Abstract
Autoimmune regulator (AIRE) enables medullary thymic epithelial cells (mTECs) to express a broad but selective repertoire of tissue-restricted antigens (TRAs), thereby supporting central T-cell tolerance. Recent studies showing that AIRE forms transcriptionally active condensates provide a framework for understanding how chromatin recognition, enhancer-associated cofactor recruitment, three-dimensional genome organization and transcriptional elongation are coordinated during TRA induction. In this view, sparse TRA expression reflects the low probability that a given locus reaches a permissive chromatin and cofactor state, whereas high output may result from local enrichment of BRD4, P-TEFb, Mediator-associated co-activators and elongation machinery after activation. Recent findings on thymic mimetic cells, human thymic spatial organization and peripheral RORγt-associated antigen-presenting cells are discussed as related contexts, with particular attention to the current lack of direct evidence for mTEC-like AIRE condensate mechanisms outside mTECs. Together, these studies place AIRE condensates within a broader regulatory landscape of chromatin gating, transcriptional kinetics and mTEC state diversity.
1 Introduction
Central T-cell tolerance depends on exposing developing thymocytes to a broad but regulated spectrum of self-antigens within the spatial and temporal constraints of the thymic medulla. This antigenic exposure supports deletion of high-affinity autoreactive clones and contributes to the generation of regulatory T cells. AIRE is indispensable to this process; loss of AIRE disrupts central tolerance and pathogenic AIRE variants cause autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy, also known as autoimmune polyendocrine syndrome type 1 (APECED/APS-1), a disorder characterized by chronic mucocutaneous candidiasis, hypoparathyroidism, Addison’s disease, and related manifestations (–).
The best-characterized site of AIRE function is the mTEC compartment. mTECs ectopically express many genes normally associated with peripheral tissues, thereby providing antigenic substrates for negative selection and regulatory T-cell induction. Population-level transcriptomic studies show that thymic epithelial cells can transcribe up to 19,293 protein-coding genes and that mature medullary epithelial cells contain roughly 3,980 AIRE-positively regulated TRA genes (). TRA expression, however, is not a uniform catalogue expressed by every mTEC. Individual antigens are detected in only ~1%-3% of mTECs, are organized into recurrent co-expression modules, and can reach markedly higher expression once activated in a single cell (, ). Thus, a central question is how AIRE distributes rare but high-amplitude antigen expression across the mTEC population.
This expression logic is difficult to explain by a simple model in which AIRE is recruited to isolated cis-regulatory elements and then activates nearby genes in a linear fashion. AIRE is not a canonical sequence-specific DNA-binding transcription factor. Instead, its activity appears to depend on chromatin state, enhancer-associated cofactor density, protein interaction networks, and transcriptional elongation. DNA topoisomerase I (TOP1), which relaxes transcription-associated DNA supercoiling and influences local chromatin topology, stabilizes parts of the AIRE interaction network: TOP1 perturbation weakens interactions between AIRE and multiple transcription-related partners (). Genome-wide localization and proteomic studies further suggest that AIRE acts preferentially within regulatory-factor-rich chromatin environments rather than at isolated recognition sites ().
These findings have shifted attention from simple recruitment models toward AIRE-associated transcriptional assemblies. Earlier reviews have established the broader background of AIRE chromatin plasticity and negative selection (, ), AIRE-associated autoimmunity (), mimetic-cell biology (–), and thymic-medulla remodeling (, , ). Building on this literature, the sections below integrate the controlled-condensate mechanism described by Huoh et al. () and its early interpretation in the field () with enhancer preference, chromatin looping, pause release, PRC2-associated restraint, and sparse yet amplified TRA expression in mTECs. This multiscale working model is summarized in Figure 1.
Figure 1
2 Central tolerance through the lens of transcriptional condensates
2.1 AIRE in the classical framework of central tolerance
In the classical framework, AIRE is the key regulator of ectopic TRA expression in mTECs and a core determinant of central immune tolerance (, ). Broader reviews of thymic tolerance and AIRE-dependent negative selection provide the classical background (, ). Recent work further emphasizes that central tolerance emerges from functionally diverse mTEC subsets rather than from one uniform epithelial population (). Pathogenic AIRE mutations across diverse populations reinforce its upstream importance in preventing APS-1 (, ). Functional genomic studies show that AIRE controls a large but selective transcriptional program (, ). A frequently cited human mTEC transcriptomic study reported that sorted human mTEC populations analyzed by bulk RNA-seq expressed an average of 20,426 genes (). This population-level result should therefore not be interpreted as evidence that individual AIRE-positive mTECs express approximately 20,000 genes; the analyzed populations were not restricted to AIRE-positive cells. At single-cell resolution, the defining feature of AIRE-dependent antigen display is sparse, heterogeneous, and modular TRA output rather than per-cell comprehensiveness.
2.2 A condensate-based view of AIRE-dependent TRA expression
AIRE function therefore appears to extend beyond recruitment to individual cis-elements. Available studies indicate that AIRE preferentially associates with extended chromatin intervals enriched for regulatory factors, including regions with super-enhancer-like features that may organize both nearby and distal transcription start sites (, , ). The relevant unit of regulation is therefore better viewed as a local regulatory environment rather than a single promoter or binding event.
The single-cell phenotype makes this point especially clear. An individual TRA is typically detected in only ~1%-3% of mTECs (), yet AIRE-induced genes are not expressed as random noise. Approximately 51% of Aire-induced transcripts can be assigned to 19 recurrent co-expression clusters, each containing about 33–114 transcripts (median, 57), with strong within-cluster correlation and reduced clustering in the absence of Aire (). Once an AIRE-dependent gene is activated, expression can be much higher than the population mean; in one analysis of 174 individual mTECs, detected AIRE-induced genes were expressed at approximately 16-fold higher levels than the population average (). These features point to a thresholded system in which rare activation events are locally amplified.
Super-enhancer biology and transcriptional condensate models provide a useful framework for interpreting this threshold-like behavior. Super-enhancers are extended enhancer clusters with high co-activator density; by concentrating BRD4, MED1, and other transcriptional regulators, they may increase both the probability and amplitude of transcription from selected loci (, ). Huoh et al. showed that AIRE can assemble enhancer-associated condensates that promote local transcriptional activity and connect genomic loci across chromosomes (). This controlled assembly requires coordinated action of the CARD, PHD1, and C-terminal tail (CTT) domains. These findings connect AIRE condensate formation to chromatin topology, elongation control, and single-cell TRA output without requiring every AIRE-bound region to be interpreted as a condensate.
3 Structural basis of AIRE condensate assembly
3.1 Domain architecture of AIRE and its assembly interfaces
AIRE has a modular architecture that supports multivalent interactions and regulated nuclear assembly. Human AIRE is approximately 545 amino acids long and contains an N-terminal caspase recruitment domain (CARD), a SAND domain, two plant homeodomain (PHD) zinc fingers, and a C-terminal tail (CTT), separated by low-complexity regions that contribute to nuclear localization, chromatin recognition, and partner recruitment (). The CARD provides a self-assembly interface that supports AIRE nuclear puncta formation; mutations that disrupt CARD-mediated assembly impair nuclear localization and transcriptional output (, ). CARD filament mutants such as R16A/E18A and D35A/D37A lose nuclear foci and transcriptional activity, indicating that polymerization and function are tightly coupled in these experimental systems (). Pathogenic APS-1/APECED variants across conserved domains, including CARD and PHD1, further link structural integrity to downstream TRA regulation (, ).
Recent mechanistic work places these domains into a coherent assembly model. AIRE can form enhancer-associated condensates that increase local transcriptional activity and connect loci across chromosomes (). CARD-mediated polymerization supplies the assembly interface, PHD1 contributes chromatin-state engagement, and the CTT binds co-activators such as CBP/p300 to bias AIRE toward appropriate regulatory sites. This domain cooperation explains why AIRE assembly must be controlled rather than simply maximized: productive condensates require both multimerization capacity and correct chromatin/cofactor context.
3.2 Histone sensing and chromatin interactions underlying targeting
AIRE targeting is therefore best understood as chromatin-context sensing rather than canonical sequence-specific DNA recognition (). AIRE-PHD1 preferentially recognizes unmethylated H3K4 (H3K4me0), with reported binding affinities in the micromolar range (Kd approximately 5 μM) (, ). In vivo and genome-wide analyses indicate that the histone H3-binding module is required for AIRE-dependent gene regulation and tolerance induction, although it almost certainly operates with additional targeting mechanisms (). Functionally, PHD1 links AIRE to relatively silent or insufficiently activated loci, providing an epigenetic entry point for TRA activation in mTECs.
Histone recognition is only one layer of targeting. AIRE recruitment is also coordinated with DNA damage repair-related pathways. DNA-dependent protein kinase (DNA-PK) is a nuclear serine/threonine kinase composed of the catalytic subunit DNA-PKcs and the DNA-end-binding Ku70/Ku80 heterodimer, and it acts as a core effector of non-homologous end joining and the broader DNA double-strand break repair response. H3K4me0 and DNA-PK together promote AIRE recruitment to TRA loci, and loss of DNA-PKcs disrupts AIRE assembly and transcriptional activity at chromatin and episomal targets (). Recent work suggests that Z-DNA-associated features may contribute to AIRE target selection during thymic T-cell tolerization (). Conversely, although AIRE contains a SAND domain, mechanistic work indicates that this domain lacks residues required for direct canonical DNA recognition (). The emerging picture is therefore a layered targeting system in which histone state, DNA structure, DNA-PK-associated repair pathways, enhancer context and cofactor availability jointly influence where productive AIRE assemblies may form.
3.3 Gating multimerization and condensate assembly
CARD-mediated polymerization provides the structural basis for AIRE nuclear puncta, but productive condensate formation depends on how this polymerization is gated. Broad PHD1-mediated engagement with H3K4me0 may keep AIRE distributed across many potential genomic sites, limiting inappropriate polymerization. In contrast, CTT interaction with co-activators such as CBP/p300 biases AIRE toward enhancer-rich environments and permits controlled CARD-dependent assembly (). Thus, CARD supplies assembly capacity, PHD1 contributes chromatin dispersal and targeting, and the CTT links AIRE to co-activator-rich regulatory neighborhoods. This division of labor explains how AIRE can combine broad genomic reach with selective local activation.
Assembly is coupled to transcriptional kinetics. AIRE promotes the transition from promoter-proximal pausing to productive elongation by engaging BRD4 and positive transcription elongation factor b (P-TEFb), a CDK9-containing kinase complex that releases paused RNA polymerase II into elongation (). This coupling depends on post-translational modification of AIRE within the CARD region and is directly linked to tolerance induction. APECED-associated mutations can disrupt AIRE-P-TEFb coupling and impair downstream transcriptional output (). Thus, AIRE condensates may be viewed as regulatory states that integrate partner recruitment, chromatin context, and elongation control rather than as passive nuclear clusters (Figure 2).
Figure 2
4 AIRE condensates in super-enhancer contexts
4.1 Preference for super-enhancer occupancy in mTECs
Functionally relevant AIRE condensates are most likely to occur within defined regulatory contexts rather than as nonspecific nuclear aggregates. ChIP-seq profiling of mTEChi cells identified 42,124 AIRE-occupied sites, with more than 75% overlap between biological replicates (). Their distribution supports the view that AIRE enrichment is shaped less by classical sequence recognition than by chromatin environments with locally accumulated regulatory factors (, ).
Super-enhancer analyses support this interpretation. H3K27ac-based ROSE analysis identified 1,170 super-enhancer-like regions in mTEChi cells, with an average length of approximately 30 kb. AIRE density was higher within these regions than within conventional enhancers (, ), and loss of AIRE reduced H3K27ac density and chromatin accessibility at the same regulatory regions. In comparative ATAC-seq analyses, the accessibility difference between Aire+/+ and Aire-/- mTEChi cells at super-enhancers was highly significant (P = 6.6 x 10-52) (). These data support a close association between AIRE and high-density regulatory platforms, although they do not by themselves prove that every AIRE-occupied super-enhancer forms a condensate.
These enhancer-rich regions may provide a favorable environment for assembly of the AIRE partner network. AIRE-associated complexes include transcriptional, chromatin, topological, and RNA-processing components rather than a single uniform cofactor set. TOP1 appears to act as an important stabilizing component of the AIRE-associated interaction network: shRNA-mediated TOP1 knockdown weakens interactions between AIRE and many transcription-related partners while having a smaller effect on splicing-related factors. TOP1 and γH2AX also show partial co-localization with AIRE at enhancer-rich regions in mTEChi cells, suggesting that these loci may integrate AIRE enrichment, partner assembly, and local topological regulation (, –).
4.2 Three-dimensional genome remodeling and enhancer–promoter coupling
AIRE-associated super-enhancer regions also connect AIRE function to three-dimensional genome organization. In general models of genome organization, loop anchors and topological hubs can increase enhancer-promoter contact frequency and support transcriptional output. Mechanistic and modeling studies of loop extrusion provide the broader physical context for this idea (, ). Developmental and in vitro studies further support the physical plausibility of cohesin- or condensin-driven extrusion (, ). In mTECs, AIRE-associated super-enhancers are linked to local and distal transcription start sites, and AIRE can reshape chromatin looping by reducing CTCF occupancy at domain boundaries while favoring cohesin accumulation on super-enhancers (, ). Thus, the general topology literature supplies context, whereas the mTEC-specific inference rests on AIRE super-enhancer and looping studies.
The interaction between AIRE and the cohesin loader NIPBL provides one mechanistic route into genome folding. AIRE can recruit NIPBL to super-enhancer-rich regions and thereby promote chromatin looping and activation of distant loci (). This observation supports a model in which AIRE-containing assemblies nucleate within enhancer-dense domains and couple co-activator recruitment to local loop remodeling. The ability of AIRE condensates to bridge loci on different chromosomes () suggests that AIRE-dependent activation may operate across multiple spatial scales, from enhancer-promoter proximity to broader nuclear-contact organization. How these spatial scales are used at individual TRA loci remains unclear.
4.3 Control of transcriptional kinetics and pause release
The condensate model must also account for transcriptional kinetics, not only chromatin proximity. AIRE-regulated genes are enriched for promoter-proximal paused RNA polymerase II, and AIRE can promote release of paused polymerase into productive elongation (). Promoter-proximal pausing occurs when RNA polymerase II initiates transcription but remains transiently restrained near the transcription start site until elongation factors are recruited. Many AIRE targets reside in relatively inactive chromatin and may therefore require both enhancer engagement and a kinetic switch. AIRE provides such a switch by engaging BRD4 and positive transcription elongation factor b (P-TEFb), a CDK9-containing complex that phosphorylates the pause-release machinery and supports productive elongation (, ).
This kinetic view helps explain how sparse detection can coexist with high output at activated loci. Sparse detection may reflect the low probability that a given TRA locus reaches a permissive combination of chromatin accessibility, enhancer contact, AIRE nucleation and cofactor availability in an individual mTEC. Once this threshold is crossed, local enrichment of AIRE, BRD4, P-TEFb, MED1-associated co-activator assemblies, and paused polymerase could increase the probability of pause release, extend the duration of an active burst, and permit repeated rounds of polymerase re-initiation from the same locus (, ). In this scenario, a rare activation event is converted into a high-output transcriptional episode, while most loci in most cells remain below threshold. Direct measurements of nascent transcription, AIRE residence time, enhancer-contact persistence, and burst kinetics in primary mTECs will be needed to test this model.
5 Epigenetic gating and stochastic TRA expression
5.1 Polycomb-associated repression and an inducible context
Sparse TRA expression is shaped by the repressive chromatin landscape on which AIRE acts. In general terms, epigenetic gating describes how access to the genome is controlled by DNA methylation and the associated histone modifications, which together determine whether a given locus is permissive or refractory to transcription. Here, epigenetic gating refers to a repressive but inducible chromatin state that shapes the probability of TRA activation at individual loci. Many AIRE-dependent TRA loci are marked by Polycomb repression, especially H3K27me3, the trimethylation of histone H3 lysine 27 (–). AIRE-dependent transcription therefore emerges not from a uniformly permissive genome, but from loci that are restrained before activation.
PRC2, the enzyme complex that deposits H3K27me3, provides functional support for this gating model. Developmental ChIP-seq analyses indicate that tissue-specific genes in mTECs acquire H3K27me3 during transition from immature to mature states, with particular enrichment at genes induced during late maturation (). Genetic and recent mechanistic studies further indicate that PRC2 shapes both ordered stochastic expression of Aire targets and development of mimetic cells, although the exact molecular handoff between Polycomb repression and AIRE condensate activation remains incompletely resolved (, ). PRC2 may therefore shape the probability of AIRE-dependent activation rather than functioning as a simple binary antagonist of AIRE.
The gate is also dynamic. The H3K27 demethylase Kdm6b is required for mTEC homeostasis and function; loss of Kdm6b reduces mature mTEC numbers, impairs TRA expression, and compromises immune tolerance (). These findings suggest that AIRE-dependent activation is shaped by the balance between Polycomb-mediated methylation and demethylation rather than by a fixed repressive state. Changes in chromatin-modifying enzymes may therefore shift the probability that an AIRE-associated assembly becomes transcriptionally productive.
5.2 mTEC subset diversity and AIRE-dependent regulatory modes
AIRE operates within a heterogeneous epithelial compartment. Single-cell studies support a continuum of mTEC differentiation states and regulatory modules rather than a single terminal population (–). Within this continuum, AIRE intersects with other lineage regulators, including FEZF2/Fezf2 (FEZ family zinc finger 2), that contribute to the overall self-antigen landscape (). Consequently, the same AIRE-centered machinery may produce different outputs depending on cell state, chromatin accessibility, and cofactor availability.
AIRE-dependent output is therefore not deployed identically across all mTEC subsets. Integrated single-cell transcriptomic and DNA methylomic analyses resolve mature mTECs into populations with overlapping but non-identical antigen-presentation programs (). Some subsets appear enriched for strong Aire-associated promiscuous gene expression, whereas others preferentially deploy alternative lineage-defining programs. Single-cell heterogeneity therefore reflects structured deployment of antigen-expression programs across differentiated epithelial states rather than random noise around a homogeneous cell type.
Mimetic cells provide another layer of this cell-state dependence. Thymic tuft-cell studies first established that the medulla contains epithelial subsets with specialized peripheral-like features (, ). Broader single-cell and enhancer studies then showed that mTECs can co-opt lineage-defining transcription-factor programs to generate tissue-like mimetic states, including gut-, muscle-, neuroendocrine-, and other lineage-associated programs (–). These subsets suggest that tolerance is built not only through stochastic activation of isolated TRAs but also through controlled deployment of tissue-like transcriptional modules. AIRE may contribute to this architecture, but current evidence does not support treating it as the universal master regulator of all mimetic identities.
5.3 A gating model for stochastic activation and modular output
Together, PRC2-associated restraint and AIRE-dependent activation yield a testable gating model. Most tissue-specific loci remain in repressed or low-probability states across the mTEC population. A subset of cells transiently acquires the chromatin context, cofactor availability, and AIRE assembly state needed for productive elongation. Once that threshold is crossed, transcription is locally amplified, producing high-level expression of a limited set of TRA modules. This model helps explain why TRA expression is rare, modular, and high-amplitude at single-cell resolution (, ) and is summarized conceptually in Figure 3.
Figure 3
Pathological failure could occur on either side of this gate. If Polycomb-associated repression is too rigid or improperly resolved, AIRE may not gain productive access to selected loci, reducing antigen coverage. Conversely, if AIRE multimerization, partner recruitment, or elongation control is impaired, loci that are otherwise poised for activation may fail to enter high-output states. In both cases, the defect is not simply a lower number of expressed genes; it is a breakdown in the probabilistic architecture that distributes self-antigen display across the mTEC population.
6 Single-cell atlases and the emergence of mimetic cells
6.1 mTEC heterogeneity and the definition of mimetic cells
Single-cell omics has transformed mTEC heterogeneity from a descriptive observation into a map of cell states, trajectories, and antigen-output programs (, ). Functional studies further indicate that distinct mTEC subsets cooperate in central tolerance rather than simply representing redundant versions of the same cell state (). The medullary compartment contains specialized subsets that differ in maturation stage, transcription-factor usage, and antigenic repertoire. This atlas places AIRE activity within a diversified epithelial compartment rather than within a uniform terminal cell type.
Mimetic cells include tuft-like, muscle-like, neuroendocrine-like, epithelial, and other specialized subsets that reproduce selected peripheral tissue features within the thymus (, ). The emergence of thymic mimetic cells as a distinct conceptual category has also been highlighted in field commentary (). Developmental conversion of embryonic thymic epithelial states provides an additional route by which antigen-displaying medullary programs emerge (). These programs appear to be actively specified thymic epithelial states rather than passive leakage of peripheral identity (, ). They provide a second organizing principle for tolerance: tissue-like modules can complement stochastic TRA activation by presenting self-information in coordinated cellular contexts.
Functionally, these subsets appear to cooperate rather than simply duplicate one another. CCL21-associated mTECs support thymocyte migration and medullary organization, tuft-like mTECs shape cytokine environments, and other mimetic lineages contribute distinct antigenic repertoires (, ). Central tolerance can therefore be viewed as an emergent property of the medullary network. AIRE remains a major axis of antigen diversification, but its output is embedded within a broader multicellular architecture.
6.2 Network constraints and allocation of mimetic lineages
Mimetic cells also reveal a regulatory constraint. mTECs must deploy lineage-defining transcriptional logic strongly enough to generate tissue-like antigenic features, but not so strongly that they undergo wholesale lineage conversion. Ikaros (Ikzf1) has emerged as a principal regulator of Aire-positive mTEC homeostasis and mimetic-cell diversity, linking thymic epithelial network architecture to allocation of mimetic states ().
Insm1 provides another example of this developmental control. Insm1 deficiency impairs mTEC development and immune tolerance, suggesting that appropriate epithelial-state formation is a prerequisite for correct deployment of AIRE-dependent and AIRE-adjacent programs (). Thus, transcriptional output attributed to AIRE cannot be interpreted independently of the developmental circuitry that establishes the cellular context in which AIRE acts.
Dependence on AIRE also differs across mimetic lineages. Lineage-defining transcription factors are required for accumulation of specific mimetic subsets, whereas AIRE appears to modulate only part of the associated transcriptional repertoire (, ). New developmental studies strengthen this distinction. RUNX1 deficiency reveals thymic alveolar type 2 epithelial mimetic cells, and RUNX3 has been linked to Aire+ mTEC development, tissue-specific antigen expression, and central tolerance (, ). These data support a division of labor: lineage factors allocate epithelial state space, while AIRE contributes a condensate-linked activation module within selected states rather than acting as a universal master regulator of all mimetic programs.
6.3 Cross-species conservation of mimetic cells and human models
Mimetic cells can be viewed as a modular tolerance strategy that generates localized tissue-like antigenic domains within the thymus. Cross-species studies support conservation of this broad logic while also identifying species-specific elements (, ). Recent spatial mapping of the human thymus further supports this view by localizing lineage-associated transcription factors within rare mimetic epithelial populations (). The implication is that tolerance benefits from architectural diversification of antigen display, not merely from increasing the total number of genes expressed.
Experimentally tractable human systems now provide platforms to test these ideas directly. Reviews of pluripotent stem-cell differentiation into thymic epithelial cells frame organoid systems as potential tools for modeling APECED-relevant thymic biology (). Human pluripotent stem cell-derived thymic epithelial models recapitulate aspects of human thymic epithelial development and multilineage specification (, ). Adult and engineered thymic epithelial organoid systems provide complementary platforms for modeling thymic epithelial function and T-cell development (–). These systems may help determine how AIRE activity, mimetic-state allocation, and chromatin context are coordinated in human cells, where species differences limit direct extrapolation from mouse models (Figure 4).
Figure 4
7 Perspectives: peripheral AIRE biology and open questions
Peripheral AIRE biology is best discussed as a related but distinct area of tolerance research (, ), rather than as a direct extension of the mTEC condensate mechanism. Direct evidence for AIRE condensate assembly, enhancer-local transcriptional activation, and interchromosomal bridging is currently strongest in mTECs and reconstituted systems (). Peripheral AIRE- or RORγt-related antigen-presenting cells may share selected tolerance functions with mTECs, but their lineage identity, antigen sources, and regulatory environments differ. They are therefore best viewed as distinct APC states whose possible relationship to AIRE condensates remains an open question.
Table 1 summarizes peripheral AIRE- or RORγt-associated tolerance settings that are relevant to, but mechanistically distinct from, mTEC AIRE biology. Across these studies, the shared theme is tolerogenic antigen presentation outside the classical thymic epithelial compartment; however, the evidence mainly supports context-dependent APC states, microbiota- or food-antigen–linked tolerance, or RORγt-associated immune regulation rather than direct reuse of the mTEC AIRE-condensate program. At present, none of these peripheral systems has demonstrated AIRE-containing enhancer condensates, interchromosomal bridging, or stochastic TRA activation through an mTEC-like pause-release mechanism. Therefore, these cell populations should be discussed as comparative tolerance contexts rather than as proven extensions of the mTEC condensate model.
Table 1
| Cell type or population | Main context | Reported immune role | Evidence boundary for AIRE condensates | Representative references |
|---|---|---|---|---|
| Classical eTACs | Secondary lymphoid organs | Deletion or functional inactivation of autoreactive CD4+ T cells in selected models | Classical eTACs were originally defined as extrathymic Aire-expressing APCs, but their lineage identity and stability are context dependent. They should not be treated as simple peripheral counterparts of mTECs, and direct evidence for AIRE condensates has not been demonstrated. | (–) |
| Aire-protein ILC3-like cells | Lymph nodes | Local antigen presentation and immune regulation | These cells support the existence of an APC state with detectable Aire protein, but current evidence does not establish broad mTEC-like TRA output or nuclear AIRE condensate assembly. | () |
| AmDCs/Janus-like populations | Peripheral lymphoid tissues | Tolerogenic APC states and peripheral immune restraint | Single-cell multiomic studies define Aire-positive APC states with dendritic-cell-like and epithelial-homology features. However, lineage relationships and mechanistic AIRE dependence remain under refinement, and direct AIRE condensate evidence is lacking. | () |
| R-eTACs and Thetis cells | Gut-associated and early-life intestinal tolerance | Treg-dependent tolerance to gut microbiota and dietary antigens | These populations illustrate microbiota-, food-antigen-, and developmental-context-dependent tolerance. Their functions should not be interpreted as a general self-TRA display program or as evidence for mTEC-like AIRE condensate reuse. | (–) |
| RORγt-positive DCs/related intestinal APCs | Oral antigen and intestinal tolerance | Peripheral regulatory T-cell induction in response to oral and intestinal antigens | These cells are important for peripheral tolerance, but available studies do not show AIRE-dependent enhancer condensates, stochastic TRA expression, or pause-release mechanisms analogous to those in mTECs. | (, ) |
| Related RORγt APC/ILC3 programs | Mucosal and peripheral immune settings | Treg induction, oral tolerance, antifungal defense, and cell-state support | These programs show the functional diversity of peripheral tolerance APCs. AIRE transcript or protein involvement varies by system, and direct evidence for AIRE condensate-based regulation has not been demonstrated. | (–) |
Evidence boundaries for peripheral and non-mTEC AIRE-related tolerance contexts.
B-cell contexts raise a similar interpretive issue. Thymic B cells can participate in central tolerance through antigen capture, B-cell receptor-dependent antigen presentation, and self-antigen-driven class switching, but they are not epithelial cells and should not be treated as mTEC replicas (–). Recent commentaries further emphasize thymic B cells as important contributors to T-cell tolerance (). B-cell-mediated tolerance can also be relevant in disease-associated antigen settings, as illustrated by tolerance to the neuromyelitis optica autoantigen AQP4 (). Age-related changes in thymic B-cell Aire expression provide an additional context, but they do not establish mTEC-like condensate regulation (, ). Germinal center B-cell AIRE biology is more distant from the mTEC paradigm because it relates to antibody diversification checkpoints rather than population-wide TRA display (). Future work should test peripheral AIRE mechanisms directly through live-cell imaging, chromatin profiling, perturbation of AIRE assembly interfaces, and single-cell nascent-transcription measurements before extending the condensate model beyond mTECs.
8 Conclusion
Current evidence supports a model in which AIRE coordinates tolerogenic transcriptional assemblies in mTECs rather than acting as a conventional sequence-specific transcription factor. The strongest mechanistic evidence links AIRE-containing enhancer assemblies with three-dimensional genome contacts, pause-release kinetics and PRC2/H3K27me3-associated chromatin gating, while emerging work on Z-DNA-associated targeting adds an additional layer to AIRE locus selection. This view helps explain how rare TRA activation events can become high-output antigen-expression episodes while preserving the single-cell sparsity that characterizes mTEC antigen display. Recent studies of spatial human thymus organization, RUNX-dependent mimetic-cell programs, RORγt-associated peripheral APCs, thymic B cells, and germinal center B cells extend the biological context of AIRE-related tolerance, but they do not yet demonstrate shared condensate-based TRA regulation. Future work should directly measure AIRE residence time, condensate composition, enhancer-contact dynamics and nascent transcription in defined mTEC states and, where relevant, in peripheral AIRE-related cells.
Statements
Author contributions
JH: Writing – original draft, Writing – review & editing. WX: Conceptualization, Writing – review & editing. JZ: Supervision, Writing – review & editing. YL: Funding acquisition, Supervision, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the National Natural Science Foundation of China (Nos. 82160523 and 82560564), the West Light Foundation of the Chinese Academy of Sciences, and Leading Talent under the 2025 Qinghai Province “Kunlun Elite: High-End Innovation and Entrepreneurship Talent” High-Level Talent Program.
Acknowledgments
The Figures were illustrated using Adobe Illustrator.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. Generative AI tools were used during the preparation of this manuscript. OpenAI ChatGPT was used to assist with language polishing, wording refinement, structural editing, and preparation of responses to reviewers. Google Gemini was used to assist with the generation and refinement of schematic figures. All AI-assisted text and figures were reviewed, edited, verified, and approved by the authors. The authors take full responsibility for the final content of the manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AndersonMSVenanziESKleinLChenZBerzinsSPTurleySJet al. Projection of an immunological self shadow within the thymus by the aire protein. Science. (2002) 298:1395–401. doi: 10.1126/science.1075958
2
Finnish-German APECED Consortium. An autoimmune disease, APECED, caused by mutations in a novel gene featuring two PHD-type zinc-finger domains. Nat Genet. (1997) 17:399–403. doi: 10.1038/ng1297-399
3
ScottHSHeinoMPetersonPMittazLLaliotiMDBetterleCet al. Common mutations in autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy patients of different origins. Mol Endocrinol. (1998) 12:1112–9. doi: 10.1210/mend.12.8.0143
4
SansomSNShikama-DornNZhanybekovaSNusspaumerGMacaulayICDeadmanMEet al. Population and single-cell genomics reveal the Aire dependency, relief from Polycomb silencing, and distribution of self-antigen expression in thymic epithelia. Genome Res. (2014) 24:1918–31. doi: 10.1101/gr.171645.113
5
RattayKMeyerHVHerrmannCBrorsBKyewskiB. Evolutionary conserved gene co-expression drives generation of self-antigen diversity in medullary thymic epithelial cells. J Autoimmun. (2016) 67:65–75. doi: 10.1016/j.jaut.2015.10.001
6
BansalKYoshidaHBenoistCMathisD. The transcriptional regulator Aire binds to and activates super-enhancers. Nat Immunol. (2017) 18:263–73. doi: 10.1038/ni.3675
7
BansalKMichelsonDARamirezRNVinyADLevineRLBenoistCet al. Aire regulates chromatin looping by evicting CTCF from domain boundaries and favoring accumulation of cohesin on superenhancers. Proc Natl Acad Sci USA. (2021) 118:e2110991118. doi: 10.1073/pnas.2110991118
8
KaiserCBraduAGambleNCaldwellJAKohAS. AIRE in context: Leveraging chromatin plasticity to trigger ectopic gene expression. Immunol Rev. (2022) 305:59–76. doi: 10.1111/imr.13026
9
PassosGASpeck-HernandezCAAssisAFMendes-da-CruzDA. Update on Aire and thymic negative selection. Immunology. (2018) 153:10–20. doi: 10.1111/imm.12831
10
PetersonP. AIRE's complex role beyond promiscuous gene expression. Immunol Rev. (2026) 337:e70091. doi: 10.1111/imr.70091
11
MichelsonDAMathisD. Thymic mimetic cells: Tolerogenic masqueraders. Trends Immunol. (2022) 43:782–91. doi: 10.1016/j.it.2022.07.010
12
AbramsonJ. Reframing thymic biology: From AIRE to mimetics, thymflammation, regeneration, and beyond. Immunol Rev. (2026) 339:e70121. doi: 10.1111/imr.70121
13
HuismanBDZuoCMathisD. Thymic mimetic cells: Evolutionarily ancient mirrors of the periphery. Immunol Rev. (2025) 331:e70028. doi: 10.1111/imr.70028
14
MichalakiCSharjeelMCowanJE. From development to aging: Dynamic roles of the thymic medulla in T cell biology. Immunol Rev. (2026) 339:e70118. doi: 10.1111/imr.70118
15
UshioAMatsuda-LennikovMKalle-YoungoueFShimizuAAbdelmaksoudAKellyMCet al. Functionally diverse thymic medullary epithelial cells interplay to direct central tolerance. Cell Rep. (2024) 43:114072. doi: 10.1016/j.celrep.2024.114072
16
HuohYSZhangQTörnerRBacaSCArthanariHHurS. Mechanism for controlled assembly of transcriptional condensates by Aire. Nat Immunol. (2024) 25:1580–92. doi: 10.1038/s41590-024-01922-w
17
WaterfieldMRAndersonMS. A balancing act in the control of self-antigen expression. Nat Immunol. (2024) 25:1520–1. doi: 10.1038/s41590-024-01940-8
18
AndersonMSVenanziESChenZBerzinsSPBenoistCMathisD. The cellular mechanism of Aire control of T cell tolerance. Immunity. (2005) 23:227–39. doi: 10.1016/j.immuni.2005.07.005
19
InglesfieldSCoswayEJJenkinsonWEAndersonG. Rethinking thymic tolerance: Lessons from mice. Trends Immunol. (2019) 40:279–91. doi: 10.1016/j.it.2019.01.011
20
NishijimaHMatsumotoMMorimotoJHosomichiKAkiyamaNAkiyamaTet al. Aire controls heterogeneity of medullary thymic epithelial cells for the expression of self-antigens. J Immunol. (2022) 208:303–20. doi: 10.4049/jimmunol.2100692
21
CarterJAStrömichLPeaceyMChapinSRVeltenLSteinmetzLMet al. Transcriptomic diversity in human medullary thymic epithelial cells. Nat Commun. (2022) 13:4296. doi: 10.1038/s41467-022-31750-1
22
MiragaiaRJZhangXGomesTSvenssonVIlicicTHenrikssonJet al. Single-cell RNA-sequencing resolves self-antigen expression during mTEC development. Sci Rep. (2018) 8:685. doi: 10.1038/s41598-017-19100-4
23
MeredithMZemmourDMathisDBenoistC. Aire controls gene expression in the thymic epithelium with ordered stochasticity. Nat Immunol. (2015) 16:942–9. doi: 10.1038/ni.3247
24
SabariBRDall'AgneseABoijaAKleinIACoffeyELShrinivasKet al. Coactivator condensation at super-enhancers links phase separation and gene control. Science. (2018) 361:eaar3958. doi: 10.1126/science.aar3958
25
LovénJHokeHALinCYLauAOrlandoDAVakocCRet al. Selective inhibition of tumor oncogenes by disruption of super-enhancers. Cell. (2013) 153:320–34. doi: 10.1016/j.cell.2013.03.036
26
BezPCeraudoMVianelloFRattazziMScarpaR. Where AIRE we now? Where AIRE we going? Curr Opin Allergy Clin Immunol. (2024) 24:448–56. doi: 10.1097/ACI.0000000000001041
27
HuohYSWuBParkSYangDBansalKGreenwaldEet al. Dual functions of Aire CARD multimerization in the transcriptional regulation of T cell tolerance. Nat Commun. (2020) 11:1625. doi: 10.1038/s41467-020-15448-w
28
FergusonBJAlexanderCRossiSWLiivIRebaneAWorthCLet al. AIRE's CARD revealed, a new structure for central tolerance provokes transcriptional plasticity. J Biol Chem. (2008) 283:1723–31. doi: 10.1074/jbc.M707211200
29
BjörsesPHalonenMPalvimoJJKolmerMAaltonenJEllonenPet al. Mutations in the AIRE gene: Effects on subcellular location and transactivation function of the autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy protein. Am J Hum Genet. (2000) 66:378–92. doi: 10.1086/302765
30
OftedalBEHellesenAErichsenMMBratlandEVardiAPerheentupaJet al. Dominant mutations in the autoimmune regulator AIRE are associated with common organ-specific autoimmune diseases. Immunity. (2015) 42:1185–96. doi: 10.1016/j.immuni.2015.04.021
31
OrgTChignolaFHetényiCGaetaniMRebaneALiivIet al. The autoimmune regulator PHD finger binds to non-methylated histone H3K4 to activate gene expression. EMBO Rep. (2008) 9:370–6. doi: 10.1038/sj.embor.2008.11
32
MuscoGPetersonP. PHD finger of autoimmune regulator: An epigenetic link between the histone modifications and tissue-specific antigen expression in thymus. Epigenetics. (2008) 3:310–4. doi: 10.4161/epi.3.6.7182
33
KohASKingstonREBenoistCMathisD. Global relevance of Aire binding to hypomethylated lysine-4 of histone-3. Proc Natl Acad Sci USA. (2010) 107:13016–21. doi: 10.1073/pnas.1004436107
34
ŽumerKLowAKJiangHSakselaKPeterlinBM. Unmodified histone H3K4 and DNA-dependent protein kinase recruit autoimmune regulator to target genes. Mol Cell Biol. (2012) 32:1354–62. doi: 10.1128/MCB.06359-11
35
FangYBansalKMostafaviSBenoistCMathisD. AIRE relies on Z-DNA to flag gene targets for thymic T cell tolerization. Nature. (2024) 628:400–7. doi: 10.1038/s41586-024-07169-7
36
YoshidaHBansalKSchaeferUChapmanTRiojaIProektIet al. Brd4 bridges the transcriptional regulators, Aire and P-TEFb, to promote elongation of peripheral-tissue antigen transcripts in thymic stromal cells. Proc Natl Acad Sci USA. (2015) 112:E4448–57. doi: 10.1073/pnas.1512081112
37
ŽumerKPlemenitašASakselaKPeterlinBM. Patient mutation in AIRE disrupts P-TEFb binding and target gene transcription. Nucleic Acids Res. (2011) 39:7908–19. doi: 10.1093/nar/gkr527
38
WhyteWAOrlandoDAHniszDAbrahamBJLinCYKageyMHet al. Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell. (2013) 153:307–19. doi: 10.1016/j.cell.2013.03.035
39
HniszDAbrahamBJLeeTILauASaint-AndréVSigovaAAet al. Super-enhancers in the control of cell identity and disease. Cell. (2013) 155:934–47. doi: 10.1016/j.cell.2013.09.053
40
GoldfarbYGivonyTKadouriNDobešJPeligero-CruzCZalayatIet al. Mechanistic dissection of dominant AIRE mutations in mouse models reveals AIRE autoregulation. J Exp Med. (2021) 218:e20201076. doi: 10.1084/jem.20201076
41
MahLJEl-OstaAKaragiannisTC. gammaH2AX: A sensitive molecular marker of DNA damage and repair. Leukemia. (2010) 24:679–86. doi: 10.1038/leu.2010.6
42
MadabhushiRGaoFPfenningARPanLYamakawaSSeoJet al. Activity-induced DNA breaks govern the expression of neuronal early-response genes. Cell. (2015) 161:1592–605. doi: 10.1016/j.cell.2015.05.032
43
Herrero-RuizAMartínez-GarcíaPMTerrón-BautistaJMillán-ZambranoGLiebermanJAJimeno-GonzálezSet al. Topoisomerase IIα represses transcription by enforcing promoter-proximal pausing. Cell Rep. (2021) 35:108977. doi: 10.1016/j.celrep.2021.108977
44
SanbornALRaoSSHuangSCDurandNCHuntleyMHJewettAIet al. Chromatin extrusion explains key features of loop and domain formation in wild-type and engineered genomes. Proc Natl Acad Sci USA. (2015) 112:E6456–65. doi: 10.1073/pnas.1518552112
45
FudenbergGImakaevMLuCGoloborodkoAAbdennurNMirnyLA. Formation of chromosomal domains by loop extrusion. Cell Rep. (2016) 15:2038–49. doi: 10.1016/j.celrep.2016.04.085
46
GasslerJBrandãoHBImakaevMFlyamerIMLadstätterSBickmoreWAet al. A mechanism of cohesin-dependent loop extrusion organizes zygotic genome architecture. EMBO J. (2017) 36:3600–18. doi: 10.15252/embj.201798083
47
GolfierSQuailTKimuraHBruguésJ. Cohesin and condensin extrude DNA loops in a cell cycle-dependent manner. eLife. (2020) 9:e53885. doi: 10.7554/eLife.53885
48
GiraudMYoshidaHAbramsonJRahlPBYoungRAMathisDet al. Aire unleashes stalled RNA polymerase to induce ectopic gene expression in thymic epithelial cells. Proc Natl Acad Sci USA. (2012) 109:535–40. doi: 10.1073/pnas.1119351109
49
FujinagaKHuangFPeterlinBM. P-TEFb: The master regulator of transcription elongation. Mol Cell. (2023) 83:393–403. doi: 10.1016/j.molcel.2022.12.006
50
MatsumotoMYoshidaHTsuneyamaKOyaTMatsumotoM. Revisiting Aire and tissue-restricted antigens at single-cell resolution. Front Immunol. (2023) 14:1176450. doi: 10.3389/fimmu.2023.1176450
51
HuCZhangKJiangFWangHShaoQ. Epigenetic modifications in thymic epithelial cells: an evolutionary perspective for thymus atrophy. Clin Epigenet. (2021) 13:210. doi: 10.1186/s13148-021-01197-0
52
BarthlottTHandelAETehHYWirasinhaRCHafenKŽuklysSet al. Indispensable epigenetic control of thymic epithelial cell development and function by polycomb repressive complex 2. Nat Commun. (2021) 12:3933. doi: 10.1038/s41467-021-24158-w
53
HandelAEShikama-DornNZhanybekovaSMaioSGraedelANZuklysSet al. Comprehensively profiling the chromatin architecture of tissue restricted antigen expression in thymic epithelial cells over development. Front Immunol. (2018) 9:2120. doi: 10.3389/fimmu.2018.02120
54
MatsumotoMYoshidaMOyaTTsuneyamaKMatsumotoMYoshidaH. Role of PRC2 in the stochastic expression of Aire target genes and development of mimetic cells in the thymus. J Exp Med. (2025) 222:e20240817. doi: 10.1084/jem.20240817
55
LiuZZhangHHuYLiuDLiLLiCet al. Critical role of histone H3 lysine 27 demethylase Kdm6b in the homeostasis and function of medullary thymic epithelial cells. Cell Death Differ. (2020) 27:2843–55. doi: 10.1038/s41418-020-0546-8
56
DhallaFBaran-GaleJMaioSChappellLHolländerGAPontingCP. Biologically indeterminate yet ordered promiscuous gene expression in single medullary thymic epithelial cells. EMBO J. (2020) 39:e101828. doi: 10.15252/embj.2019101828
57
BautistaJLCramerNTMillerCNChavezJBerriosDIByrnesLEet al. Single-cell transcriptional profiling of human thymic stroma uncovers novel cellular heterogeneity in the thymic medulla. Nat Commun. (2021) 12:1096. doi: 10.1038/s41467-021-21346-6
58
KadouriNNevoSGoldfarbYAbramsonJ. Thymic epithelial cell heterogeneity: TEC by TEC. Nat Rev Immunol. (2020) 20:239–53. doi: 10.1038/s41577-019-0238-0
59
TakabaHMorishitaYTomofujiYDanksLNittaTKomatsuNet al. Fezf2 orchestrates a thymic program of self-antigen expression for immune tolerance. Cell. (2015) 163:975–87. doi: 10.1016/j.cell.2015.10.013
60
MillerCNProektIvon MoltkeJWellsKLRajpurkarARWangHet al. Thymic tuft cells promote an IL-4-enriched medulla and shape thymocyte development. Nature. (2018) 559:627–31. doi: 10.1038/s41586-018-0345-2
61
BornsteinCNevoSGiladiAKadouriNPouzollesMGerbeFet al. Single-cell mapping of the thymic stroma identifies IL-25-producing tuft epithelial cells. Nature. (2018) 559:622–6. doi: 10.1038/s41586-018-0346-1
62
MichelsonDAHaseKKaishoTBenoistCMathisD. Thymic epithelial cells co-opt lineage-defining transcription factors to eliminate autoreactive T cells. Cell. (2022) 185:2542–2558.e18. doi: 10.1016/j.cell.2022.05.018
63
MichelsonDAZuoCVerziMBenoistCMathisD. Hnf4 activates mimetic-cell enhancers to recapitulate gut and liver development within the thymus. J Exp Med. (2023) 220:e20230461. doi: 10.1084/jem.20230461
64
NusserAThomasOSZhangGNagakuboDArrigoniLKrauthBet al. Developmental trajectory and evolutionary origin of thymic mimetic cells. Nature. (2025) 643:1066–75. doi: 10.1038/s41586-025-09148-y
65
KamarajUSChenYLeiJGautamPHorcharoensukPClementeCKMet al. Spatial cartography of human thymus enables the geopositioning of lineage transcription factors in rare mimetic thymic epithelial cells. Nat Commun. (2026) 17:3721. doi: 10.1038/s41467-026-68596-w
66
VisanI. Thymus mimetic cells. Nat Immunol. (2022) 23:1134. doi: 10.1038/s41590-022-01282-3
67
OhigashiIWhiteAJYangMTFujimoriSTanakaYJacquesAet al. Developmental conversion of thymocyte-attracting cells into self-antigen-displaying cells in embryonic thymus medulla epithelium. eLife. (2024) 12:RP92552. doi: 10.7554/eLife.92552
68
JamesKDCoswayEJParnellSMWhiteAJJenkinsonWEAndersonG. Assembling the thymus medulla: Development and function of epithelial cell heterogeneity. Bioessays. (2024) 46:e2300165. doi: 10.1002/bies.202300165
69
MuroRNittaT. It's time to unite: Diversity and coordination of thymic stromal cells for T cell selection and organ integrity. Immunol Rev. (2025) 332:e70040. doi: 10.1111/imr.70040
70
SinJHSucharovJKashyapSWangYProektILiuXet al. Ikaros is a principal regulator of Aire+ mTEC homeostasis, thymic mimetic cell diversity, and central tolerance. Sci Immunol. (2023) 8:eabq3109. doi: 10.1126/sciimmunol.abq3109
71
TaoWYeZWeiYWangJYangWYuGet al. Insm1 regulates mTEC development and immune tolerance. Cell Mol Immunol. (2023) 20:1472–86. doi: 10.1038/s41423-023-01102-0
72
SinJHBowmanCJGerminoJPhillipsSHSucharovJLiuXet al. Thymic alveolar type 2 epithelial mimetic cells revealed by RUNX1 deficiency. Nat Immunol. (2026) 27:1462–75. doi: 10.1038/s41590-026-02536-0
73
SinJHPhillipsSHParentAVWaterfieldMR. Runx3 instructs Aire+ mTEC development, TSA gene expression, and central tolerance. J Autoimmun. (2026) 161:103573. doi: 10.1016/j.jaut.2026.103573
74
ProvinNGiraudM. Differentiation of pluripotent stem cells into thymic epithelial cells and generation of thymic organoids: Applications for therapeutic strategies against APECED. Front Immunol. (2022) 13:930963. doi: 10.3389/fimmu.2022.930963
75
RamosSAArmitageLHMortonJJAlzofonNHandlerDKellyGet al. Generation of functional thymic organoids from human pluripotent stem cells. Stem Cell Rep. (2023) 18:829–40. doi: 10.1016/j.stemcr.2023.02.013
76
PretemerYGaoYKanaiKYamamotoTKometaniKOzakiMet al. An iPSC-based in vitro model recapitulates human thymic epithelial development and multi-lineage specification. Nat Commun. (2025) 16:7680. doi: 10.1038/s41467-025-62523-1
77
LimSJ F van SonGWisma Eka YantiNLAndersson-RolfAWillemsenSKorvingJet al. Derivation of functional thymic epithelial organoid lines from adult murine thymus. Cell Rep. (2024) 43:114019. doi: 10.1016/j.celrep.2024.114019
78
MajorVPalmerSRousePMorysJHendersonTHübscherTet al. Establishment of a microwell-array-based miniaturized thymic organoid model suitable for high-throughput applications. Cell Rep. (2025) 44:115579. doi: 10.1016/j.celrep.2025.115579
79
HübscherTLorenzo-MartínLFBarthlottTTillardLLangerJJRousePet al. Thymic epithelial organoids mediate T-cell development. Development. (2024) 151:dev202853. doi: 10.1242/dev.202853
80
van LaarGGvan HamburgJPTasSW. Extrathymic AIRE-expressing cells: Friends or foes in autoimmunity and cancer? Autoimmun Rev. (2022) 21:103141. doi: 10.1016/j.autrev.2022.103141
81
MetzgerTCAndersonMS. Control of central and peripheral tolerance by Aire. Immunol Rev. (2011) 241:89–103. doi: 10.1111/j.1600-065X.2011.01008.x
82
GardnerJMDevossJJFriedmanRSWongDJTanYXZhouXet al. Deletional tolerance mediated by extrathymic Aire-expressing cells. Science. (2008) 321:843–7. doi: 10.1126/science.1159407
83
GardnerJMMetzgerTCMcMahonEJAu-YeungBBKrawiszAKLuWet al. Extrathymic Aire-expressing cells are a distinct bone marrow-derived population that induce functional inactivation of CD4+ T cells. Immunity. (2013) 39:560–72. doi: 10.1016/j.immuni.2013.08.005
84
PolianiPLKisandKMarrellaVRavaniniMNotarangeloLDVillaAet al. Human peripheral lymphoid tissues contain autoimmune regulator-expressing dendritic cells. Am J Pathol. (2010) 176:1104–12. doi: 10.2353/ajpath.2010.090956
85
YamanoTDobešJVobořilMSteinertMBrabecTZiętaraNet al. Aire-expressing ILC3-like cells in the lymph node display potent APC features. J Exp Med. (2019) 216:1027–37. doi: 10.1084/jem.20181430
86
WangJLareauCABautistaJLGuptaARSandorKGerminoJet al. Single-cell multiomics defines tolerogenic extrathymic Aire-expressing populations with unique homology to thymic epithelium. Sci Immunol. (2021) 6:eabl5053. doi: 10.1126/sciimmunol.abl5053
87
AkagbosuBTayyebiZShibuGPaucar IzaYADeepDParisottoYFet al. Novel antigen-presenting cell imparts Treg-dependent tolerance to gut microbiota. Nature. (2022) 610:752–60. doi: 10.1038/s41586-022-05309-5
88
KedmiRNajarTAMesaKRGraysonAKroehlingLHaoYet al. A RORγt+ cell instructs gut microbiota-specific Treg cell differentiation. Nature. (2022) 610:737–43. doi: 10.1038/s41586-022-05089-y
89
CabricVFranco ParisottoYParkTAkagbosuBZhaoZLoYet al. A wave of Thetis cells imparts tolerance to food antigens early in life. Science. (2025) 389:268–74. doi: 10.1126/science.adp0535
90
MintonK. RORγt+ antigen-presenting cells mediate food tolerance. Nat Rev Immunol. (2025) 25:479. doi: 10.1038/s41577-025-01195-w
91
RodriguesPFWuSTrsanTPandaSKFachiJLLiuYet al. Rorγt-positive dendritic cells are required for the induction of peripheral regulatory T cells in response to oral antigens. Cell. (2025) 188:2720–2737.e22. doi: 10.1016/j.cell.2025.03.020
92
FuLLittmanDR. Specialized dendritic cells mediating peripheral tolerance to intestinal antigens. Immunol Rev. (2026) 337:e70082. doi: 10.1111/imr.70082
93
SunIHQuallsAEYinHSWangJArvedsonMPGerminoJet al. RORγt eTACs mediate oral tolerance and Treg induction. J Exp Med. (2025) 222:e20250573. doi: 10.1084/jem.20250573
94
LyuMSuzukiHKangLGaspalFZhouWGocJet al. ILC3s select microbiota-specific regulatory T cells to establish tolerance in the gut. Nature. (2022) 610:744–51. doi: 10.1038/s41586-022-05141-x
95
DobešJBen-NunOBinyaminAStoler-BarakLOftedalBEGoldfarbYet al. Extrathymic expression of Aire controls the induction of effective TH17 cell-mediated immune response to Candida albicans. Nat Immunol. (2022) 23:1098–108. doi: 10.1038/s41590-022-01247-6
96
YamanoTNedjicJHinterbergerMSteinertMKoserSPintoSet al. Thymic B cells are licensed to present self antigens for central T cell tolerance induction. Immunity. (2015) 42:1048–61. doi: 10.1016/j.immuni.2015.05.013
97
PereraJMengLMengFHuangH. Autoreactive thymic B cells are efficient antigen-presenting cells of cognate self-antigens for T cell negative selection. Proc Natl Acad Sci USA. (2013) 110:17011–6. doi: 10.1073/pnas.1313001110
98
PereraJZhengZLiSGudjonsonHKalininaOBenichouJICet al. Self-antigen-driven thymic B cell class switching promotes T cell central tolerance. Cell Rep. (2016) 17:387–98. doi: 10.1016/j.celrep.2016.09.011
99
Lombard-VadnaisFLesageS. Highlight of 2023: Thymic B cells-important players in the establishment of T-cell tolerance. Immunol Cell Biol. (2024) 102:448–51. doi: 10.1111/imcb.12761
100
AfzaliAMNirschlLSieCPfallerMUlianovOHasslerTet al. B cells orchestrate tolerance to the neuromyelitis optica autoantigen AQP4. Nature. (2024) 627:407–15. doi: 10.1038/s41586-024-07079-8
101
CepedaSCantuCOrozcoSXiaoYBrownZSemwalMKet al. Age-associated decline in thymic B cell expression of Aire and Aire-dependent self-antigens. Cell Rep. (2018) 22:1276–87. doi: 10.1016/j.celrep.2018.01.015
102
WedemeyerSAGriffithAV. Thymic B cells in aging and autoimmune disease. Front Immunol. (2025) 16:1595805. doi: 10.3389/fimmu.2025.1595805
103
ZhouJZHuangBPeiBSunGWPawlitzMDZhangWet al. A germinal center checkpoint of AIRE in B cells limits antibody diversification. eLife. (2026) 15:RP110530. doi: 10.7554/eLife.110530.1
Summary
Keywords
AIRE, autoimmunity, central tolerance, medullary thymic epithelial cells, transcriptional condensates
Citation
He J, Xue W, Zhang J and Li Y (2026) AIRE transcriptional condensates in central tolerance: a multiscale mechanistic perspective. Front. Immunol. 17:1859704. doi: 10.3389/fimmu.2026.1859704
Received
19 April 2026
Revised
29 June 2026
Accepted
01 July 2026
Published
15 July 2026
Volume
17 - 2026
Edited by
Ludger Klein, Ludwig Maximilian University of Munich, Germany
Reviewed by
Dominik Filipp, Institute of Molecular Genetics (ASCR), Czechia
Tomoyoshi Yamano, Kanazawa University, Japan
Updates
Copyright
© 2026 He, Xue, Zhang and Li.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jun Zhang, zhangjun@qhuah.com; Yan Li, liyan@qhuah.com
†These authors have contributed equally to this work
‡ORCID: YanLi, orcid.org/0000-0002-9447-7154
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.