The four-year, $14.4 million RosBREED plant breeding project, funded in 2009 under the Specialty Crop Research Initiative, is coming to a new stage, winding down, wrapping up, summarizing, and telling people what’s been accomplished. But it’s also ramping up to convince the public to continue the project—because there’s still more to come from supporting this kind of intensive, collaborative research.

The project generated 14 new PhDs, 12 of them young plant breeder trainees looking for jobs in which they can carry the work forward. These young Turks fully understand what the project was about and are able to use what it accomplished, says Dr. John Clark, the peach breeder at University of Arkansas and a mentor to three of them.

The project’s goal was to enable marker-assisted breeding in plants of the Rosaceae family, which includes apples, peaches, sweet and tart cherries, and strawberries. These plants share many genes. What makes an apple red makes cherries, peaches, and strawberries red.

Drs. Jim Luby and Kate Evans, apple breeders from the University of Minnesota and Washington State University, respectively, made a brief presentation at the RosBREED annual get-together at Michigan State University in late May. They now know what parents to choose if they want yellow apples—or if they want to avoid them—and what crosses are likely to produce high or low acidity.

They closed in on those genes that give Honeycrisp that wonderful, crisp texture, which, Evans said, “raised the bar” for breeders. It’s almost mandatory now that new apples have that crunch. It’s associated with a group of alleles at the Ma locus.

Luby said the work enables breeders to tell whether a seedling carries a desirable trait without waiting seven years to see the fruit. DNA testing is remarkably cheap—maybe $5 a sample, he said—but it’s unlikely that thousands of seedlings will be tested. The tool will be best used in choosing parents for making crosses, knowing that many of the seedling offspring will carry desirable traits.

Dr. Amy Iezzoni, the Michigan State University tart cherry breeder, is overall director of the massive project that includes breeders from South Carolina, Arkansas, Texas, California, New York, Washington, Minnesota, Michigan, Oregon, and New Hampshire and European partners from the Netherlands as well.

Tart cherry impact

As the lone tart cherry breeder in the nation, she thinks she greatly benefitted from the project. “If it wasn’t for RosBREED, where would tart cherries be?” she asked. Her goal is to create a clear-juice tart cherry, like Montmorency, that has large fruit and is resistant to cherry leaf spot. She has found that gene, but the fruit that came with it isn’t desirable.

DNA tests were developed to predict cherry pitmand fruit size, cherry skin and flesh color, ability to self-pollinate, as well as resistance to cherry leaf spot.

“These genetic tests enable tart cherry breeders to determine the best parents to combine and the best seedlings to advance, reducing the need to grow out and sort through hundreds of seedlings with unacceptable fruit quality,” she said.

Some results: “About 30 percent of seedlings from one experiment were discarded prior to field planting because DNA markers identified them as derived from unintended parentage.

“Twenty-five crosses were designed based on their increased potential for transmitting red skin and light red clear juice color to the progeny.”

Sweet cherry impact

“Genetic tests have been developed to predict cherry skin and flesh color, fruit size, firmness, and ability to self-pollinate,” according to an “impact statement” developed about progress in sweet cherry breeding.

Washington State University’s Dr. Nnadozie Oraguzie reported that as a result of genetic testing of 7,000 seedlings, 42 were chosen as parents based on their genetic abilities to pass on desirable traits, 146 crosses were made using them, 16 advanced selections have been examined more closely, and 3,735 seedlings were discarded prior to field planting based on the prediction that fruit would be soft or flowers self-sterile.

“Without marker-assisted breeding, approximately 3.5 acres of sweet cherry seedlings with no commercial potential would have been planted and evaluated,” he said. “The projected cost to the Washington State ­University program of $75,000 was avoided.”


Four peach breeding programs came together under the RosBREED umbrella, bringing in processing peaches from California and fresh market peaches from South Carolina, Arkansas, and Texas.

At Clemson University in South Carolina, Dr. Ksenija Gasic worked to develop genetic tests to predict peach maturity date and fruit quality traits such as texture, flavor, size, and color. She tested 310 peach selections, choosing 24 to use as parents based on their potential to transmit freestone, melting, or slow-softening flesh type, high blush, and bacterial spot tolerance. Forty crosses were made using them.

At the University of Arkansas, John Clark had similar goals and used new DNA tests to select four cultivars and advance 16 selections and choose 136 seedlings for field testing. He believes some real breakthroughs were made.

“New peach flesh types were discovered using a genetic test for detecting the two common melting and non-melting types,” he said. “One of the new types, the ‘non-softening’ flesh type, maintains its firmness and quality for three or more weeks in post-harvest storage and the fruit is more resistant to mechanical damage.

“Another discovery, the ‘slow-melting’ type, also maintains its firmness longer, but upon full ripeness, melts and reaches the same texture as melting types.”

At Texas A & M, breeder Dr. David Byrne also focused on finding endoPG loci, those genes that deal with flesh type. Researchers have located genes for traits called soft, ­melting, crisp-firm, stony hard, and rubbery, as well as clingstone and freestone, and have developed DNA tests for ­knowing if a seedling carries them.

At the University of California, Davis, Tom Gradziel focuses on peaches for processing, most of which are non-melting clingstones. The processing industry has used a very narrow base of genetics, ­making it difficult to find new genetic solutions. The RosBREED work helps broaden the base.


DNA tests have been developed to predict apple skin color and amount, firmness, crispness, juiciness, acidity, and storability.

In the Washington State University breeding program, parents are now chosen based on their genetic potential, and 23 crosses were designed based on this information.

The program saved some $98,000 in 2011 and 2012 by discarding 7,000 seedlings prior to field planting based on the prediction they would have inferior texture and storability. In 2013, 9,000 seedlings were screened using DNA testing and genetic markers, according to Kate Evans.

Marker-assisted breeding, or MAB, helped the University of Minnesota breeders choose sets of parents with predicted potential to transmit skin color, bitter pit, titratable acidity, crispness, storage potential related to ethylene production, and apple scab resistance.

“Between May and July this year, we will screen 1,500 seedlings for markers for crispness, skin color, storage potential, and apple scab, at a projected cost of about $5,000,” Luby said. “We anticipate culling at least 1,000 seedlings.

Our cost for carrying a seedling to fruiting is about $20, so this culling will yield a net savings of about $15,000.”

Dr. Susan Brown, the apple breeder at Cornell University, said that “know­ledge of key traits and their inheritance has changed our choices in the selection of parents and should be reflected in many less undesirable seedlings and a greater chance of genetic gain, resulting in high quality advanced selections and tomorrow’s future varieties.”

The researchers compiled a database of phenotypic values on several hundred plants of each of the five Rosaceae and have made it available to plant breeders at the Web site.

Standardized phenotyping

Dr. Cameron Peace at Washington State University is co-director of the ­RosBREED project with Iezzoni, and is ­directing marker-assisted breeding “pipelining”—helping breeding programs so they have a common focus on fruit quality traits. All breeders are being helped to see things the same way, or speak the same language using standardized tests and rating ­systems.

Phenotype refers to the way a plant looks, and that is determined by its ­genotype. The new protocols tell breeders what to look for, how to look for it, and how to evaluate what they find.

Iezzoni said a RosBREED priority for the near future is to take the lessons of RosBREED to the horticultural breeding community—telling them how to use the genetic tests that identify functional alleles and how to use them in breeding, showing them how QTL (quantitative trait loci) information can be translated into marker-assisted breeding.