http://www.currentprotocols.com/WileyCDA/CPUnit/refId-im0409.html. I do think this protocol is very good.

Hi Fabienne,

Im currently running CFSE assays on T-cells but they are CD4 isolated and CD4 naive isolated cells so Im not sure if doing whole unsorted spleenocyte cultures is a problem? Some one else might be able to answer that. I seed my cells at 1 or 2x106/ml with 1 ml in a 24 well plate. they are stimulated with plate bound anti-cd3 and anti cd28. I take samples every day but only see proliferation after day 2. I run the experiment for 4-5 days and usually use day 4 which has nice clean peaks and good viability.

First thing I would suggest is to optimise your analysis. Gate out the debris and do dead cell exclusion by adding PI. This really cleans up the peaks. You can also do doublet exclusion to further clean up the peaks. Id have a look at that before messing with the conditions of the assay but as I said I have no experience with whole spleenocyte cultures.

I also see that you add the CFSE at 5mM? That should be 5uM. CFSE is toxic to the cells, the rest of your staining protocol looks good.

You should not have serum in your label media. The CFSE will stick to the protein components and give you non-uniform labeling. The labeling media should be warmed to 37 degree before mixing with cells. We generally do cells at 10million/mL and label at 37 for 5 minutes. We use CFSE stock concentration at 5mM. We make a labeling solution in the label media at 1:350 and then add that mixture 1:1 to cells. You need to vortex well before placing cells at 37 degree in water bath. Make sure that neither cells nor label media have any serum. the remainder of the protocol seems reasonable. Dynabeads should be effective. You can do this in any plate size and at an array of cell concentrations. 72h should be sufficient.

Hi Joseph, thanks for your answer - I will try a time course also and see what my cells will look like after more than 72h culture. Can you comment a bit on how you do your CFSE incubation?

Re. the analysis - I use DAPI, and first gate on DAPI neg. cells, then on singlets using SSC-W/SSC-A, and then on CD3/CD8 double positive cells - this cleans up the peaks for all my controls, but the "picure booK" CFSE peaks I would like to see are still missing...

Hi Fabienne,

the problem with dilution peaks can often occur when CFSE is old (even 1 year old). I don't know why... Anyway, the labelling protocol I use for in vitro proliferation assay is as follow: Incubate cells at the concentration of 10E6 cell/ml in PBS 1%FCS with 1uM CFSE (final concentration) for 10' at 37°C, 5%CO2. Wash 2X in cold PBS 1%FCS (5' at 1500 rpm and 4°C). Re-count the cells (you might lose some) and seed. For stimulation, it really depends on your setting. You should include a strong positive control, such as PMA. 72h is enogh if you're looking at a secondary response, otherwise wait until 5 days. I don't use Dynabeads, but Miltenyi microbeads. Are you sure Duynabeads don't interfere with T cell function? The otimal thing would be to sort untouched T cells so that you don't harness them.

Good luck! Giovanna

You could stain 10x10e6 at same time, it should work, however I use to have my SPLCs in 1 ml ice cold-PBS and add the CFSE ( 10mM) i.e. 2x stronger concentration also ice cold: 1:1 my cells: CFSE solution to achieve 5mM at final conc. incubate 5 min at 37 C incubator ( not longer time) take out and add 1 ml RPMI-10%-FCS and fill up with PBS ca 10 ml, centrifuge at 1200 rpm 5 min. Add 5 ml medium and place in incubator for at least 1 hour, wash again and count, resuspend in desired cell concentration and plate.

How u stimulate your cells depends on ur assay, I use to have antigen specific stimulation,

Remember also that u should plate ur stained cells with Unstained SPLCs that serve as normal environment cells, and u have to calculate how many of ur stained SPLCs are CD4/8 respectively ( use ur old FACS data).

I would however recommend u to anyway sort ur CD4/8 or at least T cells stain them and do ur assay.

Good luck

Wiaam

1x10E6 in 1ml PBS/0.1%BSA with final concentration of 5mM CFDA-SE

incubate for 15min in 37℃

quench with ice cold medium after incubation and keep sample on ice for 5 min afterwards

cells number should be based on your research, your could get some useful information from the related articles. I use 10E5 for suppression assay in 96wp with round bottom

I use CD3/CD28 coated beads to stimulate T cells, and its ratio should be based on the culture time and the MFI of CFSE

no agitation during the following 72h culture

to get tidy peaks, just remember :mix the CFDA-SE with the cells immediately

Hi Fabienne,

I've been meddling with this for a while now. I'll try to answer ur questions in an orderly manner:

- how many cells do you label? Which concentration of CFSE gives you the best result? I currently label 10x10E6 in 1ml PBS/5%FCS with 5mM CFSE stock diluted in 110ul PBS.

2-5 X 10E8 cells/ml with 5mM of CFSE is best! the concentration that you use is ok too so long as there is FCS. however, to obtain the best tightest peak, always make sure you vortex for 5-10 seconds once the CFSE solution enters the PBS/cell mixture.

- how long do you incubate after CFSE labeling? I do 5 minutes at RT in the dark, but have seen protocols asking for 10min at 37C.

For this, I normally use 7 minutes in the dark at 37oC. However, to ensure better staining, I repeat the staining step twice!

- do you quench with ice cold medium after incubation and do you keep sample on ice for 5 min afterwards? I just add 5x volume warm PBS/5%FCS and wash 2 times afterwards.

I don't really like putting warm cells into cold media. I always use warm media + 10%FCS for 30 minutes at the end of my 2 rounds of staining.

- how many cells do you put in culture and which culture dish are you using? I have used 96wp with WAY too many cells, and am currently using 12wp with 10x10E6 cells in 1ml T-cell medium...

I think this part is OK! the cell numbers are pretty high for 1ml media. if they grow with the stimulation you are causing them to not unhappily.

- how do you stimulate your cells? I am using CD3/CD28 Dynabeads in 1:1 ratio. Any experiences with lower ratios? I tried up to 1:3... Plate bound CD3 and soluble CD28 have been suboptimal for me.

The ratios are fine so long as the growth media are fine. Its important to remember they need space to grow!

- silly one: do cultures need agitating during the culturing period to prevent dynabeads from settling to the bottom of the culture dish?

I've tried agitating them before, no difference! but make sure you mix them well and gentlely at the begining

- how long do you culture cells for? I have done 72h but only seem to get two very broad and untidy peaks, even with very strict gating on CD3+CD8+DAPI- lymphocytes...

72 hours is pretty good, it could go up to 7 days! try to get a read out every day with a tiny amount of cells. I dont know if you want to try, but you can always add some IL-2 (50U/ML) after 72 days of stimulation to watch them grow further and make sure that the media is changed if it ever gets yucky

We culture mouse splenocytes in a 1:3 cells:CD3/CD28 beads ratio and use 1x10E6 cells/ml in a 6well plate. And we add IL-2. But there are also other agents to stimulate your cells with for proliferation assays. It depends on what you want to look at. Have a look at this protocol: Proliferative Assays for T cell function by Kruisbeek et al, 2004 (Curr Protoc Immunol). It might give you some useful information.

The time point to analyze your cells also depends on what you want to look at. I would take out a small number of cells from the wells every day and analyze them. That way you get a feeling for how these cells proliferate.

I use the CellTrace Violet Proliferation Kit for my T cell cultures as I need the FITC channel for something else. It gives a really nice signal in vitro and in vivo, in case you want to change the proliferation assay.

Try this video protocol, it worked fine for me.

Http://www.jove.com/video/2259/the-use-carboxyfluorescein-diacetate-succinimidyl-ester-cfse-to

Hi, I did lots of splenocytes labeling with CFSE.

how many cells do you label? Which concentration of CFSE gives you the best result? I currently label 10x10E6 in 1ml PBS/5%FCS with 5mM CFSE stock diluted in 110ul PBS.

Normally, I will resuspend splenocytes in PBS with the concentration of 2 M/ml, then label the with 0.5 uM CFSE. The CFSE stock concentration is 5 mM in DMSO.

- how long do you incubate after CFSE labeling? I do 5 minutes at RT in the dark, but have seen protocols asking for 10min at 37C.

I will incubate 15 minutes in 37 c waterbath.

- do you quench with ice cold medium after incubation and do you keep sample on ice for 5 min afterwards? I just add 5x volume warm PBS/5%FCS and wash 2 times afterwards.

No. Just wash the cells by adding 5x volume PBS afterwards and spin cells down.

- how many cells do you put in culture and which culture dish are you using? I have used 96wp with WAY too many cells, and am currently using 12wp with 10x10E6 cells in 1ml T-cell medium...

- how do you stimulate your cells? I am using CD3/CD28 Dynabeads in 1:1 ratio. Any experiences with lower ratios? I tried up to 1:3... Plate bound CD3 and soluble CD28 have been suboptimal for me.

If you want to use 96-well plate, you need to put 1x105 splenocytes into each well and stimulate the cells with soluble CD3 (1 ug/ml)/CD28 (5 ug/ml).

- silly one: do cultures need agitating during the culturing period to prevent dynabeads from settling to the bottom of the culture dish?

- how long do you culture cells for? I have done 72h but only seem to get two very broad and untidy peaks, even with very strict gating on CD3+CD8+DAPI- lymphocytes...

After three days culture, you will see very strong cell division (around 6 division)

I am not used to work with mouse spleen cells. i work with human lymphocytes (PBMC). Usually, I label lymphocytes (or PBMC) ten milion of cells/ml with 1micromolar of CFSE for 15-30min at 37°C in complete medium (RPMI1640 plus 10% FCS). I think you can do it also in an appropriate buffer without FCS, but you should test which is the best condition for your purposes. After the incubation time, I make 4 washes with complete medium at RT and before the last one I rest cells for 10-15min at RT.

Then, I use the cells in my experiments seeding 100.000 cells/well (96wells plate U-bottomed).

Please take into account that CFSE can go outside of labelled cells thus it is difficult to get the same staining (evaluated at FACS) you got the first day just after the labelling.

To maintain the same staining you should put the cells 24h in incubator (at 37°C) and then the day after whash tem and use them in your experimental setting. For human cells this is OK, I do not know if it is the same for mouse cells.

I am not used to perform experiments with beads because usually in PBMC you have also APC and thus it is not necessary to stimulate cells with anti-Cd3 and anti-Cd28.

Usually, using 10-50ng/ml of anti-Cd3 you get a good stimulation in 3-5 days, you can also add IL2 but I think it is better do it after at least 24-48hrs after stimulation to boost proliferation.

I suspend cells at 10 million per ml in PBS and add equal volume of 2uM CSFE diluted in PBS, mix and incubate at 37C for 15 minutes in a water bath, mix every 5 minutes. 37C incubation is important, because CSFE is a succinyl ester form and need cellular enzyme to break it down, otherwise the reaction is very slow. After labeling, you can washed once or twice with large volume (20 to 50X of labeling volume) of medium with 10% FCS to inactivate remaining CSFE, alternatively, if you use 100% serum, for less volume.

If you dynabead, you could use 96 well U bottom plate. I will use 200,000 cells mixed with 100,000 to 400,000 beads. It is better to titrate you beads since every lots are slightly different. That has to do the efficiency of aCD28 coating. Do not agitate pale during stimulation.

For CD8 72 hours is good enough, for CD4 use 96 hours.

Clearly, there are many ways to skin this cat. My lab's protocol is found in this article: https://www.researchgate.net/publication/235442806_Homeostatic_proliferation_of_mature_T_cells?ev=prf_pub

When it comes to "wide" peaks, one thing worth considering is that a consistent sample is important for consistent labeling. That is, if you add the CFSE to the cells in 4C labeling buffer and then put the sample at RT or into a 37C bath, there will be variations in temperature across the sample and thus variations in the rate of labeling. Homogeneous labeling of your cells is important for "tight" peaks. This is why we pre-warm our sample before we add the CFSE.

I think it is also worth mentioning my theory that the DMSO you use may also be an important variable to control. I have resuspended "old" CFDA-SE (aka "CFSE") in new DMSO and generated CFSE peaks with outstanding CVs (aka tight peaks). DMSO is very hygroscopic and the longer you leave DMSO around (with our without CFSE) the more water you may accumulate therein. I have a hunch that this contributes to the issues with using "old" CFSE. Also, as mentioned by Kathrin Warner, CellTrace Violet (we call it CTV) is really nice and I suspect that at least part of the consistency and quality of this reagent is due to the fact that you get a fresh aliquot of DMSO with each kit. Even with CTV we have noticed that the peak CVs deteriorate over time once the reagent is put into solution with DMSO.

Article Homeostatic Proliferation of Mature T Cells

Fabienne, whichever 'DCT' staining you would use, CFSE, Violet etc., it is toxic to your cells, as it permanently binds to some cellular proteins and incapacitates them. thus, it is paramount to use the lowest concentration of CFSE (or alike) possible (i.e. still giving the discernible (higher than background) fluorescence signal after say 8-10 generations. Thus, you should titrate your batch of CFSE (they DO differ!) and choose the optimal staining concentration.

Then, the question is what are you staining with CFSE (or alike) for? Do you want to assess the maimal proliferative potential of your splenocytes? then of course you stimulate with optimal anti-CD3+ anti-CD28 (again, titrate, whether you are using atibodies immobillized to the well bottom or to the beads!), and would add IL-2. Or, do you want to assess the actual proliferation in response to - say - an antigen; there, you would NOT add any external cytokines, as your stimulated splenocytes woulld make their own. Considering the latter, you want your initial culture density to be large enough to rapidly provide IL2 etc to proliferating lymphos, but not too dense, as they would eat all available cytokine before they would make enough to sustain themselves. For human PBMC, I use 1-2 x 10e6 per 1-2 ml of complete medium. I stain for 15 minutes with 0.1-0.2 micromolar CFSE at 37 and then wash in FCS-contining medium (NO cold quenching! It may prevent at least some of the cellsyour stain from entering the division!) - staining them such I can discern up to 10 peaks corresponding to consecutive dividing generations after 5 days in culture (120 hours; this is for human PBMC; murine T cells cycle a bit fasterthan human ones, so if you optimiz the staining/culture conditions, you should be able to see twelve divisions of the CD4+ splenocytes or LN cells after that time (CD8+ are a bit slower). You may want to look up my paper on CFSE stainig method (Witkowski JM. Advanced application of CFSE for cellular tracking. Curr Protoc Cytom. 2008 Apr;Chapter 9:Unit9.25. doi: 10.1002/0471142956.cy0925s44. PubMed PMID: 18770652). Good luck!

Make sure the cells are in single cell suspension. I resuspend cells in 5% FBS for staining with 5 micro-Molar CFSE (final concentration) for 10 mins at room temp and then stop the reaction with plain RPMI at RT.

Make sure you gently shake the tube twice during the ten mins of incubation.

To get one peak the key is to have not more than 10 million cells in one ml during staining.

Proliferation is the best in 96 well round bottom plates with 0.5 million cells per well in 10% RPMI. Do not agitate the cells, they'll be fine. Also avoid looking too much under the microscope. Culture for 72 h is usually the best however if you are not getting many peaks then check your activation stimuli for any endotoxin contamination. Buy the endotoxin free reagents. LPS contamination usually gives such broad proliferation peaks. Also titrate the concentration of dyanabeads or anti CD3/ CD28 used. 10ug/ml anti CD3 (plate bound) and 4ug/ml anti CD28 (soluble) works best for me for T cell proliferaiton.

Good Luck

Thank you all SO MUCH guys, you have all pointed out very important issues! Looking through your answers, I believe the vortexing/ mixing during the labelling is probably (hopefully) the solution to my problem... I will have another go tomorrow! Keep your fingers crossed :-)

you should avoid FCS in the PBS during the staining, add it only to stop the reaction

my sugestionis , because CFSE binds to proteins; stain cells in buffers that have decreased serum concentration(0.1% FBS). Cells which incorporate too much CSFE will have reduced protein function and therefore reduced viability. A short efflux step using 100% serum is included to allow cells to “purge” some CFSE and washing the cells at least 3 times helps produce a tighter initial CFSE peak. CFSE should be prepared in DMSO and stored in small aliquots at -80. It decays quickly, yellow discoloration indicates the CFSE will no longer conjugate to proteins.CFSE is a dye that binds covalently to proteins inside a cell. When a cell divides, each daughter cell contains half the amount of CFSE as the parent cell, which can be read in the FL-1 channel of a flow cytometer. Briefly, pre incubated cells at 37°C for 5 mins, labeled with CFSE at 37°C for 7 minutes, washed and distributed to 96 well U-bottom plates. Peptides or peptide pools are added to each well in duplicate, and tests include positive controls such as SEB and animals with known proliferative responses, and negative controls including CFSE labeled unstimulated cells.

Hi Fabienne,

As we can see from the answers above there are many ways to do this. I work mainly with NOD and BALB splenocytes and have found that different subpopulations of cells label differently with CFSE. Some label more strongly and some less. So therefore you'll have different peaks to begin with which could then "cloud" your readout. If at all possible I think you would benefit from enriching for T cells first.

I agree with those that are recommending not including FCS or BSA in the labeling step but to use the addition of that to "quench" the labeling. Briefly, here's my protocol:

1-2 x 10^7 cells per ml in PBS. To 1 ml of cells, add 1 ml of 10 uM CFSE in PBS (made freshly by adding 50 ul of 200 uM CFSE (stock in DMSO) to 1 ml PBS) while vortexing on medium speed. Incubate on bench top for 5 minutes. Add 10ml of PBS+0.5% BSA and mix to "quench". Pellet the cells and wash one more time with PBS+BSA.

For stimulation of the cells try CD3 alone as well as CD3+CD28. I see proliferation in response to CD3 alone using 1 X 10^6 cells per well in 96-well plate (flat-bottom) and 1 ug/ml of biotinylated anti-CD3 (145.2C11) followed by 1 ug/ml of streptavidin to crosslink. (Higher concentrations of CD3 antibody do not induce proliferation in my hands.) Also, try longer times than 72 hours as all cells are not necessarily behaving the same way. Sometimes the age of the donor mouse matters.......

Another suggestion: Try CD3+CD40 stimulation. (1ug/ml of CD3 and 5 ug/ml of anti-CD40 (1C10)). CD40 is emerging as a costimulus for T cells that is as powerful as CD28 if not better sometimes.

Good Luck!!

Hi Fabienne,

Looks like you have some great advice already, and I think you have hit the nail on the head with your "vortexing/mixing" solution. Thought I would add a couple of references we have now published that pretty much covers all the essential issues here...even a video of how to do the procedure to get tight peaks! :)

J Immunol Methods. 2012 May 31;379(1-2):1-14. doi: 10.1016/j.jim.2012.02.012. Epub 2012 Feb 21.

New and improved methods for measuring lymphocyte proliferation in vitro and in vivo using CFSE-like fluorescent dyes.

Quah BJ, Parish CR.

and

http://www.jove.com/video/2259/the-use-carboxyfluorescein-diacetate-succinimidyl-ester-cfse-to

All the best,

Ben

Sorry to those following the thread. Ben and I were just chatting and deleted our conversation as it doesn't enhance this thread.

Hi, I think I did something very similar.

About stimulation, I coat a 96 wells plate one day before with anti-CD3 (diluted in PBS), 5 microgram/ml, 150microliter/ well at 4 degree overnight. The next day, aspirate PBS and the plate is ready to use.

I use fresh splenocytes and have 3 million in 500 micro liter medium (CLICK's medium with 10% FCS), then quickly mixed with 500 micro liter PBS containing CFSE (10microM) and incubate for 5 min at RT. Wash with medium and resuspend cells with medium contaning anti CD28 (2microgram/ml, 300 microliter) then load 1.5 million cells each in 96 well plates in 150 microliter medium (duplicate wells).

Then incubate cells for 72 hours.

So hopefully it will help.

Hi everybody,

so I have run an CFSE optimization experiment using WT spleen MNC based on your suggestions - the conditions included staining with and without 5%FCS, stimulation with CD3/CD28 beads (bead/cell ratio 1:1), harvesting cells at 24h, 48h, 72h, 96h and 120h and staining with CD3, 8 and 44. Now I just wanted to share my main findings (again, this is all for mouse T-cells):

- Ben's method as suggested in his paper, diluting CFSE in 110ul PBS in a horizontally inverted tube is perfect for the low amount of cells I have been staining (~30x10E6 cells in 1ml PBS). The key for me was to keep the tube inverted, wrapped in foil, until transfering it to the vortexer and vortex at slow speed for 15 seconds, then incubate for 5 minutes at RT and dilute with 10X the volume of 37C PBS/5% FCS followed by two more washes. All CFSE peaks came out nice and tight.

- Re. +/- 5%FCS during the CFSE staining: in -FCS conditions, CFSE is a lot brighter at 24 and 48h, and the proportion of DAPI- cells is slightly higher at 24h. At later timepoints, I saw no difference in terms of number of peaks or cell death.

- CD8 TC started proliferating at 48h, and the optimal time point was 96h. For 96h and 120h I added 1ml of full T-cell medium halfway through the timecourse.

- Adding a CD19 flow-antibody to the surface staining cocktail to create a dump gate did not further improve the peaks.

I am now repeating this timecourse in my disease model using purified T-cells instead of whole spleen. Thanks again for all your valuable suggestions!

CFSE may be tricky and the plots qould be hard to discern dividing generations. Use VPD 450 instead, or separate populations of interest.