Yesterday scientitsts announced in a quartet of papers in the journal Science that the draft genome of bread wheat -- Triticum aestivum -- had been decoded and mapped. Together with barley, wheat is the crop on which civilization rose in the Fertile Crescent and Egypt some 10,000 years ago. With theses grasses and the help of wild yeast, humans created bread and beer and have rarely looked back (Prohibition and the current gluten-free fad being notable exceptions). I covered the story over at National Geographic.
The content of the genome was not a surprise, Robert Bowden, supervisory research plant pathologist at the U.S. Department of Agriculture's Hard Winter Wheat Genetics Research Unit in Manhattan, Kansas, told me. What was unexpected, he said, was what the genome told scientists about the evolution of wheat, as detailed in a second paper released concurrently with the genome by Marcussen et al.
In the genome, "we found pretty much what were expecting," Bowden said. "The second paper was the one was the one that was kind of shocking, because we thought we understood a lot abut the evolution of wheat.”
Indeed, scientists did understand a lot about the evolution of wheat. But they didn't know everything, hampered by a lack of wheat fossils and by the intractably large and repetitive wheat genome, which had resisted sequencing. You can read more about that story over at Nat Geo.
For example, scientists had known for some time that wheat is a triple "polyploid", a hybrid of three parent species of wheat who through two accidents of biology had merged two genomes into one to produce emmer or durum wheat (used primarily for pasta today, though probably for different purposes by the ancients), and then two into three to produce bread wheat with a genome three times as big as that of its ancestral genomes. You can read more about this process in a blog post I wrote about polyploidy in plants here.
But without a map of the genome, answering questions about how the three parent species of wheat were related to each other (they were presumably close relatives) was difficult to impossible. Then along came the draft wheat genome, and suddenly lots of things were possible.
Thomas Marcussen, Odd-Arne Olsen, and Simen Sandve of the Norwegian University of Life Sciences and their colleagues in Norway, Germany, and the UK initially set out to date the two known polyploidy events and find out how the three wheat parents were related to one another – a topic that had been controversial for some time due to the fossil and genome void. They expected a bifurcating tree in which two of the parent species -- they were not sure which two -- were more closely related to each other than the third. Instead, they discovered a more complex situation. Instead of two hybridizations in wheat's past, there now appeared to be three.
“We really couldn't make a model that look liked a normal bifurcting tree based on our data,” Sandve said. “We had to try to make it into a network to make an evolutionary model that would fit the data.”
Wheat parents Triticum urartu and Aegilops speltoides were equally closely related to Parent #3 Aegilops tauschii and more closely related to A. tauschii than to each other. That could only make sense if the ancestors of T. urartu and A. speltoides had hybridized to produce the ancestor of A. tauschii via a process called monoploid hybridization. This type of hybridization can only take place between two very closely related species. It happens when a normal egg cell from one species meets a normal sperm cell from another and the species are not so distantly related that the sperm can't fertilize the egg. A familiar example is the production of a mule from the mating of a donkey to a horse -- two different species. In that case, the mule is usually infertile, but in the case of wheat, A. tauschii was evidently good to go.
In short, the wheat family tree is beginning to look distressingly similar to the Hapsburgs'.
The discovery fits with other data in plants like sunflowers that seem to show that this type of direct hybridization that seems to have produced A. tauschii – called homoploid hybridization – may be more common in plants than previously thought, Sandve said.
Bowden said he was taken aback by these results. “Instead of there being two speciation by hybridization events in the evolution of wheat, there's three, which is shocking. I don't think anybody was expecting that,” he said. “If it's true, and I think it is true, it's a really really unexpected result and shows the power of this method of analysis and leveraging all the data that [the draft genome] produced.”
He said they would have to rethink how they approached mining the A. tauschii genome for useful traits for new wheat varieties. Although it no doubt still contains valuable traits, it is no where near as old or independent of the T. urartu and A. speltoides subgenomes as they had thought, the said.
In addition, from an evolutionary perspective, he said, it's intriguing that the ancestral T. urartu and A. speltoides genomes hybridized twice in two different ways – once to make homoploid hybrid A. tauschii, and once in polyploid fashion to produce emmer/durum wheat (T. turgidum).
“It's very interesting that these two different kinds of event happened with the same two species to start with,” he said. “and it also says 'Wow, this is not very rare. It happened three times just in the evolution of wheat!'”