Construction of Iron Metabolism-related Prognostic Features of Gastric Cancer Based on RNA Sequencing and TCGA Database

doi:10.21203/rs.3.rs-3125640/v1

Download PDF

Research Article

Construction of Iron Metabolism-related Prognostic Features of Gastric Cancer Based on RNA Sequencing and TCGA Database

https://doi.org/10.21203/rs.3.rs-3125640/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 13 Nov, 2023

Read the published version in BMC Cancer →

You are reading this latest preprint version

Background: Researches have manifested that the disorder of iron metabolism is participated in Gastric cancer (GC), but whether iron metabolism-relevant genes (IMRGs) is related to the survival outcome of GC and the molecular mechanism remain unknown.

Methods:Eleven tumor as well as nine adjacent normal tissues from GC patients were performed mRNA sequencing. Additionally, the mRNA data, clinicopathological data and DNA methylation data related to GC were acquired from publically digital database. Then, the relevant differentially expressed IMRGs (DE-IMRGs) were selected. Cox analyses were applied to select the signature DE-IMRGs and build a IMRGs signature. The receiver operating characteristic (ROC) as well as survival curves were produced to appraise the prognostic signature. Furthermore, a nomogram was established. The infiltration profiling of immune cell of cases was estimated via single-sample GSEA (ssGSEA) algorithm. Meanwhile, the relationship between signature genes (DOHH, P4HA3 and MMP1) and 24 immune cells were investigated. Finally, the abundance of the signature genes were measured via the real-time fluorescence quantitation experiment.

Results:58 DE-IMRGs were obtained, and a IMRGs signature according to the expression and corresponding coefficient of DOHH, P4HA3 and MMP1 was constructed. The curves of survival manifested that the prognostic outcome of patients in the high-risk group was terrible. The values of area under the ROC curves (AUCs) confirmed that the IMRGs signature presented good efficiency for predicting GC prognosis. Moreover, a nomogram with the treatment type and risk-score was established and performed well. Furthermore, 11 immune cells were distinguishing in the high- and low-risk groups, and P4HA3, MMP1, and DOHH were associated with these immune cells. Finally, qRT-PCR revealed that the abundance of DOHH, P4HA3 and MMP1 were high in tumor cases.

Conclusion: In conclusion, an iron metabolism-related signature was constructed and has significant values for foretelling the OS of GC.

Gastric cancer

prognostic signature

iron metabolism-related

mRNA sequencing

TCGA Database

Gastric cancer (GC) is a malignancy springing from the mucosal epithelial cells of stomach wall, which is one of the common digestive tract tumors. By 2020, the incidence of GC ranks fifth, and the mortality rate of GC ranks fourth ^[1]. At present, the treatment of gastric cancer is still mainly surgical treatment, combined with systemic chemotherapy, immunotherapy, targeted therapy, radiotherapy, and other auxiliary treatment mode. However, these treatments did not raise the five-year survival rate of patients with advanced GC^[2]. In addition, GC patients are characterized by low early diagnosis rate, low radical resection rate and low 5-year survival rate ^[4-6]. And relevant research results show that advanced GC patients' median survival time is less than 1 year, and the five-year survival rate is about 18%^[5-8]. The prognosis of metastatic GC patients is very poor, as the median of survival time is just 4-9 months^[9]. Therefore, we need to find biomarkers with good prognosis prediction for GC, and to provide a new basis for clinical gene detection, targeted therapy and individualized treatment of GC.

Iron is an essential nutrient for promoting cell metabolism, proliferation and growth. The hydrogen peroxide was catalyzed by ferrous iron to generate reactive oxygen species through Fenton reavtion, which not only causes lipid and protein damage, but also oxidative damage DNA, inducing mutations and facilitating the emergence and preservation of tumors. Iron metabolism disorder is engaged in tumor occurrence, angiogenesis, invasion as well as metastasis, which is a general appearance in many tumors. Related studies have found that there are abnormal iron metabolism in lung cancer, prostate cancer, liver cancer, breast cancer, and,kidney cancer ^[10-13]. Some researches have also manifested that iron metabolism disorder is engaged in the process of gastric cancer, but whether it is linked to the prognosis of GC and the specific molecular mechanism are still unknown.

In this study, we first screened the DE-IMRGs from the GC expression profile of The Cancer Genome Atlas (TCGA) database, then constructed the prognostic signature of iron metabolism related GC, and probed the linkage between the immune cells and signature genes. Finally, the expression of signature genes was confirmed through external datasets and RT-qPCR, which is a great significance for GC to explore potential therapeutic targets and molecular mechanisms.

2.1 Specimen acquisition & sequencing

This research was allowed by the ethical committee of The First Affiliated Hospital of Kunming Medical University. All patients in this study signed written informed consent documents.

11 tumor tissues and 9 normal tissues from GC patients were included in mRNA-seq analysis, where normal tissues were selected at least 1.0 cm from the tumor margin. These specimens were all gathered from The First Affiliated

Hospital of Kunming Medical University and were freshly frozen and reserved at -80 °C immediately after surgery. Afterward, the total RNA from tissue cases was extracted using TRIzol reagent (Invitrogen, CA, USA). Then a library with a fragment size of 300bp ± 50bp by PCR was estimated. Finally, the mRNA sequencing was run relying on the illumina Novaseq™ 6000 (LC Bio Technology CO.,Ltd. Hangzhou, CHN).

Additionally, the mRNA data, clinicopathological data and DNA methylation information were regained from the TCGA database, including 373 tumor cases and 32 normal cases. The iron metabolism-relevant gene cohorts (v7.4) was downloaded from the Molecular Signatures Database (MSigDB) database and retrieved 428 IMRGs.

2.2 Identification of differentially expressed genes (DEGs)

The mRNA data from the sequencing data were selected for identification of the DEGs through edgeR package^[14-15]. Similarly, the mRNA data from the 373 tumor and 32 normal cases of TCGA database were used to identify the DEGs via edgeR package ^[14-15]. The DEGs were displayed in the volcano map by ggplot2 package. The overlapping genes of the 428 iron metabolism-relevant genes (IMRGs), the DEGs of TCGA and the DEGs of sequencing data were selected by Venn tool. These overlapping genes were DE-IMRGs.

2.3 The correlation between DEGs and CpG site methylation

The CpG site methylation levels of DE-IMRGs were extracted from the TCGA, and the relationship were estimated between the DE-IMRGs and CpG site methylation levels via psych^[16]in R (|cor| > 0.25).

2.4 Construction and evaluation of a IMRGs signature

The cox regression analysis were run to select the signature genes relying on p < 0.05. After that, the 345 cancer cases with complete survival data in the TCGA database were classified into two cohorts, namely training cohort (242) and validation cohort (103), according to 7:3. The formula for calculation of risk score was ‘h0(t)*exp(β1X1+β2X2+…+βnXn)’ via survival in R. The cases were separated into low- and high-risk group based on the median of risk-score. Besides, the Kaplan-Meier survival curve as well as the ROC curve were produced to evaluate the signature, and were drawn via survival and the survivalROC in R software.

2.5 Independent prognostic analysis of IMRGs signature

All of the clinicopathological factors of 345 cancers with complete clinical information were subjected to the model for COX independent prognostic analysis to assess the independent clinical prognostic factors. Afterward, the nomogram was built for GC patients to predict the survival of via the R package rms.

2.6 Correlation analysis between the IMRGs signature and clinical characteristics

The clinical information of GC data in the TCGA database, as well as the classification of low- and high-risk groups and signature genes were applied to establish a heatmap of expression using the R packages pheatmap and magrittr.

2.7 Immune infiltration analysis

The ssGSEA algorithm and Wilcoxon test were applied to estimate the infiltration profile of immune cell and the differences in immune infiltration between two risk subgroups. Meanwhile, the relation between the 24 immune cells and signature genes were investigated based on the ssGSEA algorithm.

2.8 Collaborative gene analysis of signature

In order to study the genes that cooperate with the signature genes, R package psych ^[17] was run to count the pearson correlation between the three signature genes with all genes, and then performed the correlation analysis according to |Pearson coefficient value| > 0.6 with FDR < 0.05 threshold.

The function annotation of different collaborative genes were analyzed via the ClusterProfiler^[18]. Additionally, enrichment analysis was performed separately for each collaborative gene based on a significance threshold p < 0.05 and visualized by ggplot2 in R.

2.9 qRT-PCR

The total RNA of the tumor cases (10) and the normal cases (10) were extracted using TRIzol Reagent (Invitrogen, CA, USA). The sweScript RT I First strand cDNA SynthesisAll-in-OneTM First-Strand cDNA Synthesis Kit (Servicebio, WuHan, CHN) was used for reverse transcription. The primer sequences were showed in the Table S1. Finally, the relative abundance of signature genes was detected by the 2xUniversal Blue SYBR Green qPCR Master Mix (Servicebio, WuHan, CHN) and standardized with the 2^−ΔΔct method.

3.1 Identification of DEGs

In the sequencing data, 11481 DEGs were screened from the tumor and normal cases (Figure 1A).In TCGA database, 5062 DEGs were screened from the tumor and normal cases (Figure 1B). Finally, 58 DE-IMRGs were obtained from 5062 DEGs of TCGA, 11481 DEGs of sequencing data and 428 IMRGs (Figure 1C).

3.2 The linkage between DEGs and methylation of CpG site

10 DEGs were obtained whose expression were associated with CpG sites to explore the potential regulation of DNA methylation on gene expression. These 10 DEGs were SLC11A1 (Exp-logFC = 1.603420664, cor = 0.5011, p = 0.0368268), FANCG (Exp-logFC = 0.669374024, cor = -0.3843059, p = 2.87E-13), CCND1 (Exp-logFC = 0.578168838, cor = -0.3466985, p = 6.35E-11), ALKBH2 (Exp-logFC = 0.705188653, cor = -0.3018947, p = 1.65E-08), CDK5RAP1 (Exp-logFC = 0.588938787, cor = -0.2978591, p = 2.60E-08), CDO1 (Exp-logFC = -1.393788957, cor = -0.2903042, p = 6.00E-08), HMBS (Exp-logFC = 0.889997626, cor = -0.2693556, p = 5.39E-07), NTHL1 (Exp-logFC = 0.755057612, cor = -0.2686109, p = 5.81E-07), FXN (Exp-logFC = 0.698112433, cor = -0.2547962, p = 2.23E-06) and KIF23 (Exp-logFC = 1.595989666, cor = -0.2520228, p = 2.90E-06) (Table 1).

Table 1. DEG-CpGs
Symbol	logFC-exp	cor	p-value
SLC11A1	1.603420664	0.5011	0.036827
FANCG	0.669374024	-0.3843059	2.87E-13
CCND1	0.578168838	-0.3466985	6.35E-11
ALKBH2	0.705188653	-0.3018947	1.65E-08
CDK5RAP1	0.588938787	-0.2978591	2.6E-08
CDO1	-1.39378896	-0.2903042	6E-08
HMBS	0.889997626	-0.2693556	5.39E-07
NTHL1	0.755057612	-0.2686109	5.81E-07
FXN	0.698112433	-0.2547962	2.23E-06
KIF23	1.595989666	-0.2520228	2.9E-06
PLOD3	0.856492524	-0.2372911	1.1E-05
GLRX3	0.51508111	-0.2336994	1.51E-05
RAD51	1.522265757	-0.2295502	2.15E-05
ATP6V1C1	0.516632482	-0.2265069	2.78E-05
PUS1	0.939137431	-0.2249831	3.15E-05
FANCI	1.382012951	-0.2147057	7.27E-05
XRCC2	1.635300181	-0.2004757	0.000217
P4HA3	1.973571416	0.1896872	0.000473
ATP6V1C2	1.244956964	-0.183886	0.000707
CYP27B1	1.336683143	-0.183343	0.000733
UBE2T	1.532447267	-0.1635036	0.002646
PRIM2	0.761406132	-0.1614566	0.002997
PALB2	0.654905467	-0.1583963	0.003602
HYAL2	0.650940956	-0.1528398	0.00499
POLA1	0.703782535	-0.1504755	0.005714
ALOX12B	1.390150295	0.1495621	0.006019
CYP4B1	-1.92512197	-0.1439937	0.008208
SCARA5	-1.92784436	0.1277368	0.01916
CCNB1	1.735808613	-0.1267312	0.02014
REP15	-2.2991481	-0.1202464	0.02753
RTEL1	0.967633929	-0.1037375	0.05749
ABCE1	0.893800372	0.1032949	0.05857
POLD1	0.879078201	-0.0991999	0.06936
TMPRSS6	-1.56762532	0.09883517	0.0704
RFWD3	0.760886638	-0.09304779	0.08858
FANCB	1.564025495	-0.09022061	0.09874
FANCA	1.248125352	-0.08374725	0.1255
PLOD1	0.844171575	-0.07410696	0.1754
COL7A1	1.812713075	-0.06719322	0.2193
DNA2	0.987347716	0.06629113	0.2255
FANCM	0.640403824	0.06592408	0.2281
PPEF1	1.314169478	0.06155626	0.2605
DOHH	0.680727916	-0.05825401	0.287
MMP1	1.558222571	-0.05223964	0.3398
CYP26B1	1.312969294	-0.04155381	0.4477
PPAT	1.01160099	0.03998513	0.4651
BRIP1	0.876330825	-0.03070913	0.5748
FANCE	0.816669612	-0.02599784	0.6349
YARS2	0.514504526	-0.00627551	0.9088
BRCA1	1.050006084	0.000138788	0.998

3.3 Construction and evaluation of a IMRGs signature

The 58 DE-IMRGs of the TCGA database were imported to univariate COX regression and 5 prognosis related genes (P4HA3, DOHH, POLD1, MMP1 and FANCE) were obtained (Figure 2A). Afterward, these 5 signature genes were evaluated as signature genes for multivariate Cox regression analysis. Finally, DOHH (coef = -0.37939, HR = 0.6843, p = 0.05429), P4HA3 (coef = 0.30383, HR = 1.3550, p = 0.08584) and MMP1 (coef = 0.07982, HR = 1.0831, p = 0.10920) were selected as signature genes (Figure 2B).

The patients in training cohort were classified into low- and high-risk groups (Figure 3A). In addition, the signature genes of P4HA3 and MMP1 were positively linked to risk-score, but the DOHH was negatively associated with risk-score (Figure 3B). The Kaplan-Meier curve showed that patients with higher risk had a poorer prognosis with p = 0.0028 (Figure 3C). In details, the AUCs for OS (1-, 2-, 3-, 4- and 5-years) were 0.617, 0.604, 0.629, 0.653 and 0.711 (Figure 3D).

Likewise, the GC patients of validation cohort were segmented into low- and high-risk groups (Figure 4A). The correlation of risk-score and signature genes (Figure 4B), and the result of Kaplan-Meier curve were similar with training cohort (Figure 4C). Moreover, the AUCs for 1-year was 0.675; 2-years was 0.649; 3-years was 0.609; 4-years was 0.602 and 5-years was 0.602 (Figure 4D).

3.4 Independent prognostic analysis of IMRGs signature

The 345 patients of TCGA with complete information including treatment type, risk-score, ajcc pathologic stage, gender, ajcc pathologic m, age at index and ajcc pathologic t were included for univariate Cox regression analysis. As shown in Figure 5A, treatment type and risk-score were related with risk model. Moreover, treatment type and risk-score were selected as independent prognostic factors by multivariate Cox regression analysis (Figure 5B). Next, treatment type and risk-score were included to establish a nomogram and the C-index of nomogram model was 0.596096 (Figure 5C). Furthermore, the calibration effect of 1- and 3-years in calibration curve were performed well (Figure 5D). Decision curve analysis (DCA) shows that the nomogram model achieves better net benefit than 1-year OS rate (Figure 5E).

3.5 Correlation analysis between the IMRGs signature and clinical characteristics

The expression of the P4HA3 and MMP1 were high in GC patients of high-risk group. However, the abundance of the DOHH was high in GC of low-risk group (Figure 6A). Moreover, the expression of MMP1 was higher in GC patients at age >= 60 (Figure 6B).

3.6 Immune infiltration analysis

For studying the imparity of immune infiltration in patients with differing risks, the spearman correlations for 24 immune cells were calculated. The 11 immune cells were different between the two risk GC subgroups (Figure 7A). The relationship between the signature genes and 24 immune cells and signature genes shown that P4HA3 was significantly associated with 17 immune cells; MMP1 was significantly associated with 12 immune cells; DOHH was significantly related to 18 immune cells (Figure 7B).

3.7 Collaborative gene analysis of signature genes

To investigate genes that act synergistically with signature genes, the pearson correlations were calculated between pairwise genes (Figure 8A). The P4HA3 were associated with 49 genes, such as ADAMTS2, ANTXR1 and APCDD1L (Figure 8A). The DOHH were associated with 33 genes, such as ADAMTS2, ANTXR1 and APCDD1L (Figure 8A). The MMP1 were associated with 2 genes including IL24 and MMP3 (Figure 8A). The GO and KEGG results of 84 collaborative genes were shown in the Figure 8B and Figure 8C. The top 8 GO terms were collagen-containing extracellular matrix and extracellular matrix organization etc. The KEGG pathways of enrichment were‘Protein digestion and absorption’ as well as ‘AGE-RAGE signaling pathway in diabetic complications’.

3.8 qRT-PCR

The expression of DOHH, P4HA3 and MMP1 were detected via qRT-PCR. The abundance of these genes were higher in tumor cases than normal cases (Figure 9).

People with GC have a high-level morbidity and mortality^[19], and prognostic biomarkers are pivotal for GC patients to improve the awful clinical consequences. Biomarkers related to GC diagnosis and prognosis are urgently needed to be uncovered.

Iron is required for cell proliferation and growth, as well as promote the formatio of toxic-free radical species. Besides, iron metabolism-relevant pathways, containing uptake-export, storage, and regulation processes, may be aberrantly mediated during the course of cancer progression^[20]. For instance, transferrin receptor 1 (TFR1), which is complicated in the regulation of cell growth as well as iron uptake, is aberrantly expressed in tumors and is intimately linked to tumour proliferation as well as metastasis^[21-23]. This has led to TFR1 being an ideal target for cancer preventionand cure^[24]. Prior investigations have shown that iron metabolism probably contribute to a number of cancers, namely lung cancer^[25], leukemia^[26], prostate cancer^[27], and kidney cancer^[28]. Nevertheless, present researches has centered on the involvement of iron metabolism in cancer progression and management, with hardly ever discussing the value of iron metabolism-related genes in cancer prognosis foretelling. Thus, the prognostic signature of iron metabolism in GC required to be adequately probed. In the present study, we found that IMRGs are highly associated with immune cells in GC. In addition, three prognostic features related to iron metabolism were constructed, which could effectively distinguish patients into the high- and low-risk groups, and had a better prognostic effect in GC. As a HEAT-repeat protein, DOHH has eight tandem helical hairpins on a symmetric dyad. DOHH encompasses two possible iron coordination sites (one on each dyad) which comprised of two rigorous conserved His-Glu motifs, and the activity of DOHH activity could be recovered nothing but by the appending of Fe²⁺ to the apoenzyme^[29]. DOHH catalyzes the last process of maturation of eIF5A, an momentous protein in proliferation of eukaryotic cell ^[30-33]. The metal chelating compounds could efficiently suppress deoxyhypusine hydroxylation in eIF5A, and arrest the progression of cell cycle in mammalian cells, including human cancer cells and HUVEC cells, at the boundary of G1/S^[34-35]. P4HA3 is expressed at very low levels in normal adults and fetal tissues^[36]. A recent research relying on the bioinformatic analysis and TCGA database found that up-regulation of P4HA3 was highly linked to genes responding to ECM generation in breast cancer, and higher expression of P4HA3 is relevant to worse prognosis^[37]. Previous studies have shown P4HA3 is significantly up-regulated in GC, and up-regulation of P4HA3 is epigenetically activated by Slug, which is correlated with GC metastasis and poor survival^[38]. MMP-1 have been authenticated to have agonist activity against PAR1, which is expressed in most human tissues, containing the majority of cell types in the blood vessel wall, platelets, and inflammatory cells, and is tought to be the primary enzyme responsible for collagen degradation^[39-43]. Although basal expression of MMP-1 is widespread, some disease states lead to further increasing expression of MMP-1, a result that is usually linked to adverse outcomes. Combined with these studies, the prognosis prediction of GC was composed of three iron metabolism related genes, including DOHH, P4HA3 and MMP1. The external verification set further proved the good performance of three gene signature in predicting GC prognosis Jianming Wei et al found that LC22A17 associated with poor overall survival in GC^[44]. Yuehong Cui et al found that NOTCH1 and NOTCH3 associated with poor overall survival and low expression of NOTCH2 associated with poor overall survival in GC^[45]. In the current research, we postulated that the iron metabolism-relevant signature for OS can adequately foretell the clinical consequences of GC patients.

The concept of immunity promoting or suppressing the tumours is widely accepted, and one of the most influential anticancerous therapies that have been exploited in these years is the therapy targeting immune checkpoints. And the development of tumours is not only regulated by the intrinsic variations in cancerous cells but also dependent on the lymphocyte infiltration and activation^[46]. Our results uncovered that the immune statue was notably disparate between the two risk GC patients, containing the Macrophages, dendritic cells (DC), Eosinophis, Immature dendritic cells (DCi), Neutrophils, Natural killer cells (NK), Mast cells, Plasmacytoid dendritic cells (DCp), Tem, Gamma delta T cells (γδT) and Thi cells. Huang XM et al found that ICOS(+) Tregs associated with poor overall survival in GC and pDCs play a underlying role in recruiting ICOS(+) Tregs^[47]. Sammarco G et al found that mast cells exerted a cancer promotion role in GC through the release of angiogenic and lymphangiogenic factors ^[48]. Eosinophils have either anti-tumor impacts or stimulate the development of tumour by secreting various cytokines and factors containing eosinophil-derived neurotoxin, peroxidase, eosinophil cationic protein, and major basic protein^[49]. Macrophages was comprised of two main types, namely M1 and M2. Macrophage M2 exert an momentous role in tumor progression, facilitating pro-angiogenic and immunosuppressive signal in the tumor while M1 macrophage infiltration may be linked to a favorable survival rate^[50]. People suffering from GC displayed a notably higher neutrophil infiltration in GC tumuor tissues. These tumour-infiltrating neutrophils revealed a phenotype of activated CD54⁺ and expressed high level programmed death-ligand 1 (PD-L1), which was an immunosuppressive molecule and relevant to disease progression and lowered GC patient survival^[51]. As vital lymphocytes in innate immunity, NK cells exert important impact in restraining GC initiation, progression, and metastases, and can improve NK cells’ killing activity toward GC. Gene therapy have been uncovered to directly or indirectly activate NK cells^[52]. The memory T cell (Tm) comprises of two populations, namely the effector memory T cells (Tem) and central memory T cells (Tcm) , and the ratio of CD4(+)/CD8(+) Tem were notably increased in GC than healthy controls^[53]. In addition, the prognostic genes (DOHH, P4HA3 and MMP1) have remarkable linkages with immune cell, suggesting the complexity between iron metabolism and immunity. Moreover, GO and KEGG analyses performed using these 84 collaborative genes suggested that these 84 collaborative genes were primarily engaged in ‘AGE-RAGE signaling pathway in diabetic complications’ and ‘protein digestion and absorption’. In fact, previous studies revealed that ‘AGE-RAGE signaling pathway in diabetic complication constitute the main mechanisms of vascular oxidative stress and enable the activation of NADPH oxidase (Nox) and NF-κB, thus incepting a vicious cycle of oxidative stress and inflammation^[54-59]. Previous studies have demonstrated that the receptor activator of NF-κB ligand signaling pathway can promote the metastasis of tumor cells.^[60-61] The ‘protein digestion and absorption’ pathway had been found to be linked to pancreatic neuroendocrine tumours as well as breast cancer^[62-63]. Expression of DOHH, P4HA3 and MMP1 was assessed with qRT-PCR in tumor cases and the normal cases. The results demonstrated that the abundance of DOHH, P4HA3 and MMP1 were higher in tumor cases than normal cases. The experiment verified the aforementioned conclusion.

All in all, this study manifested that the IMRGs (DOHH, P4HA3 and MMP1) presented good efficiency for predicting GC prognosis, implying that this gene signature relevant to iron metabolism was a promising biomarker in foretelling the prognosis of GC, which would provide new idea on the digging of underlying predictive biomarkers for GC patients. Nevertheless, this study had certain limitations. For instance, public data may have certain boundedness when we analyze the prognostic performance of gene signatures. Additional data, including primary data from patients with GC, are required to affirm the predictive effect of these gene signature.

GC Gastric cancer

IMRGs iron metabolism-relevant genes

DE-IMRGs differentially expressed IMRGs

ROC receiver operating characteristic

ssGSEA single-sample GSEA

AUCs area under the ROC curves

OS overall survival

MSigDB Molecular Signatures Database

DEGs differentially expressed genes

TFR1 transferrin receptor 1

TCGA The Cancer Genome Atlas

DC dendritic cells

DCi Immature dendritic cells

NK Natural killer cells

DCp Plasmacytoid dendritic cells

ΓδT Gamma delta T cells

PD-L1 programmed death-ligand 1

Tm memory T cell

Tem effector memory T cells

Tcm central memory T cells

Nox NADPH oxidase

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of The First Affiliated Hospital of Kunming Medical University. Written informed consent was obtained from all participantsfor the use of tissue samples and clinical data for research purpose.

Consent for publication

Not applicable.

Availability of data and materials

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors’ contributions

Hongbin Zhang conceived the idea for the study. All authors approved the manuscript and provided relevant suggestions.

Acknowledgements

We acknowledge the support from Department of Oncology First Affiliated Hospital of Kunming Medical University. We are grateful to Professor Wen Zhengqi and his team because they provided the data of cases from the General Surgery Department. We must give our sincere thanks to the GC patients who took part in this research.

Authors' information

¹ Kunming Medical University, Kunming, Yunnan, CHINA. ²Department of Oncology First Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, CHINA. ³Department of Pediatric Surgery First Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, CHINA

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): 209‐249. DOI:10.3322/caac.21660.
Oh SC, Sohn BH, Cheong JH, et al. Clinical and genomic landscape of gastric cancer with a mesenchymal phenotype. Nat Commun. 2018;9(1):1777. doi:10.1038/s41467-018-04179-85.
Allemani C, Weir HK, Carreira H, et al. Global surveillance of cancer survival 1995–2009: analysis of individual data for 25,676,887 patients from 279 population-based registries in 67 countries (CONCORD-2). Lancet. 2015;385(9972):977–1010. doi:10.1016/ s0140-6736(14)62038-9.
SHAFABAKHSH R，YOUSEFI B，ASEMI Z，et al. Chitosan：Acompound for drug delivery system in gastric cancer--A review［J］. Carbohydr Polym，2020，242：116403.
DAN J，WANG J，WANG Y，et al. LncRNA-MEG3 inhibits pro-liferation and metastasis by regulating miRNA-21 in gastric cancer［J］. Biomed Pharmacother，2018，99：931-938.
CATALANO V，LABIANCA R，BERETTA GD，et al. Gastric cancer［J］. Crit Rev Oncol Hematol，2005，54（3）：209-241.
ASPLUND J，KAUPPILA JH，MATTSSON F，et al. Survival trends in gastric adenocarcinoma：A population-based study in Sweden［J］. Ann Surg Oncol，2018，25（9）：2693-2702.
LIN JX，LIN JP，XIE JW，et al. Prognostic value and association of sarcopenia and systemic inflammation for patients with gastric cancer following radical gastrectomy［J］. Oncologist，2019，24（11）：e1091-e1101.
VAN CUTSEM E ，SAGAERT X，TOPAL B，et al. Gastric can⁃cer［J］. Lancet，2016，388（10060）：2654-2664.
PinnixZK, MillerLD, WangW, et al. Ferroportin and iron regulation in breast cancer progression and prognosis[J]. Sci Transl Med,2010, 2(43): 43ra56.
LouandreC, EzzoukhryZ, GodinC, et al. Iron-dependent cell death of hepatocellular carcinoma cells exposed to sorafenib[J]. Int J Cancer, 2013, 133(7): 1732-1742.
GreeneCJ, AttwoodK, SharmaNJ, et al. Transferrin receptor 1 upregulation in primary tumor and downregulation in benign kidney is associated with progression and mortality in renal cell carcinoma patients[J]. Oncotarget, 2017, 8(63): 107052-107075.
TesfayL, ClausenKA, KimJW, et al. Hepcidin regulation in prostate and its disruption in prostate cancer[J]. Cancer Res, 2015, 75(11): 2254-2263.
Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010 Jan 1;26(1):139-40. doi: 10.1093/bioinformatics/btp616. Epub 2009 Nov 11.
McCarthy DJ, Chen Y, Smyth GK. Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Res. 2012 May;40(10):4288-97. doi: 10.1093/nar/gks042. Epub 2012 Jan 28.
Thornley B, Adams CE, Awad G. Chlorpromazine versus placebo for schizophrenia. Cochrane Database Syst Rev. 2000;(2):CD000284. doi: 10.1002/14651858.CD000284. Update in: Cochrane Database Syst Rev. 2003;(2):CD000284.
Thornley B, Adams CE, Awad G. Chlorpromazine versus placebo for schizophrenia. Cochrane Database Syst Rev. 2000;(2):CD000284. doi: 10.1002/14651858.CD000284. Update in: Cochrane Database Syst Rev. 2003;(2):CD000284.
Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, Fu X, Liu S, Bo X, Yu G. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021 Jul 1;2(3):100141. doi: 10.1016/j.xinn.2021.100141. PMID: 34557778; PMCID: PMC8454663.
Allemani C, Weir HK, Carreira H, et al. Global surveillance of cancer survival 1995–2009: analysis of individual data for 25,676,887 patients from 279 population-based registries in 67 countries (CONCORD-2). Lancet. 2015;385(9972):977–1010. doi:10.1016/ s0140-6736(14)62038-9
Forciniti S, Greco L, Grizzi F, Malesci A, Laghi L. Iron Metabolism in Cancer Progression. Int J Mol Sci (2020) 21(6):2257. doi: 10.3390/ijms21062257
Adachi M, Kai K, Yamaji K, Ide T, Noshiro H, Kawaguchi A, et al. Transferrin receptor 1 overexpression is associated with tumour de-differentiation and acts as a potential prognostic indicator of hepatocellular carcinoma. Histopathology (2019) 75:63–73. doi: 10.1111/his.13847
Cui C, Cheng X, Yan L, Ding H, Guan X, Zhang W, et al. Downregulation of TfR1 promotes progression of colorectal cancer via the JAK/STAT pathway. Cancer Manage Res (2019) 11:6323–41. doi: 10.2147/cmar.S198911
Jeong SM, Hwang S, Seong RH. Transferrin receptor regulates pancreatic cancer growth by modulating mitochondrial respiration and ROS generation. Biochem Biophys Res Commun (2016) 471:373–9. doi: 10.1016/j.bbrc.2016.02.023
Clarke SL, Thompson LR, Dandekar E, Srinivasan A, Montgomery MR. Distinct TP53 Mutation Subtypes Differentially Influence Cellular Iron Metabolism. Nutrients. 2019;11(9):9. doi:10.3390/nu11092144.
Sukiennicki GM, Marciniak W, Muszyńska M, et al. Iron levels, genes involved in iron metabolism and antioxidative processes and lung cancer incidence. PLoS One. 2019;14(1):e0208610. doi:10.1371/journal.pone.0208610
Tsuma-Kaneko M, Sawanobori M, Kawakami S, et al. Iron removal enhances vitamin C-induced apoptosis and growth inhibition of K-562 leukemic cells. Sci Rep. 2018;8(1):17377. doi:10.1038/ s41598-018-35730-8.
Iron VD. Metabolism in Prostate Cancer; From Basic Science to New Therapeutic Strategies. Front Oncol. 2018;8:547. doi:10.3389/ fonc.2018.00547.
Linehan WM, Schmidt LS, Crooks DR, et al. The Metabolic Basis of Kidney Cancer. Cancer Discov. 2019;9(8):1006–1021. doi:10.1158/ 2159-8290.Cd-18-1354.
Kim YS, Kang KR, Wolff EC, Bell JK, McPhie P, Park MH. Deoxyhypusine hydroxylase is a Fe(II)-dependent, HEAT-repeat enzyme. Identification of amino acid residues critical for Fe(II) binding and catalysis [corrected]. J Biol Chem. 2006 May 12;281(19):13217-13225.
Park MH. J Biochem (Japan) 2006;139:161–169. [PubMed: 16452303].
Chen KY, Liu AY. Biol Signals 1997;6:105–109. [PubMed: 9285092].
Caraglia M, Marra M, Giuberti G, D'Alessandro AM, Baldi A, Tassone P, Venuta S, Tagliaferri P, Abbruzzese A. J Biochem (Tokyo) 2003;133:757–765. [PubMed: 12869532].
Hanauske-Abel HM, Park MH, Hanauske AR, Popowicz AM, Lalande M, Folk JE. Biochim Biophys Acta 1994;1221:115–124. [PubMed: 8148388].
Hanauske-Abel HM, Park MH, Hanauske AR, Popowicz AM, Lalande M, Folk JE. Biochim Biophys Acta 1994;1221:115–124. [PubMed: 8148388]
Clement PM, Hanauske-Abel HM, Wolff EC, Kleinman HK, Park MH. Int J Cancer 2002;100:491– 498. [PubMed: 12115536]
Kukkola L, Hieta R, Kivirikko KI, Myllyharju J. Identification and characterization of a third human, rat, and mouse collagen prolyl 4-hydroxylase isoenzyme. J Biol Chem. 2003;278(48):47685–47693.
Winslow S, Lindquist KE, Edsjo A, Larsson C. The expression pattern of matrix-producing tumor stroma is of prognostic importance in breast cancer. BMC Cancer. 2016;16(1):841.
Song H, Liu L, Song Z, Ren Y, Li C, Huo J. P4HA3 is Epigenetically Activated by Slug in Gastric Cancer and its Deregulation is Associated With Enhanced Metastasis and Poor Survival. Technol Cancer Res Treat. 2018 Jan 1;17:1533033818796485.
Galt SW, Lindemann S, Allen L, et al. Outside-in signals delivered by matrix metalloproteinase-1 regulate platelet function. Circ
Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest. 1994;94(6):2493–2503. Res. 2002;90(10):1093–1099.
[41] Brinckerhoff CE, Rutter JL, Benbow U. Interstitial collagenases as markers of tumor progression. Clin Cancer Res. 2000;6(12):4823–4830.
De Garavilla L, Vergnolle N, Young SH. et al. Agonists of proteinase-activated receptor 1 induce plasma extravasation by a neurogenic mechanism. Br J Pharmacol. 2001;133:975–87.
Giambernardi TA, Grant GM, Taylor GP, et al. Overview of matrix metalloproteinase expression in cultured human cells. Matrix Biol. 1998;16(8):483–496.
Wei J, Gao X, Qin Y, Liu T, Kang Y. An Iron Metabolism-Related SLC22A17 for the Prognostic Value of Gastric Cancer. Onco Targets Ther. 2020 Dec 14;13:12763-12775. doi: 10.2147/OTT.S287811. PMID: 33363382; PMCID: PMC7751842.
Cui Y, Li Q, Li W, Wang Y, Lv F, Shi X, Tang Z, Shen Z, Hou Y, Zhang H, Mao B, Liu T. NOTCH3 is a Prognostic Factor and Is Correlated With Immune Tolerance in Gastric Cancer. Front Oncol. 2021 Jan 5;10:574937. doi: 10.3389/fonc.2020.574937. PMID: 33479597; PMCID: PMC7814877.
Wang M, Zhao J, Zhang L, Wei F, Lian Y, Wu Y, et al. Role of tumor microenvironment in tumorigenesis. J Cancer (2017) 5:761–73. doi: 10.7150/jca.17648.
Huang XM, Liu XS, Lin XK, Yu H, Sun JY, Liu XK, Chen C, Jin HL, Zhang GE, Shi XX, Zhang Q, Yu JR. Role of plasmacytoid dendritic cells and inducible costimulator-positive regulatory T cells in the immunosuppression microenvironment of gastric cancer. Cancer Sci. 2014 Feb;105(2):150-8. doi: 10.1111/cas.12327. Epub 2014 Jan 4. PMID: 24261990; PMCID: PMC4317822.
Sammarco G, Varricchi G, Ferraro V, Ammendola M, De Fazio M, Altomare DF, Luposella M, Maltese L, Currò G, Marone G, Ranieri G, Memeo R. Mast Cells, Angiogenesis and Lymphangiogenesis in Human Gastric Cancer. Int J Mol Sci. 2019 Apr 29;20(9):2106. doi: 10.3390/ijms20092106. PMID: 31035644; PMCID: PMC6540185.
Sakkal S, Miller S, Apostolopoulos V, Nurgali K. Eosinophils in Cancer: Favourable or Unfavourable? Curr Med Chem. 2016;23(7):650-66. doi: 10.2174/0929867323666160119094313. PMID: 26785997.
Wang XL, Jiang JT, Wu CP. Prognostic significance of tumor-associated macrophage infiltration in gastric cancer: a meta-analysis. Genet Mol Res. 2016 Dec 2;15(4). doi: 10.4238/gmr15049040. PMID: 27966749.
Wang TT, Zhao YL, Peng LS, Chen N, Chen W, Lv YP, Mao FY, Zhang JY, Cheng P, Teng YS, Fu XL, Yu PW, Guo G, Luo P, Zhuang Y, Zou QM. Tumour-activated neutrophils in gastric cancer foster immune suppression and disease progression through GM-CSF-PD-L1 pathway. Gut. 2017 Nov;66(11):1900-1911. doi: 10.1136/gutjnl-2016-313075. Epub 2017 Mar 8. PMID: 28274999; PMCID: PMC5739867.
Du Y, Wei Y. Therapeutic Potential of Natural Killer Cells in Gastric Cancer. Front Immunol. 2019 Jan 21;9:3095. doi: 10.3389/fimmu.2018.03095. PMID: 30719024; PMCID: PMC6348255.
Zhang R, Li F, Li H, Yu J, Ren X. The clinical significance of memory T cells and its subsets in gastric cancer. Clin Transl Oncol. 2014 Mar;16(3):257-65. doi: 10.1007/s12094-013-1066-5. Epub 2013 Jun 21. PMID: 23793812.
Giacco F., Brownlee M. Oxidative stress and diabetic complications.Circulation Research. 2010;107(9):1058–1070.doi: 10.1161/CIRCRESAHA.110.223545.
Goldin A., Beckman J. A., Schmidt A. M., Creager M. A. Advanced glycation end products: sparking the development of diabetic vascular injury. Circulation. 2006;114(6):597–605. doi: 10.1161/CIRCULATIONAHA.106.621854.
Yan S. F., Ramasamy R., Schmidt A. M. The RAGE axis: a fundamental mechanism signaling danger to the vulnerable vasculature. Circulation Research. 2010;106(5):842–853. doi: 10.1161/CIRCRESAHA.109.212217.
Bierhaus A., Chevion S., Chevion M., et al. Advanced glycation end product-induced activation of NF-κB is suppressed by α-lipoic acid in cultured endothelial cells. Diabetes. 1997;46(9):1481–1490. doi: 10.2337/diab.46.9.1481.
Wautier M.-P., Chappey O., Corda S., Stern D. M., Schmidt A. M., Wautier J.-L. Activation of NADPH oxidase by AGE links oxidant stress to altered gene expression via RAGE. American Journal of Physiology-Endocrinology and Metabolism. 2001;280(5):E685–E694. doi: 10.1152/ajpendo.2001.280.5.E685.
Arai M., Nihonmatsu-Kikuchi N., Itokawa M., Rabbani N., Thornalley P. J. Measurement of glyoxalase activities. Biochemical Society Transactions. 2014;42(2):491–494. doi: 10.1042/BST20140010.
Lazennec G, Richmond A. Chemokines and chemokine receptors: new insights into cancer-related inflammation. Trends Mol Med. 2010;16(3):133-144. doi: 10.1016/j.molmed.2010.01.003.
Chen Q, Sun W, Liao Y, et al.. Monocyte chemotactic protein-1 promotes the proliferation and invasion of osteosarcoma cells and upregulates the expression of AKT. Mol Med Rep. 2015;12(1):219-225. doi: 10.3892/mmr.2015.3375.
Akkiprik M, Peker I, Özmen T, Amuran GG, Güllüoğlu BM, Kaya H, Özer A. Identification of differentially expressed IGFBP5-related genes in breast cancer tumor tissues using cDNA microarray experiments. Genes (Basel) 2015;6:1201–1214. doi: 10.3390/genes6041201.
Wang DD, Liu ZW, Han MM, Zhu ZM, Tu YL, Dou CQ, Jin X, Cai SW, Du N. Microarray based analysis of gene expression patterns in pancreatic neuroendocrine tumors. Eur Rev Med Pharmacol Sci. 2015;19:3367–3374.

No competing interests reported.

TableS1.Primerinformation.docx

Download PDF

Journal Publication

published 13 Nov, 2023

Read the published version in BMC Cancer →

Editorial decision: Major revision
18 Aug, 2023
Reviews received at journal
03 Aug, 2023
Reviewers agreed at journal
24 Jul, 2023
Reviewers invited by journal
24 Jul, 2023
Editor assigned by journal
24 Jul, 2023
Editor invited by journal
23 Jul, 2023
Submission checks completed at journal
23 Jul, 2023
First submitted to journal
29 Jun, 2023

You are reading this latest preprint version

Construction of Iron Metabolism-related Prognostic Features of Gastric Cancer Based on RNA Sequencing and TCGA Database

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials and methods

3. Results

4. Discussion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1