Genetic Resources and Vulnerabilities of Major Cucurbit Crops.

The Cucurbitaceae family provides numerous important crops including watermelons (Citrullus lanatus), melons (Cucumis melo), cucumbers (Cucumis sativus), and pumpkins and squashes (Cucurbita spp.). Centers of domestication in Africa, Asia, and the Americas were followed by distribution throughout the world and the evolution of secondary centers of diversity. Each of these crops is challenged by multiple fungal, oomycete, bacterial, and viral diseases and insects that vector disease and cause feeding damage. Cultivated varieties are constrained by market demands, the necessity for climatic adaptations, domestication bottlenecks, and in most cases, limited capacity for interspecific hybridization, creating narrow genetic bases for crop improvement. This analysis of crop vulnerabilities examines the four major cucurbit crops, their uses, challenges, and genetic resources. ex situ germplasm banks, the primary strategy to preserve genetic diversity, have been extensively utilized by cucurbit breeders, especially for resistances to biotic and abiotic stresses. Recent genomic efforts have documented genetic diversity, population structure, and genetic relationships among accessions within collections. Collection size and accessibility are impacted by historical collections, current ability to collect, and ability to store and maintain collections. The biology of cucurbits, with insect-pollinated, outcrossing plants, and large, spreading vines, pose additional challenges for regeneration and maintenance. Our ability to address ongoing and future cucurbit crop vulnerabilities will require a combination of investment, agricultural, and conservation policies, and technological advances to facilitate collection, preservation, and access to critical Cucurbitaceae diversity.

Genetic characterization of melon accessions in the U.S. National Plant Germplasm System and construction of a melon core collection.

Melon (C. melo L.) is an economically important vegetable crop cultivated worldwide. The melon collection in the U.S. National Plant Germplasm System (NPGS) is a valuable resource to conserve natural genetic diversity and provide novel traits for melon breeding. Here we use the genotyping-by-sequencing (GBS) technology to characterize 2083 melon accessions in the NPGS collected from major melon production areas as well as regions where primitive melons exist. Population structure and genetic diversity analyses suggested that C. melo ssp. melo was firstly introduced from the centers of origin, Indian and Pakistan, to Central and West Asia, and then brought to Europe and Americas. C. melo ssp. melo from East Asia was likely derived from C. melo ssp. agrestis in India and Pakistan and displayed a distinct genetic background compared to the rest of ssp. melo accessions from other geographic regions. We developed a core collection of 383 accessions capturing more than 98% of genetic variation in the germplasm, providing a publicly accessible collection for future research and genomics-assisted breeding of melon. Thirty-five morphological characters investigated in the core collection indicated high variability of these characters across accessions in the collection. Genome-wide association studies using the core collection panel identified potentially associated genome regions related to fruit quality and other horticultural traits. This study provides insights into melon origin and domestication, and the constructed core collection and identified genome loci potentially associated with important traits provide valuable resources for future melon research and breeding.

Cucurbit Genomics Database (CuGenDB): a central portal for comparative and functional genomics of cucurbit crops.

The Cucurbitaceae family (cucurbit) includes several economically important crops, such as melon, cucumber, watermelon, pumpkin, squash and gourds. During the past several years, genomic and genetic data have been rapidly accumulated for cucurbits. To store, mine, analyze, integrate and disseminate these large-scale datasets and to provide a central portal for the cucurbit research and breeding community, we have developed the Cucurbit Genomics Database (CuGenDB; http://cucurbitgenomics.org) using the Tripal toolkit. The database currently contains all available genome and expressed sequence tag (EST) sequences, genetic maps, and transcriptome profiles for cucurbit species, as well as sequence annotations, biochemical pathways and comparative genomic analysis results such as synteny blocks and homologous gene pairs between different cucurbit species. A set of analysis and visualization tools and user-friendly query interfaces have been implemented in the database to facilitate the usage of these large-scale data by the community. In particular, two new tools have been developed in the database, a 'SyntenyViewer' to view genome synteny between different cucurbit species and an 'RNA-Seq' module to analyze and visualize gene expression profiles. Both tools have been packed as Tripal extension modules that can be adopted in other genomics databases developed using the Tripal system.

Emergence and epidemiology of Cucurbit yellow stunting disorder virus in the American Desert Southwest, and development of host plant resistance in melon.

Cucurbit yellow stunting disorder virus (CYSDV), emerged in the Sonoran Desert region of the southwestern USA in 2006 and has become well established. Symptoms induced by CYSDV infection include a striking interveinal chlorosis or yellowing and reduced yield and quality. The virus is transmitted by Bemisia tabaci, and the cryptic species MEAM1 has been present in the region since the early 1990s. CYSDV has now become the most economically important of the viruses affecting cucurbit production in the southwestern US. Here, we present a review of recent studies on CYSDV in the southwestern US, with implications for management of this virus throughout the world. Field surveys have established that CYSDV results in late-season infection of spring melon crops with limited economic impact; however, all summer and fall cucurbits become infected shortly after emergence due to high B. tabaci populations and abundant sources of inoculum. Studies have also demonstrated that CYSDV has an extensive host range among crops and weeds prevalent in the region. Recent studies demonstrated considerable variation in virus accumulation and transmission rates among the host plants evaluated as potential reservoirs. Cucurbit hosts had the highest CYSDV titers, were efficient sources for virus acquisition, and showed a positive correlation between titer in source plants and transmission to cucurbit plants. Non-cucurbit hosts had significantly lower CYSDV titers and varied in their capacity to serve as sources for transmission. Experiments demonstrated that multiple factors influence the efficiency with which a host plant species will be a reservoir for vector transmission of CYSDV to crops. Melon PI 313970 was identified as a new source of host plant resistance to CYSDV, in addition to the previously identified TGR 1551 (=PI 482420) and TGR 1937 (=PI 482431). Potential new sources of CYSDV resistance were identified by field screening of ca. 500 melon accessions with naturally occurring inoculum from 2007 through 2012. Host plant resistance to B. tabaci has also been identified in melon germplasm resistant to CYSDV and could be an important factor in reducing losses to CYSDV. Resistance to CYSDV is being transferred to US western shipping type cantaloupe and honeydew.

Genome-Wide Differentiation of Various Melon Horticultural Groups for Use in GWAS for Fruit Firmness and Construction of a High Resolution Genetic Map.

Melon (Cucumis melo L.) is a phenotypically diverse eudicot diploid (2n = 2x = 24) has climacteric and non-climacteric morphotypes and show wide variation for fruit firmness, an important trait for transportation and shelf life. We generated 13,789 SNP markers using genotyping-by-sequencing (GBS) and anchored them to chromosomes to understand genome-wide fixation indices (Fst) between various melon morphotypes and genomewide linkage disequilibrium (LD) decay. The FST between accessions of cantalupensis and inodorus was 0.23. The FST between cantalupensis and various agrestis accessions was in a range of 0.19-0.53 and between inodorus and agrestis accessions was in a range of 0.21-0.59 indicating sporadic to wide ranging introgression. The EM (Expectation Maximization) algorithm was used for estimation of 1436 haplotypes. Average genome-wide LD decay for the melon genome was noted to be 9.27 Kb. In the current research, we focused on the genome-wide divergence underlying diverse melon horticultural groups. A high-resolution genetic map with 7153 loci was constructed. Genome-wide segregation distortion and recombination rate across various chromosomes were characterized. Melon has climacteric and non-climacteric morphotypes and wide variation for fruit firmness, a very important trait for transportation and shelf life. Various levels of QTLs were identified with high to moderate stringency and linked to fruit firmness using both genome-wide association study (GWAS) and biparental mapping. Gene annotation revealed some of the SNPs are located in ß-D-xylosidase, glyoxysomal malate synthase, chloroplastic anthranilate phosphoribosyltransferase, and histidine kinase, the genes that were previously characterized for fruit ripening and softening in other crops.

Cucumis melo endornavirus: Genome organization, host range and co-divergence with the host.

A high molecular weight dsRNA was isolated from a Cucumis melo L. plant (referred to as 'CL01') of an unknown cultivar and completely sequenced. Sequence analyses showed that dsRNA is associated with an endornavirus for which a name Cucumis melo endornavirus (CmEV) is proposed. The genome of CmEV-CL01 consists of 15,078 nt, contains a single, 4939 codons-long ORF and terminates with a stretch of 10 cytosine residues. Comparisons of the putative CmEV-encoded polyprotein with available references in protein databases revealed a unique genome organization characterized by the presence of the following domains: viral helicase Superfamily 1 (Hel-1), three glucosyltransferases (doublet of putative capsular polysaccharide synthesis proteins and a putative C_28_Glycosyltransferase), and an RNA-dependent RNA polymerase (RdRp). The presence of three glycome-related domains of different origin makes the genome organization of CmEV unique among endornaviruses. Phylogenetic analyses of viral RdRp domains showed that CmEV belongs to a specific lineage within the family Endornaviridae made exclusively of plant-infecting endornaviruses. An RT-PCR based survey demonstrated high incidence of CmEV among melon germplasm accession (>87% of tested samples). Analyses of partial genome sequences of CmEV isolates from 26 different melon genotypes suggest fine-tuned virus adaptation and co-divergence with the host. Finally, results of the present study revealed that CmEV is present in plants belonging to three different genera in the family Cucurbitaceae. Such diverse host range is unreported for known endornaviruses and suggests a long history of CmEV association with cucurbits predating their speciation.

Host-Specific Relationship Between Virus Titer and Whitefly Transmission of Cucurbit yellow stunting disorder virus.

Cucurbit yellow stunting disorder virus (CYSDV; genus Crinivirus, family Closteroviridae) was identified in the melon (Cucumis melo) production regions of the desert southwestern United States in fall 2006. It is now well established in the region, where it is transmitted efficiently by the sweet potato whitefly, Bemisia tabaci biotype B (MEAM1). In order to evaluate the spread and establishment of the virus, nearly all spring and fall cucurbit fields planted in the Imperial Valley of California from 2007 to 2009 were surveyed and representative plants were tested for CYSDV infection. Incidence of CYSDV in spring melon fields was initially low and limited to a small number of fields in 2007 but increased to 63% of fields by spring 2009. Virus incidence in fall melon fields was 100% in each year. These results suggested that the virus had become established in native vegetation, weeds, and other crop species, and represented an increasing threat to melon production in the southwestern United States. Therefore, a select set of weed and crop species which grow or are cultivated in the Imperial Valley were evaluated as CYSDV reservoir hosts. For each species, we determined the capacity of CYSDV to accumulate, the relationship between virus titer in these source plants and transmission by whiteflies, as well as subsequent accumulation in inoculated cucurbit plants. Among these hosts, there was considerable variation in virus accumulation and transmission rates. Cucurbit hosts had the highest CYSDV titers, were efficient sources for virus acquisition, and showed a positive correlation between titer in source plants and transmission. Noncucurbit hosts had significantly lower CYSDV titers and varied in their capacity to serve as sources for transmission. CYSDV titers in some noncucurbit source plants, specifically common bean (Phaseolus vulgaris) and shepherd's purse (Capsella bursa-pastoris), were not positively correlated with transmission, demonstrating that additional environmental, physical, or biochemical factors were involved. These results demonstrate that multiple factors influence the efficiency with which a host plant species will be a reservoir for vector transmission of virus to crops.

Responses of Nasonovia ribisnigri (Homoptera: Aphididae) to susceptible and resistant lettuce.

Nymphs and alates of aphid Nasonovia ribisnigri (Mosley) (Homoptera: Aphididae) were tested on 10 lettuce cultivars with N. ribisnigri resistance gene Nr and 18 cultivars without the resistance gene in various bioassays. Bioassays used whole plants, leaf discs, or leaf cages to determine susceptibility of commercial lettuce cultivars to N. ribisnigri infestation and to evaluate screening methods for breeding lettuce resistance to N. ribisnigri. Resistant and susceptible plants were separated in 3 d when using whole plant bioassays. Long-term (> or =7 d) no-choice tests using leaf cages or whole plants resulted in no survival of N. ribisnigri on resistant plants, indicating great promise of the Nr gene for management of N. ribisnigri. Effective screening was achieved in both no-choice tests where resistant or susceptible intact plants were tested separately in groups or individually and in choice tests where susceptible and resistant plants were intermixed. Leaf discs bioassays were not suitable for resistance screening. All lettuce cultivars without the resistance gene were suitable hosts for N. ribisnigri, indicating the great importance of this pest to lettuce production and the urgency in developing resistant lettuce cultivars to manage N. ribisnigri.

Effect of Planting Date, Cultivar, and Stage of Plant Development on Incidence of Fusarium Wilt of Lettuce in Desert Production Fields.

Fusarium wilt of lettuce, first recognized in Japan in 1955, has since been discovered in the United States (California in 1990, Arizona in 2001), Iran (1995), Taiwan (1998), and Italy (2001). In Arizona, the causal agent, Fusarium oxysporum f. sp. lactucae, has been recovered from lettuce plants in 27 different lettuce fields during the 2001 to 2003 production seasons. Studies were initiated to examine the impact of planting date, cultivar, and stage of plant development on the incidence of disease in the field. In 2002 and 2003, tested lettuce cultivars were sown in at least one of the following planting windows; early-season (September), mid-season (October), and late-season (December). Within each planting window, significant differences in disease incidence among lettuce cultivars were noted at plant maturity. The mean incidence of Fusarium wilt on cultivars sown in September, October, and December was 92.3, 15.1, and 2.0%, respectively, in 2002 and 74.2, 5.1, and 0.7%, respectively, in 2003. The mean soil temperatures at the10-cm depth during the September, October, and December plantings for both years were 26, 14, and 14°C, respectively. Initial symptoms of Fusarium wilt were apparent as early as 14 days after seeding, with increasing incidence of disease noted as the crop developed and reached maturity. Among all lettuce cultivars planted in September, only one and two cultivars of romaine in 2002 and 2003, respectively, reached maturity with ≤5% incidence of Fusarium wilt, whereas the lowest incidence of disease among crisphead, green leaf, red leaf, or butterhead cultivars was 73.7, 27.0, 20.2, and 65.7%, respectively, in 2002 and 62.1, 29.0, 100, and 100%, respectively, in 2003. For October plantings, all romaine cultivars had ≤5% incidence of Fusarium wilt at maturity, whereas disease incidence among tested cultivars of crisphead lettuce in 2002 and 2003 ranged from 0.8 to 66.8% and 0.3 to 43.3%, respectively. When planted in December, 82 and 88% of tested cultivars, including all romaine entries, reached maturity with ≤1% incidence of Fusarium wilt. Selection of appropriate lettuce cultivars and planting times should allow successful production of lettuce in the southwestern Arizona production region with minimal or no incidence of disease in fields infested with F. oxysporum f. sp. lactucae. On the other hand, successful production of lettuce in infested fields when temperatures favor disease development will not be possible until lettuce cultivars are developed that possess high tolerance or resistance to the pathogen.