Investigation of novel regulatory mechanisms of TRIM29 in BC, GC, and CRC patients: systems biology approach to find novel non-coding interactions

doi:10.21203/rs.3.rs-5092400/v1

Download PDF

Research Article

Investigation of novel regulatory mechanisms of TRIM29 in BC, GC, and CRC patients: systems biology approach to find novel non-coding interactions

https://doi.org/10.21203/rs.3.rs-5092400/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

The rising cancer mortality and increasing incidence demand further investigation, particularly for breast cancer (BC), the leading cause of cancer deaths in women, gastric cancer (GC), among the top five global cancers, and colorectal cancer (CRC), the third most common in men and second in women. Numerous studies have shown that mRNAs and lncRNAs play key roles in cancer regulation. Dysregulation of lncRNAs like NORAD, MIR497-HG, and the TRIM29 gene has been reported in various cancers. This research aimed to explore their potential as biomarkers and tumor suppressors in BC, GC, and CRC.

Methods

High-throughput gene expression analysis was conducted using R Studio (v4.4.1) with datasets GSE134359, GSE54129, and GSE81558 from GEO. Data normalization and visualization were done with gplots, ggplot2, factoextra, reshape2, EnhancedVolcano, VennDiagram, and pheatmap gplots. PPI networks were sourced from STRING, and pathway enrichment was analyzed via Enrichr and Reactome. Gene ontology and expression analysis were performed using Enrichr and ENCORI, while GEPIA2 was used for correlation and survival analysis. Data visualization was done through NetworkAnalyst and R Studio. qRT-PCR validated the findings in BC, GC, and CRC samples, with data analyzed via the ddCt method using GraphPad Prism (v10.3.1).

Results

Bioinformatics and qRT-PCR analyses revealed TRIM29 was downregulated in BC and upregulated in CRC, but It reduced in GC despite microarray data suggesting otherwise. TRIM29 showed significant interaction with hsa-miRNA-3940-5p. MIR497-HG expression was notably reduced in BC, GC, and CRC across both microarray and qRT-PCR. It also exhibited strong links with MUC2, MUC4, MUC5AC, and MUC5B. While ENCORI indicated a slight decrease in NORAD expression in BC, qRT-PCR results were not significant. However, NORAD was significantly upregulated in GC and CRC, interacting with key cancer-related genes like MUC4, MUC2, MUC16, MUC3A, and MUC5AC.

Conclusion

Our findings highlight TRIM29's significant involvement in interferon gamma signaling and Interferon Signaling pathways, where dysregulation can contribute to tumorigenic processes cancer in BC, GC, and CRC. MIR497-HG and NORAD seem to interact with mRNAs and indirectly contribute to signaling pathways that impact tumorigenesis in these cancers. TRIM29, NORAD, and MIR497-HG are potential diagnostic biomarkers in GC and CRC. However, for BC just TRIM29 and MIR497-HG Show diagnostic significance. Our study found strong positive correlations between TRIM29 and the lncRNAs MIR497-HG and NORAD in BC, GC, CRC. The robust associations, particularly between TRIM29 and MIR497-HG in BC and GC, suggest that these interactions may play a role in tumorigenesis.

Cancer Biology

Bioinformatics

lncRNA

TRIM29

NORAD

MIR497-HG

Microarray

cancer biomarkers

Worldwide, Breast Cancer (BC) constitutes the primary cause of cancer-derived fatalities between females. As highlighted in the 2018 Cancer Statistics, it ranks as the most prevalent neoplasm in women and the leading cause of cancerous demise, with over 2,100,000 new diagnoses annually and beyond Sixty-two thousand deceased [1]. Five intrinsic/molecular subgroups of BCs can be distinguished by the existence or lack of three hormone receptors, including human epidermal Growth Factor 2 (HER2), Progesterone Receptor (PR), Estrogen Receptor (ER), and along with the proliferation index marker Ki67[2, 3]. Breast cancer subtypes encompass HER2-positive, basal-like, and normal-like tumors, luminal A and luminal B [3, 4]. Despite enhancements in outcomes for Breast Cancer patients due to early diagnosis and the latest anti-cancer medications, including surgery confined to the region, standard chemotherapeutic treatments, targeted radiotherapy, hormone-based therapy, and monoclonal antibody administration, this disease still has a significant recurrence rate and threatens the lives of a great number of people around the world [5–8]. Thus, the analysis of gene expression associated with BC, comprehending gene expression patterns in various clinical and pathological circumstances related to it, and detecting diagnostic and prognostic biomarkers can yield significant awareness of the disease and assisting in its precaution [9].

Gastric Cancer (GC) is a substantial health crisis globally. Helicobacter pylori contamination and food intake habits play a major role in determining the epidemiology of this disease[10, 11]. With over 1 million cases and more than 768,000 casualties in 2020, GC is considered the fifth most widespread cancer diagnosis and the third predominant cause of cancer-associated deaths [1, 12]. Based on the molecular classification by the Asian Cancer Research Group (ACRG), four subtypes have been identified, including i) Microsatellite instability in gastric carcinomas (MSI GCs), ii) Gastric carcinomas demonstrating epithelial-to-mesenchymal transition (EMT GCs), iii) Microsatellite stable gastric cancers with preserved TP53 activity (MSS/TP53+) and iv) Microsatellite-stable gastric tumors exhibiting TP53 inactivity (MSS/TP53-)[13]. A distinguishing aspect of the ACRG Sorting, is its potential for relating each molecular category to therapeutic results and specific relapse models [14]. The latest advancements in genome analysis have revealed numerous biomarkers involving genes, lncRNAs, and miRNAs. The emergence of these biomarkers is expected to greatly influence cancer progression, effective treatment selection, and follow-up practices[15, 16].

Colorectal cancer (CRC) is excessive cell proliferation or abnormal cell growth in the colon, rectum, or appendix[17]. The incidence of this disease has increased worldwide and is now the third most prevalent tumorous malignancy in men and the second most prevalent in women. Furthermore, 10% of cancer fatalities are attributed to CRC[18]. Approximately nineteen million CRC occurrences and 10 million casualties due to this cancer were reported in 2020[19]. By 2030, the international incidence of CRC is anticipated to elevate by 60% leading to approximately two million two hundred thousand new diagnoses and 1,100,000 fatalities [20]. The majority of patients receive a diagnosis of colorectal cancer at an advanced stage, leading to a poor prognosis[21]. At this stage, malignant proliferation and significant metastasis have already taken root[22]. Statistics show that 50% of the patients with CRC lose their lives due to the spread of the disease to distant organs. Colorectal cancer initiation and progression is a multifaceted, multi-phase procedure involving various biological pathways and genetic changes. Consequently, it is crucial to seek pioneering biomarkers which are able to either stimulate or prohibit CRC progression[23].

lncRNAs (Long non-coding RNAs) are classified as a type of RNA that does not encode proteins. They are longer than 200 base pairs and do not possess open reading frames[24]. Compared to protein coding genes, lncRNAs usually contain reduced number of exons, are less conserved, and are present in lower quantities[25]. The importance of lncRNAs not only in transcriptional but also in post-transcriptional regulation of genes has been demonstrated[25]. They achieve this by interacting physically with various Surfaces, including proteins (transcriptional co-factors), RNAs (microRNA sponges), and DNA (chromatin interaction frameworks)[26]. Meanwhile, lncRNAs could be encapsulated within lipid vesicles, like exosomes and apoptotic bodies, or firmly attach to particular proteins. They are then emitted in to serum, urine, and various body fluids, facilitating their acquisition and detection. [27]. These molecules have a half-life of around 3.5 hours, surpassing that of many mRNAs[28]. lncRNAs are recognized as crucial regulators of the biological behavior of cells, playing essential roles in cancer initiation, development, and metastasis[29, 30]. Additionally, their stable expression and easy acquisition from body fluids make them viable candidates for tumor diagnosis and prognosis[31].

A sophisticated relationship between miRNAs and lncRNAs has been distinguished, suggesting that specific lncRNAs serve as host genes for miRNAs and are designed to generate miRNAs that inhibit intended genes [32, 33]. Findings indicate that lncRNA MIR497HG, positioned on chromosome 17p, is downregulated and acts as a tumor suppressor in bladder cancer[34, 35]. Progressive investigations reveal that the MIR497HG-derived miR-497/195 cluster is diminished and serves as a cancer-inhibiting factor in various human neoplastic diseases, such as hepatic malignancy, bladder cancer, lung cancer, glioma, colon cancer, ovarian cancer, acute lymphoblastic leukemia, and breast cancer[32].NORAD (LINC00675) is triggered by DNA damage and takes part in preserving genome steadiness and normal mitosis[36]. This lncRNA is extensively expressed and thoroughly preserved across mammalian species. In the human DNA sequence, the NORAD gene consists of a single exon, yielding a 5.3 kb transcript, and is found at Chr20q11.23. NORAD is primarily distributed within the cytoplasmic area of human cells[37, 38]. It ensures genomic stability by interacting with Pumilio proteins, restricting the inhibition of their aimed mRNAs. Thus, the deactivation of NORAD results in chromosomal instability and aneuploidy, contributing to the buildup of genetic mutations and cancer development. NORAD has been observed in multiple forms of cancers such as breast cancer[39].TRIM29 also referred to, as the ataxia-telangiectasia group D complementing gene (ATDC), is part of the TRIM protein family, locating on chromosome 11q[40]. Increased amounts of theTRIM29 expression have been detected in lung, bladder, pancreatic, gastric, prostate, and nasopharyngeal carcinomas[41–46]. TRIM29 has been shown to enhance malignant cell proliferation and invasiveness via the stabilization of β-catenin[47]. The mechanisms could involve suppressing p53 activity and enhancing cell vitality by downregulating apoptosis-promoting genes that are modulated through p53[48]. However, reduced expression of it has been reported in prostate cancers[49, 50]. Hence, its function is intricate, relying on the specific cancer subtype and the microenvironmental context[51].

Numerous efforts have been directed towards pinpointing lncRNAs as innovative active genetic sequences involved in cancerous tumorigenesis and understanding their potential as novel biomarkers for BC, GC, and CRC diagnosis and prognosis. In this study, we concentrated on two novel lncRNAs (MIR497HG, NORAD) and a gene (TRIM29) to investigate their expression levels in BC, GC, and CRC via laboratory experiments. Simultaneously, diverse bioinformatic analyses were employed to investigate the roles and significance of the three mentioned elements in BC, GC, and CRC.

2.1. Microarray analysis

R Studio (v4.4.1) was used for conducting microarray analysis on the datasets GSE134359, GSE54129, and GSE81558 (Table 1). The affy package facilitated the preliminary processing of raw microarray data, with normalization implemented using the Robust Multi-array Average (RMA) method included within the package. The expression data from microarrays were statistically scrutinized using the limma package, which, along with the affy package, was obtained from Bioconductor.org. The gplots, ggplot2, factoextra, reshape2, EnhancedVolcano, VennDiagram, and pheatmap packages were utilized to generate visual representations of the microarray analysis results.

Dataset	tumor samples	control samples	platform	disease
GSE134359	74	12	GPL570	breast cancer
GSE54129	111	21	GPL570	gastric cancer
GSE81558	42	9	GPL15207	colorectal cancer

2.2. Bioinformatics Analyses

PPI networks for TRIM29-associated genes were sourced from the STRING database[52]. For conducting pathway enrichment analysis, the Enrichr[53] as well as the Reactome [54]database was utilized. Gene ontology analysis was implemented through the Enrichr online database. Besides, the process of Gene Expression analysis was completed with the ENCORI[55] online database. This database supported the execution of correlation analysis too. The GEPIA2 online resource[56] enabled the conduct of survival analysis. The miRwalk[57] online interface was applied to investigate miRNA and mRNA interaction dynamics. For understanding the interplay between lncRNAs and mRNAs, the lncRRlsearch [58] web tool was utilized. Visual data representation was achieved via the NetworkAnalyst platform[59] and Rstudio software(v4.4.1).

Table 1: Features of the GSEs investigated in this study

2.3. Tissue collection and ethics statement:

Approval for all methodologies employed in this research project, including human tissue samples, was granted by the Ethics Committee of Al-Zahra Hospital, Isfahan University of Medical Sciences, and all patients acknowledged by signing written consent forms according to institutional guidelines. No radiation or chemotherapy had ever been administered to the patients. For each of the three cancer types—breast cancer, gastric cancer, and colorectal cancer—twenty pairs of tumors and neighboring non-malignant tissues were acquired from Al-Zahra Hospital (Isfahan, Iran) and Sayyed al-Shohada (Omid) Hospital (Isfahan, Iran). Following a distilled water rinse, the tissue samples were submerged in RNA later solution (Invitrogen, USA) and then frozen in liquid nitrogen for further pathological examination.

Table 2

Clinicopathological features of BC samples
Characteristic	Status	Number of patients (%)
Stage	I II III Unknown	1(5%) 7(35%) 10(50%) 2(10%)
Age	< 50 > 50	9(45%) 11(55%)
Tumor size	< 5cm > 5cm Unknown	3(15%) 14(70%) 3(15%)
Menopausal status	Yes No Unknown	11(55%) 7(35%) 2(10%)
Lymph node	Unknown Yes, Nos No	1(5%) 8(40%) 11(55%)
ER receptor	positive negative Unknown	11(55%) 6(30%) 3(15%)
PR receptor	positive negative Unknown	8(40%) 9(45%) 3(15%)
HER2/neu receptor	positive negative Unknown	7(35%) 10(50%) 3(15%)

Table 3

Clinicopathological features of GC samples
Characteristic	Status	Number of patients (%)
Stage	I II III IV	1(5%) 6(30%) 4(20%) 9(45%)
Sex	Male Female	14(70%) 6(30%)
Perineural invasion	Yes No	14(70%) 6(30%)
Histology	Adenocarcinoma Signet ring carcinoma Mucinous Adenocarcinoma	18(90%) 1(5%) 1(5%)
Smoking	DX-smoker at diagnosis but discontinued Nonsmoker Ex-Smoker	2(10%) 17(85%) 1(5%)
Family History	Yes No	5(25%) 15(75%)
Nodal Extension	Yes No	3(15%) 17(85%)

Table 4

Clinicopathological features of CRC samples
Characteristic	Status	Number of patients (%)
Stage	I II III IV	2(10%) 3(15%) 7(35%) 8(40%)
Age	< 50 > 50	15(75%) 5(25%)
Sex	Male Female	9(45%) 11(55%)
Tumor size	< 5cm > 5cm	6(30%) 14(70%)
Perineural invasion	Yes No	12(60%) 8(40%)
Lymphatic Invasion	Unknown Yes, Nos No	1(5%) 8(40%) 11(55%)
Histology	Adenocarcinoma Mucinous (Colloid) Adenocarcinoma	17(85%) 3(15%)
Smoking	DX-smoker at diagnosis but discontinued Nonsmoker smoker	1(5%) 16(80%) 3(15%)
SOP	Ascending Colon Rectosigmoid Sigmoid Colon Rectum Cecum Colon, NOS	1(5%) 4(20%) 4(20%) 4(20%) 2(10%) 5(25%)

2.4. RNA extraction, cDNA synthesis, and QRT-PCR experiment

YTzol pure RNA reagent (Yekta Tajhiz Azma, Tehran. Iran) was utilized to extract the RNA from both tumorous and normal tissues. Subsequently, cDNA synthesis was carried out using the RT-ROSE cDNA synthesis kit as per the manufacturer's instructions (ROJE Technologies, Tehran, Iran). Real-time PCR was conducted using SYBR green qPCR master mix 2x (Yekta Tajhiz Azma, Tehran. Iran). A Magnetic Induction Cycler (Biomolecular Systems, Australia) was employed to perform the RT-qPCR experiment. The device, controlled by the program, was used to amplify cDNA at three different temperature settings: first at 95°C for 15 minutes, then at 95°C for 15 seconds, at 60°C for 20 seconds, and finally at 72°C for 20 seconds. In more detail, from the second 95°C step onward, 40 cycles were executed. The primer sequences for the target sequences in this research are listed in Table 1. The expression levels of TRIM29, MIR497HG, and NORAD were

calculated using 2 ^−ΔΔCt method[60]. As an internal control, the related expressions were normalized using the quantity of housekeeping gene GAPDH.

Table 5

Features of the primers
Gene	RNA Type	Primer (5' –> 3')	Forward/Reverse
TRIM29	mRNA	CCCAGCAAGCCTGAAACAG TCAGCCTTGGTGCCATTCTC	F R
MIR497HG	lncRNA	GAGCACCTTCTCTTCTGAAACC TCGTCCTCAGAACTCCTTTCC	F R
NORAD	lncRNA	GGGCGGTTGGTCTTCATTC AGAGGGTGGTGGGCATTTC	F R
GAPDH	mRNA	AGCCACATCGCTCAGACAC GCCCAATACGACCAAATCC	F R

2.5. Statistical analyses

Real-time PCR data analysis via the ddCt method was performed with GraphPad Prism (version 10.3.1), and the resulting expression charts were generated using GraphPad Prism software too. The normality of the expression data was evaluated using the Kolmogorov-Smirnov test. Besides, the Wilcoxon test and paired t-test were then applied to the expression data to determine the significance of our results. A receiver operating characteristic (ROC) analysis was conducted to determine the diagnostic effectiveness of the tumor samples. The ROC results are evaluated by examining the area under the curve (AUC), where an AUC between 0.7 and 0.8 signifies an acceptable biomarker, 0.8 to 0.9 indicates an excellent biomarker, and 0.9 to 1 reflects an exceptional diagnostic biomarker.

3.1. Microarray Analysis

This study involved the analysis of gene expression profiles from breast, gastric, and colorectal cancers using data derived from three Gene Expression Series (GSE134359, GSE54129, and GSE81558). To assess the microarray samples' quality, principal component analysis (PCA) was applied. The PCA results for BC (Figure. 1(a)), GC (Figure. 1(b)), and CRC (Figure. 1(c)) demonstrated that all samples were sufficiently robust for subsequent analyses. Differential expression analysis (DEA) was performed to identify novel, common diagnostic biomarkers for BC, GC, and CRC, with results showing that TRIM29 is overexpressed in BC (logFC: 0.26753, adj. P. Value: 0.001241). The expression level of TRIM29 has increased by a factor of approximately 1.2 (since 2^0.26753 ≈ 1.2). This indicates that the gene is up-regulated by about 20%. Conversely, in GC (logFC: -0.64095, adj. P. Value: 0.002168) the expression level of TRIM29 has decreased by approximately 60% compared to the control condition (since 2^ (-0.64095) ≈ 0.65). This indicates that the gene is down-regulated. In CRC (logFC: -1.14673, adj. P. Value: 0.007131) the expression level of TRIM29 has decreased to about 45% of its original level (since 2^ (-1.14673) ≈ 0.45). This represents a down-regulation, with a decrease in gene expression by approximately 55% compared to the control condition (Fig. 2). Furthermore, Figs. 3(a), 3(b), and 3(c) illustrate the most highly differentially expressed genes (DEGs) in tumor versus normal samples across each cancer type. To visualize the overlap of the significant DEGs (adj. P. Value < 0.05) among the three cancer types, a Venn diagram was constructed (Fig. 4). To explore the differential gene expression patterns in BC, GC, and CRC, we generated violin plots for the top 100 differentially expressed genes (DEGs) based on microarray data. These plots illustrate the expression levels of these critical genes in normal versus tumor samples across the three cancer types, as shown in Fig. 5.

3.2. Protein-Protein Interaction (PPI), Pathway Enrichment, and Gene

Ontology (GO)Analyses

Genes perform their biological activities through mutual interactions. As per the information from STRING database we identified that BBOX1, TRAT1, TRIM23, ATM, DES, DSG3, HINT1, KAT5, KRT5, and DVL2, which are co-expressed with TRIM29, demonstrate robust interactivity with each other (Fig. 6). Using Enrichr for pathway enrichment analysis, we investigated cancer-associated signaling routes modulated by TRIM29 and its interactive protein network. The analysis confirmed that TRIM29 and its network directly influence Interferon Gamma Signaling Pathway (adjusted p-value < 0.05, Fig. 7(a)) by participating in the expression of IFNG-stimulated genes and GAF binds the GAS promoter elements in the IFNG-regulated genes. Additionally, it plays role in Interferon Signaling Pathway (adjusted p-value < 0.05, Fig. 7(b)). In accordance with the Reactome database, Table 3-a and Table 3-b compile pathways that are directly and indirectly pertinent to TRIM29. Additionally, in our study, we performed a detailed Gene Ontology (GO) enrichment analysis, by Enrichr, to determine key biological processes, cellular components, and molecular functions correlated with diversely expressed genes in breast, gastric, and colorectal cancers. The analysis revealed significant enrichment in several GO terms. Among the biological processes, epidermis development (GO:0008544) and intermediate filament organization (GO:0045109) were notably enriched, indicating their crucial roles in maintaining epithelial integrity and structural organization in these cancers. For the cellular component category, the intermediate filament cytoskeleton (GO:0045111) was significantly represented, highlighting its importance in providing mechanical support and stability to the cancer cells. Additionally, in the molecular function category, calcium ion binding (GO:0005509) emerged as a significant term, suggesting its involvement in various signaling pathways and regulatory mechanisms critical for cancer cell proliferation and metastasis. The entirety of Gene Ontology analysis information (adjusted p-value < 0.05) is included in Table <link rid="tb9">4</link>-a, <link rid="tb9">4</link>-b, and <link rid="tb9">4</link>-c.

Table 3

a: Reactome Pathway Enrichment (TRIM29 direct association)
Term	P-value	Adjusted P-value	Combined Score
Interferon Gamma Signaling R-HSA-877300	0.004449938	0.017799751	107815.3803
Interferon Signaling R-HSA-913531	0.009999899	0.019999798	91182.56991
Cytokine Signaling in Immune System R-HSA-1280215	0.035099812	0.04679975	64639.7992

Table 3

b: Reactome Pathway Enrichment (TRIM29 indirect association)
Term	P-value	Adjusted P-value	Combined Score	Genes
Keratinization R-HSA-6805567	2.62E-22	7.42E-20	1316.044725	KRT80; KRT4; KLK5; KRT13; KRT7; KRT5; EVPL; PPL; KRT17; KRT16; KRT15; KRT14; PKP1; PRSS8; PKP3; DSG3; KRT6C; SPRR1A; KRT6B; SPRR1B; IVL; KRT6A
Formation Of Cornified Envelope R-HSA-6809371	8.94E-12	1.26E-09	788.1150524	KLK5; PKP1; PRSS8; DSG3; PKP3; EVPL; PPL; SPRR1A; IVL; SPRR1B
Developmental Biology R-HSA-1266738	3.15E-09	2.98E-07	98.11316789	GRB7; KRT80; KRT4; KLK5; KRT13; KRT7; KRT5; EVPL; PPL; KLF5; KRT17; KRT16; KRT15; KRT14; PKP1; PRSS8; PKP3; DSG3; KRT6C; SPRR1A; KRT6B; SPRR1B; IVL; KRT6A
Type I Hemidesmosome Assembly R-HSA-446107	1.07E-07	7.56E-06	2409.003511	COL17A1; LAMB3; ITGB4; LAMC2
Cell Junction Organization R-HSA-446728	4.72E-07	2.67E-05	248.2571248	COL17A1; CLDN4; LAMB3; ITGB4; LAMC2; NECTIN4; CLDN1
Cell-Cell Communication R-HSA-1500931	4.48E-06	2.11E-04	146.4855796	COL17A1; CLDN4; LAMB3; ITGB4; LAMC2; NECTIN4; CLDN1
Gap Junction Assembly R-HSA-190861	8.60E-05	0.0034786	401.2837151	GJB4; GJB3; GJB5
Laminin Interactions R-HSA-3000157	2.31E-04	0.008171618	245.5215128	LAMB3; ITGB4; LAMC2
Gap Junction Trafficking R-HSA-190828	3.41E-04	0.010718872	202.1566037	GJB4; GJB3; GJB5
Gap Junction Trafficking and Regulation R-HSA-157858	4.30E-04	0.011314005	179.8983478	GJB4; GJB3; GJB5
Sensing Of DNA Double Strand Breaks R-HSA-5693548	4.43E-04	0.011314005	710.8779085	KAT5; ATM
MET Promotes Cell Motility R-HSA-8875878	4.80E-04	0.011314005	170.2579955	LAMB3; LAMC2; TNS4
Anchoring Fibril Formation R-HSA-2214320	6.18E-04	0.013458792	544.1581961	LAMB3; LAMC2
Non-integrin membrane-ECM Interactions R-HSA-3000171	0.00148214	0.027784121	95.41656808	LAMB3; ITGB4; LAMC2
Collagen Formation R-HSA-1474290	0.001539135	0.027784121	56.27963508	COL17A1; LAMB3; ITGB4; LAMC2
Apoptotic Cleavage of Cell Adhesion Proteins R-HSA-351906	0.001596101	0.027784121	263.45155	PKP1; DSG3

Table 4

*a: TRIM29 and its interactome GO (*biological *process) results furnished by Enrichr.*
Term	P-value	Adjusted P-value	Combined Score	Genes
Epidermis Development (GO:0008544)	4.41E-24	2.68E-21	3113.035397	DSP; COL17A1; CALML5; LAMB3; LAMA3; KLK5; LAMC2; GRHL3; KRT5; OVOL1; EVPL; FOXN1; SCEL; KRT15; GJB5; KRT14; SPRR1A; SPRR1B
Intermediate Filament Organization (GO:0045109)	9.48E-19	2.88E-16	2223.101245	DSP; KRT80; KRT4; KRT13; KRT7; KRT5; DES; KRT17; KRT16; KRT15; KRT14; PKP1; KRT6B; KRT6A
Epidermal Cell Differentiation (GO:0009913)	5.45E-10	1.11E-07	721.6756717	DSP; SCEL; KRT16; OVOL1; EVPL; SPRR1A; IVL; SPRR1B
Supramolecular Fiber Organization (GO:0097435)	2.13E-09	3.24E-07	188.858869	DSP; KRT80; KRT4; KRT13; KRT7; KRT5; DES; KRT17; KRT16; KRT15; KRT14; PRSS8; KRT6B; KRT6A
Keratinocyte Differentiation (GO:0030216)	2.89E-09	3.51E-07	758.0554529	DSP; SCEL; KRT16; EVPL; SPRR1A; IVL; SPRR1B
Skin Development (GO:0043588)	3.64E-09	3.68E-07	503.7165659	DSP; SCEL; ITGB4; KRT16; EVPL; SPRR1A; IVL; SPRR1B
Epithelial Cell Differentiation (GO:0030855)	4.77E-08	4.15E-06	241.4044373	EHF; KRT4; KRT17; KRT16; KRT15; KRT14; KRT13; IRF6; OVOL1
Peptide Cross-Linking (GO:0018149)	3.37E-07	2.56E-05	640.9516848	DSP; EVPL; SPRR1A; IVL; SPRR1B
Epithelium Development (GO:0060429)	2.17E-06	1.47E-04	138.3092007	EHF; KRT4; KRT17; KRT16; KRT15; KRT14; KRT13; KRT6A
Establishment Of Skin Barrier (GO:0061436)	3.15E-06	1.92E-04	633.4033721	CLDN4; KRT16; SFN; CLDN1
Skin Epidermis Development (GO:0098773)	4.82E-06	2.67E-04	540.03792	CLDN4; KRT16; SFN; CLDN1
Negative Regulation of Endopeptidase Activity (GO:0010951)	3.14E-05	0.001592208	160.4879983	SERPINB3; ANXA8L1; SERPINB13; ANXA8; SERPINB5
Wound Healing (GO:0042060)	7.95E-05	0.00371946	118.7566716	DSP; CLDN1; EVPL; PPL; KRT6A
Positive Regulation of Peptidase Activity (GO:0010952)	3.41E-04	0.014804056	202.1566037	SERPINB3; CLDN4; PRSS22
Desmosome Organization (GO:0002934)	4.43E-04	0.017964292	710.8779085	DSP; PKP3
Cell-Cell Junction Organization (GO:0045216)	5.38E-04	0.018794752	87.99532342	DSP; PKP1; PKP3; CLDN1
Negative Regulation of Phospholipase Activity (GO:0010519)	6.18E-04	0.018794752	544.1581961	ANXA8L1; ANXA8
Intermediate Filament Bundle Assembly (GO:0045110)	6.18E-04	0.018794752	544.1581961	KRT14; PKP1
Regulation Of Phospholipase A2 Activity (GO:0032429)	6.18E-04	0.018794752	544.1581961	ANXA8L1; ANXA8
Negative Regulation of RNA Catabolic Process (GO:1902369)	6.18E-04	0.018794752	544.1581961	PKP1; PKP3
Positive Regulation of Cell Migration (GO:0030335)	7.70E-04	0.022294067	36.08182257	SERPINB3; GRB7; CLDN4; RAB25; ATM; LAMC2; CLDN1
Positive Regulation of Cell Motility (GO:2000147)	0.001402353	0.038755938	34.68433457	SERPINB3; GRB7; CLDN4; ATM; LAMC2; CLDN1

Table 4

*a: TRIM29 and its interactome GO (*biological *process) results furnished by Enrichr.*
Term	P-value	Adjusted P-value	Combined Score	Genes
Epidermis Development (GO:0008544)	4.41E-24	2.68E-21	3113.035397	DSP; COL17A1; CALML5; LAMB3; LAMA3; KLK5; LAMC2; GRHL3; KRT5; OVOL1; EVPL; FOXN1; SCEL; KRT15; GJB5; KRT14; SPRR1A; SPRR1B
Intermediate Filament Organization (GO:0045109)	9.48E-19	2.88E-16	2223.101245	DSP; KRT80; KRT4; KRT13; KRT7; KRT5; DES; KRT17; KRT16; KRT15; KRT14; PKP1; KRT6B; KRT6A
Epidermal Cell Differentiation (GO:0009913)	5.45E-10	1.11E-07	721.6756717	DSP; SCEL; KRT16; OVOL1; EVPL; SPRR1A; IVL; SPRR1B
Supramolecular Fiber Organization (GO:0097435)	2.13E-09	3.24E-07	188.858869	DSP; KRT80; KRT4; KRT13; KRT7; KRT5; DES; KRT17; KRT16; KRT15; KRT14; PRSS8; KRT6B; KRT6A
Keratinocyte Differentiation (GO:0030216)	2.89E-09	3.51E-07	758.0554529	DSP; SCEL; KRT16; EVPL; SPRR1A; IVL; SPRR1B
Skin Development (GO:0043588)	3.64E-09	3.68E-07	503.7165659	DSP; SCEL; ITGB4; KRT16; EVPL; SPRR1A; IVL; SPRR1B
Epithelial Cell Differentiation (GO:0030855)	4.77E-08	4.15E-06	241.4044373	EHF; KRT4; KRT17; KRT16; KRT15; KRT14; KRT13; IRF6; OVOL1
Peptide Cross-Linking (GO:0018149)	3.37E-07	2.56E-05	640.9516848	DSP; EVPL; SPRR1A; IVL; SPRR1B
Epithelium Development (GO:0060429)	2.17E-06	1.47E-04	138.3092007	EHF; KRT4; KRT17; KRT16; KRT15; KRT14; KRT13; KRT6A
Establishment Of Skin Barrier (GO:0061436)	3.15E-06	1.92E-04	633.4033721	CLDN4; KRT16; SFN; CLDN1
Skin Epidermis Development (GO:0098773)	4.82E-06	2.67E-04	540.03792	CLDN4; KRT16; SFN; CLDN1
Negative Regulation of Endopeptidase Activity (GO:0010951)	3.14E-05	0.001592208	160.4879983	SERPINB3; ANXA8L1; SERPINB13; ANXA8; SERPINB5
Wound Healing (GO:0042060)	7.95E-05	0.00371946	118.7566716	DSP; CLDN1; EVPL; PPL; KRT6A
Positive Regulation of Peptidase Activity (GO:0010952)	3.41E-04	0.014804056	202.1566037	SERPINB3; CLDN4; PRSS22
Desmosome Organization (GO:0002934)	4.43E-04	0.017964292	710.8779085	DSP; PKP3
Cell-Cell Junction Organization (GO:0045216)	5.38E-04	0.018794752	87.99532342	DSP; PKP1; PKP3; CLDN1
Negative Regulation of Phospholipase Activity (GO:0010519)	6.18E-04	0.018794752	544.1581961	ANXA8L1; ANXA8
Intermediate Filament Bundle Assembly (GO:0045110)	6.18E-04	0.018794752	544.1581961	KRT14; PKP1
Regulation Of Phospholipase A2 Activity (GO:0032429)	6.18E-04	0.018794752	544.1581961	ANXA8L1; ANXA8
Negative Regulation of RNA Catabolic Process (GO:1902369)	6.18E-04	0.018794752	544.1581961	PKP1; PKP3
Positive Regulation of Cell Migration (GO:0030335)	7.70E-04	0.022294067	36.08182257	SERPINB3; GRB7; CLDN4; RAB25; ATM; LAMC2; CLDN1
Positive Regulation of Cell Motility (GO:2000147)	0.001402353	0.038755938	34.68433457	SERPINB3; GRB7; CLDN4; ATM; LAMC2; CLDN1

Table 4

*a: TRIM29 and its interactome GO (*biological *process) results furnished by Enrichr.*
Term	P-value	Adjusted P-value	Combined Score	Genes
Epidermis Development (GO:0008544)	4.41E-24	2.68E-21	3113.035397	DSP; COL17A1; CALML5; LAMB3; LAMA3; KLK5; LAMC2; GRHL3; KRT5; OVOL1; EVPL; FOXN1; SCEL; KRT15; GJB5; KRT14; SPRR1A; SPRR1B
Intermediate Filament Organization (GO:0045109)	9.48E-19	2.88E-16	2223.101245	DSP; KRT80; KRT4; KRT13; KRT7; KRT5; DES; KRT17; KRT16; KRT15; KRT14; PKP1; KRT6B; KRT6A
Epidermal Cell Differentiation (GO:0009913)	5.45E-10	1.11E-07	721.6756717	DSP; SCEL; KRT16; OVOL1; EVPL; SPRR1A; IVL; SPRR1B
Supramolecular Fiber Organization (GO:0097435)	2.13E-09	3.24E-07	188.858869	DSP; KRT80; KRT4; KRT13; KRT7; KRT5; DES; KRT17; KRT16; KRT15; KRT14; PRSS8; KRT6B; KRT6A
Keratinocyte Differentiation (GO:0030216)	2.89E-09	3.51E-07	758.0554529	DSP; SCEL; KRT16; EVPL; SPRR1A; IVL; SPRR1B
Skin Development (GO:0043588)	3.64E-09	3.68E-07	503.7165659	DSP; SCEL; ITGB4; KRT16; EVPL; SPRR1A; IVL; SPRR1B
Epithelial Cell Differentiation (GO:0030855)	4.77E-08	4.15E-06	241.4044373	EHF; KRT4; KRT17; KRT16; KRT15; KRT14; KRT13; IRF6; OVOL1
Peptide Cross-Linking (GO:0018149)	3.37E-07	2.56E-05	640.9516848	DSP; EVPL; SPRR1A; IVL; SPRR1B
Epithelium Development (GO:0060429)	2.17E-06	1.47E-04	138.3092007	EHF; KRT4; KRT17; KRT16; KRT15; KRT14; KRT13; KRT6A
Establishment Of Skin Barrier (GO:0061436)	3.15E-06	1.92E-04	633.4033721	CLDN4; KRT16; SFN; CLDN1
Skin Epidermis Development (GO:0098773)	4.82E-06	2.67E-04	540.03792	CLDN4; KRT16; SFN; CLDN1
Negative Regulation of Endopeptidase Activity (GO:0010951)	3.14E-05	0.001592208	160.4879983	SERPINB3; ANXA8L1; SERPINB13; ANXA8; SERPINB5
Wound Healing (GO:0042060)	7.95E-05	0.00371946	118.7566716	DSP; CLDN1; EVPL; PPL; KRT6A
Positive Regulation of Peptidase Activity (GO:0010952)	3.41E-04	0.014804056	202.1566037	SERPINB3; CLDN4; PRSS22
Desmosome Organization (GO:0002934)	4.43E-04	0.017964292	710.8779085	DSP; PKP3
Cell-Cell Junction Organization (GO:0045216)	5.38E-04	0.018794752	87.99532342	DSP; PKP1; PKP3; CLDN1
Negative Regulation of Phospholipase Activity (GO:0010519)	6.18E-04	0.018794752	544.1581961	ANXA8L1; ANXA8
Intermediate Filament Bundle Assembly (GO:0045110)	6.18E-04	0.018794752	544.1581961	KRT14; PKP1
Regulation Of Phospholipase A2 Activity (GO:0032429)	6.18E-04	0.018794752	544.1581961	ANXA8L1; ANXA8
Negative Regulation of RNA Catabolic Process (GO:1902369)	6.18E-04	0.018794752	544.1581961	PKP1; PKP3
Positive Regulation of Cell Migration (GO:0030335)	7.70E-04	0.022294067	36.08182257	SERPINB3; GRB7; CLDN4; RAB25; ATM; LAMC2; CLDN1
Positive Regulation of Cell Motility (GO:2000147)	0.001402353	0.038755938	34.68433457	SERPINB3; GRB7; CLDN4; ATM; LAMC2; CLDN1

3.3. Gene Expression, Correlation (co-expression) and Survival Analyses

We performed gene expression analyses of MIR497HG, NORAD, and TRIM29 using the ENCORI bioinformatics tool, comparing their expression levels between cancerous and normal tissues in BC, GC, and CRC. The expression levels were measured in fragments per kilobase of transcript per million mapped reads (FPKM) and log2-transformed for normalization (adjusted p-value < 0.001). There is a significant downregulation of MIR497HG in tumorous tissues compared to non-cancerous tissues for all the mentioned cancers. Furthermore, in GC and CRC a marked upregulation of NORAD expression in malignant tissues compared to healthy tissues is witnessed while there is a negligible reduction of NORAD in breast cancer-affected tissues versus normal tissues. In the same way, TRIM29 levels are markedly higher in neoplastic tissues in comparison to normal tissues in both GC and CRC, whereas breast cancer tissues exhibit a slight decrease in TRIM29 compared to non-cancerous tissues (Fig. 8). ENCORI-based co-expression analysis was employed to investigate the correlation between TRIM29 and lncRNAs (MIR497HG and NORAD) expression levels in BC, GC, and CRC. In BC, the co-expression analysis shows a weak positive correlation between TRIM29 and MIR497HG (r = 0.132, p = 1.16e-05) and a stronger negative correlation between TRIM29 and NORAD (r = -0.307, p = 1.48e-25), indicating potential complex regulatory interactions of TRIM29 with these non-coding lncRNAs (Fig. 9(a)). In GC, the analysis revealed a weak negative correlation between TRIM29 and MIR497HG (r = -0.108, p = 3.65e-02) and an insignificant correlation between TRIM29 and NORAD (r = 0.035, p = 5.04e-01). These results suggest a potential inverse relationship between TRIM29 and MIR497HG expression and probably no relationship between TRIM29 and NORAD expression in GC (Fig. 9(b)). In CRC, we observed a weak negative correlation between TRIM29 and MIR497HG (r = -0.032, p = 4.92e-01) and a weak negative correlation between TRIM29 and NORAD (r = -0.094, p = 4.19e-02). These findings suggest minimal inverse relationships between TRIM29 and both MIR497HG and NORAD expression levels in CRC (Fig. 9(c)). Survival analysis indicates that the high-expression of TRIM29, MIR497HG, and NORAD exhibits a significantly better overall survival compared to the low-expression of them (HR = 0.77, Logrank p = 0.01) in BC, GC, and CRC. This suggests that higher expression of these gene signatures is associated with improved survival outcomes (Fig. 10).

3.4. mRNA-miRNA Interaction, lncRNA-mRNA Interaction,

and lncRNA-miRNA Interaction Analyses

Through miRNA-mRNA interaction analysis via miRwalk, novel regulatory RNAs modulating TRIM29 were identified. miRNAs with lower interaction energy, a binding probability score of 1, and seed region interaction were chosen. miR-3940 (score: 1, energy: -39.6) demonstrates meaningful interaction with the 3’UTR of TRIM29 mRNA. The foremost 25 regulatory miRNAs for TRIM29 are enumerated in Table 5. Figure 11 presents this data visualized with the aid of NetworkAnalyst. The lncRRlsearch database identified significant interactions between lncRNAs and mRNAs, emphasizing strong binding energies. For NORAD, the strongest binding was with MUC4, showing the lowest energy levels, indicating a highly stable interaction. Other significant partners for NORAD included MUC2, MUC16, and MUC3A, suggesting NORAD's broad regulatory role. MIR497HG showed notable interactions with MUC2 and MUC4, and MUC3A (Table 6) (Fig. 12(a)). The Venn diagram in Fig. 12(b) illustrates the overlap and distinction between these two sets of interacting lncRNAs.

Table 5

The principal 25 regulatory miRNAs for TRIM29
miRNA ID	gene symbol	bindingp	energy	seed	position
hsa-miR-3940-5p	TRIM29	1	-39.6	1	3UTR
hsa-miR-6724-5p	TRIM29	1	-36.8	1	3UTR
hsa-miR-4763-3p	TRIM29	1	-36.6	1	3UTR
hsa-miR-4763-3p	TRIM29	1	-36.6	1	3UTR
hsa-miR-4739	TRIM29	1	-36.5	1	3UTR
hsa-miR-4739	TRIM29	1	-36.5	1	3UTR
hsa-miR-1228-5p	TRIM29	1	-36.2	1	3UTR
hsa-miR-1207-5p	TRIM29	1	-35.4	1	3UTR
hsa-miR-7110-3p	TRIM29	1	-34.1	1	3UTR
hsa-miR-328-5p	TRIM29	1	-34.1	1	3UTR
hsa-miR-6749-5p	TRIM29	1	-32.8	1	3UTR
hsa-miR-6749-5p	TRIM29	1	-32.8	1	3UTR
hsa-miR-3679-5p	TRIM29	1	-32.4	1	3UTR
hsa-miR-6830-3p	TRIM29	1	-32.2	1	3UTR
hsa-miR-6830-3p	TRIM29	1	-32.2	1	3UTR
hsa-miR-6765-5p	TRIM29	1	-32.1	1	3UTR
hsa-miR-6765-5p	TRIM29	1	-32.1	1	3UTR
hsa-miR-6731-5p	TRIM29	1	-31.8	1	3UTR
hsa-miR-6731-5p	TRIM29	1	-31.8	1	3UTR
hsa-miR-663b	TRIM29	1	-31.6	1	3UTR
hsa-miR-1343-5p	TRIM29	1	-31.3	1	3UTR
hsa-miR-6746-5p	TRIM29	1	-31.1	1	3UTR
hsa-miR-1183	TRIM29	1	-31.1	1	3UTR
hsa-miR-6746-5p	TRIM29	1	-31.1	1	3UTR
hsa-miR-4685-3p	TRIM29	1	-31	1	3UTR

Table 6

lncRNA-mRNA Interaction analysis of NORAD and MIR497HG by lncRRlsearch
lncRNA	Transcript Name	Min of Energy	Sum of Energy
NORAD	MUC4	-37.83	-10,943.20
NORAD	MUC2	-38.55	-9,903.22
NORAD	MUC16	-34.18	-8,325.98
NORAD	MUC3A	-42.9	-8,046.08
NORAD	MUC5AC	-35.72	-6,666.40
NORAD	ATXN3	-111.46	-6,260.74
NORAD	MUC5B	-30.29	-6,042.52
NORAD	PRR36	-44.63	-5,637.22
NORAD	KMT2D	-34.96	-4,823.60
NORAD	LRRK1	-48.33	-4,599.99
MIR497HG	MUC2	-41.86	-11,208.80
MIR497HG	MUC4	-27.45	-9,173.99
MIR497HG	MUC3A	-33.16	-6,969.16
MIR497HG	EMC10	-35.62	-6,911.54
MIR497HG	MUC5AC	-28.61	-6,898.45
MIR497HG	SNX8	-53.51	-6,550.73
MIR497HG	PRR36	-31.97	-5,430.48
MIR497HG	MUC5B	-28.32	-5,398.53
MIR497HG	GOLGA6L22	-29.74	-5,220.11
MIR497HG	ABI2	-41.14	-5,208.24

3.4. Expression analysis of TRIM29, MIR497HG, and NORAD

The qRT-PCR experiment results are consistent with the findings of the bioinformatics analyses, except in two cases. In more details, qRT-PCR analysis revealed a significant reduction in TRIM29(logFC: – 6.042, P.value < 0.0001) and MIR497-HG (logFC: – 6.651, P.value < 0.0001) expression levels in human BC patients (Fig. 13(a)) resemble results from ENCORI database (Fig. 8(a)), whereas NORAD's outcome was non-significant because of a high P.value (P.value: 0.8122) (Fig. 13(a)) in our study. Concerning GC, although the qRT-PCR study demonstrated a marked decline in TRIM29 expression (logFC: – 3.300, P.value: 0.0017) (Fig. 13(b)), based on ENCORI it is highly expressed (Fig. 8(b)). MIR497-HG is significantly downregulated (logFC: – 4.600, P.value: 0.0002), but NORAD experienced a substantial increase in expression (logFC: – 3.803, P.value: 0.0009) (Fig. 13(b)), aligning with findings from ENCORI(Fig. 8(b)).Regarding CRC, our data indicates a remarkable upregulation in TRIM29(logFC: 4.739, P.value < 0.0001) and NORAD (logFC: 4.497, P.value: 0.0006) (Fig. 13(c)), paralleling ENCORI observations(Fig. 8(c)). A meaningful downregulation of MIR497-HG (logFC: − 3.903, P.value < 0.0001) was detected (Fig. 13(c)), corroborating ENCORI’s data (Fig. 8(c)). ROC analysis suggests that TRIM29 could serve as an outstanding biomarker for diagnosing BC and a good biomarker for GC and CRC. Meanwhile, MIR497-HG has been validated as a good diagnostic biomarker across BC, GC, and CRC. In contrast, NORAD, while adequate for GC and CRC, failed to meet the criteria of a diagnostic biomarker in our research for BC because of its elevated P.value (P.value: 0.6456) (Fig. 14). To authenticate the bioinformatics correlation analysis (Fig. 9), an equivalent analysis was executed on qRT-PCR data. In BC and GC specimens, TRIM29 is strongly positively correlated with MIR497-HG and NORAD. However, a moderate positive correlation exists between TRIM29 and both MIR497-HG and NORAD in CRC samples (Fig. 15).

To ensure a rational hypothesis and an appropriate experiment design, we performed a high-throughput bioinformatics investigation to pinpoint pivotal biomarkers with potential diagnostic and prognostic relevance in BC, GC, and CRC. In this study, network analysis was employed to map out co-expressed genes alongside our target genes and evaluate the mutual influences of coding and non-coding RNAs. Our investigation focused on mutual interactions and biological pathways regulating cancer-signaling pathways modulated by TRIM29, MIR497HG, and NORAD and their protein interactions. In parallel, In vitro experiments using qRT-PCR were performed to validate the findings from the previous phase, analyzing the expression levels of TRIM29, MIR497HG, and NORAD as potential regulatory prognostic biomarkers. A receiver operating characteristic (ROC) analysis was carried out to determine the diagnostic capabilities of the tumor samples, and TRIM29’s correlation with lncRNAs (MIR497HG and NORAD) was explored to reinforce our hypothesis.

Previous studies indicated that while TRIM29 (ATDC) has different roles depending on the cancer type, it appears to behave as a tumor suppressor in BC[49]. TRIM29 is reported to exhibit reduction in expression in nearly half of primary breast tumors in comparison to normal tissue[61]. Our results are aligning with this finding and indicates that the expression of TRIM29 in cancerous breast tissues is reduced by approximately 64-fold compared to normal tissues (logFC: – 6.042, P.value < 0.0001). Moreover, this study is the first to reveal this gene’s significant diagnostic capacity for differentiating breast cancer from normal samples via ROC analysis. Besides, our bioinformatics findings suggest TRIM29 significantly engages in interactions with various proteins such as BBOX1, TRAT1, TRIM23, ATM, DES, DSG3, HINT1, KAT5, KRT5, and DVL2. Interferon-gamma (IFN-gamma) is part of the type II interferon family and is primarily secreted by activated immune cells, including T and NK cells, but also B-cells and antigen-presenting cells (APCs). IFN-gamma exerts its influence by binding to the IFN-gamma receptor (IFNGR), which is made up of two subunits: IFNGR1 (also known as the alpha chain) and IFNGR2 (the beta chain). IFNGR1 is the receptor responsible for ligand binding, but it cannot transduce signals alone, while IFNGR2 does not bind to IFN-gamma independently but plays a crucial role in signaling, often acting as the limiting factor in IFN-gamma responsiveness. Neither subunit of the IFNGR complex possesses intrinsic kinase or phosphatase activity, requiring proteins like Janus-activated kinase 1 (JAK1), JAK2, and signal transducer and activator of transcription 1 (STAT1) for signal transduction. In its resting state, the IFNGR complex forms a preassembled tetramer, which undergoes a conformational shift upon IFN-gamma binding. This shift triggers the phosphorylation and activation of JAK1, JAK2, and STAT1, ultimately leading to the activation of genes with the gamma-interferon activation sequence (GAS) in their promoters [62]. Our bioinformatics investigation indicates that, Interferon Gamma Signaling Pathway is significantly influenced by TRIM29 because throughout this pathway TRIM29 involves in the expression of IFNG-stimulated genes (such as IFI30, MT2A, SP100, PML, FCGRA). Moreover, TRIM29 assists GAF in binding to the GAS promoter elements in the IFNG-regulated genes in this pathway. The precise function of TRIM29 in gastric cancer (GC) remains unclear so far. However, Jiang et al. reported that TRIM29 was expressed at lower levels in GC tumor tissues than in the paired normal tissues [63]. Analogously, our data demonstrate that TRIM29 is significantly suppressed in the GC samples relative to adjacent normal tissues. The expression level of it in cancerous tissues is reduced by a factor of approximately 2^-3.3008 ≈ 9.85 times compared to normal tissues. Besides, our ROC analysis illustrates this gene's sufficient diagnostic effectiveness in GC. Elevated TRIM29 expression was found in primary CRC tumors by Jiang et al., and this overexpression correlated with poor overall and disease-free survival. TRIM29 delimitation also impaired RKO cell proliferation and colony formation [64]. Han et al. observed that TRIM29 expression was significantly higher in CRC tissues than in normal tissues, and this upregulation was linked to suboptimal clinical results, advanced tumor progression, and lymph node metastasis, especially in right-sided CRC cases [65]. Evidences from our study confirm that TRIM29 is increased in CRC samples. However, based on ROC analysis, TRIM29 is considered as a reliable diagnostic biomarker. Du et al. state that a significant portion of CRC patients shows resistance to immune checkpoint blockade (ICB). The interferon gamma (IFNγ) signaling pathway promotes natural and ICB-induced antitumor immune responses [66]. During our bioinformatics investigations, we identified TRIM29 to be significantly impactful in not only interferon-gamma (IFNγ) signaling pathway but also interferon signaling pathway. Matboli et al. discovered that the lncRNA SNHG14, hsa-miRNA-3940-5p, and NAP1L2 mRNA exhibited outstanding performance in distinguishing CRC from control samples [67]. Our bioinformatics analysis uncovers a significant interaction between TRIM29 and hsa-miRNA-3940-5p.

Cheng et al. indicate that MIR497-HG is significantly under expressed in BC samples [68]. Our analysis affirms diminished MIR497-HG levels in BC samples. Additionally, our ROC analysis also classifies MIR497-HG as a good diagnostic biomarker for BC. Furthermore, the principal role of mucins is to shield and lubricate the luminal surfaces of epithelium-lined ducts in the human body. Several studies have demonstrated varying levels of expression in both membrane-associated mucins (MUC16, MUC4, MUC1) and secreted mucins (MUC6, MUC5B, MUC5AC, MUC2) in breast cancer tissues as opposed to non-neoplastic breast tissue. Various researches have highlighted numerous specific roles of mucins in BC, including their impact on tumor cell proliferation, invasion, and metastasis [69]. According to our lncRNA-mRNA analysis, MIR497-HG exhibits a substantial relationship with MUC2, MUC4, MUC5AC, and MUC5B. There is still insufficient information available regarding the role of MIR497-HG in (GC). In this study, for the first time, we measured the expression levels of MIR497-HG in GC samples compared to adjacent normal tissues. Our results indicate that the expression level of this lncRNA is decreased by about 24 times in cancerous samples than in non-cancerous ones (2^|4.6007| ≈ 24.52). Besides, the ROC analysis identifies MIR497-HG as a potentially efficient biomarker for GC. MIR497HG has been discovered to participate in multiple cancers and considered as a viable therapeutic target. Zhuang et al. state that bladder cancer progression is accelerated by MIR497HG knockdown, which strengthens interaction between the Hippo/Yap and TGF-β/Smad pathways [70]. Tang et al. found that the expression of MIR497HG was diminished in CRC tissues relative to normal ones [71]. Our results reveal that MIR497-HG is expressed at lower levels in CRC samples than normal ones, and our ROC analysis establishes it as a novel effective diagnostic biomarker for CRC.

The newly identified lncRNA NORAD is activated by DNA damage through p53 signaling and is essential for preserving genomic stability by limiting Pumilio protein activity on target mRNAs [72]. Consequently, the disruption of NORAD leads to genomic imbalances and aberrant ploidy. Alves-Vale et al. reported overexpressed NORAD levels in a batch of human epithelial BC samples [72]. However, our qRT-PCR data results were not adequately significant to fully support this claim. Chen et al. demonstrated that NORAD expression is upregulated in GC tissues and cell lines, and the inhibition of NORAD curbed GC cell proliferation via cell-cycle arrest. Their findings also highlighted the interaction between NORAD and miR-204-5p [73]. Similarly, our findings indicate that NORAD is overexpressed in GC samples compared to normal tissues, and ROC analysis identifies it as an adequate diagnostic biomarker for GC. Our bioinformatics analysis shows that NORAD interacts with MUC4, MUC2, MUC16, MUC3A, and MUC5AC, which are key genes involved in cancer related pathways like Defective GALNT12 Causes CRCS1, Defective GALNT3 Causes HFTC, and Defective C1GALT1C1 Causes TNPS. Wang et al. observed a significant elevation of NORAD in CRC tissues when compared to adjacent non-malignant tissues. Their research shows that higher concentrations of NORAD is tied to a greater extent of CRC tumor invasion, positive lymph node status, and a higher TNM stage [74]. Zhang et al. reported a positive correlation between NORAD and HIF-1α in CRC tissues. CRC cells subjected to hypoxic conditions demonstrated an increased capacity for vasculogenic mimicry (VM) and resistance to 5-fluorouracil (5-FU), which was associated with elevated NORAD expression [75]. Our validation of lncRNA NORAD's upregulation in CRC tissues agrees with earlier observations. Moreover, the ROC analysis shows that NORAD is a valuable diagnostic biomarker for CRC.

In our study, we identified positive correlations between TRIM29 and the lncRNAs MIR497-HG and NORAD across BC, GC, and CRC. The results showed robust associations between TRIM29 and MIR497-HG in BC and GC, and a moderate correlation in CRC. Similarly, strong correlations were observed between TRIM29 and NORAD in all three cancers. These findings imply that TRIM29 might interact with MIR497-HG and NORAD in cancer pathways, potentially influencing tumorigenesis. Further experimental validation is required to confirm the functional implications of these correlations.

Our study offers significant insights into the role of these genes in BC, GC, and CRC, laying a solid foundation for future research. Although we did not assess epigenetic modifications or environmental factors, these areas present promising opportunities for further exploration. Future research can build on our findings by translating these genomic insights into innovative diagnostic tools or therapeutic targets. Longitudinal studies will be particularly important for understanding how gene expression evolves over time and across different cancer stages. Additionally, while we faced limitations due to the lack of access to interaction assay tools, this underscores the potential for more in-depth investigations to further enhance our understanding and applications of these findings.

In conclusion, the findings of this study highlight the significant role of TRIM29 in interferon signaling, interferon gamma signaling, and cytokine signaling in immune system pathways. Any dysregulation in its expression can lead to oncogenic outcomes in BC, GC, and CRC. MIR497-HG and NORAD appear to interact with mRNAs, and indirectly participate in signaling pathways, which influence tumorigenesis in these malignancies. TRIM29, NORAD, and MIR497-HG hold potential as diagnostic biomarkers for GC and CRC, but only TRIM29 and MIR497-HG exhibit significant diagnostic relevance in BC. Our findings demonstrate strong correlations between TRIM29 and the lncRNAs MIR497-HG and NORAD across breast, gastric, and colorectal cancers, with the strongest associations between TRIM29 and MIR497-HG in breast and gastric cancers, hinting at a role in cancer progression.

Ethics approval: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of the Ethics Committee of Isfahan University of Medical Sciences.

Consent for publication: Informed consent was obtained from all individual participants included in the study.

Availability of data and materials: The datasets generated or analyzed during the current study are available in the GEO repository, GSE134359, GSE54129, and GSE81558.

Conflicts of interest: The authors declare that they have no competing interests.

Financial support and sponsorship: Not applicable.

Authors’ contribution: Nafiseh Sharifi, Ghazal Delgoshae, Behnaz Saeidi Palomi, Pooria Parvaz, Danial Khezrian, Niloofar Nasr Esfahani, Helia Ebrahimi, and Erfan Dehghan: Software, Formal Analysis, Investigation, Data Curation, Writing – Original Draft, Visualization; Mohammad Rezaei and Seyedeh Zahra Shirdeli: Writing – Review & Editing, Conceptualization, Methodology, Validation, Supervision; Mansoureh Azadeh: Writing – Review & Editing, Conceptualization, Methodology, Validation, Resources, Project Administration, Resources. Nafiseh Sharifi, Ghazal Delgoshae, Behnaz Saeidi Palomi, and Pooria Parvaz equally contributed to this study as the first authors. Danial Khezrian, Niloofar Nasr Esfahani, Helia Ebrahimi, and Erfan Dehghan equally contributed to this study as second authors. Mohammad Rezaei and Seyedeh Zahra Shirdeli equally contributed to this study as supervisors and third authors.

Bray F et al (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68(6):394–424
Latgé G et al (2018) Natural Antisense Transcripts: Molecular Mechanisms and Implications in Breast Cancers. Int J Mol Sci, 19(1)
Arabpour M et al (2021) The potential roles of lncRNAs DUXAP8, LINC00963, and FOXD2-AS1 in luminal breast cancer based on expression analysis and bioinformatic approaches. Hum Cell 34(4):1227–1243
Dai X et al (2015) Breast cancer intrinsic subtype classification, clinical use and future trends. Am J Cancer Res 5(10):2929–2943
Kim YA et al (2018) Doxorubicin-induced heart failure in cancer patients: A cohort study based on the Korean National Health Insurance Database. Cancer Med 7(12):6084–6092
Li N et al (2019) Global burden of breast cancer and attributable risk factors in 195 countries and territories, from 1990 to 2017: results from the Global Burden of Disease Study 2017. J Hematol Oncol 12(1):140
Wang Q et al (2020) Bioinformatics analysis of prognostic value of PITX1 gene in breast cancer. Biosci Rep, 40(9)
Siegel RL, Miller KD, Jemal A (2020) Cancer statistics, 2020. CA Cancer J Clin 70(1):7–30
Azadeh M et al (2023) Integrated High-Throughput Bioinformatics (Microarray, RNA-Seq, and RNA Interaction) and qRT-PCR Investigation of BMPR1B Axis as a Potential Diagnostic Biomarker of Isfahan Breast Cancer. Adv Biomed Res, 12: p. 120
Tirado-Hurtado I et al (2019) Helicobacter pylori: History and facts in Peru. Crit Rev Oncol Hematol 134:22–30
López MJ et al (2023) Characteristics of gastric cancer around the world. Crit Rev Oncol Hematol 181:103841
Sung H et al (2021) Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71(3):209–249
Puliga E et al (2021) Microsatellite instability in Gastric Cancer: Between lights and shadows. Cancer Treat Rev 95:102175
Cristescu R et al (2015) Molecular analysis of gastric cancer identifies subtypes associated with distinct clinical outcomes. Nat Med 21(5):449–456
Wagner AD et al (2017) Chemotherapy for advanced gastric cancer. Cochrane Database Syst Rev, 8(8): p. Cd004064.
Barani A et al (2023) Transcription Analysis of the THBS2 Gene through Regulation by Potential Noncoding Diagnostic Biomarkers and Oncogenes of Gastric Cancer in the ECM-Receptor Interaction Signaling Pathway: Integrated System Biology and Experimental Investigation. Int J Genomics, 2023: p. 5583231
Torre LA et al (2015) Global cancer statistics, 2012. CA Cancer J Clin 65(2):87–108
Li G, Zhu J, Zhai L (2023) Exploring molecular markers and drug candidates for colorectal cancer through comprehensive bioinformatics analysis. Aging 15(14):7038–7055
Siegel RL et al (2022) Cancer statistics, 2022. CA Cancer J Clin 72(1):7–33
Arnold M et al (2017) Global patterns and trends in colorectal cancer incidence and mortality. Gut 66(4):683–691
Seymour MT et al (2011) Chemotherapy options in elderly and frail patients with metastatic colorectal cancer (MRC FOCUS2): an open-label, randomised factorial trial. Lancet 377(9779):1749–1759
Walsh JM, Terdiman JP (2003) Colorectal cancer screening: Sci Rev Jama 289(10):1288–1296
Hou P et al (2021) LINC00460/DHX9/IGF2BP2 complex promotes colorectal cancer proliferation and metastasis by mediating HMGA1 mRNA stability depending on m6A modification. J Exp Clin Cancer Res 40(1):52
Van Grembergen O et al (2016) Portraying breast cancers with long noncoding RNAs. Sci Adv 2(9):e1600220
Modi A et al (2023) Integrative Genomic Analyses Identify LncRNA Regulatory Networks across Pediatric Leukemias and Solid Tumors. Cancer Res 83(20):3462–3477
Gil N, Ulitsky I (2020) Regulation of gene expression by cis-acting long non-coding RNAs. Nat Rev Genet 21(2):102–117
Shi T, Gao G, Cao Y (2016) Long Noncoding RNAs as Novel Biomarkers Have a Promising Future in Cancer Diagnostics. Dis Markers, 2016: p. 9085195
Hewson C, Morris KV (2016) Form and Function of Exosome-Associated Long Non-coding RNAs in Cancer. Curr Top Microbiol Immunol 394:41–56
Xu M et al (2018) The long noncoding RNA SNHG1 regulates colorectal cancer cell growth through interactions with EZH2 and miR-154-5. Mol Cancer 17(1):141
Liu PY et al (2019) The long noncoding RNA lncNB1 promotes tumorigenesis by interacting with ribosomal protein RPL35. Nat Commun 10(1):5026
Wang W et al (2020) Non-coding RNAs shuttled via exosomes reshape the hypoxic tumor microenvironment. J Hematol Oncol 13(1):67
Tian Y et al (2023) MIR497HG-Derived miR-195 and miR-497 Mediate Tamoxifen Resistance via PI3K/AKT Signaling in Breast Cancer. Adv Sci (Weinh) 10(12):e2204819
Lu Y et al (2017) lncRNA MIR100HG-derived miR-100 and miR-125b mediate cetuximab resistance via Wnt/β-catenin signaling. Nat Med 23(11):1331–1341
Zhuang C et al (2020) Silencing of lncRNA MIR497HG via CRISPR/Cas13d Induces Bladder Cancer Progression Through Promoting the Crosstalk Between Hippo/Yap and TGF-β/Smad Signaling. Front Mol Biosci 7:616768
Eissa S et al (2019) Measurement of Urinary Level of a Specific Competing endogenous RNA network (FOS and RCAN mRNA/ miR-324-5p, miR-4738-3p, /lncRNA miR-497-HG) Enables Diagnosis of Bladder Cancer. Urol Oncol, 37(4): p. 292.e19-292.e27
Sur S et al (2018) Association between MicroRNA-373 and Long Noncoding RNA NORAD in Hepatitis C Virus-Infected Hepatocytes Impairs Wee1 Expression for Growth Promotion. J Virol, 92(20)
Tichon A et al (2016) A conserved abundant cytoplasmic long noncoding RNA modulates repression by Pumilio proteins in human cells. Nat Commun 7:12209
Bian W et al (2020) Downregulation of LncRNA NORAD promotes Ox-LDL-induced vascular endothelial cell injury and atherosclerosis. Aging 12(7):6385–6400
Alves-Vale C et al (2023) Expression of NORAD correlates with breast cancer aggressiveness and protects breast cancer cells from chemotherapy. Mol Ther Nucleic Acids 33:910–924
Hatakeyama S (2011) TRIM proteins and cancer. Nat Rev Cancer 11(11):792–804
Hawthorn L et al (2006) Characterization of cell-type specific profiles in tissues and isolated cells from squamous cell carcinomas of the lung. Lung Cancer 53(2):129–142
Dyrskjøt L et al (2004) Gene expression in the urinary bladder: a common carcinoma in situ gene expression signature exists disregarding histopathological classification. Cancer Res 64(11):4040–4048
Wang L et al (2009) Oncogenic function of ATDC in pancreatic cancer through Wnt pathway activation and beta-catenin stabilization. Cancer Cell 15(3):207–219
Kosaka Y et al (2007) Tripartite motif-containing 29 (TRIM29) is a novel marker for lymph node metastasis in gastric cancer. Ann Surg Oncol 14(9):2543–2549
Kanno Y et al (2014) TRIM29 as a novel prostate basal cell marker for diagnosis of prostate cancer. Acta Histochem 116(5):708–712
Santin AD et al (2004) Gene expression profiles in primary ovarian serous papillary tumors and normal ovarian epithelium: identification of candidate molecular markers for ovarian cancer diagnosis and therapy. Int J Cancer 112(1):14–25
Hu T, Li C (2010) Convergence between Wnt-β-catenin and EGFR signaling in cancer. Mol Cancer 9:236
Sho T et al (2011) TRIM29 negatively regulates p53 via inhibition of Tip60. Biochim Biophys Acta 1813(6):1245–1253
Ai L et al (2014) TRIM29 suppresses TWIST1 and invasive breast cancer behavior. Cancer Res 74(17):4875–4887
Yanagi T et al (2018) Loss of TRIM29 Alters Keratin Distribution to Promote Cell Invasion in Squamous Cell Carcinoma. Cancer Res 78(24):6795–6806
Yi Q et al (2023) TRIM29 hypermethylation drives esophageal cancer progression via suppression of ZNF750. Cell Death Discov 9(1):191
Szklarczyk D et al (2019) STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res 47(D1):D607–d613
Kuleshov MV et al (2016) Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 44(W1):W90–W97
Jassal B et al (2020) The reactome pathway knowledgebase. Nucleic Acids Res 48(D1):D498–d503
Li JH et al (2014) starBase v2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data. Nucleic Acids Res 42(Database issue):D92–D97
Tang Z et al (2019) GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res 47(W1):W556–w560
Sticht C et al (2018) miRWalk: An online resource for prediction of microRNA binding sites. PLoS ONE 13(10):e0206239
Fukunaga T et al (2019) LncRRIsearch: A Web Server for lncRNA-RNA Interaction Prediction Integrated With Tissue-Specific Expression and Subcellular Localization Data. Front Genet 10:462
Zhou G et al (2019) NetworkAnalyst 3.0: a visual analytics platform for comprehensive gene expression profiling and meta-analysis. Nucleic Acids Res 47(W1):W234–W241
Livak KJ, Schmittgen TD (2001) Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2 – ∆∆CT Method. Methods 25(4):402–408
Nacht M et al (1999) Combining serial analysis of gene expression and array technologies to identify genes differentially expressed in breast cancer. Cancer Res 59(21):5464–5470
Schroder K et al (2004) Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol 75(2):163–189
Jiang T et al (2024) TRIM29 promotes antitumor immunity through enhancing IGF2BP1 ubiquitination and subsequent PD-L1 downregulation in gastric cancer. Cancer Lett 581:216510
Jiang T et al (2013) Up-regulation of tripartite motif-containing 29 promotes cancer cell proliferation and predicts poor survival in colorectal cancer. Med Oncol 30(4):715
Han J et al (2021) Transcriptional dysregulation of TRIM29 promotes colorectal cancer carcinogenesis via pyruvate kinase-mediated glucose metabolism. Aging 13(4):5034–5054
Du W et al (2022) IFNγ signaling integrity in colorectal cancer immunity and immunotherapy. Cell Mol Immunol 19(1):23–32
Matboli M et al (2021) Role of extracellular LncRNA-SNHG14/miRNA-3940-5p/NAP12 mRNA in colorectal cancer. Arch Physiol Biochem 127(6):479–485
Cheng Q et al (2024) The mechanism and therapeutic potential of lncRNA MIR497HG/miR-16-5p axis in breast cancer. BMC Womens Health 24(1):379
Mukhopadhyay P et al (2011) Mucins in the pathogenesis of breast cancer: implications in diagnosis, prognosis and therapy. Biochim Biophys Acta 1815(2):224–240
Zhuang C et al (2020) Silencing of lncRNA MIR497HG via CRISPR/Cas13d Induces Bladder Cancer Progression Through Promoting the Crosstalk Between Hippo/Yap and TGF-β/Smad Signaling. Front Mol Biosci 7:616768
Tang G et al (2022) LncRNA MIR497HG inhibits colorectal cancer progression by the miR-3918/ACTG2 axis. J Genet, 101
Alves-Vale C et al (2023) Expression of NORAD correlates with breast cancer aggressiveness and protects breast cancer cells from chemotherapy. Mol Ther Nucleic Acids 33:910–924
Chen X et al (2022) LncRNA NORAD mediates KMT2D expression by targeting miR-204-5p and affects the growth of gastric cancer. J Gastrointest Oncol 13(6):2832–2844
Wang L et al (2018) Overexpression of long noncoding RNA NORAD in colorectal cancer associates with tumor progression. Onco Targets Ther 11:6757–6766
Zhang L et al (2022) Induction of lncRNA NORAD accounts for hypoxia-induced chemoresistance and vasculogenic mimicry in colorectal cancer by sponging the miR-495-3p/ hypoxia-inducible factor-1α (HIF-1α). Bioengineered 13(1):950–962

The authors declare no competing interests.

Download PDF

Version 1

posted

You are reading this latest preprint version

Investigation of novel regulatory mechanisms of TRIM29 in BC, GC, and CRC patients: systems biology approach to find novel non-coding interactions

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials and Methods

2.1. Microarray analysis

2.2. Bioinformatics Analyses

2.3. Tissue collection and ethics statement:

2.4. RNA extraction, cDNA synthesis, and QRT-PCR experiment

2.5. Statistical analyses

3. Results

3.1. Microarray Analysis

3.2. Protein-Protein Interaction (PPI), Pathway Enrichment, and Gene

3.3. Gene Expression, Correlation (co-expression) and Survival Analyses

3.4. mRNA-miRNA Interaction, lncRNA-mRNA Interaction,

3.4. Expression analysis of TRIM29, MIR497HG, and NORAD

4. Discussion

5. Conclusion

Declarations

References

Additional Declarations

Status:

Version 1