- Email: [email protected]
Low level of gene transfer to and engraftment of murine bone marrow cells from long-term bone marrow cultures
Experimental Hematology 28 (2000) 373–381
Low level of gene transfer to and engraftment of murine bone marrow cells from long-term bone marrow cultures Thomas Relander, Cecilia Fahlman, Stefan Karlsson, and Johan Richter Department of Molecular Medicine and Gene Therapy, Wallenberg Neuroscience Center, University of Lund, Lund, Sweden (Received 22 September 1999; revised 16 December 1999; accepted 28 December 1999)
Objective. We wanted to determine whether the long-term bone marrow culture (LTBMC) transduction system would lead to efficient gene transfer and engraftment of murine repopulating hematopoietic stem cells (HSC), particularly in nonablated recipients. Materials and Methods. Congenic mouse strains expressing Ly 5.1 or Ly 5.2 and the GP⫹E86 cell line producing the MGirL22Y vector carrying the gene for enhanced GFP were used. Murine LTBMCs were established and demi-depopulated on days 7 and 14 with addition of vector supernatant on days 8 and 15. Results. Cell recovery on day 21 was 21.3% ⫾ 3.8% of input cells and CFU-C recovery was 9.7 ⫾ 3.4% as compared with CFU-C of input cells. In vitro transduction efficiency determined by CFU-C expressing GFP was 22.2% ⫾ 1.6%. In irradiated (950 cGy) mice transplanted with 2 ⫻ 106 LTBMC cells, 94% of nucleated cells in the blood at week 16 were of donor origin. However, GFP was only detected at low level in a few animals at week 4 and not later. Analysis of bone marrow from these mice at week 20 did not show any GFP expression and semiquantitative PCR revealed a transgene level of ⬍ 1%. When 3.5–20.8 ⫻ 106 LTBMC cells (corresponding to 20–100 ⫻ 106 fresh cells) were transplanted to nonablated recipients, no engraftment or GFP expression were detected. Competitive repopulation experiments showed that the longterm repopulation ability (LTRA) of the LTMC cells was only 7% of fresh cells. Conclusion. These results indicate that LTBMC transduction of murine cells leads to lowlevel transduction of progenitors, no gene transfer to repopulating stem cells, and reduction in LTRA in ablated and nonablated recipients. © 2000 International Society for Experimental Hematology. Published by Elsevier Science Inc. Keywords: Gene transfer—Long-term culture—Engraftment—Bone marrow
Introduction Retroviral-mediated gene transfer into hematopoietic stem cells (HSC) has the potential to provide permanent correction of genetic diseases affecting the hematopoietic system [1]. In murine models it has been possible to show efficient gene transfer and expression into long-term repopulating hematopoietic cells. In contrast, results from human clinical trials and studies using large nonhuman primates have, in general, been disappointing, with only a very low level of gene transfer into long-lived hematopoietic cells [2–4]. The reasons for this discrepancy are only partially known. One factor is thought to be that the level of the relevant retroviral receptor is very low on human HSC [5]. AnOffprint requests to: Thomas Relander, M.D., Department of Molecular Medicine and Gene Therapy, Wallenberg Neuroscience Center, Sölvegatan 17, 223 62 Lund, Sweden; E-mail: [email protected]
other factor of importance is the fact that most human HSC are quiescent and nondividing, and that cell division is required for retroviral vectors to integrate into the genome of the infected cells [6,7]. This requirement for cell division has been the major reason to include stimulatory cytokines into the transduction protocols used so far. Thus most preclinical and clinical gene transfer protocols have included a period of prestimulation with cytokines followed by exposure to the retroviral vector 2–4 times during a period of another 2–3 days, most often in the presence of the same cytokine combination. Despite these efforts, only low levels of gene transfer to human HSC have been achieved. The reasons for this failure can be many, one being that even if a majority of the cells under such conditions are triggered into cell cycle, it is not unlikely that the small minority representing the true HSC may still be quiescent after only a few days of cytokine stimulation. Experimental evidence to sug-
0301-472X/00 $–see front matter. Copyright © 2000 International Society for Experimental Hematology. Published by Elsevier Science Inc. PII S0301-472X(00)0 0 1 3 1 - 4
374
T. Relander et al./Experimental Hematology 28 (2000) 373–381
gest that at least some human HSC may require as much as 9 days of ex vivo culture to undergo cell division has been presented [8]. Another possibility is that ex vivo exposure to at least some cytokines leads to altered function of human HSC, such as diminished self-renewal capability or decreased ability to engraft when reinfused into the host. Data from experiments using NOD/SCID mice support this concept as there was a decline in the number of SCID repopulating cells (SRC) when cells were cultured ex vivo with cytokines [9,10]. However, it should be noted that other studies have shown that SRC can be expanded ex vivo under certain conditions [11–13]. Some investigators have hypothesized that transduction and preservation of HSC would be more efficient in a microenvironment simulating the normal conditions in the bone marrow. Indeed it has been shown that exposure of hematopoietic cells to retroviral vectors in the presence of marrow-derived stromal cells leads to an increase in transduction efficiency [14–16]. This concept was taken even further by Dubé and coworkers [17], who adopted the Dexter-type long-term bone marrow culture (LTBMC) system for gene transfer purposes. In a canine model system bone marrow was set up in LTBMC and transduced by weekly exposures to a retroviral vector and after 3 weeks reinfused into nonablated autologous hosts [17]. Reporter genemarked progenitors were detected at approximately the 5% level for up to 2 years. This level of gene marking was surprisingly high, especially considering that it was obtained without any myeloablative regimen given prior to reinfusion of the manipulated cells. Based on these encouraging data, a human marking trial in patients with multiple myeloma was initiated and the results have recently been published [18,19]. In most genetic diseases affecting the hematopoietic system the use of myeloablative conditioning of the patient prior to reinfusion of transduced HSC is not justified. Clinical trials involving patients with Gaucher disease and chronic granulomatous disease have shown that to obtain significant engraftment of ex vivo manipulated HSC in the nonablated recipient poses an additional challenge to the already mentioned problem of low transduction [20,21]. In the mouse, it has been possible to get long-term engraftment of transduced cells in a nonablated recipient, albeit at a low level [22]. It was in this context we wanted to explore whether the LTBMC transduction system would lead to efficient gene transfer and engraftment of murine repopulating HSC. We were particularly interested to ask whether gene marked repopulating cells could be detected in nonablated mouse recipients as has been reported for the canine model [17]. However, the results demonstrate that the LTBMC transduction leads to poor engraftment of gene marked repopulating cells in murine recipients. Furthermore, gene transfer efficiency into repopulating HCSs is very low by this method, as judged by transplantation into ablated recipients.
Materials and methods Mouse strains The two congenic mouse strains C57BL/6 (Thy-1.2, Ly5.2) and C57BL/6-Ly5.1-Pep3b (Thy-1.2, Ly5.1) originally from Jackson Laboratories (Bar Harbor, ME) and hereafter called Ly5.2 and Ly5.1, respectively, were bred at the animal facility of Department of Tumor Immunology, University of Lund, Sweden. All animals were housed in ventilated racks and given acidified water and food ad libitum. Marrow harvest and establishment of LTBMCs Fresh bone marrow cells were obtained from 8- to 12-week-old donor animals. Mice were not given any treatment prior to bone marrow harvest. Femurs and tibias were removed and bone marrow cells were flushed aseptically using syringes and needles. Cells were dispersed into a single-cell suspension by repeated flushing and then counted and washed before they were resuspended in Myelocult (HCC 5300, Stem Cell Technologies, Vancouver, Canada) with Hydrocortisone 10⫺7 M (H4001; Sigma Chemical, St. Louis, MO), hereafter referred to as complete Myelocult. For establishment of LTBMCs, 20 ⫻ 106 cells in 10 mL of complete Myelocult were seeded into vented 25 cm2 flasks (Falcon 009-3108; Becton-Dickinson Labware, Bedford, MA) or 60 ⫻ 106 cells in 30 mL of complete Myelocult were seeded into 75 cm2 flasks (Falcon 009-3110). Flasks were maintained at 37⬚C in a humidified 5% CO2 in air environment. Retroviral vector and producer cell line The ecotropic packaging cell line GP⫹E86 producing the MGirL22Y retroviral vector was kindly supplied by Derek Persons [23]. This bicistronic vector contains the gene for enhanced GFP, an internal ribosomal entry site from the encephalomyocarditis virus and a mutated variant of the human dihydrofolate reductase gene. The titer of the viral supernatant when tested in our lab was 2–3 ⫻ 106 infectious particles per milliliter when assessed by FACS for transfer of the green fluorescent protein to NIH3T3 cells. This is in accordance with the titer previously reported. The producer cell line was maintained in DMEM medium with 10% FCS and penicillin/streptomycin at 37⬚C, except 48 hours prior to harvest of virus for the LTBMCs when it was moved to 33⬚C. Twenty-four hours prior to harvest, medium was changed to fresh complete Myelocult. Supernatants were filtered (0.45 m MillexHV, Low protein binding; Millipore, Bedford, MA) and then used immediately without freezing. Transduction of LTBMC cells Once LTBMCs were established, they were maintained in the standard way by weekly (day 7 and 14) removal of one half of the media, including cells therein, followed by addition of fresh complete Myelocult. On days 8 and 15, demi-depopulations of the nonadherent cells were also performed but this time the cultures were replenished with Myelocult conditioned on the vector producing cell line (5 mL for 25 cm2 and 15 mL for 75 cm2 flasks). In accordance with the protocol used by Dubé and coworkers [17], protamine sulphate or polybrene was not added. On day 21, LTBMC adherent layers were enzymatically removed by incubation with 0.25% trypsin-1 mM EDTA in HBSS without Ca and Mg (GIBCO BRL, Paisley, Scotland). Care was taken to ensure that the cells were dispersed into a single cell suspension. Recovered cells were washed once and resuspended in PBS for injection into mice or PBS, with
T. Relander et al./Experimental Hematology 28 (2000) 373–381
1% FCS for FACS analysis, or resuspended in medium before plating in semisolid methylcellulose. Four-day standard transduction Fresh bone marrow cells were prestimulated with 50 U/mL murine IL-6 (generously provided by Genetics Institute, Cambridge, MA) and 100 ng/mL rat stem cell factor (kind gift from Amgen, Thousand Oaks, CA) in DMEM supplemented with 15% FCS for 48 hours and then cocultured with the producer cell line in the above culture media supplemented with 4 g/mL protamine sulphate (Sigma-Aldrich Sweden, Lund, Sweden). Forty-eight hours later, nonadherent bone marrow cells were gently rinsed off the producer cells, washed, and resuspended in PBS for injection into recipient mice or resuspended in medium for in vitro clonogenic progenitor assays. A small aliquot of cells was cultured another 24 hours in the same medium as above and analyzed by FACS for GFP expression. Analysis of transduced cells CFU-C assays were performed using methylcellulose optimized to support clonogenic proliferation and differentiation of granulocyte-macrophage colony-forming-units (Methocult HCC3534; Stem Cell Technologies). Cultures were incubated at 37⬚C in 5% CO2/95% air in a humidified chamber for 6–7 days. The number of colonies generated from LTBMCs was compared with the number of colonies generated from the same starting cell population on the day of bone marrow harvest in order to determine the recovery of CFU after LTBMC. Plates containing hematopoietic colonies growing in methylcellulose were also directly evaluated for GFP expression by visualization using a fluorescence microscope. Transplantation of transduced cells and fresh cells Recipient mice, 8–10 weeks weeks old, were irradiated at 950 cGy total dose from a dual 137Cs source (Gammarad 900; Scanditronix Inc., Uppsala, Sweden), followed by tail-vein injection of either transduced cells, fresh bone marrow cells or, for competitive repopulation experiments, a mixture of both.
375
ing cells divided by the sum of Ly5.2 and Ly5.1 staining cells when Ly5.2 mice were the donors. GFP expression is shown as the number of cells expressing GFP and Ly5.1 or Ly5.2 divided by the sum of Ly5.2 and Ly5.1 staining cells. Semiquantitative PCR DNA was isolated from harvested bone marrow cells using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN). DNA (0.25 g) from each sample was amplified in separate tubes using primers either for GFP or mouse -actin. The GFP primers were 5⬘-GAG CTG GAC GGC GAC GTA AAC G-3⬘, and 5⬘CGC TTC TCG TTG GGG TCT TTG CT-3⬘, yielding a 595 bp fragment. Mouse -actin primers were 5⬘-CAT TGT GAT GGA CTC CGG TGA CGG-3⬘ and 5⬘-CAT AGC ACA GCT TCT CTT TGA TG-3⬘, yielding a 206 bp fragment. A Peltier Thermal Cycler (PTC-200; MJ Research Inc., Watertown, MA, USA) was used, and the conditions for the GFP PCR were: 95⬚C for 5 minutes and then 32 cycles of 95⬚C for 1 minute; 60⬚C for 1 minute; and 72⬚C for 1 minute. For the actin PCR: 95⬚C for 5 minutes and then 32 cycles of 95⬚C for 1 minute; 55⬚C for 1 minute; and 72⬚C for 2 minutes. For both reactions the above cycles were followed by an 8-minute extension at 72⬚C. The resulting PCR products were run on an ethidiumbromide containing agarose gel (1%), directly visualized under UV light and also transferred to nylon membranes using established Southern blotting techniques. A standard curve was established by extracting DNA from the MGirL22Y producing GP⫹E86s serially diluted into nontransfected GP⫹E86. The MGirL22Y producing GP⫹E86 cell line has previously been reported to contain 15 copies of the vector [23]. Negative controls in every reaction set included no DNA and DNA extracted from nontransfected GP⫹E86 cells. Histopathology Tissues were preserved in 4% paraformaldehyde (PFA) overnight, embedded in paraffin, sectioned and stained with hematoxylin and eosin before examination by light microscopy.
Results Analysis of engraftment Peripheral blood from experimental animals was analyzed at 4-, 10-, and 16-weeks after transplantation. In some experiments bone marrow was analyzed at 20 weeks post-transplant when the mice were sacrificed. Peripheral blood (approximately 250 L) was obtained from the retroorbital sinus. Whole blood was incubated with mouse whole IgG to block unspecific binding prior to antibody staining. After two washes with PBS with 1% FCS, cells were incubated on ice for 20–30 minutes with anti-CD45.1-PE and antiCD45.2-biotin (Pharmingen) and then washed twice before incubation with streptavidin-APC for 20 minutes, after which RBC lysis (BD-lysis; Becton-Dickinson, San Jose, CA) was performed followed by two more washes. Flow cytometry was performed using a FACS Calibur (Becton Dickinson) using excitation at 488 nm and fluorescence detection at 530 nm (for GFP), 585 nm (for phycoerytrin), and excitation 633 nm and detection 650 nm (for APC). Ten thousand events were collected for each sample. For each mouse, engraftment was measured as the number of Ly5.1 staining cells divided by the sum of Ly5.2 and Ly5.1 staining cells when Ly5.1 mice were used as donors and as the number of Ly5.2 stain-
Standard 4-day transduction As a control to test the viral producer line and the transplantation procedure, a standard 4-day transduction was performed. Bone marrow cells harvested from Ly5.1 mice were prestimulated with SCF and IL-6 for 48 hours and then cocultivated with the MGirL22Y producing cell line for another 48 hours in the presence of the same cytokines and protamine sulphate 4 g/mL. Lethally irradiated Ly5.2 mice were then transplanted with 0.5 ⫻ 106 transduced cells each. FACS analysis 24 hours post-transduction showed that 45.2% of the cells expressed the GFP protein (Fig. 1). Cells were also plated into methylcellulose and the frequency of GFP positive hematopoietic colonies was 89% as determined by fluorescence microscopy (Fig. 1). Peripheral blood from the transplanted mice was analyzed after 4, 10, and 16 weeks to determine level of engraftment and expression of the marker gene. The contribution of donor cells to the hematopoiesis in the mice increased from 70.4% (SD
376
T. Relander et al./Experimental Hematology 28 (2000) 373–381
1.5%) at 4 weeks post-transplant to 89.3% (SD 11.3%) at 16 weeks. The level of GFP expressing cells did not change significantly from week 4 to 16 and was approximately 50%. The mice were sacrificed 22 weeks after transplantation, and analysis of bone marrow at this time point, with regard to level of donor cells and GFP expression, did not differ significantly from the levels in the peripheral blood seen at week 16 (Fig. 1). These data thus confirm previous findings that the vector and producer cell line used in this work are able to transduce long-term repopulating murine bone marrow cells to a high degree and that the gene is expressed at the same level for at least 5 months [23]. In vitro analysis of LTBMC transduced cells In four separate experiments bone marrow from non-pretreated mice were set up in LTBMC in complete Myelocult. The protocol used closely followed the modified LTBMC transduction protocol developed by Dubé and coworkers [17,18], which was subsequently used in their clinical trial. LTBM cultures were demi-depopulated and replenished with new complete Myelocult on days 7 and 14. On days 8 and 15, second rounds of demi-depopulation were performed but this time the cultures were replenished with complete Myelocult freshly conditioned on the MGirL22Y producer cell line. On day 21, all adherent cells were harvested after trypsin treatment of the flasks. Cells from each flask were counted, analyzed by FACS for GFP expression, and a small aliquot was also plated in methylcellulose. The number of cells harvested was compared with input number
of cells, and the cell recovery was determined (Table 1). Overall recovery in all four experiments was 21.6%, with little variation from one experiment to another. The yield of colonies from each LTBMC was compared with the number of colonies obtained from methylcellulose cultures set up from the same cell population on the day when the LTBMC was initiated. This figure also takes into consideration the reduction in total cell number. Overall recovery of CFU in experiments 2–4 was 9.7% (Table 1). FACS analysis was done on the adherent LTBMC cells from each flask and the frequency of GFP expressing cells was determined by gating on the hematopoietic cells; 19.8% of cells (SD 6.0%) were GFP positive. The frequency of GFP–positive hematopoietic colonies derived from LTBMCs was 22.2% (SD 1.6%) as determined by fluorescence microscopy (Table 1). Transplantation of LTBMC transduced cells into myeloablated recipients In two separate experiments, lethally irradiated mice were transplanted with 1–2 ⫻ 106 transduced LTBMC cells each and then monitored by analysis of peripheral blood at 4, 10, and 16 weeks post-transplant. The results in Figure 2 show that the LTBMC cells were able to reconstitute the myeloablated mice. At 16 weeks, 94% of the cells in peripheral blood were of donor origin. However, GFP-expressing cells over the detection limit could be detected only at 4 weeks and at a very low level. At 10 and 16 weeks, no GFP expression above baseline was detected. Furthermore, when the mice were sacrificed at 20 weeks, GFP-expressing cells
Figure 1. Transduction and engraftment of bone marrow cells after a 4-day standard transduction. Day 5 ⫽ percentage of GFP positive mononuclear cells analyzed by FACS 5 days after start of transduction; CFU ⫽ percentage of GFP positive colonies in methylcellulose analyzed by fluorescence microscopy. Weeks 4, 10, and 16 ⫽ peripheral blood was analyzed by FACS for expression of donor and recipient phenotype and GFP at the time points indicated. Week 22 ⫽ analysis of bone marrow 22 weeks post-transplant. Error bars denote SD.
T. Relander et al./Experimental Hematology 28 (2000) 373–381 Table 1. In vitro data from murine LTBMC transduction Exp. 1
Exp. 2
Exp. 3
Exp. 4
Mean
23% (cells 25.8% pooled) SD 12.8 n⫽3 Recovery of CFU-C, 8.2% mean SD 2.5 % GFP positive 19.4% adherent cells SD 3.2 day 21 n⫽3 % GFP positive CFU 21.3% SD 2.8
19.0% SD 8.3 n⫽6 7.3% SD 1.5 26.0% SD 8.6 n⫽6 24.1% SD 10.5
18.5% SD 6.8 n ⫽ 12 13.5% SD 2.4 14.1% SD 7.7 n⫽9 21.3% SD 10.5
21.6% SD 3.5 (4 exp.) 9.7% SD 3.4 19.8% SD 6.0
Recovery of cells, mean
22.2% SD 1.6
In vitro data from four experiments: separate and pooled. Abbreviations: n ⫽ number of flasks analyzed; SD ⫽ standard deviation; exp. ⫽ experiment.
could not be detected by flow cytometry in the bone marrow either. Semiquantitative PCR analysis of DNA extracted from the bone marrow of these mice showed a transgene level of approximately 0.1–1% (Fig. 3). In contrast, PCR analysis of bone marrow from one of the mice who had received cells transduced by the standard transduction method revealed a transgene level of close to 100%.
Figure 2. Engraftment of donor cells and expression of GFP in myeloablated mice. Irradiated mice were transplanted with 1–2⫻106 LTBMC transduced cells. Peripheral blood was analyzed by FACS at 4, 10, and 16 weeks post-transplant for expression of Ly5.1, Ly5.2, and GFP. Error bars denote SD.
377
Transplantation of LTBMC transduced cells into nonablated recipients In order to investigate the ability of LTBMC transduced cells to engraft in nonablated recipients, Ly5.1 mice were transplanted with LTBMC cells. In three separate experiments mice were given LTBMC cells generated from 20 or 100 million marrow cells (for actual number of cells infused see Table 2). The higher dose was split and given on two consecutive days. For comparison, control mice were injected with 20 or 100 million fresh cells. The level of engraftment was determined by FACS analysis of the peripheral blood at weeks 4, 10, and 16. While mice given fresh cells showed a increased level of chimerism over time (Fig. 4), engraftment of LTBMC cells in nonablated recipients could not be detected. Competitive repopulation In order to estimate the long-term repopulating activity (LTRA) of the LTBMC cells, competitive repopulation experiments were performed. In two separate experiments a mixture of 0.5⫻106 LTBMC cells of Ly5.2 origin and an equal amount of fresh Ly5.1 bone marrow cells were transplanted into both Ly5.1 recipients and Ly5.2 recipients. Blood was obtained from all animals 16 weeks after transplantation and FACS analysis was performed (Fig. 5). The results show that in the Ly5.1 animals LTBMC cells were responsible for 22.6% of the chimerism seen in the blood, while for the Ly5.2 animals this figure was 26.9%. Because there was no significant difference between those two numbers, this shows that the recipient type did not influence the result of the competitive repopulation. Thus the average contribution of the LTBMC cells was 24.8% and the LTRA of the output LTBMC cells was calculated to be approximately 1/3 (24.8%/75.2%) of the LTRA of fresh cells when equal amounts of cells were compared. However, if one also considers that there was only a 21.6% recovery of cells from the LTBMC as compared with input, this means that the LTRA of the cells maintained in LTBMC, as determined in the peripheral blood of mice 16 weeks post-transplant, was approximately 7% of fresh cells. Also, GFP expression was undetectable in these mice. Transplant-related mortality after infusion of LTBMC cells In the group of nonmyeloablated mice injected with high numbers of LTBMC cells, there was a considerable transplant related mortality despite great care taken to disperse the harvested cells into a single cell suspension. In total, 17 mice received cell amounts corresponding to 20–100 million input cells from LTBMCs (Table 2). Out of these, five mice died in close association to the transplantation and more mice showed signs of respiratory distress. Autopsy was performed on one mouse showing abundant thromboembolic material deposited in blood vessels of the lungs, and in some of these deposits, bone marrow cells were seen. Some areas with embolic bone marrow cell clusters only
378
T. Relander et al./Experimental Hematology 28 (2000) 373–381
Figure 3. Semiquantitative PCR of bone marrow from myeloablated mice transplanted with cells transduced according to LTBMC protocol and standard 4-day protocol. Bone marrow was harvested from mice sacrificed 20–22 weeks after transplant. DNA was purified and PCR performed as described in the materials and methods section. (A)–(E): bone marrow from five mice transplanted with LTBMC cells, GP⫹E86 cells packaging cell line without vector, standard curve from dilutions of MGirL22Y vector producing GP⫹E86 cells in GP⫹E86 cells corresponding to 1 copy of vector/cell, 0.1 copy, 0.01 copy, 0.001 copy, and 0.0001 copy. (F) Bone marrow from mouse transplanted with cells from standard transduction.
were also found. There was no pathology detected in the heart. In the control group of 13 mice receiving fresh bone marrow cells (20–100 million cells), no animals died in connection to the cell transfer and the animals showed no signs of distress after injection. Of the 22 mice receiving lower doses of LTBMC cells (0.5–2 million), no mice died in conjunction to the transplant.
Discussion It is for many reasons a theoretically very attractive idea to use the Dexter-type LTBMC system for gene transfer to hematopoietic cells. These cultures can be maintained for many weeks during which proliferation of primitive cells can take place in a milieu imitating the bone marrow microenvironment [24]. No stimulatory cytokines that might prematurely force the HSC down the road of differentiation are added to these cultures. The more differentiated cells will be removed when the cultures are demi-depopulated,
while the primitive hematopoietic cells adhering to the stromal cells will remain in the culture and can be repeatedly exposed to vector supernatant. It has also been shown that the replacement of half of the medium with fresh medium stimulates the high proliferative potential hematopoietic progenitors in the adherent layer to go into cell cycle, possibly by removing TGF- [25,26]. Thus, this is a timepoint when primitive cells in the culture might be susceptible to retroviral gene transfer. Our experimental set up was modeled to closely match the 3-week LTBMC transduction described by Dubé and coworkers [17]. In their initial experiments they exposed the bone marrow cells to viral supernatant three times, on days 1, 8, and 15. However, in a later modification of the protocol, cells were exposed only twice to the vector, and in our experiments we settled for a two-cycle transduction procedure [27]. The overall recovery of cells from the LTBMCs on day 21 was 21.6% in our experiments, with little variation from experiment to experiment. This figure closely correlates to
Table 2. Transplantation of bone marrow cells to nonablated recipients Fresh cells 20 million Experiment 1 Experiment 3 Experiment 4
LTBMC
100 million
5 4
20 million input
100 million input
5* (4.6 million output cells) 4
5** (mean output cell number 3.5 million/mouse)
3** (14.3 million output cells) 3** (mean output cell number 20.8 million/mouse)
Transplantation of bone marrow cells to nonablated recipients. Number of mice and transplanted cells in different experiments. For LTBMC input cells refer to number of cells used to set up the LTBMC, while actual number of cells at harvest day 21 is given in brackets. In mice transplanted with 100 million input LTBMC cells, the cells were injected on two consecutive days. Abbreviations: LTBMC ⫽ long-term bone marrow culture transduced cells (number of input cells). *Two mice died within a few days after transplantation; **one mouse died immediately following injection of cells.
T. Relander et al./Experimental Hematology 28 (2000) 373–381
Figure 4. Engraftment of donor cells transplanted to nonmyeloablated recipients analyzed in peripheral blood at 4, 10, and 16 weeks post-transplant. 䊊 ⫽ 100 million fresh cells transplanted; 䊉 ⫽ 20 million fresh cells transplanted; 䉭 within 䊏 (baseline) ⫽ LTBMC cells generated from 20 or 100 million fresh cells, respectively. For actual cell number transplanted, see Table 2. Data from three separate experiments. Error bars denote SD.
the recovery reported from the application of this protocol to human (28%) and canine cells (10% and 15%) [17, 27,28]. Also, in terms of recovery of progenitor cells, our yield of 9.7% closely matches the results obtained with the human protocol (11%) and in two canine models (8% and 17%) [17,27,28]. When it comes to in vitro determination of transduction efficiency, we found that 19.8% of cells coming out of the LTBMC were GFP positive when analyzed by FACS, and that 22.2% of colonies were GFP positive when these cells were plated in methylcellulose. This is in the lower range but still compatible with what was observed in the preclinical experiments with human cells and about the same as was noted in the canine model of mucopolysaccharidosis I [27,28]. Even though 22.2% of progenitors coming out of the LTBMC were GFP positive, very few more long-lived hematopoietc cells were transduced. After transplantation to ablated recipients, GFP positive cells could be detected only at 4 weeks post-transplant. This is in sharp contrast to the data obtained with a standard 2-day prestimulation 2-day
379
Figure 5. Reduced long-term repopulating ablility of LTBMC cells. Equal amounts of fresh bone marrow cells and LTBMC transduced cells (0.5⫻106 cells of each) of different Ly 5.1/Ly 5.2 phenotype were transplanted to myeloablated (950 rad) Ly 5.1 or Ly 5.2 mice, and peripheral blood was analyzed for expression of Ly5.1 and Ly5.2 at 16 weeks posttransplant. Expression of GFP could not be detected. Data from two separate experiments. Error bars denote SD.
transduction protocol where a majority of donor-derived cells were GFP positive at all time points investigated. This shows that the low transduction efficiency obtained with the LTBMC protocol was not due to an inadequate vector or producer cell line. To rule out that the low or nonexistent level of GFP expression in mice transplanted with LTBMC cells was due to transcriptional silencing, semiquantitative PCR was performed on DNA isolated from bone marrow cells from these mice. The results confirmed that the level of gene transfer was in the range of 0.1–1%, which was in sharp contrast to the standard transduction procedure. Our competitive repopulation experiments showed that the cells coming out of LTBMC only maintained 7% of the LTRA as compared to fresh cells. This figure is in close accordance with previously published results where murine bone marrow cells were cultured for 3 weeks in LTBMC and then competed with fresh cells in a sex-mismatch transplantation model where the recipient mice were irradiated with 825 rad [29]. Four months post-transplant the LTRA of cultured cells, as determined in the peripheral blood of the recipients, was 4% and after 6 months it was 7%. It is clear,
380
T. Relander et al./Experimental Hematology 28 (2000) 373–381
from both previous data and our experiments, that culture of murine bone marrow cells in LTBMC leads to a dramatic loss of their long-term repopulating ability as determined in ablated recipients. In our experiments, we did not analyze the LTRA of mock-transduced cells. Thus we cannot say if the exposure of the LTBMC cells to virus containing supernatant per se has influenced the repopulating capacity of the cells. However, we deem it unlikely that two exposures to virus harvested in the same medium as used for the LTBMC had any effect on the LTRA of the cells. In a series of publications, Quesenberry and coworkers [30,31] have been able to show that it is possible to get engraftment of bone marrow cells in nonablated recipients if the cell dose is high enough. There seems to be a clear correlation between cell dose and the level of engraftment obtained [30]. Using fresh cells we were able to confirm these data, reaching 3.2% engraftment when 20 million cells were transplanted and 18.3% when 100 million cells were transplanted over two consecutive days. However, when LTBMC cells were transplanted to nonablated recipients, significant engraftment was not observed at any of the times points investigated. If one considers that the LTRA of cells coming out of the LTBMC transduction was 7% of fresh cells, one would have expected a 1–2% engraftment in the nonablated mice receiving the higher dose of LTBMC cells if their capacity to home and engraft was the same in ablated and nonablated recipients. As we saw no significant engraftment in these mice, one can draw the conclusion that the capability of LTBMC cells to home and engraft in the nonablated recipient is reduced at least to the same level as in ablated recipients. The mortality seen in our mice transplanted with high doses of LTBMC cells is a matter of serious concern when it comes to application of this protocol to humans. Compared with previously published canine studies and the human marking study, the cell dose in this part of our protocol was at least 10- to 100-fold higher, which might explain why no immediate harmful effects have been observed previously. That the dose of cells is important is also supported by the fact that none of our mice injected with only 0.5–2 million LTBMC cells died. On autopsy we found abundant thromboembolic material and bone marrow cells deposited in blood vessels of the lungs. It is noteworthy that Dubé et al. [27] in their studies where they injected human LTBMC transduced cells into SCID mice found human DNA in the lung tissue of the mice in many cases. They speculated that the inclusion of large numbers of stromal elements in the infused cells suspension could lead to inappropriate homing of the cells to the lung vasculature. Our findings further support this and also point to the possible acute effects of infusing high doses of LTBMC cells. In addition to the direct harmful effect of the cells being maintained in the lungs, this might of course also be one factor explaining the low level of bone marrow engraftment. We have conducted a systematic study of the use of LTBMC for gene transfer to murine bone marrow cells. In
both ablated and nonablated recipients and in a competitive repopulation setting, gene transfer to cells with repopulating ability and engraftment was very low, even though our in vitro data regarding recovery of cells and CFU and in vitro gene transfer were similar to what has been earlier reported for human and canine cells. The reasons for this discrepancy between the studies with human cells and those in large animal models on one hand and our murine experiments are not clear. In standard short-term transduction protocols, it has been much easier to transduce murine long-term repopulating cells than their human counterparts. A difference between the cycling activity of human and murine HSC is one factor put forward to explain this. While the more active cycling of murine HSC might be of advantage in a short transduction protocol, it can of course not be ruled out that the LTBMC transduction protocol is less well suited for murine cells than HSC from large animals and humans as indicated by our findings. In summary, our data show that the LTBMC transduction protocol is ineffective for repopulating HSC in the murine setting, both when it comes to transduction efficiency and engraftment.
Acknowledgments We thank Lilian Wittmann and Karin Olsson for expert technical assistance, Professor Unne Stenram for help with histopatholoigical analysis, and Eva Thorén for taking care of the animals used in this study. Financial support was obtained from the Swedish Medical Research Council (K99-31X -11620-04A), Cancerfonden (3652-B98-04XBC), the John Persson Foundation, the Georg Danielsson Foundation, and Foundations of Lund University Hospital. A grant (ALF grant) to S.K. from the Swedish Government to Clinical Researchers in the Public Health Service is also acknowledged.
References 1. Karlsson S (1991) Treatment of genetic defects in hematopoietic cell function by gene transfer. Blood 78:2481 2. Dunbar CE, Cottler-Fox M, O’Shaughnessy JA, et al. (1995) Retrovirally marked CD34-enriched peripheral blood and bone marrow cells contribute to long-term engraftment after autologous transplantation. Blood 85:3048 3. Brenner MK, Rill DR, Moen RC, Krance RA, Mirro J, Jr., Anderson WF, Ihle JN (1993) Gene-marking to trace origin of relapse after autologous bone-marrow transplantation. Lancet 341:85 4. Kiem HP, Darovsky B, von Kalle C, et al. (1994) Retrovirus-mediated gene transduction into canine peripheral blood repopulating cells. Blood 83:1467 5. Orlic D, Girard LJ, Jordan CT, Anderson SM, Cline AP, Bodine DM (1996) The level of mRNA encoding the amphotropic retrovirus receptor in mouse and human hematopoietic stem cells is low and correlates with the efficiency of retrovirus transduction. Proc Natl Acad Sci U S A 93:11097 6. Miller DG, Adam MA, Miller AD (1990) Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol 10:4239 7. Abkowitz JL, Persik MT, Shelton GH, Ott RL, Kiklevich JV, Catlin
T. Relander et al./Experimental Hematology 28 (2000) 373–381
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
SN, Guttorp P (1995) Behavior of hematopoietic stem cells in a large animal. Proc Natl Acad Sci U S A 92:2031 Petzer AL, Eaves CJ, Barnett MJ, Eaves AC (1997) Selective expansion of primitive normal hematopoietic cells in cytokine-supplemented cultures of purified cells from patients with chronic myeloid leukemia. Blood 90:64 Rebel VI, Tanaka M, Lee JS, et al. (1999) One-day ex vivo culture allows effective gene transfer into human nonobese diabetic/severe combined immune-deficient repopulating cells using high-titer vesicular stomatitis virus G protein pseudotyped retrovirus. Blood 93:2217 Gan OI, Murdoch B, Larochelle A, Dick JE (1997) Differential maintenance of primitive human SCID-repopulating cells, clonogenic progenitors, and long-term culture-initiating cells after incubation on human bone marrow stromal cells. Blood 90:641 Conneally E, Cashman J, Petzer A, Eaves C (1997) Expansion in vitro of transplantable human cord blood stem cells demonstrated using a quantitative assay of their lympho-myeloid repopulating activity in nonobese diabetic-scid/scid mice. Proc Natl Acad Sci U S A 94:9836 Piacibello W, Sanavio F, Severino A, et al. (1999) Engraftment in nonobese diabetic severe combined immunodeficient mice of human CD34(⫹) cord blood cells after ex vivo expansion: evidence for the amplification and self-renewal of repopulating stem cells. Blood 93:3736 Bhatia M, Bonnet D, Kapp U, Wang JC, Murdoch B, Dick JE (1997) Quantitative analysis reveals expansion of human hematopoietic repopulating cells after short-term ex vivo culture. J Exp Med 186:619 Moore KA, Deisseroth AB, Reading CL, Williams DE, Belmont JW (1992) Stromal support enhances cell-free retroviral vector transduction of human bone marrow long-term culture-initiating cells. Blood 79:1393 Xu LC, Kluepfel-Stahl S, Blanco M, Schiffmann R, Dunbar C, Karlsson S (1995) Growth factors and stromal support generate very efficient retroviral transduction of peripheral blood CD34⫹ cells from Gaucher patients. Blood 86:141 Wells S, Malik P, Pensiero M, Kohn DB, Nolta JA (1995) The presence of an autologous marrow stromal cell layer increases glucocerebrosidase gene transduction of long-term culture initiating cells (LTCICs) from the bone marrow of a patient with Gaucher disease. Gene Ther 2:512 Bienzle D, Abrams-Ogg AC, Kruth SA, et al. (1994) Gene transfer into hematopoietic stem cells: long-term maintenance of in vitro activated progenitors without marrow ablation. Proc Natl Acad Sci U S A 91:350 Stewart AK, Dubé ID, Kamel-Reid S, Keating A (1995) A phase I study of autologous bone marrow transplantation with stem cell gene marking in multiple myeloma. Hum Gene Ther 6:107
381
19. Stewart AK, Sutherland DR, Nanji S, et al. (1999) Engraftment of gene-marked hematopoietic progenitors in myeloma patients after transplant of autologous long-term marrow cultures. Hum Gene Ther 10:1953 20. Malech HL (1999) Progress in gene therapy for chronic granulomatous disease. J Infect Dis 179(Suppl 2):S318[/3a] 21. Dunbar CE, Kohn DB, Schiffmann R, et al. (1998) Retroviral transfer of the glucocerebrosidase gene into CD34⫹ cells from patients with Gaucher disease: in vivo detection of transduced cells without myeloablation. Hum Gene Ther 9:2629 22. Schiffmann R, Medin JA, Ward JM, Stahl S, Cottler-Fox M, Karlsson S (1995) Transfer of the human glucocerebrosidase gene into hematopoietic stem cells of nonablated recipients: successful engraftment and long-term expression of the transgene. Blood 86:1218 23. Persons DA, Allay JA, Allay ER, Smeyne RJ, Ashmun RA, Sorrentino BP, Nienhuis AW (1997) Retroviral-mediated transfer of the green fluorescent protein gene into murine hematopoietic cells facilitates scoring and selection of transduced progenitors in vitro and identification of genetically modified cells in vivo. Blood 90:1777 24. Dexter TM, Allen TD, Lajtha LG (1977) Conditions controlling the proliferation of haemopoietic stem cells in vitro. J Cell Physiol 91:335 25. Cashman J, Eaves AC, Eaves CJ (1985) Regulated proliferation of primitive hematopoietic progenitor cells in long-term human marrow cultures. Blood 66:1002 26. Cashman JD, Eaves AC, Raines EW, Ross R, Eaves CJ (1990) Mechanisms that regulate the cell cycle status of very primitive hematopoietic cells in long-term human marrow cultures. I. Stimulatory role of a variety of mesenchymal cell activators and inhibitory role of TGFbeta. Blood 75:96 27. Dubé ID, Kruth S, Abrams-Ogg A, et al. (1996) Preclinical assessment of human hematopoietic progenitor cell transduction in long-term marrow cultures. Hum Gene Ther 7:2089 28. Lutzko C, Kruth S, Abrams-Ogg AC, et al. (1999) Genetically corrected autologous stem cells engraft, but host immune responses limit their utility in canine alpha-L-iduronidase deficiency. Blood 93:1895 29. van der Sluijs JP, van den Bos C, Baert MR, van Beurden CA, Ploemacher RE (1993) Loss of long-term repopulating ability in long-term bone marrow culture. Leukemia 7:725 30. Rao SS, Peters SO, Crittenden RB, Stewart FM, Ramshaw HS, Quesenberry PJ (1997) Stem cell transplantation in the normal nonmyeloablated host: relationship between cell dose, schedule, and engraftment. Exp Hematol 25:114 31. Stewart FM, Crittenden RB, Lowry PA, Pearson-White S, Quesenberry PJ (1993) Long-term engraftment of normal and post-5-fluorouracil murine marrow into normal nonmyeloablated mice. Blood 81:2566
Low level of gene transfer to and engraftment of murine bone marrow cells from long-term bone marrow cultures Thomas Relander, Cecilia Fahlman, Stefan Karlsson, and Johan Richter Department of Molecular Medicine and Gene Therapy, Wallenberg Neuroscience Center, University of Lund, Lund, Sweden (Received 22 September 1999; revised 16 December 1999; accepted 28 December 1999)
Objective. We wanted to determine whether the long-term bone marrow culture (LTBMC) transduction system would lead to efficient gene transfer and engraftment of murine repopulating hematopoietic stem cells (HSC), particularly in nonablated recipients. Materials and Methods. Congenic mouse strains expressing Ly 5.1 or Ly 5.2 and the GP⫹E86 cell line producing the MGirL22Y vector carrying the gene for enhanced GFP were used. Murine LTBMCs were established and demi-depopulated on days 7 and 14 with addition of vector supernatant on days 8 and 15. Results. Cell recovery on day 21 was 21.3% ⫾ 3.8% of input cells and CFU-C recovery was 9.7 ⫾ 3.4% as compared with CFU-C of input cells. In vitro transduction efficiency determined by CFU-C expressing GFP was 22.2% ⫾ 1.6%. In irradiated (950 cGy) mice transplanted with 2 ⫻ 106 LTBMC cells, 94% of nucleated cells in the blood at week 16 were of donor origin. However, GFP was only detected at low level in a few animals at week 4 and not later. Analysis of bone marrow from these mice at week 20 did not show any GFP expression and semiquantitative PCR revealed a transgene level of ⬍ 1%. When 3.5–20.8 ⫻ 106 LTBMC cells (corresponding to 20–100 ⫻ 106 fresh cells) were transplanted to nonablated recipients, no engraftment or GFP expression were detected. Competitive repopulation experiments showed that the longterm repopulation ability (LTRA) of the LTMC cells was only 7% of fresh cells. Conclusion. These results indicate that LTBMC transduction of murine cells leads to lowlevel transduction of progenitors, no gene transfer to repopulating stem cells, and reduction in LTRA in ablated and nonablated recipients. © 2000 International Society for Experimental Hematology. Published by Elsevier Science Inc. Keywords: Gene transfer—Long-term culture—Engraftment—Bone marrow
Introduction Retroviral-mediated gene transfer into hematopoietic stem cells (HSC) has the potential to provide permanent correction of genetic diseases affecting the hematopoietic system [1]. In murine models it has been possible to show efficient gene transfer and expression into long-term repopulating hematopoietic cells. In contrast, results from human clinical trials and studies using large nonhuman primates have, in general, been disappointing, with only a very low level of gene transfer into long-lived hematopoietic cells [2–4]. The reasons for this discrepancy are only partially known. One factor is thought to be that the level of the relevant retroviral receptor is very low on human HSC [5]. AnOffprint requests to: Thomas Relander, M.D., Department of Molecular Medicine and Gene Therapy, Wallenberg Neuroscience Center, Sölvegatan 17, 223 62 Lund, Sweden; E-mail: [email protected]
other factor of importance is the fact that most human HSC are quiescent and nondividing, and that cell division is required for retroviral vectors to integrate into the genome of the infected cells [6,7]. This requirement for cell division has been the major reason to include stimulatory cytokines into the transduction protocols used so far. Thus most preclinical and clinical gene transfer protocols have included a period of prestimulation with cytokines followed by exposure to the retroviral vector 2–4 times during a period of another 2–3 days, most often in the presence of the same cytokine combination. Despite these efforts, only low levels of gene transfer to human HSC have been achieved. The reasons for this failure can be many, one being that even if a majority of the cells under such conditions are triggered into cell cycle, it is not unlikely that the small minority representing the true HSC may still be quiescent after only a few days of cytokine stimulation. Experimental evidence to sug-
0301-472X/00 $–see front matter. Copyright © 2000 International Society for Experimental Hematology. Published by Elsevier Science Inc. PII S0301-472X(00)0 0 1 3 1 - 4
374
T. Relander et al./Experimental Hematology 28 (2000) 373–381
gest that at least some human HSC may require as much as 9 days of ex vivo culture to undergo cell division has been presented [8]. Another possibility is that ex vivo exposure to at least some cytokines leads to altered function of human HSC, such as diminished self-renewal capability or decreased ability to engraft when reinfused into the host. Data from experiments using NOD/SCID mice support this concept as there was a decline in the number of SCID repopulating cells (SRC) when cells were cultured ex vivo with cytokines [9,10]. However, it should be noted that other studies have shown that SRC can be expanded ex vivo under certain conditions [11–13]. Some investigators have hypothesized that transduction and preservation of HSC would be more efficient in a microenvironment simulating the normal conditions in the bone marrow. Indeed it has been shown that exposure of hematopoietic cells to retroviral vectors in the presence of marrow-derived stromal cells leads to an increase in transduction efficiency [14–16]. This concept was taken even further by Dubé and coworkers [17], who adopted the Dexter-type long-term bone marrow culture (LTBMC) system for gene transfer purposes. In a canine model system bone marrow was set up in LTBMC and transduced by weekly exposures to a retroviral vector and after 3 weeks reinfused into nonablated autologous hosts [17]. Reporter genemarked progenitors were detected at approximately the 5% level for up to 2 years. This level of gene marking was surprisingly high, especially considering that it was obtained without any myeloablative regimen given prior to reinfusion of the manipulated cells. Based on these encouraging data, a human marking trial in patients with multiple myeloma was initiated and the results have recently been published [18,19]. In most genetic diseases affecting the hematopoietic system the use of myeloablative conditioning of the patient prior to reinfusion of transduced HSC is not justified. Clinical trials involving patients with Gaucher disease and chronic granulomatous disease have shown that to obtain significant engraftment of ex vivo manipulated HSC in the nonablated recipient poses an additional challenge to the already mentioned problem of low transduction [20,21]. In the mouse, it has been possible to get long-term engraftment of transduced cells in a nonablated recipient, albeit at a low level [22]. It was in this context we wanted to explore whether the LTBMC transduction system would lead to efficient gene transfer and engraftment of murine repopulating HSC. We were particularly interested to ask whether gene marked repopulating cells could be detected in nonablated mouse recipients as has been reported for the canine model [17]. However, the results demonstrate that the LTBMC transduction leads to poor engraftment of gene marked repopulating cells in murine recipients. Furthermore, gene transfer efficiency into repopulating HCSs is very low by this method, as judged by transplantation into ablated recipients.
Materials and methods Mouse strains The two congenic mouse strains C57BL/6 (Thy-1.2, Ly5.2) and C57BL/6-Ly5.1-Pep3b (Thy-1.2, Ly5.1) originally from Jackson Laboratories (Bar Harbor, ME) and hereafter called Ly5.2 and Ly5.1, respectively, were bred at the animal facility of Department of Tumor Immunology, University of Lund, Sweden. All animals were housed in ventilated racks and given acidified water and food ad libitum. Marrow harvest and establishment of LTBMCs Fresh bone marrow cells were obtained from 8- to 12-week-old donor animals. Mice were not given any treatment prior to bone marrow harvest. Femurs and tibias were removed and bone marrow cells were flushed aseptically using syringes and needles. Cells were dispersed into a single-cell suspension by repeated flushing and then counted and washed before they were resuspended in Myelocult (HCC 5300, Stem Cell Technologies, Vancouver, Canada) with Hydrocortisone 10⫺7 M (H4001; Sigma Chemical, St. Louis, MO), hereafter referred to as complete Myelocult. For establishment of LTBMCs, 20 ⫻ 106 cells in 10 mL of complete Myelocult were seeded into vented 25 cm2 flasks (Falcon 009-3108; Becton-Dickinson Labware, Bedford, MA) or 60 ⫻ 106 cells in 30 mL of complete Myelocult were seeded into 75 cm2 flasks (Falcon 009-3110). Flasks were maintained at 37⬚C in a humidified 5% CO2 in air environment. Retroviral vector and producer cell line The ecotropic packaging cell line GP⫹E86 producing the MGirL22Y retroviral vector was kindly supplied by Derek Persons [23]. This bicistronic vector contains the gene for enhanced GFP, an internal ribosomal entry site from the encephalomyocarditis virus and a mutated variant of the human dihydrofolate reductase gene. The titer of the viral supernatant when tested in our lab was 2–3 ⫻ 106 infectious particles per milliliter when assessed by FACS for transfer of the green fluorescent protein to NIH3T3 cells. This is in accordance with the titer previously reported. The producer cell line was maintained in DMEM medium with 10% FCS and penicillin/streptomycin at 37⬚C, except 48 hours prior to harvest of virus for the LTBMCs when it was moved to 33⬚C. Twenty-four hours prior to harvest, medium was changed to fresh complete Myelocult. Supernatants were filtered (0.45 m MillexHV, Low protein binding; Millipore, Bedford, MA) and then used immediately without freezing. Transduction of LTBMC cells Once LTBMCs were established, they were maintained in the standard way by weekly (day 7 and 14) removal of one half of the media, including cells therein, followed by addition of fresh complete Myelocult. On days 8 and 15, demi-depopulations of the nonadherent cells were also performed but this time the cultures were replenished with Myelocult conditioned on the vector producing cell line (5 mL for 25 cm2 and 15 mL for 75 cm2 flasks). In accordance with the protocol used by Dubé and coworkers [17], protamine sulphate or polybrene was not added. On day 21, LTBMC adherent layers were enzymatically removed by incubation with 0.25% trypsin-1 mM EDTA in HBSS without Ca and Mg (GIBCO BRL, Paisley, Scotland). Care was taken to ensure that the cells were dispersed into a single cell suspension. Recovered cells were washed once and resuspended in PBS for injection into mice or PBS, with
T. Relander et al./Experimental Hematology 28 (2000) 373–381
1% FCS for FACS analysis, or resuspended in medium before plating in semisolid methylcellulose. Four-day standard transduction Fresh bone marrow cells were prestimulated with 50 U/mL murine IL-6 (generously provided by Genetics Institute, Cambridge, MA) and 100 ng/mL rat stem cell factor (kind gift from Amgen, Thousand Oaks, CA) in DMEM supplemented with 15% FCS for 48 hours and then cocultured with the producer cell line in the above culture media supplemented with 4 g/mL protamine sulphate (Sigma-Aldrich Sweden, Lund, Sweden). Forty-eight hours later, nonadherent bone marrow cells were gently rinsed off the producer cells, washed, and resuspended in PBS for injection into recipient mice or resuspended in medium for in vitro clonogenic progenitor assays. A small aliquot of cells was cultured another 24 hours in the same medium as above and analyzed by FACS for GFP expression. Analysis of transduced cells CFU-C assays were performed using methylcellulose optimized to support clonogenic proliferation and differentiation of granulocyte-macrophage colony-forming-units (Methocult HCC3534; Stem Cell Technologies). Cultures were incubated at 37⬚C in 5% CO2/95% air in a humidified chamber for 6–7 days. The number of colonies generated from LTBMCs was compared with the number of colonies generated from the same starting cell population on the day of bone marrow harvest in order to determine the recovery of CFU after LTBMC. Plates containing hematopoietic colonies growing in methylcellulose were also directly evaluated for GFP expression by visualization using a fluorescence microscope. Transplantation of transduced cells and fresh cells Recipient mice, 8–10 weeks weeks old, were irradiated at 950 cGy total dose from a dual 137Cs source (Gammarad 900; Scanditronix Inc., Uppsala, Sweden), followed by tail-vein injection of either transduced cells, fresh bone marrow cells or, for competitive repopulation experiments, a mixture of both.
375
ing cells divided by the sum of Ly5.2 and Ly5.1 staining cells when Ly5.2 mice were the donors. GFP expression is shown as the number of cells expressing GFP and Ly5.1 or Ly5.2 divided by the sum of Ly5.2 and Ly5.1 staining cells. Semiquantitative PCR DNA was isolated from harvested bone marrow cells using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN). DNA (0.25 g) from each sample was amplified in separate tubes using primers either for GFP or mouse -actin. The GFP primers were 5⬘-GAG CTG GAC GGC GAC GTA AAC G-3⬘, and 5⬘CGC TTC TCG TTG GGG TCT TTG CT-3⬘, yielding a 595 bp fragment. Mouse -actin primers were 5⬘-CAT TGT GAT GGA CTC CGG TGA CGG-3⬘ and 5⬘-CAT AGC ACA GCT TCT CTT TGA TG-3⬘, yielding a 206 bp fragment. A Peltier Thermal Cycler (PTC-200; MJ Research Inc., Watertown, MA, USA) was used, and the conditions for the GFP PCR were: 95⬚C for 5 minutes and then 32 cycles of 95⬚C for 1 minute; 60⬚C for 1 minute; and 72⬚C for 1 minute. For the actin PCR: 95⬚C for 5 minutes and then 32 cycles of 95⬚C for 1 minute; 55⬚C for 1 minute; and 72⬚C for 2 minutes. For both reactions the above cycles were followed by an 8-minute extension at 72⬚C. The resulting PCR products were run on an ethidiumbromide containing agarose gel (1%), directly visualized under UV light and also transferred to nylon membranes using established Southern blotting techniques. A standard curve was established by extracting DNA from the MGirL22Y producing GP⫹E86s serially diluted into nontransfected GP⫹E86. The MGirL22Y producing GP⫹E86 cell line has previously been reported to contain 15 copies of the vector [23]. Negative controls in every reaction set included no DNA and DNA extracted from nontransfected GP⫹E86 cells. Histopathology Tissues were preserved in 4% paraformaldehyde (PFA) overnight, embedded in paraffin, sectioned and stained with hematoxylin and eosin before examination by light microscopy.
Results Analysis of engraftment Peripheral blood from experimental animals was analyzed at 4-, 10-, and 16-weeks after transplantation. In some experiments bone marrow was analyzed at 20 weeks post-transplant when the mice were sacrificed. Peripheral blood (approximately 250 L) was obtained from the retroorbital sinus. Whole blood was incubated with mouse whole IgG to block unspecific binding prior to antibody staining. After two washes with PBS with 1% FCS, cells were incubated on ice for 20–30 minutes with anti-CD45.1-PE and antiCD45.2-biotin (Pharmingen) and then washed twice before incubation with streptavidin-APC for 20 minutes, after which RBC lysis (BD-lysis; Becton-Dickinson, San Jose, CA) was performed followed by two more washes. Flow cytometry was performed using a FACS Calibur (Becton Dickinson) using excitation at 488 nm and fluorescence detection at 530 nm (for GFP), 585 nm (for phycoerytrin), and excitation 633 nm and detection 650 nm (for APC). Ten thousand events were collected for each sample. For each mouse, engraftment was measured as the number of Ly5.1 staining cells divided by the sum of Ly5.2 and Ly5.1 staining cells when Ly5.1 mice were used as donors and as the number of Ly5.2 stain-
Standard 4-day transduction As a control to test the viral producer line and the transplantation procedure, a standard 4-day transduction was performed. Bone marrow cells harvested from Ly5.1 mice were prestimulated with SCF and IL-6 for 48 hours and then cocultivated with the MGirL22Y producing cell line for another 48 hours in the presence of the same cytokines and protamine sulphate 4 g/mL. Lethally irradiated Ly5.2 mice were then transplanted with 0.5 ⫻ 106 transduced cells each. FACS analysis 24 hours post-transduction showed that 45.2% of the cells expressed the GFP protein (Fig. 1). Cells were also plated into methylcellulose and the frequency of GFP positive hematopoietic colonies was 89% as determined by fluorescence microscopy (Fig. 1). Peripheral blood from the transplanted mice was analyzed after 4, 10, and 16 weeks to determine level of engraftment and expression of the marker gene. The contribution of donor cells to the hematopoiesis in the mice increased from 70.4% (SD
376
T. Relander et al./Experimental Hematology 28 (2000) 373–381
1.5%) at 4 weeks post-transplant to 89.3% (SD 11.3%) at 16 weeks. The level of GFP expressing cells did not change significantly from week 4 to 16 and was approximately 50%. The mice were sacrificed 22 weeks after transplantation, and analysis of bone marrow at this time point, with regard to level of donor cells and GFP expression, did not differ significantly from the levels in the peripheral blood seen at week 16 (Fig. 1). These data thus confirm previous findings that the vector and producer cell line used in this work are able to transduce long-term repopulating murine bone marrow cells to a high degree and that the gene is expressed at the same level for at least 5 months [23]. In vitro analysis of LTBMC transduced cells In four separate experiments bone marrow from non-pretreated mice were set up in LTBMC in complete Myelocult. The protocol used closely followed the modified LTBMC transduction protocol developed by Dubé and coworkers [17,18], which was subsequently used in their clinical trial. LTBM cultures were demi-depopulated and replenished with new complete Myelocult on days 7 and 14. On days 8 and 15, second rounds of demi-depopulation were performed but this time the cultures were replenished with complete Myelocult freshly conditioned on the MGirL22Y producer cell line. On day 21, all adherent cells were harvested after trypsin treatment of the flasks. Cells from each flask were counted, analyzed by FACS for GFP expression, and a small aliquot was also plated in methylcellulose. The number of cells harvested was compared with input number
of cells, and the cell recovery was determined (Table 1). Overall recovery in all four experiments was 21.6%, with little variation from one experiment to another. The yield of colonies from each LTBMC was compared with the number of colonies obtained from methylcellulose cultures set up from the same cell population on the day when the LTBMC was initiated. This figure also takes into consideration the reduction in total cell number. Overall recovery of CFU in experiments 2–4 was 9.7% (Table 1). FACS analysis was done on the adherent LTBMC cells from each flask and the frequency of GFP expressing cells was determined by gating on the hematopoietic cells; 19.8% of cells (SD 6.0%) were GFP positive. The frequency of GFP–positive hematopoietic colonies derived from LTBMCs was 22.2% (SD 1.6%) as determined by fluorescence microscopy (Table 1). Transplantation of LTBMC transduced cells into myeloablated recipients In two separate experiments, lethally irradiated mice were transplanted with 1–2 ⫻ 106 transduced LTBMC cells each and then monitored by analysis of peripheral blood at 4, 10, and 16 weeks post-transplant. The results in Figure 2 show that the LTBMC cells were able to reconstitute the myeloablated mice. At 16 weeks, 94% of the cells in peripheral blood were of donor origin. However, GFP-expressing cells over the detection limit could be detected only at 4 weeks and at a very low level. At 10 and 16 weeks, no GFP expression above baseline was detected. Furthermore, when the mice were sacrificed at 20 weeks, GFP-expressing cells
Figure 1. Transduction and engraftment of bone marrow cells after a 4-day standard transduction. Day 5 ⫽ percentage of GFP positive mononuclear cells analyzed by FACS 5 days after start of transduction; CFU ⫽ percentage of GFP positive colonies in methylcellulose analyzed by fluorescence microscopy. Weeks 4, 10, and 16 ⫽ peripheral blood was analyzed by FACS for expression of donor and recipient phenotype and GFP at the time points indicated. Week 22 ⫽ analysis of bone marrow 22 weeks post-transplant. Error bars denote SD.
T. Relander et al./Experimental Hematology 28 (2000) 373–381 Table 1. In vitro data from murine LTBMC transduction Exp. 1
Exp. 2
Exp. 3
Exp. 4
Mean
23% (cells 25.8% pooled) SD 12.8 n⫽3 Recovery of CFU-C, 8.2% mean SD 2.5 % GFP positive 19.4% adherent cells SD 3.2 day 21 n⫽3 % GFP positive CFU 21.3% SD 2.8
19.0% SD 8.3 n⫽6 7.3% SD 1.5 26.0% SD 8.6 n⫽6 24.1% SD 10.5
18.5% SD 6.8 n ⫽ 12 13.5% SD 2.4 14.1% SD 7.7 n⫽9 21.3% SD 10.5
21.6% SD 3.5 (4 exp.) 9.7% SD 3.4 19.8% SD 6.0
Recovery of cells, mean
22.2% SD 1.6
In vitro data from four experiments: separate and pooled. Abbreviations: n ⫽ number of flasks analyzed; SD ⫽ standard deviation; exp. ⫽ experiment.
could not be detected by flow cytometry in the bone marrow either. Semiquantitative PCR analysis of DNA extracted from the bone marrow of these mice showed a transgene level of approximately 0.1–1% (Fig. 3). In contrast, PCR analysis of bone marrow from one of the mice who had received cells transduced by the standard transduction method revealed a transgene level of close to 100%.
Figure 2. Engraftment of donor cells and expression of GFP in myeloablated mice. Irradiated mice were transplanted with 1–2⫻106 LTBMC transduced cells. Peripheral blood was analyzed by FACS at 4, 10, and 16 weeks post-transplant for expression of Ly5.1, Ly5.2, and GFP. Error bars denote SD.
377
Transplantation of LTBMC transduced cells into nonablated recipients In order to investigate the ability of LTBMC transduced cells to engraft in nonablated recipients, Ly5.1 mice were transplanted with LTBMC cells. In three separate experiments mice were given LTBMC cells generated from 20 or 100 million marrow cells (for actual number of cells infused see Table 2). The higher dose was split and given on two consecutive days. For comparison, control mice were injected with 20 or 100 million fresh cells. The level of engraftment was determined by FACS analysis of the peripheral blood at weeks 4, 10, and 16. While mice given fresh cells showed a increased level of chimerism over time (Fig. 4), engraftment of LTBMC cells in nonablated recipients could not be detected. Competitive repopulation In order to estimate the long-term repopulating activity (LTRA) of the LTBMC cells, competitive repopulation experiments were performed. In two separate experiments a mixture of 0.5⫻106 LTBMC cells of Ly5.2 origin and an equal amount of fresh Ly5.1 bone marrow cells were transplanted into both Ly5.1 recipients and Ly5.2 recipients. Blood was obtained from all animals 16 weeks after transplantation and FACS analysis was performed (Fig. 5). The results show that in the Ly5.1 animals LTBMC cells were responsible for 22.6% of the chimerism seen in the blood, while for the Ly5.2 animals this figure was 26.9%. Because there was no significant difference between those two numbers, this shows that the recipient type did not influence the result of the competitive repopulation. Thus the average contribution of the LTBMC cells was 24.8% and the LTRA of the output LTBMC cells was calculated to be approximately 1/3 (24.8%/75.2%) of the LTRA of fresh cells when equal amounts of cells were compared. However, if one also considers that there was only a 21.6% recovery of cells from the LTBMC as compared with input, this means that the LTRA of the cells maintained in LTBMC, as determined in the peripheral blood of mice 16 weeks post-transplant, was approximately 7% of fresh cells. Also, GFP expression was undetectable in these mice. Transplant-related mortality after infusion of LTBMC cells In the group of nonmyeloablated mice injected with high numbers of LTBMC cells, there was a considerable transplant related mortality despite great care taken to disperse the harvested cells into a single cell suspension. In total, 17 mice received cell amounts corresponding to 20–100 million input cells from LTBMCs (Table 2). Out of these, five mice died in close association to the transplantation and more mice showed signs of respiratory distress. Autopsy was performed on one mouse showing abundant thromboembolic material deposited in blood vessels of the lungs, and in some of these deposits, bone marrow cells were seen. Some areas with embolic bone marrow cell clusters only
378
T. Relander et al./Experimental Hematology 28 (2000) 373–381
Figure 3. Semiquantitative PCR of bone marrow from myeloablated mice transplanted with cells transduced according to LTBMC protocol and standard 4-day protocol. Bone marrow was harvested from mice sacrificed 20–22 weeks after transplant. DNA was purified and PCR performed as described in the materials and methods section. (A)–(E): bone marrow from five mice transplanted with LTBMC cells, GP⫹E86 cells packaging cell line without vector, standard curve from dilutions of MGirL22Y vector producing GP⫹E86 cells in GP⫹E86 cells corresponding to 1 copy of vector/cell, 0.1 copy, 0.01 copy, 0.001 copy, and 0.0001 copy. (F) Bone marrow from mouse transplanted with cells from standard transduction.
were also found. There was no pathology detected in the heart. In the control group of 13 mice receiving fresh bone marrow cells (20–100 million cells), no animals died in connection to the cell transfer and the animals showed no signs of distress after injection. Of the 22 mice receiving lower doses of LTBMC cells (0.5–2 million), no mice died in conjunction to the transplant.
Discussion It is for many reasons a theoretically very attractive idea to use the Dexter-type LTBMC system for gene transfer to hematopoietic cells. These cultures can be maintained for many weeks during which proliferation of primitive cells can take place in a milieu imitating the bone marrow microenvironment [24]. No stimulatory cytokines that might prematurely force the HSC down the road of differentiation are added to these cultures. The more differentiated cells will be removed when the cultures are demi-depopulated,
while the primitive hematopoietic cells adhering to the stromal cells will remain in the culture and can be repeatedly exposed to vector supernatant. It has also been shown that the replacement of half of the medium with fresh medium stimulates the high proliferative potential hematopoietic progenitors in the adherent layer to go into cell cycle, possibly by removing TGF- [25,26]. Thus, this is a timepoint when primitive cells in the culture might be susceptible to retroviral gene transfer. Our experimental set up was modeled to closely match the 3-week LTBMC transduction described by Dubé and coworkers [17]. In their initial experiments they exposed the bone marrow cells to viral supernatant three times, on days 1, 8, and 15. However, in a later modification of the protocol, cells were exposed only twice to the vector, and in our experiments we settled for a two-cycle transduction procedure [27]. The overall recovery of cells from the LTBMCs on day 21 was 21.6% in our experiments, with little variation from experiment to experiment. This figure closely correlates to
Table 2. Transplantation of bone marrow cells to nonablated recipients Fresh cells 20 million Experiment 1 Experiment 3 Experiment 4
LTBMC
100 million
5 4
20 million input
100 million input
5* (4.6 million output cells) 4
5** (mean output cell number 3.5 million/mouse)
3** (14.3 million output cells) 3** (mean output cell number 20.8 million/mouse)
Transplantation of bone marrow cells to nonablated recipients. Number of mice and transplanted cells in different experiments. For LTBMC input cells refer to number of cells used to set up the LTBMC, while actual number of cells at harvest day 21 is given in brackets. In mice transplanted with 100 million input LTBMC cells, the cells were injected on two consecutive days. Abbreviations: LTBMC ⫽ long-term bone marrow culture transduced cells (number of input cells). *Two mice died within a few days after transplantation; **one mouse died immediately following injection of cells.
T. Relander et al./Experimental Hematology 28 (2000) 373–381
Figure 4. Engraftment of donor cells transplanted to nonmyeloablated recipients analyzed in peripheral blood at 4, 10, and 16 weeks post-transplant. 䊊 ⫽ 100 million fresh cells transplanted; 䊉 ⫽ 20 million fresh cells transplanted; 䉭 within 䊏 (baseline) ⫽ LTBMC cells generated from 20 or 100 million fresh cells, respectively. For actual cell number transplanted, see Table 2. Data from three separate experiments. Error bars denote SD.
the recovery reported from the application of this protocol to human (28%) and canine cells (10% and 15%) [17, 27,28]. Also, in terms of recovery of progenitor cells, our yield of 9.7% closely matches the results obtained with the human protocol (11%) and in two canine models (8% and 17%) [17,27,28]. When it comes to in vitro determination of transduction efficiency, we found that 19.8% of cells coming out of the LTBMC were GFP positive when analyzed by FACS, and that 22.2% of colonies were GFP positive when these cells were plated in methylcellulose. This is in the lower range but still compatible with what was observed in the preclinical experiments with human cells and about the same as was noted in the canine model of mucopolysaccharidosis I [27,28]. Even though 22.2% of progenitors coming out of the LTBMC were GFP positive, very few more long-lived hematopoietc cells were transduced. After transplantation to ablated recipients, GFP positive cells could be detected only at 4 weeks post-transplant. This is in sharp contrast to the data obtained with a standard 2-day prestimulation 2-day
379
Figure 5. Reduced long-term repopulating ablility of LTBMC cells. Equal amounts of fresh bone marrow cells and LTBMC transduced cells (0.5⫻106 cells of each) of different Ly 5.1/Ly 5.2 phenotype were transplanted to myeloablated (950 rad) Ly 5.1 or Ly 5.2 mice, and peripheral blood was analyzed for expression of Ly5.1 and Ly5.2 at 16 weeks posttransplant. Expression of GFP could not be detected. Data from two separate experiments. Error bars denote SD.
transduction protocol where a majority of donor-derived cells were GFP positive at all time points investigated. This shows that the low transduction efficiency obtained with the LTBMC protocol was not due to an inadequate vector or producer cell line. To rule out that the low or nonexistent level of GFP expression in mice transplanted with LTBMC cells was due to transcriptional silencing, semiquantitative PCR was performed on DNA isolated from bone marrow cells from these mice. The results confirmed that the level of gene transfer was in the range of 0.1–1%, which was in sharp contrast to the standard transduction procedure. Our competitive repopulation experiments showed that the cells coming out of LTBMC only maintained 7% of the LTRA as compared to fresh cells. This figure is in close accordance with previously published results where murine bone marrow cells were cultured for 3 weeks in LTBMC and then competed with fresh cells in a sex-mismatch transplantation model where the recipient mice were irradiated with 825 rad [29]. Four months post-transplant the LTRA of cultured cells, as determined in the peripheral blood of the recipients, was 4% and after 6 months it was 7%. It is clear,
380
T. Relander et al./Experimental Hematology 28 (2000) 373–381
from both previous data and our experiments, that culture of murine bone marrow cells in LTBMC leads to a dramatic loss of their long-term repopulating ability as determined in ablated recipients. In our experiments, we did not analyze the LTRA of mock-transduced cells. Thus we cannot say if the exposure of the LTBMC cells to virus containing supernatant per se has influenced the repopulating capacity of the cells. However, we deem it unlikely that two exposures to virus harvested in the same medium as used for the LTBMC had any effect on the LTRA of the cells. In a series of publications, Quesenberry and coworkers [30,31] have been able to show that it is possible to get engraftment of bone marrow cells in nonablated recipients if the cell dose is high enough. There seems to be a clear correlation between cell dose and the level of engraftment obtained [30]. Using fresh cells we were able to confirm these data, reaching 3.2% engraftment when 20 million cells were transplanted and 18.3% when 100 million cells were transplanted over two consecutive days. However, when LTBMC cells were transplanted to nonablated recipients, significant engraftment was not observed at any of the times points investigated. If one considers that the LTRA of cells coming out of the LTBMC transduction was 7% of fresh cells, one would have expected a 1–2% engraftment in the nonablated mice receiving the higher dose of LTBMC cells if their capacity to home and engraft was the same in ablated and nonablated recipients. As we saw no significant engraftment in these mice, one can draw the conclusion that the capability of LTBMC cells to home and engraft in the nonablated recipient is reduced at least to the same level as in ablated recipients. The mortality seen in our mice transplanted with high doses of LTBMC cells is a matter of serious concern when it comes to application of this protocol to humans. Compared with previously published canine studies and the human marking study, the cell dose in this part of our protocol was at least 10- to 100-fold higher, which might explain why no immediate harmful effects have been observed previously. That the dose of cells is important is also supported by the fact that none of our mice injected with only 0.5–2 million LTBMC cells died. On autopsy we found abundant thromboembolic material and bone marrow cells deposited in blood vessels of the lungs. It is noteworthy that Dubé et al. [27] in their studies where they injected human LTBMC transduced cells into SCID mice found human DNA in the lung tissue of the mice in many cases. They speculated that the inclusion of large numbers of stromal elements in the infused cells suspension could lead to inappropriate homing of the cells to the lung vasculature. Our findings further support this and also point to the possible acute effects of infusing high doses of LTBMC cells. In addition to the direct harmful effect of the cells being maintained in the lungs, this might of course also be one factor explaining the low level of bone marrow engraftment. We have conducted a systematic study of the use of LTBMC for gene transfer to murine bone marrow cells. In
both ablated and nonablated recipients and in a competitive repopulation setting, gene transfer to cells with repopulating ability and engraftment was very low, even though our in vitro data regarding recovery of cells and CFU and in vitro gene transfer were similar to what has been earlier reported for human and canine cells. The reasons for this discrepancy between the studies with human cells and those in large animal models on one hand and our murine experiments are not clear. In standard short-term transduction protocols, it has been much easier to transduce murine long-term repopulating cells than their human counterparts. A difference between the cycling activity of human and murine HSC is one factor put forward to explain this. While the more active cycling of murine HSC might be of advantage in a short transduction protocol, it can of course not be ruled out that the LTBMC transduction protocol is less well suited for murine cells than HSC from large animals and humans as indicated by our findings. In summary, our data show that the LTBMC transduction protocol is ineffective for repopulating HSC in the murine setting, both when it comes to transduction efficiency and engraftment.
Acknowledgments We thank Lilian Wittmann and Karin Olsson for expert technical assistance, Professor Unne Stenram for help with histopatholoigical analysis, and Eva Thorén for taking care of the animals used in this study. Financial support was obtained from the Swedish Medical Research Council (K99-31X -11620-04A), Cancerfonden (3652-B98-04XBC), the John Persson Foundation, the Georg Danielsson Foundation, and Foundations of Lund University Hospital. A grant (ALF grant) to S.K. from the Swedish Government to Clinical Researchers in the Public Health Service is also acknowledged.
References 1. Karlsson S (1991) Treatment of genetic defects in hematopoietic cell function by gene transfer. Blood 78:2481 2. Dunbar CE, Cottler-Fox M, O’Shaughnessy JA, et al. (1995) Retrovirally marked CD34-enriched peripheral blood and bone marrow cells contribute to long-term engraftment after autologous transplantation. Blood 85:3048 3. Brenner MK, Rill DR, Moen RC, Krance RA, Mirro J, Jr., Anderson WF, Ihle JN (1993) Gene-marking to trace origin of relapse after autologous bone-marrow transplantation. Lancet 341:85 4. Kiem HP, Darovsky B, von Kalle C, et al. (1994) Retrovirus-mediated gene transduction into canine peripheral blood repopulating cells. Blood 83:1467 5. Orlic D, Girard LJ, Jordan CT, Anderson SM, Cline AP, Bodine DM (1996) The level of mRNA encoding the amphotropic retrovirus receptor in mouse and human hematopoietic stem cells is low and correlates with the efficiency of retrovirus transduction. Proc Natl Acad Sci U S A 93:11097 6. Miller DG, Adam MA, Miller AD (1990) Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol 10:4239 7. Abkowitz JL, Persik MT, Shelton GH, Ott RL, Kiklevich JV, Catlin
T. Relander et al./Experimental Hematology 28 (2000) 373–381
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
SN, Guttorp P (1995) Behavior of hematopoietic stem cells in a large animal. Proc Natl Acad Sci U S A 92:2031 Petzer AL, Eaves CJ, Barnett MJ, Eaves AC (1997) Selective expansion of primitive normal hematopoietic cells in cytokine-supplemented cultures of purified cells from patients with chronic myeloid leukemia. Blood 90:64 Rebel VI, Tanaka M, Lee JS, et al. (1999) One-day ex vivo culture allows effective gene transfer into human nonobese diabetic/severe combined immune-deficient repopulating cells using high-titer vesicular stomatitis virus G protein pseudotyped retrovirus. Blood 93:2217 Gan OI, Murdoch B, Larochelle A, Dick JE (1997) Differential maintenance of primitive human SCID-repopulating cells, clonogenic progenitors, and long-term culture-initiating cells after incubation on human bone marrow stromal cells. Blood 90:641 Conneally E, Cashman J, Petzer A, Eaves C (1997) Expansion in vitro of transplantable human cord blood stem cells demonstrated using a quantitative assay of their lympho-myeloid repopulating activity in nonobese diabetic-scid/scid mice. Proc Natl Acad Sci U S A 94:9836 Piacibello W, Sanavio F, Severino A, et al. (1999) Engraftment in nonobese diabetic severe combined immunodeficient mice of human CD34(⫹) cord blood cells after ex vivo expansion: evidence for the amplification and self-renewal of repopulating stem cells. Blood 93:3736 Bhatia M, Bonnet D, Kapp U, Wang JC, Murdoch B, Dick JE (1997) Quantitative analysis reveals expansion of human hematopoietic repopulating cells after short-term ex vivo culture. J Exp Med 186:619 Moore KA, Deisseroth AB, Reading CL, Williams DE, Belmont JW (1992) Stromal support enhances cell-free retroviral vector transduction of human bone marrow long-term culture-initiating cells. Blood 79:1393 Xu LC, Kluepfel-Stahl S, Blanco M, Schiffmann R, Dunbar C, Karlsson S (1995) Growth factors and stromal support generate very efficient retroviral transduction of peripheral blood CD34⫹ cells from Gaucher patients. Blood 86:141 Wells S, Malik P, Pensiero M, Kohn DB, Nolta JA (1995) The presence of an autologous marrow stromal cell layer increases glucocerebrosidase gene transduction of long-term culture initiating cells (LTCICs) from the bone marrow of a patient with Gaucher disease. Gene Ther 2:512 Bienzle D, Abrams-Ogg AC, Kruth SA, et al. (1994) Gene transfer into hematopoietic stem cells: long-term maintenance of in vitro activated progenitors without marrow ablation. Proc Natl Acad Sci U S A 91:350 Stewart AK, Dubé ID, Kamel-Reid S, Keating A (1995) A phase I study of autologous bone marrow transplantation with stem cell gene marking in multiple myeloma. Hum Gene Ther 6:107
381
19. Stewart AK, Sutherland DR, Nanji S, et al. (1999) Engraftment of gene-marked hematopoietic progenitors in myeloma patients after transplant of autologous long-term marrow cultures. Hum Gene Ther 10:1953 20. Malech HL (1999) Progress in gene therapy for chronic granulomatous disease. J Infect Dis 179(Suppl 2):S318[/3a] 21. Dunbar CE, Kohn DB, Schiffmann R, et al. (1998) Retroviral transfer of the glucocerebrosidase gene into CD34⫹ cells from patients with Gaucher disease: in vivo detection of transduced cells without myeloablation. Hum Gene Ther 9:2629 22. Schiffmann R, Medin JA, Ward JM, Stahl S, Cottler-Fox M, Karlsson S (1995) Transfer of the human glucocerebrosidase gene into hematopoietic stem cells of nonablated recipients: successful engraftment and long-term expression of the transgene. Blood 86:1218 23. Persons DA, Allay JA, Allay ER, Smeyne RJ, Ashmun RA, Sorrentino BP, Nienhuis AW (1997) Retroviral-mediated transfer of the green fluorescent protein gene into murine hematopoietic cells facilitates scoring and selection of transduced progenitors in vitro and identification of genetically modified cells in vivo. Blood 90:1777 24. Dexter TM, Allen TD, Lajtha LG (1977) Conditions controlling the proliferation of haemopoietic stem cells in vitro. J Cell Physiol 91:335 25. Cashman J, Eaves AC, Eaves CJ (1985) Regulated proliferation of primitive hematopoietic progenitor cells in long-term human marrow cultures. Blood 66:1002 26. Cashman JD, Eaves AC, Raines EW, Ross R, Eaves CJ (1990) Mechanisms that regulate the cell cycle status of very primitive hematopoietic cells in long-term human marrow cultures. I. Stimulatory role of a variety of mesenchymal cell activators and inhibitory role of TGFbeta. Blood 75:96 27. Dubé ID, Kruth S, Abrams-Ogg A, et al. (1996) Preclinical assessment of human hematopoietic progenitor cell transduction in long-term marrow cultures. Hum Gene Ther 7:2089 28. Lutzko C, Kruth S, Abrams-Ogg AC, et al. (1999) Genetically corrected autologous stem cells engraft, but host immune responses limit their utility in canine alpha-L-iduronidase deficiency. Blood 93:1895 29. van der Sluijs JP, van den Bos C, Baert MR, van Beurden CA, Ploemacher RE (1993) Loss of long-term repopulating ability in long-term bone marrow culture. Leukemia 7:725 30. Rao SS, Peters SO, Crittenden RB, Stewart FM, Ramshaw HS, Quesenberry PJ (1997) Stem cell transplantation in the normal nonmyeloablated host: relationship between cell dose, schedule, and engraftment. Exp Hematol 25:114 31. Stewart FM, Crittenden RB, Lowry PA, Pearson-White S, Quesenberry PJ (1993) Long-term engraftment of normal and post-5-fluorouracil murine marrow into normal nonmyeloablated mice. Blood 81:2566