Re: [DNA Staining of cells with intracellular bacteria]

Mike Herron wrote, in response to my response to Todd Parker:

>Some of the live DNA stains partition preferentially into some bacteria 
>growing in some host cells.  (How is that for waffling?)  The same goes 
>for for  the cell tracking dyes.  I think you need to approach the problem 
>empirically, though I am all for throwing in some theory too!

What Todd Parker wanted to do was not detect the bacteria, as I had 
suspected, but measure DNA content of the cells for cell cycle analysis, 
which should be easier, if only because isolating the nuclei should 
eliminate most of the bacteria and their DNA.

>I do this sort of thing every day with the bacteria that I study, but I 
>usually do it microscopically rather than by flow.

Most of the live "DNA" stains are cyanines; they are lipophilic and 
positively charged and will therefore partition preferentially into any 
intracellular membrane-bounded compartment across the membrane of which 
there is an interior-negative electrical potential. That would include 
mitochondria, which contain only a very small amount of DNA, which yields 
very little fluorescence from the bound dye, and intracellular bacteria, 
which will show substantial fluorescence from both DNA and rRNA which bind 
dyes such as the SYTO series. There are complicating factors; bacteria have 
a lot of extrusion pumps, many of which don't exist in mammalian cells, and 
quenching may occur when dye becomes highly concentrated. And you need 
microscopy or image analysis rather than flow cytometry to distinguish 
staining of cytoplasmic organelles and intracellular bacteria from staining 
of the nucleus. If you've published on this, could you share the references 
with us?

Thanks,

-Howard


>>Todd Parker wrote-
>>
>>>Does anyone have experience performing DNA staining using cells infected 
>>>with intracellular bacteria (i.e., Preferrable dyes, combination of 
>>>dyes).  All information is welcome
>>
>>What is the objective of the staining? If it is to detect the 
>>intracellular bacteria, there is going to be a substantial 
>>signal-to-noise-problem using almost all known DNA stains, because 
>>bacterial genomes are typically well under a hundredth the size of the 
>>human genome. Unless there are a lot of bacteria in each infected cell, 
>>the overall DNA content won't increase enough to make the infected cell 
>>detectable, even with an instrument CV of 2 per cent or so.
>>
>>Bacteria exhibit a wide range of ratios of A+T/G+C, whereas, in human 
>>cells, the ratio is approximately 1:1. If there were a fairly large 
>>number of bacteria in a cell, staining with a combination of A-T 
>>selective and G-C selective dyes, e.g., Hoechst 33258 and chromomycin A3, 
>>might show a different fluorescence ratio in infected and uninfected 
>>cells, possibly improving specificity of detection - although this 
>>wouldn't work for the many bacterial species with ratios similar to the 
>>1:1 value for humans.
>>
>>You would almost certainly have to use highly DNA-selective stains, e.g., 
>>the Hoechst dyes and DAPI, as bacteria typically contain about 5 times as 
>>much double-stranded ribosomal RNA as DNA. The RNA stains with less 
>>DNA-selective dyes, e.g., ethidium or propidium and the SYTO dyes, and it 
>>is harder to get rid of using RNAse than is intracellular RNA in 
>>mammalian cells, because getting RNAse through the bacterial cell wall is 
>>difficult.
>>
>>If you needed to do this without fixation, you'd pretty much have to use 
>>Hoechst 33342 as the stain, and there is no guarantee it would get into 
>>the bacteria. On the other hand, if you can fix the cells, a better 
>>approach to detection might be to use PNA probes to detect ribosomal RNA 
>>sequences specific to the bacterial genus, species, or strain in the 
>>infected mammalian cells. The rRNA probes offer better specificity than 
>>you would usually get using antibodies.
>>
>>All of the above is predicated on measurements made with flow cytometry 
>>or low-resolution imaging. It is relatively easy to detect bacteria - or 
>>even mycoplasmas and, in some cases, viruses - in infected cells by 
>>looking at (or imaging) a DAPI-stained slide at relatively high resolution.
>>
>>I'd be happy to expand on this given more information about the desired 
>>end results, provided you don't have to shoot me if you tell me.
>>
>>-Howard
>>
>>
>>
>>
>---
>
>Michael J. Herron,  U of MN, Dept. of Entomology
>   herro001@umn.edu
>      612-624-3688 (office) 612-625-5299 (FAX)