TIME was when tiny islands and remote mountain tops were the only places for evolutionary biologists to develop their theories. In quiet isolation, they contemplated the curve of a finch’s beak or the colour of a salamander’s tail. Now a new breed of researchers are taking over. They have ditched their binoculars and sketchbooks in favour of white coats and test tubes. They play God in the laboratory, creating communities of bacteria that replicate and mutate over days and weeks, evolving into new strains right before their eyes.
This approach has already allowed biologists to test ideas about how individual organisms evolve in response to change. It has shown that microbes can adapt quickly to new environments, can produce new strains to fit into new niches, and can evolve to cope if conditions turn hostile. Now test-tube evolutionary research has become more ambitious, and its practitioners have begun to manipulate their glass-encased worlds like stage directors, bringing together larger casts of characters and creating ingenious set designs in which their roles can unfold.
Their work is starting to reveal how diverse ecological communities can change and adapt over time, putting to the test ideas developed out in the field about how organisms adapt to one another as well as to their environment. It may ultimately even help us understand one of evolution’s most puzzling transitions: the creation of multicellular organisms from single cells (see “Getting it together”).
Advertisement
“Slowly evolving strains always came out on top in the end, when in competition with their quickly evolving counterparts”
Pioneers of experimental evolution include Richard Lenski from Michigan State University in East Lansing and Paul Rainey from the University of Oxford, UK, and the University of Auckland in New Zealand. Lenski’s landmark studies began in the late 1980s with a long-running experiment to look at how strains of the gut bacterium Escherichia coli evolve to become more competitive than their ancestors. A decade later, Rainey’s team revealed that a single strain of the leaf-dwelling bacterium Pseudomonas fluorescens differentiates into a variety of different morphs simply by being allowed to grow in an environment with several ecological niches. Studies such as these showed that evolution was open to experimental analysis, including controlled investigations and replicated studies. For the first time, biologists were able to rewind the tape of life and watch how things would evolve when evolutionary history was played over and over again (New Scientist, 13 February 1999, p 28).
With these experimental building blocks in place, researchers have been moving on to the next step. “Now that we know the basic dynamics, we can make [the systems] more complex and look at how communities evolve, as opposed to how single organisms evolve,” says Michael Travisano from the University of Houston, Texas, who worked with Rainey on his original experiments. The size and complexity of natural ecosystems makes it difficult to determine whether a particular change is due to variations in the environment, pressures from neighbouring species or more random factors such as genetic drift. In the lab, by contrast, biologists can exert exquisite experimental control over their microbial populations, which allows them to pin down the specific factors involved in a particular evolutionary trajectory, Travisano says.
One strand of investigation is looking at how the environment in which organisms find themselves affects the way they evolve. These studies test the 70-year-old “shifting balance” theory, which attempts to describe how natural selection and more arbitrary forces interact to produce organisms that are either more or less well adapted to their environment. Evolutionary biologists like to imagine an organism’s genetic possibilities as a mountainous landscape: lofty peaks represent super-survivor combinations of genes, while deep valleys equate to poorly adapted genotypes whose owners will eventually die off. If all organisms were perfectly adapted they would settle on the highest peaks. This clearly doesn’t happen and to understand why means looking at the forces shaping the organism’s evolution.
One explanation might be the conditions in which organisms evolve. In small, isolated groups, where they experience little or no competition, local populations should be free to explore their genetic possibilities through random genetic drift as well as natural selection. This gives them the best chance of attaining their full evolutionary potential. Conversely, organisms that interact with large numbers of other species will not have time to explore the evolutionary landscape in this leisurely way. Here, the balance of evolutionary forces will be tipped away from the aimless meandering of genetic drift and towards the more directed force of selection. The pressures of competition will force them to evolve fast, with natural selection immediately lighting on any mutations that confer a marginal advantage. As a result, species are unlikely to reach the highest peaks of evolution but instead get stuck on lower peaks representing genetic combinations that are good enough for survival, but not optimal.
Global versus local
Until recently, there was little experimental evidence to confirm this prediction arising from the shifting balance theory. But earlier this year, Brendan Bohannan from Stanford University in California described microbial evolution studies that seem to do just that. Using E. coli, Bohannan’s team grew the same strains in two different kinds of environments. The first setting, which they described as a “local” environment, was provided by the surface of agar gel in a Petri dish. This allowed colonies derived from the starting population of microbes to interact in two dimensions only. In the second, “global”, environment microbes mixed freely in a three-dimensional, watery world, like ingredients in a bowl of soup.
The researchers isolated new strains that developed in each environment and pitted them against both their own ancestors and each other. Successive competitions with their own ancestors revealed that strains reared in the local environment went on to evolve more slowly than those in the global environment. Despite this, the slowly evolving strains always came out on top in the end, when in competition with their quickly evolving counterparts. It seems that slower evolution does indeed allow the bacteria to explore more genetic combinations, ultimately reaching a fitness peak that is at a higher level than that achieved by those forced to evolve quickly simply to survive. “These results strongly suggested that localisation made a difference,” Bohannan says.
“Predator-prey interactions can drive microbes to evolve along divergent and unpredictable pathways”
This may not always be the case, though. “We have now looked at two more strains,” Bohannan adds, “and results from these do not support our preliminary conclusions.” He suspects the difference may lie in the precise nature of the strains he starts out with. But even if the effects of local and global environments on evolution are not universal, they do have some interesting practical implications.
For example, they could signal some good news in the fight against drug-resistant bacteria. “If you stop using a particular drug, you should be able to get rid of resistant strains because the sensitive ones are fitter,” Bohannan says. “Sensitive strains can then re-establish themselves and you can start using the drug again.” To be set against that, however, is the unfortunate, if unlikely, possibility that as the microbes evolve rapidly to become drug-resistant they may hit the jackpot by chancing upon an evolutionary Everest. If that happens, they will have no trouble outcompeting all comers – including more slowly adapting non-resistant strains. In that case, the drug to which they had become resistant would have to be abandoned for ever.
Mixed-up world
Then there is the wider issue of humans meddling in the biosphere on a global scale. What do Bohannan’s findings mean for a world in which humans transport organisms across the globe, creating a kind of universal mix of species? “As we enter the ‘homogeneous era’, we alter the dynamics of evolution in ways we can’t anticipate,” Bohannan says. Could this be compromising the ability of organisms to reach their full evolutionary potential and so be making species less fit and more susceptible to extinction? It’s too early to say, Bohannan cautions, but it is possible.
Bohannan and his colleagues do have some compelling evidence that the fate of competing organisms is affected by the environment in which they interact (Nature, vol 418, p 171). They began with three strains of E. coli: one that produces a toxin, one that is resistant to the toxin, and one that is sensitive to the toxin. In a global environment where the strains can interact freely, the toxin-resistant strain quickly triumphed, outcompeting all others. But when the bacteria were restricted to a two-dimensional environment, all three strains survived.
What happens here is that the different strains take it in turns to dominate, but there are always enough individuals left from the other strains for them to triumph once conditions change. First, the toxin-producing strain kills off its toxin-sensitive neighbours, while nearby toxin-resistant individuals survive. But manufacturing toxin costs the producing strain extra energy, allowing the resistant strain – which has more resources to devote to reproducing – to triumph eventually. The toxin disappears as the producing strain dies off, but the resistant strain continues wastefully churning out toxin-resistant proteins that are no longer necessary. In the new environment, the more efficient toxin-sensitive strain triumphs, and the cycle begins again. So the three kinds of microbe end up chasing each other around the surface of the Petri dish in a bacterial version of the rock-paper-scissors game.
An experiment like this shows how organisms within a given ecosystem can reach a kind of equilibrium. But in nature, the interactions between coexisting organisms are not always so stable. Other studies in experimental evolution reveal how the presence of one microbe may influence the evolution of another in ways that are difficult to predict. Rainey and his Oxford colleague Angus Bucking, for example, have recently shown how predator-prey interactions can drive microbes to evolve along divergent paths. Growing 12 populations of a single strain of P. fluorescens together with one of its parasites, a bacteriophage called SBW25f2, they found that after 50 generations the genetic variation between the resulting populations was greatly increased. The new differences were primarily due to the fact that each strain had evolved a distinct way to resist the phage (Nature, vol 420, p 6915).
In another experiment, Travisano and his colleagues showed how it is not just hard times that apply evolutionary pressure: unusual abundance can also do the trick. They took a flask of glucose solution – the traditional home of lab bacteria – and pumped it full of extra glucose. With a glut of food, the E. coli flourished, their numbers increasing to such an extent that before long there was strong competition for the glucose. But instead of the weaker individuals going to the wall, some of them evolved the ability to feed on the leftovers from the banquet. As a result, while one group of microbes fed in the normal way by splitting glucose molecules in two and extracting the energy, the second group evolved to survive by chopping up the by-products still further. “This is an example of how a very simple ecosystem might evolve,” Travisano says.
Building on experiments like these, Travisano and others hope their studies with microbes will yield important insights into the factors that influence how complex natural environments evolve. Their studies certainly add experimental rigour in an area that is otherwise notoriously difficult to analyse.
Test-tube evolution is an ideal way to test theories about evolution inspired by fieldwork in the world’s remotest spots. And its potential extends far beyond just helping us understand the history of life on Earth. Evolution is still happening all around us, and experimental evolution can help unravel how human activity may be shaping life now and in years to come.
Getting it together
IT IS one of biology’s hottest questions: how did multicellular life emerge? And the practitioners of experimental evolution seem to be onto an answer.
The first step in the evolution of complex life is to get single-celled organisms to work together. While nature is rife with instances of cooperation, the puzzle has been to work out how such a seemingly selfless strategy could evolve in a dog-eat-dog world.
“It is costly to cooperate,” says Paul Rainey from the University of Oxford in the UK and the University of Auckland, New Zealand, “so there must be some overall benefit to being in a group.” The discovery that bacteria can easily be coaxed to work together in the lab leaves no doubt that the balance often does tip in favour of cooperation.
The strength of the cooperative urge was cleverly demonstrated by Greg Velicer and his colleague Yuen-tsu Yu from the Max Planck Institute for Developmental Biology in Tübingen, Germany. They looked at the behaviour of the bacterium Myxococcus xanthus. This is a species that cannot spread across solid surfaces unless individuals cooperate by sprouting tiny appendages known as pili and then swarming en masse. Velicer and Yu created a strain that could not produce pili, and then challenged these mutants to re-evolve the ability to swarm. What emerged were two new strains that could swarm, but in a novel way. Instead of growing pili, they excreted a slimy, fibrous matrix that bound the cells together and allowed them to move across the surface using a cellular motor that normally drives individual cells (Nature, vol 425, p 75). “The results show that fundamental transitions to primitive cooperation can readily occur at that level,” Velicer says.
Rainey, too, has seen cooperation evolve. He found that a particular strain of Pseudomonas fluorescens, which he called the fuzzy spreader, is able to colonise the interface between air and liquid in a flask full of nutrient broth by producing a glue-like material that allows individual bacteria to form a raft. The glue costs the bacteria energy to produce, and this slows their growth rate. So what’s the pay-off? The raft gives them access to unlimited amounts of oxygen from the air, which is not available to microbes living within the liquid. “This is the clearest example of evolutionary transition,” Rainey says. “Simple transitions are probably occurring all the time. They could be the precursors to multicellular life.”