Relationship between microRNA-29c expression and clinicopathological features of gastric cancer

ZHANG Yong-fang, LI Jing-cheng, XIE Na,3, CHEN Yan, CHEN Jun-ming, HUANG You-sheng?

1.Department of Pathology, the First Affiliated Hospital of Hainan Medical University,Haikou 570100, China

2.Department of Clinical Laboratory, the First Affiliated Hospital of Hainan Medical University,Haikou 570100, China

3.Department of Oral Histopathology, Hainan Medical College,Haikou 570100, China

Keywords:

ABSTRACT Objective:To explore the relationship between the expression of miR-29c and clinicopathological characteristics of gastric cancer and predict the biological functions of target genes of miR-29c.Methods: The expression and clinical data of microRNA of gastric cancer were downloaded from The Cancer Genome Atlas (TCGA) to analyze the expression of miR-29c in gastric cancer tissues and normal gastric mucosal tissues.Paraffin specimens of gastric cancer and normal tissues were collected to make tissue microarray.The expression of miR-29c in gastric cancer tissues and normal tissues were detected by miRNAscope technique.The relationship between the expression of miR-29c and clinicopathological characteristics of gastric cancer was analyzed by chi-square test.Target genes of miR-29c were predicted by TargetScan, Pictar, miRTarbase and Starbase databases,and enrichment analysis were performed on potential target genens.Results: TCGA analysis showed that the expression of miR-29c was significantly lower in gastric cancer tissue (n=434) than normal tissues (n=41, P＜0.001).miRNAscope analysis showed that the expression of miR-29c was significantly lower in gastric cancer tissue than normal tissues(P＜0.001), and was significantly associated with TNM staging, lymph node metastasis and histological type of gastric cancer patients(P＜0.05).Database predicted 18 potential target genes that were enriched in DNA methyltransferase activity and cancer-associated miRNA signal pathways.Conclusion: miR-29c was expressed at a low level in gastric cancer tissue and associated with TNM staging,lymph node metastasis and histological type, and could become a potential prognosis marker of gastric cancer and a new target of molecular treatment.

1.Introduction

In the 2020 Global Cancer Statistics, gastric cancer ranks fifth in the global incidence of cancer and fourth in the mortality rate.At present, the numbers of the new cases and fatal cases of gastric cancer in China are above the world average[1].The development of gastric cancer involves various genetic and epigenetic changes, and exploring the molecular mechanism of the development of gastric cancer is helpful to search for biomarkers with effective therapies in clinic.According to recent research results, microRNA (miRNA) is involved in the occurrence and progression of tumors and is a key factor in cancer epigenetic changes[2], and has become a promising diagnostic and prognostic marker and potential therapeutic target[3].MiRNA are non-coding RNA that perform their biological functions at the RNA level and do not directly encode proteins.They exert their functions mainly by binding with their target genes completely or incompletely, mediating mRNA degradation or inhibiting target mRNA translation, and thus negatively regulating gene expression[4].miR-29c, miR-29a and miR-29b belong to the miR-29 family.Previous studies have shown that miR-29c is down-regulated in most human cancers, including nasopharyngeal carcinoma[5],lung cancer[6,7],gastric cancer[8,9],cervical cancer[10],and hepatocellular carcinoma[11].Yoshimasa Saito et al found that the downregulation of miR-29c is related to the progression of gastric cancer, and was more prominent in advanced gastric cancer than in gastric adenoma and early gastric cancer[12].However,there are studies show that the expression level of miR-29c has no correlation with clinicopathological parameters[13].Tae-SuHan et al.analyzed FFEP tissues of 113 cases of gastric cancer and concluded that the loss of miR-29c expression was an early event in the occurrence of gastric cancer[9].Therefore, the relationship between miR-29c expression and clinicopathological features of gastric cancer is still unclear.In different studies, different research objects and methods may result in inconsistent results.In previous studies,fresh tissues and paraffin tissues was mostly used, but fresh tissues are difficult to obtain, and the degradation rate of RNA in paraffin tissues was relatively fast.In this study, the miRNAscope technique was used to solve this problem, and its unique probe design greatly improved the sensitivity and specificity of RNA detection in paraffin tissue, while simultaneously observing tissue morphology and signal expression.

Therefore, this study firstly used TCGA database to analyze the expression of miR-29c in gastric cancer tissues and adjacent tissues,then verified the results of biogenic analysis by miRNAscope technology, and further analyzed the relationship between miR-29c expression and clinical characteristics of gastric cancer.It lays a theoretical foundation for elucidating the molecular mechanism of the occurrence and development of gastric cancer and developing new targeted drugs for gastric cancer.

2.Materials and Methods

2.1 General information

Obtaining the miRNA expression data and clinical data of gastric cancer based on The Cancer Genome Atlas (TCGA) database, and the miRNA expression data of 434 cases of gastric cancer were obtained, of which 41 cases had expression data of matched normal tissues.After excluding cases with duplicate specimens, nonprimary gastric cancer, and a history of preoperative radiotherapy and chemotherapy, a total of 372 gastric cancer specimens were included in the clinical feature correlation analysis, and totally 405 samples were included in the survival analysis removing samples with incomplete survival data.

Gastric cancer surgical specimens of the First Affiliated Hospital of Hainan Medical College from January 2020 to June 2022 were collected, including 73 primary gastric cancer tissues and 71 normal tissues, among which 71 gastric cancer tissues had paired normal tissues, and were all fixed by neutral formalin to make paraffin specimens.None of the included patients received preoperative radiotherapy or chemotherapy, and all cases have confirmed by pathological examination in our hospital.TNM staging was performed according to the criteria of the American Joint Committee on Cancer (AJCC, 8th edition).Table 1 summarizes the clinical factors of 73 patients with gastric cancer, including age, sex, tissue type, and lymph node metastasis.The use of specimens in this study was discussed and approved by the Ethics Committee of the First Affiliated Hospital of Hainan Medical College, and all gastric cancer patients participating in this study signed written informed consent.

2.2 Main Reagents

miRNAScopeTM miR-29c probe (item number: 1058391),miRNAscope?H2O2 and Protease Kit (item number: 322381),miRNAscopeTM HD Test Kit (red, item number:324510) were purchased from ACD Corporation, USA.

2.3 Bioinformatics analysis

2.3.1 Difference analysis of miR-29c

According to the grouping of gastric cancer tissue and normal gastric tissue, the “edgeR” package was used for differential expression analysis, and the significance threshold was set as the absolute value of the difference multiple 1, and the False discovery rate (FDR) ＜ 0.05.The paired rank sum test was performed on the paired gastric cancer and normal tissue.P＜ 0.05 was statistically significant.

2.3.2 Clinical characteristics and prognosis analysis of miR-29c

The optimal cutoff value of miR-29c was calculated according to the “survival” package, and the samples were divided into miR-29c high expression group and miR-29c low expression group, and the statistical differences of various clinical factors (tumor site, depth of invasion, lymph node metastasis, etc.) among the groups were analyzed.Kaplan-Meier method was used to describe the survival data, and log-rank test was used to evaluate the statistical difference in overall survival between the groups with high and low miR-29c expression.

2.3.3 Prediction and enrichment analysis of miR-29c target genes

The downstream genes of miR-29c were predicted on the online website mirtarbase[14], and miR-29c target genes supported by at least two evidences were downloaded.At the same time, TargetScan[15],Pictar[16] and Starbase[17] online software were used to predict the downstream genes that miR-29c may regulate, and the binding sites between miR-29c and these genes were found using TargetScan.Four overlapping target genes selected by online website(https://bioinformatics.psb.ugent.be/webtools/Venn/) were used as candidate miR-29c target genes.Input the candidate target genes into the online website DAVID (The Database for Annotation, Visualization and Integrated Discovery) for enrichment analysis, including GO functional annotation and KEGG pathway analysis to explore the possible biological functions of miR-29c.

2.4 Experimental methods

2.4.1 Tissue Microarray (TMA) Construction

FFPE (formalin-fixed paraffin-embedded) gastric cancer and normal tissue specimens were sectioned and stained with hematoxylin and eosin (HE).Representative areas of tissue were selected under a microscope.The paracancerous normal tissues are located within 5 cm of the cancerous tissue.Using a hollow core needle, cylindrical cores with a diameter of approximately 1.4mm were extracted from the representative regions of each tissue block and embedded into pre-drilled empty paraffin blocks.The paraffin blocks were then placed on a mold, and the surface was coated with molten paraffin to seal the edges.Subsequently, they were incubated in an oven at 60 ℃ for 1 hour to slightly melt the paraffin and enclose the tissue cores.After cooling at room temperature and further chilling in a refrigerator at 4 ℃ for 1 hour, the blocks were demolded.

2.4.2 miRNAscope In Situ Hybridization for miR-29c Expression Detection

Tissue microarray sections, 3.5 μm in thickness, were baked at 60 ℃ for 1.5 h, deparaffinized in fresh xylene, dehydrated in fresh ethanol, fixed in 12% formaldehyde at room temperature for 2 h, and endogenous peroxidase was blocked using hydrogen peroxide.The tissue sections were then subjected to target retrieval at boiling point(98 ℃ to 102 ℃) for 15 min, followed by drawing a hydrophobic barrier with an ImmedgeTM hydrophobic pen, proteinase III treatment, and incubation in a HybEZ hybridization oven at 40 ℃for 30 min.The tissue was then covered with miRNAScopeTM miR-29c probe and hybridized for 2 h.Signal detection was performed using the miRNAscopeTM HD detection kit, with signal amplification using Amp1-Amp6, and colorimetric visualization with a mixture of Fast-RED solution A and B (60:1).Finally, the sections were counterstained with hematoxylin, air-dried in a fume hood, and mounted with neutral resin.Hs-RNU6 served as the positive control,while sramble was used as the negative control.The miRNAscope results were observed under a standard optical microscope.

2.5 Interpretation of Results

miRNAscope results are divided into 4 levels:0 points (the average staining point per cell is 0 or 1 point /10 cells, 40× field of view),1 point (2 to 10 points per cell, no or few cell clusters, 40 field of view), 2 points (11 to 20 points per cells and or ＜ 25% of positive cells have punctated clusters, 20～40 field of view),Scores of 3(＞ 20 points per cell and or ＞ 25% of positive cells had punctated clusters, 20 field of view), the scores of 0 and 1 are classified as low expression, and scores of 2 and 3 are classified as high expression.

2.6 Statistical Analysis

R4.0.3 and SPSS26.0 statistical software are used for analysis, Chisquare test was used for categorical variables comparison, paired rank sum test was used for paired continuity variables comparison,and log-rank test was used for survival analysis.P＜ 0.05 is considered statistically significant.

3.Result

3.1 Difference analysis of miR-29c in TCGA database

The results showed that compared with normal gastric mucosa tissues (median expression 12.51054,n=41), the expression of miR-29c in gastric cancer tissues (median expression 10.87942,n=434)was significantly down-regulated (P＜ 0.001).In 41 matched tissues,miR-29c expression in gastric cancer tissues was also significantly down-regulated (P＜ 0.001).See Figure 1.

Fig 1 Expression diagram of miR-29c-3p in TCGA data

3.2 Relationship between miR-29c expression in TCGA database and clinical features and prognosis

There was no significant correlation between the expression of miR-29c and the age, gender, location, grade, TNM stage, T stage and N stage of patients with gastric cancer (P＞ 0.05).Analysis of the expression of miR-29c and the overall survival of patients with gastric cancer showed that the survival rate of the group with high expression of miR-29c was above that of the group with low expression, but the difference was not significant (P=0.141).

3.3 Prediction and enrichment analysis of miR-29c target genes

In order to further explore the biological function of miR-29c (FIG.2A), this project predicted downstream genes regulated by miR-29c according to an online database, and the prediction threshold of miRNA was set as follows: when Pictar score was greater than 2,449 candidate target genes were obtained; The number of TargetScan conservative binding sites was ＞ 0 and the total score was ＜ -0.4, 365 candidate target genes were obtained;Starbase CLIP Data number3,Pan-Cancer number 10, 809 candidate target genes were obtained.In addition, 242 confirmed downstream mRNA of miR-29c were downloaded from the mirtarbase website, of which 161 were supported by two or more evidences.The intersection of prediction results from the online website was used to obtain 18 candidate target genes (Figure 2B).GO functional annotation results showed (Figure 2C) that the 18 potential target genes mainly included biological functions such as extracellular matrix structural components, DNA methyltransferase activity, protein binding.The KEGG pathway is mainly concentrated in signaling pathways such as miRNAs in cancer, pathways in cancer, PI3K-AKT signaling pathway and ECMreceptor interaction.(Figure 2D).

Tab 1 Relationship between expression of miR-29c and clinicopathological features in gastric cancer patients

3.4 Verification of miR-29c expression in gastric cancer tissues and its relationship with clinicopathological features of patients

Fig 2 Target genes prediction and enrichment analysis of miR-29c

In order to further confirm the results of TCGA data analysis,miRNAscope technology was used in this project to detect the expression level of miR-29c.The results showed that among 73 gastric cancer tissues, there were 29 cases in the group with high expression of miR-29c and 44 cases in the group with low expression, and the high expression rate was 39.7% (29/73).Among the 71 normal tissues, 64 cases were in the high expression group of miR-29c, and 7 cases were in the low expression group,with the high expression rate of 90.14% (64/71).Compared with normal tissues, the expression of miR-29c was significantly downregulated in gastric cancer tissues (χ2=39.995,P＜0.001).See Figure 3.Correlation analysis of miR-29c expression and clinicopathological features in 73 patients with gastric cancer showed that the expression level of miR-29c was related to TNM stage, lymph node metastasis and tissue type of patients with gastric cancer (allP＜0.05), but not related to age, sex, location, depth of invasion and vascular invasion.See Table 1.

Fig 3 Expression of miR-29c in gastric cancer and normal tissues(200×)

4.Discussion

Gastric cancer is a prevalent form of malignant tumor globally,and the use of endoscopic resection has emerged as an effective treatment approach for early gastric cancer.However, due to latestage diagnosis and limited treatment options, the overall survival rate remains low, posing a significant threat to human health and generating substantial economic burdens.In recent years, targeted therapy for tumors has made rapid progress.For instance, targeting the HER2 gene with trastuzumab therapy has demonstrated improved survival rates in patients with breast and gastric cancer[18,19].Nonetheless, targeted drugs still have limitations, such as narrow applicability and susceptibility to drug resistance.As a result,there is considerable emphasis on exploring new and more precise biomarkers in current research efforts.

miRNAs, crucial factors in epigenetic alterations, act as endogenous suppressors of multiple target genes and exhibit aberrant expression in various malignant tumors[20].Previous studies have revealed differential expression of various miRNAs, including miR-29c, in gastric cancer tissues.These miRNAs play roles in regulating cancer cell adhesion, migration, gene expression, and the activation of signaling pathways such as AKT, MAPK, and p53[21].Research has also indicated that miR-29c is involved in regulating the migration and apoptosis of gastric cancer cells through the RhoA/ROCK signaling pathway[22],Analysis of fresh tissue samples from 60 gastric cancer patients by Li Liu et al.showed statistically significant differences in miR-29c expression among different tumor stages (P= 0.049)[23],Tae-Su Han[9] et al.observed downregulation of miR-29c expression in gastric cancer tissues but did not establish a direct relationship with the clinical pathological features of gastric cancer.However, it’s important to note that RNA degradation is severe in paraffin-embedded tissues, and the abundance of miR-29c in cancer tissues is very low, which might impact the accuracy of conclusions drawn from this method.

Our project also attempted to analyze miR-29c expression by extracting RNA from paraffin-embedded tissues.However, only a few tissues could detect miR-29c expression, with CT values all above 32 and unsatisfactory melting curves (unpublished data).To enhance the detection rate of miR-29c expression in gastric cancer tissues, our project utilized the RNAscope technology developed by ACD.RNAscope is an in situ hybridization technique specifically designed for paraffin-embedded tissues.Its unique double Z probe design enables signal amplification up to 400 times while simultaneously suppressing background noise[24].The application of this method significantly increased the detection rate of miR-29c, with enhanced specificity, and allowed for the simultaneous observation of tissue morphology and signal expression.

Through the use of RNAscope technology, our project detected miR-29c in gastric cancer patients and analyzed its relationship with clinical factors.We found that the expression of miR-29c in gastric cancer tissues was significantly lower than in normal tissues(P＜ 0.001), consistent with the analysis results from the TCGA database.The decreased expression of miR-29c was significantly associated with the pathological staging, lymph node metastasis, and histological typing of gastric cancer patients (allPvalues ＜ 0.001),consistent with the conclusions of other researchers[25], These research findings suggest that miR-29c may become a diagnostic,prognostic biomarker, and therapeutic target for gastric cancer.Through enrichment analysis of potential target genes of miR-29c, our study found that miR-29c may be involved in biological functions such as DNA methylation activation and interference with protein synthesis, participating in the activation of signaling pathways such as PI3K-AKT.Many studies have confirmed these cellular functions and signaling pathways involving miR-29c [6,26-28], but our study still has some limitations in that it only demonstrates the close relationship between miR-29c expression and clinical pathological features at the histological level and does not further explore its mechanisms at the cellular level.

In conclusion, miRNAscope technology can sensitively detect the expression of trace miRNAs in paraffin-embedded tissues.miR-29c is underexpressed in gastric cancer tissues and significantly correlates with the TNM staging, lymph node metastasis, and histological typing of patients.It is possible that miR-29c is involved in the occurrence and development of gastric cancer, and further in-depth research may confirm its potential as a diagnostic and prognostic marker for gastric cancer patients, as well as a potential molecular therapeutic target.

5.Conflict of Interest Author Statement

They have no competing interests.

6 Author Contributions

All authors contributed to the manuscript: Huang Yousheng designed the research, Zhang Yongfang completed the experiment and wrote the paper, Li Jingcheng analyzed the data, Xie Na participated in the research idea and manuscript writing, Huang Yousheng, Chen Yan,and Chen Junming helped revise the article, and all authors approved the final version of the article.