CBFB-MYH11 hinders early T-cell development and induces massive cell death in the thymus

Abstract

Recent studies suggest that the chromosome 16 inversion, associated with acute myeloid leukemia M4Eo, takes place in hematopoietic stem cells. If this is the case, it is of interest to know the effects of the resulting fusion gene, CBFB-MYH11, on other lineages. Here we studied T-cell development in mice expressing Cbfb-MYH11 and compared them with mice compound-heterozygous for a Cbfb null and a hypomorphic GFP knock-in allele (Cbfb(-/GFP)), which had severe Cbfb deficiency. We found a differentiation block at the DN1 stage of thymocyte development in Cbfb-MYH11 knock-in chimeras. In a conditional knock-in model in which Cbfb-MYH11 expression was activated by Lck-Cre, there was a 10-fold reduction in thymocyte numbers in adult thymus, resulting mainly from impaired survival of CD4+CD8+ thymocytes. Although Cbfb-MYH11 derepressed CD4 expression efficiently in reporter assays, such derepression was less pronounced in vivo. On the other hand, CD4 expression was derepressed and thymocyte development was blocked at DN1 and DN2 stages in E17.5 Cbfb(-/GFP) thymus, with a 20-fold reduction of total thymocyte numbers. Our data suggest that Cbfb-MYH11 suppressed Cbfb in several stages of T-cell development and provide a mechanism for CBFB-MYH11 association with myeloid but not lymphoid leukemia.

Figure 1

Lck-Cre–mediated conditional Cbfb-MYH11 expression in adult thymus. (A) The Cbfb^56M allele contains an insertion in intron 4 that includes exons 5 and 6 of Cbfb flanked by 2 lox sites. It also contains the knocked-in MYH11 cDNA in exon 5. Cre-mediated excision of the exons 5/6 cassette led to the restoration of the knock-in Cbfb-MYH11 fusion (Cbfb^56M excised allele). PCR primers used for quantitative PCR to determine the efficiency of Cre-mediated deletion were: e5 and e6 primers for the Cbfb^56M allele; neo primers for both the Cbfb^56M and the Cbfb^56M excised alleles. (B) Total thymocyte numbers from Tg(Lck-Cre) and Tg(Lck-Cre)Cbfb^+/56M embryos at E17.5 (n = 4 for each genotype) and E18 (n = 3 for Tg(Lck-Cre) and n = 6 for Tg(Lck-Cre)Cbfb^+/56M). (C) Representative contour plots of cell surface marker expression of thymocytes from Tg(Lck-Cre)/Cbfb^+/56M and littermate control Tg(Lck-Cre) embryos at E18.0. The cells were stained with CD4, CD8, CD44, and CD25. The top panel shows CD4 and CD8 distribution, and the bottom panel shows CD44 and CD25 staining of DN cells.

Figure 2

T-cell developmental defects in the adult Tg(Lck-Cre)/Cbfb^+/56M mice. (A) Tg(Lck-Cre)/Cbfb^+/56M mice had smaller thymi (right) and their thymic architecture appeared to be homogeneous, as compared to the control mice (left). C indicates cortex; M: medulla. Left panel (gross view): pictures were taken with an Olympus SZ-40 (Tokyo, Japan) dissecting microscope camera with zoom setting at 2.5×. For the panel on the right, the 2 pictures on the left were taken with an Olympus SZ-40 dissecting microscope camera with zoom setting at 4×; the 2 pictures on the right were taken with a Nikon ECLIPSE E800 (Tokyo, Japan) microscope at 100× (10×/0.45 NA objective and 10× eyepiece). (B) Total and DP thymocyte numbers from adult thymi are graphed. (C) Numbers of DN, CD4⁺ SP, and CD8⁺ SP thymocytes are graphed. For panels B and C: □, littermate controls (n = 9); ■, Tg(Lck-Cre)/Cbfb^+/56M mice (n = 10). Statistically significant P values (Student t test) are indicated. (D) Representative contour plots of cell surface marker expression of thymocytes from Tg(Lck-Cre)/Cbfb^+/56M and littermate control mice. The cells were stained with CD4, CD8, CD44, and CD25. The top panels show CD4 and CD8 distribution, and the bottom panels show CD44 and CD25 staining of DN cells. (E) Representative contour plots of CD4 and CD8 expression of spleen and peripheral blood cells from Tg(Lck-Cre)/Cbfb^+/56M and littermate control mice.

Figure 3

BrdU incorporation by thymocytes in the Tg(Lck-Cre)/Cbfb^+/56M mice. (A) Percentages of BrdU⁺ cells among total thymocytes, DP cells, and DN cells in the littermate control (□, n = 3) and the Tg(Lck-Cre)/Cbfb^+/56M mice (■, n = 6). *Only the difference in DP BrdU⁺ cells between the littermate control and the Tg(Lck-Cre)/Cbfb^+/56M mice reached statistical significance (P = .05). (B) Distribution of BrdU⁺ cells relative to CD4 and CD8 expression.

Figure 4

Increased apoptosis of thymocytes in the Tg(Lck-Cre)/Cbfb^+/56M mice. (A) Dot plots of annexin V and 7AAD staining in DN, DP, and CD4 and CD8 SP cells of the Tg(Lck-Cre), Tg(Lck-Cre)/Cbfb^+/56M, and Tg(Lck-Cre)/Cbfb^+/56M/Tg(Lck-hBcl-2) mice (all littermates). (B) Contour plots of cell surface marker expression on the thymocytes of mice with the genotypes as shown. The total thymocyte numbers for each genotype are also given.

Figure 5

Effects of Cbfb-MYH11 on CD4 expression in vivo and in vitro. (A) CD4 and CD8 distribution of TCRβ^hiHSA^lo thymocytes from Tg(Lck-Cre) and Tg(Lck-Cre)/Cbfb^+/56M mice. (B) CD4 expression in immature thymocytes. The left panel shows the histogram overlay of CD4 expression in TCRβ^loFSC^hiCD8^−/lo thymocytes from Tg(Lck-Cre) (n = 10, gray line) and Tg(Lck-Cre)/Cbfb^+/56M mice (n = 5, black line). The right panel shows the mean fluorescence intensity (MFI) of CD4 expression in TCRβ^loFSC^hiCD8^−/lo cells. (C) CD4 silencer reporter assay. The indicated cDNA constructs were transfected into a CD4⁻ Jurkat cell clone (D1.1) and CAT assays were performed. The CAT expression from the construct with no CD4 silencer was set at 100% and used to normalize the other results. No CD4 silencer: a construct with the CAT gene driven by CD4 enhancer/promoter. CD4 silencer p131-265: the above construct with 3 copies of the CD4 silencer core sequence. Runx1: Runx1 full-length cDNA. Cbfb: Cbfb full-length cDNA. Inv(16): CBFB-MYH11 cDNA.

Figure 6

Developmental defects of thymocytes in Cbfb-MYH11 chimeras. (A) Identification of Ly9.1⁺ and Ly9.1⁻ thymocytes from a Cbfb-MYH11 chimera. (B) Analysis of thymocyte development in the Cbfb-MYH11 chimeras by FACS. Top panels show CD4 and CD8 distribution of Ly9.1⁺ and Ly9.1⁻ cells. Bottom panels show CD44 and CD25 distribution of DN cells from Ly9.1⁺ and Ly9.1⁻ cells. (C) The annexin V and 7AAD stainings of Ly9.1⁺ and Ly9.1⁻ cells are displayed. (D) CD4 expression in immature thymocytes. The left panel shows a representative histogram overlay of CD4 expression in TCRβ^loFSC^hiCD8^−/lo thymocytes from Ly9.1⁺ and Ly9.1⁻ cells. The right panel shows the MFI of CD4 expression in TCRβ^loFSC^hiCD8^−/lo cells.

Figure 7

T-cell development alterations in Cbfb^−/GFP and Cbfb^GFP/GFP embryos. (A) Representative contour plots of cell surface marker expression of thymocytes from Cbfb^+/+, Cbfb^+/GFP, and Cbfb^−/GFP embryos at E17.5. The cells were stained with CD4, CD8, CD44, and CD25. The top panels show CD4 and CD8 distribution, and the bottom panels show CD44 and CD25 staining of DN cells. (B) CD4 expression in immature thymocytes from E17.5 embryos. The left panel shows the histogram overlay of CD4 expression in TCRβ^loFSC^hiCD8^−/lo thymocytes of wild-type (filled gray line), Cbfb^−/GFP, and Cbfb ^GFP/GFP mice (black line). The right panel shows the MFI of CD4 expression in TCRβ^loFSC^hiCD8^−/lo cells from wild-type (n = 7), Cbfb^−/GFP (n = 6), and Cbfb^GFP/GFP (n = 8). (C) Total thymocyte numbers from embryos at E17.5. The panel shows the means and SDs of wild-type (WT, n = 4), Cbfb^+/GFP (+/GFP, n = 8), Cbfb^+/−(+/−, n = 3), Cbfb^GFP/GFP (GFP/GFP, n = 6), and Cbfb^−/GFP (−/GFP, n = 6) embryos. (D) Summary of T-cell phenotypes of Cbfb^GFP/GFP, Cbfb^−/GFP, Cbfb-MYH11 chimera, and Tg(Lck-Cre)/Cbfb^+/56M mice. The top panel shows the mouse thymocyte developmental diagram. The crosses in the lower panel indicate block of differentiation at the corresponding stages. The arrows in the lower panel indicate increase or decrease in percentage of the population, or apoptosis; − indicates no change.