This tutorial describes the process of designing a construction file for a plasmid using Golden Gate assembly. Golden Gate cloning and related methods use a type IIs restriction endonuclease, usually BsaI, BsmBI, or AarI, to make scarless junctions between two DNA ends. This can be done from plasmids or PCR products; we discuss the latter scenario herein.
Type IIs Restriction Endonucleases
Type IIs Restriction Endonucleases, like your more typical EcoRI and BamHI type enzymes, recognize a specific sequence but the sequence is not palindromic, and the cut site is outside of the recognition sequence a few bases away. Cutting with these enzymes results in a sticky end derived from adjacent sequence, and the enzyme is unopinionated about the identity of that sequence. Thus, any 4bp sequence can become the sticky end. This is nice because the 4bp sequence, which I'll refer to as the ligation junction, doesn't have to be palindromic such that there is only one way the fragments can ligate. It also enables the generation of scarless junctions without the need to introduce recognition sequences into the final product.
The golden gate assembly reaction is carried out in one pot with both the type IIs enzyme and T4 DNA ligase. You typically will set these up in a PCR tube at about 10 uL volume with nanogram quantities of DNA, and then run a temperature program on the thermocycler. Many other guides online cover specific protocol implementations.
I recommend reading Thermofisher Explanation for a better idea of how this works.
Example Golden Gate Construction File
In this example, we are trying to replace the spectinomycin resistance gene in the CRISPR plasmid pTargetF with an ampicillin resistance gene. The ampR gene is PCR amplified from a template plasmid p20N5 resulting in PCR product fragA. Everything but the specR gene in pTargetF is amplified by PCR for fragB. The two fragments are then joined in the assembly reaction with BsaI and T4 DNA Ligase, and then transformed into Mach1 cells. The product plasmid is called pTarg1
//Replace the SmR gene in pTargetF by Bla pcr yyBla-F/yyBla-R on p20N5 (1183 bp, fragA) pcr yyEI-F/yyEI-R on pTargetF (1174 bp, fragB) assemble fragA, fragB (BsaI, pTarg1) Transform pTarg1 (Mach1, Amp) --------------------------------------------------- >yyBla-F Forward BsaI for Bla gene in p20N5 atgctGGTCTCactcgctgaaattctgcctcgtg >yyBla-R Reverse BsaI for Bla gene in p20N5 agctaGGTCTCgattatcaaaaaggatcttcacc >yyEI-F Forward BsaI for pTargetF template agtcaGGTCTCataatcgcgtaaaggatctaggtga >yyEI-R Reverse BsaI for pTargetF template tggtgGGTCTCtcgagtagggataacagggt
Try simulating this construction file and predict the sequence of pTarg1. You can then compare your answer to the model provided.
Designing a Golden Gate Construction File
- Construct the sequences of the fragments you are joining. Make a file of the fragments in your sequence editor
- Identify 4 bp as the ligation junction. You might add 4 bp between the two joined sequences, you can pick the last 4 bp of one of the two fragments, or have it flank the junction. In the latter two strategies, it will be scarless.
- Find a ~20 bp annealing region on the fragment ends adjacent to the ligation junction
- Copy and paste the BsaI sites and 5’ tails on the 5' end of your sequence. An example sequence is ccataGGTCTCa.
There will be two oligos associated with each golden gate junction. In both cases the BsaI site will be on the 5' end, and a region identical to the template DNA will be on the 3' end. The most challenging thing about all this is keeping your wits about you with respect to the reverse/complement orientation of the annealing region. Suffice it to say, the PCRs have to "work" and thus when you check your construction file you will see whether you have it right or not.
Your construction file will involve a PCR step for each fragment to be joined, and then a single golden gate assemble step to stitch them together.