A Minimal Model of Functional Redundancy as a Route to Enzyme Specialisation

Enzymes are the tiny machines that keep life running, speeding up the chemical reactions that build cells, digest food, and repair DNA. For a long time, scientists believed that for these machines to become highly efficient at one specific job, they usually needed to split apart. The standard story was that a gene would copy itself, giving the organism two versions of the same tool. One copy could keep doing the original work, while the other was free to change and specialize in a new task. This process, known as gene duplication, has long been seen as the primary way nature creates new, specialized enzymes. But this raises a tricky question: what if an organism doesn't have the luxury of making a copy? What if it must make do with the proteins it already has? Can a single protein become more focused on one job without losing the ability to help with others that the organism still needs?

A team of researchers set out to answer this by building a digital model of evolution to watch how proteins change when they are forced to share the workload. They did not look at real bacteria or yeast in a lab; instead, they created a simplified computer simulation where a set of proteins had to collectively manage 120 different chemical reactions. In their world, the "fitness" or success of the organism depended entirely on how well the entire group of proteins performed these reactions together. The researchers started with proteins that were generalists, meaning each one could help with many different reactions, but none were perfect at any single one. They then let these proteins evolve over time, allowing them to mutate and change their activity levels, but with one strict rule: no new genes were allowed to be created. The number of proteins remained fixed throughout the entire experiment.

The goal was to see if these proteins would naturally become more specialized, focusing their energy on a few key tasks, even though the organism still needed all 120 reactions to happen. The researchers tested this under many different conditions. They varied how many proteins were available to do the work, how likely mutations were to increase or decrease activity, and how much the structure of the protein limited which activities could happen at the same time. In one version of the simulation, mutations were more likely to reduce activity than to increase it, mimicking a scenario where it is easier to break a function than to improve it. In another version, increases and decreases were equally likely. They ran hundreds of these simulations to see what would happen.

The results showed that specialization happened even without gene duplication. Across almost every scenario, the proteins became more focused. They stopped spreading their efforts thinly across many reactions and instead concentrated their activity on a smaller set of tasks. The researchers measured this by looking at the "effective breadth," a statistical measure of how broadly a protein's activity is distributed. In the beginning, the average protein had an effective breadth of approximately 13.5. After the simulation ran its course, this number dropped significantly. The mean reduction in breadth ranged from about 1.6 to 4.0 units under the loss-biased mutation scenario, and from 6.3 to 8.2 units under the symmetric mutation scenario. At the same time, the share of a protein's total effort dedicated to its single strongest job grew larger. This means the proteins were not just losing functions randomly; they were actively reorganizing to become better at their primary tasks.

Crucially, this specialization did not come at the cost of the organism's survival. Even as the proteins became more focused, the weakest reactions in the system did not collapse. Because there were many proteins working together, if one protein lost its ability to help with a specific reaction, the others were still there to pick up the slack. The system was buffered by this redundancy. The researchers found that the performance of the most difficult reactions remained relatively stable, dropping only slightly or even improving in some cases. This suggests that the pressure to specialize did not force the organism to sacrifice its ability to handle its basic needs. The proteins could become more efficient specialists while the group as a whole continued to supply everything the organism required.

One of the most surprising findings was that this shift toward specialization did not require a bias toward losing functions. Many scientists assumed that for a protein to become specialized, it would need a mutational pressure that favored breaking down extra abilities. However, the researchers found that specialization occurred even when mutations were equally likely to increase or decrease activity. In fact, the simulations where increases and decreases were balanced produced an even stronger shift toward specialization than the ones where losing activity was more common. This indicates that the drive to specialize comes from the way the system is organized and how success is measured, rather than from a specific type of mutation. The proteins naturally found a way to concentrate their efforts because it was the most effective strategy for the group to succeed.

The study also revealed that the physical structure of the proteins played a complex role. The researchers included a rule in their model that prevented a single part of a protein from being too good at two different things at the same time, simulating a real-world physical limit. They found that this limit did not simply push proteins to specialize more and more as it got stricter. Instead, the effect was non-linear. If the limit was too strict, it actually made it harder for proteins to specialize because it blocked certain evolutionary paths. If the limit was too loose, the proteins didn't feel enough pressure to change. There was a middle ground where the physical constraints worked best to encourage specialization. This suggests that the physical architecture of enzymes is a fine-tuned factor in how they evolve, not just a simple barrier.

Throughout the simulations, the total amount of chemical activity in the system actually increased. The proteins became more powerful overall, but this was not because the model forced them to be more efficient in a way that saved resources. The model did not penalize the organism for making more active enzymes. Instead, the increase in total activity was a side effect of the proteins finding better ways to work together. The key takeaway was not that the system became more efficient in terms of energy use, but that the distribution of work changed. The proteins stopped being generalists who did a little bit of everything and became specialists who did a lot of a few things, while the group as a whole maintained its ability to handle all the necessary tasks.

This work challenges the idea that gene duplication is the only way for enzymes to become specialized. While duplication is a well-known and important process in nature, this study shows that it is not a logical requirement for specialization to occur. Even when the number of proteins is fixed, the existing tools can reorganize themselves. The proteins can shift their focus, becoming more dedicated to specific jobs, while the collective effort of the group ensures that no critical function is left undone. The researchers conclude that the path to specialization is broader than previously thought. It can happen through the reorganization of existing contributions, driven by the need to balance performance across a complex network of reactions.

The implications of this finding are significant for how we understand the history of life. It suggests that early organisms, which may have had very few genes and no capacity for duplication, could still have evolved highly specialized enzymes. They could have done this by tweaking the activities of the few proteins they had, allowing them to become more efficient without needing to copy their genetic code. The study provides a clear, simulated example of how nature can solve the problem of specialization through cooperation and redistribution, rather than just through multiplication. It shows that the evolution of complexity does not always require new parts; sometimes, it just requires the old parts to find a better way to work together.

In the end, the researchers demonstrated that specialization is a natural outcome of a system where many agents share a workload and where success depends on the collective performance. The proteins in their model did not need a master plan or a specific type of mutation to become specialists. They simply evolved toward a state where they were better at what they did best, while relying on their neighbors to cover the rest. This process happened consistently across different conditions, proving that the ability to specialize is a robust feature of biological systems, even in the absence of gene duplication. The study offers a new perspective on how life might have evolved from simple, generalist beginnings into the complex, specialized machinery we see today.