In 2004, the chicken became the first bird to have its genome fully sequenced. Its DNA revealed something odd—or rather, an odd lack of something. It was missing a gene called T1R2, which we and other mammals need to taste sweet foods. Chickens, it seemed, can’t taste sweets.
They aren’t alone. Maude Baldwin from Harvard University and Yasuka Toda from the University of Tokyo looked at the genomes of 10 different birds, from falcons to finches and ducks to doves. None of them had T1R2. Alligators do, and they’re some of the closest living relatives of birds. So at some point, as birds evolved from small dinosaurs, they lost their sweet tooth.
What about hummingbirds?
Hummingbirds feed largely on nectar, the sweet liquid that flowers produce. They love the stuff and the sweeter the better; they’ll actually reject flowers whose nectar isn’t sweet enough. They lack the T1R2 gene, but they can clearly taste sugar.
Baldwin and Toda have now discovered their workaround: they repurposed two other taste genes that are normally responsible for detecting savoury tastes. On a hummingbird’s tongue, these savoury sensors are sugar sensors too.
Most back-boned animals (vertebrates) have three taste genes: T1R1, T1R2, and T1R3. Each of these builds a protein of the same name, and the proteins combine in pairs. T1R2 and T1R3 fuse to create a sensor that recognises sugar molecules. When it sticks to one, it sets off a chain reaction that ends with a signal going to our brains—a signal that tells us we’ve just tasted something sweet. T1R3 can also fuse with T1R1 to create a sensor for amino acids—that’s what allows us to recognise savoury or “umami” tastes.
The team reasoned that if hummingbirds had lost T1R2, perhaps the T1R1-T1R3 savoury sensor could detect sugars instead. They tested the version in Anna’s hummingbird. They were right. It detects simple sugars like glucose and fructose and some sweeteners like sorbitol and erythritol. It can still detect amino acids*; it just acquired a new ability on top of that, sometime during the last 42 to 72 million years.
The two proteins changed dramatically in that time. Proteins are made of chains of amino acids, and the hummingbird versions of T1R1 and T1R3 had many different links compared to their chicken counterparts. To work out which of these changes were important, the team spliced the chicken and hummingbird versions together in different combinations, and tested their responses to sugars.
They identified 19 amino acids, dotted throughout one part of T1R3, which changed during hummingbird evolution, warped the shape of the proteins, and allowed them to stick to sugars as well as amino acids. Baldwin thinks there many more mutations were also important—after all, the team only looked at one small region of one of the two partners. This wasn’t a simple evolutionary step. It was an extremely complicated set of them.
Many animals have lost one or more of the three T1R genes. The giant panda has lost T1R1; savoury flavours are irrelevant when you only eat bamboo. Cats, Asian otters, spotted hyenas, sea lions, dolphins, and vampire bats all have broken versions of T1R2 and can’t taste sweets—possibly because they eat nothing but meat (or blood). Losing a taste gene isn’t unusual.
Gaining one, on the other hand, is very rare. Except for a few fish, we don’t know of any vertebrates that expanded on their basic trio. That’s a little strange, given that smell-related genes duplicate and diversifysmell-related genes duplicate and diversify very readily. These senses are fundamentally the same—both taste and smell involve proteins that detect chemicals in the environment. And yet, smell genes diversify like mad, but taste genes (or at least those for sweet and savoury) don’t at all. No one knows why. The genes don’t seem to gain new functions either—hummingbirds provide the first example of that. As far as we know, they’re unique in having regained a sweet tooth after their ancestors had lost theirs.
Why? That’s perhaps an easy question: nectar is a rich source of energy, and a sweet tooth allowed hummingbirds to successfully tap into it.
The better question might be: how? It’s easy to think that once hummingbirds started drinking nectar, their taste receptors would quickly have evolved to better sense the sugars they were after. But since the ancestral hummingbirds couldn’t taste sweets, why did they start drinking nectar at all? “This is a fantastic question and gets to the heart of a lot of debates I have,” says Baldwin. Here’s what she thinks happened.
The closest living relatives of hummingbirds are the swifts. Hummers use short, rotating wings to hover near flowers and drink nectar; swifts use long, pointed wings to scythe through open skies in search of insects. Thanks to some recently discovered fossils, we know that the both groups evolved from ancestors that had a mix of both traits—more swift-like in build, but with shorter hummingbird-sized wings. It almost certainly ate insects.
Baldwin thinks that some of these birds started spending time near flowers to catch insects that landed there. Perhaps they started to hover to snag these visitors more effectively (hummingbirds today still catch insects to supplement their diets). If this change brought the birds into regular contact with flowers, it would have given evolution something to work with. Now, individuals with changes in T1R1 and T1R3 might have been able to taste a little bit of sugar, and could have sipped some nearby nectar. Nectar means nutrients, so these sporadic sippers did better than their peers. That provided the evolutionary pressure for changing the proteins even further. “You don’t know how it begins,” says Baldwin. “But once it does, there’s selection to reinforce it and make it stronger.”
For now, this is just a story in need of confirmation—probably through finding more fossils of early hummingbirds. In the meantime, the team is already working with other scientists to see if birds like lorikeets and honeyeaters, which also drink nectar, have also regained their taste for sugar in similar ways.
They also want to know which order the 19 mutations that restored the hummingbird’s sweet tooth appeared in. Did they appear one at a time, or in batches? Are they all directly involved in detecting sugars? It’s unlikely. Some might have stabilised the proteins to allow mutations to persist. Some may have been unimportant at first, but paved the way for later adaptive changes.
There are ways of working this out. Several groups have worked out ways of reconstructing the amino acid sequences of ancient proteins, based on the sequences of their modern descendants. They can then build these ancestral molecules, resurrecting them millions of years after they were last seen. The team might try that for the hummingbird taste receptors, recreating the steps along the way from savoury to sweet-and-savoury. “This has the possibility of answering bigger questions in evolutionary biology,” says Baldwin.
* The hummingbirds can respond to both sweet and umami tastes, but it’s not clear if they can actually tell the difference between them. Perhaps it all just tastes the same to them.
Reference: Baldwin, Toda, Nakagita, O’Connell, Klasing, Misaka, Edwards & Liberles. 2014. Evolution of sweet taste perception in hummingbirds by transformation of the ancestral umami receptor. Science http://dx.doi.org/10.1126/science.1255097