Hematopoietic stem cell engraftment: a direct comparison between intramarrow and intravenous injection in nonhuman primates

Experimental Hematology 35 (2007) 1132–1139

Hematopoietic stem cell engraftment: a direct comparison between intramarrow and intravenous injection in nonhuman primates Chul Won Junga,*, Brian C. Bearda,*, Julia C. Morrisa, Tobias Neffa, Katherine Beebea, Barry E. Storera,b, and Hans-Peter Kiema,b a

Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, Wash., USA; Department of Medicine, University of Washington School of Medicine, Seattle, Wash., USA

b

(Received 14 August 2006; revised 5 April 2007; accepted 6 April 2007)

Objective. Recent studies in the mouse model have shown improved engraftment of repopulating cells when cells were administered by intramarrow (IM) vs intravenous (IV) injection. Here we wished to determine if IM injection was feasible and would result in improved engraftment in a clinically relevant large animal model. Materials and Methods. We used a competitive repopulation assay to directly compare IM vs IV injection in four baboons. CD34+ autologous bone marrow cells were split into two equal fractions and transduced with either green fluorescent protein (GFP) or yellow fluorescent protein (YFP). Gene-marked cells were infused by IM or IV administration after myeloablative irradiation. Results. Peripheral blood granulocyte marking peaked at 2 to 3 weeks after transplantation and decreased thereafter before stabilizing. In all animals, marking levels of IM-injected cells (GFP) were lower than those of IV-injected cells (YFP) early after transplantation. However, in two of the four monkeys, GFP marking steadily increased after 2 months resulting in higher marking levels from IM-injected cells. In one animal, this trend sustained up to the last followup at 1 year after transplantation, with marking levels of 63.4% and 9.7% from IM- and IVinjected cells, respectively. Transplantation of both IM- and IV-injected CD34+ cells resulted in polyclonal multilineage engraftment. Conclusion. Our data show efficient gene marking after IM injection and suggest a different engraftment profile for IM- vs IV-injected repopulating cells. Ó 2007 International Society for Experimental Hematology. Published by Elsevier Inc.

Intravenously injected hematopoietic stem cells (HSCs) have to travel through the blood circulation and lungs, which can lead to a significant loss of HSCs [1–3]. In an effort to improve engraftment efficiency several groups have explored direct injection of hematopoietic repopulating cells into the bone marrow. Yahata et al. [4] reported that direct intramarrow (IM) injection of human cord blood CD34þ CD38– cells into nonobese diabetic/severe combined immune-deficient (NOD/SCID) mice was associated with a 15-fold increase of NOD/SCID repopulating cells and higher engraftment in secondary transplantation compared to intravenous injection. Also, Mazurier et al. [5] found that direct intrafemoral injection could induce more

rapid myeloerythroid repopulation than intravenous (IV) injection in NOD/SCID mice. More efficient engraftment would be important for stem cell transplantation and stem cell gene therapy applications, in particular in settings where only low numbers of HSCs are available for transplantation. Thus, we have studied the engraftment dynamics of baboon CD34þ cells that were administered by direct IM injection and IV injection into the same baboons. To allow comparison to studies in the immune-deficient mouse model, we have also injected gene-modified cells, from three baboons, into NOD/SCID mice via IM and IV injection.

Materials and methods Offprint requests to: Hans-Peter Kiem, M.D., Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue N, D1-100, P.O. Box 19024, Seattle, WA, 98109-1024; E-mail: [email protected] *Drs. Jung and Beard contributed equally to this work.

Retrovirus vectors We used two gamma-retroviral vectors in a competitive repopulation assay with different markers, MNDMFG.P140K.ires.EGFP

0301-472X/07 $–see front matter. Copyright Ó 2007 International Society for Experimental Hematology. Published by Elsevier Inc. doi: 10.1016/j.exphem.2007.04.004

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

vector encodes a bicistronic mRNA with the P140K mutant form of methylguanine methyltransferase (P140K) and enhanced green fluorescent protein (EGFP) under the control of the 50 -modified viral long terminal repeat and MNDMFG.EYFP vector encodes EYFP under the control of the same viral long terminal repeat. Gibbon ape leukemia virus (GALV)-pseudotyped gamma-retrovirus stocks were produced harvesting virally conditioned medium (VCM) from Phoenix-GALV-MNDMFG.P140K.ires.EGFP c32 and Phoenix-GALV-MNDMFG.EYFP c10 packaging cells as described previously [6]. Virus-containing medium was filtered through a 0.45-mM filter and stored at 70 C. Virus harvests had titers of approximately 1  105 IU/mL as evaluated by transduction of HT1080 cells. Gene transfer into baboon CD34-enriched cells CD34þ cells were enriched from baboon marrow leukocytes using immunoglobulin M monoclonal antibody 12.8 with an immunomagnetic column technique (Miltenyi Biotec, Auburn, CA, USA) as described [6–8]. Equal numbers of CD34-enriched cells were prestimulated for 48 hours in tissue culture treated on 75-cm2 canted-neck flasks (Corning, Corning, NY, USA) in Iscove’s medium containing 10% fetal bovine serum (Hyclone, Logan, UT, USA) in the presence of stem cell factor (SCF), granulocyte colony-stimulating factor (G-CSF), interleukin (IL)-3, IL-6 (no IL6 was present for M00228 and J01174 for prestimulation or virus transduction), FMS-like tyrosine kinase 3-ligand (FLt-3), and megakaryocyte growth and development factor (MGDF) at 100 ng/mL each. After prestimulation, cells were transferred to nontissue culture treated 75-cm2 canted-neck flasks (Corning) that had been coated with CH-296 (RetroNectin; Takara Shuzo, Otsu Japan) at 2 mg/cm2 and preloaded twice with virus-containing medium [9]. Cells were exposed to virus containing medium for 4 hours supplemented with the same serum and cytokines as stated previously, then collected and resuspended in fresh media with cytokines and incubated overnight at 37 C in a 5% CO2 humidified incubator. The following day, another 4-hour exposure to virus was performed as stated previously before the cells were harvested. An aliquot of each transduction culture was removed for injection into NOD/SCID mice before the cells were reinfused into the irradiated baboon (1020 cGy total body irradiation). Autologous baboon transplantation by IM or IV injection of CD34þ-enriched cells Healthy juvenile baboons were housed at the University of Washington Regional Primate Research Center under conditions approved by the American Association for the Accreditation of Laboratory Animal Care. Studies were conducted under protocols approved by the Institutional Review Board and Animal Care and Use Committees. Autologous baboon transplantations and all other procedures were performed as described previously [8]. Animals were administered daily subcutaneous injections of recombinant human G-CSF (rhG-CSF) (100 mg/kg; Amgen, Thousand Oaks, CA, USA). After 5 days of growth factor mobilization, 70 to 80 mL marrow was aspirated from the humeri and/or femora and collected in preservative-free heparin. In preparation for transplantation, all animals received a myeloablative dose of total body irradiation, 1020 cGy, administered from a linear accelerator at 7 cGy/minute as two equally divided doses 24 hours apart. After the second day of radiation, PhGALV-MNDMFG.P140K. ires.EGFP transduced cells were injected directly into the femoral

1133

bone marrow (IM) and PhGALV-MNDMFG.EYFP transduced cells were infused through an inserted central vein catheter. For direct IM injection, 1 to 3 mL of transduced cells suspended in 2% human albumin/Hank’s salt solution were slowly injected over 20 minutes through a bone marrow aspiration needle that was introduced into the proximal end of the femur after general anesthesia. In animals J01174 and M00165 a tourniquet was applied immediately prior to cell infusion and maintained for approximately 20 minutes after infusion to temporarily interrupt blood flow to the injected bone region. Blood flow interruption was verified by ultrasound. After transplantation, animals were given 100 mg/kg rhG-CSF intravenously once daily, starting at day 0 and continuing until their peripheral blood neutrophil counts were more than 1000/mL. Transplantation of NOD/SCID-b2m/ mice with baboon CD34þ cells For SCID repopulating cell assay, retroviral transduced baboon CD34þ cells were transplanted into irradiated NOD/SCID and NOD/SCID-b2m/ mice. Xenotransplantation into NOD/SCID mice was similar to a published standard protocol [10]. All mice were offspring of breeders purchased from The Jackson Laboratory (Bar Harbor, ME, USA). All animals were handled under sterile conditions and maintained in microisolators. Before transplantation, 6- to 8-week-old mice underwent total body irradiation with 375 cGy at 20 cGy/minute from a linear accelerator source. Within 24 hours, 1  106 CD34þ enriched baboon cells that had been transduced either with PhGALV-MNDMFG.P140K.ires.EGFP or with PhGALV-MNDMFG.EYFP were administered to the irradiated mice by direct IM injection and by IV injection, respectively. For intramarrow injection, a 29-gauge needle was introduced into the distal femur through the knee joint and 30 mL of transduced baboon CD34þ cells were injected into the bone marrow cavity. For IV injection, a 200 mL cell suspension was infused through the tail vein. As a control, in additional mice received 2  106 mock-transducted cells. Flow-cytometric analysis of baboon hematopoietic cells Leukocytes, isolated by ammonium chloride red cell lysis from heparinized peripheral blood and bone marrow samples, drawn at multiple time points after transplantation, were analyzed for green fluorescent protein (GFP)/yellow fluorescent protein (YFP)-expression on a FACS Vantage or FACS Calibur (Becton Dickinson, San Jose, CA, USA) as described. Transgene-expression in granulocyte, monocyte, and lymphocyte populations was determined either by gating based on forward and right-angle light scatter characteristics or based on expression of lineage-specific CD markers (data not shown). Baboon hematopoietic cell engraftment after IV vs IM injection in murine bone marrow Six or 7 weeks after transplantation, mice were euthanized and bone marrow was harvested from the injected right femur (injected bone marrow) and left femur and both tibiae pooled together (noninjected bone marrow). Red blood cells were lysed by adding 2.5 mL lysis buffer (155 mM NH4Cl, 10 mM KHCO3, and 0.1 mM ethylenediamine tetraacetic acid), and the remaining white cells were washed in phosphate-buffered saline containing 2% fetal bovine serum. To assess overall engraftment (Supplemental Fig. 3) and gene marking, aliquots containing at least 5  105 cells

1134

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

were stained with a CD45 monoclonal antibody conjugated to peridinin chlorophyll protein (clone D058-1283; Becton Dickinson, San Jose, CA, USA) as a panleukocytic marker for baboon hematopoietic cells. Immediately before flow cytometric analysis (LSR II, Becton Dickinson), 40 ,6-diamidino-2-phenylindole dihydrochloride hydrate (Sigma, St Louis, MO, USA) was added to the cells and used as a viability marker. GFP/YFP gene marking analysis of the IV- vs IM-injected baboon cells was performed by gating on the double-positive GFPþ or YFPþ and CD45þ bone marrow cells from injected or noninjected murine femurs. Analysis of EGFP/EYFP expression in a colony-forming unit assay Transduced and mock-transduced CD34-selected cells were cultured (1000 cells per 35-mm plate) in triplicate in a double-layer agar culture system as described [8]. Briefly, isolated cells were cultured in a-minimal essential medium supplemented with 15% fetal bovine serum (Hyclone), 0.1% bovine serum albumin (fraction V; Sigma), 0.3% (wt/vol) agar (BioWhittaker, Rockland, ME, USA) overlaid on medium with 0.5% agar (wt/vol) containing 100 ng/mL SCF, IL-3, IL-6 (except M00228 and J01174), granulocyte macrophage colony stimulating factor (GM-CSF), G-CSF, and 4 U/mL erythropoietin (provided by G. Molineux, Amgen). Cultures were incubated at 37 C in a 5% CO2 humidified incubator. Colonies were evaluated for EGFP/EYFP expression at day 14 of culture using an inverted fluorescence microscope. Real-time polymerase chain reaction assay Polymerase chain reaction (PCR) amplification and analysis of the EYFP and EGFP gene were performed by using a quantitative real-time PCR assay (TaqMan). DNA (300 ng) was amplified at least in duplicate with EYFP-specific primers (50 -GGA TTG CAC GCA GGT TCT C-30 and 50 -AGA GCA GCC GAT TGT CTGTT-30 ) and a fluorescence-tagged probe (50 -FAM-TGC CCA GTC ATA GCC GAA TAG CCT CTC CAT-TAMRA-30 ; Synthegen, Houston, TX). For EGFP, the specific primers 50 -TAC ACA AAT CGC CCG CAG A-30 and 50 -AGC CTG GTC GAA CGC AGAC-30 were used with the probe 50 -FAM-CGA CTT CTA CAC AGC CAT CGG TCC AGA-TAMRA-30 . These primers and probes were designed using Primer Express software

(Perkin-Elmer, Foster City, CA, USA). Standards consisted of dilutions of DNA extracted from cell lines transduced with a single copy of the EGFP or EYFP vector. Negative controls consisted of DNA extracted from peripheral blood mononuclear cell obtained preinfusion or from control animals or water. A b-globin–specific primer/probe combination (50 -CCT ATC AGA AAG TGG TGG CTG G-30 , 50 -TTG GAC AGC AAG AAA GTG AGC TT-30 , probe 50 -TGG CTA ATG CCC TGG CCC ACA AGT A-TAMRA-30 ) was used to adjust for equal loading of DNA per reaction. Reactions were run using the ABI master mix (Applied Biosystems, Branchburg, NJ, USA) on the ABI Prism 7700 sequence detection system (Applied Biosystems) using the following thermal cycling conditions: 50 C for 2 minutes and 95 C for 10 minutes, followed by 40 cycles of 95 C for 15 seconds and 60 C for 1 minute. Linear amplification–mediated PCR of retrovirus integration sites One-hundred nanograms of DNA served as template for linear amplification–mediated (LAM)-PCR that was performed as described previously with modifications [11–14]. PCR reactions are separated on a 4% to 20% TBE gradient acrylamide XCell SureLock Mini-Cell (Invitrogen, Carlsbad, CA, USA).

Results Transduction of CD34þ cells We used a competitive repopulation assay in four baboons to study engraftment of CD34þ cells administered by either IM or IV injection (Fig. 1 and Table 1). Marrow CD34þ cells were split into two equal fractions and then transduced with either GFP- or YFP-containing gamma-retrovirus vectors. One fraction was then administered IV and the other fraction was administered IM injection. Multiplicity of infection was similar between the two arms and in the range of 0.2 to 0.4. Transduction efficiency before infusion of CD34þ cells was between 76% and 94% and was similar between GFP and YFP arms in all the four baboons. There

Figure 1. Experimental scheme. Autologous transplantation of four baboons was performed by intramarrow and intravenous injection of bone marrow cells that had been transduced with two different marker genes. An aliquot of cells was also transplanted into irradiated nonobese diabetic/severe combined immune-deficient (NOD/SCID)-b2m/ mice by the two routes. GFP 5 green fluorescent protein; YFP 5 yellow fluorescent protein.

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

1135

Table 1. Transduction characteristics of the four baboons that received autologous gene-modified CD34þ cells

Animal

Weight (kg)

Route of administration

Transgene

MOI

Total cells infused (107/kg)

Gene marked cells infuseda (107/kg)

M00228

6.4

M00165

6.8

J01174

7.6

M01277

6.4

IM IV IM IV IM IV IM IV

GFP YFP GFP YFP GFP YFP GFP YFP

0.2 0.2 0.4 0.3 0.3 0.2 0.3 0.2

1.22 1.08 1.06 1.15 1.08 0.97 0.27 0.27

1.01 0.93 1.00 1.02 0.85 0.74 ND ND

GFP 5 green fluorescent protein; IM 5 intramarrow; IV 5 intravenous; MOI 5 multiplicity of infection; ND 5 not determined; YFP 5 yellow fluorescent protein. a Calculated based on the percentage of positive cells from flow cytometric analysis.

was also no difference in colony-forming cells between the two groups (data not shown). Engraftment profiles of IM vs IV injected CD34þ cells in baboons Peak marking levels of GFP- and YFP-modified cells were observed at 2 to 3 weeks after transplantation and decreased thereafter before stabilizing (Fig. 2A and Fig. 3). Marking levels in granulocytes (as determined by flow-cytometric analysis) derived from IM-injected CD34þ cells were lower than from IV-injected cells in all four animals early after transplantation. In two of four animals, marking from the IM-injected cells increased, which resulted in higher marking levels for IM- vs IV-injected cells at 1 and 3 months after transplantation (Fig. 2). These gene-marking levels were confirmed by real-time PCR demonstrating that the trends were actually a result of engraftment kinetics and not gene silencing (Supplemental Fig. 5). Similar engraftment profiles with a reduced magnitude were also observed for lymphocytes (Supplemental Fig. 1), T and B lymphocytes, monocytes, platelets, and bone marrow cells (data not shown). Granulocyte engraftment profiles of IM- vs IV-injected cells in baboons Minor variations in day-to-day gene-marking levels complicated the analysis of early and late engraftment patterns, so we attempted to differentiate between trends in gene marking of IM- and IV-injected cells at times early (before day 100) and late (after day 100) after transplantation. We took the average granulocyte gene marking level (as determined by FACS analysis) early and late for IM- and IV-injected cells and the ratio of these two values. As seen in Table 2, in three of four animals the IM-injected cell gene marking level was higher after day 100 than before day 100, while in all of the animals the gene marking level of IV-injected cells was lower after day 100 (p 5 0.07 calculated using a paired t-test). Also, we used the granulocyte marking (as determined by flow-cytometric analysis) and fitted

a line representing a cubic spline smoothing of the data, and again this shows in two of four animals a modest trend of increased gene marking over time with IM-injected cells, while IV-injected cells decreased (Fig. 3). Retrovirus integration site analysis With gene-marking levels exceeding 25% in each arm of several animals, we evaluated the overall clonal contribution using LAM-PCR to determine if a dominant clone or clones could be detected. In animal M00228 peripheral blood DNA was isolated at 251 days after transplantation and LAM-PCR samples were run in triplicate. As seen in Figure 4, the high marking seen in this animal appears to be from multiple clones with no trend toward oligoclonal or monoclonal outgrowth. Distribution of gene-marked IM- and IV-injected cells Persistent engraftment of IM- and IV-injected long-term repopulating cells was observed in all four animals (Fig. 2). GFP marking (IM) of the bone marrow cells compared to YFP marking (IV) is shown in Table 3 (data is reported from flow-cytometric analysis) for the injected bone marrow (I-BM) and the contralateral noninjected bone marrow (NI-BM). Baboons that showed superior GFP vs YFP gene marking (or vice versa) in peripheral blood cells maintained this difference in I-BM and NI-BM granulocytes (Table 3). Also, to confirm this data we stained for CD34þ cells from the I-BM and NI-BM (see Supplemental Fig. 2) at the same time points described in Table 3 to eliminate the possibility of contamination from gene marked peripheral blood cells. The gene marking analysis of CD34þ cells from I-BM and NI-BM recapitulated the data in Figure 2 and Table 3 in that a baboon with higher overall I-BM GFP marking (M00228) vs a baboon with lower overall I-BM GFP marking (M00165) showed similar trends of gene marked CD34þ cells from I-BM and NI-BM. As expected, the overall ratio of GFP/YFP was generally higher in the injected marrow compared to the noninjected marrow (Table 3).

1136

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

Figure 2. Engraftment of gene-marked CD34þ cells after autologous transplantation. Granulocyte marking (determined by flow-cytometric analysis) of intramarrow (IM)-injected green fluorescent protein–positive cells (C) and intravenous (IV)-injected yellow fluorescent protein–positive cells (B) of the four baboons plotted as days after transplant. (A) First 100 days, (B) complete follow-up.

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

1137

Table 2. Average granulocyte marking before and after day 100 Before day 100

After day 100

%After/%before 100

Animal

IM

IV

IM

IV

IM

IV

M00228 M00165 J01174 M01277

23.2 32.0 37.5 22.6

26.8 50.6 41.6 35.0

34.3 33.5 43.9 19.5

12.7 44.6 25.4 19.9

148 105 117 86

47 88 61 57

IM 5 intramarrow; IV 5 intravenous. p 5 0.07.

Figure 3. Trend of improved gene marking with intramarrow injection of gene-modified cells. Granulocyte marking (determined by flow-cytometric analysis) of intravenous (dotted line) and intramarrow (black line) after transplantation.

Engraftment and gene marking of baboon CD34þ cells in NOD/SCID and NOD/SCID-b2m/ mice To investigate the behavior of xenotransplanted baboon CD34þ cells, we administered the same number (1  106 cells, each) of GFP- and YFP-transduced CD34þ enriched baboon bone marrow cells by direct IM and IV injection, respectively, to immunodeficient mice. We evaluated engraftment in both NOD/SCID (NS) and NOD/SCIDb2m/ (B2m) mice. There was a tendency of higher overall engraftment efficiency in B2m 14.2% vs NS 2.6%, (p 5 0.22) (see Supplemental Fig. 3) and higher engraftment of lymphoid/monocytoid compartments that might contain HSCs (44.8% (B2m) vs 16.8% (NS), p 5 0.048) suggesting that xenotransplantation of baboon cells may be more successful in B2m. In both mouse models w7 weeks after

Figure 4. Multiple gene-marked clones contribute to hematopoiesis. Representative gel of linear amplification–mediated polymerase chain reaction (LAM-PCR) samples amplified in triplicate from the same day after transplantation. Samples are run beside a 50-bp ladder and Tas I refers to the restriction enzyme used during LAM-PCR and the internal product from the retrovirus vector is noted with a black arrow.

transplantation GFP marking (IM) was higher than YFP marking (IV) in the I-BM (NS; p 5 0.08 and B2m p 5 0.07), YFP marking (IV) was similar to GFP marking (IM) in NI-BM in NS, while in the B2m mice YFP marking (IV) was higher than GFP marking (IM) in the NI-BM (Fig. 5).

Discussion Here we directly compared the engraftment profiles of gene-marked CD34þ cells after IM vs IV injection in baboons that received myeloablative radiation. Baboon CD34þ cells were transduced with retroviral vectors carrying GFP or YFP for IM and IV injection, respectively. We found that engraftment of GFPþ cells (IM-injected cells) was initially lower than that of the YFPþ cells (IV-injected cells) in all animals. However, marking of IM-injected cells increased after transplantation in three of the animals, resulting in higher marking levels after day 100 compared to marking before day 100. In addition, this increase in marking eventually led to higher gene-marking levels of IM-injected cells than IV-injected cells in two of four animals. These data show efficient engraftment of genemarked CD34þ cells after IM injection and suggest differential engraftment kinetics for IM- vs IV-injected CD34þ cells. The higher marking levels observed in peripheral blood granulocytes with IM-injected cells were also seen in lymphocytes, monocytes, and platelets suggesting that the preferential engraftment of IM-injected cells occurred at the level of multipotent hematopoietic stem cells. To rule out gene silencing for the differences in gene marking over time, we performed real-time PCR analysis on peripheral blood cells in all baboons. In all animals real-time PCR correlated well with the flow cytometric results confirming that the changes in gene-marking were due to differential engraftment pattern of IM vs IV cells rather than due to gene silencing. The finding that engraftment levels between IM- and IVinjected CD34þ cells changed substantially over the first few months may reflect the fluctuation of the stem cell pool early after transplantation. Clonal fluctuation has been known to precede stable hematopoiesis in allogeneic

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

1138

Table 3. Marking levels of baboon marrow cells I-BM Baboon M00228

M00165

J01174 M01277

GFP/YFP ratio

NI-BM

GFP/YFP ratio

Time after transplant (mo)

GFPþ

YFPþ

I-BM

GFPþ

YFPþ

NI-BM

1 3 9 1 3 9 1 5 1 3

16.4 19.8 8.6 13.9 18.2 20.0 36.4 26.9 34.0 6.9

16.9 4.7 4.5 13.9 28.3 27.2 13.8 13.8 11.5 9.4

1.0 4.2 1.9 1.0 0.6 0.7 2.6 1.9 3.0 0.7

14.7 8.9 ND 8.2 19.8 21.2 24.0 23.9 14.6 10.8

18.2 4.7 ND 16.0 33.8 27.6 25.3 16.4 32.8 16.9

0.8 1.9 ND 0.5 0.6 0.8 1.0 1.5 0.4 0.6

GFP 5 green fluorescent protein; I-BM 5 injected bone marrow; IM 5 intramarrow; IV 5 intravenous; MOI 5 multiplicity of infection; ND 5 not determined; NI-BM 5 noninjected bone marrow; YFP 5 yellow fluorescent protein.

transplantation in mice and cats [15,16]. We performed LAM-PCR in peripheral blood cells after transplantation to analyze the clonal composition of IV- and IM-injected grafts. We found that large numbers of retrovirally marked stem cells were transplanted into the recipients and contribute to long-term hematopoiesis and the increase in gene marking in three animals does not appear to be the outgrowth of dominant clones. The preferential engraftment of long-term repopulating cells with IM-injected cells was also recapitulated in SCID repopulating cell assay. We first tested engraftment of baboon cells in two different immunodeficient mouse models. The average repopulation efficiency of baboon CD34þ cells was 5.5-fold higher when transplanted into B2m than NS. The proportion of CD45þ lymphoid/monocytoid cells that might contain HSCs was also higher in B2m than NOD/SCID mice (data not shown). That finding

was consistent with another report [17] that demonstrated short-term repopulating cells appeared to engraft B2m mice more efficiently than NS mice. IM-injected cells repopulate preferentially in the injected bones both in B2m and NS. IV-injected cells engrafted at similar levels as IMinjected cells in NI-BM NS mice and more efficiently in the NI-BM in B2m. These data suggest that the direct IM injection may be more advantageous for homing and lodging of HSCs. Direct IM injection into the femur might have delivered more HSCs into the physiologic ‘‘niche’’ than IV injection. [18]. In conclusion, we present evidence that in a competitive repopulation assay direct intrafemoral injection of genemodified long-term repopulating cells compared to standard IV injection leads to efficient engraftment and gene marking of both populations. In two of four animals studied, the gene marking levels of IM-injected cells increased to

Figure 5. Green fluorescent protein (GFP) and yellow fluorescent protein (YFP) marking in nonobese diabetic/severe combined immune-deficient (NOD/ SCID)-b2m/ (B2m) and NOD/SCID (NS). Bars represent the mean percentage of GFP (-) or YFP (,) marking levels of CD45þ baboon cells (M00228) in injected BM (I-BM) and noninjected BM (NI-BM) in the two types of immunodeficient mice and error bars represent standard errors. *p 5 0.07 (I-BM vs NI-BM of B2m), **p 5 0.08 (I-BM vs NI-BM of NS), ***p 5 0.04 (GFP vs YFP in I-BM of NS).

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

higher levels than the IV-injected cells. These increases have remained stable in all animals more than 1 year after transplantation. All blood cell counts are within the normal range, and in the animals with increasing gene marking levels there appears to be no progression toward hematopoietic abnormalities.

Acknowledgments We would like to thank Helen Crawford and Bonnie Larson for assistance with the preparation of the article. We would also like to thank Mike Gough, James Fletcher, and the staff of the University of Washington National Primate Research Center for assistance with the animals. Finally, we would like to thank Amgen Corporation for the provision of G-CSF and SCF. This work was supported by National Institutes of Health grants HL54881, HL53750, DK47754, DK56465, and RR00166. Hans-Peter Kiem is a Markey Molecular Medicine investigator.

References 1. Henschler R, Fehervizyova Z, Bistrian R, Seifried E. A mouse model to study organ homing behaviour of haemopoietic progenitor cells reveals high selectivity but low efficiency of multipotent progenitors to home into haemopoietic organs. Br J Haematol. 2004;126:111–119. 2. Cui J, Wahl RL, Shen T, et al. Bone marrow cell trafficking following intravenous administration. Br J Haematol. 1999;107:895–902. 3. Imai K, Kobayashi M, Wang J, et al. Selective transendothelial migration of hematopoietic progenitor cells: a role in homing of progenitor cells. Blood. 1999;93:149–156. 4. Yahata T, Ando K, Sato T, et al. A highly sensitive strategy for SCIDrepopulating cell assay by direct injection of primitive human hematopoietic cells into NOD/SCID mice bone marrow. Blood. 2003;101: 2905–2913. 5. Mazurier F, Doedens M, Gan OI, Dick JE. Rapid myeloerythroid repopulation after intrafemoral transplantation of NOD-SCID mice reveals a new class of human stem cells. Nat Med. 2003;9:959–963. 6. Horn PA, Topp MS, Morris JC, Riddell SR, Kiem H-P. Highly efficient gene transfer into baboon marrow repopulating cells using GALVpseudotype oncoretroviral vectors produced by human packaging cells. Blood. 2002;100:3960–3967.

1139

7. Kiem H-P, Heyward S, Winkler A, et al. Gene transfer into marrow repopulating cells: comparison between amphotropic and gibbon ape leukemia virus pseudotyped retroviral vectors in a competitive repopulation assay in baboons. Blood. 1997;90:4638–4645. 8. Kiem H-P, Andrews RG, Morris J, et al. Improved gene transfer into baboon marrow repopulating cells using recombinant human fibronectin fragment CH-296 in combination with interleukin-6, stem cell factor, FLT-3 ligand, and megakaryocyte growth and development factor. Blood. 1998;92:1878–1886. 9. Hanenberg H, Xiao XL, Dilloo D, Hashino K, Kato I, Williams DA. Colocalization of retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nat Med. 1996;2:876–882. 10. Larochelle A, Vormoor J, Hanenberg H, et al. Identification of primitive human hematopoietic cells capable of repopulating NOD/SCID mouse bone marrow: implications for gene therapy. Nat Med. 1996; 2:1329–1337. 11. Neff T, Horn PA, Peterson LJ, et al. Methylguanine methyltransferasemediated in vivo selection and chemoprotection of allogeneic stem cells in a large-animal model. J Clin Invest. 2003;112:1581–1588. 12. Schmidt M, Zickler P, Hoffmann G, et al. Polyclonal long-term repopulating stem cell clones in a primate model. Blood. 2002;100:2737– 2743. 13. Wu X, Li Y, Crise B, Burgess SM. Transcription start regions in the human genome are favored targets for MLV integration. Science. 2003;300:1749–1751. 14. Schmidt M, Carbonaro DA, Speckmann C, et al. Clonality analysis after retroviral-mediated gene transfer to CD34(þ) cells from the cord blood of ADA-deficient SCID neonates. Nat Med. 2003;9: 463–468. 15. Harrison DE, Astle CM, Lerner C. Number and continuous proliferative pattern of transplanted primitive immunohematopoietic stem cells. Proc Natl Acad Sci U S A. 1988;85:822–826. 16. Abkowitz JL, Catlin SN, Guttorp P. Evidence that hematopoiesis may be a stochastic process in vivo. Nat Med. 1996;2:190–197. 17. Glimm H, Eisterer W, Lee K, et al. Previously undetected human hematopoietic cell populations with short-term repopulating activity selectively engraft NOD/SCID-beta2 microglobulin-null mice. J Clin Invest. 2001;107:199–206. 18. Nilsson SK, Haylock DN, Johnston HM, Occhiodoro T, Brown TJ, Simmons PJ. Hyaluronan is synthesized by primitive hemopoietic cells, participates in their lodgment at the endosteum following transplantation, and is involved in the regulation of their proliferation and differentiation in vitro. Blood. 2003;101:856–862.

1139.e1

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

Supplemental materials and methods Baboon bone marrow CD34 staining and flow cytometry analysis Standard flow cytometry of leukocytes prepared from injected and noninjected bone marrow cells except that cells were stained with CD34 monoclonal antibody (mAb) 12.8 (Fred Hutchinson Cancer Research Center [FHCRC]) with streptavidin-allophycocyanin [APC] (Becton Dickinson). Isotype control mAbs were used to control for background staining and gating (Supplemental Fig. 2). Flow cytometric subset analysis of baboon hematopoietic cells in murine recipients Subset analysis of the engrafted baboon cells was performed by gating on CD45þ (CD45-phycoerythrin) cells

and by using APC-conjugated lineage-specific antibodies. Antibodies used were CD34 mAb 12.8 (FHCRC) or clone 563 (Becton Dickinson) with streptavidin-APC; (Becton Dickinson), CD20 mAb (clone B9E9; Immunotech, West¨ K4, Immunotech), brook, ME, USA), CD14 mAb (clone TU ¨ K1; Caltag, Burlingame, CA, or CD13 mAb (clone TU USA). 40 ,6-diamidino-2-phenylindole dihydrochloride hydrate (Sigma) was used as a viability marker, and four-color flow cytometric analysis was performed on an LSR II (Becton Dickinson). For each mouse analyzed, an aliquot of cells labeled with conjugated isotype control mAb was used as a control. In addition, bone marrow cells from NOD/SCID-b2m/ mice that did not undergo transplantation were stained with the same mAbs (Supplemental Fig. 4).

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

Supplemental Figure 1. Engraftment of gene-marked lymphocytes after autologous transplantation. Lymphocyte marking (determined by flow-cytometric analysis) of intramarrow-injected green fluorescent protein–positive cells (C) and intravenous-injected yellow-fluorescent protein–positive cells (B) of the four baboons were plotted as days after transplantation.

1139.e2

1139.e3

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

Supplemental Figure 2. Engraftment of gene-marked CD34þ bone marrow cells after autologous transplantation. Green fluorescent protein (GFP)þ or yellow-fluorescent protein (YFP)þ and CD34þ bone marrow (BM) cells from the injected (I-BM) and noninjected (NI-BM) femurs of two baboons plotted as days after transplantation. Black bars (GFPþ/CD34þ BM cells) and white bars (YFPþ/CD34þ BM cells).

Supplemental Figure 3. Engraftment of M00228 baboon cells in nonobese diabetic/severe combined immune-deficient (NOD/SCID)-b2m/ (B2m) and NOD/SCID (NS). Each point represents the percentage of CD45þ baboon cells in each mouse and bar represents mean value of engraftment in the two types of immunodeficient mice.

Supplemental Figure 4. Preferential engraftment of intramarrow (IM)-injected M00165 cells in the injected bone marrow (BM) compared to noninjected BM in B2m. Mean marking level in various SCID repopulating cell subsets were depicted with standard errors. I-BM 5 injected bone marrow, NI-BM 5 noninjected bone marrow; *p ! 0.05; **p 5 0.066.

C.W. Jung et al./ Experimental Hematology 35 (2007) 1132–1139

Supplemental Figure 5. Engraftment of gene-marked CD34þ cells after autologous transplantation. Gene marking determined by quantitativepolymerase chain reaction of nucleated white blood cells with gammaretrovirus–specific primers and probes (dashed line, closed circles [C]), green fluorescent protein–specific primers and probes (solid line, closed squares [-]), and yellow fluorescent protein–specific primers and probes (solid line, open squares [,]) plotted as days after transplantation.

1139.e4