Hi Peter

I think many researchers have encountered this problem. 8kb gene have exceeded the capacity of most common viral vectors. Effective promoters and expression regulatory elements are essential for gene expression，there is only 6kb gene insertion left for adenovirus, 4kb for lentivirus and 2kb for AAV. However, baculovirus is a good choice for gene expression in mammalian cells，its capacity is very large, even 10kb coding sequence work well in this vector.

Genemedi is experienced in virus production and large gene expression. You could find detail information on this website: www.genemedi.net

There is a publication that might help: http://www.ncbi.nlm.nih.gov/pubmed/17533614

I assume you package in lentiviral vectors? What is the original backbone you use, what lab is it from? These have impact too.

And your target gene can impact the packaging as well of course, if it interferes with the very packaging process or stresses the packaging cell. I have packaged viral genomes for vectors only up to 6kB so far so can't talk about your size. But I'm regularly packaging HIV itself with genomes of 9.7kB without problems - and the titer depends significantly on the strain used, so the DNA sequence plays a major role there.

Hi Peter,

Titer decrease with size. However, packaging 8-10 kb should be possible and still quite efficient as long as your vector contains the rev responsive element (RRE) ? Do your packaging plasmids include a REV encoding plasmid ?

It should work ; check this ref: Kumar, M. et al, HGT 2001, 12:1893-1905.

Do you have a polyA within the 9kb cloned in the sens of the vector ? If so, It should be cloned antisens.

What might also happen is that you have criptic splice sites within your sequence and vRNA is processed before encpsidation.

Hi Peter,

what matters for LV vector packaging capacity is not the size of the coding sequence, but the size of the viral genome, i.e. the transcript expressed by your LV expression plasmid. This is the part that will end up as genomic RNA inside the particles. I suppose you are working with SIN vectors, so the two LTR's in your expression plasmid are:

5'-LTR: U3 (or CMV, or RSV)-R-U5

3'-LTR: (delta)U3-R

R is the repetitive element flanking the genomic transcript:

5'-cap - R-U5 - psi - RRE - cPPT -(Promoter-insert)- WPRE - deltaU3-R - AAAAAAAAAAA - 3'

Sequence of HIV R:

gggtctctctggttagaccagatctgagcctgggagctctctggctaactagggaacccactgcttaagcctcaataaagcttgccttgagtgcttca

... so you may locate it in your vector.

The size of the 5'-R to 3'-R part is your vector genome size. HIV has a native genome size of ca. 10.2 kb. HIV particles are near-spherical, and will no longer assemble efficiently with larger genomes. How much insert you can "afford", basically depends the size of your internal promoter.

@ Ulrike:

Luckily, few genes actually stress the packaging cells so much that the titer drops to useless levels. Of > 100 genes we have packaged, there were only one or two "killers". Of course it depends on the type of genes you're looking at.

The E. coli strain used to grow the plasmid (and hence plasmid quality) matters a lot for viral titers. But how any E. coli strain would enable HIV to package larger genomes.... I can't quite imagine. Viral packaging is an RNA-dependent process.

@ Emmanuel:

10 kb genome size (R to R) is possible, absolutely. We've made LV vectors expressing 3 genes, with a genome size of slightly over 10 kb. The titers were not even markedly lower than what we usually get.

Packaging limit for lentivector (and retro I guess) is close to 11kb-12kb and decreases dramatically if longer RNA is tested (a paper describes this study: Hum Gene Ther. 2001 Oct 10;12(15):1893-905.

Systematic determination of the packaging limit of lentiviral vectors.

Kumar M, Keller B, Makalou N, Sutton RE.

They seem here to say that vectors with a 18kbRNA are produced indeed but with a low titer. Perhaps is ti your case ? That would invite you to increase the concentration fold (if you concentrate) or your scale of production.

COnsidering the fact that a classic vector backbone is around 4-5kb, limit for the Promoter-transgene cassette should be close to 7kb in a lenti. For a 9 kb transgene, lenti . AAV or Adenovector should help you.

Alternatively you can try to alter the vector design to reduce the vector backbone size (eliminating non essential sequences like WPRE or SA) to have a complete RNA smaller than 12kb(from R5' to R3').

Perhaps also can you imagine a way to encode your transgene by two different lentivectors, one carrying a 5'portion and the other a 3'portion (with a sort of Intein-based strategy. The complete protein would be reconstitued in the cell cotransduced by the two lentivectors) but that sounds quite complex.

Many thanks to all for your valuable contributions

Hi Peter,

You could try also to use Adeno-Associated Viruses (AAV), but you need to transplice 2 AAV vectors together each containing half of your gene of interest. What applications/target gene(s) are you interested to express?

Use gutless adenovirus vectors..

Hi Peter

I think many researchers have encountered this problem. 8kb gene have exceeded the capacity of most common viral vectors. Effective promoters and expression regulatory elements are essential for gene expression，there is only 6kb gene insertion left for adenovirus, 4kb for lentivirus and 2kb for AAV. However, baculovirus is a good choice for gene expression in mammalian cells，its capacity is very large, even 10kb coding sequence work well in this vector.

Genemedi is experienced in virus production and large gene expression. You could find detail information on this website: www.genemedi.net

About AAV, lentivirus and adenovirus packaging process, you can see a lot of details in these protocols.

There are other vectors with greater payload size like HSV vector. Checking duration of it's trans-gene expression for your aim seems to be necessary.