Design Pitfall: Using The Same Promoter To Drive Two Genes In Lentiviral Vector

Incorporating two gene cassettes into a lentiviral vector is a common strategy for achieving the expression of both a gene of interest (GOI) and a marker gene, such as an antibiotic resistance or fluorescence reporter gene. However, a frequent pitfall is using the same promoter for both cassettes, which can result in silencing of the upstream gene. Here, we constructed a series of lentiviral vectors expressing two fluorescence reporters to demonstrate the silencing effect.

When both cassettes were driven by the CMV promoter (568 bp), the upstream reporter was expressed in only about 12% of stably transduced cells, whereas the downstream reporter was expressed in 97% of the cells (Fig. 1A). This phenomenon was not dependent on the specific reporter gene, as swapping the positions of the reporters yielded only 14% expression of the upstream reporter (Fig. 1B). Extending the length of the common promoter and using EF1⍺ promoter (1,179 bp) made the situation even worse; only ~4% of cells expressed the upstream reporter (Fig. 1C). In contrast, when two different promoters, EF1⍺ and CMV, were used for the two reporters, they were expressed in 96% and 91% of cells, respectively (Fig. 1D).

This silencing effect is likely related to the lifecycle of human immunodeficiency virus type 1 (HIV-1), from which the lentiviral vector is derived. Each HIV-1 virion contains two single-stranded RNA genomes that recombine during reverse transcription in the host cell. Thus, in lentiviral vectors with identical promoters, recombination between these promoters can lead to the loss of the upstream open reading frame, resulting in silencing of the upstream reporter gene.

To investigate whether the length of the identical sequence in the promoters affects recombination, we used the CBA and CMV promoters, which share a 279-bp common region, for the upstream and downstream cassettes, respectively. This configuration allowed approximately 32% of transduced cells to express the upstream gene (Fig. 1E), which is higher than with two CMV promoters (12-14%, Fig. 1A and B) or two EF1⍺ promoters (4%, Fig. 1C), yet lower than when using two different promoters (96%, Fig. 1D). These results suggest that the likelihood of recombination is proportional to the length of the identical sequences in the promoters.

More broadly, our findings indicate that any identical sequence of sufficient length may cause recombination and expression silencing. For example, in the vector where both reporters were normally expressed using the EF1⍺ and CMV promoters (Fig. 1D), inserting a 249-bp non-coding stuffer sequence between the promoter and reporter of each cassette reduced the expression of the upstream reporter from 96% to 19% of cells (Fig. 1F). These results underscore the detrimental effects of repetitive sequences in lentiviral vectors and highlight the importance of avoiding such sequences in vector design.

Figure 1. Loss of the upstream cassette expression due to potential recombination of lentivirus delivered sequences. 293T cells were transduced with lentiviral vectors (A-F) at MOI 10 and subjected to puromycin selection. After the stably transduced cells were established, the expression of two fluorescence markers was quantified using FACS.

Comments/Discussion section

Comments can be submitted by registered VectorBuilder accounts and are reviewed by VB vector experts before being posted.