Transcriptome sequencing of neonatal thymic epithelial cells (original) (raw)

In order to gain novel insights into thymus biology, we analysed the whole transcriptome of cortical and medullary thymic epithelial cells (cTECs and mTECs) and of skin epithelial cells (ECs). Consistent with their ability to express ectopic genes, mTECs expressed more genes than other cell populations. Out of a total of 15,069 genes expressed in TECs, 25% were differentially expressed by at least 5-fold in cTECs vs. mTECs. Genes expressed at higher levels in cTECs than mTECs regulate numerous cell functions including cell differentiation, cell movement and microtubule dynamics. Many positive regulators of the cell cycle were overexpressed in skin ECs relative to TECs. Our RNA-seq data provide novel systems-level insights into the transcriptional landscape of TECs, highlight substantial divergences in the transcriptome of TEC subsets and suggest that cell cycle progression is differentially regulated in TECs and skin ECs. I n all vertebrates, the thymus is necessary and sufficient for production of classic adaptive T cells 1,2. There are no thymus substitutes in the animal kingdom and T cells generated extrathymically (e.g., in oncostatin Mtransgenic mice) are poorly functional: they cause severe autoimmunity and are unable to eliminate pathogens 2-5. The key components of the thymus are cortical and medullary thymic epithelial cells (cTECs and mTECs) which play several essential functions 6. During their intrathymic journey, which takes around 3 weeks, thymocytes undergo numerous reciprocal interactions with TECs located in seven functional zones 7-9. TECs produce chemokines that attract bone marrow derived hematopoietic progenitors, as well as interleukin (IL)-7 and the notch ligand DLL4 that induce thymocyte proliferation and differentiation 10,11. Furthermore, cTECs and mTECs express unique sets of ligands that mould the repertoire of antigen receptors expressed by thymocytes 6,12,13. The cells that induce the positive selection of thymocytes are primarily cTECs, whereas mTECs are instrumental in negative selection. Recent studies have highlighted several factors that regulate TEC development, maintenance and function including microRNAs, the transcription factor Foxn1 and Wnt signaling 14-18. Nonetheless, despite the capital role of TECs, our understanding of TEC biology is quite rudimentary. For instance, it is not yet known whether cTECs and mTECs are maintained by unipotent or bipotent progenitors in postnatal thymi, and what might be the extent of divergence in the functional program of these two TEC populations 6,7. In addition, while TECs display considerable proliferative potential 19 , it remains unclear why the number of TECs decreases rapidly with age, thereby leading to progressive thymic insufficiency 5,20,21. The transcriptome is a critical component of systems-level understanding of cell biology and it can be reliably tackled in its entirety in freshly harvested primary cells. In line with this, microarray analyses of several immune cell populations by the Immunological Genome Project Consortium have yielded fundamental insights into the biology of lymphocytes, dendritic cells and macrophages (http://www.immgen.org/index\_content.html). As a first step to gain novel insights into TEC biology, we therefore decided to analyse the whole transcriptome of cTECs, mTECs and skin epithelial cells (ECs). We inferred that including skin ECs in our analyses would enable us to better appreciate the extent of divergence between the transcriptomes of mTECs and cTECs. In addition, we surmised that comparing the transcriptome of skin ECs vs. TECs might yield some clues as to why precocious agerelated hypocellularity impinges on TECs but not skin ECs. We elected to analyse gene expression using RNA-seq rather than microarrays because RNA-seq has higher sensitivity and dynamic range coupled to lower technical variations 22,23. We report that the transcriptomes of mTECs and cTECs present numerous substantial differences that may have far-reaching biological consequences. In addition, we found that many positive regulators of cell division are repressed in TECs relative to skin ECs. Our RNA-seq data offer a valuable resource to the community that can be mined to explore multiple questions.