Modern synthetic biology involves not only the construction of existing genes to elicit their functions, but also designed mutation and sequence shuffling to create new, functional gene constructs which perform as biomolecular machines. Generally, synthesis of gene size fragments (500∼5000bp) typically begins with oligonucleotides (oligos) as building blocks and there now exists several methods for assembling these gene-size fragments into much longer synthetic DNA constructs[1-4].
A successful de novo gene synthesis strategy from oligos must overcome two primary obstacles:
- Relatively high cost and low-throughput oligo synthesis methods
- Errors introduced during the oligo synthesis process
Multiplex, parallel DNA construction on a large scale requires huge pools of large numbers (often tens of thousands) of short synthetic oligos, and individual synthesis of each sequence using conventional CPG solid phase synthesis technology (~$0.40-$1.00/bp) is simply not feasible.
The use of microarray technology has provided part of the solution by enabling the synthesis of large numbers of oligos in massive parallel on a microchip at a very low cost per sequence (<$0.01/bp). Designed sequences can be simultaneously synthesized and then assembled by joining the short synthetic oligos with multiplex reactions. However, the challenges associated with synthesis errors have remained and the development of economical de novo gene synthesis methods using microarray-synthesized oligos has been limited by their high error rates. Consequently, oligo quality has become the primary obstacle to de novo gene synthesis.
There have been some oligo purification strategies (such as HPLC and PAGE) applied to error correction of complex pools of microarray synthesized oligos, however these methods are ineffective at removing substitution errors or when oligos in the pool vary greatly in length and Tm. Enzymatic mismatch cleavage methods have also been applied, however these methods are expensive and time-consuming for scalable multi-gene treatments and have only been validated on assembled products, not on the initial oligo pools. Therefore, these existing approaches for error correction of DNA are not suitable for low-cost and high-throughput error removal from microarray synthesized oligos that form a complex pool consisting of oligos with high error rates.
Recently, researchers at the University of Science and Technology of China have developed a high-throughput and cost-effective error-removal method that can conveniently remove errors from microarray synthesized oligo pools or assembled fragments. This method involves the use of a cellulose purification column containing immobilized mismatch binding protein MutS which can specifically recognize and bind to all possible single-base mismatches, as well as 1-5 bases insertion or deletion loops. The researchers have developed a recombinant MutS fusion protein consisting of CBM3, EGFP, MutS and a 6-His tag. The recombinant MutS fusion proteins can be immobilized on cellulose via CBM3, and the protein purification and immobilization of the constructed fusion proteins can be monitored via the fluorescence of EGFP. Initial DNA material is loaded onto the MutS-immobilized cellulose column. Error-containing DNA is speciﬁcally retained while error-depleted DNA flows through and is collected in the eluate for downstream gene assembly. The researchers applied their error correction procedure at both the oligo pool and assembled fragment stages and demonstrated significant decrease in errors at both stages
This method improved a population of synthetic enhanced green ﬂuorescent protein (720 bp) clones from 0.93% to 83.22%, corresponding to a decrease in the error frequency from 11.44/kb to 0.46/kb. A parallel multiplex error removal strategy was also evaluated in assembling 11 genes encoding ∼21 kb of DNA from 893 oligos. The error frequency was reduced by 21.59-fold (from 14.25/kb to 0.66/kb), resulting in a 24.48-fold increase in the percentage of error-free assembled fragments (from 3.23% to 79.07%).
During de novo gene synthesis, the greatest concern is the number of clones that must be analyzed to identify at least one error-free sequence. These researchers report that screening only one to three clones was required to obtain at least one 1 kb error-free synthetic gene (probability >90%) when using this error correction method. In contrast, without error removal, 47 to 48 clones must be analyzed to obtain a 1 kb error-free gene, which means a vast waste of materials, time, effort and ultimately money.
Furthermore, this simple MutS column error-removal process could be completed within 1.5 hours at a cost of less than $1.00 per column.
- Matzas M, Stähler PF, Kefer N, Siebelt N, Boisguérin V, Leonard JT, Keller A, Stähler CF, Häberle P, Gharizadeh B, Babrzadeh F, Church GM. (2010) High-fidelity gene synthesis by retrieval of sequence-verified DNA identified using high-throughput pyrosequencing. Nat Biotechnol 28(12), 1291-94.
- Borovkov AY, Loskutov AV, Robida MD, Day KM, Cano JA, Le Olson T, Patel H, Brown K, Hunter PD, Sykes KF. (2010) High-quality gene assembly directly from unpurified mixtures of microarray-synthesized oligonucleotides. Nucleic Acids Res 38(19), e180.
- Zhou, X, Cai S, Hong A, Yu P, Sheng N, Srivannavit O, Yong Q, Muranjan S, Rouillard J M, Xia Y, Zhang X, Xiang Q, Ganesh R, Zhu Q, Makejko A, Gulari E, Gao X. (2004) Microfluidic picoarray synthesis of oligodeoxynucleotides and simultaneously assembling of multiple DNA sequences. Nucleic Acids Res 32, 5409-17.
- Tian J, Gong H, Sheng N, Zhou X, Gulari E, Gao X, Church G. (2004) Accurate multiplex gene synthesis from programmable DNA chips. Nature 432, 1050-54.
- Ellington A, Pollard JD. (2001) Introduction to the synthesis and purification of oligonucleotides. Curr Protoc Nucleic Acid Chem Appendix 3C.
- Saaem,I.,Ma,S.Y.,Quan,J.Y.andTian,J.D. (2012) Error correction of microchip synthesized genes using Surveyor nuclease. Nucleic Acids Res 40, e23.
- Wan W, Li L, Xu Q, Wang Z, Yao Y, Wang R, Zhang J, Liu H, Gao X, Hong J. (2014) Error removal in microchip-synthesized DNA using immobilized MutS. Nucleic Acids Res 42(12):e102.