RE: Baclight membrane potential kit

:-)

I did respond directly to Paul Thomson. Hopefully we'll receive some useful,
interesting feedback.

-----Original Message-----
From: Howard Shapiro [mailto:hms@shapirolab.com] 
Sent: Saturday, May 07, 2005 9:01 PM
To: cyto-inbox
Subject: Re: Baclight membrane potential kit

Paul Thomson wrote:

>Hi All, I'm considering using the BacLight Bacterial Membrane Potential Kit
>to estimate concentrations of viable bacteria in seawater.
>
>Does anyone have experience with this kit they can share with me?

I have had no experience with the kit, and would also be interested in 
others' experience with it. However, since the methodology was developed 
and is regularly used in my lab, with Dave Novo and Nancy Perlmutter having 
done most of the experimental work, I feel qualified to comment.

The details of three papers are relevant. The first (Novo D, Perlmutter NG, 
Hunt RH, Shapiro HM: Accurate flow cytometric membrane potential 
measurement in bacteria using diethyloxacarbocyanine and a ratiometric 
technique. Cytometry 1999; 35:55-63) describes the ratiometric membrane 
potential measurement technique using DiOC2(3) with a 488 nm laser; the 
second (Novo D, Perlmutter NG, Hunt RH, Shapiro HM: Multiparameter flow 
cytometric analysis of antibiotic effects on membrane potential, membrane 
permeability, and bacterial counts of Staphylococcus aureus and Micrococcus 
luteus. Antimicrob Agents Chemother 2000; 44:827-834) illustrates 
simultaneous measurement of membrane potential (with DiOC2(3) and the 488 
nm laser) and permeability (with TO-PRO-3 and a red laser). The third 
(Shapiro HM, Nebe-Von-Caron G: Multiparameter flow cytometry of bacteria. 
Methods Mol Biol 2004; 263:33-44) provides a detailed protocol.

The Data Sheet for the kit can be downloaded from 
http://probes.invitrogen.com/media/pis/mp34950.pdf.

Flow cytometry of membrane potentials was, as far as I know, first 
described in 1979 (Shapiro HM, Natale PJ, Kamentsky LA: Estimation of 
membrane potentials of individual lymphocytes by flow cytometry. Proc Natl 
Acad Sci USA 1979; 76:5728-5730). Note that the title of this paper 
describes "estimation" of membrane potentials, while that of the 1999 Novo 
et al paper describes their "measurement". This is not accidental; the 1979 
paper showed, among other things, that mean dihexyloxacarbocyanine 
(DiOC6(3)) fluorescence intensity in lymphocytes, measured at 530-580 nm 
with 488 nm excitation, was lower in electrically depolarized and higher in 
electrically hyperpolarized cells than in controls, and that lectins such 
as phytohemagglutinin appeared to depolarize only some of the lymphocytes. 
Since there was substantial overlap of the fluorescence distributions from 
control and treated cell populations in all cases, we characterized what we 
were doing as estimation of membrane potentials rather than as measurement. 
However, at least some of the many investigators who have since used flow 
cytometry of cyanine dye fluorescence intensity in studies of cytoplasmic 
membrane potential in prokaryotic and eukaryotic cells and of mitochondrial 
membrane potential in eukaryotic cells make the mistake of describing the 
procedure as "measuring" potential.

The studies of lymphocytes reported in 1979 were done in the hope that we 
might be able to detect small numbers of specifically activated lymphocytes 
based on differences in membrane potential; once it became clear that 
fluorescence intensity itself would not provide a basis for such 
discrimination, we moved on to analyses of early activation antigen 
expression, and restricted our work on membrane potential to analyses of 
bacteria. We and others soon found that neither cationic (cyanines, 
rhodamines) nor anionic (oxonols) dyes could provide reliable 
discrimination at the single cell level between metabolically active and 
electrically depolarized bacteria on the basis of fluorescence intensity 
measurements.

The underlying Nernstian theory predicts that it is the concentration, 
rather than the amount, of dye in a cell that most accurately reflects 
membrane potential. The ratiometric technique we reported in 1999 uses a 
much higher concentration of dye than is typically used for flow cytometry 
of membrane potentials (30 uM vs. tens of nm). Under these conditions, the 
green (530 nm region) fluorescence of DiOC2(3) in single bacteria or clumps 
of bacteria becomes independent of membrane potential, but remains strongly 
dependent on size. As intracellular concentration of dye increases, the 
fluorescence emission spectrum shifts toward the red, reflecting increasing 
interactions between the orbital systems of neighboring dye molecules, some 
of which may be due to dye aggregate formation. The concentration of dye is 
dependent on membrane potential; the intensity of red fluorescence (at 
600-700 nm) from a bacterial cell or clump reflects both the dye 
concentration (and therefore the potential) and the size of the cell or 
clump. The ratio of red to green fluorescence, in which the numerator is 
both potential- and size-dependent, while the denominator is 
potential-independent and size dependent, primarily reflects membrane 
potential. We have established that, in at least some microorganisms, this 
ratio provides a calibratable measurement of potential in the same range as 
can be obtained by other, noncytometric measurement techniques. 
Fluorescence ratio values for individual aerobic Gram-positive organisms 
from depolarized and control cultures typicaly differ by one to two orders 
of magnitude, providing clear discrimination between populations. By 
contrast, there is substantial overlap between the dye fluorescence 
intensity distributions of similarly treated populations, obtained using 
the lower dye concentrations normally employed for potential estimation.

I am flattered to see Molecular Probes/Invitrogen marketing a bacterial 
membrane potential kit (I make no money on this; I did have patents on 
cytometric membrane potential measurement, but they expired years ago). 
However, DiOC2(3) has been in the in the Molecular Probes catalogue for 
some time, and, although the kit provides the necessary reagents in a 
convenient, ready-to-use form, it is not particularly difficult to prepare 
all of the necessary stock and working solutions from scratch.

We are confident that the method works as originally described with 
Gram-positive organisms, including Mycobacteria. The intact outer membrane 
of most Gram-negative species presents a permeability barrier to lipophilic 
compounds such as DiOC2(3), but the addition of a low concentration (1-5 
mM) of EDTA allows the dye to be taken up, and, although it has not been 
possible to calibrate the measurement in EDTA-treated Gram-negative 
organisms, the fluorescence ratio does reflect differences in membrane 
potential and clearly discriminates depolarized from control Gram-negative 
bacteria.

However, although the ratiometric method arguably represents the most 
accurate and precise method for flow cytometry of bacterial membrane 
potentials, and the only one that can legitimately be described as 
providing a measurement, rather than an estimate, of membrane potential at 
the single cell level, a PubMed search today ("flow cytometry" AND 
"membrane potential" AND bacteria) did not turn up any publications other 
than ours in which the method was used, although we are aware of a number 
of articles in which our work has been cited.

The primary context in which we have used the ratiometric method is 
analysis of drug effects on cells, carried out using pure cultures. As the 
illustrations from our papers, those in Figure 7-32 (p. 400) and on the 
back cover of the 4th Edition of Practical Flow Cytometry, and Figure 5 of 
the kit data sheet show, there is excellent discrimination between control 
(metabolically active) and depolarized populations under these conditions. 
How suitable the method would be for estimating viability of organisms in 
seawater remains to be determined. One would typically be dealing with a 
mixed population of organisms present at relatively low concentrations, 
including Gram-positive and Gram-negative species, some of which might be 
autofluorescent in the spectral regions used for the measurement. If major 
components of the bacterial population could be gated for further analysis 
on the basis of scatter signatures, the membrane potential measurements 
would probably be cleaner than if one simply compared fluorescence ratios 
for the whole population. However, simply comparing two-dimensional 
distributions of red vs. green fluorescence for ungated populations with 
and without CCCP added as a depolarizing agent should provide an overall 
indication of the fraction of "viable" organisms. This assumes that 
"viability" is equated with having a membrane potential; it is known that 
some bacterial species can enter a state of dormancy, from which they can 
be resuscitated, losing membrane potential while they are dormant.

In our experience, the combined potential/permeability measurement is 
substantially more informative than the potential measurement alone; we 
would use 488 nm excitation for DiOC2(3), measuring green fluorescence 
through a 20-30 nm bandpass filter centered at 520-530 nm and red 
fluorescence through a 20 nm bandpass filter centered at 610 nm, and excite 
TO-PRO-3 with a red He-Ne or diode laser, measuring fluorescence using a 
680-695 nm long pass filter.

We'd be happy to take a look at the results and provide advice, so it seems 
as if the logical thing to do is  get the reagents, either in the kit or 
individually, put on a CD with some good background music, and try the 
experiment.

-Howard