Genesis redux

IN THE end there was no castle, no thunderstorm and definitely no hunchbacked cackling lab assistant. Nevertheless, Craig Venter, Hamilton Smith and their colleagues have done for real what Mary Shelley merely imagined. On May 20th, in the pages of Science, they announced that they had created a living creature.

Like Shelley's protagonist, Dr Venter and Dr Smith needed some spare parts from dead bodies to make their creature work. Unlike Victor Frankenstein, though, they needed no extra spark of Promethean lightning to give the creature its living essence. Instead they made that essence, a piece of DNA that carries about 1,000 genes, from off-the-shelf laboratory chemicals. The result is the first creature since the beginning of creatures that has no ancestor. What it is, and how it lives, depends entirely on a design put together by scientists of the J. Craig Venter Institute and held on the institute's computers in Rockville, Maryland, and San Diego, California. When the first of these artificial creatures showed that it could reproduce on its own, the age of artificial life began.

The announcement is momentous. It is not unexpected. Dr Venter's ambition to create a living organism from close to scratch began 15 years ago, and it has been public knowledge for a decade. After so much time, there is a temptation for those in the field to say “show us something we didn't know.” Synthetic DNA is, after all, routinely incorporated into living things by academics, by biotech companies, even by schoolchildren. Dr Venter—a consummate showman—and the self-effacing Dr Smith (uncharacteristically in the foreground in the picture of the two above) have merely done it on a grand scale.

Craig's parts list

But if it is a stunt, it is a well conceived one. It demonstrates more forcefully than anything else to date that life's essence is information. Heretofore that information has been passed from one living thing to another. Now it does not have to be. Non-living matter can be brought to life with no need for lightning, a vital essence or a god. And this new power will allow the large-scale manipulation of living organisms. Hitherto, genetic modification has been the work of apprentices and journeymen. This new step is, in the true and original sense of the word, a masterpiece. It is the demonstration that the practitioner has mastered his art.

The journey to mastery has been a long one. Originally, not wishing to set himself a more difficult task than necessary, Dr Venter found the smallest living thing he could and set about making it smaller still. His chosen bug was Mycoplasma genitalium, a creature that lives in genital tracts. With just 485 genes, it is the tiniest known free-living bacterium. He then knocked out the bacterium's genes one by one to see which it could live without, in the hope of making a yet smaller organism he could then use as a model for synthesis.

This was something of a dead-end. Though there turned out to be 100 genes M. genitalium can do without, at least in the cushy conditions of a laboratory, it could not do without all of them at once. And finding which smaller genomes worked best took a lot of time, because M. genitalium grows rather slowly.

On top of that, the reason for wanting a very small genome started to fade away. DNA synthesis techniques were getting better and better, a fact reflected in their ever decreasing price (see chart). So Dr Venter changed tack, and decided to go with a lightly modified version of the entire M. genitalium genome.

Around the same time, in 2003, he synthesised the genome of a virus, Phi-X174, which has a mere 11 genes. It was not the first artificial virus; a team at the State University of New York, in Stony Brook, had made a copy of the polio virus the previous year. But theirs was a feeble thing, only just capable of reproducing. Dr Venter's was the real McCoy: when he put the viral DNA into host cells they started to spit out new viruses just as self-destructively as cells infected with the natural Phi-X174.

The idea behind the efforts to make an artificial bacterium was, in essence, to treat a large synthetic genome as a giant version of Phi-X174 and use it to hijack a cell which had had all its DNA removed. The difference was that this time the result would not be a cell that produced more viruses, but a cell that produced more cells. By the time the hijacked cell had undergone a few divisions, all trace of its previous self would have been erased; its several-times-great-granddaughters would have transformed themselves into the new species.

Synthesising the genome proved reasonably easy. It was divided in “cassettes” about 1,000 base pairs long (a base pair being one of the genetic “letters” of which DNA is composed). These were put together by normal chemistry. The team then enlisted the help of yeast cells to link the cassettes in the correct order to produce the finished genomes.

At this point it was necessary to prepare the cadavers, which proved rather trickier. It wasn't just a matter of taking a bacterium closely related to M. genitalium and scooping out its DNA. Bacteria have defences against viruses in the form of chemicals called restriction enzymes, which chop up foreign DNA. These enzymes (discovered in the 1970s by Dr Smith, in work that won him a Nobel prize) would lurk in the DNA-free cadaver and cut up the synthetic genome before it was able to do its stuff. So the last step on the winding road was the creation of a bacterial strain without any restriction-enzyme genes, and thus without restriction enzymes, so that the team could have a purified reaction vessel in which the new genome could do its thing.

Or almost the last step. M. genitalium still had a slow-growth problem, so the team swapped bugs, lighting on its cousin, Mycoplasma mycoides. This has twice as much DNA, but that no longer mattered. To make the new bacterium recognisably different Dr Venter and his colleagues deleted 14 genes they thought unnecessary from M. mycoides, and added some DNA designed from scratch in a process Dr Venter refers to as “watermarking”.

This was an opportunity for some fun. The watermark, Dr Venter says, includes a cipher which contains the URL of a website and three quotations, if you can work out how to decode it. The plaintext part of the watermark brands the bug as Dr Venter's own, encoding its serial number as JCVI-syn1.0. (A plan to refer to the result as Mycoplasma laboratorium and have it recognised as a completely new species seems to have been abandoned for the moment.)

The watermarking is not just a fancy signature. It means that if, despite precautions, the Frankenbug does get out, its entirely harmless presence would be detectible in any given sample by straightforward DNA amplification technology of the sort used in genetic fingerprinting. It might also trap thieves. Dr Venter has offered his invention for patenting—an action that is sure to be controversial—and the watermark will thus stake out what he hopes will become the property of his firm, Synthetic Genomics.

Once the finished genome was inserted into the genome-free bacteria, the work regressed to the sort of microbiology that would have been familiar to the science's 19th-century pioneers. The fluid with the bacteria in it was dotted on to agar plates. Spots showed up on the agar as individual bacteria grew and multiplied. As a check, the researchers sequenced the DNA from some of the flourishing spots (a Mycoplasma genome is the sort of thing a modern sequencing set-up can knock off before its morning coffee). The colonies did, indeed, have the synthetic genomes. The masterpiece was alive.

Radicalism and ribosomes

Other journeymen, though, are hot on Dr Venter's heels. And some have different ideas on how to go about the problem of making life, concentrating on things which Dr Venter's hack-a-cadaver approach allows him to gloss over.

A minimal genome is one thing. At Harvard Medical School, Jack Szostak is working on a minimal cell, the components of which might be quite unlike those of any modern life form. Dr Szostak is interested in the origin of life, and wants to develop something analogous to what he imagines life's earliest days were like: a reaction vessel in which a self-sustaining cycle of chemical reactions can reproduce itself.

In a modern cell, such as a bacterium, instructions from the DNA are transcribed into a related molecule called RNA. The RNA messenger molecules relay them to structures known as ribosomes that read them and make proteins accordingly. The whole process also involves a lot of proteins called enzymes to act as catalysts to the reactions.

Meet the new bug: JCVI-syn1.0

Many biologists—and Dr Szostak is one of them—think that life had a simpler early stage in which the varied tasks now carried out by DNA, RNA and proteins were all achieved by RNA alone. Even today, RNA molecules are not only messengers; they are also fetchers and carriers of amino acids, the building blocks of proteins. And they can catalyse reactions, as proteins do, too. In principle, then, RNA could act as both a cell's genetic material and its self-assembly mechanism.

If this idea is true, it should be possible to make a cell using just a membrane to hold things in place, some RNA, ingredients for more RNA, and an energy source. This comes in the form of an energy-rich molecule, ATP, which is what modern cells use to move energy from where it is generated to where it is used. Dr Szostak has already made a range of “ribozymes”, as catalytic pieces of RNA are known in the trade, and some of them are ATP-powered. He does not, yet, have a system that is capable of replicating itself. But that is his goal.

Dr Szostak's cell, if it does come to pass, will be quite different from the protein- and DNA-based life familiar to biologists. It would in some ways be a greater achievement than Dr Venter's, in that it would create something truly from scratch; but it would be of less practical importance, since that something would be very primitive compared even with a bacterium.

George Church, a colleague of Dr Szostak's at Harvard, dreams instead of making something intensely practical that Dr Venter has left out: a ribosome. The Venter shortcut—booting up a bacterial cadaver—means that the new-minted bug has to rely on ribosomes from its dead host to make the proteins its genome describes. It has the genes with which to make its own ribosomes, though, and as time goes by it will do so, diluting out the legacy that got it started. Dr Venter calculates that once JCVI-syn1.0 has undergone 30 divisions, all trace of the original cell will have disappeared. But that does not address the point that the new cells have relied on the output of genes from the old one to get going in the first place.

Dr Church is working on making ribosomes—complex contraptions with dozens of protein and RNA components—from scratch. He has managed to synthesise all the RNA components in such a way that, when they are mixed with natural ribosome proteins, they form working ribosomes. Making the proteins from scratch is more difficult, because their shape is crucial to their function, so it is not clear whether he will bother to do so.

Although he is interested in chalking up firsts, Dr Church focuses mainly on making tools. Artificial ribosomes, he thinks, could be specially crafted to add new capabilities to biotechnology—higher-than-natural protein productivity, for example. And that, for all the brouhaha which rightly accompanies the passage from journeyman to master, is the ultimate point: practical control over what life can be made to do.

Another avowedly practical approach is that taken by Drew Endy, a researcher at Stanford University. Dr Endy wants to make the way that cells process genetic information more like the way that familiar computers do. Just as computers are built from electronic components that (at least in the days before integrated circuits and silicon chips) could be ordered from a catalogue by engineers and enthusiasts alike, so Dr Endy is trying to build up a catalogue of components he calls biobricks that, when linked together, will form useful biological “circuits”. Synthetic biologists will be able to order stretches of DNA that encode biobricks and link them together to do their bidding.

Dr Endy's approach is intriguing. His plan to “reimplement” life shows an engineer's desire to replace biology's unruly heritage—kludge built on kludge for billions of years—with something designed to be fit for a physicist's practical purpose. Whether it will work remains to be seen. But a less thoroughgoing approach to modular design underlies the next stage of Dr Venter's plans, too.

The constant gardener

Biotechnology can sometimes resemble that rather older interaction with nature, gardening. It relies quite heavily on pruning and grafting. Gene-by-gene biotechnology constantly comes up against the problem that living organisms like to plough their own furrow, regardless of what their human “masters” might desire. The pruning part of biotechnology involves eliminating proclivities that might be useful to a wild organism, but drain its energy and metabolic effort away from the task at hand. The grafting part is adding new characteristics from elsewhere to the well-trained root stock.

Dr Venter wants to get back to his original idea of creating a minimal genome in a peculiarly complete and rational act of pruning in order to be able to do a much more thorough job than has been previously possible of grafting in new stock. It is this ambition that makes his work something more than just a breathtaking novelty, positioning it as a milestone on the road from the craft of biotechnology, which manipulates genes one at a time, to the industry of synthetic biology, which aims to make wholesale changes to living things.

In this, Dr Venter seems to be going with the grain of nature, as wise gardeners do. Over the past decade it has become clear that bacteria are already well disposed to the idea of interchangeable parts. Each member of a bacterial species, or group of species, has a subset of genes (numbering hundreds, or a few thousand) drawn from a pool containing many thousands. Comparing lots of different but related bacteria can thus reveal a “core competence” similar in concept to a minimal genome. In seeking to build useful bacteria (ones that can, say, produce particular drugs in quantity) Dr Venter's thoroughgoing root-and-graft approach may be tidying up a strategy that has been used for 4 billion years, perhaps even returning it to its basics.

Bring on the empty algae

He does not plan to stick to bacteria, though. The other challenge, besides the minimal genome, is to repeat the trick with single-celled algae.

The step from single-celled bacteria to single-celled algae may sound like a short one. But algae are on the other side of the great dividing line of life, that between creatures with a simple, single genome which is just a big loop of DNA sitting in the cell and those with genomes that are for the most part sequestered in a nucleus set aside for them, and cut up into multiple chromosomes. This second group includes animals, plants, fungi and algae. With no disrespect towards bacteria, which are remarkably innovative and spectacularly durable, the creatures that have taken the nuclear route are much more interesting—not least because Homo sapiens is himself one of them.

Algae, though, are interesting for other reasons. Many people—including Dr Venter—want to use them to produce biofuels. They would turn carbon dioxide from the atmosphere (or, better, from power-station exhaust) into petrol or diesel by photosynthesis. At the moment, the microbes which make biofuels almost all do so through fermentation. The photosynthesis is done by plants such as sugar cane and the sugar is transformed into fuel by engineered bugs of one sort or another. Using algae would cut out the middleman.

The life to come

All of this activity, however, relies on one thing: that the price of synthesising DNA continues to fall. In a way analogous to Gordon Moore's famous law about the improvement of computers, both the price of sequencing DNA and the price of making it have plummeted over the past decade. The former means that the world's databases are filling up with genes from every part of the tree of life. The latter means those genes can be cut and pasted together with greater and greater ease.

If synthetic biology is to take off as a technology, that is not merely good, it is essential. There will be a lot of trial and error in the process of creating new, useful organisms. Evolution by artificial selection is likely to prove almost as wasteful as the kind by natural selection. But there are those that worry about the proliferation of gene synthesis. Noting the propensity of computer-hackers to turn out what have been dubbed, by analogy, software viruses, they worry that hackers of the future may turn to synthetic biology and turn out real viruses.

It is a risk, no doubt. But almost all technologies can be used for ill as well as good. Approaches that can create pathogens to order can create vaccines, too—and it is not too rose-tinted to think that the will to do good, often harnessed to the desire to make money, will attract many more people than the dark side will. They could create new crops, new fuels, new ways of investigating diseases and new drugs to treat them. They might do other, wilder things as well.

A more recent piece of science fiction than Shelley's, Michael Crichton's “Jurassic Park”, conceived of the resurrection of dinosaurs. No DNA survives that would allow that to be done directly. But the ability to make genomes, coupled to a far greater understanding of how they lead to the structures of complex organisms, could one day allow simulacra of such creatures to be made by synthetic biology.

In any case, though dinosaurs have left no usable DNA, other more recently departed creatures have been more generous. Imagine, say, allying synthetic biology with the genome of Neanderthal man that was described earlier this year. There is much excitement at the idea of comparing this with the DNA of modern humans, in the hope of finding the essential differences between the two. How much more exciting, instead, to create a Neanderthal and ask him.

And if that seems too morally fraught, may we interest you in a mammoth?