Genetic diversity, population structure, and selection of breeder germplasm subsets from the USDA sweetpotato (Ipomoea batatas) collection.

Sweetpotato (Ipomoea batatas) is the sixth most important food crop and plays a critical role in maintaining food security worldwide. Support for sweetpotato improvement research in breeding and genetics programs, and maintenance of sweetpotato germplasm collections is essential for preserving food security for future generations. Germplasm collections seek to preserve phenotypic and genotypic diversity through accession characterization. However, due to its genetic complexity, high heterogeneity, polyploid genome, phenotypic plasticity, and high flower production variability, sweetpotato genetic characterization is challenging. Here, we characterize the genetic diversity and population structure of 604 accessions from the sweetpotato germplasm collection maintained by the United States Department of Agriculture (USDA), Agricultural Research Service (ARS), Plant Genetic Resources Conservation Unit (PGRCU) in Griffin, Georgia, United States. Using the genotyping-by-sequencing platform (GBSpoly) and bioinformatic pipelines (ngsComposer and GBSapp), a total of 102,870 polymorphic SNPs with hexaploid dosage calls were identified from the 604 accessions. Discriminant analysis of principal components (DAPC) and Bayesian clustering identified six unique genetic groupings across seven broad geographic regions. Genetic diversity analyses using the hexaploid data set revealed ample genetic diversity among the analyzed collection in concordance with previous analyses. Following population structure and diversity analyses, breeder germplasm subsets of 24, 48, 96, and 384 accessions were established using K-means clustering with manual selection to maintain phenotypic and genotypic diversity. The genetic characterization of the PGRCU sweetpotato germplasm collection and breeder germplasm subsets developed in this study provide the foundation for future association studies and serve as precursors toward phenotyping studies aimed at linking genotype with phenotype.

Large-Scale Seedling Grow-Out Experiments Do Not Support Seed Transmission of Sweet Potato Leaf Curl Virus in Sweet Potato.

Sweet potato leaf curl virus (SPLCV) threatens global sweet potato production. SPLCV is transmitted by Bemisia tabaci or via infected vegetative planting materials; however, SPLCV was suggested to be seed transmissible, which is a characteristic that is disputed for geminiviruses. The objective of this study was to revisit the validity of seed transmission of SPLCV in sweet potato. Using large-scale grow-out of sweet potato seedlings from SPLCV-contaminated seeds over 4 consecutive years, approximately 23,034 sweet potato seedlings of 118 genotype entries were evaluated. All seedlings germinating in a greenhouse under insect-proof conditions or in a growth chamber were free of SPLCV; however, a few seedlings grown in an open bench greenhouse lacking insect exclusion tested positive for SPLCV. Inspection of these seedlings revealed that B. tabaci had infiltrated the greenhouse. Therefore, transmission experiments were conducted using B. tabaci MEAM1, demonstrating successful vector transmission of SPLCV to sweet potato. Additionally, tests on contaminated seed coats and germinating cotyledons demonstrated that SPLCV contaminated a high percentage of seed coats collected from infected maternal plants, but SPLCV was never detected in emerging cotyledons. Based on the results of grow-out experiments, seed coat and cotyledon tests, and vector transmission experiments, we conclude that SPLCV is not seed transmitted in sweet potato.

Genetic Diversity and Population Structure of the USDA Sweetpotato (Ipomoea batatas) Germplasm Collections Using GBSpoly.

Sweetpotato (Ipomoea batatas) plays a critical role in food security and is the most important root crop worldwide following potatoes and cassava. In the United States (US), it is valued at over $700 million USD. There are two sweetpotato germplasm collections (Plant Genetic Resources Conservation Unit and US Vegetable Laboratory) maintained by the USDA, ARS for sweetpotato crop improvement. To date, no genome-wide assessment of genetic diversity within these collections has been reported in the published literature. In our study, population structure and genetic diversity of 417 USDA sweetpotato accessions originating from 8 broad geographical regions (Africa, Australia, Caribbean, Central America, Far East, North America, Pacific Islands, and South America) were determined using single nucleotide polymorphisms (SNPs) identified with a genotyping-by-sequencing (GBS) protocol, GBSpoly, optimized for highly heterozygous and polyploid species. Population structure using Bayesian clustering analyses (STRUCTURE) with 32,784 segregating SNPs grouped the accessions into four genetic groups and indicated a high degree of mixed ancestry. A neighbor-joining cladogram and principal components analysis based on a pairwise genetic distance matrix of the accessions supported the population structure analysis. Pairwise FST values between broad geographical regions based on the origin of accessions ranged from 0.017 (Far East - Pacific Islands) to 0.110 (Australia - South America) and supported the clustering of accessions based on genetic distance. The markers developed for use with this collection of accessions provide an important genomic resource for the sweetpotato community, and contribute to our understanding of the genetic diversity present within the US sweetpotato collection and the species.

Insect resistance in traditional and heirloom sweetpotato varieties.

Fifty-nine sweetpotato cultivars, including 16 heirlooms, 11 near-heirlooms (developed in the 1960s and 1970s), 19 cultivars from the 1980s, and 13 modern varieties (since 1990), were evaluated for resistance to soil insects in field experiments during 2010-2011 at the U.S. Department of Agriculture-Agriculture Research Service, U.S. Vegetable Laboratory (USDA-ARS, USVL), Charleston, SC. These experiments included two insect-susceptible control cultivars ('Beauregard' and 'SC1149-19') and four insect-resistant control cultivars ('Charleston Scarlet,''Regal,' 'Ruddy,' and 'Sumor') that were developed by the USDA-ARS, USVL sweetpotato breeding program. Sweetpotato genotypes differed significantly in resistance measured by the overall percentage of injured roots, WDS (Wireworm, Diabrotica, and Systena) index, the percentage of roots damaged by the sweetpotato weevil (Cylas formicarius F.), the percentage of roots damaged by the sweetpotato flea beetle (Chaetonema confinis Crotch), and the percentage of roots damaged by white grub larvae (including Plectris aliena Chapin and Phyllophaga spp.). Twenty-three sweetpotato cultivars had a lower percentage of injured roots than the susceptible control genotype, SC1149-19, while 14 varieties had a lower percentage of injured roots than Beauregard, one of the leading commercial orange-fleshed cultivars in the United States. Over the 2-yr period, Ruddy (7.6%) had the lowest percentage of injured roots and 'Carolina Ruby' (84.6%) the highest percentage of injured roots. Carolina Ruby (1.07) also had the highest WDS index, but 15 genotypes had a significantly lower WDS index than either susceptible control, SC1149-19 (1.03) or Beauregard (0.82). Ruddy (0.07) and 'Murasaki-29' (0.09) had the lowest WDS indices. Forty-five genotypes had a significantly lower percentage infestation by flea beetles than SC1149-19 (12.3%), and the highest level of flea beetle infestation was for 'Bonita' (18.9%). The highest percent white grub infestation was for 'Caromex' (19.6%), however none of the genotypes had significantly less white grubs than the susceptible controls. The highest infestation of sweetpotato weevils was observed for SC1149-19 (17.9%), while 29 genotypes had significantly lower percentage of sweetpotato weevil infestation than SC1149-19. The moderate to high levels of resistance to soil insect pests exhibited by many of these traditional and heirloom cultivars may provide useful sources of germplasm for sweetpotato breeding programs.

Insect resistance in sweetpotato plant introduction accessions.

Fifty-five sweetpotato cultivars, experimental breeding clones, and plant introduction (PI) accessions were evaluated in 17 field experiments at the USDA, ARS, U.S. Vegetable Laboratory (Charleston, SC; 12 evaluations, 1997-2010), the Clemson University, Edisto Research and Education Center (Blackville, SC; two evaluations, 1998-1999), and the University of Florida, Tropical Research and Education Center (Homestead, FL; three evaluations, 2005-2007). These experiments included two insect-susceptible control entries ('Beauregard' and 'SC1149-19') and three insect-resistant control cultivars ('Regal,' 'Ruddy,' and 'Sumor'). At each location, genotypes differed significantly in the percentage of uninjured roots WDS (wireworm, Diabrotica, Systena) index, the percentage of roots damaged by the sweetpotato weevil (Cylas formicarius (F.)), the percentage of roots damaged by the sweetpotato flea beetle (Chaetocnema confinis Crotch), and the percentage of roots damaged by white grub larvae (including Plectris aliena Chapin and Phyllophaga spp.). 'SC1149-19' had a significantly lower percentage of uninjured roots, a significantly higher WDS index rating, and significantly higher percentages of infestation by flea beetles, grubs, and sweetpotato weevils than most other sweetpotato genotypes in this study. In addition, 43 of 55 genotypes had significantly less overall insect damage than 'Beauregard,' one of the leading commercial orange-fleshed cultivars in the United States. Ten genotypes had significantly less insect injury than 'Picadito,' a commercial boniato-type sweetpotato grown extensively in southern Florida. Many of these sweetpotato genotypes have high levels of resistance to soil insect pests, and they may be useful as sources of insect resistance for use in sweetpotato breeding programs.

Effects of a killed-cover crop mulching system on sweetpotato production, soil pests, and insect predators in South Carolina.

Sweetpotatoes, Ipomoea batatas (L.) Lam. (Convolvulaceae), are typically grown on bare soil where weeds and erosion can be serious problems. Conservation tillage systems using cover crop residues as mulch can help reduce these problems, but little is known about how conservation tillage affects yield and quality of sweetpotato or how these systems impact populations of beneficial and pest insects. Therefore, field experiments were conducted at the U.S. Vegetable Laboratory, Charleston, SC, in 2002-2004 to evaluate production of sweetpotatoes in conventional tillage versus a conservation tillage system by using an oat (Avena sativa L. (Poaceae)-crimson clover (Trifolium incarnatum L.) (Fabaceae) killed-cover crop (KCC) mulch. The four main treatments were 1) conventional tillage, hand-weeded; 2) KCC, hand-weeded; 3) conventional tillage, weedy; and 4) KCC, weedy. Each main plot was divided into three subplots, whose treatments were sweetpotato genotypes: 'Ruddy', which is resistant to soil insect pests; and 'SC1149-19' and 'Beauregard', which are susceptible to soil insect pests. For both the KCC and conventional tillage systems, sweetpotato yields were higher in plots that received hand weeding than in weedy plots. Orthogonal contrasts revealed a significant effect of tillage treatment (conventional tillage versus KCC) on yield in two of the 3 yr. Ruddy remained resistant to injury by soil insect pests in both cropping systems; and it consistently had significantly higher percentages of clean roots and less damage by wireworm-Diabrotica-Systena complex, sweetpotato flea beetles, grubs, and sweetpotato weevils than the two susceptible genotypes. In general, injury to sweetpotato roots by soil insect pests was not significantly higher in the KCC plots than in the conventionally tilled plots. Also, more fire ants, rove beetles, and carabid beetle were captured by pitfall traps in the KCC plots than in the conventional tillage plots during at least 1 yr of the study. This study suggests that a sweetpotatoes can be successfully grown under a killed-cover crop production system.

Resistance of sweetpotato genotypes to adult Diabrotica beetles.

Production of sweetpotatoes, Ipomoea batatas (L.) Lam. (Convolvulaceae), is limited by several insect pests, including Diabrotica spp. (Coleoptera: Chrysomelidae), and new integrated pest management (IPM) techniques for this crop are needed. Host plant resistance is one attractive approach that fits well into IPM programs. A host plant resistance research program typically depends on reliable bioassay procedures to streamline evaluation of germplasm. Thus, a bioassay technique was developed for evaluating sweetpotato germplasm by using adults of the banded cucumber beetle, Diabrotica balteata LeConte, and spotted cucumber beetle, Diabrotica undecimpunctata howardi Barber. A single beetle was placed on a piece of sweetpotato peel (periderm and cortex with stele removed) that was embedded periderm-side up in plaster in a petri dish. Feeding and longevity of insects on 30 sweetpotato genotypes were evaluated in two experiments by using this procedure. Adult longevity ranged from 7 to 11 d for starved individuals to 211 d for beetles fed a dry artificial diet. Longevity of banded cucumber beetles that fed on sweetpotato peels ranged from 12 d for the most-resistant genotype to 123 d for SC1149-19, a susceptible control cultivar. Longevity of spotted cucumber beetles was slightly shorter than longevity of banded cucumber beetles. For the most resistant sweetpotato genotypes, both Diabrotica species exhibited a significant delay in initiation of feeding, and more beetles died on these genotypes before they had fed. Both antibiosis and nonpreference (antixenosis) are important mechanisms of resistance in sweetpotato genotypes. This bioassay was consistent with field results, indicating that this technique could be useful for evaluating resistance to Diabrotica spp. in sweetpotato genotypes.

Improved dry-fleshed sweetpotato genotypes resistant to insect pests.

Thirty-five mostly dry-fleshed sweetpotato, Ipomoea batatas (L.) Lam. (Convolvulaceae), genotypes from the USDA-ARS/Clemson University sweetpotato breeding program were evaluated in nine field experiments at the U.S. Vegetable Laboratory, Charleston, SC, from 1998 to 2004. There were highly significant entry effects for percentage of uninjured roots; wireworm, Diabrotica, and Systena (WDS) index; percentage of roots damaged by sweetpotato weevil, Cylas formicarius elegantulus (Summers); percentage of roots damaged by sweetpotato flea beetle, Chaetocnema confinis Crotch); and percentage of roots damaged by white grub larvae (primarily Plectris aliena Chapin). The susceptible control, 'SC1149-19', had a significantly lower percentage of uninjured roots, a significantly higher WDS rating, and higher percentage infestations of flea beetle, grubs, and sweetpotato weevils than all other sweetpotato entries in this study. Twenty-seven genotypes had significantly less insect damage than 'Beauregard', the leading commercial orange-fleshed cultivar in the United States. In addition, 11 genotypes had significantly less insect injury than 'Picadito', a commercial boniato-type sweetpotato grown extensively in southern Florida. Overall, no genotypes were more resistant to soil insect pests than the resistant checks 'Sumor' and 'Regal'. Many of the advanced dry-flesh sweetpotato genotypes had high levels of resistance to soil insect pests, and they represent a useful source of advanced germplasm for use in sweetpotato breeding programs.

Monitoring cucumber beetles in sweetpotato and cucurbits with kairomone-baited traps.

Seven kairomone formulations (Trécé, Inc., Salinas, CA) were evaluated for their effectiveness as attractants for luring three species of cucumber beetles into Pherocon CRW traps (Trécé, Inc.) in cucurbit and sweetpotato fields. The spotted cucumber beetle, Diabrotica undecimpunctata howardi Barber; the banded cucumber beetle, Diabrotica balteata LeConte; and the striped cucumber beetle, Acalymma vittatum (F.), were captured in this study. TRE8276 (TIC mixture: 500 mg of 1,2,4-trimethoxybenzene, 500 mg of indole, and 500 mg of trans-cinnamaldeyde) and TRE8336 (500 mg of 1,2,4-trimethoxybenzene, 500 mg of trans-cinnamaldeyde, 500 mg of 4-methoxyphenethanol) were the most effective lures for spotted and striped cucumber beetles. None of the kairomone lures was very effective for attracting banded cucumber beetles. Three population peaks of spotted cucumber beetles were observed in cucurbit and sweetpotato fields at the U.S. Vegetable Laboratory (Charleston, SC). The efficacy of TRE8276 declined rapidly after 2 wk in the field. An improved design of the Pherocon CRW trap, with a yellow bottom and more-tapered top section, was more effective for capturing cucumber beetles than the original trap design made entirely of clear plastic. Banded cucumber beetles were not captured in sweetpotato fields at inland locations in North Carolina or South Carolina.

Quantity and potential biological activity of caffeic acid in sweet potato [Ipomoea batatas (L.) Lam.] storage root periderm.

The caffeic acid content of storage root periderm and cortex tissues of genetically diverse sweet potato [Ipomoea batatas (L.) Lam.] cultivars and breeding clones was quantified by high-performance liquid chromatography. Periderm caffeic acid content of the clones ranged from 0.008 to 7.97 mg/g dry weight, whereas the highest cortex content was 0.047 mg/g. Clones varied greatly in periderm caffeic acid content in all experiments, but there were also differences between experiments in content averaged for all clones. This indicates that periderm caffeic acid content is subject to genetic and environmental influences. Caffeic acid inhibited the growth of four sweet potato pathogenic fungi and germination of proso millet seeds in bioassays. Inhibitory activity in the bioassays suggests that high periderm caffeic acid levels contribute to the storage root defense chemistry of some sweet potato genotypes.

Increase in Populations of Rhizoctonia solani and Wirestem of Collard with Velvet Bean Cover Crop Mulch.

Velvet bean has been used traditionally as a summer cover crop in the southeastern United States. We investigated the use of killed velvet bean as a cover crop mulch left on the soil surface before collard was transplanted in the fall. Control treatments were weed-free fallow and velvet bean that was killed and disked into the soil before transplanting. Incidence of wirestem, caused by Rhizoctonia solani, reached a maximum of 25% in 2000 but only 4% in 2001 in cover crop mulch treatments. Nevertheless, in both years, the infection rate, area under the disease progress curve, and final incidence were significantly greater with cover crop mulch than in the fallow or disked treatments. Wirestem incidence did not differ between the disked and fallow treatments in either year. Populations of R. solani in soil were greater after cover crop mulch than in fallow plots in both years and greater in the disked treatment than in fallow soil in 2000 but not 2001. Velvet bean does not appear to be suitable as an organic mulch for fall collard production, but could be used as a summer cover crop if disked into the soil before transplanting collard.

Survival and development of Heliothis virescens (Lepidoptera: Noctuidae) larvae on isogenic tobacco lines with different levels of alkaloids.

Levels of pyridine alkaloids were measured in 18 tobacco, Nicotiana tabacum L., entries from three parental isolines ('NC 95', 'SC 58', and 'Coker 139'), grown at Tifton, GA, Florence, SC, and Oxford, NC, in 1991. Levels of alkaloids in bud leaves (first fully unfolded leaf below the apical leaf bud) were negatively correlated to natural infestation ratings of tobacco budworm larvae, Heliothis virescens (F.), 7 wk after transplanting. For artificially infested bud leaves at Oxford, there was a significant negative correlation between levels of total alkaloids and larval weights after 1 wk of feeding. In 1992, four entries from the 'NC 95' isoline were grown at Oxford, and samples for alkaloid analyses were taken every 2 wk at several leaf positions on each plant. During weeks 4, 8, 12, and 16, second instar tobacco budworms were caged on individual, intact leaves inside perforated plastic bags in the field. The survival and development of tobacco budworm larvae after 1 wk were negatively correlated with levels of alkaloids at the various leaf positions. Larvae survived better and grew faster on the bud leaves of each entry where alkaloid levels were lower than they did on leaves further down the stalk where alkaloid levels were higher. More larvae survived on the lower leaves of the low alkaloid lines than on the lower leaves of the high alkaloid lines. Even moderate increases in pyridine alkaloids had negative effects on tobacco budworm survival and development. Nicotine constituted >97% of the pyridine alkaloids in the 'NC95' isoline each year.