Designs for life

[More matters arising from the Greenland conference: in this case, a paper that John Glass of the Venter Institute discussed, and which is now published in Science. It has had a lot of press, and rightly so. Here is the article I have written for Nature's News & Views section, which will appear in next week's issue.]

The genome of one bacterium has been successfully replaced with that of a different bacterium, transforming one species into another. This development is a harbinger of whole-genome engineering for practical ends.

If your computer doesn’t do the things you want, give it a new operating system. As they describe in Science [1], Carole Lartigue and colleagues at the J. Craig Venter Institute in Rockville, Maryland, have now demonstrated that the same idea will work for living cells. In an innovation that presages the dawn of organisms redesigned from scratch, the authors report the transplantation of an entire genome between species. They have moved the genome from one bacterium, Mycoplasma mycoides, to another, Mycoplasma capricolum, and have shown that the recipient cells can be ‘booted up’ with the new genome — in effect, a transplant that converts one species into another.

This is likely to be a curtain-raiser for the replacement of an organism’s genome with a wholly synthetic one, made by DNA-synthesis technology. The team at the Venter Institute hopes to identify the ‘minimal’ Mycoplasma genome: the smallest subset of genes that will sustain a viable organism [2]. The group currently has a patent application for a minimal bacterial genome of 381 genes identified in Mycoplasma genitalium, the remainder of the organism’s 485 protein-coding genes having been culled as non-essential.

This stripped-down genome would provide a ‘chassis’ on which organisms with new functions might be designed by combining it with genes from other organisms — for example, those encoding cellulase and hydrogenase enzymes, for making cells that respectively break down plant matter and generate hydrogen. Mycoplasma genitalium is a candidate platform for this kind of designer-genome synthetic biology because of its exceptionally small genome [2]. But it has drawbacks, particularly a relatively slow growth rate and a requirement for complex growth media: it is a parasite of the primate genital tract, and is not naturally ‘competent’ on its own. Moreover, its genetic proof-reading mechanisms are sloppy, giving it a rapid rate of mutation and evolution. The goat pathogens M. mycoides and M. capricolum are somewhat faster-growing, dividing in less than two hours.

Incorporation of foreign DNA into cells happens naturally, for example when viruses transfer DNA between bacteria. And in biotechnology, artificial plasmids (circular strands of DNA) a few kilobases in size are routinely transferred into microorganisms using techniques such as electroporation to get them across cell walls. In these cases, the plasmids and host-cell chromosomes coexist and replicate independently. It has remained unclear to what extent transfected DNA can cause a genuine phenotypic change in the host cells — that is, a full transformation in a species’ characteristics. Two years ago, Itaya et al. [3] transferred almost an entire genome of the photosynthetic bacterium Synechocystis PCC6803 into the bacterium Bacillus subtilis. But most of the added genes were silent and the cells remained phenotypically unaltered.

Genome transplantation in Mycoplasma is relatively easy because these organisms lack a bacterial cell wall, having only a lipid bilayer membrane. Lartigue et al. extracted the genome of M. mycoides by suspending the bacterial cells in agarose gel before breaking them open, then digesting the proteinaceous material with proteinase enzymes. This process leaves circular chromosomes, virtually devoid of protein and protected from shear stress by the agarose encasement. This genetic material was transferred to M. capricolum cells in the presence of polyethylene glycol, a compound known to cause fusion of eukaryotic cells (those with genomes contained in a separate organelle, the nucleus). Lartigue et al. speculate that some M. capricolum cells may have fused around the naked M. mycoides genomes.

The researchers did not need to remove the recipient’s DNA before adding that of the donor; instead, they added an antibiotic-resistance gene to the M. mycoides donor genome. With two genomes already present, no replication was needed before the recipient cells could divide: one daughter cell had the DNA of M. capricolum, the other that of M. mycoides. But in the presence of the antibiotic, only the latter survived. Some M. capricolum colonies did develop in the transplanted cells after about ten days, perhaps because their genomes recombined with the antibiotic-resistant M. mycoides. But most of the cells, and all of those that formed in the first few days, seemed to be both genotypically and phenotypically M. mycoides, as assessed by means of specific antibodies and proteomic analysis.

The main question raised by this achievement is how much difference a transplant will tolerate. That is, how much reprogramming is possible? The DNA sequences of M. mycoides and M. capricolum are only about 76% the same, and so it was by no means obvious that the molecular machinery of one would be able to operate on the genome of the other. Yet synthetic biology seems likely to make possible many new cell functions, not by whole-genome transplants but by fusing existing ones. When John I. Glass, a member of the Venter Institute’s team, presented the transplant results at a recent symposium on the merging of synthetic biology and nanotechnology [4], he also described the institute’s work on genome fusion (further comments on matters arising from the symposium appeared in last week’s issue of Nature [5].

One target is to develop a species of an aerobic Clostridium bacterium that will digest plant cellulose into ethanol, thus generating a fuel from biomass. Cellulose is difficult to break down — which is why trees remain standing for so long — but it can be done by Clostridium cellulolyticum. However, this creates glucose. Clostridium acetobutylicum, meanwhile, makes butanol and other alcohols, but not from cellulose. So a combination of genes from both organisms might do the trick. For such applications, it remains to be seen whether custom-built vehicles or hybrids will win the race.

1. Lartigue, C. et al. Science Express doi:10.1126.1144622 (2007).
2. Fraser, C. M. et al. Science 270, 397–403 (1995).
3. Itaya, M. et al. Proc. Natl Acad. Sci. USA 102, 15971–15976 (2005).
4. Kavli Futures Symposium The Merging of Bio and Nano: Towards Cyborg Cells 11–15 June 2007, Ilulissat, Greenland.
5. Editorial Nature 447, 1031–1032 (2007).