Method for determining limit and breadth of contamination. - (Mar/07/2008 )
Does anyone have any advice on the best way to approach the detection of a wide range of contaminating microorganisms in a concentrated sample of a single microorganism (in this case a concentrated sample of Salmonella)?
We are attempting to respond to an FDA request for more information regarding a IND application. We are applying to use Salmonella as a vector in a cancer therapy trial. Among other things, we need to demonstrate that our cell bank is free of contaminating microorganisms, essentially looking for needles in a haystack. Our bug requires glucose and will not grow on most other carbon sources because it lacks the cyclic amp response protein necessary to turn on genes for alternative carbon source utilization. Therefore we are considering taking advantage of that fact and plating out concentrated amounts of Salmonella from our cell bank on plates with defined carbon sources and look for growth of contaminating organisms. It would seem that the best plates for these experiments would be Niedhart's defined rich medium for enterobacteria with carbon sources like maltose and/or citrate. The question of limits of detection will be easily solved by doping with a contaminating E. coli. However, the question of "breadth of detection" is more difficult. Does anyone know of a list of microorganisms that will grow on this medium?
Thanks for reading and considering this.
Another general approach would be to insert a conditionally lethal gene into your strain, turn it on, and detect any survivors. With this strategy, you could plate on a wide variety of plates and detect most anything that can be cultured. The conditional lethal would be a feature for your application, also. Choose something safe as an inducer, like Tet, and you could kill your organisms in vivo.
I second phage434's suggestion. It is quite elegant. Although I would use an inducer that did not have antimicrobial activity.
Indeed, an inducer such as 3-oxo C6 homoserine lactone, aka Vibrio autoinducer (VAI), aka acyl-homoserine lactone (AHL) might be ideal. I don't know if it is toxic or immunogenic in mammals.
Thanks. That is an interesting idea and I will consider it, especially if we decided to just go with the defined rich medium plates and the FDA remains concerned about the breadth of contaminating organisms we would see. My guess is that the Neidhardt's defined rich medium with multiple carbon sources will grow a wide variety of microorganisms, although it would be good to find some published empirical data concerning this point. We are running separate tests for fungal contamination using inhibitory mold agar.