Research ArticleGraft-Versus-Host Disease

RIG-I/MAVS and STING signaling promote gut integrity during irradiation- and immune-mediated tissue injury

+ See all authors and affiliations

Science Translational Medicine  19 Apr 2017:
Vol. 9, Issue 386, eaag2513
DOI: 10.1126/scitranslmed.aag2513
  • Fig. 1. Endogenous RIG-I/MAVS signaling reduces intestinal tissue damage in mice.

    (A) Representative images of tissue damage in hematoxylin and eosin–stained small intestine biopsies from mice after an 11-Gy TBI conditioning regimen. White asterisks, villus stunting; black arrowhead, crypt apoptosis; black arrows, granulocyte infiltration. (B) Histopathological score from (A); pooled data from two independent experiments. (C) Number of leukocytes infiltrating the mouse gut lamina propria after TBI (11 Gy) analyzed by flow cytometry. Pooled data from four independent experiments. Survival (D) and weight loss (E) in mouse allo-HSCT recipients transplanted with 5 × 106 BM ± 2 × 106 T cells (donor BALB/c WT and C57BL/6 MAVS+/+ or MAVS−/− recipients). Pooled data from four independent experiments. (F) FITC-dextran concentrations in plasma after allo-HSCT [described in (D) and (E)]. Pooled data from two independent experiments. (G) Survival of Rig-I−/+ and Rig-I−/− mouse recipients after transplant with 5 × 106 BM and 1 × 106 T cells (donor C57BL/6 WT, recipient 129/sv Rig-I−/+ or Rig-I−/−). Animal numbers per group (n) are depicted in the figure panels. Survival was analyzed using the log-rank test. All other experiments were analyzed using two-tailed unpaired t test. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM. n.s., not significant.

  • Fig. 2. MAVS signaling in nonhematopoietic cells maintains intestinal barrier function.

    (A) Average relative abundance of bacterial genera in the intestinal microbiota of cohoused Mavs+/+ (n = 5) and Mavs−/− (n = 5) littermates. One representative experiment of two independent experiments. (B) Survival of C57BL/6 BM chimeric mice, which were Mavs-deficient either in the hematopoietic or in the nonhematopoietic compartment. These mice were analyzed after a second allo-HSCT with BM and T cells from B10.BR WT donor mice. (C) GVHD histopathological score for small intestine biopsies from C57BL/6 BM chimeric mice, which were Mavs-deficient either in the hematopoietic or in the nonhematopoietic compartment. These mice were analyzed after a second allo-HSCT with BM and T cells from BALB/c WT donor mice. Pooled data from two independent experiments. (D) Quantitative polymerase chain reaction (qPCR) of Itgb6 and RegIIIγ expression in the small intestine after allo-HSCT with BM and T cells from BALB/c WT mouse donors into C57BL/6 MAVS+/+ or MAVS−/− recipients. Pooled data from five independent experiments. Animal numbers per group (n) are depicted above panels. Survival was analyzed using the log-rank test. Other experiments were analyzed using ordinary one-way analysis of variance (ANOVA) for multiple comparisons or two-tailed unpaired t test. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM.

  • Fig. 3. RIG-I/MAVS pathway activation protects from intestinal tissue damage after TBI.

    Survival (A) and weight loss (B) after allo-HSCT involving transplant of 5 × 106 BM ± 1 × 106 T cells from C57BL/6 WT donor mice into BALB/c WT recipients with or without 3pRNA treatment on day −1. Pooled data from four independent experiments. (C) Histopathological score of small intestine tissue after allo-HSCT from C57BL/6 WT donor mice to BALB/c WT mouse recipients with or without 3pRNA treatment on day −1. Pooled data from two independent experiments. (D) Weight loss in mouse recipients from (C) after allo-HSCT with or without 3pRNA treatment on day 0 (d0) or day +1 (d+1). Pooled data from three independent experiments. (E) FITC-dextran concentrations in plasma after allo-HSCT from C57BL/6 WT donor mice to BALB/c WT recipient mice with or without 3pRNA treatment on day −1. (F) qPCR of RegIIIγ expression in mouse gut epithelial cells after allo-HSCT from C57BL/6 WT donor mice into BALB/c WT recipient mice with or without 3pRNA treatment on day −1. Pooled data from three independent experiments. (G) Bacterial colony-forming units (CFU) in sera from mouse recipients after allo-HSCT [described in (F)]. Pooled data from three independent experiments. (H) Leukocytes in the small intestine lamina propria of BALB/c mice after TBI (9 Gy) analyzed by flow cytometry. Pooled data from two independent experiments. (I) Leukocytes in the small intestine lamina propria of C57BL/6 mice after treatment with chemotherapy (doxorubicin) analyzed by flow cytometry. Pooled data from three independent experiments. (J) Weight loss in C57BL/6 mice that received doxorubicin (20 mg/kg). Pooled data from six independent experiments. (K) FITC-dextran concentrations in plasma from C57BL/6 mice after doxorubicin treatment (20 mg/kg). One representative experiment of four independent experiments is shown. Animal numbers per group (n) are depicted above figure panels. 3pRNA treatment was always performed on day −1 in indicated groups except for (D). All experiments were analyzed using one-tailed (G) or two-tailed unpaired t test or ordinary one-way ANOVA for multiple comparisons. Survival was analyzed using the log-rank test. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM.

  • Fig. 4. RIG-I–induced type I IFN signaling mediates intestinal tissue protection.

    (A) Survival and percent of initial weight after allo-HSCT. BM cells (5 × 106) plus T cells (1 × 106) from C57BL/6 WT mice were transplanted into BALB/c recipient mice with or without 3pRNA treatment on day −1 before allo-HSCT and treatment with IFNAR1-blocking antibody (α-IFNAR1) or immunoglobulin G1 (IgG1) isotype control on day −2 before allo-HSCT. Pooled data from three independent experiments. (B) Weight loss after allo-HSCT. BM cells (5 × 106) plus T cells (1 × 106) from C57BL/6 WT mice were transplanted into BALB/c recipient mice with or without 3pRNA treatment on day −1 before allo-HSCT and treatment with IFNAR1-blocking antibody (α-IFNAR1) 48 hours before allo-HSCT (−48 h before TX), at the time of allo-HSCT (0 h before TX), or 24 hours after allo-HSCT (24 h after TX). (C) FITC-dextran concentrations in plasma of BALB/c mouse recipients after allo-HSCT, with or without 3pRNA treatment on day −1 before allo-HSCT, and treatment with IFNAR1-blocking antibody (α-IFNAR1) or IgG1 isotype control (IC) on day −2 before allo-HSCT. Pooled data from three independent experiments. (D and E) qPCR of RegIIIγ and Itgb6 expression in small intestine biopsies from BALB/c recipient mice after allo-HSCT, with or without 3pRNA treatment on day −1 before allo-HSCT, and treatment with IFNAR1-blocking antibody (α-IFNAR1) or IgG1 isotype control on day −2 before allo-HSCT. Pooled data from three (D) and six (E) independent experiments. (F) Flow cytometry analysis of leukocyte infiltration into the small intestine lamina propria of BALB/c mice after TBI (9 Gy), with or without 3pRNA treatment on day −1 before TBI, and treatment with IFNAR1-blocking antibody (α-IFNAR1) or IgG1 isotype control on day −2 before TBI. Pooled data from two independent experiments. (G) Survival and weight loss in BALB/c mouse recipients lacking IL-22 (Il-22−/−) after allo-HSCT with BM and T cells from C57BL/6 WT mice with or without 3pRNA treatment on day −1 before allo-HSCT. Pooled data from two independent experiments. Animal numbers per group (n) are depicted above figure panels. Treatment time points for 3pRNA and α-IFNAR1 antibody are indicated. All experiments were analyzed using two-tailed unpaired t test or ordinary one-way ANOVA for multiple comparisons. Survival was analyzed using the log-rank test. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM.

  • Fig. 5. RIG-I–induced type I IFN signaling in nonhematopoietic cells promotes regeneration of the ISC compartment.

    (A) Weight loss after allo-HSCT. BM cells (5 × 106) plus T cells (5 × 106) from BALB/c donor mice transplanted into C57BL/6 recipient mice (genotypes are indicated in the figure) with or without 3pRNA treatment on day −1. Pooled data from two independent experiments. (B) Number of organoids after 5 days in culture with or without the addition of 3pRNA (2 μg/ml) on day 1 of culture; organoids were derived from C57BL/6 IFNAR1+/+ or IFNAR1−/− mice. One representative experiment of three independent experiments. (C) Representative images of organoids derived from C57BL/6 WT mice after 5 days in culture with or without the addition of 3pRNA (2 μg/ml) on day 1 of culture. Organoid area is shown. (D) Measurement of organoids from C57BL/6 WT mice after 5 days in culture with or without the addition of 3pRNA (2 μg/ml) or α-IFNAR1–blocking antibody (10 μg/ml) on day 1 of culture. One representative experiment of three independent experiments. (E) Size of organoids derived from C57BL/6 WT mice after 7 days in culture with or without the addition of recombinant murine IFN-β (20 U/ml) on day 1 of culture. One representative experiment of three independent experiments. (F) qPCR of RegIIIγ expression in organoids 24 hours after stimulation with indicated combinations of 3pRNA and α-IFNAR1–blocking antibody or IgG1 isotype control. Pooled data from three independent experiments. All experiments were analyzed using two-tailed unpaired t test or ordinary one-way ANOVA for multiple comparisons. Survival was analyzed using the log-rank test. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM.

  • Fig. 6. RIG-I activation protects ISCs after allo-HSCT.

    (A) qPCR expression of Lysozyme P and Lgr5 in small intestine biopsies from MAVS+/+ or MAVS−/− C57BL/6 recipient mice after allo-HSCT with 5 × 106 BM cells and 2 × 106 T cells from BALB/c WT donor mice. Pooled data from five independent experiments. (B) Analysis of allo-HSCT BALB/c recipients in the presence or absence of 3pRNA treatment on day −1. Immunohistochemistry showing lysozyme staining of the small intestine of BALB/c recipients 8 days after allo-HSCT. Lysozyme-positive gut Paneth cells are indicated by black arrowheads. Histogram shows number of Paneth cells per crypt. Representative images and pooled data from three independent experiments. (C) qPCR showing expression of Lysozyme P and Lgr5 in IECs from the small intestine of BALB/c recipient mice 8 days after allo-HSCT with or without 3pRNA treatment on day −1 before allo-HSCT. Pooled data from three independent experiments. (D) qPCR of Lysozyme P and Lgr5 expression in the small intestine of BALB/c recipient mice after allo-HSCT with or without 3pRNA treatment on day +1. Data from one experiment. (E) Number of organoids derived from C57BL/6 recipient mice on day 8 after allo-HSCT with or without 3pRNA treatment on day −1 before allo-HSCT. Pooled data from four independent experiments. Animal numbers per group (n) are depicted above figure panels. All experiments were analyzed using two-tailed unpaired t test or ordinary one-way ANOVA for multiple comparisons. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM.

  • Fig. 7. STING pathway protects mouse recipients from GVHD after allo-HSCT.

    (A) Survival of cohoused C57BL/6 WT or Stinggt/gt mice that received 5 × 106 BM cells plus 1 × 106 T cells from WT B10.BR donor mice. Pooled data from two independent experiments. (B) Relative abundance of bacterial genera in the intestinal microbiota of cohoused WT and Stinggt/gt mice. One representative experiment of two independent experiments. Survival (C) and weight loss (D) for C57BL/6 mouse recipients after allo-HSCT from B10.BR donor mice with or without IFN-stimulatory DNA (ISD) treatment on day −1. Pooled data of two (C) or three (D) independent experiments. (E) FITC-dextran concentrations in plasma of BALB/c mouse recipients after allo-HSCT from C57BL/6 donors with or without calf thymus DNA (CT DNA) or IFN-stimulatory DNA treatment on day −1. Pooled data from two independent experiments. (F) Number of organoids derived from WT or Stinggt/gt C57BL/6 mouse small intestine after 5 days in culture with or without the addition of IFN-stimulatory DNA (2 μg/ml) on day 1 of culture. Pooled data from three independent experiments. (G) Organoids derived from C57BL/6 mouse small intestine. Measurement of organoid size after 5 days in culture with or without the addition of IFN-stimulatory DNA (2 μg/ml), α-IFNAR1–blocking antibody (10 μg/ml), or IgG1 isotype control on day 1 of culture. The experiment was performed three times, and images of one representative experiment are shown. (H) qPCR of RegIIIγ expression in organoids derived from C57BL/6 WT mice 24 hours after stimulation with indicated combinations of IFN-stimulatory DNA, α-IFNAR1–blocking antibody, or IgG1 isotype control. Pooled data from three independent experiments. (I) DNA in plasma of BALB/c mice 24 hours after TBI (9 Gy) or allo-HSCT. Pooled data from three independent experiments. Animal numbers per group (n) are depicted above figure panels. All experiments were analyzed using two-tailed unpaired t test or ordinary one-way ANOVA for multiple comparisons. Survival was analyzed using the log-rank test. *P < 0.05, **P < 0.01, and ***P < 0.001. Data are means ± SEM.

Supplementary Materials

  • www.sciencetranslationalmedicine.org/cgi/content/full/9/386/eaag2513/DC1

    Materials and Methods

    Table S1. Antibodies.

    Fig. S1. Endogenous RIG-I/MAVS signaling reduces intestinal tissue damage caused by conditioning therapy and attenuates GVHD.

    Fig. S2. Donor-derived T cells show enhanced alloreactivity in Mavs−/− allo-HSCT recipients.

    Fig. S3. RIG-I ligands have to be applied before or during allo-HSCT to exert their protective effects and do not affect GVL.

    Fig. S4. RIG-I–induced treatment effects are mediated by IFN-Is.

    Fig. S5. RIG-I–induced IFN-Is enhance epithelial regeneration through stimulation of the ISC compartment.

    Fig. S6. MAVS-deficient mice do not display an inherent defect in organoid formation or in the number of Paneth cells.

    Fig. S7. TBI and IFN-stimulatory DNA induce a systemic IFN-I response, and feces-derived RNA triggers a RIG-I–dependent IFN-I response in IECs.

  • Supplementary Material for:

    RIG-I/MAVS and STING signaling promote gut integrity during irradiation- and immune-mediated tissue injury

    Julius C. Fischer, Michael Bscheider, Gabriel Eisenkolb, Chia-Ching Lin, Alexander Wintges, Vera Otten, Caroline A. Lindemans, Simon Heidegger, Martina Rudelius, Sébastien Monette, Kori A. Porosnicu Rodriguez, Marco Calafiore, Sophie Liebermann, Chen Liu, Stefan Lienenklaus, Siegfried Weiss, Ulrich Kalinke, Jürgen Ruland, Christian Peschel, Yusuke Shono, Melissa Docampo, Enrico Velardi, Robert R. Jenq, Alan M. Hanash, Jarrod A. Dudakov, Tobias Haas, Marcel R. M. van den Brink,* Hendrik Poeck*

    *Corresponding author. Email: hendrik.poeck{at}tum.de (H.P.); m-van-den-brink{at}ski.mskcc.org (M.R.M.v.d.B.)

    Published 19 April 2017, Sci. Transl. Med. 9, eaag2513 (2017)
    DOI: 10.1126/scitranslmed.aag2513

    This PDF file includes:

    • Materials and Methods
    • Table S1. Antibodies.
    • Fig. S1. Endogenous RIG-I/MAVS signaling reduces intestinal tissue damage caused by conditioning therapy and attenuates GVHD.
    • Fig. S2. Donor-derived T cells show enhanced alloreactivity in Mavs−/− allo-HSCT recipients.
    • Fig. S3. RIG-I ligands have to be applied before or during allo-HSCT to exert their protective effects and do not affect GVL.
    • Fig. S4. RIG-I–induced treatment effects are mediated by IFN-Is.
    • Fig. S5. RIG-I–induced IFN-Is enhance epithelial regeneration through stimulation of the ISC compartment.
    • Fig. S6. MAVS-deficient mice do not display an inherent defect in organoid formation or in the number of Paneth cells.
    • Fig. S7. TBI and IFN-stimulatory DNA induce a systemic IFN-I response, and feces-derived RNA triggers a RIG-I–dependent IFN-I response in IECs.

    [Download PDF]

Related Content