Year: 2026
Type: dissertation
Abstract: Colorectal cancer (CRC) is ranked as the third most commonly diagnosed cancer worldwide and represents the second leading cause of cancer-related mortality. Despite significant progress in CRC treatment, patient outcomes are frequently compromised by intertumoral heterogeneity and limited therapeutic specificity, highlighting the need to identify molecular drivers of CRC and to elucidate the mechanisms underlying tumor initiation and progression. While extensive insights into CRC pathogenesis have been obtained through the analysis of protein-coding genes, the regulatory and functional contributions of long non-coding RNAs (lncRNAs) remain comparatively underexplored. LncRNAs, typically defined as transcripts longer than 200 nucleotides that lack protein-coding potential, represent promising targets for exploration. LncRNAs modulate gene expression, genome organization, and cellular architecture, thereby influencing a broad range of biological processes, and their dysregulation has been implicated in disease development, including cancer. Although differential gene expression studies have successfully identified functional lncRNAs, such approaches are biased toward abundant transcripts and often lack experimental validation. Consequently, the functional repertoire of lncRNAs in CRC remains largely uncharacterized. This project aimed at identifying functional lncRNAs driving CRC growth. To address this objective, a lncRNA-focused CRISPR interference (CRISPRi) dropout screen was performed in CRC cell lines and patient-derived 3D spheroid models representing diverse molecular backgrounds, using a library targeting 675 lncRNA loci broadly expressed across cancer cell lines. Through this approach, 125 growth-regulatory lncRNAs were identified, of which 11 were consistently depleted across all screened models, suggesting their universal role as subtype-independent regulators of CRC cell growth. To benchmark screen specificity and enable robust target prioritization, locus composition and neighboring gene essentiality were systematically assessed, providing a framework to distinguish lncRNA-driven phenotypes from effects mediated by adjacent protein-coding genes. Among the depleted candidates, three lncRNAs with previously unknown functions in CRC, namely PARTICL, ADNP-AS1, and PDCD4-AS1, were validated using CRISPRi, antisense oligonucleotides (ASOs), and RNA interference. Based on a consistent growth dependency across models, a mixed nuclear–cytoplasmic distribution, and genomic context, PDCD4-AS1 was prioritized for in-depth functional analysis. The functional and mechanistic role of the prioritized candidate PDCD4-AS1 was investigated using translational CRC models, transcriptomic and proteomic profiling, and RNA pulldown analyses. Here, PDCD4-AS1 was validated as a CRC growth regulator independent of its neighboring protein-coding gene PDCD4. Moreover, PDCD4-AS1 depletion revealed disruption of mitosis-related transcriptomic and proteomic programs in patient-derived 3D models. These effects were recapitulated through functional assays, which demonstrated G2/M cell cycle arrest and apoptosis induction following shRNA-mediated PDCD4-AS1 knockdown. Mechanistically, an optimized RNA pulldown approach revealed direct interaction of PDCD4-AS1 with TPX2 mRNA, which encodes a common essential gene and a mitotic spindle assembly factor, and might serve as important downstream target of PDCD4-AS1. This observation was further supported by in silico RNA-RNA interaction predictions. Importantly, dysregulation of TPX2 provides a plausible mechanistic basis for the mitotic defects observed upon PDCD4-AS1 depletion. Consistently, PDCD4-AS1 knockdown resulted in reduced TPX2 protein levels, and PDCD4-AS1 expression positively correlated with TPX2 protein abundance across CRC cell lines, supporting a PDCD4-AS1:TPX2 axis operating in trans. Finally, PDCD4-AS1 inhibition was shown to suppress the growth of patient-derived organoids and to impair tumor development in vivo. Accordingly, a high PDCD4-AS1/PDCD4 expression ratio, together with elevated TPX2 protein levels, delineated a highly proliferative CRC subtype associated with poor clinical outcomes. In summary, this study demonstrates how CRISPRi screens, coupled with rigorous functional validation, can uncover bona fide lncRNA dependencies in CRC. By delineating PDCD4-AS1's role in multiple preclinical models and revealing mechanistic insights into mitotic control, a rationale for potential clinical translation of lncRNA-based interventions in CRC is provided.
Author: Kufrin, Vida
Institution: Technische Universität Dresden
Language: English
Subfield: Cancer Research
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7164235978
