An efficient large-scale retroviral transduction method involving preloading the vector into a RetroNectin-coated bag with low-temperature shaking

Abstract

In retroviral vector-mediated gene transfer, transduction efficiency can be hampered by inhibitory molecules derived from the culture fluid of virus producer cell lines. To remove these inhibitory molecules to enable better gene transduction, we had previously developed a transduction method using a fibronectin fragment-coated vessel (i.e., the RetroNectin-bound virus transduction method). In the present study, we developed a method that combined RetroNectin-bound virus transduction with low-temperature shaking and applied this method in manufacturing autologous retroviral-engineered T cells for adoptive transfer gene therapy in a large-scale closed system. Retroviral vector was preloaded into a RetroNectin-coated bag and incubated at 4°C for 16 h on a reciprocating shaker at 50 rounds per minute. After the supernatant was removed, activated T cells were added to the bag. The bag transduction method has the advantage of increasing transduction efficiency, as simply flipping over the bag during gene transduction facilitates more efficient utilization of the retroviral vector adsorbed on the top and bottom surfaces of the bag. Finally, we performed validation runs of endoribonuclease MazF-modified CD4(+) T cell manufacturing for HIV-1 gene therapy and T cell receptor-modified T cell manufacturing for MAGE-A4 antigen-expressing cancer gene therapy and achieved over 200-fold (≥ 10(10)) and 100-fold (≥ 5 × 10(9)) expansion, respectively. In conclusion, we demonstrated that the large-scale closed transduction system is highly efficient for retroviral vector-based T cell manufacturing for adoptive transfer gene therapy, and this technology is expected to be amenable to automation and improve current clinical gene therapy protocols.

Figure 1. Comparison of preloading methods for retroviral gene transfer assisted by RN.

(A) Outline of the experiment. MT-MFR3 vector was preloaded into each well of an RN-coated 24-well plate. The plates were incubated at 4°C on a reciprocating shaker at 100 rpm for (1) 16 to 48 h, (2) 12 to 24 h, (3) 12 to 72 h, and (4) 8 to 20 h (RBV-LTS). For comparison, the plate was incubated at 25°C for 3 h (RBV-Static) (4). After the preloading, SUP-T1 cells were transduced via each method. (B) Retroviral gene transfer efficiencies into SUP-T1 cells under RBV-LTS and RBV-Static conditions. After the transduction, SUP-T1 cells were collected, genomic DNA was extracted, and gene transfer efficiency was determined using the qPCR method by measuring the proviral copy number of the transduced cells. All data represent mean ± SD. Statistical analysis was performed by Student t-test (^** p<0.01, ^*** p<0.001). RN, RetroNectin; RBV, RN-bound virus; LTS, low-temperature shaking; IFU, infection-forming units.

Figure 2. Optimal amount of time for preloading using a scaled-up transduction procedure with the RBV-LTS method.

(A) Outline of the experiment. MT-MFR3 vector was preloaded into an RN-coated PL325 bag. 180 ml of MT-MFR3 vector was preloaded into the bag, which was then incubated at 4°C on a reciprocating shaker at 50 rpm. The optimal preloading period in the PL325 bag was examined at the time intervals of 8 to 48 h. For comparison, transduction into SUP-T1 cells in RN-coated 24-well plates under RBV-Static and RBV-Spin conditions was also performed in parallel. (B) Retroviral gene transfer efficiencies into SUP-T1 cells under the large-scale RBV-LTS condition. After the transduction, SUP-T1 cells were collected, genomic DNA was extracted, and gene transfer efficiency was determined using the qPCR method. All data represent mean ± SD. Statistical analysis was performed by Student t-test (^*** p<0.001). N/A, not applicable.

Figure 3. The optimal time for flipping over the bag in a scaled-up transduction procedure using the RBV-LTS method.

(A) Outline of the experiment. MT-MFR3 vector was preloaded into an RN-coated PL325 bag. The bag was incubated at 4°C on a reciprocating shaker at 50 rpm for 16 h. After preloading, the bag was rinsed once, SUP-T1 cells were added to the bag, and the optimal time to flip the bag over was examined at time intervals of 1 to 8 h. For comparison, transduction into SUP-T1 cells in an RN-coated 24-well plate under RBV-LTS was also performed in parallel. (B) Retroviral gene transfer efficiencies into SUP-T1 cells under the large-scale RBV-LTS condition. After transduction, SUP-T1 cells were collected, genomic DNA was extracted, and gene transfer efficiency was determined using the qPCR method. All data represent mean ± SD. Statistical analysis was performed by Student t-test (^** p<0.01, ^*** p<0.001). w/o, without.

Figure 4. Validation runs for manufacturing endoribonuclease MazF-modified human CD4⁺ T cells.

(A) Schematic diagram of the experiment. Primary human CD4⁺ T cells were stimulated, transduced with MT-MFR3 vector using the RBV-LTS method, and further expanded. (B) Growth curve of two validation runs of healthy donors' CD4⁺ T cells. Arrows indicate the time of the transduction. (C) Retroviral gene transfer efficiencies into CD4⁺ T cells under the large-scale RBV-LTS condition. Gene transfer efficiency was determined using the qPCR method to determine the proviral copy numbers (D) The percentage of CD3 positive cells among the expanded cells was measured by gating the live cells based on the FSC/SSC parameters during the flow cytometry analysis (left panel). After gating for CD3, cells were analyzed for CD4⁺ or CD8⁺ expression (middle panel). CD3⁺ CD4⁺ cells were also analyzed for CD45RA and CCR7 expression using flow cytometry (right panel). CM, central memory; EM, effector memory; TDEM, terminally differentiated effector memory.

Figure 5. Validation runs for manufacturing TCR-modified human T cells.

(A) Schematic diagram of the experiment. Primary human PBMCs were stimulated, transduced with MS-MA24-siTCR vector using the RBV-LTS method, and further expanded. (B) Growth curve of three validation runs of healthy donors' PBMCs. EXP. 1 and 2 are the results from Donor TC2900 and EXP 3 is from Donor TC1900. Arrows indicate the time of the transduction. (C) Retroviral gene transfer efficiencies into PBMCs under the large-scale RBV-LTS condition. Gene transfer efficiency was determined based on the percentages of tetramer-positive cells among the CD8⁺ cells. (D) The percentage of CD3 positive cells among the expanded cells was measured by gating the live cells based on the FSC/SSC parameters during the flow cytometry analysis (upper left panel). After gating for CD3, cells were analyzed for CD4⁺ or CD8⁺ expression (upper right panel). CD3⁺ cells were also analyzed for CD45RA and CCR7 expression using flow cytometry (bottom panel). (E) Antigen specific IL-2 and IFNγ production achieved by stimulating the gene-modified cells with MAGE-A4 peptide-pulsed T2A24 cells. IL-2, interleukin-2; IFNγ, interferon-gamma; Approx., approximately; EXP., Experiment.