Innovations in berry varieties have been key to grower success over the years. If we look back at the origin of berry varieties, we must go back to wild plants that either were selected and became varieties or were used as parents in subsequent breeding. Formal breeding of most crops, including berries, began in the early 1900s after the rediscovery of Gregor Mendel’s findings on the inheritance of traits in peas in 1905.
As time progressed, more breeding programs developed across the U.S., mainly at universities and agricultural experiment stations and various USDA-ARS locations. These programs used traditional breeding techniques that included parent selection to provide complementary traits yielding progeny that were hoped to be improved.
The mechanics of breeding generally included pollen collection from male parents, with this pollen then being applied to emasculated flowers of the female parents. In this approach, parental selection depends largely on its pedigree, and more importantly, the characteristics of the parent. These parental characteristics are formally called the phenotype.
Molecular Level Breeding
We know little about the actual genes, or genotype, unless the trait’s inheritance is already known and can be observed in the plant. With this method, a tremendous amount of plant observation in the breeding program is needed to fully learn the potential of the plants as parents, in addition to considering the plant as a new variety for release.
We have been hearing about molecular applications in plant breeding for many years. The first activity was in the late 1970s to 1980s, when early DNA characterization was first performed mainly to determine genetic differences in plants at the molecular level. It was interesting to learn where some varieties came from (their parents), such as some of the popular wine grapes.
Not long after this, transformation technology emerged, leading to the hot topic of genetically modified organisms (GMOs). We have heard a lot about GMOs in recent years, though generally in berries we don’t talk much about this innovation; in fact, for my discussion, I want to set it aside. Molecular research has provided much more to plant breeding than to GMOs, and this technology is now making its way into berry breeding in a substantial way.
With the major row crops such as corn and soybeans, molecular methods are routine in breeding. With berries, as well as all other fruits, we are behind in this area — mainly due to much less planting acreage and lower overall crop value to the national economy.
As a “traditional” plant breeder, my career has not included molecular techniques. I have relied on the older methods, with most decisions based on the phenotype and very little emphasis on the genotype present in the plant. However, times are changing, and I want to share a little on this topic to offer a further understanding of what this is all about.
Sunshine State Strawberries
Florida has long been a strawberry-producing state. Back in the 1980s and early 1990s, the industry relied primarily on California-developed varieties. Dr. Craig Chandler came to the University of Florida, based in the Tampa area, in the late 1980s. He continued to evaluate selections made prior to his arrival, and from these, he released ‘Sweet Charlie’ in 1992. This was the first variety from the Florida program to have a big impact on the industry, providing a flavorful, early variety with disease resistance. His release of ‘Festival’ in 2000 led to the Florida strawberry industry doubling in size to 10,000 acres.
Dr. Vance Whitaker took over the breeding program in 2009. He has continued this vigorous program with subsequent releases of several varieties, notably ‘Florida Brilliance,’ an exceptional variety for the region. Whitaker continued the traditional breeding approach but also has added newer molecular techniques. Among the berry-breeding community, I consider him to be very successful in this, and I want to use his approach and success to expand an understanding of what this really means to berry variety improvement.
The first area that Whitaker focused on in the molecular area was the development of molecular markers for key traits of importance that are controlled by one or two genes, particularly disease resistance (anthracnose, Phytophthora crown rot and bacterial angular leaf spot) and perpetual flowering. A key individual who works with Vance is Dr. Seonghee Lee, who joined the program about four years ago. Together they have achieved something almost all breeders desire — the use of molecular markers to screen seedlings when they are very small plants. This saves a tremendous amount of time and effort. A breeder is able to identify and discard the undesirable plants just after the seeds germinate.
Whitaker and his team aim to attain about 250,000 seeds from crosses each year, and from this produce 40,000-50,000 seedlings to subject to marker screening. Through a tedious process, they take a very small piece of leaf from these seedlings, analyze the DNA, and then sort the plants out to retain those that have the desired genes for the screened traits.
Plant breeding is all about numbers. In this procedure, instead of evaluating 10,000 to 12,000 unscreened seedlings in the field, he plants the seedlings that have been evaluated using this system. In other words, he is increasing his chances of success by three to five times. He throws out a lot of trash before he goes to the field!
This is a dream of all breeders — to increase the number of valuable genetic recombinants to select from. Key points of success here are committed colleagues like Dr. Lee who help carry out these procedures, plus royalty proceeds from the success of the Florida varieties that pay for this expensive exercise.
Whitaker’s innovations do not stop here. He also is conducting a procedure called genomic selection. It has taken a while for me to fully understand how to apply this to breeding.
“We collect data on fruit size, yield, soluble solids, and other traits on about 450 selections in our program every year — anything that is worth measuring objectively,” Vance explains. “We also collect leaves from these plants for DNA analysis (genotyping).”
From the phenotypic and genotypic data, Whitaker can calculate breeding values for his selections for a more precise prediction of their success as parents. He explains that when his team began replacing pedigrees with the marker data in these analyses, estimating the parental value of selections became more accurate. He can also predict the parental performance of selections that don’t yet have any phenotypic data by training a genomic selection model using data from previous year’s trials.
A statistical model is trained with genome-wide marker data and phenotypic data from the previous year’s trials, and then predictions are made for brand new selections that only have marker data. He then uses some of the new selections with high predicted performance in crosses right away, rather than having to wait two to three years for further phenotypic observations. He feels he can reduce the breeding cycle by one third using this method for some crosses.
Anything that speeds breeding is critical, particularly for perennial crops.
I came away from my visit with Whitaker with a much greater understanding of his process and procedures and feel I have truly seen an optimum example of the incorporation of molecular methods in berry breeding progress. We have many new technologies on the horizon, so let’s get ready to see them pay off for our berry industries.