Synthetic biology has historically relied on bacteria as a testing ground for engineering cell behavior through genetic signals. But a small group of researchers have their sights set on

redesigning mammalian cells, which have more complex genetic machinery. DANIEL GRUSHKIN meets the scientists aiming to reprogram our bodies' cells for a new generation of tailor-made

treatments. You have full access to this article via your institution. Download PDF The idea of using a virus to deliver a missing piece of DNA into cells is the simple concept that spurred

the field of gene therapy. But to this day the discipline remains haunted by the 1999 death of Jesse Gelsinger, a teenager who died after undergoing experimental gene therapy for a rare

safety. They are developing souped-up versions of gene therapy that apply what's known as 'synthetic biology', an approach involving a series, or circuit, of genes working

together with controls and—crucially—'off' switches. Credit: Matt Hansen There are early signs of progress. Two years ago, Christina Smolke, a synthetic biologist at Stanford

University in California, developed a control system1 for immune cells in mice. Smolke's system involves introducing a circuit of genes that, once within a cell, perform various

functions within a set sequence. The DNA introduced into the cell produces a molecular complex consisting in part of RNA that is shaped to detect the presence of the antibiotic tetracycline

or the asthma drug theophylline. Once it binds this molecule, a switch is turned on: the RNA complex cleaves, allowing for the translation of a gene that produces a chimeric antigen

receptor. CARs, as they're called, are customized synthetic proteins that allow T cells to recognize cancer cells. Smolke's team observed that these T cells survived when injected

in mice, thus signaling a new direction for synthetic biology—cell therapeutics. “Ultimately, we're hoping to take this into clinical trials,” Smolke says. Although the system currently

works at a cellular level with tetracycline, the idea is that the human T cells used in clinical treatments would flip on the CAR gene therapy in the presence of a less toxic compound

prescribed by a doctor. If such treatment for some reason failed to work or—worse—caused as adverse reaction, the patient would cease taking the compound and the gene therapy would be

rendered inactive. Smolke is not alone in her excitement about synthetic biology. For the last decade, synthetic biologists like her have amassed a toolbox of simple DNA-based commands that

do seemingly impossible things when combined into circuits in bacteria and yeast. From creating bacteria that fluoresce with an oscillating rhythm2 to developing biofuels in bacteria, yeast

and algae that are metabolically reengineered to secrete diesel-like compounds, synthetic biologists have built genetic circuits with increasingly relevant applications. Companies such as

LS9, Solazyme and Amyris, all based in the San Francisco area, have developed microbes encoded to produce a widening suite of chemicals such as fatty acids and organic compounds such as

isoprenoids to make ingredients for medicines and even cosmetics. PROGRAMMED FATE: Smolke sees potential in using synthetic biology to train immune cells. Credit: Linda A. Cicero By moving

beyond bacteria, researchers like Smolke are beginning to show how synthetic biology can introduce gene circuits into mammalian systems to perhaps make gene therapies safer in the future.

“The field is moving pretty aggressively toward mammalian systems,” explains James Collins, a biomedical engineer at Boston University who is considered a father of the field. “Until

recently, the efforts were primarily focused on microbes—_Escherichia coli_ and yeast specifically—because they're easy to use and fast to grow. But I think we're going to see many

of the interesting applications in the next few years emerging in therapeutics,” Collins says. Although synthetic biology has yet to deliver on its clinical promise, Collins and others see

a huge potential for the technology to help in healing one day. His lab is working on how to send a synthetic liposome carrying a set of DNA instructions to a wound site, which would induce

some cells to become stem cells and then morph into a needed cell type, such as peripheral nerve cells. “We've not yet integrated it into a single unit,” he says, but he is quick to add

that individual genetic components have worked individually _in vitro_3. AWAKE AT THE SWITCH The field of synthetic biology erupted about a decade ago when scientists intimated that they

could coax living organisms into behaving as predictably and reliably as electronics. A crucial point came in 2000, when Collins led a team that created the first genetic ON-OFF switch in

_E. coli_. Using two repressible promoters that inhibited each other, he was able to make the bacteria glow green or turn back to normal4. Bacterial cells lack components such as a nucleus,

thereby making them easier to 'hack' with the genes used in synthetic biology. But a milestone for the hacking of mammalian cells came in 2004, when Martin Fussenegger, a

bioengineer at the Swiss Federal Institute of Technology in Zurich, reported that his team had created a genetic switch in Chinese hamster ovary cells that could turn the secretion of the

enzyme alkaline phosphatase on and off5. Much progress has taken place since then. “Now that we're able to design really complex circuitry in mammalian cells, we're not just

copying the design of the circuits that people have pioneered in prokaryotes,” Fussenegger says. This June, Fussenegger's lab took mammalian cell programming to a new level when he

demonstrated a set of biological commands in what he calls a “mammalian biocomputer.”6. Synthetic biologists have already replicated basic computing functions at the cellular level, such as

NOT, AND and NAND. NOT, for example, is a command that acts like a reverser—the 'off' state can be turned on, and vice versa. AND combines multiple inputs, whereas NAND (the

combination of NOT and AND) decouples them. Fussenegger used DNA to produce the more sophisticated N-IMPLY function, which in his system involves complexes that include transcription

proteins. In his group's experiment, the introduction of these 'half-adder' complexes (and related 'half-subtractor' complexes) into human embryonic kidney cells

equipped them to fluoresce according to the addition and removal of the flavonoid compound phloretin and the antibiotic erythromycin. Crucially, the cells fluoresced green if only one of the

two compounds was present, and red if both were there. These experiments lay the groundwork for a blue-sky hope that cells might be programmed to take readings of multiple disease symptoms

before producing a response. Similar systems have been developed using RNA interference where scientists engineered circuits to identify whether a cell had become cancerous before causing it

to self-detonate. Just like a key that fits the pins of a lock, the circuit measured the levels of six different microRNAs in the cell and, if they all matched the profile seen in cervical

cancer, signaled cell death7. GREEN-LIGHTED PLAN: Fussenegger's cells fluoresce. Credit: Dominic Büttner Fussenegger has already shown that bioengineered cells implanted inside mice can

read high uric acid levels in the bloodstream and will pump out the enzyme urate oxidase to neutralize the acid until levels return to normal8. The cells were sealed within porous alginate

capsules to protect them from the immune system and injected into the mice's body cavities. The system may one day combat gout but is really a proof of principle. “This is honestly

where we need to have synthetic biologists working,” he says. “Generating potential therapeutics is the most relevant thing we can contribute to society.” COURTING THE FINANCIERS For all its

potential, fewer than ten synthetic biology labs work on clinical applications. “There has to be more investment from the venture community and other players to support this research, and

that hasn't happened yet,” says Sridaran Natesan, head of external innovation and partnering for the French pharma giant Sanofi in the US northeast. “It's too long term, at least

in people's minds, so it's a risky investment.” Still, he adds that Sanofi is “actively thinking about investing in some promising areas and working with key guys in the field.”

Earlier this year, the company granted $150,000 to Ron Weiss's synthetic biology lab at the Massachusetts Institute of Technology (MIT) in Cambridge, Massachusetts, as part of its

Biomedical Innovation Program with the university. Weiss, director of the MIT Synthetic Biology Center, plans to create gene circuits that can be introduced into human fibroblasts to make

them insulin-producing beta cells in response to a specific signal. Meanwhile, Noubar Afeyan, chief executive of Flagship Ventures, a Cambridge, Massachusetts–based venture capital firm and

a founder of the growing biofuel company LS9, is now searching for the right therapeutic to take on. He estimates that a biotech startup would need at least $100 million to spin an academic

synthetic therapeutic into a technology with commercial value—and that's well before a pharmaceutical company can develop it into a product ready for clinical trials. So, Afeyan is

cautious. “What we would need is to feel better about the safety prospect and find a compelling first application,” he explains. HACKING A WAY FORWARD Many synthetic biologists echo

Afeyan's apprehensions. Chris Voigt, a bioengineer at MIT and one of the leaders in the field, says that synthetic biology still has to prove it can work consistently in mammalian

cells. “One of the core principles of synthetic biology is that if you characterize part A and characterize part B, you can predict how they'll behave together, and in mammalian cells

that's pretty tricky,” he says. Placing a single gene in an exact desired spot in the human genome can be difficult even with the latest technologies. And trying to then insert multiple

genes that work together makes the chances for success even more difficult. “The way you deal with mammalian cells now is by luck,” says Pamela Silver, a system biologist at Harvard Medical

School in Boston. Almost all the research in the field of synthetic biology remains on the level of proof of concept. For now, many experts are watching to see how the US Food and Drug

Administration evaluates gene therapy products for a signal of how regulators might view even more elaborate therapies involving entire gene circuits down the road. “I think once the FDA

gets comfortable with and eventually approves a gene therapy approach, then it's an incremental step to increasingly put in more than just single genes,” Afeyan says. REFERENCES * Chen,

Y.Y. et al. _Proc. Natl. Acad. Sci. USA_ 107, 8531–8536 (2010). Article  CAS  Google Scholar  * Danino, T. et al. _Nature_ 463, 326–330 (2010). Article  CAS  Google Scholar  * Warren, L. et

al. _Cell Stem Cell_ 7, 618–630 (2010). Article  CAS  Google Scholar  * Gardner, T.S. et al. _Nature_ 403, 339–342 (2000). Article  CAS  Google Scholar  * Kramer, B.P. et al. _Nat.

Biotechnol._ 22, 867–870 (2004). Article  CAS  Google Scholar  * Ausländer S. et al. _Nature_ 487, 123–127 (2012). Article  Google Scholar  * Xie, Z. et al. _Science_ 333, 1307–1311 (2011).

Article  CAS  Google Scholar  * Kemmer, C. et al. _Nat. Biotechnol._ 28, 355–360 (2010). Article  CAS  Google Scholar  Download references AUTHOR INFORMATION AUTHORS AND AFFILIATIONS *

Daniel Grushkin is a science journalist in Brooklyn, New York, and the cofounder of Genspace, a community lab and education space that focuses on synthetic biology., Daniel Grushkin Authors

* Daniel Grushkin View author publications You can also search for this author inPubMed Google Scholar RIGHTS AND PERMISSIONS Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE

Grushkin, D. The new drug circuit. _Nat Med_ 18, 1452–1454 (2012). https://doi.org/10.1038/nm1012-1452 Download citation * Published: 05 October 2012 * Issue Date: October 2012 * DOI:

https://doi.org/10.1038/nm1012-1452 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently

available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative