Somatic gene mutations expose cytoplasmic DNA to co-opt the cGAS/STING/NLRP3 axis in myelodysplastic syndromes

Abstract

NLRP3 inflammasome and IFN-stimulated gene (ISG) induction are key biological drivers of ineffective hematopoiesis and inflammation in myelodysplastic syndromes (MDSs). Gene mutations involving mRNA splicing and epigenetic regulatory pathways induce inflammasome activation and myeloid lineage skewing in MDSs through undefined mechanisms. Using immortalized murine hematopoietic stem and progenitor cells harboring these somatic gene mutations and primary MDS BM specimens, we showed accumulation of unresolved R-loops and micronuclei with concurrent activation of the cytosolic sensor cyclic GMP-AMP synthase. Cyclic GMP-AMP synthase/stimulator of IFN genes (cGAS/STING) signaling caused ISG induction, NLRP3 inflammasome activation, and maturation of the effector protease caspase-1. Deregulation of RNA polymerase III drove cytosolic R-loop generation, which upon inhibition, extinguished ISG and inflammasome response. Mechanistically, caspase-1 degraded the master erythroid transcription factor, GATA binding protein 1, provoking anemia and myeloid lineage bias that was reversed by cGAS inhibition in vitro and in Tet2-/- hematopoietic stem and progenitor cell-transplanted mice. Together, these data identified a mechanism by which functionally distinct mutations converged upon the cGAS/STING/NLRP3 axis in MDS, directing ISG induction, pyroptosis, and myeloid lineage skewing.

Keywords: Hematology; Leukemias; Oncology.

Figure 1. Cytosolic DNA is increased in MDS bone marrow and murine SGM models.

(A) Released caspase-1 was significantly increased in SGM mutant cell lines compared with WT controls; assessed by Caspase-Glo 1 Inflammasome Assay (n = 3 each). (B) Percentage of cells with micronuclei was significantly increased in SGM cell lines compared with WT controls (n = 3 each), as well as in MDS patient BM-MNCs (n = 8) compared with BM donors (n = 6) (C). (D) Representative immunofluorescence images of micronuclei in MDS (n = 8) compared with donor BM-MNCs (n = 6). (E) R-loop expression measured by S9.6 flow cytometry was significantly increased in SGM cell lines compared with WT controls (n = 3 each) as well as in MDS BM-MNCs (n = 6) compared with controls (n = 3): representative histograms. (F) Quantification of immunofluorescence images (G) of R-loops (shown in green) in MDS (n = 6) compared with donor BM-MNCs (n = 3); scale bars: 10 μm. Data are presented as mean ± SEM, Student’s t test, *P ≤ 0.05, **P ≤ 0.01, ****P ≤ 0.0001.

Figure 2. cGAS is activated in SGM models and MDS bone marrow cells.

(A) Representative immunofluorescence images of SGM mutant cell lines and MDS BM-MNCs with colocalization of cGAS to micronuclei with compromised nuclear membranes. Scale bar: 8 μm. (B) Representative immunofluorescence images of colocalization of cGAS and R-loops in SGM mutant cell lines and MDS BM-MNCs. Scale bar: 10 μm. (C) Aggregation of cGAS foci was increased in mutant immortalized SGM cells compared with WT controls (cGAS in green, DAPI in blue; original magnification, 3780×). Scale bar: 10 μm. (D) Aggregation of cGAS in MDS BM-MNCs (original magnification, 630×) compared with donor BM-MNCs (original magnification, 2940×). Scale bar: 50 μm. (E) Phospho-IRF3 was increased in SGM models compared with WT controls; representative flow histograms (n = 3 each). (F) ISG expression was increased in the immortalized SGM cells compared with WT controls (n = 3 each). (G) ISG and IFNB1 expression was significantly increased in MDS BM-MNCs (n = 213) compared with age-matched donors (n = 20). Data are presented as mean ± SEM. Images are representative of at least 2 independent experiments; Student’s t test for in vitro data; Mann-Whitney U test for G; *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 3. cGAS activation licenses NLRP3 inflammasome activation.

(A) Treatment with 1 μM of RU.521 for 24 hours decreased Ccl5 expression nearly to WT levels in both SGM cell lines (n = 3 each). (B) Phospho-IRF3 assessed by flow cytometry with representative histograms and ISG expression (n = 3). (C) IFNB1, CXCL10, and SAM9DL expression decreased in low-risk MDS BM-MNCs treated in vitro with 1 μM RU.521 for 48 hours (n = 3). (D) Cleaved caspase-1 decreased in SGM models with RU.521 treatment. (E) cGAS Western blotting of cells treated with scrambled sgRNA (WT and Tet2^–/– immortalized cells) or cGAS sg_3 or sg_4 (Tet2^–/–). (F) CRISPR knockout of cGAS in the Tet2 SGM cell line decreased Ccl5 expression (n = 4) and nuclear IRF3 (G) compared with scrambled control. (H) Active caspase-1 assessed by flow cytometry was decreased with cGAS knockdown by CRISPR (n = 3). Data are represented as mean ± SEM. Western blots are representative of at least 2 independent experiments. Student’s t test; *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 4. cGAS attenuation inhibits inflammasome activity in primary MDS cells.

(A) Low-risk MDS BM-MNCs treated in vitro with 1 μM RU.521 for 48 hours (n = 3) had decreased active caspase-1, IL-1β, and intracellular ASC specks measured by imaging flow cytometry. (B) Representative image of cell with ASC speck. (C) Decreased cGAS expression in GFP⁺ primary BM cells with representative flow histogram. (D) Decreased caspase-1 activity in 3 MDS BM-MNCs treated with cGAS shRNA with representative flow histogram. (E) Average reduction in caspase-1 level of 3 pooled primary MDS specimens treated with shRNA. Data are represented as mean ± SEM. Student’s t test; *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 5. RNAP3 inhibition suppresses R-loop accumulation and inflammasome activity.

(A) SGM models were treated with ML-60218 (RNAP3 inhibitor, RNAP3i) for 72 hours, and R-loop (S9.6) expression was assessed by flow cytometry (n = 2 each for Tet2 cells and n = 3 each for Srsf2 cells) with representative flow histograms. (B) Representative IF images of R-loop reduction in Tet2^–/– immortalized cells treated with RNAP3i (scale bar: 10 μm) and (C) MFI quantitation of immunofluorescence (n = 10 vehicle-treated cells and n = 12 RNAP3i-treated cells). (D) Caspase-1 activity by flow cytometry with representative histograms (n = 3 each). (E) Low-risk MDS BM-MNCs treated in vitro with 10 μM RNAP3i for 72 hours (n = 3), resulting in a decrease in R-loops with representative histogram, ISGs (n = 3 each) (F), and inflammasome markers (caspase-1, IL-1β, and ASC specks (n = 3 each) (G), with representative flow histograms. Data are presented as mean ± SEM. Student’s t test; *P ≤ 0.05, **P ≤ 0.01, ****P ≤ 0.0001.

Figure 6. Treatment with cGAS inhibitor restores erythroid differentiation.

(A) Treatment with the cGAS inhibitor increased GATA1 transcription (n = 4) in MDS BM-MNCs, differentiation evidenced by Wright-Giemsa staining (n = 2) (B), and CD71 expression (n = 3); representative flow histogram (C) in MDS BM-MNCs. The original image was taken with a 40× objective on EVOS FL Auto microscope. (D) BM-MNC ASC specks decreased in Tet2^–/– CD45.2 BM-MNCs treated in vivo for 6 weeks with RU.521 compared with vehicle (n = 3, each); (E) increased erythroid islands in the bone marrow from mice treated with RU.521compared with mice treated with vehicle; and (F) increased hemoglobin, hematocrit, and lymphocytes and decreased monocytes in treated mice (n = 8) compared with vehicle (n = 9). Data are presented as mean ± SEM. Student’s t test; *P ≤ 0.05.

Figure 7. Autologous cytoplasmic DNA activates the cytoplasmic sensor, cGAS, directing inflammasome activation and differentiation defects in MDS.

Schematic summarizing cGAS activation as a result of somatic gene mutations that cause accumulation of autologous cytoplasmic DNA and activate cGAS; subsequent inflammasome activity leads to proinflammatory cytokine elaboration, pyroptotic cell death, and HSPC differentiation defects. IFM, inflammasome.