Still sifting through genes, but…

Plant breeding has always been about testing to see how one gene interacts with another, but today it can be done much more quickly

Today’s new generation of plant breeders are often called “gene jockeys,” although they’re actually more like cowboys rounding up “genotypes” into a common corral called a “genome.” Then they look for other genomes to add to the corral so they can improve the herd.

Previous generations of plant breeders did the same thing, but they relied more on trial and error. Today’s breeders know much more about the workings of the genes, how they affect each other and how they affect the performance of the plant. They’ve gone beyond genetics and into the realm of genomics.

Related Articles

“The genomic technologies that are coming online are quite exciting,” says University of Saskatchewan wheat breeder Curtis Pozniak. “It’s giving us an endless supply of new DNA markers that we can use to tag important genes and then follow those within the breeding program.”

Farmers were the first plant breeders and have have been selecting and reseeding the best plants for thousands of years. Charles Darwin observed how the traits of the parents shaped their offspring, but couldn’t explain how it happened. Then, 30 years after Darwin published The Origin of Species, England’s John Garton became perhaps the first commercial plant breeder, showing that many crops were self-pollinating, and inventing the process of crossbreeding. The Garton Brothers’ seed company released Abundance Oat, their first commercial variety, in 1892.

John Garton is credited as the first scientist to show that the common grain crops and many other plants are self-fertilizing. He also invented the process of multiple cross-fertilization of crop plants.
photo: Supplied

Then, two generations later, Crick and Watson described the structure of DNA in 1953, and we began to learn the mechanics of plant breeding.

Precision tools

“Now there are tools that help you make better and more efficient crosses so you can focus your resources or impose selection earlier, based on lab data,” explains Rob Duncan, a canola breeder at the University of Manitoba.

Old-school breeding meant making controlled crosses of several sets of parents, observing the agronomics as the progeny grew in plots, analyzing the seeds produced and then discarding the huge numbers of failures. Now Duncan can look for certain genes or groups of genes in the lab and make his first round of decisions based on what he finds there. He doesn’t need to go to the field, which was essentially using “visual phenotyping” to see if the genes were likely there. Today, molecular tools prove they are.

One of these tools is the SNP chip (called the snip chip) — a glass or silicon chip with thousands of tiny portals. Each portal has a strand of synthetically produced DNA, not the full helix that you see in the pictures. When breeders want to know what genes are in the parent plants, they tease out the DNA, dissolve it into a solution and drop it onto the SNP chip. The strands of DNA from the plant seek out and bond to the compatible strands permanently etched into the chip.

A rendering of the “double helix” structure of DNA discovered by Watson and Crick in 1953. Plant breeders can now isolate strands of the structure and identify which contain desirable characteristics.

Now they have DNA that can be analyzed for “marker genes” with desired characteristics.

“Over the last number of years we’ve developed the 60 k brassica SNP chip and that’s 60,000 markers on one chip,” Duncan says. “Now we can select for thousands of markers across the whole genome of whatever cross you’re talking about and select the best combination on a genome-wide basis.”

That means that instead of painstakingly making thousands of crosses and observing the result, breeders in the lab can narrow their search to a few likely combinations.

“It’s not going to be accurate every time but it allows you to focus your resources,” says Duncan. “Rather than taking all of them to the field to evaluate them, you can do it in a more efficient manner in the lab.”

Still the final test

However, the field trial is still the last word on breed evaluation. Even though the understanding of genetics and genomics is as good as it’s ever been, the test plot is where the plant encounters the real world.

“Linking the genetic differences between cultivars and breeding lines to actual phenotypes is the next frontier,” Pozniak says. “We need to take to take full advantage of the genomic technology and using genomic tools really revolves around using marker-assisted selection.”

The marker provides a general idea of where the gene is located. If breeders understand that certain genes, or combinations of genes, give a plant characteristics such as rust resistance, then they know to look for those markers in the parent plant’s genome.

Breeders are now finding some traits are actually influenced by combinations of genes found in different locations throughout the genome. To make things more complicated, many crop plants have multiple sets of chromosomes. Bread wheat, for example has six.

As in so many other applications, computers can ease the job of sifting through data.

“The cost of high-throughput genotyping and phenotyping is going to improve and these tools are going to become more efficient and more cost effective and efficient,” Duncan says. “It’s going to provide the breeder with quicker, better, cheaper information to predict some of those best parental combinations. Our ability to use gene editing where we have a more controlled ability to edit or modify genomes is going to improve as well as our ability to evaluate how those genes are actually expressed.”

But the ultimate test is still in real soil under real weather conditions.

“So there’s still a lot of boots in the field in plant breeding and I don’t think that will ever go away,” Pozniak says. “Plant breeding is really about sifting through the massive numbers of individuals that we need to look at to identify the vast gene combinations. The technologies we’re using are really about helping us sift through those numbers more efficiently.”

About the author

Comments

explore

Stories from our other publications