NEW YORK (GenomeWeb News) – Researchers from Harvard University and Agilent Technologies have developed a strategy for using DNA microchips to ramp up — and reduce the cost of — gene synthesis.
The new approach relies on selective enrichment of long oligonucleotides from DNA microchips, allowing researchers to synthesize and assemble many oligonucleotides simultaneously. That, in turn, provides an opportunity to scale up and streamline the gene synthesis process. A study demonstrating the feasibility of the approach appears in the current issue of Nature Biotechnology.
"[W]e use high-fidelity DNA microchips, selective oligonucleotide pool amplification, optimized gene assembly protocols, and enzymatic error correction to develop a method for highly parallel gene synthesis," senior author George Church, a genetics researcher affiliated with Harvard Medical School and the Wyss Institute for Biologically Inspired Engineering, and co-authors wrote.
"This innovative new solution could bring us several steps closer to realizing the full potential of synthetic biology to address real-world medical needs and real-world
environmental issues," Wyss Institute founding director Donald Ingber, who was not directly involved in the study, said in a statement, calling the work "a major breakthrough in gene synthesis technology development."
Although the ability to synthesize genes and gene networks holds promise for facilitating synthetic biology, genetic engineering, and other applications, the researchers noted, synthesizing pieces of DNA of a sufficient length to produce complete genes is a pricey prospect.
DNA microchips, which accommodate as many as a million small pieces of DNA or oligonucleotides at once, offer a potential solution, they added, but have proven difficult to use in large scale gene synthesis because of cross-hybridization associated with such chips, as well as other issues.
"Oligonucleotides from DNA microchips can reduce costs by at least an order of magnitude, yet efforts to scale their use have been largely unsuccessful owing to the high error rates and complexity of the oligonucleotide mixtures," they explained.
To overcome such complications, the team attempted to exploit oligo library synthesis or OLS pools containing relatively high quality oligonucleotides synthesized under conditions that minimize depurination-related errors.
The researchers made two mixtures of oligonucleotides using specially designed 130-mer or 200-mer oligonucleotides that were initially printed onto high-fidelity DNA chips. These oligos were liberated from the chip to create OLS pools that were subsequently used to amplify specific sub-sets of the oligos needed for putting together specific DNA sequences.
After processing these small pieces of DNA and removing primers using restriction enzymes, the oligos of interest were then assembled into genes in PCR reactions, the team explained, noting that mismatched repair enzymes were used in some instances to help minimize errors in the newly synthesized sequences before cloning the assemblies into expression vectors.
Despite the complicated nature of these mixtures, which included nearly 13,000 oligos coding for millions of DNA bases, the team found that they could accurately construct specific gene-length sequences. For instance, they noted, the pieces of gene-length DNA constructed from 130-mer OLS pools had completely correct sequences more than half of the time.
The researchers also tested the method by attempting to synthesize 47 genes. Among them: 42 genes coding for the variable regions of therapeutically useful antibodies containing repeat rich linker sequences.
Although they found that the antibody sequences were more challenging to synthesize and assemble than some other genes, sequencing analyses of 15 antibody genes showed that they had, indeed, cloned 14 of these antibody sequences successfully.
Given their findings so far, the researchers argue that their new methodology can be used to construct gene-sized stretches of DNA using microchips, provided appropriate tools are available.
"A number of key features are important to make the process work, including the use of low-error starting material, well chosen orthogonal primers, sub-pool amplification of individual assemblies, optimized assembly methods, and enzymatic error correction," the authors explained. "Together, these features enabled gene assembly from oligonucleotide pools containing at least 50 times more sequence than previously reported."