Two Waves of Thymic Mimetic Cells Reveal How Immune Tolerance Evolved

Central immune tolerance — the process by which the thymus eliminates self-reactive T cells — depends partly on specialized 'mimetic cells' that adopt the identity of peripheral tissues within the thymic microenvironment. Despite being recognized histologically for over a century, the developmental origins and evolutionary history of these cells were largely unknown. This landmark Nature study by Nusser, Thomas, Zhang, and colleagues systematically maps how and when thymic mimetic cells arise in mice, and traces their evolutionary roots across vertebrate species.

Using single-cell RNA sequencing (scRNA-seq), bulk RNA-seq with competitive enrichment analysis, CRISPR–Cas9 lineage tracing, and RNA in situ hybridization, the authors profiled EPCAM+CD45− thymic epithelial cells (TECs) across embryonic day 15.5, birth (P0), and postnatal day 28 (P28) in mice. They identified 11 mimetic cell types and found they appear in two temporally distinct waves. Prenatal mimetic cells — including those mimicking muscle, ionocytes, goblet cells, and ciliated cells — are detectable around birth and even embryonically. Postnatal mimetic cells — including those resembling enterohepatic and skin keratinocyte populations — emerge only after birth, correlating with the expansion of postnatal TEC progenitor pools.

To dissect the genetic requirements, the team interrogated several genetic models: conditional deletion of Foxn1 and Ascl1, a hypomorphic FOXN1 variant, and overexpression of BMP4 and FGF7. These manipulations selectively altered postnatal mimetic cell populations while leaving prenatal populations largely intact, confirming the two waves respond to distinct molecular signals. Lineage barcoding showed that while both canonical and mimetic TECs express Foxn1 during development, postnatal mimetic cells are preferentially derived from postnatal progenitors, whereas early mimetic cells like ciliated and muscle cells branch earlier in development and are less dependent on FOXN1.

A compelling evolutionary dimension emerged from experiments replacing mouse Foxn1 with orthologous genes from the cephalochordate amphioxus (Foxn4) and a cartilaginous fish (Foxn4 and Foxn1). Prenatal mimetic cells such as ciliated cells formed even in the absence of FOXN1, whereas postnatal mimetic cells like enterohepatic cells required vertebrate-specific FOXN1. Strikingly, analysis of the thymus of cartilaginous fishes and the thymoid of lampreys — jawless vertebrates with an alternative (VLR-based) adaptive immune system — revealed cells expressing liver-specific genes such as transthyretin, suggesting that tolerance to evolutionarily ancient tissues predates the jawed vertebrate lineage.

Collectively, these findings propose an evolutionary model in which successive reprogramming of thymic epithelial genetic networks — anchored by the emergence and specialization of FOXN1 — enabled the gradual incorporation of new tissue-mimetic identities into the thymic microenvironment. This would have allowed immune tolerance to keep pace with vertebrate-specific tissue innovations such as the liver. For longevity and autoimmune medicine, understanding how mimetic cell populations are established and maintained may inform strategies to restore or enhance central tolerance in aging thymi or in autoimmune disease.