Aire-Deficient C57BL/6 Mice Mimicking the Common Human 13-Base Pair Deletion Mutation Present with Only a Mild Autoimmune Phenotype

Mice

All mice were C57BL/6 (B6) and bred and maintained at the Walter and Eliza Hall Institute for Medical Research. Aire knockout mice (C57BL6) were generated by a targeted disruption of the Aire gene. C57BL/6 mice were used at 6–10 wk of age unless indicated otherwise. All studies were performed according to protocols approved by the Melbourne Health Animal Ethics Committee.

Reagents

Abs were purchased from BD Pharmingen and were FITC-conjugated anti-Ly51 (clone 6C3) and anti-CD4 (clone L3T4); PE-conjugated anti-MHC class II (MHC-II) I-A^b,d,q I-E^d,k (clone M5/114.15.2), anti-CD25 (clone PC61.5), and anti-CD3 (clone 145-2C11); biotin-conjugated anti-Ly51 (clone 6C3), anti-CD69 (clone H1.2F3), and anti-CD44 (clone IM7); allophycocyanin-conjugated anti-CD45 (clone 30F11), anti-CD8 (clone 53-6.7), and anti-CD4 (clone L3T4); and streptavidin-PerCp5.5 used to reveal the biotin staining. Rat anti-Aire mAb was produced in the Monoclonal Antibody Facility at the Walter and Eliza Hall Institute of Medical Research (14). FITC anti-Vβ5 (clone MR9-4), Vβ8.1/8.2 (clone MR5-2), Vβ11 (clone RR3-15), and Vβ13 (clone MR12-3) Abs were purchased from BD Pharmingen, and biotin-conjugated anti-Vβ2, -Vβ3, -Vβ4, -Vβ6, -Vβ7, -Vβ9, -Vβ10, -Vβ12, and -Vβ14 were kindly provided by F. Carbone (Melbourne University, Melbourne, Australia).

Cell preparations

mTEC were enriched from the thymus as previously described (30). Briefly, thymi were collected into MT-RPMI and, following brief agitation using a wide bore glass pipette, the sample was subjected to enzymatic digestion. Thymic fragments were incubated in 5 ml of 0.125% (w/v) collagenase D with 0.1% (w/v) DNase I (Boehringer Mannheim) in MT-RPMI at 37°C for 15 min. Cells released into suspension were removed after the fragments had settled. This was repeated three to four times with fresh medium, and in the final digest the collagenase D was replaced with trypsin. Each cell suspension was counted and the final two or three enrichments were pooled. A negative depletion was performed to enrich for CD45⁻ cells using CD45 microbeads (Miltenyi Biotec) and the autoMACS system (Miltenyi Biotec), as per the manufacturer’s instructions. Cells were stained with Ly51-FITC (clone 6C3), MHC-II-PE (clone M5/114.15.2), and CD45-allophycocyanin (clone 30F11) mAbs, and CD45⁻MHC-II^highLy51^low cells (mTEC) were sorted on a FACSAria instrument (BD Biosciences).

Flow cytometric analysis

The fluorescently-labeled preparations were analyzed on a FACSCalibur instrument (BD Biosciences). Color compensation was checked using appropriately stained cell controls. Four fluorescent channels were used for the immunofluorescence staining (FL1 for FITC, FL2 for PE, FL3 for PerCP5.5, and FL4 for allophycocyanin). mTEC preparations were labeled with anti-CD45 (clone 30F11), anti-MHC-II (clone M5/114.15.2), and anti-Ly51 (clone 6C3). The Aire intracellular staining was performed using the BD Cytofix/Cytoperm kit (BD Biosciences) according to the manufacturer’s instructions.

TCR repertoire analysis

The immunoscope technique was used to evaluate TCR repertoire diversity by amplification of rearranged TCR Vβ-chain cDNAs (31). Each TCR Vβ family was assessed individually using forward TCR Vβ-specific primers and reverse constant β region (Cβ)-specific primers. The products were visualized by an elongation reaction using a nested fluorescent Cβ primer. The elongation products containing the CDR3 were separated by capillary electrophoresis (Applied Biosystems 3730 DNA analyzer). Size standard LIZ 500 (Applied Biosystems) was applied to each run. The PCR product was denatured by addition of formamide and heat to produce single-stranded products. The results were analyzed using GeneMapper software version 3.7 (Applied Biosystems) and are shown as graphic distribution of size peaks centered on a peak corresponding to a CDR3 of 8–10 amino acids. Normal profiles have a Gaussian-like distribution of an average of six to eight peaks per TCR Vβ family (polyclonal Gaussian). The peaks are spaced by three nucleotides. Comparisons with size standards from WT mice showed that they corresponded to in-frame transcripts. Alterations in the normal profile result in a polyclonal skewed profile, causing a shift in the center of the CDR3 distributions or a disruption of the normal peak distribution due to clonal T cell expansions. Data are expressed as percentage of oligoclonal, polyclonal skewed, or polyclonal Gaussian Vβ families per total number of Vβ families amplified from a sample. A unique length of the CDR3 is not necessarily associated with the same sequence, and the number of transcripts with a given length of CDR3 is proportional to the area under the peak. Then, for each CDR3 length profile experiment, the normalized profile was compared with the sample one, and the difference was plotted on a graph, according to the CDR3 length and percentage of perturbation. This first analysis step only gives qualitative information, because there is no indication of the amount of Vβ altered. Percentages of CDR3-length distribution alterations are represented as a color code, from deep green to dark red. The x-axis displays the mouse Vβ families; the y-axis gives the CDR3 lengths.

Western blot analysis

After homogenization on ice, testis, liver, lung, salivary gland, pancreas, and eye tissue from Rag^−/− mice were lysed and boiled in a buffer containing 1% SDS, 10% glycerol, 10% 2-ME, 40 mM Tris (pH 6.8), and 0.01% bromophenol blue. Equal amounts of cell lysate were run on a 12% Tris-glycine polyacrylamide gel (Invitrogen) and then transferred to a Hybond-ECL Western membrane (Amersham) at 50 V for 40 min. The membrane was blocked for 4 h in a 5% milk solution in PBS and was incubated overnight at 4°C with sera from Aire-deficient or WT mice diluted 1/500. The membrane was washed several times with PBS and 0.1% Tween 20 and then incubated with secondary Ab, HRP-conjugated sheep anti-mouse IgG (1/1000; Amersham), for 1 h. Detection of staining was performed using ECL Western blotting detection reagents (Amersham) according to the manufacturer’s instructions.

Radioimmunoassays

Abs to insulin were detected as described in Yu et al. (32) with slight modifications. Briefly, ¹²⁵I-insulin (Amersham) (∼20,000 cpm) was incubated with 3 μl of serum with or without unlabeled human insulin at 1/5 dilution of serum for 3 days at 4°C in buffer A (20 mM Tris-HCl buffer (pH 7.4) containing 150 mM NaCl, 1% BSA, 0.15% Tween 20, and 0.1% sodium azide) on a end-over-end shaker. Fifty microliters containing 50% protein A-Sepharose and 8% protein G-Sepharose (Amersham) was added and incubated for 45 min at 4°C on an end-over-end shaker. The samples were washed four times with 1 ml of cold buffer B (20 mM Tris-HCl buffer (pH 7.4) containing 150 mM NaCl, 0.15% Tween 20, and 0.1% sodium azide) and transferred to scintillation tubes with 3× 100 μl of buffer B.

Mouse Cyp1a2 and Gad65 cDNAs were amplified with primers mCYP-F1 (ATGGCGTTCTCCCAGTACATCT) and mCYP-R1 (TCACTTGGAAAAGCGTGGC) and mGAD-F1 (ATGGCGTCTCCTGGCTCC) and mGAD-R1 (TTACAAATCTTGTCCGAGGCG), respectively. The full-length cDNAs were cloned into pCR2.1-TOPO vector using a TOPO TA cloning kit (Invitrogen). The plasmids containing Cyp1a2 and Gad65 cDNAs were in vitro transcribed and translated using a TNT-coupled transcription and translation kit (Promega) using [³⁵S]methionine. In vitro translated mouse Gad65 and Cyp1a2 were purified through the Sephadex G-25 DNA grade column. The Ags (∼40,000 cpm) were suspended in 50 μl of RIP buffer (150 mM NaCl, 10 mM Tris-HCl (pH 7.4), 0,005% Tween 20, aprotinine 1/1000, leupeptine 1/1000), and 50 μl of 1/10 diluted mouse serum was added and incubated for 30 min at room temperature. Subsequently, 10 μl of protein G-Sepharose (Amersham Biosciences) was added and incubated for 1 h at room temperature. The samples were washed four times with RIP buffer. Abs to C. albicans enolase and Hsp90 were detected by radioimmunoassay using in vitro translated/transcribed proteins as described earlier.

In all radioimmunoassays, 3 ml of liquid scintillation cocktail (Optiphase HiSafe3; PerkinElmer) was added for signal detection and counted in a scintillation counter. The results are expressed as absolute values. The value of positive control for insulin (a human serum from a person with type 1 diabetes) was ∼4000 cpm.

Antiviral IFN neutralization assays

For detection of mouse IFN-αA (Mu-IFN-αA; PBL Laboratories), mouse L-929 cells (American Type Culture Collection CCL-1) were pretreated with dilute Mu-IFN-αA (10 laboratory units (LU)/ml) that had been preincubated for 2 h with serial dilutions of test sera or neutralizing rat anti-Mu-IFN-α mAb (PBL Laboratories). Following incubation at 37°C for 24 h, the cells were challenged with encephalomyocarditis virus for 24 h, stained with 0.05% amido blue-black, fixed with 4% formaldehyde in acetic acid buffer, and stain-eluted with 0.15 ml of 0.05 M sodium hydroxide before absorbances were read at 620 nm. The neutralizing Ab titer was the reciprocal of the dilution of serum that reduces 10 LU/ml Mu-IFN-αA to 1 LU/ml (the normal endpoint of antiviral assays) (33, 34). The cutoff for positivity was a titer of 40.

Expression analysis

RNA extraction from WT and Aire-deficient mTEC was performed using the RNeasy Micro kit (Qiagen) according to the manufacturer’s instructions. RNA amplification was done at the Australian Genome Research Facility, following the Affymetrix manual (701725 rev5). Briefly, 100 ng of total RNA was amplified using T7-oligo(dT) and the MEGAscript T7 kit (Ambion). A second round of cDNA synthesis was performed using first round-amplified RNA. Amplified RNA was biotin-labeled using the GeneChip IVT labeling kit. Fifteen micrograms of labeled RNA was then fragmented as described in the Affymetrix manual. The Mouse Genome 430 2.0 arrays were hybridized overnight, washed, and scanned using a GeneChip scanner 3000 (Affymetrix). For analysis of the Affymetrix GeneChips, the intensities for each probe set were normalized and summarized using the robust multiarray analysis algorithm (35). Analysis of the microarray and PCR expression data was undertaken using the limma software package of the Bioconductor Project (36). Differential expression between WT and Aire-deficient samples was assessed using linear modeling and empirical Bayes moderated t statistics (37). Values of p were adjusted using the Benjamini and Hochberg method to control the false discovery rate at 0.05. The microarray experiment contained three WT samples, two Aire-deficient samples, and three other closely related sample,s which were included in the linear model but not used to compute the t statistics. Array data were deposited into the Gene Expression Omnibus public database (www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc, accession no. GSE14365).

Reverse transcription was performed using SuperScript III RNase H-reverse transcriptase (Invitrogen), with 250 ng of random primers (Promega) per reaction according to the manufacturer’s instructions. Quantitative real-time PCR was performed on the ABI Prism 7900HT sequence detection system using prevalidated TaqMan assays and TaqMan Universal Master Mix with AmpliTaq Gold (Applied Biosystems) as per the manufacturer’s instructions. Ct values were determined using ABI Prism SDS version 2.2 (Applied Biosystems), and relative expression was calculated using the 2^−ΔΔCt method. The PCR experiment contained two WT and two Aire-deficient samples. In this case, moderated paired t statistics were used to account for any differences between two processing times.

Infertility test

Sperm collected from mice was incubated for 2 h in mT6 before in vitro fertilization. Oocytes were collected from (CD57BL/6 × CBA)F₁ mice and incubated for 6 h with sperm from the above mice. Fertilization and different stages of embryo development (two cells, four cells, morulae, and blastocyst) were enumerated (38).

Histology

Dissected tissues were immersed in Bouin’s fixative overnight and then washed with 70% ethanol. Fixed tissues were then embedded in paraffin using standard method and then sectioned, at 2-μm thickness, onto glass slides and stained with H&E.

Organ infiltration

The organs from 5-mo-old mice were digested three times at 37°C in fresh medium containing 1 mg/ml collagenase D and 2 μg/ml DNaseI (Roche). Cells were filtered through a 100-μm mesh to remove clumps, and samples were treated with 0.15 M NH₄Cl for lysis of erythrocytes. Cell preparations were labeled with anti-CD4, anti-CD8, and anti-CD3 and either anti-CD69 or anti-CD44.

Immunofluorescence

Acetone-fixed 5-μm thymus sections from WT, Aire^+/−, and Aire^−/− mice were stained with rat anti-Aire-FITC (clone 5H12) and rabbit anti-β-galactosidase (β-gal; a kind gift from E. Handman, Walter and Eliza Hall Institute, Melbourne) with Alexa Fluor 546 donkey anti-rabbit IgG (Invitrogen). After washing, sections were stained with Hoechst 33342 (Sigma-Aldrich) for 5 min and washed a further three times. Sections were mounted and cover-slipped using fluorescent mounting medium (Dako) before viewing with confocal or fluorescent microscopy.

Yeast infection

C. albicans isolate 3630 was obtained from the Mycology Reference Laboratory at the Royal North Shore Hospital, Sydney, Australia (a kind gift from Dr. R. B. Ashman) and stored at −70°C in Sabouraud’s broth/15% (v/v) glycerol. Strain 3630 was isolated from the nail of a patient with cutaneous candidiasis. For use, yeasts were grown in Sabouraud’s broth for 48 h at room temperature with continuous agitation on a magnetic stirrer. Blastospores were washed in PBS and adjusted to the appropriate concentration for inoculation. For oral infection, mice were inoculated orally with 10⁸ live C. albicans in 20 μl of sterile PBS. The infection was monitored by swabbing the oral cavity on days 1, 4, 8, 14, 18, and 21, with sterile cotton swabs moistened with sterile PBS, and directly plating on Sabouraud’s agar plates. Agar plates were incubated for 48 h at 30°C. All inoculation and sampling procedures were conducted under anesthesia. CFU were counted on Sabouraud’s agar plates and the counts were assigned into five groups correlating with the level of recoverable yeast from the oral cavity, as previously described (39). This provided a semiquantitative measure of the level of floridity of the infection. The scoring system used was as follows: 0, no detectable yeast; 1, 1–10 CFU/plate; 2, 11–100 CFU/plate; 3, 101-1000 CFU/plate; 4, ≥1001 CFU/plate. For systemic infection, mice were injected with 3 × 10⁵ C. albicans 3630 in 200 μl of PBS via the tail vein. Mice were sacrificed on day 5 after infection. The kidneys were harvested, weighed, suspended in 1 ml of sterile PBS, and homogenized. Then, 100 μl of suspension was titrated in duplicate on Sabouraud’s agar plates, incubated at 30°C for 48 h, and colonies were counted. The results were expressed as log₁₀ CFU/g tissue.

Disruption of the Aire gene in exon 8

To investigate the biological functions of Aire, we generated mice deficient for this gene via homologous recombination of targeting vectors in mouse embryonic stem cells (Fig. 1⇓). Insertion of the Aire targeting vector disrupted exon 8 and brought the LacZ reporter gene under the control of the endogenous Aire promoter, creating an Aire-LacZ fusion gene. A phosphoglycerate kinase neomycin (PGK-Neo) cassette was used to select positive recombination events and was later removed using the flanking LoxP sites and Cre recombinase (Fig. 1⇓, A and B). This insertion was designed to interrupt the PHD1 domain, reflecting the common 13-bp deletion in human APECED patients, localized after the SAND domain and within the PHD1 domain (c.1094_1106del, p.L323fsX373) (25). Three stop codons and a polyadenylation signal were also included to prevent any transcription of full-length Aire (Fig. 1⇓C). We verified the targeted disruption by three-primer PCR, showing a band at 202 bp for the WT locus and a band at 334 bp for the knockout locus (Fig. 2⇓A). As previously described (14), Aire protein expression was not detected by flow cytometry, confirming Aire deficiency (Fig. 2⇓B). Examination of CD45⁻ cells from WT mice for expression of MHC-II and Aire showed that only MHC-II^high cells expressed Aire (14). Interestingly, we observed a greater proportion of these cells in Aire-deficient mice compared with WT mice (Fig. 2⇓B). While these findings demonstrate that native Aire is absent in Aire-deficient mice, they did not exclude the expression of a truncated Aire protein. Insertion of a LacΖ reporter into the Aire gene generates an Aire-β-gal fusion protein. Staining with an anti-β-gal Ab revealed β-gal expression in Aire^+/− and Aire^−/− thymi (Fig. 2⇓C), as previously reported (14). Interestingly, β-gal colocalized with native Aire as shown in Fig. 2⇓C in Aire^+/− thymi, suggesting an ability of the fusion protein to correctly localize to nuclear bodies and perhaps to dimerize. However, in the thymi of our Aire-deficient mice, β-gal was expressed within the nucleus to a lesser extent than seen in the heterozygote and did not show the typical speckled staining pattern of Aire, but rather appeared to form several larger speckles that may have be an aggregation of nonfunctional fusion protein.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

The Aire gene knockout construct. A, Plasmid construct used for the generation of Aire knockout mice. B, The Aire gene was disrupted by homologous recombination with the targeting vector in C57BL/6 ES cells. C, The PGK-Neo cassette was used to select positive recombination events and was later removed using the flanking LoxP sites and Cre recombinase. This recombination event places the LacZ reporter gene under the control of the endogenous Aire promoter in fusion with exon 8, disrupting exon 8 and thus the first PHD finger domain. Three stop codons and a polyadenylation signal (PolyA) were included to prevent any transcription of full-length Aire. This encodes a truncated Aire protein size of 1328 aa containing the first 310 aa of native Aire. Restriction endonuclease sites: BG, BsrG1; M, Mlu1; S, SalI. D, Schematic representation of normal Aire protein (top) and truncated Aire fusion protein (bottom).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Analysis of Aire expression and the thymic compartment of Aire^−/− mice. A, Four- to 6-wk-old C57BL/6 mice were genotyped by PCR from tail genomic DNA. B, Aire is expressed by specific nonhematopoietic thymic cells: CD45⁻MHC-II^highLy51^low. A negative depletion was performed to enrich for CD45⁻ cells and analyzed for their expression of CD45, MHC-II, Ly51, and Aire. After gating on CD45⁻ cells, mTEC (boxed areas) are distinguished from other cell subsets according to their expression of MHC-II and Ly51 (top row). Aire expression is compared with MHC-II expression on CD45⁻ cells. C, Aire immunofluorescence on thymic sections. Tissues from WT, Aire heterozygous, and Aire-deficient mice were stained with an anti-Aire Ab (green), anti-β-gal Ab (red), and Hoechst (blue). The immunofluorescence was visualized by laser scanning confocal microscopy. Each row represents different magnifications with scales indicated. D, Estimation of the number of CD45⁻ cells and CD45⁻MHC-II^highLy51^low cells in single 6-wk-old thymus: each single thymus was digested and stained for analysis of expression of CD45, MHC-II, and Ly51. Sample data from 2 × 10⁴ CD45⁻ cells were acquired and results are expressed on a logarithmic y-axis as mean ± SD of eight mice per group.

Thymic epithelial cell compartment is disturbed

It has been established that cortical and medullary epithelial cells can be distinguished from cells of hematopoietic origin by their lack of expression of CD45 and can be identified according to the expression of MHC-II and Ly51, with both populations expressing high levels of MHC-II but only cortical thymic epithelial cells (cTEC) expressing Ly51 (14). Thus, after the elimination of contaminant CD45⁺ cells, four major populations of CD45⁻ cells were distinguished: MHC-II^highLy51^low (mature mTEC), MHC-II^lowLy51^low (progenitor or immature mTEC), MHC-II⁻Ly51^low (fibroblast and endothelial cells), and MHC-II^highLy51⁺ (cTEC). Because the enrichment procedure used above for preparing CD45⁻ cells may have affected the proportion of different subpopulations of TECs, we decided to examine the phenotype of these subpopulations within nonenriched thymus populations, after simple enzymatic digestion and red cell elimination. Despite TEC populations being very rare, it was still feasible to enumerate CD45⁻ cells and their subpopulations in WT and Aire^−/− mice (Fig. 2⇑D). Based on these analyses, we estimated that CD45⁻ cells represented 1 ± 0.4% and 0.8 ± 0.2% of a whole thymus in WT and Aire^−/− mice, respectively. Of CD45⁻ cells, the MHC-II^highLy51^low subset made up 0.5 ± 0.2% in WT and 2.1 ± 0.3% in Aire^−/− mice (Fig. 2⇑D). Thus, Aire^−/− mice had ∼4-fold more MHC-II^high mTEC than did WT littermates. Complementary to this, the immature mTEC (imTEC) compartment, considered to be CD45⁻MHC-II^lowLy51^low, was reduced by 25% in the Aire^−/− mice. Thus, the mTEC/imTEC ratio was 0.3 ± 0.16 and 1.3 ± 0.3 in WT and Aire^−/− mice, respectively. The cortical compartment remained consistent between WT and Aire^−/− mice, representing 2.5 ± 0.8% of CD45⁻ cells, with the total epithelial cell compartment representing only ∼5% of the CD45⁻ thymic stromal cell population. In summary, although the architecture of the thymus appeared to be largely normal and unmodified (14), as reported in other Aire-deficient models, the epithelial compartment appeared to be biased toward mature mTEC in Aire^−/− mice. These data establish that the impact of Aire deficiency extends beyond the lack of expression of a subset of TSA in mature mTEC and affects changes in the thymic environment that would be compatible with a role for Aire in mTEC differentiation and/or survival.

Central T cell compartments appear normal

Aire is an important mediator of central tolerance that promotes the promiscuous expression of TSA in the thymus (40), shown to be associated with the negative selection of autoreactive T cells (4, 6). Recently, Aire has also been reported to have a role in the generation of regulatory T cells within the thymus (7). Furthermore, our demonstration that the ratio of mature-to-immature mTEC is altered suggests other potential effects on T cell education. Given these results, we decided to examine the T cell compartment in different immunological organs (thymus, spleen, and lymph node). After staining thymocytes for CD4, CD8, and Foxp3, we discriminated CD4⁺ and CD8⁺ single-positive and double-positive T cells and identified regulatory CD4⁺ T cells by their expression of Foxp3 (Fig. 3⇓). No significant difference was observed for any of these populations, although a trend toward a decreased proportion of CD4⁺, CD8⁺, and CD4⁺ Foxp3⁺ T cells was noted. Moreover, the level of Foxp3 staining was slightly, although not significantly, reduced in Aire^−/− mice (Fig. 3⇓A). When the double-negative (CD4⁻CD8⁻) compartment was analyzed for expression of CD44 and CD25, no significant difference was observed, demonstrating minimal impact of Aire on early thymocyte development (Fig. 3⇓D). Taken together, these results show that in terms of proportion and number, the T cell compartment is not significantly modified, suggesting a minimal impact of Aire on thymocyte maturation, as has been previously reported by others (41). However, this does not exclude a significant role for Aire in the fate of certain TSA-specific cells. Indeed, Aire was previously found to be necessary for normal clonal deletion in OT-II/RIP-mOVA double-transgenic animals as well as in other TCR neo-self-Ag double-transgenic systems (6). Using the same system, our Aire-deficient mice show a severe defect in their ability to delete CD4⁺ autoreactive thymocytes (F. X. Hubert, H. S. Scott, W. R. Heath, unpublished data).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Thymic T cell compartment in Aire-deficient mice (open bars) vs WT mice (filled bars). Each thymus was crushed through a sieve and the cells were stained with CD4-FITC, Foxp3-PE, and CD8-allophycocyanin and acquired on a FACSCalibur. Four subsets were distinguished: CD4⁺CD8⁻ T cells, CD4⁻CD8⁺ T cells (A and B), double-positive CD4⁺CD8⁺ thymocytes (DP) (A and C), and double-negative CD4⁻CD8⁻ thymocytes (DN) (A and D). Foxp3 expression was analyzed on CD4⁺CD8⁻ T cells (A and B). The same cells were stained with CD4-FITC, CD25-PE, CD44-biotin, and CD8-allophycocyanin. After gating on CD4⁻CD8⁻ cells, four double-negative thymocyte subsets were distinguished according the expression of CD25 and CD44: DN1 (CD44⁺CD25⁻), DN2 (CD44⁺CD25⁺), DN3 (CD44⁻CD25⁺), and DN4 (CD44⁻CD25⁻) (D); y-axis as mean ± SD of six mice.

Increased number of peripheral activated T cells in Aire^−/− mice

Although the mTEC compartment was perturbed and there was a defect in the negative selection process in the thymus of our Aire^−/− mice, an analysis of peripheral lymphoid organs revealed no obvious differences in the number or proportion of CD4⁺ T cells, CD4⁺ Foxp3⁺ T cells, or CD8⁺ T cells in the spleen and lymph node of Aire^−/− mice compared with WT mice (Fig. 4⇓A). As Aire is involved in T cell selection and plays an important role in preventing autoimmunity (3, 4, 6), we analyzed whether Aire^−/− mice may have a greater proportion of activated T cells (Fig. 4⇓B), potentially reflecting the presence of autoimmunity. Examination of CD69 expression on CD4⁺ T cells as an early activation marker showed no difference in the spleen (23 ± 3% in WT mice, 20 ± 5% in Aire^−/− mice), but a 5% increase was observed in lymph nodes of Aire^−/− mice (10 ± 0.6% in WT mice, 15 ± 0.5% in Aire^−/− mice). Analysis of the proportion of CD44^highCD4⁺ T cells (Ag-experienced cells) in the same organs (Fig. 4⇓B) revealed a significant increase in these cells in the spleen (27 ± 0.8% in WT mice, 32 ± 1.2% in Aire^−/− mice) and lymph node (9 ± 0.3% in WT mice, 11 ± 0.4% in Aire^−/− mice) of Aire^−/− mice. Interestingly, examination of the same activation markers on CD8⁺ T cells did not reveal any significant differences (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Peripheral T cell compartment in Aire-deficient mice vs WT mice. A, Spleen or lymph node organs were crushed through a sieve and the cells were stained with CD4-FITC, Foxp3-PE, and CD8-allophycocyanin and acquired on a FACSCalibur. Two subsets were distinguished: CD4⁺CD8⁻ T cells and CD4⁻CD8⁺ T cells. Foxp3 expression was analyzed on CD4⁺CD8⁻ T cells; y-axis as mean ± SD of six mice. B, Activated CD4⁺ T cells in periphery in Aire-deficient mice (open bars) vs WT mice (filled bars) were analyzed according to CD69 or CD44 expression. Cells were then stained with CD4-FITC, CD8-allophycocyanin, and CD69-biotin or CD44-biotin and acquired on a FACSCalibur. The proportion of early activated CD69⁺CD4⁺ T cells or of activated/memory CD44⁺CD4⁺ T cells is estimated in a bulk of CD4⁺CD8⁻ T cells; y-axis as mean ± SD of six mice.

Vβ repertoire of CD4⁺ T cells

The modest increase of activated CD4⁺ T cells in Aire^−/− mice could be the result of an overrepresentation of T cells that are normally destined for negative selection. To examine whether this possibility is reflected by an alteration in the Vβ repertoire, we analyzed TCR Vβ expression on thymic and splenic CD4⁺ T cells using flow cytometry (Fig. 5⇓A). While the Vβ repertoire differed somewhat between thymic and peripheral CD4 T cells, no difference was observed between WT and Aire^−/− mice. This indicated that there was no overt difference in Vβ selection in mutant mice. The same observations were made for thymic and splenic CD8⁺ T cells (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Vβ repertoire of CD4⁺ T cell from Aire-deficient and WT mice. A, Vβ repertoire of CD4⁺ T cell from Aire-deficient mice (open bars) vs WT mice (filled bars). Thymus or spleen organs were crushed through a sieve and cells were then stained to CD4, CD8, CD25, and different Vβ and acquired on a FACSCalibur. The proportion of different Vβ⁺CD4⁺ T cells is estimated in a bulk of CD4⁺CD8⁻ T cells; y-axis as mean ± SD of six mice. B, Variations in both length and distribution of the most variable region of the TCR Vβ-chain, the CDR3, using the immunoscope method. Examples of immunoscope of major Vβ families from CD4⁺ T cells according to their different CDR3 lengths in thymus, spleen, and lymph node are shown. The alteration was determined by comparing the immunoscope profile of each population of CD4⁺ T cells from WT mice to CD4⁺ T cells from each Aire-deficient mouse. C, Views representing the percentage of perturbation in each different CDR3 length found in each Vβ family. Comparisons with size standards from WT mice show that they correspond to in-frame transcripts. Alterations of the normal profile result in a polyclonal skewed profile, leading to a shift in the center of CDR3 distribution or a disruption of the normal peak distribution due to clonal T cell expansions (see Materials and Methods).

To more specifically monitor alterations in the TCR repertoire, we analyzed variations in both length and distribution of the most variable region of the TCR Vβ-chain, the CDR3. For this we used a RT-PCR approach that subdivides a bulk CD4⁺ T cell population (i.e., from thymus, spleen, lymph nodes) based on rearranged Vβ/Jβ gene segments and the resulting length of the TCR’s third CDR (CDR3). This technique, referred to as immunoscope analysis (31), has the potential to identify clonal expansions. It allows identification of modifications in the CDR3 length Gaussian distribution patterns typically observed in WT animals. These CDR3 length alterations observed in some Vβ families can be visualized graphically (Fig. 5⇑B), allowing global assessment of total Vβ alterations in each sample. In our case, only Gaussian distributions were observed in both WT and Aire^−/− mice, as illustrated by the profiles of the major Vβs of CD4⁺ T cells (Fig. 5⇑B), suggesting polyclonal expansion. Thus, no obvious oligoclonal expansions dominated the repertoire of Aire^−/− mice. Using these data, Fig. 5⇑C shows the percentage perturbation of different CDR3 lengths for each Vβ (difference between the normalized WT profile and the sample for each CDR3 length). This analysis confirmed that the variation was generally low (i.e., <10%). Occasionally, greater alterations of between 10% and 30% were observed, but these were not reproducibly found in every Aire^−/− mouse. The exception to this was Vβ18, which was perturbed by >10% in all Aire^−/− thymi, but not in the other lymphoid organs. As a generalization, these results suggest that each individual Aire^−/− mouse developed its own polyclonal repertoire of autoreactive T cells.

Classic and major autoantibodies found in APECED are not present in Aire-deficient mice

APECED patients have high levels of serum Abs reacting specifically with components in the affected organs. A number of the autoantigens, which are predominantly intracellular enzymes, have been identified such as ATP4a, CYP17, CYP21, CYP11A1, CYP1A2, and GAD65 (summarized in Ref. 42). Radioimmunoassay was used to screen Aire-deficient mice for the presence of autoantibodies against mouse Gad65, Cyp1a2, insulin, Atp4a, enolases, and Hsp90 with negative results (data not shown). Recently, high-titer neutralizing IgG autoantibodies reactive to most IFN-α subtypes have been found in APS1 patients (20, 43). Sera from 33 Aire-deficient mice and 12 normal mice were therefore screened for the presence of such autoantibodies to Mu-IFN-α but, in contrast to a rat anti-Mu-IFN-α mAb with a neutralizing titer of 1280, none of these sera was positive (titer <40). This suggests that Aire-deficient mice do not have the same autoantibody pattern as do human patients.

Expression of APECED autoantigens in Aire-deficient mTEC

Several studies have shown a correlation between loss of intrathymic expression of certain TSAs and the development of autoimmunity targeted toward these Ags (4, 44, 45, 46). However, Aire-deficient mice can also develop autoimmunity against Ags whose intrathymic expression is unaltered in the absence of Aire (19, 23). To determine whether the autoantigens commonly targeted in APECED patients were Aire activation targets, Affymetrix expression arrays were performed on WT and Aire^−/− mTEC. Results show that Nalp5, Cyp11a1, Ins2, Th, and Pah were significantly down-regulated in Aire^−/− mTEC, with the 3.9-fold reduction in expression of Pah being the largest significant fold change observed (Table I⇓). Quantitative real-time RT-PCR was employed to confirm Affymetrix results for these lowly expressed TSA genes. Results confirmed the significant down-regulation of Nalp5, Cyp11a1, Th, Pah and Ins2 in Aire^−/− mTEC (Table I⇓). However, Gad2 and Cyp1a2 were only found to be down-regulated in Aire^−/− mTEC by quantitative RT-PCR. Taken together, these results suggest that Aire controls the expression of some, but not all, of the genes whose products are commonly targeted Ags in APECED patients.

Infertility is highly prevalent in male Aire-deficient mice

Infertility problems have been described in human APECED disease with a high prevalence in females (60% vs 14% of males) (1). The infertility problems previously noticed in Aire-deficient mice (3, 19, 22) were also observed in our Aire-deficient mice. Indeed, when intercrossed, Aire-deficient mice failed to produce any litters. Interestingly, this issue was most often the result of a fertility defect in the Aire-deficient males and not in the females. To confirm their sterility, we assessed the quality and the capacity of sperm from Aire-deficient mice to fertilize an oocyte via in vitro fertilization. We observed a 73% fertilization success rate with Aire-deficient sperm compared with 93% for a WT sperm. Of particular note, there was significant mouse-to-mouse variation in the quality (as indicated by mobility parameters and ultimately reflected in fertilization rates) and quantity of sperm retrieved from our Aire^−/− males. This leads us to suspect the presence of anti-sperm Abs. The testis histology of Aire^−/− mice was qualitatively identical to that of WT mice. At the ages observed, there were no differences in testis weights between genotypes.

Autoantibodies are detected against different organs

Although we observed only a minor increase in activated CD4⁺ T cells, slight perturbations to the Vβ repertoire, and no autoantibodies against the known major targets in APECED, we analyzed autoantibody production against antigenic tissue proteins by our Aire-deficient mice. Characteristically, sera of APECED patients contain specific autoantibodies indicative of an ongoing autoimmune response against target organs. To investigate the range and the specificity of the autoantibodies elicited by the Aire deficiency, we probed immunoblots of protein extracts from several tissues (testis, lung, liver, salivary gland, eye, and pancreas) with sera from individual Aire-deficient mice (Fig. 6⇓A). These analyses revealed sporadic and distinct patterns for each mouse, with some common bands. While sera from mice 1 and 3 displayed strong reactivity against the liver and pancreas, mouse 1 also reacted to components of testis and salivary gland, whereas mouse 3 reacted to proteins from the lung and eye. Interestingly, sera from all four mice had autoantibodies against liver and pancreas, while three reacted against testis, lung, salivary gland, and eye. Despite the presence of autoantibodies against the pancreas in all four Aire^−/− mice, diabetes was not detected in any of these animals (data not shown). Taken together, these results showed that even Aire^−/− mice of the same genetic background develop autoimmunity against different organs and autoantigens. These data provide a potential explanation for the observed mouse-to-mouse variations in male fertility with the presence of anti-testis Abs in the serum.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Evidence of autoimmunity in Aire-deficient mice. A, Autoantibodies generated in Aire-deficient mice. Each lane of the immunoblot shows autoantibodies in the serum from 20- and 25-wk-old individual Aire KO mice or from a WT mouse immunoblotted against testis (1), lung (2), salivary gland (3), liver (4), eye (5), and pancreas (6) Ags. B, Eye and retinal sections, showing uveoretinitis in Aire-deficient mice, reveal development of spontaneous autoimmunity in eyes. Tissue sections from WT and Aire-deficient mice are stained with H&E. The images show the different severity grade observed in each group of mice: normal (no disease), mild (sporadic infiltration), moderate (clear infiltrate with many foci), and severe (many extensive areas of infiltration). Marked as: l, lens; c, cornea; i, iris; r, retina; n, optic nerve; v, vitreous; 1, pigmented epithelium; 2, photoreceptor layer; 3, outer nuclear layer; 4, outer plexiform layer; 5, inner nuclear layer; 6, inner plexiform layer; 7, ganglion cell layer and inner limiting membrane. C, The graph gives an overview of frequency and severity of autoimmunity in the eye. D, Salivary gland sections, showing infiltration (arrows) in Aire-deficient mice, reveal development of spontaneous autoimmunity in submandibular glands. Tissue sections from WT and Aire-deficient mice are stained with H&E. The images show the different severity grade observed in each group of mice: normal (no disease), mild (sporadic infiltration), moderate (clear infiltrate with many foci), and severe (many extensive areas of infiltration). E, Overview of frequency and severity of autoimmunity in the salivary gland. F, Proportion of the T cell subsets in the lung, liver, salivary gland, and pancreas from 12 WT and 12 Aire-deficient mice.

Histology

To confirm the development of autoimmunity in Aire-deficient mice, histological analysis was performed on several organs, including salivary glands and eyes, of 3- and 7-mo-old mice. It has been shown that Aire-deficient mice develop a spontaneous autoimmune uveitis that increases in frequency and severity with age and is characterized by mononuclear infiltrate and autoantibodies specific to the photoreceptor layer of the retina (3, 44). The same observations were made in our Aire-deficient mice. Indeed, irrespective of age, uveoretinitis was present in most Aire-deficient mice but not in WT mice, with severe retinal destruction in 40% of animals (Fig. 6⇑, B and C). A mild phenotype was observed in most Aire-deficient mice with distortion of the outer and inner nuclear layer and a reduction in size of the photoreceptor layer (Fig. 6⇑, B and C). In some cases, vitritis was also found in Aire-deficient mice (data not shown).

Salivary glands consist of exocrine tissue whose main function is to secrete saliva, thereby keeping the mouth and other parts of the digestive system moist (47). Aire-deficient mice have been reported to exhibit extensive lymphocytic infiltration of salivary and lacrimal glands and destruction of epithelial cells (3, 19), as seen in primary Sjögren’s syndrome (48). The histological phenotype in the submaxillary gland was ranked according to four degrees of severity (Fig. 6⇑D). Analysis of the extent of the infiltration in our Aire-deficient mice showed more severe infiltration than that seen in normal mice (Fig. 6⇑E). We also observed an effect of age on the degree of lymphocytic infiltration of exocrine tissues. Moreover, when we analyzed the parotid and sublingual glands in Aire-deficient mice, we observed sporadic infiltration in the sublingual gland (40% in Aire^−/− mice vs 0% in WT mice), while most Aire-deficient mice analyzed showed severe infiltration in the parotid gland (80% in Aire^−/− mice vs 0% in WT mice) (data not shown). The pancreas, liver, stomach, kidney, epididymis, thyroid, parathyroid, testis, ovary, and epidermis, also affected in APECED disease, appeared histologically normal in our Aire-deficient mice (data not shown). Flow cytometry experiments revealed an abnormal infiltration by activated CD69⁺ and CD44⁺ CD8⁺ T cells in the salivary glands of Aire-deficient mice, but not in the lungs, liver, or pancreas (Fig. 6⇑F).

Aire-deficient mice are not susceptible to Candida infection

APECED patients develop candidiasis (1), an infection caused by the fungus C. albicans, usually before 5 years of age, with the nails, skin, tongue, and mucous membranes being the major sites of infection. Additionally, the role of neutralizing IgG autoantibodies reactive to most IFN-α subtypes in APS1 syndrome is also unclear. One interesting possibility was a potential link between these autoantibodies and the propensity to develop susceptibility to Candida infections in these patients, as suggested by Meager et al. (43). We tested the susceptibility of our Aire-deficient mice to C. albicans after oral or systemic infection. In Fig. 7⇓, we show that Aire-deficient mice respond equivalently to WT mice to oral Candida infection, irrespective of whether mice were young (1 wk) or old (11–14 mo). The same observations were made for systemic infection (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Aire-deficient mice are not susceptible to oral infection of C. albicans. Oral infection with C. albicans 3630 in Aire-deficient mice and WT mice. Data points represent the score (mean ± SEM) for a minimum 10 mice per time point in each group. All experiments were repeated at least once for pups and three times for adults.