Neurobiology:
Study Sees Brain in Process of Seeing

Cell Biology:
Finding NEMO: Latest Crohn’s Disease Clue

Public Health:
Prevention May Double the Effectiveness of Global HIV/AIDS Treatment

Genetics:
Do-It-Yourself DNA Poised to Remake Molecular Biology

Inflammation’s Other Face: Repairing Injury to the Brain

RNA–Protein Binding Makes Pathogen Irresistible

Down Syndrome Diagnosis Need Not Be Negative Experience for Mothers

Proceedings of the Harvard Medical School Faculty Council

Cambridge Health Alliance Gains HMS Affiliation

Fineberg Professorship Established to Advance Practice of Public Health

New Faculty Appointments to Full and Named Professorships

Connelly Named Assistant Dean for Faculty Affairs

News Brief

Ed Reform at a Critical, Malleable Phase

Front Page

Do-It-Yourself DNA Poised to Remake Molecular Biology

Off the shelf or made to order? In the lab, as in the store, this decision usually comes down to economics. Faced with the prospect of spending a lot of time and money trying to synthesize large pieces of DNA, scientists almost always make do with prét â porter, taking what’s available and tailoring it themselves.

Developed in George Church’s lab, a rapid, accurate, and inexpensive method for synthesizing DNA is set to transform the way researchers conduct molecular biology. (Photo by Graham Ramsay)

That may be about to change. In the Dec. 23/30 Nature, HMS professor of genetics George Church and colleagues report a new way to synthesize not just DNA, but possibly whole genomes. Because the technique is highly accurate, costs a fraction of current methods, and can synthesize more than 10 kilobases at a time—in comparison to current technologies that manage about 100 base pairs—it is set to revolutionize synthetic biology.

DNA from a Chip?

“Almost everybody in these biological research buildings is making DNA constructs to test their hypotheses, and these are made semi-synthetically because it is too expensive to do it any other way. But with this technology, they could sit at their computer, design the sequence they want, and have it synthesized automatically,” said Church. For molecular biologists, who spend hours going through rounds of cutting, splicing, and propagating DNA to get the desired construct, the benefits would be enormous. As proof of principle, research fellow Jingdong Tian synthesized in one fell swoop all 21 genes of the small ribosomal subunit from Escherichia coli.

Church was hesitant to pinpoint any one application for this technology, but possibilities include synthesizing souped-up ribosomes for the production of therapeutic proteins and vaccines in vitro or completely re-engineering biotech workhorses like E. coli. Though these advances are theoretically possible now, in reality it would take years before they could be made with current technology. The efficiency and accuracy of the new method changes that.

“With this technology, molecular biologists could sit at their computer, design the sequence they want, and have it synthesized automatically.”

Called multiplex gene synthesis, the technique builds on DNA chip technology. Chips can contain thousands of different oligonucleotides of about 50 base pairs in length. Because these oligos can be designed to have complementary ends that can link the segments together in specific sequences, chips can serve as ideal templates for synthesis of long stretches of DNA. The only snag is that chips, by their very nature, have such miniscule amounts of DNA, that when they are released into solution, there is insufficient mass action to drive the assembly of different oligonucleotides.

The obvious solution is to amplify the DNA first, which Church’s group found works quite well. But there is a problem—synthesis is notoriously error prone. To get over this hurdle, Church and colleagues have fashioned a second set of “selection” chips. These are printed with oligonucleotides that bind to amplification products from the “construction” chips—a selection oligo for each construction oligo. When these selection nucleotides are amplified, immobilized, and used to pull construction oligos out of solution through stringent hybridization, error-free oligonucleotides are overwhelmingly retained over those with mistakes. In the absence of this selection process, the error rate is about 1 in 160 base pairs, with selection this number becomes about 1 in 1,400.

Automated Assembly

Tian, Church, and colleagues, including Xiaolian Gao at the University of Houston, have developed a way to put this whole process—amplification, selection, and polymerase extensions—into a single polymerase assembly multiplexing reaction, or PAM for short. The power of this one-stop shop is emphasized by the synthesis of the 21 E. coli ribosomal genes, linked through sequential PAM reactions to give a 14.6 kilobase segment of DNA, a feat that was considered challenging before this paper was published. The versatility of PAM is exemplified by the ability to then alter these genes so they are translated much more efficiently in vitro. The cost benefits speak for themselves. With improvements in the technique, Church estimates that DNA could be synthesized for as little as $1 for 20 kilobases, compared to the current cost of $1 for 9 bases.

If there is a downside, it is the possibility that the technology could be put to sinister use, a threat that has been widely publicized. Pathogen genomes, for example, could be synthesized with this technology if they are small and the sequence publicly available. Acutely aware of this, Church has written a white paper calling for discussion of ways to assure appropriate uses of the technology. Because the tools needed for PAM DNA synthesis are rare—only about three laboratories worldwide have the capability—a window of opportunity exists for protocols to be put in place to ensure that the science can move forward openly and that investigators worldwide are able to avail themselves of tailor-made DNA.

—Tom Fagan