Fig 2: Regnase-1-deleted HSPCs show a phenotype and gene expression profile similar to leukemia. a Representative field of Wright-Giemsa-stained BM smears of 8-week-old control (fl/fl) and Regnase-1-KO (?/?) mice. The scale bar shows 30 µm. b Quantification of frequencies of blasts and other MNCs in PB on Wright-Giemsa-staining (n = 3 per group; 3 independent experiments). c Quantification of abnormal cells in the PB. d Peripheral white blood cell counts in blood samples from 8-week-old control (fl/fl) or Regnase-1-KO (?/?) mice (n = 3 per group). e Wright-Giemsa staining of CD34- HSCs. The scale bars represent 30 µm. f Scatter-plot representation of the transcriptional differences between control Reg1flox/flox (fl/fl) and Vav1-iCre; Regnase-1flox/flox(?/?) CD34- HSCs. The color indicates >2-fold differential gene expression (n = 3 per group and n = 2 per group; 2 independent experiments; the average of the two is shown). g Gene set enrichment analysis comparing CD34- HSCs from control (fl/fl) and Regnase-1-KO (?/?) mice with the association between genes upregulated following Regnase-1 deletion and Pu.1 deletion. h Human HSC signatures of AML compared with Regnase-1-deficient CD34- HSCs. The normalized enrichment scores (NES), P-value and q-value (FDR) are indicated on each plot. i Kaplan-Meier plot of AML patient survival data based on similarity of gene expression profiles of loss of Regnase-1. P values were calculated by the logrank test. j Regnase-1 mRNA expression (upper panel) and proliferation (lower panel) of THP1 and HL60 leukemic cells transfected with mouse Regnase-1 cDNA expression vector (n = 3 per group; 3 independent experiments). Error bars indicate mean ± SD. *p < 0.05; **p < 0.01, ***p < 0.005, log rank test in i, two-sided t-test in c, d, j

Fig 3: Loss of Regnase-1 expression results in accelerated CD34- HSC cell cycle progression. a Cell-cycle analysis of control (fl/fl) or Regnase-1-KO (?/?) BM CD34- HSC and CD34+ HSC populations using EdU/7AAD staining. Dot plots indicating the frequency of HSPCs in G0/G1 (EdU- 7AAD-), G2/M (EdU+), and S (EdU+ 7AAD+) phase of cell cycle (n = 3 per group; 3 independent experiments). b Quantification of HSPCs as shown in a. c Detection of proliferative cells in BM section of control (fl/fl) or Regnase-1-KO (?/?) mice by EdU labeling. Representative image showed EdU staining together with c-Kit and Nuclei (TOPRO3) counter-staining. Scale bars represent 50 µm. d Quantification of the ratio of EdU-positive proliferating cells in c-Kit-positive cells. e Representative cell cycle plots of CD34- HSC populations in BM from control (fl/fl) or Regnase-1-KO (?/?) mice using Hoechst 33342/pyronin Y staining (n = 3 per group; 3 independent experiments). f Representative data showing number and ratio of cells in the G0, G1, and S/G2/M phases as shown in e. g qRT-PCR analysis of cell cycle regulatory genes in CD34- HSCs from control (fl/fl) and Regnase-1-KO (?/?) mice (n = 3 per group). h Flow cytometric analysis of apoptotic populations in CD34- HSCs and CD34+ HSCs of control (fl/fl) or Regnase-1-KO (?/?) mice BM with a combination of annexin V and SYTO Green staining (n = 3 per group; 3 independent experiments). i Ratio of apoptotic cells (annexin V+ SYTO Green+). Data are the means ± SD. j Kaplan-Meier survival curve of control (fl/fl) and Regnase-1-KO (?/?) mice after weekly administration of 5-FU (150 mg/kg) (n = 7). P values were calculated by the logrank test. k Representative dot plot depicting frequency of phenotypically defined HSPCs at the indicated days following a single injection of 5-FU. l Absolute LSK cell numbers from control (fl/fl) and Regnase-1-KO (?/?) mice at the indicated days after 5-FU treatment (n = 3). m Expression Regnase-1 mRNA in LSK cells of control (fl/fl) mice at the indicated days after 5-FU treatment (n = 3). Error bars indicate mean ± SD. *p < 0.05; **p < 0.01, ***p < 0.005, two-sided t-test in b, d, f, g, i, m, l

Fig 4: The effect of Regnase-1 on HSPCs is mediated through Gata2 and Tal1. a Relative expression of Gata2, Tal1, and Bmi1 in CD34- HSCs infected with shRNA virus targeting Gata2, Tal1, Bmi1 or scrambled control (shCtrl). Bmi1 was used as a non-candidate negative control. Data are expressed as fold-change relative to control shRNA infection (n = 3 per group). b Colony-forming assay of Regnase-1-deficient CD34- HSCs infected with either an shCtrl, shGata2, or shTal1 shRNA virus (left), and control (fl/fl) or Regnase-1-KO (?/?) CD34- HSCs infected with control or shGata2+ shTal1 shRNA virus (right). G indicates CFU-granulocyte; GM CFU-granulocyte/monocyte; M CFU-monocyte; E CFU-erythroid; and mix mixed CFU-granulocytes, monocyte, erythroid, and megakaryocyte (n = 4 per group; 4 independent experiments). c Regnase-1-KO (?/?) CD34- HSCs infected with shCtrl or shGata2+ shTal1 shRNA virus were transplanted and analyzed by flow cytometry (n = 3 per group; 3 independent experiments). d–f Numbers and populations of LSK-HSCs or CD34- HSCs in total BM assessed by flow cytometry in c. Data represent the means ± SD. g Representative gross appearance of the spleen and mesenteric lymph nodes (MLNs) collected from bone marrow-transplanted mice as shown in c. The scale bars represent 1 cm. h Cell-cycle analysis of control BM CD34- HSCs in c by flow cytometry using EdU/7AAD staining. Dot plots indicate the frequency of CD34- HSCs in G0/G1 (EdU- 7AAD-), G2/M (EdU+), or S (EdU+ 7AAD+) phase of the cell cycle (n = 3 per group; 2 independent experiments). Error bars indicate mean ± SD. *p < 0.05; **p < 0.01, ***p < 0.005, two-sided t-test in a, d, e, g, h. Regnase-1 is known to mediate post-trasncriptional regulatory activity through degradation of target mRNAs. Here, the authors show that Regnase-1 regulates self-renewal of haematopoietic stem and progenitor cells through modulation of the stability of Gata2 and Tal1 mRNA

Fig 5: Regnase-1 deficiency leads to expansion of immature HSPCs. a Representative flow cytometric analysis of HSPC populations in the FL of E15.5 embryo, BM of 1-day-old neonate, BM of 2-week-old control (fl/fl) or Regnase-1-KO (?/?) mice (n = 3 per group). b The numbers of LSK-HSCs or CD34- HSCs in the total FL or BM cells assessed by flow cytometry. c Representative flow cytometric analysis of LSK cells for CD48 and CD150 expression of 8-week-old control (fl/fl) or Regnase-1-KO (?/?) (n = 3 per group). d Percentage of CD150+ CD48- cells, CD150+ CD48+ cells, and CD150- CD48+ cells of control (fl/fl) and Regnase-1-KO (?/?) mice. e Colony-forming ability of LSK-HSCs and CD34- HSC from control (fl/fl) or Regnase-1-KO (?/?) mice BM. G CFU-granulocyte; GM CFU-granulocyte/monocyte; M CFU-monocyte; E CFU-erythroid; and mix mixed CFU-granulocytes, monocyte, erythroid, and megakaryocyte. Data show the means ± SD (n = 5; 3 independent experiments). f Competitive repopulation assay of lethally irradiated (10 Gy) WT recipients (CD45.1) transplanted with BM of either control (fl/fl) or Regnase-1-KO (?/?) mice. Chimerism was analyzed in recipient mice 16 weeks after injection of 1 × 103 of CD34- HSCs from control or Regnase-1-KO mice (CD45.2) with 5 × 105 competitor BM cells (CD45.1). Donor-derived cell ratios (CD45.1-) in BM HSPC and CD34- HSC populations were determined by flow cytometry (n = 5 per group). g The number of LSK or CD34- LSK cells in the BM of mice 16 weeks after transplantation. h Hematopoietic engraftment of transplanted HSPCs for reconstitution of peripheral blood was analyzed by flow cytometry at 16 weeks after transplantation (n = 5 per group). i The contribution of transplanted donor-derived cells (CD45.1-) to B-cells (B220+), T-cells (CD3+), and myeloid cells (CD11b+) from control (fl/fl) or Regnase-1-KO (?/?) HSPCs are shown. Data summated from 3 independent experiments and an average of 6 recipient mice in each group with SD. j Gene ontology (GO) analysis of genes upregulated in the CD34-HSCs from Regnase-1-KO BM based on the GO biological process annotations. Error bars indicate mean ± SD. *p < 0.05; **p < 0.01, ***p < 0.005, two-sided t-test in b, d, g, i