Lentiviral packaging for Christmas

A large part of my PhD project revolves around creating and using various lentiviral particles for gene transfer into mammalian cells. The principle of lentiviral packaging and transduction goes something like this. You transfect your packaging cell line (I use HEK-293T) with plasmids encoding the envelope and capsid virus proteins along with the transcriptase and integrase. That will be sufficient to create your pseudotyped particles. Lastly, you include your transgene or shRNA for expression/knockdown in your target cells post transduction. 

 

www.amsbio.com

The number of plasmids used for transfection can vary but in my project I use four; one with my shRNA (so far only scrambled), MDL (HIV gag-pol), VSV-g (the glycoprotein from the Vesicular Stomatitis Virus for target cell entry) and Rev (HIV rev). Transfection of those four plasmids will result in the packaging of lentiviral particles containing my scrambled shRNA. The vectors I use belong to a 3rd generation lentiviral packaging system. This system is safer than the 2nd generation system but slightly more cumbersome as it requires the transfection of four separate plasmids (more information about packaging plasmids). The resulting lentiviral particels are replication deficient and are only capable of one round of infection and integration.

When I started I was using a roughly 5kb scrambled shRNA plasmid with a GFP reporter gene. Originally I had some problems collecting supernatant with high viral titer. That problem I could easily solve by using younger cells. Generally it's not recommended to use HEK-293T cells after passage 30. Mine had not quite reached that age but it seemed they were no longer up for the daunting task of assembling my particles. Once I changed the cells everything went somewhat smoothly. Or at least as smoothly as things can go in science. I optimized the transduction of my target cells and determined that the methods I was using were compatible with our specialized culture system.

Then we had the brilliant idea to change the shRNA vector. We found this extremely neat little vector called CherryPicker. It is basically an mCherry fluorescent protein coupled to a truncated transferrin receptor. Both the mCherry and the receptor reside on the outside of the cell allowing for cell sorting using either the fluorescent tag or the receptor.
 

 

www.clontech.com

This can come in handy when you don't have access to a FACS sorter. Cell sorting using cell surface receptors and magnetic beads require no advanced equipment and should be quite straight forward. I say "should be" because, currently, my magnetic sorting does not work. Maybe I'll write about that later.

Once I attempted to produce lentiviral particles using the CherryPicker vector instead of the GFP one the titers went way down. I made three different stocks and titrated them on Huh7 cells using the protocol by Bard, Salmon and Trono from 2010 (Pubmed). This titration method had worked great for the GFP particles but only one of my three CherryPicker stocks had a measurable titer and that one was definitely not something to be proud of. Then I thought that maybe the ratio of my packaging/shRNA mix weren't optimal for the CherryPicker. That vector is over 11kb, so almost twice the size as the GFP-shRNA vector. Maybe I needed to add more? Or less? 

No. My current ratio (which is roughly 1:1:1:1 for all plasmids) turned out to be the the best one for the CherryPicker. However, the amount of the HIV p24 capsid protein (measured using p24 ELISA) for CherryPicker particles was only about 2/3 of the amount of p24 for GFP particles produced in the same way, with the same ratio. Then I started thinking that maybe the packaging of CherryPicker particles was delayed, since the shRNA vector is quite large. Maybe it was slowing down the process?

So I set up an experiment where I collected particle containing supernatant at 24, 36, 48, 60 and 72 hours post transfection (I even chronicled my excitement for you). I did this both for the traditional HEK-293T cell culture medium (DMEM+10%FBS) and also for the medium I use on my target cells (hence the two tubes). Serum can have an adverse affect on some cell types so it would be optimal to be able to collect the particles in either serum free medium or your target cell medium. Traditionally, harvesting of lentiviral particles in done around 48h post infection and that's what I've been doing so far. However, for the CherryPicker I found that there is a big jump in the amount of p24 capsid between hours 48 and 60 while the amount remained quite stable between hours 60 and 72. That indicated that the packaging of my CherryPicker particles in indeed delayed compared to the GFP vector and the amount of p24 is also higher when they are collected in my target cell medium! 

All in all, if it actually turns out to hold true when repeated, I may have solved my lentiviral packaging crisis. Just in time for Christmas. 

-Hüldi