Hi Berislav

Generally speaking, amount of antibody is dependent on concentration and not cell number. So you probably won’t need to change volume or antibody amount, just increase cell number in same volume.

Antibodies should always be titrated in any case: see http://depts.washington.edu/flowlab/Cell%20Analysis%20Facility/Anitobody%20Titration.pdf by Mario Roederer.

Hope that helps

Barry.

The number of cells and stain concentration are generally not a proportional thing... You should probably increase the volume to get more cells, but with the data you provided, the cell concentration is the same 1x10E7 cells/mL. Giving that the antibodies are usually added in excess amounts, I would give it a try with the Abs concentrations you have been using.

I typically do this. For flow sorting with higher cell numbers I will do the staining in 1 ml. The concentration of your antibodies will stay the same adjusted for 1 ml as the total volume. You will need the larger volume for the flow sorter as the machine will have reduced purity of you sorted populations if the concentration of the cells are too high.

However the Ab are in excess in terms of concentration when used for flow cytometry studies. From my experience if the the stained molecule is plenty in terms of cellular density and you have a good staining with your Ab it doesn't matter if you stain 106 or 107. But, if you have not an abundant density of molecule you ned to increase accordingly the concentration. I would not use, though 1ml of cell suspension because you dilute the Abs availability. Let's say, we are using for a ten fold cell concentration increase a doubling of the volume. I do really hope that you will make some use of my information and you will succeed in your flow cytometry testing.

I do not fully agree with Filomena. Both the concentration and cell to Ab ratio matter - (that's why we titrate, to optimise staining!), but it also depends on the quality of your ab. I'd test it first, therefore, checking with different quantities of cells (and increasing volumes) whether you have the same ratio of positive cells.

Dear Berislav,

I agree with Filomena. You mean you need more cells for sorting (in terms of number of cells)...but if you have 1x10E6 cells in 100 ul (microliter, right?), it is in fact equal to 1x10E7 cells in 1 mL in terms of concentration. Thus, you just need to combine together 10 samples of 100 µl in order to get 1 mL of sample with a total of 1x10E7 cells.

If your antibody is correctly titrated, you should be able to stain 5 to 10 times more cells in the same volume without any problem. That is in your case, 5 to 10x10E6 cells in 100µl.

Hi Berislav

Generally speaking, amount of antibody is dependent on concentration and not cell number. So you probably won’t need to change volume or antibody amount, just increase cell number in same volume.

Antibodies should always be titrated in any case: see http://depts.washington.edu/flowlab/Cell%20Analysis%20Facility/Anitobody%20Titration.pdf by Mario Roederer.

Hope that helps

Barry.

I would adjust it for the volume and keep the cell concentration/ml constant. However, if you have very high no of cells to be stained for sorting (e.g. >100 Mio) without MACS reduction in advance you should check for higher cell no per ml to save money. since you titrated the antibody you should know to which extent it is dependent on cell no and volume.

in general the amount of Ab needed depends on a no a variables inlc type of antibody, the flourocrome, the antigen you are targeting....

There is no right way of predicting what will happen. The staining depends on the quality of the antibody and the level of expression of the antigen. For each of us, it us different and hence I would suggest you try titrating the antibodies four larger number of cells in a larger volume. Alternatively, to take the easy way out, just increase the amount of antibody you use for staining.

For cell sorting, despite the cell concentration being the same, increasing the volume will definitely help.

Hi,

I do agree with Celine , ratio of cells with respect to Ab does matter in immunostaining. now the question is why you want to increase the no? is it because the sorted cells are low in number for further experiments? in that case you have to increase ab in proportion to the cells. for eg, 1:100 for 1 million than 2:200 for 2 million . after staining might be you can add extra diluent or sheath to dilute the cells for proper sorting.

I also agree with several previous comments.

Titration is fundamental, based on the antibody concentration, not directly linked to the number of cells. Moreover, the conditions of staining, such as time and temperature of staining, as well as the fluorochrome stain index, number of receptors per cell (FMI), the conditions of titration (saturation X separation of "positive" and "negative" cells) etc. So I think that you might experiment a few variations of your staining conditions to sort your cells later.

I normally stain cells up to 1x10^6 in 100ul, up to 1x10^7 in 200ul, and if more than that up the volume to 300 or 400ul. We normally use our (mouse) antibodies between 1:200 and 1:400 dilution. I wouldn't recommend staining in 1ml, unless you don't do much Facs as the amount of antibodies needed for this volume is quite high, and would quickly become a financial issue for most groups.

For sorting, the volume you resuspend your cells in depends on the machine you're using, and the speed at which you can sort your cells. If you are sorting a very numerous cell population you can go a bit faster as it doesn't matter if you end up loosing some your cells. But if the population you're after is very small, as said above, you don't want to go too fast. You can always resuspend say in 500-600ul, acquire a small sample to check the acquisition rate and add some buffer if too fast. Hope this helps!

Hi,

I like the answers of Gérald and Monica! For your initial questions, it would be very helpful - to give more "exact" answers - if you give the antigens you want to stain: because some you are able also to increase the cells and volume without nearly an effect (see Monica).. but always it depends! @ Stephane: how do you characterize "correctly titrated"?! Are you referring to Current Protocols in Cytometry (see attachment)?! Because often we use a little bit lower amounts of antibodies for e.g. CD4 / CD8 etc (to save money!!): the negative cells can clearly be differentiated from the positive ones.

dear Berislav,

I agree with most colleagues: probably you won't need to increase the ab concentration if you keep working in 100µL. Anyway, do some controls as suggested by Mario R (mentioned by Barry)

Hello,

The efficiency of staining depends on antibody concentration. To spare Abs, try to stain in the least volume and high (titrated) Ab concentration. We mostly use BL antibodies (best value for quality/price), and mostly stain in 50ul/0.5ul Ab (of course, it works for most Abs, but for some we need increase amount of Ab according to titration). For HSC analysis, we stain 10 mil. cells in volume 50-100 ul, the Ab concentration same as for 1 mil. cells. The increase of staining volume has no sense, the cells sediment and staining depends on Ab concentration in the vicinity of cells. So decrease volume, increase cell and Ab concentration according your titration. Do not worry with higher cell concentration.

Hello Berislav,

If there was a linear relationship between cell number and antibody titre, required to adequately resolve populations for sorting most of my users would be unable to afford to sort!

You should be able to make a guesstimate on how many cells your current antibody concentration can optimally stain from your titration data. The maximal staining index is usually 3-10x the concentration of antibody that will give you resolution of a positive and negative population. Stephane is probably pushing the envelope at x10 cell number I usually find x3-5 retains sufficient resolution for sorting. However this will be effected by the percentage of positive cells, the density of the antigen in the population of interest. I would not have more than 5million cells in 100µl without gentle mixing

According to some previous replies, I think that the amount of antibody you should use is dependent on the amount of antigen (as usually). If you stain more cells (and thus more antigen) the amount of antibody should be increased accordingly. usually, one does not know how much antigen is present on a cell. To get an optimal staining are relevant also the avidity and affinity of the antibody (and those of the second reagent you use if you do indirect immunofluorescence staining). In my experience, you can stain 10x106 cells in 200microliters no more. If you use a commercial antibody, usually with 50microliters or less (considering the whole cell population as positive for the antigen) of undiluted directly conjugated antiboy in 200 microliters of medium you should get a good staining. Titration of antibody is really important, usually one can diluite an IgG1 at 2micrograms of antibody/ml in medium and use 20-40ul to stain 1x106 cells in al volume of 100ul. Usually, a nice antibody works fine at 20-2-0.2ug/ml without relevant differences, thus one has a wide range of concentrations with optimal staining.

Hello,

From my experience, it will depend on your antibody.

I would recommend first for you to check what is stated in the technical data sheet of that antibody.

Since you are already familiar with the antibody expression profile in particular to those cells, you would probably need to add more antibody if you have a weak expression and keep the antiody amount the same if you have a strong clear shift.

I do not think you should chnnge your working volume because you will dilute more the antibodies concentration. Make sure your cell pellet/suspension does not have clumps when you add the antibody. Off course I agree with everyone that titration would be your ideal choice but I do not this it is absolutely necessary in your case.

Also I would recommend when you are sorting to make sure you filter first. You probably should gate on your singlet as your cells will tend to clump and stick together at higher densities and that could give incorrect expression profiles.

Hope this helps

I agree with Gérald Grégori in that you can stain 10 samples & then mix them for sorting. Because I belive the concentration of Antibodies to cells is suppose to be in the same proportion. Otherwise why would we go into the process of titration?

@ Jochen. For me correctly titrated is by using the stain index method as described for example in this application note from BD

What is the volume required for sorting? If it is not stated in the protocol, I would scale up the antibody, without adding extra volume. If you need extra volume for the sorting, than add it right before you start the procedure. Staining usually decreases in big volumes.

Most of the suggestions are quiet relevant. I'd like to add a message although you might be aware!

The dilution depends on whether you are using direct or indirect fluorescence. If your primary antibody is conjugated directly to a fluorophore, the signal to noise ratio will be minimal and by increasing the cell number by doubling, will get you the similar result considering, you are adding excess antibody to the available antigen present in the cell (which we usually do).

On the other hand, if you are using a secondary antibody conjugated with the fluorophore, you have to work out a right titration to avoid non-specific binding depending on your samples- you have to have a systematic approach. Some cells may express more antigen than other cells (especially, if they are primary cells-you are bound to find variations) in which case, you might have to increase the antibody proportionately (remember, the volume the cells are suspended determines the final dilution of the antibody (ug/? - check the manufacture's sheet . Please keep in mind that too less an antibody to antigen could result in false negative and too much could end up being false positive or non-specific binding.Whatever you decide to do, keep a log! G'Luck

In my experience, the concentration of Ab should not be changed if you only need differentiate positive and negative cells. However, if your experiments require to find out whether the expression of a protein increase and if this increase occurs in a differential way in different cell populations you should make sure that enough Ab is being used.

I will echo Brahma Mulugu. Most of the suggestions make sense. It all depends on your antibody and the condition you do the experiemnt. We once used a primary antibody conjugated to fuorophore and did not increase the antibody with doubled cell numbers in the same volume. The results were good.

Hi

I have found that you don't need to increase the amount of Ab or staining volume in direct proportion to the increase in cell number to be stained. For 10 million cells I would resuspend in 0.5ml staining buffer and add 2 to 3x the amount of Ab you would normally use for 1 million cells. Checking that the peak/median values for your larger cell number is comparable with that using standard conditions should reassure you that you're in the right ball park

cheers

kay

The amount of antibody normally you used in small scale is supposed to be excess. So, there is no need to proportional increase in volume and microgram of Ab with the cell number. You can manage to stain 10-fold more cells in 250 to 300 micro-liter incubation volume using 5x antibody that used earlier (same lot).

Alongside what Barry says, in general I found that I can use the same amount of antibody that I would use for a 10^6 cells for up to 2x10^7 cells (note that I stain directly on pellets) without any major failure using many different commercial antibodies. There may be minor differences from antibody to antibody depending on the epitope expression, antibody quality, cell-type you are trying to stain, etc.

As somebody said quite rightly, in certain circumstances this may affect the capacity to distinguish between cell populations which are expressing the marker of interest at different degree; and if you go for a marker which is weakly expressed (and/or the antibody is not one of the best) you may experience few problems (like high background staining, poor population discrimination, etc). I usually take some precautions to tackle these problems: I found very useful to do a couple of sorting trials to make sure everything looks fine and I am detecting the population that I want, before starting with the actual samples. Frequently, if not always I also run a parallel stain with few cells and compare it to the high cell density stained unsorted sample on a different machine just to make sure you can detect all the populations that you would normally expect, as a formal validation; I normally run that control in parallel to the final purity checks on the sorted population. Finally, if you decide to stain on pellet it may be worth having 2 mM EDTA or similar (see MACS buffer) added into your washing buffer and filter through a 40 micron nylon screen the final sample, prior to the run in the cell sorter. Hope that helps!

Antibody concentration in a solution is not dependent on the amount of cell but is rather dependent on the concentration of the antigen on the cell surface. You must know your proteins characteristics and concentration before you can determine amount of antibody

Dear Berislav and all, I have read all of your remarks with interest and I have to say that apparently everybody has their own standards... which work for them ;-). But seriously:

Berislav, I take it that you want to sort a positive population, so your goal is to distinguish it from the negative one(s), and the actual hight of the +ve signal is not important, as long as you can safely distinguish the +ves from negatives, right? OK, if so, and if you had titrated your antibody/fluorochrome conjugate against your cells of interest when staining your million cells, you should know how much Ab to use for decent +/- separation. Titration (if done right, which is now more often not the case) should give you the rigt proportion of Ab molecules to the numbers of detected cellular antigen molecules, with an excess ( not all Ab would bind the ag at any time, it depends on the dissociation constant (inter alia). Eventually, in a propoerly titrated sample after staining and before washing, the majority of Ag molecules should be bound by the Ab conjugate, and some conjugate should remain free. Thus, if you add more cells, that free conjugate would bind them and show the positive cells. So, in practice, if you want THE SAME level of staining of 10 millions of your +ves, you need to use somewhere between 5 to 10 times the amount of Ab conjugate you used for a million cells, and do it in the aproppraitely increased volume of the cell suspension (keeping the conjuugate concentration same; that way the probability a conjugate molecule would meet 'its' antigen molecule will be the same as initial, and you would not have to say prolong the staining time. If however you would be satisfied with lower (but still distinguishing + from -) staining, you can probably stick to the original volume of conjugate and of (in theis case 10x denser) cell suspension, and it should still work (and, which may be also important, will save you some money ;-).

Hi all, thanks a lot for your comments! They all really helped a lot!

To sum-up while assuming that all Abs are titrated for initial staining procedures (and we are trying very hard to titrate them correctly):

1. I could increase cell concentration for 5-10x, keeping the volume for staining and Ab concentration the same.

2. If needed, volume for staining can be increased up to 500 ul, depending on the number of cells (i.e. 200 ul for 10E7 cells, 400 ul for 10E8 cells), but the concentration of the Ab should stay the same.

3. After staining with larger cell number and/or volume: it is not a problem If the signal for my positive cells is not as strong as for cells stained for normal protocol, as long as negative and positive populations are nicely separated.

4. This should in general work for highly abundant Ag (i.e. CD3 or CD4), while for markers expressed at low level I should check staining with larger cell number and, if needed, titrate Abs for larger cell number and/or volume to get optimal staining.

Once again, thanks a million to everybody who contributed!

Just to add up, if you have marker expressed at low level, try to traget them with antibodies coupled to strong fluorophores, this will increase the signal to noise ratio and provide a better population discrimination (ie go for PE or APC coupled to low level expressed markers, rather than with APCCy7 or PerCp).

And as mentionned before if your original staining provides a good separation between the autofluorescence and the signal (check with your titration conditions), increasing the cell number will reduce the number of antibodies per cell, and results in a signal decreases, but if the original signal was strong enough, it will still be OK to differenciate your population. And if you're not sure, before doing your sort, test your conditions on an analyzer and don't loose precious samples (and money) on a big experiment if you are not sure about the expected results. (use lower cell number along with smaller staining volume to mimick the cell-staining ratio you plan to use with your sort and run some preliminary tests...)

I hope this helps,

Fred

Nice trick for staining test, Frederic! Thanks a lot for a suggestion!

Hi Berislav,

It looks like you are in a good hand and everyone here have given you very helpful suggestions. I hope that your experiment goes well. Anyway, I would like to add a small note based on my own experience. When I used too high of cell density for my FACS sorting, I noticed that the cells tended to aggregate to each other. I was recommended a commercial FACS dilution buffer to prevent the cell aggregation, as this will yield a lower sorted cell number. Hopefully this doesn't happen to you. But in case you are experiencing a similar problem, you might want to consider the use of the dilution buffer. Good luck!

Cheers,

Sylvia

depends on your source of antibodies as some need scaling up while other do not. Go back to your antibody data sheet.

Hi Berislav,

In my experience if you significantly increase the number of cells and require the same level of staining you need to increase the amount of antibody but as mentioned by other researchers it isn't a directly proportional relationship. If staining level isn't important and all you need to do is distinguish +ve from -ve you can use less antibody. A preliminary test is a good idea.

Regards,

Michelle.

Berislav,

lots of good advice; one other important parameter not addressed at all. Here are my suggestions:

1. increase volume ensuring cells are well suspended for sorting (s. Sylvia). Stain at least 200-250 ul/10E7 cells.

2. Staining depends on cell number and optimal concentration (s. multiple advice above) but also on the percentage of cells you are trying to stain. For instance, if you stain 80% of cells, and if staining is bright more Ab will be needed for saturation, you definitively adjust Ab amount probably above your typical concentration ( x/100ul for 10E6, then approx. 4-5x/200-250 ul). However, if you stain only 2% of cells you don't need to change anything (keep 2x/200-250ul). Reasonable guessing - based on the straightforward criteria mentioned above - should work.

Klaus

The best thing to do is to titrate again your antibody for the new cell number, and not use just "10 times more antibody" to 10 more cells. Sometimes double antibody is enough for 10 times cells. Thus, my suggestion... just titrate it!.

Thank you all for useful additional comments!

As a final comment - we re-titrated all Abs we need for cell sorting on scaled-down cell sample/volume according to suggestion by Frederic Grosjean. It turned out that in our hands for our samples. the concentration of Abs for highly expressed proteins (i.e. CD3, CD4) did not change [although we will increase amount of Ab as we increase the volume for staining], but for Ags expressed at the low level we had to completely re-adjust the concentrations of Ab.

Once again, thank you all for your comments and suggestions - they were really useful!