Applied and Environmental Microbiology, January 2005, p. 58-64, Vol. 71, No. 1
Effect of Culture Conditions on Microorganism Identification by Matrix-Assisted Laser Desorption Ionization Mass Spectrometry
Nancy Valentine, Sharon Wunschel, David Wunschel, Catherine Petersen, and Karen Wahl
The Big Picture:
If you want to know what type of bacteria you are dealing with, you face three problems:
1. figuring out what part of the taxonomic scheme it fits into
2. figuring out whether the taxonomic scheme needs to be changed
3. figuring out whether the quality you care about is represented in the scheme.
One example: You pull up a bacteria from a culture plate where there are many. You have a patient who is ill and want to know if an antibiotic will do him good. Well, you need to know if this is the pathogen, and if so, will the antibiotic kill it. You put it through biochemcial tests, but it doesn't match any of the obvious ones. But Bergey's even says that 'negative' is only true 95% of the time, same with 'postive' not to mention 'w.'
So, you do molecular work and identify it as E. coli. Or so you think. It could be a different related species, but nobody has studied such odd variants of E. coli in sufficient detail. Finally, even if it is E. coli, there are plenty of non-pathogenic E. coli, how do you know this is the one you care about, the pathogen in this patient?
Unforunately, the pathogen may not have grown on your initial plate, and even if it did, this is all taking so much time, day after day of culturing, that the patient may be long gone.
An approach people have tried is MALDI, a form of mass spectometry that burns up a sample and pushes it through the spec to identify pieces of materials it is made from. These peak patterns, for single proteins, help identify the protein. If they can identify the protein, then presumably they can help identify cells. In practice, in the lab, this has worked reasonably well. It is certainly remarkably fast, just a few minutes per sample.
It has also started being used to identify bioterrorist agents wihthout trying to culture them or taking the time to grow them.
However, bacteria are not always composed of the same proteins, nor are those proteins always in the same ratio. Recognizing this, the authors have decided to test MALDI on strains grown in different media. They use four different media and create fingerprints for each of 3 strains. In each case, the fingerprint changes substantially, but the computerized database, using statistics, is successful at the identification.
This will not always be the case, as more strains and species enter the database, and identifications need to be more and more precise. However, it was encouraging for now. The real problems will come with the bioterrorism samples - which may nor may not have the signitures of the common strains that are being constantly scanned for by the devices.
Still, an interesting methods paper.
Effect of Culture Conditions on Microorganism Identification by Matrix-Assisted Laser Desorption Ionization Mass Spectrometry
Nancy Valentine, Sharon Wunschel, David Wunschel, Catherine Petersen, and Karen Wahl
The Big Picture:
If you want to know what type of bacteria you are dealing with, you face three problems:
1. figuring out what part of the taxonomic scheme it fits into
2. figuring out whether the taxonomic scheme needs to be changed
3. figuring out whether the quality you care about is represented in the scheme.
One example: You pull up a bacteria from a culture plate where there are many. You have a patient who is ill and want to know if an antibiotic will do him good. Well, you need to know if this is the pathogen, and if so, will the antibiotic kill it. You put it through biochemcial tests, but it doesn't match any of the obvious ones. But Bergey's even says that 'negative' is only true 95% of the time, same with 'postive' not to mention 'w.'
So, you do molecular work and identify it as E. coli. Or so you think. It could be a different related species, but nobody has studied such odd variants of E. coli in sufficient detail. Finally, even if it is E. coli, there are plenty of non-pathogenic E. coli, how do you know this is the one you care about, the pathogen in this patient?
Unforunately, the pathogen may not have grown on your initial plate, and even if it did, this is all taking so much time, day after day of culturing, that the patient may be long gone.
An approach people have tried is MALDI, a form of mass spectometry that burns up a sample and pushes it through the spec to identify pieces of materials it is made from. These peak patterns, for single proteins, help identify the protein. If they can identify the protein, then presumably they can help identify cells. In practice, in the lab, this has worked reasonably well. It is certainly remarkably fast, just a few minutes per sample.
It has also started being used to identify bioterrorist agents wihthout trying to culture them or taking the time to grow them.
However, bacteria are not always composed of the same proteins, nor are those proteins always in the same ratio. Recognizing this, the authors have decided to test MALDI on strains grown in different media. They use four different media and create fingerprints for each of 3 strains. In each case, the fingerprint changes substantially, but the computerized database, using statistics, is successful at the identification.
This will not always be the case, as more strains and species enter the database, and identifications need to be more and more precise. However, it was encouraging for now. The real problems will come with the bioterrorism samples - which may nor may not have the signitures of the common strains that are being constantly scanned for by the devices.
Still, an interesting methods paper.
posted by BenK at 9:22 AM
0 Comments:
Post a Comment
<< Home