100 years: Plant breeder reflects on past, future innovations

JOHNSTON, Iowa — It was early in the biotech era when Dave Bubeck began his career in 1991 as a corn breeder for the historic Northrup King seed brand that became part of Novartis.

Bubeck, now Corteva Agriscience research director and Global Breeding Alliances lead, joined Pioneer Hi-Bred in February 1998 as corn product development research coordinator and became research director two years later.

During a break in the annual Corteva Media Day, Bubeck reflected on the 100-year history of Pioneer and the future of plant breeding technology.

What are your thoughts about the 100th anniversary of Pioneer?

Bubeck: It’s wonderful to be here and be part of the celebration of the 100-year recognition. It’s a rather novel experience. Not a lot of corporations make it to a 100. It’s unique, and it’s unique in agriculture to be around for this long as a corporation.

My background is corn breeding, and the roots of the very beginning of Pioneer was hybrid corn breeding and delivery of hybrid corn to the marketplace. So, it goes back to the very beginning.

The other really fun thing for me personally is I grew up on a farm in north-central Iowa, and we grew Pioneer products. So, it’s not entirely coincidence that I ended up here. I really was hoping to eventually after grad school in plant breeding get a job at Pioneer Hi-Bred.

What do you think about the history of corn breeding, dating back thousands of years, when they were selecting characteristics from plants to produce better plants?

Bubeck: I wouldn’t have called it corn breeding. It was human intervention on selecting for plants that performed better than their prior ones.

It is interesting to think about in the history of the corn lineup that we looked at going back to teosinte, I think I could say all plant species that are cultivated today have a history that connects back to a wild progenitor, a wild ancestor of some sort.

When you look at the brassicas, and a lot of people use this example, cauliflower, radishes, broccoli, there’s a tremendous diversity in the brassicas. There are different species involved there, but they all have wild progenitors, and corn has a wild progenitor, too.

Among the scientists in the academics, even when I came into graduate school in the late 1980s, there was a lot of debate about what was the true origin of Zea mays or the cultivated corn that we commercialize today.

The majority thought that it was teosinte, but there were some debates about, was it Zea diploperennis or there’s a whole plethora, a whole host of other theories of where domesticated corn came from?

Everybody believes this didn’t happen by natural selection, by microevolution, or evolutionary patterns. It actually happened by human intervention, selecting something that was better than just teosinte itself.

Corn enjoys a wonderful history of diversity, though, beyond just teosinte. There are somewhere around 300 races of maize.

In the United States, we largely have leveraged the diversity of two of those races. So, when you think long-term about what could be done, there’s a tremendously untapped commercialization of the diversity that exists in the corn genome.

We don’t know what we don’t know today, things about how our diseases, how our plant pathogens, how our corn pathogens gonna evolve over time. I firmly believe that technology will enable us to protect for the long-term against those disease patterns that happen.

In fact, one of the agronomists talked to me this morning about tar spot a little bit. Tar spot for corn is largely a tropical disease.

As a long-term corn breeder, I would have never predicted that tar spot would come into the upper Midwest, upper part of the Corn Belt, and have an impact, but it has.

The pathogens are evolving, changing all the time, and so as we incorporate research efforts, we have to be ready for those sorts of things.

What were some of the genetics over the years that were real game changers?

Bubeck: I’ll just note a few of the game changers of technology along the way. When Pioneer started breeding corn aggressively, Pioneer 3780 was a 1971 commercial release, and for more than a decade it sold 1 million units or more per year. I don’t remember exactly where it peaked, but several million units at the peak of its lifecycle.

Other than good, sound testing and selection in the field, there was heavy use of statistical, experimental design in the field, the small plots, replication, being able to identify and select the best, data information management, recordkeeping.

Those were all important things to make that selection, but other than rigorous statistical capabilities, there were not a lot of other technologies at that point that came alongside.

Yet, breeders were making genetic gain. All the way from the 1930s on, they made tremendous genetic gain.

The lineup immediately after 3780, 3394, was released in 1991. That was a game changer for yield. It brought a lot of things together that had not been done before, and part of the reason for that is the breeding for scale. 3780 enabled Pioneer to start growing their breeding programs.

They invested more money and capital in research and development in their corn breeding programs. They had more corn breeding programs across the United States, began to initially expand outside the U.S. and breed in other countries where corn was important. That breeding the scale really brought on genetics like 3394.

So, that was a lot about the size and number of things. Even at that point, it still wasn’t about technology.

GMOs then came in 1996. Beyond the ability to protect for insects, and have convenient means of herbicide control in crop plants, use of glyphosate and other chemistries, there were also genomic capabilities that came in.

The human genome was published in the year 2000. A lot of those DNA-based and genomic technologies also were available in plants.

Post-2000 we had the ability to improve the resolution of our genome knowledge. So, we’re increasing the density of DNA-based molecular marker systems.

The value of that to breeders was that now we could begin to predict what the performance of an inbred or hybrid would be based on its base genetics. That molecular footprint, call it DNA fingerprint, gave us a lot of ability to predict.

We now deploy advanced breeding technologies like genome editing, and because of that knowledge of the genome that we understood now for the last 25 years, we improved our biological resolution of the understanding of how the genome works.

How did the breeding change from when Pioneer founder Henry A. Wallace selected male and female plants with certain characteristics to improve the corn plant through utilizing the DNA-based molecular marker systems?

Bubeck: The things that people would say, even academically, about Henry Wallace and the very first corn-breeder he hired, Raymond Baker, was that plant breeding was an art and a science.

If you ask me today, is it still an art and still a science? Well, there is still some artistic component of it. Some breeders have a better eye for what they’re looking at than others. There still is an art component.

But if you said in 1929 that art was 50% and science was 50%, today it’s maybe more 10% art and 90% science. The science has become a lot more contributed to from a lot of different areas.

Of course, genetics is core to plant breeding. Statistics is core. Quantitative genetics is kind of the combination of those two.

Molecular biology is important because we need to understand the DNA contribution of what we’re looking at. Molecular biology is important.

Entomology is important because we want to be able to manage insect right out of a farmer’s field. Disease pressure, plant pathology is a real critical thing to add to the components of what we do.

There’s engineering, as well. A question that we’ve asked is any particular phenotypic note, a characteristic that a human would collect from the field, could it be collected in a digital and automated way?

Remote sensing, aerial-based, or ground-based, anything that the human eye can see and score, or measure, could it be measured in an automated way?

What do you foresee may happen in the next 50 years in plant breeding? What would you like to see happen?

Bubeck: I would say a lot of the intense technology. If you look at the last 100 years for Pioneer Hi-Bred, the escalation and the rate of change of technology, not only discovery of technologies, but then the eventual implementation of those technologies has become accelerated in the last 30 years.

I’ve been in the industry 35 years. There’s been more change in how breeders breed in the last 35 years than the first 65 of Pioneer’s existence, and so the last one-third was really critical.

What’s going to happen over the next 30? I don’t see the rate of change going down. It’s going to continue to accelerate.

We have to continue to hire scientists that are very forward thinking, willing to try a lot of things, be highly innovative and willing to make change.

I think things like genome editing will become even more important, how we accelerate the speed of breeding through generational artificial intelligence is going to be core to how we think about things going forward.

And even using AI to determine, is this new technology justified to implement, yes or no? So, I think a lot of these multiple technologies coming together.

I think what will really drive us into what I’ll term as a post-genomic era will be, can we create new variation that is beyond positive variation that’s beyond the performance of what we would normally do with today’s technologies?

Less than 1% of what we create ever becomes a commercial product. The inevitable outcome of that is that we’re narrowing up the germplasm diversity.

So, in order to continue to make genetic gains for characteristic traits that our customers care about, as well as fighting against these disease pathogens as they emerge and change, that we have to stay ahead of that.

I think the question will be, can we really use the genome knowledge to create plant performance that’s beyond what we could do with the current state of technology?