BEACON Researchers at Work: Measuring fitness in the Long Term Evolution Experiment

This week’s BEACON Researchers at Work post is by Michigan State University graduate student Mike Wiser.

If there’s one thing you can really depend on about life, it’s that it’s constantly changing.

Many of us learned in our biology classes that evolution is basically a change in a population or a species across generations. This change is something that takes some degree of time to happen, but can produce powerful results — dinosaurs changing into birds, for example. But it’s easy to fall into the trap of thinking that evolution is something that happened. That’s true, but it’s incomplete. Not only did evolution happen, but it’s happening right now. You just generally need a number of generations before you can see things change. This is why I work with bacteria; more generations in a shorter time gives us more to look at. But, first, a non-bacterial example:

If, like me, you’re not really careful about keeping your yard perfect, you’ll end up with some weeds mixed in with the grass. Dandelions, for example, are all over the place. But not all dandelions are the same. Sometimes, they flower and produce seeds when they’re much taller than the grass. Some of them, though, do so when they’re the same height as freshly mowed grass. Seeds coming off the taller dandelions might manage to spread much further than seeds coming off the shorter ones, which makes being tall good for having offspring launch out into the neighbors’ lawn. Short dandelions, though, might be more likely to survive long enough to produce seeds than their taller cousins, who have to face the mower. The frequency with which you mow the lawn will tend to select for one type or another. Assuming that dandelion height can be inherited across generations, you have the ability to control the way in which the population is likely to evolve. Even though you’re not consciously trying to make the dandelions taller or shorter, your actions will influence their growth, survival, and reproduction, and thus help shape future generations.

I work in the lab of Dr. Richard Lenski, with a project called the Long Term Evolution Experiment (LTEE). At its core, this project involves 12 different flasks of bacteria. Every day, we take 1% of the liquid volume of a flask and transfer it to a fresh, sterile flask, clean our tools, and repeat for each of the 12 flasks. Therefore, every day, each population has a chance to grow by 100 fold before it runs out of nutrients, which allows us to know just how many generations a flask can have each day: about 6 and 2/3. Every 75 days, which is every 500 generations, we take the part of the culture we didn’t transfer and freeze it, so we have frozen samples of each lineage that we can revive and study at any time we like.

The LTEE started back in February of 1988. That means we’re now more than 50,000 generations into the project, and all of those frozen samples are available to work with. There’s a lot that can be done with this project, so my work is just a subset of it. Broadly speaking, the questions I’m interested in all concern the evolutionary process itself: how repeatable it is and how predictable it is. Because of the unique nature of the LTEE in terms of how many generations it covers, and how the old samples can be revived for later analysis, it provides an unmatched system in which to look at the long term evolutionary dynamics of a single species adapting to a very simple environment.

In terms of evolutionary dynamics, what matters the most is fitness. Now, the grad students in my lab are pretty physically active: half of the lab bike or walk to work, one person has run marathons, some of us swim or work with weights or play sports pretty regularly, and we’re generally in our 20s and 30s. But from an evolutionary perspective, none of us are fit: we don’t have children, and haven’t found other ways to pass on our genes like helping a sibling care for our nieces or nephews. Thankfully, the way our LTEE system is designed, we can directly measure the evolutionary fitness of our bacteria.

Measuring fitness in our system is a fairly simple, albeit a little time consuming, process. I take two different strains out of the freezer, and revive them separately. After I give them a couple of transfers to shake off the effect of the freezer and physiologically adjust to the normal growth media, I then take a small sample of each strain and put them together into the same flask. I immediately take out a sample of this culture, dilute it, and spread it onto a Petri dish. That dish is filled with a chemical mixture that allows cells to grow into colonies, and that causes one of the strains to become red, and the other strain to become light pink. This difference is totally irrelevant in the environment in which they’ve been evolving — it depends on chemicals that aren’t present there, so it’s neutral there — but it merely allows us to distinguish the two types from each other in a head-to-head competition. The rest of the culture goes into the incubator for 24 hours, at the end of which I again sample the culture, dilute it, and spread it on a Petri dish, and allow that to grow into colonies. By counting the number of colonies of each of the two types at the start and at the end of the competition, I can calculate how many generations each of the two strains underwent over the course of the competition, which is a very direct measure of their evolutionary fitness.

For my work, I’m systematically going through each of the populations and measuring fitness at dozens of points along the time course from the start to generation 50,000. Each of the populations is being compared to the ancestors, so I can see how much the population has changed from the start of the experiment. I’m also comparing pairs of populations through time, to see how much they’re diverging from each other. These data will allow me to address a set of interrelated questions about how these populations change over time. Are things still changing substantially after 50,000 generations, or has adaptation to this system essentially stopped? What mathematical function best predicts change in fitness over time? Can I accurately predict fitness at all? Is there enough consistency between populations to make useful general predictions, or are the predictions from one population so different from another that each case is completely different? Is the variance in fitness between populations more consistent with populations finding the same overall solution to their challenges (though at different rates), or with them finding very different ways to deal with their environment?

And that’s one of the great things about science: while I have some ideas of what I think might happen, for some of my questions I have no strong intuition at all. Research is all about finding out things that not only did you not know, but no one else did either. Many people like to ask why a question is even worth answering, so I’ve learned to find justifications for my work. For example, our ecological models tend to assume that species out in nature are already perfectly adapted to their environment unless there’s been a recent disturbance, but if I show that a population is still adapting to a simple environment after 50,000 generations — a longer time frame than almost any population will be subject to a consistent environment and no evolving prey or predators or diseases — that calls into question the assumption of everything already being perfectly adapted to its natural environment. But at the core, I’m really an ivory tower academic, and to me, the mere fact that no one knows the answer is reason enough to find out. Evolutionary biology, therefore, is a great place for someone like me. We’ve got enough of a theoretical understanding that there are good mathematical predictions for many things, but there are still a lot of experiments waiting to be done to either support or refute the existing theories.

For more information about Mike’s work, you can contact him at mwiser at msu dot edu.

This entry was posted in BEACON Researchers at Work and tagged , , , , , . Bookmark the permalink.

Comments are closed.