杏吧原创

Read our related editorial: Uprooting Darwin鈥檚 tree

IN JULY 1837, Charles Darwin had a flash of inspiration. In his study at his house in London, he turned to a new page in his red leather notebook and wrote, 鈥淚 think鈥�. Then he drew a of a tree.

As far as we know, this was the first time Darwin toyed with the concept of a 鈥渢ree of life鈥� to explain the evolutionary relationships between different species. It was to prove a fruitful idea: by the time he published On The Origin of Species 22 years later, Darwin鈥檚 spindly tree had grown into a mighty oak. The book contains numerous references to the tree and its is of a branching structure showing how one species can evolve into many.

鈥淭he affinities of all the beings of the same class have sometimes been represented by a great tree. I believe this simile largely speaks the truth鈥︹��

The tree-of-life concept was absolutely central to Darwin鈥檚 thinking, equal in importance to natural selection, according to biologist W. Ford Doolittle of Dalhousie University in Halifax, Nova Scotia, Canada. Without it the theory of evolution would never have happened. The tree also helped carry the day for evolution. Darwin argued successfully that the tree of life was a fact of nature, plain for all to see though in need of explanation. The explanation he came up with was evolution by natural selection.

Ever since Darwin the tree has been the unifying principle for understanding the history of life on Earth. At its base is LUCA, the Last Universal Common Ancestor of all living things, and out of LUCA grows a trunk, which splits again and again to create a vast, bifurcating tree. Each branch represents a single species; branching points are where one species becomes two. Most branches eventually come to a dead end as species go extinct, but some reach right to the top 鈥� these are living species. The tree is thus a record of how every species that ever lived is related to all others right back to the origin of life.

鈥溾��The green and budding twigs may represent existing species, and those produced during each former year may represent the long succession of extinct species鈥�

For much of the past 150 years, biology has largely concerned itself with filling in the details of the tree. 鈥淔or a long time the holy grail was to build a tree of life,鈥� says Eric Bapteste, an evolutionary biologist at the Pierre and Marie Curie University in Paris, France. A few years ago it looked as though the grail was within reach. But today the project lies in tatters, torn to pieces by an onslaught of negative evidence. Many biologists now argue that the tree concept is obsolete and needs to be discarded. 鈥淲e have no evidence at all that the tree of life is a reality,鈥� says Bapteste. That bombshell has even persuaded some that our fundamental view of biology needs to change.

So what happened? In a nutshell, DNA. The discovery of the structure of DNA in 1953 opened up new vistas for evolutionary biology. Here, at last, was the very stuff of inheritance into which was surely written the history of life, if only we knew how to decode it. Thus was born the field of molecular evolution, and as techniques became available to read DNA sequences and those of other biomolecules such as RNA and proteins, its pioneers came to believe that it would provide proof positive of Darwin鈥檚 tree of life. The basic idea was simple: the more closely related two species are (or the more recently their branches on the tree split), the more alike their DNA, RNA and protein sequences ought to be.

It started well. The first molecules to be sequenced were RNAs found in ribosomes, the cell鈥檚 protein-making machines. In the 1970s, by comparing RNA sequences from various plants, animals and microorganisms, molecular biologists began to sketch the outlines of a tree. This led to, among other successes, the unexpected discovery of a previously unknown major branch of the tree of life, the unicellular archaea, which were previously thought to be bacteria.

By the mid-1980s there was great optimism that molecular techniques would finally reveal the universal tree of life in all its glory. Ironically, the opposite happened.

The problems began in the early 1990s when it became possible to sequence actual bacterial and archaeal genes rather than just RNA. Everybody expected these DNA sequences to confirm the RNA tree, and sometimes they did but, crucially, sometimes they did not. RNA, for example, might suggest that species A was more closely related to species B than species C, but a tree made from DNA would suggest the reverse.

Which was correct? Paradoxically, both 鈥� but only if the main premise underpinning Darwin鈥檚 tree was incorrect. Darwin assumed that descent was exclusively 鈥渧ertical鈥�, with organisms passing traits down to their offspring. But what if species also routinely swapped genetic material with other species, or hybridised with them? Then that neat branching pattern would quickly degenerate into an impenetrable thicket of interrelatedness, with species being closely related in some respects but not others.

We now know that this is exactly what happens. As more and more genes were sequenced, it became clear that the patterns of relatedness could only be explained if bacteria and archaea were routinely swapping genetic material with other species 鈥� often across huge taxonomic distances 鈥� in a process called horizontal gene transfer (HGT).

At first HGT was assumed to be a minor player, transferring only 鈥渙ptional extra鈥� functions such as antibiotic resistance. Core biological functions such as DNA replication and protein synthesis were still thought to be passed on vertically. For a while, this allowed evolutionary biologists to accept HGT without jeopardising their precious tree of life; HGT was merely noise blurring its edges. We now know that view is wrong. 鈥淭here鈥檚 promiscuous exchange of genetic information across diverse groups,鈥� says Michael Rose, an evolutionary biologist at the University of California, Irvine.

From tree to web

As it became clear that HGT was a major factor, biologists started to realise the implications for the tree concept. As early as 1993, some were proposing that for bacteria and archaea the tree of life was more like a web. In 1999, Doolittle made the provocative claim that 鈥渢he history of life cannot properly be represented as a tree鈥� (). 鈥淭he tree of life is not something that exists in nature, it鈥檚 a way that humans classify nature,鈥� he says.

Thus began the final battle over the tree. Many researchers stuck resolutely to their guns, creating ever more sophisticated computer programs to cut through the noise and recover the One True Tree. Others argued just as forcefully that the quest was quixotic and should be abandoned.

The battle came to a head in 2006. In an ambitious study, a team led by Peer Bork of the European Molecular Biology Laboratory in Heidelberg, Germany, examined 191 sequenced genomes from all three domains of life 鈥� bacteria, archaea and eukaryotes (complex organisms with their genetic material packaged in a nucleus) 鈥� and identified 31 genes that all the species possessed and which showed no signs of ever having been horizontally transferred. They then generated a tree by comparing the sequences of these 鈥渃ore鈥� genes in everything from E. coli to elephants. The result was the closest thing yet to the perfect tree, Bork claimed ().

Other researchers begged to differ. Among them were Tal Dagan and William Martin at the Heinrich Heine University in D眉sseldorf, Germany, who pointed out that in numerical terms a core of 31 genes is almost insignificant, representing just 1 per cent of a typical bacterial genome and more like 0.1 per cent of an animal鈥檚. That hardly constitutes a mighty oak or even a feeble sapling 鈥� more like a tiny twig completely buried by a giant web. Dagan dubbed Bork鈥檚 result 鈥渢he tree of 1 per cent鈥� and argued that the study inadvertently provided some of the best evidence yet that the tree-of-life concept was redundant ().

The debate remains polarised today. Bork鈥檚 group continue to work on the tree of life and he continues to defend the concept. 鈥淥ur point of view is that yes, there has been lots of HGT, but the majority of genes contain this tree signal,鈥� Bork says. The real problem is that our techniques are not yet good enough to tease that signal out, he says.

Meanwhile, those who would chop down the tree of life continue to make progress. The true extent of HGT in bacteria and archaea (collectively known as prokaryotes) has now been firmly established. Last year, Dagan and colleagues examined more than half a million genes from 181 prokaryotes and found that 80 per cent of them showed signs of horizontal transfer ().

Surprisingly, HGT also turns out to be the rule rather than the exception in the third great domain of life, the eukaryotes. For a start, it is increasingly accepted that the eukaryotes originated by the fusion of two prokaryotes, one bacterial and the other archaeal, forming this part of the tree into a ring rather than a branch ().

The neat picture of a branching tree is further blurred by a process called endosymbiosis. Early on in their evolution, eukaryotes are thought to have engulfed two free-living prokaryotes. One of these gave rise to the cellular power generators called mitochondria while the other was the precursor of the chloroplasts, in which photosynthesis takes place. These 鈥渆ndosymbionts鈥� later transferred large chunks of their genomes into those of their eukaryote hosts, creating hybrid genomes. As if that weren鈥檛 complicated enough, some early eukaryotic lineages apparently swallowed one another and amalgamated their genomes, creating yet another layer of horizontal transfer ().

This genetic free-for-all continues to this day. The vast majority of eukaryote species are unicellular 鈥� amoebas, algae and the rest of what used to be known as 鈥減rotists鈥� (). These microscopic beasties have lifestyles that resemble prokaryotes and, according to Jan Andersson of the University of Uppsala in Sweden, their rates of HGT are often comparable to those in bacteria. The more we learn about microbes, the clearer it becomes that the history of life cannot be adequately represented by a tree.

Hang on, you may be thinking. Microbes might be swapping genes left, right and centre, what does that matter? Surely the stuff we care about 鈥� animals and plants 鈥� can still be accurately represented by a tree, so what鈥檚 the problem?

Well, for a start, biology is the science of life, and to a first approximation life is unicellular. Microbes have been living on Earth for at least 3.8 billion years; multicellular organisms didn鈥檛 appear until about 630 million years ago. Even today bacteria, archaea and unicellular eukaryotes make up at least 90 per cent of all known species, and by sheer weight of numbers almost all of the living things on Earth are microbes. It would be perverse to claim that the evolution of life on Earth resembles a tree just because multicellular life evolved that way. 鈥淚f there is a tree of life, it鈥檚 a small anomalous structure growing out of the web of life,鈥� says John Dupr茅, a philosopher of biology at the University of Exeter, UK.

More fundamentally, recent research suggests that the evolution of animals and plants isn鈥檛 exactly tree-like either. 鈥淭here are problems even in that little corner,鈥� says Dupr茅. Having uprooted the tree of unicellular life, biologists are now taking their axes to the remaining branches.

For example, hybridisation clearly plays an important role in the evolution of plants. According to Loren Rieseberg, a botanist at the University of British Columbia in Vancouver, Canada, around 14 per cent of living plant species are the product of the fusion of two separate lineages.

Hybrid humans

Some researchers are also convinced that hybridisation has been a major driving force in animal evolution (see 鈥淣atural born chimeras鈥�, and 鈥淭wo into one鈥�), and that the process is ongoing. 鈥淚t is really common,鈥� says James Mallet, an evolutionary biologist at University College London. 鈥淭en per cent of all animals regularly hybridise with other species.鈥� This is especially true in rapidly evolving lineages with lots of recently diverged species 鈥� including our own. There is evidence that early modern humans hybridised with our extinct relatives, such as Homo erectus and the Neanderthals ().

Hybridisation isn鈥檛 the only force undermining the multicellular tree: it is becoming increasingly apparent that HGT plays an unexpectedly big role in animals too. As ever more multicellular genomes are sequenced, ever more incongruous bits of DNA are turning up. Last year, for example, a team at the University of Texas at Arlington found a peculiar chunk of DNA in the genomes of eight animals 鈥� the mouse, rat, bushbaby, little brown bat, tenrec, opossum, anole lizard and African clawed frog 鈥� but not in 25 others, including humans, elephants, chickens and fish. This patchy distribution suggests that the sequence must have entered each genome independently by horizontal transfer ().

Other cases of HGT in multicellular organisms are coming in thick and fast. HGT has been documented in insects, fish and plants, and a few years ago a piece of snake DNA was found in cows. The most likely agents of this genetic shuffling are viruses, which constantly cut and paste DNA from one genome into another, often across great taxonomic distances. In fact, by some reckonings, 40 to 50 per cent of the human genome consists of DNA imported horizontally by viruses, some of which has taken on vital biological functions (New 杏吧原创, 27 August 2008, p 38). The same is probably true of the genomes of other big animals. 鈥淭he number of horizontal transfers in animals is not as high as in microbes, but it can be evolutionarily significant,鈥� says Bapteste.

Nobody is arguing 鈥� yet 鈥� that the tree concept has outlived its usefulness in animals and plants. While vertical descent is no longer the only game in town, it is still the best way of explaining how multicellular organisms are related to one another 鈥� a tree of 51 per cent, maybe. In that respect, Darwin鈥檚 vision has triumphed: he knew nothing of micro-organisms and built his theory on the plants and animals he could see around him.

Even so, it is clear that the Darwinian tree is no longer an adequate description of how evolution in general works. 鈥淚f you don鈥檛 have a tree of life, what does it mean for evolutionary biology?鈥� asks Bapteste. 鈥淎t first it鈥檚 very scary鈥� but in the past couple of years people have begun to free their minds.鈥� Both he and Doolittle are at pains to stress that downgrading the tree of life doesn鈥檛 mean that the theory of evolution is wrong 鈥� just that evolution is not as tidy as we would like to believe. Some evolutionary relationships are tree-like; many others are not. 鈥淲e should relax a bit on this,鈥� says Doolittle. 鈥淲e understand evolution pretty well 鈥� it鈥檚 just that it is more complex than Darwin imagined. The tree isn鈥檛 the only pattern.鈥�

Others, however, don鈥檛 think it is time to relax. Instead, they see the uprooting of the tree of life as the start of something bigger. 鈥淚t鈥檚 part of a revolutionary change in biology,鈥� says Dupr茅. 鈥淥ur standard model of evolution is under enormous pressure. We鈥檙e clearly going to see evolution as much more about mergers and collaboration than change within isolated lineages.鈥�

Rose goes even further. 鈥淭he tree of life is being politely buried, we all know that,鈥� he says. 鈥淲hat鈥檚 less accepted is that our whole fundamental view of biology needs to change.鈥� Biology is vastly more complex than we thought, he says, and facing up to this complexity will be as scary as the conceptual upheavals physicists had to take on board in the early 20th century.

If he is right, the tree concept could become biology鈥檚 equivalent of Newtonian mechanics: revolutionary and hugely successful in its time, but ultimately too simplistic to deal with the messy real world. 鈥淭he tree of life was useful,鈥� says Bapteste. 鈥淚t helped us to understand that evolution was real. But now we know more about evolution, it鈥檚 time to move on.鈥�

Read our related editorial: Uprooting Darwin鈥檚 tree