Experimental Hematology 28 (2000) 1451–1459 Autonomous behavior of hematopoietic stem cells Leonie M. Kammingaa, Imre Akkermana, Ellen Weersinga, Albertina Ausemaa, Bert Dontjea, Gary Van Zantb,c, and Gerald de Haana a Department of Stem Cell Biology, University of Groningen, Groningen, The Netherlands; Division of Hematology/Oncology, Blood and Marrow Transplant Program, Lucille P. Markey Cancer Center, Lexington, Ky., USA; cDepartment of Physiology, University of Kentucky Medical Center, Lexington, Ky., USA b (Received 19 May 2000; revised 24 July 2000; accepted 27 July 2000) Objective. Mechanisms that affect the function of primitive hematopoietic stem cells with long-term proliferative potential remain largely unknown. Here we assessed whether properties of stem cells are cell-extrinsically or cell-autonomously regulated. Materials and Methods. We developed a model in which two genetically and phenotypically distinct stem cell populations coexist in a single animal. Chimeric mice were produced by transplanting irradiated B6D2F1 (BDF1) recipients with mixtures of DBA/2 (D2) and C57BL/ 6 (B6) day-14 fetal liver cells. Results. We determined the mobilization potential, proliferation, and frequency of D2 and B6 stem and progenitor cells in animals with chimeric hematopoiesis. After granulocyte colonystimulating factor (G-CSF) administration, peripheral blood D2 colony-forming units granulocyte-macrophage were fourfold to eightfold more numerous than B6 progenitors. We determined that D2 and B6 progenitors maintained their genotype-specific cycling activity in BDF1 recipients. Chimeric marrow was harvested and D2 and B6 cell populations were separated by flow cytometry. Cobblestone area-forming cell (CAFC) analysis of sorted marrow showed that the number of late appearing CAFC subsets within the D2 cell population was zthreefold higher than within the B6 fraction. We performed secondary transplantation using unfractionated chimeric marrow, which was given in limiting doses to lethally irradiated BDF1 recipients. Comparison of the proportion of animals possessing D2 and/or B6 leukocytes 5 months after transplant revealed that the frequency of D2 LTRA was z10-fold higher than B6 LTRA numbers. Conclusion. Our data demonstrate that genetically distinct stem cell populations, coexisting in individual animals, independently maintain their parental phenotypes, indicating that stem cell properties are predominantly regulated cell-autonomously. © 2000 International Society for Experimental Hematology. Published by Elsevier Science Inc. Keywords: Stem cells—Genetics—Transplantation—Fetal liver—Chimerism Introduction Peripheral blood cell numbers are homeostatically maintained within a narrow range to ensure that essential physiological processes are carried out effectively. Studies of mice bearing mutant genes that encode specific lineage-restricted growth factors or receptors have conclusively demonstrated that the pool size of mature blood cell subsets is regulated by extrinsically acting hematopoietic growth factors [1–3]. Whereas the number of circulating peripheral blood cells is Offprint requests to: G. de Haan, Ph.D., Department of Cell Biology, University of Groningen, A. Deusinglaan 1, 9713 AV Groningen, The Netherlands; E-mail: g.de.haan@med.rug.nl similar among normal healthy humans and in most inbred strains of mice, the frequency of primitive hematopoietic stem cells varies widely [4–7]. This suggests that regulatory mechanisms affecting the frequency of stem cells in the bone marrow are distinct, and at least partly independent, from those that maintain normal levels of mature cells in the blood. Given the expanding diversity of potential clinical applications for which hematopoietic stem cells, and indeed also stem cells derived from other somatic tissues [8–10], can be of use to the patient, the issue as to how stem cell identity is regulated is of fundamental importance. Pertinent to the question of which roles stem cells may play in clinical applications is the extent to which these cells can be manipulated extrinsically, either in vivo or in vitro. 0301-472X/00 $–see front matter. Copyright © 2000 International Society for Experimental Hematology. Published by Elsevier Science Inc. PII S0301-472X(00)0 0 5 4 3 - 9 1452 L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 Conceptually, stem cell homeostasis could be achieved by cell-extrinsic or cell-intrinsic mechanisms, and these two pathways need not be mutually exclusive. To address this issue, we studied qualitative and quantitative competitive repopulation potential of genetically distinct stem cell populations, coexisting in a common environment in vivo. This was accomplished by transplanting stem cells obtained from fetal livers of gestational day 14 D2 and B6 embryos into lethally irradiated BDF1 recipients. To avoid “graft-vs-graft” reactions, we elected to use fetal liver, instead of adult bone marrow, as a source of stem cells. At this stage of ontogeny, no mature T and NK cells are present in the fetal liver [11], rendering both genetically distinct grafts immunologically compatible. We and others have shown that D2 mice have a large stem cell pool with actively cycling progenitors, whereas B6 mice have fewer stem cells with slowly proliferating progenitors [4,12–14]. In addition, possibly as a consequence of distinct stem cell pool sizes, D2 progenitor cells are much more readily mobilized to the peripheral blood after G-CSF administration than B6 cells [5,15]. Using a variety of in vitro and in vivo techniques, we show that the two stem cell populations maintain their parental phenotype within the common environment of the chimera. These data support the notion that normal hematopoietic stem cells, not unlike their malignant counterparts [16], autonomously regulate their own cell number. Materials and methods Mice Female B6D2F1OlaHsD and timed pregnant C57BL/6JOlaHsD and DBA/2OlaHsD mice were purchased from Harlan, Horst, The Netherlands. BDF1 mice were used as recipients at an age of 10– 12 weeks. Production of fetal liver chimeras BDF1 mice were lethally irradiated with 9.5 Gy using an IBL 637 Cesium-137 g source (CIS Biointernational, Gif-sur-Yvette, France), 24 hours prior to transplantation. Pregnant B6 and D2 mice (14 days postcoitus) were sacrificed by cervical dislocation and fetal livers were isolated. A single-cell suspension was prepared by repeated flushing through increasingly thinner needles and cell yield was counted using a Coulter Counter (Coulter Electronics, LTD, Dunstable, England). BDF1 mice were transplanted with either D2 cells alone (n 5 4), B6 cells alone (n 5 4), or a 1:4 mixture of D2 and B6 cells respectively (n 5 3). All mice received a total of 2 3 106 cells, injected intravenously into the retro-orbital plexus. Fourfold more fetal liver cells from B6 than D2 were transplanted since we expected D2 fetal liver cells to have superior repopulating potential compared to B6 cells, as the frequency of primitive stem cell subsets in the fetal liver of D2 mice is significantly higher [17]. Determination of peripheral blood cell chimerism using flow cytometry At various times after transplantation, approximately 50 ml blood was drawn from the retro-orbital plexus of each mouse and erythro- cytes were lysed by hypotonic shock (0.16 M NH4Cl, 1.0 3 1024 M EDTA, 0.017 M NaCl, 5 minutes at room temperature). Leukocytes were washed and stained with fluorescein isothiocyanate (FITC)-labeled anti H-2Kd-antibody (Pharmingen, San Diego, CA), and biotinylated anti H-2Kb (clone B8-24-3, a gift from Dr. S.J. Szilvassy) cells were incubated for 40 minutes at 48C, washed twice, and subsequently incubated an additional 40 minutes at 48C with Streptavidin-phycoerythrin (Str-PE) (Pharmingen, San Diego, CA). After washing, cells were finally resuspended in 200 ml phosphatebuffered saline (PBS) 1 2% fetal calf serum (FCS). Two-color analysis (FACSCalibur, Becton-Dickinson, Palo Alto, CA) was performed to assess the percentages of leukocytes derived from D2, B6, and BDF1. D2 cells are H-2Kd1 whereas B6 cells are H-2Kb1. BDF1 cells possess both H-2K isotypes, allowing accurate quantification of leukocyte contribution of all three genotypes. Progenitor mobilization with SD/01 All transplanted mice, including four additional mice of each parental strain, received a single subcutaneous injection of 25 mg polyethyleneglycol-conjugated recombinant human G-CSF (PEGG-CSF, referred to as Sustained Duration/01, kindly provided by Amgen Inc., Thousand Oaks, CA) dissolved in 200 ml PBS with 2% FCS. SD/01 has been demonstrated to potently mobilize stem and progenitor cells from the bone marrow to the blood [18]. Progenitor cells were harvested from the peripheral blood three days after administration of SD/01, a time point coinciding with peak increase in progenitor cell numbers in the blood [18]. Hydroxyurea suicide technique To measure the number of cells in S phase, the fraction of cells killed by a 1-hour incubation with hydroxyurea (Sigma, St. Louis, MO, 200 mg/mL) was determined as we have reported before [4]. Bone marrow cells were obtained from anesthesized mice by inserting a 30G needle into the knee joint [19]. 30 ml of medium was first injected into the marrow cavity and cells were subsequently aspirated. The average yield of these aspirates was z1 3 106 cells. Cell suspensions were diluted to a volume of 1 mL, and divided in two samples. Hydroxyurea was added to one sample, and incubated with the control sample for 1 hour at 378C. Both cell suspensions were then washed, and CFU-GM cultures were initiated. The fraction of cells killed by hydroxyurea was calculated and was considered to reflect the percentage of cells in S phase. CFU-GM culture assay CFU-GM were determined using standard methylcellulose cultures (0.8% methylcellulose, 30% FCS in alpha medium). Cultures, supplemented with 10 ng/mL recombinant mouse GM-CSF (Behringwerke, Marburg, Germany) and 100 ng/mL recombinant rat PEG-SCF (Amgen), were inoculated with 5–15 ml of whole heparinized blood/mL. Colonies were cultured at 378C and 5% CO2 and scored after 6 days. Genotyping individual CFU-GM colonies To genotype individual CFU-GM, colonies were picked from the culture using a micropipet and suspended in 100 ml PBS. Cells were washed and the pellet was resuspended in 10 ml DNA isolation buffer (50 mM KCl, 1.5 mM MgCl2, 10 mM Tris HCl, pH 8.5, 0.01% gelatin [Sigma, St. Louis, MO], 0.45% Igepal [Sigma], 0.45% Tween 20 [Sigma], and 100 mg/mL proteinase K [GibcoBRL]) to release DNA from the cells. Samples were incubated at 558C for two hours and subsequently at 958C for 15 minutes to in- L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 activate proteinase K. Samples were stored at 2208C until use in PCR. The genotype of each colony was verified using primers (MapPairsTM, Research Genetics Inc., Huntsville, AL) that amplify the simple sequence length polymorphism D14Mit266; the D2 allele of this microsatellite is 180 bp and the B6 allele is 148 bp. A 10 ml PCR reaction was performed according to the protocol provided by the manufacturer using a PerkinElmer GeneAmp PCR System 9700 (PerkinElmer Corp., Norwalk, CT). Amplified DNA was resolved on a 3,5% ethidium bromide–stained agarose gel. Sorting of D2 and B6 bone marrow cells from chimeric mice Twenty weeks after transplantation chimeric mice were sacrificed and bone marrow cells were isolated from both femurs. After staining total bone marrow cells with anti–H-2Kb and anti–H-2Kd antibodies as described, B6 and D2 cell populations were separated using a MoFlo flow cytometer (Cytomation, Ft. Collins, CO). D2 and B6 cell populations were collected and separate cobblestone areaforming cell (CAFC) assays were performed on both samples. Cobblestone area-forming cell assay Stem cell activity was assessed using in vitro limiting dilution type long-term bone marrow cultures. The cobblestone area-forming cell assay was performed as described previously [4]. For each of the six cell dilutions used, 20 replicate wells were tested. Early appearing CAFC day-7 correspond to relatively committed progenitor cells, whereas late appearing CAFC day-35 reflect more primitive cell subsets [20]. CAFC frequencies were calculated using maximum likelihood analysis [21]. Secondary transplantation Twenty weeks after the initial fetal liver stem cell transplantation, a single chimeric mouse was sacrificed by cervical dislocation. Cells were isolated from both femurs and transplanted in limited dilution into a new cohort of lethally irradiated BDF1 recipients. Recipients (5 per group) received 2.0 3 106, 1.0 3 106, 0.5 3 106, or 0.2 3 106 chimera-derived bone marrow cells. Engraftment of B6 and D2 leukocytes was assessed using H-2K phenotyping. Using Poisson statistics the proportion of secondary recipient animals that showed B6 or D2 leukocytes 12 weeks after transplantation was used to calculate the frequency of B6 and D2 long-term repopulating stem cells. 1453 three cell populations derived from B6, D2, and BDF1 stem cells can easily be distinguished. Figure 2 shows the contribution of stem cells from each genotype to the total leukocyte pool. Panel A depicts chimerism after injection of 2 3 106 D2 fetal liver cells. After 2 months almost all leukocytes (.95%) originated from donor stem cells, whereas only a few cells were host-derived. Similarly, leukocyte reconstitution in BDF1 mice injected with 2 3 106 B6 cells was almost entirely of B6 origin (Panel B). Importantly, no differences in either engraftment kinetics or engraftment level were observed between these two cell sources, demonstrating that there was no preferential rejection by BDF1 recipients of stem cells of either B6 or D2 genotype. Panel C shows engraftment kinetics in recipients transplanted with mixed D2/B6 stem cells. Although mice received fourfold more B6 fetal liver cells, D2 peripheral blood leukocytes were twofold to threefold more abundant than B6 leukocytes for the first 100 days after transplant. This indicates that D2 fetal liver cells have an eightfold to 12-fold higher repopulating potential compared to B6 cells. However, it should be noted that DBA reconstitution decreased with time, and B6 leukocytes became more prominent at later time points after transplant. This has also been observed in B6 ↔ D2 embryo aggregation chimeras, although in these animals skewing occurred only after 12–18 months [23]. D2 and B6 progenitor cell cycling in chimeras We have previously shown that D2 progenitors have a higher steady state cycling activity than B6 cells [12,13,17]. We isolated bone marrow cells from BDF1 mice transplanted 5 months earlier with only D2 or only B6 fetal liver cells (Fig. Results Development of radiation chimeras Chimeric mice were produced by transplanting D2, B6, or a mixture of D2 and B6 fetal liver cells in lethally irradiated BDF1 recipients. To ascertain that genotype-restricted hybrid resistance (i.e., preferential rejection of donor cells of either parent by the F1 recipient) did not bias our results, we included two groups of recipients that received only D2 cells or only B6 cells (2 3 106 cells). Chimeric mice were created by transplanting 0.4 3 106 D2 fetal liver cells, mixed with 1.6 3 106 B6 fetal liver cells. A 1:4 ratio of D2 vs B6 cells was used because we expected D2 cells to have a better repopulating ability [22] and because we had previously demonstrated that the frequency of primitive cells in the D2 fetal liver is zthreefold higher than in the B6 liver [17]. The dot plot shown in Figure 1 demonstrates that all Figure 1. Dot plot showing the presence of D2 (H-2Kd1) and B6 (H-2Kb1) leukocytes in the peripheral blood of lethally irradiated BDF1 mice (H2Kbd1), transplanted with 2.0 3 106 fetal liver cells in a 1:4 D2:B6 ratio. 1454 L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 Figure 3. Proliferation of progenitors. The percentage of CFU-GM in S phase in the bone marrow of lethally irradiated BDF1 recipients transplanted with D2 fetal liver cells only (D2 . F1) or B6 fetal liver cells only (B6 . F1). Figure 2. Chimerism in peripheral blood. Panel A shows the percentages of donor-derived D2 leukocytes in the peripheral blood of lethally irradiated BDF1 mice after transplantation of 2.0 3 106 D2 fetal liver cells. Panels B and C show chimerism after transplantation of 2.0 3 106 B6 fetal liver cells and 0.4 3 106 D2 1 1.6 3 106 B6 fetal liver cells, respectively. Values are given 61 SEM. 2A and B) and determined the percentage of CFU-GM in S phase. Progenitors in mice transplanted with D2 stem cells had an average percentage of 28% in S phase, whereas only 9% of the progenitors in mice transplanted with B6 cells were in S phase (Fig. 3). These values were identical to those typically observed in the parental strains [13,17]. Genotype-restricted mobilization potential of hematopoietic progenitors In humans and in various inbred strains of mice, the number of primitive cells mobilized from the bone marrow into the peripheral blood after growth factor administration varies widely [5,7,15,24]. We have previously postulated that differential mobilization responses reflect distinct marrow pool sizes [6]. To test this hypothesis we administrated SD/ 01, a pegylated formulation of rhG-CSF with enhanced biological activity [18], to parental and transplanted chimeric mice. The number of CFU-GM/mL blood, three days after injection of SD/01 to parental D2 and B6 mice, is shown in Figure 4A. The large strain difference is apparent and is in full agreement with previously reported data [5,15]. Mobilization potential of irradiated F1 animals, transplanted with 2 3 106 B6 or 2 3 106 D2 fetal liver cells, is depicted in Panel B of Figure 4. No significant differences compared to parental strains were observed, indicating that transplantation and engraftment of donor stem cells does not detrimentally affect subsequent mobilization potential. Since B6 and D2 progenitor cells maintained their characteristic parental mobilization pattern in an F1 microenvironment, these data additionally demonstrate that genotype-restricted differences in mobilization response after G-CSF represent an intrinsic stem cell trait. Fifteen weeks after transplantation, a time point when D2-derived leukocytes were ztwofold more frequent than B6 leukocytes in recipients of mixed B6 and D2 fetal liver cells (Fig. 2C), SD/01 was administered to the chimeras, and three days later CFU-GM were cultured from peripheral blood. Individual CFU-GM colonies were genotyped using primers that amplify the polymorphic microsatellite marker D14Mit266. Figure 4C shows an example of a genotyping experiment in which seven D2, four B6, and one BDF1 colonies were detected. Fifty colonies were analyzed in each L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 1455 Figure 4. Progenitor mobilization potential. Number of CFU-GM per mL blood three days after administration of SD/01 to parental strains (Panel A), or to chimeric mice transplanted with only D2 or only B6 fetal liver cells (Panel B). Mobilized colonies in the peripheral blood of the chimeras were genotyped using informative SSLP-PCR. Panel C depicts the amplified PCR products (D2 allele is 180 bp and B6 allele is 148 bp). Panel D shows the number and genotype of CFU-GM per mL peripheral blood in three BDF1 recipients that received 0.4 3 106 D2 1 1.6 3 106 B6 fetal liver cells. chimera. The total number of B6, D2, or BDF1-derived CFU-GM/mL blood from each chimeric mouse is depicted in Panel D of Figure 4. Endogenous BDF1-derived colonies were observed rarely. The frequency of D2-derived CFUGM was fourfold to eightfold higher than B6-derived progenitors. The skewed prevalence of D2 progenitors exceeded by severalfold the relative frequency of total D2 leukocytes, suggesting either that D2 progenitor cells were more efficiently mobilized than were B6 cells, or, alternatively, that the marrow frequency of D2 stem and progenitor cells was significantly higher than B6 cells. D2 and B6 stem cell frequencies in chimeric mice; a limiting dilution in vitro analysis To assess whether the overrepresentation of D2 progenitors after mobilization resulted from a higher D2 stem cell frequency in the bone marrow of the chimeras, we measured the pool size of D2 and B6 stem cells. First, an in vitro cob- 1456 L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 blestone area-forming cell (CAFC) assay was performed using FACS-sorted D2 and B6 cell populations obtained from bone marrow of the chimeras. Importantly, at the time point of sacrifice (140 days after transplant) B6 and D2 leukocyte populations in the peripheral blood were equally frequent in these chimeric mice (Fig. 2C). After staining bone marrow cells with anti H-2Kd and anti H-2Kb antibodies, D2 and B6 cell fractions were separated using flow cytometry, and independent CAFC assays were performed on both populations. In the chimeric bone marrow, 56% of the cells were D2derived and 33% were B6-derived (Fig. 5A). CAFC day-7 frequency was twofold higher in the H-2Kd1-cell fraction compared to cobblestone area-forming activity in the H-2Kb1 cell fraction. (Fig. 5B). For comparison, the CAFC day-7 frequency in unfractionated parental D2 and B6 control bone marrow is shown as well. CAFC day-35 were threefold more frequent in the H-2Kd1cell fraction compared to the H-2Kb1 cell fraction. (Fig. 5C and D). These values were similar to the stem cell frequency in the parental strains (Fig. 5C). D2 and B6 stem cell frequencies in chimeric mice; a limiting dilution in vivo analysis To confirm that the distinct cobblestone area-forming activity of H-2Kd1 and H-2Kb1 cell fractions properly reflected stem cell activity, and thus to reinforce our in vitro CAFC data, we performed a limiting dilution long-term repopulating ability (LTRA) assay. To this end, bone marrow was isolated from a chimeric animal and transplanted without further manipulation into a secondary cohort of lethally irradiated BDF1 mice in four cell doses (2.0 3 106, 1.0 3 106, 0.5 3 106, or 0.2 3 106 cells). At the time of bone marrow harvest (18 weeks after transplantation), 45% of the leukocytes were D2 derived and 40% were B6 derived. Figure 6A shows the percentage of secondary F1 recipients that failed to show more than 0.5% (detection limit) B6 or D2 lymphocytes or granulocytes 5 months posttransplant, as a function of the number of cells transplanted. In all four groups, each consisting of five animals, were mice that did not have B6 leukocytes. Only a single recipient, transplanted with the lowest cell doses of 0.2 3 106 cells, did not show long-term D2 granulocyte contribution (chimera 20, Fig. 6B). In contrast to the initial fetal liver chimeras, this cohort of secondary recipients showed considerable variation in levels of chimerism within each group (Fig. 6B). Using Poisson statistics, the frequency of B6 repopulating stem cells was estimated to be 0.08 per 105 bone marrow cells, whereas the frequency of D2 repopulating stem cells was 10-fold higher, 0.8 per 105 (Fig. 6C). In good agreement with the in vitro CAFC data, these in vivo results show that D2 stem cells are much more frequent than B6 stem cells in the common environment of the F1 chimera. Discussion In this study we wished to determine whether specific functional characteristics of hematopoietic progenitor and stem Figure 5. Cobblestone forming activity in chimeric bone marrow. Percentage D2- and B6-derived bone marrow cells in chimeras, 140 days posttransplant, is shown in panel A. Panels B and C show the frequency of CAFC day-7 and day-35 respectively, within the H-2Kd1 and H-2Kd1 cell fraction, isolated from the chimeric bone marrow by flow cytometry. Also shown in panels B and C are values for these cell stages as measured in parental control animals. Panel D illustrates the presence of cobblestone areas within the H-2Kb1 and H-2Kd1 cell fractions after 35 days of culture as a function of the number of cells inoculated per well. L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 1457 Figure 6. LTRA measurements using chimeric bone marrow graft. Percentage of secondary recipients that failed to show multilineage B6 or D2 contribution after transplantation of chimeric marrow obtained 5 months after primary fetal liver transplantation is shown in panel A. Panel B depicts the percent B6 and D2 granulocytes in the peripheral blood of each of the 20 chimeras as a function of transplanted cell dose. Panel C shows the LTRA frequency estimates of the corresponding limiting dilution analysis. 1458 L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 cells are defined cell-autonomously or by signals from the microenvironment. Cell traits that we analyzed included the rate of proliferation, the ability to form colonies in longterm stroma-associated cultures, the ability to migrate to the peripheral blood after cytokine challenge, and, finally, the ability to reconstitute the hematopoietic system of a lethally irradiated recipient. We took advantage of the fact that different inbred strains of mice have been shown to display a widely varying range of phenotypes related to these aspects of stem cell biology. Compared to B6 mice, D2 mice have more stem cells [6,14 22] and a higher proportion of progenitor cells in S phase [12,13,25], which are more easily mobilized to the peripheral blood [5,15]. Using a competitive analysis in which D2 hematopoietic stem cells derived from the fetal liver were allowed to compete with similarly obtained B6 stem cells, we show conclusively that all strain-specific characteristics were maintained with high fidelity within the common environment of the chimeras. Thus, all data presented in this study argue strongly for an intrinsic modulation of stem cell function. One of us (G. Van Zant) has reported extensively on hematopoietic chimerism in B6 ↔ D2 embryo aggregation chimeras [23,26]. Stem cell proliferation, stem cell frequency, and stem cell mobilization were never directly measured in those chimeras. However, when bone marrow cells obtained from these chimeras were transplanted into lethally irradiated F1 animals, the large majority of leukocytes were derived from D2 stem cells, very similar to what we have observed in the current study using fetal liver radiation chimeras. This competitive advantage was considered to be a consequence of the higher proliferative fraction of D2 cells compared to B6 cells. Although differential cell proliferation probably contributes to this process, our present studies show in fact that D2 stem cells are substantially more abundant than B6 stem cells in the chimeras. Our current results are in good agreement with those recently published by Muller-Sieburg et al., who also used the above mentioned embryo-aggregation technique. LTC-IC measurements in chimeric mice revealed consistently a higher frequency of D2 stem cells than expected based upon chimerism in other somatic tissues [27]. We have recently shown that the in vitro proliferative potential of a highly enriched single Lin2Sca-11 c-kit1 D2 stem cell is substantially higher compared to its B6 counterpart [22]. Although our present data at first sight also suggest that D2 stem cells are “superior” compared to B6 cells, production of mature cells by D2 stem cells is not very efficient in the chimeras, since a large D2 stem cell compartment produces an equal number of leukocytes as a small B6 compartment. This paradox is also observed in parental mice; in B6 mice few stem cells and few actively cycling progenitors maintain normal blood cell values as efficiently as D2 mice with many stem cells and highly cycling progenitors. This suggests that down-stream mechanisms, such as distinct rates of apoptosis of committed cells or the num- ber of amplification divisions, may further regulate peripheral blood cell production. The extent to which stem cell population size in vivo is affected by cues from the systemic or local environment potentially predicts limitations to ex vivo expansion protocols. Attempts to multiply and manipulate hematopoietic stem cells are currently being explored in many laboratories. If stem cell function is predominantly intrinsically regulated, as our data imply, it may prove difficult to expand these cells in vitro using exogenously added growth factors. It may therefore be significant that hematopoietic stem cells have been notoriously difficult to maintain in culture. Rather, altering the intrinsic genetic program of stem cells may prove to be more fruitful. First examples of genetic (re)programming are provided by studies showing that stem cells overexpressing HoxB4 or bcl-2 have enhanced repopulating activity, whereas cells deficient of the cyclin-dependent kinase inhibitor p21 have reduced stem cell activity [28–30] Recently it has been argued that cell-autonomous behavior is a property of malignantly transformed leukemic cells [16]. Our data indicate that normal primitive hematopoietic stem cells also display autonomous characteristics. At present, the molecular nature of such stem cell autonomy has yet to be defined. Truly intrinsic effectors, such as differentially expressed transcription factors, may play a role. Alternatively, however, secreted glycoproteins expressed by stem cells may affect stem cell pool size in an autocrine fashion. We, and others, have previously mapped chromosomal regions associated with variation in stem cell frequency and mobilization potential [6,13–15,31]. Although the underlying genes have not yet been identified, our current experiments indicate that these genes must be expressed by the stem cells themselves. Acknowledgments G. deHaan is a fellow of the Netherlands Organization for Scientific Research (NWO) and of the Royal Netherlands Academy of Arts and Sciences (KNAW). This work was supported by funds provided by NIH grant RO1 AG16653 to G. Van Zant, and the Lucille P. Markey Cancer Center, the University of Kentucky Hospital, and the Department of Internal Medicine. The authors would like to thank Geert Mesander for excellent assistance with flow cytometry and Dr. Piet Wierenga for intravenous injection of fetal liver cells. References 1. Lieschke GJ, Grail D, Hodgson G, et al. (1994) Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization. Blood 84:1737 2. Gurney AL, Carver-Moore K, de Sauvage FJ, Moore MW (1994) Thrombocytopenia in c-mpl-deficient mice. Science 265:1445 3. Kieran MW, Perkins AC, Orkin SH, Zon LI (1996) Thrombopoietin rescues in vitro erythroid colony formation from mouse embryos lacking the erythropoietin receptor. Proc Natl Acad Sci U S A 93:9126 L.M. Kamminga et al./Experimental Hematology 28 (2000) 1451–1459 4. de Haan G, Nijhof W, Van Zant G (1997) Mouse strain–dependent changes in frequency and proliferation of hematopoietic stem cells during aging: correlation between lifespan and cycling activity. Blood 89:1543 5. Roberts AW, Foote S, Alexander WS, Scott C, Robb L, Metcalf D (1997) Genetic influences determining progenitor cell mobilization and leukocytosis induced by granulocyte colony-stimulating factor. Blood 89:2736 6. de Haan G, Van Zant G (1997) Intrinsic and extrinsic control of hemopoietic stem cell numbers: mapping of a stem cell gene. J Exp Med 186:529 7. Roberts AW, DeLuca E, Begley CG, Basser R, Grigg AP, Metcalf D (1995) Broad inter-individual variations in circulating progenitor cell numbers induced by granulocyte colony-stimulating factor therapy. Stem Cells (Dayt) 13:512 8. Taniguchi H, Toyoshima T, Fukao K, Nakauchi H (1996) Presence of hematopoietic stem cells in the adult liver (see comments). Nat Med 2:198 9. Bjornson CR, Rietze RL, Reynolds BA, Magli MC, Vescovi AL (1999) Turning brain into blood: a hematopoietic fate adopted by adult neural stem cells in vivo (see comments). Science 283:534 10. Jackson KA, Mi T, Goodell MA (1999) Hematopoietic potential of stem cells isolated from murine skeletal muscle (see comments). Proc Natl Acad Sci U S A 96:14482 11. Spits H, Lanier LL, Phillips JH (1995) Development of human T and natural killer cells. Blood 85:2654 12. Van Zant G, Eldridge PW, Behringer RR, Dewey MJ (1983) Genetic control of hematopoietic kinetics revealed by analyses of allophenic mice and stem cell suicide. Cell 35:639 13. De Haan G, Van Zant G (1999) Genetic analysis of hemopoietic cell cycling in mice suggests its involvement in organismal life span. Faseb J 13:707 14. Muller-Sieburg CE, Riblet R (1996) Genetic control of the frequency of hematopoietic stem cells in mice: mapping of a candidate locus to chromosome 1. J Exp Med 183:1141 15. Hasegawa M, Baldwin TM, Metcalf D, Foote SJ (2000) Progenitor cell mobilization by granulocyte colony-stimulating factor controlled by loci on chromosomes 2 and 11. Blood 95:1872 16. Jiang X, Lopez A, Holyoake T, Eaves A, Eaves C (1999) Autocrine production and action of IL-3 and granulocyte colony- stimulating factor in chronic myeloid leukemia. Proc Natl Acad Sci U S A 96:12804 17. de Haan G, Van Zant G (1999) Dynamic changes in mouse hematopoietic stem cell numbers during aging. Blood 93:3294 18. Molineux G, Kinstler O, Briddell B, et al. (1999) A new form of 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 1459 Filgrastim with sustained duration in vivo and enhanced ability to mobilize PBPC in both mice and humans. Exp Hematol 27:1724 Verlinden SF, van Es HH, van Bekkum DW (1998) Serial bone marrow sampling for long-term follow up of human hematopoiesis in NOD/SCID mice. Exp Hematol 26:627 Ploemacher RE, van der Sluijs JP, van Beurden CA, Baert MR, Chan PL (1991) Use of limiting-dilution type long-term marrow cultures in frequency analysis of marrow-repopulating and spleen colony-forming hematopoietic stem cells in the mouse. Blood 78:2527 Fazekas de St G (1982) The evaluation of limiting dilution assays. J Immunol Methods 49:R11 de Haan G, Szilvassy SJ, Meyerrose TE, Dontje B, Grimes B, Van Zant G (2000) Distinct functional properties of highly purified hematopoietic stem cells from mouse strains differing in stem cell number. Blood 96:1374 Van Zant G, Holland BP, Eldridge PW, Chen JJ (1990) Genotyperestricted growth and aging patterns in hematopoietic stem cell populations of allophenic mice. J Exp Med 171:1547 Bowen S, Tare N, Inoue T, et al. (1999) Relationship between molecular mass and duration of activity of polyethylene glycol conjugated granulocyte colony-stimulating factor mutein. Exp Hematol 27:425 de Haan G, Van Zant G (1999) Dynamic changes in mouse hematopoietic stem cell numbers during aging. Blood 93:3294 Van Zant G, Scott-Micus K, Thompson BP, Fleischman RA, Perkins S (1992) Stem cell quiescence/activation is reversible by serial transplantation and is independent of stromal cell genotype in mouse aggregation chimeras. Exp Hematol 20:470 Muller-Sieburg CE, Cho RH, Sieburg HB, Kupriyanov S, Riblet R (2000) Genetic control of hematopoietic stem cell frequency in mice is mostly cell autonomous. Blood 95:2446 Sauvageau G, Thorsteinsdottir U, Eaves CJ, et al. (1995) Overexpression of HOXB4 in hematopoietic cells causes the selective expansion of more primitive populations in vitro and in vivo. Gene Develop 9:1753 Domen J, Cheshier SH, Weissman IL (2000) The role of apoptosis in the regulation of hematopoietic stem cells: overexpression of bcl-2 increases both their number and repopulation potential. J Exp Med 191:253 Cheng T, Rodrigues N, Shen H, et al. (2000) Hematopoietic stem cell quiescence maintained by p21cip1/waf1. Science 287:1804 Roberts AW, Hasegawa M, Metcalf D, Foote SJ (2000) Identification of a genetic locus modulating splenomegaly induced by granulocyte colony-stimulating factor in mice. Leukemia 14:657