Gaurab Kumar Jha, Azima Fatima, Alok Srivastava, Swathi Kaliki, Rani Pallavi
Transcriptomic heterogeneity in RB appears to be shaped by epifactor expression, suggesting that RB heterogeneity may be encoded within its epigenetic machinery. Epifactor profiling, clustering, and complex-level analysis identified key epigenetic factors that could serve as potential biomarkers and therapeutic interventions.
PURPOSE: Transcriptional heterogeneity observed in retinoblastoma (RB) is unexplained by the limited genetic alterations identified. Epifactors, which can modulate transcriptional activity and oncogenic pathways, may contribute to this heterogeneity. However, knowledge regarding the role of epigenetic machinery in RB remains limited. Here, we analyze epifactors' expression to uncover dysregulations and their contribution to RB tumor heterogeneity.
METHODS: Transcriptomic datasets of primary RB tumors and normal fetal retina were obtained from the NCBI Gene Expression Omnibus (NCBI-GEO). Differentially expressed epifactors were analyzed using edgeR, and the identified epifactors were functionally annotated. Non-negative matrix factorization (NMF) was used for clustering based on epifactor's expression.
RESULTS: We found 133 epifactors differentially expressed in RB compared with the healthy retina. In particular, epifactors involved in histone phosphorylation and RNA degradation were predominantly upregulated, whereas those involved in RNA methylation regulation were largely downregulated. Based on epifactor expression, NMF-based clustering stratified patients with RB into two clusters with distinct biological signatures. Cluster 1 patients showed enrichment for cell-cycle and MYC programs, and cluster 2 showed enrichment for interferon- and inflammatory-related pathways. Clusters were found to be associated with specific epigenetic factors that formed cluster-specific functional complexes, such as centromere/chromosomal passenger complexes in the proliferative cluster 1 and nuclear receptor/interferon-associated complexes in the immunologically active cluster 2.
CONCLUSIONS: Transcriptomic heterogeneity in RB appears to be shaped by epifactor expression, suggesting that RB heterogeneity may be encoded within its epigenetic machinery. Epifactor profiling, clustering, and complex-level analysis identified key epigenetic factors that could serve as potential biomarkers and therapeutic interventions.