Mad2 is required for optimal hematopoiesis: Mad2 associates with c-Kit in MO7e cells

Abstract

Mitotic arrest deficiency 2 (Mad2) is a component of mitotic spindle checkpoint proteins and is essential for accurate chromosome segregation. We investigated a role for Mad2 in hematopoiesis using Mad2-haploinsufficient (Mad2+/-) mice. Mad2+/- bone marrow (BM) and spleen manifested decreased absolute numbers and cycling status of immature, but not mature, hematopoietic progenitor cells. Mad2+/- BM granulocyte-macrophage colony-forming units (CFU-GMs) did not manifest synergistic proliferation in response to stem cell factor (SCF) plus GM-CSF. The percentage of annexin V+ cells was higher in Mad2+/- than Mad2+/+c-Kit+lin- BM after culture with SCF and GM-CSF. However, no significant difference in phosphorylation of extracellular signal-related kinase (Erk1/2) at Thr202/Tyr204 and Akt at Ser473 between Mad2+/- and Mad2+/+BM c-Kit+lin- cells was observed. Immunoprecipitation assays performed in human MO7e cells demonstrated physical association of c-Kit with Mad2. Moreover, stimulation with SCF plus GM-CSF led to dissociation of Mad2 from c-Kit. Confocal microscopy demonstrated that Mad2 colocalized with c-Kit in the cytoplasm of MO7e cells. These results suggest that Mad2 is involved in synergistic growth of immature hematopoietic progenitor cells in response to SCF plus GM-CSF, effects that may be mediated via physical association of Mad2 with c-Kit.

Figure 1

Mad2 protein expression in bone marrow cells of Mad2^+/− and Mad2^+/+ mice. Lineage-depleted (lin⁻) and -positive (lin⁺) cells from Mad2^+/+ and Mad2^+/− mice BM were obtained by magnetic separation. Total-cell lysates were dissolved in SDS-PAGE. Membrane was immunoblotted with anti-Mad2 and antiactin antibodies. +/+ indicates Mad2^+/+; +/−, Mad2^+/−.

Figure 2

Comparative analysis of hematopoietic progenitor cells in bone marrow and spleen of Mad2^+/− and Mad2^+/+ mice. Absolute numbers of immature hematopoietic progenitor cells (responsive to stimulation by multiple cytokines (5% [vol/vol] PWMSCM, 50 ng/mL rmuSCF, and 1 U/mL rhuEpo in vitro) in bone marrow (A) and spleen (B), and the percentage of BM progenitor cells in S phase (C). Absolute numbers of immature (responsive to 50 ng/mL rmuSCF and 10 ng/mL rmuGM-CSF) and mature (responsive to stimulation by 50 ng/mL rmuSCF or 10 ng/mL rmuGM-CSF) CFU-GM in BM (D). ■ and □ indicate Mad2^+/+ and Mad2^+/−, respectively. Data represent the average of a total of 5 mice of each phenotype from 2 independent experiments. Error bars represent 1 SD. *P < .05 (+/− vs +/+).

Figure 3

Absolute numbers of KSL cells and expression of c-Kit on Sca1⁺lin⁻ cells. Absolute numbers of KSL cells in BM. (A) Representative cytograms of forward scatter (FSC) versus side scatter (SSC), lineage marker (top panels), and Sca-1 versus c-Kit (bottom panels) for measurements of percentages of KSL cells in Mad2^+/+ and Mad2^+/− BM cells are shown. The percentage of c-Kit⁺Sca-1⁺ cells (R4) was analyzed on gated lin⁻ cells (M1 + R1). Numbers indicate the percentage of KSL cells. (B) Data show KSL cell numbers per femur. Error bars indicate mean ± SD of measurements of 6 mice each. (C) Representative histograms of lineage marker and Sca-1 (top panels) for gating the Sca-1⁺lin⁻ cell population, and c-Kit expression (bottom panels) are shown. c-Kit expression was analyzed on gated lin⁻ cells (M1) and Sca-1⁺ cells (M2). The shaded curves show staining with anti–c-Kit antibody; the open curves show the isotype control (bottom panels). Numbers indicate the mean fluorescence intensity (MFI) of c-Kit. (D) Expression level of c-Kit on Sca-1⁺lin⁻ cells in BM. Error bars indicate mean ± SD of measurements of 4 mice each. NS indicates not significant.

Figure 4

Cell death 6 days after culture with SCF plus GM-CSF. (A) c-Kit⁺lin⁻ bone marrow cells (2 × 10⁵) from wild-type and Mad2^+/− mice were cultured in a 12-well tissue-culture plate in the presence of SCF (50 ng/mL) plus GM-CSF (10 ng/mL) for 6 days. Representative histograms of Mad2^+/+ and Mad2^+/− BM cells are shown. Numbers in histograms correspond to the percentage of annexin V⁺ cells. (B) Data represent the average of triplicate samples, and error bars represent 1 SD from the mean for triplicate wells. Result shown is representative of 2 independent experiments. *P < .01 (+/− vs +/+).

Figure 5

Mad2 physically associates with c-Kit in MO7e cells. Total-cell lysates (TCL) were immunoprecipitated with either control IgG, anti-Mad2, or anti–c-Kit Abs, and then immunoblotted with anti–c-Kit Abs (A) or anti-Mad2 Abs (B). The results of 1 representative of 3 experiments are shown. IB indicates immunoblot; IP, immunoprecipitation.

Figure 6

Mad2 dissociates from c-Kit after stimulation with SCF plus GM-CSF in MO7e cells. (A) After serum starvation, cells were stimulated with 50 ng/mL SCF plus 2 ng/mL GM-CSF for 3 hours. (B) Cells were unstimulated or stimulated with 50 ng/mL SCF, 2 ng/mL GM-CSF, or SCF plus GM-CSF for 3 hours. Total-cell lysates (TCL) were separated into 2 tubes for immunoblotting and immunoprecipitation, respectively. Cell lysates were immunoprecipitated with control IgG or anti–c-Kit Abs, and then immunoblotted with anti–c-Kit Abs (left panel, top blot) or anti-Mad2 Abs (left panel, bottom blot). To compare the Mad2 proteins, separated cell lysates also were immunoblotted with anti-Mad2 Abs (right panel). The results of 1 representative of 2 experiments are shown. IB indicates immunoblot; IP, immunoprecipitation; −, no cytokine stimulation; S, SCF; G, GM-CSF; and S/G, SCF plus GM-CSF.

Figure 7

Mad2 colocalizes with c-Kit in cytoplasm in MO7e cells. Top (A-D) and bottom (E-H) panels show unstimulated and cytokine-stimulated MO7e cells, respectively. Images were acquired by confocal microscopy with an oil immersion lens (magnification, × 100). c-Kit is shown in green (A,E), Mad2 in red (B,F), and nucleus in blue (C,G). Merged images are shown in panels D and H. Arrows indicate representative colocalization of Mad2 with c-Kit.