Help -> Gibson

Gibson Assembly

Gibson Assembly, as described by Daniel Gibson in Nature, 2009, allows you to create large constructs of DNA from an arbitrary number of fragments in one step without scarring.

Gibson Assembly assumes that there is an overlap between adjacent strands of DNA. This is of course not normally the case and so an initial stage of extension must be performed to create this region of homology. The primers for this is what this site will help you design

For a slightly more rigorous explanation, have a look at this page on synbio.org.uk

Design of primers is covered in more detail in this section

It is most likely that your DNA will not already overlap, so you'll need to add the overlap yourself. A simple PCR with a primer comprised of two bits of DNA - one from the target DNA (the annealing portion) and one from the adjacent DNA (the flappy end).

This means, to join two strands of DNA (A and B) with an overlap of 40, A will need to be extended as follows

  1. Primer for top strand of A: Comprises the complement of the last 20 bp of the top strand of A, and the complement of the first 20 bp of the top strand of B, all reversed.
  2. Primer for bottom strand of A: Comprises the last 20 bp of the top strand of B, and the first 20 bp of the top strand of A.

These are all written and read 5' to 3', and the primers for B are more or less the reverse complement of the primers for A.

This is better explained with an example, using an overlap of 8

Strand A: 5' ATGGTACCGGA 3'     Strand B: 5' GGTACCAGTTC 3'
          3' TACCATGGCCT 5'               3' CCATGGTCAAG 5'
			
           3' GCCTCCAT 5'
    5' ATGGTACCGGA 3'    
    3' TACCATGGCCT 5'    
5' GTTCATGG 3'           

Strand A: Top:    5' TACCGCCT 3'
          Bottom: 5' GTTCATGG 3'

Strand B: Top:    5' CGGAGGTA 3'
          Bottom: 5' CCATGAAC 3'

It is likely that you will be using DNA that is already in a vector of some sorts - the procedure is exactly the same as the diagrams below show:

(1) PCR cycle 1: melting

A single strand of the vector containing RFPm

(2) PCR cycle 1: annealing

One of the Gibson primers attaches by the annealing portion to the RFP. The flappy end does nothing.

(3) PCR cycle 1: extension

The primer is extended around the full vector.

(4) PCR cycle 2: melting

We now have a linear single strand of DNA with one extension.

(5) PCR cycle 2: annealing

The other Gibson primer attaches by the annealing portion to the other end of the RFP. The flappy end does nothing.

(6) PCR cycle 2: extension

The primer is extended to the end of the RFP plus the short extension for the overlap.

(7) All further PCR cycles

PCR now continues with the Gibson primers acting as normal primers. Both the annealing portions and flappy ends anneal to the extended DNA.

The principle at one join is as follows, with the two strands in black and red, and enzyme action in blue:

5'...ATCGAGGCTGTTAGGAGTATTACGTATTCGAGGATTCGAGC 3'                      
3'...TAGCTCCGACAATCCTCATAATGCATAAGCTCCTAAGCTCG 5'                      

                      5' TACGTATTCGAGGATTCGAGCAGTCGATCAGGATTCGATTC...3'
                      3' ATGCATAAGCTCCTAAGCTCGTCAGCTAGTCCTAAGCTAAG...5'

Stage 1: two DNA strands with a 40bp overlap (shown here as 20bp)

5'...ATCGAGGCTGTTAGGAGTATTACGTATTCGAGGATTCGAGC 3'                      
3'...TAGCTCC 5' <-- exonuclease                                        
                                        exonuclease --> 5' TCGATTC...3'
                      3' ATGCATAAGCTCCTAAGCTCGTCAGCTAGTCCTAAGCTAAG...5'

Stage 2: As the sample is heated to 50°C, an exonuclease chews back 5' to 3'

				                          
                                               polymerase              
5'...ATCGAGGCTGTTAGGAGTATTACGTATTCGAGGATTCGAGC 3' -->   5' TCGATTC...3'
3'...TAGCTCC 5'   <-- 3' ATGCATAAGCTCCTAAGCTCGTCAGCTAGTCCTAAGCTAAG...5'
              polymerase

Stage 3: At 50°C, the sticky ends anneal, and a polymerase begins repairing the gap 3' to 5'

                                                        ligase         
5'...ATCGAGGCTGTTAGGAGTATTACGTATTCGAGGATTCGAGCAGTCGATCAGGATTCGATTC...3'
3'...TAGCTCCGACAATCCTCATAATGCATAAGCTCCTAAGCTCGTCAGCTAGTCCTAAGCTAAG...5'
         ligase

Stage 4: Finally, a ligase repairs the backbone and the seamless join is complete

Both of these protocols were found to work well by the 2010 Cambridge iGEM team.

Assembly

For assembly, the following was found to be highly reliable

  1. Prepare the Assembly Mix on ice
  2. Incubate for one hour at 50°C
  3. Optional gel purification
  4. Transform
Assembly MixVolume (μl)
Gibson Master Mix15
Total DNA5
Total20

For instance, if you are joining 4 bits of DNA you would use about 1.25μl of each. There is a limit at which point the concentrations and thus yield becomes unhelpfully low, but we have not found it yet.

Gibson Master MixVolume (μl)
Taq ligase (40u/µl)50
5x isothermal buffer100
T5 exonuclease (1u/µl)2
Phusion polymerase (2u/µl)6.25
Nuclease-free water216.75
Total375

Whilst the mixture can be refrozen, for best performance you should aliquot it into 25x15μl.

5x isothermal bufferVolume (µl)
25% PEG-80000.75g
500 mM Tris-HCl pH 7.51500
50mM MgCl275
50mM DTT150
1mM dATP30
1mM dTTP30
1mM dCTP30
1mM dGTP30
5mM NAD300
Nuclease-free water...
Total3000

Use nuclease-free water to make up to 3000μl. In this case, very little water is needed as the volume is 2145μl + 0.75g.

Extension

In time, the protocol for extension will be provided in detail by Gibthon; in the mean time here is a template from which to work.

Note that Tm for the primer here refers to the annealing portion of the primer.

TemperatureDuration
Initial Melting98°C30s
Begin cycle: Repeat ~30 times
Melting98°C10s
AnnealingTm°C15s
Elongation72°C45s/kbp DNA
End cycle
Final elongation72°C7m30s
Final hold4°C∞