Bacteria microscopy

[mo24]

What techniques could be used to reliably identify bacteria in the blood under a microscope?

 

There are tests such as the Western Blot which identify antibodies in the blood, but I am interested in the possibility of a procedure to identify the antigen itself.

 

I imagine the bacteria would have to somehow be isolated from the blood first of all by some method. Is this correct?

 

I assume this must be difficult or unreliable to do, since if there was an accurate way to identify a bacteria antigen in the blood, that would become the standard method for diagnosing illness.

 

Could anyone who is in this type of field elaborate on the subject? Thanks.

[CraigD]

This is outside of my real expertise – I have a student/hobbyist microscope too weak to resolve any but the largest bacteria – but I’ll give your question my best shot.

 

The best optical microscope, using an oil emersion lens, should be able to at least detect the presence of even the smallest bacterium – it can resolve down to about 2e-7 meters, while bacteria are 5e-7 + in diameter, most a good bit bigger. Most bacteria have been identified this way. It’s necessary to stain the slide to have much hope of spotting anything, and some bacteria are stain-resistant, so there could be some problems with this approach. Also, it’s slow, and requires an expert microscopist.

 

If you could dry and secure the sample, lots of non-optical (electron, tunneling, laser or atomic force can give you so much detail you could literally see the DNA in the bacteria. Awful difficult, time consuming, and expensive for a diagnostic procedure, though.

 

I’d say you best bet is to forget about microscopes, and try to ID the wee beasties’ proteins or DNA. The technique would be similar to the Western blot, a protein-sorting technique, but you’d want to purify the sample and look for bacterial, not antibody, proteins. There might be a way to fracture all the DNA in the sample, get a blot of it, and subtract the known DNA to get a “gene fingerprint” of the bacteria.

 

This is way outside of my experience, so my guesses could be way off.

[MortenS]

Well, it is not my speciality, but I found a great site that you probably could use.

 

http://www.bmb.leeds.ac.uk/mbiology/ug/ugteach/dental/tutorials/introduction/bacteria.html

 

In general you can use a microscope in combination with the gram-staining test to do a very rough classification of the bacteria in question based on shape and whether it is gram-positive or gram-negative.

 

http://www.bmb.leeds.ac.uk/mbiology/ug/ugteach/dental/tutorials/classification/head.html

 

Based on the result of these tests, you go further with various tests, checking for absence/presence of various enzymes, motility etc:

 

http://www.bmb.leeds.ac.uk/mbiology/ug/ugteach/dental/tutorials/tests/head.html

[Biochemist]

There are tests such as the Western Blot which identify antibodies in the blood, but I am interested in the possibility of a procedure to identify the antigen itself....

I assume this must be difficult or unreliable to do, since if there was an accurate way to identify a bacteria antigen in the blood, that would become the standard method for diagnosing illness...

Clinically, most folks don't usually care about any particular antigen. The usual approach is to use morphology to identify the strain of bug of interest. Automated devices expedite morphology assessment. The simple approach is to culture the bug (usually on agar plates) and use colony morphology to identify the pathogen. Some folks also use a secondary culture to establish drug sensitivity as well. Treatment is based on the the endemic sensistiviy of particular pathogens in particular areas, and/or based on minimum inhibitory concentrations (MICs) of antibiotics that are extrapolated from the behavior of colonies in response to antibiotic-treated disks on the agar plates.

 

This field is clinical bacteriology/microbiology/medical technology.