Articles

By Jennifer Van Brunt, Editor

Synthetic biology is a new discipline that brings together engineering, biology, chemistry, and information technology to take genetics and genomics to the next level. Cloning and genetic engineering have been used for decades now to study the molecular mechanisms of basic cellular processes as well as create transgenic organisms that make therapeutic proteins (think Epogen and Enbrel). But these efforts have always been hit or miss, depending on the many variables inherent in biological systems. Now, fledgling efforts in synthetic biology intend to replace random events with predictable ones – by creating a catalog of synthetic parts (genes, operons, chromosomes, for instance) that can be used in protein production, metabolic engineering, even environmentally friendly fuel production. It's building organisms for specific purposes, rather than modifying existing ones to fit the need.

Synthetic biology: It's a sexy phrase, all right, but just exactly what does it mean? Is synthetic biology a new branch of science dedicated to creating life de novo? Is it about designing organisms to carry specific traits? Is it just genetic engineering parading under a new banner? Well, the answer depends on whom you ask.

In general, though, the relatively new branch of research brings together science and engineering in an attempt to design new biological functions and systems. The idea is to create a catalog of parts – genes, operons, whole chromosomes, on/off switches and the like – that will allow researchers to pick and choose which to study. While so-called traditional genetic engineering is a haphazard affair, synthetic biology promises to impose a sense of order to the entire process. The mysterious workings of living systems will at long last become known. Plus, the potential applications of this new science are virtually endless, and range from drug design to hydrogen production for fuel.

Sound too far-fetched? Well, it's just this sort of sweeping vision that fires the imagination of at least some venture capitalists, and in 2005 not one, but three synthetic biology start-ups attracted venture funding.

Starting from scratch
The biggest chunk of change went to Synthetic Genomics Inc.: In late June, the Rockville, MD firm raised $30 million from VC Draper Fisher Jurvetson and two other investors. This new outfit is the latest brainchild of J. Craig Venter, of human genome fame – and his company's immediate goal is to create new bacteria that will churn out ethanol or hydrogen as alternate sources of fuel.

How will this occur? The idea is to harness the vast amounts of genomic information now available -- more than 300 genomes have been sequenced -- to create microbes with new metabolic pathways, a sort of mix and match approach. Researchers will devise new systems with specific functions that can then be used as "parts" that can be inserted into cells. The cells, of course, will also have to be turned into tiny factories to make this plan work, but a number of very good hosts are already available.

Not surprisingly, Synthetic Genomics will collaborate with the J. Craig Venter Institute, where researchers have already synthesized the genome of the bacteriophage Phi-X174, a DNA virus that has a single stranded circular genome. The genome is small – only 5,386 base pairs – but the fact that it's been synthesized is an important first step in creating more complicated life forms.

This proof of principal was enough to convince Venter and his colleagues, including Nobel laureate Hamilton Smith, to form the new company – and it may also have been sufficient to kindle the interest of some energy companies, with which Synthetic Genomics is reported to be in talks.

Stellar line up
Venter's company is now working on developing what it calls a minimal genome – one containing the least number of genes to allow it to survive in a specific environment. Then, researchers will add the specific genetic information of interest (say, a photosynthetic pathway) and insert the whole thing into a so-called synthetic cell, which replicates and metabolizes in a particular manner.

While Venter's approach may be unique, his young firm is not the only one intent on designing and synthesizing biological parts – from genes to entire genomes. But for Codon Devices, the approach takes a slightly different turn. Whereas Synthetic Genomics is starting with a minimal genome, upon which it can build, Codon Devices is focused on synthesizing very long pieces of specified DNA – kilobase to megabase length -- more rapidly and at a lower cost than what's available today. These various pieces of DNA, in turn, can be used to form a library of engineered parts, explained John Danner, the company's new president and CEO.

Providing such services is the short-term value proposition for the firm, Danner said, and includes applying the technology to protein engineering, metabolic engineering, cell factories, protein production and the like. Longer-term, the technology could be used to improve vaccines and agricultural products as well as create biorefineries, among other uses.

Once again, it's the availability of vast amounts of sequence data – not only from the human genome but also from the genomes of many other organisms – that underlies the concept of biological parts. When once scientists were hard at work deciphering the genetic code, now they can begin to write it. But before they can do that, they need better methods for synthesizing the lengthy stretches of DNA that comprise a genome – not hundreds of bases, but rather millions or billions.

Codon Devices, of Cambridge, MA, claims to be the very first venture backed synthetic biology company: At the beginning of June 2005 it raised $13 million from Flagship Ventures, Alloy Ventures, Kleiner Perkins Caufield & Byers and Vinod Khosla (founding CEO of Sun Microsystems). That's an impressive line up of VCs, but it's matched by the pedigrees of the academic founders, who are among the Who's Who of synthetic biology practitioners. These include George Church of Harvard, Drew Endy and Joseph Jacobson of MIT, and Jay Keasling of the University of California at Berkeley.

Parts catalog
According to Samir Kaul, a principal at founding investor Flagship Ventures and Codon Devices' founding CEO, "synthetic biology is taking true engineering and applying it in a biological setting." And, he added, with a "catalog of biological parts" available, it's possible to "pick and choose." This catalog could include genes, proteins, regulatory elements, and other parts, he said. A researcher can test a model, refine it, and test it again. This whole process should take days instead of months, Kaul added.

But while Codon Devices is now offering its services to researchers, current president and CEO Danner emphasized that the company's business "Is not to do one or two constructs. We will typically have large customers or high volume customers." For instance, Jay Keasling, director of the Berkeley Center for Synthetic Biology at UC Berkeley – in collaboration with his company Amyris Biotechnologies Inc. and the Institute for One World Health -- is engineering cells to churn out a precursor to the malaria drug artemisinin, and the project "needs a tremendous amount of synthesized material," Danner said.

Moreover, since Keasling has put together genes from different organisms – a plant, a yeast and a bacterium – there's a hitch in synthesizing the entire length of DNA. This stems from the fact that the DNA code is degenerate: There are multiple codons coding for the same amino acid. And different organisms make different choices of which codon to use, so when the gene of one organism is put into another, expression levels may be affected. "Codon usage in plants is quite different from E. coli," Kaul said. But Codon Devices' technology platform allows the synthesis of codon-optimized DNA, so the process now become "trivial," Kaul said. "Our technology platform doesn't care what it's doing," whereas cloning is biased by what's available in nature, he added.

Engineering a pathway
In fact, Keasling's work on artemisinin is a perfect example of how synthetic biology is supposed to work. And it doesn't hurt that the malaria project is funded by a generous $42.5 million grant from The Bill & Melinda Gates Foundation. The importance of the program was also highlighted last month by the World Economic Forum, which honored Amyris Biotechnologies as one of 36 technology pioneers for 2006.

Artemisinin has already proved itself to be a powerful antimalarial drug, and it works against all parasitic strains so far. That very property has rung the alarms of the World Health Organization, though, which is fearful that resistance will soon result if the drug is used as a monotherapy instead of the recommended combination regimen with older medicines. And, apparently, many drug makers are selling artemisinin as a monotherapy – a practice WHO wants to stop. Resistance may be an especially thorny problem because the next generation of malaria drugs is based on artemisinin, too.

Adding to the problem, artemisinin itself is relatively scarce, meaning that its cost is too steep for purchase by the very countries that are most affected by the disease. Currently, artemisinin is extracted from the dry leaves of the sweet wormwood tree, which grows in many locales, but only produces the drug when grown in China and Vietnam.

However, Keasling's group has engineered a new metabolic pathway in E. coli that allows the rapid and inexpensive production of amorphadiene, a precursor to the drug. They've accomplished this by introducing genes from the sweet wormwood tree (Artemisia annua) and Saccharomyces cerevisae into the bacteria. The next step is to clone the remaining genes that will complete the entire synthetic pathway.

The malaria drug is a great start, no question. But there's much more in the offing: Artemisinin is a terpinoid, a member of the large class of chemicals called isoprenoids – which already constitute drugs like Taxol and are ripe for further exploitation in the type of bacterial system Keasling's group is now designing. And even with the isoprenoid precursor, it's now at least theoretically possible to make a wide variety of final products by use of the appropriate synthase gene.

Keasling's group is not the first to graft more than one foreign gene into E. coli, though – Amgen Inc. researchers did it ages ago when they engineered microbes to make indigo, the dye used for blue jeans. But the yields were low, according to Codon Devices' Kaul, meaning the process wasn't cost-effective. The technology developed by Keasling and associates is intended to solve that problem.

Genes to order
Still, companies like Amyris Biotechnologies and Synthetic Genomics are relatively rare in the synthetic biology arena. More common are those firms – like Codon Devices – that are creating faster, cheaper ways to make long strands of DNA or particular genes to specifications. They include, among others, Bothell, WA-based Blue Heron Biotechnology Inc.; DNA 2.0 of Menlo Park, CA; and Geneart GmbH of Regensburg, Germany.

Each has its own approach, and they all differ from that taken by CODA Genomics Inc., which is located in Irvine, CA. CODA Genomics was the third synthetic biology company to receive venture funding in 2005: In October, it raised about $0.8 million in Series A funding from the Tech Coast Angels, which claims to be the largest U.S.-based angel investor network.

The fledgling firm was started in 2004 to commercialize DNA assembly and protein expression work, developed under an Information Technology Research award from the National Science Foundation, from the UC Irvine labs of Richard Lathrop and G. Wesley Hatfield. In essence, the technology, computationally optimized DNA assembly (CODA), allows large-scale production of synthetic genes that are optimized to make only one correct protein product.

Again, we face the problem of codon usage or distribution: Different organisms have different preferences, so when one creates a transgenic system for making proteins, expression levels may be low or the resulting protein product, which is actually a mixture of different forms, may lack solubility or some other function as a result. According to CODA Genomics, optimizing codon usage is really insufficient to overcome this protein expression problem, so the company goes one step further to control the translation kinetics, i.e., regulating the speed of protein elongation during translation. The end result, it says, is a thermodynamically favored, perfectly correct protein.

"Initially, [biotechnology] created great protein drugs," explained CEO Robert Molinari, but they were "developed by chance. Researchers had to tweak the cloned gene to produce something active in humans. This was the exception, not the rule," he added. No wonder: According to Molinari, because of codon usage (in which there is variability [a.k.a. wobble] in the third base that codes for a particular amino acid) there is "a huge probability of variations to make a particular protein," even one that is composed of 500 amino acids (1,500 base pairs).

"Now we design a gene to be expressed in a particular organism and we have to take into account evolutionary preferences," Molinari said. "But even when we do that, we can fail to get the correct protein." Synthetic biology changes all that by eliminating the variability. "The only way to make changes is to synthesize the gene from scratch," he said. Then it's possible to make genes that are optimized for heterologous protein production. CODA Genomics employs "computational assembly mechanisms that get rid of all the errors along the way."

New Vision
"Synthetic biology is a new vision for industry," explained Molinari. And the term encompasses several disciplines. On the one hand, "Codon Devices, Jay Keasling and that whole group are engineering organisms and components not unlike integrated circuit components that can slide in and out of biological systems," he said. They are trying to create the equivalent of standardized I/O circuits.

But there is a "whole other branch," Molinari said. This is the synthetic gene business, which has been up and running for several years already and is intended to lower the price of synthetic DNA. It's got another goal, too, and that's to eliminate the random aspect of transgenic protein production.

As well, he said, it will be possible to synthesize genes to make therapeutic proteins without deleterious side effects. Interleukin-2, for instance, "has very serious side effects. To fix them, we will start changing the amino acids. The only efficient way to do that is to synthesize the gene."

To make any comments on this article, or to ask a question of the author, please contact the publisher. If you would like to submit an article please subscribe to our Pharmalicensing Intelligence service.

The opinions expressed in the articles published in this section do not necessarily reflect those of Innovaro Pharmalicensing or Innovaro Corporation. No actions including proposals to or agreements with other companies should be taken by any reader without obtaining specific business or legal advice. Neither the publisher nor the authors accept any liability for any actions or activities undertaken by any reader or other third party as a consequence of these articles or for any errors or omissions therein.