Consent forms detailing the use of human cord blood (CB) and bone marrow (BM) in our experimental protocol were approved by the American University of Beirut Medical Center (AUB-MC) Institutional Review Board (IRB). All animal experiments were approved by the AUB Institutional Animal Care and Use Committee (IACUC). Animals were handled under pathogen-free sterile conditions, maintained under micro-isolators, and fed sterile food. Finally, this study was performed in accordance with the principles of the declaration of Helsinki.

Samples of CB were obtained from discarded placental and umbilical tissues, after informed consent from delivering mothers attending the AUB-MC. BM cells were collected after informed consent of patients attending the hospital for BM aspirations. Bone marrow samples were derived from patients diagnosed with Hodgkin disease or Non-Hodgkin Lymphomas, who had routine BM assessment as staging for their disease and who were found to have a normal bone marrow. CB and BM were collected in tubes containing EDTA, sodium citrate or heparin, and were processed within 24 hours.

Fresh samples of CB were diluted 1∶3 in Iscove’s modified Dulbecco’s medium (IMDM, GIBCO-BRL, Life Technologies, Saint Aubin, France) containing 10% fetal bovine serum (FBS, Life Technologies). Samples were enriched for mononuclear cells (MNC) by centrifugation on Ficoll/Hypaque (GE Healthcare Life Sciences, Uppsala, Sweden). This step of density gradient centrifugation would allow the purification of MNCs from red blood cells and neutrophils. These cells were either used immediately for the experiments described below or cryopreserved in 10% dimethylsulfoxide (DMSO, Sigma GmbH, Roedermark, Germany) in FBS. Later, cryopreserved cells were thawed at 37°C, washed in phosphate-buffered saline (PBS), and resuspended in medium for the specific experiment. After thawing, cell viability was monitored by Trypan blue dye exclusion.

Enrichment of HSCs was obtained by isolation of the CD34⁺ fraction from BM cells collected from free of disease patients after informed consent. MS columns and µMACS magnetic beads (Miltenyi Biotec, Surrey, UK) or EasySep columns (Stem Cell Technologies, Vancouver, Canada) were used for this purpose. Using a magnetic separation procedure, the CD34⁺ fraction of HSCs was positively selected by labeling with a CD34 antibody that is anchored to magnetic beads. As a result, about one million CD34 positively selected cells were isolated from 100 million BM MNCs. Purity of the isolated fraction was measured by flow cytometry and was always determined to be greater than 90%. Data acquisition and analysis was performed with CellQuest or Diva softwares from Becton Dickinson (BD, New Jersey, USA) or with FlowJo software (Treestar, Oregon, USA).

Extracts are sterile acellular lyophilizates from bovine BM matrix and were supplied by Ossacur AG, (Oberstenfeld, Germany). This new technology is based on the extraction of extracellular matrix proteins from animal tissue and contains a mixture of all growth factors and cytokines within the matrix. Briefly, extracts were prepared, aseptically, from bone marrow of disease-free, young (<12 months old) calves of closed herds. Bone marrows were pulverized and delipidated with acetone for 60 min at 4°C. The resulting particles were demineralized in 0.6 N hydrochloric acid for 60 min at 4°C. Particles were then washed in deionized water, extracted with 4 M guanidine hydrochloride and ultrafiltered using 3 K nominal molecular cut-off membranes. Extracts were then reconstituted in water from the initial stock solution in order to have a final concentration of 0.0138 mg/ml. This concentration was selected based on previous liquid culture experiments, which showed optimum differentiation results (data not shown).

MNCs were cultured, in 25 cm² Falcon dishes, in Iscove’s modified Dublecco’s medium (IMDM, GIBCO-BRL, Life Technologies,) supplemented with 10% fetal bovine serum (FBS) for 7 days. On the other hand, about 20,000 CD34⁺ cells were seeded in 6-well plates and suspended at 10⁵ cells/mL in serum-free StemSpan H3000 culture medium (Stem Cell Technologies). The CD34+ cells were supplemented with SCF (25 ng/mL) and G-CSF (10 ng/mL) recombinant human factors for 13 days (Invitrogen, Life Technologies). Cells were treated with BME, diluted with culture medium to have a final concentration of 0.0138 mg/ml. Control cells were left un-treated, i.e. without the addition of BME. Cell suspensions were incubated at 37°C in a 5%CO2/95% air atmosphere, after which they were collected, washed in IMDM, counted by Trypan blue exclusion, analyzed for progenitor and stem cells, and engrafted into NOD-SCID or NSG mice.

CD34⁺ cells from each CB sample were suspended in IMDM/10% FBS and plated on methylcellulose-based media (HSC002, HSC003, and HSC004, R & D systems, Lille, France), with or without the addition of BME. The human methylcellulose base media HSC002 contains 1.3% methylcellulose, 25% FBS, 2% BSA, 2 mM L-glutamine, and 5×10⁻⁵ M beta-mercaptoethanol. The human methylcellulose complete media HSC003 contains the basic HSC002 media and is supplemented by rh-SCF (50 ng/mL), rh-GM-CSF (10 ng/mL), rh-IL-3 (10 ng/mL), and rh-Epo (3 U/mL). The human methylcellulose complete media HSC004 is similar to HSC003 media but without Epo. Cells were plated, in duplicate, at a final concentration of 100 cells/9.6 cm² dish. Colonies consisting of CFU-GM (Granulocyte Macrophage), CFU-E/BFU-E (Erythroid), and CFU-GEMM (Granulocyte Erythroid Macrophage Megakaryocyte) were counted at days 14–16 under an inverted microscope.

NOD/SCID (NOD.CB17-Prkdc^scid/J) and NSG mice (NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ) were obtained from the Jackson Laboratory (Maine, USA).

We used the alkylating agent, busulfan (Busilvex, Pierre Fabre, Boulogne, France), as conditioning regimen for transplantation. Busulfan is a cell cycle non-specific drug that functions similarly to total body irradiation (TBI). It provides equivalent sensitivity to the standard irradiation regimen conditioning protocol in immune-compromised mice, as demonstrated by HSC detection, evaluated by limiting dilution analysis of SRC [10]. Mice aged 6–10 weeks received intra-peritoneal (ip) injection of Busulfan at a dose of 22 mg/kg, for 3 consecutive days before transplantation. HSCs are then injected intravenously.

Purified cell populations (MNCs or CD34⁺), at the indicated dose, were transplanted by tail-vein intravenous (iv) injection into Busilvex treated mice according to standard protocols. Briefly, human cells were suspended in 250 µl IMDM tissue culture medium (Sigma) containing 1% human serum albumin (HSA, ZLB Behring GmbH, Marburg, Germany). Groups of five mice each, received a dose of viable human CB MNCs (1.5×10⁶, 3×10⁶, or 4×10⁶ cells/mouse). If purified CD34⁺ cells were used, mice were transplanted with 50×10³ cells/mouse. The effect of BME on engraftment was studied by culturing the same amount of cells with or without BME for 7 days.

Mice were sacrificed at 3 or 6 weeks after transplantation. Mice were euthanized by Isoflurane inhalation (Abbot Laboratories, Kent, UK) then cervical dislocation. BM was collected from the long bones of the two posterior limbs (femurs, tibiae, and iliac crests) by flushing with IMDM containing 10% FBS. Single cell suspensions were prepared by drawing the BM cells through a 27-gauge needle, then expelling them back through the needle and through a nylon mesh (70 µ strainers, BD).

Collected BM samples from NOD/SCID or NSG mice were stained with various human specific monoclonal antibodies and analyzed by flow cytometry, after 3 or 6 weeks of transplantation. Briefly, approximately 10⁵ cells were suspended in 50 µl of PBS/5% FCS/5% human serum for 30 min at 4°C, in order to block Fc receptors. Cells were then washed in PBS containing 10%FBS, and incubated with monoclonal antibodies at a concentration of 5 µg/ml for 30 min at 4°C. Cells were stained with human specific FITC-conjugated anti-CD33, PE-conjugated anti-CD19 and Phycoerythrin-Cyanin 5 (PE-Cy5/PC5)-conjugated anti-CD45. Cells were also stained with human specific FITC-conjugated anti-CD235a (Glycophorin A), PE-conjugated anti-CD36 and Peridinin-Chlorophyll-Protein complex (PerCP)-conjugated anti-CD45 antibodies. Following staining, cells were washed in PBS/5% FBS and analyzed on a flow cytometer (BD FACSCalibur or BD FACSCanto, New Jersey, USA). Gates were set up to exclude nonviable cells and debris. For each mouse analyzed, an aliquot of cells was also stained with isotype control mouse immunoglobulin G (IgG) conjugated to FITC, PE, PC5 or PerCP (mouse IgG1, IgG2a, IgG2b, or IgM; Becton-Dickinson). BM cells from an untransplanted mouse were stained in parallel as a negative control. Fluorescence levels excluding greater than 99% of the cells in these negative controls were considered to be positive and specific for human staining. About 50,000 events were collected for data analysis. Data acquisition was done using SciQuest or Diva softwares from Becton Dickinson while data analysis was performed on flowJo (Treestar, USA). The level of engraftment (i.e. % of human cells) in the BM of NOD/SCID or NSG mice was then measured by adding the engraftment levels from the human leukocytes population (CD45⁺ cells) in addition to immature erythroblasts (CD45⁻CD36⁺CD235a⁺). Furthermore, multi-lineage differentiation of human cells in murine BM was also determined. Indeed, human CD45⁺ cells were further gated based on characteristic forward and side scatter properties into lymphoid and myeloid populations. Multi-lineage engraftment is defined by the presence of separate CD45⁺CD33⁺ (myeloid) and CD45⁺CD19⁺ (lymphoid) populations with appropriate scatter characteristics. Moreover, erythroid populations made up of immature erythroblasts (CD45⁻CD36⁺CD235a⁺) or mature RBC’s (CD45⁻CD36⁻CD35a⁺) were also determined as explained in previous publications [10], [11]. Finally, in order to confirm human engraftment, murine BM cells were also stained with murine specific FITC-conjugated anti-CD45. The engraftment pattern of mice transplanted with purified CD34⁺ cells was similar to what is observed after transplantation of unsorted CB cells. CD45 and CD34 antibodies were purchased from Becton Dickinson (NJ, USA). All other antibodies were from Beckman Coulter (California, USA). It’s important to note that myeloid and B lymphoid reconstitution can be easily attained in NOD-SCID and NSG mice by the transplantation of human HSCs. Several reports have suggested that T cells may be able to develop in the thymus of NOD-SCID mice transplanted with human CD34 cells. In particular, van der Loo et al. succeeded in repopulating a mouse thymus with human cells by administering G-CSF and stem cell factor. However, most of the T cells were double positive (DP) cells, and a functional analysis of these T cells has not been completed [12]. Recently, Yahata et al reported a new model for the study of T-Lymphocytes in the NOD/Shi-scid, IL2-Receptor gamma null mouse background [13].

Results are expressed as individual data or as the mean±SD. Statistical comparisons were performed using the Student’s t-test in order to determine statistical significance. Paired t-test was used for methylcellulose experiments. The p value was determined and values for p<0.05 or p<0.001 were considered significant. The Microsoft Excel and SPSS Packages were used to perform statistical analysis.

We first investigated the effect of acellular BME on the proliferation and differentiation of CB hematopoietic progenitors using colony-forming assays in methylcellulose. A total of 20,000 CB MNCs from each patient (n = 6) were plated on different types of methylcellulose-based media (HSC002, HSC003, and HSC004), in the presence or absence of BME (CB+BME or CB, respectively). In HSC002 media, which does not contain any added growth factors, the addition of BME increased the number of CFU-G/M colonies by 1.7-fold (Figure 1A, upper panel). In HSC003 media, which contains SCF, GM-CSF, IL3 and Epo, the addition of BME resulted in a significant increase of approximately 1.5 fold in the number of myeloid colonies (CFU-G/M), erythroid colonies (BFU-E) and mixed colonies (CFU-GEMM) (p<0.05 for all colony types) (Figure 1A, middle panel). Finally, in HSC004 which contains SCF, GM-CSF and IL3, BME increased the number of myeloid colonies by at least 2-fold (Figure 1A, lower panel). Similar results were obtained when positively selected CB CD34⁺ cells were used. Indeed, when 100 CD34⁺ cells from each CB sample (n = 4), were plated on the basic methylcellulose media without any added factors (HSC002), myeloid colonies (CFU-G/M) appeared only in the presence of BME (Figure 1B, upper panel). Moreover, addition of BME resulted in a significant increase of myeloid colonies (2 fold) in HSC003 and HSC004 (p<0.001), of erythroid colonies in HSC003 (p<0.001), and of GEMM in HSC003 (p<0.05) (Figure 1B, middle and lower panels). Since HSC002 and HSC004 media presented neither CFU-GEMM nor BFU-E (erythroid) colonies, we may speculate that BME do not contain any EPO activity. Overall, BME increased the number of myeloid and erythroid colonies in methylcellulose culture from both CB MNC and CB CD34⁺ cells.

The current models for assessment of transplantable human HSCs are based on the use of immune-deficient mice such as NOD/SCID [14] (NOD.CB17-Prkdc^scid/J) and NSG [15] (NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ) mice strains. NOD/SCID mice are characterized by low NK cell activity [16], in addition to B-cell, T-cell, and complement defects. NSG mice represent a more permissive xeno-transplantation model generated from NOD/SCID and the IL-2 receptor gamma c (CD132) knock-out mice and characterized by complete lack of NK activity.

A representative example of the data in an NSG mouse is shown in Figure 2. At six weeks, the percentage of engraftment (CD45⁺ fraction) increased from 5% in an NSG mouse injected with untreated CB MNC cells to about 24% in a mouse injected with cells treated with BME. This increase in engraftment was mainly due to an increase in the number of myeloid cells (CD33⁺, R4 region), which represented more than 90% of engrafted cells. Finally, the percentage of engraftment due to immature erythroblasts increased from around 1.11% in the mouse injected with untreated CB MNC cells to around 2.92% in the mouse injected with CB MNC treated with BME (Figure 2).

Two groups of NOD/SCID mice (n = 5 each) received CB MNCs that were put in culture for 7 days, in the presence or absence of BME. A third group (n = 5) served as a negative control and was injected a saline solution. Mice were sacrificed six weeks after transplantation. The percentage of engraftment, as determined by the human pan leucocytic CD45 marker, significantly increased from 3% in the mice that have received untreated CB MNC cells to more than 10% in the mice which received CB MNC cells cultured in the presence of BME (p<0.05) (Figure 3, upper panel). When subpopulations were examined, the increase in engraftment was significant for the myeloid populations (p<0.05) but not for lymphoid cells (Figure 3, lower panel). Similar results were obtained in NSG mice (n = 4 for each group) which received CB MNCs that were put in culture for 7 days, in the presence or absence of BME. The percentage of engraftment, at 6 weeks of transplantation, significantly increased from 5% in the mice that have received cells cultured without BME to 20% in the mice that received cells cultured with BME (p<0.05) (Figure 4, upper panel). When subpopulations were examined, the increase in engraftment was significant for both myeloid (p<0.05) and lymphoid (p<0.05) populations (Figure 4, lower panel).

In order to assess erythroid engraftment, the percentage of engraftment was measured at three weeks instead of six weeks. The time point at 3 weeks allows seeing engraftment due to immature erythroblasts and RBC’s. NSG mice (n = 3 for each group) received CB MNCs that were put in culture for 7 days, in the presence or absence of BME. In these conditions, the percentage of engraftment significantly increased from 12% in the mice that have received cells cultured without BME to 27% in the mice that received cells cultured with BME (p<0.05) (Figure 5, upper panel). When subpopulations were examined, the increase in engraftment was only significant for immature erythroblasts (CD45⁻CD36⁺, p<0.05) (Figure 5, lower panel).

We then investigated the effect of BME on positively selected CD34 cells derived from normal human bone marrow, using the immune-deficient NSG mice model. Two groups of NSG mice (n = 5 each) received 50×10³ CD34 positive cells that were put in culture for 13 days in the presence or absence of BME. A third group (n = 5) served as a negative control and was injected a saline solution. Mice were sacrificed six weeks after transplantation. The percentage of engraftment, as determined by the human pan leucocyte CD45 marker, significantly increased from 8% in the mice that have received untreated CD34 positive cells to more than 31% in the mice that received CD34 positive cells cultured in the presence of BME (p<0.05) (Figure 6, upper panel). When subpopulations were examined, the increase in engraftment was significant for both myeloid (p<0.05) and immature erythroblasts (p<0.05) populations but not for lymphoid cells (Figure 6, lower panel).