I am attempting to construct a concatamer - three glutamate transporters linked together (without stop codons) in order to express a tri-mer, and I am having a devil of a time with it. Beacuase they subunits are identical, and I have limited restriction sites, I must build the plasmid sequentially - first the monomer - then a dimer, then the trimer.

It appears that *sometimes* after culture in bacteria, I frequently lose on or more of my subunits. That is, I may ligate one transporter into the dimer, but when I screen colonies, I find evidence of the monomer! The first few times it happened, I assumed I had made some mistakes and had some contamination issues, but it has happened too many times for my technique to be that sloppy...

Has anyone ever encountered anything like this or do you have any other possible explanantion besides recombinant excision in e. coli? (BTW, I am using DH5alphas, which are recA-, supposedly.) More importantly, does anyone have any ideas how to solve this quandary?

Cheers,

Audra

As I'm sure you know, cloning direct repeats is difficult. I used to have problems using retroviral vectors based on HTLV-I as the long terminal repeats were each about 500bp. Even using AAV vectors which have 145bp terminal repeat encoding sequences is tricky but using a good E.Coli strain prevents problems with both the HTLV and AAV. I used SURE cells from Stratagene. (Surpressing Unwanted Recombination Events - I think).

I'm guessing that your monomers are significantly larger than this though. So just choosing good cells will not solve the problem. There are RecA dependent and RecA independent recombinations and it is the latter that are the problem. During plasmid replication the plasmid forms a figure 8 structure and if the sequences on each strand at the cross-over are similar there will be recombination and replication of only one half of the "8". You might be able to get more stable multimer containing plasmids if you insert a large amount of non-coding stuffer sequence that just acts to separate the repeats from one another during replication. The multimers would then all be on the top or the bottom of the "8". I've never worked with cosmids but this might be worth looking in to. There might also be ways to reduce recombination by dropping the temperature of the E.Coli cultures when growing up the plasmids.


Good luck

Before dropping this project, you may consider different E. coli strains: Stbl2 (Invitrogene) and Sure (and Able or XL10Gold) (Staratagene)