What kind of measure do you want to calibrate? DNA content, enzymatic activity, ...?

Usually you do not add a calibrator to the sample. You want to standardize your cytometer PMT settings. Once that is done, your data is what it is and differences are either due to patient differences or procedural issues. How you calibrate your machine depends on the type of machine you have. Here are a few references:

www.ncbi.nlm.nih.gov/pubmed/22948490

www.dako.com/08065_15dec05_guide_to_flow_cytometry_standardization__calibration_and_quality_control_chapter7.pdf

www.bangslabs.com/sites/.../pdf/SS%20Flow%20QC%20%26%20Std.pdf

You should calibrate your flow cytometer with raimbow beads.

The optimal calibrator depends upon what you are measuring. Many of our assays at Trillium Diagnostics are built around the concept that internal calibrators and controls are the most informative and precise method of doing quantitative measurements, especially if you are crossing different lots of reagents. My suggestion would be dependent upon your assay panel, but feel free to contact us at www.trilliumdx.com.

Regards,

Bruce Davis

What you can use is special beads that you can found to BD or Beckman coulter.

First time, you made PMT seetings and you pass a sample with this beads.

Then you have the MFI of your beads for eauch color. You pass your patients samples and at the end you pass the beads again, then you should found the same result as before.

3 month later, you pass the beads with the same protocol, and you have MFI, if it the same, it's ok. If it's different you should adjust your paramaters (PMT) to have the same brightness.

At each time with thissystem,, you will have a reproductible results.

Beads are Flow set for beckman.

Every time you should control the alignement with flow check.

Some program in the flow cytometer computer are design to check your settings.

For the calibration of the flow cytometer, fluorescently labelled beads can be used. Basically calibration is done to change the PMT settings to avoid later problems in the analysis like bleed through of the fluorescence of one fluorophore into another filter.

Remember, it is better to calibrate the instrument before every experiment and the PMT voltage settings once set during the calibration must be same through out the acquisition of that particular experiment.

As for the controls, you must have an unstained sample to do an initial gating for the forward and side scatter range of your cell type. And you must have an internal negative control generally referred to as "isotype control" where you will have an antibody which is the same isotype as your test antibody and is tageed with the same fluorophore as your test antibody. The only difference between the isotype antibody and the test antibody is the test antibody is specific to the marker you are looking at (assuming you are doing surface marker staining) and the isotype is simply a non specific antibody but the same isotype and same fluorophore as the test antibody.

So if there is a fluorescence signal in your isotype treated cells, it is due to the non specific binding of the antibody. essentially at the end, you have to subtract the no of cells positive in isotype control from the number of cells positive in the test sample, which will give you the cell count actually positive for the marker you are looking at.

Thank you very much your responses.

To be more precise, we want to reach a quantitative determination of a specific surface and intracellular antigen by flow cytometry. We want to compare the quantity of membrane surface and intracellular antigen present in cells obtained from different patients, before and after treatment. So far, we are using a monoclonal mouse primary antibody and phycoerythrin coupled secondary antibody. An internal or external point of comparison would be of particul help for us to compare the assay in samples obtained from the same patient and/or between patients from the same group.

Again thank you all for your advices.

The results obtained from different patient samples may be different by total number of analyzed cells and by the level of the fluorescent signal from the cells of interest.

In the last case the level of cells fluorescence might easy be compared with the signal from calibrated fluorescent microspheres (MS). Several companies produce such MS of different size and fluorescent properties. For example, some time ago I have successfully used for similar purpose latex AlignFlow microspheres produced by Invitrogen (probes.invitrogen.com) due to very good CV of their fluorescent signal. However, application of MS does not avoid possible mistakes of the registered numbers of cells of interest in the patient sample, as it needs to add each time exactly the same number of MS. To reduce this mistake one might carefully vortex MS and divide them into equal aliquots for each series of examinations mixing in the same proportion with patient samples.

For the flow instrument setting i use either stained beads or i use unstained and single stained samples.

Hi Romuald, for the alignment of your instrument I think that you need to refer to the Instrument manufacturer (Beckman, BD, Partec, etc.). For example the flow check beads from Beckman are not useful for Partec Instruments, etc. for your determination, I think it is enough to mark with and without permeabilization your cells with only the secondary antibody to assess any nonspecific binding it with cellular antigens. If you don't know the normal expression of the antigens, could be useful to use in the experimental sessions a basal control, unstimulated, as for the preliminary tests, in order to verify the repeatability and stability of your settings. I control alignment of my instruments with beads at least once a week. All the best.

The best procedure is to run commercial beads (like rainbow beads or similar) and adjust you PMT voltages. Perform a good calibration at the beggining of your project and use the control beads to maintain the PMT voltages within acceptable values (you can create a Levy-jennings graph to control this). Once you do this and use the same antibodies (take care with lots especially if using tandem conjugates like PE-cy7) you are totally able to compare the results over time. You may find usefull to visit the web page of Euroflow (www.euroflow.org). It is a group which dedicated to standardize immunophenotyping panels for hematological malignancies. They published a set of papers in Leukemia with the panels and also standard operating procedures for what you want. They provide the papers free. I hope this was helpfull.

In addition to Jose very good answer I might suggest that some commercial company can provide you with standardised preserved samples (human, if this applies to you) which you might want to run in parallel to your samples to check that your staining protocol as well as your instrument set up don't drift from day to day. A simple Google search will bring up the product but please feel free to contact me if you are unable to find it (I do not like to endorse commercial companies in an open forum). This is OK if you look for an antigen which is moderately to highly expressed and present on non-rare cell populations. If you are working with rare populations or a low expressed antigen you might have to accept that your data will be less precise since small variation in your sample handling can have a larger impact on your final data.

Thank you all for your advices

Romuald

You could use a cell line. With the cell line you could maintain it and label it each time you do the experiment, or you could fix a batch in methanol at -20C. If you aliquot it you could run the same fixed sample each time you run your patient samples.

Hi Romuald,

Yes, I did a calibration in flow cytometer. The signals are possible to check using microbeads, and there are many types of beads available. You can find some advices in these articles:

1.Sutherland DR, Marsh JCW, Davison J et al. Differential sensitivity of CD34 epitopes to cleavage by Pasteurella haemolytica glycoprotease: implications for purification of CD34-positive progenitor cells. Exp Hematol 1992; 20:590-9.

2. Schwartz A, Marti GE, Poon R, Gratama JW, Fernández-Repollet E. Standardizing flow cytometry: a classification system of fluorescence standards used for flow cytometry Cytometry 1998; 33:106-14.

3. Kraan J, Gratama JW, Keeney M, D’Hautcourt JL. Setting up and calibration of a flow cytometer for multicolour immunophenotyping. J Biol Regul Homeost Agents 2003; 17: 223-33.

Good luck

Hi Romuald,

Quantitative Flow CytoMetry (QFCM) has been my favorite approach for over 30 years. I may not be able to explain all tricks in the forum (lack of space possibly) but I can confirm your question is easily solved. Since I understand you are french, I attach a french speaking review article from 2004 (with abstract in english). My first paper on that subject was: Poncelet P & Carayon P., J Immunol Methods 1986.

At this time, we used cell lines expressing diverse levels of a cell-surface antigen (in practice CD5) that was measured using radiolabelled anti-CD5 monoclonal antibodies (MAbs). We compared the mean Fluoroescence intensity (MFI) obtained after saturation of these cell lines using indirect IF (with anti-mouse IgG-FITC at that time - later we showed that other flluorochromes or biotin + streptavidin-fluor would also fit). with the amount of bound 125I-MAb measured by cell RIA. Then we demonstrated that this calibration based on mouse IgG2a CD5 MAbs was also true for other cell surface antigens (e.g. CD4, CD8, CD45 ...) and for all mouse IgG isotypes (not IgM though!). I refered this technology as Quantitative Indirect Fluorescence Immuno (QIFI) assay .

Then, we later developped mouse IgG-coated beads, themselves calibrated against CD5 cell clones to render the calibration easier for routine application. This was called QIFIkit(r), made of a series of individual beads with known amounts of mIgG. At that time I created a biotech company named BioCytex (Marseille, F) to facilitate commercialization (see first historical flyer attached).

Even easier to use was a mixture of 5 levels altogether also called QIFIkit and commercially available from DAKO for about 20 years now. You can find a number of papers looking for this term QIFIkit. Some are listed in the attached review chapter (Ann Biol Clin 2004, 62 : 53-7). One I wrote with co-authors including JC Chermann and F. Kourilsky in 1991 in J Immunol made the basis for a well-known biological reference i.e. ~50,000 CD4 molecules per CD4+ lymphocyte, independent of AIDS virus infection. Also CD45 level on human lymphocyte (~200,000 molecules/cell) is a very stable biological standard (see Bikoue, Janossy et al, J Immunol Methods 2002, 266, 19-32).

A good source for QFCM is dated 1998, a suppl volume 33 of the journal Cytometry that was devoted to Quantitative Fluorescence. See in particular Serke et al, Cytometry 1998, which compared QIFIkit with a few other approaches proposed.

If you have access to a popular book in FCM labs, Immuno-phenotyping, CC Stewart & JKA Nicholson (eds), Wiley-Liss 2000,I wrote a review chapter exactly on your question:

Poncelet P et al Clinical applications of Quantitative Immuno-Phenotyping, p 49-104.

Later we have adapted this QFCM approach to the measurement of antigens on human platelets (see www.biocytex.fr, Platelet kits and ref 14-34 in Ann Biol Clin 2004).

If you have access to a popular book in FCM labs, Immuno-phenotyping, CC Stewart & JKA Nicholson (eds), Wiley-Liss 2000, we wrote a review chapter exactly on your question: see Poncelet P et al Clinical applications of Quantitative Immuno-Phenotyping, p 49-104.

Basically, we listed there a huge number of examples where QFCM allows antigen quantitation on an absolute basis allowing i) comparison of density data in time, ii) among different patients or patient groups iii) in aging iv) modifications by physiology (e.g. activation or apoptosis or senescence ...) v) modulations by pathologies (e.g. CD38 over-expression in AIDS and other viral infections) vi) modulations induced by therapy and the potential of QFCM to help monitoring clinical trials with e.g. therapeutic MAbs. This is our actual main activity under contract with major Biotech Pharma companies (see e.g. Lapusan et al, Invest New Drugs 2012).

And to prove that , despite being first described 30 years ago (1st poster in 1982) our QFCM approach is still in fashion, I attach an abstract presented as a poster and discussed in a workshop (Receptor Occupancy based on FCM) at CYTO 2013, the congress of ISAC in San Diego at the exact minute I write this message.

If I can write this message at the same time is simply because I had an accident last week and I am stuck at home with a broken leg at the time my colleagues discuss that point in San Diego -USA).

Let me now thank you for giving me the opportunity to advocate (again) for cell-surface antigen quantitation with flow cytometry. And as the well-known guru of FCM wrote once: "Put the Metry into Flow Cyto-Metry !"

Hi again Romuald,

Sorry for few bugs in my previous answer. This is my first answer in Researchgate and I was not used to do it.

The FCM guru I wanted to cite is Howard Shapiro, who wrote a series of authoritative books (bibles) on FCM.

At disposal to anybody wants to reliably apply QFCM in basic or medical research.

visit www.biocytex.fr looking for either CellQuant Calibrator or Platelet kits.

For special questions, contact me at: [email protected]

Bonne chance ! and Go with the Quantitative Flow !

Philippe

I fully support the excellent work of Philippe and he is truly a pioneer in this area. Another approach was championed by Dr. Abe Schwartz whose concept still lives with his IP being acquired by Bangs Laboratories (www.bangslaboratories.com). I have further evolved this concept to using FITC coated stabilized beads as internal calibrators for an assay for infection/sepsis measuring neutrophil expression of CD64, complete with software for listmode file analysis in a near automated fashion. More information can be easily found on our website (see Leuko64 assay at www.trilliumdx.com).

I too second Philippe's recommendation to the reading of Howard Shapiro's authoritative textbook. If you have not read at least one edition, you are missing true knowledge.

My clear bias is that for clinical cytometry to find it's full potential in cellular diagnostics, especially in laboratory hematology, assays should have internal calibrators, internal controls if feasible, high quality reagents, and automated software for data analysis.

Look forward to a Cytometry B publication this summer with the publication of a special issue featuring the ICSH/ICCS Guideline for the Validation of Cell-based Fluorescence Assays. The proper use of beads for both instrument set-up and QC and assay calibration for quantitative measurements of molecules on cells is thoroughly discussed.

Regards,

Bruce

The discussion is slightly bringing together the ones who spent a significant part of their life trying to bring QFCM to real clinical applications. Leuko64 is typically one excellent example of what can be achieved with real diagnostic value. And as Bruce wrote, any IVD approach should include calibrators AND controls, calculation software (or pieces of ... I mean Excel spreadsheets) and optimal reagents fitted with the rest. If your question still remains at the level of initial clinical research, you may however begin with calibrators and well selected (SATURATING) MAbs.

Nice to hear from you Bruce !

Philippe (again).

To be clear with everybody, my first-ever registration on researchgate was associated with a bug in my profile. I am not linked to Warwick Univ (although it may be exciting to be) but with BioCytex (a stago group company) located in Marseille, France.

www.biocytex.fr

Philippe

Philippe - Does that mean you also disavow that rather impressive rating of 71.45? We mere mortals are very impressed.

HI bruce,

I don't know the rules of the game and would not be able to say where this 71.45 which seems impressive comes from. The only information I get is that this forum interests good people, since you are here.

Presently, in San Diego is being discussed a workshop on "(Q)FCM-based receptor occupancy approaches in pharmaco-dynamic monitoring" and that is also a nice application for absolute antigen/receptor quantitation. Tell how much target is present on (circulating) target cells, how much of the target Ag is covered by a therapeutic MAb at any specific time after infusion, how (and if?) this target Ag is modulated by the effect of treatment and from that on, help finding the most appropriate dosage for a new drug in early clinical trial.

I could not be more incentive, for all who did not yet take this train, it is not too late:

Go with the QFCM !

Philippe

Bruce, Philippe,

Thanks a lot for your answers and advices. I sincerely hope we can join you all in such an interesting topic soon.

Romuald