Can FACS and Immunoflourescence data be correlated?

Nebe-Von-Caron, G g.nebe-von-caron at
Wed Jan 2 05:44:57 EST 2008

A happy and hopefully healthy new year to everyone and thanks to the
busy elf's keeping this list going.
In the light of Pauls end of year message let us rephrase the question
"Can flow and and image fluorescence data be correlated" - as both
techniques use immunofluorescence.
It goes a bit like:
In theory there is no difference between theory and practice - but in
practice there is!
To keep it short just a few points to consider
In principle flow and image analysis values should agree. The key
measurement differences are that FCM in most cases gives you an
integrated signal for the whole cell being illuminated whilst ICM gives
you localised information and depending on the setup of the imaging
system object size only a section of the cell might be in focus.
Other differences arise from the variation in excitation spectrum and
the perceived emission which can give rise to different levels of
perceived signal and autofluorescence.
Assuming you use the same reagents, the key differences between the two
technologies lies in the way the cells are handled. With cell disruption
for connective tissue or cell dissociation you may have selective cell
loss with the preparation.Apart from cell loss, problems will arise with
the indirect staining technique which can give more NSB thus washing
steps and dilutions become more critical. If looking at live cells one
has to be aware of physiological changes of attachment / cell - cell
contact. Sometimes it may be required to coat the glass slides to avoid
denaturation on the glass.
As a final point we have to consider statistics and subjectivity and
signal quantification. Counting error goes with the square root of the
cells counted. Having only a few cells remains a problem with either
technique. The subjectivity of the selection of the file of view is more
a problem in ICM. The reason you have to count the different squares in
a haemocytometer is that the distribution of the cells can be
inhomogeneous. However FCM can suffer from cell sedimentation or
selective adherance. Operator bias tends to be less in FCM but still
depends on thresholding strategies and the limitations in signal
quantification mentioned in the first point 
All the best


From: ddey at [mailto:ddey at]
Sent: Sun 23/12/2007 14:22
To: cyto-inbox
Subject: Can FACS and Immunoflourescence data be correlated?

Dear Users,
I had a very fundamental question about quantification of flourescence
data, as obtained by FACS v/s immunocytochmemistry (ICC) data as
observed under the microscope. We have been doing staining of mammary
epithelial cells with BrdU using protocols that we have standardised
for both FACS and ICC using a secondary conjugated to FITC.

With ICC, we count the number of green cells under the microscope as
positive. With FACS, we get the number as given by the machine. In a
one time experiment done parallely, we could not get the same positive
cell count when compared between the 2 techniques. I am not sure if
researchers always find the quantitaive FACS and ICC data the same?

We work with primary human breast cancer cells. Hence cell number is
our biggest limitation in all experiments. Therefore we have been
majorly using ICC to count BrdU positivity. We wanted to test if the
quantification we are doing using ICC is correct. Now, I am not sure if
we can & should expect an exact corrrelation between numbers obtained
with ICC and FACS.

I would be very grateful and helped for any insight in this direction.

Best wishes and wishing each of you a Very Happy New Year!


This message was sent using IMP, the Internet Messaging Program.

This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

More information about the Cytometry mailing list