Lentivirus Transduction of Human Osteoclast Precursor Cells

Fig 2. GFP+ osteoclast. Lentivirus Transduction of Human Osteoclast Precursor Cells. Ramnaraine ML, + Clohisy DR, Mathews W +University of Minnesota, Minneapolis, MN. Senior author [email protected] INTRODUCTION Recent advances in understanding the pathophysiology of bone cancer have confirmed the pathologic role of osteoclasts [1]. In addition, the use of suicide gene delivery to tumors by accessory cells has been demonstrated in treatment for gliomas and lung cancer using neural [2] or endothelial precursor cells [3]. Osteoclasts form from recruited precursor cells in high numbers at sites of bone cancer and stimulate tumor progression. One step toward developing a novel gene therapy for treating bone metastases, is the engineering of osteoclast precursor cells (OcP) to serve as a cellular gene delivery system capable of killing cancer cells. A previous study has demonstrated that murine OcP can provide a basis for bone cancer-targeted suicide gene therapy [4]. This study represents a crucial step toward human therapy by demonstrating successful transduction of human OcP. METHODS OcP isolation. Human peripheral blood was purchased from Memorial Blood Center and the mononuclear cells (PBMC) isolated by density gradient centrifugation. CD14 cells were purified to >98%+ using Miltenyi MACS magnetic separation columns. Samples from frozen PBMC were primed by 3 days of culture in MEM α with 10% FBS and 25ng/ml rHuM-CSF (MØ media) before CD14 purification. Four different donors were used for each experiment with an average CD14+ recovery of 5% (range 2-8%) using fresh PBMC and 5% (range 1-9%) using frozen PBMC. Transduction of OcP. Transduction of 2 to 6 × 10 CD14+ cells /ml was performed in 6-well Corning ultra-low attachment plates (to prevent adherence) using MØ media. CSIIeG lentivirus, containing the eGFP marker (M.O.I = 10) was added to a total volume of 2 ml. Charged bridge, either 2.5μg/ml DEAE-dextran (d-d), 10μg/ml protamine sulfate or none, was also added. Cells were grown at 37°C in 5% CO2 for up to 96 hours. Aliquoits were removed for cell analysis, flow cytometry (FACS) and osteoclastogenesis at 24h intervals. Cell analysis and FACS. Transduction was evaluated for cell yield and viability using a ViCell counter. Cells were stained with CD14-PE and CD11b-APC to monitor OcP lineage. Analysis was performed using a FACScaliber cytometer for acquisition and FloJo software used for quantitation of percentage OcP and marker gene expression. Backgating on CD11b onto the forward/side scatter plot defined the OcP population of viable, single cells, denoted R1. In vitro osteoclastogenesis. Transduced cells were cultured in quadruplicate in 96-well plates at 5 × 10 cells/well with MØ media for a total of 4 days (including transduction time). Cultures were then switched to osteoclast media, MØ media containing 50 ng/ml rHuRANKL and incubated at 37°C in 5% CO2 for 14 days. Half-volume media changes to replace cytokines occurred every 3 days. Imaging of osteoclasts. Cultured osteoclasts were fixed and either stained with tartrate-resistant acid phosphatase (TRAP), a marker for osteoclasts or counterstained using DAPI in a slo-fade mounting media for fluorecent microscopy. Brightfield microscopy was used to visualize TRAP-stained osteoclasts. Confocal microscopy was used to visualize green (GFP+) osteoclasts. Statistics. Data is presented as means ± SD. Statistical significance was determined by Student’s t-test. A P value of less than 0.05 was considered statistically significant. RESULTS Yield, viability and OcP markers after transduction. The number of OcP decreased after transduction to 50% of the plating number after 24h and to 33% after 48h, where it held stable up to 96h. There were significant differences between either charged bridge or none at 24 and 48h. Viability ranged between 60-70% at 24h, dropped to 50-60% at 4872h and recovered to 60% by 96h. There were significant differences between protamine and none at 24-72h. The OcP marker, CD11b was virtually unchanged throughout the transduction period, with a low of 78% at 24h and high of 94% at 96h. CD14 in contrast decreased to 18% by 24h and did not recover throughout the transduction period, reaching a low of 12% at 96h. R1, the cells comprising the sub-population of viable OcP was 48% at 24h, 55% at 72 and 85% at 96h. There were no significant differences between either charged bridge or no charged bridge for CD11b, CD14 and R1. A charged bridge enhances OcP transduction. At every time point there were significantly more GFP+ (transduced) cells in both the total population and the OcP sub-population (R1) using a charged bridge compared to none. At 24h total GFP was 4.8 ±2.3%, 8.6 ±1.8% and 13.2 ±5.6% for no bridge, protamine and d-d, respectively. Similar percentages were measured in R1. At 96h the total GFP had increased to 19.7 ±2.2%, 25.8 ±3.9% and 26.9 ±5.9%, and GFP in R1 to 30.5 ±3.6%, 39.3 ±6.1% and 42.1 ±10%. Transduced OcP proliferate in vitro. Proliferation was determined by calculating the absolute number of GFP+ OcP using the following: #viable cells x %R1 x %GFP in R1. GFP+ OcP increased by 1.8-fold/day for all three conditions († p< 0.05), a total expansion from 24 to 96h of 6-fold. At every time point there were significantly more GFP+ OcP using d-d (Fig 1). Both fresh and frozen PBMC can be used to generate transduced OcP. A repeat experiment using frozen PBMC as the source of OcP demonstrated similar results; GFP, GFP in R1 and GFP+ OcP increased dramatically from 24 to 96h. Due to lower number of initial CD14+ cells, the N was 2 instead of 4, and showed trends, not significant differences. All three conditions showed a 6-fold increase in GFP+ OcP. OcP differentiate into gfp-expressing osteoclasts. Examination of day 14 osteoclast cultures showed abundant mono-, bi and multi-nucleated TRAP+ cells. An example of a GFP+ multi-nucleated (blue = DAPI) osteoclast is shown in Figure 2.