Unstable MFs

Part of my work involves using the gpt delta transgenic mouse model, which was first described by Nohmi et al. 1996. Well Nohmi has made this into a veritable industry churning out a bunch of papers with a whole host of collaborators. There are even spin-off gpt delta transgenic rat model and the gpt delta transgenic cell line! Watch out Big Blue and MutaMouse!

Well this post has seen the light of day, perhaps not so much for its relevance to microbes (oops), but because it’s something I have to deal with. Anyway, if you are familiar with Big Blue, you’ll understand exactly how this works. Otherwise it’s convoluted and I normally have to recur to many diagrams to get it across. The gist of it is a rodent that has had bacteriophage genomes knocked in at a certain intron. The bacteriophage are not actively used by the rodent but are present in every cell of its body. It’s believed that the phage genome can be used as a reporter of the background levels of mutations perceived by the rodent.

The DNA from the phage can be separated from the mouse DNA and it carries 2 sets of reporters. The first set, consisting of the gpt gene [similar to the HPRT gene in humans] and CAT cassette, is on a plasmid transfected into E. coli [microbes, yay!]. Since gpt incorporates the toxic 6-thioguanine, only bacteria with a mutated gpt gene survive, so it’s useful for looking at small mutations that inactivate the gene. There’s also a set of genes, red/gam, used by bacteriophage to transfect bacteria. E. coli previoulsy transfected by another phage target red/gam products and prevent the entrance of other phages. If the DNA recovered from the mouse suffered a large deletion that excised red/gam then plaques form in a lawn of E. coli telling you about the existence of mutants (The ability to measure larger deletions is gpt delta’s claim to fame). If you are really interested in how this works and what people do with it, here’s a review written by Nohmi himself.

People who really buy into these models use them to determine how carcinogenic compounds are when given to a mouse. People who don’t buy into these models, raise a whole host of problems concerning the relevance of untranscribed, unrepaired, highly methylated, phage DNA in telling us about the levels of DNA damage in the host.

Anyway, I won’t get into that right now, but I actually wrote this entry to mention an interesting paper, where the treatment was given as smaller doses over several days (as opposed to the normal bolus approach) and the gpt delta mice were looked at 1- and 4-weeks after the end of treatment. The normal spike in MF seen is pretty normal when you give a putative chemical carcinogen to a rodent, but the interesting thing was the follow-up where they saw that at 4-weeks the levels were back to normal. It was surprising to see the follow-up as it’s not common practice in many of these gpt delta rodent/chemical carcinogenesis experiments. I had the naive impression that the spike in MF set up the mice for a lots of mutations, but maybe a lot are repaired or occur in cells that are going to be destroyed anyway.

PS here’s Nohmi’s official site with protocols