Dual functions of Aire CARD multimerization in the transcriptional regulation of T cell tolerance

Abstract

Aggregate-like biomolecular assemblies are emerging as new conformational states with functionality. Aire, a transcription factor essential for central T cell tolerance, forms large aggregate-like assemblies visualized as nuclear foci. Here we demonstrate that Aire utilizes its caspase activation recruitment domain (CARD) to form filamentous homo-multimers in vitro, and this assembly mediates foci formation and transcriptional activity. However, CARD-mediated multimerization also makes Aire susceptible to interaction with promyelocytic leukemia protein (PML) bodies, sites of many nuclear processes including protein quality control of nuclear aggregates. Several loss-of-function Aire mutants, including those causing autoimmune polyendocrine syndrome type-1, form foci with increased PML body association. Directing Aire to PML bodies impairs the transcriptional activity of Aire, while dispersing PML bodies with a viral antagonist restores this activity. Our study thus reveals a new regulatory role of PML bodies in Aire function, and highlights the interplay between nuclear aggregate-like assemblies and PML-mediated protein quality control.

Fig. 1. Filament assembly of Aire CARD mediates nuclear foci formation and transcriptional activity.

a Schematic of murine Aire domain architecture with previously characterized functions of individual domains. b Representative EM image of wild-type (WT) mAire CARD. While fully formed CARD filaments (i) were major species, thin protofilaments (ii) were also observed. Right, representative images of the two types of filaments. 2D class averages were also shown for fully formed CARD filaments. c, d Representative absorption spectra of Congo Red (CR, 15 μM) (c) or fluorescence emission spectra (excitation at 430 nm) of Thioflavin (ThT, 50 μM) (d) in the presence of WT mAire CARD filaments or known amyloids of RIP1/3-RHIM. 5 μM monomeric concentration was used for both proteins. The experiments were performed two independent times. e 3D homology model of mAire CARD using the program FUGUE 2.01. Putative CARD:CARD interfaces were identified from the sequence alignment with other CARDs (Supplementary Fig. 1a) and are highlighted green. f Representative EM images of recombinant mAire CARD. WT protein and those with mutations in the putative CARD:CARD interfaces in (e) were compared. g Transcriptional activity of WT mAire and mutants, as measured by the relative mRNA levels of Aire-dependent genes (represented by CD4, IGFL1, KRT14 and S100A9). An Aire-independent gene, HPRT1, was also examined as a negative control. Aire variants were transiently expressed (using 1.25 μg/ml DNA) in 293 T cells for 48 h prior to RT-qPCR analysis. Data are representative of at least three independent experiments and presented as mean ± s.d., n = 3. Right, western blot (WB) showing the expression levels of FLAG-tagged mAire compared to endogenous levels of Histone H3 (anti-H3). See Supplementary Fig. 1c for experiments in 4D6. P-values (two-tailed t-test) were calculated in comparison to WT mAire. *p < 0.05; **p < 0.01; p > 0.05 is not significant (ns). Exact p-values are provided in the Source Data File. h Representative fluorescence images of FLAG-tagged mAire in 293 T cells using anti-FLAG. See Supplementary Fig. 1d for experiments in 4D6.

Fig. 2. Chemically induced multimerization partially restores the transcriptional activity of AireΔCARD.

a Schematic of mAire∆CARD fused with four tandem repeats of FKBP (FKBP4-ΔCARD). Proteins with FKBP repeats are known to multimerize upon addition of chemical dimerizer AP1903. b Transcriptional activity of ∆CARD fused with 1–4 tandem repeats of FKBP (FKBP1–4) in the presence and absence of AP1903. Fusion constructs were transiently expressed (using 1.25 μg/ml DNA) in 293 T cells and DMSO or AP1903 (5 μM) was added 24 h later. Cells were harvested 48 h after transfection, followed by RT-qPCR of the respective Aire-dependent genes (represented by CD4 and IGFL1). The relative mRNA level of an Aire-independent gene, HPRT1, was also shown as a negative control. Data are representative of at least three independent experiments and presented as mean ± s.d., n = 3. See Supplementary Fig. 2a, b for additional target genes and western blot (WB) showing protein expression levels. P-values (two-tailed t-test) were calculated in comparison to empty vector (EV) + AP1903. *p < 0.05; **p < 0.01; p > 0.05 is not significant (ns). Exact p-values are provided in the Source Data File. c, d Representative fluorescence microscopy images of FLAG-tagged ∆CARD fusion variants in 4D6. Note that 4D6 cells were used as these cells are flatter than 293 T, allowing more robust analysis of Aire nuclear localization. Fusion constructs were transiently expressed and treated with DMSO or AP1903 as in (b). 48 h after transfection, cells were immunostained with anti-FLAG (mAire) and anti-PML. See Supplementary Fig. 2c, d for endogenous PML immunostaining.

Fig. 3. Aire mutants with increased PML association are impaired in transcriptional activity and are hyper-SUMOylated.

a Transcriptional activity of mAire and Sp110-CARD swap, where Aire CARD was swapped with Sp110 CARD. Experiments were performed as in Fig. 1g and presented as mean ± s.d., n = 3. Bottom right, WB of mAire and histone H3. b Representative fluorescence microscopy images of FLAG-tagged WT, Sp110-CARD swap and K53A/E54A mAire variants in 4D6 cells. Note that 4D6 cells were used as these cells show more distinct PML bodies and are flatter than 293 T, allowing more robust analysis of Aire foci and their endogenous PML body localization. Cells were immunostained with anti-FLAG (mAire) and anti-PML. Right, quantification of the Aire foci colocalized with endogenous PML bodies from automated image analysis (see Supplementary Fig. 3a, b for the definition of colocalization). n = number of Aire foci examined per sample. Statistical significance comparisons were calculated using a two-tailed Student’s t-test for two population proportions where each population consists of all individual Aire foci examined. ***p = 5.57e-26 and 6.46e-15 for Sp110-CARD swap and K53A/E54A, respectively. See also Supplementary Fig. 3c for FLAG-tagged hAire stably expressed in 4D6 cells. c SUMO modification analysis of WT mAire and K53A/E54A. FLAG-tagged mAire was co-expressed with HA-SUMO1 or -SUMO2 in 293 T cells. Cells were treated with MG132 (10 μM) for 24 h before harvesting. mAire proteins were immunoprecipitated (IPed) using anti-FLAG beads under semi-denaturing condition and analyzed by anti-HA WB. d Schematic of chromatin fractionation analysis of Aire. 293 T cells were transfected with mAire expressing plasmids for 48 h before harvesting. Solubility of Aire and chromatin (as measured by Histone H3) before and after MNase treatment was analyzed by WB. e Chromatin fractionation analysis of mAire WT and K53A/E54A. Experiments were performed as in (d).

Fig. 4. Directing Aire to PML bodies inhibits transcriptional activity, while dispersing PML bodies increases Aire activity.

a Representative fluorescence microscopy images of FLAG-tagged WT mAire and mAire N-terminally fused with SUMO interaction motifs (SIMs) from RNF4 (SIM-mAire). Right, quantification of Aire foci colocalized with endogenous PML bodies. n = number of Aire foci examined per sample. ***p (two-tailed Student’s t-test) = 1.68e-71. b Transcriptional activity of mAire and SIM-mAire. Experiments are presented as mean ± s.d., n = 3. P-values (two-tailed t-test) were calculated in comparison to WT mAire where *p = 0.031; **p = 0.0035 and 0.0014 for KRT14 and S100A9 respectively; ***p = 0.0004; p = 0.68 is not significant (ns). c Representative fluorescence microscopy images of WT mAire-FLAG with and without co-expression with SIM-mAire (no tag) in 4D6 cells. Right, quantification of the Aire foci colocalized with endogenous PML bodies as in (a). ***p (two-tailed Student’s t-test) = 4.39e-7. d Transcriptional activity of mAire (black circle) with and without co-expression of SIM-mAire (green circle) in 293 T cells. Each circle represents 0.6 μg/ml DNA transfected. Experiments are presented as mean ± s.d., n = 3. See Supplementary Fig. 4a for additional Aire-dependent target genes. P-values (two-tailed t-test) were calculated in comparison to WT mAire where **p = 0.002; p = 0.88 is not significant (ns). e Representative fluorescence microscopy images of FLAG-tagged mAire co-expressed with CMV IE1 (0.5 μg/ml DNA each construct) in 4D6 cells. Cells were immunostained with anti-FLAG and anti-PML. See Supplementary Fig. 4c for the impact of IE1 on PML body dispersion. f Transcriptional activity of mAire, K53A/E54A and SIM-mAire (0.5 μg/ml DNA) with an increasing concentration of IE1 (0, 0.17, 0.5 μg/ml) in 293 T cells. Values are normalized against WT mAire without IE1. Experiments are presented as mean ± s.d., n = 3. Fold-changes in transcriptional activities upon addition of IE1 are indicated. P-values (two-tailed t-test) were calculated in comparison to no IE1. *p < 0.05; **p < 0.01; p > 0.05 is not significant (ns). Exact p-values are provided in the Source Data File.

Fig. 5. Isolated Aire CARD multimers associate with PML bodies.

a Representative fluorescence microscopy images (3 independent experiments) of mAire CARD fused with monomeric GFP (CARD-mGFP) transiently expressed in 4D6 cells. Note that mAire residues 1–174 containing both Aire CARD and nuclear localization signal (NLS) were used in the fusion construct. GFP fluorescence and anti-PML staining were used for imaging CARD-mGFP and PML bodies, respectively. b Chromatin fractionation analysis of NLS-mGFP (mAire aa 105–174 fused with mGFP) and CARD-mGFP. Experiments were performed as in Fig. 3d. c Representative fluorescence microscopy images (two independent experiments of CARD-mGFP and HA-tagged mAire upon their co-expression in 4D6 cells. GFP fluorescence was used for imaging CARD-mGFP and immunostained with anti-HA and anti-PML. d Transcriptional activity of mAire (black circle) and its changes upon co-expression with CARD-mGFP (red circle) in 293 T cells. Each circle represents 0.6 μg/ml DNA transfected. Experiments were performed as in Fig. 1g and presented as mean ± s.d., n = 3. P-values (two-tailed t-test) were calculated in comparison to WT mAire. *p < 0.05; p > 0.05 is not significant (ns). Exact p-values are provided in the Source Data File.

Fig. 6. APS-1 mutations C302Y and C311Y increase PML association.

a 3D structure of Aire PHD1 domain bound to histone H3-derived peptide (PDB ID: 2KFT [10.2210/pdb2KFT/pdb]). C302 and C311, which coordinate Zn²⁺ ions, are mutated in APS-1 patients (highlighted in orange). b Transcriptional activity of hAire WT, C302Y and C311Y. Experiments were performed as in Fig. 1g and are presented as mean ± s.d., n = 3. c Representative fluorescence microscopy images of hAire C302Y and C311Y variants in 4D6 cells. Right, quantitation of Aire foci colocalized with PML bodies. n = number of Aire foci examined per sample. Statistical significance comparisons were calculated using a two-tailed Student’s t-test for two population proportions where each population consists of all individual Aire foci examined. ***p = 2.6e-18 (C302Y) and 0.0004 (C311Y). d SUMOylation analysis of hAire WT, C302Y (top) and C311Y (bottom). Experiments were performed as in Fig. 3c. e Chromatin fractionation analysis of hAire WT, C302Y and C311Y. Experiments were performed as in Fig. 3d.

Fig. 7. PML colocalization explains the dominant negative effect of C302Y and C311Y.

a Transcriptional activity of hAire (black circle) and its changes upon co-expression with C302Y (orange circle) and C311Y (blue circle) in 293 T cells. Each circle represents 0.6 μg/ml transfected DNA. Data are presented as mean ± s.d., n = 3. b Representative fluorescence microscopy images of hAire WT-FLAG co-expressed with hAire C302Y-HA (0.6 μg/ml DNA each) in 4D6 cells. Cells were immunostained with anti-FLAG (WT hAire) and anti-PML. Right, quantitation of Aire foci colocalized with PML bodies. n = number of Aire foci examined per sample. Statistical significance comparison was calculated using a two-tailed Student’s t-test for two population proportions where each population consists of all individual Aire foci examined. ***p = 1.24e-9 (WT + C302Y) and 5.45e-05 (WT + C311Y). c Representative fluorescence microscopy images of hAire WT-FLAG co-expressed with hAire C302Y-HA (0.3 μg/ml DNA each) in the presence or absence of IE1-HA (0.5 μg/ml) in 4D6 cells. Cells were immunostained with anti-FLAG (WT hAire) and anti-PML. d Transcriptional activity of hAire WT (0.3 μg/ml) co-expressed with empty vector (0.3 μg/ml left panel) or hAire C302Y (0.3 μg/ml, right panel) and the impact of IE1 co-expression (0, 0.17, 0.5 μg/ml). Data are presented as mean ± s.d., n = 3. Fold-changes in transcriptional activities with IE1 are indicated. P-values (two-tailed t-test) were calculated in comparison to WT mAire. **p < 0.01; ***p < 0.001. Exact p-values are in the Source Data File. Right, western blot showing the expression levels of FLAG-tagged hAire variants co-expressed with or without IE1. e Model for PML body-mediated regulation of Aire function. Proper Aire function requires large homo-multimerization, but this property inevitably subjects Aire to PML body-associated protein quality control and transcriptional regulation. Note that PML body localization does not require protein mis-folding of Aire, and that the locations of Aire foci are not pre-defined. Instead, PML body localization can be induced by multiple factors, including altered multimerization property (through mutations in CARD), loss of chromatin interaction (through mutations in PHD1) or by interaction with PML colocalizing alleles.