Activation of autoreactive lymphocytes in the lung by radioresistant cells expressing a STING gain-of-function mutation

Abstract

Gain-of-function mutations in the dsDNA sensing adaptor STING lead to a severe autoinflammatory syndrome known as STING-associated vasculopathy with onset in infancy (SAVI). Patients with SAVI develop interstitial lung disease (ILD) and produce autoantibodies that are commonly associated with systemic autoimmune diseases. Mice expressing the most common SAVI mutation, STING V154M (VM), similarly develop ILD but exhibit severe T and B cell lymphopenia and low serum Ig titers, and they lack autoantibodies. Importantly, lethally irradiated VM hosts reconstituted with WT stem cells (WT→VM) still develop ILD. In this study, we find that WT→VM chimeras had restored B cell function, produced autoantibodies, and thereby recapitulated the loss of tolerance seen in patients with SAVI. Lymphocytes derived from both WT and BCR or TCR transgenic (Tg) donors accumulated in the extravascular lung tissue of WT+Tg→VM mixed chimeras, but lymphocyte activation and germinal center formation required WT cells with a diverse repertoire. Furthermore, when T cells isolated from the WT→VM chimeras were adoptively transferred to naive Rag1-deficient secondary hosts, they trafficked to the lung and recruited neutrophils. Overall, these findings indicated that VM expression by radioresistant cells promoted the activation of autoreactive B cells and T cells that then differentiated into potentially pathogenic effector subsets.

Keywords: Adaptive immunity; Autoimmune diseases; Autoimmunity; Inflammation; Innate immunity.

Figure 1. B cell intrinsic VM expression impairs antibody production, proliferation, and survival.

(A) Serum titers of IgM, IgG₁ and IgG_2c in age- and sex-matched WT (n = 7) and VM (n = 7–14) mice at 3–4 months of age, or from WT→WT (n = 10–11) and WT→VM (n = 13–15) chimeric mice at 8 weeks after reconstitution. (B–G) Splenic B220⁺ B cells from biologic replicates of WT (n = 6) versus VM (n = 6) B cells and WT→WT (n = 4) and WT→VM (n = 4) B cells labeled with violet proliferative dye (VPD) were stimulated for 72 hours with BLyS alone (NS) or anti≠IgM F(ab’)₂ antibody and BLyS (+αIgM), and they were then stained with TOPRO to assess cell death. Cell division is shown by representative VPD histograms (B) and division index (C and D). Cell death is shown by representative TOPRO histograms (E) and percentage of TOPRO^– live cells (F and G). Nonparametric Mann-Whitney U tests were used for pair-wise comparisons to determine statistical significance (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 2. WT→VM mice form germinal centers in the lung.

Data obtained from 3- to 4-month-old age- and sex-matched WT (n = 3–10) and VM (n = 5–13) mice. Chimeric WT→WT (n = 7–13) and WT→VM (n = 10–14) mice were evaluated 8–9 weeks after reconstitution. (A) Percentage of IgD^–Fas⁺GL7⁺ germinal center (GC) B cells within the CD45⁺CD19⁺ B cell compartment of WT and VM spleen or within the donor-derived B cells in WT→WT and WT→VM chimeric spleens. (B) Percentage of GC B cells within the CD45⁺CD19⁺ lung extravascular (EV) B cell compartment of in WT and VM mice and within the total donor-derived lung EV B cells in WT→WT and WT→VM chimeras. (C) Immunofluorescence imaging of WT→VM chimeric mouse lungs stained for B220 (yellow) and peanut agglutinin (PNA, magenta) to identify B220⁺PNA⁺ GC B cells. Two B cell follicles containing GCs are highlighted with white squares. Magnification (4×) of the highlighted sections are shown in the 2 panels to the right, and a dotted circular outline is used to identify the germinal center. In the middle panel, DAPI, B220, and PNA are shown together, and the right panel shows DAPI and PNA alone. Data in C are representative of 2 biologic replicates. Nonparametric Mann-Whitney U tests were used for pairwise comparisons to determine statistical significance (*P < 0.05, **P < 0.01, ***P < 0.001). Scale bars: 200 μm.

Figure 3. Formation of GC B cells in VM lung requires an unrestricted repertoire.

(A) Six-week-old age- and sex-matched CD45.1/CD45.1 WT and VM littermates were lethally irradiated and reconstituted with a mixture of BM stem cells comprised of 20% CD45.1/CD45.2 WT and 80% CD45.2/CD45.2 MD4 Rag1^–/– cells. WT+MD4→WT (n = 4–8) and WT+MD4→VM (n = 6–9) chimeric mice were then evaluated 8–9 weeks later. (B) Percentage of MD4 donor–derived cells within all CD45⁺ cells in BM, BM B220⁺ B cells, or splenic CD19⁺ B cells, or within the lung extravascular (EV) B cells of either the WT hosts (gray bar) or VM hosts (red bar), after excluding any CD45.1 single-positive radio-resistant host cells. (C) Total number of lung EV B cells derived from either the WT or MD4 donor in either the WT or VM hosts. (D–G) Percentage of IgD⁺ naive B cell, IgD^– Fas⁺ GL7⁺ germinal center (GC) B cells, B220^– CD138⁺ PC cells, and IgD^– Fas⁺ GL7^– B cells derived from either the WT or MD4 donor within the EV B cell compartment of either WT or VM host. (H–K) Percentage of naive B cells, GC B cells, PCs, and IgD^–Fas⁺GL7^– B cells derived from WT or MD4 donor within the splenic B compartment of WT or VM hosts. For C–K, data points in columns 1 and 2 represent paired WT and MD4 donors from a shared WT host. Similarly, data points in columns 3 and 4 represent paired WT and MD4 donors from a shared VM host. To determine statistical significance, nonparametric Mann-Whitney U tests were used for pair-wise comparisons between columns 1 and 3 and between columns 2 and 4. A Wilcoxon matched pairs signed rank test was used for pairwise comparisons between columns 3 and 4 (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 4. T lymphocyte recruitment and activation in VM ILD requires a diverse repertoire.

(A) Six-week-old sex-matched CD45^.1/.1 WT and VM littermates were lethally irradiated and reconstituted with a mixture of BM stem cells comprised of 20% WT CD45^.1/.2, 40% OT-I Rag2^–/– CD45^.2/.2, and 40% OT-II Rag2^–/– CD45^.2/.2 donor BM. WT+OT-I/-II→WT (n = 4–10) and WT+OT-I/-II→VM (n = 4–8) chimeric mice were evaluated 8–9 weeks later. (B) Percentage of OT-I/-II donor-derived cells within the donor-derived CD45⁺ BM compartment, CD45⁺ thymus, CD3⁺TCRb⁺ splenic T cells, and CD3⁺ TCRb⁺ lung extravascular (EV) T cells. (C) Total number of lung EV T cells derived from either the WT or OT-I/-II donors in WT and VM hosts. (D–G) Percentage of CD44^–CD62L⁺ naive T cells and percentage of CD44⁺CD62L^– effector T cells within lung EV T cells, percentage of CD127⁺CD69^– CD4 effector memory cells within CD4⁺ lung EV T cells, and percentage of PD-1⁺CD69⁺ CD8 PD-1 effector cells within CD8⁺ lung EV T cells derived from WT and OT-I/-II donors in WT and VM hosts. (H–K) Percentage of naive and effector T cells in splenic T cells, percentage of CD4 effector memory T cells within CD4⁺ splenic T cells, and percentage of CD8 PD-1 effector cells within CD8⁺ splenic T cells derived from WT and OT-I/-II donors in WT and VM hosts. For C–K, data points in columns 1 and 2 represent paired WT and OT-I/-II donors from a shared WT host. Similarly, data points in columns 3 and 4 represent paired WT and OT-I/-II donors from a shared VM host. To determine statistical significance, nonparametric Mann-Whitney U tests were used for pair-wise comparisons between columns 1 and 3 and between columns 2 and 4. A Wilcoxon matched pairs signed rank test was used for pairwise comparisons between columns 3 and 4. Statistical significance in B was determined using a multiple Mann-Whitney U test with FDR correction (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 5. WT→VM mice produce IgM autoantibodies.

(A and B) Immunofluorescence staining of HEp2 slides with representative sera of the indicated strains or chimeric mice with representative WT (n = 5) and VM (n = 3) mice as well as WT→WT (n = 16) and WT→VM (n = 17) chimeric mice. (A) Autoantibodies were detected with a Dylight 488–conjugated anti-IgM antibody (A) or anti-IgG antibody (B). 7D7 is an IgM anti-DNA antibody, and PL2-3 is an IgG_2a anti-chromatin antibody. Scale bars: 200 μm. (C and D) Sera from WT (n = 4), VM (n = 4), WT→WT (n = 7), WT→VM (n = 9), BALB/c (n = 1), and BALB/c mice treated with pristane (n = 1) were assessed by an autoantigen array as negative and positive reference controls. Data are column normalized to the highest and lowest signal observed for each autoantigen. Autoantigens are grouped according to the disease they are associated with: gastroenteric and endocrine autoimmunity (GI/endocrine), nonspecific inflammation and stress, myositis and overlap syndromes, scleroderma and systemic sclerosis, and SLE/Sjögren’s syndrome. C shows the reactivity of sera IgM, and D shows the reactivity of sera IgG.

Figure 6. WT→VM autoantibodies target lung proteins.

(A) Immunofluorescence staining of Rag1^–/– lung sections with serum samples diluted 1:40. A representative result is shown from WT (n = 3) and VM (n = 2) mice and from WT→WT (n = 10) and WT→VM (n = 10) chimeric mice. Slides were then developed with DAPI and Alexa Fluor 488–conjugated anti-IgG (top row) or anti-IgM (bottom row) antibody. Scale bars: 200 μm. (B and C) Lung reactive IgM and IgG antibodies were detected in sera from WT→WT or WT→VM chimeric mice (diluted 1:500) by Western blot against lysates from Rag1^–/– lungs. Numbers on the left side correspond to the mass in kDa of bands from the protein ladder. Data are representative of a total of n = 9 biologic replicates per group. The lanes in B and C were run on the same gel but were noncontiguous.

Figure 7. WT lymphocytes from WT→VM chimeras induce lung inflammation in Rag1-KO mice.

(A) Six-week-old CD45^.2/.2 WT and VM mice were lethally irradiated and reconstituted with CD45^.1/1 WT donor BM to generate WT→WT and WT→VM chimeric mice. Splenocytes from these chimeric mice were then harvested 8 weeks after BM engraftment and adoptively transferred into secondary recipient 8-week-old Rag1^–/– recipients to generate (WT→WT)→Rag1^–/– (n = 9) and (WT→VM)→Rag1^–/– (n = 11) mice and then evaluated 8–9 weeks later. (B) Representative H&E lung histology from (WT→WT)→Rag1^–/– and (WT→VM)→Rag1^–/– mice captured with a 4× objective. A region of interest, as indicated by dotted red box, was further imaged with a 10× objective and shown 5× magnified over the original image. Data in B are representative of n = 9 (WT→VM)→Rag1^–/– mice and n = 11 (WT→WT)→Rag1^–/– mice. (C) Lungs from (WT→WT)→Rag1^–/– and (WT→VM)→Rag1^–/– mice, stained for DAPI (gray), CD3 (yellow), B220 (cyan), and LYVE-1 (magenta). Data in C are representative of n = 3 (WT→VM)→Rag1^–/– mice and n = 2 (WT→WT)→Rag1^–/– mice. (D) Total number of EV T and B cells in the lungs of (WT→WT)→Rag1^–/– and (WT→VM)→Rag1^–/– mice. (E) Percentage of donor-derived cells in the CD11b⁺ or CD11c⁺ lung EV myeloid compartment. (F) Percentage of CD11b⁺Ly6G⁺ neutrophils from the host and donor lung myeloid cell compartment. (G) Number of neutrophils from host lung EV myeloid compartment. Nonparametric Mann-Whitney U tests were used for pair-wise comparisons to determine statistical significance (*P < 0.05, **P < 0.01). Scale bar: 200 μm.