Nature Cell Biology7, 172 - 178 (2005)
Published online: 16 January 2005; | doi:10.1038/ncb1214
Negative cell-cycle regulators cooperatively control self-renewal and differentiation of haematopoietic stem cells
Carl R. Walkley1, 2, 6, Matthew L. Fero4, Wei-Ming Chien4, Louise E. Purton1, 5, 6
& Grant A. McArthur1, 2, 3, 5
1 Research Division, Peter MacCallum Cancer Centre, Victoria 3002, Australia.
2 Department of Medicine, St. Vincent's Hospital, University of Melbourne, Victoria 3065, Australia.
3 Department of Haematology and Medical Oncology, Peter MacCallum Cancer Centre, Victoria 3002, Australia.
4 Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA 98109-1024, USA.
5 These authors contributed equally to this work.
6 Present addresses: Department of Hematology-Oncology, Children's Hospital and Department of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA 02115, USA (C.R.M.) and Center for Regenerative Medicine and Technology, Massachusetts General Hospital, Boston, MA 02129, USA (L.E.P.).
Haematopoietic stem cells (HSCs) are capable of shifting from a state of relative quiescence under homeostatic conditions to rapid proliferation under conditions of stress. The mechanisms that regulate the relative quiescence of stem cells and its association with self-renewal are unclear, as is the contribution of molecular regulators of the cell cycle to these decisions. Understanding the mechanisms that govern these transitions will provide important insights into cell-cycle regulation of HSCs and possible therapeutic approaches to expand HSCs. We have investigated the role of two negative regulators of the cell cycle, p27Kip1 and MAD1, in controlling this transition. Here we show that Mad1-/-p27Kip1-/- bone marrow has a 5.7-fold increase in the frequency of stem cells, and surprisingly, an expanded pool of quiescent HSCs. However, Mad1-/-p27Kip1-/- stem cells exhibit an enhanced proliferative response under conditions of stress, such as cytokine stimulation in vitro and regeneration of the haematopoietic system after ablation in vivo. Together these data demonstrate that the MYC-antagonist MAD1 and cyclin-dependent kinase inhibitor p27Kip1 cooperate to regulate the self-renewal and differentiation of HSCs in a context-dependent manner.
In vivo HSCs have a cell division cycle in the order of 2−4 weeks, a characteristic considered to be critically important for their biological function1,
2. In addition to the slow time of the cell division cycle of HSCs in vivo, isolated stem cells in vitro exhibit a significant delay in committing to their first cell division, but exhibit a high proliferative potential3,
4,
5. Several families of genes have been found to have central roles in the regulation of the cell cycle during both the entry to and exit from quiescence, in particular the Max-network and the CIP/KIP family of CDK inhibitors6,
7,
8. As relative quiescence is a defining characteristic of the HSC, we sought to determine if, as for terminal differentiation, the MYC-antagonist MAD1 and the CDKi p27Kip1 has a role in regulating the quiescence of the stem cell6,
9,
10.
Analysis of the more mature progenitor compartment revealed a p27Kip1-dependent increase in the functional day-12 colony-forming-unit spleen cells (CFU-S12) from whole bone marrow11,
12 (see Supplementary Information, Fig. S1a). The frequency of the common myeloid, granulocyte/monocyte, megakaryocyte/erythroid and common lymphoid progenitor cells was not significantly different between genotypes13,
14 (see Supplementary Information, Fig. S1b and data not shown).
To further study the effects of loss of p27Kip1 and MAD1, the stem-cell-enriched lineage negative (lin-), c-Kit+, Sca-1+ (lin-c-Kit+Sca-1+ or LKS+) population was isolated12,
15. The non-HSC-containing lin-c-Kit+Sca-1- (LKS-) fraction was also isolated15. Quantitative real-time PCR with wild-type LKS+ and LKS- populations revealed a marked difference in expression of Mad1, with the LKS+ population expressing over 700-fold more Mad1 mRNA than the LKS- population (Fig. 1a). The p27Kip1 transcript was expressed in both LKS+ and LKS- fractions, with fourfold more expression in the LKS- population (Fig. 1a).
Figure 1. Expansion of the phenotypic stem cell population in Mad1-/- p27Kip1-/- bone marrow.
(a) Expression of Mad1 and p27Kip1 cDNA in the stem-cell-enriched LKS+ and non-stem-cell-containing LKS- population as determined by quantitative real-time PCR (n = 3). (b) Frequency of the LKS+ and LKS- populations. Data expressed relative to the frequency of wild-type LKS+/LKS- cells (normalised to one; n 4). (c) Quantification of the frequency of the LKS+ CD34-/low cells (n 3). (d) Quantification of the absolute number of the LKS+ CD34-/low cells. Data are expressed as individual (open circls) and mean values s.e.m. (closed squares) (n 3). All data are expressed as mean values s.e.m.
Loss of either p27Kip1 or MAD1 alone did not alter the relative frequency of LKS+ or LKS- cells relative to wild-type mice. Notably, however, loss of both MAD1 and p27Kip1 resulted in a 3.3-fold increase in the frequency of the HSC-containing LKS+ population (P 0.05; Fig. 1b). By contrast, despite the expanded progenitor populations in both the p27Kip1-/- and the Mad1-/-p27Kip1-/- bone marrow, there was no increase in the relative frequency of the LKS- populations6,
9 (Fig. 1b). The discrepancy between the expanded CFU-S12 progenitors and that of phenotypic evaluation of progenitors (CMP, LKS+, LKS-) in the p27Kip1-/- and Mad1-/-p27Kip1-/- mice may be accounted for by an increased proliferative capacity of p27Kip1-mutant progenitors despite an unaltered frequency as assessed phenotypically, or due to different cell types being assessed in these assays16. We also determined the frequency of the highly HSC-enriched LKS+CD34-/low population17. Loss of both MAD1 and p27Kip1 significantly increased both the frequency (1.7-fold compared with wild-type; P = 0.05; Fig. 1c) and absolute number (2.6-fold compared with wild-type; P = 0.01; Fig. 1d) of this phenotypic HSC population, consistent with the analysis of the LKS+ population. Interestingly, loss of MAD1 alone was sufficient to increase the absolute numbers of LKS+CD34-/low cells, despite no increase in the relative frequency of the LKS+ population.
Next we performed competitive bone marrow transplant to determine if loss of MAD1 and p27Kip1 resulted in an increased frequency of functional HSCs. Equal numbers of wild-type and Mad1-/-p27Kip1-/- whole bone marrow were mixed with wild-type competitor bone marrow and transplanted into lethally irradiated recipients. Mad1-/-p27Kip1-/- bone marrow contributed significantly more to haematopoiesis than wild-type cells at 6-months post-transplantation (total leukocytes, P = 0.02; myeloid, P = 0.03; lymphoid, P = 0.06; Fig. 2a). These data suggested that loss of MAD1 and p27Kip1 resulted in either an increased frequency of HSCs or a competitive advantage for these cells. To distinguish between these possibilities, quantitative long-term competitive repopulation (LTRC) analysis was performed next using serial dilution to determine the frequency of HSCs in p27Kip1-/-, Mad1-/- and Mad1-/-p27Kip1-/- whole bone marrow18,
19. Analysis of short-term repopulating cells (STRC) at 3-months post-transplantation showed that loss of p27Kip1 alone or in combination with MAD1 expanded the short-term repopulating cell compartment16 (see Supplementary Information, Fig. S2a). Analysis of long-term repopulating cell (LTRC) frequency revealed no change in HSC frequency following loss of MAD1 (ref. 18, 19) (Fig. 2b and Table 1). Loss of p27Kip1 resulted in a 2.1-fold increase in the numbers of HSCs (Fig. 2b and Table 1). When both MAD1 and p27Kip1 were deleted there was a 5.7-fold increase in the frequency of multi-lineage reconstituting stem cells in the bone marrow (Fig. 2b and Table 1). This increase in HSC frequency in Mad1-/-p27Kip1-/- animals was also seen when the number of competitive repopulating units were calculated and normalized for bone marrow cellularity (Table 1). All mutant HSCs tested equally contributed to multi-lineage haemopoiesis at 6-months post-transplantation, and no evidence of proliferative disorders or leukaemia was observed in any transplant recipient up to 7-months post-transplantation (data not shown). Together, two independent competitive repopulation experiments with appropriate statistical power highlighted a competitive advantage of bone marrow derived from Mad1-/-p27Kip1-/- animals, consistent with an increase in the number of HSCs in the Mad1-/-p27Kip1-/- animals.
Figure 2. Increased numbers of long-term repopulating cells following loss of MAD1 and p27Kip1.
(a) 5 105 pooled bone marrow cells were mixed with 5 105 wild-type whole bone marrow cells and injected into recipients (n = 10). Data are expressed as means s.e.m. at 6-months post-transplant (*P 0.03, using Wilcoxon test). Myeloid, CD11b; T Lymphoid, CD3 (P = 0.06). (b) Long-term competitive repopulation analysis was performed using serial dilutions. Data shown are the percentage of CD45.2 (donor cell)-positive PB leukocytes from 2 105 WBM (whole bone marrow) cell dose 6-months post-transplant. All mice received 2 105 WBM competitor. The mean values are indicated (-); n = 8 recipients. (c) Secondary transplant analysis results. Data shown is from recipients of 2 106 cells. The mean values are indicated (-); n = 8 recipients. Wild-type cells were transplantable but no recipients achieved 5% test-derived peripheral blood cells on secondary transplantation. (d) Tertiary transplant analysis results. Data shown is from recipients of 2 106 cells. The mean values are indicated (-); n = 8 recipients. Hash indicates that recipients died before 3 months.
To determine the qualitative nature of the HSCs, serial transplant analysis was performed. Secondary analysis, undertaken using a serial dilution approach, revealed that loss of p27Kip1 and loss of both MAD1 and p27Kip1, resulted in an increased number of secondarily transplantable HSCs (Fig. 2c, Table 1). Tertiary transplantation of the wild-type, p27Kip1-/- and Mad1-/-p27Kip1-/- HSCs was further performed. Only Mad1-/-p27Kip1-/- HSCs were tertiary transplantable, averaging 2.1 0.9% of peripheral blood leukocytes at 3-months post-transplantation (Fig. 2d). Serially transplantable HSCs represent the most primitive populations within the HSC hierarchy, and these data demonstrate that loss of both MAD1 and p27Kip1 expands this primitive HSC population.
To determine if the increased frequency of HSCs in Mad1-/-p27Kip1-/- bone marrow was due to an increase in the number of HSCs within the bone marrow, as suggested phenotypically, or secondary to increased repopulating potential, competitive transplant analysis using 1,000 wild-type or Mad1-/-p27Kip1-/- LKS+ cells was performed. At 6-months post-transplantation there was no difference in the repopulating potential of wild-type and Mad1-/-p27Kip1-/- HSCs on a per cell basis (see Supplementary Information, Fig. S2c). Collectively, these data demonstrate that loss of MAD1 and p27Kip1 results in the increased frequency of long-term repopulating serially transplantable HSCs, and that the increased repopulation is not a result of different functional capacities of the HSCs but due to an increased number of HSCs. The LKS+ population was chosen for this and subsequent analysis in preference to the LKS+CD34-/low population as CD34 expression is altered in cycling/activated HSCs20,
21. Moreover, there was a discrepancy between the frequency of HSCs as defined phenotypically for Mad1-/- and p27Kip1-/- animals and that defined using bone marrow transplantation (Figs 1 and 2).
Next we directly assessed the cell-cycle status of the HSC compartment under homeostatic conditions. The stem-cell-enriched lineage-Sca-1+ (lin-Sca-1+) cells were predominantly found in the G0/1 phase of the cell cycle in wild-type bone marrow, whereas the non-stem-cell-containing lin-Sca-1- population was actively cycling1,
16,
22,
23,
24 (Fig. 3a). No differences were observed with regard to the percentage of G0/1-phase HSC populations derived from any of the mutants (Fig. 3b). By contrast, we observed an increased fraction of cycling lin-Sca-1- cells from p27Kip1-/- animals16 (Fig. 3b). We further discriminated the G0/1-phase, HSC-containing population (lin-Sca-1+) into the quiescent G0 cells based on the intensity of staining for pyronin Y, reflective of cellular RNA content23,
25. No difference was observed between wild-type and p27Kip1-/- HSCs; however, loss of MAD1 or both MAD1 and p27Kip1 resulted in an increased fraction of pyronin Ylow-staining cells, suggestive of more quiescent HSCs (Fig. 3c, d).
Figure 3. Increased quiescent stem cell pool in Mad1-/- and Mad1-/- p27Kip1-/- bone marrow under homeostatic conditions.
(a) Representative plots showing cell-cycle status of the lineage- Sca-1+ and the non-stem-cell-containing lin-Sca-1- as determined by Hoechst 33342 staining of DNA. (b) Percentage of the lin-Sca-1+ and lin-Sca-1- in the G0/G1 phase of the cell cycle (n 4). (c) Representative FACS plots showing fractionation of the G0/G1 stem cell containing lin-Sca-1+ based on intensity of RNA staining (PyroninY). G0-phase cells are PyYlow. (d) Percentage of G0 lin-Sca-1+ cells (n 4). (e) Single LKS+ cells were sorted into 60 microwell plates containing media and cytokines. The cumulative percentage of wells to have undergone at least one cell division after 24 h. (f) The cumulative percentage of wells to have undergone at least one cell division after 48 h (n 4) and 120 cells mouse-1. All data are expressed as the mean s.e.m.
To determine if MAD1 and p27Kip1 have roles in regulating the recruitment of HSCs into the cell cycle in response to stimulation, single LKS+ cells were cultured and monitored to determine the time taken for cells to undergo division after cytokine stimulation. Wild-type and p27Kip1-/- LKS+ cells underwent their first cell division at a similar rate (Fig. 3e, f). Loss of MAD1 alone did not affect the time taken to undergo division in the first 24 h, but by 48 h Mad1-/- LKS+ HSCs were more likely to have undergone division than wild-type HSC, (P = 0.03). Notably, loss of both MAD1 and p27Kip1 resulted in a decrease in the time taken to undergo division at both 24 h (P = 0.05) and 48 h (P = 0.005) compared with wild-type LKS+ cells under conditions of cytokine stimulation (Fig. 3e, f). These data indicate a role for MAD1 in regulating the rate of entry into the cell cycle of HSCs under non-steady-state conditions, and that loss of both MAD1 and p27Kip1 co-operate to allow rapid entry of HSCs into the cell cycle after stimulation.
To evaluate the in vivo consequences of loss of p27Kip1, MAD1 or both on the response of the HSC to physiological stimulation/stress, we analysed the recovery of the haematopoietic systems following haemo-ablation with 5-fluorouracil (5-FU). 5-FU kills cycling cells, ablating the progenitor pool but sparing the pool of quiescent stem cells26,
27. Loss of MAD1 alone did not alter the rate of leukocyte recovery compared with the wild-type, whereas loss of p27Kip1 led to a greater reduction in leukocyte counts within 7 days of 5-FU treatment than wild-type animals (Fig. 4a, b). In contrast to all other genotypes, Mad1-/-p27Kip1-/- animals had significantly accelerated multi-lineage recovery with 7 days after 5-FU ablation (Fig. 4a, b; also see Supplementary Information, Table 1). Several possibilities may account for the enhanced recovery of the Mad1-/-p27Kip1-/- animals. Loss of MAD1 and p27Kip1 may alter the response of cells to 5-FU; however, 3 days after 5-FU treatment there was an equivalent reduction between wild-type and Mad1-/-p27Kip1-/- animals (Fig. 4a; also see Supplementary Information, Table 1). Alternatively, increased recruitment of early progenitor cells may enhance recovery of Mad1-/-p27Kip1-/- animals. However, p27Kip1-/- animals have increased numbers of progenitors, but fewer HSCs than Mad1-/-p27Kip1-/- animals, and did not recover more rapidly than wild-type animals (Figs 2 and 4a, b). Furthermore the enhanced recovery in Mad1-/-p27Kip1-/- animals was observed across all haematopoietic lineages, more reflective of recruitment of multi-lineage repopulating cells (see Supplementary Information, Table 1). These data suggest that the enhanced recovery of Mad1-/-p27Kip1-/- animals is due to a more rapid entry into the cell cycle and subsequent expansion of Mad1-/-p27Kip1-/- HSCs in response to stress, consistent with the shortened time required for Mad1-/-p27Kip1-/- HSCs to enter the cell cycle in ex vivo culture (Fig. 3e, f) and the accelerated recovery of haematopoiesis after transplantation of Mad1-/-p27Kip1-/- HSCs (Fig. 4c).
Figure 4. Loss of both MAD1 and p27Kip1 enhanced recovery of haemopoiesis in vivo under conditions of stress.
(a) Kinetics of recovery following haemo-ablation with a single dose of 5-fluorouracil (n 3 mice per genotype). (b) Day 7 PB leukocyte counts expressed as a percentage of Day 0 counts for each of the indicated genotypes. Statistical significance was as indicated (n 3 mice per genotype). (c) Repopulation of the haematopoietic system at 5-weeks post-transplant of 1,000 wild-type or Mad1-/- p27Kip1-/- LKS+ cells. Recipients (CD45.1+) received 1,000 LKS+ of the indicated genotype and 2 105 WBM from a CD45.1+/CD45.2+ competitor. Data are expressed as the percentage of CD45.2+ PB leukocytes and represented as individual (open circles) and as means s.e.m. (closed squares) (n = 7 recipients). All data are expressed as the mean s.e.m.
The recovery of the haematopoietic compartment of mice transplanted with 1,000 wild-type or Mad1-/-p27Kip1-/- LKS+ cells further demonstrated a role for MAD1 and p27Kip1 in regulating HSC function during recovery after haematopoietic stress. Similar levels of contribution to haematopoiesis were observed 3- and 6-months after transplantation of 1,000 LKS+ cells in a competitive repopulation assay, and these populations also contain the same CFU-S12 potential (see Supplementary Information, Fig. S2). However, analysis at 5-weeks post-transplantation revealed that Mad1-/-p27Kip1-/- HSCs could more rapidly contribute to the regeneration of the haematopoietic compartment than wild-type HSCs. Five weeks after bone marrow transplantation, wild-type LKS+ cells contributed 27.6 6.2% of peripheral blood leukocytes compared with 47.6 7.4% for Mad1-/-p27Kip1-/- LKS+ cells (P = 0.03; Fig. 4c).
In contrast to previous reports, we report a role for p27Kip1-/- in regulating the size of the HSC compartment, having performed in vivo quantitative long-term competitive repopulation analysis using serial dilution to determine the frequency of HSCs rather than cobblestone-area-forming cell analysis16,
18,
19. Despite this discrepancy in the reported frequency of HSCs between the studies, analysis of the cycling status of progenitors, response to 5-FU and the progenitor populations is consistent between the studies, possibly highlighting the choice of assays of HSCs as the underlying source of the difference16. Loss of both MAD1 and p27Kip1 resulted in an expansion of the HSC pool under homeostatic conditions, similar to loss of p21Cip1. However, unlike p21Cip1, loss of MAD1 and p27Kip1 resulted in an increase in the size of the quiescent pool of HSCs under homeostatic conditions and a more rapid entry into the cell cycle under conditions of stimulation or stress22. The data suggest that MAD1 and p27Kip1 normally co-operate by distinct mechanisms to restrict entry of HSCs into the cell cycle. These data also reveal a novel role for MAD1 as a regulator of stem-cell quiescence.
MAD1 and p27Kip1 seem to regulate HSC self-renewal and differentiation decisions in a context-dependent manner. Under conditions of homeostasis, we observed an increased frequency of HSCs in Mad1-/-p27Kip1-/- bone marrow, whereas under conditions of stress we observed a significantly more rapid entry into the cell cycle. One idea that would reconcile both sets of data is that loss of MAD1 and p27Kip1 results in HSCs that are more likely than wild-type cells to enter the cell cycle, independent of context (see Supplementary Information, Fig. S3). Under homeostatic conditions, this predisposition to enter the cell cycle is balanced by the requirement for more differentiated cells with the HSC either choosing a self-renewal or apoptotic fate rather than differentiation. As we observed an increased number of HSCs in the Mad1-/-p27Kip1-/- background, loss of these genes seems to favour self-renewal under homeostatic conditions. Under conditions of haematopoietic stress, when more cells are required, or during ex vivo cytokine stimulation, predominantly functioning to enhance differentiation of HSCs, the enhanced capacity of the Mad1-/-p27Kip1-/- HSCs to enter the cell cycle allows for a more rapid generation of differentiated cells. Collectively, these data demonstrate a cooperative role for these negative regulators of the cell cycle in controlling the self-renewal and differentiation decisions of HSCs under both homeostatic and stress conditions.
Methods Mice. p27Kip1-mutant mice were provided by J. Roberts9 (Fred Hutchinson Cancer Research Center, Seattle, USA). Mad1-mutant and Mad1p27Kip1-double-mutant mice have been described6,
10. All wild-type mice were derived from within the colony and all mice were on the same C57Bl/6J background (five generations backcrossed onto the C57BL/6J background from C57Bl/6J 129/SV hybrid, CD45.2). Wild-type C57Bl/6J mice (CD45.2) used for CFU-S assays and the congenic B6.SJL-PtprcaPep3b/BoyJ (Ptprca or CD45.1) mice were purchased from Animal Resource Centre (ARC, Perth, Australia). For HSC analysis, cells derived from the first generation cross of C57Bl/6J (PMCC) and Ptprca mice were used as the source of competitor bone marrow cells (CD45.1/CD45.2 heterozygous). For the results described in Fig. 2a only, donor cells were derived from mice of a 129S4 background, competitor cells from B6.SJL animals and transplanted into wild-type C57Bl/6Jx129 F1 hybrids. Where applicable, genotyping was performed as described9,
28 using standard PCR conditions. All mice used were 8−12 weeks old. All animal experiments were conducted with the approval of and in accordance with guidelines from the Peter MacCallum Cancer Centre or FHCRC Animal Ethics Committee.
Analysis of common myeloid and lymphoid progenitors in bone marrow. Analysis of the clonogenic common myeloid (CMP) and common lymphoid progenitor (CLP) was performed essentially as described13,
14. In vitro validation of the behaviour of these cells populations gave similar results to published studies13,
14.
Colony-forming unit spleen assays. CFU-S Day 12 assays were performed using the spleen colony forming assay with whole bone marrow, LKS+ and LKS- cells from mice of the indicated genotypes11,
12.
Isolation and enrichment of haematopoietic precursor cells. Enrichment of haematopoietic stem and primitive progenitor populations (lin-c-kit+Sca-1+, LKS+; lin-c-kit+Sca-1-, LKS-) from bone marrow cells and depletion of lineage-positive cells was performed essentially as described12. Magnetic cell separation was performed as described by the manufacturer (MACS column, Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were sequentially selected for sorting as being 7-AAD-, lineage-/low, c-kit+ then Sca-1+. For analysis of the LKS+CD34-/low phenotype, cells were prepared as described above and stained with allophycocyanin-conjugated anti-CD117, fluoroscein-isothiocyanate-conjugated anti-CD34 (clone RAM34), phytoerythrin-conjugated lineage antibodies and biotin-conjugated anti-Ly6A/E (antibodies from PharMingen). Cells were then washed and stained with streptavidin-conjugated PE−Cy5.5 (Caltag). Cells were analysed on a LSRII Flow Cytometer interfaced with Cell Quest software (Becton Dickinson, San Diego, CA). Fluorescence-activated cell sorting was performed on a FACSStarplus cell sorter (Becton Dickinson) either into collection tubes or as single cells into 60-well micro-well plates using the CloneCyt direct cloning application (Becton Dickinson).
Expression analysis of Mad1 and p27Kip1 in primitive progenitor populations. LKS+ and LKS- fractions were isolated from wild-type bone marrow. Expression of Mad1 and p27Kip1 was determined by quantitative real-time PCR analysis and relative expression levels normalized to vimentin levels. Quantitative real-time PCR was performed after isolation of the LKS+ and LKS- fractions (n = 3 independent cDNA sets). Cells were lysed in Trizol reagent (Invitrogen, Carlsbad, CA) and RNA prepared using standard procedures. cDNA was prepared as described28. Primer sequences are available on request.
Short and long-term competitive repopulating analysis. Female Ptprca (Ly5.1) mice were irradiated with a total of 9.5Gy -radiation (2 4.75Gy, 3 h apart) on the day of transplantation. For transplant analysis in Fig. 2a, whole bone marrow from age- and sex-matched mice of each genotype (CD45.2, 129S4 strain, n = 5) was pooled and 5 105 cells were pooled with 5 105 competitor cells (CD45.1, B6.SJL, n = 5) and injected into C57x129 F1 hybrid recipients (n = 10). Whole bone marrow was isolated from wild-type, p27Kip1-/-, Mad1-/- and Mad1-/-p27Kip1-/- mice and a single-cell suspension was prepared (referred to as test cells). 2 105 whole bone marrow competitor cells pooled from wild-type x Ptprca heterozygous mice (CD45.1+CD45.2+ by surface phenotype) were co-injected with increasing numbers of test cells (donor cells; 5 104, 2 105 and 2 106). Eight recipients per cell dose per genotype were transplanted18,
19,
29. Recipients were analysed at 3 and 6 months post transplant. Poisson statistical analysis was performed to allow quantification of the frequency of long-term repopulating haematopoietic stem cells within the test-cell populations. Data was expressed as the percentage of negative recipients of a defined test-cell dose cohort/genotype, where negative was defined as <5% of peripheral blood leukocytes derived from Test cells. Quantification of HSCs was performed using Poisson statistical analysis of the data. The mathematical line of best fit was applied to the data and the frequency of HSCs was calculated by determining the cell dose at which 36.8% (P0) of the recipient mice were negative for Test-cell contribution to recipient haematopoiesis. (n = 8 recipients/genotype/cell dose18,
19,
29).
For secondary bone marrow transplantation, bone marrow from all individual recipients in the 2 105 cell group was pooled and then 5 104, 2 105 and 2 106 bone marrow cells were injected into cohorts of eight female lethally irradiated Ptprca recipients per cell dose. Recipients were analysed at 5 weeks and 3 months post-transplant. Tertiary analysis was performed as indicated. For tertiary analysis, 2 106 pooled bone marrow cells from the 2 106 cell dose secondary transplant recipients were injected into eight female lethally irradiated Ptprca recipients. Tertiary recipients were analysed at 5 weeks and 3 months post-transplant. For transplants using LKS+ cells, recipient mice received 1,000 freshly isolated cells and were analysed at 5 weeks, 3 months and 6 months.
Analysis of transplant recipients. Peripheral blood from each individual recipient was obtained from the retro-orbital plexus at 5 weeks, 3-months and 6-months post-transplant was analysed for lineage marker expression and CD45 congenic status by flow cytometry.
Suspension culture of haematopoietic primitive progenitor cells. Single-cell analysis was undertaken in 60 microwell plates in 15 l well-1. Single LKS+ cells were deposited into each well and cultured in cytokines (SCF, IL-6, IL-11. Flt-3l) as described12. For each experiment 120 microwells per mouse were established. At least 3 h after sorting, each well was individually viewed for the presence of a single cell, and wells in which there was no cell or more than one cell were excluded from analysis. Wells were monitored at 24-h intervals with the number of cells per well recorded over a period of 4 days.
Cell-cycle analysis of primitive progenitor populations. The cell-cycle status of freshly isolated primitive progenitor cells was determined using Hoechst 33342 (Molecular Probes, Eugene, OR) and Pyronin Y staining (Sigma, St Louis, MO). The co-staining with the RNA dye Pyronin Y allows for the separation of G0 and G1 cell populations, based on the low RNA staining of G0 cells relative to G1 cells16,
22,
23,
25. Lineage-negative progenitor cells were prepared, then resuspended at 1 106 cells ml-1 in PBS/2% FCS containing 10 M Hoechst 33342. Cells were incubated at 37 °C for 30 min then washed and resuspended at 5 106 cells ml-1 in PBS supplemented with 20 mM HEPES, 1 mg ml-1 glucose and 10% FCS containing 10 M Hoechst 33342 and 1 g ml-1 Pyronin Y23,
25. Cells were incubated for a further 30 min at 37 °C. After incubation, cells were washed and stained with biotin-conjugated anti-Ly6A/E (Sca-1), then stained with streptavidin-conjugated APC (Pharmingen, San Diego, CA). The Lin-Sca-1+ surface phenotype is enriched for haematopoietic stem cells24,
30. After staining, cells were analysed on an LSRII Flow Cytometer (Becton Dickinson). Pyronin Y fluorescence was detected at 575 nm in the linear range.
Bone marrow haemoablation. 5-Fluorouracil (David-Bull Laboratories, Australia) was administered intra-peritoneally at 150 mg kg-1 per dose. Differential counts and lineage reconstitution was performed on intra-orbital blood obtained at time of injection and at weekly intervals for the next five weeks. Peripheral blood leukocytes were analysed for granulocytic (CD11b/Gr-1), monocytic (F4/80), B lymphoid (B220/IgM) and T lymphoid (CD4/CD8a) markers (antibodies from PharMingen, except F4/80 from Caltag).
Statistical analyses. Statistical analyses were performed using the paired and unpaired Student's t test, the Mann-Whitney Rank Sum test or Wilcoxon test as indicated. The standard error of the mean was used to determine the deviation from the mean.
Received 18 October 2004; Accepted 3 December 2004; Published online: 16 January 2005.
REFERENCES
Bradford, G. B., Williams, B., Rossi, R. & Bertoncello, I. Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment. Exp. Hematol.25, 445−453 (1997). | PubMed | ISI | ChemPort |
Lemischka, I. R., Raulet, D. H. & Mulligan, R. C. Developmental potential and dynamic behaviour of hematopoietic stem cells. Cell45, 917−927 (1986). | Article | PubMed | ISI | ChemPort |
Jordan, C. T., Yamasaki, G. & Minamoto, D. High-resolution cell cycle analysis of defined phenotypic subsets within primitive human hematopoietic cell populations. Exp. Hematol.24, 1347−1355 (1996). | PubMed | ChemPort |
Leary, A. G., Zeng, H. Q., Clark, S. C. & Ogawa, M. Growth factor requirements for survival in G0 and entry into the cell cycle of primitive human hemopoietic progenitors. Proc. Natl Acad. Sci. USA89, 4013−4017 (1992). | PubMed | ChemPort |
Li, C. L. & Johnson, G. R. Stem cell factor enhances the survival but not the self-renewal of murine hematopoietic long-term repopulating cells. Blood84, 408−414 (1994). | PubMed | ISI | ChemPort |
McArthur, G. A. et al. MAD1 and p27KIP1 cooperate to promote terminal differentiation of granulocytes and to inhibit Myc expression and cyclin E-CDK2 activity. Mol. Cell. Biol.22, 3014−3023 (2002). | Article | PubMed | ISI | ChemPort |
Zezula, J. et al. p21cip1 is required for the differentiation of oligodendrocytes independently of cell cycle withdrawal. EMBO Rep.2, 27−34 (2001). | Article | PubMed | ChemPort |
Holzel, M. et al. Myc/Max/Mad regulate the frequency but not the duration of productive cell cycles. EMBO Rep.2, 1125−1132 (2001). | Article | PubMed | ChemPort |
Fero, M. L. et al. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27Kip1-deficient mice. Cell85, 733−744 (1996). | Article | PubMed | ISI | ChemPort |
Foley, K. P. et al. Targeted disruption of the MYC antagonist MAD1 inhibits cell cycle exit during granulocyte differentiation. EMBO J.17, 774−785 (1998). | Article | PubMed | ChemPort |
Till, J. & McCulloch, E. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat. Res.14, 213−222 (1961). | PubMed | ISI | ChemPort |
Purton, L. E., Bernstein, I. D. & Collins, S. J. All-trans retinoic acid delays the differentiation of primitive hematopoietic precursors (lin-c-kit+Sca-1+) while enhancing the terminal maturation of committed granulocyte/monocyte progenitors. Blood94, 483−495 (1999). | PubMed | ChemPort |
Akashi, K., Traver, D., Miyamoto, T. & Weissman, I. L. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature404, 193−197 (2000). | Article | PubMed | ISI | ChemPort |
Kondo, M., Weissman, I. L. & Akashi, K. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell91, 661−672 (1997). | Article | PubMed | ISI | ChemPort |
Okada, S. et al. In vivo and in vitro cell function of c-kit- and Sca-1-positive murine hematopoietic cells. Blood80, 3044−3050 (1992). | PubMed | ISI | ChemPort |
Cheng, T., Rodrigues, N., Dombkowski, D., Stier, S. & Scadden, D. Stem cell repopulation efficiency but not pool size is governed by p27kip1. Nature Med.6, 1235−1240 (2000). | Article | PubMed | ISI | ChemPort |
Osawa, M., Hanada, K., Hamada, H. & Nakauchi, H. Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science273, 242−245 (1996). | PubMed | ISI | ChemPort |
Szilvassy, S. J., Humphries, R. K., Lansdorp, P. M., Eaves, A. C. & Eaves, C. J. Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy. Proc. Natl Acad. Sci. USA87, 8736−8740 (1990). | PubMed | ChemPort |
Taswell, C. Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis. J. Immunol.126, 1614−1619 (1981). | PubMed | ISI | ChemPort |
Sato, T., Laver, J. H. & Ogawa, M. Reversible expression of CD34 by murine hematopoietic cells. Blood94, 2548−2554 (1999). | PubMed | ISI | ChemPort |
Tajima, F., Sato, T., Laver, J. H. & Ogawa, M. CD34 expression by murine hematopoietic stem cells mobilized by granulocyte colony-stimulating factor. Blood96, 1989−1993 (2000). | PubMed | ChemPort |
Cheng, T. et al. Hematopoietic stem cell quiescence maintained by p21cip1/waf1. Science287, 1804−1808 (2000). | Article | PubMed | ISI | ChemPort |
Gothot, A., Pyatt, R., McMahel, J., Rice, S. & Srour, E. F. Functional heterogeneity of human CD34+ cells isolated in subcompartments of the G0/G1 phase of the cell cycle. Blood90, 4384−4393 (1997). | PubMed | ISI | ChemPort |
Spangrude, G. J. & Scollay, R. A simplified method for enrichment of mouse hematopoietic stem cells. Exp. Hematol.18, 920−926 (1990). | PubMed | ISI | ChemPort |
Gothot, A., van der Loo, J. C., Clapp, D. W. & Srour, E. F. Cell cycle-related changes in repopulating capacity of human mobilized peripheral blood CD34+ cells in non-obese diabetic/severe combined immune-deficient mice. Blood92, 2641−2649 (1998). | PubMed | ISI | ChemPort |
Lerner, C. & Harrison, D. E. 5-Fluorouracil spares hemopoietic stem cells responsible for long-term repopulation. Exp. Hematol.18, 114−118 (1990). | PubMed | ISI | ChemPort |
Hodgson, G. S., Bradley, T. R. & Radley, J. M. The organisation of the hemopoietic tissue as inferred from the effects of 5-fluorouracil. Exp. Hematol.10, 26−35 (1982). | PubMed | ChemPort |
Walkley, C. R. et al. Identification of the molecular requirements for an RAR-mediated cell cycle arrest during granulocytic differentiation. Blood103, 1286−1295 (2004). | Article | PubMed | ISI | ChemPort |
Harrison, D. E., Jordan, C. T., Zhong, R. K. & Astle, C. M. Primitive hemopoietic stem cells: direct assay of most productive populations by competitive repopulation with simple binomial, correlation and covariance calculations. Exp. Hematol.21, 206−219 (1993). | PubMed | ISI | ChemPort |
Spangrude, G. J., Heimfeld, S. & Weissman, I. L. Purification and characterization of mouse hematopoietic stem cells. Science241, 58−62 (1988). | PubMed | ISI | ChemPort |
Acknowledgements We thank E. Sitnicka and C. Li for discussion and critical comment, Peter MacCallum Cancer Centre Animal Facility Staff for care of experimental animals, FACS staff for assistance with FACS sorting. C.R.W. is a recipient of an Australian Postgraduate Award. This work was supported by grants from the National Health & Medical Research Council of Australia (NHMRC) to G.A.M and L.E.P.
Competing interests statement:
The authors declare that they have no competing financial interests.
4). (c) Quantification of the frequency of the LKS+ CD34-/low cells (n 
s.e.m. (closed squares) (n 
0.05; 
105 pooled bone marrow cells were mixed with 5 
-radiation (2
l well-1. Single LKS+ cells were deposited into each well and cultured in cytokines (SCF, IL-6, IL-11. Flt-3l) as described
-mediated cell cycle arrest during granulocytic differentiation. Blood 103, 1286−1295 (2004). | 
