Imaging hematopoietic precursor division in real time

Abstract

Stem cells are thought to balance self-renewal and differentiation through asymmetric and symmetric divisions, but whether such divisions occur during hematopoietic development remains unknown. Using a Notch reporter mouse, in which GFP acts as a sensor for differentiation, we image hematopoietic precursors and show that they undergo both symmetric and asymmetric divisions. In addition we show that the balance between these divisions is not hardwired but responsive to extrinsic and intrinsic cues. Precursors in a prodifferentiation environment preferentially divide asymmetrically, whereas those in a prorenewal environment primarily divide symmetrically. Oncoproteins can also influence division pattern: although BCR-ABL predominantly alters the rate of division and death, NUP98-HOXA9 promotes symmetric division, suggesting that distinct oncogenes subvert different aspects of cellular function. These studies establish a system for tracking division of hematopoietic precursors and show that the balance of symmetric and asymmetric division can be influenced by the microenvironment and subverted by oncogenes.

Figure 1. GFP⁺ Cells Are Phenotypically More Immature Than GFP⁻ Cells

(A–C) GFP⁺KLS cells from TNR mice are highly enriched for hematopoietic stem cells. Five-hundred GFP⁺KLS cells were transplanted into recipient mice (six per cohort), and peripheral blood was analyzed for donor derived chimerism ([A], left) by FACS at 15 weeks. (A) Right, two million bone marrow cells from primary transplanted mouse were transplanted to recipient mice (six mice per cohort), and total donor chimerism was analyzed at 15 weeks. (B and C) Multilineage repopulation was determined by analysis of peripheral blood with antibodies to distinct lineage markers. Primary transplantation results shown are an average of four independent experiments. Secondary transplantation results shown are an average of three independent experiments. (D) FACS analysis of GFP levels of GFP⁺KLSC cells sorted from TNR mice before (gray) and after (black) culture with stromal cells. (E–H) Expression of GFP, Lineage, and c-kit on GFP⁺KLSC cells before (E and F) and after (G and H) culture. Error bars represent standard deviation.

Figure 2. GFP⁺ Cells Are Functionally More Immature Than GFP⁻ Cells

(A) GFP⁺KLSC cells were cocultured with 7F2 cells. Subsequently, single-sorted GFP⁺ or GFP⁻ cells were cultured in methylcellulose in individual wells of a 96-well plate and colonies counted after 7 days. Average colony number per plate is shown. Results shown are an average from four independent experiments (p = 0.007). (B and C) GFP⁺KLSC cells were cocultured with 7F2 cells, and subsequently, 800 GFP⁺ or GFP⁻ cells were transplanted into lethally irradiated mice to assess their ability to form d12 colony-forming unit-spleen (CFU-S). Representative spleens from mice transplanted with GFP⁻ and GFP⁺ cells (B) and average colonies per spleen in each group are shown (C) (p < 0.0001). Results are representative of three different experiments with three or five mice per group. (D–F) GFP⁺KLSC cells were cocultured with 7F2 cells for 2 days and transplanted in limiting numbers to determine the relative transplantation efficiency of GFP⁺ and GFP⁻ cells. 40, 20, and 10 GFP⁺ or GFP⁻ cells were sorted and transplanted into sublethally irradiated Rag2^−/− IL2rγ^−/− mice. Average donor chimerism in mice is shown at 13–15 weeks after transplantation. (Results are representative of two independent experiments with three mice per cell dose for a total of 18 mice per experiment [p < 0.05].) (G) Comparative repopulation ability of postculture GFP⁺ cells to freshly isolated KLSC cells. Forty KLSC and 40 postculture GFP⁺ cells (also shown in [D]) were transplanted into recipient mice (three or six mice per cohort), and the average donor-derived chimerism was determined by analyzing the peripheral blood at 15 weeks. Data are representative of two independent experiments. Error bars represent standard deviation.

Figure 3. Time-Lapse Imaging of Hematopoietic Precursor Divisions

(A–C) Schematic shows three possible division patterns of hematopoietic precursors: symmetric commitment (A), symmetric renewal (B), and asymmetric division (C). (D–U) Representative still frames of GFP⁺KLSC cells undergoing distinct patterns of divisions (corresponding movies shown in Movies S1–S6). GFP⁺KLSC cells were infected with MSCV-IRES-CFP virus, KLS CFP⁺ GFP⁺ resorted, and plated on 7F2s for imaging. Upper frames show GFP images, and lower frames show CFP images. (D–I) Symmetric commitment division. A mother cell, shown in (D) (pixel intensity unit [PIU] = 179), divides to produce two daughters (E), which downregulate GFP ([F], PIU = 46, 42). CFP intensity of the daughters (G–I) remains similar to their mother. (J–O) Symmetric renewal division. A mother cell ([J], PIU = 189) divides to produce two daughters (K) that retain GFP levels equivalent (L, PIU = 183, 180) to that of their mother. CFP intensity of the daughters (M–O) remains similar to their mother. (P–U) Asymmetric cell division. A mother cell ([P], PIU = 220) divides to produce two daughters with equivalent GFP intensity ([Q], PIU = 210); subsequently, the GFP expression of one of the daughters is downregulated ([R], PIU = 71) relative to the mother and the other daughter (PIU = 237). CFP intensity of the daughters (S–U) remains similar to their mother. Bar graphs show GFP PIU of mother cell in first panel and PIU of daughters in last panel.

Figure 4. Numb Is Asymmetrically Distributed and Segregated during Hematopoietic Precursor Division

(A–D) GFP⁺KLSC cells were cultured overnight, treated with nocodazole, and subsequently stained to visualize Numb distribution. (A) Symmetric distribution of Numb in cells undergoing division (representative of 56% of dividing cells, n = 48). DAPI staining is shown in (B). (C) Asymmetric distribution of Numb in cells undergoing division (representative of 44% of dividing cells, n = 48). DAPI staining is shown in (D). (E and F) GFP⁺KLSC cells were cultured overnight, treated with nocodazole, and stained for Numb (red, left panels) and pan-cadherin (blue, right panels), n = 57 from three independent experiments. (G and H) GFP⁺KLSC cells were cultured on 7F2 cells and subsequently stained with antibodies to Numb (G) and GFP (H). (I and J) GFP⁺KLSC cells were infected for 48 hr with MSCV-Numb::CFP or MSCV-CFP as a control. After infection, cells were cocultured with 7F2 cells for 3 days and subsequently analyzed by FACS to determine GFP reporter expression in cells expressing either control (I) or Numb (I) vectors. (K–P) Increased repression of Notch reporter activity with increasing levels of ectopic Numb expression. GFP levels in cells that were either negative (K and L), low (M and N), or high (O and P) for control vector or Numb expression. Results are representative of two independent experiments. (Q) KTLS cells from wild-type mice were infected for 36 hr with MSCV-Numb::GFP or MSCV-GFP as a control. After infection, GFP⁺KLS cells were resorted, cocultured with 7F2 cells, and lineage marker expression analyzed. Average frequency of Lin⁻ cells over four independent experiments (p = 0.02). Error bars represent standard deviation.

Figure 5. Distinct Stromal Cells Differentially Influence the Balance of Asymmetric and Symmetric Division of Hematopoietic Precursors

(A) GFP⁺KLSC cells were sorted and plated on 7F2 or OP9 cells for 72 hr, and their phenotype was analyzed by FACS. (B) Rates of cell divisions of KLSC cells on 7F2 and OP9 were monitored by counting the number of times a cell divided in a given time period using time-lapse imaging. The division time was averaged from scoring 112 cells cultured in 7F2 and 104 cells cultured in OP9 cells (from three independent experiments). (C) Apoptosis rates of KLSC cells on 7F2 and OP9 were determined by analyzing by time-lapse microscopy the number of cells that underwent cell fragmentation postdivision. (D–F) Frequency of cells undergoing asymmetric, symmetric renewal or symmetric commitment division on 7F2 or OP9 cells scored over a period of 72 hr by time-lapse microscopy. Total numbers of cells tracked in 7F2 cultures were 112, and in OP9 cultures were 104. Data shown are an average of three independent experiments. The difference in the frequency of symmetric renewal and asymmetric divisions between OP9 and 7F2 was significant (p = 0.012 and p = 0.00019, respectively). Relative ratios of symmetric:asymmetric division on 7F2 or OP9 cells (F). Error bars represent standard deviation.

Figure 6. BCR-ABL Expression Does Not Influence the Balance between Asymmetric and Symmetric Division

GFP⁺KLSC cells were sorted from TNR mice and infected with viruses carrying control vector-IRES-CFP or BCR-ABL-IRES-CFP. After infection, KLS CFP⁺ GFP⁺ cells were resorted, plated on 7F2 cells, and imaged for 72 hr. (A) Rate of cell division of vector or BCR-ABL-infected KLS cells were monitored by counting the number of times a given cell divided in a 72 hr period. The division time was averaged from 78 and 89 cells respectively (n = 3). Frequency of divisions of BCR-ABL-infected KLSC cells was significantly higher than vector-infected KLSC cells (*p = 0.038). (B) Apoptosis rate of vector-infected or BCR-ABL-infected KLSC cells was determined as in Figure 5C (n = 3, p = 0.040). (C and D) Frequency of vector-infected or BCR-ABL-infected cells undergoing asymmetric, symmetric renewal or symmetric commitment division were scored over a period of 72 hr by time-lapse microscopy. Data shown are derived from 78 vector-infected and 89 BCR-ABL-infected cells, and is an average of three independent experiments. The difference in frequency of symmetric renewal divisions between vector and BCR-ABL is not significant (p = 0.392). Error bars represent standard deviation.

Figure 7. NUP98-HOXA9 Expression Alters the Balance between Asymmetric and Symmetric Division

(A) Rates of cell divisions of vector-infected or NUP98-HOXA9-infected KLSC cells on 7F2 were monitored by counting the number of times a cell divided in a period of 72 hr by time lapse microscopy. The division time was averaged from 78 and 63 cells, respectively (n = 3). (B) Apoptosis rate of vector-infected or NUP98-HOXA9-infected KLSC cells were determined as in Figure 5C. (C and D) Frequency of vector-infected or NUP98-HOXA9-infected cells undergoing asymmetric, symmetric renewal or symmetric commitment divisions were scored over 72 hr by time-lapse microscopy. Data shown are derived from 78 vector-infected and 63 NUP98-HOXA9-infected cells, and are obtained over three independent experiments. The difference in frequency of symmetric renewal divisions between vector-infected and NUP98-HOXA9-infected cells is significant (p = 0.007). (E) Relative ratios of symmetric:asymmetric division after expression of BCR-ABL or NUP98-HOXA9. Error bars represent standard deviation.