Optimization of transposon-mediated gene transfer into PBL. T3 transposon (pT3) was previously described²¹ and modified to contain the MSCV-U3 promoter, multiple cloning site and bovine growth hormone polyadenylation signal. The green fluorescent protein (GFP) reporter gene was inserted into the multiple cloning site. The Sleeping Beauty (SB) transposase was derived plasmid HSB5 as described,³⁸ directionally cloned into T7-promoter expression plasmid pGEM4Z/GFP/64A, replacing GFP to yield pT7-HSB5-64A. In vitro transcription of RNA was previously described,³⁹ and gene transfer of human peripheral blood mononuclear cell (PBMC) was performed using the nucleofection electroporation-procedure for human T cells as recommended by the supplier (using program U14; Amaxa Inc., Gaithersburg, MD, USA). Cells were activated next day in AIM-V medium (Invitrogen, Carlsbad, CA, USA) supplemented with 5% human serum, 50 ng ml^-1 OKT3 and 6000 IU ml^-1 interleukin-2 (IL2). (a) Either 10 μg or 20 μg HSB5 RNA was coelectroporated into PBMC with varying amounts of pT3-GFP transposon plasmid DNA and GFP expression measured at 10 days after gene transfer by fluorescence-activated cell sorting (FACS). (b) pT3-GFP plasmid DNA or HSB5 RNA plus pT3-GFP plasmid were coelectroporated into PBMC followed the next day by OKT3/IL2 stimulation. FACS was performed at the times indicated to measure GFP expression. (c) A fixed amount of pT3-GFP plasmid DNA (20 μg) was coelectroporated with increasing amount of HSB5 transposase plasmid DNA or in vitro transcribed HSB5 mRNA and GFP expression measured at 10 day after gene transfer by FACS. The differences in gene transfer efficiency comparing HSB5 DNA and HSB5 RNA were statistically significant (P<0.05).

Comparison of green fluorescent protein (GFP) reporter expression in γ-retroviral, lentiviral or transposon-mediated genetically modified PBL. Identical expression gene expression cassettes (MSCV-U3-GFP) were cloned into each vector system and produced as described.^40,41 (a) Peripheral blood mononuclear cell (PBMC) were electroporated with transposon system on day 0 or transduced with viral vectors following 2 days of stimulation with interleukin-2 (IL2; 300 IU ml^-1) and OKT3 (50 ng ml^-1). On day 14, GFP-positive cells were isolated by fluorescence-activated cell sorting (FACS) to normalize for differences in gene transfer efficiency. GFP-positive population of cells in each vector group was assayed weekly by FACS and on day 30 subject to a further expansion using the rapid expansion protocol (REP).⁴² Shown were the percent positive cells and mean fluorescence intensity (MFI) of each group measured at the times indicated. Transgene copy number was determined by quantitative PCR after normalization with β-actin. β-Actin and GFP primer-probes (Tamra) were from Applied Biosystems Oligonucleotide Factory (Foster City, CA, USA). Reactions performed on Applied Biosystems 7500 Fast Real-time PCR System. (b) Cryopreserved PBMC were resuspended in AIM-V medium and culture overnight. The next day, cells were subject to electroporation with the Sleeping Beauty (SB) transposon alone (DNA control) or transposon plus transposase RNA (Transposon) or transduction with γ-retroviral or lentiviral vectors. IL2 (300 IU ml^-1) was added to the cultures after gene transfer and the following day T cells were stimulated with OKT3 (50 ng ml^-1). The number GFP-positive cells in each vector group were determined by FACS at the time points indicated.

Transposon-mediated transfer of antitumor antigen T-cell receptors (TCR). pT3-p53 was constructed by cloning of the anti-p53 TCR previously reported⁴³ with the α- and β-chains being linked by a T2A ribosomal skip peptide.⁴⁴ pT3-MART-1 was engineered by cloning of the σ-T2A-β-MART-1-specific TCR DMF5⁴⁵ containing murine constant regions⁴⁶ into the pT3 transposon. (a) Peripheral blood mononuclear cell (PBMC) were coelectroporated with pT3-p53 or pT3-MART-1 and HSB5 RNA (20 μg each) and expression measured at 10 days after gene transfer by fluorescence-activated cell sorting (FACS) using MART-1 tetramer (Beckman Coulter, San Jose, CA, USA) or p53 pentamer (ProImmune, Oxford, UK). (b) Left panel: interferon-γ (IFN-γ) production assayed by enzyme-linked immunosorbent assay (ELISA; Endogen, Cambridge, MA, USA) of supernatant from overnight coculture of pT3p53/HSB5 RNA-electroporated PBMC and T2 cells pulsed with p53:264–272 peptide at labeled concentrations. Right panel: interferon-γ production assayed by ELISA of supernatant from overnight coculture of pT3-p53/HSB5 coelectroporated cells and listed tumor lines (H2087 HLA A2+/p53+, Saos2 A2+/p53 , 888 A2-/p53–). (c) Left panel: interferon-γ production assayed by ELISA of supernatant from overnight coculture of pT3-MART-1/HSB5 RNA-electroporated PBMC and melanoma tumor cell lines (888, MART-1+ /HLA-A2–, 938 MART-1+/HLA-A2–, 624 HLA-A2+/MART-1+, 528 HLA-A2+/MART-1+).

Comparison of tumor-antigen reactivity between γ-retroviral and transposon T-cell receptor (TCR) vectors. The anti-MART-1 TCR expression cassette was engineered into both γ-retrovirus and transposon vectors. Peripheral blood mononuclear cell (PBMC) were genetically modified by the two vector systems and comparisons made in TCR expression and functional reactivity. γ-Retrovirus transduction resulted in 28%, whereas transposon resulted in 36% MART-1 tetramer staining by fluorescence-activated cell sorting (FACS) analysis (not shown). (a) Both γ-retrovirus and transposon modified lymphocytes demonstrated specific reactivity to HLA-A2+ and MART-1+ melanoma tumor cell lines by interferon-γ (IFN-γ) enzyme-linked immunosorbent assay (ELISA). MART-1 TCR transposon (b) and γ-retroviral vector (c) modified lymphocytes demonstrated increased specific cytolytic activity of HLA-A2+ and MART-1+ tumor cell lines in a chromium release assay (performed as previously described⁴⁷).