Development of Amino Acid Metabolism-Related Prognostic Model and Immune Infiltration Analysis in Patients with Stomach Adenocarcinoma

doi:10.21203/rs.3.rs-2754183/v1

Download PDF

Research Article

Development of Amino Acid Metabolism-Related Prognostic Model and Immune Infiltration Analysis in Patients with Stomach Adenocarcinoma

https://doi.org/10.21203/rs.3.rs-2754183/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Stomach adenocarcinoma (STAD) is a major contributor to cancer mortality worldwide. Alterations in amino acid metabolism have been reported in various tumors. However, the prognostic value of amino acid metabolism-related genes in STAD deserves to be further elucidated. In this study, we constructed a prognostic risk model consisting of 3 amino acid metabolism-related genes (SERPINE1, NRP1, MATN3) in STAD. Based on the median risk score, STAD patients were divided into high- and low-risk groups. The patients with high-risk scores had a worse prognosis. A nomogram consisting of risk score and various clinical characteristics accurately predicted the 1-, 3-, and 5-year survival time of STAD patients. Notably, KEGG pathway enrichment analysis indicated immune-related pathways enriched in the high-risk group. High-risk scores were significantly related to C6 (TGF-β dominant type), while low-risk scores were significantly related to C4 (lymphocyte-depleted type). The higher risk score was associated with higher immune infiltration, immune-related function, lower tumor purity and worse response to immunotherapy. In addition, the model genes were correlated with antitumor drug sensitivity. Finally, functional assays confirmed that interference of model gene MATN3 inhibited the proliferation and migration of STAD cells. In conclusion, the amino acid metabolism-related prognostic model might be used as a biomarker to predict the prognosis and guide immunotherapy for STAD patients.

Bioinformatics

Amino acid metabolism

Stomach adenocarcinoma

Prognosis

Immunity

Drug sensitivity

MATN3

According to the Global Cancer Statistics 2020, the incidence of gastric cancer ranked fifth and mortality ranked third worldwide, with over 1 million new cases and about 769,000 deaths in 2020^[1]. Despite the diagnosis and treatment of gastric cancer having made great progress in recent decades, a lot of patients still failed to be diagnosed until advanced stages, with a 5-year survival rate of around 30%, while it reaches 60% in the early-to-middle stage patients ^{[2, 3]}, hence it is an urgent need for sensitive biomarkers for early diagnosis and prognostic prediction.

Amino acids, the basic units of proteins, are involved in energy production, hormones and neurotransmitters generation, and methylation^[4]. In the late 1930s, Wargurg first discovered the phenomenon of glucose metabolism reprogramming in tumor cells, namely the Warburg effect, which opened up a new direction for the study of tumor development from the perspective of metabolism^[5]. The in-depth study of tumor metabolism found that tumor cells usually reprogram substance and energy metabolism to meet the energy demand of proliferation^[6]. In addition to glucose metabolism providing energy for tumor cells proliferation, amino acid metabolism reprogramming also plays a crucial role in tumor growth. Amino acid metabolic reprogramming refers to the abnormal process of amino acid uptake rate, metabolites, metabolic key enzymes or amino acid metabolic pathway in tumor cells^[7]. In recent years, more and more studies have focused on the amino acid metabolism of tumors, including stomach cancer, colon cancer, hepatocellular carcinoma, kidney clear cell carcinoma, breast cancer, ovarian cancer, etc^[7–12]. Amino acids promote cancer cells survival and proliferation under oxidative, genotoxic, and nutritional stress^[13]. Therefore, targeting amino acid metabolism is emerging as a potential therapeutic measure for cancer patients.

In stomach cancer, various amino acids metabolism is abnormal. The levels of L-glutamine, L-isoleucine and serine were elevated in gastric cancer tissue compared to non-malignant tissue, and higher levels of L-cysteine and L-tyrosine were detected in the invasive (T3 and T4) gastric cancers than non-invasive (T1 and T2) cancers^[14]. The activation of kynurenine pathway in tryptophan metabolism was reported in gastric cancer and kynurenic acid in gastric juice and serum might serve as a biomarker for gastric cancer^[15]. Aromatic amino acids in gastric juice could serve as potential diagnostic biomarkers to monitor gastric malignancies^[16]. Proline and serine metabolisms played an important role in gastric cancer metastasis^[17]. Significantly higher levels of amino acids (e.g., alanine, glycolate, arginine, methionine, phenylalanine, glycine, and tyrosine) were excreted in GC patients urine and the levels of alanine, phenylacetylglycine, arginine and glycine increased significantly with T stage^[18].

Changes in amino acid metabolism in gastric cancer have been supported by many studies, but the mechanism remains unclear. Alanine-serine,-cysteine transporter 2 (SLC1A5; ASCT2), the main glutamine transporter, determines intracellular glutamine level^[19]. Knockdown of SLC1A5 in gastric cancer cells suppressed cell proliferation, migration and invasion partly by inactivated mTOR/p-70S6K1 signaling pathway and also inhibited the relative volume of xenografted tumor in vivo^[20]. Cai et al. also found lactate dehydrogenase A (LDH-A) downregulation and pyruvate dehydrogenase B (PDH-B) overexpression could inhibit gastric cancer cell growth and migration by forcing pyruvate into the Krebs cycle instead of the glycolysis process^[21].

Furthermore, amino acid metabolism has also been reported to play an important role in the regulation of tumor immunity^[22–25]. Immune cells have specific requirements for amino acid. The activation of CD8 + T cells in the presence of interleukin (IL)-2 resulted in increased cell surface densities of Slc1a5 and Slc7a5^[26]. Glutamine is necessary for T cell activation, as T cells cannot proliferate and produce IL-2 or IFN-γ when cultured without glutamine^[27]. Myc-dependent natural killer cell activation induced by IL-2/IL-12 requires Slc7a5-mediated glutamine uptake^[28]. Lipopolysaccharide supports cytokine production in proinflammatory macrophages by increasing Slc7a5 expression and leucine uptake^[29]. Gls1 deficiency impairs the differentiation of Th17 cell through limiting αKG supply^[30]. Thus, targeting amino acid metabolism might be helpful to guide the immunotherapy and improve existing therapeutic methods.

Based on the importance of amino acid metabolism in tumor development and immunity, in this paper, we focused on the role of amino acid metabolism-related genes in STAD and its involvement in tumor immunity to provide potential biomarkers for prognostic prediction, guide for immunotherapy and develop new therapeutic targets for further breakthroughs in the field of amino acid metabolism in STAD.

2.1. Data Collection

We downloaded transcriptomic gene expression data and corresponding clinical information of 380 patients with STAD (Table 1) from the TCGA database (https://portal.gdc.cancer.gov/)^[31]. The data were from STAD patients after primary surgical resection (patients with missing clinical data were excluded). The gene expression data (GSE84433 series matrix) and corresponding clinical information of 357 patients with STAD from the GEO database (https://www.ncbi.nlm.nih.gov) were employed to perform model validation^[32]. The samples in this study were initial STAD after primary surgical resection (samples with missing clinical information were excluded). The flowchart of the data analysis was shown in Fig. 1.

Table 1

Clinical characteristics of STAD patients in the TCGA database.
Characteristics		Total	%
All		380	100.00
Age (y)	≥ 65	218	57.37
	< 65	162	42.63
Gender	Male	239	62.89
	Female	141	37.11
Grade	G1	8	2.11
	G2	129	33.95
	G3	243	63.95
Stage	I	50	13.16
	II	120	31.58
	III	169	44.47
	IV	41	10.79
T stage	T1	18	4.74
	T2	77	20.26
	T3	180	47.37
	T4	105	27.63
M stage	M0	353	92.89
	M1	27	7.11
N stage	N0	122	32.11
	N1	100	26.32
	N2	79	20.79
	N3	79	20.79

Table 2. Multivariate Cox regression analysis results of model genes.
Gene	Coef	HR	HR.95L	HR.95H	P value
SERPINE1	0.0224577911629626	1.199456	1.094436	1.314553	0.0001
NRP1	0.0463621851393052	1.970686	1.412392	2.749666	6.56E-05
MATN3	0.0531597548577127	1.883591	1.396202	2.541119	3.40E-05

2.2. Acquisition of Differentially Expressed Amino Acid Metabolism Genes (DEAAMGs)

Amino acid metabolism-related genes were extracted from GSEA (http://www.gsea-msigdb.org/gsea/index.jsp)^[33]. To obtain differentially expressed amino acid metabolism-related genes, we employed the “bioconductor limma” R package to analyze the STAD tumor tissues and normal gastric tissues. The fold change > 0.5 and false discovery rate < 0.05 were considered as the criteria for DEAAMGs.

2.3. Construction and Validation of Amino Acid Metabolism Genes-Related Prognostic Model

We used the LASSO regression analysis method to construct the best prognostic model^[34]. Univariate cox regression analysis was used to screen for DEAAMGs related to OS through the “survival” R package. P < 0.05 was a criterion for the prognosis-related gene. Then, we used the LASSO algorithm to avoid overfitting the model and screen out the optimal prognostic genes for constructing the amino acid metabolism-related prognostic model. The risk score in the model is the sum of the expression of each prognosis-related gene, multiplied by the corresponding regression coefficient. All STAD patients in TCGA were divided into high- and low-risk groups according to the median risk score. Next, we performed PCA and t-SNE analysis to investigate the distribution of patients in the high- and low-risk groups using “ggplot2” and “Rtsne” R packages. KM analysis was used to analyze the OS between the high- and low-risk groups. Furthermore, a nomogram comprised age, gender, grade, stage and risk score was plotted using “regplot” and “rms” R packages. The calibration curve was used to detect the predictive accuracy of the nomogram. The relationship of risk score with tumor grade and stage was analyzed using Spearman correlation analysis.

2.4. Functional Enrichment Analysis

To further investigate the potential mechanism of model genes affecting the prognosis of STAD patients, we performed Kyoto Gene and Genome Encyclopedia (KEGG) pathway enrichment analysis by GSEA 4.1.0 software^[33]. Firstly, we downloaded “c2.cp.kegg.v7.1.symbols.gmt” set from Molecular Signatures Database. Then, we used “GSEA” software to analyze the enriched pathways in the high- and low-risk groups. The “ggplot2” R package was applied to visualize the top 5 significantly enriched biological pathways in the high- and low-risk groups. The pathways at p < 0.05 were considered statistically significant.

2.5. Immune Infiltration Analysis

We applied the “GSVA” R package for single sample genomic enrichment analysis (ssGSEA) to analyze the immune cell infiltration and immune-related function^[35]. A box plot was drawn to visualize the differences in the immune infiltration subtypes between high- and low-risk groups. 2-way ANOVA analysis was used to test the relationship between risk score and immune infiltration subtypes. In addition, we used Spearman correlation analysis to determine the correlation of risk scores with immune cell scores, stromal cell scores, ESTIMATE scores and tumor purity ^[36].

2.6. Immunotherapy Response Evaluation and Model Comparison

To further investigate the efficacy of our prognostic risk model for assessing immunotherapy response, we analyzed immunotherapy biomarkers based on the TCGA data. We uploaded TCGA sample expression profiles to the TIDE database (http://tide.dfci.harvard.edu/) to obtain dysfunction, exclusion, TIDE, and microsatellite instability (MSI) scores for each sample^[37]. Next, we used the ROC curve to compare the predictive efficacy of the risk score with the TIDE and tumor inflammatory marker (TIS) score^[38].

2.7. Drug Sensitivity Analysis

We downloaded transcriptomic gene expression data and FDA-certified drug sensitivity data from the CellMiner database (https://discover.nci.nih.gov/cellminer/) and used Pearson correlation analysis to investigate the correlation of the sensitivity of 216 FDA-certified drugs (Supplementary Table 1) with prognosis-related gene expression.

2.8. Cell Culture and siRNA Treatment

Gastric mucosal epithelial cells (GES1), STAD cells (AGS, MKN45, BGC-823, SNU-216, MGC-803, SGC-7901, N87 and HGC-27) were from the American Type Culture Collection (Manassas, VA, USA). Lipofectamine 3000 (Invitrogen, California, USA) was used to transfect MATN3 siRNA (RiboBio, Guangzhou, China) into cells.

2.9. WB Analysis

We extracted proteins from BGC-823 and AGS cells using RIPA lysis buffer (Servicebio, China) and separated the protein samples using 10% SDS-polyacrylamide gel electrophoresis^[39]. Then we transferred the protein to polyvinylidene fluoride membranes. After being blocked with skim milk, membranes were incubated overnight at 4°C with the following primary antibodies: MATN3 (Abclonal, China), GAPDH (Proteintech, USA), then followed by the secondary antibody at 25°C for 1 hour. Finally, the proteins were detected by West Pico Plus Chemiluminescent Substrate (Thermo Fisher Scientific, USA).

2.10. qRT-PCR

TRIzol reagent (TaKaRa, Japan) was used to extract cellular RNA. Hi Script II QRT SuperMix (Vazyme, China) was employed to synthesize cDNA, then qRT-RCR was performed using ChamQ Universal SYBR qPCR Master Mix (Vazyme, China). The primers were as follows: MATN3-Forward: ACATGGCGTCCCTCAAGATG; MATN3-Reverse: GCACAGAAGGTTTCCTGGAATC; GAPDH-Forward: GACCACAGTCCATGCCATCA; GAPDH-Reverse: GTCAAAGGTGGAGGAGTGGG.

2.11. CCK8 and Colony Formation Assays

After being transfected with MATN3 siRNA for 48 h, BGC-823 and AGS cells were seeded into 96-well plates. Then we added the CCK8 reagent (MCE, USA) according to the manufacturer's instructions and detected the OD450 values using a microplate reader (BioTek, USA). For colony formation assay, BGC-823 and AGS cells were inoculated in 6-well plates and cultured for 2 weeks, and the number of cell clones per well was counted after crystalline violet staining.

2.12. Wound Healing Assay

BGC-823 and AGS cells were seeded into a 24-well-plate and then scraped with a 10 µl pipette tip. Images of cell migration were captured at 0, 24, and 48 hours after scratching.

2.13. Transwell Migration Assay

Cells were cultured in the upper chamber with 200 µl medium without serum, and the lower chamber was added to 500 µl medium containing 20% fetal bovine serum. After incubation for 24 h at 37°, the cells in the lower chamber were fixed with methanol and stained with 1% crystal violet, and then the cells were counted with light microscopy.

2.14. Statistical Analysis

The gene differences between tumor tissue and normal tissue were analyzed by the Wilcoxon test. R software (version 4.0.4) and GraphPad Prism 7 software (version 7.0) were used to perform the statistical analysis. P < 0.05 was considered statistically significant.

3.1. Acquisition of Differentially Expressed Amino Acid Metabolism Genes (DEAAMGs) related to OS in TCGA Cohort

Differential expression analysis of the gene expression and clinical data in the TCGA cohort identified a total of 2495 DEAAMGs between normal tissue and tumor tissue of STAD patients, including 574 down-regulated and 1921 up-regulated genes in STAD tissue (Fig. 2, Supplementary Table 2). Next, 294 DEAAMGs were found to be related to OS by Univariate cox regression analysis, including 161 high-risk genes and 133 low-risk genes (Supplementary Table 3).

3.2. Construction of an Amino Acid Metabolism Genes-Related Prognostic Model in TCGA Cohort

Through analyzing the 294 prognosis-related DEAAMGs and excluding the relatively unimportant genes from the model through LASSO Cox regression analysis, we screened 3 genes (SERPINE1, NRP1, MATN3) to develop a prognostic risk model (Fig. 3A, B). Risk score = 0.022458 × SERPINE1 + 0.046362 × NRP1 + 0.053160 × MATN3 (Table 2). In this model, SERPINE1, NRP1 and MATN3 were indicators of the poor prognosis. The median cut-off value of risk score divided STAD patients into high-risk and low-risk groups (Fig. 3C). The results of PCA and t-SNE analyses demonstrated that the high-risk and low-risk groups could be separated based on the expression of 3 prognostic DEAAMGs (Fig. 3D, E). In addition, the survival status scatter plots suggested that high-risk patients were more likely to have a poor prognosis (Fig. 3F). Similarly, KM survival analysis results also demonstrated that patients in the low-risk group had a longer OS (p < 0.0001) (Fig. 3G). Furthermore, we plotted a nomogram to precisely predict the prognosis for each patient according to the risk score and clinical characteristics. For example, the total points of the patient in the nomogram were 327, which represented the 1, 3 and 5-year survival rates were 75.4%, 43.4% and 31.2%, respectively (Fig. 3H). The calibration curve showed the high predictive accuracy of the nomogram (Fig. 3I).

3.3. Validation of the 3-Gene Prognostic Model in the GEO Cohort

To test the predictive stability of this prognostic model, we downloaded transcriptomic gene expression data and clinical data of STAD patients from the GEO database as external validation. Similarly, we divided STAD patients into high- and low-risk groups based on the above median cut-off value of risk score (Fig. 4A). PCA and t-SNE analysis revealed that the expression of the model genes could separate the high- and low-risk groups in the GEO cohort well (Fig. 4B, C). Besides, scatter plots and KM survival analyses showed that high-risk patients were prone to have a poorer prognosis (Fig. 4D) and shorter OS (p < 0.001) compared to low-risk patients (Fig. 4E). In addition, we analyzed the clinical characteristics and found that the risk scores were higher in the patients with tumor grade 3 compared to those with grade 1–2 (p < 0.01) (Fig. 4F), and patients with tumor stage III-IV also had higher risk scores than those with stage I-II (p < 0.01) (Fig. 4G).

3.4. KEGG Pathway Enrichment Analysis

To explore the potential mechanism of the model genes regulating the prognosis, we used GSEA to perform the KEGG pathway enrichment analysis and found that the focal adhesion and ECM receptor interaction pathways were significantly enriched in the high-risk group, while the low-risk group was mainly enriched in the spliceosome and oxidative phosphorylation pathways (Fig. 5A). In addition, we found that tumor-associated pathways were mainly enriched in the high-risk group, such as the MAPK signaling pathway, JAK-STAT signaling pathway and TGF-β signaling pathway (Fig. 5B-D). Notably, we noticed 4 immune-related pathways, such as cytokine-cytokine receptor interaction, toll like receptor (TLR) signaling pathway, chemokine signaling pathway and nod like receptor signaling pathway were also significantly enriched in the high-risk group (Fig. 5E-H), indicating that amino acid metabolism genes might be associated with the progression of STAD through regulating the immune-related pathways.

3.5. Immune Infiltration Analysis

Thus, to further explore the correlation between amino acid metabolism and immune infiltration in STAD, we performed immune cell subtypes and immune-related functions analysis by ssGSEA. We found that aDCs, DCs, iDCs, pDCs, B cells, CD8 + T cells, macrophages, mast cells, neutrophils, T helper cells, Tfh, Th1 cells, TIL and Treg cells were significantly increased in the high-risk group than low-risk group (p < 0.01) (Fig. 6A). In addition, immune-related function analysis showed that inflammation-promoting, parainflammation, APC co-inhibition, APC co-stimulation, T cell co-inhibition, T-cell co-stimulation, CCR, check-point, cytolytic activity, HLA, Type I IFN response and Type II IFN response were significantly increased in the high-risk group than low-risk group (Fig. 6B). There were six types of immune infiltration in human tumors, namely C1 (wound healing type), C2 (IFN-gamma dominant type), C3 (inflammatory type), C4 (lymphocyte depleted type), C5 (immunologically quiet type) and C6 (TGF-β dominant type)^[40]. Among them, the STAD patients belonging to the C5 immune subtype in the TCGA dataset were not sufficient for study, therefore, we analyzed the other five immune subtypes. We found that high-risk scores were significantly related to C6, while low-risk scores were significantly related to C4 (Fig. 6C). Furthermore, Spearman correlation analysis showed that high risk score was associated with high immune score (p < 0.0001) (Fig. 6D), ESTIMATE score (p < 0.0001) (Fig. 6E), stromal score (p < 0.0001) (Fig. 6F), and low tumor purity (p < 0.0001) (Fig. 6G), suggesting that the higher immune composition and lower tumor purity predicted the poorer prognosis for STAD patients.

3.6. Immunotherapy Response Evaluation and Model Comparison

Given that the risk score was related to immune cell infiltration according to the above results, we analyzed immunotherapy biomarkers based on the TCGA dataset to further investigate whether our prognostic model could be used to predict immunotherapy response. The results showed a higher dysfunction score (p < 0.001) (Fig. 7A), exclusion score (p < 0.001) (Fig. 7B), TIDE score (p < 0.001) (Fig. 7C) and a lower MSI score (p < 0.001) (Fig. 7D) in high-risk group, suggesting that high-risk group had a higher immune escape potential and a less benefit from immunotherapy. Furthermore, compared with the TIDE score and TIS score, our risk score had the largest area under the curve value, indicating a higher prognostic predictive efficacy than the TIDE score and TIS score (Fig. 7E).

3.7. Drug Sensitivity Analysis

To investigate the availability of our prognostic model in chemotherapy and targeted therapy, we analyzed the correlation between our model genes and FDA-approved drug sensitivity. The results demonstrated that the expression of NRP1 was positively correlated with the treatment sensitivity of dasatinib, simvastatin, idelalisib and midostaurin (p < 0.001) (Fig. 8), and negatively correlated with the treatment sensitivity of tamoxifen (p < 0.001) (Fig. 8). The expression of MATN3 was negatively correlated with the treatment sensitivity of eribulin mesilate, vinblastine, pipamperone, paclitaxel and vinorelbine (p < 0.001) (Fig. 8) and positively correlated with the treatment sensitivity of erlotinib and dacomitinib (p < 0.001) (Fig. 8). SERPINE1 expression was negatively correlated with tamoxifen and nilotinib sensitivity (p < 0.001) (Fig. 8), and positively correlated with simvastatin and lenvatinib sensitivity (p < 0.001) (Fig. 8).

3.8. Interference with MATN3 Expression Inhibits Proliferation and Migration of STAD Cells

To further increase the credibility of our prognostic model, we conducted functional validation through in vitro experiments. Since previous studies have reported the role of SERPINE1 and NRP1 in promoting the proliferation and migration of gastric cancer cells, thus, we selected MATN3 for subsequent functional validation^{[41, 42]}. We detected the expression of MATN3 mRNA in gastric mucosal epithelial cell line GES1 and STAD cell lines AGS, BGC-823, SNU-216, MKN45, SGC-7901, MGC-803, N87 and HGC-27. The results showed that MATN3 expression was highest in AGS (p < 0.05) and BGC-823 cells (Fig. 9A), thus we used AGS and BGC-823 cells for subsequent experiments. The expression of MATN3 in BGC-823 and AGS cells was successfully interfered with siRNA and confirmed by qRT-PCR (p < 0.05) and WB (Fig. 9B, C). The CCK8 assay showed that MATN3 interference significantly inhibited BGC-823 and AGS cell proliferation (p < 0.05) (Fig. 9D). The colony formation assay revealed that MATN3 interference significantly inhibited colony formation in BGC-823 and AGS cells (p < 0.05) (Supplementary Fig. 1A, B). Subsequently, we used wound healing and transwell migration assays to explore the effect of MATN3 interference on the migration ability of STAD cells. The results showed that interference of MATN3 significantly inhibited the migration ability of BGC-823 and AGS cells (p < 0.001) (Fig. 9E-H). Taken together, interference of MATN3 expression inhibited STAD cell proliferation and migration, further confirming the predictive credibility of our prognostic model.

The important role of metabolism in tumors is gradually being recognized. Increasing evidence has proven that abnormal metabolic features exist in tumors^{[43, 44]}. Dysfunctional metabolic activities can promote tumor cell proliferation, recurrence and metastasis, affecting the immunosuppressive properties and malignant phenotype of tumors^[45–48]. With the rapid development of bioinformatics technology, more and more studies have illustrated the metabolism-related risk profile of the tumor, including STAD^[49–51]. However, there is still a lack of study on amino acid metabolism-related genes in STAD. Thus, in this study, we analyzed the characteristic of amino acid metabolism-related genes in STAD.

We firstly extracted prognosis-related amino acid metabolism genes by analyzing transcriptomic gene expression data and clinical parameters of STAD patients in TCGA and successfully constructed an amino acid metabolism genes-related prognostic model, consisting of 3 genes (SERPINE1, NRP1, MATN3). SERPINE1, a member of the serine protease inhibitor family, has been reported to promote tumor progression and metastasis^[52–54]. Recent studies have reported that SERPINE1 is a reliable prognostic marker for gastric cancer, breast cancer, ovarian cancer, colorectal cancer, bladder cancer and many other cancers^[55–61]. Our study identified SERPINE1 as a prognostic marker in STAD and found the potential role of SERPINE1 in regulating the immune cells infiltration, which is consistent with previous studies^{[62, 63]}. NRP-1, a transmembrane protein, acts as a multifunctional coreceptor involved in cancer initiation, growth and metastasis^[64]. NRP-1 is a potential target in cancer therapy and negatively correlates with the prognosis of esophageal squamous cell carcinoma, hepatocellular carcinoma and colorectal cancer^[65–67]. In gastric cancer, NRP-1 promotes the gastric cancer cells proliferation and migration of and is associated with the clinicopathological staging^[42]. In our study, we further confirmed its role in predicting the prognosis and immunotherapy efficacy of STAD patients. MATN3, a protein-encoding gene, has been mainly focused on its role in cartilage and skeletal development in past studies^[68]. Recently, MATN3 was reported to be highly expressed in STAD patients^[69]. However, the role of MATN3 in the prognosis of STAD patients was still unclear. In our study, we found that MATN3 might influence the prognosis of STAD patients by involving in the metabolism acid metabolism and immune infiltration in the tumor microenvironment. In addition, we further confirmed the importance of MATN3 in promoting the proliferation and migration ability in gastric cancer through in vitro studies.

Immunometabolism provides an understanding of the relationship between immune response and metabolism^[70]. Amino acids, as the main raw materials for protein synthesis, regulate the immune function through energy metabolism, redox balance, epigenetic modification and protein post-translational modification^[71]. Tumor cells require a continuous supply of amino acids, which leads to a metabolic competition between tumor cells and immune cells^[22]. Metabolic remodeling of immune cells in the tumor microenvironment plays a potential role in the progression and metastasis of the tumor^{[24, 72]}. We explored the role of amino acid metabolism-related model in immune infiltration in STAD. In this study, we found immune-related pathways, such as cytokine-cytokine receptor interaction, TLR signaling pathway, chemokine signaling pathway and nod like receptor signaling pathway were significantly enriched in the high-risk group. Further analysis demonstrated that the high-risk scores were associated with higher immune cell infiltration and TGF-β dominant immune subtype, demonstrating that amino acid metabolism-related genes might influence STAD patient prognosis through regulating immune-related pathways and TGF-β dominant immune subtype cell infiltration. In gastric cancer, CXC chemokines and their receptors regulate cell transport into and out of tumor microenvironment, influence tumorigenesis indirectly by regulating tumor growth, survival, transformation, invasion, and metastasis, as well as indirectly by modulating tumor-leukocyte interactions and angiogenesis^[73]. TLRs are key innate immunopathogenic sensors mediating chronic inflammation and carcinogenic responses. TLR9 promotes gastric cancer initiation and the gastric inflammation and hyperplasia induced by Helicobacter pylori^[74]. Nod like receptor family are patterns recognition receptors related with innate immunity. NLRX1, a member of the nod like receptor family, is reported to be associated with increased risk of Helicobacter pylori infection, which is common risk factor for developing gastric cancer^[75]. TGF-β, a key factor of immune homeostasis and tolerance, inhibits the expansion and function of the immune system^[76]. TGF-β signaling has dual roles in the development and progression of gastrointestinal tumor as both a suppressor and promoter^[77]. Furthermore, we revealed that the high-risk group had a higher immune escape potential, indicating a poorer immunotherapy effect, which provided a potential reference for applying immunotherapy in STAD patients.

KEGG functional enrichment analysis revealed the possible mechanisms of amino acid metabolism genes in STAD. We found that the tumor-associated pathways, such as the JAK-STAT signaling pathway, TGF-β signaling pathway and MAPK signaling pathway were enriched in the high-risk group, suggesting that amino acid metabolism genes may affect STAD prognosis through these pathways. The relationship between MAPK signaling pathway and amino acid metabolism has been reported. A previous study showed that L-Glutamine deficiency disturbed amino acid metabolism and attenuated mTOR and MAPK/ERK signaling pathways, thereby affecting protein synthesis and cell proliferation^[78]. The JAK-STAT signaling pathway is related with varieties of physiological processes such as cell proliferation, immune response, and stem cell self-renewal^[79–81]. Aberrant activation of the JAK/STAT pathway has also been reported to be related to gastric cancer progression^[82]. Inhibition of STAT3 significantly reduces the expression of the anti-apoptotic protein survivin, which promotes gastric cancer cell death^[83]. TGF-β signaling pathway can affect cancer progression by regulating the immune response, tumor microenvironment, epithelial-mesenchymal transition, and cancer cell stemness^[84–86]. A previous study showed that TGF-β expression was elevated in the tumor tissue of gastric cancer patients and was associated with a poorer prognosis^[87].

To further increase the clinical translational value of our study, we performed the FDA-approved drug sensitivity analysis. By analyzing data from the CellMiner database, we found that our model genes were associated with the patient’s sensitivity to many FDA-approved drugs. The expression of SERPINE1 was positively correlated with the therapeutic sensitivity to lenvatinib. It has been reported that the combination of lenvatinib with pembrolizumab has promising antitumor activity in patients with advanced gastric cancer^[88]. In addition, our study found that MATN3 expression was positively correlated with the therapeutic sensitivity to dacomitinib. In a previous clinical trial, dacomitinib showed good efficacy in HER2-positive gastric cancer patients^[89]. Therefore, our study may provide a potential reference for drug selection in STAD patients.

In this study, we developed a prognostic model of amino acid metabolism in STAD and demonstrated its high efficiency through the external validation. KEGG pathway enrichment analysis, immune infiltration analysis, immunotherapy efficacy prediction and drug sensitivity analysis further revealed that this model is highly correlated with immunity, discovered the potential mechanism, and enhanced the clinical application value. Finally, in vitro experiment demonstrated that model gene MATN3 promoted the proliferation and migration in gastric cancer, providing the basis for further study on the amino acid metabolism in STAD.

In conclusion, our study revealed that targeting amino acid metabolism might be a potential therapeutic method to improve STAD patients’ prognosis and provides a personalized predictive tool for prognosis and immunotherapy response in STAD.

Differentially expressed amino acid metabolism genes (DEAAMGs)

Gene Expression Omnibus (GEO)

Gastric mucosal epithelial cells (GES1)

Gene Set Enrichment Analysis (GSEA)

Indoleamine 2,3-dioxygenase (IDO)

Kyoto Gene and Genome Encyclopedia (KEGG)

Kaplan-Meier (KM)

Least absolute shrinkage and selection operator (LASSO)

Microsatellite instability (MSI)

Quantitative real-time polymerase chain reaction (qRT-PCR)

Single sample genomic enrichment analysis (ssGSEA)

Stomach adenocarcinoma (STAD)

The Cancer Genome Atlas (TCGA)

Tumor immune dysfunction and exclusion (TIDE)

Tumor inflammatory marker (TIS)

Western blot (WB)

Authors' contributions

Guangyuan Hu, Xiaohong Peng and Yuanyuan Zhang conceived of the study. Wenjun Zhu, Min Fu, Qianxia Li and Xin Chen searched the literature, conducted data analysis and experiments, and produced the manuscript. Xiaoyu Li, Na Luo, Wenhua Tang, Feng Yang, Ziqi Chen and Yiling Zhang revised the manuscript. All authors have read and approved the manuscript.

Acknowledgments

Thanks to the TCGA and GEO databases for the availability of the above data. Thanks to the Huazhong University of Science & Technology Analytical & Testing center for the technical support.

Funding

The study was funded by the National Natural Sciences Foundation of China (Grant No. 82003312, 82173311).

Availability of data and materials

All data generated and analyzed in this study are included in this article and its supplementary files. The transcriptomic gene expression data (TCGA and GSE84433 series matrix) and corresponding clinical information were downloaded from the TCGA database (https://portal.gdc.cancer.gov/) and GEO database (https://www.ncbi.nlm.nih.gov/).

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

Supplementary information

All the supplementary files have been properly named and uploaded.

Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries [J]. CA Cancer J Clin. 2021;71(3):209–49. 10.3322/caac.21660.
Smyth EC, Nilsson M, Grabsch HI, et al. Gastric cancer [J]. Lancet (London England). 2020;396(10251):635–48. 10.1016/S0140-6736(20)31288-5.
Karimi P, Islami F, Anandasabapathy S, et al. Gastric cancer: descriptive epidemiology, risk factors, screening, and prevention [J]. Cancer Epidemiol Biomarkers Prev. 2014;23(5):700–13. 10.1158/1055-9965.EPI-13-1057.
Butler M, Van Der Meer LT, Van Leeuwen FN. Amino Acid Depletion Therapies: Starving Cancer Cells to Death [J]. Trends Endocrinol Metab. 2021;32(6):367–81. 10.1016/j.tem.2021.03.003.
Warburg O. On the origin of cancer cells [J]. Science. 1956;123(3191):309–14.
Hanahan D. Hallmarks of Cancer: New Dimensions [J]. Cancer Discov. 2022;12(1):31–46. 10.1158/2159-8290.CD-21-1059.
Najumudeen AK, Ceteci F, Fey SK, et al. The amino acid transporter SLC7A5 is required for efficient growth of KRAS-mutant colorectal cancer [J]. Nat Genet. 2021;53(1):16–26. 10.1038/s41588-020-00753-3.
Xu L, Chen J, Jia L, et al. SLC1A3 promotes gastric cancer progression via the PI3K/AKT signalling pathway [J]. J Cell Mol Med. 2020;24(24):14392–404. 10.1111/jcmm.16060.
Okuda K, Umemura A, Kataoka S, et al. Enhanced Antitumor Effect in Liver Cancer by Amino Acid Depletion-Induced Oxidative Stress [J]. Front Oncol. 2021;11(758549). 10.3389/fonc.2021.758549.
Qu Y-Y, Zhao R, Zhang H-L, et al. Inactivation of the AMPK-GATA3-ECHS1 Pathway Induces Fatty Acid Synthesis That Promotes Clear Cell Renal Cell Carcinoma Growth [J]. Cancer Res. 2020;80(2):319–33. 10.1158/0008-5472.CAN-19-1023.
Zhang L, Han J. Branched-chain amino acid transaminase 1 (BCAT1) promotes the growth of breast cancer cells through improving mTOR-mediated mitochondrial biogenesis and function [J]. Biochem Biophys Res Commun. 2017;486(2):224–31. 10.1016/j.bbrc.2017.02.101.
Wang Z-Q, Faddaoui A, Bachvarova M, et al. BCAT1 expression associates with ovarian cancer progression: possible implications in altered disease metabolism [J]. Oncotarget. 2015;6(31):31522–43. 10.18632/oncotarget.5159.
Wei Z, Liu X, Cheng C, et al. Metabolism of Amino Acids in Cancer [J]. Front Cell Dev Biol. 2020;8(603837). 10.3389/fcell.2020.603837.
Wu H, Xue R, Tang Z, et al. Metabolomic investigation of gastric cancer tissue using gas chromatography/mass spectrometry [J]. Anal Bioanal Chem. 2010;396(4):1385–95. 10.1007/s00216-009-3317-4.
Choi JM, Park WS, Song KY, et al. Development of simultaneous analysis of tryptophan metabolites in serum and gastric juice - an investigation towards establishing a biomarker test for gastric cancer diagnosis [J]. Biomed Chromatogr. 2016;30(12):1963–74. 10.1002/bmc.3773.
Deng K, Lin S, Zhou L, et al. Three aromatic amino acids in gastric juice as potential biomarkers for gastric malignancies [J]. Anal Chim Acta. 2011;694(1–2):100–7. 10.1016/j.aca.2011.03.053.
Chen J-L, Tang H-Q, Hu J-D, et al. Metabolomics of gastric cancer metastasis detected by gas chromatography and mass spectrometry [J]. World J Gastroenterol. 2010;16(46):5874–80.
Jung J, Jung Y, Bang EJ et al. Noninvasive diagnosis and evaluation of curative surgery for gastric cancer by using NMR-based metabolomic profiling [J]. Ann Surg Oncol, 2014, 21 Suppl 4(S736-S742. doi:10.1245/s10434-014-3886-0
Esslinger CS, Cybulski KA, Rhoderick JF. Ngamma-aryl glutamine analogues as probes of the ASCT2 neutral amino acid transporter binding site [J]. Bioorg Med Chem. 2005;13(4):1111–8.
Lu J, Chen M, Tao Z, et al. Effects of targeting SLC1A5 on inhibiting gastric cancer growth and tumor development in vitro and in vivo [J]. Oncotarget. 2017;8(44):76458–67. 10.18632/oncotarget.19479.
Cai Z, Zhao J-S, Li J-J, et al. A combined proteomics and metabolomics profiling of gastric cardia cancer reveals characteristic dysregulations in glucose metabolism [J]. Mol Cell Proteomics. 2010;9(12):2617–28. 10.1074/mcp.M110.000661.
Kelly B, Pearce EL. Amino Assets: How Amino Acids Support Immunity [J]. Cell Metabol. 2020;32(2):154–75. 10.1016/j.cmet.2020.06.010.
Bian Y, Li W, Kremer DM, et al. Cancer SLC43A2 alters T cell methionine metabolism and histone methylation [J]. Nature. 2020;585(7824):277–82. 10.1038/s41586-020-2682-1.
Wu D. Innate and Adaptive Immune Cell Metabolism in Tumor Microenvironment [J]. Adv Exp Med Biol. 2017;1011:211–23. 10.1007/978-94-024-1170-6_7.
Brochez L, Chevolet I, Kruse V. The rationale of indoleamine 2,3-dioxygenase inhibition for cancer therapy [J]. Eur J Cancer. 2017;76:167–82. 10.1016/j.ejca.2017.01.011.
Howden AJM, Hukelmann JL, Brenes A, et al. Quantitative analysis of T cell proteomes and environmental sensors during T cell differentiation [J]. Nat Immunol. 2019;20(11):1542–54. 10.1038/s41590-019-0495-x.
Carr EL, Kelman A, Wu GS, et al. Glutamine uptake and metabolism are coordinately regulated by ERK/MAPK during T lymphocyte activation [J]. J Immunol (Baltimore Md: 1950). 2010;185(2):1037–44. 10.4049/jimmunol.0903586.
Loftus RM, Assmann N, Kedia-Mehta N, et al. Amino acid-dependent cMyc expression is essential for NK cell metabolic and functional responses in mice [J]. Nat Commun. 2018;9(1):2341. 10.1038/s41467-018-04719-2.
Yoon BR, Oh Y-J, Kang SW, et al. Role of SLC7A5 in Metabolic Reprogramming of Human Monocyte/Macrophage Immune Responses [J]. Front Immunol. 2018;9(53). 10.3389/fimmu.2018.00053.
Xiao M, Yang H, Xu W, et al. Inhibition of α-KG-dependent histone and DNA demethylases by fumarate and succinate that are accumulated in mutations of FH and SDH tumor suppressors [J]. Genes Dev. 2012;26(12):1326–38. 10.1101/gad.191056.112.
Hutter C, Zenklusen JC. Cell. 2018;173(2):283–5. 10.1016/j.cell.2018.03.042. The Cancer Genome Atlas: Creating Lasting Value beyond Its Data [J].
Barrett T, Troup DB, Wilhite SE, et al. NCBI GEO: mining tens of millions of expression profiles–database and tools update [J]. Nucleic Acids Res. 2007;35(Database issue):D760–5.
Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles [J]. Proc Natl Acad Sci U S A. 2005;102(43):15545–50.
Jiang J, Liu D, Xu G, et al. TRIM68, PIKFYVE, and DYNLL2: The Possible Novel Autophagy- and Immunity-Associated Gene Biomarkers for Osteosarcoma Prognosis [J]. Front Oncol. 2021;11(643104). 10.3389/fonc.2021.643104.
Zuo S, Wei M, Wang S, et al. Pan-Cancer Analysis of Immune Cell Infiltration Identifies a Prognostic Immune-Cell Characteristic Score (ICCS) in Lung Adenocarcinoma [J]. Front Immunol. 2020;11(1218). 10.3389/fimmu.2020.01218.
Yoshihara K, Shahmoradgoli M, Martínez E, et al. Inferring tumour purity and stromal and immune cell admixture from expression data [J]. Nat Commun. 2013;4(2612). 10.1038/ncomms3612.
Jiang P, Gu S, Pan D, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response [J]. Nat Med. 2018;24(10):1550–8. 10.1038/s41591-018-0136-1.
Danaher P, Warren S, Lu R, et al. Pan-cancer adaptive immune resistance as defined by the Tumor Inflammation Signature (TIS): results from The Cancer Genome Atlas (TCGA) [J]. J Immunother Cancer. 2018;6(1):63. 10.1186/s40425-018-0367-1.
Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications [J]. Proc Natl Acad Sci U S A. 1979;76(9):4350–4.
Lin Z, Xu Q, Miao D, et al. An Inflammatory Response-Related Gene Signature Can Impact the Immune Status and Predict the Prognosis of Hepatocellular Carcinoma [J]. Front Oncol. 2021;11(644416). 10.3389/fonc.2021.644416.
Chen S, Li Y, Zhu Y et al. SERPINE1 Overexpression Promotes Malignant Progression and Poor Prognosis of Gastric Cancer [J]. J Oncol, 2022, 2022(2647825. doi:10.1155/2022/2647825
Li L, Jiang X, Zhang Q, et al. Neuropilin-1 is associated with clinicopathology of gastric cancer and contributes to cell proliferation and migration as multifunctional co-receptors [J]. J Exp Clin Cancer Res. 2016;35(16). 10.1186/s13046-016-0291-5.
Boroughs LK, Deberardinis RJ. Metabolic pathways promoting cancer cell survival and growth [J]. Nat Cell Biol. 2015;17(4):351–9. 10.1038/ncb3124.
Armitage EG, Ciborowski M. Applications of Metabolomics in Cancer Studies [J]. Adv Exp Med Biol. 2017;965:209–34. 10.1007/978-3-319-47656-8_9.
Li H, Ning S, Ghandi M, et al. The landscape of cancer cell line metabolism [J]. Nat Med. 2019;25(5):850–60. 10.1038/s41591-019-0404-8.
Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation [J]. Cell. 2011;144(5):646–74. 10.1016/j.cell.2011.02.013.
Chang C-H, Qiu J, O'sullivan D, et al. Metabolic Competition in the Tumor Microenvironment Is a Driver of Cancer Progression [J]. Cell. 2015;162(6):1229–41. 10.1016/j.cell.2015.08.016.
Ho P-C, Bihuniak JD, Macintyre AN, et al. Phosphoenolpyruvate Is a Metabolic Checkpoint of Anti-tumor T Cell Responses [J]. Cell. 2015;162(6):1217–28. 10.1016/j.cell.2015.08.012.
Zhao Y, Zhang J, Wang S, et al. Identification and Validation of a Nine-Gene Amino Acid Metabolism-Related Risk Signature in HCC [J]. Front Cell Dev Biol. 2021;9(731790). 10.3389/fcell.2021.731790.
Zhang Y, Liu W, Feng W, et al. Identification of 14 Differentially-Expressed Metabolism-Related Genes as Potential Targets of Gastric Cancer by Integrated Proteomics and Transcriptomics [J]. Front Cell Dev Biol. 2022;10(816249). 10.3389/fcell.2022.816249.
Ye Z, Zheng M, Zeng Y, et al. A 13-Gene Metabolic Prognostic Signature Is Associated With Clinical and Immune Features in Stomach Adenocarcinoma [J]. Front Oncol. 2021;11(612952). 10.3389/fonc.2021.612952.
Declerck PJ, Gils A. Three decades of research on plasminogen activator inhibitor-1: a multifaceted serpin [J]. Semin Thromb Hemost. 2013;39(4):356–64. 10.1055/s-0033-1334487.
Seker F, Cingoz A, Sur-Erdem İ, et al. Identification of as a Regulator of Glioblastoma Cell Dispersal with Transcriptome Profiling [J]. Cancers (Basel). 2019;11(11). 10.3390/cancers11111651.
Hanekom GS, Stubbings HM, Kidson SH. The active fraction of plasmatic plasminogen activator inhibitor type 1 as a possible indicator of increased risk for metastatic melanoma [J]. Cancer Detect Prev. 2002;26(1):50–9.
Xu B, Bai Z, Yin J, et al. Global transcriptomic analysis identifies as a prognostic biomarker associated with epithelial-to-mesenchymal transition in gastric cancer [J]. PeerJ. 2019;7(e):7091. 10.7717/peerj.7091.
Jevrić M, Matić IZ, Krivokuća A, et al. Association of uPA and PAI-1 tumor levels and 4G/5G variants of PAI-1 gene with disease outcome in luminal HER2-negative node-negative breast cancer patients treated with adjuvant endocrine therapy [J]. BMC Cancer. 2019;19(1):71. 10.1186/s12885-018-5255-z.
Harbeck N, Kates RE, Gauger K, et al. Urokinase-type plasminogen activator (uPA) and its inhibitor PAI-I: novel tumor-derived factors with a high prognostic and predictive impact in breast cancer [J]. Thromb Haemost. 2004;91(3):450–6.
Sakakibara T, Hibi K, Koike M, et al. Plasminogen activator inhibitor-1 as a potential marker for the malignancy of colorectal cancer [J]. Br J Cancer. 2005;93(7):799–803.
Nakatsuka E, Sawada K, Nakamura K, et al. Plasminogen activator inhibitor-1 is an independent prognostic factor of ovarian cancer and IMD-4482, a novel plasminogen activator inhibitor-1 inhibitor, inhibits ovarian cancer peritoneal dissemination [J]. Oncotarget. 2017;8(52):89887–902. 10.18632/oncotarget.20834.
Chan OTM, Furuya H, Pagano I, et al. Association of MMP-2, RB and PAI-1 with decreased recurrence-free survival and overall survival in bladder cancer patients [J]. Oncotarget. 2017;8(59):99707–21. 10.18632/oncotarget.20686.
Becker M, Szarvas T, Wittschier M, et al. Prognostic impact of plasminogen activator inhibitor type 1 expression in bladder cancer [J]. Cancer. 2010;116(19):4502–12. 10.1002/cncr.25326.
Huang Z, Liu X, Wu C, et al. A New Strategy to Identify ceRNA-Based CCDC144NL-AS1/SERPINE1 Regulatory Axis as a Novel Prognostic Biomarker for Stomach Adenocarcinoma High Throughput Transcriptome Data Mining and Computational Verification [J]. Front Oncol. 2021;11(802727). 10.3389/fonc.2021.802727.
Wang S, Pang L, Liu Z, et al. SERPINE1 associated with remodeling of the tumor microenvironment in colon cancer progression: a novel therapeutic target [J]. BMC Cancer. 2021;21(1):767. 10.1186/s12885-021-08536-7.
Prud'homme GJ, Glinka Y. Neuropilins are multifunctional coreceptors involved in tumor initiation, growth, metastasis and immunity [J]. Oncotarget. 2012;3(9):921–39.
Bergé M, Allanic D, Bonnin P, et al. Neuropilin-1 is upregulated in hepatocellular carcinoma and contributes to tumour growth and vascular remodelling [J]. J Hepatol. 2011;55(4):866–75. 10.1016/j.jhep.2011.01.033.
Alattar M, Omo A, Elsharawy M, et al. Neuropilin-1 expression in squamous cell carcinoma of the oesophagus [J]. Eur J Cardiothorac Surg. 2014;45(3):514–20. 10.1093/ejcts/ezt380.
Staton CA, Koay I, Wu JM, et al. Neuropilin-1 and neuropilin-2 expression in the adenoma-carcinoma sequence of colorectal cancer [J]. Histopathology. 2013;62(6):908–15. 10.1111/his.12098.
Klatt AR, Nitsche DP, Kobbe B, et al. Molecular structure and tissue distribution of matrilin-3, a filament-forming extracellular matrix protein expressed during skeletal development [J]. J Biol Chem. 2000;275(6):3999–4006.
Wu P-L, He Y-F, Yao H-H, et al. Martrilin-3 (MATN3) Overexpression in Gastric Adenocarcinoma and its Prognostic Significance [J]. Med Sci Monitor: Int Med J Experimental Clin Res. 2018;24:348–55.
O'neill LJ, Kishton RJ, Rathmell J. A guide to immunometabolism for immunologists [J]. Nat Rev Immunol. 2016;16(9):553–65. 10.1038/nri.2016.70.
Lieu EL, Nguyen T, Rhyne S, et al. Amino acids in cancer [J]. Exp Mol Med. 2020;52(1):15–30. 10.1038/s12276-020-0375-3.
Uyttenhove C, Pilotte L, Théate I, et al. Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase [J]. Nat Med. 2003;9(10):1269–74.
Lee HJ, Song I-C, Yun H-J, et al. CXC chemokines and chemokine receptors in gastric cancer: from basic findings towards therapeutic targeting [J]. World J Gastroenterol. 2014;20(7):1681–93. 10.3748/wjg.v20.i7.1681.
Tang K, Mcleod L, Livis T, et al. Toll-like Receptor 9 Promotes Initiation of Gastric Tumorigenesis by Augmenting Inflammation and Cellular Proliferation [J]. Cell Mol Gastroenterol Hepatol. 2022;14(3):567–86. 10.1016/j.jcmgh.2022.06.002.
Pickering RJ, Booty LM. NLR in eXile: Emerging roles of NLRX1 in immunity and human disease [J]. Immunology. 2021;162(3):268–80. 10.1111/imm.13291.
Batlle E, Massagué J. Transforming Growth Factor-β Signaling in Immunity and Cancer [J]. Immunity. 2019;50(4):924–40. 10.1016/j.immuni.2019.03.024.
Luo J, Chen X-Q, Li P. The Role of TGF-β and Its Receptors in Gastrointestinal Cancers [J]. Transl Oncol. 2019;12(3):475–84. 10.1016/j.tranon.2018.11.010.
Zhu Y, Lin G, Dai Z, et al. L-Glutamine deprivation induces autophagy and alters the mTOR and MAPK signaling pathways in porcine intestinal epithelial cells [J]. Amino Acids. 2015;47(10):2185–97. 10.1007/s00726-014-1785-0.
Darnell JE, Kerr IM, Stark GR. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins [J]. Science. 1994;264(5164):1415–21.
Aaronson DS, Horvath CM. A road map for those who don't know JAK-STAT [J]. Science. 2002;296(5573):1653–5.
Rawlings JS, Rosler KM, Harrison DA. The JAK/STAT signaling pathway [J]. J Cell Sci. 2004;117(Pt 8):1281–3.
Giraud AS, Menheniott TR, Judd LM. Targeting STAT3 in gastric cancer [J]. Expert Opin Ther Targets. 2012;16(9):889–901. 10.1517/14728222.2012.709238.
Kanda N, Seno H, Konda Y, et al. STAT3 is constitutively activated and supports cell survival in association with survivin expression in gastric cancer cells [J]. Oncogene. 2004;23(28):4921–9.
Massagué J. TGFbeta in Cancer [J]. Cell. 2008;134(2):215–30. 10.1016/j.cell.2008.07.001.
Fabregat I, Fernando J, Mainez J, et al. TGF-beta signaling in cancer treatment [J]. Curr Pharm Design. 2014;20(17):2934–47.
David CJ, Massagué J. Contextual determinants of TGFβ action in development, immunity and cancer [J]. Nat Rev Mol Cell Biol. 2018;19(7):419–35. 10.1038/s41580-018-0007-0.
Hu W-Q, Wang L-W, Yuan J-P, et al. High expression of transform growth factor beta 1 in gastric cancer confers worse outcome: results of a cohort study on 184 patients [J]. Hepatogastroenterology. 2014;61(129):245–50.
Kawazoe A, Fukuoka S, Nakamura Y, et al. Lenvatinib plus pembrolizumab in patients with advanced gastric cancer in the first-line or second-line setting (EPOC1706): an open-label, single-arm, phase 2 trial [J]. Lancet Oncol. 2020;21(8):1057–65. 10.1016/S1470-2045(20)30271-0.
Oh D-Y, Lee K-W, Cho JY, et al. Phase II trial of dacomitinib in patients with HER2-positive gastric cancer [J]. Gastric Cancer. 2016;19(4):1095–103. 10.1007/s10120-015-0567-z.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Development of Amino Acid Metabolism-Related Prognostic Model and Immune Infiltration Analysis in Patients with Stomach Adenocarcinoma

Status:

Version 1

Abstract

Figures

Introduction

Material And Methods

2.1. Data Collection

2.2. Acquisition of Differentially Expressed Amino Acid Metabolism Genes (DEAAMGs)

2.3. Construction and Validation of Amino Acid Metabolism Genes-Related Prognostic Model

2.4. Functional Enrichment Analysis

2.5. Immune Infiltration Analysis

2.6. Immunotherapy Response Evaluation and Model Comparison

2.7. Drug Sensitivity Analysis

2.8. Cell Culture and siRNA Treatment

2.9. WB Analysis

2.10. qRT-PCR

2.11. CCK8 and Colony Formation Assays

2.12. Wound Healing Assay

2.13. Transwell Migration Assay

2.14. Statistical Analysis

Results

3.1. Acquisition of Differentially Expressed Amino Acid Metabolism Genes (DEAAMGs) related to OS in TCGA Cohort

3.2. Construction of an Amino Acid Metabolism Genes-Related Prognostic Model in TCGA Cohort

3.3. Validation of the 3-Gene Prognostic Model in the GEO Cohort

3.4. KEGG Pathway Enrichment Analysis

3.5. Immune Infiltration Analysis

3.6. Immunotherapy Response Evaluation and Model Comparison

3.7. Drug Sensitivity Analysis

3.8. Interference with MATN3 Expression Inhibits Proliferation and Migration of STAD Cells

Discussion

Conclusions

Abbreviations

Declarations

Authors' contributions

Acknowledgments

Funding

Availability of data and materials

Ethics approval and consent to participate

Consent for publication

Competing interests

Supplementary information

References

Additional Declarations

Supplementary Files

Status:

Version 1