I’m trying to assemble 4 fragments (2.1kb vector + 3 inserts, each around 1.1kb) using Gibson assembly. All three inserts have the same sequence at the beginning and at the end (200bp promoter and 200bp terminator). So I added custom overlaps, each between 30-40bp to ensure the proper order of the assembled fragments.
The problem is that I’m not able to obtain colonies with the correct insert (= vector with all the 3 inserts in the proper order). I’m getting colonies with 1 or 2 inserts, sometimes even more, but it’s a random mix of insert fragments in the vector.
After the assembly, I can see very strong bands that correspond to each of the fragments, and weak bands for combinations of 2 fragments, but no bands for 3 or 4 fragments together.
I’ve optimized the overhangs in order to remove all potential secondary structures and also the GC content is below 50%. Using the same overlaps, I’m able to join each of the insert fragments into different vectors.
The Tm temperature of the overhangs is 60C, except for the first insert, where the Tm is 67C. Do you have experience with the Tm of the overhangs? NEB recommends Tm > 48C, overhangs in the example in their manual have Tm=50C, so maybe my Tm is very high for assembly of more fragments than 2? Incubation time is 1h at 50C.
Thanks!
Break it down into individual reactions... Gibson is sensitive to secondary structures (terminators!) and repeats independently. Putting them together in one reaction... sounds like a bad idea. You'll still have trouble if you break it down into 3 individual assemblies. But you chance of having one correct clone at all goes up exponentially at that point.
Also, if you just care that 3 proteins are expressed equally, consider other ways of achieving that, which don't involve repeats, like T2A peptides.
Hi,
in general you can try to do an 'overlap extension' PCR if Gibson doesn't work. I always use that as a backup strategy. However, since your final product would be quite long (around 3.3 kb) a PCR might be tricky here...
Which ratios (backbone:insert) did you use? How much DNA? Which transformation method?
Thank you for your tips! I've been also thinking about an overlap extension PCR, so I'll give it a try, together with breaking down the whole reaction into individual assemblies.
Generally, I use equimolar amounts of all the fragments (maybe I should try to increase the amount of inserts?) - 100ng of a vector. The final product is transformed by electroporation to the NEB E5a EC cells.
Yes I think more than equimolar inserts is an excellent idea too. Try that first, it might do the trick.
Gibson cloning works best with ratios (backbone to insert) of 1:8 or 1:4
So for your case try increasing the amounts of inserts multiple times, e.g. 1:4:4:4:4
- Badges
- Science topic