Establishing bioinformatic benchmarks for in vitro engineered thymus from human pluripotent stem cells

Abstract

Advancements in regenerative medicine have facilitated the development of fully human artificial thymic organoids (ATO) derived from human pluripotent stem cells (hPSCs) in vitro. This study presents a comprehensive bioinformatic framework for characterizing the engineered thymic tissue. The computational analyses employed include Gene Set Enrichment Analysis (GSEA), Principal Component Analysis (PCA), and Transcription Factor Network (TFN) analysis. These techniques are applied to evaluate the transcriptional profile of the ATO, comparing it to its native counterpart. Our results demonstrate that while GSEA had limited benefit, it still provides insights in the analysis of apoptosis over time, indicating that the neural crest cells (NCC) and hPSC-hNCC-derived mesenchyme (NCC-M) are generally enhancing cell survival. Additionally, PCA reveals that our thymic epithelial progenitor cells (TEPC) are progressing along a trajectory towards primary thymic epithelial cells (TEC), and that our NCC and NCC-M samples are transitioning towards the primary mesenchyme over time. TFN analysis uncovers underlying molecular mechanisms governing cell identities. Our discoveries emphasize the distinct benefits and drawbacks of each methodology, and all three approaches contribute to different aspects of guiding the tissue engineering process, addressing questions about long-term survival, lineage commitment, and the fundamental cellular identity. These integrated methods not only enhance our comprehension of thymus tissue dynamics, but also hold potential for broader applications in tissue engineering and regenerative medicine.