PI staining live cells

From: Kelly Hardwicke (kellyh@ccmi.salk.edu)
Date: Mon May 17 1999 - 15:42:06 EST


ON FRIDAY, MAY 14 CANDACE ENOCKSON WROTE:
>Kely, I have a tangential question related to a comment you made in the
>response to Sraboni.  I am new at this and have seen what I think you refer
>to as "aggregate tails". Are these cell aggregate or are they reagent
>aggregates?  How do you gate them out? Do you back gate from a dot plot?
>Thank you in advance for your response.
>
>Candace Enockson
>Flow Cytometry Shared Facility
>Medical University of South Carolina
>Charleston, SC


Hi Candace,
One of the points that I was trying to make to Sraboni is that the
information one can attain from a FSC vs. SSC plot is rather limited,
even though it is a conventional flow cytometry strategy to gate from
this plot.  A particle found in the "aggregate tail"  located to the
upper right of the elipsoid "cell" population could be ANY particle
large enough, granular enough, and of a certain shape to fit the high
FSC and SSC requirements for this tail.  That is, a particle in this
tail ma be a cell clump, but it may also be a very large single cell, or
in some very improbable cases, a (VERY) large clump of reagent (larger
than cells themselves!).  The converse holds true for the debris tail,
located to the lower left of the elipsoid "cell" population,  where a
particle appearing here may be membrane bits, dead cells, random debris
from sample preparation, reagent aggregates, or a very small live cell.
The theory behind using this imprecise FSC/SSC plot to gate from is
that even though you may be gating away a few misshapen or oddly-sized
live cells that you want, you are also gating away a large percentage of
obvious debris and aggregates that you know for certain you do NOT want,
as dead cells and cell aggregates often exhibit "strange" fluorescence
characteristics due to changes in membrane permeability (dead cells),
quenching, or just plain more fluorophores per particle (aggregates).
However, because this method of determining what cells are singlet and
live and/or intact IS so imprecise, one should use alternative methods
for determining viability I usually gate as I acquire the data file
from a FSC vs SSC dot plot.  I use CellQuest, though, and I don't know
how or if one even can do this from other flow  programs, although it
seems relatively straitforward so I would assume one can.  However,
I do back-gate sometimes, by collecting all of my data with no gate,
and then during analysis making a FSC vs. SSC dot plot. I outline the
elipsoid region that I feel looks best like a cohesive population of
singlet cells.  If your FSC and SSC voltages and amp gains are set
correctly, and the FSC axis is set to LINEAR mode, this region should
be near the middle of the plot, and wider than it is tall.   (SSC axis
set to LOG mode, and FSC on X-axis)  If it isn't located there, you can
play around with the voltages and amp gains to get it
(e.g. propidium iodide) and/or single cells (e.g. pulse-width processing) when it is really important to know these things.

there, unless it is the occasional badly prepared sample that consists entirely of either debris or aggregates.

Good luck,
Kell



This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:53:30 EST