|Home | About | Journals | Submit | Contact Us | Français|
One of major objectives of crop breeding is conferring resistance to diseases and pests. However, large-scale phenotypic evaluation for many diseases and pests is difficult because strict controls are required to prevent their spread. Detection of disease resistance genes by using DNA markers may be an alternative approach to select potentially resistant accessions. Potato (Solanum tuberosum L.) breeders in Japan extensively use resistance gene H1, which confers nearly absolute resistance to potato cyst nematode (Globodera rostochiensis) pathotype Ro1, the only pathotype found in Japan. However, considering the possibility of accidental introduction of the other pathotypes, breeding of resistant varieties is an important strategy to prevent infestation by non-invading pathotypes in Japan. In this study, to evaluate the prevalence of resistance genes in Japanese genetic resources, we developed a multiplex PCR method that simultaneously detects 3 resistance genes, H1, Gpa2 and Gro1-4. We revealed that many Japanese varieties possess not only H1 but Gpa2, which are potentially resistant to other pathotypes of potato cyst nematode. On the other hand, no genotype was found to have the Gro1-4, indicating importance of introduction of varieties having Gro1-4. Our results demonstrate the applicability of DNA-marker assisted evaluation of resistant potato genotypes without phenotypic evaluation.
Conferring resistance to diseases and pests is a major goal of crop breeding. Traditionally, resistance is assessed by inoculating test plants with a disease or pest or by planting them in an infested field. However, the handling of many diseases and pests, which are not found or exist only in restricted areas in a country, is strictly restricted by quarantine. Therefore, large-scale phenotypic evaluation is impractical for breeding novel resistant varieties.
The use of DNA markers for detecting disease resistance genes may be an alternative approach to phenotypic evaluation (Collard et al. 2005, Jena and Mackill 2008). DNA markers, which are closely linked to disease resistance genes, enable the evaluation of resistance genes without inoculation. Such evaluation provides information about the prevalence of disease resistance genes in genetic resources and facilitates breeding of novel resistant varieties.
Potato cyst nematode (PCN) comprises 2 species (Globodera rostochiensis and G. pallida) and 8 pathotypes (Ro1 to Ro5 of G. rostochiensis and Pa1 to Pa3 of G. pallida). Infested potato plants show retarded growth and decreased yield compared with healthy plants. PCN causes yield losses ranging from 20% to 70% and up to 30% worldwide (Oerke et al. 1994). Thus these species are one of the most important pests of potato (Solanum tuberosum L.) and are therefore included in the list of quarantine pathogens in many countries. It can be controlled by crop rotation, chemical soil disinfestation and cultivation of resistant varieties. However, because of the formation of cysts, PCN can survive in soil for many years in the absence of host plants, making crop rotation unattractive for potato farmers. Chemical control of PCN involves unspecific and harmful pesticides. Because of increasing concern about environmental issues and governmental regulations, this method has been almost abandoned in many countries. Therefore, breeding and cultivation of resistant varieties are becoming increasingly important, and hence, scientific studies of the resistance genes and their underlying mechanisms are of great interest.
Several genes and quantitative trait loci (QTLs) of PCN resistance derived from other cultivated species and wild species have already been identified in potato. Thirteen resistance loci confer partial resistance: Gro1.2, Gro1.3, Gro1.4, Gpa, Grp1, Gpa5, Gpa6, GpaM1, GpaM2, GpaM3, RGp5-vrn HC, GpaXIltar and GpaIVSadg (Caromel et al. 2003, Kreike et al. 1993, 1994, 1996, Moloney et al. 2010, Sattarzadeh et al. 2006, Tan et al. 2009, van der Voort et al. 1998, 2000). Further, 4 loci (Gro1, H1, GroVI and Gpa2) (Barone et al. 1990, Gebhardt et al. 1993, Jacobs et al. 1996, van der Voort et al. 1997) and the combination of GpaVsspl and GpaXIsspl (Caromel et al. 2005) confer nearly absolute resistance to one or more pathotypes. Gro1.2, Gro1.3, Gro1.4, Gpa, GpaM1, GpaM2, GpaM3 and Gro1 were derived from S. spegazzinii Bitt.; RGp5-vrn HC and GroVI were derived from S. vernei Bitt. et. Wittm.; GpaXIltar was derived from S. tarijense Hawkes; GpaIVSadg, H1 and Gpa2 were derived from S. tuberosum ssp. andigena Hawkes; GpaVsspl and GpaXIsspl were derived from S. sparsipilum (Bitt.) Juz. et. Buk.; Grp1, Gpa5 and Gpa6 were derived from interspecific hybrid clones of S. tuberosum and several wild species including S. oplocense Hawkes, S. spegazzinii, S. tuberosum ssp. andigena, S. vernei and S. vernei ssp. ballsii.
Breeders have succeeded in producing potato varieties resistant to G. rostochiensis by introducing H1. H1 confers hypersensitive resistance to G. rostochiensis pathotypes Ro1 and Ro4 (Bakker et al. 2004, Gebhardt et al. 1993). However, widespread and extensive use of varieties with H1 has inadvertently caused rapid worldwide increase in G. pallida (Bradshaw et al. 1998, Bryan et al. 2002). Therefore, great emphasis is now placed on breeding of varieties resistant to a broad spectrum of PCN populations.
Recently, a set of DNA markers closely linked to H1, N146 and N195, was developed; these markers sandwich H1 with recombination frequencies of 0.109% and 0.207%, respectively and have been used for selecting PCN-resistant lines (Mori et al. 2011, Takeuchi et al. 2008). DNA marker Gro1-4 can detect Gro1-4 (Biryukova et al. 2008, Gebhardt et al. 2006). Gro1-4 is a member of the Gro1 locus, which confers nearly absolute resistance to all pathotypes of G. rostochiensis, and is therefore considered a useful resistance gene (Barone et al. 1990, Paal et al. 2004). Gpa2 provides dominant resistance to distinct G. pallida populations (Bakker et al. 2003, van der Voort et al. 1997, 1999); because this resistance gene has already been isolated and well characterized, markers directly amplifying the gene can be developed on the basis of its sequence information. Broad-spectrum resistance to G. rostochiensis and G. pallida is conferred by the Grp1 locus and detected by cleaved amplified polymorphic sequence (CAPS) marker TG432 (Finkers-Tomczak et al. 2009). CAPS marker C237 is linked to GpaIVSadg, which provides partial resistance to G. pallida Pa2/3 (Moloney et al. 2010). Single-nucleotide polymorphism (SNP) marker HC can detect RGp5-vrnHC, which confers partial resistance to G. pallida Pa2/3 (Sattarzadeh et al. 2006).
In Japan, breeding of PCN-resistant varieties was started just after pathotype Ro1 was discovered in 1972, and since then, achieving Ro1 resistance in new potato varieties has become the priority. Among the 8 pathotypes of PCN, only Ro1 has been found so far (Inagaki 1984, Kushida and Momota 2005). Because H1 provides perfect Ro1 resistance and has not been overcome by G. rostochiensis populations (Kushida and Momota 2005), breeders exclusively use this gene for breeding of Ro1-resistant varieties (Mori et al. 2007). However, despite the strict quarantine procedures in Japan, the possibility of accidental introduction of the other PCN pathotypes cannot be completely excluded. Therefore, selection of potato genotypes potentially resistant to several pathotypes and breeding of resistant varieties are an important strategy to prevent infestation by other pathotypes.
In this study, we tested published DNA markers and newly developed markers. Then, to identify candidate PCN-resistant accessions, we evaluated our genetic resources by a novel multiplex PCR method developed using selected DNA markers. Our results revealed the current status of resistance genes in our resources. Further, we discuss the possibility of evaluation of genetic resources for potential resistance to diseases and pests before actual invasion by using DNA markers.
Our genetic resources contain 812 varieties and lines including wild relatives, other cultivated species, landraces, modern varieties, and breeding lines derived from crosses among our resources and cover almost all varieties used in Japanese potato breeding. A list of the varieties and lines used in this study is shown in Supplementary Table 1. Total DNA was isolated by the hexadecyltrimethylammonium bromide method (Doyle and Doyle 1987).
To develop DNA markers and to determine the PCR conditions, samples were chosen on the basis of previous researches, evaluation data published in The European Cultivated Potato Database (http://www.europotato.org/menu.php), pedigree records and breeder’s evaluation.
Kita-akari, Touya and Haruka, which are known to have H1, were used to determine the PCR condition for N146 and N195. For developing Gpa2 markers, Marijke which has Gpa2 (van der Voort et al. 1999), and Aiyutaka and Touya, which are progenies of Gpa2 donor S. tuberosum ssp. andigena CPC 1673, were used. For Gro1-4, Alwara which possesses Gro1-4 (Biryukova et al. 2008) and W872204, W872204-1 and W872204-2, which are progenies of S. spegazzinii, were used. The PCR condition for GpaIVSadg was determined with Eden, which is known to have GpaIVSadg (Moloney et al. 2010) and Irida, P10111-1, W804421-3 and W862208-1, which are progenies of S. tuberosum ssp. andigena. Because we do not have varieties and lines known to have Grp1 and RGp5-vrnHC, we chose, Astarte, Elles, Magic Red, Mara and Starter, as they are progenies of several wild species, and show broad-spectrum resistance. Irish Cobbler and May Queen, which are considered to have no PCN resistance genes, were always used as PCN-susceptible genotypes. Because Marijke, Alwara and Eden do not exist in Japan, DNA of these varieties were provided by Dr. John Bamberg in USDA, ARS, University of Wisconsin.
DNA markers used in this study are shown in Table 1. Although Gpa2 has already been isolated and well characterized, DNA markers only linked to Gpa2 have been reported (van der Voort et al. 1997, 2000). Therefore, we developed DNA markers directly amplifying Gpa2. Gpa2 and its related genes (RGC1, RGC3, PSH-RGH6, PHS-RGH7 and Gpa2-like NBS-LRR protein) (van der Vossen et al. 2000) were aligned using CLUSTALW, followed by manual alignment. We developed 2 markers (Gpa2-1 and Gpa2-2) amplifying specific regions of Gpa2. In addition to previously published marker, Gro1-4, we developed a more specific marker, Gro1-4-1, by aligning nucleotide sequences of Gro1-4 and genes of the Gro1 family (Gro1-1, Gro1-2, Gro1-3, Gro1-5, Gro1-6, Gro1-8, Gro1-10, Gro1-11, Gro1-12 and Gro1-14) (Paal et al. 2004) using CLUSTALW.
The total volume in the individual PCR assays was 10 μL, including 2 μL template DNA, 5 μL gene amplification reagent (Ampdirect® Plus, Shimadzu Corp., Kyoto, Japan), 0.25 units Taq DNA polymerase (BIOTAQTM HS; Bioline, London, UK) and the related primer pair shown in Table 1. Thermal cycling for Gro1-4-1, Gpa2-1, Gpa2-2 and GBSS1-3 was performed by using a 96-well thermal cycler (Veriti®; Applied Biosystems, Life Technologies, Carlsbad, CA). The PCR conditions consisted of 1 cycle of 10 min at 94°C, followed by 35 cycles of 30 s at 94°C, 30 s at 60°C and 1 min at 72°C and finally, 1 cycle of 5 min at 72°C. The PCR conditions for Gro1-4, HC, N146, N195, TG432 and C237 were as previously described (Finkers-Tomczak et al. 2009, Moloney et al. 2010, Paal et al. 2004, Sattarzadeh et al. 2006, Takeuchi et al. 2008). PCR products of TG432 and C237 were digested with restriction endonucleases RsaI and TaqI, respectively. All PCR products were separated by electrophoresis on 1.2–2% agarose gels in 1 × TAE buffer and visualized with SYBR SafeTM DNA gel stain (Invitrogen, Life Technologies, Carlsbad, CA) and UV.
N146 and N195 amplified expected-size bands (506 and 337 bp) in resistant varieties, Kita-akari, Touya and Haruka but not in Irish Cobbler and May Queen (Fig. 1A). Further, Gpa2-1 and Gpa2-2 amplified the expected-size bands (1120 and 452 bp, respectively) in Marijke, Touya and Aiyutaka but not in Irish Cobbler and May Queen (Fig. 1B). The expected-size band (602 bp) was amplified using published marker Gro1-4 only in Alwara; however, a slightly larger band was amplified in W872204 and W872204-2 (Fig. 1C). As the size difference is very slight, there is possibility that varieties or lines that have the larger band might be falsely judged as positive for Gro1-4. To avoid such erroneous decision, we developed a new PCR marker that specifically amplify Gro1-4 as described in Materials and Methods. The newly developed marker Gro1-4-1 amplified the expected 602-bp band only in Alwara (Fig. 1C). TG432, CAPS marker for Grp1, was tested with five resistant varieties (Astarte, Mara, Elles, Magic Red and Starter). However, we could not determine which band correlated with resistance (Fig. 2A). For the CAPS marker C237, Irida, P10111-1 and W804421-3 showed the same band pattern as positive control Eden; however, the negative control May Queen also showed the same pattern (Fig. 2B). HC did not yield a stable result; positive bands were also amplified in Irish Cobbler and May Queen irrespective of the PCR condition (Fig. 2C). Therefore, we finally used N146, N195, Gpa2-2 and Gro1-4-1 for evaluating the genetic resources.
To avoid erroneous judgment by PCR errors, a marker amplifying granule-bound starch synthase 1 (GBSS1), GBSS1-3, was combined with N146, N195, Gpa2-2 and Gro1-4-1. GBSS1 is shared in all potato species (Spooner et al. 2008), thus, it functions as a positive control to check whether the PCR was performed correctly (Mori et al. 2011). DNAs isolated from Touya (H1 and Gpa2) and Alwara (Gro1-4 and Gpa2) were equally mixed (DNA mixture), and was used as samples. The multiplex PCR conditions were optimized by varying the primer and Taq DNA polymerase concentrations and annealing temperatures. Consequently, simultaneous amplification was achieved with the primer concentrations shown in Table 1, 0.5 units of Taq DNA polymerase and annealing temperature at 60°C (Fig. 3).
To determine detection sensitivity of the multiple PCR method, the DNA mixture and DNA isolated from Irish Cobbler were assayed by the multiplex PCR method for a series of template DNA concentrations (60, 40, 20, 10, 5, 2.5, 1.3 and 0.6 ng per 10 μL reaction). In the DNA mixture, all marker bands were detected at 1.3 ng or higher concentrations, and Gro1-4-1 was not detectable at 0.6 ng. In Irish Cobbler, GBSS1-3 band was detected at concentrations as low as 0.6 ng (Fig. 4).
The DNA concentration of 30 randomly selected samples ranged from 3.1 ng/μL to 41.7 ng/μL, with an average of 14.1 ng/μL (SD = 9.1). Because the DNA concentration of the smallest sample was much higher than the detectable range, 2 μL of these DNAs were used as template DNA in a 10 μL PCR without adjusting the concentration.
The genetic resources were evaluated by the multiplex PCR method. As expected, N146 and N195 co-segregated perfectly. Of 812 varieties and lines, 264 varieties and lines were positive for the H1 markers (Table 2). Especially, these markers were detected in almost all varieties for which breeding program started after PCN invasion of Japan in 1972 (Table 3). We found 142 varieties and lines positive for the Gpa2 marker (Table 2). Of these, 117 varieties and lines were positive for both H1 and Gpa2 markers (Table 2). Among the varieties and a line mainly used as Japanese H1 resources (Tunika, Atlantic and R392-50), the Gpa2 marker was also detected in Atlantic and R392-50 but not in Tunika (Table 3, Fig. 5 and Supplemental Table 1). No accession was positive for the Gro1-4 marker (Table 2). None of the 3 markers were amplified in varieties such as Astarte, Mara and Magic Red, which are categorized as PCN resistant in The European Cultivated Potato Database or breeder’s evaluation (Supplemental Table 1).
As expected, one-third of the varieties and lines in our genetic resources possess H1. Especially, most varieties recently bred in Japan possess H1 (Tables 2, ,33 and Supplemental Table 1). Because Ro1 resistance is the priority for potato varieties in Japan, many varieties and lines inevitably have H1. Interestingly, many varieties and lines also possess Gpa2, although no selection for G. pallida resistance has been performed. Both H1 and Gpa2 were originally derived from the same S. tuberosum ssp. andigena CPC 1673 accession (van der Voort et al. 1997). Atlantic (Wauseon × Lenape) and R392-50 (Hudson × Wauseon) both having Gpa2, probably derived from Wauseon, and their progenies are frequently used for breeding of recent resistant varieties. Their frequent use might have resulted in the presence of Gpa2 in many recent varieties and lines. Alternatively, Gpa2 is closely located (less than 200 kb apart) to the Potato virus X (PVX) resistance gene Rx1, and therefore, these genes are genetically linked (van der Vossen et al. 2000). Along with PVX resistance, Gpa2 might also have been unconsciously introduced into many varieties. In fact, all varieties and a line with Rx1 (Aiyutaka, Atlantic, Kintoki-imo, Saco, Sayaka, Touya and Chokei 108) (Ohbayashi et al. 2010), also have Gpa2 (Supplemental Table 1).
In contrast to Gpa2, we could not find varieties and lines having Gro1-4, although our samples included PCN-resistant S. spegazzinii (PI 275143), a different accession of the S. spegazzinii and its progeny (W792224-1) (Supplemental Table 1). This fact emphasizes the importance of introduction of varieties or lines having Gro1-4 with a priority for our future safety. Our results revealed that only a very narrow genetic background in regard to the potato cyst nematode resistance resource has been used in Japanese potato breeding. Such a narrow genetic background of PCN resistance resources is disadvantageous from the viewpoint of breeding of novel PCN resistance varieties, because this may cause a bottleneck effect in the genetic background, and may result in the eventual genetic vulnerability of crops to the pest. Besides the 3 resistance genes analyzed in this study, we found other resistance genes in foreign varieties which are categorized as PCN resistant in the database and breeder’s evaluation. Their utilization and breeding of other novel resistant varieties should be promoted to broaden the genetic background in Japanese potato breeding. However, reliable DNA markers to detect these resistance genes have not developed yet. In fact, DNA markers TG432, C237 and HC were not applicable to the evaluation of resistant accessions in our study (Fig. 2). This is probably because of the complex genetic background of potato as pointed out by Milczarek et al. (2011). They tested the suitability of published molecular markers and suggested that most markers are inapplicable because of complex genetic backgrounds (Milczarek et al. 2011). DNA markers that accurately identify these resistance should be developed for efficient breeding.
Application of DNA markers for breeding has numerous advantages such as time saving, consistency, biosafety, efficiency and more accurate selection of complex traits (Collard et al. 2005, Jena and Mackill 2008). In this study we focused on the biosafety to prepare for PCN pathotypes not yet invading Japan. Usually, breeding programs target resistance to diseases and pests that are already epidemic in the region. In such cases, DNA markers are often used to reduce cost and time, to precisely and effectively evaluate complex traits and to simultaneously evaluate several resistance genes and pyramiding them. In fact, several studies that use marker-assisted selection target diseases already found in the area, and use DNA markers for such purpose (Biryukova et al. 2008, Gebhardt et al. 2006, Mori et al. 2011, Tan et al. 2010). To the best of our knowledge, this is first attempt in potato breeding to use DNA markers for preparations to prevent diseases and pests from invading the country. If functions of other gene/genes such as complementary gene are essential for expression of the resistance, some varieties selected based on only DNA markers evaluation may not show resistance. Therefore, it is necessary to confirm whether some of the candidate varieties and lines show resistance by inoculating other pathotypes, but no other pathotypes exist in Japan at this time. However, if we breed varieties possessing resistance genes, we will be able to choose resistance varieties among them when other pathotypes invade. These varieties enable us to quickly respond to invasion by other pathotypes. Our study demonstrates that DNA markers can be used to select potentially resistant accessions in potato breeding before infestation by new pathogens. Indeed, in addition to G. rostochiensis Ro1 resistance, we found many varieties and lines possibly resistant to distinct populations of G. pallida. This approach may be applicable to other quarantine diseases such as Columbia root-knot nematode and potato wart, DNA markers of which resistance have already developed (Gebhardt et al. 2006, Zhang et al. 2007). Moreover, because we evaluated many varieties used in potato breeding in Japan and several major varieties in the world, evaluation data in this study are very informative and useful for breeders. Breeders can devise schemes to breed novel resistance varieties based on these data. Furthermore, the multiplex PCR method developed in this study will contribute to efficient selection of ideal genotypes and facilitate breeding of novel resistant varieties.
We thank Dr. J. Bamberg in USDA, ARS, University of Wisconsin for providing DNA samples. We are also grateful to N. Taira, T. Yamada and A. Morizumi for field maintenance, S. Soma, H. Suzuki, H. Onuma, K. Okamoto, M. Okada and H. Ozaki for technical assistance and Dr. K. Hosaka and Dr. M. Mori for helpful suggestions. Almost genetic resources used in this study were distributed by the NIAS Genebank Project of the National Institute of Agro-biological Sciences.