Hedgehog modulates cell cycle regulators in stem cells to control hematopoietic regeneration

Hh Activation Expands Primitive Bone Marrow (BM) Cells.

The BM compartment of ptc-1+/− mice was found to contain significantly greater numbers of hematopoietic progenitors (CFU) compared with WT littermates (Fig. 1a). Ptc-1+/− and WT progenitors had similar capacity to form erythroid, granulocyte-macrophage, and multimyeloid colonies (Fig. 1b), indicating that Hh activation does not preferentially expand progenitors of a specific hematopoietic lineage. In adult mice, the spleen serves as a secondary hematopoietic organ (12). The spleens of ptc-1+/− mice were found to contain similar numbers of hematopoietic progenitors (CFU) compared with WT littermates (Fig. 5, which is published as supporting information on the PNAS web site), suggesting that activation of Hh target genes preferentially expands primitive hematopoietic cells within the BM compartment. Additionally, Hh activation did not affect mature BM populations in terms of total numbers (Fig VSports. 1c) or lineage composition (Fig. 1d).

Primitive hematopoietic cells (including HSCs) in the mouse are phenotypically defined as expressing the c-Kit and Sca-1 cell surface markers, but not mature hematopoietic lineage (Lin) markers, and are termed Lin- Sca-1+ c-Kit+ (LSK) cells (13). The total number and frequency of LSK BM cells was significantly higher in ptc-1+/− compared with WT mice (Fig. 1e) V体育官网. Increased Hh target gene activation in ptc-1+/− LSK BM cells was quantitated, demonstrating a 5-fold greater expression of the target gene Gli-1 (Fig. 1f). These data indicate that increased Hh signaling has little or no effect on mature hematopoietic cell composition but, however, leads to a larger LSK compartment and functional increase in progenitor capacity in the BM under homeostatic conditions.

Hh Activation During Regeneration Exhausts HSC Self-Renewal.

Although LSK serves as a strong surrogate marker for HSCs, an increase in the LSK compartment and progenitor (CFU) capacity is not necessarily indicative of increased functional capacity of HSCs. Because HSCs are defined by their continued self-renewal and multilineage differentiation properties, only damage-induced BM repopulation and regeneration assays are able to assess the functional properties of HSCs.

To examine the impact of increased Hh signaling on the endogenous regenerative ability of HSCs, WT and ptc-1^+/− littermates were administered a single dose of 5-fluorouracil (5-FU), which acts to ablate the hematopoietic system by killing actively cycling cells while sparing quiescent subsets (1). Ptc-1^+/− mice demonstrated accelerated recovery of peripheral blood leukocytes compared with WT mice (Fig. 2a), suggesting that ptc-1^+/− mice have greater numbers of noncycling primitive hematopoietic cells (surviving 5-FU ablation) that are capable of cell cycle reentry for hematopoietic recovery.

Within the BM compartment, populations of stromal cells comprise a specialized niche for HSCs and have an important role in maintenance of the HSC population (14, 15). To test the autonomous function of ptc-1^+/− HSCs distinguished from the effect of ptc-1^+/− cells within the BM microenvironment, HSCs were challenged to regenerate the hematopoietic system of ptc-1^+/+ recipients that have undergone irradiation-induced hematopoietic damage. Primitive (Lin-) BM cells from donor ptc-1^+/− or WT littermates, expressing CD45.2, were isolated and transplanted into irradiated NOD/SCID mice that express CD45.1, thus allowing donor and recipient cells to be distinguished (Fig. 2b). This model follows an appropriate order of regeneration; acute hematopoietic regeneration followed by sustained hematopoiesis that can be serially passaged into secondary recipients. Importantly, the same LSK subset of BM cells is detected in this assay (recipient NOD/SCIDs) as in other repopulation systems (Fig. 6, which is published as supporting information on the PNAS web site). Ptc-1^+/− hematopoietic cells were found to possess greater short-term BM regeneration capacity compared with their WT counterparts at 5 weeks after transplant (Fig. 2c). This result is consistent with an increase in the LSK compartment (Fig. 1e) and progenitor (CFU) capacity (Fig. 1a) of ptc-1^+/− BM observed under steady-state conditions and more rapid recovery of ptc-1^+/− mice after 5-FU hemoablation (Fig. 2a). However, by 8 weeks after transplant, the regenerative capacity of ptc-1^+/− HSCs was significantly reduced, amounting to a 6.7-fold decrease in frequency of donor cells in recipient BM compared with mice transplanted with WT HSCs (Fig. 2d). The hematopoietic graft generated from ptc-1^+/− HSCs at 8 weeks after transplant was comprised of all mature hematopoietic lineages, similar to WT cells (Fig. 2e).

To further examine the regenerative capacity of ptc-1^+/− HSCs engrafting the primary host, we performed serial transplantation assays by using secondary recipients. At 8 weeks, equal numbers of ptc-1^+/− and WT CD45.2+ cells isolated from primary reconstituted mice were used for secondary transplantation and measure of CFU capacity arising from ptc-1^+/− and WT HSCs (Fig. 2b). Ptc-1^+/− HSCs gave rise to significantly fewer progenitors (CFU) compared with WT (Fig. 2f). In secondary recipients, serially transplanted ptc-1^+/− HSCs had significantly reduced regenerative ability (Fig. 2 g and h) and reduced progenitor (CFU) output (Fig. 2i) compared with WT HSCs. Two factors known to be important for stem cell self-renewal, Bmi-1 (16, 17) and Bmp-4 (6), have been shown to be induced by Hh signaling. Expression of the Bmp-4 transcript was increased in ptc-1^+/− primitive Lin- Sca-1+ repopulating cells, whereas expression of the Bmi-1 transcript was not significantly altered (Fig. 7, which is published as supporting information on the PNAS web site). These data indicate that the reduced regenerative capacity of HSCs was not due to a failure to induce Bmp-4, and although we did not find Bmi-1 to be up-regulated, similar levels of expression between ptc-1^+/− and WT indicate that the ptc-1^+/− repopulating cells do not have defects in expression of this important HSC regulator.

Taken together, the loss of one copy of ptc-1 leads to (i) an increase in the LSK compartment and functionally increases (ii) progenitor capacity, (iii) noncycling cells capable of hematopoietic regeneration, and (iv) short-term reconstituting ability. The increase in these functional properties seems to be at the expense of HSCs, resulting in an overall loss in HSC regenerative ability as evidenced by the inability to (i) sustain hematopoiesis, (ii) produce progenitors, and (iii) regenerate hematopoiesis upon secondary challenge.

Hh Activation Modulates Specific Cell Cycle Regulators Leading to HSC Exhaustion.

Because Hh signaling has been closely associated with cell cycling (3), we examined whether abnormal cell cycle regulation contributes to the loss of ptc-1^+/− HSC regenerative ability. Detailed examination of the in vivo proliferative capacity of LSK cells in ptc-1^+/− BM under homeostatic conditions (Fig. 3a) revealed an increase in the number of cycling LSK cells compared with WT (Fig. 3b). Concomitantly we observed increased expression of CyclinD1 (Fig. 8a, which is published as supporting information on the PNAS web site), which is involved in G₁ phase progression and is induced in certain tissues by Hh stimulation (18). When ptc-1^+/− primitive (Lin-) hematopoietic cells were transplanted into recipient mice after hematopoietic damage, the total number of cycling ptc-1^+/− Lin- Sca-1+ cells was greater than WT at 5 weeks after transplant (Fig. 3c), consistent with the greater short-term regenerative capacity of ptc-1^+/− hematopoietic cells (Fig. 2c). By 8 weeks after transplant, significantly fewer ptc-1^+/− Lin- Sca-1+ cells were cycling as compared with WT (Fig. 3d), consistent with reduced hematopoietic regeneration (Fig. 2d). Our data suggest that increased Hh signaling results in increased cycling and expansion of LSK cells in adult BM under homeostatic conditions and during acute hematopoietic regeneration. However, continued Hh activity results in exhaustion of the cycling capacity of HSCs and failure of cells to regenerate hematopoiesis upon secondary transplant.

To identify cell cycle regulators affected by the ptc-1^+/− genotype, primitive Lin- CD45.2+ WT or ptc-1^+/− donor cells were isolated at 8 weeks after transplant and the expression pattern of 96 cell cycle-related genes (Table 1, which is published as supporting information on the PNAS web site) were compared by array hybridization (Fig. 3e). The analysis identified six cell-cycle regulators (Fig. 3f) and three genes involved in DNA damage repair (Fig. 3g) that differ in expression by >3-fold. Transcripts for all nine differentially expressed genes were found to be lower in ptc-1^+/− primitive Lin- repopulating cells compared with WT. These transcripts were not differentially expressed in de novo isolated WT and ptc-1^+/− LSK cells (Fig. 3 f Inset and g Inset). Comparing the kinetics of expression of these nine genes in WT and ptc-1^+/− primitive cells during BM homeostasis and during hematopoietic regeneration after transplant revealed two subcategories of genes (Fig. 9, which is published as supporting information on the PNAS web site). The first category, including Tfdp2, Skp1a, CyclinA2, and Rad51, are down-regulated in both WT and ptc-1^+/− cells during hematopoietic regeneration but are down-regulated to a greater degree in ptc-1^+/− cells compared with WT (Fig. 9a). This difference indicates that ptc-1^+/− cells are unable to sustain expression levels necessary for hematopoietic regeneration. The second category, including CyclinH, Cdc16, Cdc2a, Mre11a, and Rpa3, are up-regulated in WT but down-regulated in ptc-1^+/− cells during hematopoietic regeneration (Fig. 9b), suggesting that ptc-1^+/− cells are unable to up-regulate expression of these genes as necessary for hematopoietic regeneration. These data identify a specific profile of cell cycle-related genes that are deregulated in ptc-1^+/− HSCs challenged to regenerate hematopoiesis, thereby providing a molecular basis to further characterize this HSC defect.

In Vivo Rescue of ptc-1^+/− HSC Functional and Transcriptional Defects By Using Cyclopamine.

To evaluate whether regenerative deficiencies and cycling of ptc-1^+/− HSCs were a specific consequence of Hh pathway activation, we carried out in vivo chemical “complementation” by using cyclopamine, a plant-derived teratogen that binds to the Smo protein and inhibits Hh signal transduction (19). i.p. injection of cyclopamine was effective in reducing Hh signaling in ptc-1^+/− LSK cells (Fig. 4a Inset) and reduced the CFU capacity of ptc-1^+/− BM cells to within WT levels (Fig. 10, which is published as supporting information on the PNAS web site), demonstrating that the expansion of progenitors in ptc-1^+/− BM is a result of increased Hh signaling. Because Hh signaling has been proposed to interact with other signaling pathways important for HSC proliferation, including Wnt and Notch (20), we examined the consequence of altered Hh signaling activity on these pathways. Administration of cyclopamine did not alter activation of the Wnt pathway (Fig. 4a Inset). The expression of Wnt and Notch pathway target genes, Axin2 and Hes-1, respectively, in WT and ptc-1^+/− LSK cells were similar (Fig. 8b), illustrating both cyclopamine specificity and that modulation of Hh signaling does not impact Axin2 and Hes-1 in this cell subset. This finding is consistent with basal skin cells, where elimination of ptc-1 does not promote nuclear translocation of β-catenin or induce ectopic activation or expression of Notch pathway components (21).

To determine the effects of reduced Hh signaling on ptc-1^+/− HSC cycling, self-renewal, and regenerative capacity, recipient mice transplanted with ptc-1^+/− cells were administered cyclopamine twice per week for the 8-week transplant period (Fig. 4a). Cyclopamine administration significantly restored the regenerative capacity of ptc-1^+/− HSCs, rescuing total reconstitution in recipient mice to 79% of levels observed for WT (Fig. 4b) without affecting total numbers of endogenous (CD45.1+) hematopoietic cells (Fig. 4b Inset). The total number of cycling ptc-1^+/− Lin- Sca-1+ cells at 8 weeks after transplant also was restored by cyclopamine treatment to 60% of WT levels (Fig. 4c). Ptc-1^+/− HSCs treated in vivo with cyclopamine in primary transplant recipients possessed significantly higher secondary regenerative ability in the absence of continued cyclopamine treatment (Fig. 4d). Furthermore, expression of the nine cell cycle-related genes down-regulated in ptc-1^+/− repopulating cells were restored to WT levels after cyclopamine treatment in vivo (Fig. 4 e and f). These results demonstrate that administration of cyclopamine is sufficient to rescue ptc-1^+/− HSC regenerative ability and provides molecular and cellular evidence for the specificity of Hh signaling in regulating cell cycle machinery that controls HSC regenerative capacity in vivo.

Mice. (V体育平台登录)

ptc-1^+/− (9), C57BL/6, TOP-gal (25), and NOD/SCID mice were used at 8–12 weeks of age. Ptc-1^+/− mice were housed in the University of Western Ontario (UWO) animal care facility and remaining mice housed in the Robarts Research Institute barrier facility (London, ON, Canada). Procedures and protocols were approved by the UWO Animal Care Council.

BM and Spleen Analysis.

Cells were harvested from the femurs, tibiae, and spleen of mice. Mononuclear cells were counted, stained with fluorescent-conjugated antibodies against hematopoietic cell surface markers, and analyzed on a FACSCalibur (Becton Dickinson, Franklin Lakes, NJ). Lineage-depleted (Lin-) cells were purified from BM by using StemSep murine hematopoietic progenitor enrichment kit (StemCell Technologies, Vancouver, BC, Canada). Cells expressing high levels of c-Kit and Sca-1 were isolated by FACSVantage SE (Becton Dickinson). For the CFU assay, equal numbers of BM or spleen cells were plated into MethoCult GF M3434 (StemCell Technologies), and colonies were scored after 10–12 days incubation at 37°C and 5% CO₂.

Real-Time PCR.

Total RNA was extracted from >30,000 LSK cells using the RNeasy kit (Qiagen, Mississauga, ON, Canada). First-strand cDNA synthesis was performed (First-Strand cDNA Synthesis Kit; Amersham Biosciences, Piscataway, NJ), and the resulting cDNA was analyzed for differential gene expression by using SYBRGREEN double-stranded DNA binding dye and the Mx4000 Multiplex Quantitative PCR System (Stratagene, La Jolla, CA). Comparative quantitation of transcripts was assessed relative to GAPDH. Primers are listed in Table 2, which is published as supporting information on the PNAS web site.

5-FU Hemoablation.

5-FU (Sigma, Oakville, ON, Canada) was administered to mice by i.p. injection at 150 mg/kg body weight. Leukocyte counts were performed on peripheral blood taken from the tail vein before injection, 24 h and 9 days after injection.

In Vivo HSC Repopulation Assay.

One hundred thousand Lin- donor CD45.2+ cells from ptc-1^+/− or WT littermates were transplanted by tail vein injection into sublethally irradiated (350 rads, ¹³⁷Cesium) NOD/SCID mice. Mice received i.p. injections of 25 mg/kg body weight cyclopamine (Toronto Research Chemicals, North York, ON, Canada) or vehicle control (DMSO) twice per week for the duration of the transplant period. Mice were killed at 5 and 8 weeks after transplantation. Engraftment was assessed by the frequency of CD45.2+ donor cells, along with analysis of multilineage hematopoietic cell surface makers, on a FACSCalibur (Becton Dickinson). 7AAD (Becton Dickinson) was used to assess frequency of dead cells. CD45.2+ cells were isolated by FACSVantage (Becton Dickinson) at 8 weeks after transplant and equal numbers of cells were plated in the CFU assay. For secondary repopulation, 250,000 donor CD45.2+ cells were transplanted by tail vein injection into secondary sublethally irradiated NOD/SCID mice, and engraftment was assessed at 6 weeks after transplant.

Cell Cycle and Gene Expression Analysis. (V体育2025版)

Mice were administered 200 μl of 10 mg/ml BrdU (Becton Dickinson) by i.p. injection at 12 and 24 h before killing. Isolated CD45.2+ Lin- BM cells were stained with fluorescent-conjugated antibodies against c-Kit and Sca-1, BrdU (Becton Dickinson), 7-AAD, and analyzed on a FACSCalibur. For cell cycle array analysis, Lin- donor CD45.2+ cells were isolated by FACSVantage (Becton Dickinson) at 8 weeks after transplant representing three groups: WT donor, ptc-1^+/− donor, and ptc-1^+/− donor with cyclopamine-treated recipient (three biological replicates per treatment). Total RNA was extracted from >10,000 cells by using the RNeasy kit (Qiagen). Two rounds of RNA amplification were performed by using the MessageAmp II aRNA Kit (Ambion, Austin, TX). Amplified RNA was biotin-labeled and hybridized to the GEArray Q Series Mouse Cell Cycle Gene Array (SuperArray Bioscience Corp., Frederick, MD). For analysis, background levels of hybridization were subtracted, and gene expression was normalized to levels of the housekeeping gene GAPDH in each individual sample. Gene expression of candidates also was quantitated by real-time PCR in three technical replicates per sample, also normalized to GAPDH. Data points shown represent mean values.

"VSports在线直播" Fluorescein di-β-d-Galactopyranoside (FDG) Staining.

TOP-gal (for Wnt signaling) and ptc-1^+/− (for Hh signaling) mice were administered 25 mg/kg cyclopamine (Toronto Research Chemicals) or vehicle control by i.p. injection at 12 and 24 h before killing. Lin- BM cells were isolated in PBS/5% FBS and incubated 2 min at 37°C with an equal volume of 2 mM FDG reagent (Marker Gene Technologies Inc., Eugene, OR) in distilled water. FDG loading was terminated by addition of cold PBS/FBS. Cells then were stained with fluorescent-conjugated antibodies against c-Kit and Sca-1, 7-AAD (Becton Dickinson) and analyzed on a FACSCalibur (Becton Dickinson). The mean fluorescence intensity of FDG was examined to determine the level of Wnt- and Hh-signaling activation in 7AAD- KLS cells.

"V体育2025版" Statistical Analysis.

Data were analyzed by paired, two-tailed Student t tests, and results were considered significant when P ≤ 0.05. Error bars represent SEM.