Follow Up: Bacterial Analysis

From: R. Thomas Leonard (tleonard@mail.wistar.upenn.edu)
Date: Wed Sep 25 2002 - 16:09:48 EST


Thanks to everyone for your helpful replies to my earlier request (see text
below). Accordingly, I feel compelled to share our results.

Based on your suggestions, we've (the operator and I) concluded that the
discriminator setting makes all the difference when attempting to quantify
bacteria. While we obtained very poor quantitative results triggering on
forward scatter, triggering on log side scatter worked beautifully in terms
of accurately quantifying bacteria. Even triggering on PI yielded
acceptable results.

As for viability discrimination, we found wonderful separation between
independent live and dead controls, yet when the two populations were mixed
the separation appeared to become more of a gradient. Nonetheless, when
"live" and "dead" regions established from the controls were actually
sorted, about 80% of the "live" cells were live and less than 5% of the
"dead" sorted cells were live. Can't complain about that (or at least I
won't).

Thanks Folks!

Tom Leonard







I'm on a mission to determine bacterial viability as well as quantify
bacteria using Molecular Probes' BacLight viability kit. Using an EPICS
Elite, I have routinely found bacterial counts to be 10% of those
determined with fluorescence microscopy. The accuracy of the microscopy
counts have been verified by other labs. I recognize that the organism has
mechanisms to pump out fluorochrome, but to the extent that 90% of bacteria
go undetected? We're analyzing on the log scale. Our operators are
accustomed to analyzing mammalian cells, but claim they have done all the
"tweaking" possible to optimize bacterial analysis.

Why such a disparity in counts? My conclusion is that the instrument simply
can't detect the smaller organisms in the population (of E. coli).

Also, we can see distinct populations when analyzing live controls and dead
controls independently, but when we mix the live and dead populations we
are often unable to distinguish between live and dead populations.
Sometimes we can, other times we can't. It is not only strange but
frustrating.

Has anyone had similar experiences? Any thoughts on what might be
contributing to these unexpected results? Thanks in advance.

Tom Leonard



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:26:26 EST