Cecile L. Tondo

Genomics of Tropical Fruit Tree Crops

The genetic improvement of tropical fruit trees is limited when compared to progress achieved in temperate fruit trees and annual crops. Tropical fruit tree breeding programs require significant resources to develop new cultivars that are adapted to modern shipping and storage requirements. The use of molecular markers in tropical fruit tree breeding is greatly assisting in solving a number of difficult challenges for breeders such as the development of complex family structures for recombination mapping and for recurrent selection. A review of the literature on molecular markers development and new techniques for increasing single-nucleotide polymorphic markers is discussed. The development of marker-assisted breeding for these tropical tree crops is also discussed.

Selection of Highly Informative Microsatellite Markers to Identify Pollen Donors in ‘Hass’ Avocado Orchards

Abstract‘Hass’ is the most popular avocado (Persea americana Mill.) cultivar in the world. It has been characterized as a crop requiring cross-pollination. However, the potential extent of self-pollination and the most effective pollen donors (best cross-pollinizing cultivars) have not been determined. In this study, 56 markers were screened against ‘Hass’ and nine commonly used pollinizing cultivars grown in southern California: ‘Bacon,’ ‘Ettinger,’ ‘Fuerte,’ ‘Harvest,’ ‘Lamb Hass,’ ‘Marvel,’ ‘Nobel,’ ‘Sir Prize,’ and ‘Zutano.’ Seventeen microsatellite, i.e., simple sequence repeat (SSR) markers, were found to be very promising for paternity analysis. Four highly informative SSR markers were selected to accurately and unequivocally identify pollen parents of ‘Hass’ fruit from an orchard interplanted with these pollinizing cultivars. From 2003 to 2006, 7,984 ‘Hass’ fruit were analyzed for their paternity. Overall, the pollen parents of 99.55% of the analyzed fruit could be unequivocally identified with a single multiplex polymerase chain reaction (PCR). Only 36 fruits (<0.45%) required a second PCR reaction to reach unequivocal identification of the pollen parents.

Re-evaluation of the roles of honeybees and wind on pollination in avocado.

Summary Avocado (Persea americana Mill.) flowers, with their synchronously dichogamous behaviour, are considered to be pollinated by honeybees, despite the lack of any direct evidence. Results in southern Florida showed that avocado pollen was transferable by wind, and was dispersed over a brief period of time (15 – 60 min) each day. Ten ‘Hass’ avocado orchards in the Santa Clara River Valley, CA, USA, planted far from any known ‘Zutano’ polliniser trees, were selected to investigate the impact of honeybees on pollen transfer. ‘Zutano’ pollen (5 g per insert) was placed at the entry to beehives (approx. eight beehives per orchard) and refreshed four-to-five times during the flowering season. Successful pollinations were determined by parental analysis of harvested ‘Hass’ fruit from trees located at various distances from the beehives, and at three different stages of fruit development, using microsatellite DNA markers. The results showed no significant difference in the proportions of ‘Zutano’-pollinated fruit with respect to distance and/or development stage between orchards provided with beehives containing ‘Zutano’ pollen and those without supplemented pollen. This strongly suggests that honeybees are not the major pollinators of avocado, and that most avocado flowers are self-pollinated by wind.

Genetic differentiation, races and interracial admixture in avocado (Persea americana Mill.), and Persea spp. evaluated using SSR markers

Avocado (Persea americana Mill.) is a subtropical domesticated fruit tree indigenous to Mesoamerica. It is a member of the Lauraceae family and is separated into three horticultural races (Guatemalan, Mexican, and West Indian) mainly corresponding to their ecological adaptation, botanical, and physiological traits. Main objectives of this study were to characterize the population structure, genetic diversity, and horticultural race of a total of 354 Persea spp. trees whose origin is as follow: 221 trees [P. americana, (218), P. nubigena (2) and P. krugii (1)] from the USDA-ARS-Subtropical Horticultural Research Station, Miami; 105 trees from the Fairchild Farm [P. americana (104) and P. schiedeana (1)], and 28 trees collected in Mexico [P. schiedeana (23) and P. americana (5)]. The complexity of their interracial admixture; as well as mislabeling frequency was also evaluated. Molecular marker analysis utilizing a set of 55 simple sequence repeat (SSR) markers amplified a total of 869 alleles with a mean number of alleles per locus of 15.8 and average polymorphism information content value of 0.71, indicating a high variability in the allele frequency for the collection. Significant deviations from Hardy–Weinberg equilibrium were identified after Bonferroni correction for a large number of loci (48; 87%) due to the presence of null alleles. The main source of variation for this population was found to be within individuals (66.84%), with 19.30% variation among populations, and 13.86% variation among individuals within populations. Moreover, population specific inbreeding indices (FIS) were calculated for West Indian, Guatemalan, and Mexican [(0.1918; p value 0.0000), (0.1879; p-value 0.0000), (0.0925; p-value 0.0022)], respectively. Bayesian analysis divided the individual genotypes into groups associated with the Guatemalan, Mexican, West Indian races; interracial admixture; complex hybrids and P. schiedeana species. Also, results of the multivariate clustering method (PCA) and genetic distance analyses calculated among all possible individual combinations within the SSR diversity data agreed with Bayesian or Structure analyses results. The 55 SSRs provided complete resolution of all individuals and the estimated mislabeling error was approximately 0.28%.