A stromal inflammasome Ras safeguard against Myc-driven lymphomagenesis

Abstract

The inflammasome plays multifaceted roles in cancer, but less is known about its function during premalignancy upon initial cell transformation. We report a homeostatic function of the inflammasome in suppressing malignant transformation through Ras inhibition. We identified increased hematopoietic stem cell (HSC) proliferation within the bone marrow of inflammasome-deficient mice. HSCs within an inflammasome-deficient stroma expressed a Ras signature associated with increased Ras pathway- and cancer-related transcripts and heightened levels of cytokine, chemokine and growth factor receptors. Stromal inflammasome deficiency established a poised Ras-dependent mitogenic state within HSCs, which fueled progeny B cell lymphomagenesis upon Myc deregulation in a spontaneous model of B cell lymphoma, and shortened its premalignant stage leading to faster onset of malignancy. Thus, the stromal inflammasome preserves tissue balance by restraining Ras to disrupt the most common oncogenic Myc-Ras cooperation and establish a natural defense against transition to malignancy. These findings should inform preventative therapies against hematological malignancies.

Extended Data Figure 1

Extended Data Figure 2

Extended Data Figure 3

Extended Data Figure 4

Extended Data Figure 5

Extended Data Figure 6

Extended Data Figure 7

Extended Data Figure 8

Extended Data Figure 9

Extended Data Figure 10

Figure 1. Inflammasome impairment expands circulating and bone marrow B cell progenitors in premalignant Eμ-myc mice.

a-c, Representative images and plots from 4-week-old pre-malignant mice of each genotype and malignant Eμ-myc mice. a, Gross pathology of mesenteric, axial, and inguinal lymph nodes. b, Giemsa-stained peripheral blood smears showing degrees of leukocytosis. c. Flow cytometry of peripheral blood pre-gated on CD19⁺ cells, stained with the B cell markers IgM and GL-7. Data in a-c represent at least 10 mice/genotype. d-f, Quantification of CD19⁺GL-7⁺IgM^– frequency (d), total leukocyte counts (e), and frequency of CD19⁺ cells (f) in the blood. Data in d-f represent at least 10 mice/genotype. g, Representative flow cytometry of bone marrow CD19⁺ cells stained for GL-7 and IgM. h-j, Quantification of total bone marrow (h), and absolute numbers of CD19⁺ (i), and GL-7⁺IgM⁻ (j) cells from conventional flow gating strategy. Data in h-j represent at least 4 mice/genotype. g, Cytokine concentration measurements on bone marrow supernatants determined by cytometric bead array. Data represent four independent experiments, at least eight mice per group. k, Representative CyTOF Sunburst diagrams where the % of the circumference for a given cell gate corresponds to the frequency of that group out of the total CD45⁺ cells in the bone marrow. Data in (k) represent 6 mice/genotype. l, Quantification of GL-7⁺ cells out of total cells in the bone marrow from our CyTOF data (6 mice per group). Each symbol represents an individual mouse. Data represent at least three independent experiments. p^*≤0.05, p**≤0.01, p***≤0.001, p****≤0.0001 calculated by two-tailed student’s t-test or 1-Way ANOVA with Sidak’s posttest. All quantified data are presented as mean values +/- SEM.

Figure 2. Stromal inflammasome deficiency accelerates malignant disease onset in Eμ-myc mice.

a, Kaplan-Meier curves of cohorts of mice of each genotype indicating age at first pathology. b, Kaplan-Meier curves of mixed fetal liver chimeras. In a 9:1 ratio, either μMT or μMT/ Casp1^−/−Casp11^−/− bone marrow mixed with either Eμ-myc or Eμ-myc Casp1^−/−Casp11^−/− fetal liver cells were used to reconstitute WT and Casp1^−/−Casp11^−/− recipient mice, as indicated. In resultant chimeras, Casp1 and Casp11 deficiency was isolated to either the B cell compartment (red), the non-B cell hematopoietic compartment (yellow), the stromal compartment (green), or both the stromal and non-B cell hematopoietic compartments (blue). Control chimeras with no Casp1 or Casp11 deficiency shown as black. a,b, Mouse numbers in each group indicated in parentheses. p*≤0.05, p**≤0.01, calculated by Mantel-Cox Log-Rank test.

Figure 3. Inflammasome impairment accelerates bone marrow stem and B cell progenitor proliferation in Eμ-myc mice.

a, Representative SPADE tree analysis of CyTOF data depicting the gating strategy for the B cell lineage, from hematopoietic stem cell (HSC) through fully differentiated B cell (LT-HSC = long-term HSC; ST-HSC = short-term HSC; MPP/MLP = multipotent progenitor/multilineage progenitor; CLP = common lymphoid progenitor). b, Representative heatmap of the B cell lineage from the gating strategy in (a) showing relative staining intensity of the indicated markers for each population in the bone marrows of 4-week-old mice. c, Heatmap of the raw % IdU positive cells for each population for each genotype (3 mice per group). d, Heatmap of relative %IdU positive cells per population, normalized within each population column across the four genotypes (3 mice per group). e, Representative bidirectional plots of CyTOF IdU staining within the LT-HSC population from the bone marrow of each indicated genotype. f, Bar chart depicting the % increase in IdU⁺ cell frequency within each indicated population comparing Eμ-myc Casp1^−/−Casp11^−/− to Eμ-myc. Significance determined by student t-test of IdU⁺ cell frequency within each population (3 mice per group). g, Pre-B cell colony formation (CFC) assay using bone marrow from the indicated genotypes, 5 biological replicates/genotype. h, Cell viability in the pre-B CFC assays from (g) as determined by Annexin-V⁺7-AAD⁺ staining using flow cytometry, 3 biological replicates/genotype. i, HSC colony formation in complete methylcellulose media without Epo with plated bone marrow from the indicated genotypes and cytokine treatment groups, n = a single 2cm x 2cm area of formed colony counts for each of the 5 biological replicates per genotype. j, IL-1β concentrations in bone marrow supernatants determined by cytometric bead array, 5 biological replicates/genotype. k, IL-18 concentrations in bone marrow supernatants determined by ELISA, 6 biological replicates/genotype p*≤0.05, p**≤0.01, p***≤0.001, calculated by two-tailed unpaired student’s T-test. All quantified data are presented as mean values +/- SEM.

Figure 4. A heightened HSC Ras signaling signature upon inflammasome deficiency.

a, Principal components analysis (PCA) plot of RNAseq-derived transcriptomes of sorted HSCs. b, Volcano plots showing differential expression due to Casp1^−/−Casp11^−/− in HSCs of Eμ-myc (top) and wild type (bottom) mice. Positive log fold changes indicate higher expression in Casp1^−/−Casp11^−/−. Points representing significantly differentially expressed genes (adjusted p < 0.1) are colored according to the Venn diagrams, which indicate the numbers of differentially expressed genes shared in the wild type and Eμ-myc backgrounds. c, Volcano plots showing differential expression due to Eμ-myc in Casp1^−/−Casp11^−/− (top) and wild type (bottom) mice. Differentially expressed genes are colored according to the Venn diagrams, which show differentially expressed genes shared in the Casp1^−/−Casp11^−/− and wild type backgrounds. d, Gene Ontology (GO) biological process terms most significantly enriched among genes differentially expressed due to caspase 1 and caspase 11 deficiency in both Eμ-myc and wild type backgrounds (common genes indicated in purple in (b)). (b,d) Enrichment p values were corrected using the method of Benjamini and Hochberg. e, Heatmaps showing normalized expression levels of differentially expressed genes from selected enriched GO terms. Genes shown are differentially regulated by Casp1^−/−Casp11^−/− in both Eμ-myc and wild type backgrounds. Color scale indicates row z score of normalized read counts. f, Representative histograms of normalized Ras target expression in Lin^–Sca1⁺c-Kit⁺CD48^– HSC from bone marrow of adult male Wt, Casp1^−/−Casp11^−/−, Eμ-myc and Eμ-myc Casp1^−/−Casp11^−/− mice. g, Quantification of gMFI of (f); n=6 biological replicates per genotype; p*≤0.05, p**≤0.01, p***≤0.001, p****≤0.0001; 2 WAY Anova. h, Quantitative real time PCR showing fold change expression of Ras signaling associated, and cancer related genes derived from RNA-Seq data relative to WT (normalized to β2M at n=1) of enriched HSCs from bone marrow of adult male WT, Casp1^−/−Casp11^−/−, Eμ-myc and Eμ-myc Casp1^−/−Casp11^−/− mice; n=3 biological replicates per genotype; p*≤0.05, p**≤0.01, p***≤0.001, p****≤0.0001; 2 WAY Anova. All data are presented as mean values +/-SD.

Figure 5. The stromal inflammasome constrains the Ras pathway in HSCs.

a, Schematic of bone marrow chimeric mouse experimental design to investigate the role of Caspase-1/11 in bone marrow stroma on wild type bone marrow derived HSC phenotype and gene expression. Created with BioRender.com and exported under a paid subscription to BioRender. b, Representative histograms of HSC (Lin^–Sca1⁺c-Kit⁺CD48^–) Ras normalized target expression of wild type WT donor CD45.1⁺ CD45.2^– bone marrow from chimeric WT CD45.2⁺ recipient vs Casp1^−/−Casp11^−/− CD45.2⁺ recipient. c, Quantification of gMFI of (b); n=6 biological replicates per genotype, p**≤0.01, p***≤0.001; 2 WAY Anova. d, Quantitative real time PCR showing the fold change expression of Ras signaling associated, and cancer related genes derived from RNA-Seq data relative to WT (normalized to β2M at n=1) of enriched wild type donor CD45.1⁺ CD45.2^– bone marrow HSC from chimeric WT CD45.2 recipient vs Casp1^−/−Casp11^−/− CD45.2⁺ recipient; n=3 biological replicates per genotype; p**£0.01, p***£0.001; ordinary 2 WAY Anova with Sidak’s multiple comparison posttest comparing the mean of each genotype row. e, Representative histograms of cell trace violet (CTV) proliferation dye labeled HSCs (CD45⁺ Sca1⁺ c-Kit⁺ CD48^–) at 8 h timepoint from coculture of WT HSC or 4 h Ras inhibitor treated WT HSC with enriched stroma (CD45^–Sca-1^–c-Kit^–Lin^–) from WT, Casp1^−/−Casp11^−/− mice or WT stroma treated with pan caspase inhibitor QVD-OPH for 2 h at 1:2 ratio (HSC:Stroma). f, Quantification of (e) as the frequency (%) of total HSCs that have undergone at least 4 rounds of proliferation (CTV⁺ left gate = proliferation dye dilution); n=4 biological replicates per condition, p**≤0.01, p****≤0.0001; 2 WAY Anova. g, Quantification of the proliferation and replication (expansion) index from (e); n=3 biological replicates per condition, p*≤0.05, p**≤0.01, p****≤0.0001; 2 WAY Anova. h, Representative histograms of enriched HSC (CD45⁺Sca1⁺c-Kit⁺CD48^–) Ras normalized target expression from 16 h coculture of HSC and stroma as described in (e). i, Quantification of gMFI of (h); n=3 biological replicates per condition, p*≤0.05, p**≤0.01, p***≤0.001, p****≤0.0001; ordinary 2 WAY Anova with Tukey’s multiple comparisons posttest comparing simple effects within condition rows. For f,g, only p values ≤0.05 are shown. All data are presented as mean values +/- SD.

Figure 6. The stromal inflammasome controls HSC surface TNFR1/II and MIP receptors via Ras.

a, Representative cytokine array blots of WT (left) vs Casp1^−/−Casp11^−/− (right) bulk bone marrow after 6 h incubation including array controls of kit Membrane C3. b, Concentration of soluble sTNFR1 and sTNFRII, TNF and indicated chemokines measured by ELISA at 6 h post ex vivo culture of bone marrow from adult male WT, Casp1^−/−Casp11^−/−, and Eμ-myc Casp1^−/−Casp11^−/− mice or WT bone marrow treated with pan caspase inhibitor QVD-OPH for 2 h. n=3 biological replicates per genotype or condition, ns, p>0.05, p*≤0.05, p**≤0.01, p***≤0.001, p****≤0.0001, 2 WAY Anova. c, Representative histograms of indicated receptor expression on HSC (CD45⁺Lin^–Sca1⁺c-Kit⁺CD48^–) from (b) at 6 h post ex vivo culture. d, Quantification of gMFI of (c). e, Representative histograms of indicated receptor expression on enriched HSC (CD45⁺ Lin^–Sca1⁺c-Kit⁺CD48^–) at 16 h of coculture of WT HSC or 4 h Ras inhibitor treated WT HSC with enriched stroma (CD45^–Sca-1^–c-Kit^–Lin^–) from WT, Casp1^−/−Casp11^−/− mice or WT stroma treated with pan caspase inhibitor QVD-OPH for 2 h at 1:2 ratio (HSC:Stroma); n=3 biological replicates per genotype or condition, ns, p>0.05, p****≤0.0001, 2 WAY Anova. f, Quantification of gMFI of (e); n=3 biological replicates per genotype or condition, p*≤0.05, p****≤0.0001, ordinary 2 WAY Anova with Tukey’s multiple comparisons posttest comparing simple genotype column effects within rows. g, Concentration of sTNFR1, sTNFRII, and TNF from (e) cocultures as measured by ELISA; n=3 biological replicates per genotype or condition, ns, p >0.05, p*≤0.05, p**≤0.01, p****≤0.0001, 2 WAY Anova. h, Concentration of indicated MIP family chemokines from (e) cocultures as measured by ELISA; n=3 biological replicates per genotype or condition, ns, p>0.05, p****≤0.0001, 2 WAY Anova. b,d, Both share the same legend shown in b. g,f,h, All share the same legend shown in f. For f, only p values ≤0.05 are shown. All data are presented as mean values +/- SD.

Figure 7. Canonical NLRP3 inflammasome components control HSC Ras signaling in trans.

a, Schematic of bone marrow chimeric mouse experimental design to investigate the role of various bone marrow stromal inflammasome components as indicated on wild type bone marrow derived HSC phenotype and gene expression. b, Representative histograms of HSC (Lin^–Sca1⁺c-Kit⁺CD48^–) Ras normalized target expression of WT donor CD45.1⁺ CD45.2^– bone marrow from chimeric WT recipient vs knockout recipient. c, Quantification of gMFI of (b); n=5 biological replicates per genotype recipient, p*≤0.05, p***≤0.001, p****≤0.0001, 2 WAY Anova. d, Quantitative real time PCR showing the fold change expression of Ras signaling associated, and cancer related genes derived from RNA-Seq data relative to WT (normalized to β2M at n=1) of enriched wild type donor CD45.1⁺ CD45.2^– bone marrow from chimeric WT recipient vs knockout recipients; n=3 biological replicates per genotype recipient, p*<0.05, p**≤0.01, p****≤0.0001, ordinary 2 WAY Anova with Dunnett’s multiple comparisons posttest posttest comparing simple genotype column effects within rows. e, Representative histograms of TNF receptor and MIP receptor expression on HSCs from WT donor CD45.1⁺ CD45.2^– bone marrow from chimeric WT recipient vs knockout recipient. f, Quantification of gMFI of (e); n=5 biological replicates per genotype recipient, p**≤0.01, p****≤0.0001, 2 WAY Anova. g, Representative contour plots of proliferation marker BrDU and DNA synthesis marker PI on HSCs from WT donor CD45.1⁺ CD45.2^– bone marrow from chimeric WT recipient vs knockout recipient. h, Quantification of the frequency of viable HSCs from (g); n=5 biological replicates per genotype recipient, ns, p>0.05, p****≤0.0001, ordinary 2 WAY Anova with Sidak’s multiple comparisons posttest comparing simple genotype column effects within rows. i. Representative histograms of HSC (Lin^–Sca1⁺c-Kit⁺CD48^–) proliferation marker expression of WT donor CD45.1⁺ CD45.2^– bone marrow from chimeric WT recipient vs knockout recipient. j, Quantification of gMFI of (i); n=5 biological replicates per genotype recipient, ns, p>0.05, p**≤0.01, p****≤0.0001, 2 WAY Anova. For c, d, f, only p values ≤0.05 are shown. All data are presented as mean values +/-SD.

Figure 8. Stromal-derived soluble factor and direct stroma-HSC contact mediate stromal inflammasome control of HSCs.

a. Schematic of experimental transwell setup showing HSC in the top insert separated from bone marrow stroma in the lower base by a 3μM pore size allowing exchange of soluble factors and HSC migration towards the stroma. b. Quantification of viable Sca1⁺ cKit⁺ cells with (right) Ras inhibitor treatment and untreated (left) of total CD45⁺ cells in the upper insert and lower base every 8hrs during coculture with WT stroma, Casp1^−/−Casp11^−/− stroma or WT stroma treated with pan caspase inhibitor QVD-OPH, n=3 biological replicates per condition, ns, p>0.05, p*≤0.05, p**≤0.01, 2 WAY Anova. c, Representative histograms of WT HSC (Lin^–Sca1⁺c-Kit⁺CD48^–) Ras normalized target expression during coculture with WT stroma, Casp1^−/−Casp11^−/− stroma or WT QVD-OPH treated stroma. d, Quantification of gMFI of (c), n=3 biological replicates per condition, p*≤0.05, p**≤0.01, p****≤0.0001, 2 WAY Anova. e, Representative histograms of TNF and MIP receptor expression on HSCs during coculture with WT stroma, Casp1^−/−Casp11^−/− stroma or WT QVD-OPH treated stroma. f, Quantification of gMFI of (e), n=3 biological replicates per condition, p*≤0.05, p****≤0.0001, 2 WAY Anova. g, Quantification of cytokines - soluble TNFR and MIP family in the upper HSC (top) and lower stroma (bottom) portion of transwell, n=3 biological replicates per condition, p**≤0.01, p***≤0.001, p****≤0.0001, 2 WAY Anova. h, Representative histograms of TNF and MIP receptor expression on WT and Ras inhibitor treated HSCs during coculture with WT stroma, Casp1^−/−Casp11^−/− stroma or WT QVD-OPH treated stroma at 8hr (top) and 20hr (bottom). i, Quantification of proliferation index of (h), n=3 biological replicates per condition, p*≤0.05, p****≤0.0001, 2 WAY Anova. For d, f, g, i, only p values ≤0.05 are shown. All data are presented as mean values +/- SD.