Deletion of Tet2 in mice leads to dysregulated hematopoietic stem cells and subsequent development of myeloid malignancies

Zhe Li, Xiaoqiang Cai, Chen-Leng Cai, Jiapeng Wang, Wenyong Zhang, Bruce E. Petersen, Feng-Chun Yang and Mingjiang Xu

Data supplements

Article Figures & Data


  • Figure 1

    Generation of Tet2:nlacZ/nGFP knock-in mice and evaluation of the levels of GFP (Tet2) expression in different hematopoietic cell populations. (A) A nlacZ/nGFP-FRTNeoFRT cassette was introduced into 6bp upstream of Tet2 start codon (exon 3). (B) Southern blot of ES cell DNA digested with ScaI (S) and hybridized with a genomic fragment external to the 5′ arm displayed a wild-type (WT) band of 6.3 kb and a recombinant band of 4.0 kb, and ES cell DNA digested with EcoRI (E) and hybridized with a probe external to the 3′ arm displayed a WT band of 10.3 kb and a recombinant band of 8.9 kb. Square bars indicate exons. (C) BM cells from 6- to 8-week-old heterozygous Tet2:nGFP mice were separated into GFPhi and GFPlo/− cells and Tet2 expression levels were measured by quantitative real-time PCR. (D) GFP (Tet2) expression levels in various hematopoietic cell populations of BM cells from a representative 7-week-old heterozygous Tet2:GFP mice.

  • Figure 2

    Analyses of Tet gene expression and 5-hmC levels in BM cells of WT, Tet2+/−, and Tet2−/− mice. (A) Analysis of the mRNA expression levels of each Tet genes in BM cells of 6- to 7-week-old WT (n = 4), Tet2+/− (n = 6) and Tet2−/− (n = 6) mice by quantitative real-time PCR. The relative mRNA expression of Tet2, Tet1, and Tet3 was determined using ß-actin as internal calibrator. The mRNA expression levels are reported as relative expression units to the respective Tet expression in WT mice. (B-D) Genomic DNA was extracted from BM cells of 6- to 7-week-old WT (n = 3), Tet2+/− (n = 4), and Tet2−/− (n = 4) mice and blotted onto nitrocellulose membrane after 2-fold serial dilution. 5-hmC (B) and 5-mC (C) levels were detected with an anti–5-hmC (ActiveMotif, #39791) or 5-mC (Calbiochem; NA#81) antibody. Methylene blue staining was performed to ensure equal spotting of total DNA on the membranes. Quantification of the signal density (D) was shown. *P < .05, **P < .01 ***P < .001, ****P < .0001 (E-H) Tet2−/− mice developed a phenotype resembling characteristics of CMML as early as 2-4 months of age. WT (n = 18), Tet2+/− (n = 24), and Tet2−/− (n = 28) mice were killed at 2-4 months of age and were analyzed for PB WBC (E), monocyte (MO, F), and RBC (G) counts. (H) Representative May-Giemsa stained PB smears prepared from WT, Tet2+/−, and Tet2−/− mice at 3-4 months of age. Black arrows indicate monocytes, and red arrows, indicate neutrophils.

  • Figure 3

    Tet2−/− mice evolved to lethal myeloid malignancies. (A) Kaplan-Meier survival curve of WT (n = 48), Tet2+/− (n = 66), and Tet2−/− (n = 62) mice up to 1 year of age. (B) Appearance of a representative moribund Tet2−/− mouse (2A48) with distended abdomens (i) and pale footpads (ii), the gross morphology of spleen (iii), and liver (iv) of this moribund Tet2−/− mice are drastically different from WT control. (C) Spleen and liver weights of moribund/deceased Tet2−/− (n = 18) and Tet2+/− (n = 5) mice as well as age matched WT controls. (D) H&E staining of paraffin-embedded sections of spleen, liver, and femurs from representative deceased/moribund Tet2−/− (2A116, 2A57, 2A45, 2A19) and Tet2+/− (2A102) mice. (E) May-Giemsa–stained BM cytospin-preparations from representative deceased/moribund Tet2−/− (2A116, 2A45, 2A19) or Tet2+/− (2A102) mice. (F) Lineage distribution of BM and spleen cells of representative deceased/moribund Tet2−/− (2A116, 2A57, 2A45) and Tet2+/− (2A102) mice. (G) Flow cytometric analysis of monocytic lineages in representative deceased/moribund Tet2−/− (3G53) and Tet2+/− (2A102) mice. (H) Flow cytometric analysis of LSK (LinSca-1+Kit+) and LK (LinSca-1Kit+) cell population in representative deceased/moribund Tet2−/− (2A19 and 2A116) and Tet2−/− (2A102) mice. The numbers indicate the percentages of cells in each cell population. ***P < .001.

  • Figure 4

    Deletion of Tet2 alters HSC compartments in mice before their development of myeloid malignancies as well as HSC proliferation and differentiation potential in vitro. (A) Flow cytometric analysis of LSK (LinSca-1+Kit+) HSC and LK (LinSca-1Kit+) progenitor cell population in a representative preleukemic 6- to 7-week-old Tet2−/− mice. (B) The percentage of LSK cells within the Lin cell populations in the BM of preleukemic Tet2−/−, Tet2+/−, and WT control mice (average ± SD of 4-8 animals). (C-E) The proliferation and differentiation potential of various genotypes of LSK cells were examined by culturing 500 LSK cells in the presence of 4 growth factors and assaying their ability to generate total cells (C) and CFCs (D) after 7 days of culture. CFCs in each culture were evaluated by colony assay of a fraction of the cultures. Cytospin preparations of the progenies generated from each genotype of LSK cells after 7 days of culture were stained with May-Giemsa, a 200-cell differential was performed (E). Representative data from 2 separate experiments are shown. ***P < .001.

  • Figure 5

    Deletion of Tet2 promoted the proliferative capacity of BM cells in vivo and Tet2 deletion induced-phenotype is cell-autonomous. Recipient mice receiving 1 × 106 BM cells each from B6SJL mice (CD45.1) and 6- to 7-week-old WT, Tet2+/− or Tet2−/−, mice (CD45.2) were killed 2 or 6 months after transplantation. (A) The ratio of CD45.2+ versus CD45.1+ BM cells in each recipient mice were examined at 2 or 6 months after transplantation (mean ± SD of 3-7 animals). (B-C) The lineage distribution within the CD45.2+ cells in the BM of the recipient mice 6 months after receiving WT, Tet2+/−, or Tet2−/− BM cells (B). Flow cytometric analysis of representative recipient mice 6 months after receiving WT or Tet2−/− BM cells (C). Histogram shows the percentage of CD45.2+ cells in the BM. (D-E) blood smear (May-Giemsa staining, D) and spleen weight/size (E) of representative mice receiving WT or Tet2−/− (arrows indicate monocytes). BM cells are shown.


  • Table 1

    Increased colony forming capacity of BM or spleen cells from Tet2−/− mice

    WT Control (n = 6)BM32 ± 22 ± 11 ± 135 ± 5
    SP10 ± 21 ± 1011 ± 3
    Tet2−/− (2-4 months; n = 8)BM52 ± 12*3 ± 23 ± 158 ± 13*
    SP64 ± 8**4 ± 2*0 ± 168 ± 10**
    Tet2−/− (erythroid infiltration; n = 6)BM1 ± 1**12 ± 8**013 ± 7
    SP6 ± 222 ± 10**028 ± 14
    Tet2−/− or Tet2−/− myeloid infiltration (n = 5)BM48 ± 10**02 ± 150 ± 11**
    SP26 ± 7**0 ± 1027 ± 8**
    • In vitro hematopoietic colony-forming assays were performed with BM (1 × 104) and spleen (SP; 1 × 105) cells from indicated genotypes of mice in the presence of mSCF, mIL-3, IL-6, and EPO. Colonies were enumerated on day 8 of culture. Data are presented as mean ± SEM.

    • * P < .05,

    • ** P < .01.

  • Table 2

    Diagnosis and subclassification of the myeloid malignancies in moribund/deceased Tet2−/− and Tet2+/− mice up to 1 year of age

    GenotypeSurvival, dFrequencyNecropsy and other findingsBlasts*, %TransplantabilityDiagnosis and subclasification
    Tet2−/−92-19612/21Hepatosplenomegaly anemia< 20%NoMDS with erythroid predominance
    Tet2−/−208-2984/21Hepatosplenomegaly monocytosis< 20%NDCMML
    Tet2−/−378-3604/21Hepatosplenomegaly neutrophilia< 20%YesMPD-like Myeloid leukemia
    Tet2−/−2061/21Hepatosplenomegaly neutrophilia> 20%YesMyeloid leukemia with maturation
    Tet2+/−316-3652/5Hepatosplenomegaly neutrophilia (1 had myeloid sarcoma)< 20%YesMPD-like Myeloid leukemia
    Tet2+/−238-3653/5Splenomegaly monocytosis< 20%NDCMML
    • ND indicates not done.

    • * Percentage of blasts in the BM.