Accelerating Citrus Breeding Efforts

Josh McGillBreeding

By John M. Chater

Professors Jude Grosser and Fred Gmitter have been breeding improved citrus varieties at the University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) Citrus Research and Education Center (CREC) for decades. Over time, they have released several of their advanced selections. Many of these selections are sweet oranges. There is a subset of these creations that appear to have some appreciable level of HLB tolerance under the right environmental and cultural conditions (Figure 1).

Figure 1. N13-32 Hamlin sweet orange displays good health in a commercial grove.

Today, there are even more promising materials in the breeding pipeline either under evaluation or awaiting field trials. Some of these are sweet oranges, and others are fresh market fruit selections. The program has also selected sweet orange-like hybrids that have brilliant color, high Brix and exceptional orange-like flavor.

From a breeder’s standpoint, the program has been hugely successful in creating and identifying both scions and rootstocks of diverse and clever genetic backgrounds for Florida’s growers. The list is long and full of potential winners. Currently, this program’s successes have led to a healthy backlog of citrus material from Gmitter and Grosser’s breeding program that needs thorough evaluation. Therefore, work is underway to characterize germplasm as quickly as possible.

In plant breeding, finding new and improved, rapid methods to phenotype (or characterize) plant material for important traits of value to industry is helpful because there can be thousands, even tens of thousands of individuals or more, to evaluate across multiple sites. Being able to determine important traits for HLB tolerance, such as canopy size and health status, of 10,000 or more trees in minutes with a drone can assist breeders in the selection process and in phenotyping during the course of field trials.

In horticulture and other disciplines, the act of collecting data quickly from a large population is called “high-throughput phenotyping.” Some have defined high-throughput image-based phenotyping as characterizing minimally “hundreds of plants per day.” The ability to characterize large populations allows for quick phenotyping or for the best individuals to be selected as parents and/or for more advanced field trial stages. It also opens the door to 21st century advanced breeding methods using genomics and other tools.

One of these methods, marker assisted selection, would allow a breeder to select individuals for desired traits based on molecular markers. This means he or she would be able to use the genetic or biochemical composition of the leaf of a seedling to determine its value downstream. These advanced methods could reduce the costs of phenotyping substantially. Imagine only having to phenotype 12 individuals for HLB tolerance, yield and pounds solids instead of 30,000 trees.

In most breeding programs, phenotyping is typically the most expensive and time-consuming process in the system. Hundreds of thousands to millions of dollars per year are required to support a premier tree-fruit breeding program. These costs involve personnel; space in the laboratory, greenhouse, and field; materials and supplies; vehicle upkeep and much more.

For a species like citrus, the costs are extraordinary. The space is on the acreage scale, the time required is long, and the path to success can be pushed at a moment’s notice. Florida Citrus Hall of Fame Inductee and Professor Emeritus Bill Castle would say that a trial should last at least eight years at multiple sites. This timeline of eight years makes good sense and comes from a place of wisdom considering the challenges that the industry has faced with rootstock incompatibility. Such incompatibility can take years to manifest, as was seen in Roble on Swingle, Carrizo and C-35 citranges. Fruit disorders, like fruit split on Early Pride, and other undesirable outcomes can result from unfortunate scion and rootstock selection.

There are currently dozens of trials and other plantings that have trees with evaluations ongoing. Many of these trees are mature now and have several years of data collection. Other trials are newer, and some are just being budded this season. Data are continuously being compiled on a website for the rootstock trials that has long been overseen by Castle. But with new faculty hires, the passing of the torch is underway and writing up the results of the trials in papers has been delegated.

To create a snapshot of the ongoing citrus trials and plantings, the Citrus Research and Development Foundation sponsored a project titled “Utilizing high-throughput phenotyping to screen germplasm and ongoing field trials for promising citrus accessions in HLB-endemic Florida” to evaluate all of the UF/IFAS citrus material in trials. The purpose of this project was not only to get a status update from the plant improvement team, but to create inventories and lists of top performers with potential to serve the industry.

This work was commissioned around the time of the Donaldson selection, and stakeholders wished to know if there is a similarly healthy selection in the UF/IFAS citrus breeding germplasm. Drones were flown over 91 sites comprised of plantings and trials. Some of these were replicated trials of over 70 acres with dozens of rootstock-scion combinations. Others were germplasm blocks, and some were smaller plantings, many on growers’ land.

Due to the timeliness of work, it was determined by stakeholders that larger, fruit-bearing trees should be evaluated and not young trees. An interim report has been submitted that identified top performers among all trials (such as some of the OLL lines, Figure 2). The report provides stakeholders with some baselines for future comparisons.

Figure 2. Jude Grosser stands in front of his OLL lines in St. Cloud, Florida.

Before this drone work even started, some selections were already being described by experienced growers as HLB tolerant. These selections are actively being planted by growers as resets or in solid blocks.

For example, there is some promise in N13-32, a somaclone of Hamlin, as a healthier alternative scion for early sweet oranges. At one site in Lake Alfred, juice quality and yield have been rather similar to Hamlin. Tree health seems to be going strong at some sites and trials. However, more data are needed to know if this selection is a good choice for resets or solid blocks as some, including the breeder, would say cropping may be delayed by about one season. Whether there is a tradeoff between this reported delay (notedly unsubstantiated by data) in cropping versus the building of biomass over time to make a large tree that can bear more boxes of fruit versus conventional Hamlin 1-4-1 remains to be seen. There are trees in some groves that are performing quite well for health. At other sites, trees are of typical health for Hamlin in HLB-endemic Florida. An important next step is discovering why there are differences among sites for this and other selections. Such a research direction is being pursued for funding. As Castle would say, “The beat goes on.”

John M. Chater is an assistant professor at the UF/IFAS CREC in Lake Alfred.

Share this Post

Sponsored Content