Pulmozyme® Treatment of Vectors Produced by Transient Transfection Reduces Residual Plasmid DNA on Human CD34+ Hematopoietic Progenitor Cells without Loss of Transduction or Engraftment Efficiency

Abstract

A significant percentage of clinical grade safety-enhanced gamma-retrovirus and lentivirus vector supernatants are currently produced by transient transfection. These products contain high concentrations of plasmid DNA that pose the risk of transfer into subjects, and preclude accurate estimation of transduction efficiency. Most envelope pseudotyped vectors are too labile for DNAse treatment or other purification steps and adding non-clinical grade reagents into any production step further complicates product qualification. Here we present development of a method to remove residual plasmid as a part of the transduction procedure of human CD34+ cells or mouse bone marrow progenitors, using the DNAse Pulmozyme®. Using a standard clinical transduction protocol, CD34+ were transduced with SRS11.EFS.IL2RG.pre* or SRS11.EFS.EGFP. pre* gamma-retroviral vectors pseudotyped with GALV envelope or a mock control. Retronectin coated flasks preloaded with vector were either treated with Pulmozyme® or no Pulmozyme®. Cells were transduced on two consecutive days and collected 4 –24 hrs after the 2nd transduction (day1). CFU were plated and enumerated and proportion of transduced progenitors quantified by real time PCR (qPCR) using wPRE primers. For the EGFP containing vector, gene transfer was determined by flow cytometry in cultures that were continued for up to 14 days. DNA was extracted and tested by qPCR on days 1, 7, and 14. Residual plasmid DNA was quantified using GALV primers. Transduction was also estimated on the bulk cultures using qPCR. In all qPCR reactions human ApoB was quantified concurrently to asses the cell number. Pulmozyme® treatment reduced GALV copy numbers in cells one day following transduction by up to 20-fold. After 7 days in culture, there was no residual plasmid detectable in all groups. More importantly, Pulmozyme® treatment did not alter the transduction efficiency of the progenitors in any of the experiments, as evaluated by either wPRE qPCR on Day-7 DNA (4.3 vs. 4.2, 6.8 vs. 7.6 and 2.0 vs. 1.9 vector copies per cell for Pulmozyme® treated and non-treated respectively), colony PCR (69.4 vs. 66.7% vector positive CFUs) or flow cytometry (81.8 vs. 85.7% EGFP positive cells). We also analyzed the engraftment potential of cells transduced using Pulmozyme®-treated versus non-treated vector after transplanting cells in NOD/SCID and NOD/SCID-IL2Rg null (NOG) mice. Analysis of mice at 6 weeks post transplantation showed no significant reduction in human cell engraftment in the Pulmozyme® and the average copy number per human cell in NOG bone marrow was 0.65 vs. 0.52 and in NOD/SCID bone marrow 1.03 vs. 0.67 (Pulmozyme® treated versus non-treated, respectively). We also tested Pulmozyme® treatment in mouse experiments using gamma-retroviral vector pseudotyped with ecotropoc envelope with a similar result. We conclude that treatment of gamma-retroviral vector supernatant with Pulmozyme® after vector preload on Retronectin® coated plates reduces vector and packaging plasmids and does not inhibit transduction, clonogenic potential, or engraftment. The removal of excess plasmid from transiently produced virus supernatant without affecting the virus transducibility allows accurate estimation of transduction efficiencies prior to transplant and will conceivably reduce the toxicity to primary cells from excess plasmid. These studies can be directly translated to preclinical animal safety studies and clinical trials.