, Wang L, Wu Z. Comprehensive Landscape of RRM2 with Immune Infiltration in Pan-Cancer. Cancers (Basel). 2022;14(12):2938. Published 2022 Jun 14. doi:10.3390/cancers14122938
Li W, Zhang G, Li X, et al. Thyroid hormone receptor interactor 13 (TRIP13) overexpression associated with tumor progression and poor prognosis in lung adenocarcinoma. Biochem Biophys Res Commun. 2018;499(3):416-424. doi:10.1016/j.bbrc.2018.03.129
Zhang Q, Dong Y, Hao S, Tong Y, Luo Q, Aerxiding P. The oncogenic role of TRIP13 in regulating proliferation, invasion, and cell cycle checkpoint in NSCLC cells. Int J Clin Exp Pathol. 2019;12:3357–3366.
Ma X, Xie M, Xue Z, et al. HMMR associates with immune infiltrates and acts as a prognostic biomaker in lung adenocarcinoma. Comput Biol Med. 2022;151(Pt A):106213. doi:10.1016/j.compbiomed.2022.106213
He R, Zuo S. A robust 8-gene prognostic signature for early-stage non-small cell lung cancer. Front Oncol. 2019;9:693. doi: 10.3389/fonc.2019.00693.
Lei X, Zhang M, Guan B, Chen Q, Dong Z, Wang C. Identification of hub genes associated with prognosis, diagnosis, immune infiltration and therapeutic drug in liver cancer by integrated analysis. Hum Genomics. 2021;15(1):39. Published 2021 Jun 29. doi:10.1186/s40246-021-00341-4
Otto T, Sicinski P. Cell cycle proteins as promising targets in cancer therapy. Nat Rev Cancer. 2017;17(2):93-115. doi:10.1038/nrc.2016.138
Vincenzi B, Schiavon G, Silletta M, et al. Cell cycle alterations and lung cancer. Histol Histopathol. 2006;21(4):423-435. doi:10.14670/HH-21.423
Cancer Genome Atlas Research Network. Comprehensive molecular profiling of lung adenocarcinoma [published correction appears in Nature. 2014 Oct 9;514(7521):262. Rogers, K [corrected to Rodgers, K]] [published correction appears in Nature. 2018 Jul;559(7715):E12]. Nature. 2014;511(7511):543-550. doi:10.1038/nature13385
Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci U S A. 1996;93(12):5925-5930. doi:10.1073/pnas.93.12.5925
Marquez-Garban DC, Mah V, Alavi M, et al. Progesterone and estrogen receptor expression and activity in human non-small cell lung cancer. Steroids. 2011;76(9):910-920. doi:10.1016/j.steroids.2011.04.015
He Q, Zhang M, Zhang J, et al. Correlation between epidermal growth factor receptor mutations and nuclear expression of female hormone receptors in non-small cell lung cancer: a meta-analysis. J Thorac Dis. 2015;7(9):1588-1594. doi:10.3978/j.issn.2072-1439.2015.09.04
Clara JA, Monge C, Yang Y, Takebe N. Targeting signalling pathways and the immune microenvironment of cancer stem cells - a clinical update. Nat Rev Clin Oncol. 2020;17(4):204-232. doi:10.1038/s41571-019-0293-2
Hu Chan, Fan Yi, Xu Yuan, Hu Zhijian, Zeng Yiming. Research progress of lipid metabolism in the field of occurrence, development, diagnosis and treatment of lung cancer. Journal of Shanghai Jiaotong University (Medical Edition) [J], 2022,42 (12): 1766-1771 DOI: 10.3969/J.ISSN.
Apte RS, Chen DS, Ferrara N. VEGF in Signaling and Disease:Beyond Discovery and Development. Cell, 2019, 176(6):1248-1264.
Kaiser U, Hofmann J, Schilli M, et al. Steroid-hormone receptors in cell lines and tumor biopsies of human lung cancer. Int J Cancer. 1996;67(3):357-364. doi:10.1002/(SICI)1097-0215(19960729)67:3<357::AID-IJC9>3.0.CO;2-Q
Rades D, Setter C, Dahl O, Schild SE, Noack F. The prognostic impact of tumor cell expression of estrogen receptor-α, progesterone receptor, and androgen receptor in patients irradiated for nonsmall cell lung cancer. Cancer. 2012;118(1):157–163. doi: 10.1002/cncr.26282.
Ma WL, Jeng LB, Lai HC, Liao PY, Chang C. Androgen receptor enhances cell adhesion and decreases cell migration via modulating β1-integrin-AKT signaling in hepatocellular carcinoma cells.Cancer Lett. 2014;351(1):64-71. doi:10.1016/j.canlet.2014.05.017

Tables 1-6 is available in the Supplementary Files section.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Genes associate with Immunity and Amino Acid Metabolism in Lung adenocarcinoma (LUAD): A bioinformatic analysis

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials and Methods

Data Acquisitions

Protein-protein interaction (PPI)

Gene Ontology (GO) Enrichment Analysis

Gene Set Enrichment Analysis (GSEA)

Gene Set Variation Analysis (GSVA)

Least Absolute Shrinkage and Selection Operator (LASSO)

Cox regression analysis

Statistical Analyses

Results

Analysis of Differentially Expressed Genes(DEGs) Related to Immunity-Acid Metabolism Between the Normal and LUAD Groups

PPI Interaction Network Analysis of Immune-amino Acid Metabolism Related Genes

GO/KEGG enrichment analysis of genes related to immune-amino acid metabolism

GSEA of immune-amino acid metabolism related genes in high and low risk groups

GSVA of immune-amino acid metabolism related genes in high and low risk groups

Screening Prognostic Related Genes by Lasso Regression Analysis

Correlation between the expression of key genes and the occurrence of LUAD

Multivariate Cox regression analysis of clinical correlation between expression of key genes and prognosis

PLK1 has a higher expression level in LUAD tumor tissues

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1