Genetic Resources
View abstracts in the conference PDF booklet.
Where in The New Melon Classification Schemes Does Cucumis melo ssp. agrestis var. texanus Belong?
Kaori Ando1, Xin Wang2, Zhangjun Fei3,2, William M. Wintermantel1, James D. McCreight1
1U.S. Department of Agriculture, Agricultural Research Service, Crop Improvement and Protection Research Unit, Salinas, CA, USA. 2Boyce Thompson Institute, Ithaca, NY, USA.
Melon, Cucumis melo L., is one of the most important fruit crops worldwide, and is highly diverse for many vegetative characters, fruit color, size, shape, soluble solids, flavor, aroma, and ripening. Thirteen schemes for melon classification have been proposed since 1859, the most recent of which recognized 19 horticultural groups. Wild North American melon populations were recognized as C. melo ssp. agrestis var. texanus in 2002, but were overlooked in 3 subsequent revisions to melon classification. This wild melon is found in southeastern U.S. and Mexico, is weedy and bears many small, non-sweet, often bitter, yellow fruit, and is a potential source of resistances to powdery mildew and nematodes. It is similar for some traits to groups Chito (syn. ssp. melo var. chito), which is found in Central America and Caribbean countries, and Dudaim (syn. ssp. melo var. dudaim), which is cultivated in Central Asia, but were considered to be distinctively different from texanus based on expression of 43 morphological and 2 phenological quantitative traits, and a total of 40 RAPD and SSR markers. We analyzed genetic variation among more than 2,000 C. melo accessions in the U.S. National Plant Germplasm System, including 44 texanus, 4 putative Chito and 2 putative Dudaim accessions, as well as heirloom varieties and important cultivars using genome-wide SNP markers that were derived via the Genotyping-by-Sequencing (GBS) method. The texanus accessions clustered together, isolated from other Groups in phylogenetic and principal component analyses (PCA). The Chito accessions fell in 2 different groups distinct from texanus. Dudaim clustered with 3 of the Chito accessions. Nine additional Chito accessions were, therefore, obtained from M. Pitrat, Institut National de la Recherche Agronomique (INRA), Avignon-Montfavet, France for an ongoing comparative GBS and phenotypic analyses with the texanus accessions in order to more clearly characterize Chito and determine whether texanus constitutes a distinct Group or is a sub-group of Chito. Several of the Chito reference accessions used to delimit texanus were also included in this analysis. Partial Funding provided by the Specialty Crop Block Grant Program at the U.S. Department of Agriculture (USDA) through Grant 14-SCBGP-CA-0006 (The CucCAP Project)
Diversity of Lesser Known Cultivated Cucurbita from Latin America: Landraces to Locally Grown Cultivars
Thomas Andres
The New York Botanical Garden, Science, Bronx, NY, USA
There are perhaps over a thousand named varieties, or cultivars, of Cucurbita, which represent an astounding diversity. The exact number is difficult to ascertain since names change over time and from region to region, and there are many synonyms, as well as new varieties released by breeders every year. Taxonomically, these fall into multiple species from the Americas, including three primary species, i.e., Cucurbita pepo, Cucurbita maxima, and Cucurbita moschata; two species of lesser worldwide economic importance, i.e., Cucurbita argyrosperma and Cucurbita ficifolia; and landraces from an additional species, Cucurbita ecuadorensis. Seed catalogs today generally list far more cultivars than in the past. But whether this represents an increase in genetic diversity is doubtful. For example, what was once a landrace, may have been selected for a number of named cultivars, but with an overall erosion in total genetic diversity. While there are fewer major seed houses than in the past, this is not a major issue as far as a loss in squash diversity, since many alternative online sources of seeds have arisen, from start- up private seed companies to seed saver organizations to e-commerce, such as eBay and Etsy. In addition to these seed sources, there are noteworthy locally grown cultivars. These are not more widely distributed for a number of reasons. Perhaps the main reason is they are locally adapted to certain growing conditions, such as short-day photoperiodism in the tropics, and thus will not set fruit in temperate regions. Most of the cultivars of C. ficifolia, as well as many moschata, are short-day sensitive, to the extent that there is far more morphological variation within both of these species than is generally recognized. Some cultivars are selected for uses not popular elsewhere, such as their edible leaves and shoot tips, sprouts, flowers, or seeds. The C. moschata ‘loche’ is grown only on the northern coast of Peru in a special way that helps concentrate its flavor, so that it becomes more of a seasoning than a vegetable.
Development of Novel Sets of Reciprocal Introgression-line Collections in Melon
Jason Argyris1,2, Lara Pereira2, Marta Pujol1,2, Jordi Garcia-Mas1,2
1Institut de Recerca i Tecnologies Agroalimentàries, Campus UAB, Cerdanyola del Vallès, Barcelona, Spain. 2Centre de Recerca en Agrigenòmica (CRAG) Consorci CSIC-IRTA-UAB-UB, Campus UAB, Cerdanyola del Vallès, Barcelona, Spain
Introgression line (IL) collections are composed of individuals sharing a high proportion of genetic background from a recurrent parent, differing only in a specific region of a genomic introgression inherited from the donor parent. Two sets of reciprocal ILs were generated from the intraspecific cross between a non-climacteric, inodorus ‘Piel de Sapo’ (PS) type (C. melo ssp. melo group inodorus) and the highly climacteric Charentais-type ‘Vedrantais’ (VED) (C. melo ssp. melo group cantalupensis). Marker assisted backcrossing and selection followed by selfing was employed for up to six generations to develop the core IL sets. The VED IL collection is composed of 38 lines with a mean introgression size of 15.6 Mb covering 91% of the VED genome in the PS genetic background. The PS IL collection was also comprised of 38 lines with introgression lengths ranging from 2.4 to 33 Mb and covering 95% of the PS genome in the VED genetic background. A preliminary phenotypic analysis of the IL collections in autumn of 2017 showed a high level of segregation for different traits related to plant architecture and vigor, flowering time, fruit set rate, and especially for fruit quality and ripening behavior. Several QTL that were mapped in a VED x PS RIL population were successfully validated in the ILs. In summer of 2018 exhaustive phenotypic analyses for n=5 fruits of each line of both populations will be performed. These collections should serve as an outstanding resource for trait diversification in melon, and fine-mapping and cloning of QTL for traits related to fruit quality.
Exploring Spanish Watermelon Diversity for Resistance to Fungal Diseases
Ana Garcés-Claver1, Vicente González1, Gorka Perpiñá2, Eva Martínez2, Belen Picó2, Maria Luisa Gómez-Guillamón3
1Centro de Investigación y Tecnología Agroalimentaria de Aragón, Zaragoza, 50059, Spain. 2Instituto de Conservación y Mejora de la Agrodiversidad- Universitat Politecnica de Valencia , Valencia, Spain. 3Instituto de Hortofruticultura Subtropical y Mediterránea La Mayora, UMA- CSIC, Algarrobo, Málaga, Spain
Watermelon [Citrullus lanatus (Thumb.)Matsum. and Nakai] crops are affected by several soil- borne fungi as Fusarium oxysporum f. sp. niveum (Fon), Monosporascus cannoballus and Macrophomina phaseolina and more recently, by powdery mildew caused by Podosphaera xanthii which has become a concern among growers. In Spain, little research has been done until now searching for resistance sources, in spite of the occurrence and spread of these diseases and the existing and unexplored Spanish watermelon diversity. One hundred and twenty one Spanish watermelon accessions were artificially inoculated with the isolate RZ1 belonging to P. xanthii race 1W by using a conidia suspension (4,2 x 104 cel/mL). No immune watermelon accession to powdery mildew was found and most of them were highly susceptible, with profuse sporulation of the fungus. However two accessions of C. lanatus citroides (NC079249 and NC0100745) and two of C. lanatus lanatus (NC054847 and NC082460) showed very low level of sporulation. Sixty-four of these accessions were also inoculated with an isolate of Fon race 2 by using a conidial suspension (3 x 106 cel/mL). Most of the accessions were susceptible and only four accessions of C. lanatus lanatus, NC042492, NC026156, NC054866 y NC047502 showed a high level of resistance. This subset was also evaluated against M. cannonballus and M. phaseolina, inoculating the roots with an aggressive isolate of each pathogen, grown in wheat seeds (200 gr of infected wheat seeds/kg of peat) and with the toothpick/stem method, respectively. M. phaseolina was less aggressive in our conditions, and six highly resistant accessions were found. Interestingly, one of them was NC079249, also resistant to powdery mildew, and a second citroides accession NC100274. Two of the accessions resistant to Fon, NC042492 and NC026156, were also moderately resistant to M. phaseolina. M. cannonballus was very aggressive and severe root and hypocotyl damage was found in most accessions; the citroides accession NC100274 was one of the few highly resistant to M. cannonballus. Results allowed the selection of accessions multi-resistant to the main fungi affecting watermelon that will be used in breeding programs. This work was partially funded by Spanish grants AGL2017-85563-C2-1-R and AGL2017-85563-C2-2-R.
Oleaginous Potential of Seeds of Cucurbita moschata and C. argyrosperma
Magda Valdés Restrepo1, Sanin Ortiz Grisales2
1Universidad de San Buenaventura, Cali, 237, Colombia. 2Universidad Nacional de Colombia, Palmira, 237, Colombia
Cucurbita species are New World cucurbits that have been grown since pre-Columbian times for their fruit pulp and seeds. Of the five cultivated species of Cucurbita, the two best adapted for growing in the warmest parts of the Americas are C. moschata Duchesne (tropical pumpkin) and C. argyrosperma Huber (silverseed gourd). The purpose of the present investigation was to evaluate the seeds of these two species as energy sources, focusing on ethereal extract (EE) content. We evaluated 394 accessions, 295 of C. moschata and 99 of C. argyrosperma subsp. sororia, from the Cucurbita collection maintained at the Vegetable Program of the Universidad Nacional de Colombia, Palmira. Five plants of each accession were grown out in a completely randomized design, within- and between-row spacing 3 m in a 1.6-hectare plot. A weighted selection index (WSI) was used for each variable: EE (34.53 ± 15.0%), Seed production per fruit (SPPF) (63.29 ± 20.0 g) and Number of fruits per plant (NFP) (5.3 ± 2.0). All three variables expressed significant differences (P < 0.05) between and within species and a highly significant (P < 0.01) positive correlation was observed between Seed weight per plant (PSP) and Total ethereal extract per plant (TEEPP) (r2 = 0.95). C. moschata accessions that were identified by WSI as having the highest oleaginous potential were numbers 308, 129, 142, 144, 136, and 160. The most promising C. argyrosperma subsp. sororia accessions were numbers 256, 140, 260, 132, 107, and 68. The composition of the oil from the EE was predominantly polyunsaturated fatty acids, 70.1% and 68.3% from C. moschata and C. argyrosperma, respectively.
Reactions of Cornell Melon Breeding Lines to Cucumber mosaic virus and Their Horticultural Qualities
Mikyeong Kim1, Michael Mazourek2, James D. McCreight3, William M. Wintermantel3
1RDA, National Institute of Sciences, Wanju, Korea, Republic of. 2Cornell University, Ithaca, NY, USA. 3U.S. Department of Agriculture, Agricultural Research Service, Crop Improvement and Protection Research Unit, Salinas, CA, USA
Cucumber mosaic virus (CMV) has been classified into major subgroups I and II. CMV subgroup I has reemerged as a concern for melon (Cucumis melo L.) production in California and Arizona. Twenty-five melon breeding lines developed by Cornell University for resistance to CMV were, therefore, evaluated for their reactions to CMV in a greenhouse test, and for horticultural quality and adaption to the Central Valley of California in a field test. Cotyledons of melon seedlings were mechanically inoculated with a CMV subgroup I isolate collected from infected melon plants in California. Plants were individually evaluated for their reaction to CMV as evidenced by mosaic symptoms, and by ELISA using commercial antiserum against CMV for quantification of virus titer. Eleven lines were susceptible, with rates of infection ranging from 78% to 100% (n ranged from 8 to 10). Eight lines may be considered to be segregating for resistance, with 1 to 6 infected plants (n ranged from 5 to 9). Six lines were uniformly resistant, with zero infected among 5 to 9 plants. The lines were planted 25 June 2018 in a field at University of California, Westside Research and Extension Center, Five Points, CA for evaluation of adaptation (plant size and condition) and 11 fruit quality traits in late August and early September. Fruit of one of the lines resembled Group Chandalak subgroup zami, while those of three other lines were characteristic of Group Inodorus subgroup honeydew. Fruit of most of the other lines could be considered members of Group Cantalupensis but could not readily be categorized as members of a particular subgroup. One of the honeydew lines segregated for resistance; the other two were susceptible. Fruit of the segregating honeydew line were smaller than desired for commercial production. Fruit of the six CMV-resistant Group Cantalupensis- type lines resembled subgroup American eastern, and were undersized compared with standard western shipping type cantaloupe (Group Cantalupensis subgroup American western). These tests identified the best CMV subgroup I-resistant melon breeding lines from the 25 Cornell CMV-resistant lines for continued introgression of CMV resistance into western U.S. shipping type cantaloupe and honeydew. Partial Funding provided by the Specialty Crop Block Grant Program at the U.S. Department of Agriculture (USDA) through Grant 14-SCBGP-CA-0006 (The CucCAP Project)
The Mis-Understood History of the Cucumber, Cucumis sativus
Harry Paris1, Jules Janick2
1Agricultural Research Organization, Ramat Yishay, 3009500, Israel. 2Purdue University, West Lafayette, IN, USA
Cucumbers, Cucumis sativus L., are among the most cosmopolitan and widely consumed vegetables, and are eaten fresh, pickled, or cooked. Cucumbers are native to the Indian sub- continent, but the history of their westward diffusion from that region is widely misunderstood. On-line sites and literature are riddled with misinformation, in which there is inane repetition of claims that the ancient Egyptians, Greeks, and Romans were fond of cucumbers. Much of this misunderstanding is derived from a combination of two factors: confusing cucumbers with long-fruited melons, Cucumis melo L., and lack of familiarity of the scientific community with the latter because most of this community, historically, hailed from cooler climates where these melons are ill-adapted for producing good crops. Young, long-fruited melons do closely resemble cucumbers, but have two outstanding traits not possessed by cucumbers: their surfaces are hairy and they are usually longitudinally furrowed. The sikyos of Greek writers, cucumis (or cucumeres) of Latin writers, and qishu’im of Hebrew writers were described as hairy and ancient Egyptian depictions of elongate cucurbits show them to be furrowed. Hence, the ancient Egyptians, Greeks, Romans, and Jews were growing long-fruited melons, not cucumbers. There is no substantiated evidence for the presence of cucumbers, Cucumis sativus, in Europe prior to the medieval period. Also, in contrast to what is stated in English dictionaries, the word “gherkin,” for a small or pickled cucumber, can be traced all the way back to the Indian sub-continent, through late medieval German kychern and Latin chache and circea, to medieval Arabic and Persian khiyar, to Urdu khira and Hindi k(h)ira.
Genome-Wide Diversity for Worldwide Watermelon Collections: Analysis of Population Structure, Haplotype Networks, Selective Sweeps and LD Decay to Characterize Domestication Signals
Umesh Reddy1, Takshay Patel2, Carlos Ortiz1, Venkata Lakshmi Abburi1, Kan Bao3, Xin Wang3, Luis Rivera-Burgos4, Zgangjun Fei3, Todd Wehner2, Padma Nimmakayala1, Shaker Kousik4, W. Patrick Wechter4, Sandra Branham4, Kai-shu Ling4, Robert Jarret5, Cecilia McGregor6, Amnon Levi4
1West Virginia State University, Institute, WV, USA. 2North Carolina State University, Raleigh, NC, USA. 3Boyce Thompson Institute, Ithaca, NY, USA. 4U.S. Department of Agriculture, Agricultural Research Service, U.S. Vegetable Laboratory, Charleston, SC, USA. 5Plant Genetic Resources Conservation Unit, Griffin, GA, USA. 6University of Georgia, Athens, GA, USA
Genotyping by sequencing (GBS) technology was employed to differentiate among 1259 watermelon (C. lanatus) and egusi (C. mucasospermus) accessions. Among these accessions, 1052 United States Plant Introductions (PIs) maintained at the USDA, ARS, Plant Genetic Resources Conservation Unit, Griffin, GA (https://npgsweb.ars- grin.gov/gringlobal/site.aspx?id=22) were sampled to represent the geographical diversity of these taxa. The GBS analysis produced 64,205 SNPs. A minor allele frequency cutoff of >0.01 and a call rate of >70% retained 17,558 SNPs for use in genetic diversity and population structure analyses. Neighbor joining analysis produced a phylogenetic tree that resolved all genotypes into distinct clusters. Population structure analysis confirmed the clustering patterns produced by the neighbor joining and the principal component (PCA) analyses. A core collection containing 384 accessions was identified and selected. Tajima’s D windows for various chromosomes were used to identify selective sweeps. Linkage disequilibrium (LD) decay was estimated across chromosomes and haplotypes were deduced for various LD blocks and selective sweeps. Haplotype networks were generated for various LD blocks characterized across chromosomes. These networks are analogous to genealogical histories and will be useful in efforts to resolve reticulation histories for the materials analyzed. Partial Funding provided by the Specialty Crop Block Grant Program at the U.S. Department of Agriculture (USDA) through Grant 14-SCBGP-CA-0006 (The CucCAP Project)
Characterization of Bitter Gourd Breeding Lines Developed at World Vegetable Center
Suwannee Laenoi, Narinder Dhillon
World Vegetable Center, East and Southeast Asia/Oceania, Kasetsart University, Kamphaeng Saen, Nakhon Pathom 73140, Thailand
The bitter gourd breeding program of the World Vegetable Center (WorldVeg) is globally recognized for development of unique products which are shared with its stakeholders in the private and public sectors. Understanding of different market segments (based on fruit length, shape, color, skin pattern) of WorldVeg’s bitter gourd breeding lines helps seed requesters to select appropriate lines for their breeding program. We evaluated the fruit traits of 100 WorldVeg bitter gourd breeding lines at the World Vegetable Center, East and Southeast Asia/Oceania, Research and Training Station, Kasetsart University, Kamphaeng Saen, Nakhon Pathom, Thailand. Entries were planted in a single block with row spacing of 1.6 m, within-row spacing of 1 m and five plants per plot. Five marketable fruit of each entry were harvested for evaluation of five fruit traits: 1) shape = cylindrical, spindle, elongated; 2) color = green, light green, medium green, dark green; 3) skin pattern = smooth, spiny; 4) length = short (20 cm); 5) bitterness = low, medium, high. Three types of fruit shape were identified among the lines. The majority of lines produced spindle-shaped fruit (69%); cylindrical and elongated fruit were produced by 25% and 6% of the lines. Four distinct fruit skin colors were observed among lines: green (46%), light green (26%), medium green (11%) and dark green (17%). Two fruit skin patterns were observed among lines: spiny (90%) and smooth (10%). Three classes of bitterness among the lines were noted based on taste panels: low (40%), medium (35%) and high (25%). Lines were categorized into three market segments based on the fruit length: short (10%), medium (53%) and long (37%). Fruit pictures of lines were also recorded. The bitter gourd fruit database is available to seed requesters to facilitate ordering lines from the WorldVeg genebank.