Technical Report Optimization of Culture Conditions to Enhance Transfection of Human CD34؉ Cells by Electroporation

Home , Stem cell factor, Thrombopoietin

Bone Marrow Transplantation (2001) 27, 1201–1209  2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00 www.nature.com/bmt Technical report Optimization of culture conditions to enhance transfection of human CD34؉ cells by electroporation

MH Wu1,* SL Smith1,2,*, GH Danet3, AM Lin1, SF Williams1,4, DN Liebowitz3,5 and ME Dolan1

1Department of Medicine, Section of Hematology-Oncology and Cancer Research Center, University of Chicago, Chicago, IL; and 3Howard Hughes Medical Institute and Abramson Family Cancer Research Institute, University of Pennsylvania, Philadelphia, PA, USA

Summary: stable as determined by EGFP+/LTC-IC (long-term colony-initiating cells), at 30–40 positive colonies (16 The ability to culture CD34+ stem cells, while main- ؎ 7%) per 1 × 105 electroporated CD34+ cells. Bone taining their pluripotency, is essential for manipu- Marrow Transplantation (2001) 27, 1201–1209. lations such as gene transfection for therapeutic trials. Keywords: CD34 cells; TPO; SCF; Flt-3L; cell cycle; Human peripheral blood (PB) CD34+ cells (у90% EGFP purity) were cultured for up to 4 days in serum-free culture medium supplemented with thrombopoietin (TPO), stem cell factor (SCF), Flt-3 ligand (Flt-3L), with or without PIXY321 (IL-3 /GM-CSF fusion The clinical efficacy of transplanting CD34+ stem cells with protein) and human serum. The CD34 mean fluor- modified genetic material is rapidly becoming an area of escence intensity (MFI) and cell cycle status were intense interest for both future and on-going therapeutic evaluated daily using flow cytometry and hypotonic trials in humans. Although the mode of gene delivery has -propidium iodide. Prior to culture (day 0), 97.0 ؎ been diverse, hematopoietic stem cell studies have pri -and 1.0 ؎ 0.6% of the selected marily used either retroviral or adenoviral vectors.1–5 Over 0.3% ؎ 1.9 ,0.9% CD34+ cells were in G0–G1, S-phase, or G2–M, all, there has been low stem cell transduction efficiency and respectively. After 2–4 days in culture with TPO/SCF/ some controversy as to the safety when using these vectors. Flt-3L, there was an increase in the percent of cells in For many current modes of gene transfer (with the excep- S-phase to 26.4 ؎ 0.1% without significant loss of tion of adenoviral and lentiviral systems), cell division is CD34 MFI. The addition of PIXY321 increased the required. Previous studies have demonstrated that electro- percentage of CD34+ cells in S-phase to 36.3 ؎ 4.0%, poration can be a reasonable means of hematopoietic cell but the CD34 MFI and numbers of CFU (colony-for- gene transfer, and the genes delivered are integrated into ming units) were significantly decreased at day 3 when the host chromosome.6–10 Although cell synchronization cultured with PIXY321 or various recombinant cyto- and electroporation studies have led to controversial con- kine combinations that included IL-3 and IL-6. There clusions on the requirement for S-phase,11–13 it appears that is an increase from day 0 to day 4 in the percentages efficient electroporation of hematopoietic cells requires pre- of CD34+ with CD38؊,HLA-DR؊, and c-kitlow, but not stimulation, which results in an apparent increase in the Thy-1+ cells. Electroporation with EGFP reporter S-phase sub-population.6,14,15 gene showed that 1–2 days of pre-stimulation in X- Ideal culture conditions would promote a high efficiency VIVO 10 supplemented with TPO/SCF/Flt-3L was of gene transfer to ensure not only short-term, but also necessary and sufficient for efficient transfection. Flow long-term engraftment and subsequent bone marrow cytometry analysis demonstrated that 22% of the repopulation with each hematopoietic lineage carrying the viable cells are CD34+/EGFP+ 48 h post electropor- transgene. In vitro manipulation of human stem cells for ation. The introduced reporter gene appears to be gene transfer by electroporation involves three primary goals: (1) optimization of short-term culture conditions that will increase the percentage of cells in S-phase and enhance successful incorporation of exogenous gene delivered; (2) Correspondence: ME Dolan, Section of Hematology-Oncology, University maintenance of the pluripotent stem cells for full bone mar- of Chicago, 5841 S Maryland Avenue, Box MC2115, Chicago, IL 60637, USA row engraftment and long-term reconstitution after trans- Present addresses: 2Institute for Transfusion Medicine, Clinical Services- plant; and (3) optimization of the electroporation conditions Stem Cell Department, 1205 N Milwaukee Ave, Glenview, IL 60025, for efficient transfection. It has been shown that many cul- USA; 4Hematology Oncology Associates of Illinois, 676 N St Clair, Suite ture conditions will support stem cell division. However, 2140, Chicago, IL 60611, USA; 5Immunex Corp., 51 University St, Seattle, WA 98101, USA this usually leads to loss of pluripotency and diminished *MHW and SLS contributed equally to this work short-term and/or long-term reconstituting ability as a Received 11 August 2000; accepted 1 March 2001 consequence of expansion and differentiation.16,17 Human CD34+ cell culture optimization for efficient electroporation MH Wu et al 1202 Optimal ex vivo culture conditions and/or recombinant Mg2+-free Dulbecco’s phosphate-buffered saline (D- growth factor combinations that will promote survival, pro- PBS) and X-VIVO 10 serum-free culture medium liferation, and self-renewal of the most primitive CD34+ (supplemented with 2 mml-glutamine) were from stem cell, yet prevent differentiation into committed pro- BioWhittaker (Walkersville, MD, USA). Human genitors are being explored.18–21 Various studies have been immunoglobulin solution (Gammagard) and human performed using numerous combinations of early-acting serum albumin (HSA) were from Baxter Healthcare cytokines that affect post-G0 phase exit, and initiation of (Deerfield, IL, USA). Flt-3L and PIXY321 (IL-3/GM- cell-cycling. These include thrombopoietin (TPO), stem CSF fusion protein) were kindly provided by Immunex cell factor (SCF), Flt-3 ligand (Flt-3L), interleukin-3 (IL- (Seattle, WA, USA). Thrombopoietin (TPO), stem cell 3), IL-12, IL-6, IL-1 and granulocyte–macrophage colony- factor (SCF), interleukin-3 and 6 (IL-3 and IL-6) were stimulating factor (GM-CSF), as well as late-acting cyto- purchased from R&D Systems (Minneapolis, MN, USA). kines (affecting proliferation and maturation) such as Trypan blue, propidium iodide (1 ␮g/␮l solution), hydro- erythropoietin (EPO), G-CSF, and TGF-beta.22–31 The dif- cortisone, bovine serum albumin (BSA), and pooled ferences between early- and late-acting cytokines are often human AB serum were from Sigma Chemical Company complicated by the fact that the effects of an individual (St Louis, MO, USA). MethoCult H4433 (methylcellulose cytokine in combination with others can be multiple, containing PHA-LCM and erythropoietin) and MyeloCult depending on the stage of differentiation of a particular H5100 (long-term culture medium) were from Stem Cell stem cell.32,33 In addition to recombinant cytokines, the role Technologies (Vancouver, BC, Canada). Miltenyi CD34 of stromal contact,17,34–36 stroma-conditioned media,37 fib- isolation kit and separation columns were from Miltenyi ronectin35 and the use of endothelial layers38 to maintain Biotec (Auburn, CA, USA). The FACScan and FACSCali- stem cells in a primitive state has been evaluated. Among bur Flow Cytometers, CellQuest, and Modfit LT softwares the most frequently used cytokine combinations, studies were from Becton Dickinson. Statistical significance was have consistently shown that TPO in various cocktails with determined by the Student’s t-test (P р 0.05) using Stat- the tyrosine kinase receptor ligands, Flt-3L and SCF, is cap- View (SAS Institute, Cary, NC, USA). able of maintaining phenotypic self-renewal of primitive stem cells.22,25,39–41 Numerous studies have shown that IL- 3 has an important role in survival42 and ex vivo expansion CD34+ cell selection of primitive cells.25,43 We chose to study culture conditions that would maintain the primitive state of stem cells yet Fresh or cryopreserved peripheral blood apheresis products maximize the proportion of CD34+ cells in S-phase in order from non-cancerous donors or ovarian and breast cancer to determine the optimal time for gene transfer by electro- patients, treated with prior chemotherapy and/or recombi- poration. Our goal was to maintain CD34 pluripotency and nant granulocyte colony-stimulating factor (G-CSF), were optimize the culture conditions for electroporation using used as the source of cells for CD34+ cell selection. Periph- enhanced green fluorescence protein (EGFP) as a reporter eral blood apheresis mononuclear cells were resuspended gene. The effect of culturing CD34+ cells with TPO, Flt- at a final concentration of 2 × 107 cells/ml in selection 3L, and SCF, with or without IL-3 and PIXY321 (GM- buffer (0.4% sodium citrate, 1% HSA in D-PBS) sup- CSF/IL-3 fusion protein) in addition to other combinations plemented with 0.5% human immunoglobulin solution and of recombinant cytokines was explored. We also studied incubated for 15 min at room temperature. CD34 selection the effect of stroma-conditioned media and human serum reagents were obtained from Baxter Healthcare and used on the transfection efficiency in CD34+ cells. The cell-cycle as described by the manufacturer. Anti-CD34 monoclonal (%S-phase cells), EGFP expression in CD34+ cells, colony- antibody (9C5, mouse IgG1) was added at 0.5 ␮g/106 cells forming units (CFU), and long-term culture-initiating cells and the cells were mixed by end-over-end rotation for 30 (LTC-IC) were used to determine optimal reporter gene min at room temperature. The cells were then washed twice delivery with electroporation. by centrifugation for 10 min at 250 g and resuspended at 2 × 107 cells/ml in selection buffer. Paramagnetic beads coated with sheep anti-mouse IgG1 were added to the cell Materials and methods suspension at a cell:bead ratio of 1:1 and rotated for 30 min at room temperature. The bead–cell complexes were Materials isolated using a magnetic tube holder and washed twice with selection buffer. CD34+ cells were released from the The HPCA-2 monoclonal antibody (anti-CD34 PE or magnetic beads by addition of 1:5 dilution of CD34 peptide APC), CD38 PE, CD117 (c-kit) PE and HLA-DR FITC release agent diluted in selection buffer, followed by an were from Becton Dickinson (San Jose, CA, USA). Thy-1 end-over-end rotation for 30 min at room temperature. The monoclonal antibody (anti-CD90 PE) was from Phar- beads were then removed on the magnet and the released Mingen (San Diego, CA, USA). Enhanced green fluor- CD34+ cells were isolated in selection buffer. CD34+ purity escence protein (EGFP) used in this study is encoded was evaluated by flow cytometry using anti-CD34-PE on the plasmid pEGFP-N1 (ClonTech Laboratories, Palo (HPCA-2). The Miltenyi CD34 isolation kit, MidiMacs, Alto, CA, USA). The control plasmid pcDNA3 was from and SuperMacs separation columns were also used accord- Invitrogen (Carlsbad, CA, USA). The ECM 600 Electro- ing to the manufacturer’s instructions. CD34+ cell purities poration System (an exponential decay electroporator) were more than 90% as assessed by flow cytometry after was from Genetronics (San Diego, CA, USA). Ca2+ and selection.

Bone Marrow Transplantation Human CD34+ cell culture optimization for efficient electroporation MH Wu et al 1203 Culture conditions for CD34+ cells minimum of 50000 events was acquired on a FACScan flow cytometer and analyzed using CellQuest. CD34+ cells were cultured in tissue culture flasks at 5 × 105 cells/ml in X-VIVO 10 serum-free cell culture medium supplemented with 2 mm of fresh l-glutamine. The culture Electroporation medium was supplemented with various combinations of Electroporation efficiency determination was performed as recombinant human growth factors including: 20 ng/ml IL- previously described.6 Immediately after electroporation, 3, 40 ng/ml IL-6, 50 ng/ml Flt-3L, 20 ng/ml stem cell factor + the CD34 cells were resuspended in cell culture medium (SCF), and 50 ng/ml TPO. Some cultures received 20 ng/ml containing TPO/SCF/Flt-3L with or without IL-3, PIXY 321 (IL-3/GM-CSF fusion protein) or 10% pooled PIXY321, CM, and serum in 24-well plates at 1 × 106 human AB serum. Additional cultures were supplemented cells/well. The cells were incubated at 37°C and 5% CO with 20% stroma-conditioned medium (CM) prepared from 2 for an additional 48 h prior to transfection efficiency and confluent monolayers of allogeneic bone marrow. Stroma- viability analyses by flow cytometry. conditioned medium was prepared from bone marrow mononuclear cells isolated (from 5 to 10 ml) over a Ficoll gradient and plated in 75 cm2 flasks supplemented with 20 Colony-forming unit (CFU) assay mls of MyeloCult H5100. Confluent stromal monolayers + were formed after 1 month of culture. CM was prepared by CD34 cells after 0, 1, 2, 3 and 4 days in culture were removing the stroma culture medium (containing serum), plated in CFU assays. Each time point was evaluated in rinsing once with D-PBS, and replacing it with X-VIVO duplicate by plating 5000 cells per 35 mm dish in 1 ml of MethoCult H4433 (PHA/LCM with erythropoietin). The 10 serum-free culture medium. The flasks were incubated ° for an additional 7 days. CM was removed and filtered CFU assays were scored after 14 days incubation at 37 C, through a 0.2 ␮m filter and supplemented into fresh X- 5% CO2 for colony-forming unit granulocyte–macrophage (CFU-GM), burst-forming unit erythroid (BFU-E), and VIVO 10 at a final concentration of 20% (v/v) containing + recombinant growth factors. mixed (CFU-Mix) colonies. Electroporated CD34 cells, after 0, 1, 2, 3 and 4 days in culture, and 48 hours post electroporation were plated in CFU assays. The EGFP- CD34 mean fluorescence intensity and subset analysis positive colonies were scored on a Nikon inverted fluor-

+ + escent microscope after 14 days in culture. The frequency Freshly isolated CD34 cells and cultured CD34 cells were of EGFP colonies is expressed as the number of EGFP- analyzed for CD34 mean fluorescence intensity (MFI) on positive colonies relative to the total number of colonies. day 0, 1, 2, 3 and 4 on a FACScan or FACScalibur flow cytometer. The cells were washed 1 × in PBS supplemented with 0.5% BSA and then incubated with 20 ␮l of HPCA- Long-term culture-initiating cell (LTC-IC) assay 2 monoclonal antibody (anti-CD34 PE) for 30 min on ice. + CD34 cells after 0, 1, 2, 3 and 4 days in culture were After incubation, the cells were washed 1 × and resus- plated in LTC-IC assays at 1 × 105 cells per well on pre- pended in 0.5–1.0 ml of PBS supplemented with 0.5% formed irradiated (25 Gy) allogeneic bone marrow stroma BSA. Propidium iodide was added at 10 ␮g/ml to each tube + in 24-well plates. Non-electroporated cells were plated at containing CD34 cells immediately prior to flow acqui- + 2 × 104 CD34 cells per well. Each well contained 2 ml of sition to exclude non-viable cells during analysis. The MFI MyeloCult H5100 supplemented with 0.2 ng/ml of hydro- of CD34+ PE cells (FL2) was determined for gated viable cortisone. The plates were incubated at 37°C, 5% CO for cells (propidium iodide negative) on a dot plot displaying 2 5–6 weeks. At weekly intervals, half of the culture medium FL1 (x-axis) vs FL2 (y-axis). CD34 subset analysis was was replaced with fresh culture medium. At the end of the done on days 0, 2, and 4 of culture. CD34 subset labeling culture period, the supernatant was removed, and cells were was done using anti-CD34 APC in combination with CD38, trypsinized by rinsing with D-PBS prior to adding 200 ␮l HLA-DR, c-kit, or Thy-1 PE, while gating on viable cells. of 0.05% trypsin/0.53 mm EDTA solution. After 10–15 min of incubation, the trypsinized cells were combined with the Cell cycle analysis using propidium iodide culture supernatant and centrifuged at 250 g for 10 min.

+ + The total cells from each well were then plated into CFU Freshly isolated CD34 cells and cultured CD34 cells were assays as previously described and scored after 14 days. analyzed for stages of cell cycling (G0/G1, S-phase, and EGFP-positive colonies and cloning efﬁciencies were G2/M) with propidium iodide on days 0, 1, 2, 3 and 4 using determined as described previously for the CFU assay. a FACScan ﬂow cytometer. Approximately 5 × 105 cells were centrifuged at high speed for 5–10 min. The pellet was resuspended in 1 ml of hypotonic propidium iodide Results solution prepared in 100 ml sterile water containing 0.1 g Na citrate, 5 mg propidium iodide, 0.1 ml Triton X-100 Effect of culture conditions on CD34 MFI, CFU and and 10 mg boiled RNAse A. The cells were incubated on LTC-IC ice for 20–60 min until analysis. The G0/G1 peak was set at channel 400 (FL-2 A) and on channel 200 (FL-3 H) as The CD34 MFI and CFU frequency after a 3 day culture displayed on a histogram. FL-2 A (linear) was used for the period are shown in Figure 1. Cultures supplemented with analysis and cell cycle determination using ModFIT-LT. A TPO/SCF/Flt-3L maintained a higher frequency for both

Bone Marrow Transplantation Human CD34+ cell culture optimization for efﬁcient electroporation MH Wu et al 1204 CFU per 104 cells 104 Day 0 50 100 150 200 250 300 103 Day 0 102

TPO,SCF,Flt-3L CD34 PE 101 IL-3,TPO,SCF,Flt-3L

100 IL-3,SCF,Flt-3L,IL-6 100 101 102 103 104 FL1-Height IL-3,SCF,Flt-3L

PIXY321,TPO,SCF

4 PIXY321,SCF,Flt-3L 10 Day 3 3GF PIXY321,SCF 103 CFU PIXY321 MFI 102

50 100 150 200 250 300 CD34 PE 101 MFI of CD34+ cells

Figure 1 Mean ﬂuorescence intensity (MFI) and colony-forming unit 100 0 1 2 3 4 (CFU) frequency of CD34+ cells cultured for 4 days in serum-free medium 10 10 10 10 10 supplemented with TPO/SCF/Flt-3L, or various recombinant growth factor FL1-Height combinations, as depicted. The MFI was determined from the geometric mean of ﬂuorescent cells that stained positive for CD34-PE (HPCA-2) on days 0 and 4. CFU per 10000 cells were scored from colony assays set up on days 0 and 4 of culture. Each data point represents the average of 104 104 two independent experiments. Day 3 Day 3 3 GF 3 GF 103 +PIXY321 103 +Serum

102 102 CD34 PE

CD34 MFI and CFU compared to those with other recombi- CD34 PE nant cytokine combinations, including IL-3, IL-6 and 101 101 PIXY321. This is clearly illustrated in a dot plot shown in Figure 2, in which CD34+ cells cultured in TPO/SCF/Flt- 100 100 0 1 2 3 4 3L sustained the highest CD34 brightness, whereas the 100 101 102 103 104 10 10 10 10 10 FL1-Height addition of PIXY321 or serum caused a dramatic decline FL1-Height in CD34 brightness during the same culture period. In Figure 2 Representative dot plots showing CD34+ cells cultured in the addition, the colonies derived from TPO/SCF/Flt-3L cul- presence of TPO/SCF/Flt-3L alone (3GF), with PIXY321, or with 10% tures consisted of all types of colonies including CFU-GM, human AB serum for 3 days. BFU-E, and CFU-Mixed, whereas colonies from those with additional serum or PIXY321 contained predominantly CFU-GM (data not shown). Cell cycle analysis of cultured CD34+ cells The number of LTC-IC (long-term culture-initiating cells) derived from cultures containing TPO/SCF/Flt-3L Using hypotonic propidium iodide, we observed 27 Ϯ 1.9% was superior to those with additional serum or PIXY321 CD34+ cells in S-phase after 2 days in cultures sup- (Figure 3). The LTC-IC frequency per 20000 cells peaked plemented with TPO/SCF/Flt-3L (Figure 4b). The percent- at day 2 in cultures supplemented with TPO/SCF/Flt-3L age of S-phase cells was signiﬁcantly increased by the and was higher than at day 0. In particular, the LTC-IC addition of PIXY321 (41.1 Ϯ 4.4%, P р 0.01) after 2 days frequency from the TPO/SCF/Flt-3L culture was 5-fold in culture. However, PIXY321 reduced CD34 brightness higher than at day 0, whereas those from cultures with and resulted in a lower number of LTC-IC at day 2 as additional PIXY321 or serum were only 2-fold higher than shown in Figures 2 and 3. The higher percentage of S-phase at day 0. Similar to our observed CFU results, the LTC-IC cells in the presence of PIXY321 was accompanied by less colonies from the TPO/SCF/Flt-3L culture consisted of two cells in G0 (Figure 4a, P р 0.01). No change in the percent types of colonies (CFU-GM and BFU-E) whereas those of cells in G0–G1, S-phase and G2–M was observed when supplemented with PIXY321 or serum were all CFU-GM serum was supplemented into cultures with TPO/SCF/Flt- colonies (data not shown). As shown in Figure 3 (inset), 3L (Figure 4). There was little or no increase in total cell the CFU per 10000 cells peaked at days 1 and 2, then numbers during the 4-day culture period using dropped to the levels similar to non-stimulated cells at days TPO/SCF/Flt-3L, yet a slight increase in total cell numbers 3 and 4. was observed when serum and PIXY321 was added.

Bone Marrow Transplantation Human CD34+ cell culture optimization for efﬁcient electroporation MH Wu et al 1205 500 100 a 80 TPO/SCF/Flt-3 400 90 +Serum 70 300 CFU +PIXY321 80 200 60 70 cells 4 01 234 50 Days in culture 10 60 ¥

40 % Cells in G0/G1 50 TPO/SCF/Flt-3L +Serum 30 40 +PIXY321 20 LTC-IC per 2 LTC-IC b 10 50

0 1 2 3 4 40

Days in culture 30 Figure 3 LTC-IC and CFU frequency of CD34+ cells per 20000 cells. CD34+ cells were cultured in serum-free cell culture medium sup- 20 plemented with TPO/SCF/Flt-3L (circles), with 10% human AB serum % Cells in S-phase (triangles), or with PIXY321 (squares) for 4 days. LTC-IC assays were 10 setup on days 0, 1, 2, 3 and 4 of culture and scored after 5 weeks on irradiated allogeneic bone marrow stroma. Inset: CFU per 10000 cells. CFU assays were set up on days 0, 1, 2, 3 and 4 of culture with TPO/SCF/Flt-3L and scored after 2 weeks. Each data point represents the 10 c mean (Ϯ s.e.m.) of three independent experiments. 8 Effect of TPO/SCF/Flt-3L on CD34+ subsets 6 To determine if a short-term culture period (4 days) in the presence of TPO/SCF/Flt-3L could sustain or expand the + 4

more primitive CD34 cell subsets, we examined the % Cells in G2/M expression of CD38, Thy-1, c-kit, and HLA-DR by flow cytometry. Figure 5 shows representative profiles for uncul- 2 tured (day 0) and cultured (day 4) CD34+ cells. These results indicate that, at least by phenotype, TPO/SCF/Flt- 012 34 3L exerts a differential effect on these subsets. The fre- + quency of Thy-1 cells after 4 days appears to be decreased Days in culture whereas the CD38Ϫ, c-kitlow and HLA-DRϪ subpopulations Figure 4 CD34+ cell cycle analysis by hypotonic propidium iodide. have expanded. The apparent selective expansion of these + CD34 cells were cultured in X-VIVO 10 supplemented with primitive subsets is at least partially reflected in the TPO/SCF/Flt-3L alone (circles), plus 10% human AB serum (triangles), increased frequency of LTC-IC after a 4-day culture period or plus PIXY321 (squares) for 4 days. Cell cycle status was evaluated at in the presence of TPO/SCF/Flt-3L. These results also sug- time points indicated. Shown are the percentages of CD34+ cells in G0/G1 gest that this cytokine combination can induce cell cycling (a), S-phase (b), or G2/M (c). Each data point represents the mean (Ϯ and therefore stable integration of exogenous DNA in the s.e.m.) of three independent experiments. more primitive CD34+ cell subsets attributed to long-term hematopoietic reconstitution. day 3 may be due, in part, to differentiation as suggested by the decrease in CD34 MFI (Figure 7). The decline of 6 Transfection of CD34+ cells by electroporation CD34 MFI may also explain our previous observation that the transient EGFP+/CD34+ peaked at day 2 while the %S- We have previously reported a correlation between elec- phase was maintained at Ͼ25% for up to 4 days. troporation efficiency and %S-phase CD34+ cells,6 showing The electroporation efficiencies for cells cultured under that 24–48 h incubation of CD34+ cells in X-VIVO 10 sup- different conditions for CD34, CFU, and LTC-IC are sum- plemented with TPO/SCF/Flt-3L resulted in 28% S-phase marized in Table 1. With TPO/SCF/Flt-3L alone, the trans- and 20% EGFP/CD34+ cells. A representative dot plot fection efficiencies per 100000 cells as detected by EGFP showing CD34+ and EGFP expression 2 days post electro- expression was approximately 20000 EGFP+ CD34+ cells poration (4 days total in culture) is shown in Figure 6. In (48 h post electroporation), 1800 EGFP+ CFU (16 days post this study, we extended our investigation on these cells by electroporation), and 40 EGFP+ LTC-IC (5 weeks post evaluating LTC-IC (Figure 7). The peak %EGFP+/LTC-IC electroporation). This indicates a high transient expression was observed when CD34+ cells were electroporated of %EGFP (20 Ϯ 1%) with long-term transfection as indi- between days 1 and 2 of culture. The LTC-IC decline after cated by LTC-IC CD34+ cells (16 Ϯ 7%). We observed

Bone Marrow Transplantation Human CD34+ cell culture optimization for efﬁcient electroporation MH Wu et al 1206 Day 0 Day 4 a

104 2%

103 CD38

102 CD34 PE

101

100 100 101 102 103 104

Thy-1 EGFP

104 22%

103 c-kit 102 CD34 PE

101

100 100 101 102 103 104 EGFP

+ HLA-DR Figure 6 EGFP expression in CD34 cells 48 h post electroporation (550 V/cm, 38 ms, 30 ␮g/500 ␮l DNA). CD34+ cells were cultured for 48 h in serum-free culture medium supplemented with TPO/SCF/Flt-3L prior to electroporation. The electroporated cells were cultured for an additional 48 h before analysis by flow cytometry. Shown are the expression of EGFP CD34 (FL1-Height) on the x-axis and CD34 (PE) on the y-axis for CD34+ cells electroporated with pcDNA3 as a control (a) or with pEGFP-N1 (b). Figure 5 Analysis of CD34+ cell subsets at day 0 and after 4 days of culture in serum-free culture medium supplemented with TPO/SCF/Flt- 3L. Shown are the dot plots for the expression of CD34 vs CD38, Thy- + 1, c-kit and HLA-DR. an enhancement of EGFP CD34 transient expression while stroma-conditioned medium from donor 2 did not. As shown, the frequency of EGFP+/CFU and EGFP+/LTC-IC that 56 Ϯ 15% of the CFU were EGFP+ indicating that the were not significantly enhanced by the addition of IL-3 or committed progenitor cells were also transfected short term. stroma-conditioned medium. Considering total cell output, for cultures supplemented with TPO/SCF/Flt-3L, the transfection efficiency was 20%, 1.8%, and 0.04% for EGFP+CD34+ cells, CFU and LTC- Discussion IC, respectively. We attempted to further enhance the electroporation Short-term culture of CD34+ stem cells could provide a efficiency by adding IL-3 or stroma-conditioned medium means for gene modification of hematopoietic cell precur- to TPO/SCF/Flt-3L. The electroporation results in Table 1 sors. However, there is a critical demand in maintaining show that IL-3 enhanced the EGFP+/CD34+ cells from 20 the stem cell pluripotency, which is essential for stable Ϯ 1% to 41 Ϯ 2%, however, the percentage of EGFP+/CFU long-term gene expression in hematopoietic lineages. In and EGFP+/LTC-IC did not increase. The conditioned this study, we demonstrated that CD34+ cells cultured in medium from two different donors gave inconsistent results serum-free medium supplemented with TPO/SCF/Flt-3L with stroma-conditioned medium from donor 1 providing for 2–4 days sustain a small loss in CD34 brightness and

Bone Marrow Transplantation Human CD34+ cell culture optimization for efﬁcient electroporation MH Wu et al 1207 80 800 genic progenitor cells but not the more primitive un- % EGFP+/LTC-IC 70 700 CD34 MFI committed LTC-IC cells. Moreover, our results suggest that 60 600 TPO/SCF/Flt-3L alone are sufﬁcient for maintaining and expanding the more primitive CD34+ cell subsets (CD38Ϫ, 500 50 Ϫ low

/LTC-IC HLA-DR and c-kit cells). + 40 400 The overall effects of TPO/SCF/Flt-3L on CD34+ cells 30 300 CD34 MFI are apparently combinatorial, although each of the compo- % EGFP 20 200 nents plays its own role, yet to be defined. TPO has been 10 100 shown to maintain primitive cell survival and to suppress apoptosis.44 These effects are augmented when used in 01234 combination with other early-acting cytokines such as SCF, Days in culture Flt-3L and/or c-kit ligand, leading to more effective expansion of primitive cells ex vivo.39,43,45,46 Flt-3L has a direct Figure 7 % EGFP+/LTC-IC and MFI derived from CD34+ cells electro- effect on hematopoietic cells, is synergistic with other porated at day 0 to day 4 in culture with TPO/SCF/Flt-3L. MFI was growth factors, and the addition of Flt-3L in cultures has determined at the time indicated. LTC-IC assays were set up 2 days after 25,47–49 + been shown to increase the yield of primitive cells. electroporation. The percent of EGFP /LTC-IC colonies is relative to total + Ϫ number of colonies × 100. Each data point represents the mean of three Flt-3L was also shown to recruit more CD34 /CD38 cells independent experiments (Ϯ s.e.m.). into division and enhance retroviral transduction of the expanded cells.41 SCF has been demonstrated to promote the viability of progenitor cells in culture and may promote exhibit a 5-fold higher frequency of LTC-IC compared to the cycling of G0-CD34+ cells.22,41,50,51 Goff et al40 was day 0. We also provided evidence for expansion of the + Ϫ low able to identify candidate combinations of cytokines that more primitive CD34 cells (CD38 , c-kit and HLA- Ϫ Ϫ promote maintenance of lin cells by TPO, or recruit primi- DR ) in this cytokine combination. Cell cycle analysis of + tive cells to divide and phenotypic self-renewal by TPO/ CD34 cells, as evaluated by flow cytometry, revealed that Flt-3L and TPO/SCF. Piacibello et al52 demonstrated exten- %S-phase cells increased from 3% at day 0 to 28% at day + 2. We observed in this study that the %EGFP+/LTC-IC sive self-renewal of LTC-IC cord blood CD34 cells using × 6 peaked at day 2, suggesting that the transient expression TPO/Flt-3L. CFU-GM were expanded 2 10 -fold over may be a good predictor of long-term transfection stability. the initial number after 6 months in culture. In their studies, The percentage of EGFP+/CD34+ cells and EGFP+/LTC-IC the addition of SCF reduced the number of LTC-IC while declined at days 3 and 4 of culture although the percent of the addition of IL-3 completely depleted LTC-IC after only cells in S-phase did not. These data suggest that S-phase 2 weeks. Our studies show the addition of IL-3 had no may not be critical for transfection as measured by LTC-IC. effect on LTC-IC cells. Other recombinant cytokine combinations (including High efficiency EGFP reporter gene transfection by elec- IL-3) resulted in a greater loss of CD34 MFI and frequency troporation has recently been reported with bone marrow + of CFU after a 4-day culture period. IL-3 is a well known CD34 cells15 and with dendritic cells (DC) derived from + stimulatory and survival factor for early hematopoietic stem cultured CD34 cells.14 In comparison, CD34-derived DC cell.25,42,43 However, the addition of IL-3 to cultures con- had a 16% transfection efficiency in contrast to monocyte- taining TPO/SCF/Flt-3L, while useful for enhancing the derived DC (Ͻ2%). Murray et al46 used retroviral transduc- transient expression of EGFP+ CD34+ cells, did not enhance tion and cycling CD34+ Thy-1+ cells to predict a 6-fold the percentage of EGFP+/LTC-IC, compared to TPO/ improvement in the transduction efficiency with a short- SCF/Flt-3L alone. Our results suggest that IL-3 can only term culture using TPO/SCF/Flt-3L compared to IL-3, enhance transient EGFP expression in the committed clono- IL-6, and leukemia inhibitory factor (LIF). It has been

Table 1 CD34+ cell transfection efﬁciency using EGFP

Culture conditions CD34 MFI EGFP+ CD34+ EGFP+ CFU EGFP+ LTC-IC

TOPO/SCF/Flt-3 alone 445 ± 36 20770 ± 1450 1800 ± 200 39 ± 12 (20 ± 1%) (56 ± 15%) (16 ± 7%) Plus stroma-conditioned medium from donor 1 414 ± 46 28970 ± 2990 1300 ± 200 37 ± 15 (29 ± 3%) (57 ± 20%) (14 ± 3%) Plus stroma-conditioned medium from donor 2 396 ± 24 14220 ± 5040 1900 ± 600 41 ± 9 (14 ± 5%) (60 ± 17%) (15 ± 5%) Plus IL-3 418 ± 50 41100 ± 2250 1900 ± 300 44 ± 9 (41 ± 2%) (62 ± 14%) (14 ± 4%)

Data in parentheses indicate % EGFP+ CD34+ cells, CFU and LTC-IC. Mean (± s.e.m.) of three experiments showing MFI and EGFP expression in CD34+ cells, CFU and LTC-IC per 105 cells.CD34+ cells were cultured with TPO/SCF/Flt-3 alone, plus stroma-conditioned medium (donor 1 or 2), or IL-3 for 48 h before and after electroporation. Shown are the CD34 MFI and EGFP+ CD34 cells (2 days post electroporation), EGFP+/CFU (2 weeks) and EGFP+/LTC-IC (7 weeks).

Bone Marrow Transplantation Human CD34+ cell culture optimization for efficient electroporation MH Wu et al 1208 reported by several investigators that CD34bright cells are 12 Takahashi M, Furukawa T, Tanaka I et al. Gene introduction responsible for more long-term stable hematopoiesis while into granulocyte–macrophage progenitor cells by electropor- the CD34dim cells are more differentiated and less likely ation: the relationship between introduction efficiency and the to provide long-term stable hematopoiesis.53–56 It has been proportion of cells in S-phase. Leukemia Res 1992; 16: shown that short-term culture periods (2–6 days) in serum- 761–767. 13 Yorifuji T, Tsuruta S, Mikawa H. The effect of cell synchron- free or serum-reduced conditions can increase the SCID- ization on the efficiency of stable gene transfer by electropor- repopulating cell frequency and enhance the efficiency of ation. FEBS Lett 1989; 245: 201–203. EGFP reporter genes in repopulating NOD/SCID 14 Van Tendeloo VF, Snoeck HW, Lardon F et al. Nonviral mice.16,57,58 Considering our results, we conclude that short- transfection of distinct types of human dendritic cells: high- term culture of CD34+ cells with TPO/SCF/Flt-3L alone efficiency gene transfer by electroporation into hematopoietic can increase the percentage of cycling CD34+ cells progenitor- but not monocyte-derived dendritic cells. Gene resulting in enhanced expression of EGFP reporter genes Therapy 1998; 5: 700–707. by electroporation. 15 Van Tendeloo VF, Willems R, Ponsaerts P et al. High-level transgene expression in primary human T lymphocytes and adult bone marrow CD34+ cells via electroporation-mediated gene delivery. Gene Therapy 2000; 7: 1431–1437. Acknowledgements 16 Bhatia M, Bonnet D, Kapp U et al. Quantitative analysis reveals expansion of human hematopoietic repopulating cells This work was supported by United States Public Health Service after short-term ex vivo culture. J Exp Med 1997; 186: 619– grants CA57725 (MED) and CA71627 (MED), and the breast can- 624. cer planning grant awarded to University of Chicago Cancer 17 Tisdale JF, Hanazono Y, Sellers SE et al. Ex vivo expansion Center CA661324 (MED). We would like to thank Susan Barker of genetically marked rhesus peripheral blood progenitor cells and Kahadijah James from the Apheresis Facility and Kristi Hol- results in diminished long-term repopulating ability. Blood lingsworth and Janine Kobar from the Stem Cell Cryolab at the 1998; 92: 1131–1141. University of Chicago for providing the apheresis blood products 18 Scheding S, Kratz-Albers K, Meister B et al. Ex vivo expan- used in these studies. sion of hematopoietic progenitor cells for clinical use. Semin Hematol 1998; 35: 232–240. 19 von Kalle C, Glimm H, Schulz G et al. New developments in References hematopoietic stem cell expansion. Curr Opin Hematol 1998; 5: 79–86. 1 Bertolini F, Corsini C, Lazzari L et al. Gene transfer-mediated 20 Brugger W, Kanz L. Ex vivo expansion of hematopoietic pre- generation of drug-resistant hemopoiesis. Leuk Lymphoma cursor cells. Curr Opin Hematol 1996; 3: 235–240. 1996; 21: 17–23. 21 Ogawa M. Differentiation and proliferation of hematopoietic 2 Srivastava A. Parvovirus-based vectors for human gene ther- stem cells. Blood 1993; 81: 2844–2853. apy. Blood Cells 1994; 20: 531–536; discussion 536–538. 22 Gothot A, Pyatt R, McMahel J et al. Assessment of proliferat- 3 Fruehauf S, Boesen JJ, Breems DA et al. Retroviral transfer ive and colony-forming capacity after successive in vitro of the multidrug resistance-1 gene into lineage-committed and divisions of single human CD34+ cells initially isolated in G0. primitive hemopoietic cells. Stem Cells 1995; 13 (Suppl. 3): Exp Hematol 1998; 26: 562–570. 93–99. 23 David S, Biancone L, Caserta C et al. Alternative pathway 4 Baum C, Eckert HG, Stockschlader M et al. Improved retrovi- complement activation induces proinflammatory activity in ral vectors for hematopoietic stem cell protection and in vivo human proximal tubular epithelial cells. Nephrol Dial Trans- selection. J Hematother 1996; 5: 323–329. plant 1997; 12: 51–56. 5 Frey BM, Hackett NR, Bergelson JM et al. High-efficiency 24 Birkmann J, Oez S, Smetak M et al. Effects of recombinant gene transfer into ex vivo expanded human hematopoietic pro- human thrombopoietin alone and in combination with erythro- genitors and precursor cells by adenovirus vectors. Blood poietin and early-acting cytokines on human mobilized pur- 1998; 91: 2781–2792. ified CD34+ progenitor cells cultured in serum-depleted 6 Wu MH, Liebowitz DN, Smith SL et al. Efficient expression medium. Stem Cells 1997; 15: 18–32. of foreign genes in human CD34+ hematopietic precursor cells 25 Petzer AL, Zandstra PW, Piret JM et al. Differential cytokine using electroporation. Gene Therapy 2001; 8: 384–390. effects on primitive (CD34+CD38−) human hematopoietic 7 Toneguzzo F, Hayday AC, Keating A. Electric field-mediated cells: novel responses to Flt3-ligand and thrombopoietin. J DNA transfer: transient and stable gene expression in human Exp Med 1996; 183: 2551–2558. and mouse lymphoid cells. Mol Cell Biol 1986; 6: 703–706. 26 Brugger W, Mocklin W, Heimfeld S et al. Ex vivo expansion 8 Toneguzzo F, Keating A. Stable expression of selectable genes of enriched peripheral blood CD34+ progenitor cells by stem introduced into human hematopoietic stem cells by electric cell factor, interleukin-1 beta (IL-1 beta), IL-6, IL-3, inter- field-mediated DNA transfer. Proc Natl Acad Sci USA 1986; feron-gamma, and erythropoietin. Blood 1993; 81: 2579–2584. 83: 3496–3499. 27 McKenna HJ, de Vries P, Brasel K et al. Effect of flt3 ligand 9 Toneguzzo F, Keating A, Glynn S et al. Electric field- on the ex vivo expansion of human CD34+ hematopoietic pro- mediated gene transfer: characterization of DNA transfer and genitor cells. Blood 1995; 86: 3413–3420. patterns of integration in lymphoid cells. Nucleic Acids Res 28 Sui X, Tsuji K, Tanaka R et al. gp130 and c-Kit signalings 1988; 16: 5515–5532. synergize for ex vivo expansion of human primitive hemopo- 10 Keating A, Toneguzzo F. Gene transfer by electroporation: a ietic progenitor cells. Proc Natl Acad Sci USA 1995; 92: model for gene therapy. Prog Clin Biol Res 1990; 333: 2859–2863. 491–498. 29 Bertolini F, Soligo D, Lazzari L et al. The effect of interleu- 11 Takahashi M, Furukawa T, Nikkuni K et al. Efficient introduc- kin-12 in ex-vivo expansion of human haemopoietic progeni- tion of a gene into hematopoietic cells in S-phase by electro- tors. Br J Haematol 1995; 90: 935–938. poration. Exp Hematol 1991; 19: 343–346. 30 Garbe A, Spyridonidis A, Mobest D et al. Transforming

Bone Marrow Transplantation Human CD34+ cell culture optimization for efficient electroporation MH Wu et al 1209 growth factor-beta 1 delays formation of granulocyte–macro- cytokines to directly promote survival and suppress apoptosis phage colony-forming cells, but spares more primitive pro- of human primitive CD34+CD38− bone marrow cells with genitors during ex vivo expansion of CD34+ haemopoietic multilineage potential at the single-cell level: key role of progenitor cells. Br J Haematol 1997; 99: 951–958. thrombopoietin. Blood 1997; 90: 2282–2292. 31 Kobari L, Giarratana MC, Poloni A et al. Flt 3 ligand, MGDF, 45 Kobayashi M, Laver JH, Kato T et al. Thrombopoietin sup- Epo and G-CSF enhance ex vivo expansion of hematopoietic ports proliferation of human primitive hematopoietic cells in cell compartments in the presence of SCF, IL-3 and IL-6. synergy with steel factor and/or interleukin-3. Blood 1996; 88: Bone Marrow Transplant 1998; 21: 759–767. 429–436. 32 Ihle JN. Interleukin-3 and hematopoiesis. Chem Immunol 46 Murray LJ, Young JC, Osborne LJ et al. Thrombopoietin, flt3, 1992; 51: 65–106. and kit ligands together suppress apoptosis of human mobil- 33 Gabbianelli M, Pelosi E, Montesoro E et al. Multi-level effects ized CD34+ cells and recruit primitive CD34+ Thy-1+ cells of flt3 ligand on human hematopoiesis: expansion of putative into rapid division. Exp Hematol 1999; 27: 1019–1028. stem cells and proliferation of granulomonocytic 47 Dooley DC, Xiao M, Oppenlander BK et al. Flt3 ligand progenitors/monocytic precursors. Blood 1995; 86: 1661– enhances the yield of primitive cells after ex vivo cultivation 1670. of CD34+ CD38dim cells and CD34+ CD38dim CD33dim 34 Young JC, Varma A, DiGiusto D et al. Retention of quiescent HLA-DR+ cells. Blood 1997; 90: 3903–3913. hematopoietic cells with high proliferative potential during ex 48 Petzer AL, Hogge DE, Landsdorp PM et al. Self-renewal of vivo stem cell culture. Blood 1996; 87: 545–556. primitive human hematopoietic cells (long-term culture-initiat- 35 Dao MA, Hashino K, Kato I et al. Adhesion to fibronectin ing cells) in vitro and their expansion in defined medium. Proc maintains regenerative capacity during ex vivo culture and Natl Acad Sci USA 1996; 93: 1470–1474. transduction of human hematopoietic stem and progenitor 49 Zandstra PW, Conneally E, Petzer AL et al. Cytokine manipu- cells. Blood 1998; 92: 4612–4621. lation of primitive human hematopoietic cell self-renewal. 36 Breems DA, Blokland EA, Siebel KE et al. Stroma-contact Proc Natl Acad Sci USA 1997; 94: 4698–4703. prevents loss of hematopoietic stem cell quality during ex vivo 50 Li CL, Johnson GR. Stem cell factor enhances the survival expansion of CD34+ mobilized peripheral blood stem cells. but not the self-renewal of murine hematopoietic long-term Blood 1998; 91: 111–117. repopulating cells. Blood 1994; 84: 408–414. 37 Bhatia R, McGlave PB, Miller JS et al. A clinically suitable 51 Leary AG, Zeng HQ, Clark SC et al. Growth factor require- ex vivo expansion culture system for LTC-IC and CFC using ments for survival in G0 and entry into the cell cycle of primi- Proc Natl Acad Sci USA stroma-conditioned medium. Exp Hematol 1997; 25: 980–991. tive human hemopoietic progenitors. 1992; 89: 4013–4017. 38 Brandt JE, Galy AH, Luens KM et al. Bone marrow repopu- 52 Piacibello W, Sanavio F, Garetto L et al. Extensive amplifi- lation by human marrow stem cells after long-term expansion cation and self-renewal of human primitive hematopoietic culture on a porcine endothelial cell line. Exp Hematol 1998; stem cells from cord blood. Blood 1997; 89: 2644–2653. 26: 950–961. 53 Tjonnfjord GE, Steen R, Veiby OP et al. Lineage commitment 39 Ramsfjell V, Borge OJ, Cui L et al. Thrombopoietin directly of CD34+ human hematopoietic progenitor cells. Exp Hematol and potently stimulates multilineage growth and progenitor + − 1996; 24: 875–882. cell expansion from primitive (CD34 CD38 ) human bone 54 Tjonnfjord GE, Steen R, Evensen SA et al. Characterization marrow progenitor cells: distinct and key interactions with the of CD34+ peripheral blood cells from healthy adults mobilized ligands for c-kit and flt3, and inhibitory effects of TGF-beta by recombinant human granulocyte colony-stimulating factor. and TNF-alpha. J Immunol 1997; 158: 5169–5177. Blood 1994; 84: 2795–2801. 40 Goff JP, Shields DS, Greenberger JS. Influence of cytokines 55 Knudsen LM, Rasmussen T, Jensen L et al. Reduced bone on the growth kinetics and immunophenotype of daughter marrow stem cell pool and progenitor mobilisation in multiple + cells resulting from the first division of single CD34( )Thy- myeloma after melphalan treatment. Med Oncol 1999; 16: + − 1( )lin cells. Blood 1998; 92: 4098–4107. 245–254. 41 Haylock DN, Horsfall MJ, Dowse TL et al. Increased recruit- 56 Osawa M, Hanada K, Hamada H et al. Long-term ment of hematopoietic progenitor cells underlies the ex vivo lymphohematopoietic reconstitution by a single CD34− expansion potential of FLT3 ligand. Blood 1997; 90: 2260– low/negative hematopoietic stem cell. Science 1996; 273: 2272. 242–245. 42 Dao MA, Nolta JA. Inclusion of IL-3 during retrovirally- 57 Conneally E, Cashman J, Petzer A et al. Expansion in vitro mediated transduction on stromal support does not increase of transplantable human cord blood stem cells demonstrated the extent of gene transfer into long-term engrafting human using a quantitative assay of their lympho-myeloid repopulat- hematopoietic progenitors. Cytokines Cell Mol Ther 1997; 3: ing activity in nonobese diabetic-scid/scid mice. Proc Natl 81–89. Acad Sci USA 1997; 94: 9836–9841. 43 Ohmizono Y, Sakabe H, Kimura T et al. Thrombopoietin aug- 58 Dorrell C, Gan OI, Pereira DS et al. Expansion of human ments ex vivo expansion of human cord blood-derived hemato- cord blood CD34(+)CD38(−) cells in ex vivo culture during poietic progenitors in combination with stem cell factor and retroviral transduction without a corresponding increase in flt3 ligand. Leukemia 1997; 11: 524–530. SCID repopulating cell (SRC) frequency: dissociation of SRC 44 Borge OJ, Ramsfjell V, Cui L et al. Ability of early acting phenotype and function. Blood 2000; 95: 102–110.

Bone Marrow Transplantation