Criteria for Choosing a Transposon System

Researchers can choose from among several in vitro transposition systems and two in vivo systems. Epicentre has hyperactive Tn5-based EZ-Tn5™ Transposon Systems and hyperactive phage Mu-based HyperMu™ Transposon Systems for both in vitro and in vivo applications. The properties of these and other commercial systems are summarized in the following table and important criteria for choosing between them are discussed below.

Artificial Transposon Transposition
In vitro
Efficiency
In vivo*
Randomness # Basepairs in a Transposon End Target Sequence Duplication (bp) # Transposase Proteins Needed
EZ-Tn5™ High High Demonstrated ME, 19 9 1 (Tnp)
HyperMu™** High Medium Demonstrated ~51 5 1 (MuA)
Tn7-Type Medium Not Detectable Demonstrated Tn7L, 165 Tn7R, 90-198 5 3 (TsnABC)
Ty1 Low Not Done Demonstrated 4 5 Virus-Like
Particle Complex

* The "in vivo transposition frequency" data here refers to the relative number of colonies obtained containing transposon insertions following electroporation of electrocompetent E. coli cells with a synaptic complex (Transposome™ Complex) formed between a transposon and the respective transposase.

** HyperMu™ MuA Transposase can insert transposons with Mu Ends having free 3'-termini into another DNA, but is unable to excise the transposon from another DNA.


Transposition efficiency is one important criterium for judging a transposon system. High transposition efficiency is especially desirable for insertion of primer-binding sites into large BAC or fosmid clones in vitro in order to determine its complete sequence. High transposition efficiency is also important for insertion of a transposon into DNA in a living cell in vivo. EZ-Tn5 Systems have very high transposition efficiencies for in vitro insertion, as well as for in vivo insertion of an EZ-Tn5 Transposome™ Complex, the synaptic complex formed between an EZ-Tn5 Transposase and an EZ-Tn5 Transposon. Although Epicentre's HyperMu™ MuA Transposase is at least 50 times more active for in vitro transposition than the MuA transposase from other suppliers and almost as active as the EZ-Tn5 System in vitro, the HyperMu Transposome™ Complex generates 10-100 times fewer insertions than the EZ-Tn5 System in vivo using E. coli. Tn7-based transposon systems are quite efficient for in vitro insertion, but have undetectable activity as a synaptic complex, probably because the active Tn7 transposase consists of 3 proteins, which may be difficult to electroporate into the cell as a complex. The Ty1 system has very low transposition efficiency.

For most applications, it is also important that transposition is random. Based on our experience with different systems, we believe that the EZ-Tn5 System is the most random. However, all four transposon systems seem to be sufficiently random for most applications, including for determining the complete sequence of BAC or fosmid clones. Although Tn7 insertion is highly specific in the presence of the TsnE Protein, target specificity is not observed using a transposase consisting of only the TsnABC proteins. A study comparing all systems using transposons with the same selectable marker and other identical transposon features into identical target sequences has not been published. Comparisons using different selectable markers or other transposon features, or different targets are flawed because variations in transposon features and in the DNA target chosen can affect which insertions are viable and selected for analysis.

In general, it is desirable that the length of the transposase recognition sequences in the "transposon ends" are short when primer binding sites in the transposon are used for sequencing. Shorter transposon ends maximize sequence data obtained from the target DNA and minimize undesired sequence data from the end of the primer to the end of the transposon.

Approximate Amount of Template (in nanograms) Required for DNA Sequencing Reactions
Template Required (fmoles) Template Size (insert + vector) in Kilobases
1 kb 3 kb 5 kb 7 kb 9 kb
25
17 ng 50 83 120 150
50 33 100 165 230 300
75
50 150 250 350 450
100 66 200 330 460
600
150 100 300 500 700 900
200 135 400 660 930 1200
250 165 500 830 1200 1550
300 200 600 1000 1400 1800
Estimating Molar Amounts of Template
To estimate the weight of 100 fmoles of double-stranded DNA template, multiply the number of kilobases by 66 ng. For example, 100 fmoles of a 2.7 kb plasmid is 178 ng (2.7 x 66 ng = 178 ng).

[top of page]