Cell Isolation and Culture
Institutional Review Board approval was obtained prior to collection of human tissue. Human adipose tissue was collected from the infrapatellar fat pad of healthy patients undergoing total knee arthroplasty (TKA) for osteoarthritis. The fat pad is partially excised and typically discarded in a TKA. The fat pad was processed using a previously published protocol to obtain the stromal vascular fraction (SVF).(5–7) Dulbecco’s PBS (DPBS, Lonza, Basel, CH) was used to extensively wash the adipose tissue to remove debris and red blood cells. The tissue was then mechanically digested with scissors and enzymatically digested with 0.1% collagenase type 1A (C2674, Sigma-Aldrich, St. Louis, MO) at 37° C for 120 minutes. The digested fat tissue solution was then filtered with a 100µm strainer and washed three times with DPBS. The SVF including human ASCs was then resuspended with Dulbecco’s modified eagle medium (DMEM; Corning, Corning, NY) which contained 10% fetal bovine serum (Gibco, Amarillo, TX) and an antibiotic mix containing 100 unit/mL of penicillin, 100 µg/mL streptomycin and 250 ng/mL Amphotericin B (Lonza, Basel, CH), and plated on a 100mm dish for culture.(7) The adherent ASCs were split weekly and culture media was changed every 3–4 days.
Viral Vector Generation and Transduction
ASCs were transduced with a TSTA vector system, as described in previously published protocols.(8–10) The TSTA system contains two lentiviral vectors including the GAL4-VP16 transactivator vector (LV-RhMLV- GAL4-VP16) and a transgene expression vector encoding BMP-2 or eGFP (LV-G5-BMP2 or LV-G5-eGFP. As previously described, lentiviral vectors were generated by transfecting 293 T cells (American Type Culture Collection, Manassas, VA).(10, 11) To quantify the titer of vectors, HT-29 cells (ATCC, Manassas, VA) were transduced with diluted vector prep and genomic DNA was extracted at 3 days after transduction. Human cell number and lentiviral vector copy number were quantified using QX200 droplet digital PCR system (Biorad, Herclus, CA) and HIV-1 psi region and human Syndecan-2 (SDC-4) primers and probes. LV titer was then calculated from vector copy number per cell.
At passage 3, ASCs were plated at a density of 1 x 106 cells/dish. The following day cells were co-transduced with two viral vectors (LV-RhMLV-GAL4-VP16 with LV-G5-BMP2 or -eGFP) at MOI of 3 each, then incubated overnight. The following morning the transduced cells were washed with DPBS to remove extracellular virus, and incubated another overnight in fresh media prior to implantation.
In Vivo Implantation of Transduced ASCs
Animal care and procedures were performed in accordance with the Institutional Animal Care and Use Committee guidelines after approval of the study protocol. 12-week-old male athymic rats (Charles River Laboratories, Wilmington, MA) were used in this study. A critical size femoral defect of 6 mm was created according to well established protocols.(8, 12, 13)
Briefly, the rats were anesthetized using inhalation anesthesia of 2–3% isoflurane in an induction chamber and then maintained with a nose cone. The left leg was shaved and prepped in a sterile fashion. The femur was exposed circumferentially and stripped of all soft tissue including periosteum. A polyethylene plate was then secured with four threaded wires (Zimmer Biomet, Warsaw, IN) and two stainless steel cerclage wires and a 6 mm defect was created with a burr. Lastly, the exposed medullary canals were flushed with copious normal saline.
Transduced cells were harvested, mixed with 30 µl of DPBS, and were then pipetted onto all surfaces of a compression resistant matrix (Medtronics Inc., Memphis, Tenn.) measuring 6 x 3 x 3 mm. The saturated scaffold was placed into the defect site. A 4 − 0 vicryl suture was used to close fascia and skin. Subcutaneous buprenorphine SR injection was given post-operatively for pain relief.
Study Group
There were 3 groups studied; LV-TSTA-BMP-2, LV-TSTA-GFP, and nontransduced cells. The two LV-TSTA groups were evaluated in two doses which included 5 million cells and 15 million cells. Five million cells transduced with BMP-2 has reliably healed a 6 mm critical size defect and is therefore considered the standard cell dose for gene therapy. The 15 million cell dose was chosen to simulate a supratherapeutic cell dose or “overdose” scenario, which could occur when bone grafting defects in patients. The study groups consisted of 5 rats per study group per time point: High-dose (HD) LV-TSTA-BMP-2 (15 million cells), standard-dose (SD) LV-TSTA-BMP-2 (5 million cells), HD LV-TSTA-GFP (15 million cells), SD LV-TSTA-GFP (5 million cells), and non-transduced (NT) ASCs (5 million cells). Each study group was evaluated at the following time points post implantation: 4, 14, 56, and 84 days.
Organ Harvest:
Animals were sacrificed at their respective timepoint from cell implantation. Deep anesthesia was provided with 3–4% isoflurane. The rats were positioned supine and a midline incision was used to expose the thoracic and abdominal cavity. A 22 gauge intravenous (IV) catheter was inserted into the left ventricle and 2–3 mL of blood was aspirated in a sterile syringe. Next, 240 cc of PBS was slowly infused into the IV catheter to perfuse all organs. A small nick in the left ventricle and right atria were created to allow for outflow. Once the perfusion was complete and the organs were pale, the following organs were harvested: heart, lung, liver, spleen, kidney, testis, defect site from operative femur, ipsilateral tibia, bone marrow from contralateral femur, brain, and GI tract (from mid-jejunum).
Pathology
Tissue specimens from all ten organs and the defect site in all study groups and all time points were evaluated by a board-certified pathologist (S.C.). All organ tissue (non-bone) specimens were fixed in 10% formalin for 24 hours at room temperature. The bone specimens (operative femurs and ipsilateral tibias) were placed in 10% formalin for 1 week at 4° C and decalcified in 10% EDTA solution for 2 months at room temperature with gentle mechanical stirring. Samples were then dehydrated in ethanol and embedded in paraffin. Paraffin blocks were then sectioned and hematoxylin and staining was performed following standard protocols.
LAM – PCR and Integration Site Analysis
Locations of vector integration sites are read out by Illumina paired end DNA sequencing and alignment to the human genome as described(14–19). Briefly, samples of animal DNA are cleaved by sonication, and DNA adaptors ligated onto the broken ends. PCR is then used to amplify from the viral DNA end to the adaptor. All samples are analyzed in quadruplicate independently to suppress founder effects in the PCR. Different adaptors are used for each sample to suppress PCR cross-over(20, 21). A well-developed software pipeline allows automated sequence work up, report generation, and comparison of new sequence profiles to user-defined datasets from earlier studies(14–21).
The SonicAbundance method was used to infer the abundance of cell clones from integration site data(15). Simply counting the number of sequences reads per integration site after PCR is known to yield distorted abundance estimates(22, 23). The SonicAbundance method takes advantage of the random fragmentation of DNA molecules by sonication prior to PCR amplification steps. In a DNA sample from cells containing integration sites, DNA from an expanded cell clone will be represented by many DNA chains. Random fragmentation by sonication followed by linker ligation results in many linkers joined at different positions near the integrated sequence from the expanded clone. PCR amplification and paired end sequencing results in recovery of many different sites of linker ligation. The number of these positions (adjusted to account for the probabilities of random overlaps) provides an abundance score based on a mark introduced into DNA prior to any PCR steps.
Genomic DNA Extraction
Hard tissue (healed defect site) was crushed into small fragments in a stainless-steel jar with a 20mm grinding ball using Retsch MM400 mixer (Verder Scientific, Newtown, PA), then the powdered bone was decalcified by 0.5M, pH8.0 EDTA (BP2482, Fisher Scientific, Waltham, MA) on a shaker at 37° C for 5 days. Organ tissue was disrupted in a 2.0mL centrifuge tube with a 5mm stainless steel bead using MM400 mixer prior to DNA extraction. Genomic DNA was extracted from decalcified bone pellet or disrupted tissue using Qiagen DNeasy Blood and Tissue kit according to manufacturer’s protocol (QUIAGEN, Valenica, CA). Genomic DNA concentration was measured by NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA).
Vector copy number quantification by Droplet Digital PCR (ddPCR)
Vector copy and human cell numbers in organ genomic DNA samples were quantified by with the QX200 droplet digital PCR system (Bio-Rad, Hercules, CA) as per the manufacturer’s instructions. PCR reaction mixtures were prepared consisting of 25 ml volumes containing 1x ddPCR supermix for probes (Bio-Rad), primers and probe specific to the HIV-1
Psi region, for detection of both vectors, and human Syndecan-4 (SDC-4) internal control gene (400 nM and 100 nM for primers and probe, respectively), DraI (40 U; New England Biolabs, Ipswich, MA), and 100ng of the genomic DNA sample (Supplemental Table 1).
Sample droplets were generated by the QX200 droplet generator (Bio-Rad), then transferred onto a 96-well PCR plate. PCR amplification of droplets was performed by a C1000 touch thermal cycler (Bio-Rad). Thermal cycling conditions consisted of 95° C 10 minutes, 94° C 30 seconds and 60° C 1 minute (55 cycles), 98° C 10 minutes (1 cycle), and 4° C hold. HIV Psi (vector copy) or SDC-4 positive droplets were counted by the QX200 droplet reader operated by manufacture’s QuantaSoft software. Human cell number in genomic DNA samples was calculated from SDC-4 copy number, which was known to be present in two copies in a diploid genome.
Copy number per cell was subsequently determined by dividing VCN by human cell number. Viral copy number and human cell number per 100 ng of were set at threshold of 0.9 copies/100ng genomic DNA and 0.45 cells/100ng genomic DNA to eliminate background detection of the droplet digital PCR. These values were selected based on data in the non-transduced samples.