Author: Kyle J. Wang, Yading Yuan 👨🔬
Affiliation: Bergen County Technical High School, Columbia University Irving Medical Center, Department of Radiation Oncology 🌍
Purpose: Genotoxic cancer therapies inevitably damage normal cells, particularly circulating hematopoietic cells, posting a risk for therapy-induced leukemia. This study aims to develop an automated image analysis pipeline to quantitatively assess the effects of genotoxic cancer therapies, including radiation, on rRNA transcription in a population of normal cells, providing a sensitive measurement for the hematopoietic impact of cancer therapy.
Methods: DDX21 is an RNA helicase associated with newly synthesized rRNA. Using DDX21 as a marker for ongoing rRNA transcription, we measured the time-dependence change of rRNA transcription following ionizing radiation where diffusion and decrease of DDX21 staining indicate a disruption in rRNA transcription. Cellpose, a foundational AI model for cellular segmentation, was used to segment DAPI-stained nuclei on DAPI images. Then, the segmented masks were transferred to co-registered DDX21 images, and rule-based image analysis was developed to exclude the overlapping and partial nuclei. Finally, DDX21 positive-nuclear-area ratio and integrated intensity were determined at time points before radiation, 0, 0.5, 1, 3, or 6 hours after 4 Gy of fractionated radiation, respectively.
Results: The pipe enables high-throughput analyses of >100 cells within 1 minute (compared to hours of manual quantification) and reveals a heterogeneous distribution of DDX21 positive nucleoli area and DDX21 intensity before radiation. These slightly increased immediately after radiation, gradually decreased at 0.5 hour, further decayed to less than 10% of normal levels at 1hour, and recovered by 3-6 hours after radiation. The results revealed a delayed impact of 4 Gy radiation on rRNA synthesis.
Conclusion: This automated image analysis platform enables rapid, unbiased quantification of rRNA transcriptional activity across hundreds of cells and can be adapted for other nuclear biomarkers (e.g., EU or NPM1) and offers a robust approach for assessing how cancer therapies impact ribosomal biogenesis, paving the way for evaluating broader therapeutic effects on hematopoietic health.