This month's question from the Molecular Biology Forums (online at molecularbiology.forums.biotechniques.com) comes from the “General Methods” section. Entries have been edited for concision and clarity. Mentions of specific products and manufacturers have been retained from the original posts, but do not represent endorsements by, or the opinions of, BioTechniques.
Molecular Biology Techniques Q&A What is the importance of filtration during lentivirus production? (Thread 23111)Q I have a general question about lentivirus production. After I collect the supernatant containing the lentivirus from the transfected 293T cells, all protocols I have seen state that I should spin down the supernatant (to remove debris and dead cells, I presume) and then pass it through a .45/.22-µm filter. Is the filtration step necessary? I would prefer to omit this step, but will that cause problems for me later when I try to concentrate the virus by ultracentrifugation?
A The filtration step is needed to remove multivesicular bodies (MVBs) and very small membrane parts that can form micelles. Most MVBs are ~300 nm in diameter. If you omit the filtration step you will definitely end up with a lot of impurities.
In my experience, a 220-nm filter leads to a purer virion preparation than a 450-nm filter, although you will collect fewer virions in the end. The purity difference is significant enough to be seen in a Coomassie-stained gel. Some bands representing proteins of cellular origin will not appear when using a 220-nm filter and ultracentrifugation.
Q The reason I would like to omit the filtering step is that a 0.45-µm filter placed on a 50-mL syringe got clogged after only 10 mL of my 300 mL collected media. I skipped filtering the rest and ended up with a dirty and insoluble pellet after ultracentrifugation. Maybe I should try a larger filtration unit with a vacuum pump.
A You should preclear the supernatant prior to filtration by low-speed centrifugation to avoid clogging the filter. However, 300 mL is too much media to pass through a single 450-nm filter. If you try to get a solution through an already clogged filter, it often foams and you will end up pressing impurities into the filtrate. If you want to use a filter with a syringe, you should split your collected medium into 20–50 mL volumes, pre-clear the aliquots (1500–2000 rpm for 10 min) in Falcon tubes, and then finally pass each volume through a separate 450-nm filter. Depending on your ultracentrifuge, you can either recombine the aliquots prior to spinning or combine the resuspended pellets after centrifugation.
Q I tried that method, but after ultracentifugation I could not dissolve the pellets in PBS + 1 mM EDTA, even after I let them stand overnight at 4°C. The solution just became slimy. I centrifuged and did the titration on the supernatant anyway, but the yield was not good.
A I once had problems passing my lentivirus-containing media through the filter because I screwed the filter onto the syringe to tightly. You should make sure the filter is secure, but it should not be to tight or it can interfere with the passage of the media.
A You can store the precleared supernatant at 4°C overnight without a loss of infectivity. Stored viruses will work for simple reinfections to propagate the virus even after several weeks at 4°C. However, if you need to maintain high infectivity, it is better to freeze aliquots of ultracentrifuged samples at −70°C/−80°C or even in nitrogen. If high purity is not needed, ultracentrifugation is usually carried out through a 20% (w/v in PBS) sucrose cushion.
Slimy pellets often come from serum protein in the cell culture supernatant that gets concentrated after passing through the sucrose cushion.
Q I used 2% fetal calf serum (FCS) in the medium and didn't use the sucrose cushion. I am thinking of trying the Vivaspin 500 devices with the polyethersulfone (PES) membrane (100–600 µL). I think the 100-kD cutoff will be suitable.
A I haven't used those columns, but one of my labmates used it to concentrate his engineered retrovirus and got repeated low titers. In contrast to viruses such as HPV, you always have to consider that the physicochemical properties of the retrovirion can differ depending on the cell line (and thus cell membrane) in which it was produced. This is particularly true for virions that have an altered molecular surface (i.e., HIV particles containing surface fusion proteins instead of the cognate Env). These properties, along with contamination, can interfere with the function of the column.
You should always use a sucrose cushion. Without one, you will see a huge BSA band on a Coomassie-stained gel and a lot of other heterologous protein bands even when the pellet is not slimy.
Q To prepare the sucrose cushion in a 35-mL tube, should I pipet 10 mL of 20% sucrose, add the supernatant on top, and spin?
A I usually use open polypropylene tubes in an Beckman swinging bucket rotor (SW42Ti). Given that I have 8–10 mL precleared supernatant after low-speed centrifugation and 200/450-nm filtering, I add 2 mL sucrose solution (20% w/v in PBS) and then carefully overlay the supernatant without disturbing the interface. After 2 h at 28,000–36,000 rpm at 4°C, the phases will mix, but most of the impurities will stick in the sucrose. The virions will be in an invisible pellet at the bottom of the tube. Depending on what I need them for, I either resuspend the pellet directly in SDS-PAGE sample buffer (for SDS-PAGE and various immunoblot analyses) or in PBS (for cell culture purposes).
Q I use the 70Ti fixed angle rotor which holds 35 mL tubes. For lentiviruses, we use tubes that seal (like the ones used for CsCl/EtBr plasmid isolation). I have access to the rotor you mentioned and can use it for small experiments, but I usually need virus for in vivo work where I need quantities of 10 × 108 IU, so I have volumes greater than 300 mL of supernatant.
A I prefer swinging bucket rotors because the invisible pellet will always be at the bottom of the tube. There is a swinging bucket rotor for larger volumes called SW32 from the same company. I think it holds ~35 mL.
