Hi Jonathan Elegheert ,

As I feel, Since you are planning to add poly A sequence to the antisense cassette, it is not necessary to include an additional WPRE element. It will make your cargo bigger. Since polyA is in the opposite direction, it will not interfere with the transcription of the full viral genome and the subsequent viral titer.

Also, why don't you arrange both cassettes into one direction (5' to 3')?

See this popular addgene LV vector (pLenti CMV GFP Hygro (656-4 )) having dual cassettes. You can easily replace your inducible cassette with the CMV-EGFP & constitutive cassette with the PGK-HygR.

I'm just a beginner in this field. So I hope this will help you at least a little.

Good luck !!!

Hi Jonathan

Adding an antisense transgene into a lentiviral vector backbone will definitely reduce your titre, due to transcriptional collision when producing the vector genome. But the titre loss can be minimised by correct design. This recent paper shows how these vectors can be designed well.

https://www.nature.com/articles/s41434-018-0039-2

It looks as though your decision to use an inducible promoter for the antisense transgene is a good one.

WPRE is generally needed for efficient production because it helps termination at the pA motif in the 3’LTR. This has benefits both during production of the vector and after transduction.

Article Improving Transcriptional Termination of Self-inactivating G...

I hope this helps with your design.

Best

Thanks John, really useful references!

Indeed, my plan was to include the TRE3GS promoter in antisense direction to minimize or avoid transcriptional collision and titer loss.

The general problem with these dual-pomoter designs however is the limited packaging capacity left for the transgene and the impact of the large proviral DNA on the titer. I am wondering whether it is not more worthwhile to switch to e.g. the PiggyBac tranposon for these reasons. Thanks Jonathan

Packaging capacity is quite broad for lentivirus. I’ve Packaged 16kb.

https://www.ncbi.nlm.nih.gov/m/pubmed/28250438/

Titre was low, but titre should hold ok up to 8kb. There’s a figure in this paper showing efficiency with increasing payloads.

Thanks for the link! Yes, I knew about another paper where this was systematically investigated (https://www.ncbi.nlm.nih.gov/pubmed/11589831) but didn't really find any real-life examples. When I look at my current design I'm at ~6.5 kb without the inducible transgene; I'm just afraid that it will limit efficient use of the system to transgenes of max. 3 kb, before the efficiency really takes a hit.