RAPD markers have been used successfully in genetic analysis of several crop plants. This method poses difficulties with a highly heterozygous species such as Theobroma cacao because of the dominant phenotypic expression of bands. A backcross family derived from ctultivars Catongo and Pound 12 was analyzed to determine the efficacy of RAPD markers in analyzing cacao populations. A preliminary screen of the parents and the F1 plant used as the backcross parent was conducted with 180 RAPD primers; of these, 26% were polymorphic and reproducible and produced 104 storable loci. Genomic DNA from 54 individuals of the backcross population was then amplified with these primers; 68.3 % of the loci segregated as expected in a Mendelian fashion. Separation of RAPD fragments on acrylamide revealed an additional polymorphic locus from one primer that was indistinguishable on agarose. The results demonstrated that RAPD markers can be used to study the cacao genome.
C.M. Ronning, R.J. Schnell and D.N. Kuhn
C.M. Ronning, R.J. Schnell and S. Gazit
The native American genus Annona contains many species that are cultivated for their edible fruit, including the custard apple (A. reticuluta L.), soursop (A. muricata L.), cherimoya (A. cherimola L.), sugar apple (A. squamosa L.), and interspecific hybrids, the atemoyas. RAPD analysis of A. cherimola. `Campa' and `Jete,' A. squamosa `Lessard,' and the atemoyas `Ubranitzki,' `Malali,' and `Kaspi' resulted in very distinctive patterns, indicating that RAPD markers, may be an efficient method of fingerprinting genotypes within and between Annona species. All 15 primers used generated repeatable, polymorphic patterns. An F1 population of `Jete' × `Lessard' was analyzed to determine the inheritance of the RAPD banding patterns. Fifty-two polymorphic loci were identified, which segregated in an expected Mendelian fashion.
R.J. Schnell, C.M. Ronningl and C Degani
To investigate the usefulness of RAPDs for determining parentage in mango, progeny arising from caged trees of the cultivars Keitt (FS1) and Kent (FS2) were analyzed. In the FS1, 110 bands were generated, of which 78 (70.2%) were repeatable. Of these 78 loci, 23 were variable and segregated 3:1 as expected. In the FS2, 142 bands were generated, of which 57 (40%) were not repeatable, 6 (4.2%) were present in the progeny but not in the parent, and 79 (56%) were repeatable. Thirty-nine of these loci were variable; however, only 21 segregated as expected. Apparently, the progeny arising from the caging of `Keitt' are the result of self-pollination, while those arising from caged `Kent' are not. The six bands found in the FS2 but not in `Kent' are reproducible and, along with the 46% anomalous segregation, indicate that cross-pollination did occur. The implications for mango breeding efforts are discussed.
R.J. Knight Jr., J.A. Payne, R.J. Schnell and A.A. Amis
R. J. Knight Jr., R. J. Schnell, J. A. Payne and A. A. Amis
Efforts to obtain edible-fruited passion vines through hybridization of Passiflora edulis and P. incarnata produced a population of tetraploid plants that survive freezing winters in central Georgia belowground, and grow aboveground in warm weather. One selection from this population blooms from late June or early July through October, yielding quantities of flowers from 8.5 to 9.6 cm diameter that have deep blue-colored sepals and petals surmounted by a disc of heavily-crimped filaments that are white at the outer margin. The nectar has proven a good food resource for the ruby-throated hummingbird, which breeds in much of the southeastern U.S. This clone is highly self-incompatible and sets no fruit when grown apart from cross-compatible clones. Its vines are vigorous, growing to 5 meters or more, and have dark green, markedly denticulate trilobed leaves 13 to 24.5 cm long by 15 to 25 cm wide. These afford a nursery habitat for caterpillars of 3 native butterflies, the zebra and the Gulf and variegated fritillaries. Because of its ease of culture and wide adaptation, this vine is recommended to plant in the continental U.S. for environmental enhancement.
R.J. Schnell, R.J. Knight, D.M. Harkins and Gary Zill
The ability to eliminate zygotic seedlings from the polyembryonic mango (Mangifera indica L.) rootstock `Turpentine' by visual roguing was investigated. Four selected populations, A) randomly selected plants, B) plants selected as off-types, C) seedlings that were of `Turpentine' phenotype, and D) seeds where a single seedling emerged, were examined using electrophoretic analysis and five enzyme systems. Significant differences (χ2 = 39.63, P< 0.001) were found among the four categories, with 28% of the random, 66% of the off-type, 10% of the true-to-type, and 54% of the monoembryonic seedlings being zygotic. These data indicate that visual selection for trueness-to-type and roguing for off-types is useful in reducing the frequency of zygotic seedlings among `Turpentine' rootstock plants.
R.J. Schnell, J.S. Brown, C.T. Olano, A.W. Meerow, R.J. Campbell and D.N. Kuhn
Mango (Mangifera indica L.) germplasm can be classified by origin with the primary groups being cultivars selected from the centers of diversity for the species, India and Southeast Asia, and those selected in Florida and other tropical and subtropical locations. Accessions have also been classified by horticultural type: cultivars that produce monoembryonic seed vs. cultivars that produce polyembryonic seed. In this study we used 25 microsatellite loci to estimate genetic diversity among 203 unique mangos (M. indica), two M. griffithii Hook. f., and three M. odorata Griff. accessions maintained at the National Germplasm Repository and by Fairchild Tropical Botanic Garden in Miami, Fla. The 25 microsatellite loci had an average of 6.96 alleles per locus and an average polymorphism information content (PIC) value of 0.552 for the M. indica population. The total propagation error in the collection (i.e., plants that had been incorrectly labeled or grafted) was estimated to be 6.13%. When compared by origin, the Florida cultivars were more closely related to Indian than to Southeast Asian cultivars. Unbiased gene diversity (Hnb) of 0.600 and 0.582 was found for Indian and Southeast Asian cultivars, respectively, and both were higher than Hnb among Florida cultivars (0.538). When compared by horticultural type, Hnb was higher among the polyembryonic types (0.596) than in the monoembryonic types (0.571). Parentage analysis of the Florida cultivars was accomplished using a multistage process based on introduction dates of cultivars into Florida and selection dates of Florida cultivars. In total, 64 Florida cultivars were evaluated over four generations. Microsatellite marker evidence suggests that as few as four Indian cultivars, and the land race known as `Turpentine', were involved in the early cultivar selections. Florida may not represent a secondary center of diversity; however, the Florida group is a unique set of cultivars selected under similar conditions offering production stability in a wide range of environments.
C.M. Ronning, D.M. Harkins, R.J. Schnell and L.H. Purdy
Cacao is an important crop in the tropics, but its breeding has been hampered by a lack of understanding of its genetics. One result of this has been the introduction of “hybrid” trees which did not perform predictably under various environmental conditions. We are studying the inheritance of isoenzyme, RFLP, and Random Amplified Polymorphic DNA (RAPD™) markers in order to estimate the genetic relationships among and between populations. Our objectives include determining if any linkage exists between these molecular markers and witches' broom (Crinipellis perniciosa) resistance, a major disease of cacao.
R.J. Schnell, C.T. Olano, J.S. Brown, A.W. Meerow, C. Cervantes-Martinez, C. Nagai and J.C. Motamayor
Commercial production of cacao in Hawaii is increasing, and this trend is expected to continue over the next several years. The increased acreages are being planted with seedlings from introduced and uncharacterized cacao populations from at least three initial introductions of cacao into the islands. Productive seedlings have been selected from a planting at Waialua, Oahu. The parents of these selections were believed to be the population at the Hawaii Agriculture Research Center (HARC) at Kunia; however, potential parental populations also exist at Univ. of Hawaii research stations at Waimanalo and Malama Ki. Using microsatellite markers, we analyzed the potential parental populations to identify the parents and determine the genetic background for 99 productive and 50 unproductive seedlings from the Waialua site. Based on 19 polymorphic microsatellite loci the parental population was identified as trees from Waimanalo and not trees from Malama Ki or Kunia. The Kunia and Malama Ki populations were very similar with low allelic diversity (A = 1.92) and low unbiased gene diversity (Hnb) of 0.311 and 0.329, respectively, and were determined to be Trinitario in type. The Waimanalo, productive seedling, and unproductive seedling populations had much higher levels of genetic diversity with Hnb of 0.699, 0.686, and 0.686, respectively, and were determined to be upper Amazon Forastero hybridized with Trinitario in type. An additional 46 microsatellite markers were amplified and analyzed in the Waimanalo parents, productive, and unproductive seedlings for a total of 65 loci. Seventeen loci contained alleles that were significantly associated with productive seedlings as determined by Armitage's trend test. Of these, 13 loci (76.4%) co-located with previously reported quantitative trait loci for productivity traits. These markers may prove useful for marker assisted selection and demonstrate the potential of association genetic studies in perennial tree crops such as cacao.
J. Steven Brown, R.J. Schnell, J.C. Motamayor, Uilson Lopes, David N. Kuhn and James W. Borrone
A genetic linkage map was created from 146 cacao trees (Theobroma cacao), using an F2 population produced by selfing an F1 progeny of the cross Sca6 and ICS1. Simple sequence repeat (SSR) markers (170) were used principally for this map, with 12 candidate genes [eight resistance gene homologues (RGH) and four stress related WRKY genes], for a total of 182 markers. Joinmap software was used to create the map, and 10 linkage groups were clearly obtained, corresponding to the 10 known chromosomes of cacao. Our map encompassed 671.9 cM, approximately 100 cM less than most previously reported cacao maps, and 213.5 cM less than the one reported high-density map. Approximately 27% of the markers showed significant segregation distortion, mapping together in six genomic areas, four of which also showed distortion in other cacao maps. Two quantitative trait loci (QTL) for resistance to witches' broom disease were found, one producing a major effect and one a minor effect, both showing important dominance effects. One QTL for trunk diameter was found at a point 10.2 cM away from the stronger resistance gene. One RGH flanked the minor QTL for witches' broom resistance, implying possible association. QTLs mapped in F2 populations produce estimates of additive and dominance effects, not obtainable in F1 crosses. As dominance was clearly shown in the QTL found in this study, this population merits further study for evaluation of dominance effects for other traits. This F2 cacao population constitutes a useful link for genomic studies between cacao and cotton, its only widely grown agronomic relative.