I've been gone for a week and things have really heated up! I've heard
from several people that disagree with me. Any supporters? Anyway...
Maryalice,
You say that you see a pattern of annexinV+PI- cells early and then a
pattern develops where there is a creeping up into the double positive.
(Once again we see the shortcomings of quad stats. But I digress...)
There should still be a definite very bright PI+, dead population. It is
no secret that apoptotic cells become leaky and can be PI dim+. As the
membrane integrity breaks down, PI fluorescence increases. In fact, there
is an apoptosis technique that relies on this phenomenon. Let me ask you
this: Have you ever permeablized cells and stained with Annexin V? Cells
with compromised membranes can stain very positive for Annexin V. I agree
that it is easy to infer that the cells that stain AnnexinV+PI- will be
double positive at a later time point. Let me make a few points about
this.
1) Yes, late apoptotic cells will be double positive. However, necrotic
cells will also be double positive. In fact, based on just this staining
pattern, they are indistinguishable. When using AnnexinV, we are supposed
to be looking at early apoptosis. It is not a good method for late
apoptosis for this very reason. The proper "window" of usefulness must be
determined by experiment. We sort of have the chicken and the egg quandary
here. Are the double positive cells PI+ because they are late apoptotic or
are they AnnexinV+ because they are necrotic? My view is that if the cells
are PI+, the cell membrane is compromised and they are best excluded from
the apoptotic population. I suppose that is, obviously, open for argument.
2) Keeping the cells on ice after staining, and running them as soon as
possible is very important for similar reasons. The PI staining will
increase over time. They need to be run in a timely fashion to maximize
the separation of "live" from "dead". Even "live" cells stained with PI
fluoresce more brightly than unstained cells. Try this, you may be
surprised.
3) For whatever reason, I find that AnnexinV-FITC often requires more
compensation than a FITC-Ab. (I have seen significant AnnexinV-FITC
fluorescence spectral overlap even when using FL3 instead of FL2.) Bright
AnnexinV+ cells drifting into the double positive quadrant could be
undercompensated. (However, as I mentioned above, PI fluorescence in
apoptotic cells can increase as measure of loss of membrane integrity).
Undercompensation is very possible if you don't have a bright AnnexinV+
control. I recommend two AnnexinV+ single color controls. One as a
positive apoptotic control and one as a compensation control. For the
apoptotic control, use whatever stimulus gives you good results. For
compensation, use permeablized cells stained with AnnexinV to simulate dead
cells or kill them with drug treatment, if you prefer.
4) As someone else metioned. Make sure you don't gate out cells of
interest based on light scatter. I do use a light scatter gate to try to
remove debris.
I do not claim to be an expert in the field of apoptosis, but I have played
around with several assays. Some worked for me; some didn't work so well.
Don't take my word for it, or anyone else's for that matter. But don't
dismiss me out of hand either. Try some things and see what you think. If
we never tried something new or tried to optimize our techniques we would
still be using leeches to bleed people. Wait a minute....aren't leeches
making a comeback????
---------------------- Forwarded by David C McFarland/DEV/PHRD/SB_PLC on
03/26/2001 09:44 ---------------------------
stetler@box-s.nih.gov on 21-Mar-2001 10:09
To: cyto-inbox
cc: (bcc: David C McFarland/DEV/PHRD/SB_PLC)
Subject: Re: AnnexinV
This has been discussed before. When one looks at apoptosis by
Annexin V one often sees a pattern of Annexin V+ PI- cells early and
then a pattern develops where there is a creeping up into the double
positive quadrant- a smear that extends from the Annexin V+ PI-. We
all know that these cells are late apoptotic cells. Although one
needs to see a population of Annexin V+ PI- cells to conclude there
is apoptosis, if one sets the analysis quadrants and cranks out
numbers for this population only, they are inaccurate.
Maryalice
>I've been following the discussions on AnnexinV and decided to put in my
two
>cents. A concept that a lot of people seem to miss is that AnnexinV is
more
>than useless when used alone, it is misleading. You have to use it in
>conjunction with a dead cell discriminator. Dead cells stain very
>brightly for
>Annexin V. And no, AnnexinV+/PI+ double positives do not imply
>late apoptosis
>or apoptotic death. The PI is only used so that you don't count
>those cells as
>apoptotic. They are simply dead and that is all you can say about
>them. Also,
>timing is everything. AnnexinV is only useful for EARLY apoptosis. You
may
>have to do some experimentation to determine the proper time point for
>measurement. If you find you are really interested in late apoptosis, use
a
>tunel kit Sometimes you must come to the realization that AnnexinV is not
the
>best apoptosis assay for your system. I find, as someone else
>pointed out, that
>adherent cells tend to give a high percent of background. Physical
>perturbation of the cells can cause PS flipping and false positives.
Controls
>are paramount. You need to have a non-apoptotic control that is
>indeed AnnexinV
>negative, or at least dim. You also need to have a positive apoptotic
control
>that is indeed Annexin V positive. If you can't get AnnexinV
>positivity with a
>strong apoptotic stimulus and Annexin V negativity with no apoptotic
stimulus,
>how can you trust your results? Your actual test samples may be somewhere
in
>between and you will have to decide how to interpret them based on
>your control.
>It is not always black and white. And finally, try to correlate your
results
>with another assay. (Caspase 3 is another early apoptosis method.)
>I find that
>too often experimenters doing flow neglect the microscope. An easy
>fluorescence
>microscopy technique uses ethidium bromide and acridine orange to
distinguish
>viable, early apoptotic, late apoptotic and necrotic cells. (Sorry
>I don't have
>the reference at my fingertips). I will reiterate what I said before,
one
>apoptosis assay is not going to work under all experimental
>conditions. You may
>actually have to try a few things and see what works for you.
>
>
>And now, just to be Devil's advocate, check these out:
>
>Annexin V Binds to Positively Selected B Cells. Stacey R. Dillon, et al.
J.
>Immun. 2001, 166: 58-71.
>
>Annexin V Binds to Viable B Cells and Colocalizes with a Marker of Lipid
Rafts
>upon B Cell Receptor Activaiton. Dillon, et al. J. Immun. 2000, 164:
>1322-1332.
>
>
>
>David McFarland
>
>GlaxoSmithKline
Maryalice Stetler-Stevenson
Director Flow Cytometry Unit
Laboratory of Pathology, NCI, NIH
"Learn the rules so you know how to break them properly."
The Dalai Lama
This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:57:31 EST