After over a decade of seeding small(er) scale 293T fresh from ATCC I can say with confidence that Invitrogen (HeLa based) chart never really worked out for me to obtain 70-80% for 293T, the significantly smaller cell body size affects the seeding a lot.
I've now developed counting-based seeding for most small scales that are highly efficient but I would be curious to know if people have experience (from counting) of densities that work for real large scale volumes and compare. I'm currently playing with a lot of formats and among them are multi-layer flasks like the Nunc triple layer flasks (500 cm2 area). Talked with people who work with the Corning 10 layer stacks which you can't check under the microscope anymore - and their numbers seem extremely high for seeding.
So for those of you who do counting based in large volumes & areas (lets say at least 150 mm plates): could you give a number of how many cells you seed in what volume of media for 70-80% confluency 18-24h later? What about 70-80% 42-48 hours later? Are your 293T fresh from ATCC or are they long term circulating in your lab (cryo, freezing, thawing again, starting batch number at 1 again maybe)?
Thanks!
Dear Ulrike,
Typically, for lentiviral production, the recommend cell density of 293T/T17 (ATCC) is 80% confluency, which means roughly 240,000-280,000 cells/cm2. To get this confluency, 140,000cells/cm2 should be seeded in DMEM with 10% FBS at 24h in advance, while 70,000cells/cm2 at 48h ahead.
If you are going to package lentivirus, I recommend you visit this website.：www.genemedi.net/i/lentivirus-packaging
I agree, 293T need much higher seeding density than invitrogen say is average.
I use 3 million cells per round 10cm dish (55cm2). 11.25 million per round 15cm dish (150cm2). Works very well (using PEI).
As you may know, cell growth in the presence of serum varies with the lot of serum. Also, small differences in temperature among incubators can affect your cell growth and confluency may be estimated differently by different people. My suggestion is to evaluate at small scale using a defined number of cells per cm2 and looking at your desired endpoint (i.e. amount of virus produced). This can then be translated to larger flasks. We use 8 x 10e4 cells/cm2 to get to the optimal density at 18-24 hours for our application based on our endpoint.
When I am generating lentivirus, I typically use a 10 cm dish. I seed at 4.6 x 10^6 cells on day 1, transfect the cells day 2, wash and add 6 mL medium for collection day 3, harvest and replenish medium day 4, and harvest day 5. On days 4 and 5 I am infecting my cell of choice. Usually, I place the filtered viral supe on and leave it over night.
The density of the cells is pretty high the day of transfection. By the final harvest, the cells are densely packed and the medium is turning from red to orangey-yellow. Based on the difference in surface area (~3 fold), I would recommend seeding a 15 cm dish at 13.5 million cells and harvesting in 18 mL medium.
This protocol, has worked well for infecting 293s, Hela, and THP-1s. Hope it helps.
Hi Ulrike Jung,
I used HEK 293T cell line for transfection of HIV-1 type B (pNL4-3) viral plasmid. The seeding was done in 6 well plate as well as in T-25 flask. In 6 well plate (10 cm2), the seeding density for 2ml media was 5 x 10^5 cells for 50-60% confluency after 24 hours. We are not using 70-80% confluency as the same cell are used for transfection and harvesting virus for 24 hours, 36 hours and 48 hours. Similarly for T-25 (25 cm^2) the seeding was done with 1.5 x 10^6 cells for 5 ml. For T-75 (75 cm2) the seeding was done with 5 x 10^6 cells for 13 ml. But for 70-80% confluency in 18-24 hours, there should be more cells as compared to above. The cells were fresh cell passaged for 4-5 times.
Sher, a 15 cm dish is 3x larger than a 10 cm dish. So she would need to increase volume 3 fold as well.
We were using 293T cell line to produce lentiviruses. And for 150 mm dishes we passage our cells at the ratio 1/4. We passage 3 confluent 150 mm dishes to 12 x 150 mm dishes 24 hours before transfection. So at the transfection time we have 12 petri dishes which have % 80 confluence each.
Thanks everyone for their seeding input. I was hoping more people have experience with larger scales but have found decent values now for the 500cm2 triple flasks (ca. 55.000/cm2 18-24h in advance). Yes, the serum lot of course plays a role, thanks Johannes for that reminder, I had not considered that. That might explain reporting differences too.
And no, you don't have to increase the volume in a linear fashion, that is one of the points of large scale production: higher titer concentration at larger scales while not needing so much volume. You just need enough that the peak production is maintained and the area is covered! The amount of media/cell added prior to packaging at the seeding time point seems to indeed affect that number to some degree at least even over short periods. Which is why after adding the packaging mix I use larger volumes of media than I do for seeding. But no, linear is not necessary in my experience.
We use Corning HyperFlasks (1720 cm2 area) and plate 230 million 293T cells in 560 ml media.
Ulrike, since you have done a lot of small scale seeding, could you help me with this? I want to do transfections in HEK293T cells. Can you help me with the cell seeding density for 60mm petri dish ? I need to plate cells on day 1, transfect on day 2 and harvest cells on day 4. Thanks !
By now I recommend in DMEM with 10% FBS with 293T/T17 (ATCC) to seed for 80% density (which means roughly 240,000-280,000 cells/cm2, ca. 300,000-333,333/cm2 at 100%): 24h ahead seeding at 140,000/cm2, 48h ahead seeding at 70,000/cm2. Hope that helps.
Thanks to all scientists for sharing their expertise but especially to Dr Jung for testing all advice and getting back with concrete advice on seeding number and corresponding confluency at desired time point.
- Badges
- Science topic
- Similar topics
- Methodology
- Laboratory
- Cellular Biology Techniques