Does anyone of you know how many cells is needed to run a cell cycle distribution analysis using flow cytometry (PI staining)? Mouse ES cells have been treated with a certain agent and I expect the cell cycle distribution of the cells to be affected.
Thanks in advance,
Sharif
Hi, Sharif,
Depends how the reagents would affect your cell cycle stages and which stage you are look into. Stem cells do have relatively higher S/G2 compare to other cell types.
In my cycle analysis for mouse bone marrow cells, I need collect at least 50000 events from the sorted population, to make sure after discard all unwanted events I would still have at least 20000 single cell events in the analysis gate. Of course the starting material will be at least 2 million cells, as the fixing and harvest process will result cell loss.
If you are to pick up minor changes or low frequency events from drug treatment such as apoptosis, you will need a lot more cells to start with and collect a lot more total events in the FACS, I would start with 10 million cells and collect 200000 events at least.
My FACS samples have always been in 10-20 million cells per ml, otherwise it will take forever to process the sample through FACS machine.
Hope this is useful for you.
To check the effect of agent on cell cycle specially on sub Go/G1 population (apoptotic cells), adjust cell density 1.5* million cells/ml for treatment. For FACS analysis, adjust at least 10000 single cell events in the analysis gate.
Have a look at the attached paper. Hope u will get some idea from this.
Article Deltamethrin induced an apoptogenic signalling pathway in mu...
Hi, Sharif,
Depends how the reagents would affect your cell cycle stages and which stage you are look into. Stem cells do have relatively higher S/G2 compare to other cell types.
In my cycle analysis for mouse bone marrow cells, I need collect at least 50000 events from the sorted population, to make sure after discard all unwanted events I would still have at least 20000 single cell events in the analysis gate. Of course the starting material will be at least 2 million cells, as the fixing and harvest process will result cell loss.
If you are to pick up minor changes or low frequency events from drug treatment such as apoptosis, you will need a lot more cells to start with and collect a lot more total events in the FACS, I would start with 10 million cells and collect 200000 events at least.
My FACS samples have always been in 10-20 million cells per ml, otherwise it will take forever to process the sample through FACS machine.
Hope this is useful for you.
I agree with Anoop for 10 000 as a minima for analysis after all gated parameters setted. By the way and If you are not limited acquire at least 100 000 events.
Thomas
The minimum number of gated , single cells to model is 10k...accuracy increases with higher numbers in the 50-100k range. Excellent source ; go to vsh,com and follow the resource links
1) It depends on the G1 CV and the expected fraction of cells in sub-G1, G0/G1, S, or G2+M. If your CV's are good (e.g., 1-3%), then 400 events in whatever fractions you are interested in is a good place to start as a minimum e.g., if you are interested in G2+M and G2+M =10%, then 4000 cells is a safe minimum . 2) More important than the number of events is the induced fractional changes. If you are anticipating small fractional chances, then you need more replicates, not more events. Also, I agree with Keith. Consult VSH.com
Dear Sharif Moradi,
In 2014 I have analyzed a lot of FACs stainings for the HSC cell cycle. My cells were frozen (in liquide nitrogen) and I wanted to see/count cells in the G0, G1, S, G2, M phases of cell cycle + apoptotic cells. If I had 5-10 mln of frozen cells, I could get nice results. If I had less than 5 mln to begin with, the cell subpopulations were not so easy to distinguish. If you don't want/need to distinguish between G0/G1 and/or G2/M? you can use less cells. Also it depends from your treatment. If you expect that you will get less cells (in the specific phase of the cell cycle) due to your treatment, I would suggest to take more cells for your FACS staining.
I would suggest to start with the "normal" (withour any treatment) cells in 3-5 different amounts and setup your staining. In this way you will find your own optimum. Take into account that some cells you would "loose" due to fixation and permeabilization procedures. Some cells you might loose due to additional stainings.
Thums up for James.
BTW, I would recommend you Pulse your cells with Edu/Brdu. Click-it kit form Life Sciences works very well in my hands (Edu+PI). In that case, you could clearly distinguish the cells in G0/G1, S, G2/M (PI by itself doesn't do a good job, the cells in different phases will overlap).
Since this method needs permeabilization and fixation (you will lose cells). You need to start at ~0.3 million cells (1million recommended). But for Flow Analysis, as long as you have several hundreds cell in each gate, It should be fine.
It depends on the number of cells in each stage of cell cycle you expect. Keeping in mind that the p is proportional to 1/sqrt (N) there N is a number of events in a population of interest. p 400 and p10000 for a smallest stage (by number of events).
You should analyze 500 000 - 1 000 000 cells. If you determine influence of detrimental factors on the cell cycle, the cell count can be reduced ...
Dear all,
Considering your useful comments (thank you very much!), I did my flow cytometry yesterday. Here is what I found:
Not all of the samples need to contain millions of cells. I realized that for setting up the FACS machine, I need to have only 'one' sample with a lot of cells (millions) stained with PI (in my case). If the other samples contain around 50,000-100,000 cells, you should be able to analyze at least 20,000 events in addition to what you need for playing around with the settings (e.g., voltage), provided that you have already properly set up the machine.
To minimize the cell loss during permeabilization/fixation, you can use low-binding tips and tubes. Being careful while handling the cells during preparation (e.g. not sucking part of the cell pellets when removing the supernatant), will reduce cell loss.
I also realized that at least 10,000 events for cell cycle analysis should be enough, as many of you suggested.
Thank you all again for your help.
Sharif
- Badges
- Science method