Unfortunately you can not beat the counting statistic. In fact you can test
counting statistics by making a big file and then analysing fractions of it
from different time intervals with different stop counts. Another problem to
add in particular if only parts of the sample are analysed is the sampling
statistic. If you look at 100cells / ml (=0.1cell /ul) and you sample 10ul you
might not have a cell there at all as it has it's own mind on where to be at
the time. This is where you have to start thinking about physical, chemical or
biological pre-amplification.
The other limiting factors are the accuracy of your volume determination (for
ratiometric counts the variation increases as the bead count adds another
counting error) and the effects of the vanishing particle. The most accurate
counting method does not help if your particle sticks to the wall of the tube.
A little bit of detergent (0.05% Tween 20) gave most consistent recovery in our
hands. You might also want to consider mechanical disaggregation
(see http://www.elsevier.com/homepage/sah/mimet/speciss/1378.pdf
from the Journal of Microbiological Methods special issue free full-text
Volume 42/1, September 2000, Microbial Analysis at the Single Cell Level,
guest-edited by:
L. Alberghina, D. Porro, H. Shapiro, F. Srienc, H. Steen.
http://www.elsevier.com/homepage/sah/mimet/speciss/toc42_1.htm )
Hope this helps
Regards
Gerhard
-----Original Message-----
From: Daniel Hoefel [SMTP:daniel.hoefel@sawater.sa.gov.au]
Sent: Monday, March 05, 2001 1:41 AM
To: Cytometry Mailing List
Subject: FACSCalibur limit of detection analysing bacteria
I am interested in enumeration and identification of bacteria from water samples
using flow cytometry. I recently attempted to determine the limit of detection
on
a FACSCalibur for enumeration of E.coli cells suspended in MilliQ water
(filtered
distilled water). The experiment was carried out on an overnight broth culture
culture
of E.coli stained with SYTO9. MilliQ water was used as sheath fluid. I was
able to
get down to determining the concentration of 500 E.coli cells/ml where below
this
number results became inaccurate. I was looking to enumerate down to about 100
E.coli
cells/ml and I was wondering if anyone has done this type of experiment before
and if
they were able to accurately detect less than 500 cells/ml. Am I asking too
much from
the FACSCalibur and/or are there any secrets to improving the limit of
detection?
Input from anyone on this matter would be greatly appreciated.
Daniel Hoefel
Email: daniel.hoefel@sawater.sa.gov.au
This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:08 EST