Hello. I am using human peripheral lymphocytes for an assay of genetic
damage (the 'Comet' assay). I have been isolating PBL from 50-100
microlitres blood from a finger-prick, on a Ficoll-Paque gradient in
microcentrifuge tubes (1.5 ml). My problem is this - I have been
incubating the isolated PBL with various concentrations of a genotoxic
agent, then performing the assay. The data shows a lot of variation - a
dose-response can be plotted, but at each concentration there are some
cells which show a great deal of damage, and some which show very little
damage, with responses in between. Several Questions :
1. Could the heterogeneity be due to the different cell types in my
(rather crude) PBL isolation preps?
2. I know that some lymphocytes are long-lived - could any of the
variation be due to genetic damage prior to the assay, giving rise to
variability?
3. For conducting an assay of this type (single cells), how many nuclei
should I be scoring,for ststistical purposes?
4. If the heterogeneity is due to biological factors (e.g. subpopulations
of cells that are more sensitive to damage), rather than random variation
in the technique, then am I correct in assuming that increasing the
numbers of cells scored (beyond a certain point) will yield diminishing
returns in terms of signal-to-noise?
Thanks in advance for any and all considerations!! Greg :-)
