Send to printer »

Tutorials : Oct 15, 2010 (Vol. 30, No. 18)

Next-Generation DNA Assembly Tools

Emergence of Synthetic Biology Necessitates New Solutions to Generate Molecules
  • Lansha Peng
  • ,
  • Billyana Tsvetanova, Ph.D.
  • ,
  • Xiquan Liang, Ph.D.
  • ,
  • Federico Katzen

Genome engineering has expanded to cover myriad applications including the analysis of complex pathways, the construction of new biological parts, and the re-design of existing biological systems. All these areas require the precise and concerted assembly of multiple DNA fragments of various sizes, including chromosomes.

Current commercial cloning products are not robust enough to support the assembly of very large or very small genetic elements or a combination of both. In addition, current strategies are not flexible enough to allow further modifications to the original design without having to undergo complicated cloning strategies.

In this article we present a set of technologies from Invitrogen, part of Life Technologies, that allow the seamless, simultaneous, flexible, and highly efficient assembly of genetic material, designed for a wide dynamic size range. The assembly can be performed either in vitro using an enzymatic mix or within the living cells.

In Vivo Recombineering

The in vivo approach requires the transformation into yeast of two or more linear DNA molecules that share unique common end sequences. End-homology can even be provided in trans by overlapping oligonucleotides. If one of the molecules harbors replication and selection elements for yeast and E. coli, then circular episomes up to 110 kilo base-pairs (kbp) in length can be assembled in yeast and transferred to E. coli for downstream applications.

For this purpose a new vector, pYES1L, was constructed (Figure 1A). The plasmid contains a S. cerevisiae’s centromere and the trp1 gene for selection in yeast. In addition F´ ori, the spectinomycin resistance gene, an oriT, and the bacteriophage λ COS site were added for manipulation in E. coli. Rare restriction sites were added for cloning and mapping. We also designed a cassette to convert any E. coli plasmid into a YAC episome, with yeast features similar to pYES1L plus the URA3 gene as an optional counter-selectable marker (Figure 2B).

The system was tested by simultaneously assembling up to 10 linear fragments in a predetermined order into pYES1L (Figure 1C). Adjacent DNA fragments, which shared 30 to 80 base pairs (bp) of end-homology, had been previously cloned into pACYC184 or directly amplified by PCR.

DNA (up to 200 ng) was combined with S. cerevisiae MaV203 competent cells (trp1-901; URA3) and plated onto tryptophan-free CMS agar medium. Clones were verified by colony-PCR and sequencing and/or transferred back to E. coli by directly electroporating colony lysates for restriction profiling. Cloning efficiencies ranged from 58 to 100%.

The approach has also been applied to recombine fragments that do not share end-homology. Here, homology is provided in trans by complementary oligonucleotides that overlap both fragments (Figure 1D). The method is advantageous for re-using fragments in a new sequence context, for cloning DNA targets that cannot be readily amplified by PCR, or for editing the fragment junctions generating end imperfections.

In Vitro Recombineering

In a parallel approach we sought to accomplish DNA fragment assembly in vitro where the assembled molecules are directly selected in E. coli. For this purpose we developed a highly efficient enzymatic mix that promotes homologous recombination of up to four DNA fragments plus a vector with 15 bp end-homology. (Note: we were able to prove assembly of up to seven fragments, but the mix is optimized for up to four fragments plus vector).

DNA molecules (20 to 500 ng per fragment in a 2:1 insert:vector molar ratio) were incubated with the enzyme for 30 minutes  at room temperature, transformed into TOP10 chemically competent cells, and selected on LB agar plates supplemented with the corresponding antibiotic. Cloning efficiencies ranged from 40% to >90% depending on the number and quality of the DNA fragments (Figure 2A).

Recombination can occur not only at the end of the fragments, but it also works at least up to 32 bp away from their ends (Figure 2B). This attribute is useful for generating cloning variants using a single linearized vector.

The enzyme described above can also  be used for recombining and editing the ends of a single DNA molecule, thereby enabling a highly efficient site-directed mutagenesis approach (Figure 3A). Two complementary oligonucleotides with centrally located mutation sites were used.

The DNA methylation and amplification steps were combined into a single reaction. No in vitro digestion or DNA purification was required after methylation or mutagenesis. The system can generate base substitutions, deletions, or insertions of up to 12 nucleotides in plasmids as large as 14 kbp (Figure 3B).


In anticipation of an imminent paradigm shift, largely due to the emergence of the synthetic biology field, our ultimate goal is to reach a comprehensive solution to generate any DNA molecule up to high-level genetic systems starting from digital sequences stored in a computer. The approaches presented in this article are relevant not only for the area of synthetic biology but they also have remarkable implications for the current cloning standards.