posted over 2 years ago

CRISPR library transduced cell numbers I need to keep during passage after puromycin selection

I am a beginner of CRISPR pooled screen. I read the paper titled "Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening" (Julia Joung et al., 2017). In the paper, for a library size of 100,000 unique sgRNAs, we transduce 1.67 × 10^8 cells at an MOI of 0.3 at the start. After puromycin addition, I need to passage the cells when they reach full confluency, I wonder how many cells I need in the following passages? My calculation is that for 1 sgRNA I need 500 cells, so the cells I need is 100,000 * 500 = 50, 000, 000 (50 million). Am I on the right track?

My second question is that for cells I don't need in my following passages after puromycin selection, can I freeze these live cells (for example, freeze 40--50 million live cells in freezing media in liquid nitrogen) for a future culture so that I do not need to go back to the CRISPR lentiviral VLP transduction step again?

over 2 years ago

You calculation 100,000 * 500 = 50, 000, 000 is assuming all your cells are already CRISPR+. By my experience, you should maintain your cells at the same number when you start puro selection( 50, 000, 000/0.3, 1.6x10^8 in you hands) until they fully pass puro selection(normally at least 7 days).

2. No. Freeze/thaw is a selection by itself and you will not be able to get a good screen result for the phenotype you interested in.

over 2 years ago

I see. Thanks a lot!

over 2 years ago

Following Xiaos’ question. I am working in a tamoxifen-inducible SAM CRISPR/Cas9 library. Without tamoxifen there is no gene expression activation. In this case, would be necessary to keep the infection at x500 cell/ guide or would be fine if I do x100. Then, would be possible to freeze live cells (10 millions) after puro selection without tamoxifen for future cultures?

I agree that I would need to grow the cells to 50 milions just before adding tamoxifen and starting the screening.

Thank you very much for your help.

over 2 years ago

1. You still need 300-500x. The 300-500x coverage is required to avoid causing big variation on any of your sg reads. Imagine you only start with 100x and some sg lose 50x during your puro selection, it will cause much greater difference comparing to losing 50x in a 500x population.
2. Ideally you can freeze-thaw your cells. However, you should avoid any manipulation that will give unnecessary variation during your screen. Freeze/thaw is definitely a process that will change your cell population in some way. So it's not recommended.

over 2 years ago

I understand you but if we confirm that the system is tight and there is no CRISPR/Cas9 mediated activation without tamoxi we should consider that all cells infected with the library are the same although they incorporated a single guide. Am I right?

I don't see the difference between having sg lose of x50 in a library representation of 500x or 100x. If you grow cells after puro selection without tamoxi to have a coverage of x1000 in both cases you will have a representation of 20x of this specific guide, right?