 |
 |
 |
 |
To do this requires several pieces of information, some of which may be hard
to estimate accurately.
Let's assume the simplest case: a ligand that is directly labeled by FITC.
If you know the number of FITC per ligand, the binding multiplicity of ligand
per receptor, and if you have a calibration of your digital signal in terms
of equivalent FITC per channel, then you can calculate the number of ligands
bound per channel. The FITC per channel can be done using beads which have a
known number of molecules per bead. The ligand/fluorochrome and
ligand/receptor stoichiometry are a bit harder to estimate, but too hard.
Lastly, let's assume the cells are spherical--like billiard balls. It's been
shown somewhere (by Gary Salzman I think some 10 years ago) the forward
scatter (FSC) is reasonably proportional to the cross sectional area of a
particle. So FSC will be proportional to the square of the cell's dimension.
That is, the same proportionality as area. So, to be crude (but perhaps
effective), you could divide fluorescence by the FSC on a cell by cell basis
to obtain an estimate of number ligands per area. At the least, this will
show if there are receptor density differences among cells in the population
(i.e., the presence of multi-modal populations), or it may show a single,
normally distributed class.
This method has lots of assumptions each of which affect accuracy. In fact,
the measurement is only a relative one since there are several scalar values
missing in the assumptions of proportionality. Moreover, if the cell surface
has folds or other structure, then the estimate of surface area could be
much larger than the area of an equivalent sphere. Of course, there are ways
to improve accuracy: volume measurement, axial extinction, or time of flight
can provide a measure of cellular dimension; membrane area might be measured
by using a dye that partitions only into plasma membrane, binding
stoichiometry can be measured by Scatchard plots.
Dave Coder
dcoder@u.washington.edu
Begin forwarded message:
Date: Wed, 5 Apr 1995 11:31:25 +0200 (MET DST)
From: Bjarne Moeller <bkm@biobase.dk>
Subject: Fluorescence intensity from large cells
To: cytometry@flowcyt.cyto.purdue.edu
I have a simple question to the community of flowcytometric oracles.
We are currently analysing rat peritoneal macrophages for surface
membrane receptors after staining with fluorescein-conjugated ligands.
"Resident" macrophages are often larger than 60um, thus considerably larger
than the 20um x64um laser beam at excitation focus of our FACScan & FACSort
flowcytometers. The fluorescence intensity from the cells suggestedly
mirrors the receptor density at the exposed part of the cell membrane,
but hardly reflects the total receptor pool in these cells, since
variation in surface area might contribute. Comparison with binding of
radio-iodinated ligands does not exhibit convincing correlations.
It might be common knowledge, but can someone give me a clue to
correction for surface area by arithmetic on FSC, or would it be far
reached to introduce a "topical intensity times FSC" parameter in huge cells?
|
|
|
|