Opinion: Evolving Engineering

Exploiting the unique properties of living systems makes synthetic biologists better engineers.

Oct 1, 2011
George M. Church

George Church with a tape for measuring his facial morphology for a research project PERSONAL GENOME PROJECT

After 10 years of tinkering with biological circuits, we need to explain—once again and clearly—the rationale for doing synthetic biology. Despite the musings of some, the field is not limited to toy projects. Metabolic engineers have clearly articulated their goals of manufacturing cheap alkane fuels and much-needed drugs, such as the antimalarial drug artemisinin.[1. D.K. Ro et al., "Production of the antimalarial drug precursor artemisinic acid in engineered yeast," Nature 440: 940-43, 2006.] (See “Tinkering with Life.”) But for building DNA nanostructures or whole bacterial genomes, the rationales have been less clear—initially confined to cartoonish shapes and watermark sequences, respectively. Recent advances, however—such as a DNA nanostructure that combines cell targeting, molecular logic, and cancer-fighting ability[2. S.M. Douglas et al., “A logic-gated DNA capsule for targeted transport of molecular payloads,” Nature Biotech (revised version in review), 2011.], and a new E. coli genome well on its way to possessing multivirus resistance[3. F.J. Isaacs et al., “Precise manipulation of chromosomes in vivo enables genome-wide codon replacement,” Science, 333:348-53, 2011.]—have demonstrated the discipline’s incredible potential.

Much of the progress can be credited to engineers who have developed a deeper appreciation of life’s power. While synthetic biology has brought a welcome injection of rigorous engineering principles to biology, including hierarchical abstractions, computer-aided design (CAD), and interoperable parts, biological mechanisms also offer some distinctive qualities of their own—a handful of underexploited strategies previously rare in engineering fields, such as replication at low cost and natural selection.

In just 6 years, researchers have reduced the cost of genome reading by a millionfold, and we are now accomplishing a similar effort in writing DNA—thanks to a new technique for harvesting synthetic oligonucleotides from chips, which can generate 60 million linked bases for just $900.[4. S. Kosuri et al., “Scalable gene synthesis by selective amplification of DNA pools from high-fidelity microchips,” Nat Biotechnol, 28:1295-99. 2010.] Moreover, we can now create expansive genetic libraries, generating billions of genome variants per day using targeted mutagenesis. Those combinations can then be pitted against each other in an evolutionary footrace, allowing researchers to quickly ferret out the good gene combinations—for example, those that yield high levels of a desired compound—from the bad.

Biologically inspired devices also offer other advantages that may increasingly allow them to compete with silicon-based electronics. DNA is over a billion times more compact per bit than the densest electronic storage or Blu-ray Disc, and polymerase steps are 10 million times more energy efficient than conventional computer unit operations. Indeed, these properties are allowing hybrid bio/optical/electronic systems to grow in diversity and complexity (See, for example, “The Birth of Optogenetics,” The Scientist, July 2011).

Lab evolution and synthetic biology are about embracing the outliers and creating the occasional hopeful monster, just as evolution has done for millions of years.

Synthetic biology will also increasingly be applied in a clinical setting. Semisynthetic organs, such as bladders, have been in medical use for years. But while classical tissue-engineering techniques used simple polymers and natural tissues, researchers are now designing gene circuits to build scaffolds and reprogram cell fates, which will enable the construction of human organs with new, customizable properties. Furthermore, as vaccine and drug production declines and antibiotic resistance increases, researchers may increasingly turn to syn bio-aided alternatives, such as targeted genome engineering.

The new techniques can be used to correct problems common to most human genomes, rather than simply fixing rare mutations. For example, Tim Brown was simultaneously cured of both leukemia and HIV-AIDS after receiving a transplant of histocompatible stem cells, which also happened to lack the HIV receptor protein CCR5.[5. G. Hütter et al., “Long-term control of HIV by CCR5 ?32/?32 stem-cell transplantation,” N Engl J Med, 360:692-98, 2009.] He is still disease free 4 years later. That serendipitous result has led to a so-far-successful clinical trial of a zinc finger nuclease therapy aimed at eliminating the human CCR5 gene from HIV-infected individuals.

This is a significant departure from the traditional notion of gene therapy: rather than “normalize” the mutated genome of someone with a rare disease, we are utilizing rare (or synthetic) genotypes to “cure” or “improve” normal genotypes. Other examples of future contributions based on synthetic biology might include therapeutic human immune cells that make a neutralizing antibody observed very rarely in our global population—a possible alternative to potentially fruitless searches for an HIV vaccine aimed at evoking such an uncommon antibody. It is quickly becoming clear that “normalization” is not always best. Lab evolution and synthetic biology are about embracing the outliers and creating the occasional hopeful monster, just as evolution has done for millions of years.

George M. Church is a professor of genetics at Harvard Medical School and director of the university’s Center for Computational Genetics. Additional reading and citations can be found in the online version of this article.