Wnt Signaling in the Niche Enforces Hematopoietic Stem Cell Quiescence and Is Necessary to Preserve Self-Renewal In Vivo

Cell Stem Cell

Article Wnt Signaling in the Niche Enforces Hematopoietic Stem Cell Quiescence and Is Necessary to Preserve Self-Renewal In Vivo Heather E. Fleming,1,3 Viktor Janzen,1,3 Cristina Lo Celso,1,3 Jun Guo,2 Kathleen M. Leahy,1 Henry M. Kronenberg,2 and David T. Scadden1,3,* 1Center

for Regenerative Medicine Unit Massachusetts General Hospital, Boston, MA 02114, USA 3Harvard Stem Cell Institute, Cambridge, MA 02138, USA *Correspondence: [email protected] DOI 10.1016/j.stem.2008.01.003 2Endocrine

SUMMARY

Wingless (Wnt) is a potent morphogen demonstrated in multiple cell lineages to promote the expansion and maintenance of stem and progenitor cell populations. Wnt effects are highly context dependent, and varying effects of Wnt signaling on hematopoietic stem cells (HSCs) have been reported. We explored the impact of Wnt signaling in vivo, specifically in the context of the HSC niche by using an osteoblast-specific promoter driving expression of the paninhibitor of canonical Wnt signaling, Dickkopf1 (Dkk1). Here we report that Wnt signaling was markedly inhibited in HSCs and, unexpectedly given prior reports, reduction in HSC Wnt signaling resulted in reduced p21Cip1 expression, increased cell cycling, and a progressive decline in regenerative function after transplantation. This effect was microenvironment determined, but irreversible if the cells were transferred to a normal host. Wnt pathway activation in the niche is required to limit HSC proliferation and preserve the reconstituting function of endogenous hematopoietic stem cells. INTRODUCTION The regulation of hematopoietic stem cell function is a complex and balanced process that requires coordinated input from inherent HSC programs and moderating signals provided by the surrounding microenvironment. Together, these signals permit the maintenance of the stem cell pool for the life of the organism, while also allowing for sufficient steady-state and injury-responsive blood cell production. These somewhat dichotomous aspects of HSC function require mechanisms that both preserve a quiescent population of stem cells and also promote their activation, expansion, differentiation, and circulation under appropriate conditions (Akala and Clarke, 2006; Scadden, 2006). The morphogen family of signaling molecules has been identified as a prominent player in the function of numerous stem cell types, including the hematopoietic lineage. The wingless (Wnt) 274 Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc.

pathway has been studied extensively in the context of hematopoiesis, and the combined impact of multiple family members binding to a range of receptors leads to activation of canonical and noncanonical signaling pathways (Nemeth and Bodine, 2007). Canonical signals are mediated by TCF/LEF transcription factor activity (Daniels and Weis, 2005), and are considered to be largely dependent on the accumulation of nuclear b- (and/or g-) catenin (Nemeth and Bodine, 2007). Wnt signals have been implicated in mammalian hematopoiesis by studies not intended to assess normal physiology in which Wnt activation had a strong expansive effect on reconstituting HSCs and multipotent progenitors (Baba et al., 2006; Murdoch et al., 2003; Reya et al., 2003; Trowbridge et al., 2006b). With enforced, persistent Wnt activation, however, engineered mice developed hematopoietic failure with impaired differentiation of HSC (Kirstetter et al., 2006; Scheller et al., 2006). In contrast, deletion of members of the Wnt/b-catenin cascade under homeostatic conditions had little to no effect on blood cell production by HSCs (Cobas et al., 2004; Jeannet et al., 2008; Koch et al., 2008), raising the question of what physiological role, if any, Wnt signaling has on this cell type. Some of the variation observed may reflect differing influences exerted by canonical versus noncanonical Wnt signals, particularly given a recent report indicating that Wnt5a can modulate canonical signals mediated by Wnt3a (Nemeth et al., 2007). Wnt signals are also regulated by a host of soluble inhibitors that may interact directly with Wnt ligands, such as the frizzled-related proteins (sFRP), or by preventing Wnt binding to its receptors (Kawano and Kypta, 2003). The Dickkopf (Dkk) family of Wnt inhibitors falls into this latter category, by binding the Wnt coreceptor LRP5/6 in combination with a Kremen receptor, and leading to internalization of the complex (Mao et al., 2001, 2002). In order to specifically examine the impact of Wnt activation in an in vivo microenvironment that has been shown to regulate HSC number and function, we utilized mice engineered to overexpress the Wnt inhibitor, Dkk1, under control of the osteoblastspecific 2.3 kb fraction of the collagen1a (Col1a) promoter. This promoter has been previously shown to direct transgene expression to osteoblastic cells, resulting in changes in the number and function of HSCs (Calvi et al., 2001, 2003). We noted very little overt phenotype in the hematopoietic compartment of the Dkk1 tg mice at steady state, and confirmed that transgene expression did not extend to the primitive

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

hematopoietic fraction itself. Clear alterations of bone morphology were observed, however, including a 20% decrease in trabecular bone (J.G., M. Liu, D. Yang, M. Bouxsein, H. Saito, R. Baron, R. Bringhurst, and H.M.K., unpublished data). Despite the absence of a steady-state hematopoietic phenotype, TCF/ LEF activity was specifically reduced within the HSC-containing fraction of Dkk1 transgenic (Dkk1 tg) mice, and stem cell function was altered under specific conditions. For example, a highly significant defect in the maintenance of reconstitution potential of HSC was observed, either in settings of serial transplant, or following secondary transplantation of wild-type donor cells previously used to reconstitute Dkk1 tg hosts. In agreement with the functional data, HSC populations had a marked reduction of cells within the G0 fraction of the cell cycle, and displayed enhanced sensitivity to 5-fluorouracil treatment. Wnt signals therefore appear to participate in mediating HSC quiescence in vivo, a result that was largely unpredicted from previous studies, although recent analysis of Hmgb3 mutant mice also supports this conclusion (Nemeth et al., 2006). Our results highlight the importance of studying the impact of a signaling pathway over long-term experiments, and in a physiologic context when seeking to resolve the effects of manipulations on HSC function. In that context, Wnt signaling plays an unanticipated role in maintaining HSC quiescence, which may underlie its requirement in preserving the self-renewing capability of HSCs. RESULTS Osteoblast Expression of Dkk1 Does Not Affect Blood or Marrow Primitive Hematopoietic Cell Populations at Steady State The Wnt inhibitor, Dkk1, has been shown to play an important role in bone formation during development (Niehrs, 2006) and is normally expressed by osteoblasts (Grotewold et al., 1999; MacDonald et al., 2004), hence it may have regulatory roles as part of the endosteal HSC niche. To examine the impact of Wnt inhibition on hematopoietic stem cells localized to the periendosteal region, Dkk1 was overexpressed within osteoblastic lineage cells under the control of the truncated 2.3 kb Col1a promoter (J.G., M. Liu, D. Yang, M. Bouxsein, H. Saito, R. Baron, R. Bringhurst, and H.M.K., unpublished data). Resulting Col1a2.3Dkk1 tg mice were backcrossed for at least five generations to the C57Bl/6 background and examined for bone and blood phenotypic alterations. No significant differences in peripheral white or red blood cell counts were observed (see Figure S1A available online). Bone marrow (BM) cellularity and spleen cellularity were also unchanged when Dkk1 tg mice and their littermates were compared, although a slight but not significant trend toward reduced body weight and BM cellularity was apparent in transgenic mice (Figure S1B and data not shown). In contrast, significant alterations in bone morphology were observed, as is reported elsewhere (J.G., M. Liu, D. Yang, M. Bouxsein, H. Saito, R. Baron, R. Bringhurst, and H.M.K., unpublished data; and Li et al. [2006]). Of note, trabecular bone volume was reduced by 20%, whereas cortical bone was unaffected in Dkk1 tg mice (data not shown). Trabecular bone has been shown by us and others to affect HSC number and function (Adams et al., 2007; Calvi et al., 2003; Jung et al., 2007; Zhang et al., 2003). A panel of antibodies using seven different fluorochromes was used for

multiparametric analysis of primitive precursors within the BM of Dkk1 tg mice and their littermates, including populations of LT-HSC, ST-HSC, CMP, GMP, MEP, and CLP (Figures 1A and 1C). Subpopulations containing primitive HSCs were not significantly altered at steady state (Figure 1B). However, additional cell surface markers revealed a slight but significant increase in the population containing phenotypically defined common lymphoid progenitors (Figure 1D). The calculated absolute cell numbers based on these frequencies indicated a similar pattern of results (Figure S2). Despite the elevation of early lymphoid progenitors in the BM of Dkk1 tg mice, no significant changes were observed in the relative proportion of early B lineage progenitor subsets in the BM (data not shown). Dkk1 tg HSCs Exhibit Impaired Wnt Signaling in a Non-Cell Autonomous Manner To confirm that the transgenic expression of Dkk1 leads to the inhibition of Wnt/b-catenin signaling in the Dkk1 tg mice, HSCcontaining populations were isolated from Dkk1 tg mice that had been intercrossed with the Topgal reporter strain. In these Topgal mice (DasGupta and Fuchs, 1999), multiple TCF/LEF binding sites have been inserted to control the expression of the reporter gene, b-galactosidase. Reporter activity using this construct has been shown to correlate with canonical Wnt signaling. Of note, TCF/LEF transcription has recently been shown to proceed even with the combined loss of b-catenin and g-catenin, suggesting that canonical Wnt signals can be transduced by alternate intermediates (Jeannet et al., 2008). Reporter activity was examined within the LK+S+ (Lineage-cKit+Sca1+), HSCcontaining population, and the LK+S population that is devoid of LT-HSC potential. When the Wnt reporter activity detected in each of these populations was compared, a dramatic reduction (>100-fold reduction) in b-catenin activation was observed in the HSC-containing LK+S+ population isolated from Dkk1 tg mice (Figure 2A). A more modest reduction (<5-fold reduction) was observed in the less actively signaling LK+S fraction. This finding indicates that despite the unchanged frequency of phenotypically defined HSC-containing populations in unmanipulated Dkk1 tg animals, there is evidence that these cells are molecularly altered by osteoblast expression of the Wnt inhibitor. These data provide evidence for direct inhibition of Wnt signaling in the HSC population in addition to any effects that might be mediated by decreased trabecular bone mass. Wnt signaling is regulated, in part, via a negative feedback loop by TCF/LEF-dependent transcription of endogenous Dkk1 (Niida et al., 2004). Consistent with the decrease in Topgal reporter activity, expression of endogenous Dkk1 was also inhibited in the LK+S+ population of Dkk1 tg mice (Figure 2B). Using primers specific for the Dkk1 tg, and in comparison with its expression in WT and Dkk1 tg tibiae, sorted LK+S+ cells do not express the Dkk1 transgene (Figure 2C). Together, these results confirm that Dkk1 tg mice inhibit Wnt signaling specifically within the HSC compartment in a non-cell autonomous manner. Functional Impact of the Dkk1 Transgene on BM Reconstitution Analysis of stem/progenitor activity cannot rely exclusively on the quantitation of precursors according to phenotypically defined parameters. Using functional measures, we detected a consistent Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc. 275

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

Figure 1. Seven-Color FACS Analysis of Primitive Populations in WT and Dkk1 tg BM BM from Dkk1 tg and littermates was assayed by multiparameter FACS for relative proportion of primitive HSC populations. BM was stained with antibodies against Lineage markers, cKit, Sca-1, CD34, Flk2, CD16/32, and CD127 and gated as shown in (A) and (C). At least ten mice per genotype were compared, over at least three separate experiments. The proportion of BM corresponding to the HSC-containing LK+S+ fraction (A, blue gate) is shown in (B) (left axis), and is subsectioned according to CD34 and Flk2 expression to yield phenotypic assessments of LT-HSC and ST-HSC fractions (B, right axis). More differentiated progenitors gated in the LK+S population (A, left, green gate) were subsectioned based on CD16/32 and CD34 expression to compare CMP, GMP, and MEP progenitors as shown in (C) (left panel). Frequencies of each population, from the same samples quantified for HSC frequency in

276 Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc.

defect in multilineage and myeloid colony formation on a per-cell basis in BM isolated from Dkk1 tg mice (Figure 3A). This result was despite the absence of significant alteration of myeloid and more primitive progenitors by immunophenotype, possibly reflecting the elevated lymphoid fraction, whose progeny are not read out under these culture conditions. In vitro methods such as the CFU assay offer an entry-level analysis of hematopoietic activity; however, functional reconstitution in vivo more accurately examines true HSC function (Purton and Scadden, 2007). Therefore, in order to better assess the functional capacity of HSCs isolated from the Dkk1 tg environment, BM was transplanted from WT or Dkk1 tg littermates with an equivalent dose of competing marrow from congenic donor mice into lethally irradiated recipients. Donor marrow was isolated from a single WT or transgenic mouse to assess any individual-to-individual variation. Following 6 months of engraftment, no significant changes in reconstitution were observed across the groups of recipients receiving BM isolated from individual WT or Dkk1 tg environments, although a range of reconstitution capacity was apparent in both groups (Figure S3A). Using a limiting dilution assay to determine the frequency of repopulating cells present in BM isolated from individual Dkk1-expressing animals revealed a 2-fold elevation in the number of functional reconstituting HSCs (Figure 3B). These transplant results indicate that cells isolated from the Dkk1-expressing niche are capable of reconstituting irradiated recipients and appear to be present at a higher frequency when Wnt has been inhibited in this location. An important additional parameter to test when investigating HSC function is their longevity, or ability to respond to repeated rounds of expansion stress. To assay the longevity of HSCs isolated from Dkk1 tg mice, noncompetitive serial transplants were performed. As expected from the previous transplant experiments, Dkk1 tg BM was able to completely reconstitute WT irradiated recipients (data not shown). However, following two subsequent transplantations, each after a period of 14 weeks of reconstitution, Dkk1 tg-derived HSCs were unable to rescue irradiated hosts under the same conditions supported by littermate BM donors (Figure 3C). Highly significant differences in the survival curves of the tertiary transplanted recipients are depicted, and were similar in an independent serial transplant experiment using multiple donors. Consistent with the decrease in survival after multiple transplants, a reduction in the frequency of phenotypic LT-HSCs was also detected in the BM of Dkk1 tg secondary hosts used as donors for tertiary transplants (Figure 3D). Together, these results demonstrate that HSCs are functional following isolation from a Dkk1-expressing endosteal environment, but their long-term function is not maintained over sequential transplantations. A forced, repeated expansion requirement imposed on cells that have once been exposed to the Wnt-inhibited environment results in their early expiration. Effect of Temporary Exposure to Endosteal Dkk1 on HSC Function Transplantation of BM cells from unmanipulated Dkk1 tg donors allows the assessment of the functional capacity of HSCs (B), are shown in (D) (left axis). The CLP fraction, gated on LKloSlo in (A) (red gate), and gated further on CD127+ cells in (C) (right panel) are quantified in (D) (right axis). Significance was determined by a Student’s 2-tailed t test. Error bars indicate the SE of the mean.

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

derived cells detected within the circulation of reconstituted animals (Figure 4A and data not shown). These findings suggest that HSC function is not prevented in a setting that inhibits Wnt via Dkk1 overexpression within the HSC niche. In stark contrast with the competent function of donor cells in primary Dkk1 tg recipients, BM isolated from individual reconstituted hosts was unable to efficiently competitively repopulate lethally irradiated secondary WT recipients (Figure 4B). Consistent with the loss of HSC function following only 14 weeks of exposure to the Dkk1 tg host environment, reconstituted BM cells that had been resident in Dkk1 tg hosts did not yield as many primitive myeloid colonies in a CFC assay (Figure S3B). Therefore, the adverse effect of Dkk1 exposure was not limited to cells that developed within the transgenic environment, but could also be induced by transplantation of adult cells into an adult BM niche that overexpressed the Wnt inhibitor.

Figure 2. Assessment of Canonical Wnt Signal Activity in HSCContaining Populations of Dkk1 tg Mice Dkk1 tg mice were crossbred with Topgal transgenic mice. Three pairs of topgal+ littermates, with and without the Dkk1 tg, were used to isolate the HSCcontaining LK+S+ primitive fraction, and more differentiated LK+S cells. FACS-sorted populations were used to isolate mRNA and assessed for expression of the Topgal reporter. Data are expressed as the relative Topgal mRNA expression, compared to the levels present in LK+S+ WT cells (A). Three pairs of Dkk1 tg and WT littermates were also used to isolate the LK+S+ fraction of BM, from which mRNA was isolated (non-Topgal transgenics) (B). mRNA was also isolated from the tibiae (bone + marrow) of WT and Dkk1 tg mice. Quantitative PCR was performed using primers specific for the Dkk1 transgene. Data are expressed as the relative amount of Dkk1 mRNA compared to that found in WT tibiae (C), and error bars indicate SE of the mean.

following chronic exposure to a Wnt-inhibited environment, as the assayed cells have been exposed to the transgenic osteoblastic cells throughout development. To investigate the impact of temporary exposure to osteoblast-produced Dkk1 on HSC function in a situation of stress, WT or Dkk1 tg mice were lethally irradiated and then transplanted with WT BM from CD45 congenic donors. Consistent with the results obtained in the transplant of chronically Wnt-inhibited HSCs, Dkk1 tg mice were fully reconstituted by WT donor cells, as measured by peripheral white blood cell counts and the percent of multilineage, donor-

Wnt-Inhibited HSC-Containing Populations Are Less Quiescent The long-term function of HSCs that permits them to support hematopoiesis even after several rounds of stress-induced expansion, as mimicked by a serial transplant, is thought to depend on maintaining a fraction of the population in a highly quiescent state (Cheng et al., 2000; Hock et al., 2004; Yilmaz et al., 2006). Multiparameter FACS analysis revealed that the HSCcontaining, LK+S+ population isolated from Dkk1 tg mice is specifically depleted of cells in the quiescent, G0 stage of the cell cycle (Figure 5A). Indeed, greater fractionation of this population by gating on LK+S+CD34low cells also revealed that Dkk1 tg mice did not retain quiescent cells (Figure 5B). A similar trend was observed when WT cells were examined following transplantation and temporary exposure to the Wnt-inhibited environment (data not shown). Cell-cycle analysis at a single time point may not fully reflect the tendency of a population to initiate divisions over time (Janzen et al., 2006). To examine the frequency of entry into cell cycle based on the initiation of DNA synthesis, WT and Dkk1 tg mice were exposed to BrdU for a period of 16–18 hr, followed by analysis of BrdU incorporation within primitive BM populations. When the LK+S+ population was fractionated into LTand ST-HSC populations, the proportion of BrdU-containing cells was elevated in tg cells, significantly in the CD34+Flk2 population (Figure S4). These results suggest that Wnt-inhibited HSCs are unable to maintain a population of quiescent cells, and may provide an explanation for the reduced longevity of HSC function observed in serial transplants. An additional method that can assay for the presence of a population of quiescent BM progenitors is to track the recovery of peripheral blood following treatment with the S phase-specific nucleoside analog, 5-fluorouracil (5-FU) (Randall and Weissman, 1997; Van Zant, 1984). Depletion of cycling cells following 5-FU treatment is followed by a spike in newly generated whole blood cells (wbc) emerging from a more quiescent, and therefore spared, stem/ progenitor pool. The time to recovery correlates with the relative preservation of function of the stem/progenitor pool (Yeager et al., 1983). In animals reconstituted with Dkk1 tg BM, there was a significant lag in the ability of BM to replace the depleted neutrophil pool (Figure 5C). The impact on total wbc was also significant, if dampened (data not shown), due to the presence of long-lived circulating lymphocytes. These findings suggest Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc. 277

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

Figure 3. Functional Assessment of HSCs Isolated from Dkk1 tg Mice (A) BM from eight pairs of WT and Dkk1 tg mice was plated in methylcellulose with growth factors (SCF, IL-3, IL-6, Epo) and scored for CFU-C (combined scoring for BFU-E, CFU-GM, and CFU-GEMM colonies) after 12 days. All live colonies of more than 30 cells were counted for each of three wells plated per sample. Data are shown as mean colonies per well for each of eight mice studied over three individual experiments. Significance was determined using a two-tailed Student’s t test. (B) Limiting dilution experiments were performed using three doses of test marrow (CD45.1) transplanted with 5 3 105 competing cells (CD45.2) into groups of at least nine recipients (CD45.2) per dose. Test marrow was isolated from two WT and two Dkk1 tg mice, and the Dkk1 tg donors shown here were transplanted into separate groups of irradiated recipients. Data points are plotted as the percent of recipients per group that did not exhibit at least 1% multilineage PB engraftment at 6 months (percent unreconstituted). LT-HSC frequency and significance were determined using Poisson statistics: WT, 1 in 63,000 versus tg, 1 in 31,500 or 1 in 37,000; p < 0.02. Similar results were obtained in an independent assessment of two Dkk1 tg donors. (C) Noncompetitive serial transplants were initiated by transplanting 1 3 106 whole BM pooled from three WT or Dkk1 tg donors (CD45.1) into each of ten irradiated recipients (CD45.2). Secondary and tertiary transplants were performed after 14 weeks of engraftment by pooling BM from three to four reconstituted recipients to transplant 1 3 106 whole BM into new groups of ten irradiated CD45.2 recipients. The Kaplan-Meier survival graph depicts the survival of tertiary recipients, mice receiving BM from Dkk1 tg mice (solid line) or WT controls (dashed line). Similar results were obtained in an independent assessment of two WT and 2 Dkk1 tg mice. (D) Prior to transplant into tertiary recipients, BM from five secondary recipients of both genotypes was assayed by FACS for the frequency of LT-HSCs (LK+S+CD34lowFlk2). Error bars indicated SD of the mean, and significance was determined by a two-tailed t test.

that the stem/progenitor population in Dkk1 tg-exposed BM exhibits increased sensitivity to the toxic effects of 5-FU: a phenomenon expected with increased cell cycling. Indeed, only 40 hr after exposure to 5-FU, Dkk1 tg BM contains fewer phenotypically primitive cells (Figure S4B), as measured by lineage and SLAM family markers CD150 and CD48 (Kiel et al., 2005). No differences in the frequency of cells measured by cKit expression were observed, likely due to modulation of this protein during 5-FU exposure (Randall and Weissman, 1997). Consistent with the observed effect of Dkk-1 on stem cell cycling, we detected a decrease in expression of the cyclin-dependent kinase inhibitor, p21Cip1 within the Dkk1 tg LK+S+ population (Figure 6). This cell-cycle regulator has been previously noted by us to regulate stem cell cycling and, in its absence, an acceleration in stem cell exhaustion (Cheng et al., 2000; Miyake et al., 2006). Though not all studies agree, this effect of p21Cip1 has also been documented in other stem cell systems (Kippin et al., 2005; van Os et al., 2007). A pathway demonstrated under some circumstances to be affected by Wnt is Notch (Duncan et al., 2005), and we also observed a decline in expression of the Notch pathway target, Hes-1 (Figure 6). Changes in Hes-1 expression have been shown by others to directly affect HSC selfrenewal (Kunisato et al., 2003), perhaps by regulating p21 (Yu et al., 2006). A similar reduction in Hes-1 expression was also observed in the LK+S+ fraction of control BM used to reconstitute 278 Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc.

Dkk1 tg hosts (Figure S3B). In contrast, other transcription factors were unaffected by the presence of the osteoblastic-derived Wnt inhibition (Figure 3C). These included Gfi-1, Bmi-1, and HoxB4, each of which has been shown previously to play a role in HSC self-renewal programs (Hock et al., 2004; Iwama et al., 2004; Kyba et al., 2002; Park et al., 2003; Zeng et al., 2004). DISCUSSION Understanding the role of specific signals in the varied regulatory functions of HSC activities is crucial for designing and developing therapeutic interventions involving these cells. The impact of the Wnt family on the expansion and regulation of hematopoietic cells has been examined in a variety of studies. However, the physiologic effects of this pathway remain somewhat ill defined, with often contradicting results. Some have demonstrated that Wnt cascade activation promotes the proliferation of HSCs and their progeny while maintaining at least short-term functional activity (Baba et al., 2006; Murdoch et al., 2003; Reya et al., 2003; Trowbridge et al., 2006b). Others, employing persistent genetic activation of the pathway, have also demonstrated an increase in proliferation of cells with an HSC immunophenotype, but with marked impairment of HSC differentiation resulting in animal death (Kirstetter et al., 2006; Scheller et al., 2006). However, induced deletion of b-catenin, the primary downstream mediator

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

Figure 4. Transplant Analysis of HSC Function Following Residence in a Dkk1 tg Environment Control, 1 3 106 WT whole BM cells (CD45.2) were transplanted into each of five WT or Dkk1 tg irradiated recipients (CD45.1). Peripheral blood was assayed by FACS for engraftment by CD45.2 staining of wbc after 14 weeks (A). Contribution to both B220+ and Mac1+ populations was confirmed. BM was harvested from two engrafted recipients for each genotype, and groups of five irradiated WT CD45.1 secondary hosts received 9 3 105 CD45.2 cells from individual primary recipients, mixed with 3 3 105 fresh CD45.1 WT BM. PB engraftment of secondary recipients was assayed after 6 months (B). Data shown compare the reconstitution from ten recipients of WT-resident BM versus groups of five WT recipients transplanted with BM from two individual Dkk1 tg hosts. Error bars are SD of the means in each case, with significance determined by two-tailed t tests. Similar results were obtained in a second experiment of two independently transplanted and retransplanted Dkk1 tg hosts.

of the Wnt cascade, resulted in no apparent impact on HSC activity, even in a reconstitution assay that required expansion of b-catenin null transplanted HSCs (Cobas et al., 2004). Furthermore, recent combined deletions of both b-catenin and its homolog, g-catenin, also maintain HSC function under steadystate and primary reconstitution conditions (Jeannet et al., 2008; Koch et al., 2008). All of these studies have assayed Wnt activity in either broad over- or under-stimulation settings, and the manipulations have been performed on the HSCs themselves, or broadly applied to recipient animals. The context in which morphogens are present is highly relevant to their effect and not previously studied for Wnt effects on hematopoiesis (Trowbridge et al., 2006a). Indeed, Wnt ligands can modulate signaling initiated by other Wnt family members, underscoring the concept that context and different signaling intermediates may have a strong impact on functional outcome (Nemeth et al., 2007). In the present study, we have established a system that permits the analysis of localized Wnt inhibition, offering the opportunity to assay the impact of chronic or temporary exposure to this inhibited environment. In particular, we have directed expression of the Wnt inhibitor, Dkk1, to a cell population that has been previously demonstrated to exert a regulatory function over HSC activity, and which normally express Dkk-1, albeit at lower levels (Grotewold et al.,1999; MacDonald et al., 2004). It should be noted that while an increasing number of reports suggest that phenotypically identified HSCs inhabit additional physical locations within the BM environment (Kiel et al., 2005; Sipkins et al., 2005), the promoter used in our study has proven to functionally impact the number and activity of HSCs when used to direct modifying signal expression to a population of osteoblastic cells. Given that expression of Dkk1 also results in alterations to bone morphology itself, there is likely to be a dual effect of Dkk1: one altering the niche architecture and the other affecting Wnt signaling in stem/progenitor cells. Our studies demonstrated an effect of Dkk1 overexpression by non-HSCs on Wnt signaling in hematopoietic stem/progenitors, suggesting that this is at least a contributing factor to the phenotype observed. This observation that TCF/LEF reporter activity is reduced, as is expression of endogenous Dkk1, itself a Wnt signaling target (Niida et al., 2004) in BM cells of the transgenic mice, indicates altered canonical Wnt signaling. It does not rule out that Dkk1 may exert additional Wnt-independent functions. The results presented here also indicate

that the reduced longevity of HSCs does not require constant exposure to exogenous Dkk1, given that we were unable to detect Dkk1 tg expression within populations of primitive hematopoietic cells, and therefore the functional impact on transplanted cells is observed in a Dkk tg-free environment. It is important to note that transplantation of whole BM populations is generally not effective at engrafting nonhematopoietic cells (Koc et al., 1999).

Wnt Mediates HSC Quiescence and Maintains Reconstitution Function In Vivo The results presented here establish a role for Wnt, in the maintenance of a quiescent fraction of functional HSCs in BM. This was associated with evidence of increased stem cells on limit dilution transplant analysis. However, the ability of the same cells to function after serial rounds of transplantation was drastically reduced. The ability of stem cells to persist under the stress conditions of transplantation requires self-renewal capability that is compromised after Dkk1 exposure. The studies of inducible deletion of b- and g-catenin noted that they were dispensable for HSC function; however, they did not include sequential transplants out to the extent where we observed our most dramatic phenotype (Cobas et al., 2004; Jeannet et al., 2008; Koch et al., 2008). Alternatively, it is possible that Dkk1 interferes with HSC function through a process that does not depend on b- or g-catenin signaling (Jeannet et al., 2008; Niehrs, 2006). Our results emphasize the importance of studying pathways within the context of other signals present in the natural microenvironment, and underscore the potential for unanticipated functional roles. It is clear that different combinations of signals may have a range of effects depending on the context in which they are received. Indeed, we observed an impact of Wnt inhibition on the activation of the Notch target, Hes-1, raising the possibility that Notch and Wnt coordinate in vivo to maintain quiescence of HSCs, rather than participating in expansive and/or self-renewal functions (Duncan et al., 2005). Notably, elevated Hes-1 and p21 expression have recently been shown to correlate with the maintenance of quiescence and repopulating function of primitive HSCs (Yu et al., 2006). We noted a highly specific impact of the Dkk1 tg on the stem cell-enriched LK+S+ fraction in Wnt-dependent pathway activation and inhibition and the Notch target, Hes-1, or the cell-cycle regulator, p21 expression. Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc. 279

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

Figure 5. Examination of Cell-Cycle Status of Primitive BM in WT and Dkk1 tg Mice BM was isolated from three pairs of WT and Dkk1 tg mice and subjected to FACS analysis after surface staining and treatment with Hoechst 33342 and Pyronin Y. Proportion of LK+S+ gated (A) or LK+S+CD34low gated (B) cells present in gates corresponding to G0, G1, or S-G2-M phases of the cell cycle (see inset diagram) are depicted for a representative pair of WT (left) and Dkk1 tg (right) mice. (C) To functionally assay the preservation of a quiescent BM fraction, control recipient mice were reconstituted with WT or Dkk1 tg cells and allowed to engraft for 16 weeks. Peripheral neutrophil (NE) counts were monitored by weekly cbc analysis following intraperitoneal injection of 5-FU at day 0 and day 42 (injections indicated by downward arrows). Data indicate mean ± SE for groups of five recipients per genotype (WT, dashed line; Dkk1 tg, solid line) and are representative of two independent experiments (**p < 0.005, *p < 0.05, two-tailed t test).

is a biphasic response of cell cycling to the Wnt pathway and that proper control of stem cell quiescence requires a fine-tuned modulation of intermediate Wnt signaling intensity. This has implications for the potential use of Wnts as mediators of stem cell expansion ex vivo and for interruption of this pathway as an antileukemic intervention. In sum, niche-related expression of Dkk1 reveals a role for Wnt signaling in the physiologic regulation of the hematopoietic compartment, altering stem cell cycling and longevity following repeated expansion, or self-renewal. The phenotype observed was sufficiently distinct from what cell-autonomous modifications of the pathway would have predicted to argue for niche-specific modeling of exogenous factors’ effects on stem cells. This may be particularly true for members of the locally acting morphogen group of cell modifiers. EXPERIMENTAL PROCEDURES

The effects of Dkk1 on cell cycling were unanticipated given previous reports of constitutively active b-catenin inducing increased stem/progenitor cell proliferation (Kirstetter et al., 2006; Scheller et al., 2006). However, others found that with deletion of the chromatin binding protein, Hmgb3, Wnt signaling was increased, yet stem cells more readily returned to quiescence after 5-FU challenge than controls (Nemeth et al., 2006). Both increased and decreased activation of the pathway may therefore alter HSC cycling kinetics. This may again be due to the context differences observed with a microenvironmentally provided signal in the current study contrasted with cell autonomous activation of the pathway in the prior reports. Alternatively, it may be an example of the complex effects of morphogens, which have dose-dependent actions (Delaney et al., 2005; Kielman et al., 2002; MacDonald et al., 2004). It may be that there 280 Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc.

Mice C57Bl/6 WT, CD45.1 congenic (Jackson), and Col1a-Dkk1 tg mice were bred in-house in a pathogen-free environment. Unless otherwise stated, male Dkk1 tg mice and WT littermates were studied between 8 and 12 weeks of age. Transplant recipients were female mice at least 12 weeks old. The Subcommittee on Research Animal Care of the Massachusetts General Hospital (MGH) approved all animal work according to federal and institutional policies and regulations. Cells and Cell Culture BM was harvested as previously described (Walkley et al., 2007) and cultured in CFU-C assays according to the manufacturers’ protocols (Stem Cell Technologies). Flow Cytometric Analysis and Sorting of Subpopulations Lineage staining utilized a cocktail of biotinylated anti-mouse antibodies to Mac-1 (CD11b), Gr-1 (Ly-6G and 6C), Ter119 (Ly-76), CD33, CD4, CD8a (Ly-2), and B220 (CD45R) (BD Biosciences). For detection and sorting, we used streptavidin conjugated with either PE/Cy7, APC/Cy7, or PerCP (BD Biosciences). Directly conjugated antibodies used were as follows: cKitAPC (CD117) and Flk2-PE (CD135) (BD Biosciences); Sca1-PE/Cy5.5 (Ly 6A/E, Caltag); and CD34-FITC, CD117-APC/Cy7, CD16/32-Alexa700, and

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

Figure 6. Gene Expression by Quantitative PCR of Sorted Primitive Populations BM was isolated from three independent pairs of WT and Dkk1 tg mice, and LK+S+ (white) or LK+S (gray) populations were sorted prior to mRNA isolation. Gene expression was assayed by qPCR for the indicated targets. Data are expressed as the delta-CT between the gene of interest and the Hprt1 levels expressed in the same Dkk1 tg sample, relative to the delta-CT observed in the paired WT sample. Means from three independent experiments are depicted with SE (Hes-1, p21 p < 0.02).

IL-7Ra-Pacific Blue (eBioscience). Primitive subpopulations were gated as shown in Figures S2A–S2C. For congenic strain discrimination, anti-CD45.1PE and anti-CD45.2 FITC antibodies (BD Biosciences) were used in combination with CD11b-APC and B220-Pe/Cy5. For BrdU incorporation, we used the APC-BrdU Flow Kit (BD Biosciences) according to the manufacturer’s instructions. BM was assayed 16–18 hr after a single intraperitoneal injection of BrdU (100 ul of a 10 mg/ml solution) with subsequent access to 1 mg/ml BrdU in drinking water. Surface staining for lineage markers was performed as above, with Lin-btn/saPerCP, Sca1-PE/Cy5.5, cKit-APC/Cy7, CD135-PE, and CD34FITC. Staining for the primitive BM populations following 5-FU exposure utilized CD150-PE and CD48-FITC conjugates in combination with biotinylated Lineage cocktail and saAPC.

Transplantation Assays For serial transplantation, 1 3 106 whole BM cells from 8 to 12 week old WT and Dkk1 tg (CD45.1) littermates were injected into lethally irradiated (9.5 Gy) female C57Bl/6 (CD45.2) recipients. Reconstitution was monitored by FACS analysis of peripheral blood at 4, 8, and 14 weeks posttransplantation. Fourteen weeks posttransplantation, recipients were used as donors for the subsequent transplantation cycle. Analysis of surviving fractions was calculated using a Kaplan-Meier plot with Prism graphing software. For limiting dilution/ CRU transplants, doses of 1.5 3 105, 5 3 104, or 1.67 3 104 test cells (CD45.1) were mixed with 3.0 3 105 competing WT BM cells and transplanted intravenously into groups of at least nine lethally irradiated WT recipients, as previously described (Janzen et al., 2006). Reconstitution was assessed at weeks 12 and 24 posttransplantation and analyzed with L-Calc software (Stem Cell Technologies). For competitive transplants, 5 3 105 BM cells from individual Dkk1 tg or littermate mice (CD45.1) were transplanted with the same number of WT BM into each of 8–10 lethally irradiated WT recipients. Repopulation was assessed by flow cytometry at 8, 16, and 24 weeks by FACS analysis of peripheral blood. For transplantation to assess acute Dkk1 exposure, 1 3 106 WT BM cells were transplanted into five individual Dkk1 tg or control (CD45.1) lethally irradiated recipients. Reconstitution was assessed by FACS analysis of peripheral blood at 4, 8, and 14 weeks. After 14 weeks, groups of five secondary WT CD45.1 recipients received 9 3 105 donor cells (CD45.2) from an individual primary recipient mixed with 3 3 105 fresh CD45.1 BM. Reconstitution was assessed at 4, 12, and 24 weeks.

5-Fluorouracil Assay for Quiescence To test for retention of a quiescent pool of hematopoietic progenitors, groups of five WT recipients were lethally irradiated and transplanted with 1 million BM cells from individual WT or Dkk1 tg donors (CD45.1) and left to engraft for 16 weeks. A single 150 mg/kg dose of 5-FU (Pharmacia) was administered intraperitoneally on days 0 and 42. Peripheral blood cell composition was monitored by tail vein sampling using a Hemavet 850 (Drew Scientific) at day 0, and every 7 days thereafter. Prior to each 5-FU injection, mice were assayed to confirm complete peripheral blood reconstitution by the original donor CD45.1 population (data not shown). For BM analysis, mice were sacrificed 40–42 hr after 5-FU injection. BM cells were isolated and stained for FACS analysis of primitive BM populations. PCR and Gene Expression Analyses Gene expression was determined as described previously (Janzen et al., 2006). Briefly, RNA was isolated from sorted BM subpopulations using the PicoPure Kit (Arcturus Bioscience). Predesigned assays for Hprt1, Bmi-1, Gfi-1, Hes1, and p21 were purchased from Applied Biosystems (assay IDs Mm00446968, Mm00776122, Mm00515853, Mm00468601, and Mm00432448, respectively). For analysis of Topgal reporter expression, LK+S+ and LK+S populations were sorted from three mice, transgenic either for Topgal alone, or Topgal and Dkk1 tg. Topgal reporter activity was assayed using Applied Biosystems Sybr green kit with the primers (FW) 50 -TAATGTTGATGAAAGCTGGCT-30 and (RV) 50 -ATGCGCTCAGGTCAAATTCAG-30 . Data are presented as the delta-CT between the gene of interest and Hprt1 in a given population of Dkk1 tg cells, relative to the same comparison performed for WT cells. For endogenous and transgenic Dkk1 expression, in addition to sorted LK+S+ populations, total RNA was extracted from tibiae (including marrow cells) using Trizol reagents (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. cDNA was prepared using the SuperScript First-Strand Synthesis system for RTPCR (Invitrogen) and analyzed in triplicate with the SYBR Green Master Mix system (QIAGEN). Primers specific for Dkk1 tg were (FW) 50 -AACCTTACTT CTGTGGTGTGACAT-30 and (RV) 50 -AACCTTACTTCTGTGGTGTGACAT-30 . Primers detecting both endogenous and tg Dkk1 were (FW) 50 -CTTGCGCT GAAGATGAGGAGT-30 and (RV) 50 -GAGGGCATGCATATTCCATTT-30 . Statistical Analysis Where not otherwise stated, the Student’s t test was used to determine the significance, set at p < 0.05. SUPPLEMENTAL DATA Supplemental Data include four figures and can be found with this article online at http://www.cellstemcell.com/cgi/content/full/2/3/274/DC1/. ACKNOWLEDGMENTS

Hoechst/Py Cell-Cycle Analysis Unmanipulated WT and Dkk1 tg BM was incubated with 10 ug/ml Hoechst 33342 for 30 min at 37 C. Cells were washed and stained with antibodies against CD34, cKit, Sca1, and Lin. The cells were fixed in 5% PFA, and incubated with an RNA dye, Pyronin Y (PY, 1 mg/ml), prior to flow cytometry. The proportion of cells in G0 (PYlow, Hoechstlow) was measured in the LK+S+ and LK+S+CD34low gated populations.

The authors would like to thank D. Dombkowski for cell sorting and assistance with multiparameter flow cytometry, R. Klein for mouse colony maintenance, and L. Purton for critical reading of the manuscript. The authors gratefully acknowledge the support of the NHLBI Specialized Center for Cell-based Therapy (HL081030), NHLBI HL044851 and NIDDK DK050234. In addition, H.E.F. was funded by a Ruth L. Kirschstein National Research Service Award

Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc. 281

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

(HL07623). D.T.S. is a founder, consultant, and shareholder of Fate Therapeutics, Inc. Received: October 5, 2007 Revised: December 3, 2007 Accepted: January 2, 2008 Published: March 5, 2008

Jung, Y., Wang, J., Song, J., Shiozawa, Y., Wang, J., Havens, A., Wang, Z., Sun, Y., Emerson, S., Krebsbach, P., and Taichman, R. (2007). Annexin II expressed by osteoblasts and endothelial cells regulates stem cell adhesion, homing, and engraftment following transplantation. Blood 110, 82–90. Kawano, Y., and Kypta, R. (2003). Secreted antagonists of the Wnt signalling pathway. J. Cell Sci. 116, 2627–2634. Kiel, M.J., Yilmaz, O¨.H., Iwashita, T., Yilmaz, O.H., Terhorst, C., and Morrison,

REFERENCES

S.J. (2005). SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 121, 1109–1121.

Adams, G.B., Martin, R.P., Alley, I.R., Chabner, K.T., Cohen, K.S., Calvi, L.M., Kronenberg, H.M., and Scadden, D.T. (2007). Therapeutic targeting of a stem cell niche. Nat. Biotechnol. 25, 238–243.

Kielman, M.F., Rindapaa, M., Gaspar, C., van Poppel, N., Breukel, C., van Leeuwen, S., Taketo, M.M., Roberts, S., Smits, R., and Fodde, R. (2002). Apc modulates embryonic stem-cell differentiation by controlling the dosage of beta-catenin signaling. Nat. Genet. 32, 594–605.

Akala, O.O., and Clarke, M.F. (2006). Hematopoietic stem cell self-renewal. Curr. Opin. Genet. Dev. 16, 496–501. Baba, Y., Yokota, T., Spits, H., Garrett, K.P., Hayashi, S., and Kincade, P.W. (2006). Constitutively active beta-catenin promotes expansion of multipotent hematopoietic progenitors in culture. J. Immunol. 177, 2294–2303. Calvi, L.M., Sims, N.A., Hunzelman, J.L., Knight, M.C., Giovannetti, A., Saxton, J.M., Kronenberg, H.M., Baron, R., and Schipani, E. (2001). Activated parathyroid hormone/parathyroid hormone-related protein receptor in osteoblastic cells differentially affects cortical and trabecular bone. J. Clin. Invest. 107, 277–286. Calvi, L.M., Adams, G.B., Weibrecht, K.W., Weber, J.M., Olson, D.P., Knight, M.C., Martin, R.P., Schipani, E., Divieti, P., Bringhurst, F.R., et al. (2003). Osteoblastic cells regulate the haematopoietic stem cell niche. Nature 425, 841–846. Cheng, T., Rodrigues, N., Shen, H., Yang, Y., Dombkowski, D., Sykes, M., and Scadden, D.T. (2000). Hematopoietic stem cell quiescence maintained by p21cip1/waf1. Science 287, 1804–1808.

Kippin, T.E., Martens, D.J., and van der Kooy, D. (2005). p21 loss compromises the relative quiescence of forebrain stem cell proliferation leading to exhaustion of their proliferation capacity. Genes Dev. 19, 756–767. Kirstetter, P., Anderson, K., Porse, B.T., Jacobsen, S.E., and Nerlov, C. (2006). Activation of the canonical Wnt pathway leads to loss of hematopoietic stem cell repopulation and multilineage differentiation block. Nat. Immunol. 7, 1048–1056. Koc, O.N., Peters, C., Aubourg, P., Raghavan, S., Dyhouse, S., DeGasperi, R., Kolodny, E.H., Yoseph, Y.B., Gerson, S.L., Lazarus, H.M., et al. (1999). Bone marrow-derived mesenchymal stem cells remain host-derived despite successful hematopoietic engraftment after allogeneic transplantation in patients with lysosomal and peroxisomal storage diseases. Exp. Hematol. 27, 1675– 1681. Koch, U., Wilson, A., Cobas, M., Kemler, R., Macdonald, H.R., and Radtke, F. (2008). Simultaneous loss of {beta}- and {gamma}-catenin does not perturb hematopoiesis or lymphopoiesis. Blood 111, 160–164.

Cobas, M., Wilson, A., Ernst, B., Mancini, S.C., MacDonald, H.R., Kemler, R., and Radtke, F. (2004). {beta}-catenin is dispensable for hematopoiesis and lymphopoiesis. J. Exp. Med. 199, 221–229.

Kunisato, A., Chiba, S., Nakagami-Yamaguchi, E., Kumano, K., Saito, T., Masuda, S., Yamaguchi, T., Osawa, M., Kageyama, R., Nakauchi, H., et al. (2003). HES-1 preserves purified hematopoietic stem cells ex vivo and accumulates side population cells in vivo. Blood 101, 1777–1783.

Daniels, D.L., and Weis, W.I. (2005). Beta-catenin directly displaces Groucho/ TLE repressors from Tcf/Lef in Wnt-mediated transcription activation. Nat. Struct. Mol. Biol. 12, 364–371.

Kyba, M., Perlingeiro, R.C., and Daley, G.Q. (2002). HoxB4 confers definitive lymphoid-myeloid engraftment potential on embryonic stem cell and yolk sac hematopoietic progenitors. Cell 109, 29–37.

DasGupta, R., and Fuchs, E. (1999). Multiple roles for activated LEF/TCF transcription complexes during hair follicle development and differentiation. Development 126, 4557–4568.

Li, J., Sarosi, I., Cattley, R.C., Pretorius, J., Asuncion, F., Grisanti, M., Morony, S., Adamu, S., Geng, Z., Qiu, W., et al. (2006). Dkk1-mediated inhibition of Wnt signaling in bone results in osteopenia. Bone 39, 754–766.

Delaney, C., Varnum-Finney, B., Aoyama, K., Brashem-Stein, C., and Bernstein, I.D. (2005). Dose-dependent effects of the Notch ligand Delta1 on ex vivo differentiation and in vivo marrow repopulating ability of cord blood cells. Blood 106, 2693–2699.

MacDonald, B.T., Adamska, M., and Meisler, M.H. (2004). Hypomorphic expression of Dkk1 in the doubleridge mouse: dose dependence and compensatory interactions with Lrp6. Development 131, 2543–2552.

Duncan, A.W., Rattis, F., Di, M., Mascio, L.N., Congdon, K.L., Pazianos, G., Zhao, C., Yoon, K., Cook, J.M., Willert, K., et al. (2005). Integration of Notch and Wnt signaling in hematopoietic stem cell maintenance. Nat. Immunol. 6, 314–322. Grotewold, L., Theil, T., and Ruther, U. (1999). Expression pattern of Dkk-1 during mouse limb development. Mech. Dev. 89, 151–153. Hock, H., Hamblen, M.J., Rooke, H.M., Schindler, J.W., Saleque, S., Fujiwara, Y., and Orkin, S.H. (2004). Gfi-1 restricts proliferation and preserves functional integrity of haematopoietic stem cells. Nature 431, 1002–1007. Iwama, A., Oguro, H., Negishi, M., Kato, Y., Morita, Y., Tsukui, H., Ema, H., Kamijo, T., Katoh-Fukui, Y., and Koseki, H. (2004). Enhanced self-renewal of hematopoietic stem cells mediated by the polycomb gene product Bmi-1. Immunity 21, 843–851. Janzen, V., Forkert, R., Fleming, H.E., Saito, Y., Waring, M.T., Dombkowski, D.M., Cheng, T., DePinho, R.A., Sharpless, N.E., and Scadden, D.T. (2006). Stem-cell ageing modified by the cyclin-dependent kinase inhibitor p16INK4a. Nature 443, 421–426. Jeannet, G., Scheller, M., Scarpellino, L., Duboux, S., Gardiol, N., Back, J., Kuttler, F., Malanchi, I., Birchmeier, W., Leutz, A., et al. (2008). Long-term, multilineage hematopoiesis occurs in the combined absence of {beta}-catenin and {gamma}-catenin. Blood 111, 142–149.

282 Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc.

Mao, B., Wu, W., Li, Y., Hoppe, D., Stannek, P., Glinka, A., and Niehrs, C. (2001). LDL-receptor-related protein 6 is a receptor for Dickkopf proteins. Nature 411, 321–325. Mao, B., Wu, W., Davidson, G., Marhold, J., Li, M., Mechler, B.M., Delius, H., Hoppe, D., Stannek, P., Walter, C., et al. (2002). Kremen proteins are Dickkopf receptors that regulate Wnt/beta-catenin signalling. Nature 417, 664–667. Miyake, N., Brun, A.C., Magnusson, M., Miyake, K., Scadden, D.T., and Karlsson, S. (2006). HOXB4-induced self-renewal of hematopoietic stem cells is significantly enhanced by p21 deficiency. Stem Cells 24, 653–661. Murdoch, B., Chadwick, K., Martin, M., Shojaei, F., Shah, K.V., Gallacher, L., Moon, R.T., and Bhatia, M. (2003). Wnt-5A augments repopulating capacity and primitive hematopoietic development of human blood stem cells in vivo. Proc. Natl. Acad. Sci. USA 100, 3422–3427. Nemeth, M.J., and Bodine, D.M. (2007). Regulation of hematopoiesis and the hematopoietic stem cell niche by Wnt signaling pathways. Cell Res. 17, 746– 758. Nemeth, M.J., Kirby, M.R., and Bodine, D.M. (2006). Hmgb3 regulates the balance between hematopoietic stem cell self-renewal and differentiation. Proc. Natl. Acad. Sci. USA 103, 13783–13788. Nemeth, M.J., Topol, L., Anderson, S.M., Yang, Y., and Bodine, D.M. (2007). Wnt5a inhibits canonical Wnt signaling in hematopoietic stem cells and enhances repopulation. Proc. Natl. Acad. Sci. USA 104, 15436–15441.

Cell Stem Cell Endosteal Wnt Signaling Induces HSC Quiescence

Niehrs, C. (2006). Function and biological roles of the Dickkopf family of Wnt modulators. Oncogene 25, 7469–7481. Niida, A., Hiroko, T., Kasai, M., Furukawa, Y., Nakamura, Y., Suzuki, Y., Sugano, S., and Akiyama, T. (2004). DKK1, a negative regulator of Wnt signaling, is a target of the beta-catenin/TCF pathway. Oncogene 23, 8520–8526. Park, I.K., Qian, D., Kiel, M., Becker, M.W., Pihalja, M., Weissman, I.L., Morrison, S.J., and Clarke, M.F. (2003). Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 423, 302–305. Purton, L.E., and Scadden, D.T. (2007). Limiting factors in murine hematopoietic stem cell assays. Cell Stem Cell 1, 263–270. Randall, T.D., and Weissman, I.L. (1997). Phenotypic and functional changes induced at the clonal level in hematopoietic stem cells after 5-fluorouracil treatment. Blood 89, 3596–3606.

Trowbridge, J.J., Xenocostas, A., Moon, R.T., and Bhatia, M. (2006b). Glycogen synthase kinase-3 is an in vivo regulator of hematopoietic stem cell repopulation. Nat. Med. 12, 89–98. van Os, R., Kamminga, L.M., Ausema, A., Bystrykh, L.V., Draijer, D.P., van Pelt, K., Dontje, B., and de Haan, G. (2007). A limited role for p21Cip1/Waf1 in maintaining normal hematopoietic stem cell functioning. Stem Cells 25, 836–843. Van Zant, G. (1984). Studies of hematopoietic stem cells spared by 5-fluorouracil. J. Exp. Med. 159, 679–690. Walkley, C.R., Olsen, G.H., Dworkin, S., Fabb, S.A., Swann, J., McArthur, G.A., Westmoreland, S.V., Chambon, P., Scadden, D.T., and Purton, L.E. (2007). A microenvironment-induced myeloproliferative syndrome caused by retinoic acid receptor g deficiency. Cell 129, 1097–1110.

Reya, T., Duncan, A.W., Ailles, L., Domen, J., Scherer, D.C., Willert, K., Hintz, L., Nusse, R., and Weissman, I.L. (2003). A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature 423, 409–414.

Yeager, A.M., Levin, J., and Levin, F.C. (1983). The effects of 5-fluorouracil on hematopoiesis: studies of murine megakaryocyte-CFC, granulocyte-macrophage-CFC, and peripheral blood cell levels. Exp. Hematol. 11, 944–952.

Scadden, D.T. (2006). The stem-cell niche as an entity of action. Nature 441, 1075–1079.

Yilmaz, O.H., Valdez, R., Theisen, B.K., Guo, W., Ferguson, D.O., Wu, H., and Morrison, S.J. (2006). Pten dependence distinguishes haematopoietic stem cells from leukaemia-initiating cells. Nature 441, 475–482.

Scheller, M., Huelsken, J., Rosenbauer, F., Taketo, M.M., Birchmeier, W., Tenen, D.G., and Leutz, A. (2006). Hematopoietic stem cell and multilineage defects generated by constitutive beta-catenin activation. Nat. Immunol. 7, 1037–1047.

Yu, X., Alder, J.K., Chun, J.H., Friedman, A.D., Heimfeld, S., Cheng, L., and Civin, C.I. (2006). HES1 inhibits cycling of hematopoietic progenitor cells via DNA binding. Stem Cells 24, 876–888.

Sipkins, D.A., Wei, X., Wu, J.W., Runnels, J.M., Coˆte´, D., Means, T.K., Luster, A.D., Scadden, D.T., and Lin, C.P. (2005). In vivo imaging of specialized bone marrow endothelial microdomains for tumour engraftment. Nature 435, 969– 973.

Zeng, H., Yucel, R., Kosan, C., Klein-Hitpass, L., and Moroy, T. (2004). Transcription factor Gfi1 regulates self-renewal and engraftment of hematopoietic stem cells. EMBO J. 23, 4116–4125.

Trowbridge, J.J., Scott, M.P., and Bhatia, M. (2006a). Hedgehog modulates cell cycle regulators in stem cells to control hematopoietic regeneration. Proc. Natl. Acad. Sci. USA 103, 14134–14139.

Zhang, J., Niu, C., Ye, L., Huang, H., He, X., Tong, W.G., Ross, J., Haug, J., Johnson, T., Feng, J.Q., et al. (2003). Identification of the haematopoietic stem cell niche and control of the niche size. Nature 425, 836–841.

Cell Stem Cell 2, 274–283, March 2008 ª2008 Elsevier Inc. 283