In my previous posts, I have talked about ways of using different microorganisms to produce different things that are of use to humans. Two specific examples were diesel from fungi, or oils from algae. There are many other examples, some of which I hope to cover in my future posts. At the current time, the main challenge to many of these by-products is that they are not produced at efficiencies and quantities that allow for economical use.
So how would you make fungi produce more biodiesel? How would you make the breakdown of cellulose happen faster? How do you increase oil production in algae?
Although each of these questions, involves a deep understanding of complex cellular networks, at the base of each of these problems is a protein that carries out a given task. Increasing efficiencies of these proteins can help to increase the yields of the desired products.
Today, we will look at one way, how proteins can be "engineered" by asking:
What is Directed Evolution?
To discuss this idea, we need to briefly go over the basic concept of evolution. Without going into the philosophical implications of evolution, evolution relies on the following observations:
1.) Variation: Be it butterfly populations, birds, or humans, you will probably notice that even within one species, there are clearly observable differences ranging from appearance, or size to things less observable, like the proteins within each cell.Where the differences came from is subject to a different discussion.
2.) Natural Selection: The environment imposes certain pressures. In natural selection an environment may make it harder for predators to find and eat butterflies that have darkly colored wings compared to white wing.
3.) Adaptation: If the pressures mentioned above persist, this will lead a species to adapt. In our butterfly example, over time this allows for the darkly colored butterflies to become more dominant in the environment, and a defining character of that species.
Similarly, we can "evolve" a protein. This is done in the following ways.
1.) First, we need to introduce variations of the protein in question. To change a protein, we need to go back and change the blueprint which is the underlying DNA sequence. Although this sounds easy, choosing which mutations to try out is not an easy task.Trying out all different possible combinations is not really feasible. To give an often cited example, take a protein with just 400 amino acids. If we imagine proteins to be just letters on a string, and if at a given position, we want to just change 1 amino acid with another, there would be 20 ways of doing so. If we were to try to change just change 400 positions with any of the 21 possible amino acids, there would be 20^400 different ways of doing so. These are unimaginably large numbers. To give a comparison of how much this would be, the estimated number of molecules in the entire universe is only about 4*10^79 . So the number of possible mutations far exceeds the estimated number of molecules in the entire known universe!!! Therefore, choosing which mutations to try out gives directed evolution its direction so to speak. We will explore some ways of how mutations are chosen in future blog posts.
2.) The proteins are then expressed, and subjected to the "selection pressure" which is some desired criteria we wish the protein to have. Some examples of criteria are: specificity, how fast it processes, stability of protein, or the ability to work under certain temperature or pH conditions. Proteins that fulfill the criteria are then selected for.
Why aren't proteins designed rationally?
Let's take a chair as an example. Making a naturally occurring protein fit for a particular task would be akin to improving an imperfect chair where for example not all the legs are even. In the perfect world, we would be able to see which leg was short and fix the problem in a very direct way. In the real world, our knowledge of protein structures is very incomplete. Many structures are unknown. And it takes significant efforts to find the structure of just one protein. Making a chair more stable without knowing which leg is short is difficult. Nature's way is to randomly slightly change copies of the chair of unknown structure and then to test for the stability of the chair. In another analogy, when ants follow a certain destination along a planned route, and find the route blocked, the ants natural instincts would be to find a detour around the blockage by randomly searching around the blockage. Natural selection can be compared to the ant's way to find the way around a blockage given their limited knowledge.
Take Home Message
The use of enzymes can in many cases reduce the need for energy or replace toxic chemical reactions. By using bioprospecting, we turn to nature to find potential enzymes suitable for a given task. By using directed evolutionary algorithms, proteins can be improved for specific industrial use. Directed evolution can therefore contribute to reduction of the human impact on the environment by producing things in more sustainable ways. In future posts, we shall explore some of the algorithms with specific examples.