Vectofusin-1 – based T-cell transduction approach compared with RetroNectin-based 1 transduction for generating murine chimeric antigen receptor T-cells 2 3

Gene transfer into human and murine T-cells using viral-based approaches has several promising therapeutic applications including the production of chimeric antigen receptor T-cell (CAR-T) therapy. The generation of murine CAR-T is paramount to test and validate immunocompetent mouse models for CAR-T therapy. Several viral transduction enhancers already exist for gene therapy with few limitations. In this study, we tested vectofusin-1, a short cationic peptide, as a soluble transduction enhancer for gammaretroviral transduction for the generation of anti-CD19 murine CAR-T. We found that in comparison to Retronectin, Vectofusin-1 is an equally optimal transduction enhancer for the generation of murine CAR-T cells.


Introduction 35
Chimeric antigen receptors (CARs) are synthetic receptors containing a target binding 36 domain usually derived from a single-chain variable fragment of an antibody, a 37 hinge/transmembrane region, and a truncated CD3 zeta cytoplasmic domain with or without 38 a costimulatory domain 1 . CAR T-cell therapy represents the latest advance in the treatment 39 of hematologic malignancies, with unprecedented response rates and survival outcomes 40 seen in patients with relapsed refractory lymphomas 2 . Therefore, there is a need for 41 production of CAR T-cells for both clinical use and laboratory research, particularly mouse 42 validation studies. 43 Generation of CAR T-cells through viral transduction can be optimized by adding 44 various culture additives such as cationic polymers (polybrene) 3 , dextran 4 , or cationic lipids 45 (lipofectamine 5 ). RetroNectin, derived from fibronectin, has recently been used to generate 46 CAR T-cells for clinical applications 6,7 , and RetroNectin has primarily been used to generate 47 mouse T-cells for in vivo testing. However, RetroNectin-based transduction protocols are 48 cumbersome because RetroNectin must be surface-coated prior to use, and therefore, a 49 new soluble additive capable of enhancing infection is needed. 50 Vectofusin-1, a new cationic amphipathic peptide, is a soluble additive that has been 51 successfully used for transduction of human T-cells; however, this approach has not been 52 tested for generation of murine CAR T-cells. 1. Activation beads can also be used for activation of T-cells; however, they are costly 197 compared with coated antibody methods. We followed the manufacturer protocol, 198 with a 1:1 bead-to-cell ratio. Also, activation beads must be removed from T-cells 199 before analysis or experimental use by pipetting T-cells in a conical tube and 200 thoroughly pipetting up and down to remove the T-cells sticking to beads. If the tube 201 is exposed to a magnet for 2 minutes, the unbound T-cells can be decanted from the 202 tube. For our experiments, we obtained stable virus-producing Phoenix-Eco cell lines as a 207 generous gift from Dr. Marco L. Davila's laboratory to generate gamma retrovirus. 208 3. Human IL-2 can be used for both mouse and human T-cells.

210
Statistical analysis 211 A nonparametric two-tailed t test was used to evaluate variation between groups. 212 Differences in proportions were evaluated using a two-sided Fisher exact test. For all 213 statistical analyses, p ≤ 0.05 was considered statistically significant.

216
Transduction efficiency with activation beads compared with plated antibodies 217 First, we sought to evaluate the activation of murine T-cells and transduction efficiency with 218 beads compared with plated antibodies. We harvested mouse spleens (n = 3) and murine 219 T-cells were isolated as detailed in the protocol above. 5X10 6 T-cells were collected and 220 were individually activated with either the mouse T-cells activator beads or plated anti-221 CD3/CD28 antibodies. Subsequently, we performed RetroNectin-based transduction 222 assays with same amount of harvested viral supernatants (3 mL) in each well. Because 223 viral supernatant was obtained from stable virus-producing cell lines and the supernatant 224 was pooled before transduction, we expected the viral titers to be similar between individual 225 experiments and steps. 226 Our CAR plasmid vector is based on an SFG plasmid backbone with an anti-mouse CD19 227 CAR sequence followed by a T2A and green fluorescent protein(GFP) sequence, as shown 228 in Figure 1A. We evaluated the transduction efficiency in murine T cells by determining the 229 expression of GFP using flow cytometry on a FITC channel. We found that in RetroNectin-230 based transduction assays, similar transduction efficiency was generated with beads (mean 231 = 60.40%, n = 3) as with plated anti-CD3/CD28 antibodies (mean = 55.4%, n = 3) using the 232 T-cell activation methods shown in Figure 1B and 1C (p = 0.14, two-tailed unpaired t test).

of RetroNectin and Vectofusin-1 for generation of murine CAR T-cells 235
Next, we compared RetroNectin with Vectofusin-1 as a transduction enhancer for 236 generation of murine CAR T-cells. For direct comparison, we activated T-cells isolated from 237 mouse spleens (n = 3) with plated anti-CD3/CD28 antibodies and did viral transduction on 238 day 1 and day 2. We found that Vectofusin-1 generated lower transduction efficiency (mean 239 = 43.33%, n = 3) than RetroNectin (mean = 55.57%, n = 3; p = 0.001, paired two-tailed t 240 test; Figure 2). 241 242 CAR-T cells proliferation and expansion is optimal with Vectofusin-1 243 Next, we also evaluated the absolute number of cells at day 5 after transduction with the 244 help of Countess FL automated Cell Counters and computed the median fold expansion 245 from the baseline. We found that by the end of day 5, fold expansion of total T-cells were 246 significantly higher with Vectofusin-1 (1.9333 + 0.141 fold) compared to Retronectin (1.4667 247 +0.0706) transduction (p=0.008).

249
Discussion 250 In our study, we compared Vectofusin-1 and RetroNectin as transduction enhancers for the 251 generation of murine anti-CD19 CAR T-cells. We found that the vectofusin-1 generated 252 lesser CAR-T transduction efficiency, as assessed by GFP expression, compared to 253 Retronectin. However, the absolute number of CAR-T cells generated was approximately 254 equal between the two groups due to higher expansion and proliferation of total T-cells 255 under vectofusin-1. The reason for the increased total number of T-cells generated with 256 vectofusin-1 is unclear to us. It is possible that Retronectin labeled plates adhere to T-cells 257 more tightly after spin down post-transduction and probably limit their mobility and 258 proliferation to a small extent. In conclusion, vectofusin-1 leads to optimal murine CAR-T 259 production, like Retronectin, for performing in-vitro and in-vivo mouse validation studies. 260 261 262 Acknowledgements: We thank Erica Goodoff, Senior Scientific Editor in the Research Medical 263 Library at The University of Texas MD Anderson Cancer Center, for editing this article.  264  265  266  267  268  269  270  271  272  273  274  275  276  277  278  279  280  281  282  283  284  285  286  287  288  289  290  291  292  293  294  295  296  297  298  299  300  301  302  303  304  305  306  307  308  309  310  311