This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

Welcome to life on the tiniest scales
17 March 2007 news service

Henry Nicholls
Smash 'N' grab

At the last count 1.8 million species were known to science. That sounds like a lot, but in truth it's no big deal. We may have done a reasonable job of describing the larger stuff, but the fact remains that an average teaspoon of water, soil or ice contains millions of micro-organisms that have never been counted or named.

Not for much longer. A genetic technique called metagenomics is promising to change our view of life on Earth, revealing billions of life forms and a depth of diversity we never knew existed. It is offering up new genes, new species, new ecosystems and even the chance to add entirely new branches to the tree of life. There's more to metagenomics than numbers, though. The approach could also reveal how life survives in extreme environments, provide new molecules for the pharmaceutical industry and reconstruct the secrets of long-extinct species from surviving fragments of their DNA (see "Old bones, new tricks").

Biologists have long suspected that microbial biodiversity stretches far beyond the species they have been able to describe. Cataloguing microbes is no simple matter, however. Look at that teaspoonful of water or soil under a microscope and you'll see a lot of microbes, but telling them apart is another matter. To identify a species it helps to sequence its DNA, but it's not easy to pluck a single cell out of the soup and sequence it. First you have to grow large numbers of the same kind in laboratory cultures. Most micro-organisms simply refuse to grow in the lab, and many of those that do can't be cultured without other species around. That makes them difficult or impossible to study in any detail, let alone sequence. Consequently, microbiologists estimate that at best we have described just 1 per cent of microbial life.

Metagenomics provides a way of getting to know that silent and mysterious majority. The technique involves taking a sample from the environment, pooling the DNA from all the different species present, fracturing it into a mixture of relatively short fragments and then sequencing the lot (see Diagram). It is then possible to work backwards, piecing fragments together like a molecular jigsaw puzzle to reconstruct individual genes or, hopefully, entire genomes. From this it is possible to infer what organisms were present in the original sample. The technique also offers a rough species count, and a genetic snapshot of a microbial community that allows ecologists to monitor it for the effects of environmental change.

The idea of studying genes from mixtures of DNA was first proposed in the mid-1980s by Norman Pace, a microbiologist now at the University of Colorado at Boulder. It was nearly a decade, however, before the tools and techniques to do this kind of analysis became widely available. "The complexity of the microbial world is so far beyond anything we've had to deal with in biology that we simply didn't have the tools to describe it," says Jo Handelsman, a plant pathologist at the University of Wisconsin-Madison.

In 1998, Handelsman and four colleagues coined the term "metagenomics", literally "beyond genomics". Since then, this technique has been used to sequence billions of samples of microbial DNA, identify millions of new genes and a few complete genomes. "It will take years if not decades to unleash the promise of these data," says Handelsman.

There have already been a number of successes. In 2004 Jill Banfield of the University of California, Berkeley, became one of the first to use the technique to probe an entire microbial community. She chose an ecosystem that was likely to be fairly simple - a toxic mine in upstate California where only a small range of microbe species can survive. "The water running out of the mine is more acidic than battery acid," says Brett Baker, a researcher in Banfield's lab. "The concentration of metals in the water is so high that drinking a cupful would be lethal." It was also warm work for the researchers, with temperatures in the mine frequently reaching a sweltering 40 C. "We try not to stay in there too long," he says.

When they sequenced the mine's metagenome, the team discovered a completely new phylum of the archaea, one of the three main branches to the tree of life along with eukaryotes (complex animals and plants) and bacteria (Science, vol 314, p 1933). In earlier work on the mine ecosystem, these archaeans had gone undetected. Measuring just 200 nanometres across - about one-fifth the size of a typical bacterium - the new archaeans may turn out to be the smallest living organisms yet found, says Baker.

Metagenomics is also beginning to help the team understand the workings of the microbial community. For example, they have identified that the ecosystem depends on just one relatively rare species of bacterium that is capable of fixing nitrogen from the atmosphere into a form that the other microbes can use. "Without it, the community simply would not exist," says Baker.

They found the microbe by sifting through the sequence data for signs of nitrogen-fixing genes. They then reconstructed part of its genome from fragments in the metagenomic sample. With this information they were able to grow the microbe in isolation in the lab. "I'm not sure we would have found it without the genome data," says Baker. They also managed to reconstruct the entire genome of a couple of other microbes in the DNA mix - the first time anyone had achieved this.

While Banfield and her team were busy homing in on just a few species of bacteria, Craig Venter, the scientist-cum-entrepreneur made famous by his human genome sequencing project, had other plans for metagenomics. He was excited by the idea that countless new organisms were out there waiting to be discovered but, unlike Banfield, his goal wasn't to understand the workings of an entire ecosystem. He just wanted to find new genes. Lots of them.

In a 2003 pilot project he and his colleagues collected samples of seawater from the Sargasso Sea off Bermuda. In just a few hundred litres of water, they found more than 1.2 million new genes from thousands of different microscopic species. It was a complete surprise that the open ocean, an environment once assumed to be the marine equivalent of a desert, should harbour so much diversity.

Encouraged by these findings Venter decided to take the project one step further. In August 2003 a global ocean sampling expedition set sail aboard Venter's boat, Sorcerer 2. Its aim was to use metagenomics to map ocean life to a new level of detail. For 10 months Sorcerer 2 sailed down the east coast of the US to the Caribbean Sea, through the Panama Canal into the Pacific, around the Galapagos and finally to French Polynesia, collecting a seawater sample every few hundred kilometres.

Count 'em

It wasn't always plain sailing. Although the French Polynesian authorities approved the sampling, the central French government took issue with what it saw as bioprospecting in French territorial waters. "We were placed under house arrest and the boat was not allowed to leave the harbour," says Venter. The voyage had to be put on hold for more than a week while he thrashed out a new agreement with French officials.

The ship did set sail again, however, and this week PLoS Biology has published three papers announcing the discovery of over 6 million new genes from its seawater samples (vol 5, p e77). Most of these are similar to genes we already know about, but almost 2000 appear to be unique to this dataset.

Discoveries like these are redefining how we think about microbes. What's more, they have brought the way we classify species into question. So great is the variation being detected by metagenomics that it is hard to deal with this diversity using conventional taxonomic categories. "The whole species concept is under discussion right now," says Handelsman.

Even viruses are throwing up surprises. In November, microbial ecologist Forest Rohwer at San Diego State University in California and colleagues reported that after surveying 68 ocean sites around the world, most of the viruses they detected were completely different from their land-based counterparts: they had a distinct "marine-ness".

As well as shedding new light on old categories, metagenomics is also being tipped as the latest tool in environmental monitoring. Rohwer has been investigating how pollution may have subtle effects on microscopic life in the Northern Line Islands, a chain of five coral atolls in the Pacific that runs from the uninhabited Kingman Reef to the island of Kiritimati. Using metagenomics to track microbial communities, Rohwer's team has found clear changes in the metagenome that correlate with increasing human disturbance from north to south.

Rohwer believes that overfishing and an influx of sewage is allowing pathogenic microbes to dominate and destroy the more disturbed reefs. "Metagenomics offers us a way to understand how the micro-organisms and the macro-organisms are interacting," he says. Then it should be possible to identify marker genes in the microbial community that can predict the decline of a reef, buying time to avert and maybe reverse the trend.

Others are looking at how a similar approach could be used to explore the role of microbes in global environmental change and how such change will, in turn, affect them. Ian Joint, a marine microbiologist at Plymouth Marine Laboratory in the UK, is working on how ocean acidification will affect ocean microbes. "In the next 100 years, the oceans will be more acidic than they have been in the past 25 million years," he says. "We don't know how they will function when the pH changes."

Joint and his colleagues recently designed an experiment to find out. On a huge floating laboratory in a Norwegian fjord, the researchers simulated the effects of different levels of acidification by pumping varying amounts of carbon dioxide through containers of seawater. The team are still analysing how these changes affect the marine metagenome, but it's already clear that dramatic changes are to be expected. "We've seen large changes in the productivity of the phytoplankton," says Joint.

There is much excitement about where the technique will take us. Some researchers are using metagenomics to investigate how life has evolved around hydrothermal vents and in other sites of extreme temperatures. They may lead us to life in totally unexpected places. And each new study will bring a glut of new genes, and therefore proteins, which may lead to useful drugs. Metagenomics has the potential to influence any field where microbes are involved, says Handelsman, and that means just about every area of biology.

Many challenges remain, not the least of which is trawling through the vast amount of data generated by each sweep of the ocean or other ecosystem. Handelsman has spent the past decade trying to understand the metagenome of a soil community. She does not expect to achieve this any time soon. "Soil appears to be the most complex environment on Earth," she says, and extracting DNA and removing contaminants without damaging the sample is a major technical problem.

The biggest challenge of all, though, is only now becoming apparent. It is clear that the vast majority, possibly even 90 per cent, of all microbial diversity comes from very rare organisms. It is possible to detect them in metagenomic samples, but researchers like Venter are really only piecing together genes from the most abundant organisms, says Mitchell Sogin, director of the Josephine Bay Paul Center in Comparative Molecular Biology and Evolution at the Marine Biological Laboratory in Woods Hole, Massachusetts. The real diversity could easily be 100 times what Venter is finding, he says.

Sogin and others are beginning to find ways to home in on the DNA fragments from these rare organisms. "It's starting to happen," he says. But for now, we still know virtually nothing about these organisms except that they're out there. One thing is for sure: our present tally of 1.8 million species will soon look like a drop in the ocean.

From issue 2595 of New Scientist magazine, 17 March 2007, page 44-47

Old bones, new tricks

We may not be able to bring extinct species back to life, but metagenomics is offering us the next best thing - a look at their long-lost genomes.

Last year, an international team led by Eddy Rubin of the Lawrence Berkeley National Laboratory in California recovered and sequenced genetic material from a 38,000-year-old Neanderthal thigh bone (New Scientist, 11 November 2006, p 44). The achievement was widely reported as the Neanderthal genome project. In fact, it was more like a metagenome project.

The vast majority of the 6 million base pairs Rubin's team sequenced were from micro-organisms. Using metagenomics the researchers could sift through these fragments and pick out 65,000 base pairs of Neanderthal DNA - about 0.002 per cent of the whole genome (Science, vol 314, p 1113).

Next the team assembled the fragments into longer stretches and compared them with equivalent sequences in humans. It turned out that we share at least 99.5 per cent of our DNA with our extinct cousins. From this researchers estimate that it is 700,000 years since humans and Neanderthals last shared a common ancestor.

A similar approach is being used for other extinct species. In 2005, researchers at McMaster University in Hamilton, Ontario, Canada, pulled out about 13 million base pairs of woolly mammoth DNA from samples collected in Siberia. The same year another team did something similar for the extinct cave bear, Ursus spelaeus. But since DNA doesn't survive for long, metagenomics is unlikely to reveal much about species that died out more than a million years ago.

For the latest from New Scientiist visit

Academy disclaimer: We cannot guarantee the accuracy of information in external sites.