Type I interferon signaling before hematopoietic stem cell transplantation lowers donor T cell activation via reduced allogenicity of recipient cells

Recent studies highlight immunoregulatory functions of type I interferons (IFN-I) during the pathogenesis of graft-versus-host disease (GVHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT). We demonstrated that selective activation of IFN-I pathways including RIG-I/MAVS and cGAS/STING prior to allo-HSCT conditioning therapy can ameliorate the course of GVHD. However, direct effects of IFN-Is on immune cells remain ill characterized. We applied RIG-I agonists (3pRNA) to stimulate IFN-I production in murine models of conditioning therapy with total body irradiation (TBI) and GVHD. Using IFN-I receptor-deficient donor T cells and hematopoietic cells, we found that endogenous and RIG-I-induced IFN-Is do not reduce GVHD by acting on these cell types. However, 3pRNA applied before conditioning therapy reduced the ability of CD11c+ recipient cells to stimulate proliferation and interferon gamma expression of allogeneic T cells. Consistently, RIG-I activation before TBI reduced the proliferation of transplanted allogeneic T-cells. The reduced allogenicity of CD11c+ recipient cells was dependent on IFN-I signaling. Notably, this immunosuppressive function of DCs was restricted to a scenario where tissue damage occurs. Our findings uncover a context (damage by TBI) and IFN-I dependent modulation of T cells by DCs and extend the understanding about the cellular targets of IFN-I during allo-HSCT and GVHD.


Wild-type BM (n=3)
Wild-type BM + T cells (n=11) Wild-type BM + T cells + 3pRNA (n=11) (B,C) Overall recipient weight course after allo-HSCT and weight loss on day 8 after allo-HSCT as a surrogate marker for GVHD severity. Data are pooled from two independent experiments. Animal numbers per group (n) are depicted. Survival was analyzed using the Log-rank test. Weight loss data were analyzed using ordinary one-way ANOVA for multiple comparisons. Significance was set at p values < 0.05, p < 0.01 and p < 0.001 and was then indicated with asterisks (* , ** and ***). Data are presented as mean ± S.E.M.
www.nature.com/scientificreports www.nature.com/scientificreports/  www.nature.com/scientificreports www.nature.com/scientificreports/ Results type i interferon receptor signaling in donor t and hematopoietic cells does not contribute to the development of GVHD. First, we addressed the influence of IFN-I on transplanted donor cells. To this end, we used a major mismatch allo-HSCT mouse model and administered allogeneic bone marrow (BM) and T cells derived from either Wild-type (WT) or IFNaR1 −/− donor mice. WT recipient mice transplanted with allogeneic BM and T cell grafts derived from IFNaR1 −/− donor mice showed similar survival as compared to allo-HSCT recipients transplanted with IFNaR1 +/+ (WT) BM and T cells (Fig. 1A). Furthermore, lack of IFNaR1 in donor BM and T cell grafts did not influence weight loss of transplanted mice, which is a surrogate marker for the severity of intestinal GVHD (Fig. 1B,C).
Next, we analyzed the contribution of IFNaR1 signaling in donor cells to RIG-I mediated protection against GVHD. We have shown previously that the therapeutic effect of selective RIG-I ligation using 3pRNA in allo-HSCT recipients is mediated via IFN-I signaling 5 . Consistent with this, recipient mice transplanted with wild-type BM and T cells showed reduced weight loss and prolonged survival after injection with the specific RIG-I ligand 3pRNA one day prior to allo-HSCT, as compared to untreated control animals (Fig. 1A,B). Allo-HSCT recipients transplanted with IFNaR1 −/− BM and IFNaR1 −/− T cells showed a similar beneficial response to 3pRNA treatment with significantly reduced weight loss and mortality ( Fig. 1A-C). Taken together, these data demonstrate that neither the development of GVHD nor its amelioration by therapeutic targeting of RIG-I is modulated by IFN-I signaling in hematopoietic donor cells and donor T cells.

RIG-I activation before conditioning therapy reduces activation of transplanted donor T cells in vivo.
Based on the above findings, we hypothesized that RIG-I activation limits recipient cells to activate transplanted donor T cells after allo-HSCT. Therefore, we analyzed proliferation and cell death of transplanted donor T cells in mice that were either treated with 3pRNA before TBI or left untreated. We found that RIG-I activation with 3pRNA before TBI reduced the absolute number of CD4 + and CD8 + donor T cells after transplantation, whereas the percentage of Annexin V positive or Propidiumiodid positive T cells was unchanged between 3pRNA treated and untreated recipients ( Fig. 2A,B). Next we analyzed the proliferation of transplanted T cells and found that 3pRNA treatment reduced the proliferation of CD4 + and CD8 + donor T cells (Fig. 2C,D). Finally, we analyzed apoptosis as determined by caspase 3 activation and found that caspase 3 activation of transplanted CD4 + and CD8 + T cells was comparable in 3pRNA treated and untreated mice. In addition, we found that the percentage of T cells that did not show either proliferation or caspase 3 activation was increased after 3pRNA treatment ( Fig. 2C,D). In sum, we found reduced numbers of donor T cells after transplantation, which correlated mechanistically with reduced proliferation but not with enhanced cell death.

Induction of IFN-I before conditioning therapy with TBI reduces allogenicity of dendritic cells to activate allogeneic t cells. RIG-I-induced IFN-I release may limit recipient antigen presenting cells
(APCs) to activate donor T cells resulting in reduced activation of allogeneic T cells. To address this, we first used a conventional mixed lymphocyte reaction (MLR) with in vitro generated BM-derived dendritic cells (DCs), which are co-cultured with allogeneic CD4 + or CD8 + T cells after stimulation with 3pRNA. The advantage of this conventional MLR was to analyze direct RIG-I dependent effects on DC function independent of the pleotropic effects on DCs that may be induced by the conditioning therapy before allo-HSCT. After three to five days of co-culture, we assessed proliferation and IFN-γ production of allogeneic T cells (Fig. 3A). We did not observe significant changes in allogenic T cell activation after DC stimulation with RIG-I-MAVS activating 3pRNA in vitro ( Fig. 3B-D). Furthermore, blocking of the IFN-I receptor with anti-IFNaR1 antibody did not alter allogeneic CD4 + or CD8 + T cell activation in vitro ( Fig. 3B-D).
We therefore postulate that 3pRNA treatment before the conditioning therapy negatively regulates T cell stimulatory responses induced by conditioning. This results in reduced allogeneic T cell activation after allo-HSCT. Therefore, a conventional MLR using BM-derived dendritic cells DCs co-cultured with allogeneic T cells in vitro and in the absence of damage cannot mirror this scenario.
We therefore aimed to analyze the allogenicity of recipient DCs after in vivo 3pRNA treatment and in vivo conditioning therapy. On day 3 after allo-HSCT, high amounts of transplanted donor T cells are located within the spleen of recipient mice, before they begin to infiltrate GVHD effector organs such as the intestine 10 . We thus aimed to analyze the potency of splenic recipient CD11c + DCs to activate allogeneic T cells after conditioning therapy. Consequently, we used an ex vivo MLR to mimic the interaction of transplanted donor T cells with recipient DCs after conditioning therapy and allo-HSCT in the host. We isolated splenic CD11c + DCs on day 3 after TBI from mice that had already been treated with 3pRNA prior to irradiation (Fig. 4A). We then subjected isolated CD11c + cells to co-culture with CD4 + or CD8 + T cells isolated from allogeneic mice. After three to five days of co-culture, we assessed DC allogenicity by measuring proliferation and IFN-γ production of T cells (Fig. 4A,B). DCs isolated from irradiated mice activated allogenic CD4 + and CD8 + T cells (Fig. 4B,C). Pretreatment of DC donor mice with 3pRNA prior to radiation therapy and cell harvest significantly decreased proliferation and IFN-γ production of subsequently co-cultured allogenic T cells (Fig. 4B,C). To analyze the role of IFNaR1 activation in our model, we per group. (D) Upper panel: Percent proliferated (CellTrace − ) CD8 + donor T cells of all transplanted CD8 + donor T cells. Mid panel: Percent active caspase 3 + CD8 + donor T cells of all transplanted CD8 + donor T cells. Percent active caspase 3 negative and non-proliferated CD8 + donor T cells (CellTrace + ) of all transplanted CD8 + donor T cells. Representative FACS plots of CD8 + h2kb + T cells. Pooled data of two experiments, 12 mice per group. Data were analyzed using two-tailed unpaired t test. Significance was set at p values < 0.05, p < 0.01 and p < 0.001 and was then indicated with asterisks (* , ** and ***). Data are presented as mean ± S.E.M. www.nature.com/scientificreports www.nature.com/scientificreports/ additionally pretreated recipients with IFNaR1-blocking antibody on day -2 and before injection of 3pRNA (day -1) (Fig. 4A). Pre-treatment with IFNaR1-blocking antibody abrogated the ability of 3pRNA to limit allogeneic T cell activation in our ex vivo MLR (Fig. 4B,C). We conclude that in vivo pretreatment of mice with 3pRNA before conditioning therapy lowers the allogenicity of splenic DCs via an IFN-I dependent mechanism.
Tissue damage induced by irradiation is a pre-requisite for RIG-I induced immunosuppression. Next, we examined whether 3pRNA-induced RIG-I signaling directly influenced the allogenicity of dendritic cells in the absence of tissue damage induced by pre-transplant conditioning such as radiation therapy. Therefore, we isolated splenic CD11c + DCs from mice that had been treated with 3pRNA but did not undergo subsequent irradiation (Fig. 5A). We found that 3pRNA treatment of naive mice did not influence the potency of isolated DCs to activate allogeneic CD4 + or CD8 + T cells (Fig. 5B,C). We thus conclude that 3pRNA treatment and subsequent IFN-I release in mice per se do not directly influence the allogenicity of DCs. Rather, 3pRNA pre-treatment of mice seems to dampen the immune activating effects of tissue damage on APCs to stimulate allogeneic T cells.

Discussion
Recent studies highlighted the potential of IFN-I to modulate intestinal epithelial barrier damage and regeneration 5,11-13 . Furthermore, several independent groups have shown that IFN-I activation can ameliorate GVHD after allo-HSCT 4-6 . We reported previously that treatment of recipient mice with IFN-I inducing targeted therapy of RIG-I prior to allo-HSCT results in reduced development of GVHD. Mechanistically, IFN-I activation promoted intestinal barrier function and epithelial regeneration after conditioning therapy and during GVHD. We could link this enhanced mucosal integrity to reduced proliferation and IFN-γ production of transplanted T cells in vivo 5 . We now characterized the details of the influence of IFN-I signaling on donor T cell activation after allo-HSCT. We found that i) IFN-I signaling in donor hematopoietic and T cells does not influence GVHD after allo-HSCT and ii) targeted therapy of RIG-I ameliorates GVHD independent of IFN-I signaling in donor hematopoietic and T cells. Furthermore, RIG-I activation before allo-HSCT resulted not only in reduced proliferation as previously described 5 but also reduced apoptosis and a large fraction of inactive transplanted donor T cells. Mechanistically, we found that in vivo 3pRNA treatment reduced the potency of isolated CD11c + recipient DCs to activate allogeneic donor T cells in an IFN-I dependent manner. Importantly, this immunosuppressive www.nature.com/scientificreports www.nature.com/scientificreports/ role of IFN-I signaling seems to be restricted to a scenario of genotoxic tissue damage as neither RIG-I activation and IFN-I induction in naive (unirradiated) mice altered allogeneic T cell activation. Consistent with this, in vitro activation of the RIG-I/MAVS/IFN-I pathway in in vitro generated DCs failed to alter the activation of allogeneic CD4 + or CD8 + T cells during conventional MLRs. Our data are contradictory to the well-established role of RIG-I induced IFN-Is in mediation enhanced cross-priming of syngeneic CD8 + T cells 14,15 . However, little is currently known about context dependent roles of IFN-Is on APC function and their modulation of T cell response. Therefore, further mechanistic studies are needed to determine how genotoxic tissue damage such as irradiation modulates the effects of I RIG-I/cGAS induced IFN-I on allogeneic and syngeneic T cell cross priming.
We have previously published that allo-HSCT recipient mice with a selective deficiency for IFN-I receptor signaling in CD11c + dendritic cells (CD11cre pos IFNaR1 floxed mice) still develop reduced GVHD after treatment with 3pRNA prior to allo-HSCT 5 . Putting our new data and our previously published results in context, we conclude that 3pRNA-induced IFN-I does not directly target CD11c + DCs to reduce their allogenicity to prime donor T cells but acts via an indirect mechanism. One explanation could be that IFN-I protects the intestinal barrier function after allo-HSCT resulting in reduced APC and donor T cell activation and intestinal GVHD (Fig. 6) 5 . This conclusion is in line with a recent study showing that metabolites derived from the intestinal microbiota act via modulation of IFN-I

Material and Methods
Mice and animal studies. Animal studies were approved by the local regulatory authority (Regierung von Oberbayern, Munich). All experiments were performed in accordance with the relevant guidelines and regulations. C57BL/6 (H-2kb) and Balb/c (H-2kd) were purchased from Janvier Labs (France). IFNaR1-deficient mice (Ifnar1 −/− ) were provided by Ulrich Kalinke and were previously described 17 . Mice were between 6 and 10 weeks of age at the onset of experiments.

Induction of GVHD after allo-HSCT with myeloablative TBI. Induction of GVHD after allo-HSCT
with myeloablative TBI using a major mismatch (H-2kd/H-2kb) GVHD mouse model were performed as previously described 18 . Briefly, Balb/c recipients were injected with 5 × 10 6 BM cells directly after myeloablative TBI with 2 × 4.5 Gy. Radiation was performed using the Gulmay RS225A irradiation device (Gulmay Medical, Camberley, UK) at a dose rate of 0.95 Gy/min (15 mA, 200 keV). Some mice additionally received 0.5 × 10 6 C57BL/6 purified donor T cells. Donor T cells and bone marrow cells were isolated from either WT mice or IFNaR1-deficient mice as indicated in the figure legends. Weight loss was monitored daily after allo-HSCT.

Mixed lymphocyte reaction with bone marrow derived GMCSF dendritic cells. Mixed lympho-
cyte reaction (MLR) were performed as described previously 19 . In brief, BM cells were harvested from murine femur and tibia of C57BL/6 mice and erythrocytes were lysed with red blood cell lysis buffer. BMDCs www.nature.com/scientificreports www.nature.com/scientificreports/ mL GM-CSF. Fresh medium was added on the 3rd and 6th day after the start of the culture. GMCSF-DCs were harvested on day 7 using PBS supplemented with 2 mM EDTA. 25 × 10 3 GMSCF-DCs (C57BL/6) were plated in 96-U-well plates. DCs were treated with 3pRNA if indicated. To do so, DCs were transfected with 3pRNA (1 µg/ ml) complexed in Lipofectamin 2000 (Life Technologies, Darmstadt, Germany) according to the manufacturer's protocol. In indicated experiments GMCSF-DCs were treated with anti-IFNaR1 blocking antibody (10 ug/ml) (Clone: MAR1-5A3, BioXCell, West Lebanon, NH). On the next day, 10 5 CD4 + or CD8 + T cells extracted from the spleens of naïve WT mice (Bab/c) were added to the GMSCF-DCs (C57BL/6). Proliferation of T cells was analyzed using the CellTrace ™ CFSE Cell Proliferation Kit (Thermo Fisher Scientific). CD4 + or CD8 + T cells labeled with CFSE were analyzed on day 3 (CD8 + T cells) or day 5 (CD4 + T cells) after coculture using flow cytometry.
Mixed lymphocyte reaction with isolated CD11c + splenocytes. Spleens were cut into small pieces and incubated with PBS +Ca/+Mg supplemented with FCS (10%) and Collagenase II (200 U/ml; Worthington) at 37 °C for 45 min. Remaining tissue was mashed slightly and filtered through a 100 μm strainer (BD 352360). Afterwards, CD11c + cells were purified using CD11c + MACS positive selection. CD11c + cells were washed, resuspended in complete media and 1 × 10 4 cells were cocultured with 1 × 10 5 allogeneic T cells purified from splenocytes derived from C57BL/6 mice (CD4 or CD8 MACS positive selection). In indicated experiments, Balb/c mice were exposed to TBI with 9 Gy three days prior to isolation of CD11 + splenocytes. Proliferation of T cells was analyzed using the CellTrace ™ CFSE Cell Proliferation Kit (Thermo Fisher Scientific). CD4 + or CD8 + T cells labeled with CFSE were analyzed on day 3 (CD8 + T cells) or day 5 (CD4 + T cells) after coculture using flow cytometry. flow cytometry. All antibodies were purchased from BD Bioscience or BioLegend. We used LIVE/DEAD Fixable Dead Cell Stains and the Dead Cell Apoptosis Kit with Annexin V FITC and Propidium Iodide (PI, Thermo Fisher Scientific) according to the manufacturer's instructions. For intracellular cytokine staining, T cells were activated with 80 nM Phorbol-12-myristat-13-acetat (PMA; Sigma), 1 μM ionomycin (Merck Millipor) and Brefeldin A for 4 hours. The Foxp3 Transcription Factor Fixation/Permeabilization Kit (eBioscience) was used according to manufacturer's instructions. Data acquisition and analysis was performed using a FACSCanto II (BD Bioscience) and Flow Jo software (Tree Star).

t cell analysis in vivo.
In vivo T cell proliferation was performed as previously described 5 . T cell preparation was performed as described above. T cells were stained with CellTrace ™ Violet Cell Proliferation Kit (Thermo Fisher Scientific) according to the manufacturer's instructions. 12 minutes after staining at 37 °C, T cells were washed and counted. 15 × 10 6 stained cells were transplanted into lethally irradiated allogeneic recipients as described above. Spleens of allo-HSCT recipients were harvested on day 3 and analyzed with FACS. Caspase 3 activation was analyzed as previously described 20 . Active caspase-3 stain was performed directly before further viability and cell surface staining using the CaspGLOW Fluorescein Active Caspase-3 Staining Kit (Thermo Fisher Scientific) according to the manufacturer's instructions. Whole splenocytes were incubated with the staining reagent for 60 minutes at 37 °C. Afterwards, surface staining was performed as described above.
Statistics. Animal numbers per group (n) are depicted in the figure legends. GraphPad Prism version 6 was used for statistical analysis. Survival was analyzed using the Log-rank test. Differences between means of experimental groups were analyzed using two-tailed unpaired t test or ordinary one-way ANOVA correspondingly to the distribution shape of our observations. We used ordinary one-way ANOVA for multiple comparisons and always performed Dunnett's test for Multiple-test corrections. Applied statistical tests are indicated in the figure legends. Significance was set at p values < 0.05, p < 0.01 and p < 0.001 as indicated with asterisks (* , ** and ***, respectively). Data are presented as mean ± S.E.M.