Australasian Science: Australia's authority on science since 1938

Survival of the Different

When a simple ancestral population of bacteria is kept in a constant controlled environment, a rich mixture of types evolves rather than a single “winner”.

When a simple ancestral population of bacteria is kept in a constant controlled environment, a rich mixture of types evolves rather than a single “winner”.

By Tom Ferenci

If evolution is about survival of the fittest, why does diversity emerge instead of perfectly evolved organisms that are fit in all environments? Now the complex trade-offs that shape the evolution of diversity have been measured.

Darwin told us that evolution is about the natural selection of types best fitted to their environment. This is sometimes interpreted as the survival of the fittest, but this is not really true. The emergence of super-fit types is strongly constrained by complex evolutionary forces we do not fully understand. Rather than leading to outright winners, diversity is the most common outcome of evolution.

Plants and animals are diverse enough, but the variety is most extreme with organisms we do not even see: the microbes living in and on us. More than 10,000 different types of organisms are part of the healthy human microbiome. This richness of microbial types reinforces the need to understand two fundamental questions in evolution: how diversity arises, and how it is maintained.

It is has become possible to study both these fundamental questions using bacteria in the laboratory. We can now follow diversification in real time and analyse this fascinating process. The maintenance of diversity can also be dissected by studying the way in which co-existing organisms are generated as a result of trade-offs between evolved characteristics. These trade-offs are crucial in ensuring that perfect, super-fit organisms do not evolve in nature, and diversity is maintained instead.

Our research has demonstrated evolutionary diversification in the laboratory. When we maintained a simple ancestral population of E. coli bacteria in a constant controlled environment for extended periods, the outcome was repeatedly the evolution of a rich mixture of types rather than a single “winner”.

Many of the types that evolved were equally fit, and they co-existed over many generations in the same environment. DNA sequencing and other analyses revealed both genetic and physiological divergence in the strains that evolved. This is a clear demonstration that a single ancestor and a simple constant environment give rise to extensive diversity.

Another interesting observation was that many of the evolved types were fit in the growth environment but diminished in other environments. This demonstrates diversification instead of the emergence of super-fit individuals.

To explain this observation, as well as the natural richness of the biosphere, we need to dig a little deeper than the Darwinian ingredients of mutation and selection, and consider several constraints that limit the evolution of a single fit type.

In the past century, several theoreticians have suggested that the evolution of “winner” types is constrained by principles such as ecological trade-offs. In general, a trade-off is a negative correlation between two characteristics. For example, if I have a fixed income I may be able to afford a new computer but may have to sacrifice a holiday. But there are also intermediate options: if I buy a cheaper computer I may also afford a short holiday.

Bacteria do not deal with money but they do deal with other limiting resources and need to make resource allocation decisions. The most common limiting resource is the food they need. Most environments in nature, like soil or water, have low levels of nutrients. Whether hunting for food, scavenging minerals through roots or trying to trap nutrients in solution, this is a problem for nearly all forms of life.

Every organism has to undertake a balancing act between its capabilities. Generally, if it’s good at one thing it will be bad at another. While this notion has been established for a long time, the key mechanisms have proved elusive. As we shall see, the resource allocation settings provide a key to understanding ecological and evolutionary behaviour.

To keep things simple, let us deal with just one particular resource – the availability of a sugar like glucose, which is limiting but needs to be divided between the ability to produce lots of offspring and the ability to survive in hostile environments. Empirical observations with plants and animals have all indicated that organisms are good at one or the other but never optimal for both. Organisms that rapidly generate lots of offspring generally do not nurture or protect their young whereas organisms that protect themselves and their offspring generally produce fewer and/or more slowly maturing young.

The ecological explanation of this negative correlation is that organisms cannot do both when resources are limiting. Some have evolved to focus on rapid reproduction whereas others have evolved a protective strategy for survival. This relationship is called a multiplication–survival trade-off.

The sheer complexity of trade-offs has slowed our understanding of how resource allocation actually works. Growth and self-protection are extremely intricate processes involving large numbers of genes and proteins, particularly in plants and animals. However, bacteria are less complex so we have been examining the allocation of resources in E. coli.

A cellular protein called RpoS is the central regulator of the general stress response in bacteria. RpoS levels control how much of the cellular resources are allocated to growth or stress resistance. However, the concentration of RpoS in normal bacteria is very unstable, and is affected by many environmental variables.

To really test this hypothesis we needed to construct a set of artificial bacteria with many individually different levels of RpoS, ranging from zero to high cellular levels normally only found in highly stressed bacteria. This challenge was recently overcome by Ram Maharjan and Judy Sung in my laboratory.

Using ten strains with different RpoS levels, we observed that stress-resistant organisms (with high RpoS levels) grew slower than bacteria with low stress resistance. Thus it is clear that resource allocation is indeed controlled by RpoS, and is responsible for the multiplication-survival trade-off in E. coli.

When we plotted growth and survival against RpoS levels we found that the shapes of the curves differed remarkably for each resource and stress. They were certainly not linear relationships, so stress resistance does not increase proportionally with RpoS concentration.

Why is this important? Theoreticians have long proposed that trade-offs and their shapes affect the course of evolution. For example, mathematical ecologist Richard Levins has proposed that the shapes determine competition, and consequently types with diverse settings of a trade-off can co-exist. Is this important theory really true?

We were able to test Levins’ predictions using the trade-off shapes we had measured for our set of strains. We modelled the outcome of events when the E. coli strains were grown in environments with intermediate stress levels, and compared the results with the predictions of Levins.

Indeed, the predicted outcomes showed excellent concordance with the results of the evolution experiments we conducted in acid and salty environments. Conversely, the results did not match at all when we incorporated the wrong shapes into our model.

This proved that trade-off shapes do matter. Our results demonstrate that if trade-off shapes are accurately determined they can actually be used to predict which strains will adapt and evolve under different environmental stresses.

An important prediction from the model, and observed in the experiments, was that trade-offs can result in the co-existence of more than one type of bacterium from a single ancestor in experimental cultures. This is, of course, a requirement for the generation of diversity in an environment.

Thus an important conclusion from these studies is that we can finally explain the emergence of diversity rather than a single super-fit organism.

The ability to produce synthetic organisms and strain sets offers a major advance in the study of complex biological phenomena like trade-offs. We now have a better appreciation of why natural selection does not produce perfectly-evolved organisms that are fit in all environments. The need to balance resource allocations in different environments prevents the evolution of “Darwinian demons” that are equally fit in all situations.

Tom Ferenci is Honorary Associate Professor in the School of Molecular Biosciences at The University of Sydney.