Powdery mildew resistant roses

Abstract

The present invention relates to rose plants, such as cut roses, garden roses, and potted roses, that contain at least two genes that confer resistance to powdery mildew-causing pathogens. Specifically, the present invention relates to rose plants that are resistant to the powdery mildew-causing pathogen, Podosphaera pannosa (also known as Sphaerotheca pannosa var. rosae). The present invention also relates to a method for selecting powdery mildew-causing rose plants of the present invention. The rose plants are characterized by containing, in their nuclear genome, at least one nucleotide sequence represented by SEQ ID NO: 1 and at least one nucleotide sequence represented by SEQ ID NO: 2, and the coexistence of SEQ ID NO: 1 and SEQ ID NO: 2 confers resistance to powdery mildew.

Description

[Technical Field]

[0001] The present invention relates to rose plants, such as cut roses, garden roses, potted roses, and rose rootstocks, that contain at least two genes that confer resistance to a powdery mildew-causing pathogen. Specifically, the present invention relates to rose plants that are resistant to the powdery mildew-causing pathogen, Podosphaera pannosa (also known as Sphaerotheca pannosa var. rosae). The present invention further relates to a method for selecting powdery mildew-causing rose plants of the present invention. [Background technology]

[0002] Powdery mildew (PM) is a major foliar disease of cut flowers, potted plants, and garden roses, and is caused by the obligately biotrophic ascomycete fungus, Podosphaera pannosa (also known as Sphaerotheca pannosa var. rosae). Symptoms of powdery mildew infection include plant growth retardation and leaf deformation due to the formation of white, powdery conidia that appear on plant surfaces such as leaves, flowers, stems, and buds. There are no official figures for rose production losses due to powdery mildew, but controlling powdery mildew on three major greenhouse crops—cucumber, tomato, and rose—costs CAD $6,000 / ha in Canada.

[0003] Powdery mildew can be controlled with chemicals, but the use of fungicides is costly, labor-intensive, and environmentally damaging. Increasingly strict regulations on the use of chemicals in horticulture are being introduced by governments around the world, and combined with the factors mentioned above, this means that it is crucial to develop powdery mildew-resistant rose varieties to manage and avoid financial losses due to powdery mildew outbreaks.

[0004] Although P. pannosa can infect a wide range of hosts, including black cherry (Prunus cerasus) and sweet cherry (Prunus avium), its interaction with rose is characterized by high specificity. For example, several pathotypes of the fungus were described half a century ago, and recently, assays examining compatibility between eight monoconidial isolates and 18 host genotypes revealed a high level of specificity and varietal diversity.

[0005] Cut roses are susceptible to powdery mildew (PM). To breed PM-resistant cut roses, it is necessary to identify resistance genes. Several publications have provided evidence for the genetic basis of powdery mildew resistance in the genus Rosa, ranging from qualitative to quantitative resistance. However, commercially important cut roses (standard roses (Rosa hybrida)) are generally tetraploid, and quantitative trait loci (QTLs) for powdery mildew resistance have primarily been demonstrated in species or cultivars other than standard roses.

[0006] For example, using the diploid wild rose (Rosa multiflora) hybrid BC1 population (n = 117), which was created by crossing a resistant line (88 / 124-46) with a susceptible line (82 / 78-1) and backcrossing the F1 hybrid 95 / 13-90 with 82 / 78-1, the Rpp1 gene was found to be a main-effect gene on linkage group 3 that confers PM resistance up to 10 days post-inoculation in a dominant single-gene mode of action. Using different diploid wild rose hybrid F1 populations (n = 270), which were created by crossing a resistant line (95 / 13-39) with a susceptible line (Sp3 or 82 / 78-1) from the same open-pollinated breeding program intended to transfer genes from tetraploid horticultural rose to wild rose, QTLs were mapped for resistance in six different environments. In total, 28 different QTLs were found on linkage groups 1, 2, 3, 4, 6, and 7, with a strong clustering of QTLs on linkage groups 3 and 4, i.e., indicating a polygenic resistance mechanism. A substantial proportion (31%) of the observed phenotypic variance in resistance was inherited from the susceptible parent, and interestingly, the aforementioned effect of the Rpp1 gene was not exhibited in this population.

[0007] In a diploid cross (n=90) between the rose cultivar 'Yesterday' and Rosa wichurana, nine QTLs were found on linkage groups 2, 3, 5, and 6 for resistance to 10 days post-inoculation in the two cultivars, explaining 15% to 74% of the phenotypic variance, respectively. Of the nine QTLs, only one was detected in each cultivar. However, although the majority of resistance alleles were derived from Rosa wichurana, both parents contributed to the resistance variation.

[0008] The only example of a resistance-conferring QTL in tetraploid cut roses comes from the K5 population, an F1 population created by crossing two cultivars, P540 and P867, both of which were only partially resistant. Two different fungal monospore isolates were used to score resistance using three different disease scores: disease score 11 days postinoculation (dpi), latent period, and rate of symptom progression. For each of the three disease scores, 16 to 28 markers covering all seven linkage groups were shown to be associated with powdery mildew, although precise statistical support was not provided. While numerous markers were associated with different disease scores, no markers were detected in both isolates. Multiple regressions were performed using the four or five most significant markers for each of the six combinations of isolate and disease score. The phenotypic variance explained by the multiple regressions ranged from 10.4 to 22.3%, indicating that resistance in this cross was quantitative and controlled by multiple genes, each with only a small effect.

[0009] Thus, although several studies have published QTL for powdery mildew resistance in rose, a key theme is that there is little consistency or agreement among the published results. For example, resistance ranges from monogenic to quantitative. Furthermore, QTL are not shared across studies, suggesting that there may be heterogeneity in the underlying genetic mechanisms of fungal protein recognition, i.e., powdery mildew resistance. In part, this heterogeneity may be the result of inter- and intra-isolate genetic diversity, which, combined with gene-gene models hypothesizing interactions, explains that the location and effect size of QTLs depend on the isolate used.

[0010] Alternatively, lack of confirmation, often due to small sample size, low marker coverage, or a combination thereof, casts doubt on the robustness of the conclusions drawn. A diploid F1 population of Rosa roxburghii (cv. Guinong no. 6 × cv. Guinong no. 5) evaluated under natural disease pressure revealed that four resistance gene analogs (RGAs) that could not be assigned to linkage groups were associated with CRPM1, a major unassigned R locus in LG, explaining 72% of the phenotypic variance in powdery mildew resistance.

[0011] Powdery mildew resistance locus (MLO)-based resistance was first identified in barley (Hordeum vulgare) as a recessive allele conferring persistent resistance to all powdery mildew (Blumeria graminis f. sp. horde) isolates. MLO-based resistance is not unique to barley, as MLO orthologs have been found in many other plant genera, such as Pisum, Arabidopsis, and Solanum, and loss-of-function mutations in MLO genes have been shown to lead to broad-spectrum resistance to powdery mildew. The MLO protein family consists of seven clades and is widespread in land plants. However, the function of most MLO genes is unknown, and all MLO genes shown to be associated with powdery mildew resistance are found in clade V. In apple (Malus domestica), a member of the Rosaceae family, only three of 21 MLO genes are MLO homologs, two of which belong to clade V and are upregulated after powdery mildew infection. This suggests that detecting MLO homologs per se is not sufficient to identify genes involved in powdery mildew resistance.

[0012] Despite their durability, MLO-based resistance often has deleterious pleiotropic effects, such as necrotic leaf spot formation and reduced vigor, posing barriers to designing breeding strategies. In rose (diploid wild rose hybrid and tetraploid rose (standard) cv. Pariser Charme), four MLO genes, designated RhMLO1 through 4, were identified based on sequence homology with well-characterized sequences from barley and Arabidopsis. Each gene has two to six alleles, consists of 15 exons, and the total length of the coding sequence is approximately 1,700 bp in all cases. The MLO genes are distributed across several linkage groups, with rhMLO3 and rhMLO4 clustered on LG1 (between 40 and 45 cM), rhMLO2 on LG3 (approximately 35 cM), and rhMLO1 on LG5 at 60 cM.

[0013] Although no loss-of-function mutants are currently known in rose, all four MLO homologs have been shown to belong to clade V, the only clade in which MLO genes are known to be involved in PM resistance. Transformation of multiflora rose 'Baiyu' with antisense rhMLO1 resulted in plants with increased (but not absolute) levels of resistance to powdery mildew compared with non-transgenic control plants (up to 15 days post-inoculation), suggesting that only one of the four MLO genes found in rose is functionally associated with powdery mildew resistance. Summary of the Invention [Problem to be solved by the invention]

[0014] In view of the above, there is a need in the art for additional powdery mildew resistance-conferring genes.

[0015] It is an object of the present invention, inter alia, to fulfill the above-mentioned needs in the art.

[0016] According to the present invention, the above objects are met, inter alia, by providing a rose plant as outlined in the appended claims.

[0017] Specifically, the above-mentioned objects are met by providing a rose plant that is resistant to powdery mildew and that contains in its nuclear genome at least one nucleotide sequence represented by SEQ ID NO: 1 and at least one nucleotide sequence represented by SEQ ID NO: 2, and the coexistence of SEQ ID NO: 1 and SEQ ID NO: 2 in the nuclear genome confers resistance to powdery mildew. [Means for solving the problem]

[0018] The present inventors have surprisingly found that the combination of SEQ ID NO: 1 and SEQ ID NO: 2 confers strong resistance to powdery mildew. According to the present invention, the combination of both resistance genes is essential because SEQ ID NO: 1 does not confer detectable powdery mildew resistance in roses in the absence of SEQ ID NO: 2, whereas SEQ ID NO: 2 confers weak powdery mildew resistance in the absence of SEQ ID NO: 1. The present inventors have further surprisingly found that the resistance genes are dominant, i.e., the presence of a single copy of both genes is sufficient to confer powdery mildew resistance.

[0019] According to a preferred embodiment, the present invention relates to powdery mildew resistance to the ascomycete plant pathogen Podosphaera pannosa (also known as Sphaerotheca pannosa var. rosae). In rose, the ascomycete plant pathogen Podosphaera pannosa is the major powdery mildew-causing pathogen.

[0020] According to a further preferred embodiment, the rose plant is a standard rose plant, and the nuclear genome is a tetraploid genome. In particular, in polyploid genomes, such as diploid, triploid, tetraploid, hexaploid, or octaploid genomes, the availability of a dominant powdery mildew resistance gene offers a significant advantage in avoiding a multitude of intermediate powdery mildew phenotypes, depending on the copy number of the resistance gene present. The polyploid genomes, such as triploid, tetraploid, hexaploid, or octaploid genomes, can be obtained directly or indirectly by genome doubling, as appropriate, including SEQ ID NOs: 1 and 2. For example, tetraploids, hexaploids, and octaploids can be easily obtained from diploid genomes including SEQ ID NOs: 1 and 2, and hexaploid genomes can also be obtained by genome doubling of triploid plants.

[0021] According to a still further preferred embodiment, the present invention relates to powdery mildew resistant rose plants comprising in their nuclear genome at least one, preferably two, more preferably three, even more preferably four nucleotide sequences as set forth in SEQ ID NO: 1, such as 2, 3, 4, 5 or 6 in the case of a hexaploid genome or 2, 3, 4, 5, 6, 7 or 8 in the case of an octoploid genome, and / or at least one, preferably two, more preferably three, even more preferably four nucleotide sequences as set forth in SEQ ID NO: 2, such as 2, 3, 4, 5 or 6 in the case of a hexaploid genome or 2, 3, 4, 5, 6, 7 or 8 in the case of an octoploid genome.

[0022] According to a still further preferred embodiment, the present invention relates to powdery mildew resistant rose plants comprising in their nuclear genome at least one, preferably three, more preferably four, such as 2, 3, 4, 5 or 6 in the case of a hexaploid genome or 2, 3, 4, 5, 6, 7 or 8 in the case of an octoploid genome, nucleotide sequences as set forth in SEQ ID NO: 1, or at least one, preferably two, more preferably three, and even more preferably four, such as 2, 3, 4, 5 or 6 in the case of a hexaploid genome or 2, 3, 4, 5, 6, 7 or 8 in the case of an octoploid genome, nucleotide sequences as set forth in SEQ ID NO: 2.

[0023] According to a still further preferred embodiment, the present invention relates to powdery mildew resistant rose plants comprising in their nuclear genome at least one, preferably two, more preferably three, and even more preferably four nucleotide sequences as set forth in SEQ ID NO: 1, such as 2, 3, 4, 5 or 6 in the case of a hexaploid genome, or 2, 3, 4, 5, 6, 7 or 8 in the case of an octoploid genome, and at least one, preferably two, more preferably three, and even more preferably four nucleotide sequences as set forth in SEQ ID NO: 2, such as 2, 3, 4, 5 or 6 in the case of a hexaploid genome, or 2, 3, 4, 5, 6, 7 or 8 in the case of an octoploid genome.

[0024] According to a particularly preferred embodiment, the powdery mildew resistant rose plant is selected from the group consisting of cut roses, potted roses, rose rootstocks and garden roses, preferably cut roses.

[0025] According to yet another particularly preferred embodiment, the powdery mildew resistant rose plant exhibits a dominant phenotype.

[0026] Considering the beneficial property of powdery mildew resistance, which is conferred by a synergistic epistatic effect between two dominant genes, the present invention further provides a method for selecting a powdery mildew resistant rose plant as defined above, comprising the steps of: a) isolating nuclear genomic DNA from a rose plant; b) establishing the presence of SEQ ID NO: 1 and SEQ ID NO: 2 in the isolated nuclear genomic DNA; and c) establishing the powdery mildew phenotype of said rose plant, wherein the presence of SEQ ID NO: 1 and SEQ ID NO: 2 is indicative of a powdery mildew resistance phenotype.

[0027] The invention will now be described in further detail in the following examples, in which reference is made to the figures. [Brief explanation of the drawings]

[0028] [Figure 1]Box plot showing the effect of the resistance alleles SEQ ID NO: 1 and SEQ ID NO: 2 separately and in series. The presence of the resistance allele is indicated by a "+" and the absence of the resistance allele is indicated by a "-" symbol. Plants with both resistance alleles are highly resistant. [Figure 2] FIG. 1 shows the copy numbers of SEQ ID NO: 1 and SEQ ID NO: 2 required to confer powdery mildew resistance to rose plants. DETAILED DESCRIPTION OF THE INVENTION [Example]

[0029] Introduction Here, we investigated the number, effect sizes, and genetic locations of QTLs underlying PM resistance in a tetraploid F1 rose population (standard rose). We present findings demonstrating distinct main-effect QTLs on linkage groups 1 and 5, explaining 20% and 90% of the phenotypic variance in PM resistance, respectively. We also show that the effect of the QTLs is only seen when resistance alleles are present at both QTLs, as plants carrying resistance alleles at both QTLs are all highly resistant up to 15 weeks postinoculation, whereas plants carrying either one or no resistance alleles develop PM symptoms within this period.

[0030] method A tetraploid F1 rose (standard) population was generated by hand-pollinating the tetraploid cut rose RS-1183 ('Avalanche', hereafter referred to as P1) with pollen from a tetraploid horticultural rose. One of the resulting F1 progeny was self-pollinated to generate the F2 population. The parents, 235 F1 progeny, and 42 F2 plants were screened for resistance to PM (Podosphaera pannosa). Isolates were originally isolated from infected roses in a horticultural greenhouse, and inoculum was obtained from a previous PM assay. Inoculation was performed in a block design, with six cuttings from each cultivar randomized across six blocks. The bioassay was conducted under long-day conditions with day and night temperatures set at 20°C and 23°C, respectively. Relative humidity was maintained at 60% during the day and 85% at night. For each plant in the F1 population, the infection level was scored at 1, 3, 6, 9, 12, and 15 weeks after infection, and for each plant in the F2 population, the infection level was scored at 6 and 12 weeks after infection, and the infection level of plants in both populations was scored on a scale of 1 to 9, with 1 representing the most susceptible individual and 9 representing completely resistant individuals.

[0031] All plants were genotyped using the WagRhSNP Axiom SNP array. This chip contains 68,893 SNPs targeted by two probes from each direction. Quality control was performed using the R package FitPoly, retaining 67,779 markers across 51,685 SNPs. After removing SNPs with 5% or more missing data, 42,143 markers remained. A total of 232 F1 individuals were successfully genotyped, of which three were genetic outliers and one was removed due to missing phenotypic data. Further quality control was performed by checking for reproducibility of parental genotypes, unexpected segregation, genotypic outliers, skewed markers and null alleles, and differences between plates.

[0032] Using a previously obtained genetic map (using the K5 population), these correlated SNPs were mapped to linkage groups (LGs) and genetic locations. All associated SNPs were segregated according to a scenario in which the resistant parent was simplex and the susceptible parent was nulliplex. For chromosomes where QTLs were found, linkage maps were constructed using JoinMap with markers that were simplex at P2 and nulliplex at P1, and QTL analysis was performed using MapQTL.

[0033] For each genomic region significantly associated with PM resistance, KASP primers were designed targeting the most significantly associated SNP as well as one SNP on either side. KASP primers were designed using the flanking sequences of probes targeting the associated SNP on the WagRhSNP Axiom SNP array.

[0034] The parents and a total of 48 randomly selected F1 plants were genotyped at all SNPs using the KASP assay. Genotypes were recorded as the number of resistance alleles possessed by the individual. Because the resistant parent had one copy of the resistance allele at every relevant SNP and the susceptible parent had zero, the amount of genotypes in the F1 was restricted to 0 (null genotype of the resistance allele) and 1 (simple genotype of the resistance allele).

[0035] result A total of 267 markers were correlated with PM resistance by >0.35. All highly correlated markers were found on linkage groups 1 and 5 on the genetic map obtained using the K5 population. As with the map obtained using the K5 population, the order of markers included in our genetic map was preserved to ensure successful construction of linkage maps for these two linkage groups.

[0036] Three weeks after inoculation, SNP M23333_428 in homolog 5.2 explained up to 90% of the phenotypic variance (LOD = 114.2). A second SNP, G54183_559, was found at 60.5 cM on LG1 of homolog 1.1 (LOD = 23.1) at 15 weeks after inoculation, explaining 20.3% of the phenotypic variance. Analysis of QTLs together using multiple QTL models indicated that QTLs in homologs 5.2 and 1.1 were required for absolute resistance after 15 weeks. Using the KASP assay, a total of 86 plants with PM resistance data were genotyped, including 48 F1 progeny, 29 F2 plants (sib progeny of a single selfed F1 plant), and 4 P1 and P2 plants (including both duplicates).

[0037] We first analyzed the association between SNP genotype and PM resistance in the F1 population. Genotyping call rates varied between 87% (for G8670_490) and 100%. When examining the association between KASP genotype and PM resistance, the presence of resistance alleles at the most strongly associated SNPs in both homologs strongly indicated PM resistance at 15 weeks postinoculation. All plants with this genotype combination exhibited PM scores greater than 8 (high resistance, Figure 1), whereas plants with resistance genotypes at one locus or none at all were never highly resistant but primarily highly susceptible (Figure 2). Analysis using ANOVA showed highly significant synergistic epistatic effects (Table 1). Further genomic analysis of rose plants yielded SEQ ID NOs: 1 and 2, which are directly linked to the resistance genes underlying this resistance.

[0038] [Table 1]

[0039] After demonstrating that the presence of resistance genes at both loci is necessary to confer resistance, we next examined whether the mechanism of action at each locus is completely dominant, i.e., whether there is no difference in PM resistance between plants with a single resistance allele at each locus and plants with multiple resistance alleles at each locus. To do this, we combined data from the F2 population with data from the parents and F1 populations. The F2 population was a self-pollinated population obtained by self-pollinating F1 plants with a single resistance allele at each locus. Therefore, assuming polysomic inheritance, we would expect plants with 0, 1, and 2 copies at each locus in the resulting dataset. PM resistance in the F2 population was assayed up to 12 weeks post-inoculation. PM resistance at 12 weeks post-inoculation was strongly correlated with PM resistance at 15 weeks post-inoculation (r = 0.98), meaning that restricting our analysis to the 12-week post-inoculation data did not significantly affect our conclusions.

[0040] Indeed, it was clearly shown that one resistance allele at each locus was sufficient to confer absolute resistance at 12 weeks postinoculation, and the presence of multiple resistance genes per locus did not confer significant additional resistance (Fig. 2), providing clear evidence that resistance alleles are dominant over susceptible alleles.

Claims

[Claim 1] A method for selecting powdery mildew resistant rose plants, said method comprising: a) isolating nuclear genomic DNA from a rose plant; b) establishing the presence of SEQ ID NO: 1 and SEQ ID NO: 2 in the isolated nuclear genomic DNA; c) establishing the powdery mildew phenotype of said rose plant, wherein the presence of SEQ ID NO: 1 and SEQ ID NO: 2 indicates a powdery mildew resistance phenotype. Including, The method, wherein said rose plant is not obtained by genetic modification.

Images