Adiponectin receptors sustain haematopoietic stem cells throughout adulthood by protecting them from inflammation

Abstract

How are haematopoietic stem cells (HSCs) protected from inflammation, which increases with age and can deplete HSCs? Adiponectin, an anti-inflammatory factor that is not required for HSC function or haematopoiesis, promotes stem/progenitor cell proliferation after bacterial infection and myeloablation. Adiponectin binds two receptors, AdipoR1 and AdipoR2, which have ceramidase activity that increases upon adiponectin binding. Here we found that adiponectin receptors are non-cell-autonomously required in haematopoietic cells to promote HSC quiescence and self-renewal. Adiponectin receptor signalling suppresses inflammatory cytokine expression by myeloid cells and T cells, including interferon-γ and tumour necrosis factor. Without adiponectin receptors, the levels of these factors increase, chronically activating HSCs, reducing their self-renewal potential and depleting them during ageing. Pathogen infection accelerates this loss of HSC self-renewal potential. Blocking interferon-γ or tumour necrosis factor signalling partially rescues these effects. Adiponectin receptors are thus required in immune cells to sustain HSC quiescence and to prevent premature HSC depletion by reducing inflammation.

Extended Data Fig. 1. Adiponectin receptor deficient mice are born in normal numbers and are normal in size (related to Fig. 1).

a-c. Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice were born at the expected mendelian frequencies (a) and were grossly normal in size (b) and appearance (c) (n=4 female and n=8 male Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=10 female and n=3 male control mice in one experiment (b)). d. The number of bone marrow cells in one femur and one tibia or in the spleen of 8-10-week-old Vav1-Cre and control mice (five mice per genotype in one experiment). e. White blood cell, red blood cell, and platelet counts in the blood of 8-10-week-old Vav1-Cre and control mice (five mice per genotype in one experiment). f-i. The frequencies of HSCs (f), MPPs (g), HPCs (h), CMPs, MEPs, and GMPs (i) in bone marrow from one femur and one tibia of Vav1-Cre and control mice (five mice per genotype in one experiment). j-m. The frequencies of HSCs (j), MPPs (k), HPCs (l), CMPs, MEPs, and GMPs (m) in splenocytes from Vav1-Cre and control mice (five mice per genotype in one experiment). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using Mann-Whitney tests followed by Holm-Sidak’s multiple comparisons adjustments (b and d-e) or Student’s t-tests followed by Holm-Sidak’s multiple comparisons adjustments (f-m). All statistical tests were two-sided. Source numerical data are available in the source data files.

Extended Data Fig. 2. Flow cytometry gating strategy for the isolation of hematopoietic stem cells, progenitor cells, and differentiated cells (related to Figs. 1–6)

a-c. Representative flow cytometry gates used to identify hematopoietic stem and progenitor cells (a-b), NK cells, CD4+ T cells, and CD8+ T cells (c) in the bone marrow. The markers used to identify each of the cell populations characterized in this study are listed in Supplementary Table 1. d. Hematoxylin and Eosin stained sections from the spleens of Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice. In Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice the spleens were enlarged (Fig. 1d) and the red pulp was expanded throughout the spleen with increased numbers of myeloid, erythroid, and megakaryocytic cells. In control spleens there was limited extramedullary hematopoiesis confined to the subcapsular region. Representative images from three mice per genotype in one experiment are shown.

Extended Data Fig. 3. Adiponectin receptors are required to sustain HSC function (related to Fig. 2).

a-b. The percentages of HSCs that formed colonies (n=6 mice per genotype in two independent experiments) (a) and colony size (n=3 mice per genotype in one experiment) (b) from Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice. c. The percentage of whole bone marrow cells that formed CFU-GEMM, CFU-GM, CFU-G, CFU-M, and BFU-E colonies (n=3 mice per genotype in one experiment). d. Donor CD45+ cells, T, B, and myeloid cells from the blood of mice competitively transplanted with donor bone marrow cells from Mx1-Cre; Adipor1^fl/fl; Adipor2^fl/fl (n=27 recipients) or control (n=25 recipients) mice (five donors per genotype in five independent experiments). e. Secondary recipients of bone marrow cells from the mice in (d) (n=25 recipients from six Mx1-Cre; Adipor1^fl/fl; Adipor2^fl/fl donors and n=24 recipients from five control donors in five independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using Mann-Whitney tests (d-e), a Student’s t-test (a), a multinomial logistic regression (b), and Student’s t-tests followed by Holm-Sidak’s multiple comparisons adjustments (c), or Mann-Whitney tests (d-e). All statistical tests were two-sided. Source numerical data are available in the source data files.

Extended Data Fig. 4. Adiponectin receptors suppress the production of inflammatory factors (related to Fig. 3).

a. ELISA analysis of inflammatory cytokines in spleen lysates from Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (spleen lysates from 16 (IFNγ, TNF, IL6, and Il1β) or 13 (IFNα and IFNβ) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and 19 (IFNγ, TNF, IL6, and Il1β) or 14 (IFNα and IFNβ) control mice in two independent ELISAs per cytokine). b. qRT-PCR analysis of Tnf transcript levels in sorted bone marrow cells. Data are normalized to transcript levels in wildtype whole bone marrow (WBM) cells (cells sorted from six (WBM), three (HSC, HPC, CMP, MEP, GMP, T-cells, B-cells, and erythroid progenitors), or five (myeloid cells) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and from seven (WBM), three (HSC, HPC, CMP, MEP, GMP, and erythroid progenitors), or four (T-cells, B-cells, and myeloid cells) control mice in one experiment). c. qRT-PCR analysis of Tnf transcripts in sorted CD4+ T cells, CD8+ T cells, or NK cells from the bone marrow (cells sorted from three mice per genotype in one experiment). d. qRT-PCR analysis of Tnf transcripts in sorted macrophages, inflammatory monocytes, and neutrophils from the bone marrow. Data are normalized to transcript levels in wildtype WBM (cells sorted from seven mice per genotype in two independent experiments). e. The frequency of LepR+ stromal cells in bone marrow from 8-10 week old Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=7 mice per genotype in two independent experiments). f. RNAseq analysis of transcripts for Ifng and Tnf in LepR⁺ cells from 8-10 week old Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=3 mice per genotype one experiment). g. The frequencies of donor and competitor HSCs in the spleens of recipient mice co-transplanted with Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl or control donor cells and wildtype competitor cells 16 weeks after transplantation (n=19 recipients from Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl donors and n=20 recipients from control donors in four independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using Student’s t-tests followed by Holm-Sidak’s multiple comparisons adjustments (a-b and d), a matched samples two-way ANOVA followed by Sidak’s multiple comparisons adjustment (c), Student’s t-tests (e), or Mann-Whitney tests followed by Holm-Sidak’s multiple comparisons adjustments (a and g). All statistical tests were two-sided. Source numerical data are available in the source data files.

Extended Data Fig. 5. Adiponectin receptors promote HSC function by reducing IFNγ receptor signaling (related to Fig. 4).

a. Bone marrow (one femur and one tibia) and spleen cellularity in Vav1-Cre; Ifngr1^fl/fl and control mice (n=7 Vav1-Cre; Ifngr^fl/fl mice and n=6 control mice in three independent experiments). b-d. The frequencies of HSCs, MPPs, HPCs (b), CMPs, MEPs, GMPs (c), and differentiated T, B, myeloid, and erythroid cells (d) in the bone marrow of Vav1-Cre; Ifngr1^fl/fl and control mice (n=7 Vav1-Cre; Ifngr^fl/fl mice and n=6 control mice in three independent experiments). e-g. The frequencies of HSCs, MPPs, HPCs (e), CMPs, MEPs, GMPs (f), and differentiated cells (g) in the spleens of Vav1-Cre; Ifngr1^fl/fl and control mice (n=7 Vav1-Cre; Ifngr^fl/fl mice and n=6 control mice in three independent experiments). h. The percentage of HSCs that incorporated a 72 hour pulse of BrdU (n=3 Vav1-Cre; Ifngr^fl/fl mice and n=4 control mice one experiment). i-o. The frequencies of HSCs (i), MPPs (j), HPCs (k), CMPs (l), MEPs (m), GMPs (n) and differentiated cells (o) in the bone marrow of control, Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl, and Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/fl mice (n=15 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=13 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/f mice and n=21 control mice in five independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using a matched samples two-way ANOVA followed by Sidak’s multiple comparisons adjustment (a), Mann-Whitney tests followed by Holm-Sidak’s multiple comparisons adjustments (b-g), a Student’s t-test (h), or one-way ANOVAs followed by Tukey’s multiple comparisons adjustments (i-o). All statistical tests were two-sided. Source numerical data are available in the source data files.

Extended Data Fig. 6. Adiponectin receptors promote HSC function by reducing TNF levels (related to Fig. 5).

a. Bone marrow (one femur and one tibia) and spleen cellularity in Tnf deficient and control mice (n=5 Tnf^−/− mice and n=7 control mice in four independent experiments). b-d. The frequencies of HSCs, MPPs, HPCs (b), CMPs, MEPs, GMPs (c) and differentiated T, B, myeloid, and erythroid cells (d), in the bone marrow of Tnf deficient and control mice (n=5 Tnf^−/− mice and n=7 control mice in four independent experiments). e-g. The frequencies of HSCs, MPPs, HPCs (e), CMPs, MEPs, GMPs (f) and differentiated cells (g), in the spleens of Tnf deficient and control mice (n=5 (e and g) or n=4 (f) Tnf^−/− mice and n=7 control mice in four independent experiments). h. The percentage of HSCs that incorporated a 72 hour pulse of BrdU (n=4 Tnf^−/− mice and n=9 control mice in two independent experiments) i-o. The frequencies of HSCs (i), MPPs (j), HPCs (k), CMPs (l), MEPs (m), GMPs (n) and differentiated cells (o) in the bone marrow of control, Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl, and Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice (n=12 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=11 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice and n=12 control mice in five independent experiments). p-q. ELISA of TNF (p) or INFg (q) in blood plasma from Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/fl (p), Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− (q), Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=6 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=10 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/f mice and n=17 control mice run in one ELISA per cytokine (p) and n=13 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=6 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice and n=12 control mice in one ELISA per cytokine (q)). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using a matched samples two-way ANOVA followed by Sidak’s multiple comparisons adjustment (a), Student’s t-tests (b-h) followed by Holm-Sidak’s multiple comparisons adjustments (b-g), one-way ANOVAs followed by Tukey’s multiple comparisons adjustments (i-o and q), a Welch’s one-way ANOVA followed by Dunnett’s T3 multiple comparisons adjustment (o), or a Kruskal-Wallis test followed by Dunn’s multiple comparisons adjustment (o and p). All statistical tests were two-sided. Source numerical data are available in the source data files.

Extended Data Fig. 7. Adiponectin receptor function reduces the frequencies of splenic myeloid cells in 5-6 and 19-24 month old mice (related to Fig. 6).

a-d. The frequencies of CMPs, MEPs, GMPs (a, c), differentiated T, B, myeloid, and erythroid cells (b, d) in the bone marrow from 5-6 month old (a-b) or 19-24 month old (c-d) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl or control mice (n=7 (a-b) or n=5 (c-d) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=10 (a-b) or n=6 (c-d) control mice in two independent experiments per age group). e-n. The frequencies of HSCs (e, j), MPPs (f, k), HPCs (g, l), CMPs, MEPs, GMPs (h, m), and differentiated T, B, myeloid, and erythroid cells (i, n) in the spleens of 5-6-month-old (e-i) and 19-24 month-old (j-n) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl or control mice (n=7 (e-i) or n=5 (j-n) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n= 10 (e-i) or n=6 (j-n) control mice in two independent experiments per age group). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using Mann-Whitney tests followed by Holm-Sidak’s multiple comparisons adjustments (a-n). All statistical tests were two-sided. Source numerical data are available in the source data files.

Figure 1.. Adiponectin receptors promote HSC quiescence.

a-c. RNAseq analysis of the adiponectin receptors Adipor1, Adipor2, and Tcad in hematopoietic stem and progenitor cell populations (n=3 mice per genotype in one experiment). d. The number of cells in one femur and one tibia, the thymus, and the spleen of 8–10-week-old Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and Adipor1^fl/fl; Adipor2^fl/fl control mice (n=12 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=13 control mice in six independent experiments for bone marrow and spleen, n=3 mice per genotype in one experiment for thymus). e. White blood cell, red blood cell, and platelet counts in the blood of 8–10-week-old Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=16 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=17 control mice in seven independent experiments). f-j. The frequencies of HSCs (f), MPPs (g), HPCs (h), CMPs, GMPs, MEPs (i) and differentiated T, B, myeloid, and erythroid cells (j) in bone marrow cells from one femur and one tibia of Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=12 (f-h) or n=13 (i-j) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=13 control mice in six independent experiments). k-o. The frequencies of HSCs (k), MPPs (l), HPCs (m), CMPs, GMPs, MEPs (n) and differentiated T, B, myeloid, and erythroid cells (o) in the spleens of Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=12 (k-m) or n=13 (n-o) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=13 control mice in six independent experiments). p-r. The percentage of HSCs (p), HPCs (q), and LK myeloid progenitors (r) that incorporated a 72 (p) or 2 (q, r) hour pulse of BrdU in Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (n=4 mice per genotype for HSCs; n=5 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=13 control mice for HPCs and LKs in two independent experiments for each cell population). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using Mann-Whitney tests followed by the Holm-Sidak’s multiple comparisons adjustment (d-o), a Student’s t-test (p), or matched samples two-way ANOVAs followed by Sidak’s multiple comparisons adjustments (q-r). All statistical tests were two-sided. Source numerical data are in the source data files.

Figure 2.. Adiponectin receptors are required to sustain HSC function.

a. Donor CD45+ cells, T, B, and myeloid cells from the blood of mice competitively transplanted with donor bone marrow cells from Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl (n=29 recipients) or control (n=34 recipients) mice (four donors per genotype in four independent experiments). b. Secondary recipients of bone marrow cells from the mice in (a) (n=29 recipients from eight Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl donors and n=32 recipients from seven control donors in seven independent experiments). c. Tertiary recipients of bone marrow cells from the mice in (b) (n=25 recipients from six Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl donors and n=20 recipients from four control donors in four independent experiments). d. The number of tertiary recipients in (c) with long term (>16 weeks) multilineage reconstitution by donor cells. e. The frequencies of donor HSCs, MPPs, HPCs, and LK cells in the bone marrow of tertiary recipients in (c) 16 weeks after transplantation (n=20 mice per genotype in three independent experiments). f. Tertiary recipients of bone marrow cells from the mice in Extended Data Fig. 3e (n=8 recipients from four Mx1-Cre; Adipor1^fl/fl; Adipor2^fl/fl donors and n=17 recipients from four control donors in four independent experiments). g. The number of tertiary recipients in (f) that had long term (LT; >16 weeks) multilineage reconstitution by donor T, B, and myeloid cells. h. The frequencies of donor HSCs, MPPs, HPCs, and LK cells in the bone marrow of tertiary recipients in (f) 16 weeks after transplantation (n=8 Mx1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=12 control mice in two independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed with Mann-Whitney tests (a-c, e, f, and h) followed by Holm-Sidak’s multiple comparisons adjustments (e and h), or a Fischer’s exact test (d and g). All statistical tests were two-sided. Source numerical data are in the source data files.

Figure 3.. Adiponectin receptors suppress the expression of inflammatory cytokines in bone marrow lymphocytes.

a. Gene set enrichment analysis showing significant enrichment of genes (FDR<0.05 and NES>1.5) related to interferon signaling and cell cycle progression in Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl as compared to control HSCs (FDR: false discovery rate; GO: gene ontology; NES: normalized enrichment score). b. ELISA analysis of inflammatory cytokines in blood plasma from Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and control mice (blood plasma from 30 (IFNγ and TNF), 33 (IL6), 22 (IFNα), 17 (IFNβ), and 30 (Il1β) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and 32 (IFNγ), 28 (TNF), 22 (IL6 and IFNα), 17 (IFNβ), and 24 (Il1β) control mice in two independent ELISAs per cytokine). c. qRT-PCR analysis of Ifng transcript levels in sorted bone marrow cells. Data are normalized to transcript levels in wildtype whole bone marrow (WBM) cells (three Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and four control mice in one experiment). d-f. qRT-PCR analysis of Ifng transcripts in sorted CD4+ T cells (d), CD8+ T cells (e), or NK cells (f) from the bone marrow (three Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and four control mice in one experiment). g. Donor CD45+ cells, T, B, and myeloid cells from the blood of mice competitively transplanted with Mx1-Cre; Adipor1^fl/fl; Adipor2^fl/fl or control bone marrow cells and then treated with pIpC to delete adiponectin receptors 6 weeks after transplantation (n=21 Mx1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and n=19 control recipients with five donors per genotype in five independent experiments). h. Secondary recipients of bone marrow cells from the primary recipients in (g) (n=7 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and n=17 control recipients from four Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and three control donors in three independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed with Mann-Whitney tests (b, g, and h) followed by Holm-Sidak’s multiple comparisons adjustments (b), or Student’s t-tests (c) or Welch’s t-tests (d-f) followed by Holm-Sidak’s multiple comparisons adjustments (c-f). All statistical tests were two-sided. Source numerical data are available in the source data files.

Figure 4.. Adiponectin receptors promote HSC function by reducing IFNγ signaling.

a. The number of cells in one femur and one tibia or in the spleen of Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl, Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/f or control mice (n=12 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=10 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/f mice and n=20 control mice in five independent experiments). b-h. The frequencies of HSCs (b), MPPs (c), HPCs (d), CMPs (e), MEPs (f), GMPs (g), or differentiated T, B, myeloid, and erythroid cells (h) in the spleen of Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl, or Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/fl or control mice (n=15 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=12 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/f mice and n=20 (b-d) or n=21(e-h) control mice in five independent experiments). i. The percentage of HSCs that incorporated a 72 hour pulse of BrdU (n=3 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=7 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/f mice and n=8 control mice in two independent experiments). j. Donor CD45+, T, B, and myeloid cells from the blood of mice competitively transplanted with Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl (n=19 recipients from five donors), Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Ifngr^fl/fl (n=18 recipients from five donors), or control donor bone marrow cells (n=19 recipients from four donors) in four independent experiments. All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using one-way ANOVAs followed by Tukey’s multiple comparisons adjustments (a, d-f, h, and i), Kruskal-Wallis tests followed by Dunn’s multiple comparisons adjustments (b, g, and h), Welch’s one-way ANOVAs followed by Dunnett’s T3 multiple comparisons adjustments (c and h), or nparLD tests followed by FDR multiple comparisons adjustments (j). All statistical tests were two-sided. Source numerical data are available in the source data files.

Figure 5.. Adiponectin receptors promote HSC function by reducing TNF signaling.

a. The number of cells in one femur and one tibia or in the spleen of control, Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl, and Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice (n=12 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=11 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice and n=12 control mice in five independent experiments). b-h. The frequencies of HSCs (b), MPPs (c), HPCs (d), CMPs (e), MEPs (f), GMPs (g), and differentiated T, B, myeloid, and erythroid cells (h) in the spleen of control, Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl, and Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice (n=12 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=11 (b-d and h) or n=10 (e-g) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice and n=12 (b-d and h) or n=11(e-g) control mice in five independent experiments). i. The percentage of HSCs that incorporated a 72-hour pulse of BrdU (n=7 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice, n=4 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− mice and n=5 control mice in two independent experiments). j. Donor CD45+, T, B, and myeloid from the blood of mice competitively transplanted with Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl (n=15 recipients), Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl; Tnf^−/− (n=18 recipients) or control (n=17 recipients) donor bone marrow cells (four donors per genotype in four independent experiments). k. Donor CD45+ cells, T, B, and myeloid cells from the blood of mice competitively transplanted with bone marrow cells from 8–10 week-old septic Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl or septic control donor mice and the number of mice with long term multilineage reconstitution (LTMLR) by donor cells (n=3 donors per genotype transplanted into 15 recipients per genotype in three independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using Welch’s one-way ANOVAs followed by Dunnett’s T3 multiple comparisons adjustments (a and h), one-way ANOVAs followed by Tukey’s multiple comparisons adjustments (b-d and f-i), Kruskal-Wallis tests followed by Dunn’s multiple comparisons adjustments (e), nparLD tests followed by FDR multiple comparisons adjustments (j), or Mann-Whitney tests (k). All statistical tests were two-sided. Source numerical data are available in the source data files.

Figure 6.. Adiponectin receptors are required to sustain HSCs during aging.

a-b. The number of cells in one femur and one tibia or in the spleen of 5–6-month-old (a) or 19–24-month-old (b) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and Adipor1^fl/fl; Adipor2^fl/fl control mice (n=7 (a) or n=5 (b) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=10 (a) or n=6 (b) control mice in two independent experiments for each age group). c-d. The frequencies of bone marrow HSCs in 5–6-month-old (c) and 19–24-month-old (d) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and Adipor1^fl/fl; Adipor2^fl/fl control mice (n=7 (c) or n=5 (d) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=10 (c) or n=6 (d) control mice in two independent experiments for each age group). e. The percentage of HSCs from 19–24 month old mice that incorporated a 72-hour pulse of BrdU (n=6 Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=5 control mice in two independent experiments). f-i. The frequencies of MPPs (f, h) and HPCs (g, i) in the bone marrow from 5–6 month old (f-g) or 19–24 month old (h-i) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl and Adipor1^fl/fl; Adipor2^fl/fl control mice (n=7 (f-g) or n=5 (h-i) Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl mice and n=10 (f-g) or n=6 (h-i) control mice in two independent experiments for each age group). j. Donor cells from the blood of mice competitively transplanted with bone marrow cells from 5–6 month-old Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl (n=11 recipients) or control (n=11 recipients) mice (three donors per genotype in three independent experiments). k. Donor cells from the blood of mice competitively transplanted with bone marrow cells from 15 month old Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl (n=15 recipients) or control (n=14 recipients) mice (three donors per genotype in three independent experiments). l. Secondary recipients of bone marrow cells from the primary recipients in (k) (n=17 recipients from four Vav1-Cre; Adipor1^fl/fl; Adipor2^fl/fl donors and n=14 recipients from three control donors in three independent experiments). All data represent mean ± standard deviation and each dot reflects a different mouse. Statistical significance was assessed using matched samples two-way ANOVAs followed by Sidak’s multiple comparisons adjustments (a-b), Mann-Whitney tests (c-d, and f-l) followed by Holm-Sidak’s multiple comparisons adjustments (c-d and f-i), or a Student’s t-test (e). All statistical tests were two-sided. Source numerical data are available in the source data files.