Back

HTLV-1 Splice Sites in Prevalent Gene VectorsCause Splicing Perturbations in Transgenic Human Cells

Miskey, C.; Prommersberger, S.; Mestermann, K.; Hudecek, M.; Ivics, Z.

2023-01-30 genomics
10.1101/2023.01.28.526022 bioRxiv
Show abstract

The use of any semi-randomly integrating gene vector in a therapeutic setting is associated with genotoxic risks. The two major mechanisms of genotoxicity are disruption of a coding sequence (loss-of-function) or transcriptional upregulation of genes (gain-of-function) in the cellular genome where the genetic modifications are executed. A third, less widely recognized genotoxic risk stems from splice sites and polyadenylation sites within the vector sequences. These transcriptional elements may drive aberrant splicing and/or polyadenylation between transgene-contained and genomic sequences. A widely used promoter/enhance element present in gene vectors to ensure high transgene expression levels in mammalian cells is composed of a hybrid EF1/HTLV-1 LTR, in which the retroviral LTR contains an intron. We assessed aberrant splicing initiated from the splice donor (SD) element present in the HTLV-1 LTR in CAR-T cells that had been engineered by either lentiviral vector (LV) or Sleeping Beauty (SB) transposon-mediated gene transfer. We establish that the vector-contained canonical SD site gives rise to aberrantly spliced RNA species and thereby can cause misexpression of host gene segments that are involved in various host cell functions. This, potentially genotoxic, effect could be abrogated by mutating or completely eliminating the SD (or the entire intron) from the HTLV-1 LTR segment. CAR-T cells generated by the modified vectors are equally potent in efficiency of CAR-T cell manufacturing and in functionality. The simple genetic modifications that we describe here affecting vector design therefore enhance genomic safety while maintaining efficacy of gene-modified therapeutic cells.

Matching journals

The top 12 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.