Abstract
Single nucleotide polymorphisms (SNPs) were genotyped using a high-density array and DNAs from individual plants of important onion (Allium cepa L.) populations from major production regions and from the likely progenitor of onion, Allium vavilovii Popov et Vved. Genotypes at 1226 SNPs were used to estimate genetic relationships among these populations and revealed close associations among onions grown in Europe and those in North America, South America, and eastern Asia, supporting paths of introduction from Europe to the Americas and Asia. ‘Nasik Red’ is a population grown on the Indian subcontinent and was divergent from onions of European origin. Frequencies of SNPs among and within populations were used as a measure of informativeness, and 199 commonly polymorphic SNPs were identified distributed across the eight chromosomes of onion. These SNPs will be useful for estimations of relatedness among broader collections of onion populations, mapping of important phenotypes, fingerprinting of inbred lines and hybrids, and quality control of seed lots.
Onion is one of the world’s most widely grown and economically important vegetables. The likely progenitor of onion, A. vavilovii, grows wild in and around the Kopet Dag Mountains of Turkmenistan and Iran (Bradeen and Havey, 1995; Gurushidze et al., 2007; Hanelt, 1985; Havey, 1992; Van Raamsdonk et al., 1992). From this region, onion was introduced into the Mediterranean region. Onion cultivation subsequently spread throughout Europe; from Europe to North America, South America, Australia, sub-Saharan Africa, and New Zealand; and later from North America to Japan and eastern Asia (Goldman et al., 2001). During its worldwide dispersal, onion was selected for response to different daylength requirements for bulbing and flowering and became a biennial vegetable (Brewster, 1994). Although the length of night is important for bulbing and flowering, onion populations are classified based on the minimal daylength necessary to induce bulbing, such as short-day (≈12–13 h), intermediate-day (≈13–14 h), or long-day (>14 h) types. Production of short-day onion cultivars under longer days will produce small bulbs because of early maturation; conversely long-day cultivars grown under shorter days show little to no bulbing. Therefore, cultivars of different daylength responses may represent divergent germplasm pools within the cultivated onion.
Single nucleotide polymorphisms are robust, codominant genetic markers that commonly occur in the genomes of cultivated plants. In previous research, we completed transcriptome sequencing of onion inbreds and identified 2285 SNPs amenable for genotyping using the KASPar platform (Duangjit et al., 2013). Subsets of these SNPs were mapped in three segregating families (Damon and Havey, 2014; Duangjit et al., 2013). In this study, we genotyped 1692 of these SNPs using a high-density array and DNAs from random plants from A. vavilovii and 14 open-pollinated (OP) onion populations to determine how commonly polymorphisms exist in cultivated germplasm (informativeness) and to estimate relationships among the populations. We identified a set of commonly polymorphic SNPs and produced a consensus map of these SNPs across the eight chromosomes of onion.
Materials and Methods
Origins of onion populations are listed in Table 1. Doubled haploid (DH) CU066619 was included as a control to identify heterozygous SNPs (potentially from paralogous sequences) and its origin was reported by Hyde et al. (2012). DNAs were isolated using a midi-preparation (NucleoSpin Plant II kit; Macherey-Nagel, Düren, Germany) from five random plants from each of the 14 OP populations and two plants of A. vavilovii. DNA of DH CU066619 was isolated from pooled leaf tissue of ≈25 seedlings. DNA concentrations were determined spectrophotometrically and intactness by electrophoresis through 1% agarose gels.
Origins, cytoplasms, and percent heterozygous loci for onion populations genotyped for 1226 single nucleotide polymorphisms (SNPs).
Cytoplasm of at least 10 individual plants from each of the 14 onion populations were classified as normal (N) male-fertile or male-sterile (S) using an indel in the chloroplast accD gene described by Von Kohn et al. (2013). Twenty-microliter polymerase chain reactions used primers 5′-AGAATGAGGAGCAGGAAA and 5′-AGTCGTGATTGTTACTCTT and conditions of 50 ng of DNA, 0.25 μm of each primer, and 5x HOT FIREPol DNA Polymerase EvaGreen HRM mix (Solis BioDyne, Tartu, Estonia). Differences in melt curves were visualized using high-resolution melting (LightCycler 480 II; Roche, Indianapolis, IN).
An Infinium array (Illumina, San Diego, CA) was constructed by Eurofins-BioDiagnostics (River Falls, WI) using 1692 SNPs (Supplemental Table 1) identified from the onion transcriptome (Duangjit et al., 2013). For array hybridizations, DNA concentrations were determined using Picogreen according to the manufacturer’s instructions (Molecular Probes, Eugene, OR) and adjusted to 50 ng·μL−1. Four microliters of DNA was used for hybridization according to the Infinium HD Assay Ultra manual (Illumina). After scanning BeadChips, the raw data were analyzed using the Genome Studio genotyping module (version 2.0.3; Illumina) clustering algorithm. The auto clustering process was monitored manually and corrected when needed to produce the most accurate genotyping results. Clustering and genotype calls were proofed by a second person to eliminate any miscalling of the markers (standard protocol of Eurofins-BioDiagnostics).
Genetic distances among A. vavilovii and the OP onion populations were estimated using 1226 SNPs (Supplemental Table 2) and Nei’s 72 coefficient of genetic diversity, and a dendrogram was generated by the unweighted paired group method algorithum (UPGMA) using Numerical Taxonomy and Multivariate Analysis System (NTSYS version 2.2; Exeter Software, Setauket, NY). Commonly polymorphic SNPs across the eight onion chromosomes were identified from a consensus map created using the JoinMap software version 4.0 (Van Ooijen, 2006) and segregations from three onion families (Damon and Havey, 2014; Duangjit et al., 2013).
Results and Discussion
Five randomly selected plants from each of 14 OP onion populations and two plants from A. vavilovii, and pooled DNA from seedlings of DH CU066619 were genotyped for 1692 SNPs. For the OP populations, a sample size of five plants should reveal alleles with frequencies greater than 0.25 at the 95% confidence level (Mansur et al., 1990). Two SNPs were heterozygous in DH CU066619 and were eliminated from analyses because they cannot be allelic. Of the remaining 1690 SNPs, 378 were discarded because of frequently missing genotypes and 86 discarded because they were monomorphic across all DNAs. The remaining 1226 SNPs provided genotypes across all populations (Supplemental Table 2). Overall heterozygosity averaged across the 14 OP populations was relatively low at 23.5% (Table 1). Onion shows significant inbreeding depression (Jones and Davis, 1944) and populations are generally considered to be highly heterozygous. McCallum et al. (2008) genotyped simple sequence repeats (SSRs) using bulked DNA from onion populations and estimated median heterozygosity for OP onion populations at 70%. In contrast, Baldwin et al. (2012) isolated DNA from individual plants from diverse onion populations and reported relatively low heterozygosity at ≈22%. One explanation for the low heterozygosity revealed by this study could be the sample size of five plants per population. McCallum et al. (2008) estimated that a relatively rare allele would amplify and be detected if its frequency was greater than 5%; in our study, we would not confidently detect an allele if its frequency were less than 25%. Another explanation for the low heterozygosity observed in this study and that of Baldwin et al. (2012) may be that relatively few individuals were used during seed increases of these populations, and genetic drift may have reduced heterozygosity. However, the commercially grown cultivars Senshu-Ki Early and Red Pinoy had similar levels of estimated heterozygosity at 30.1% and 22.5%, respectively (Table 1). Allium vavilovii, the likely progenitor of onion (Fritsch et al., 2001; Gurushidze et al., 2007; Havey, 1992, 1997), possessed a similar level of heterozygosity at 25.7%. Allium vavilovii grows naturally in the Kopet Dag (Turkmen-Khorasan) mountainous region of Turkmenistan and Iran (Hanelt, 1985), and germplasm accessions may trace back to relatively few isolated plants, or bottlenecks may have occurred during seed increases of this wild species.
Plants from A. vavilovii and all but one of the OP populations possessed N cytoplasm as determined the accD polymorphism (Table 1). ‘Pukekohe Longkeeper’ (PLK) possessed only S cytoplasm as previously reported (Havey, 1993). ‘Red Creole’ and ‘White Creole’ are closely related OP populations and possessed both N and S cytoplasms. The Japanese OP population ‘Senshu-Ki Early’ possessed both N and S cytoplasms; this onion was likely introduced into Japan from North America, and North American OP populations can possess both cytoplasms (Havey, 1993).
Onion populations were selected as representative of key production regions or as important founder populations, and relationships were estimated using Nei’s 72 coefficient of genetic diversity (Table 2) and UPGMA (Fig. 1). ‘Babosa’ and ‘Valencia’ are yellow onions originating from Spain (Goldman et al., 2001). ‘Babosa’ is a short-day onion producing yellow bulbs of relatively low pungency and ‘Valencia’ (potentially synonymous with ‘Cebolla Valenciana’) produces yellow bulbs with relatively good storage ability. In 1925, a population named ‘Valencia Grano 9452’ was introduced into the United States, initially was grown and selected in New Mexico, and eventually re-named as ‘Early Grano’ (Magruder et al., 1941) or ‘New Mexico Early Grano’ (Goldman et al., 2001). Uncertainty exits whether ‘Texas Early Grano’ (TEG) traces back to the ‘Babosa’ or ‘Valencia’ type. TEG has been reported to be an early-maturing selection from ‘Early Grano’ (Ewart, 1945) or ‘Grano (Babosa)’ (Perry, 1949). Bulbs of TEG are top shaped (Magruder et al., 1941) and of lower pungency, more similar to ‘Babosa’ than ‘Valencia’ types. Genotyping of SNPs revealed a close relationship between TEG and ‘Babosa’ (Fig. 1), supporting ‘Babosa’ as the origin of TEG.
Genetic distances estimated among onion populations and Allium vavilovii using allele frequencies at 1226 single nucleotide polymorphisms and the Nei’s 72 coefficient of genetic diversity. Larger numbers indicate greater genetic distances. Abbreviations of population names are listed in Table 1.
Relationships among onion populations estimated using unweighted pair group method with arithmetic mean and allele frequencies at 1226 single nucleotide polymorphisms genotyped using DNA from individual plants of each population. Abbreviated population names are listed in Table 1.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 143, 1; 10.21273/JASHS04277-17
‘Sweet Spanish Valencia’ is representative of long-day onions grown in regions of northwestern United States that produce bulbs with relatively low pungency and good storage ability. ‘Sweet Spanish’ onions likely trace back to introduction(s) into California and Utah from Spain in the late 1800s and early 1900s (Goldman et al., 2001; Magruder et al., 1941). Consistent with this observation, ‘Sweet Spanish Valencia’ was placed close to the Spanish population ‘Cebolla Valenciana’. ‘Valcatorce’ is an important onion grown in Argentina and phenotypically similar to yellow storage onions from Spain; this OP population was placed closely to both ‘Cebolla Valenciana’ and ‘Sweet Spanish Valencia’.
‘Yellow Bermuda’ was introduced from Italy (Magruder et al., 1941) into Texas via the Canary Islands in 1898 (Perry, 1949) and is the source of the male-sterile parent for ‘Granex’ hybrids (Goldman et al., 2001; Havey and Bark, 1994). This onion has relatively flat bulbs of low pungency that are phenotypically different from the ‘Babosa’ and ‘Valenciana’ types and grouped with onions of Spanish origin (Fig. 1).
‘Red Creole’ and ‘White Creole’ are short-day onions with relatively high soluble solids and were likely introduced into southern United States from Italy (Magruder et al., 1941). ‘White Creole’ is an important founder population of short-day dehydrator onions in North America. ‘Red Creole’ and ‘White Creole’ are very closely related (Fig. 1), and ‘White Creole’ was likely selected from ‘Red Creole’ because it is homozygous recessive at the C locus (El-Shafie and Davis, 1967). ‘Red Pinoy’ is a pungent short-day onion grown in the Philippines and likely shares a common origin with ‘Red Creole’ and ‘White Creole’. However, we did not detect S cytoplasmic plants in ‘Red Pinoy’ as in closely related ‘Red Creole’ and ‘White Creole’ (Table 1).
For longer day onions, ‘Southport Yellow Globe’ (SYG) is a well-storing onion introduced into northeastern United States from northern Europe; cultivation of this type of onion spread across northern United States and southern Canada and subsequently introduced into northern Japan and eastern Asia (Goldman et al., 2001). Consistent with this scenario is the placement of ‘Early Senshu-Ki’ in this long-day group. ‘Pukekohe Longkeeper’ from New Zealand and ‘Wolska’ from Poland are both well-storing, pungent, and long-day onions, and were closely related to North American onions of similar phenotypes such as SYG (Bark and Havey, 1995).
‘Nasik Red’ is a pungent short-day onion grown in Maharashtra state of India and was placed distantly from populations of European origin. The average Nei’s 72 coefficient of genetic diversity averaged across onion populations was highest for ‘Nasik Red’ (0.246) and A. vavilovii (0.263). Baldwin et al. (2012), Khar et al. (2011), and McCallum et al. (2008) used primarily SSRs to estimate relationships among onion populations and observed that onion population(s) from the Indian subcontinent were divergent relative to onions of European origin. Together, these studies indicate that Indian onions may trace back to an independent path of introduction from central Asia and a comprehensive evaluation of onion germplasm from south Asia should be undertaken to assess genetic diversity relative to onions from Europe and related populations in the Western Hemisphere and Asia.
Genetic maps built using wide crosses may possess polymorphisms that are relatively rare in elite germplasm. We generated a consensus map (Table 3) of 199 SNPs commonly polymorphic across onion populations using segregations from three families (Damon and Havey, 2014; Duangjit et al., 2013). These SNPs are distributed across the eight chromosomes of onion and should be useful for estimation of relatedness among broader collections of onion populations, mapping of important phenotypes, fingerprinting of inbred lines and hybrids, and quality control of seed lots.
Genetic map positions of single nucleotide polymorphisms (SNPs) and mean allelic frequencies across 14 onion populations and Allium vavilovii.
Literature Cited
Baldwin, S., Pither-Joyce, M., Wright, K., Chen, L. & McCallum, J. 2012 Development of robust genomic simple sequence repeat markers for estimation of genetic diversity within and among bulb onion (Allium cepa L.) populations Mol. Breed. 30 1401 1411
Bark, O.H. & Havey, M.J. 1995 Similarities and relationships among open-pollinated populations of the bulb onion as estimated by nuclear RFLPs Theor. Appl. Genet. 90 607 614
Bradeen, J.M. & Havey, M.J. 1995 Restriction fragment length polymorphisms reveal considerable nuclear divergence within a well defined maternal clade in Allium section Cepa (Alliaceae) Amer. J. Bot. 82 1455 1462
Brewster, J.L. 1994 Onions and other vegetable alliums. CAB Intl. Univ. Press, Cambridge, UK
Damon, S. & Havey, M.J. 2014 Quantitative trait loci controlling amounts and types of epicuticular waxes in onion J. Amer. Soc. Hort. Sci. 139 597 602
Duangjit, J., Bohanec, B., Chan, A.P., Town, C.T. & Havey, M.J. 2013 Transcriptome sequencing to produce SNP-based genetic maps of onion Theor. Appl. Genet. 126 2093 2101
El-Shafie, M. & Davis, G. 1967 Inheritance of bulb color in Allium cepa Hilgardia 9 607 622
Ewart, W.H. 1945 Texas, p. 29. In: Report of the National Onion Breeding Program. U.S. Dept. Agr., Beltsville, MD
Fritsch, R.M., Matin, F. & Klaas, M. 2001 Allium vavilovii M. Popov et Vved. and a new Iranian species are the closest known relatives of the common onion A. cepa L. (Alliaceae) Genet. Resources Crop Evol. 48 401 408
Goldman, I.L., Schroeck, G. & Havey, M.J. 2001 History of public onion breeding programs and pedigree of public onion germplasm releases in the United States Plant Breed. Rev. 20 67 103
Gurushidze, M., Mashayekhi, S., Blattner, F.R., Friesen, N. & Fritsch, R.M. 2007 Phylogenetic relationships of wild and cultivated species of Allium section Cepa inferred by nuclear rDNA ITS sequence analysis Plant Syst. Evol. 269 259 269
Hanelt, P. 1985 On taxonomy, chorology and ecology of the wild species of Allium L. sect. Cepa (Mill.) Prokh Flora 176 99 116
Havey, M.J. 1992 Restriction enzyme analysis of the chloroplast and nuclear 45s ribosomal DNA of Allium sections Cepa and Phyllodolon Plant Syst. Evol. 183 17 31
Havey, M.J. 1993 A putative donor of S-cytoplasm and its distribution among open-pollinated populations of onion Theor. Appl. Genet. 86 128 134
Havey, M.J. 1997 On the origin and distribution of normal cytoplasm of onion Genet. Resources Crop Evol. 44 307 313
Havey, M.J. & Bark, O.H. 1994 Molecular confirmation that sterile cytoplasm has been introduced into open-pollinated Grano onion cultivars J. Amer. Soc. Hort. Sci. 119 90 93
Hyde, P.T., Earle, E.D. & Mutschler, M.A. 2012 Doubled haploid onion (Allium cepa L.) lines and their impact on hybrid performance HortScience 47 1690 1695
Jones, H.A. & Davis, G. 1944 Inbreeding and heterosis and their relation to the development of new varieties of onions. U.S. Dept. Agr. Tech. Bul. 874
Khar, A., Lawande, K.E. & Negi, K.S. 2011 Microsatellite marker based analysis of genetic diversity in short day tropical Indian onion and cross amplification in related Allium spp Genet. Resources Crop Evol. 58 741 752
Magruder, R., Webster, R., Jones, H.A., Randall, T., Snyder, G., Brown, H., Hawthorn, L. & Wilson, A. 1941 Descriptions of types of principal American varieties of onions. U.S. Dept. Agr. Misc. Publ. 435
Mansur, L.M., Hadder, K.M. & Suaérez, J.C. 1990 A computer program for calculating the population size necessary to recover any number of individuals exhibiting a trait J. Hered. 81 407 408
McCallum, J.A., Thomson, S., Pither-Joyce, M., Kenel, F., Clarke, A. & Havey, M.J. 2008 Genetic diversity analysis and single-nucleotide-polymorphism marker development in cultivated bulb onion based on expressed sequence tag–simple sequence repeat markers J. Amer. Soc. Hort. Sci. 133 810 818
Perry, B.A. 1949 Texas, p. 62–69. In: Report of the National Onion Breeding Program. U.S. Dept. Agr., Beltsville, MD
Van Ooijen, J.W. 2006 JoinMap 4, software for the calculation of genetic linkage maps in experimental populations. Kyazma, Wageningen, The Netherlands
Van Raamsdonk, L.W.D., Wiestma, W.A. & De Vries, J.N. 1992 Crossing experiments in Allium L. section Cepa Bot. J. Linn. Soc. 109 193 303
Von Kohn, C., Kiełkowska, A. & Havey, M.J. 2013 Sequencing and annotation of the chloroplast DNAs of normal (N) male-fertile and male-sterile (S) cytoplasms of onion and single nucleotide polymorphisms distinguishing these cytoplasms Genome 56 737 742
