Recombinant IL-7/HGFβ efficiently induces transplantable murine hematopoietic stem cells

Difficulty obtaining sufficient hematopoietic stem cells (HSCs) directly from the donor has limited the clinical use of HSC transplantation. Numerous attempts to stimulate the ex vivo growth of purified HSCs with cytokines and growth factors generally have induced only modest increases in HSC numbers while decreasing their in vivo reconstituting ability. We previously developed a recombinant single-chain form of a naturally occurring murine hybrid cytokine of IL-7 and the β chain of hepatocyte growth factor (rIL-7/HGFβ) that stimulates the in vitro proliferation and/or differentiation of common lymphoid progenitors, pre-pro-B cells, and hematopoietic progenitor cells (day 12 spleen colony-forming units) in cultures of mouse BM. Here we used the rIL-7/HGFβ in culture to induce large numbers of HSCs from multiple cell sources, including unseparated BM cells, purified HSCs, CD45^– BM cells, and embryonic stem cells. In each instance, most of the HSCs were in the G₀ phase of the cell cycle and exhibited reduced oxidative stress, decreased apoptosis, and increased CXCR4 expression. Furthermore, when injected i.v., these HSCs migrated to BM, self-replicated, provided radioprotection, and established long-term hematopoietic reconstitution. These properties were amplified by injection of rIL-7/HGFβ directly into the BM cavity but not by treatment with rIL-7, rHGF, and/or rHGFβ.

An elusive goal in BM reconstitution therapy has been the establishment of conditions under which large numbers of transplantable hematopoietic stem cells (HSCs) can be selectively generated in vitro (1–4). Two major approaches using purified HSCs have been taken. In the first, gene transfer techniques featuring the overexpression of NOTCH-1, WNT, or HOXB4 were used to influence the survival and proliferation of HSCs in vitro (5–7). Although successful in several instances, this strategy has raised safety concerns that thus far have prohibited its clinical use.

In the second approach, a number of growth factors and soluble proteins were used to expand HSCs in vitro. Of the numerous cytokines that have been tested, none has thus far been able to stimulate the generation of sufficient functional HSCs for clinical application. This negative effect has been attributed to the decline or loss of long-term in vivo repopulating abilities (4, 5, 8) secondary to the entrance of HSCs into cell cycle (9–11). Therefore, only cytokines that maintain HSCs in G₀ phase, such as SDF-1, TGF-β, FGF, and angiopoietin-1, have been found to preserve their engraftment ability (12–15).

Unfortunately, even these factors, whether used alone or in combination, were unable to stimulate sufficient HSC formation. Notably, the retroviral transduction of 5-FU–treated BM cells with the receptor for FGF-1 has shown promise (16). However, as FGF-1 has been shown to act indirectly by inducing other cells to stimulate HSCs, it is not known whether such an approach would be successful in cultures of purified HSCs.

Another approach to generating HSCs in vitro has used embryonic stem cells (ESCs), which are less prone than HSCs to replicative senescence and differentiation in vitro. However, the HSCs that have been produced by ESCs in stromal cell–dependent cultures lacked the ability to reconstitute the hematopoietic system efficiently when infused in vivo, possibly because they were not adult-type (definitive) HSCs (17).

We have demonstrated that a recombinant single-chain form of a naturally occurring, BM stromal cell–derived, hybrid cytokine consisting of murine IL-7 and the β chain of the hepatocyte growth factor (rIL-7/HGFβ) stimulates the expansion of day 12 spleen colony-forming units (CFU-S₁₂), common lymphoid progenitors (CLPs), pre-pro-B cells, and thymocytes in vitro (18) and greatly enhances the thymopoiesis and naive T cell regeneration after BM transplantation in vivo (19). In this study, we show that rIL-7/HGFβ, but not its component cytokines, also stimulated the generation of large numbers of definitive HSCs in long-term cultures of adult BM cells. The results indicated that rIL-7/HGFβ maintains the hematopoietic reconstituting potential of these HSCs by inhibiting their proliferation, reducing oxidative stress and apoptosis, and increasing CXCR4 expression. Moreover, the fact that almost all of the culture-generated HSCs were in G₀ phase raised the possibility that they had been generated by earlier precursors. This was confirmed by demonstrating that rIL-7/HGFβ induced the rapid generation of HSCs in short-term cultures of purified CD45^–Lin^–Sca-1⁺ BM cells, which have been reported to contain the most primitive cells in adult BM, including some that could serve (albeit inefficiently) as the precursors of HSCs in vivo (20–24). Furthermore, we demonstrated that rIL-7/HGFβ-responsive precursors of HSCs resided within the SSEA⁺IL-7Rα⁺c-Met⁺ subset of the CD45^–Lin^– BM cells.

In addition to stimulating the generation of HSCs from CD45^– precursors, rIL-7/HGFβ was able to prevent proliferating HSCs from losing their in vivo repopulating abilities. This permitted large numbers of transplantable HSCs to be produced in rSCF- and rFlt3L-stimulated short-term cultures of unfractionated BM cells or purified HSCs. rIL-7/HGFβ was also able to induce murine ESCs to selectively form HSCs in OP-9 stromal cell cocultures, apparently through a CD45^–Lin^– SSEA⁺IL-7Rα⁺c-Met⁺ intermediate. In all of these systems, the culture-generated HSCs competitively established prolonged, multilineage hematopoietic cell chimerism in vivo. However, the ESC-derived HSCs required the injection of IL-7/HGFβ intrafemorally (i.f.) to establish hematopoietic chimerism, whereas this procedure helped to optimize, but was not essential for, the establishment of chimerism by the adult BM cell–derived HSCs. These in vitro and in vivo effects of the rIL-7/HGFβ fusion protein appeared to be unique, as they could not be replicated in whole or part by rIL-7, rHGF, and/or rHGFβ.

rIL-7/HGFβ stimulates the in vitro production of CD45⁺Lin^–Sca-1⁺c-Kit⁺ cells in long-term BM cell cultures. As we reported (18), mouse BM cells cultured in vitro for 3 to 4 weeks in the presence of rIL-7/HGFβ generated large numbers of CFU-S₁₂, CLPs, pre-pro-B cells, and pro-B cells in vitro (data not shown). However, as shown in Figure 1, A–C, when the culture period was extended to 2 months, the number of CD19⁺ B-lineage cells decreased markedly and the numbers of CD45^–Lin^– (presumptive nonhematopoietic) cells and CD45⁺Lin^–Sca-1⁺c-Kit⁺ (CD45⁺LSK) (presumptive hematopoietic stem and progenitor) cells increased markedly (5,000 and 14,000 cells/well, respectively), exceeding those in the original BM cell inoculum by 15- to 20-fold. More than 90% of the CD45^–Lin^– cells incorporated BrdU within 45 minutes, and less than 2% were in G₀ phase when stained with Ki-67 (Figure 1D). In contrast, less than 1% of the CD45⁺LSK cells (approximately 65% having the Flt3^– phenotype of HSCs and 35% having the Flt3⁺ phenotype of multipotent progenitors [MPPs]) incorporated BrdU and were in G₀ phase (Figure 1E). These results raised the possibility that most of the HSCs in these cultures had been generated by rapidly proliferating precursors, possibly among the CD45^–Lin^– cell population, and that the MPPs had differentiated from these HSCs, possibly without proliferation. Furthermore, the effect of rIL-7/HGFβ on the generation of HSCs and MPPs in these cultures appeared to be unique, as PBS, rIL-7, and/or rHGF (or HGFβ; data not shown) supported the generation of CD45^–Lin^–, but not CD45⁺LSK, cells.

Figure 1

rIL-7/HGFβ induces the appearance of CD45⁺LSK cells in long-term BM cultures. Normal adult mouse BM cells were cultured in the presence of rIL-7/HGFβ (30 ng/ml), rIL-7 (10 ng/ml), and/or rHGF (25 ng/ml), and the total numbers of (A) CD19⁺ B cells, (B) CD45^–Lin^– cells, and (C) CD45⁺LSK cells were determined at 1 and 2 months. Results represent the mean ± SD of the total cells of each type derived from 1 × 10⁶ BM cells. *P < 0.05 as compared with PBS control. In addition, the percentages of the (D) CD45^–Lin^– and (E) CD45⁺LSK cells that incorporated BrdU (45-minute pulse) and expressed Ki-67 were determined for the 2 month rIL-7/HGFβ-containing cultures. One representative experiment out of 3 is shown.

Cells from rIL-7/HGFβ-stimulated BM cell cultures establish long-term hematopoietic chimerism in vivo. Competitive repopulating assays were conducted to determine whether the rIL-7/HGFβ-stimulated BM cells could reconstitute the hematopoietic system of lethally irradiated mice. In these experiments, 3 × 10⁴ cells from 2-month-old cultures of rIL-7/HGFβ-, rIL-7-, and/or rHGF-stimulated (or HGFβ-stimulated) EGFP Tg BM cells were mixed with 2 × 10⁵ freshly harvested WT BM cells and injected i.v. into 10 Gy–irradiated WT recipients. Twelve weeks later, the blood was analyzed for donor-origin (GFP⁺) nucleated cells. As shown in Figure 2, only mice that received cells from the rIL-7/HGFβ-containing cultures displayed multilineage hematopoietic chimerism. In other experiments (Table 1), hematopoietic chimerism (14%–22% GFP⁺ cells) was observed in a variety of tissues (e.g., BM, blood [BL], spleen [SP], LN) as late as 21 weeks after adoptive transfer of 2-month-old cultures of rIL-7/HGFβ-stimulated BM cells; and the proportions of the major donor-origin leukocyte subsets generated were similar to those obtained when freshly harvested (noncultured) GFP⁺ BM cells were used as controls. Furthermore, the donor-origin BM cells in the experimental and control recipients demonstrated equivalent hematopoietic renewal potentials 12 weeks after i.v. injection into secondary irradiated recipients (Table 2).

Figure 2

In vivo hematopoietic repopulating potential of cells from long-term rIL-7/HGFβ-containing BM cultures. 3 × 10⁴ cells from 2-month cultures of rIL-7/HGFβ, rIL-7, and/or rHGF-stimulated BM cells were mixed with 2 × 10⁵ freshly harvested WT BM cells and injected i.v. into lethally irradiated WT recipients. Twelve weeks later, the percentages of donor-origin (GFP⁺) cells in the blood of mice that had received cells from cultures containing (A) rIL-7, (B) rHGF, (C) rIL-7 plus rHGF, or (D) rIL-7/HGFβ were determined. The presence of (E) T cells (CD3⁺) and B cells (CD19⁺) and (F) myeloid cells (Gr-1^–CD11b⁺) and granulocytes (Gr-1⁺) as well as (G) the numbers of these cells among the donor-origin cells is shown. One representative experiment out of 3 is shown.

Table 1

Hematopoietic chimerism in various tissues of primary recipients of culture-generated BM cells

Table 2

Hematopoietic chimerism in various tissues of secondary recipients of culture-generated BM cells

To demonstrate more directly that the culture-generated HSCs were self-renewing, BM cells from the primary recipients were transferred at 14 weeks into irradiated secondary recipients, and the Lin^– donor-origin BM cells were examined 4 weeks later. The results in Figure 3A show that approximately 15% of the total Lin^– cells in the BM of the secondary recipients were of donor origin. This was approximately twice that present in the BM that was transferred from the primary recipients. Furthermore, the subset distribution of the donor-origin Lin^– BM cells in the secondary recipients resembled that in the BM from the primary recipients, with the exception of a significant increase in the proportion of HSCs (Figure 3B).

Figure 3

Self-renewal of HSCs in recipients of cells from long-term rIL-7/HGFβ-containing BM cultures. 20 × 10⁶ cells from 10 Gy–irradiated mice that had received mixtures of culture-generated and WT BM cells i.v. 14 weeks previously (primary recipients) were injected i.v. into irradiated WT mice (secondary recipients). (A) The percentage of Lin^– cells of donor-origin (GFP⁺) and (B) their subset distribution in the BM of the secondary recipients was determined 4 weeks later. Mean ± SD of 4 to 6 mice per group. *P < 0.05 as compared with BM cells from the primary recipients. HSC, Flt3^–LSK; MPP, Flt3⁺LSK; CLP, IL-7Rα⁺LSK; MEP, Flt3⁺Sca-1^–c-kit⁺. One representative experiment out of 2 is shown.

It should be noted that no Ly5.2⁺GFP⁺ cells were observed in any of the chimeras (data not shown), indicating that fusion between donor-origin (Ly5.1⁺GFP⁺) and recipient (Ly5.2⁺GFP^–) cells did not contribute to the observed hematopoietic reconstitution (25, 26). Importantly, no tumor formation was seen in any of the recipients observed for up to 8 months.

i.f. injection of rIL-7/HGFβ enhances the migration to BM and establishment of hematopoietic chimerism by culture-generated HSCs. Because the production of IL-7 and other cytokines is reduced in the irradiated BM (27–29), it was possible that the ability of the culture-generated HSCs to home to, implant in, and/or repopulate the host BM is suboptimal. Therefore, we determined whether the direct i.f. injection of rIL-7/HGFβ (500 ng) shortly before (and at monthly intervals thereafter) the i.v. transfer of cells from 2-month-old cultures of rIL-7/HGFβ-, rIL-7-, and/or rHGF-stimulated (or HGFβ-stimulated) BM would enhance the establishment of long-term hematopoietic chimerism. As shown in Figure 4A, the i.f. injection rIL-7/HGFβ, as opposed to PBS, caused an approximately 2-fold increase in the level of hematopoietic cell chimerism (75% vs. 35%; P < 0.05) attained 12 weeks after adoptive transfer of cells from the rIL-7/HGFβ-containing cultures. In contrast, i.f.-injected rIL-7/HGFβ did not enable BM cells from rIL-7– and/or rHGF-containing cultures to establish hematopoietic chimerism. Furthermore, the i.f. injection of rIL-7 and/or rHGF did not enhance the ability of BM cells from rIL-7/HGFβ-, rIL-7–, and/or HGF-containing cultures to do so (data not shown). Hence, the enhancing effect of i.f.-injected cytokine was unique to rIL-7/HGFβ and appeared to be specific for HSCs from rIL-7/HGFβ-containing cultures.

Figure 4

i.f. injection of rIL-7/HGFβ enhances the establishment of hematopoietic and thymic chimerism by cells from long-term rIL-7/HGFβ-containing BM cultures. GFP⁺ BM cells, cultured for 2 months in the presence of rIL-7/HGFβ, rIL-7, and/or rHGF, were mixed with freshly harvested WT BM cells and injected i.v. into lethally irradiated WT recipients that had also been injected i.f. with rIL-7/HGFβ (500 ng) initially and at monthly intervals thereafter. (A) The levels of multilineage hematopoietic chimerism in the blood and (B) thymocyte chimerism were determined 12 weeks after cell transfer. Mean ± SD from 4 to 6 mice per group. *P < 0.05 as compared with PBS control. One representative experiment out of 3 is shown.

Given our prior demonstration (19) that systemically injected rIL-7/HGFβ significantly enhanced thymocytopoiesis after syngeneic BM transplantation, we determined whether i.f.-injected rIL-7/HGFβ would have a similar effect using culture-generated HSCs. As shown in Figure 4B, i.f. rIL-7/HGFβ induced a 2-fold increase in thymocyte chimerism over i.f. PBS (27% vs. 13%; P < 0.05). This suggested that one of the supportive effects of the systemic administration of rIL-7/HGFβ on thymopoiesis occurs at least partly in the BM (19).

Enhanced migration of hematopoietic stem and progenitor cells from the blood to the BM could be one of the mechanisms by which i.f. injection of rIL-7/HGFβ increases hematopoietic and thymopoietic chimerism. To test this, mice were injected i.f. with rIL-7/HGFβ and i.v. with cultured BM cells. As shown in Figure 5, i.f. injection of rIL-7/HGFβ significantly (P < 0.05) increased the number of rIL-7/HGFβ (but not rIL-7 or rHGF) culture-generated cells in the BM by 15- to 20-fold within 24 hours.

Figure 5

i.f. injection of rIL-7/HGFβ induces the migration of cells from long-term rIL-7/HGFβ-containing BM cultures from blood to BM. Lethally irradiated WT recipients were injected i.f. with 500 ng rIL-7/HGFβ and i.v. with GFP⁺ BM cells cultured for 2 months in the presence of PBS, rIL-7/HGFβ, rIL-7, and/or rHGF (5 × 10⁵). GFP⁺ cells in the injected femurs were analyzed 24 hours later. Mean ± SD of groups of 3 mice. One representative experiment out of 3 is shown.

SCF and Flt3L accelerate the production of transplantable HSCs in short-term rIL-7/HGFβ-containing BM cultures. Early-acting cytokines, such as stem cell factor (SF) and Flt3 ligand (FL), are able to stimulate the proliferation of HSCs in vitro (30). However, these HSCs lose most of their in vivo hematopoietic repopulating capacity when they enter cell cycle. We therefore determined whether rIL-7/HGFβ, which does not induce the proliferation of HSCs in vitro (Figure 1E), could preserve the long-term repopulating capacity of HSCs generated in the presence of SF and FL. In these experiments, 1 × 10⁶ normal BM cells or 0.5 × 10³ to 2 × 10³ purified HSCs (Flt3^–LSK cells) therefrom were cultured for 7 to 12 days with a mixture of SF and FL (SF/FL) in the presence or absence of rIL-7/HGFβ, rIL-7, and/or rHGF. Under these circumstances, all of the cultures contained elevated numbers of CD45⁺LSK cells (2 × 10⁴ to 3 × 10⁴), suggesting that, unlike long-term rIL-7/HGFβ-containing BM cell cultures, most of the HSCs in the short-term SF/FL-containing BM cell cultures were generated by the proliferation of preexisting HSCs. Yet, as shown in Figure 6, only the cells from the rIL-7/HGFβ-containing cultures were able to establish long-term (22–24 weeks) hematopoietic chimerism in adoptive recipients.

Figure 6

rIL-7/HGFβ maintains the hematopoietic reconstituting activity of cells generated in short-term cultures of SF/FL-stimulated BM cells. BM was cultured with 50 ng/ml SF and FL in the presence or absence of rIL-7/HGFβ, rIL-7, and/or rHGF. The proportions of donor-origin (GFP⁺) hematopoietic cells in the blood of sublethally irradiated (6 Gy) recipients 22 to 24 weeks after the i.v. transfer of 2 × 10⁵ culture-generated and 2 × 10⁵ WT BM cells were determined. As indicated, some of the recipients also were injected i.f. with rIL-7/HGFβ. Mean ± SD of 4 to 6 mice per group. One representative experiment out of 2 is shown.

At least 4 differences between the rIL-7/HGFβ-containing cultures and those not containing rIL-7/HGFβ might help to explain these results. First, the LSK cells in the rIL-7/HGFβ-containing cultures expressed significantly lower levels of ROS than did their counterparts in non–rIL-7/HGFβ-containing cultures (Figure 7A). Second, a significantly lower proportion of the LSK cells in the rIL-7/HGFβ-containing cultures incorporated BrdU (Figure 7B). Third, a significantly lower proportion of the LSK cells in the rIL-7/HGFβ-containing cultures were annexin V–positive (Figure 7C), including propidium iodide–positive and –negative subsets. Fourth, a significantly higher proportion of the LSK cells in the rIL-7/HGFβ-containing cultures expressed CXCR4 (Figure 7D). These effects were restricted to rIL-7/HGFβ and were not seen in cultures containing the individual factors rIL-7 and/or rHGFβ (data not shown). Hence, the addition of rIL-7/HGFβ to SF/FL-containing cultures appeared to maintain the hematopoietic reconstituting potential of these proliferating HSCs, at least in part, by protecting them against oxidative stress (possibly by enabling them to revert to the quiescent state; ref. 31), enhancing their survival, and increasing their ability to home to the BM (32).

Figure 7

rIL-7/HGFβ decreases oxidative stress and proliferation of LSK cells generated in short-term cultures of SF/FL-stimulated BM cells and increases their survival and expression of CXCR4. BM was cultured for 7 days in SF/FL in the presence or absence of rIL-7/HGFβ as in Figure 6, and the (A) relative amounts of ROS per LSK cell, (B) percentage of viable LSK cells that incorporated BrdU after a 45-minute pulse, (C) percentage of total LSK cells that were annexin V positive, and (D) percentage of viable LSK cells expressing CXCR4 were determined. Results represent mean ± SD of data pooled from 3 independent experiments. *P < 0.05 as compared with SF/FL only.

rIL-7/HGFβ induces purified CD45^–Lin^–Sca-1⁺ BM cells and the CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met ⁺ subset therein to generate HSCs in vitro. To test the hypothesis first suggested by the data in Figure 1 that most of the CD45⁺LSK cells in long-term rIL-7/HGFβ-containing BM cell cultures are generated by CD45^– precursors, we purified (>98%) CD45^–Lin^–Sca-1⁺ cells from normal BM and placed them in culture for 12 days in the presence of rIL-7/HGFβ, rIL-7, and/or rHGF (or PBS). As with the long-term BM cell cultures (Figure 1C), only the rIL-7/HGFβ-containing short-term cultures of purified CD45^–Lin^–Sca-1⁺ BM cells generated CD45⁺LSK cells. However, unlike the long-term BM cell cultures, CD45⁺LSK cells comprised approximately 30% (as opposed to 1%–2%) of the total cells in the short-term cultures of purified CD45^–Lin^–Sca-1⁺ BM cells (Figure 8A). Of these, approximately 75% were putative HSCs (Flt3^–) and 25% were MPPs (Flt3⁺) (data not shown). Again, only the cells from the rIL-7/HGFβ-containing cultures competitively repopulated the hematopoietic system (Figure 8B) and provided radioprotection (data not shown) in vivo. In contrast, neither hematopoiesis nor radioprotection occurred when freshly purified CD45^–Lin^–Sca-1⁺ BM cells themselves were placed directly in vivo (Figure 8B and data not shown).

Figure 8

rIL-7/HGFβ induces the appearance of HSCs in short-term cultures of CD45^–Lin^–Sca-1⁺ cells. FACS-purified (≥98%) CD45^–Lin^–Sca-1⁺ cells, freshly harvested from normal BM, were cultured for 12 days in the presence or absence of rIL-7/HGFβ. (A) The percentages of culture-generated cells that were CD45⁺LSK cells and (B) donor-origin (GFP⁺) hematopoietic cells in the blood of lethally irradiated recipients 12 weeks after the i.v. transfer of 2 × 10³ culture-generated and 2 × 10⁵ Ly5.2 WT BM cells were determined. Mean ± SD of groups of 5 mice. *P < 0.05 versus PBS values. One representative experiment out of 2 is shown.

Consistent with published results (20), approximately 75% of the purified CD45^–Lin^–Sca-1⁺ BM cells were SSEA⁺ (data not shown), indicating that they were developmentally less mature than HSCs. In addition (and also unlike HSCs), approximately 30% of the SSEA⁺ cells expressed the IL-7Rα chain as well as c-Met, the receptor for HGF. Inasmuch as rIL-7/HGFβ acts selectively on dual receptor-expressing cells (ref. 18 and our unpublished observations), we hypothesized that, at a minimum, the SSEA⁺IL-7Rα⁺c-Met⁺ subset of CD45^–Lin^– BM cells contained the immediate precursors of the HSCs that are formed in these cultures. To test this, purified CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells were cultured for 12 days with rIL-7/HGFβ (or PBS). As shown in Figure 9A, these cultures generated HSCs, MPPs, megakaryocytic erythroid progenitors (MEPs), and CLPs in vitro. In addition, these cells competitively repopulated the hematopoietic system after i.v. injection into primary (Figure 9B) and secondary (Figure 9C) recipients. In contrast, CD45^–Lin^–Sca-1⁺SSEA⁺ BM cells that failed to express IL-7Rα and c-Met did not generate HSCs, MPPs, MEPs, and CLPs in the presence of rIL-7/HGFβ (data not shown).

Figure 9

rIL-7/HGFβ induces the appearance of HSCs in short-term cultures of CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells. FACS-purified (≥98%) CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells freshly harvested from normal BM were cultured for 12 days in the presence or absence of rIL-7/HGFβ. (A) Percentages of HSCs (CD45⁺LSK Flt3^–), MPPs (CD45⁺LSK Flt3⁺), MEPs (CD45⁺Lin^–Sca-1^–c-kit⁺Flt3⁺), and CLPs (CD45⁺LSK Flt3⁺IL-7Rα⁺) generated in vitro. (B) Donor-origin (Ly5.1) hematopoietic cells in the blood of lethally irradiated recipients (Ly5.2) 4 weeks after the i.v. transfer of 2 × 10³ culture-generated and 2 × 10⁵ Ly5.2 BM cells (or precultured cells and 2 × 10⁵ Ly5.2 BM cells as controls). Neither feeder cells nor IL-7/HGFβ were included in the transplants. (C) 5 × 10⁶ BM cells from the primary recipients in B were mixed with 5 × 10⁵ Ly5.2 BM cells and injected i.v. into lethally irradiated secondary recipients. The percentages of donor-origin LSK cells in the BM and hematopoietic cells in the blood were determined 4 weeks later. Mean ± SD of groups of 4 to 5 mice. *P < 0.05 versus control values. One representative experiment out of 2 is shown.

rIL-7/HGFβ enhances the generation of HSCs by ESCs in vitro. To further demonstrate that rIL-7/HGFβ can induce primitive precursors to generate HSCs in vitro, we cocultured undifferentiated murine ESCs (TC-1/EGFP) with OP9 cells for 12 days in the presence of equimolar amounts of rIL-7/HGFβ, rIL-7, and/or rHGF (or PBS). The cells from these cultures were then injected i.v. (2 × 10⁶) into sublethally irradiated (6 Gy) 129SVEV Tac mice. As shown in Figures 10 and 11, only the cells from the rIL-7/HGFβ-containing ESC cocultures were able to reconstitute the hematopoietic system after i.v. injection. However, unlike the HSCs generated in cultures of purified CD45^–Lin^–Sca-1⁺ cells from adult BM (Figure 8), the ESC-derived HSCs required the i.f. injection of rIL-7/HGFβ to establish chimerism. Again, the i.f. injection of rIL-7/HGFβ (or rIL-7 and/or rHGF) was without effect when cells from the IL-7– and/or HGF-containing ESC cocultures were infused i.v. (data not shown). Nonetheless, as shown in Figure 12, the cells from all of the ESC cocultures (including the PBS controls) were able to establish low levels of BM chimerism when they themselves were injected i.f., and the levels of chimerism could be increased significantly by the concurrent i.f. injection of rIL-7/HGFβ. Even under these conditions, however, the i.f. injection of cells from the rIL-7/HGFβ-containing ESC cocultures established significantly higher levels of hematopoietic chimerism than did equal numbers of cells from the rIL-7– and/or HGF-containing cultures. However, even the donor-origin BM cells from the primary recipients of cells from the rIL-7/HGFβ-containing ESC cocultures were unable to establish hematopoietic chimerism when injected i.v. into secondary recipients without i.f. injection of rIL-7/HGFβ (data not shown).

Figure 10

rIL-7/HGFβ-stimulated ESC-derived cells require the i.f. injection of rIL-7/HGFβ to engraft in BM and establish hemopoietic chimerism. 2 × 10⁶ differentiated cells from cocultures of murine ESCs and OP9 stromal cells containing rIL-7/HGFβ, rIL-7, and/or rHGF were injected i.v. into irradiated syngeneic 129SVEV Tac recipients that had also been injected i.f. with rIL-7/HGFβ. The proportions of donor-origin (EGFP⁺) cells in the BM and BL of the recipients were determined 20 weeks later. One representative experiment out of 2 is shown. Each symbol denotes the percentage of donor cells in the BM (black square) or BL (white square) of an individual recipient mouse under the indicated experimental conditions. The horizontal bars indicate the mean values in the BM (black bar) or BL (white bar) of the respective groups of recipients.

Figure 11

The donor-origin cells in the blood of the recipients of rIL-7/HGFβ-stimulated ESCs are composed of lymphoid and myeloid cells. (A) The percentages of donor-origin (GFP⁺) cells in the blood of irradiated mice were determined 20 weeks after the injection of cells from rIL-7/HGFβ-containing cocultures of murine ESCs (see Figure 10). (B) The presence of T cells (CD3⁺) and B cells (CD19⁺) and (C) myeloid cells (Gr-1^–CD11b⁺) and granulocytes (Gr-1⁺) among the donor-origin cells is shown. One representative experiment out of 2 is shown.

Figure 12

The i.f. injection of rIL-7/HGFβ enhances the engraftment of i.f.-injected ESC-derived HSCs. 5 × 10⁵ cells from rIL-7/HGFβ-, rIL-7 plus rHGF–, or PBS-containing ESC cocultures were mixed with rIL-7/HGFβ or PBS and injected i.f. into irradiated syngeneic recipients. Nine weeks later, the percentages of donor-origin (GFP⁺) cells in the BM of the recipients were determined. Mean ± SD of 4 to 5 mice per group. One representative experiment out of 2 is shown.

These results suggest that the i.f. injection of rIL-7/HGFβ enabled the ESC-derived HSCs from the rIL-7/HGFβ-containing cocultures to migrate to and implant in the host BM but was unable to do so for their counterparts from the rIL-7- and/or rHGF-containing (or PBS-containing) cocultures. However, when the need for cell migration was bypassed, the i.f. injection of rIL-7/HGFβ was also able to enhance the ability of the ESC-derived HSCs from the latter cocultures to establish hematopoietic chimerism. Hence, the ESC-derived HSCs from the rIL-7– and/or rHGF-containing cocultures appeared to be less mature on average than were those from the rIL-7/HGFβ-containing cocultures; and these in turn appeared to be less mature than were the HSCs from the rIL-7/HGFβ-containing adult BM cell cultures, many of which could migrate to and implant in the BM in the absence of i.f.-injected rIL-7/HGFβ (Figure 2).

We then determined whether CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ HSC precursors were present in the ESC cultures. We found that 10- to 12-day-old murine ESC (TC-1/EGFP)/OP9 cell cocultures contained 2%–3% CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ cells, which when purified and stimulated in vitro with rIL-7/HGFβ, were able to reconstitute the hematopoietic system in vivo (Figure 13). These data suggest that, as in adult BM, CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ HSC precursors in ESC cultures can be induced by rIL-7/HGFβ to develop into HSCs.

Figure 13

Purified CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ HSC precursors from ESC cultures reconstitute the hematopoietic system in vivo after in vitro stimulation with rIL-7/HGFβ. 2 × 10⁴ CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ HSC precursors that were purified from cocultures of murine ESCs and OP9 stromal cells were cultured with medium in the presence of PBS or rIL-7/HGFβ for 12 days. The cultured cells were then injected i.v. into irradiated 129SVEV Tac mice that had also been injected i.f. with rIL-7/HGFβ. The proportions of donor-origin cells in the BM and blood of the recipients were determined 12 weeks later. Mean ± SD of groups of 4 to 5 mice. One representative experiment out of 2 is shown.

In their aggregate, the present data indicate that rIL-7/HGFβ can induce the generation of at least 3 distinct subsets of HSCs in vitro, each of which can establish long-term multilineage hematopoietic chimerism in vivo. These HSC subsets can be distinguished operationally by their respective abilities to migrate to and/or implant in the BM of irradiated recipients in the presence or absence of i.f.-injected rIL-7/HGFβ. However, it is not known whether they represent different developmental stages of a single lineage or separate lineages that are produced at different stages of ontogeny (33).

The least mature subset of HSCs was generated by ESCs in OP-9 stromal cell cocultures that contained rIL-7 and/or rHGF (or PBS in lieu of cytokines). These HSCs did not migrate to the BM of irradiated recipients after i.v. injection, even when rIL-7/HGFβ had been injected i.f. However, they were able to establish low levels of hematopoietic chimerism when they were injected directly into the BM, and, under these circumstances, i.f.-injected rIL-7/HGFβ further enhanced their ability to do so. A subset of HSCs of intermediate maturity was generated by ESCs in OP-9 stromal cell cocultures that contained rIL-7/HGFβ. Although these cells did not migrate to the BM after i.v. injection, they could be induced to do so by the i.f. injection of rIL-7/HGFβ. Thus, even in the presence of rIL-7/HGFβ, ESCs did not generate definitive HSCs. The most mature subset of HSCs was generated by adult BM cells in cultures containing rIL-7/HGFβ. These adult BM-derived HSCs spontaneously migrated to the BM and established long-term hematopoietic chimerism after i.v. injection. In addition, the level of chimerism could be significantly enhanced by the i.f. injection of rIL-7/HGFβ.

In addition to enhancing the migration of HSCs to the BM, it is possible that locally injected rIL-7/HGFβ directly or indirectly repairs the radiation-induced damage to the BM microenvironment (26–28). This is consistent with the known ability of HGF to enhance the migration and adhesion of HSCs to BM niches (32), to diminish oxidative stress-induced apoptosis (34), and to interact synergistically with a variety of hematopoietic cytokines (35–37). It also is consistent with the antiapoptotic properties of both HGF and IL-7 and with their abilities to enhance CXCR4 signaling and to upregulate and/or activate adhesion molecules on many cell types (32, 38, 39). When combined with the unique signaling properties of rIL-7/HGFβ (see below), the results raise the additional possibility that rIL-7/HGFβ improves the BM microenvironment by increasing the local expression of molecules that regulate self-renewal, expansion, and/or fate determination of HSCs (40). One such defect may be related to the low levels of cytokine production in the irradiated BM, which adversely affect early B-lineage and T-lineage development and the mobilization of HSCs and MPPs (27–29, 41–43). Indeed, we have demonstrated here that i.f. injection of rIL-7/HGFβ significantly enhances thymocyte chimerism (Figure 4), and our recent demonstration that the systemic administration of rIL-7/HGFβ also rapidly restores thymocytopoiesis after HSC transplantation strongly supports such a mechanism (19).

Although our initial observations were made in long-term rIL-7/HGFβ-containing BM cell cultures, it was possible to greatly accelerate the generation of transplantable HSCs in these cultures (7 days vs. 2 months) when SF/FL was added. As confirmed in cultures initiated with purified HSCs, the SF/FL appeared to stimulate the proliferation of preexisting HSCs, whereas rIL-7/HGFβ helped to maintain the viability and function of these SF/FL-stimulated HSCs. The rIL-7/HGFβ acted at least in part by preventing oxidative stress and preserving (or inducing) the quiescent (G₀) state of HSCs, possibly by enabling the proliferating HSCs in these cultures to reenter G₀ phase and/or to upregulate genes implicated in HSC survival or migration (30, 31). However, as HSCs do not express the IL-7Rα chain, it is difficult to explain the superior protective ability of rIL-7/HGFβ, as compared with that of rHGF (or HGFβ) alone. One possibility is that rIL-7/HGFβ coordinately binds to and signals through c-Met and the γc chain. This is not implausible, as a functional hybrid receptor composed of the γc chain and the β chain of the GM-CSF receptor has been detected on human CD34⁺ hematopoietic cells (44). Another possibility is that the IL-7 component of rIL-7/HGFβ signals through other surface receptors, such as c-kit (45, 46). Whatever the mechanism, it is unlikely that rIL-7/HGFβ affects HSCs indirectly, as it is active in cultures of purified, SF/FL-stimulated HSCs.

In addition to preserving the hematopoietic reconstituting capacity of proliferating HSCs, it is possible that rIL-7/HGFβ further increases the size of the pool of transplantable HSCs in these short-term BM cell cultures by inducing the de novo formation of HSCs from CD45^–Lin^– precursors. This is supported by our demonstration that HSCs could be generated efficiently in short-term cultures of purified murine CD45^–Lin^–Sca-1⁺ or CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells. Although it might be argued that the expanded populations of CD45⁺LSK cells resulted from the selective outgrowth of a minor population of preexisting HSCs, this is unlikely as rIL-7/HGFβ does not stimulate the proliferation of HSCs and as SF/FL is not required in these cultures. Alternatively, it could be argued that the CD45^–Lin^–Sca-1⁺ or CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells that were induced by rIL-7/HGFβ to generate definitive HSCs were in fact embryonic or fetal-type HSCs that had not yet expressed hematopoietic-specific surface markers (17, 47). However, this is unlikely, as most embryonic and fetal-type HSCs express CD11b (48) and therefore would have been removed during the purification of the CD45^–Lin^– BM cells (see Methods).

Because many CD45^–Lin^–Sca-1⁺ BM cells express markers common to ESCs (e.g., Oct-4, CD71, SSEA-1, Nanog, Rex-1) and are reputed by some investigators to have the ability to generate tissues from all 3 germ layers, they originally were termed adult BM stem cells. However, this claim has been disputed, and it has been suggested instead that this primitive cell population has the appearance of pluripotency because it is composed of (or fuses with) a heterogenous mixture of tissue-committed stem cells (24, 25). Regardless, some CD45^–Lin^–Sca-1⁺ BM cells, including similar somatic precursor cell populations in human BM and umbilical cord blood, have been reported to have the ability to generate multilineage HSCs in vivo or in vitro, albeit inefficiently (16, 20–23, 49–51). Importantly, we have demonstrated that purified CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ cells from BM and ESC cultures can generate HSCs in the presence of rIL-7/HGFβ, suggesting these cells are HSC precursors. Based on our prior demonstration of the ability of rIL-7/HGFβ (as opposed to uncomplexed mixtures of IL-7 and HGFβ) to physically cross-link and aggregate the receptors for HGF and IL-7 on the cell surface (18, 19, 52), we predict that rIL-7/HGFβ can also cross-link the IL-7Rα and c-Met receptors on these HSC precursors, resulting in juxtacrine interactions and the appearance of phosphorylated downstream products that differ from those observed when each receptor signals independently, even if simultaneously (ref. 53 and our unpublished observations). In support of this hypothesis, we have observed the appearance of such distinct products in rIL-7/HGFβ cross-linked CLPs (unpublished observations).

It remains to be determined whether any of the short-term HSC-producing culture systems described here holds promise for clinical application when appropriately modified. At present, the cultures of adult BM cells appear to have advantages over those of ESCs in that they generate definitive HSCs, do not require stromal cell layers, do not require (but may benefit from) the injection of cytokine into the BM cavity, and do not run the risk of inducing teratomas. Although cultures using unseparated BM cells are the simplest to establish, cultures using purified HSCs or CD45^–Lin^–Sca-1⁺ or CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells have the advantage of producing high proportions of CD45⁺LSK cells, thereby enabling HSCs to be easily purified prior to transplantation. In addition, untreated cultures of self-replicating CD45^–Lin^–Sca-1⁺ or CD45^–Lin^–SSEA⁺IL-7Rα⁺c-Met⁺ BM cells have the potential to serve as reservoirs for the continuous or intermittent induction of HSCs by rIL-7/HGFβ and might also prove useful in HSC gene transfer therapy (54). Also, the production of HSCs in these cultures could be further amplified by the addition of SF/FL or other early-acting cytokines. Hence, by (a) stimulating the efficient in vitro production of transplantable HSCs from earlier precursors; (b) protecting proliferating HSCs against the loss of in vivo repopulating ability; (c) supporting the migration and engraftment of HSCs to/in the BM; and (d) enhancing thymocytopoiesis and naive T cell regeneration in irradiated recipients (unpublished observations), rIL-7/HGFβ has the potential to overcome several of the major obstacles to the widespread employment of HSC transplantation as a therapeutic modality.

This study was supported in part by NIH grant AI32752 and a grant from the Connecticut Stem Cell Research Program (10SCB12).

Address correspondence to: Irving Goldschneider, Department of Immunology, School of Medicine, University of Connecticut Health Center, 263 Farmington Ave., Farmington, Connecticut 06030, USA. Phone: 860.679.4244; Fax: 860.679.2936; E-mail: igoldsch@neuron.uchc.edu. Or to: Laijun Lai, Department of Allied Health Sciences, University of Connecticut, 1390 Storrs Road, Storrs, Connecticut 06269, USA. Phone: 860.679.6073; Fax: 860.486.6073; E-mail: laijun.lai@uconn.edu.

Conflict of interest: Laijun Lai and Irving Goldschneider hold US patent No. 6,749,847 (“Hybrid Cytokine of IL-7 and β-chain of Hepatocyte Growth Factor”) and US patent No. 7,578,998 (“Chimeric Cytokine of IL-7 and β-chain of HGF and Methods of Use”).

Reference information: J Clin Invest. 2012;122(10):3552–3562. doi:10.1172/JCI46055.

1. Sorrentino BP. Clinical strategies for expansion of haematopoietic stem cells. Nat Rev Immunol. 2004;4(11):878–888.
   View this article via: PubMed Google Scholar
2. Ross J, Li L. Recent advances in understanding extrinsic control of hematopoietic stem cell fate. Curr Opin Hematol. 2006;13(4):237–242.
   View this article via: PubMed CrossRef Google Scholar
3. Devine SM, Lazarus HM, Emerson SG. Clinical application of hematopoietic progenitor cell expansion: current status and future prospects. Bone Marrow Transplant. 2003;31(4):241–252.
   View this article via: PubMed CrossRef Google Scholar
4. McNiece I. Ex vivo expansion of hematopoietic cells. Exp Hematol. 2004;32(5):409–410.
   View this article via: PubMed CrossRef Google Scholar
5. Amsellem S, et al. Ex vivo expansion of human hematopoietic stem cells by direct delivery of the HOXB4 homeoprotein. Nat Med. 2003;9(11):1423–1427.
   View this article via: PubMed CrossRef Google Scholar
6. Reya T, et al. A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature. 2003;423(6938):409–414.
   View this article via: PubMed CrossRef Google Scholar
7. Stier S, Cheng T, Dombkowski D, Carlesso N, Scadden DT. Notch1 activation increases hematopoietic stem cell self-renewal in vivo and favors lymphoid over myeloid lineage outcome. Blood. 2002;99(7):2369–2378.
   View this article via: PubMed CrossRef Google Scholar
8. Von Drygalski A, Alespeiti G, Ren L, Adamson JW. Murine bone marrow cells cultured ex vivo in the presence of multiple cytokine combinations lose radioprotective and long-term engraftment potential. Stem Cells Dev. 2004;13(1):101–111.
   View this article via: PubMed CrossRef Google Scholar
9. Glimm H, Oh IH, Eaves CJ. Human hematopoietic stem cells stimulated to proliferate in vitro lose engraftment potential during their S/G(2)/M transit and do not reenter G(0). Blood. 2000;96(13):4185–4193.
   View this article via: PubMed Google Scholar
10. Summers YJ, Heyworth CM, de Wynter EA, Chang J, Testa NG. Cord blood G(0) CD34+ cells have a thousand-fold higher capacity for generating progenitors in vitro than G(1) CD34+ cells. Stem Cells. 2001;19(6):505–513.
    View this article via: PubMed CrossRef Google Scholar
11. Ito K, et al. Reactive oxygen species act through p38 MAPK to limit the lifespan of hematopoietic stem cells. Nat Med. 2006;12(4):446–451.
    View this article via: PubMed CrossRef Google Scholar
12. Yeoh JS, et al. Fibroblast growth factor-1 and -2 preserve long-term repopulating ability of hematopoietic stem cells in serum-free cultures. Stem Cells. 2006;24(6):1564–1572.
    View this article via: PubMed CrossRef Google Scholar
13. Arai F, et al. Tie2/angiopoietin-1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell. 2004;118(2):149–161.
    View this article via: PubMed CrossRef Google Scholar
14. Glimm H, Tang P, Clark-Lewis I, von Kalle C, Eaves C. Ex vivo treatment of proliferating human cord blood stem cells with stroma-derived factor-1 enhances their ability to engraft NOD/SCID mice. Blood. 2002;99(9):3454–3457.
    View this article via: PubMed CrossRef Google Scholar
15. Wiesmann A, et al. Modulation of hematopoietic stem/progenitor cell engraftment by transforming growth factor beta. Exp Hematol. 2000;28(2):128–139.
    View this article via: PubMed Google Scholar
16. Weinreich MA, et al. Growth factor receptors as regulators of hematopoiesis. Blood. 2006;108(12):3713–3721.
    View this article via: PubMed CrossRef Google Scholar
17. Tian X, Kaufman D. Differentiation of embryonic stem cells towards hematopoietic cells: progress and pitfalls. Curr Opin Hematol. 2008;15(4):312–318.
    View this article via: PubMed CrossRef Google Scholar
18. Lai L, Zeff RA, Goldschneider I. A recombinant single-chain IL-7/HGFβ hybrid cytοkine induces juxtacrine interactions of the IL-7 and HGF (c-Met) receptors and stimulates the proliferation of CFU-S12, CLPs, and pre-pro-B cells. Blood. 2006;107(5):1776–1784.
    View this article via: PubMed CrossRef Google Scholar
19. Jin J, Goldschneider I, Lai L. In vivo administration of the rIL-7/HGFβ hybrid cyτokine restores thymopoiesis and naïve T cell generation in lethally-irradiated mice after syngeneic bone marrow transplantation. J Immunol. 2011;186(4):1915–1922.
    View this article via: PubMed CrossRef Google Scholar
20. Kucia M, et al. A population of very small embryonic-like (VSEL) CXCR4(+)SSEA-1(+)Oct-4+ stem cells identified in adult bone marrow. Leukemia. 2006;20(5):857–869.
    View this article via: PubMed CrossRef Google Scholar
21. Serafini M, et al. Hematopoietic reconstitution by multipotent adult progenitor cells: precursors to long-term hematopoietic stem cells. J Exp Med. 2007;204(1):129–139.
    View this article via: PubMed CrossRef Google Scholar
22. Ratajczak MZ, Zuba-Surma EK, Wysoczynski M, Ratajczak J, Kucia M. Very small embryonic-like stem cells: characterization, developmental origin, and biological significance. Exp Hematol. 2008;36(6):742–751.
    View this article via: PubMed CrossRef Google Scholar
23. Jiang Y, et al. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature. 2002;418(6893):41–49.
    View this article via: PubMed CrossRef Google Scholar
24. Reyes M, Li S, Foraker J, Kimura E, Chamberlain JS. Donor origin of multipotent adult progenitor cells in radiation chimeras. Blood. 2005;106(10):3646–3649.
    View this article via: PubMed CrossRef Google Scholar
25. Wagers AJ, Weissman IL. Plasticity of adult stem cells. Cell. 2004;116(5):639–648.
    View this article via: PubMed CrossRef Google Scholar
26. Goodell MA. Stem-cell “plasticity”: befuddled by the muddle. Curr Opin Hematol. 2003;10(3):208–213.
    View this article via: PubMed CrossRef Google Scholar
27. Chung B, Barbara-Burnham L, Barsky L, Weinberg K. Radiosensitivity of thymic interleukin-7 production and thymopoiesis after bone marrow transplantation. Blood. 2001;98(5):1601–1606.
    View this article via: PubMed CrossRef Google Scholar
28. Mazo IB, Quackenbush EJ, Lowe JB, von Andrian UH. Total body irradiation causes profound changes in endothelial traffic molecules for hematopoietic progenitor cell recruitment to bone marrow. Blood. 2002;99(11):4182–4191.
    View this article via: PubMed CrossRef Google Scholar
29. Testa NG, Hendry JH, Molineux G. Long-term bone marrow damage in experimental systems and in patients after radiation and chemotherapy. Anticancer Res. 1985;5(1):101–110.
    View this article via: PubMed Google Scholar
30. Miller CL, Eaves CJ. Expansion in vitro of adult murine hematopoietic stem cells with transplantable lympho-myeloid reconstituting ability. Proc Natl Acad Sci U S A. 1997;94(25):13648–13653.
    View this article via: PubMed CrossRef Google Scholar
31. Araki H, Yoshinaga K, Boccuni P, Zhao Y, Hoffman R, Mahmud N. Chromatin-modifying agents permit human hematopoietic stem cells to undergo multiple cell divisions while retaining their repopulating potential. Blood. 2007;109(8):3570–3578.
    View this article via: PubMed CrossRef Google Scholar
32. Son BR, et al. Migration of bone marrow and cord blood mesenchymal stem cells in vitro is regulated by stromal-derived factor-1-CXCR4 and hepatocyte growth factor-c-met axes and involves matrix metalloproteinases. Stem Cells. 2006;24(5):1254–1265.
    View this article via: PubMed CrossRef Google Scholar
33. Ratajcak MZ, Machalinski B, Wojakowski W, Ratajczak J, Kucia M. A hypothesis for an embryonic origin of pluripotent Oct-4+ stem cells in adult bone marrow and other tissues. Leukemia. 2007;21(5):860–867.
    View this article via: PubMed Google Scholar
34. Kannan R, Jin M, Gamulescu MA, Hinton DR. Ceramide-induced apoptosis: role of catalase and hepatocyte growth factor. Free Radic Biol Med. 2004;37(2):166–175.
    View this article via: PubMed CrossRef Google Scholar
35. Yu SZ, et al. Stimulatory effects of hepatocyte growth factor on hemopoiesis of CSF/c-kit system-deficient mice. Stem Cells. 1998;16(1):66–77.
    View this article via: PubMed CrossRef Google Scholar
36. Galimi F, et al. Hepatocyte growth factor induces proliferation and differentiation of multipotent and erythroid hemopoietic progenitors. J Cell Biol. 1994;127(6 pt 1):1743–1754.
    View this article via: PubMed CrossRef Google Scholar
37. Kmiecik TE, Keller JR, Rosen E, Vande Woude GF. Hepatocyte growth factor is a synergistic factor for the growth of hematopoietic progenitor cells. Blood. 1992;80(10):2454–2457.
    View this article via: PubMed Google Scholar
38. Aiello FB, et al. IL-7 Induces myelopoiesis and erythropoiesis. J Immunol. 2007;178(3):1553–1563.
    View this article via: PubMed Google Scholar
39. Schmitt N, et al. Positive regulation of CXCR4 expression and signaling by interleukin-7 in CD4+ mature thymocytes correlates with their capacity to favor human immunodeficiency X4 virus replication. J Virol. 2003;77(10):5784–5793.
    View this article via: PubMed CrossRef Google Scholar
40. Zhang J, et al. Identification of the haematopoietic stem cell niche and control of the niche size. Nature. 2003;425(6960):836–841.
    View this article via: PubMed CrossRef Google Scholar
41. Foss DL, Donskoy E, Goldschneider I. The importation of hematogenous precursors by the thymus is a gated phenomenon in normal adult mice. J Exp Med. 2001;193(3):365–374.
    View this article via: PubMed CrossRef Google Scholar
42. Donskoy E, Foss DL, Goldschneider I. Gated importation of prothymocytes by adult mouse thymus is coordinated with their periodic mobilization from bone marrow. J Immunol. 2003;171(7):3568–3575.
    View this article via: PubMed Google Scholar
43. Porter DL, June CH. T-cell reconstitution and expansion after hematopoietic stem cell transplantation: ‘T’ it up! Bone Marrow Transplant. 2005;35(10):935–942.
    View this article via: PubMed CrossRef Google Scholar
44. Giron-Michel J, et al. Detection of a functional hybrid receptor gammac/GM-CSFRbeta in human hematopoietic CD34+ cells. J Exp Med. 2003;197(6):763–775.
    View this article via: PubMed CrossRef Google Scholar
45. Fry TJ, Mackall CL. Interleukin-7: from bench to clinic. Blood. 2002;99(11):3892–3904.
    View this article via: PubMed CrossRef Google Scholar
46. Cosenza L, Gorgun G, Urbano A, Foss F. Interleukin-7 receptor expression and activation in nonhaematopoietic neoplastic cell lines. Cell Signal. 2002;14(4):317–325.
    View this article via: PubMed CrossRef Google Scholar
47. McKinney-Freeman SL, et al. Surface antigen phenotypes of hematopoietic stem cells from embryos and murine embryonic stem cells. Blood. 2009;114(2):268–278.
    View this article via: PubMed CrossRef Google Scholar
48. Kim I, He S, Yilmaz OH, Kiel MJ, Morrison SJ. Enhanced purification of liver hematopoietic stem cells using SLAM family receptors. Blood. 2006;108(2):737–744.
    View this article via: PubMed CrossRef Google Scholar
49. D’Ippolito G, et al. Marrow-isolated adult multilineage inducible (MIAMI) cells, a unique population of postnatal young and old human cells with extensive expansion and differentiation potential. J Cell Sci. 2004;117(pt 14):2971–2981.
    View this article via: PubMed CrossRef Google Scholar
50. Yoon YS, et al. Clonally expanded novel multipotent stem cells from human bone marrow regenerate myocardium after myocardial infarction. J Clin Invest. 2005;115(2):326–338.
    View this article via: JCI PubMed Google Scholar
51. Kogler G, et al. A new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential. J Exp Med. 2004;200(2):123–135.
    View this article via: PubMed CrossRef Google Scholar
52. Wei C, Lai L, Goldschneider I. Pre-pro-B cell growth-stimulating factor (PPBSF) upregulates IL-7Rα chain expρession and enables pro-B cells to respond to monomeric IL-7. J Interferon Cytokine Res. 2002;22(7):823–832.
    View this article via: PubMed CrossRef Google Scholar
53. Mellado M, et al. Chemokine receptor homo-or heterodimerization activates distant signaling pathways. EMBO J. 2001;20(10):2497–2507.
    View this article via: PubMed CrossRef Google Scholar
54. Van Damme A, Driessche TV, Collen D, Chuah MKL. Bone marrow stromal cells as targets for gene therapy. Curr Gene Ther. 2002;2:195–209.
    View this article via: PubMed CrossRef Google Scholar
55. Nakano T, Kodama H, Honjo T. Generation of lymphohematopoietic cells from embryonic stem cells in culture. Science. 1994;265(5175):1098–1101.
    View this article via: PubMed CrossRef Google Scholar
56. Yahata T, et al. A highly sensitive strategy for SCID-repopulating cell assay by direct injection of primitive human hematopoietic cells into NOD/SCID mice bone marrow. Blood. 2003;101(8):2905–2913.
    View this article via: PubMed CrossRef Google Scholar