Haematopoietic stem cell number is not solely defined by niche availability

Abstract

Haematopoietic stem cells (HSCs) reside in specialized microenvironments, referred to as niches, and the classical model suggests that HSC numbers are predominantly determined by the niche size^1-5. However, the vast excess of niche cells relative to HSCs challenges this perspective. To rigorously define the role of niche size in regulating HSC numbers, we developed a femur-transplantation system, enabling us to increase available HSC niches. Notably, the addition of niches did not alter the total HSC numbers in the body, suggesting the presence of a systemic mechanism that limits HSC numbers. Additionally, HSC numbers in transplanted wild-type femurs did not exceed physiological levels when HSCs were mobilized from defective endogenous niches to the periphery, indicating that HSC numbers are constrained at the local level as well. The notion of dual restrictions at systemic and local levels was further supported by other experimental approaches, including parabiosis and non-conditioned transfer of HSCs after bone transplantation. Moreover, we found that thrombopoietin has a pivotal role in determining the total number of HSCs in the body, even in the context of increased niche availability. Our study redefines key principles underlying HSC number regulation, providing insights into this critical biological process.

Fig. 1. Normal functions of HSCs and their niches in the bone transplantation model.

a, Experimental strategy and analyses for the transplantation of WT femurs into WT mice, G-CSF administration and HSC transplantation (HSCT). s.c., subcutaneous. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. b, BM cellularity per host femur and graft. n = 8 host femurs and 8 grafts from 8 G-CSF-administered host mice. c, HSC numbers per host femur and graft. n = 6 host femurs and 6 grafts from 6 vehicle-administered hosts, and 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. d, The number of differentiated haematopoietic cells per host femur and graft. n = 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. e,f, The frequency of ECs (e) and MSCs (f) in the host femurs and the grafts. n = 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. g,h, The number of ECs (g) and MSCs (h) per host femur and graft. n = 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. i, Inflammatory cytokine (IL-1β, IL-6 and TNF) levels in BMEF of the host femurs and grafts measured by enzyme-linked immunosorbent assay (ELISA). n = 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. j, Quantification of mRNA levels of the indicated HSC niche factors in MSCs from the host femurs and grafts. n = 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. k, The frequency of quiescent (G0) and proliferating (non-G0) cells in HSCs from the host femurs and grafts. n = 8 host femurs and 8 grafts from 8 G-CSF-administered hosts. l, White blood cell (WBC) chimerism (CD45.2) in recipient mice transplanted with HSCs (CD45.2) from unperturbed WT femurs, the G-CSF-administered host femurs or the grafts mixed with competitor BM cells (CD45.1). n = 10 mice per group. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-tests (b and d–k) or one-way analysis of variance (ANOVA; c and l). Source data

Fig. 2. Unchanged total HSC numbers in the body after increased niche size.

a, Schematic of the transplantation of six WT femurs into WT mice and analyses. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. b, Quantification of mRNA levels of the indicated inflammatory cytokines in WBCs from sham-operated mice and bone transplantation hosts. n = 7 and 8 mice, respectively. c,d, The number of BM cells (c) and MSCs (d) per host femur and graft. n = 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. e, CXCL12 and SCF levels in BMEF of the host femurs and the grafts measured by ELISA. n = 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. f, Representative flow cytometry plots of HSCs from the femurs of sham-operated mice, host femurs and grafts of bone transplantation hosts (left). Gates and percentages represent the frequency of the HSC population. Right, HSC frequency in the host femurs and the grafts. n = 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. g, HSC numbers per host femur and graft. n = 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. h,i, WBC chimerism (CD45.2) in recipient mice transplanted with BM cells (h) or HSCs (i) (CD45.2) from the indicated bones mixed with competitor BM cells (CD45.1). n = 10 mice per group. j,k, HSC numbers in the entire body of hosts (excluding grafts) (j), and the sum of HSCs in the hosts and the grafts (k). n = 7 and 8 mice, respectively. Data are mean ± s.e.m. Significance was assessed using two-tailed unpaired Student’s t-tests (b, j and k) and one-way ANOVA (c–i). Source data

Fig. 3. HSC numbers are limited locally.

a, The experimental strategy and analyses for localized irradiation to limbs at a dose of 20 Gy. b,c, The number of ECs (b) and MSCs (c) in the targeted bones (limbs) after localized irradiation. n = 6 mice per group. d, HSC numbers in the targeted bones (four limbs), non-targeted regions (skull, spine, rib cage, pelvis, spleen and liver) and the sum of HSCs after localized irradiation. n = 6 mice per group. e, Schematic of the transplantation of a single WT femur into Cdh2-creER;Cxcl12^fl/fl mice and analyses. i.p., intraperitoneal. f, HSC numbers per host femur and graft of the indicated genotypes. 8 femurs from 8 sham-operated mice, 8 host femurs and 8 grafts from 8 bone transplantation hosts in both Cxcl12^fl/fl and Cdh2-creER;Cxcl12^fl/fl groups. g–i, HSC numbers in the spleens (g) and in the entire body of hosts (excluding grafts) (h), and the sum of HSCs in the hosts and the grafts (i) of the indicated genotypes. n = 8 mice per group. Data are mean ± s.e.m. Significance was assessed using two-tailed unpaired Student’s t-tests (b–d) and one-way ANOVA (f–i). The diagrams in a and e were created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. Source data

Fig. 4. Homeostatic mechanism allowing for HSC replenishment.

a, Schematic of six WT femur transplantation into SCF-deficient mice. b, HSC numbers per host femur and graft of the indicated genotypes. n = 8 femurs from 8 sham-operated mice, 8 host femurs and 48 grafts from 8 hosts in both Kitl^fl/fl and Cdh2-creER;Kitl^fl/fl groups. c,d, HSC numbers in the entire body of hosts (c) and the sum of HSCs in the hosts and the grafts (d) of the indicated genotypes. n = 8 mice per group. e, Schematic of G-CSF administration, followed by splenectomy and BMT. f, HSC numbers in the blood on the last day of G-CSF injection. n = 12 and 6 mice, respectively. g, HSPC numbers in the blood at 1 week after G-CSF administration. n = 12 and 6 mice, respectively. h,i, The HSC frequency (h) and numbers (i) in the femurs of the indicated cohorts at 2 months after G-CSF administration. n = 6 mice per group. j, WBC chimerism (CD45.2) in recipient mice transplanted with BM cells (CD45.2) from the indicated cohorts mixed with competitor BM cells (CD45.1). n = 10 mice per group. k, HSC numbers in the entire body of the indicated cohorts at 2 months after G-CSF administration. n = 6 mice per group. l, Schematic of WT femur transplantation and HSPC transfer into SCF-deficient mice. m, HSC numbers per host femur and graft of the indicated genotypes and conditions. n = 8, 8, 8, 8, 8, 8, 48, 8 and 48 bones, respectively. n, The sum of HSC numbers in the hosts and the grafts of the indicated genotypes and conditions. n = 8 mice per group. Data are mean ± s.e.m. Significance was assessed using two-tailed unpaired Student’s t-tests (f and g) or one-way ANOVA (b–d, h–k, m and n). The diagrams in a, e and l were created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. Source data

Fig. 5. TPO determines the total HSC numbers in the body in the context of increased niche availability.

a, Experimental strategy for the transplantation of six WT femurs into Tpo-knockout mice. b, HSC numbers per host femur and graft of the indicated genotypes. n = 8 femurs from 8 sham-operated Tpo^+/+ mice, 14 femurs from 14 sham-operated Tpo^+/− mice, 9 femurs from 9 sham-operated Tpo^−/− mice, 8 host femurs and 48 grafts from 8 Tpo^+/+ hosts, 12 host femurs and 72 grafts from 12 Tpo^+/− hosts, and 8 host femurs and 48 grafts from 8 Tpo^−/− hosts. c,d, HSC numbers in the entire body of hosts (excluding grafts) (c) and the sum of HSCs in the hosts and the grafts (d) of the indicated genotypes. n = 8, 14, 9, 8, 12 and 8 mice, respectively. e, TPO levels in serum of WT and Tpo-Tg mice were measured using ELISA. n = 6 mice per group. f, Schematic of transplantation of six WT femurs into Tpo-Tg mice. g, HSC numbers per host femur and graft of the indicated genotypes. n = 8 femurs from 8 sham-operated mice, 8 host femurs and 48 grafts from 8 hosts in both WT and Tpo-Tg groups. h,i, HSC numbers in the entire body of hosts (excluding grafts) (h) and the sum of HSCs in the hosts and the grafts (i) of the indicated genotypes. n = 8 mice per group. j, Schematic of HSC numbers at local (x axis) and systemic (y axis) levels in each experimental approach. Cxcl12 conditional knockout (CKO), Cdh2-creER;Cxcl12^fl/fl; Kitl CKO, Cdh2-creER;Kitl^fl/fl; Tpo KO, Tpo^+/−, Tpo^−/−. Data are mean ± s.e.m. Significance was assessed using two-tailed unpaired Student’s t-tests (e) or one-way ANOVA (b–d and g–i). The diagrams in a and f were created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. Source data

Extended Data Fig. 1. While MSCs survive in the femoral grafts, virtually no haematopoietic cells are detected shortly after bone transplantation.

a, Experimental strategy to determine the number of MSCs, HSCs and differentiated haematopoietic cells in the grafts shortly after bone transplantation from WT to WT mice. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. b, c, The number of MSCs (b) and HSCs (c) in the grafts at the indicated timepoints after bone transplantation. n = 6, 7 grafts from 6, 7 host mice, respectively, at each timepoint. d, Relative number of MSCs and HSCs in the grafts compared with non-transplanted femurs at the indicated timepoints after bone transplantation. n = 6, 7 grafts from 6, 7 host mice, respectively, at each timepoint. e, The number of differentiated haematopoietic cells in the grafts at the indicated timepoints after bone transplantation. n = 8 grafts from 8 host mice at each timepoint. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-test (d) or one-way ANOVA (b, c, e). Source data

Extended Data Fig. 2. BM stroma regeneration and haematopoietic recovery in the femoral grafts.

a, Schematic of the transplantation of Nestin-GFP femurs into Nestin-GFP mice and analyses. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. b, Representative flow cytometry plots and the quantification of overlap of CD51⁺CD140α⁺ cells and Nestin-GFP⁺ cells in CD45^–TER-119^–CD31^– fraction of the host femurs and the grafts at 5 months after bone transplantation. 8 host femurs and 8 grafts from 8 host mice. c, Representative confocal z-stack projection montages of Nestin-GFP (green) host femur and graft stained for CD31⁺CD144⁺ (white) vasculature at 1 or 5 months after bone transplantation. Scale bars, 1000 µm; four independent experiments yielded similar results. d-f, The number of BM cells (d), MSCs (e) and HSCs (f) per host femur and graft at the indicated timepoints after bone transplantation. 8 host femurs and 8 grafts from 8 host mice at each timepoint. g, The number of differentiated haematopoietic cells per host femur and graft. 8 host femurs and 8 grafts from 8 host mice. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-test. Source data

Extended Data Fig. 3. Determination of the origin of the cells in the femoral grafts.

a, Experimental strategy to determine whether MSCs in the grafts are derived from the grafts. b, Left: representative flow cytometry plots of Nestin-GFP⁺ cells in CD45^–TER-119^–CD31^–CD51⁺CD140α⁺ MSCs isolated from the host femurs (top) and the grafts (bottom) in the experiment shown in a. Right: the frequency of the Nestin-GFP⁺ population within the CD45^–TER-119^–CD31^–CD51⁺CD140α⁺ fraction in the host femurs and the grafts. 6 host femurs and 6 grafts from 6 host mice. c, Experimental strategy to determine whether MSCs in the grafts are derived from the hosts. d, Left: representative flow cytometry plots of Nestin-GFP⁺ cells in CD45^–TER-119^–CD31^–CD51⁺CD140α⁺ MSCs isolated from the host femurs (top) and the grafts (bottom) in the experiment shown in c. Right: the frequency of the Nestin-GFP⁺ population within the CD45^–TER-119^–CD31^–CD51⁺CD140α⁺ fraction in the host femurs and the grafts. 6 host femurs and 6 grafts from 6 host mice. e, Experimental strategy to determine whether ECs in the grafts are derived from the grafts. f, Confocal z-stack projection of Cdh5-CreER; iTdTomato (red) graft transplanted to WT mice and stained for CD31⁺CD144⁺ (white) vasculature. Scale bar, 1000 µm; three independent experiments yielded similar results. g, Left: representative flow cytometry plots of TdTomato⁺ cells in CD45^–TER-119^–CD31⁺SCA-1^highCD62E^low AEC fraction from the host femurs (top left) and the grafts (bottom left) and in CD45^–TER-119^–CD31⁺SCA-1^lowCD62E^high SEC fraction from the host femurs (top right) and the grafts (bottom right) in the experiment shown in e. Right: the frequency of the TdTomato⁺ cells in AEC and SEC fractions in the host femurs and the grafts. 6 host femurs and 6 grafts from 6 host mice. h, Experimental strategy to determine whether ECs in the grafts are derived from the hosts. i, Confocal z-stack projection of WT graft transplanted to Cdh5-CreER; iTdTomato (red) mice and stained for CD31⁺CD144⁺ (white) vasculature. Scale bar, 1000 µm; three independent experiments yielded similar results. j, Left: representative flow cytometry plots of TdTomato⁺ cells in CD45^–TER-119^–CD31⁺SCA-1^highCD62E^low AEC fraction from the host femurs (top left) and the grafts (bottom left) and in CD45^–TER-119^–CD31⁺SCA-1^lowCD62E^high SEC fraction from the host femurs (top right) and the grafts (bottom right) in the experiment shown in h. Right: the frequency of the TdTomato⁺ cells in AEC and SEC fractions in the host femurs and the grafts. 6 host femurs and 6 grafts from 6 host mice. k, Schematic of the determination of the origin of haematopoietic cells in the grafts. l, The frequency of CD45.1⁺ and CD45.2⁺ cells in the whole BM cells, HSCs and differentiated haematopoietic cells in the host femurs and the grafts. 6 host femurs and 6 grafts from 6 host mice. Data are mean ± s.e.m. The diagrams in a, c, e, h and k were created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. Source data

Extended Data Fig. 4. Normal BM structure and HSC niche function in G-CSF-administered femoral grafts.

a, Schematic of the transplantation of Nestin-GFP femurs into Nestin-GFP mice, G-CSF administration and analyses. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. b, Representative confocal z-stack projection montages of G-CSF-administered Nestin-GFP (green) host femurs and grafts stained for CD31⁺CD144⁺ (white) vasculature. Green arrows indicate arterioles. Scale bars, 100 µm; four independent experiments yielded similar results. c, Vasculature density in the host femurs and the grafts, as assessed by quantification of CD31⁺CD144⁺ vascular area divided by total BM area. n = 30 and 34 projections in the host femurs and the grafts, respectively; 4 host femurs and 4 grafts from 4 host mice. d, Arteriolar segment length in the host femurs and the grafts, as assessed by quantification of the length of the Nestin-GFP⁺ signal covering CD31⁺CD144⁺ arterioles. n = 60 and 55 projections in the host femurs and the grafts, respectively; 4 host femurs and 4 grafts from 4 host mice. e, Nestin-GFP density in the host femurs and the grafts, as assessed by quantification of Nestin-GFP⁺ area divided by total BM area. n = 28 and 31 projections in the host femurs and the grafts, respectively; 4 host femurs and 4 grafts from 4 host mice. f, g, BM cellularity (f) and the frequency of Nestin-GFP⁺ MSCs (g) from the host femurs and the grafts, assessed by flow cytometry. 8 host femurs and 8 grafts from 8 host mice. h, IL-1β, IL-6 and TNF-α levels in BMEF of the host femurs and the grafts measured by ELISA at the indicated timepoints after bone transplantation in the experiment shown in Fig. 1a. 8 host femurs and 8 grafts from 8 G-CSF-administered host mice. i, CXCL12 and SCF levels in BMEF of the host femurs and the grafts measured by ELISA. 8 host femurs and 8 grafts from 8 G-CSF-administered host mice. Data are mean ± s.e.m. For box plots, the box spans from the 25th to 75th percentiles and the centre line is plotted at the median. Whiskers represent the minimum to maximum range. Significance was assessed using a two-tailed unpaired Student’s t-test. Source data

Extended Data Fig. 5. G-CSF-administered grafts harbour HSCs with normal functions.

a, Mean fluorescence intensity (MFI) of cKIT, CD150 and CD41 in HSCs from the host femurs and the grafts in the experiment shown in Fig. 1a. 8 host femurs and 8 grafts from 8 G-CSF-administered host mice. b, Quantification of mRNA levels of the indicated cell cycle regulators in HSCs from the host femurs and the grafts. 8 host femurs and 8 grafts from 8 G-CSF-administered host mice. c, Blood chimerism (CD45.2) in myeloid (CD11b⁺), B (B220⁺) and T (CD3ε⁺) cells of recipient mice transplanted with HSCs (CD45.2) from G-CSF-administered host femurs or grafts in competition with CD45.1⁺ BM cells at the indicated timepoints after primary HSCT in the experiment shown in Fig. 1a. n = 10 mice per group. d, e, BM chimerism in whole BM, myeloid, B, T cells (d) and HSCs (e) at 5 months after primary HSCT. n = 10 mice per group. f, Blood chimerism (CD45.2) in total WBC, myeloid, B and T cells of recipient mice at the indicated timepoints after secondary BMT. n = 10 mice per group. g, BM chimerism in whole BM, myeloid, B and T cells at 5 months after secondary BMT. n = 10 mice per group. Data are mean ± s.e.m. For box plots, the box spans from the 25th to 75th percentiles and the centre line is plotted at the median. Whiskers represent the minimum to maximum range. Significance was assessed using a two-tailed unpaired Student’s t-test. Source data

Extended Data Fig. 6. Characterization of HSCs and their niches in the host femurs and the grafts after six femur transplantation.

a, HSC distribution in the mouse BM, as assessed by flow cytometry. The number in parentheses indicates the number of bones examined per mouse. n = 10 mice. b, Schematic of transplanted sites of femurs in the experiment shown in Fig. 2a. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. c, d, The number of BM cells (c) and ECs (d) in the grafts by their transplanted site. n = 16 grafts per transplanted site. e, EC numbers per host femur and graft. 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. f, MSC numbers in the grafts by their transplanted site. n = 16 grafts per transplanted site. g, CXCL12 and SCF levels in BMEF of the grafts by their transplanted site. n = 16 grafts per transplanted site. h, HSC numbers in the grafts by their transplanted site. n = 16 grafts per transplanted site. i, MFI of cKIT and CD150 in HSCs from the grafts by their transplanted site. n = 16 grafts per transplanted site. j, MFI of cKIT and CD150 in HSCs from the host femurs and the grafts. 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. k, Diagram showing the results of six WT femur transplantation into WT mice. Data are mean ± s.e.m. For box plots, the box spans from the 25th to 75th percentiles and the centre line is plotted at the median. Whiskers represent the minimum to maximum range. Significance was assessed using one-way ANOVA. Source data

Extended Data Fig. 7. MPP numbers in the entire body do not alter after transplantation of six WT femurs.

a, The number of the indicated MPP subsets per host femur and graft in the experiment shown in Fig. 2a. 7 femurs from 7 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts. b, c, The number of the indicated MPP subsets in the entire body of hosts (excluding grafts) (b), and the sum of MPPs in the hosts and the grafts (c). n = 7, 8 mice, respectively. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-test (b, c) or one-way ANOVA (a). Source data

Extended Data Fig. 8. Total HSC numbers in the body remain unchanged when the size of the intact niche is increased, even in mice with impaired HSC retention in endogenous BM niches.

a, Left: representative flow cytometry plots of TdTomato⁺ cells in CD51^–CD140α^– cells and CD51⁺CD140α⁺ MSCs within CD45^–TER-119^–CD31^– fraction of Cdh2-CreER; iTdTomato mice. Right: quantification of overlap of CD51^–CD140α^–, CD51⁺CD140α⁺ and TdTomato⁺ cells in the CD45^–TER-119^–CD31^– fraction of Cdh2-CreER; iTdTomato mice. n = 4 mice. b, Left: representative flow cytometry plots of TdTomato⁺ cells in Nestin-GFP^– cells and Nestin-GFP⁺ MSCs within CD45^–TER-119^–CD31^– fraction of Cdh2-CreER; iTdTomato; Nestin-GFP mice. Right: Quantification of overlap of Nestin-GFP^–, Nestin-GFP⁺ and TdTomato⁺ cells in the CD45^–TER-119^–CD31^– fraction of Cdh2-CreER; iTdTomato; Nestin-GFP mice. n = 4 mice. c-e, HSC numbers in the femurs (c), blood (d) and spleens (e) of Cxcl12^fl/fl and Cdh2-CreER; Cxcl12^fl/fl mice. n = 7, 9 mice, respectively. f, Schematic of the transplantation of six WT femurs into Cdh2-CreER; Cxcl12^fl/fl mice and analyses. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. g, HSC numbers per host femur and graft of the indicated genotypes. 8 femurs from 8 sham-operated mice, 8 host femurs and 48 grafts from 8 bone transplantation hosts in both Cxcl12^fl/fl and Cdh2-CreER; Cxcl12^fl/fl groups. h-j, HSC numbers in the spleens (h), in the entire body of hosts (excluding grafts) (i) and the sum of HSCs in the hosts and the grafts (j) of the indicated genotypes. n = 8 mice per group. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-test (c-e) or one-way ANOVA (g-j). Source data

Extended Data Fig. 9. HSC numbers are restricted at both systemic and local levels.

a, b, The number of ECs (a) and MSCs (b) in the non-targeted bone (pelvis) after localized irradiation in the experiment shown in Fig. 3a. n = 6 mice per group. c, Quantification of mRNA levels of the indicated HSC niche factors in MSCs from the non-targeted bone (pelvis) after localized irradiation. n = 6 mice per group. d, CXCL12 and SCF levels in BMEF of the non-targeted bone (pelvis) measured by ELISA after localized irradiation. n = 6 mice per group. e, HSC numbers in the indicated bones and the spleens after localized irradiation. The number in parentheses indicates the number of bones examined per mouse. n = 6 mice per group. f, Schematic of parabiosis experimental design. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. g, Percent partner-chimerism in parabionts. n.a.: not available due to expression of the same CD45 isoform. n = 8 mice per group. h, i, HSC numbers per femur (h) and spleen (i) from mice of the indicated genotypes. 8 mice and 8 partners from 8 parabionts per group. j, HSC numbers in the entire bodies per parabiont of the indicated genotypes. n = 8 parabionts per group. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-test (a-e) or one-way ANOVA (h-j). Source data

Extended Data Fig. 10. HSC numbers are not upregulated when they are reduced in specific settings.

a, HSC numbers per femur of Kitl^fl/fl and Cdh2-CreER; Kitl^fl/fl mice. n = 8 mice per group. b, HSPC numbers in the blood of Kitl^fl/fl and Cdh2-CreER; Kitl^fl/fl mice. n = 8 mice per group. c, HSC numbers in the spleens of Kitl^fl/fl and Cdh2-CreER; Kitl^fl/fl mice. n = 8 mice per group.d, Experimental strategy to determine HSC numbers in the femurs and the spleens shortly after G-CSF administration. The diagram was created using BioRender. Takeishi, S. (2025) https://BioRender.com/9d3nv16. e, f, HSC numbers per femur (e) and spleen (f) at 7 days after vehicle or G-CSF administration in the experiment shown in d. n = 6 mice per group. g-i, The number of BM cells (g), ECs (h) and MSCs (i) per femur of the indicated cohorts at 2 months after vehicle or G-CSF administration in the experiment shown in Fig. 4e. n = 6 mice per group. j, Quantification of Kitl mRNA levels in MSCs from the indicated cohorts at 2 months after vehicle or G-CSF administration. n = 6 mice per group. k, SCF levels in BMEF of the femurs from the indicated cohorts measured by ELISA at 2 months after vehicle or G-CSF administration. n = 6 mice per group. l, Quantification of mRNA levels of niche factors in MSCs from the indicated cohorts at 2 months after vehicle or G-CSF administration. n = 6 mice per group. Data are mean ± s.e.m. Significance was assessed using a two-tailed unpaired Student’s t-test (a-c, e, f) or one-way ANOVA (g-l). Source data