2020 Volume 70 Issue 4 Pages 502-507
Japanese domestic tobacco (Nicotiana tabacum L.) cultivar ‘Kokubu’ shows high powdery mildew resistance controlled by recessive alleles at two loci, and these alleles have been widely used as a resource for powdery mildew resistance in tobacco breeding. However, the introduction of this trait by conventional breeding takes much work because of the requirement for test crosses with the parental strains and inoculation tests using active fungi to confirm the introduction of two recessive alleles during back-crossing. Recently, we found that powdery mildew resistance in ‘Kokubu’ is caused by splice site mutations of two MILDEW LOCUS O genes, NtMLO1 and NtMLO2. Here, we report DNA markers that detect mutations of the NtMLO1/2 genes based on the cleaved amplified polymorphic sequence (CAPS) or allele-specific polymerase chain reaction (AS-PCR) methods. These markers can be used as co-dominant markers that detect heterozygotes of the NtMLO genes at the seedling stage in back-crossed progenies, and will contribute to the simplification of breeding.
Powdery mildew is a fungal disease affecting various grains and crops, causing serious yield losses in environments with high humidity and moderate temperatures. Infected plants display white powdery spots on the leaves and stems. Powdery mildew diseases are caused by many different species of fungi in the order Erysiphales, and each fungal species infects its specific plant host. Golovinomyces cichoracearum var. cichoracearum (syn. Erysiphe cichoracearum) is the powdery mildew pathogen of tobacco (Nicotiana tabacum L.), and causes losses of yield and quality in southern Africa, southeastern Europe, the Middle East and Asia. Several genes for powdery mildew resistance have been found in various Nicotiana species and have been used in tobacco breeding. The wild tobacco species N. debneyi, N. glutinosa, and N. tomentosiformis harbor a single dominant locus for powdery mildew resistance (Davis and Nielsen 1999). The Japanese domestic tobacco cultivar ‘Kokubu’ shows powdery mildew resistance controlled by recessive alleles at two loci (Wan 1962). These loci have been used as a genetic resource for breeding powdery mildew resistance and introduced into several tobacco cultivars in Japan (Hamamura et al. 1981). However, the introduction of this trait in breeding takes much work. Since the powdery mildew-resistant alleles of ‘Kokubu’ are recessive, the heterozygotes of these alleles, such as the progenies of crosses between resistant and susceptible plants, are susceptible to powdery mildew. Therefore, to determine the zygosity and isolate heterozygotes, test crosses (crossing between backcross progenies and the donor parent) and subsequent inoculation tests on the progeny population are required in addition to the ordinary backcrosses (crossing between backcross progenies and the recurrent parent). Moreover, maintenance of tobacco plants infected with powdery mildew is required to keep fresh conidiospores for the inoculation tests, because the powdery mildew fungi are obligate parasites, requiring live hosts to grow and reproduce.
Recently, the causative genes for powdery mildew resistance in ‘Kokubu’ were identified (Fujimura et al. 2016). This resistance is due to splice site mutations in two MLO genes, NtMLO1 and NtMLO2. The MLO genes encode a plant-specific seven-transmembrane domain protein localized in the plasma membrane (Devoto et al. 2003). MLO genes have been found in various plant species and consist of gene families. Loss-of-function mutants of MLO alleles show powdery mildew resistance in various dicot and monocot species such as pea (Pavan et al. 2011), Arabidopsis (Consonni et al. 2006), tomato (Bai et al. 2008), cucumber (Nie et al. 2015), and barley (Büschges et al. 1997). Phylogenetically, MLO proteins are divided into seven clades, and all MLO proteins from dicot species that are known to be associated with powdery mildew susceptibility belong to clade V (Acevedo-Garcia et al. 2014). Tobacco contains 15 MLO genes, which are classified into six clades, and the proteins NtMLO1 and NtMLO2 are the sole MLO proteins belonging to clade V (Appiano et al. 2015). It is expected that discovery of the DNA sequences responsible for powdery mildew resistance in ‘Kokubu’ should pave the way to developing DNA markers useful for the breeding of resistance. In this study, we developed co-dominant DNA markers using the cleaved amplified polymorphic sequence (CAPS) and allele-specific PCR (AS-PCR) methods for the detection of powdery mildew resistance genes derived from ‘Kokubu’. These easy-to-use and reliable methods should contribute to saving both labor and materials in the breeding of powdery mildew resistance in tobacco.
Japanese domestic tobacco cultivar ‘Kokubu’ and flue-cured tobacco cultivar ‘Tsukuba 1’, whose NtMLO genes contain splice site mutations, were used as the powdery mildew-resistant cultivars. As powdery mildew-susceptible cultivars, domestic tobacco cultivar ‘Ibusuki’ and flue-cured tobacco cultivar ‘K326’ were used. ‘Kokubu’ and ‘Tsukuba 1’ were crossed with ‘Ibusuki’ and ‘K326’, respectively and the F1 hybrids obtained were used as the heterozygotes of NtMLO genes. These F1 hybrids showed susceptibility to powdery mildew because the powdery mildew resistance of ‘Kokubu’ and ‘Tsukuba 1’ is controlled by two recessive loci. Genomic DNA was extracted from green leaves using a Gentra Puregene Cell kit (QIAGEN) according to the manufacturer’s instructions. The genomic DNA obtained was used directly as a template for DNA fragment amplification using the CAPS and AS-PCR methods.
Cleaved amplified polymorphic sequence (CAPS)The primer pairs were designed for the amplification of DNA fragments containing polymorphic sequences of each NtMLO gene (Table 1). Amplification of DNA fragments with the QIAGEN Multiplex PCR Kit was performed according to the manufacturer’s instructions. PCR was carried out in 10 μl reactions that consisted of 5 ng genomic DNA, 5 μl of QIAGEN Multiplex PCR Master Mix, and 0.1 μM of each primer. After preheating at 95°C for 15 min, 40 PCR cycles (94°C for 30 sec, 60°C for 90 sec and 72°C for 60 sec) were performed. The last cycle was followed by a final 72°C incubation for 10 min. Restriction enzymes that recognize the polymorphic sequences of each NtMLO gene were identified using Genetyx v. 10 software (Genetyx Corporation, Tokyo, Japan). The following enzymes were selected for the digestion of the DNA: BfaI (New England Biolabs) or BciT130I (Takara) for NtMLO1 fragment digestion; BseRI (New England Biolabs) for NtMLO2 fragment digestion. BfaI and BciT130I digest the mutant and wild-type NtMLO1, respectively; BseRI digests the mutated NtMLO2. All digestion reactions were carried out according to the manufacturer’s instructions. Ten μl of reaction solution was consisted of 3 μl of PCR product and 2.5–5.0 units of appropriate restriction enzyme. After one-hour digestion at 37°C, the restriction enzyme was denatured by heating. The DNA fragments were electrophoresed by 3% agarose gel in 0.5 × Tris-borate-EDTA (TBE) buffer, and the patterns of bands were confirmed by GelRed (Biotium Inc.) staining.
| Primer pairs for CAPS | ||
| NtMLO1 | F-primer | 5ʹ-TTCAGAATTATATTCTTCCCTCCC-3ʹ |
| R-primer | 5ʹ-TGACAGTATTGGTAAAAAGTTTCTG-3ʹ | |
| NtMLO2 | F-primer | 5ʹ-CCAGAGAGGTTCAGGTTTGCAAGAG-3ʹ |
| R-primer | 5ʹ-TTTTGAACCCCCTTCGTGAAGATCC-3ʹ | |
| Primer pairs for allele-specific PCR | ||
| NtMLO1 | F-primer | 5ʹ-TCAAATTTTCAGAATTATATTCTTCCCTCCC-3ʹ |
| Wild type R-primer | 5ʹ-CCTGATTCTGTGGAGTTAAATGTGCCTGG-3ʹ | |
| Mutant R-primer | 5ʹ-TCTGTGGAGTTAAATGTGCTAGG-3ʹ | |
| NtMLO2 | Wild type F-primer | 5ʹ-TGCTGCTCTGGATAGTCAGTC-3ʹ |
| Mutant F-primer | 5ʹ-GTGCTGCTCTGGATAGTCTC-3ʹ | |
| R-primer | 5ʹ-ACCCCCTTCGTGAAGATCC-3ʹ | |
In the AS-PCR method, PCR primer pairs were designed for the separate amplification of wild-type or mutant DNA fragments (Table 1) referring to the previous report (Liu et al. 2012). For the amplification of NtMLO1 DNA fragments, the substituted dinucleotide that was observed at the 3ʹ-end of intron 7 was placed on the 3ʹ-end of the reverse primers, and the DNA fragments were amplified separately using the same forward primer and either wild-type-specific or mutant-specific reverse primer. Amplification of NtMLO2 fragments were carried out using primer pairs that contained an allele-specific region at the 3ʹ-end of the forward primer. PCR were carried out as a duplex reaction by the two primer pairs that amplify wild-type or mutant DNA fragments of both NtMLO genes. The reaction volume was 10 μl including 5 ng genomic DNA, 5 μl of QIAGEN Multiplex PCR Master Mix, and 0.1 μM of each primer. After preheating at 95°C for 10 min, 36 PCR cycles (94°C for 30 sec, 60°C for 20 sec and 72°C for 30 sec) were performed, and followed a final 72°C incubation for 10 min. Amplified DNA fragments were separated by 0.8% agarose gel electrophoresis in 0.5 × TBE buffer and detected by GelRed staining.
The CAPS method detects polymorphism caused by nucleotide changes in the restriction endonuclease recognition sites in a gene. In this method, DNA fragments containing the polymorphic sequence of a target gene are amplified and then digested with the specific restriction enzymes. After the enzyme reaction, DNA fragments are separated by electrophoresis and the genotypes are indicated by the pattern of bands. For the detection of NtMLO1 gene polymorphism, DNA fragments of 283 bp that include the junction region of intron 7 and exon 8 of NtMLO1 were amplified by PCR, and the fragments obtained were then digested by BciT130I or BfaI. BciT130I recognizes the sequence 5ʹ-CCWGG-3ʹ and BfaI recognizes the sequence 5ʹ-CTAG-3ʹ; these recognition sequences coincide with the sequence of the junction region. Therefore, BciT130I and BfaI digest the DNA fragments amplified from the wild-type (susceptible) and mutant (resistant) NtMLO1 gene, respectively (Fig. 1). Electrophoresis results (Fig. 2) showed that the DNA fragments derived from resistant cultivars (lane 1: ‘Kokubu’ and lane 4: ‘Tsukuba 1’) were digested by BfaI and produced two fragments (183 bp and 100 bp). In contrast, the DNA fragments derived from susceptible cultivars (lane 2: ‘Ibusuki’ and lane 5: ‘K326’) were not digested by BfaI. BciT130I recognized and digested the DNA fragments of the wild-type NtMLO1 gene and generated two fragments (183 bp and 100 bp). The DNA fragments derived from the heterozygote, an F1 hybrid of powdery mildew-resistant and susceptible cultivars (lane 3: ‘Kokubu’ × ‘Ibusuki’ and lane 6: ‘Tsukuba 1’ × ‘K326’), showed three fragments of different lengths in electrophoresis after digestion by each restriction enzyme. These DNA fragments comprised one longer undigested fragment and two shorter digested fragments derived from wild-type and mutant genes. Thus, using the CAPS markers we developed, we were able to distinguish the wild-type, mutant, or heterozygote of NtMLO1 gene by the electrophoretic pattern of digested DNA fragments.
Alignment of NtMLO genomic DNA sequences and restriction enzyme recognition sites used in the cleaved amplified polymorphic sequence (CAPS) method. The NtMLO1 gene of powdery mildew-resistant cultivars harbors two-base transitions (AG to TA) at the 3ʹ end of intron 7. In the NtMLO2 gene, the fourth and fifth nucleotide (AG) of intron 6 are deleted in powdery mildew-resistant cultivars. The CAPS method detects these polymorphisms by restriction enzyme digestion patterns. BciT130I and BfaI recognize and digest the wild-type (powdery mildew-susceptible) and mutant (powdery mildew-resistant) NtMLO1 genomic DNA sequence, respectively. BseRI recognizes and digests the mutant (powdery mildew-resistant) NtMLO2 genomic DNA sequence. Intron sequences are shown in bold letters. Arrow heads indicate the digestion site of each enzyme.
Agarose gel electrophoresis profiles of NtMLO alleles detected using the cleaved amplified polymorphic sequence (CAPS) method. Amplified DNA fragments were digested by BciT130I or BfaI for NtMLO1, and BseRI for NtMLO2. BciT130I and BfaI digests the wild-type (susceptible) and mutant (resistant) NtMLO1 fragments, respectively, and BseRI digests mutant (resistant) NtMLO2 fragments. In the heterozygotes, three (digested and undigested) DNA fragments are generated by the restriction enzyme treatment. 1: ‘Tsukuba 1’ (resistant); 2: ‘K326’ (susceptible); 3: F1 hybrid of ‘Tsukuba 1’ × ‘K326’ (susceptible); 4: ‘Kokubu’ (resistant); 5: ‘Ibusuki’ (susceptible); 6: F1 hybrid of ‘Kokubu’ × ‘Ibusuki’ (susceptible).
For the detection of NtMLO2 gene polymorphism, DNA fragments that included the junction region of exon 6 and intron 6 of NtMLO2 were amplified by PCR, and the fragments obtained were digested by BseRI. BseRI recognizes the sequence 5ʹ-CTCCTC-3ʹ and digests asymmetrically at the eight-base position upstream of the recognition site. BseRI recognizes the two-base deleted sixth intron 5ʹ-splice site sequence 5ʹ-GTCTCCTC-3ʹ of the mutant (resistant) NtMLO2 gene but the same region of the wild-type (susceptible) NtMLO2 gene (5ʹ-GTCAGTCC-3ʹ) is not recognized by this enzyme (Fig. 1). Unfortunately, enzymes that specifically recognize the wild-type NtMLO2 DNA sequence were not among the commercially available restriction enzymes. Electrophoresis results (Fig. 2) showed that the DNA fragments of resistant cultivars (lanes 1 and 4) were digested by BseRI to give two fragments of 257 bp and 81 bp. The DNA fragments derived from susceptible cultivars (lanes 2 and 5) were not digested by BseRI, and the DNA fragments derived from heterozygotes (lanes 3 and 6) showed three fragments of different sizes (340 bp, 257 bp and 81 bp) in the electrophoresis profile after digestion. Thus, a co-dominant CAPS marker detecting NtMLO2 polymorphism was also successfully developed.
These results showed that resistance to powdery mildew is detectable by the electrophoresis profiles. The amplified DNA fragments derived from powdery mildew-resistant tobacco plants were digested by BfaI (NtMLO1) and BseRI (NtMLO2) and not digested by BciT130I (NtMLO1). Detection of heterozygotes is essential for the utilization of DNA markers in backcrossing. Heterozygotes of the NtMLO genes are detectable by the electrophoresis profiles that show three DNA fragments in all three enzyme digestions. As all three restriction enzyme digestions were able to distinguish the DNA fragments derived from wild-type, mutant and heterozygote of NtMLO1 or NtMLO2, these CAPS markers effectively function as co-dominant markers.
Allele-specific PCRAllele-specific PCR (AS-PCR) detects polymorphism caused by a single nucleotide change in genomic DNA by analyzing PCR products. In this approach, the allele-specific DNA fragments are amplified using specific primers differing in the 3ʹ-end nucleotides. In our assay, we designed allele-specific primers that harbored different dinucleotides in each NtMLO gene at the three- and four-base position upstream of the 3ʹ end (Fig. 3). In addition, the opposite primer for each NtMLO gene was arranged at positions that amplified DNA fragments of different sizes for NtMLO1 (134 bp) and NtMLO2 (260 bp). This primer pair arrangement enabled duplex PCR and detected both fragments in a single electrophoresis gel. Therefore, the genotypes of both NtMLO genes were distinguishable by two PCR events that detected the wild-type or mutant alleles of both genes. In powdery mildew-resistant cultivars (lane 1: ‘Kokubu’ and lane 4: ‘Tsukuba 1’ of Fig. 4), two DNA fragments of different sizes, derived from NtMLO1 and NtMLO2, were amplified using mutant NtMLO gene-specific primer pairs, and no DNA fragment was amplified by wild-type NtMLO gene-specific primer pairs. In contrast, powdery mildew-susceptible cultivars (lane 2: ‘Ibusuki’ and lane 5: ‘K326’) showed the opposite results from the resistant cultivars after duplex PCR using mutant and wild-type NtMLO gene-specific primer pairs. In the heterozygotes between powdery mildew-resistant and susceptible cultivars (lane 3: ‘Kokubu’ × ‘Ibusuki’ and lane 5: ‘Tsukuba 1’ × ‘K326’), two DNA fragments of different sizes were amplified by both primer pairs. These results suggest that the duplex PCR successfully determined the genotype for both NtMLO genes.
Alignment of the genomic DNA sequences and specific primers for the detection of resistant or susceptible alleles of NtMLO genes by allele-specific PCR. Specific primers (shaded sequences) differing in the 3ʹ-end nucleotides were used for the amplification of allele-specific DNA fragments. For each NtMLO gene, the opposite direction primers were the same sequence for the amplification of both alleles (Table 1).
Allele-specific PCR electrophoresis profiles of powdery mildew-resistant and susceptible cultivars. As the size of the amplified DNA fragment differs between NtMLO1 (134 bp) and MtMLO2 (260 bp), the determination of powdery mildew-resistant/susceptible genotype is performed by duplex PCR with wild-type-specific and mutant-specific primers, then subsequent electrophoresis. The relation between the electrophoresis profile and genotype or phenotype is summarized in the table. Lane 1: ‘Tsukuba 1’ (resistant); 2: ‘K326’ (susceptible); 3: F1 hybrid of ‘Tsukuba 1’ × ‘K326’ (susceptible); 4: ‘Kokubu’ (resistant); 5: ‘Ibusuki’ (susceptible); 6: F1 hybrid of ‘Kokubu’ × ‘Ibusuki’ (susceptible).
Recessive powdery mildew-resistant loci derived from Japanese cultivar ‘Kokubu’ have been widely used and introduced into various tobacco cultivars. However, the introduction of these loci has required considerable effort. In the traditional breeding procedure for the introduction of recessive traits, two types of crossing experiments have been performed in each backcross program, namely backcrossing with the recurrent parental cultivar for the transfer of the desired trait and test crosses with the resistant parental cultivar for the confirmation of the genotype. Then, the test-cross progenies are cultivated and the phenotype of each progeny is investigated to determine the genotype of each backcrossed plant. Theoretically, a backcrossed plant harboring two heterozygote recessive loci generates test-cross progenies showing the desired trait at a rate of one fourth. Moreover, specific examinations, such as pathogen inoculation tests, are needed for genotype determination, depending on the desired trait. Therefore, for the introduction of powdery mildew resistance by traditional breeding, special facilities such as a containment greenhouse are required to cultivate the test-crossed progenies infected artificially with powdery mildew. Thus, the spread of the disease to other tobacco plants can be prevented.
In this study, we provide two methods of detecting recessive powdery mildew-resistant loci of tobacco derived from mutations of the NtMLO1 and NtMLO2 genes, which originated in the Japanese domestic cultivar ‘Kokubu’. These two methods have several advantages compared with the standard procedure of recessive trait breeding: 1) test crosses for the pathogen inoculation tests are not needed; 2) genotypes can be determined in young seedlings; 3) assays can be carried out using common genetic experiment devices such as a PCR machine and electrophoresis equipment. In the method described in this manuscript, we used a commercially-available genomic DNA extraction kit, but the genomic DNA extracted by easier and less expensive ways that does not affect amplification reaction might be available as the template of PCR. In addition, the small amount of genomic DNA extracted from young seedlings grown on cell culture trays is sufficient for DNA amplification in our methods. Therefore, using our DNA markers, the genotype determination and the heterozygote selection are practicable at the early growth stage of young seedlings, and the space needed for the cultivation of backcrossed plants is minimized. Both DNA markers described in this manuscript are co-dominant markers that can distinguish three genotypes, namely homozygous mutant, heterozygote and wild type; this capability is useful for breeding to introduce recessive traits.
The CAPS method is superior as the co-dominant marker detecting three genotypes by one electrophoresis operation, but this method entails digestion by restriction enzymes after DNA fragment amplification. In contrast, our AS-PCR method is able to determine the genotypes for two NtMLO genes by the electrophoresis profiles of duplex PCR products. Therefore, the AS-PCR method might be simpler and less expensive than the CAPS method for application in breeding programs. However, since the primers for AS-PCR method are designed based on the two-nucleotide difference between wild-type and mutant NtMLO genes, PCR in this method is subject to the quality of genomic DNA solution. On the other hand, amplification of NtMLO DNA fragments by the CAPS method is relatively stable regardless of the genomic DNA quality. Therefore, the CAPS method seems to be more reliable than the AS-PCR method. In order to check the PCR process in the AS-PCR method, the co-amplification of known housekeeping genes designed to amplify the fragments of different size from amplified NtMLO DNA fragments will be effective. In either method, we recommend the application of the powdery mildew resistance and susceptible cultivar samples as the control to verify that each step is going well.
The two methods described in this paper will make it possible to breed tobacco powdery mildew resistance more speedily, easily, reliably and inexpensively.
TK and HU contributed the development of AS-PCR method. SS contributed the development of CAPS method. TT and MA designed this study. MA drafted the manuscript. All authors read and approved the final manuscript.
