Effect of Differential Expression of Genes Induced by Radiation Therapy In Cancer-associated broblasts Cell on Patients' Prognosis

To investigate the effect of radiation therapy on differential expression of genes in tumor-associated broblasts and prognosis of patients. Methods: The tumor-associated broblast gene expression prole data chip GSE37318 after radiotherapy treatment was retrieved from the GEO database, and the differentially expressed genes were screened using the limma R software package; GO and KEGG pathway enrichment analysis was performed using the DAVID tool; Protein interaction networks was built by String and Cytoscape software and core genes were obtained; GEPIA was used for prognostic value analysis; Immunohistochemistry was used to detect the expression of the top 5 hub genes in tumor tissues of patients in the radiotherapy and non-radiotherapy groups. Results: 144 genes were up-regulated and 54 genes were down-regulated, which were mainly enriched in functional pathways such as cell stress, DNA damage, cell cycle, aging, apoptosis, oxidative stress, and p53 signaling pathway. The protein interaction network was constructed and the top 20 hub genes were obtained. Prognostic analysis showed that: Expression of up-regulated PCNA and hub genes that were down-regulated after irradiation, such as MCM10, DLGAP5, FANCI, CENPA, CDC6, FBXO5, NCAPG, and DTL, has a negative correlation with the overall survival time of lung cancer patients (p <0.05). Immunohistochemical results showed that PCNA gene expression was up-regulated in patients with radiotherapy compared with patients without radiotherapy. The test results are consistent with the results of the biochemical analysis. Conclusion: Radiotherapy can induce differential expression of genes in tumor-associated broblasts, and these differentially expressed genes can be used as potential molecular markers for tumor radiotherapy effect and patient prognosis.

(CAF) are the main members of the tumor microenvironment. It is distributed more in the stromal tissue of most tumor types and is a key factor in tumor progression and metastasis. [5][6].
The toxic and side effects of radiation therapy on tumor microenvironment are currently less studied, and changes in tumor in ltrating broblasts in patients after radiotherapy are not clear. In view of the wide application of radiotherapy in tumor treatment and the important role of CAF in the development of cancer, this study explores the effect of radiotherapy on CAF gene expression. Through in-depth analysis of microarray data of tumor in ltrating broblasts isolated from patients with lung cancer irradiated by radiation therapy in the Gene Expression Omnibus (GEO) database, differentially expressed genes and key signaling pathways in tumor in ltrating broblasts before and after radiotherapy are identi ed in order to provide a new basis for radiotherapy of lung cancer.
1 Materials And Methods

chip data download
Radiation-treated tumor in ltrating broblast gene expression data set GSE37318 was downloaded from the GEO database for screening differentially expressed genes. GPL10191 platform was used by this set of chips and 8 groups of cells were contained. Tumor in ltrating broblasts were isolated from patients with non-small cell lung cancer. The cells were subjected to a single dose of 18 Gy ionization irradiation, and the total RNA of the cells was extracted for whole-gene transcriptome analysis after 24 h.

Identi cation of differentially expressed genes
The R software limma was used to process the chip data GSE37318, and the default Benjamini-Hochberg method was used for statistical analysis. log2 FC |> 1 and adj. p < 0.05 were used as the screening criteria (FC is the multiple of gene expression difference, adj. p is the corrected p-value); the volcano map was drawn using the ggplot2 package in the R software, and the heat map was drawn using the heatmap MEV4.9.0 (Multi Experiment Viewer) package in the R software .

GO analysis and KEGG pathway analysis
The screened differentially expressed genes were introduced into the DAVID [7] online toolkit for reannotation, and the differentially expressed genes were enriched by GO and KEGG pathway analysis.

Building protein interaction networks
The STRING database [8] was used to analyze the interaction of differentially expressed genes, the protein interaction network was constructed using Cytoscape 3.5.0 software, and the core submodule network was screened and obtained through the plug-in MCODE, and the core genes were screened using the cytoHubba plug-in, and the difference is considered statistically signi cant by p < 0.05.
1.5 Veri cation of core differential genes Patients with clinical stage II and III of non-small cell lung cancer were screened out, and 20 patients each with pre-operative radiotherapy and non-radiotherapy were screened out. See Table 1 for basic information of all patients .They were divided into radiotherapy group and non-radiotherapy group. The resected tumor tissue was taken, para n specimens were prepared and sectioned. Immunohistochemical techniques were used to detect and compare the expression of the top 5 hub genes in the two groups of patients. Using the lung cancer-related data in the TCGA database, the GEPIA online tool was used to analyze the expression of the top 10 hub genes in patients with lung adenocarcinoma, lung squamous cell carcinoma, and normal people, and to compare and analyze whether there is differential expression of the hub genes in patients with lung cancer.

2.2GO and KEGG pathway enrichment analysis
The screened differentially expressed genes were imported into the DAVID database for re-annotation, and the GO (Fig. 2a) and KEGG (Fig. 2b) pathway enrichment analysis were performed. The results show that the down-regulated genes are mainly enriched in biological processes such as DNA replication, strand shift, initiation of DNA replication, and G1 / S transition of the mitotic cell cycle, cell components such as cytoplasm, nucleus, and chromosome centromere, molecular functions such as single-stranded DNA binding, DNA binding, and ATP binding, and Fanconi anemia pathway, homologous recombination and signaling pathways such as p53; up-regulated genes are mainly enriched in biological processes such as DNA damage response, p53-mediated cell cycle arrest, positive regulation of cell proliferation, and positive regulation of GTPase activity, cell components such as nucleosomes, cell junctions, and cytoplasm, molecular functions such as growth factor activity, receptor binding, and protein heterodimerization activity, and functional pathways such as p53 signaling pathways, alcohol abuse, chronic myelogenous leukemia, axon guidance, and ErbB signaling pathways.

Protein interaction networks
The differentially expressed genes were imported into the STRING database to analyze protein-protein interaction (PPI), and the results were visualized using Cytoscape 3.5.0 software to construct a PPI network, which contains 103 nodes and 376 edges as shown in Fig. 2c.

Gene enrichment module and Hub gene analysis
MODE plug-in of Cytoscape software was used to perform module analysis on the PPI network and select 2 core modules, and the rst 3 core modules were obtained according to the scoring level. Module 1 contains 46 nodes and 286 edges (Fig. 3a). Functions are mainly concentrated in mitotic cell cycle checkpoints, cell responses to amino acid stimulation, negative regulation of neuronal apoptosis, p53 signaling pathway, and FoxO signaling pathway .
The plug-in CytoHubba was used to further analyze the constructed PPI network, and the top 10 differentially expressed hub genes were screened. The inter-gene interaction network is shown in Fig. 3d. The top 10 screened genes are NCAPG, MCM10, DTL, DLGAP5, RAD51, RFC4, MKI67, CDC6, PCNA and BLM. Of these genes, only the expression of PCNA was up-regulated after irradiation, and the expression of other genes was down-regulated.
2.5 Experimental veri cation of differential genes and their correlation with lung cancer Immunohistochemical results showed that the expression levels of the ve hub genes of MCM10, DLGAP5, CDC6, NCAPG and DTL in tumor tissues of the radiotherapy group were signi cantly higher than those of the non-radiotherapy group (Fig. 4). Among the top 10 hub genes, BLM, RFC4, and PCNA genes were expressed more in lung squamous cell carcinoma than normal people, and the remaining 7 genes were expressed more in lung squamous cell carcinoma and lung adenocarcinoma than normal people ( Fig. 5 ).

Clinical stage and prognostic value analysis
In order to further explore the clinical staging and prognosis value of the selected core genes for nonsmall cell lung cancer, we used GEPIA software combined with data from TCGA to analyze the expression of the rst 10 core genes in different clinical stages and the prognosis of patients. The results found that the expression levels of all 10 genes in patients with clinical stage II and III were higher than those in clinical stage I (Fig. 6).  (Fig. 7). The expression levels of these genes decreased after radiation therapy, suggesting that irradiation can inhibit the expression of genes related to lung cancer, which is bene cial to the survival of patients. In addition, the expression level of the only core gene PCNA that were highly expressed after radiation therapy was negatively correlated with the prognosis of patients with lung cancer (HR = 1.1, p = 0.41), suggesting that irradiation can also promote tumor cell proliferation to a certain extent.

Discussion
With the continuous development of nuclear technology, medical imaging technology and computer technology, radiotherapy technology has made great progress and has become the main method of comprehensive tumor treatment. 65-75% of cancer patients clinically need radiation therapy during treatment, and about 40% of cancer-cured patients treat radiotherapy as an important part of their disease course management [10]. Radiotherapy works on tumor cells: on the one hand, it can cause damage to biological macromolecules such as proteins, DNA and RNA, and reduce the expression of related genes, promote apoptosis, necrosis, autophagy and aging, and exert the function of tumor killing; on the other hand, tumor cells will start their own repair system after radiation damage. By regulating the expression of certain genes, they respond to radiation damage, mediate tumor radiation resistance, and promote tumor recurrence and metastasis [11][12].. Analysis of these differentially expressed genes after radiation can not only help us understand the toxic and side effects of tumor radiotherapy, but also provide new biomarkers for tumor radiotherapy effects, patient prognosis, and risk assessment of recurrence and metastasis.
The tumor microenvironment refers to the area between tumor cells and adjacent normal tissues. Its composition mainly includes extracellular matrix, soluble molecules and tumor stromal cells. It plays an important role in tumor cell immune escape, tumor growth, recurrence and metastasis [13]. CAF is a major member of the tumor microenvironment and plays a key role in the progression and metastasis of various tumors such as lung cancer [14].
In view of the current background of radiotherapy, the impact of radiation therapy on tumor-supporting microenvironment is poorly understood. This study used bioinformatics methods to explore the differentially expressed genes in CAF and the key pathways mediated by them after radiotherapy, and obtained 144 up-regulated expression genes and 54 down-regulated expression genes in CAF after irradiation. The analysis results of GO and KEGG showed that the up-regulated expression genes were mainly enriched in functional pathways such as DNA damage response, p53-mediated cell cycle arrest, positive regulation of cell proliferation, p53 signaling pathway, chronic myeloid leukemia, and ErbB signaling. And down-regulated expression genes were mainly enriched in DNA replication, strand shifts, G1 / S transition of mitotic cell cycle, Fanconi anemia pathways, homologous recombination, p53 and other signaling pathways. Radiation therapy acts on the body, causing damage to biological macromolecules such as DNA. The body will initiate damage repair responses by regulating the expression of a series of genes, induce cell cycle arrest, and promote DNA repair and regulate cell proliferation [15][16]. Of course, high-dose radiation therapy will seriously damage the tissue structure of cells, inhibit the expression of many functional genes, cause cell DNA replication and repair defects, cell cycle transition disorders, and cell maturation disorders, etc. [17][18]. However, we also found that the pathways enriched by highly expressed genes are also closely related to the occurrence and development of tumors, such as the activation of the ErbB signaling pathway and the invasion and metastasis of tumors such as lung cancer [19][20]. The p53 signal pathway enriched with low-expressed genes is a classic tumor suppressor pathway [21][22]. Changes in these pathways are likely to be an important cause of tumor recurrence and metastasis after radiotherapy.
We obtained Top 20 genes from the PPI network, which are NCAPG, MCM10, DTL, DLGAP5, RAD51, CENPA, MKI67, CDC6, FANCI, FBXO5, KIF15, HELLS, HJURP, PCNA, GTSE1, CASC5, CCNE2, CHAF1B, BLM and KIFC1. These genes all play an important role in the genesis and development of tumors. NCAPG is highly expressed in patients with prostate cancer and liver cancer, and it is involved in promoting the proliferation and migration of tumor cells. Down-regulating the expression of NCAPG can inhibit the growth of tumor cells [23][24]. MCM10 is an important protein that mediates DNA replication. It is highly expressed in a variety of tumor cells and tissues and is involved in promoting tumor cell proliferation [25][26]. DTL is a homologue of E3 ubiquitin protein ligase. It is highly expressed in liver cancer tumor tissues. Down-regulating the expression of DTL can induce cell cycle arrest and senescence, and inhibit the growth and colony formation of liver cancer cells [27]. PCNA is the only gene in this Top 20 gene that has an increased expression after radiation and is a cofactor of DNA polymerase δ. The expression of PCNA in lung cancer and other tumors is signi cantly increased, promoting tumor cell proliferation, migration and invasion [28][29]. In addition, these pivot genes are related to the poor prognosis of tumor ,which is consistent with our prognostic analysis results, such as the expression of NCAPG is signi cantly negatively correlated with the survival of prostate and liver cancer [23][24]; the expression of MCM10 is signi cantly correlated with the poor prognosis of patients with breast and prostate cancer [25][26]; the expression of DTL is negatively correlated with the prognosis of breast and lung cancers [30]. These ndings suggest that down-regulated expression of tumor-related genes after irradiation treatment determines the effect of tumor radiotherapy, while up-regulated genes are likely to play an important role in subsequent tumor recurrence.

Conclusion
The results of the study indicate that the gene expression and biological pathways in CAF have changed after radiotherapy treatment, mainly involving cell stress, DNA damage, cell cycle, aging, apoptosis, oxidative stress, and matrix remodeling. Radiation therapy has both anti-tumor effects and induction of tumor-promoting effects on CAF. The speci c mechanism is worthy of further exploration through in vivo experiments. Informed consent was obtained from all subjects.

List Of Abbreviations
All experimental protocols were approved by the Ethics Committee of the Second A liated Hospital of Nanchang University Consent for publication: Not applicable.
Availability of data and material: All datas are available. Please contact us to access if it is needed.
Competing interests: There are no con icts of interest in this study.  Heatmap (left) and volcano map (right) of differential genes; red represents gene up-regulation and blue represents gene down-regulation. Figure 2 a: GO analysis result of differentially expressed genes; b: KEGG analysis result of differential genes; c: protein-protein interaction network of differential genes. Figure 2 a: GO analysis result of differentially expressed genes; b: KEGG analysis result of differential genes; c: protein-protein interaction network of differential genes.   Comparison of the expression levels of the top ve hub genes MCM10, DLGAP5, CDC6, NCAPG and DTL in tumor tissues between the radiotherapy group and the non-radiotherapy group.  The expression levels of the rst 10 hub genes in patients with lung cancer at various clinical stages