Genetics to Breeding

Strawberry is the most consumed Rosaceae small fruit in the world where flavor is a key for the success of a new cultivar. Overall flavor in strawberry is driven by a balance between sugars, acids, and several aroma compounds. Strawberry breeders are interested in using marker-assisted selection to combine better flavor with remontancy (day-neutrality). While quantitative trait loci (QTLs) have been reported for some of these traits, markers do not exist for routine application in breeding. The RosBREED project is focused on enabling marker-assisted breeding via identification and validation of traits important to strawberry breeders. The development of the first genome-wide scanning platform (Affymetrix Axiom IStraw90TW) for the cultivated octoploid strawberry enabled high-throughput genotyping of 384 pedigree-connected strawberry accessions. Phenotypic data for titratable acidity (TA), soluble solids content (SSC) and remontancy were collected in the field in 2011 and 2012 in Oregon and Michigan. FlexQTL™ software was used for a marker consistency check and genome-wide QTL analysis for these three traits using SNP markers that were already genetically positioned using a ‘Holiday’ × ‘Korona’ linkage map. The preliminary QTL analysis resulted in 2 QTLs for TA, 1 QTL for SSC and 7 QTLs for remontancy. These results validate some of the previously reported QTLs. Estimation of haplotype effects and phenotypic variation explained by these QTLs is in progress for RosBREED strawberry germplasm. Haplotypes associated with desirable QTL alleles should enable the development of DNA-based markers and improve efficiency in the development of remontant and flavorful strawberry cultivars.

The RosBREED project (http://www.rosbreed.org/) has successfully identified functional alleles associated with nine peach fruit quality traits: Fruit resistance to bacterial spot (Xap1 and Xap6), maturity date (G4mat), fruit type (G), blush, flesh color (Y), texture type (F-M), acidity (D), acidity and soluble solid content (G7Flav) and mealiness. Forty-eight single nucleotide polymorphism (SNP) markers, significantly associated with the nine phenotypic traits and capable of distinguishing all functional alleles were selected and divided into two 24-SNP mini arrays. Two DNA service providers [University of Arizona Genetics Core (UA) and BioDiagnostics (BDI)] were charged with the design and testing of the mini SNP arrays. Plant material from the four RosBREED peach demonstration breeding programs (University of Arkansas, Clemson University, Texas A&M University and University of California, Davis) representing advanced breeding material as well as several control samples were genotyped. Genotyping data were successfully translated into trait predictions and results provided to each breeding program. Depending on the material genotyped, genotyping data acquired accurately predicted phenotypic performance in 80-90% of cases across material from all breeding programs. These results are due to the vast haplotypic diversity in U.S. peach breeding germplasm. The additional 20-10% of material required specific attention to assigning the correct haplotypes to the correct phenotypes. Feasibility of using mini SNP arrays in breeding programs for cross planning and seedling selection as well as the level of service required from providers will be discussed.

The Rosaceae family includes most of the temperate tree fruit crops such as apple, apricot, peach, pear, and cherry. Seedling selection is the process of selecting superior seedlings generated from crosses for cultivar development. Traditional seedling selection (TSS) is expensive for most rosaceous tree fruit because of long juvenility and large plant sizes. Marker-assisted seedling selection (MASS) is a promising approach to improve the efficiency of TSS with the assistance of DNA markers. A lack of decision-support tools is a major challenge in identifying cost-efficient MASS schemes. Here we report on progress made in the development of a web-based software tool Seedling Select to enable accurate and flexible cost modeling for MASS in rosaceous tree fruit.
Seedling Select consist of four modules: TSS Cost Structure (mod. 1), Seedling Family Structure (mod. 2), and DNA Test Cost Structure (mod. 3) lay the foundation of cost estimation by modeling costs in each TSS stage, seedling family, and DNA test, respectively, while MASS Cost Structure (mod. 4) calculates and compares total costs of TSS and MASS by integrating outputs from the previous modules with a user-defined MASS scheme. By comparing costs of all possible MASS schemes, this software tool will suggest cost-efficient MASS schemes for defined sets of breeding families.

Seedling Select will be hosted on the Genome Database for Rosaceae (www.rosaceae.org). Once available, Seedling Select will assist breeders to design efficient MASS schemes with a rational allocation of resources.

Red coloration has become an important commercial trait in Prunus by improving consumer acceptance and health properties as well as ornamental aspects. This pigmentation is caused by the accumulation of anthocyanins in the vacuole of the cells that form the colored tissue. The biosynthetic pathway of the anthocyanins is being activated by transcription factors of the MYB family. With the Prunus genome sequence being available and red coloration traits mapped in the Prunus reference map or elsewhere candidate genes can be found. This communication will present the identification of new candidates of the MYB transcription factor family in the peach genome and their possible implication in the expression of red coloration traits. For the candidate gene-trait co-localization we will also include our latest results on newly mapped red pigmentation traits. Finally marker development for marker assisted breeding will be discussed.

Since the late 80’s, significant increases of mean temperatures has been recorded worldwide. This change leads to earlier flowering dates in apple, possibly making frost damage and phenological disorders related to fruit setting more frequent. This situation urges studies on genetic determinism of chilling and heat requirements in apple tree leading to timing of bud phenology.
The present study aimed at clarifying the genetic determinism of timing of budbreak for five full sib families related by their pedigrees. Two of these families are located in Montpellier, South-East of France, and three in Angers, North-West of France. Up to five development stages, green point, vegetative budbreak, floral budbreak, flowering and end of flowering, were observed for the five families, over two years. The climatic years and locations were characterized by the duration of chilling and heat requirement completion for a reference cultivar ‘Golden Delicious’.
On that basis, mixed models were used to estimate the genotype effects and the GxE interaction effects. The five populations were also genotyped with the Illumina 20k SNP array within the EU-FP7 project FruitBreedomics. These SNP data and the Best Linear Unbiased Prediction values on genotypes and GXE interaction for the different phenological stages are used for pedigree-based QTL mapping with the FlexQTLTM software. The outcomes of this study will provide new insights on the genetic determinism of timing of budbreak in apple, leading to new tools for apple breeders.

The power to identify, map and quantify QTL underlying complex traits may substantially increase when considering multiple segregating progenies simultaneously. The relatedness among these progenies can be fully exploited when the pedigree is known and the ancestors are available for genotyping. Previously, we successfully used the FlexQTL software to map QTL in such a Pedigree Based Analysis (PBA) approach to a large pedigreed population in apple with only 87 SSR markers (Bink et al. 2014). The current availability of high-throughput SNP genotyping infrastructures allows marker genotyping at much higher densities, which will contribute to higher mapping resolution and more accurate QTL characterization but also poses challenges to the QTL mapping software. Within the framework of the projects SCRI-RosBREED (#2009-51181-05808) and EU-FP7 FruitBreedomics (#FP7- 265582), we have improved and extended functionality of the FlexQTL software to better handle large numbers of SNP markers with regard to computation time, the phasing of the lowly-informative SNPs in complex pedigrees (e.g., inbreeding loops). Furthermore visualization tools were added to inspect recombination events. Furthermore, highly informative multi-allelic haplotypes can now be built comprising consecutive low informative SNPs, which will reduce computational effort without losing information from the high density marker data. The progress and challenges on utilizing high density SNP data in Rosaceae crops will be highlighted via a case study, i.e., applying the new version of FlexQTL software to fruit firmness trait on a large 20K genotyped apple population as available from the FruitBreedomics project.

Reference: Bink MCAM et al (2014). Theor Appl Genet: 1-18.(doi: 10.1007/s00122-014-2281-3).

The tremendous increase in throughput of genotyping techniques opened appealing perspectives for genome-wide selection (GWS), which could enhance breeding efficiency by decreasing generation interval and increasing selection intensity and/or accuracy of breeding values. In GWS a large training population with both phenotypic and genotypic data is used to construct a statistical prediction model which is then applied to estimate Genomic Breeding Values (GBV) of individuals that only have genotypic data. In the EU-FP7 project FruitBreedomics, we performed a pilot study of GWS in apple. The two main objectives were to provide proof of principle and to evaluate the accuracy of prediction with respect to the relatedness between the test and the training populations. Hereto the phenotypic means of the 50 best and 50 worst predicted individuals of a test progeny comprising 700 individuals were compared and correlations between predicted GBV and phenotypic data were compared for smaller full-sib families of different relatedness.
The training population included 20 full-sib families comprising 992 individuals genotyped with an Illumina 20K SNP array and phenotyped for fruit quality traits. The test population comprised four progenies from commercial breeding programs totaling 1500 individuals that were genotyped with 512 SNP that had been selected among the 20K for heterozygosity in the parents of the test population and genome-wide coverage. SNP genotypes were completed through imputation and fed into the calibrated prediction model to obtain GBV on the 1500 offspring. We will present the results and discuss challenges remaining to bring GWS into practice in Rosaceae species.

The European project FruitBreedomics was conceived with the aim of improving the efficiency of peach and apple breeding. One of the strategies consists of enhancing the use of molecular markers in progeny selection (Marker Assisted Selection, MAS). In peach, many QTLs linked to more than one hundred agronomic traits have been identified and published so far, in all instances detected after analyzing single mapping populations. Despite the great amount of information, the number of molecular markers available to be used in MAS is still reduced. One of the reasons is the too large QTL intervals, which are due to low marker density and/or limited progeny size. With the final goal of reducing the confidence intervals of some relevant agronomic QTLs, in FruitBreedomics we have genotyped with the 9K peach SNP chip about 1,500 progeny from 18 peach and other Prunus crosses with phenotypic data from several years. Hereby, we present first results on fruit quality QTL mapping using pedigree-based analysis (PBA) methodology. PBA has its power in the joint analysis of progenies from different crosses at a time, enlarging the population size and consequently narrowing the candidate genomic regions. Moreover, as multiple parents with different genetic backgrounds are included in the analysis, this method may detect multiple favorable alleles. Additionally, the subsequent analyses of identity by descent (IBD) give insight into the peach founders, that brought some of the favorable QTL alleles into these breeding populations.

Climate change conditions accentuate drought since the reduction of 10% of the precipitation is translated in a reduction of 25% of the soil water. Therefore, it is indispensable the use of adapted rootstocks to water stress conditions. In order to understand their response to drought, physiological and molecular parameters of three Prunus hybrid rootstocks, the almond x peach hybrid (P. amygdalus x P. persica) ‘Garnem’, their progenies ‘P.2175’ x ‘Garnem’-3 trihybrid (P. cerasifera x [P. amygdalus x P. persica]) and ‘P.2175’ x ‘Garnem’-9 trihybrid were investigated. Plants in pots were subjected to drought conditions (35% water soil content) during one month. Subsequently, plants were submitted to re-watering period. Physiological responses were monitored through transpiration and leaf water potential showing significant differences along the experiment and among the genotypes. Gene expression analysis of four genes coding for proteins related to ABA pathway and abiotic stress were analyzed by RT-qPCR. A dehydrin (ppa005514m), a LEA protein, (ppa008651m), an A20/AN1 zinc finger (ppa012373m), and a bZIP transcription factor (ppa013046m). The profile expression of the four genes showed a correlation with physiological parameters of drought response, being higher in root than in phloem tissue. The zinc finger and bZIP transcription factors showed differences in expression relative to the target genes LEA protein and dehydrin under drought and recovery treatment, showing a regulatory response to water stress in Prunus genotypes.

Superoxide dismutase (SOD, EC 1.15.1.1) is one of the most important antioxidant enzymes and constitutes the first level of defence against superoxide radicals in plants. SOD catalyzes the dismutation of O2– to H2O2 and O2. SOD enzymes fall into three distinct families that are classified according to metal cofactors. The most abundant SODs in plants are the CuZnSODs which are found mainly in the cytosol and in the chloroplasts providing enhanced tolerance to oxidative stress. In grafted plants, it is reported that oxidative stress could trigger cell and tissue degradation processes in incompatible grafts. The aim of this study was to evaluate the gene expression of different genes encoding CuZnSOD (CSD1, CSD2, CSD3) and cytosolic SOD activity in callus unions with different graft response from pear cvs. ‘Conference’ (Co) and ‘William’ (Wi) and the quince rootstock clone ‘Ba29’ during three weeks after union. CuZnSOD mRNA transcripts were detected in all combinations throughout the graft union development However, the mRNA expression patterns varied with the kind of combination examined. CSD1 and CSD2 transcript levels showed significant differences between ‘Co/Co’ and ‘Co/BA29’ at 10 and 21 DAG, respectively. Likewise, it was observed a 4-fold increase in CSD3 gene expression in the homograft ‘Wi/Wi’ at 21 DAG compared to the heterograft ‘Wi/Ba29’. Furthermore, there was a higher SOD activity in the compatible cultivar (Co) than in the incompatible cultivar (Wi) from 1 day after wounding onwards. These results will be discussed in terms of wound/healing repair mechanisms and oxidative stress associated to grafted plants during graft union development.