Progress on Self-Cloning Crops
A Q&A with Mary Gehring, plant biologist at MIT’s Whitehead Institute, about synthetic apomixis.
My latest essay, about why it is difficult to genetically engineer rice, was published today in Works in Progress magazine. I hope you’ll read the piece and send me your thoughts. Lots of fun anecdotes didn’t make it into the text, including some notes on “synthetic apomixis,” a scientific quest to make plants that produce clones of themselves. If plant engineers figure this out, it could improve life for millions of the world’s poorest farmers. This Q&A with Mary Gehring, plant biologist at MIT, explains how it works and why it matters.
A typical farmer in India makes ₹15,000 per month, or about 180 U.S. dollars. They could make more money by growing hybrid crops, which grow faster and larger than anything else on the market. But the seeds are very expensive. They must be purchased from a large company, such as Syngenta, Bayer, or Biostadt India.
Hybrid rice (the staple crop for half of the world’s population) is often 10-20 percent bigger than even the best inbred strains. In the U.S., more than 99 percent of all corn is hybrid. But in low- or middle-income nations, less than half of farmers can afford to plant these special seeds, which are made by painstakingly taking pollen from one crop and using it to fertilize another.
Hybrid seeds are also only good for one year. They cannot be replanted for a second generation or they lose their vigor. A hybrid’s offspring, often, won’t grow as big because all the ‘good’ genes are washed out and lost.
But change is coming. We are now hurtling toward a future in which farmers could possibly buy hybrid seeds once, and then grow them forever, without relying on the big seed companies at all. It’s called synthetic apomixis.
A small number of plant biologists are making genetically engineered plants that produce clonal offspring without any fertilization. The plants don’t have sex. They just…clone themselves. If we figure this out, it would be possible to fix a plant’s traits in perpetuity, without worrying about messy genetics. It would be possible to make a hybrid plant once, and then propagate its traits forever. In the right hands, this could boost global crop yields and uplift millions of the world’s poorest. It would probably be the biggest agricultural breakthrough since the invention of hybrid rice in the 1970s.
Dandelions and hundreds of other plants naturally do apomixis, but none of the nutritional staples — like rice, corn, or soybeans — do. Mary Gehring, associate professor of plant biology at MIT’s Whitehead Institute, is trying to solve that. We recently sat down to talk about synthetic apomixis and the future of agriculture.
This interview has been edited. Additional notes, my own, are included in parentheses.
Niko McCarty: Hey, Dr. Gehring. It’s hard to understand what the impact of synthetic apomixis might be without first understanding how hybrid seeds are currently produced. Can you explain that process?
Mary Gehring: Many of the seeds that you buy from seed companies are hybrids. It’s easy to make hybrid seeds from some plants, but more difficult in others, depending on where the pollen or female parts are located. Soybeans tend to self-fertilize, so it’s relatively difficult to breed hybrid plants. Rice also self-fertilizes, so it’s tricky. They are fairly expensive to make because you need to make sterile males first, and then breed those with the other female parent, to ensure the plants don’t self-pollinate.
Niko: Do the big seed companies keep a portion of their crops each year to breed the next generation? How do they maintain all the parental lines?
Gehring: Yes, inbred lines have to be carefully maintained. You need to prevent cross-pollination, and you must be sure of the genetic purity of your line each time you plant it. Inbred lines are maintained by just crossing the same genotype to itself over and over again. And in some species, this will just happen through self-fertilization. A hybrid is created by crossing two different inbred lines.
Niko: Hybrid seeds produce larger plants because of something called hybrid vigor. But what does that mean?
Gehring: There are many, many inbred lines. These inbreds have no genetic variation; there’s no diversity. The two alleles are the same for every gene, so this genotype can be propagated indefinitely if you keep breeding that plant. And just like in animals, an inbred line is generally less vigorous than a heterozygote. Many, many years ago people realized that if you cross two inbreds, the resulting F1 will be more vigorous than either parent. That is hybrid vigor. Hybrid plants are generally larger, more robust, more resistant to stresses, and will typically have higher yields.
Niko: Got it.
Gehring: But then, if you take that hybrid plant and self-fertilize it, the progeny will lose this vigor. There'll be a segregation of traits, where some plants are vigorous, like the parent, but some will be more like the inbred grandparent, and some will be even worse than that. Uniformity is generally desired for agriculture. But if you take the progeny of a vigorous hybrid, you would just have this diversity of plant phenotypes, many of which would be negative.
Niko: Ahh, I see. So that’s why big agricultural companies keep generating these hybrids and selling the seeds?
Gehring: Exactly.
Niko: Now tell me about synthetic apomixis. How would it change how hybrid plants are made today?
Gehring: Right. The idea is this: If you could engineer seeds so that they reproduce asexually, then you could pass on the vigorous genotype year after year. You could breed a hybrid plant that's really good for a specific region, and then maintain its traits without having to generate that same hybrid over and over again.
Niko: Do seed companies want to stop research on apomixis? Are they concerned that it could slash into their monopoly on seed supplies?
Gehring: No, I mean, I think companies are actually interested in apomixis because it would make breeding programs so much faster. If you can transmit a genotype without recombination, then that is also good for seed companies because it takes so long to breed plants currently.
Niko: What is the current state of synthetic apomixis? What do we still have to figure out?
Gehring: Yeah, I mean, you've probably come across the recent work in rice. (Researchers altered two genes — called MiMe and BABYBOOM1 — in rice to make “hybrid plants that produce more than 95% of clonal seeds across multiple generations.) But there have been a couple of exciting papers over the past couple years.
There are three main components of apomixis that we need to figure out. The first step is to bypass meiosis so that you make a gamete and an egg cell that is diploid (maintains two copies of each chromosome.) It basically is like a mitotic cell rather than a product of meiosis.
Second, you have to get that diploid cell to make an embryo without fertilization, without the addition of sperm. The third component, which is probably the most complicated and where there's the least progress, is to make the embryo’s nutritive tissue, the endosperm, without fertilization.
Niko: Why is the endosperm important?
Gehring: Well, the endosperm is the part of the plant that we eat. And so the properties, the quality of endosperm, is very important. If your hybrid gives you a particular endosperm quality or phenotype, then you wanna be able to maintain that.
Niko: Got it. And so what have recent papers shown, at least for those first two steps?
Gehring: Well, this really exciting paper came out in rice this year. It showed that they could bypass meiosis in rice to make this diploid egg cell, and then they could induce parthenogenesis (develop an embryo from an unfertilized egg cell) in a hybrid plant. They showed that they could transmit that hybrid phenotype. But the third piece was missing. The endosperm was still a product of sexual fertilization.
Niko: The amazing thing about that paper, from what I understand, is that the engineered hybrid plants produced more than 95% clonal seeds, and they did it across three generations. Is that good enough for the seed companies? Do we need synthetic apomixis to be 100% efficient?
Gehring: I think 95% is amazing. It’s certainly farther than anything else, but you’d probably have to ask a seed company. If you buy a seed, and 95% of its offspring are clonal, is that good enough for farmers? I'm not an expert in agricultural economics, so I don’t know.
Niko: And the paper only did this in one specific type of rice, right? There are thousands of different rice cultivars grown around the world, and something like 3,000 species of rice in India alone. If we figure out synthetic apomixis in one plant, will it be possible to transfer it to another?
Gehring: Yeah, that's really important. Obviously the more lines you can do this in, the more impact it will have. But my guess is that, once we figure out the key genes for a particular species, it will probably be transferable to other genotypes within that species. I don't think that what we develop for rice, say, will work for soybean. I don't think there's going to be a single solution.
Niko: Do you think in 10 years we'll have this figured out?
Gehring: We will probably have it figured out in a laboratory context within 10 years or less. Getting it to farmers as a commercially viable product will take longer.
Niko: And what happens then? What impact would this have on poor farmers, who currently buy expensive seeds from agricultural monopolies?
Gehring: What I’d say is that this probably is not a magic bullet. The question is: Who has this technology? Will it be open access, or will it be in the hands of a company? That’s going to impact how this gets distributed throughout the world. In an ideal world, it would help poor farmers, but I don’t know if that will actually happen.
But synthetic apomixis does have the potential, still, to make agriculture require less inputs and be more resistant to climate variability, because of the vigorous nature of hybrids. Right now, hybrid seeds are too expensive to buy in some cases, right? But if poor farmers could get their hands on hybrid seeds with these traits, they could keep propagating them. So this could still help in places with fewer resources.
Thanks to Will Shaw and Alice Boo for helpful discussions.
Disclosure: The views expressed in this blog are entirely my own and do not represent the views of any company or university with which I am affiliated.