One of the hallmarks of the adaptive immune system is its ability to distinguish between self and nonself, in order to be able to protect the individual from invading pathogens, while avoiding a potentially-destructive immune response against the body’s own tissues. Collectively known as immune tolerance, this ability is governed by an array of tightly regulated processes that are at the heart of immunological research. Much of our understanding of the mechanisms underlying immune tolerance has come from studying various cases of autoimmune disorders, which occur as a consequence of breakdown of self-tolerance mechanisms. Autoimmune polyglandular syndrome type 1 (APS-1), also known as autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED), is a monogenic autoimmune disorder characterized by devastating multiorgan pathological manifestations. In 1997, two research consortia cloned the gene responsible for this syndrome and aptly dubbed it autoimmune regulator (AIRE) (Finnish-German 1997; Nagamine et al. 1997). The homologous murine Aire gene was isolated shortly thereafter, showing >70% nucleotide correspondence with human AIRE (Perniola and Musco 2014). This has consequently resulted in tremendous scientific interest aimed at elucidating how AIRE operates at the cellular and molecular levels. One of the key milestones in these endeavors was the generation and analysis of AIRE-deficient mice in 2002 (Anderson et al. 2002), which has provided several critical insights into how AIRE induces self-tolerance in the thymus and how its dysfunction results in multiorgan autoimmune manifestations.
The Structure of AIRE
AIRE’s Roles in Immunological Tolerance
AIRE is primarily expressed in the thymus, predominantly in the population of mature medullary thymic epithelial cells (mTECs) (Heino et al. 1999). A seminal paper from 2002 revealed that the pivotal contribution of AIRE to the induction of central tolerance stems from its ability to regulate the expression of thousands of genes encoding for tissue-restricted antigens (TRAs), such as insulin, by epithelial cells in the thymic medulla (Anderson et al. 2002). Such AIRE-driven promiscuous gene expression of TRA genes by mTECs has been shown to mediate the establishment of self-tolerance at multiple levels.
First, developing thymocytes that recognize these self-antigens with very high affinity will undergo negative selection. As a result, these self-reactive clones will be deleted from the pool of T cells released to the periphery, to prevent possible autoimmune damage of the peripheral tissues. Indeed, upon AIRE deficiency, self-reactive T cells were shown to escape negative selection and ultimately provoke autoimmune disease (Anderson et al. 2002; Mathis and Benoist 2009).
The third major mechanism by which AIRE mediates the establishment of self-tolerance has been discovered only recently and was shown to involve AIRE-dependent generation of IL-17A+Vg6+Vd1+T cells in the thymus. Specifically, it has been demonstrated that AIRE, by controlling Il7 expression in mTECs, regulates the size of thymic IL-17A+Vg6+Vd1+T cell population, which has been implicated in mediating early proinflammatory and autoimmune responses in peripheral tissues. Indeed, AIRE-deficient mice have expanded thymic and peripheral populations of perinatally generated IL-17A+Vg6+Vd1+T cells, which then promote autoimmune pathogenesis in peripheral organs (Fujikado et al. 2016).
AIRE may further contribute to the induction of tolerance through additional related aspects, having to do with the efficiency of TRA presentation by mTECs and with their turnover. Specifically, the expression of AIRE at the final maturation stages of mTECs and its function as a proapoptotic factor has been suggested to promote cross-presentation of these antigens to dendritic cells (Mathis and Benoist 2009; Perry et al. 2014).
AIRE’s Mechanism of Action
Considering that AIRE regulates the expression of several thousand genes, it seems implausible that it would act as a classical transcription factor, which would activate its target genes by binding to their corresponding promoters. Instead, it appears that AIRE is targeted to the genes it regulates via the recognition of epigenetic marks. As mentioned above, AIRE can interact with H3K4me0, a hallmark of transcriptionally repressed genes. While binding of H3K4me0 by AIRE’s PHD1 is essential for its function in cell culture, global demethylation of H3K4 was not found to affect TRA expression in such system (Koh et al. 2010). Moreover, it is unlikely that target specificity is obtained by recognizing this modification, when taking into account its rather frequent appearance in the genome. Indeed, additional modifications associated with a repressive chromatin state (such as H3K27me3 and H3K9me3) were found to be enriched in loci of AIRE-dependent genes, while active marks (e.g., H3K4me3 or acetylated H3) are less abundant. Nevertheless, none of these other histone modifications is recognized directly by AIRE, suggesting that other transcriptional regulators that do recognize them partner with AIRE to facilitate gene expression. In addition, it was found that AIRE physically associates with the transcription start sites (TSS) of most genes, regardless of whether or not they are regulated by it. Thus, it appears that AIRE becomes activated only upon receiving additional signals, e.g., from transcriptional regulators associating with inactive chromatin (reviewed in Abramson and Husebye (2016), Anderson and Su (2016)).
In addition, AIRE-mediated induction of gene expression was found to involve the release of stalled RNA polymerase II (RNAP II) and subsequent transcriptional elongation. Accordingly, AIRE was shown to associate with the pTEFb elongation complex and with Bromodomain-containing protein 4 (BRD4), which mediate the release of RNAP II (Abramson and Goldfarb 2016). Though these findings provide a mechanistic framework as to how AIRE induces transcription, they still do not explain how only a fraction of the genes occupied by AIRE and RNAP II are eventually activated, whereas others are repressed.
Furthermore, an intriguing line of investigation also aims at elucidating the potential role of the DNA damage response machinery in AIRE-mediated gene expression. Specifically, one of the most prominent partners of AIRE is DNA-dependent protein kinase (DNA-PK), which is activated by double-strand breaks in the DNA (Abramson and Goldfarb 2016). Upon its activation, DNA-PK phosphorylates and thereby activates additional proteins that mediate DNA repair, some of which have also been shown to associate with AIRE. Strikingly, the anticancer drug, etoposide, which causes DNA breaks by inhibiting topoisomerase 2, can mimic the expression pattern induced by AIRE in HEK 293 cells (Abramson et al. 2010). Together with recent reports implicating the DNA damage pathway in the activation of several transcriptional programs, these observations suggest that AIRE-mediated gene expression may also be mediated by the response to breaks in the DNA.
Finally, oscillations in the acetylation state of residues on AIRE itself also appear to play an important role in mediating its activity. This is evident by the close association of AIRE with both the acetyl transferase, CREB-binding protein (CBP) as well as with the deacetylase Sirtuin-1 (Sirt1), both of which were shown to be crucial for AIRE-mediated gene expression, at least in part through acting on lysine residues located between the NLS and SAND domains of AIRE (Pitkanen et al. 2000; Chuprin et al. 2015; Abramson and Goldfarb 2016).
While much progress has been achieved over the past 15 years in elucidating AIRE mechanism of action and the allies that mediate or facilitate different aspects of its activity, many open questions remain. Contemporary research and ongoing advances in high-throughput single-cell experimental approaches are bound to improve our understanding of the ways in which AIRE identifies and activates its targets.
Regulation of AIRE Expression
While AIRE is predominantly expressed by mTECs, its expression has recently been ascribed to additional cell populations as well, where its roles are not as clear. Nonetheless, given the robust effects of AIRE in driving global gene expression, it is clear that its expression necessitates tight regulation.
Recently, a conserved noncoding sequence (CNS1) located 3 kb upstream of Aire TSS was found to serve a critical enhancer of Aire expression via recruitment of NFκB transcription factors. Moreover, deletion of this region demonstrated that NFκB signaling is essential but probably not sufficient for AIRE expression in mTECs (Haljasorg et al. 2015; LaFlam et al. 2015). More recently, our own research has further elucidated the rather complex regulation of AIRE expression. Specifically, by comprehensive analysis of both cis-acting and trans-acting regulatory mechanisms, we found that the Aire locus is insulated by the global chromatin organizer CTCF and is hypermethylated in cells and tissues that do not express Aire. In mTECs, however, Aire expression is facilitated by concurrent eviction of CTCF, specific demethylation of exon 2 and the proximal promoter, and the coordinated action of several transcription activators, including Irf4, Irf8, Tbx21, Tcf7, and Ctcfl, which act on mTEC-specific accessible regions in the Aire locus (Herzig et al. 2017).
Aire expression was also found to be regulated at the posttranscriptional level by lysyl-hydroxylase Jmjd6, which was shown to regulate splicing of the Aire transcript. Accordingly, Jmjd6 deficiency affects the abundance of AIRE protein expressed by mTECs and evokes multiorgan autoimmunity in mice (Yanagihara et al. 2015).
As mentioned above, AIRE was first reported in the context of APS-1, which, in its classical form, is inherited in an autosomal recessive manner, with very early onset, usually observed within the first years of life. The main clinical components of APS-1 are hypoparathyroidism, adrenal insufficiency, and chronic mucocutaneous candidiasis. Although chronic candida infection is not an autoimmune condition in itself, it has, in the case of APS-1, an autoimmune context, as it is caused by autoantibodies against cytokines (i.e., IL-17 and IL-22) critical for mediating immune response against candida infection (Kisand et al. 2010; Puel et al. 2010). Most APS-1 patients typically develop at least two of these three components, which are often supplemented by additional autoimmune manifestations, such as type-1 diabetes, premature ovarian failure, enamel hypoplasia, alopecia, and others (Husebye et al. 2009).
Almost 100 different APS-1 causing mutations have been mapped to AIRE, most of which are assumed to be inherited in an autosomal recessive manner. Recently, however, dominant-negative mono-allelic mutations have also been mapped to the AIRE locus, resulting in autoimmune disorders of varying phenotypes, onset, and penetrance (Oftedal et al. 2015). Interestingly, while the majority of recessive mutations cluster within the CARD domain, thus affecting proper homo-oligomerization of AIRE, the recently identified dominant-negative mutations mapped to the PHD1 finger. These latter mutants were able to physically associate with wild-type AIRE within its characteristic nuclear speckles, suggesting that the formed homo-oligomer was dysfunctional due to faulty PHD1 folding (Oftedal et al. 2015). The relatively high frequency of mono-allelic dominant AIRE mutants in mixed populations suggests that such mutations may underlie some organ-specific autoimmune disorders with familial clustering (Oftedal et al. 2015; Abramson and Husebye 2016).
An intriguing clinical phenomenon, providing some indication that the proper function of AIRE is needed to maintain self-tolerance throughout life, is the frequent association of thymomas and autoimmunity. Various autoimmune diseases, such as myasthenia gravis, often appear in patients with thymic tumors, in conjunction with the presence of autoantibodies that are often detected in APS-1 patients. A potential mechanistic link between these clinical manifestations lies in the fact that AIRE expression is defective in over 95% of thymomas, possibly resulting in defective negative selection and the escape of self-reactive T cell clones into the periphery (Marx et al. 2010; Abramson and Husebye 2016).
Finally, the role of AIRE in promoting self-tolerance may be perceived as a double-edged sword when considering that the negative selection of self-reactive T cell clones could result in the inability to mount an effective antitumor immune response upon need. Accordingly, reduced AIRE expression in both mice and humans is associated with enhanced antitumor immune activity, pointing to the potential of short-term perturbation of thymic AIRE expression in cancer immunotherapy (recently reviewed in Anderson and Su (2016)).
The discovery of AIRE has promoted significant advances in the research on immune tolerance. Cellular and molecular mechanisms of self-tolerance in the thymus and beyond have been brought to light, teaching us fascinating lessons on the workings of the immune system, with obvious clinical implications. The unique expression pattern and transactivation properties of AIRE, which are under current investigation, provide intriguing new facets on gene regulation in general.
As is commonly the case in science, the answer to one research question is generally the basis for a myriad of new open questions. In the case of AIRE, a very partial list includes deciphering the transcriptional and epigenetic regulatory mechanisms underlying AIRE expression and control over its targets, gaining a better understanding of the extrathymic roles of AIRE, and finding ways to compensate for various modes of AIRE deficiency. Progressively developing technological advancements now enable us to tackle such questions at depth and resolution that were not previously available and hold promise that the coming years will provide exciting new insights on AIRE and beyond.
- Fujikado N, Mann AO, Bansal K, Romito KR, Ferre EM, Rosenzweig SD, et al. Aire inhibits the generation of a perinatal population of interleukin-17A-producing gammadelta T cells to promote immunologic tolerance. Immunity. 2016;45:999–1012. doi: 10.1016/j.immuni.2016.10.023.PubMedCentralCrossRefPubMedGoogle Scholar
- Herzig Y, Nevo S, Bornstein C, Brezis MR, Ben-Hur S, Shkedy A, et al. Transcriptional programs that control expression of the autoimmune regulator gene Aire. Nat Immunol. 2017;18. doi: 10.1038/ni.3638.
- Pitkanen J, Doucas V, Sternsdorf T, Nakajima T, Aratani S, Jensen K, et al. The autoimmune regulator protein has transcriptional transactivating properties and interacts with the common coactivator CREB-binding protein. J Biol Chem. 2000;275:16802–9. doi: 10.1074/jbc.M908944199.CrossRefPubMedGoogle Scholar
- Puel A, Doffinger R, Natividad A, Chrabieh M, Barcenas-Morales G, Picard C, et al. Autoantibodies against IL-17A, IL-17F, and IL-22 in patients with chronic mucocutaneous candidiasis and autoimmune polyendocrine syndrome type I. J Exp Med. 2010;207:291–7. doi: 10.1084/jem.20091983.PubMedCentralCrossRefPubMedGoogle Scholar