Exploration and validation of ceRNA regulatory networks in colorectal cancer based on associations whole transcriptome sequencing

Lulu Zhang, Yulei Li, Chao Fu, Lixia Yang, Gang Li, Yiyang Wu, Huanjun Tong, Guojiang Tian, Kaifang Wang, Jun Wang, Xiaojiang Ying, Zhenjun LiScientific 2024 · DOI

Abstract

Colorectal cancer (CRC) is a significant global health challenge characterized by high morbidity and mortality. Understanding the complex regulatory landscape of the tumor transcriptome is crucial for identifying novel biomarkers and therapeutic targets. This study utilized whole-transcriptome sequencing on 10 pairs of CRC and para-cancer (CRCP) tissues to map competing endogenous RNA (ceRNA) networks involving mRNAs, lncRNAs, circRNAs, and miRNAs. We identified 2,465 differentially expressed (DE) mRNAs, 77 DE miRNAs, 2,852 DE lncRNAs, and 1,477 DE circRNAs. Integration with TCGA and GEO databases highlighted KPNA2 as a critical oncogenic driver. Functional validation through siRNA-mediated depletion confirmed that KPNA2 is essential for CRC cell viability, proliferation, and migration, validating the reliability of our sequencing data and offering new insights into the molecular mechanisms of CRC progression.

Introduction: The ceRNA Hypothesis in Colorectal Cancer

Colorectal cancer remains a leading cause of cancer-related mortality worldwide. While traditional research often focuses on protein-coding genes, the "transcriptome dark matter"—consisting of non-coding RNAs (ncRNAs)—plays a pivotal role in gene regulation. The ceRNA hypothesis proposes that ncRNAs, such as long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs), act as "sponges" for microRNAs (miRNAs). By sequestering these miRNAs, ncRNAs prevent them from silencing their target mRNAs, effectively creating a complex regulatory crosstalk. Mapping these networks is essential for uncovering the mechanisms behind CRC development and identifying new therapeutic targets.

Methods and Materials

The study employed a multi-layered approach combining high-throughput sequencing with functional validation. Whole Transcriptome Sequencing (WTS) was performed on 10 pairs of fresh CRC and adjacent para-cancer tissues collected from Shaoxing People’s Hospital.

Clinical and pathological characterization of CRC samples, including surgical specimens, endoscopy, CT imaging, and histology.

Regulatory networks were constructed using co-expression analysis and target prediction algorithms like miranda. To ensure clinical relevance, the sequencing data were cross-referenced with public repositories, including TCGA-COAD and GEO (GSE33113 and GSE39582), to identify genes strongly correlated with overall survival (OS). Functional validation of the candidate gene KPNA2 was conducted in CRC cell lines (HCT116 and RKO) using siRNA-mediated knockdown, followed by Western blotting, CCK8 viability assays, EdU proliferation tests, and scratch assays to observe phenotypic effects.

The Transcriptomic Landscape of CRC

The sequencing effort revealed a massive shift in the RNA profile of CRC tissues compared to normal controls. The researchers identified thousands of differentially expressed transcripts, which formed the basis of the constructed ceRNA regulatory networks (mRNA-miRNA-lncRNA and mRNA-miRNA-circRNA).

Heatmap and KEGG enrichment analysis showing the transcriptomic landscape and functional associations in CRC vs. CRCP samples.

Pathway enrichment analysis (KEGG) indicated that these networks are heavily involved in the cell cycle, ECM-receptor interaction, and various metabolic pathways. These findings suggest that the identified ceRNA axes are central to the fundamental hallmarks of cancer, modulating processes such as smooth muscle contraction and plasma membrane integrity.

Key Findings: KPNA2 as a Master Regulator

Among the thousands of candidates, KPNA2 (Karyopherin Subunit Alpha 2) emerged as a top-tier oncogenic driver. Its expression was consistently elevated in CRC tissues across both the primary study cohort and public databases. High KPNA2 expression was a strong predictor of worse overall and disease-specific survival in TCGA-COAD patients.

Functional assays confirmed the bioinformatic predictions:

  • Reduced Proliferation: Silencing KPNA2 significantly decreased the viability and proliferative capacity of CRC cells, as confirmed by CCK8 and EdU assays.
  • Inhibited Migration: Scratch assays demonstrated that KPNA2 depletion hindered the migratory ability of tumor cells.
  • Data Reliability: The high degree of correlation between the RNA-seq findings and experimental results underscores the robustness of the whole-transcriptome sequencing approach.

Discussion and Future Directions

This study provides a comprehensive map of the ceRNA regulatory economy in colorectal cancer. By identifying KPNA2 and its associated regulatory axes, the researchers have pinpointed potential therapeutic vulnerabilities. However, limitations remain. The study focused primarily on the transcriptome; integrating genomic data could reveal the upstream mutations driving these RNA changes. Additionally, while in vitro results are promising, future in vivo studies using animal models are necessary to confirm the therapeutic potential of targeting these networks in a clinical setting. These findings lay the groundwork for developing more precise biomarkers and targeted treatments for CRC.

Summary updated

Abstract

Colorectal cancer (CRC) is a wide-spread gastrointestinal cancer that is associated with augmented morbidity and mortality, and we do not yet have a deep understanding of its epidemiology and carcinogenicity. The transcriptome can reveal the complexity and heterogeneity of tumors and uncover new biomarkers or treatment options. In this study, we identified messenger RNAs (mRNAs), long non-coding RNAs (lncRNAs), round RNAs (circRNAs), and microRNAs (miRNAs) using whole-transcriptome sequencing and generated competing endogenous RNA (ceRNA) modulatory axes. We conducted whole transcriptome sequencing on 10 CRC and para-cancer (CRCP) samples and discovered 2465 differentially expressed (DE) mRNAs (DEmRNAs), 77 DE miRNAs (DEmiRNAs). 2852 DE lncRNAs (DElncRNAs) and 1477 DE circRNAs (DEcircRNAs). In addition, utilizing co-DE analysis, we generated the ceRNA axis. Subsequently, we employed the ceRNA axis to identify essential genes and corresponding associations with lncRNAs, circRNAs, and miRNAs in CRC. ceRNA regulatory network including mRNA-miRNA-lncRNA and mRNA-miRNA-circRNA. These modulatory axes potentially modulate the positive regulation of smooth muscle contraction, melanosome, plasma membrane, integral plasma membrane component and so on. Finally, the results of RNA sequencing (RNA-SEQ) were combined with the TCGA and GEO databases, and the DEGs strongly correlated with the TCGA-COAD overall survival (OS) as estimated by univariate cox and logarithmic rank analyses were cross-analyzed, and the co-upregulated DEGs were screened. Among the many DEs, KPNA2 was chosen for additional analysis. Using invitro experimentations, western blot, CCK8, EdU and other experiments were performed to verify the results. We found siRNA-based KPNA2 depletion reduces bladder cancer cells' viability, migratory, and proliferative activities, which showed that the DEmRNA profiles were comparable to the sequencing information, confirming that the sequencing data were very reliable. These evidences highlight the ceRNA regulatory mechanisms in CRC and will aid future research into the molecular mechanisms behind colorectal cancer prevention and treatment.
BibTeX
@inproceedings{zhang2024exploration,
  author = {Lulu Zhang and Yulei Li and Chao Fu and Lixia Yang and Gang Li and Yiyang Wu and Huanjun Tong and Guojiang Tian and Kaifang Wang and Jun Wang and Xiaojiang Ying and Zhenjun Li},
  title = {{Exploration and validation of ceRNA regulatory networks in colorectal cancer based on associations whole transcriptome sequencing}},
  booktitle = {Scientific},
  year = {2024},
  doi = {10.1038/s41598-024-71465-5},
}
Figure