Using 0.1% saponin detergent at room temperature  would give you the better results for permeabilization step after paraformaldehyde fixation. The possibility of cell loss is higher during washing steps. Thus after centrifuge, decanting of supernatant would reduce the cell loss. However the decanting step should be quick. Single wash with 1 ml buffer is enough to remove the unbound antibodies. All these steps can be done in flow tubes. As said earlier, increase the centrifuge speed after the cells are fixed.

Thank you both! We are actually spinning pretty hard, around 1100xg. Staining is being done in 1.5 mL eppendorf tubes with 1 mL washes. Fix/perm step is being performed with the eBioscience FoxP3 staining set (30 min. fixation) because we are looking for nuclear staining. After washes, we remove the supernatant by suction, leaving ~30-50 uL behind. Has anyone found that a different nuclear staining set is better or worse than the one from eBioscience? Is 30 min. fixation optimal? How about aspiration compared to decanting, is one better than the other? I really appreciate your help!

I personally have done the nuclear staining of transcription factors in PBMCs. The protocol I followed is as follows

Take cells in the flow-tubes (0.5-1 X 10^6), Wash with PBS (1000 RPM, 10 min), Fix the cells for 15 minutes at 4 degree Celsius with paraformaldehyde (1% prepared in PBS ), wash with PBS, Permeabilize the cells with 0.1% saponin for 5 minutes at RT (Don't use chilled saponin bufffer for this step), Add saponin buffer equal to the staining volume (20 to 100 microlitre), Add the antibody, incubate it for 30 min at RT, wash with saponin buffer, resuspend in FACS buffer. 

For every washing step decant the supernatant and gently touch the rim of the tube to tissue paper.

This procedure should work with low number of starting population of cells and you can acquire more that 0.1X10^6 events.

Hi Sarah,

I also had this problem; the cells get quite slippery when fixed! I know it sounds weird, but try using different tubes (Corning: #MCT-150-L-C); you can see the pellet much better and even using methanol fixation is working fine! http://www.ncbi.nlm.nih.gov/pubmed/23449254

good luck!

For cytokines you can use self made buffers (everything on ice) we first wash with PBS, fix 4% paraformaldehyde (4 minutes); wash with 0,1% saponin/0,5% BSA, permeabilize 0,1% saponin/0,5%BSA/10% FCS 10 minutes wash (0,1% saponin/0,5% BSA) and stain intracellular.

FoxP3 in our hands does not work with this protocol but works fine with the eBioscience kit (manufacturer's protocol)

Fixation/permeabilization is harsh for PBMC methanol > intranuclear kit > PFA/saponin (last one most gentle)

We always stain in 96 well roundbottom plate, spin 2 min 440xg, remove supernatant by flicking the plate. Flicking: in a single motion, move the plate upward, turn it over, and bring it straight down, stopping above the sink, with plate still turned upside down gently push against a tissue to remove drops.

I agree with many others, eppendorf tubes do not allow good cell pellets and when you pipette off the liquid the current resuspends the cells. Change tubes to a FACS tube (can pour off supernatant easily), 2ml screw cap (pellet forms on side where straight tube becomes V rather than in bottom so you can aspirate the liquid without disturbing pellet) or V or U bottom plate (flick off liquid and tap on tissue to remove supernant). Any of these improvements should help if your loss is during the washes.

One simple way can be either extending the time or raising the speed for spinning down  the cells after each wash.

Hi Sarah,

Here are some suggestions based on my experience of developing combined 11-color transcription factors/intracellular cytokine staining panel on 0.1 - 0.2 million cells:

1) We used BD Transcription factor buffer set (#562725). It works nice with listed TF including FoxP3 and other nuclear proteins.

2) We always used V-bottom 96-well plate for staining (NerbePlus, #10-11-0000). Simply transfer your cell culture or distribute cell suspension in 96 wells and do further staining procedure with 150-200 ul wash buffer volume. For permeabilization, 100 ul of perm-buffer is enough. Buffer preparations as per kit's protocol.

3) For final acquisition, you can keep 100 ul final volume, which is enough for acquiring cells between 0.05 - 0.5 million cells (yes, I am not kidding. It works perfectly)

4) We used small tube for acquisition (Sterilin Plastic Round Base tube, Thermo Scientific, #RT15). Transfer your stained cells at the last washing step into this small tubes and put this small tubes in normal BD FACS tube.

I hope, it will help you!

Best Luck! ;-)

Jay

Wow, thanks for all the great answers! We did manage to get good numbers back this time round by doing the following:

1) Use swinging bucket rotor

2) Leave ~200 uL supernatant after 1st wash and leave ~50 uL supernatant after 2nd wash.

I see though that there are many suggestions from everyone here for how we could do things differently. I have always been hesitant to decrease fix/perm volume because both the eBio and BD sets suggest 1 mL fix and 1-2 mL perm wash buffer, which of course does not fit into a standard eppendorf tube! It is good to hear that we can decrease these volumes. I assume that spinning in plates should improve cell recovery since the depth of fluid is much less?

I agree with all the comments about the tubes eppendorfs aren't ideal.  Because i do a very large amount of samples/ tubes all at once i do all my staining in a v bottom 96 well plate (saves time and reagents!).  I also use the ebioscience Foxp3 kit and get really good results.  You wont get got intra-nuclear staining with  a 30 min perm though.  This works beautifully for cytokine staining but you need a longer perm step for nuclear antigens.  I normally do mine overnight at 4oC.  It isn't necessary to do quite such a long perm a couple of hours in the fridge will do it, but logistically overnight fits in better with our protocol.  

As already mentioned: Fixation/Permeabilization (using buffers from either eBioscience or BD) in 96 well plates with reduced volumes works perfect. In addition to such advices in sample preparation, I would like to contribute a rather ´technical´ remark regarding flow cytometrical acquisition: you should take in account that the scatter (in particular FSC) is changed considerably upon Fix/Perm, i.e. you should increase the FSC Voltage/gain - otherwise you will "loose" your cells during  acquisition...

We regularly use the BD and eBio fix/perm kits and one thing we found reduced our cell loss was simply to reduce the brake speed on the centrifuge for all the spin steps after the perm-buffer is added.  Good luck!

Dear Sarah:

    Except for above constructive suggestions, my long experience in intracellular protein staining would added several more:

     1) If you sample number is not so big, you can transfer your cells into 1.5ml eppendofs tube, perform all your wash, staining in the tube,  spin down cells in centrifuge gently, and discard your buffer by pippett. There is little chance to lose your cells then. If your sample number is very big, then you can try to do this in V shape 96 well plate. It is a little bit harder to take away buffer.

     2) After fixative treatment, especially methanol/ethanol containing fixative, the cells turned lighter, if you wash in PBS, most cells can no spin down. In this case, PBS with 10% FCS can avoid this problem.

     3) Fixation/Permeabilization  from most company will do your job, but if you want to optimize your experimental conditions, you can try to make buffer by yourself

   Good luck.

Dear All,

We are trying to perform Flow FISH to measure telomerase activity in PBMC. Unfortunately, we have problems with fixing, permeabilization and hybridization steps. We used a FITC PNA probe from pangene. But both stained and unstained cells showed the same result which means either permeabilization did not happen effectively. Probe concentration used was 0.3ug/mL from 50uM. Fixation was done with 4% formaldehyde in PBS and permeabilization solution has 100mM tris HCl, 50mM EDTA and 0.5% triton-X 100. Hybridization buffer was made with 5% dextrose, 10mM HEPES and 0.1% BSA. We do not have any commercial kit and require a protocol with these reagents.

I know this post was 3 years ago, but I had the same problem as the OP, and while many of the recommendations here seem to help, I think the answer that finally got it to work was the recommendation from Shiqiu Xiong to add 10% FCS to the PBS. Not sure how it works, but I assume the cells adsorb the proteins from the FCS and becomes denser(?). I have to highlight that I was using methanol fixation, since I am trying phospho-STAT staining (the commercial methods with fix/perm buffers already work well for other stains like Ki67). To my surprise, the cell pellet can easily be seen with the naked eye, and even looks larger than before permeabilization. Before, I was getting only something like 20K cells from 200 million starting cells.

I used to also lose cells in fixation, however if you dilute the fixative a lot before spinning down (eg. if you added 100ul fixative dilute in 1000ul PBS ) you won't lose a lot of cells and that it is because the fixative tends to be viscous slowing down the settling of the cells during centrifugation.

another solution to prevent cell loss during washing steps is to move away from pipette tips for supernatant removal and rather use a split centrifugal force device, like the one described here

http://www.siliconbiosystems.com/vrnxt