Intrinsic self-DNA triggers inflammatory disease dependent on STING

Abstract

Inflammatory diseases such as Aicardi-Goutières syndrome and severe systemic lupus erythematosus are generally lethal disorders that have been traced to defects in the exonuclease TREX1 (DNase III). Mice lacking TREX1 similarly die at an early age through comparable symptoms, including inflammatory myocarditis, through chronic activation of the stimulator of IFN genes (STING) pathway. In this study, we demonstrate that phagocytes rather than myocytes are predominantly responsible for causing inflammation, an outcome that could be alleviated following adoptive transfer of normal bone marrow into TREX1(-/-) mice. TREX1(-/-) macrophages did not exhibit significant augmented ability to produce proinflammatory cytokines compared with normal macrophages following exposure to STING-dependent activators, but rather appeared chronically stimulated by genomic DNA. These results shed molecular insight into inflammation and provide concepts for the design of new therapies.

FIGURE 1

Infiltrating immune-related cells cause STING-dependent proinflammatory gene induction in TKO mice. (A) Kaplan–Meier survival curve of TKO mice (n = 43) or STKO (n = 53) mice. Immunoblot of STING and TREX1 in heart from WT, TKO, SKO, and STKO mice. β-Tubulin was used as loading control. (B) Macroscopic view of heart of each genotype mice at the age of 10 wk. (C) Gene array analysis of hearts isolated from WT, TKO, STKO, and SKO mice at 10 wk age of mice. Total RNA was purified and transcripts analyzed by Illumina Sentrix BeadChip array (mouse WG6 version 2). The data discussed in this publication have been deposited in the National Center for Biotechnology Information’s Gene Expression Omnibus and are accessible through Gene Expression Omnibus series accession no. GSE59219 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE59219). (D) Quantitative RT-PCR (qPCR) analysis of TNF, IL-1β, IL-6, and IFN-β from heart as in (C). Data are shown as the means of four mice. (E) Immunohistochemical detection of STING, CD68, and CD11C in the hearts. Original magnification ×400. (F) Multiplex Luminex assay for IL-1β, TNF, CXCL10, and MCP-3 in the sera. Data are shown as the means of five mice. (G) ANAs in the sera from WT, TKO, STKO, and SKO mice. Original magnification ×1260. (H) qPCR analysis of IFN-β, IL-1β, and IL-6 in unperfused and perfused heart from 10-wk-old WT and TKO mice. Data are shown as the means of two mice. Error bars indicated SD. Statistical analysis was performed using a Student t test. *p ≤ 0.05, **p ≤ 0.005.

FIGURE 2

TREX1-deficient macrophages and DCs generate high background levels of proinflammatory cytokines. (A) ELISA of IFN-β in BMDMs and BMDCs treated with 4 µg/ml dsDNA90 (ISD) and cGAMP and infected with HSV-γ34.5 (multiplicity of infection of 1). Results are from one representative of at least two independent experiments. Lipo, lipofectamin only; UI, uninfected. (B) qPCR of IFIT3 and CXCL10 in macrophages isolated from WT, TKO, STKO, and SKO mice peritoneal cavities. (C) qPCR of IFIT3 and CXCL10 in PBMCs. (D) Gene array analysis by Illumina Sentrix Bead-Chip array (mouse WG6 version 2) from total RNA purified in WT, TKO, and STKO BMDMs treated with or without dsDNA90. The data discussed in this publication have been deposited in National Center for Biotechnology Information’s Gene Expression Omnibus and are accessible through Gene Expression Omnibus series accession no. GSE59219 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE59219). (E) Immunoblots show genotypes in BMDMs, BMDCs, and PBMCs using anti-STING and anti-TREX1. qPCR is shown of IFIT3 and CXCL10 mRNA level in BMDMs (F) and BMDC (G). All data are the means of duplicates for at least two mice. Error bars indicated SD. Statistical analysis was performed using a Student t test. **p ≤ 0.005.

FIGURE 3

Elimination of TREX1-deficient hematopoietic cells alleviates inflammatory cytokine production in TKO mice. Survival curves (A and B) and body weight (C and D) of TKO or WT mice transplanted with WT, TKO, or STKO bone marrow. STKO BM > TKO, TKO mice reconstituted with STKO bone marrow (n = 4); TKO BM > TKO, TKO mice reconstituted with TKO bone marrow (n = 4); TKO BM > WT, WT mice reconstituted with TKO bone marrow (n = 7); WT BM > TKO, TKO mice reconstituted with WT bone marrow (n = 4); WT BM > WT, WT mice reconstituted with WT bone marrow (n = 5). Results are shown in Kaplan–Meier curve with a mean of at least four mice from three independent experiments. Error bars indicated SEM. Statistical analysis was performed using a Student t test. x-Axes indicate weeks after bone marrow transplant. *p ≤ 0.05. (E) Representative H&E staining (HE) of heart tissue (left, original magnification ×400) and ANA assay with the sera (right, original magnification ×1260). (F) Inflammation score (left) and semiquantitation of fluorescent intensity (right) from (E). (G) Multiplex Luminex assay for IL-6, TNF, and MCP-3 in the sera from WT BM > WT (n = 4) and TKO BM > WT (n = 2) mice. (H) qPCR analysis of IFIT3 and mRNA in PBMCs from the mice same as (A). Error bars indicated SD. Statistical analysis was performed using a Student t test. *p ≤ 0.05, **p ≤ 0.005.

FIGURE 4

DNA species left over from cellular replication activates STING in Trex1 deficiency. (A) Cell cycle analysis in BMDCs isolated from WT and TKO mice at 0, 8, and 24 h after release from thymidine/nocodazole block for synchronization. (B) qPCR analysis of IFIT3 and IFN-β in BMDCs isolated from WT and TKO mice at 0, 8, and 24 h after release from thymidine/nocodazole block for synchronization. Error bars indicated SD. Statistical analysis was performed using a Student t test. *p ≤ 0.05. (C) FISH of all mouse centromere in BMDMs isolated from WT and TKO mice at 0 h after release from synchronization. DAPI was used for counterstaining. The cells were examined under Leica SP5 spectral confocal inverted microscope. Original magnification ×1260. Data are representative of at least two independent experiments.

FIGURE 5

ICAM-1 and CXCR3 in endothelial cells of heart attract inflammatory cells in TKO mice. (A) Fold changes of chemokines in TKO and STKO compared with WT macrophages from gene array analysis in Fig. 2D and fold changes of chemokine receptors and adhesion molecules in TKO and STKO heart in comparison with WT heart from gene array analysis in Fig. 1C. qPCR of ICAM-1 (B) and CXCR3 (C) in unperfused and perfused heart (p-WT and p-TKO) from 10-wk-old WT and TKO mice same as Fig. 1H. Error bars indicated SD. Statistical analysis was performed using a Student t test. *p ≤ 0.05, **p ≤ 0.005. (D) Immunohistochemical detection of ICAM1 and CXCR3 in the hearts of WT and TKO mice. Cell percentages were blind reviewed by a pathologist. Original magnification ×400.