I'm using the gibson assembly to assemble 3 fragments (1.5kb 1,5kb and 3kb) into a vector (topo2.1 4kb) to generate a 10Kb plasmid. 2 fragments and the vector are PCR amplified and purified from gel while the 3Kb fragment has been synthesized (gBlock). All fragments contain 20bp overlapping sequences.
I obtained many colonies following transformation but unfortunately I never got properly assemble vectors (digestion analyses showed abnormal digestion patterns with overall shorter or longer size compared to the expected one, as if fragments have not been included or included multiple times).
I tried to change the molar ratios between inserts and vector from 1:1 to 3:1, I also used different bacterial strain mutated in RecA genes, and tried the new HiFi DNA assembly kit from NEB but with no success.
Can anybody give me some advice for successful assembly? Any help is appreciated
Hi Paulo,
Too weird! I am sure you have already, but have you checked all your fragments to make sure they are correct? Also recheck your overlaps and make sure they are correct. The only time we have had Gibson Assembly fail to give us the correct construct has been an experiment design problem on our part! Most common for beginners (and I am not saying that you are a beginner, just saying that in my experience the first couple of times people use Gibson Assembly, this is where they go wrong) is to have the overlaps reverse complimented to the orientation it should be. Another unexpected issue we had run into was that one of our cloning strategies involves a large binary vector that contained repeats. Having used one of these repeated areas gave one of our grad students a lot of grief.
I hope you figure it out, sorry I can't be more help!
Cheers
Ron
Most likely one of your overlaps doesn't want to assemble. This isn't necessarily "your fault", as in a non-complementary sequence. The assembly isn't nearly as sequence-independent as they make you believe. Some of the most important limitations are getting under-emphasized, like the extreme sensitivity to high GC content and secondary structures. That's not necessarily only in the overlap regions. I've had reactions around secondary-structure rich features like terminators or regulatory regions fail miserably, and they were unfixable by changing the overlaps.
I think what happens here is that the ends can get obscured by the structures and become unavailable for assembly. I've even seen clear evidence of breakage and re-joining in structure-rich regions. You can't necessarily predict all of that, but sometimes you can. So look at your sequence for structural features, and try to get your overlaps to be as far away from them as possible.
You can also determine experimentally where the failure occurs by breaking the reaction down into 2-fragment assemblies. When you do 3 at a time, it can easily work with an A-T balanced, no-structure model sequence. But with a "real" sequence, containing functional features, the failure probability at each step shoots up, and when you do multiple at a time they compound against you exponentially.
Same problem with us. We'd check the over lap sequences many times and they are okay. however, whenever we do digestion of plasmid DNA we can't get the right products.
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Molecular Biological Techniques