I’ve been to fancy restaurants where dozens of cooks toil away in the kitchen. They have their own quirks and varying degrees of skill but they all produce the same plates of food—those dictated by the head chef. He or she ensures that their recipes are followed to exacting specifications. I’ve sometimes wondered what would happen if you deposed the head chef, and let the others loose to play around with the recipes. Some would produce culinary disasters, but a precious few would probably create gastronomic triumphs.
That’s roughly what a scientist named Conrad Waddington did in the 1950s, but rather than deposing chefs, he stressed some fruit fly pupae with bursts of heat or chemicals. When the adult flies emerged, they had an array of weird features, like broken veins in their wings or extra body segments.
The genes for these odd traits hadn’t arisen out of the blue; they were already there in the flies, but their effects were somehow being masked. They were like the underling cooks in the fancy kitchen, committed to enacting the same recipes regardless of their own variation. When Waddington heated or chemically treated the flies, he lifted this repression and allowed their hidden genetic variation to manifest as physical variations.
But how does genetic variation get first concealed and then later released? What’s the molecular equivalent of the draconian head chef?
Susan Lindquist at MIT found a candidate four decades after Waddington’s experiments. It’s a protein called Hsp90. When she depleted it in flies, they grew up with weird features from extra hairs to severe deformities, just like in Waddington’s studies. These changes weren’t caused by fresh mutations, but pre-existing ones that had been unmasked in Hsp90’s absence.
As I wrote for Scientific American earlier this year, Hsp90 helps other proteins to fold in the right way, and stops them from unfolding at high temperatures. This allows the proteins to tolerate and accumulate mutations that might otherwise catastrophically distort their shapes. If the temperature goes up, Hsp90 can’t meet the demand for its services. Suddenly, proteins start to fold in a variety of different ways.
For this reason, Lindquist describes Hsp90 as an evolutionary capacitor, after the devices that store and release electrical charge. It does the same, but for genetic variation.
Lindquist’s team have gone on to show that Hsp90 can store cryptic variation in many other species, including microbes, worms, mustard cress plants, and yeast. Her former postdoc Daniel Jarosz found that this capacitor conceals a whopping fifth of all the variation it the yeast genome.
It looked like Hsp90 could be a major driving force in evolution, allowing genetic variation to build up behind the scenes and unleashing it in one burst when times are tough. Suddenly, different versions of the same gene (alleles) that were once corralled down the same path can produce a smorgasbord of new traits. Natural selection gets a bonanza of physical variation to act upon.
But there were three niggling problems.
First: virtually all of the traits that Hsp90 unleashed were defective rather than adaptive—broken-veined flies are hardly evolution’s next breakout stars. Second: most of these experiments involved lab-bred creatures, so no one knew how important Hsp90 is in the wild. Finally: what is the wild equivalent of the experimental heat shocks? “It’s hard to imagine a small browsing dawn horse encountering a particularly hot summer and instantly giving birth to the modern large grazing horse,” says Clifford Tabin from Harvard Medical School.
Lindquist recognised these problems, and she started searching for a genuine case where Hsp90 unshackled some cryptic variation, which led to the evolution of adaptive traits. She asked Tabin, who had studied the development of many unusual animals, from Darwin’s finches to hopping rodents. He thought they should look at species that suddenly find themselves in a completely new environment. He thought they should look at blind cavefish.
Millennia ago, several groups of Mexican tetra—a popular aquarium fish—swam into dark caves, and eventually evolved to be blind. Today, their embryos are born with eyes that gradually waste away, leaving hollow orbits. Tabin and Lindquist’s teams, headed by postdoc Nick Rohner, have now compiled a strong case that Hsp90 was involved in this change.
They exposed the larvae of sighted, surface-dwelling tetras to a chemical that blocks Hsp90. Some larvae ended up with much larger eyes than are ever found in nature, while others had much smaller ones. Overall, the range of sizes shot up by 83 percent. So, the surface fish have a lot of cryptic variation in eye size. The team deliberately exposed this variation by with their Hsp90-blocking drug, but they think that entering dark caves would have achieved the same effect.
They examined the water in one of the caves where the blind fish live, and showed that it’s much less electrically conductive than nearby surface waters, thanks to the calcium carbonate that leaches into it from the surrounding rocks. This low conductivity can mess with a fish’s ability to control the ions in its body, creating physical stress on a par with what Waddington’s heat-shocked flies experienced. Indeed, when Rohner raised surface tetras in water with low conductivity, they developed a much larger variety of eye sizes.
So, when these fish swam into Mexican caves, they would have experienced conditions that unleashed the cryptic variation stored by Hsp90. Some would have grown up with unusually large eyes, and others with unusually small ones.
It takes energy to maintain functioning eyes—energy that’s wasted in pitch-black caves, where eyes are of little use. In these dark worlds, the small-eyed fish would probably have done better than their big-eyed peers. Rohner simulated this evolutionary pressure by taking the Hsp90-blocked fish with the smallest eyes and breeding them with normal peers. He found that their offspring, despite having fully functioning Hsp90, still had smaller-than-average eyes.
Here’s why. The initial generation had different alleles for the genes that control eye size, and blocking Hsp90 unmasked this variation. Now, Rohner could select for the small-eye alleles by picking the small-eyed fish for his breeding experiment. These alleles became more common in the next generation of fish, which developed smaller eyes on average.
If selection acts consistently enough, you’d expect the once-cryptic variation to slowly dwindle away. Only alleles that produce advantageous traits—in this case, small eyes—would remain in the population.
That’s exactly what Rohner found when he repeated his experiment in the blind cavefish. The adults lack eyes, but they still have hollow orbits where their eyes should be. And when Rohner blocked their Hsp90, the size of these orbits stayed the same. The cryptic variation in their eye size had already been exposed earlier in their evolutionary history, and they were only left with the alleles for small eyes.
Waddington did the same thing with his lab-bred flies, almost 60 years ago. At first, he needed heat or chemicals to produce the weird individuals. As he bred these mutants with each other, he eventually ended up with generations that developed their odd features under normal conditions. He had uncovered cryptic variation, and then selected for a specific set of exposed alleles. Now, Rohner has finally found a natural example of the same process.
“It’s an amazing example that proves the importance of cryptic variation in a fascinating natural context,” says Joanna Masel from the University of Arizona. “It shows that cryptic genetic variation is important, and that Hsp90 depletion can uncover that variation, but it doesn’t decisively prove that Hsp90 was the historical means by which the natural adaptation occurred.”
Joshua Plotkin from the University of Pennsylvania is more enthused. “It’s a terrific study that, at long last, demonstrates an important role for cryptic variation in the evolution of natural populations of the ‘everyday’ organisms that we can easily see in front of us,” he says. “I cannot imagine a more beautiful system in which to study this question than Mexican cavefish.”
Of course, this is just one example. We still don’t know whether Hsp90, by concealing and exposing genetic variation, steers the evolution of many species, or just a few. The process seems to rely on sudden environmental changes, and you’d certainly get those in places like deep-sea vents and glacial lakes, or at microscopic scales. But for bigger organisms, “most evolution occurs in gradually changing environments,” says Tabin. Are the blind cavefish the exceptions that prove the rule, or merely the first good example of a common phenomenon?
Finally, remember that this style of evolution is no different to what Darwin envisaged centuries ago. You still have variation that get passed from one generation to the next, and produces traits that affect the success of their owners. The only difference here is that the variation already exists in a hidden form, and gets exposed to natural selection all at once. As Plotkin once said to me: “It’s standard vanilla evolution.” It’s just faster.
Reference: Rohner, Jarosz, Kowalko, Yoshizawa, Jeffery, Borowsky, Lindquist & Tabin. 2013. Cryptic Variation in Morphological Evolution: HSP90 as a Capacitor for Loss of Eyes in Cavefish. Science http://dx.doi.org/10.1126/science.1240276.