Introduction

Sunflower (Helianthus annuus L.) is adapted to a wide variety of soils and climatic conditions and is widely grown in the world. There are two basic types of sunflowers. Approximately, 90 % of world production is oil-type sunflower, which is used as a source of high-quality vegetable oil. The remaining sunflowers grown are the confection type, which is grown for human food consumption or bird feed (Hladni 2016). Breeding for confection sunflower is different from that of oil type, focusing on large seeds as well as increased test weight, protein content and quality. The seed of confection sunflower with white striped hulls is larger than that of the oilseed types and has a lower oil content and test weight. The United States (US) is a major global producer of confection sunflower, supplying both domestic needs and export markets. The use of sunflower seed in human diets as a snack or for birdfeed has grown consistently over the past 15 years (US National Sunflower Association 2011).

Downy mildew caused by Plasmopara halstedii (Farl.) Berl. & de Toni is a yield-limiting disease that occurs in all areas of intensive sunflower production throughout the world, except Australia (Markell et al. 2015). P. halstedii overwinters as a thick-walled oospore that can survive many years in the soil. Germinated oospores produce zoosporangia which release motile zoospores that infect germinating seedlings, causing a systemic infection that kills most infected plants (Harveson et al. 2016). Infected plants that survive produce a sterile head and act as a competitor with adjacent plants in the field (Markell et al. 2015). Thus, 100 % yield loss of each systemically infected plant is expected. Downy mildew is not observed uniformly throughout fields; rather, it occurs in areas of fields prone to saturated soils (for example, low spots). As a result, nearly 100 % yield loss can occur in large patches of sunflower fields when environmental conditions are favorable for infection (Gulya et al. 2013). The frequency of fields infected by downy mildew in a region depends heavily on rainfall after planting, but can be very high. In the US state of North Dakota, which frequently leads the country in production, the prevalence of sunflower fields infected with downy mildew ranged from less than 2 % to over 70 % between 2001 and 2011, with a mean prevalence of 33.9 % over those 11 years (Gulya et al. 2013).

Confection sunflower is more vulnerable to downy mildew than oil sunflower due to a lack of resistance sources. The P. halstedii pathogen was first identified in North America (Nishimura 1922; Young and Morris 1927) and was later reported in Russia and Western Europe in approximately 1960 (Novotelnova 1960). In the 1970s, two P. halstedii races were identified, one each in Europe and North America (Zimmer 1974). To combat this destructive disease, host race-specific resistance genes (R genes), denoted as Pl, were deployed at a large scale in sunflower production, especially in the late 1990s when the P. halstedii population developed resistance to the widely used seed treatment mefenoxam (Albourie et al. 1998; Gulya et al. 1999, 2013). However, over 40 years, the P. halstedii pathogen has constantly evolved new physiological races due to its pathogenic variability and the selection pressure resulting from the use of resistant hybrids in sunflower production. To date, at least 24 P. halstedii races have been defined in Europe and 40 in the Americas (Gulya et al. 2011; Gascuel et al. 2015; Viranyi et al. 2015), posing a serious threat to global sunflower production. Gilley et al. (2016) screened 14 downy mildew-resistant lines using 185 isolates collected from US sunflower fields in 2014 and 2015, and found that only a few lines were resistant to all isolates, none of which contained known resistance genes present in confection sunflower. Thus, continuing the search for new Pl genes that provide resistance to new P. halstedii strains is a critical task for sunflower researchers to protect sunflower production from the yield loss that occurs from downy mildew.

The cultivated sunflower (2n = 2x = 34) belongs to the Helianthus genus, a member of the Compositae (Asteraceae) family, which is the largest and most diverse family of flowering plants (Funk et al. 2005). The Helianthus genus contains 53 wild species comprising 14 annual and 39 perennial species (Schilling 2006; Stebbins et al. 2013). In the last four decades, a total of 1359 annual accessions and 842 perennial accessions were collected in the US (Seiler and Marek 2014). These provide many useful genes for sunflower improvement, such as better quality and higher yield, cytoplasm male sterility, disease resistance and tolerance for biotic and abiotic stresses (Seiler and Jan 2010). Sunflower wild species are also an important source of downy mildew resistance genes. Fourteen Pl genes (Pl 1 , Pl 2 , Pl 5 –Pl 8 , Pl 13 –Pl 18 , Pl 21 , and Pl Arg ) that were previously mapped to the sunflower genome can be traced to their wild origin, most of which were derived from wild annual species, especially H. annuus (Miller and Gulya 1987, 1988, 1991; Seiler 1991, 2010; Vear et al. 2008; Mulpuri et al. 2009; Hulke et al. 2010; Qi et al. 2016). This species contributed five of the 14 Pl genes (Pl 1 , Pl 2 , Pl 6 , Pl 13 , and Pl 17 ) and is considered to be the most valuable source of Pl genes. Gulya (2005) screened an additional 286 accessions of the wild H. annuus using the most virulent P. halstedii race at the time (race 773) and found 26 accessions where >90 % plants displayed immunity to systemic infection. Among the 14 wild annual species, which are all diploid with 34 chromosomes, the same number as cultivated sunflower, H. annuus is the most diverse species of sunflower in North America. This species is found throughout the US, but is more common in the western two-thirds of the country (Rogers et al. 1982; Kantar et al. 2015). However, most of the resistant accessions identified by Gulya (2005) originated from the south central US, with a high amount of them originating from the state of Texas.

The wild H. annuus accession PI 435414 was originally collected from Paris, Texas, US in 1978. This accession was first identified to be resistant to downy mildew by Gulya (2005) and further confirmed by Markell and Humann in 2013 (personal communication). In this research, we introgressed downy mildew resistance from PI 435414 into confection sunflower and describe a new Pl gene that was discovered in wild H. annuus species and located in linkage group (LG) 4 of the sunflower genome. This gene, designated Pl 19 , confers resistance to several new virulent races of P. halstedii currently identified in North America, providing a novel source for use in confection sunflower breeding programs.

Materials and methods

Cross and backcross populations

Due to open-pollinated nature, wild H. annuus accession PI 435414 is segregating for downy mildew resistance. PI 435414 plants were screened with P. halstedii race 734, and the selected resistant plants were crossed with the cytoplasmic male sterile (CMS) CONFSCLB1 in May 2013. The selected F1-resistant plants were backcrossed twice to the maintainer CONFSCLB1. CONFSCLB1 is a confection BC1F3-derived BC1F4 maintainer selected from the cross of HA 441/ROM PH//HA 442/HA 441/4/CONF/5/CONF and is susceptible to downy mildew. The line was released by the USDA-ARS Sunflower Research Unit and the North Dakota Agricultural Experiment Station in 2006. The F1 and each of backcross generations were screened in the greenhouse for downy mildew resistance to P. halstedii race 734 as described below. The selected downy mildew-resistant BC2F1 plants were advanced to the BC2F2 generation.

Mapping population

For molecular mapping of the downy mildew resistance gene derived from PI 435414, the BC1F2 and BC1F2-derived BC1F3 populations were developed from a single resistant BC1F1 plant from the cross of CMS CONFSCLB1 and PI 435414. The BC1F2 individuals were grown in a greenhouse in the winter of 2014–2015 and were self-pollinated to produce BC1F2:3 families that were used in subsequent disease evaluation.

Downy mildew resistance evaluation

A Plasmopara halstedii isolate of race 734, which was first identified in 2009 in North America and overcame the Pl 6 and Pl 7 genes (Gulya et al. 2011), was chosen to test seedlings of each backcross generation and the mapping population of the BC1F3 families for resistance to downy mildew. Phenotypic variation was evaluated in greenhouse trials using the whole seedling immersion technique described by Gulya et al. (1999) and Qi et al. (2015). Sunflower seedlings infected by downy mildew display typical leaf chlorosis with white sporulation on the underside of cotyledons and true leaves. A plant was considered to be susceptible (S) if sporulation was observed on cotyledons and true leaves and was considered to be resistant (R) if no sporulation was observed. In addition, isolates of another five P. halstedii races, 314, 700, 710, 714, and 774, which are the most predominant and highly virulent races in North America and Europe (Gulya et al. 2011; Gascuel et al. 2015; Gilley et al. 2016), were selected to test the homozygous resistant BC1F3 families derived from the cross of CONFSCLB1 and PI 435414 and the susceptible recurrent parent CONFSCLB1 with Cargill 270 and HA-DM1 carrying the Pl 18 gene as susceptible and resistant controls, respectively.

Artificial infection of BC1F2:3 seedlings was used to determine the segregation of the downy mildew R gene in the BC1F2 population. A total of 40 seeds from each 140 BC1F2:3 families were germinated at 22–24 °C in a growth chamber, and 30 seedlings of each family were inoculated with a P. halstedii isolate of race 734. The BC1F3 families were classified as homozygous resistant if none of the seedlings had sporulation, segregating if some seedlings (about one-quarter in an F3 family) had sporulation on the cotyledons and true leaves, and homozygous susceptible if all seedlings had sporulation on cotyledons and true leaves, which represented the genotypes of downy mildew resistance in each F2 individual.

DNA extraction and SSR PCR condition

Young leaf tissues from the parents, CONFSCLB1 and PI 435414, and 140 BC1F2 plants were collected and freeze-dried. Tissues samples were also collected from the BC2F2 population. DNA extraction from lyophilized tissues was performed using a DNeasy 96 plant kit following the manufacturer’s instructions (Qiagen, Valencia, CA, USA). The quantity and quality of DNA were determined using a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The concentration of DNA was adjusted to 5 ng/µl for PCR amplification.

For the simple sequence repeat (SSR) marker, 10–20 ng of template DNA was added to a 15-µl polymerase chain reaction (PCR) mixture, including 0.75 µl of 50 mM MgCl2, 1.5 µl of 2.5 mM dNTPs, 0.9 µl of 10× PVP-40, 1.5 µl of 0.2 pmol of the forward primer with a M13 tail (CACGACGTTGTAAAACGAC) at the 5′ end, 1.5 µl of 1.0 pmol of the reverse primer, 1.5 µl of 1.0 pmol of the fluorescently labeled M13 primer, 1.5 µl of 10× PCR buffer, and 0.5 units of Taq polymerase (Bioline, Randolph, MA, USA). PCR was performed in a Peltier thermocycler (Bio-Rad Lab, Hercules, CA, USA) using the touchdown program as described by Qi et al. (2011). PCR products were detected by an IR2 4300/4200 DNA Analyzer with denaturing polyacrylamide gel electrophoresis (LI-COR, Lincoln, NE, USA).

Bulked segregant analysis and genetic mapping

A total of 860 SSR markers were used to identify polymorphisms between the parents CONFSCLB1 and PI 435414 of the mapping population. For bulked segregant analysis (BSA) (Michelmore et al. 1991), genomic DNA from BC1F2 of ten homozygous resistant plants and ten homozygous susceptible plants were pooled in equal proportions to create the resistant and susceptible bulks. SSR markers that were polymorphic between the parents were used to test the S- and R-bulks. Once the potential locations of resistance gene were identified, additional markers from these locations were used to genotype the population.

SSR marker and downy mildew phenotype data were used to construct a genetic map of the 140 BC1F2 individuals from the CONFSCLB1 × PI 435414 population. Linkage analysis was performed using JoinMap 4.0 with a regression mapping algorithm and Kosambi’s mapping function (Van Ooijen 2006). The Chi-square (χ 2) test was used to assess the goodness of fit to the expected segregation ratio for each marker. A minimum likelihood of odds (LOD) ≥3.0 and a maximum distance of ≤50 centimorgans (cM) were used to test linkages among markers.

SNP marker and genotyping

Once the Pl gene was positioned relative to the SSR markers, additional SNP markers surrounding the region were selected from the three published sunflower SNP maps, which could be used to better define the genomic position of the Pl gene. Thirteen SNPs were SFW SNPs (Bowers et al. 2012) covering the region from 5.21 to 20.78 cM, and 14 SNPs were NSA SNPs (Talukder et al. 2014; Hulke et al. 2015) covering the region from 11.04 to 21.01 cM on LG4 (Table S1) in the respective maps.

Genotyping of SNPs was performed using a strategy of converting the SNPs into length polymorphism markers as described by Qi et al. (2016). Four primers were used in each PCR: a priming-element-adjustable primer (PEA-primer 5′-ATAGCTGG-Sp9-GCAACAGGAACCAGCTATGAC-3′) with an attached fluorescence tag at the 5′ terminus as a universal primer, two-tailed forward allele-specific primers (AS-primers F1 and F2, the tail sequences were identical to the PEA-primer), and one common reverse primer. An additional 5-base oligonucleotide (5′-ATGAC-3′) was inserted between the tail and allele-specific sequences in the AS-primer F2 to produce a length difference between the two SNP alleles after amplification (Table 1). The primers were screened for polymorphisms among the parents and BC1F2 population using the methods described by Qi et al. (2016). PCR products were detected by an IR2 4300/4200 DNA Analyzer with denaturing polyacrylamide gel electrophoresis (LI-COR, Lincoln, NE, USA).

Table 1 The primer sequences of the SNPs mapped in this study
Full size table

Results

Transfer of downy mildew resistance from wild species H. annuus into confection sunflower

Confection sunflower of the cytoplasmic male sterile (CMS) line CONFSCLB1 was used as the female in an initial cross with downy mildew-resistant plants selected from H. annuus accession PI 435414 in May 2013. Not all of the F1 plants were resistant given that the wild species accessions were not homozygous for genes controlling downy mildew resistance because they have an open-pollinated nature. A total of 50 F1 plants were tested for resistance against P. halstedii race 734. Forty-two F1 plants were resistant, and eight were susceptible. The selected F1 resistant plants were used as male parents backcross to CONFSCLB1 with a normal cytoplasm in December 2013. The downy mildew test of the 82 BC1 plants against race 734 in April 2014 indicated that 46 BC1 plants were resistant and 36 were susceptible, fitting an expected 1:1 segregation ratio of the BC1 population (χ2 = 1.2195, df = 1, P = 0.2695). The resistant BC1 plants were used again as the male parent and were crossed to the recurrent parent CONFSCLB1 to produce the BC2 generation. After a test of the BC2 seedlings, the selected resistant BC2 plants were self-pollinated to produce the BC2F2 population for the further selection of homozygous resistant plants.

Inheritance of downy mildew resistance in the mapping population

The parental line CONFSCLB1 and the 140 BC1F3 families (30 seedlings for each family) were inoculated with a P. halstedii isolate of race 734 in a greenhouse under controlled conditions in June 2015. White sporulation of P. halstedii was clearly detected on the cotyledons and true leaves of CONFSCLB1. The susceptible seedlings in the BC1F3 population exhibited similar symptoms to those of CONFSCLB1, whereas the resistant seedlings had no sporulation. The segregation of the downy mildew resistance for the 140 BC1F2:3 families was 33 homozygous resistant, 78 segregating, and 29 homozygous susceptible plants and fits a 1 resistant:2 heterozygous resistant:1 susceptible segregation ratio (χ 2 = 2.0572, df = 2, P = 0.3575). The results indicated that a single dominant gene was responsible for downy mildew resistance in the PI 435414 population.

Molecular mapping of the downy mildew gene from PI 435414

Over 860 SSR markers previously mapped to the sunflower genome (Tang et al. 2002; Yu et al. 2003) were used to screen polymorphisms between two parental lines, CONFSCLB1 and PI 435414, and approximately 43 % of the SSR markers were polymorphic between the two parents. Subsequent bulked segregant analysis (BSA) (Michelmore et al. 1991) was conducted between resistant and susceptible bulks with the all of the polymorphic SSR markers. BSA analysis revealed that the downy mildew resistance from PI 435414 was associated with the SSR markers HT664, ORS499, ORS721, and ORS963 from LG4. These four SSRs plus seven additional polymorphic LG4 SSRs were then used to genotype the 140 individuals of the BC1F2 population. Among the 11 SSR markers tested, eight co-dominant SSRs segregated at the expected 1:2:1 ratio, whereas three dominant SSRs, ORS963, HT664, and ORS721, segregated at the expected 3:1 ratio. The resistant locus from PI 435414, designated Pl 19 , was mapped 1.5 cM from ORS963 and 3.2 cM from HT298 in a 4.7-cM interval in LG4 (Fig. 1a).

Fig. 1
figure 1

Genetic maps of sunflower linkage group (LG) 4. a LG4 SSR and SNP combined map of Pl 19 ; b LG4 Pl 17 map taken from Qi et al. (2015)

Full size image

Marker enrichment in the Pl 19 region

To identify additional markers in the Pl 19 region, a total of 27 SNP markers located on LG4 were used to saturate the Pl 19 genetic map (Bowers et al. 2012; Talukder et al. 2014; Hulke et al. 2015). After screening these SNP markers between the two parents, CONFSCLB1 and PI 435414, nine SNPs were identified as polymorphic and were subsequently evaluated across the entire BC1F2 population. No segregation distortion was identified for these SNP markers in the population, and the resulting map placed nine SNP markers at an interval of 10.4 cM in the upper end of the LG4 genetic map, encompassing the Pl 19 locus (Fig. 1a). This mapping narrowed the resistance gene Pl 19 from a 4.7-cM to a 1.2-cM region on LG4, flanked on one side by the SNP marker NSA_003564 and on the other side by NSA_006089 (Fig. 1a).

Physical linkage analysis of the clustered Pl 19 and Pl 17 in LG4

Pl 19 is the second downy mildew R gene that mapped to LG4 of the sunflower genome and showed a close linkage to the Pl 17 gene (Fig. 1, Qi et al. 2015). In the Pl 19 and Pl 17 genetic maps, the common markers ORS963 and NSA_003564 were mapped downstream of Pl 17 (Fig. 1b) but upstream of Pl 19 (Fig. 1a). The SSR marker ORS963 is the nearest marker that is closely linked to Pl 17 and had different PCR pattern between PI 435414 and HA 458 (Pl 17 donor in mapping population). Among five DNA markers, ORS1197, SFW04052, ORS963, SFW08268, and NSA_003564 surrounding Pl 17 , ORS1197 and SFW08268 did not have a polymorphism in the Pl 19 population. The SNP marker NSA_006089 closely linked to Pl 19 was also not mapped to the Pl 17 population (Fig. 1, Qi et al. 2015). These results indicated that Pl 19 and Pl 17 are different genes, but are closely linked to each other as a gene cluster.

Aligning ten SNPs sequences against the whole sunflower genome of HA412 v1.1. pseudomolecules available at http://sunflowergenome.org/early_access/repository/main/pseudomolecules/ indicated that the two genes are physically located at an interval of 3231,660 bp at the upper end of LG4 (Table 2). Pl 17 is located between 3,621,089 and 6,813,151 bp, whereas Pl 19 is likely located between 6,813,151 and 6,852,749 bp. The genetic order of eight SNP markers were consistent with their physical position, whereas two SNPs, NSA_03564 and SFW02206, were slightly different regarding their genetic and physical order (Table 2; Fig. 1).

Table 2 Comparison of the physical position of Pl 19 and Pl 17 in the sunflower genome
Full size table

Spectrum of downy mildew resistance from PI 435414

A BC1F3 homozygous R-family, 14-213-69, was selected to test against additional five P. halstedii races, 314, 700, 710, 714, and 774, along with race 734, which represents the most predominant and virulent races currently identified in North America (Gulya et al. 2011; Gilley et al. 2016). Seedlings of 14-213-69 along with susceptible and resistant checks, Cargill 270 and HA-DM1, respectively, and the susceptible recurrent parent CONFSCLB1 were inoculated with an isolate of each race separately. Cargill 270 is a hybrid of one of the nine standard differential lines used to identify P. halstedii races and is susceptible to all races. HA-DM1 is a recently released downy mildew-resistant line harboring the Pl 18 gene and is resistant to all NA races (Gilley et al. 2016; Qi et al. 2016; Qi and Seiler 2016). As expected, Cargill 270 and CONFSCLB1 were susceptible to all of the races tested, and abundant white sporulation was observed on the leaf surface of their seedlings. In contrast, HA-DM1 and 14-213-69 were immune to all of the races, with no sporulation on cotyledons and true leaves (Fig. 2; Table 3), indicating that the resistance from PI 435414 has a broad spectrum of activity against downy mildew.

Fig. 2
figure 2

Downy mildew evaluation of the homozygous BC1F3 family of 14-213-69. a P. halstedii race 710; b P. halstedii race 734. Abundant white sporulation was observed on the underside of the leaf surface of Cargill 270 and CONFSCLB1, whereas no sporulation was noted on HA-DM1 and 14-213-69. Cargill 270, susceptible check; CONFSCLB1, susceptible recurrent parent; HA-DM1, resistance check; 14-213-69, homozygous BC1F3 family

Full size image
Table 3 Downy mildew multi-race tests of the homozygous BC1F3 family for resistance
Full size table

Marker selection of the homozygous resistant plants from the BC2F2 population

To select homozygous resistant plants from the BC2F2 population of a CONFSCLB1/PI 435414 cross, two SNPs, SFW04052 and NSA_006089, were initially used to screen 94 BC2F2 individuals. Eighteen BC2F2 plants were identified as homozygous plants with identical PI 435414 marker alleles and were tested with the three additional markers, ORS963, NSA_03564, and SFW02206, which are closely linked to Pl 19 . As anticipated, only the PI 435414 PCR patterns of the tested markers were detected in the homozygous BC2F2 plants, which were advanced to the BC2F3 generation. To further confirm the marker–trait association, a total of 185 BC2F2:3 progenies derived from the putative homozygous resistance of BC2F2 selected by DNA markers were inoculated with race 734. All of the BC2F3 plants exhibited resistance to downy mildew, and no segregation was observed. These findings confirmed the correlation of the phenotype and genotype, indicating that these markers were associated with the downy mildew resistance derived from PI 435414.

Discussion

Downy mildew is one of the most damaging diseases for sunflower production in North America and Europe as well as Asia and South America. Historically, host genetic resistance has provided an economically and environmentally friendly method for controlling downy mildew. However, the main drawback of using R genes to control resistance is that their effects are often not durable due to the rapid evolution of P. halstedii. This limitation necessitates the continued search for new sources of resistance and the design of new strategies for more durable resistance. In this study, downy mildew resistance was successfully transferred from the wild H. annuus accession PI 435414 to confection sunflower. Seedling resistance to P. halstedii race 734 in the segregation population was conferred by a single dominant gene, designated Pl 19 . This gene was mapped to LG4 of the sunflower genome and flanked by two SNP markers, NSA_003564 and NSA_006089, in a 1.2-cM interval. These markers are suitable for marker-assisted selection, especially if Pl 19 can be used in combination with other Pl genes. In addition, Pl 19 exhibits broad-spectrum resistance against P. halstedii races, 314, 700, 710, 714, 734, and 774, which are the most predominant and virulent races currently identified in North America and Europe (Gulya et al. 2011; Gascuel et al. 2015; Gilley et al. 2016). To the best of our knowledge, this is the first downy mildew gene in the confection sunflower background, and the selected homozygous BC2F3 germplasm will be extremely useful to develop downy mildew-resistant confection hybrids for commercial production.

As the second Pl gene mapped to LG4, Pl 19 is located in a region that is closely linked to the downy mildew R gene Pl 17 . Both Pl 19 and Pl 17 originate from wild H. annuus of different accessions, PI 435414 and PI 468435, respectively. The former was collected from the US state of Texas, and latter was obtained from the US state of Idaho. (https://npgsweb.ars-grin.gov/gringlobal/search.aspx). Based on the genetic linkage analysis, Pl 19 is located downstream of Pl 17 and is separated by the common markers ORS963 and NSA_003564 in the order SFW04052-Pl 17 -ORS963/NSA_003564-Pl 19 -NSA_006089 (Fig. 1). The alleles amplified from Pl 19 and Pl 17 for the common marker ORS963 were clearly discriminated according to their PCR pattern and fragment size (data not shown). ORS963 segregated as a co-dominant marker in the Pl 17 population (Qi et al. 2015) and exhibited a dominant nature in the Pl 19 population. Additionally, the SNP marker NSA_006089 nearest to Pl 19 was not mapped to the Pl 17 population, suggesting a null allele or altered priming site. These results indicate that Pl 19 is a novel downy mildew R gene that is different from Pl 17 , and ongoing allelism test efforts will help further elucidate the relationships of these two genes. Sequence alignments of SFW04052 and NSA_006089 flanking Pl 17 and Pl 19 against sunflower whole genome sequences indicated that Pl 17 and Pl 19 are physically located at an interval of 3,231,660 bp in the upper end of LG4 (Table 2), where the recombination frequency of 0.59 Mb/cM is fourfold higher than the genomic average of 2.4 Mb/cM, making it a region that is amenable to map-based cloning of genes of interest (Talukder et al. 2014; Qi et al. 2015). The closest flanking markers can be used as starting points for the positional cloning of Pl 17 and Pl 19 in the future to further elucidate the genetic mechanism underlining this gene cluster.

Sunflower wild species of H. annuus are an important source of downy mildew resistance and other disease resistances (for review see Seiler and Jan 2010). Among 15 mapped downy mildew Pl genes (Pl 1 , Pl 2 , Pl 5 –Pl 8 , Pl 13 –Pl 19 , Pl 21 , and Pl Arg ), six (Pl 1 , Pl 2 , Pl 6 , Pl 13 , Pl 17 , and Pl 19 ) are known to originate from the wild H. annuus (Fick and Zimmer 1974; Miller and Gulya 1991; Vear et al. 2008). These Pl genes from H. annuus tend to appear as a gene cluster in the sunflower genome. Pl 1 , Pl 2 , and Pl 6 are clustered on LG8 (Bouzidi et al. 2002; Slabaugh et al. 2003). Pl 13 is part of a cluster on LG1 (Mulpuri et al. 2009; Liu et al. 2012), and Pl 19 , mapped in the current study, is in a cluster with Pl 17 on LG4 (Qi et al. 2015). Interestingly, three Pl gene clusters plus another Pl gene cluster on LG13 are all located on the distal regions of sunflower chromosomes (Slabaugh et al. 2003; Bachlava, et al. 2011; Liu et al. 2012; Qi et al. 2015). The gene clusters on LGs 1 (Pl 13 , Pl 14 , and Pl 16 ) and 13 (Pl 5 , Pl 8 , and Pl 21 ) are in the lower end of the chromosomes, whereas the clusters on LGs 4 (Pl 17 and Pl 19 ) and 8 (Pl 1 , Pl 2 , Pl 6 , P l7 , and Pl 15 ) are in the upper end of the chromosomes. The only two Pl genes that are not in the cluster, Pl Arg and Pl 18 , are located in the middle of the one arm of sunflower chromosomes 1 and 2, respectively (Wieckhorst et al. 2010; Qi et al. 2016). The evolution of the R gene clusters has been extensively investigated in Arabidopsis (Kunkel 1996; Leister 2004), flax (Ellis et al. 1995, 1997, 2000), lettuce (Kesseli et al. 1994; Michelmore and Meyers 1998), tomato (Jones et al. 1993; Kaloshian et al. 1995; Takken et al. 1999), rice (Song et al. 1995, 1997; Wang et al. 1998; Zhou et al. 2007), maize (Sanz-Alferez et al. 1995; Hulbert et al. 2001), wheat (Yahiaoui et al. 2004, 2009), and barley (Jorgensen 1994; Wei et al. 1999, 2002). The studies revealed that tandem and segmental gene duplication, recombination, diversifying selection, and sequence divergence in intergenic regions mainly contribute to the generation of novel R genes and their resistance specificity. This feature is coincident with gene evolution in flowering plants, which occurs preferentially at the ends of chromosomes and is derived by duplication and divergence associated with high rates of recombination (See et al. 2006). This feature explains why most of the R gene clusters are located in the distal regions of chromosomes.

Although many downy mildew resistance genes were identified from sunflower wild species [some of which were transferred into cultivated sunflower [(Miller and Gulya 1987, 1988, 1991; Seiler 1991; Rahim et al. 2002; Gulya 2005; Vear et al. 2008; Hulke et al. 2010; Qi et al. 2016)], none of these genes are present in confection sunflower. Pl 19 is the first downy mildew R gene directly introduced from wild species into confection sunflower and is highly resistant to a wide range of P. halstedii races tested in the current study. However, deployment of a single gene in sunflower production will place strong selection pressure on the pathogen population, as it is known that P. halstedii pathogens quickly evolve and mutate to overcome plant resistances (Gascuel et al. 2015; Viranyi et al. 2015). To enhance the durability and effectiveness of a single resistance gene, Pl 19 should be used in combination with other Pl genes. In an ongoing project, we selected three Pl genes, Pl Arg , Pl 17 , and Pl 18 , and are transferring them from oil sunflower into confection sunflower to diversify the Pl gene pool in confection sunflower. Pl 19 will easily combine with Pl Arg and Pl 18 in a single line, as the two genes are located in LGs 1 and 2, respectively (Wieckhorst et al. 2010; Qi et al. 2016). In contrast, the linkage between Pl 17 and Pl 19 appears to be tight in LG4, and a large segregating population from a cross of Pl 17 /Pl 19 is needed to recover rare recombinations. Because Pl Arg , Pl 17 , and Pl 18 are resistant to all currently identified North America P. halstedii races (Gilley et al. 2016), the selection of genotypes with a gene combination by disease screening is not possible due to the lack of pathogen isolates with a specific virulence gene. Therefore, molecular markers for each gene are required for the selection of resistant pyramids with both parent’s Pl genes from a segregation population. Such DNA markers are now available for Pl Arg , Pl 17 , and Pl 18 (Wieckhorst et al. 2010; Qi et al. 2015, 2016); thus, the tightly linked SSR and SNP markers for Pl 19 identified in this study will facilitate to pyramid this new gene with other Pl genes to aid in the development of durable downy mildew control in confection sunflower.

Author contribution statement

Conceived and designed the experiments: LLQ. Performed the experiments: ZWZ, LLQ, GJM, JZ, SGM. Analyzed data: ZWZ, LLQ. Wrote the paper: LLQ, ZWZ. Commented on the manuscript before submission: SGM, GJM.