Resistance Genes to Agrobacterium Tumefaciens Infections in Rose

Abstract

Provided herein are rose plants including one or more resistance genes to Agrobacterium tumefaciens. Also provided herein are plant parts, cells or reproductive tissue of the plants disclosed herein and to methods for identifying Agrobacterium tumefaciens resistant rose plants. Specifically, provided herein is an Agrobacterium tumefaciens resistant rose plant, wherein the Agrobacterium tumefaciens resistance is provided by one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb and especially one or more genes selected from the group consisting of SEQ ID Nos. 1 to 5.

Claims

1. An Agrobacterium tumefaciens resistant rose plant, wherein said Agrobacterium tumefaciens resistance is provided by one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb

2. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said Agrobacterium tumefaciens resistance is provided by one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb comprising one or more sequences selected from the group consisting of SEQ ID Nos. 11 to 14.

3. The Agrobacterium tumefaciens resistant rose plant according to claim 2, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 6, or sequences with 95% identity therewith, wherein the cDNA sequence of SEQ ID No. 6, or sequences with 95% identity therewith, comprise SEQ ID No. 11.

4. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 7, or sequences with 95% identity therewith, wherein the cDNA sequence of SEQ ID No. 7, or sequences with 95% identity therewith, comprise SEQ ID No. 12.

5. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 9, or sequences with 95% identity therewith, wherein the cDNA sequence of SEQ ID No. 9, or sequences with 95% identity therewith, comprise SEQ ID No. 13.

6. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 10, or sequences with 95% identity therewith, wherein the cDNA sequence of SEQ ID No. 10, or sequences with 95% identity therewith, comprise SEQ ID No. 14.

7. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 1.

8. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 2.

9. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 3.

10. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 4.

11. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb is a gene encoding the cDNA sequence of SEQ ID No. 5.

12. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said rose is a hybrid.

13. The Agrobacterium tumefaciens resistant rose plant according to wherein said rose plant comprises at least 2 copies of said one or more genes providing Agrobacterium tumefaciens resistance.

14. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said rose plant comprises a tetraploid genome and at least 3 copies, preferably 4, of said one or more genes providing Agrobacterium tumefaciens resistance.

15. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb are a gene encoding the cDNA sequence of SEQ ID No. 1 and a gene encoding the cDNA sequence of SEQ ID No. 2; or a gene comprising SEQ ID No. 11 and a gene comprising SEQ ID No. 12.

16. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb are a gene encoding the cDNA sequence of SEQ ID No. 1; a gene encoding the cDNA sequence of SEQ ID No. 2 and a gene encoding the cDNA sequence of SEQ ID No. 3; or a gene of comprising SEQ ID No. 11, a gene comprising SEQ ID No. 12 and a gene comprising SEQ ID No. 3.

17. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb are a gene encoding the cDNA sequence of SEQ ID No. 1; a gene encoding the cDNA sequence of SEQ ID No. 2, a gene encoding the cDNA sequence of SEQ ID No. 3 and a gene encoding the cDNA sequence of SEQ ID No. 4; or a gene comprising SEQ ID No. 11, a gene comprising SEQ ID No. 12, a gene comprising SEQ ID No. 13 and a gene comprising SEQ ID No. 3.

18. The Agrobacterium tumefaciens resistant rose plant according to claim 1, wherein said one or more genes located in linkage group 7 (LG7) of the diploid rose consensus map between 56 to 58.5 Mb are a gene encoding the cDNA sequence of SEQ ID No. 1; a gene encoding the cDNA sequence of SEQ ID No. 2, a gene encoding the cDNA sequence of SEQ ID No. 3 a gene encoding the cDNA sequence of SEQ ID No. 4 and a gene encoding the cDNA sequence of SEQ ID No. 5; or a gene comprising SEQ ID No. 11, a gene comprising SEQ ID No. 12, a gene comprising SEQ ID No. 14 and a gene comprising SEQ ID No. 3.

19. A plant part, cell or reproductive tissue of an Agrobacterium tumefaciens resistant rose plant according to claim 1.

20. A method for identifying an Agrobacterium tumefaciens resistant rose plant, comprising establishing the presence, or allelic copies thereof, of one or more genes encoding a cDNA sequences selected from the group consisting of SEQ ID Nos. 1 to 5 or sequences selected from the group consisting of SEQ ID Nos. 11 to 14 in the genome of said rose plant.

Description

DESCRIPTION OF THE INVENTION

[0035] The present invention will be further detailed in the example presented below.

Example

Material & Methods

Population Development and Phenotypic Screen

[0036] A tetraploid F1 Rosa hybrida population was created by hand-pollinating a tetraploid cut rose with a rose accession. This cross resulted in 353 F1s, which were tested for Agrobacterium tumefaciens resistance. The parents were also tested. The A. tumefaciens isolate that was used originated from the Netherlands and was isolated in 1992. It was multiplied on artificial medium and used at a final inoculum density of 2.6×10.sup.9 colony forming units/ml. For each accession (F1 individuals and parents) 5 rooted cuttings were transplanted in 14 cm pots, one week upon inoculation. Inoculated plants were put in randomized block design, 4 plants for each accession. One plant per accession was inoculated with water. Plants were inoculated by making an incision in the stem with a scalpel that was dipped in the inoculum. After each incision, scalpel was dipped again. The bio-assay was carried out under long-day conditions with a temperature set at 22° C. and 20° C. for day and night respectively. Relative humidity was set at 70%. For each plant, tumor presence was scored and size was measured 3, 5, 7, 9, 11, and 14 weeks after inoculation.

Phenotypic Data Analysis

[0037] A total of 8572 observations for 1412 plants of 353 unique F1 genotypes were scored for tumor size and presence. Response variables were analysed using a mixed-model framework using the sommer package (Covarrubias-Pazaran 2016) in R. Both genotype and block were fitted as random effects, and for tumor size week of observation was included as a fixed effect (covariate). Broad sense heritability was then calculated by dividing the genotypic variance by the sum of the variance components. Best Linear Unbiased Predictors (BLUP) were obtained for each genotype and used as corrected phenotypes for downstream analyses. Corrected phenotypes are centered around 0 (which defines the population mean) and deviations are expressed in the measurement units.

Genetic Analyses

[0038] Genotyping: A panel consisting of F1 plants, parents and broad germplasm accessions were genotyped using the WagRhSNP Axiom SNP array. This chip contains 68′893 SNPs which are targeted by two probes from each direction. Initial quality control and dosage calling was performed using the R package FitPoly (non-default settings: p.threshold=0.95, call.threshold=0.65, peak.threshold=0.975). Further QC was performed using custom-made scripts in R. Non-segregating SNPs, SNPs where the most common genotype had a frequency of >81% and SNP markers with more than 10% missing data were removed resulting in a dataset containing 46′539 markers. A total of 298 F1 individuals were successfully genotyped.

[0039] SNPs were mapped to the Rose Genome assembly (Raymond et al. 2018) by blasting flanking sequences against the reference using local blast (settings: −evalue 1-outfmt 6-max_target_seqs 1-max_hsps 1). Using custom scripts in R only blast hits were retained with alignment length greater than 50 and percentage of identical matches greater than 85.

[0040] Association analyses: Marker-trait analyses were performed using a mixed-model GWAS framework using the sommer package (Covarrubias-Pazaran 2016) in R where SNPs were fitted as fixed effects and a genomic relationship matrix (GRM) was fitted as a random effect to account for population structure and residual polygenic effects (genetic effects not caused by the SNP of interest). Genomic heritability was also calculated similarly, by analyses the corrected phenotypes by fitting a GRM as a random effect.

[0041] Annotation analyses: Gene predictions in QTL regions were extracted from two files. The first was the Eugene Annotation v1.1 without repeats (RchiOBHm-V2-EGN-r1.1.without_TE.gff). The second were the gene predictions obtained using blast2Go (RchiOBHm-V2-EGN-r1.blast2go.20170310.MAPPING) (Raymond et al. 2018).

[0042] A gene was considered a resistance gene if the gene predictions contains at least one of the following terms: Arabinogalactan, AtAGP17, Celluloseynthase-like, CslB-05, Celluloseynthase-like, Cs1A-09, defense, Reticulons, BTI1(AtRTNLB1), BTI2, (AtRTNLB2), BTI3, (AtRTNLB4), Rab8, GTPase, Microtubules, kinesin, Myosin, Actin, Cyclophilin, Importin, Transportin, CAK2Ms, kinase, phosphatase, (PP2C), VIP1, Caspase, GALLS, interacting, CAK2Ms, Caspase, Histones, Histone, pCsn5-1, DNA, ligase, IVa, Nucleosome, assembly, CAF-1, Histone, H3, chaperoneGA1, Histone, deacetylases, H4, H3-11, H2A, Myb, transcription, factor, wrky, wrky17 (Gelvin 2010; Lacroix and Citovsky 2013).

[0043] A total of 60 broad germplasm accessions (including the resistant source) were whole-genome-sequenced. Reads were QC-ed and SNPs were called after mapping the reads to the Rosa chinensis reference genome (Raymond et al. 2018).

Validation Using SNPs

[0044] We mined for high-utility SNPs by 1) selecting resistance genes in the QTL region; 2) identifying SNPs in those genes that showed alleles that were uniquely or almost uniquely found in the source; 3) only retaining SNPs for which KASP assays could be designed.

KASP assays were designed and first tested on a small panel of accessions to determine if they amplified. KASP assays that amplified were then run on two F1 populations sharing the same resistant parent to identify the SNPs that were most highly associated with A. tumefaciens resistance. After fine-mapping, the SNPs were used to screen a panel of broad germplasm to determine the occurrence of resistance alleles in the germplasm.

Results

[0045] Tumor incidence (the percentage of plants with visible tumors) increased from 22.5% 3WAI (Weeks After Inoculation) to 65.7% 14WAI. Of the 484 plants that were tumor-free 14 WAI, 35 genotypes were symptomless for all 4 reps (a total of 140 plants), and a further 54 genotypes were symptom-free for 3 replicates. The remaining symptom-less plants belonged to genotypes that displayed more heterogeneity in tumor score. Mean gall size increased from 5.69 3WAI to 14.66 mm 14 WAI (Table 1), with a maximum gall size of 62.65 mm Mean gall sizes (in mm) at 14WAI for the references (e.g. resistant/susceptible and used for normalization between experiments) were as follows: RS-1075: 48.8, RS-1183: 38.6, RS-1408: 36.28, RS-1418: 0, RS-9052: 15.9.

TABLE-US-00001 TABLE 1 Progression of tumor incidence during the 14 week long experiment. WAI Tumor presence Frequency Percentage Gall size (mm) .sup.a 3 1 318 22.52  5.69 (0.19) 0 1094 77.48 5 1 677 47.95  6.72 (0.26) 0 735 52.05 7 1 790 55.95  8.13 (0.31) 0 622 44.05 9 1 894 63.31 10.02 (0.35) 0 518 36.69 11 1 925 65.51 11.01 (0.36) 0 487 34.49 14 1 928 65.72 14.66 (0.43) 0 484 34.28 .sup.a Means and standard errors for gall size. Plants not showing tumors were excluded from the calculations.

[0046] When looking at the parameter estimates from the mixed model, gall size increased significantly in time with an average increase of 0.76 mm per week (Table 2) and based on visual inspection the tumor growth rates did not differ substantially between individual plants.

TABLE-US-00002 TABLE 2 Parameter estimates of fixed effects Gall Size (mm) Weeks until tumor Parameter Estimate +/− Standard Error T value Estimate +/− Standard Error T value Intercept 0.29 (0.42) 0.66 12.49 (0.53) 23.33 WAI 0.76 (0.02) 50.07

[0047] For both traits, block effects explained very little variance (<1%,Table 3) or both gall size and weeks until tumor development. Broad sense heritability ranged from 0.67 for gall size to 0.41 for weeks until tumor development (Table 3).

TABLE-US-00003 TABLE 3 Variance components for random effects. Trait Variance component Explained variance Percentage variance explained Gall size Genotype 53.74 (4.1)  66.74.sup.a Block 0.14 (0.2) 0.18 Residual 26.64 (0.41) 33.09 Weeks.till.tumor Genotype 35.29 (3.63) 40.80.sup.a Block  0.18 (0.29) 0.21 Residual 51.03 (2.2)  59.00 .sup.aBroad sense heritability

[0048] The distributions of the corrected phenotypes were very different for the two traits. Gall size showed a bimodal distribution and the parents have contrasting extreme trait values, whereas for weeks until tumor a multimodal distribution was observed. For Gall size the mode with the smallest gall sizes contained the largest number of individuals (83% if using a cutoff of 5, 72% if using a cutoff of 0).

[0049] The distribution of gall size is compatible with a scenario where a single gene underlies resistance, that this gene is dominant and the resistant parent harbours 2 copies of the resistance allele but that penetrance of the resistance alleles is not complete. Mean gall size (14WAI) for accessions with a corrected phenotype for gall size of 5 or smaller was 2.4 mm, whereas mean gall size (14 WAI) for accessions with a corrected phenotype higher than 5 was 20.8 mm.

[0050] Genomic heritability (a proxy for the narrow sense heritability) for the corrected phenotypes for gall size and weeks until tumor development were 0.57 and 0.51 respectively. This indicates that genetics explains part of the differences between individuals, but that other unknown factors e.g. the environment, also explain a proportion of the phenotypic variance. The genomic heritability provides an upper bound for the variance that can be explained by genome-wide significant markers.

[0051] Using the mixed model SNPs on three linkage groups were significant after Bonferroni correction for multiple testing (threshold for significance 1.07*10.sup.−6): on LG1, LG5 and LG7. The most significant SNPs on LG7 explained up to 53% of the phenotypic variance whereas the most significant SNPs on the other two LGs (LG1 and LG5) explained only 16% of the phenotypic variance (using linear regressions).

[0052] Looking at the allelic dosage at significant SNPs the SNPs on LG1 and LG5 co-segregate, indicating that these should be mapped to the same LG. The fact that using multiple regression only one of the three markers on LG1 and LG5 remains significant corroborates this notion. The allelic substation effects showed that three genotype classes were observed in the F1 for the most significant SNP AX-86888149 (Table 4) and that plants scoring a 2 or a 3 had almost but not entirely equally small gall sizes and that plants scoring a 4 had substantially larger gall sizes. This scenario is incompatible with both additivity and complete dominance but is compatible with incomplete dominance.

TABLE-US-00004 TABLE 4 Mean and standard errors for gall size per genotype for AX-86888149. Number/ Number/ percentage of percentage of Mean trait Standard Number of symptomless symptomless Trait Genotype score (mm) error observations accessions.sup.a accessions.sup.b Gall 2 2.21 0.54 31 10 (32.26%)  91 (73.39%) size in 3 6.37 0.45 217 18 (8.29%) 308 (35.48%) Wk 14 4 29.34 1.23 50  0 (0%)  21 (10.5%) F1 9.79 0.64 298 Mean Corrected 2 −4.46 0.22 31 gall 3 −2.47 0.23 217 size 4 13.14 0.87 50 F1 −0.06 0.41 298 Mean .sup.aNumbers are shown only for accessions for which all reps were symptomless 14WAI .sup.bNumbers are shown for all individual plants harbouring the specified genotype

[0053] In the QTL region LG7 (between 56 and 58.5 Mb) five resistance genes were found (Table 5).

TABLE-US-00005 TABLE 5 Resistance genes involved in Agrobacterium resistance near QTL peak on LG7 Start End Gene LG position position EuGene gene prediction RchiOBHmChr7g0232831 7 57010720 57017672 product = Putative transcription factor WRKY family RchiOBHmChr7g0232901 7 57087871 57092473 product = Putative P-loop containing nucleoside triphosphate hydrolase leucine-rich repeat domain RchiOBHmChr7g0232931 7 57128076 57131496 product = Putative P-loop containing nucleoside triphosphate hydrolase leucine-rich repeat domain RchiOBHmChr7g0232961 7 57174178 57174936 product = Putative leucine-rich repeat domain L domain- containing protein RchiOBHmChr7g0232471 7 56575965 56579420 Putative protein kinase RLK-Pelle-LRR-XI-1 family