HAVE you ever investigated your family tree? Whether it is to learn about our great-grandparents, discover what life was like in the past, or just satisfy our curiosity, there is a growing craze for finding out as much as possible about our ancestors. Genealogy has become one of the top reasons for using the internet (after less wholesome pursuits, of course).
Some people investigate the previous two or three generations of their family; others trace their ancestors over many centuries. But when it comes to extreme genealogy, none can compete with a group of scientists trying to find out about the ancestor of all life.
Every living thing on Earth – from humans to bacteria, from bluebells to blue whales – is thought to be descended from one single entity, a sort of primitive cell floating around in the primordial soup three or four billion years ago. So what did it look like? How did it live, and where? Named the “last universal common ancestor”, or LUCA for short, it has left no known fossil remains, nor any other physical clues to its identity. For a long time, questions about LUCA were thought beyond the reach of science.
Advertisement
But LUCA could enjoy a resurrection at last. Researchers are comparing the genes of all kinds of life to draw a portrait of this mother of us all. Their findings are the subject of huge controversy and are challenging some of the most basic assumptions about primordial life. There is evidence, for example, that early evolution was driven by forces very different to those we usually associate with natural selection. Another surprise is that DNA, life’s code book, may have evolved on two separate occasions. “This is very exciting,” says Anthony Poole of Stockholm University in Sweden. “It changes our picture of LUCA extensively.”
It was in the mid-1800s that Charles Darwin first focused attention on our distant ancestry by setting out his theory of evolution by natural selection. He proposed that similar species had common ancestors and so shared a family tree. But Darwin could not decide whether all living creatures belonged to one such tree or several.
It was only in the 1950s and 60s that efforts to probe the most basic operations of cells began unveiling the similarities that link all life forms. Almost every organism, for example, uses long strands of deoxyribonucleic acid, better known as DNA, to encode the countless proteins required to build and sustain life. They also use short lengths of a similar molecule called ribonucleic acid (RNA) for retrieving the information stored in DNA one gene at a time, to allow the proteins to be manufactured. And without exception all life forms use large and complex molecular machines known as ribosomes to use those RNA snippets as templates for assembling proteins out of amino-acid building blocks.
Perhaps the most telling evidence for a single common ancestor is the shared language of our genes. Although there seem to be no biochemical reasons why certain “letters” of DNA or RNA should encode certain amino acids, the same codes are used across the tree of life, with only a few exceptions (New Scientist, 30 August 2003, p 34).
Although LUCA was one of our earliest ancestors, it was not the very first living entity. LUCA’s DNA, RNA and ribosomes, almost certainly enclosed by a membrane-bound cell, would have been far too complex to have arisen spontaneously from the primordial soup. The first living entity would have been merely a self-replicating molecule, thought to have arisen around 4.3 billion years ago. It must have slowly evolved into a variety of primitive cells. But only one – LUCA – had descendants that were ultimately successful.
So what did LUCA look like? Our earliest insights came from the work of Carl Woese, a molecular biologist at the University of Illinois in Urbana-Champaign. In the late 1960s he developed a technique for gauging relatedness between species by comparing the sequence of a small segment of RNA found within ribosomes. Assuming genetic mutations accrue over time, the more disparate two species’ sequences are, the further back in time the species must have diverged.
Woese’s work, which spanned more than a decade, redefined how biologists classify life, in ways that would have a major impact on the search for LUCA. Previously life forms had been divided at the most basic level into two groups: eukaryotes and prokaryotes. Eukaryotes were all animals, plants, fungi and single-celled microbes such as yeast. They consisted of relatively large cells containing many complex internal structures, such as mitochondria for producing energy and a nucleus for housing their DNA, which was wrapped in protein. Prokaryotes, mainly bacteria, comprised smaller, simpler cells, lacking these internal elements but possessing a rigid cell wall made from distinctive sugary proteins called peptidoglycans, not found in eukaryotes.
Woese discovered that within prokaryotes there was actually a third type of creature: a bizarre class of microbes that he named the archaea. Although similar to bacteria in many ways, archaea lack the defining peptidoglycans and possess several eukaryote traits such as having DNA wrapped in protein. From then on scientists adopted a new classification system in which life is divided into three domains: archaea, bacteria – all the prokaryotes that were not archaea – and eukaryotes (see Graphic).
“What it really looks like is that DNA has evolved twice”
Although a major advance, Woese’s early efforts left a key question unanswered: in which order had the three domains evolved? In other words, was LUCA a bacterium, an archaean or a eukaryote? In the late 1980s further comparisons of ribosome RNA suggested bacteria were the oldest domain. Although subsequent analyses of other genes yielded conflicting results, this has remained the prevailing view.
But not everyone subscribes to it. One of the leading naysayers is Patrick Forterre at Paris-Sud University in Orsay, France. He thinks there is a major flaw in the method of such genetic analyses: they fail to take account of the different rates at which mutations can accumulate among different groups. So faster-evolving lineages such as bacteria appear older than they really are.
Bacteria are often assumed to be more primitive than eukaryote cells because they are simpler. But Forterre points out that while eukaryotes are more complex, they are also riddled with what he sees as primitive machinery. For example, eukaryote chromosomes consist of linear strands of DNA that require elaborate molecules called telomeres to protect their ends from damage during replication. Bacterial chromosomes form a loop so they have no need for telomeres.
Eukaryote genes also contain introns – non-coding and often apparently useless DNA whose corresponding sequences must be removed from the intermediary RNA molecules during protein manufacture by complex editing machines known as spliceosomes. Bacteria lack introns so they do not need spliceosomes. In comparison with eukaryotes, bacteria are sleek and efficient at making proteins. They can begin the first step on the road to protein synthesis within seconds; the same events in eukaryotes take half an hour. As such, Forterre argues that bacteria probably evolved more recently, and that LUCA was in fact a eukaryote – one of us.
Elsewhere, a different approach to uncovering LUCA’s identity came from the first genome-sequencing projects, completed in the 1990s. These allowed researchers to list the genes that were common to all life forms and so likely to have been passed down from LUCA. But to everyone’s surprise, the number of genes shared throughout the tree of life turned out to be remarkably small. The most recent comparison, for example, looked at 100 species and found only 60 genes in common (Nature Reviews Microbiology, vol 1, p 127) – far too few to maintain a cell-based life form such as LUCA, no matter how primitive. It appears that much of the evolutionary record has been erased from species’ genomes due to gene loss as organisms adapt to new conditions and ditch redundant genetic material.
Another problem is what is sometimes called horizontal gene transfer. In contrast to the vertical transfer of genes from parent to offspring, single-celled creatures such as bacteria can swap genes between individuals of one generation. Horizontal transfer seems to have played a major role in early evolution. As a result, living species are mosaics of genes with different evolutionary histories.
Hot or not?
So can we gain any clues about LUCA from where it would have lived? Certain family trees drawn up from gene comparisons suggest the earliest life forms were hyperthermophiles – organisms that live at temperatures over 80 °C. This suggested that LUCA was such a creature too, perhaps living near deep-sea hydrothermal vents, where abundant minerals would have offered an energy source on a planet then lacking oxygen.
This theory’s weak point is that life in extreme heat requires special enzymes to protect RNA and DNA from damage. It seems more probable that simpler life forms first arose at moderate temperatures before evolving the enzymes that allowed them to gradually nudge closer to hot zones. And recent research has further undermined the theory. Work done in 2000 showed that reverse gyrase, an enzyme that appears to boost DNA’s resistance to heat damage and is found only in hyperthermophiles, did not evolve until after the three domains had split. Meanwhile, other protective genes found in bacteria appear to have been horizontally acquired from archaea.
A more direct challenge to the theory that LUCA was a heat lover comes from recent work by Celine Brochier and Herve Philippe, evolutionary biologists then at the Pierre and Marie Curie University in Paris, France. They argue that previous work on ribosome RNA was flawed because it included rapidly evolving genetic material, which is more likely to have lost useful historical information. Instead they focused on the more slowly evolving parts of ribosome RNA. In 2002 they published a tree that suggests the oldest living organisms are an unusual group of bacteria known as Planctomycetales, which survive only at moderate temperatures (Nature, vol 417, p 244). These organisms lack peptidoglycan in their cell walls and their chromosomes are enclosed in a membrane, the closest thing in bacteria to a cell nucleus.
But not everyone agrees. Molecular biologist Massimo Di Giulio at the International Institute of Genetics and Biophysics in Naples, Italy, believes Brochier and Philippe failed to use enough material from the ribosome RNA to make their analysis meaningful. Di Giulio repeated the French team’s work using more genetic material and got different results. In the tree he drew, Planctomycetales were still near the base, but two groups of extreme heat lovers – Aquificales and Thermotogales – proved even more ancient. The debate is currently at an impasse, stalled by the uncertainty over whose methods are more accurate.
“There was a variety of primitive cells, but only one had successful descendants”
Even the very nature of LUCA’s genes has come into question. It has long been thought that the first self-replicating organisms could not have used today’s system of DNA and proteins because it is too complex and interdependent to have arisen spontaneously. DNA encodes the proteins that catalyse the chemical reactions that replicate DNA – one could not exist without the other. So the first genes were probably made of RNA, which, due to its different chemical properties, can catalyse some reactions without proteins. This “RNA world” would eventually have been superseded by the superior DNA/protein system, with RNA relegated to its present accessory role. LUCA, it was first thought, could only have existed after the DNA/protein stage had been reached because all life forms seemed to make and use DNA in the same way. For example, the enzyme ribonucleotide reductase, which organisms use to create the building blocks for DNA, contains a molecular core that is identical in all three domains. “It’s like looking at triplets,” says Poole.
But recent genome comparisons have uncovered a list of differences in the molecular machinery associated with DNA in the three domains. Some of the enzymes that bacteria use for replicating DNA, for example, are unrelated to the ones used by archaea and eukaryotes. Poole and others now believe the most likely explanation is that LUCA existed before the switch from RNA genes to DNA ones occurred. “What it really looks like,” says Poole, “is that DNA has evolved twice.”
If that were not heresy enough, Carl Woese, whose work identified the domain of archaea, has recently cast doubt on whether LUCA even existed. He argues that the last common ancestor was not an individual organism, but a varied community of horizontally gene-swapping primitive cells.
The earliest membrane-bound cells would have been extremely simple, Woese argues, comprising a few basic components that could function independently if their genes were acquired individually by other cells. Horizontal gene transfer would have been the main power behind evolution at this point, not vertical inheritance. “There would have been a time when gene transfer was the dominant evolutionary force,” says Woese.
As cells became more complex, individual components acquired at random could not be so easily incorporated. At this point, which Woese calls the Darwinian threshold, genomes began to depend on inheritance, and lineages with distinct identities began to emerge from the communal soup. Woese thinks this theory is the best explanation of the fact that the tree of life looks different depending on which genes you choose to analyse.
But other scientists have not given up on LUCA and are trying to understand the inconsistencies by developing better analytical tools. At the National Center for Biotechnology Information in Bethesda, Maryland, evolutionary biologist Eugene Koonin has recently renewed attempts to compile a list of LUCA’s genes. His team has developed a computer model for determining the relative roles played by gene loss and horizontal gene transfer, and so far they have identified about 600 candidate genes.
Koonin and others hope to discover the minimum genes required to run a primitive cell so this minimal genome can be assembled in the lab. “It is easy to imagine a ‘Jurassic Park’ of cellular evolution, with experimental study of various reconstructed ancestral forms,” wrote Koonin in a recent paper.
Clearly the search for LUCA is only just hotting up.
