Original Article

Leukemia (2010) 24, 1325–1334; doi:10.1038/leu.2010.97; published online 20 May 2010

Stem Cells

Regulatory functions of TRAIL in hematopoietic progenitors: human umbilical cord blood and murine bone marrow transplantation

K Mizrahi1, J Stein1,2, M Pearl-Yafe1, O Kaplan3, I Yaniv1,2 and N Askenasy1

  1. 1Frankel Laboratory, Center for Stem Cell Research, Department of Pediatric Hematology-Oncology, Schneider Children's Medical Center of Israel, Petach Tikva, Israel
  2. 2Bone Marrow Transplant Unit, Department of Pediatric Hematology-Oncology, Schneider Children's Medical Center of Israel, Petach Tikva, Israel
  3. 3Department of Surgery, Sourasky Medical Center and Tel Aviv University, Tel Aviv, Israel

Correspondence: Professor N Askenasy, Frankel Laboratory, Center for Stem Cell Research, Schneider Children's Medical Center of Israel, 14 Kaplan Street, Petach Tikva 49202, Israel. E-mail: anadir@012.net.il

Received 17 October 2009; Revised 27 January 2010; Accepted 18 February 2010; Published online 20 May 2010.



The tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) signaling pathway has selective toxicity to malignant cells. The TRAIL receptors DR4 and DR5 are expressed at low levels in human umbilical cord blood cells (3–15%) and are upregulated by incubation with the cognate ligand, triggering apoptosis in 70–80% of receptor-positive cells (P<0.001). Apoptosis is not induced in hematopoietic progenitors, as determined from sustained severe combined immunodeficiency reconstituting potential and clonogenic activity. Furthermore, elimination of dead cells after incubation with TRAIL for 72h results in a threefold enrichment in myeloid progenitors. Exposure to TRAIL in semisolid cultures showed synergistic activity of DR4 and granulocyte/macrophage colony-stimulating factor in recruiting lineage-negative (lin) and CD34+ progenitors and in promoting the formation of large colonies. In murine bone marrow, ~30% of lin cells express TRAIL-R2 (the only murine receptor), and the receptor is upregulated after transplantation in cycling and differentiating donor cells that home to the host marrow. However, this receptor is almost ubiquitously expressed in the most primitive (linSCA-1+c-kit+) progenitors, and stimulates the clonogenic activity of lin cells (P<0.001), suggesting a tropic function after transplantation. It is concluded that TRAIL does not trigger apoptosis in hematopoietic progenitors, and upregulation of its cognate receptors under stress conditions mediates tropic signaling that supports recovery from hypoplasia.


TRAIL; hematopoietic progenitors; transplantation; umbilical cord blood



Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is a member of the tumor necrosis factor (TNF) superfamily that participates in homeostatic regulation of cell death and activation.1 Although it is expressed in many tissues under physiological conditions, TRAIL is not toxic to resting parenchymal cells and selectively kills malignant cells,2, 3, 4 a differential sensitivity of neoplastic cells to TRAIL-induced apoptosis that can be used for therapeutic purposes.5, 6, 7 For example, induced expression of TRAIL in CD34+ cells8, 9 and mesenchymal stromal cells10, 11 was successfully applied to target tumor cells and induced their selective death in vitro and in vivo. In both cases, overexpression of TRAIL did not significantly affect the differentiation capacity of mesenchymal stromal cells into multiple lineages in vitro,11 and the hematopoietic function of progenitors was preserved in vivo.8 Although the intracellular signaling pathways associated with TRAIL receptor activation have been described in detail,12, 13, 14, 15, 16 the role of TRAIL under physiological, pathological and stress conditions is not completely understood.

As other members of the TNF superfamily, TRAIL participates in the development and regulation of the immunohematopoietic system.2, 3, 4 The ligand and its receptors are expressed at low levels in human hematopoietic progenitors under steady-state conditions; therefore, these cells are largely unaffected by this signaling pathway.17, 18, 19 Among the known five TRAIL receptors, DR4 (TRAIL-R1) and DR5 (TRAIL-R2) are associated with transduction of apoptotic signals in humans,16 whereas mice express only TRAIL-R2.20, 21 Although mRNA encoding TRAIL receptors DR4 and DR5 is frequently detected in CD34+ progenitors, these cells were insensitive to apoptosis22, 23 even in the presence of low doses of doxorubicin.24 Consistently, infusion of soluble TRAIL into rodents was not associated with gross abnormalities in hematopoiesis,25, 26 and TRAIL neither induced nor impaired the clonogenic activity of human hematopoietic progenitors in vitro.22, 23 These data argue that TRAIL is not a significant factor in proximal stages of hematopoietic cell differentiation. Nevertheless, administration of recombinant human TRAIL to nonhuman primates resulted in mild anemia,27 and this pathway is considered to contribute to deficient erythropoiesis in myelodysplastic syndromes.23, 28, 29 Therefore, TRAIL may serve as a negative regulator of expanding clones in the distal stages of differentiation of all hematopoietic lineages.13, 17, 18, 19, 30, 31

As umbilical cord blood (UCB) becomes an important source of hematopoietic progenitors for transplantation, we evaluated the effect of TRAIL in proximal stages of their activity. We recently showed that the Fas/Fas-ligand interaction transduces trophic signals in early stages of hematopoietic cell engraftment,32 because the most primitive progenitors are insensitive to Fas-mediated apoptosis.33 This is the apparent reason for upregulation of Fas expression in donor cells soon after homing to the host bone marrow (BM). In this study, we evaluated the signaling activity of TRAIL receptors in UCB cells and found that TRAIL is involved in complex apoptotic and tropic signaling in proximal stages of hematopoietic progenitor function. Following these observations, we monitored the dynamics of murine TRAIL receptor-2 after transplantation, this being the only known murine receptor for TRAIL that corresponds to human DR5.20, 21 We found that signaling by the murine receptor is trophic in nature.


Materials and methods

Animal preparation and transplantation

The mice used in this study were C57BL/6J (B6, H2Kb, CD45. 2), B6.SJL-Ptprca Pepcb/BoyJ (H2Kb, CD45.1), BALB/c (H2Kd) and NOD.CB17-Prkdcscid/J (non-obese diabetic severe combined immunodeficiency (NOD.SCID)), purchased from Jackson Laboratories (Bar Harbor, ME, USA). Mice were housed in a barrier facility and all procedures were approved by the Institutional Animal Care Committee. Wild-type recipients were routinely conditioned 1 day before transplantation by total body irradiation at 850rad using an X-ray irradiator (RadSource 2000; RadSource, Suwanee, GA, USA) at a rate of 106rad/min. Notably, X-ray irradiation is different from γ-irradiation in myeloablative dose and toxicity. NOD.SCID mice were conditioned with two daily doses of 25μg/g busulfan, administered 2 days before cell transplantation. Cells suspended in 0.2ml of phosphate-buffered saline (Biological Industries, Kibutz Beit Haemek, Israel) were infused into the lateral tail vein.

Cell isolation, characterization and staining

In mice, whole BM cells (BMCs) were harvested from femurs and tibia and placed in M199 (Beit Haemek, Israel) in aseptic conditions, as previously described.32, 33 The lineage/low (lin/low) subset was immunomagnetically depleted with saturating amounts of hybridoma-derived rat anti-mouse monoclonal antibodies (mAb) specific for CD5, B220, Mac-1, Gr-1 and TER-119, NK1.1 purchased from eBioscience (San Diego, CA, USA). Antibody-coated cells were incubated with sheep-anti-rat IgG conjugated to M-450 magnetic beads (Dynal Inc, Lake Success, NY, USA) and lineage-positive (lin+) cells were retained by exposure to a magnetic field. The efficiency of the lin cell separation procedure was reassessed by flow cytometry using a cocktail of fluorescein–isothyocyanate-labeled mAb against the lineage markers (eBioscience; Pharmingen, Erembodegem, Belgium; and IQ Products, Groningen, The Netherlands).

Umbilical cord blood was obtained from healthy donors pending informed consent according to institutional guidelines. UCB samples that were harvested before placental delivery with citrate–phosphate–dextrose adenine-1 were diluted twofold in phosphate-buffered saline containing 0.5% human serum albumin and 2mM ethylenediaminetetraacetic acid, and collected by centrifugation using lymphocyte separation medium (1.077–1.080g/ml; MP Biomedicals, Illkirch, France). Mononuclear cells (MNCs) were washed twice, and the lin/low subset was immunomagnetically isolated using the Human Lineage Cell Depletion Kit (Miltenyi Biotec, Bergisch Gladbach, Germany). The efficiency of lin cell separation was reassessed as detailed above using a cocktail of phycoerythrin-labeled antibodies.

For staining with an intracellular dye, cells were incubated for 7min with 2.5μM of 5-(and 6-)carboxyfluorescein diacetate succinimidyl ester (CFSE; Molecular Probes, Eugene, OR, USA), washed, resuspended and counted.

Flow cytometry

Measurements were taken with a Vantage SE flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). Peripheral murine blood and BMCs were layered over 1.5ml of lymphocyte separation medium at room temperature (Cedarlane, Burlington, ON, Canada), collected from the buffy coat and stained with primary-labeled mAb. In murine experiments, donor chimerism in syngeneic transplants was determined with mAb against minor antigens CD45.1 (clone A20; eBioscience) and CD45.2 (clone 104; eBioscience). Murine cells were assayed for progenitor markers c-kit (CD117, clone 2B8; eBioscience) and Sca-1 (clone D7; eBioscience), and for expression of TRAIL-R2 (CD262, clone MD5-1; eBioscience).

Expression of TRAIL receptors in UCB cells was determined using mAb against TRAIL-R1 (clone 69036; R&D Systems, Minneapolis, MN, USA) and TRAIL-R2 (clone 71908). Human neuroblastoma SHEP-1 cells served as positive controls for TRAIL-R2. Engraftment of UCB cells in NOD.SCID xenochimeras was determined in the BM after 12 weeks using antibodies against murine CD45 (clone 30-F11; eBioscience) and human CD45 (clone ML2; IQ Products).

Cell death and apoptosis was determined in cells incubated with 5μg/ml 7-aminoactinomycin-D (7-AAD, Sigma, St Louis, MO, USA) and Annexin-V (IQ products).

Apoptotic challenge

Murine BMCs and human UCB cells were incubated (5 × 106 cells per ml) for variable times in α-MEM culture medium (Beit Haemek) supplemented with StemPro Nutrient Supplement (Stem Cell Technologies, Vancouver, BC, Canada), 2mM L-glutamine and 50μM 2β-mercaptoethanol purchased from PeproTech (Rocky Hill, NJ, USA).32 Murine and human cells were challenged with various concentrations of soluble TRAIL (PeproTech) and evaluated for apoptotic activity in murine A20 and human Jurkat cells.

Colony-forming unit assays

Colony-forming unit (CFU) assays were performed in methylcellulose containing Iscove modified Dulbecco's medium (IMDM, Beit Haemek) supplemented with 2mM L-glutamine. In murine assays, 3 × 105 whole and 3 × 104 lin BMCs were plated in 1.2% methylcellulose containing 20% fetal calf serum, 20ng/ml recombinant mouse stem cell factor , 10ng/ml of recombinant mouse interleukin-3, 10μ/ml recombinant human erythropoietin and 5ng/ml granulocyte/macrophage colony-stimulating factor (GM-CSF) (recombinant mouse GM-CSF).

In UCB cell assays, 2.5 × 103 MNCs per well and 103 CD34+ or lin cells per well were plated in 0.9% methylcellulose containing 30% fetal calf serum , 50ng/ml stem cell factor, 10ng/ml interleukin-3 and 10ng/ml rhGM-CSF (PeproTech). TRAIL was supplemented at 200–1000ng/ml in murine cell assays and 500–2000ng/ml in human cell assays, with and without blocking anti-DR4 (clone 69036, R&D Systems) and anti-DR5 (clone 71908, R&D Systems) antibodies.

Statistical analysis

Data are presented as means±standard deviations for each experimental protocol. Results in each experimental group were evaluated for reproducibility by linear regression of duplicate measurements. Differences between the experimental protocols were estimated with a post hoc Scheffe t-test and significance was considered at P<0.05.



Expression of TRAIL receptors in human UCB cells

The dynamics of receptor–ligand interactions can be monitored in UCB cells without disrupting the close interaction between hematopoietic progenitors and stroma in the BM. Lin and CD34+ progenitors derived from fresh human UCB express relatively low levels of TRAIL receptors, predominantly DR4 (P<0.05 versus DR5+, Figure 1a), with a disproportionately higher expression of DR4 over DR5 being most accentuated in mononuclear UCB cells (P<0.001). The low-level expression of DR5 was not a result of deficient antibody-mediated labeling, as SHEP-1 cells were effectively stained for this receptor. Incubation of UCB cells in liquid culture for 4 days resulted in a gradual increase in cell death to 40%, which was not significantly affected by exposure to a toxic concentration of 1.5μg/ml TRAIL, as determined from apoptosis in Jurkat cells (Figure 1b). The apparent insensitivity of UCB cells to TRAIL-induced apoptosis corroborates previous observations.17, 18, 19, 22 Apoptosis of lin and CD34+ UCB progenitors (~25%) was slightly lower than that observed in the heterogeneous MNC population after 3 days of liquid culture (Figure 1c), whereas the highest sensitivity to apoptosis was apparent in myeloid cells (CD33+) and B-lymphocytes (CD19+). The low sensitivity of T cells (CD3+) to apoptosis may be explained by their naive nature in UCB,34 as opposed to the sensitivity of activated T cells to TRAIL-mediated apoptosis.35

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Death receptor expression and apoptosis in UCB cells. (a) DR4 and DR5 expression in fresh mononuclear cells (MNCs, n=20), CD34+ (n=9) and lin progenitors (n=8), as compared with a positive control for DR5 in SHEP-1 human neuroblastoma cells. (b) Cell death increases with time of incubation, with little influence of 1.5μg/ml TRAIL (n=6–16 for each time point). (c) Apoptosis of lin (n=6) and CD34+ progenitors (n=16), and CD3+ T cells, CD19+ B-lymphocytes and CD33+ myeloid cells (n=5) after 72h of incubation with and without exposure to 1.5μg/ml TRAIL.

Full figure and legend (134K)

Dynamic changes in TRAIL receptors during in vitro incubation

Although TRAIL had no apparent significant influence on UCB cell sensitivity to apoptosis, several parameters were modulated during incubation with the ligand for 72h. An intriguing observation was a 50% decrease in DR4 expression during control incubation in medium, resulting in low levels of expression of both TRAIL receptors (Figure 2a). In contrast, TRAIL induced a twofold increase in the expression of DR4 and DR5 (P<0.05), in particular in CD33+ cells, showing an upregulation of the cognate receptors as previously reported for Fas-ligand in murine BMCs.32 Thus, UCB cells were apparently insensitive to TRAIL-induced apoptosis, despite expression of both receptors in 15% of the cells at the end of the incubation period. The changes in TRAIL receptor expression in culture might be caused either by differential sensitivity to apoptosis of receptor-positive cells, variations in proliferation rates or true modulation of receptor expression.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Dynamics of TRAIL receptor expression in liquid culture. UCB cells were incubated for 72h in medium (n=6) and with 1.5μg/ml TRAIL (n=7) to characterize the activity of DR4 and DR5: (a) receptor expression after 72h of liquid culture in mononuclear cells (MNCs) and gated CD33+ cells, (b) fractional apoptosis within the receptor-positive subsets under various conditions (n=5–7), (c) proliferation rates of receptor-positive cells as determined from CFSE dilution (n=4).

Full figure and legend (147K)

To assess these possibilities, cell death was assayed in reference to TRAIL receptor expression in liquid culture. Approximately 70% of DR4+ and DR5+ cells succumbed to apoptosis under control conditions of incubation in medium (P<0.001 versus MNCs), as well as during incubation with TRAIL, suggesting a higher sensitivity to spontaneous death of these cells in the absence of specific signaling (Figure 2b). Thus, cells expressing TRAIL receptors were more sensitive to apoptosis in control culture, and the fraction of receptor-positive apoptotic cells was constant despite an upregulation of receptors triggered by the cognate ligand. These differential responses of subsets of receptor-positive cells to TRAIL had variable contributions to cell mortality: approximately 5% of DR4+ and DR5+ cells were apoptotic under control culture conditions (26±6% overall MNC death), whereas the DR4+ and DR5+ subsets corresponded to ~10 and ~14% of apoptotic cells, respectively, during exposure to TRAIL (33±8% overall MNC death). Under these conditions, DR5+ cells cycled faster than DR4+ cells, irrespective of the presence of the cognate ligand (Figure 2c), indicating that differential proliferation was not the primary cause for the observed variations in sensitivity to apoptosis. However, the faster cycling rates might be responsible for the apparent relative increase in DR5 as compared with DR4 on exposure to TRAIL. These data revealed that TRAIL induces expression of its cognate receptors, and modulates the pattern of apoptosis in culture because of sensitivity to apoptosis of cells responsive to the ligand.

Differential stimulation of progenitors by TRAIL

As TRAIL-mediated signaling is coupled to both apoptotic and nonapoptotic pathways,12, 13, 14, 15, 16 its effects were further assessed in clonogenic assays of UCB-derived progenitors. This ligand showed an interesting effect on progenitor engagement to clonogenic activity: TRAIL suppressed CFU-GM activity in bulk UCB cells at low concentrations (0.25–1μg/ml) and induced this activity at higher concentrations (Figure 3a). In contrast, dose-dependent activation was observed in lin and CD34+ progenitors, starting at low concentrations. The decline in the activity of progenitors in bulk cell populations is attributed to the indirect suppressive effects of dead cells on clonogenic activity.32 The constant presence of the ligand in cultures influenced two distinct events: recruitment of progenitors to perform clonogenic activity and regulation of colony size. The latter parameter was evaluated by analysis of the partition of large (>50 cells) and small (30–50 cells) colonies in individual cultures (Figure 3b). A gradual increase in colony size was observed in all cultures, indicating that TRAIL does not negatively regulate the proliferation of differentiated cells. On the contrary, TRAIL increased colony size, in addition to its synergism with GM-CSF in the activation of progenitors.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

The effect of TRAIL on hematopoietic progenitors in semisolid cultures. Mononuclear UCB cells (MNCs, n=11), CD34+ (n=5) and lineage-negative (lin, n=8) cells were evaluated in methylcellulose cultures (triplicates) stimulated with stem cell factor , interleukin-3 and GM-CSF under the influence of escalating concentrations of TRAIL. Colonies were counted after 14 days. (a) Clonogenic activity as a function of TRAIL concentration presented as normalized values against control cultures. (b) The ratio between large (>50 cells) and small (30–50 cells) colonies in cultures for the respective TRAIL concentrations. (c) Cultures of whole (MNCs) and lin UCB cells in the presence of 1.5μg/ml TRAIL and blocking antibodies to DR4 and DR5 (n=5). Data were normalized against cultures stimulated by TRAIL in the absence of blocking antibodies.

Full figure and legend (78K)

To determine the TRAIL receptor that is responsible for tropic signaling, cultures were performed in the presence of blocking antibodies to DR4 and DR5 (Figure 3c). Inhibition of DR4 resulted in a slight decrease in the activation of clonogenic activity by TRAIL, whereas inhibition of DR5 resulted in increased clonogenic activity (P<0.01). Although there were minor differences in stimulation of whole BMCs and lin progenitors, the activity of both subsets was increased by selective stimulation of DR4, the receptor that also presents more potent apoptotic activity. It is therefore an antagonistic mode of regulation of progenitor activity in which DR4 stimulates and DR5 has a mild inhibitory effect, which results in the overall stimulation of UCB progenitors by TRAIL.

Functional assessment of UCB cells exposed to TRAIL

In view of the variable effects of TRAIL on apoptosis and activity of UCB progenitors, we assessed the overall impact of this ligand in two functional assays. The activity of SCID-repopulating cells was evaluated after preincubation of UCB cells for 1–3 days with 1.5μg/ml TRAIL before transplantation into immunocompromised mice (Figure 4a). Within the wide distribution of the levels of xenogeneic chimerism of human cells from a single UCB unit, short-term incubation (24h) with TRAIL did not impair SCID-repopulating cell activity, as previously reported.23 Incubation for 3 days showed a modest decrease of 8±5% in the SCID-repopulating activity of UCB cells incubated with TRAIL, compared with control medium (n=4 individual UCB units). Evidently, prolonged exposure to TRAIL in culture has little effect on the SCID-repopulating cell activity of UCB cells, suggesting that progenitors survive these culture conditions. To better quantify the effect of TRAIL, cells incubated under the same conditions were assessed for CFU-GM activity (Figure 4b). Culture of equal numbers of viable cells showed no significant difference after incubation for 1 and 3 days. Elimination of dead cells by centrifugation over ficoll showed a remarkable increase in colony-forming cell activity when equal numbers of viable cells were plated after incubation with TRAIL for 24h (P<0.05) and 72h (P<0.001). The threefold enrichment in clonogenic activity attained by incubation with TRAIL for 72h is evidence that CFU-GM activity was spared and its fraction increased on removal of dead cells (Figure 4c). Notably, enrichment of progenitors by incubation with TRAIL is different from the effect of the ligand when continuously present in clonogenic assays of naive UCB cells. These functional assays show that TRAIL does not induce apoptosis in progenitors responsible for reconstitution of SCID mice and enriches UCB progenitors by elimination of other cell subsets.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Functional assessment of TRAIL-exposed UCB cells. (a) UCB cells were incubated for 24 and 72h with and without 1.5μg/ml TRAIL before infusion into busulfan-conditioned NOD.SCID mice. Engraftment was determined in the BM after 12 weeks using human-specific antibodies. Data are representative of four UCB units for each culture time, in which engraftment was measured in at least two recipients in the pretreated and control groups. (b) UCB cells were preincubated for 24 (n=4) and 72h (n=4) with 1.5μg/ml TRAIL, and equal numbers of viable cells were assayed in semisolid cultures in the presence of stem cell factor , interleukin-3 and GM-CSF, with and without centrifugation over ficoll. CFU activity is calculated for 103 cells at the onset of culture. (c) Enrichment was determined by normalization against clonogenic activity without removal of dead cells by ficoll (n=4).

Full figure and legend (88K)

Expression of the TRAIL receptor in murine BM and grafted cells

Considering these data acquired on surrogate assays of human UCB cells, the impact of TRAIL was evaluated in a more physiological assay of murine BM transplants. Among the five known receptors for TRAIL in humans,2, 3, 4 the dominant receptor identified in rodents corresponds to DR5 (TRAIL-R2).20, 21 This receptor is expressed at relatively low levels in murine whole BMCs; however, it is detectable in 30±3.5% of lin progenitors (Figure 5a). These levels of expression exceed those of Fas and TNF receptors.32, 33 Transplantation of lin BMCs into syngeneic (H2Kb, CD45.1 → CD45.2) and allogeneic recipients (H2Kb → H2Kd) resulted in upregulation of TRAIL-R2 in BM-homed donor cells (Figure 5b). This pattern of expression was similar to the previously reported upregulation of Fas in grafted BMCs,33 and TRAIL-R2 was coexpressed with Fas in 85% of BM-homed donor cells, suggesting a coordinated upregulation of these receptors. In variance, residual host BMCs that survived irradiation and recovered thereafter showed a low-level expression of TRAIL-R2 (Figure 5b), consistent with the polarized nature of Fas expression in donor and host-type cells coresiding in the BM after transplantation.32, 33 TRAIL-R2 was upregulated primarily in cycling cells (CFSEdim), which comprise 25–30% of BM-homed cells 48h after transplantation, and was expressed in a relatively small fraction of quiescent cells (CFSEbright, Figure 1c). The early proliferating and quiescent/slow-cycling subsets of BM-homed donor cells are dissociated in their composition, with lin cells consisting of 34±10 and 73±4% of these subsets, respectively (n=7). Consistently, TRAIL-R2 was upregulated primarily in the lin+ subset of BM-homed donor cells (Figure 1d). However, analysis for markers of primitive hematopoietic progenitors revealed high levels of receptor expression in linSCA-1+ and linSCA-1+c-kit+ donor cells (Figure 5e), which further increased in cells identified by putative progenitor markers at 6 days after transplantation (Figure 5f). Thus, under steady-state conditions, ~30% of naive lin progenitors express the TRAIL receptor, and under stress conditions of transplantation, it is ubiquitously expressed in BM-homed primitive donor hematopoietic progenitors.

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Expression of death receptors in BM cells. (a) Naive murine whole BMCs (wBMC) express low levels of TRAIL-R2, whereas ~30% of lin BMCs are positive for TRAIL-R2 (n=8). (b) After transplantation of lin BMCs, BM-homed donor cells upregulate TRAIL-R2 expression as a function of time in parallel to Fas, whereas residual host cells show low levels of receptor expression (n=5–6 at each time point). (c) TRAIL-R2 is primarily expressed in fast-cycling cells, as determined from CFSE dilution (CFSEdim) and in a small subset of mitotically quiescent and slow-cycling cells (CFSEbright) (n=5). (d) TRAIL-R2 is preferentially expressed in cells that convert to express lineage markers early after transplantation (n=6). (e) The TRAIL receptor is detected in significant fractions of linSCA-1+ and linSCA-1+c-kit+ donor cells that homed to the host BM (n=5). (f) During the early period after transplantation, the expression of TRAIL-R2 increases in linc-kit+ donor cells that reside in host BM (n=5).

Full figure and legend (143K)

It has been previously shown that overexpression of TRAIL in BMCs is not detrimental to the hematopoietic reconstituting potential of progenitors,8 and a consistently direct administration of soluble TRAIL to mice had no toxic effects.25, 26 Although TRAIL is not a significant molecular mediator of graft-versus-host disease,36 selective expression of the receptor in donor cells might mediate rejection of the cell graft. To evaluate this possibility, cells were harvested from femurs at 2 and 4 days after transplantation and were exposed to TRAIL in vitro. This apoptotic challenge resulted in an insignificant increase in apoptosis (17±2.3%), as compared with medium (14±3%), and was not attenuated by incubation with TRAIL in conjunction with either Fas-ligand or TNF-α. Similarly, day +4 BM-homed cells were insensitive to TRAIL-mediated apoptosis, which seems to be an intrinsic insensitivity of primitive progenitors to receptor-mediated apoptosis.32 We have also reported that Fas, the common executioner of apoptosis in the TNF superfamily,1 transduces tropic signals to hematopoietic progenitors that promote hematopoietic cell engraftment.33 Assessment of colony-forming cell activity in methylcellulose assays showed a mild but consistent inhibitory effect of TRAIL concentrations higher that 200ng/ml on whole BMCs (P<0.001, Figure 6a). In variance, the clonogenic activity of lin progenitors increased by 50% (P<0.001). As shown above, ~30% of these murine progenitors expressed the TRAIL receptor at culture onset. Specificity of the interaction between TRAIL and the cognate receptor was validated with a neutralizing antibody, which abolished the topic effect of TRAIL (Figure 6b). According to these data, the TRAIL/TRAIL-R2 interaction evolves as another mechanism that promotes rather than inhibits donor cell function after transplantation.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Effect of murine TRAIL receptor activation on clonogenesis. Whole BM cells whole BMCs, n=5 and lineage-negative BMCs (lin, n=6) were plated over methylcellulose in the presence of stem cell factor , interleukin -3, erythropoietin and GM-CSF (triplicates). Data were normalized against the number of colonies in control cultures (without TRAIL). (a) The medium was supplemented with increasing doses of TRAIL (n=6 independent cultures). (b) Enhanced clonogenesis of lin cells mediated by 1μg/ml TRAIL was reversed by a blocking anti-TNF-R2 antibody (n=4 independent cultures).

Full figure and legend (56K)



The TRAIL receptors evolve as a complex signaling mechanism that affects multiple hematopoietic cell functions. In general, our findings are consistent with previous reports showing that TRAIL does not negatively impact hematopoietic progenitors.9, 10, 22, 23, 24 However, detailed analysis of the dynamics of cultured cells revealed a complex pattern of cognate receptor induction, differential proliferation and sensitivity to apoptosis of cells expressing DR4 and DR5. In this study, we report that TRAIL induces expression of its two receptors involved in apoptotic signaling, triggers death of nonprogenitor UCB cells in culture and imposes both suppressive and inductive effects on clonogenic activity. In murine BM transplantation, TRAIL evolves as a stimulatory factor that promotes hematopoietic reconstitution.

The contention that resting hematopoietic progenitors are not induced to apoptosis by TRAIL receptors17, 18, 19, 22, 23, 24 best describes the preservation of the engraftment and reconstituting potential, as shown here and in previous studies.23 Human–mouse xenochimerism was not impaired by preincubation with TRAIL, showing that SCID-repopulating cells were spared and not induced into apoptosis. Our findings that exposure of UCB cells to TRAIL for 3 days had no toxic consequences extend the previous observation of insensitivity of CD34+ cells to short-term (1 day) exposure to TRAIL before transplantation.23 Although SCR cells generally relate to human reconstituting cells, their relationship to progenitors that provide short-term radioprotection and durable multilineage engraftment is uncertain.37, 38 Survival of short-term reconstituting progenitors was evident from preservation of CFU-GM activity after exposure to TRAIL. When dead cells were removed, TRAIL-mediated signaling deviates toward tropism in the subset of apoptosis-insensitive cells inducing differentiation in conjunction with GM-CSF.

Selective sensitivity to apoptosis was unrecognized when heterogeneous populations of UCB cells were examined, with prevalent apoptotic death of B-lymphocytes and myeloid cells. However, apoptosis was attenuated by dynamic changes in TRAIL receptor expression in liquid culture, with the ligand inducing expression of DR4 and DR5 and triggering death in 70–80% of these receptor-positive cells. These data confirm the contention that DR4 is associated with apoptotic signaling,3 and consistently its expression was accompanied by slower proliferation rates in the presence of the cognate ligand. TRAIL is best characterized in committed erythroid progenitors, in which it serves as a negative regulator: DR4 induces apoptosis and DR5 arrests differentiation by activation of extracellular signal-regulated protein kinase and mitogen-activated protein kinase.4, 19 In variance, TRAIL has been associated with increased maturation in the myeloid lineage.39 The dose-dependent increase in colony size is evidence that supplementation of exogenous TRAIL at saturating concentrations enhances the maturation of developing myeloid cells. As hematopoietic cells (in particular those of the myeloid lineage) secrete TRAIL in response to inflammatory cytokines,40 this ligand may serve as an autocrine positive feedback mechanism activated in response to injury and/or changing cytokine composition in the environment. In addition, synergism with GM-CSF in recruitment of progenitors to clonogenic activity was observed in human UCB progenitors, with DR4 being the primary signaling mediator of tropism in human cells and DR5 showing a mild inhibitory effect. Consistently, the only murine receptor, analog to human DR5, assumes the inductive function of DR4 in human UCB cells. TRAIL therefore participates in all stages of myeloid cell differentiation, starting at the early stage of progenitor commitment and continuously supporting the growth of developing colonies.

The TRAIL receptors are associated with Fas-associated death domain-mediated recruitment of death domains, caspase-independent transduction of apoptosis and concomitant activation of the nuclear factor-κB pathway,16 which might correspond to modulation of both apoptotic and tropic signaling in UCB cells. These distinct effects were best characterized in isolated subsets of cells, whereas in bulk preparations of human UCB cells and murine BM, TRAIL indirectly affected the assays. For example, the clonogenic activity of bulk MNCs (human and murine) was suppressed at low concentrations of TRAIL, an inhibitory effect attributed to dead cells in the culture.32 The dose-dependent increase in human colony-forming cell engagement to clonogenic activity induced by high concentrations of TRAIL is evidence that its stimulatory effect overrides the indirect inhibition associated with cell death. The threefold enrichment in clonogenic activity attained by preincubation with TRAIL was more pronounced than the 1.5-fold increase when TRAIL was constantly present in the semisolid medium, suggesting that some inhibitory cells might have been removed by this ligand.32 It is therefore speculated that TRAIL specifically removes inhibitory cellular elements, resulting in a marked increase in clonogenic activity after exposure to this ligand before culture.

Within heterogeneous cell compartments and dynamic environments it is difficult to estimate the physiological role of this molecular interaction in hematopoietic homeostasis. Approximately 10% of human lin progenitors express DR4, suggesting that a small fraction of fresh UCB cells is responsive to TRAIL-mediated signaling. However, in murine BM, ~30% of lin progenitors express TRAIL-R2, the only known receptor in rodents.20, 21 Possibly, the differences between murine BMCs and human UCB cells reflect the differential behavior of these cell subsets rather than species-specific variations. UCB cells represent a more primitive subset of circulating progenitors, whereas BM cells live in close contact with and under the control of the marrow stroma. TRAIL has multiple implications in the function of marrow stroma, including endothelial cells,3 bone41 and mesenchyme,42 and the elevated levels of receptor expression in hematopoietic cells suggest involvement of this molecular pair in cell–stroma interaction. A common pattern of upregulation of TRAIL receptors under stress conditions in murine BMCs (transplant) and human UCB cells (ex vivo incubation) renders them more responsive to this signaling pathway. Notably, expression was not induced only by injury factors released as a consequence of radiation of the marrow stroma; the murine receptor was also upregulated in donor cells that homed successfully to the BM of nonirradiated hosts, and expression was induced by interaction with the stroma. In addition, TRAIL receptors were endogenously upregulated in cultured UCB cells in the absence of inductive factors. These data exemplify different situations in which TRAIL receptors are induced by the cognate ligand, in response to culture and interaction with marrow stroma, with a significant number of hematopoietic cells becoming responsive to this signaling pathway.

Clonogenic assays attribute a primary trophic role to TRAIL signaling, which is consistent with the need to enhance progenitor activity under hypoplastic environments such as the irradiated BM. Upregulation of death receptors is not limited to transplantation, as similar dynamics have been observed on activation of murine hematopoiesis with 5-fluorouracil (Pearl-Yafe M, unpublished data). A dose-dependent stimulatory effect of TRAIL in progenitors suggests participation of the ligand in correlating hematopoietic activity to the severity of injury. It is tempting to speculate that this is the reason for inhibition of erythropoiesis,13, 14, 15, 16 while promoting the activity of colony-forming cells for other lineages.2, 3, 4, 12, 13, 14, 30, 31, 39 This pattern of behavior of death receptors in hematopoietic progenitors is not unique to TRAIL in the mouse, and is shared by Fas32, 33 and TNF receptors (Pearl-Yafe M, unpublished data). Importantly, topic signaling mediated by death receptors relies on the intrinsic insensitivity of primitive progenitors to apoptosis triggered by membranal receptors of the TNF superfamily.33

In the context of transplantation, the findings presented here have several implications. First, donor cells can be indeed purged of residual neoplastic cells if the latter are sensitive to TRAIL-induced apoptosis.24 Second, activation of short-term reconstituting cells, corresponding to committed progenitors that are active in clonogenic assays, might augment early UCB cell engraftment, which is one of the problems associated with the use of this rich source of progenitors.43 Third, overexpression of TRAIL in UCB-derived progenitors seems to be safe and might be of benefit. TRAIL is important in the initiation of graft-versus-tumor reactivity,36 and hematopoietic and mesenchymal progenitors can be used as vehicles that efficiently target neoplastic cells in vivo.8, 9, 10, 11 Thus, in addition to the selective sensitivity of neoplastic cells to TRAIL,2, 3, 4 its engraftment-promoting attributes suggest its usefulness as an immunotherapeutic adjuvant to chemotherapy.24, 44 Furthermore, membrane-bound ligands of the TNF family, such as Fas-ligand, are effective in counterattacking graft rejection in addition to graft supportive functions.45 Therefore, TRAIL evolves as a promising immunotherapeutic agent, particularly in emerging approaches, to induce graft-versus-tumor reactivity through nonlife-threatening transplantation under nonmyeloablative conditioning.46, 47, 48, 49


Conflict of interest

The authors declare no conflict of interest.



  1. Aggarwal BB. Signaling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol 2003; 3: 745–756. | Article | PubMed | ISI | ChemPort |
  2. Di Pietro R, Zauli G. Emerging non-apoptotic functions of tumor necrosis factor related apoptosis-inducing ligand (TRAIL)/Apo2L. J Cell Physiol 2004; 201: 331–340. | Article | PubMed | ISI | ChemPort |
  3. Zauli G, Secchiero P. The role of the TRAIL/TRAIL receptors system in hematopoiesis and endothelial cell biology. Cytokine Growth Factor Rev 2006; 17: 245–257. | Article | PubMed | ChemPort |
  4. Secchiero P, Zauli G. Tumor-necrosis-factor-related apoptosis-inducing ligand and the regulation of hematopoiesis. Curr Opin Hematol 2008; 15: 42–48. | Article | PubMed | ChemPort |
  5. Carlo-Stella C, Lavazza C, Locatelli A, Viganò L, Gianni AM, Gianni L. Targeting TRAIL agonistic receptors for cancer therapy. Clin Cancer Res 2007; 13: 2313–2317. | Article | PubMed | ChemPort |
  6. Henson ES, Johnston JB, Gibson SB. The role of TRAIL death receptors in the treatment of hematological malignancies. Leuk Lymphoma 2008; 49: 27–35. | Article | PubMed | ChemPort |
  7. Mahalingam D, Szegezdi E, Keane M, Jong S, Samali A. TRAIL receptor signalling and modulation: Are we on the right TRAIL? Cancer Treat Rev 2009; 35: 280–288. | Article | PubMed | ChemPort |
  8. Song K, Benhaga N, Anderson RL, Khosravi-Far R. Transduction of tumor necrosis factor-related apoptosis-inducing ligand into hematopoietic cells leads to inhibition of syngeneic tumor growth in vivo. Cancer Res 2006; 66: 6304–6311. | Article | PubMed | ChemPort |
  9. Carlo-Stella C, Lavazza C, Di Nicola M, Cleris L, Longoni P, Milanesi M et al. Antitumor activity of human CD34+ cells expressing membrane-bound tumor necrosis factor-related apoptosis-inducing ligand. Hum Gene Ther 2006; 17: 1225–1240. | Article | PubMed | ChemPort |
  10. Kim SM, Lim JY, Park SI, Jeong CH, Oh JH, Jeong M et al. Gene therapy using TRAIL-secreting human umbilical cord blood-derived mesenchymal stem cells against intracranial glioma. Cancer Res 2008; 68: 9614–9623. | Article | PubMed | ChemPort |
  11. Sasportas LS, Kasmieh R, Wakimoto H, Hingtgen S, van de Water JA, Mohapatra G et al. Assessment of therapeutic efficacy and fate of engineered human mesenchymal stem cells for cancer therapy. Proc Natl Acad Sci USA 2009; 106: 4822–4827. | Article | PubMed
  12. Dempsey PW, Doyle SE, He JQ, Cheng G. The signalling adaptors and pathways activated by TNF superfamily. Cytokine Growth Factor Rev 2003; 14: 193–209. | Article | PubMed | ISI | ChemPort |
  13. Gaur U, Aggarwal BB. Regulation of proliferation, survival and apoptosis by members of the TNF superfamily. Biochem Pharmacol 2003; 66: 1403–1408. | Article | PubMed | ISI | ChemPort |
  14. Ware CF. The TNF superfamily. Cytokine Growth Factor Rev 2003; 14: 181–184. | Article | PubMed | ChemPort |
  15. Falschlehner C, Emmerich CH, Gerlach B, Walczak H. TRAIL signalling: decisions between life and death. Int J Biochem Cell Biol 2007; 39: 1462–1475. | Article | PubMed | ChemPort |
  16. Wilson NS, Dixit V, Ashkenazi A. Death receptor signal transducers: nodes of coordination in immune signaling networks. Nat Immunol 2009; 10: 348–355. | Article | PubMed | ChemPort |
  17. De Maria R, Zeuner A, Eramo A, Domenichelli C, Bonci D, Grignani F et al. Negative regulation of erythropoiesis by caspase-mediated cleavage of GATA-1. Nature 1999; 401: 489–493. | Article | PubMed | ISI | ChemPort |
  18. Zamai L, Secchiero P, Pierpaoli S, Bassini A, Papa S, Alnemri ES et al. TNF-related apoptosis-inducing ligand (TRAIL) as a negative regulator of normal human erythropoiesis. Blood 2000; 95: 3716–3724. | PubMed | ISI | ChemPort |
  19. Secchiero P, Melloni E, Heikinheimo M, Mannisto S, Di Pietro R, Zauli G. TRAIL regulates normal erythroid maturation through an ERK-dependent pathway. Blood 2004; 103: 517–522. | Article | PubMed | ISI | ChemPort |
  20. Wu GS, Burns TF, Zhan Y, Alnemri ES, El-Deiry WS. Molecular cloning and functional analysis of the mouse homologue of the KILLER/DR5 tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) death receptor. Cancer Res 1999; 59: 2770–2775. | PubMed | ISI | ChemPort |
  21. Schneider P, Olson D, Tardivel A, Browning B, Lugovskoy A, Gong D et al. Identification of a new murine tumor necrosis factor receptor locus that contains two novel murine receptors for tumor necrosis factor-related apoptosis-inducing ligand (TRAIL). J Biol Chem 2003; 278: 5444–5454. | Article | PubMed | ISI | ChemPort |
  22. Gazitt Y. TRAIL is a potent inducer of apoptosis in myeloma cells derived from multiple myeloma patients and is not cytotoxic to hematopoietic stem cells. Leukemia 1999; 13: 1817–1824. | Article | PubMed | ISI | ChemPort |
  23. Plasilova M, Zivny J, Jelinek J, Neuwirtova R, Cermak J, Necas E et al. TRAIL (Apo2L) suppresses growth of primary human leukemia and myelodysplasia progenitors. Leukemia 2002; 16: 67–73. | Article | PubMed | ISI | ChemPort |
  24. Lee NS, Cheong HJ, Kim SJ, Kim SE, Kim CK, Lee KT et al. Ex vivo purging of leukemia cells using tumor-necrosis-factor-related apoptosis-inducing ligand in hematopoietic stem cell transplantation. Leukemia 2003; 17: 1375–1383. | Article | PubMed | ChemPort |
  25. Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M et al. Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med 1999; 5: 157–163. | Article | PubMed | ISI | ChemPort |
  26. Chinnaiyan AM, Prasad U, Shankar S, Hamstra DA, Shanaiah M, Chenevert TL et al. Combined effect of tumor necrosis factor-related apoptosis-inducing ligand and ionizing radiation in breast cancer therapy. Proc Natl Acad Sci USA 2000; 97: 1754–1759. | Article | PubMed | ChemPort |
  27. Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA, Marsters SA et al. Safety and antitumor activity of recombinant soluble Apo2 ligand. J Clin Invest 1999; 104: 155–162. | Article | PubMed | ISI | ChemPort |
  28. Campioni D, Secchiero P, Corallini F, Melloni E, Capitani S, Lanza F et al. Evidence for a role of TNF-related apoptosis-inducing ligand (TRAIL) in the anemia of myelodysplastic syndromes. Am J Pathol 2005; 166: 557–563. | PubMed | ISI | ChemPort |
  29. Kakagianni T, Giannakoulas NC, Thanopoulou E, Galani A, Michalopoulou S, Kouraklis-Symeonidis A et al. A probable role for TRAIL-induced apoptosis in the pathogenesis of marrow failure. Implications from an in vitro model and from marrow of aplastic anemia patients. Leuk Res 2006; 30: 713–721. | Article | PubMed | ChemPort |
  30. Greil R, Anether G, Johrer K, Tinhofer I. Tuning the rheostat of the myelopoietic system via Fas and TRAIL. Crit Rev Immunol 2003; 23: 301–322. | Article | PubMed | ChemPort |
  31. Testa U. Apoptotic mechanisms in the control of erythropoiesis. Leukemia 2004; 18: 1176–1199. | Article | PubMed | ISI | ChemPort |
  32. Pearl-Yafe M, Stein J, Yolcu ES, Farkas DL, Shirwan H, Yaniv I et al. Fas transduces dual apoptotic and trophic signals in hematopoietic progenitors. Stem Cells 2007; 25: 3194–3203. | Article | PubMed
  33. Pearl-Yafe M, Yolcu ES, Stein J, Kaplan O, Shirwan H, Yaniv I et al. Expression of Fas and Fas-ligand in donor hematopoietic stem and progenitor cells is dissociated from the sensitivity to apoptosis. Exp Hematol 2007; 35: 1601–1612. | Article | PubMed | ChemPort |
  34. Harris DT, LoCascio J, Besencon FJ. Analysis of the alloreactive capacity of human umbilical cord blood: implications for graft-versus-host disease. Bone Marrow Transplant 1994; 14: 545–553. | PubMed | ChemPort |
  35. Song K, Chen Y, Göke R, Wilmen A, Seidel C, Göke A et al. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is an inhibitor of autoimmune inflammation and cell cycle progression. J Exp Med 2000; 191: 1095–1104. | Article | PubMed | ISI | ChemPort |
  36. Schmaltz C, Alpdogan O, Kappel BJ, Muriglan SJ, Rotolo JA, Ongchin J et al. T cells require TRAIL for optimal graft-versus-tumor activity. Nat Med 2002; 8: 1433–1437. | Article | PubMed | ISI | ChemPort |
  37. Eaves C, Glimm H, Eisterer W, Audet J, Maguer-Satta V, Piret J. Characterization of human hematopoietic cells with short-lived in vivo repopulating activity. Ann NY Acad Sci 2001; 938: 63–70. | PubMed | ChemPort |
  38. Manz MG. Human-hemato-lymphoid-system mice: opportunities and challenges. Immunity 2007; 26: 537–541. | Article | PubMed | ChemPort |
  39. Secchiero P, Gonelli A, Mirandola P, Melloni E, Zamai L, Celeghini C et al. Tumor necrosis factor-related apoptosis-inducing ligand induces monocytic maturation of leukemic and normal myeloid precursors through a caspase-dependent pathway. Blood 2002; 100: 2421–2429. | Article | PubMed | ISI | ChemPort |
  40. Griffith TS, Wiley SR, Kubin MZ, Sedger LM, Maliszewski CR, Fanger NA. Monocyte-mediated tumoricidal activity via the tumor necrosis factor-related cytokine, TRAIL. J Exp Med 1999; 189: 1343–1354. | Article | PubMed | ISI | ChemPort |
  41. Zauli G, Rimondi E, Stea S, Baruffaldi F, Stebel M, Zerbinati C et al. TRAIL inhibits osteoclastic differentiation by counteracting RANKL-dependent p27Kip1 accumulation in pre-osteoclast precursors. J Cell Physiol 2008; 214: 117–125. | Article | PubMed | ChemPort |
  42. Secchiero P, Melloni E, Corallini F, Beltrami AP, Alviano F, Milani D et al. Tumor necrosis factor-related apoptosis-inducing ligand promotes migration of human bone marrow multipotent stromal cells. Stem Cells 2008; 26: 2955–2963. | Article | PubMed | ChemPort |
  43. Brown JA, Boussiotis VA. Umbilical cord blood transplantation: basic biology and clinical challenges to immune reconstitution. Clin Immunol 2008; 127: 286–297. | Article | PubMed | ChemPort |
  44. Baumann S, Zhu JY, Giaisi M, Treiber MK, Mahlknecht U, Krammer PH et al. Wogonin sensitizes resistant malignant cells to TNF{alpha}- and TRAIL-induced apoptosis. Blood 2006; 108: 3700–3706. | Article | PubMed | ChemPort |
  45. Pearl-Yafe M, Yolcu ES, Stein J, Kaplan O, Yaniv I, Shirwan H et al. Fas ligand enhances hematopoietic cell engraftment through abrogation of alloimmune responses and nonimmunogenic interactions. Stem Cells 2007; 25: 1448–1455. | Article | PubMed | ChemPort |
  46. Burroughs L, Storb R. Low-intensity allogeneic hematopoietic stem cell transplantation for myeloid malignancies: separating graft-versus-leukemia effects from graft-versus-host disease. Curr Opin Hematol 2005; 12: 45–54. | Article | PubMed | ISI
  47. Resnick IB, Shapira MY, Slavin S. Nonmyeloablative stem cell transplantation and cell therapy for malignant and non-malignant diseases. Transpl Immunol 2005; 14: 207–219. | Article | PubMed | ISI | ChemPort |
  48. Yaniv I, Stein J. Reduced-intensity conditioning in children: a reappraisal in 2008. Bone Marrow Transplant 2008; 41 (Suppl 2): S18–S22. | Article | PubMed
  49. Yaniv I, Ash S, Farkas DL, Askenasy N, Stein J. Consideration of strategies for hematopoietic cell transplantation. J Autoimmun 2009; 33: 255–259. | Article | PubMed | ChemPort |


This work was funded by grants from the Frankel Trust for Experimental Bone Marrow Transplantation (JS, IY and NA) and the United States–Israel Binational Science Foundation (2003276 to NA and IY). We thank the technical staff for their outstanding support: Mrs Natalia Binkovsky, Mrs Ela Zuzovsky and Mrs Ana Zemlianski.