Spectrum of germline AIRE mutations causing APS-1 and familial hypoparathyroidism

Abstract

Objective: The autoimmune polyendocrine syndrome type 1 (APS-1) is an autosomal recessive disorder characterised by immune dysregulation and autoimmune endocrine gland destruction. APS-1 is caused by biallelic mutations affecting the autoimmune regulator (AIRE) gene on chromosome 21q22.3, which facilitates immunological self-tolerance. The objective was to investigate >300 probands with suspected APS-1 or isolated hypoparathyroidism for AIRE abnormalities.

Methods: Probands were assessed by DNA sequence analysis. Novel variants were characterised using 3D modelling of the AIRE protein. Restriction enzyme and microsatellite analysis were used to investigate for uniparental isodisomy.

Results: Biallelic AIRE mutations were identified in 35 probands with APS-1 and 5 probands with isolated hypoparathyroidism. These included a novel homozygous p.(His14Pro) mutation, predicted to disrupt the N-terminal caspase activation recruitment domain of the AIRE protein. Furthermore, an apparently homozygous AIRE mutation, p.Leu323fs, was identified in an APS-1 proband, who is the child of non-consanguineous asymptomatic parents. Microsatellite analysis revealed that the proband inherited two copies of the paternal mutant AIRE allele due to uniparental isodisomy. Hypoparathyroidism was the most common endocrine manifestation in AIRE mutation-positive probands and >45% of those harbouring AIRE mutations had at least two diseases out of the triad of candidiasis, hypoparathyroidism, and hypoadrenalism. In contrast, type 1 diabetes and hypothyroidism occurred more frequently in AIRE mutation-negative probands with suspected APS-1. Around 30% of AIRE mutation-negative probands with isolated hypoparathyroidism harboured mutations in other hypoparathyroid genes.

Conclusions: This study of a large cohort referred for AIRE mutational analysis expands the spectrum of genetic abnormalities causing APS-1.

Figure 1

In silico and structural analysis of AIRE mutations. (A) Schematic representation of the 545-amino acid AIRE protein predicted to comprise an N-terminal caspase activation recruitment domain (CARD; residues 1–105) involved in protein multimerization (10, 20); a monopartite nuclear localization sequence (NLS; residues 131–133) (32); a SAND domain (named after proteins harbouring this domain: Sp100, AIRE, NucP1/P75, DEAF-1; residues 189–290) involved in protein–protein interactions and DNA binding (3, 18); plant homeodomain 1 (PHD1; residues 299–340), which interacts with histone H3 (19); and plant homeodomain 2 (PHD2; residues 434–475), which may interact with protein complexes promoting transcriptional elongation (33). The location of the 20 different mutations identified in 40 probands is shown. The novel p.His14Pro variant is shown in red. (B) Multiple protein sequence alignment of the N-terminal α1-helix of AIRE orthologs showing conservation of the WT His14 (bold) residue in mammals and in the homologous apoptotic protease-activating factor 1 (Apaf-1) CARD. The mutant Pro14 residue identified in an APS-1 proband (Table 1) is shown in red. Conserved or homologous residues are shown in red boxes. (C) AlphaFold 3D structure of the AIRE CARD (16). The AIRE CARD is predicted to comprise six alpha helices (α1–α6), and the location of the mutated His14 (H14), Arg15 (R15), Leu28 (L28), and Leu81 (L81) residues identified in this study are shown. (D) The WT His14 residue (green) is predicted to form interactions with Leu10 and Glu17 α1-helix residues. (E) These interactions are disrupted by the introduction of the mutant Pro14 residue (cyan). (F) Graph showing predicted effect of all AIRE CARD missense mutations on CARD domain stability (9, 21). Mutations are plotted according to residue number (x-axis) and the predicted stability difference score (pseudo ∆∆G) is shown on the y-axis. Mutations with negative pseudo ∆∆G values are predicted to impair protein stability (17). Neutral, hydrophilic, and hydrophobic residues mutated in APS-1 are shown in black, blue, and red, respectively.

Figure 2

Identification of 21q uniparental isodisomy. (A) DNA sequence analysis of the AIRE gene in proband 19 (Table 1) revealed a 13-bp deletion (c.967_979del13, p.Leu323fs), which is predicted to result in the loss of a BsrBI restriction enzyme site. (B) PCR and BsrBI digestion confirmed that this proband (individual II.1, arrow) is homozygous for the p.Leu323fs mutation, whereas the father (individual I.1) is heterozygous for the p.Leu323fs mutation and the mother (individual I.2) is unaffected. (C) Restriction enzyme map showing that BsrBI digestion would result in two products of 140 bp and 89 bp from the 229-bp WT sequence but would not affect the 216-bp mutant (m) sequence, as reported (14). (D) Microsatellite analysis of chromosome 21q in the proband and parents. An analysis of nine markers across 21q21.1-21q22.3 (http://genome.ucsc.edu) showed the proband to be homozygous at all loci tested. Two markers (D21S1409 & D21S1280 (in bold and shaded grey)) were informative and demonstrated that there was no maternally derived allele. These results are consistent with paternal uniparental isodisomy (UPiD). (E) Schematic representation of the potential mechanism for paternal UPiD of the AIRE mutation in the proband, which may have been caused by generation of a nullisomic oocyte during meiosis and the rescue of the monosomic conceptus by duplication of the paternally derived chromosome 21 (shaded grey), which is carrying the mutant AIRE allele (represented by an ‘X’). Germ cells/gametes are represented by dashed ellipses, and the conceptus is represented by a solid ellipse.