In our previous (July 12) blog post, we began a discussion of the creation of the “synthetic cell” by J. Craig Venter and his colleague at the J. Craig Venter Institute (JCVI).  In this study, the researchers produced a synthetic version of the genome of bacterium Mycoplasma mycoides. They then transferred the synthetic M. mycoides genome into the closely related bacterium M. capricolum, where then new genome took over the cells, resulting in bacteria that expressed the proteome of M. mycoides. The resulting cells were dubbed “synthetic cells”.

We said that we agreed with scientists (including working synthetic biologists) who said that the “synthetic cell” project did very little to advance synthetic biology, beyond the demonstration of technical virtuosity. Here’s why we agree.

JCVI’s “synthetic genome” is merely a copy of the natural M. mycoides genome, with a few minor differences. Copying the natural genome does not advance our understanding of how the various genes of M. mycoides function. Even in the case of such a simple organism as M. mycoides, only about half the genes in the genome are understood in terms of their biological function.

Copying the natural genome of M. mycoides also does not tell us how the genes of this organism work together in biological processes. Researchers find it difficult to construct even simple synthetic biology devices, because it is difficult to get artificially-introduced genes to work together as planned even in prokaryotic systems. For example, in 2000 researchers constructed a simple biological clock based on a three-gene “repressilator network” in E. coli. The clock was designed so that the cells flash on and off, based on oscillating expression of a fluorescent protein. However, the operation of this simple clock was “noisy”; i.e., the period of these oscillations was highly variable. However, natural biological clocks are not noisy. Other researchers therefore designed a more complex synthetic clock to damp down the noise of the original clock. This example illustrates how the complexity of biology can affect the ability of researchers to construct even simple biological devices. However, constructing such devices and then working to improve them can help researchers gain a greater understanding of how genes work together in natural networks and processes.

Every year since 2004, there has been an international undergraduate synthetic biology competition, run out of MIT, called the International Genetically Engineered Machine (iGEM) competition. In this competition, students use “biological parts” from a “kit”, and/or develop new “parts”, to design a synthetic biology device. Many teams that participate in this competition find out that designing new synthetic biology systems is not a simple case of putting together “biological parts” like Lego blocks or components of an electrical circuit. Not all the “parts” deposited in MIT’s Registry of Standard Biological Parts (including the ones designed by iGEM competition participants themselves) are well characterized, and even if they are, the “parts” may not work together with each other, or with the organism that contains them, as planned.

A January 2010 news feature in Nature, entitled “Five hard truths for synthetic biology”, details the many challenges to the field of synthetic biology. The construction of the “synthetic cell” does little or nothing to help synthetic biologists to meet these challenges.

Our main interest in synthetic biology is in metabolic engineering for use in drug discovery and in production of natural product drugs that are difficult to obtain cheaply from their natural plant sources and are difficult to synthesize chemically. We have two articles on synthetic biology on this blog pervious to the July 12, 2010 article–one dated July 28, 2009 and the other dated September 7, 2009 .

We are much more impressed by some of the work discussed in the September 7, 2009 article than by the creation of the “synthetic cell”. These include, for example, the metabolic engineering studies of Jay Keasling, such as his engineering of yeast and E coli to produce precursors of the antimalarial drug artemisinin. (For example, see this 2006 Nature article.) Dr. Keasling and his colleagues had to contend with several genes that did not work together as envisioned, and to find ways to optimize their expression so that they would work together to produce significant quantities of the desired product. Dr. Keasling says that it took approximately 150 person-years to complete this project. The pharmaceutical company Sanofi-Aventis has been scaling up Dr. Keasling’s artemisinin yeast production system, and plans to sell its product cheaper than artemisinin produced by current methods (i.e., from its natural plant source) by 2012.

We are also impressed by George Church’s automated method for optimizing metabolic engineering pathways, which was the focus of our September 7, 2009 article. This might make the work of metabolic engineers, epitomized by the Keasling groups’s work on artemisinin, faster, easier, and less expensive.

Finally, engineered mycoplasma are unsuitable for making medicines. They are too fragile, and their genomes are too small. For making medicines, one needs a “workhorse” organism that can take the extensive genetic manipulation needed to engineer new metabolic pathways, without losing viability and while maintaining a reasonable growth rate. And cultures of engineered microorganisms must be capable of being scaled up to industrial levels. Metabolic engineers have used such “workhorse” organism as E. coli, Saccharomyces cerevisiae, Streptomyces lividans, and Myxococcus xanthus.

Dr. Venter, through his company Synthetic Genomics, Inc. (SGI, La Jolla, CA) intends to work on synthetic genomics solutions for the production of biofuels. SGI has an alliance with the Exxon Mobil Research and Engineering (EMRE) group to generate biofuels from algae. Dr. Venter claims that SGI will make use of the tools and technologies from the synthetic genome research (together with more traditional metabolic engineering methods) to build “an entire algae genome so we can vary the 50 to 60 different parameters for algae growth to make superproductive organisms.” An alga that would be capable of being scaled up to produce commercial amounts of biofuels would have to be a real “workhorse” organism. And engineering of such algae would be an impressive achievement, beyond the level of a mere technical tour de force as with the “synthetic” mycoplasma cell. We (along with the rest of the world) await further progress on this project.

In late May of 2010–as most of you probably know–J. Craig Venter announced that he and his colleague at the J. Craig Venter Institute (JCVI) (Rockville, MD) had created a “synthetic cell”. This was a bacterial cell containing an entire genome that had been synthesized based on a nucleotide sequence stored on a computer, which had then been inserted into the cytoplasm of another bacterial cell. The JCVI paper was first published online in Science on May 20. It has now appeared in print in the July 2, 2010 issue of Science, where it made the cover of the journal.

Commentaries on the JCVI paper were published in Nature (here and here) and in the New York Times, among other places.

To summarize the JCVI report: the researchers assembled sets of chemically synthesized oligonucleotides, in stages, into a 1.08 million base pair DNA molecule with the slightly modified sequence of the genome of the bacterium Mycoplasma mycoides. Among the slight modifications were “watermarks”, i.e., four added sequences in nonessential genome regions that identify the researchers, and enable researchers to differentiate synthetic genomes from natural ones. The researchers then transferred the synthetic M. mycoides genome into the closely related bacterium M. capricolum, where then new genome took over the cells, resulting in bacteria that expressed the proteome of M. mycoides. The resulting cells were dubbed “synthetic cells”.

The work of the JCVI researchers was a technical tour de force. It required the accurate sequencing of the M. mycoides genome, as well as advanced technologies for accurate chemical synthesis of oligonucleotides, oligonucleotide extraction and assembly, and transplantation into recipient bacterial cells. There were many stumbling blocks in this process. The final stumbling block was a single base pair deletion in an essential gene for chromosomal replication. Once they discovered and corrected this error, the researchers were able to successfully transplant the synthetic genome and get it to commandeer the recipient M. capricolum cells, resulting in expression only of M. mycoides proteins as directed by the synthetic genome.

In evaluating the importance of the “synthetic cell” beyond this technical virtuosity, there are two perspectives–philosophical and scientific/technological. From the philosophical point of view, some commentators hailed this work as the final refutation of vitalism, i.e., the contention that there is something special about processes in living organisms that cannot be artificially created from nonliving systems. Interestingly, the commentators who expressed this conclusion are philosophers and bioethicists. (See, for example, the statements of philosopher Mark Bedau and bioethicist Arthur Caplan in the 27 May Nature discussion article.)

However, as one can see from the same discussion article, scientists–including leading working synthetic biologists–know better. The old biologists’ dictum, “all life comes from life”, espoused by Louis Pasteur among others, still holds. Not only were living M. capricolum cells required for the creation of “synthetic” M. mycoides cells, but the researchers utilized the yeast homologous recombination system in vivo to assemble their “synthetic” genome.

From the scientific and technological perspective, what has the “synthetic cell” project done to advance synthetic biology, beyond the demonstration of technical virtuosity?  The answer of most leading scientific commentators, including working synthetic biologists, is “not very much”.

We agree. We’ll discuss why in Part 2 of this article.