RIPK1 blocks early postnatal lethality mediated by caspase-8 and RIPK3

Abstract

Receptor-interacting protein kinase (RIPK)-1 is involved in RIPK3-dependent and -independent signaling pathways leading to cell death and/or inflammation. Genetic ablation of ripk1 causes postnatal lethality, which was not prevented by deletion of ripk3, caspase-8, or fadd. However, animals that lack RIPK1, RIPK3, and either caspase-8 or FADD survived weaning and matured normally. RIPK1 functions in vitro to limit caspase-8-dependent, TNFR-induced apoptosis, and animals lacking RIPK1, RIPK3, and TNFR1 survive to adulthood. The role of RIPK3 in promoting lethality in ripk1(-/-) mice suggests that RIPK3 activation is inhibited by RIPK1 postbirth. Whereas TNFR-induced RIPK3-dependent necroptosis requires RIPK1, cells lacking RIPK1 were sensitized to necroptosis triggered by poly I:C or interferons. Disruption of TLR (TRIF) or type I interferon (IFNAR) signaling delayed lethality in ripk1(-/-)tnfr1(-/-) mice. These results clarify the complex roles for RIPK1 in postnatal life and provide insights into the regulation of FADD-caspase-8 and RIPK3-MLKL signaling by RIPK1.

Figure 1. Lethality of ripk1^−/− mice is rescued by concomitant ablation of ripk3 and caspase-8 or fadd

(A) Compilation of pup survival of the given genotypes from various crosses. See Figure S1B,D,F for survival data from the individual crosses. The numbers in brackets refer to the number of individual pups of each genotype. (B) Expected and observed frequency of genotypes in offspring at weaning from crosses of ripk1^+/− ripk3^−/− casp8^−/− and ripk1^+/− ripk3^−/− casp8^+/− mice (p=0.8465 for comparison of ripk1 status among ripk3^−/− casp8^−/− genotypes). (C) Expected and observed frequency of genotypes in offspring at weaning from crosses of ripk1^+/− ripk3^−/− fadd^−/− and ripk1^+/− ripk3^−/− fadd^+/− mice (p=0.9592 for comparison of ripk1 status among ripk3^−/− fadd^−/− genotypes). (D) Weight vs. age of male ripk1^−/− ripk3^−/− casp8^−/− or control littermates (p=0.8539). (E) 5-week old ripk1^−/− ripk3^−/− casp8^−/− mouse (arrow) and littermate control. (F) 5-week old ripk1^−/− ripk3^−/− fadd^−/− mouse (arrow) and littermate control. (G) As in (D), but for ripk1^−/− ripk3^−/− fadd^−/− males or control littermates (p=0.2787). See also Figure S1.

Figure 2. Aging ripk1^−/− ripk3^−/− casp8^−/− animals develop an ALPS-like phenotype, yet they have a normal immune system when young

(A) Lymphoid organs at 16 weeks. (B) FACS analysis of cells stained with anti-CD3 and anti-B220 taken from 16-week old animals of the indicated genotypes. (C, D) Percentage of CD3⁺B220⁺ cells among peripheral blood mononuclear cells in mice of the indicated genotypes and ages. (E) Percentage of CD62L^hiCD44⁻ cells in CD4⁺ and CD8⁺ T cell populations from spleens of 4-week old mice of the indicated genotypes. (F) CFSE dilution of 3-week old anti-CD3 and anti-CD28-activated splenic naïve CD4⁺ and CD8⁺ T cells. Blue line corresponds to ripk1^+/+ripk3^−/− casp8^−/− cells and red line corresponds to ripk1^−/− ripk3^−/− casp8^−/− cells.

Figure 3. Loss of RIPK1 potentiates TNF-mediated apoptosis, which is independent of RIPK3

(A) Schematic of the role of RIPK1 in cell death in response to TNF. (B) Cell death assessed by propidium iodide and annexin V (dark grey bars) or cleaved caspase-3 staining (light grey bars) of primary MEF of the specified genotypes treated in the presence or absence of TNF (10 ng/mL), zVAD (50 μM), CHX (25 ng/mL), and/or Nec1 (30 μM) as indicated. Data is representative of three independent experiments. Error bars, s.d. (C) Basal or (D) induced (10 ng/mL TNF for 6h) expression of cFLIP_L determined via qPCR using two separate sets of primers (set #1 and set #2, see Methods for sequences). (E,F) Expression of cFLIP_L determined via qPCR using primer set #1 from the (E) spleen or (F) lung of P1 pups of the indicated genotypes. For qPCR data, relative expression (arbitrary units) is expressed as a ratio of cFLIP_L versus control signal (L32), while fold increase represents ratio of relative expression from end of treatment with TNF (T6h) versus start of experiment (T0h). See also Figure S2.

Figure 4. Ablation of tnfr1 rescues the perinatal lethality of ripk1^−/− ripk3^−/− mice

(A) Survival of pups from intercrosses of ripk1^+/− ripk3^−/− tnfr1^+/− animals. ripk1^−/− ripk3^−/− tnfr1^−/− survival was compared to ripk1^−/− ripk3^−/− tnfr1^+/+ (p<0.0001) and ripk1^−/− ripk3^−/− tnfr1^+/− (p<0.0001). ripk1^−/− ripk3^−/− tnfr1^−/− survival was also compared to ripk1^+/+ripk3^−/− tnfr1^all (p<0.0001) and ripk1^+/− ripk3^−/− tnfr1^all (p<0.0001). (B) Expected and observed frequency of genotypes in offspring at weaning from intercrosses of ripk1^+/− ripk3^−/− tnfr1^+/− mice (p<0.0001 vs. expected Mendelian ratios). (C) Photograph of 6-week old ripk1^−/− ripk3^−/− tnfr1^−/− (arrow) and littermate control. (D–F) Hematoxylin and eosin-stained, cleaved caspase-3 immunostaining, and TUNEL staining of sections of (D) duodenum, (E) spleen and (F) lung from ripk1^−/− ripk3^−/− or control animals. Images are representative of 3 independent experiments. See also Figure S3.

Figure 5. TNFR1-induced RIPK1 activity causes E10.5 lethality in caspase-8-deficient mice

(A) Expected and observed frequency of genotypes in embryos between E14 and E17 from intercrosses of casp8^+/− tnfr1^−/− mice (p= 0.5658 vs. expected Mendelian ratios). (B) Photograph of E16.5 casp8^−/− tnfr1^−/− embryo and casp8^+/+ tnfr1^−/− littermate control. (C) Hematoxylin and eosin-stained sections of whole fixed E16.5 casp8^−/− tnfr1^−/− and casp8^+/+ tnfr1^−/− embryos. (D) Photograph of vascularization of yolk sac from E16.5 casp8^−/− tnfr1^−/− embryo. (E) Expected and observed frequency of genotypes in embryos between E14 and E17 from intercrosses of cflar^+/− tnfr1^−/− ripk3^−/− mice (p= 0.2068 vs. expected Mendelian ratios). (F) Photograph of E16.5 cflar^−/− tnfr1^−/− ripk3^−/− embryo and cflar^+/+tnfr1^−/− ripk3^−/− littermate control. (G) Hematoxylin and eosin-stained sections of whole fixed E16.5 cflar^−/− tnfr1^−/− ripk3^−/− and cflar^+/+tnfr1^−/− ripk3^−/− embryos. (E) Expected and observed frequency of genotypes in embryos between E14 and E17 from intercrosses of fadd^+/− tnfr1^−/− mice (p= 0.9858 vs. expected Mendelian ratios). (I) Expected and observed frequency of genotypes in embryos between E14 and E17 from intercrosses of cflar^+/− tnfr1^−/− mice (p= 0.1897 vs. expected Mendelian ratios).

Figure 6. In the absence of RIPK1, RIPK3-MLKL-dependent cell death is potentiated in response to poly I:C

(A) Model for the activation of RIPK3 via a RIPK1-independent mechanism. (B) Survival of pups from intercrosses of ripk1^+/− trif^−/−, ripk1^+/− tnfr1^−/− or ripk1^+/− tnfr1^−/− trif^−/− animals. Ripk1^−/− tnfr1^−/− trif^−/− survival was compared to ripk1^−/− tnfr1^−/− (p=0.008) and ripk1^−/− trif^−/− (p<0.0001). ripk1^−/− tnfr1^−/− survival was compared to ripk1^−/− trif^−/− (p<0.0001). (C) 3T3-SA cells stably expressing 2vFV-TRIF-HA were transfected with either scrambled or RIPK1 siRNA and treated with AP-1 dimerizer (10nM) in the absence or presence of zVAD (25 μM). Cell death was assessed by Sytox Green uptake using an Incucyte Kinetic Live Cell Imager. Data is representative of three independent experiments. Error bars, s.d. (D) 3T3-SA expressing 2Fv- TRIF-HA were subjected to immunoprecipitation (IP) of HA following the transfection of either scrambled or RIPK1 siRNAs and 45 min treatment with AP-1 (10 nM) in absence or presence of zVAD (50 μM). (E) Mlkl^−/− MEF expressing DOX-inducible tagged MLKL (N-term FLAG-MLKL, left panel and C-Term MLKL-FLAG, right panel) were transfected with either scrambled or RIPK1 siRNAs and then treated with DOX (1 μg/ml), poly I:C (50 μg/mL), zVAD (50 μM) and/or Nec1 (30 μM) as indicated. Cell death assessed by Sytox Green uptake using an Incucyte Kinetic Live Cell Imager. Data is representative of two independent experiments. Error bars, s.d. (F) Cell death of cells treated in (E) by propidium iodide uptake using flow cytometry 10h after treatment. Data is representative of two independent experiments. Error bars, s.d. (G) In the same experiment as (E,F), MLKL was immunoprecipitated with anti-FLAG and co-immunoprecipiated proteins assessed by western blot. IP of MLKL via FLAG (N-Term, left panel, C-Term, right panel) of mlkl^−/− MEF treated as indicated for 4h. (H) IP of endogenous RIPK1 in mlkl^−/− MEF expressing FLAG-MLKL (N-Term) treated as in (E). For all cases, immune complexes were detected by Western blotting with the indicated antibodies. See also Figure S4.

Figure 7. In the absence of RIPK1, RIPK3-dependent cell death is potentiated in response to interferons

(A) Survival of pups from intercrosses of ripk1^+/− ifnar^−/−, ripk1^+/− tnfr1^−/− or ripk1^+/− tnfr1^−/− ifnar^−/− animals. ripk1^−/− tnfr1^−/− ifnar^−/− survival was compared to ripk1^−/− tnfr1^−/− (p=0.0001) and ripk1^−/− ifnar^−/− (p<0.0001). ripk1^−/− tnfr1^−/− survival was compared to ripk1^−/− ifnar^−/− (p=0.0079). (B) Cell death assessed by propidium iodide uptake via flow cytometry of primary MEF of the specified genotypes treated in the presence or absence of Interferon-γ (IFNγ)(100 ng/mL) or Interferon-β (IFNβ) (1000 U/mL). Data is representative of three independent experiments. Error bars, s.d. (C,D) Cell death assessed by propidium iodide uptake via flow cytometry of primary ripk1^+/+ or ripk1^−/− MEF treated in the presence or absence of (C) IFNγ (100 ng/mL), zVAD (25 μM), and/or Nec1 (30 μM) or (D) IFNβ (1000 U/mL), zVAD (25 μM), and/or Nec1 (30 μM) as indicated. Data is representative of three independent experiments. Error bars, s.d. (E) Cell death assessed by propidium iodide uptake via flow cytometry of primary mlkl^+/+ or mlkl^−/− MEF treated in the presence or absence of IFNγ (100 ng/mL), IFNβ (1000 U/mL), TNF (10ng/mL), and/or zVAD (25 μM) as indicated. Data is representative of two independent experiments. Error bars, s.d. (F) Model of RIPK1 function in cell death. See also Figure S5–7.