Sunflower Resistance to Broomrape (Orobanche cumana) Is Controlled by Specific QTLs for Different Parasitism Stages

Abstract

Orobanche cumana (sunflower broomrape) is an obligatory and non-photosynthetic root parasitic plant that specifically infects the sunflower. It is located in Europe and in Asia, where it can cause yield losses of over 80%. More aggressive races have evolved, mainly around the Black Sea, and broomrape can rapidly spread to new areas. Breeding for resistance seems to be the most efficient and sustainable approach to control broomrape infestation. In our study, we used a population of 101 recombinant inbred lines (RILs), derived from a cross between the two lines HA89 and LR1 (a line derived from an interspecific cross with Helianthus debilis). Rhizotrons, pots and field experiments were used to characterize all RILs for their resistance to O. cumana race F parasitism at three post vascular connection life stages: (i) early attachment of the parasite to the sunflower roots, (ii) young tubercle and (iii) shoot emergence. In addition, RIL resistance to race G at young tubercle development stage was evaluated in pots. The entire population was genotyped, and QTLs were mapped. Different QTLs were identified for each race (F from Spain and G from Turkey) and for the three stages of broomrape development. The results indicate that there are several quantitative resistance mechanisms controlling the infection by O. cumana that can be used in sunflower breeding.

Keywords: Orobanche cumana; QTL mapping; broomrape; candidate genes; parasitic weeds; plant–plant interaction; resistance; sunflower.

FIGURE 1

Life cycle of O. cumana. Stage 1 corresponds to broomrape seed germination induced by sunflower root exudates. During stage 2, O. cumana radicle attaches, invades and connects the vascular system of the sunflower root by means of haustorium. Once O. cumana is attached and the nutrient diversion towards the parasite is established, O. cumana develops a tubercle (stage 3). During the stage 4, the stem emerges through the soil surface with the subsequent onset of flowering. The red arrows show putative resistance mechanism to O. cumana at stage 2 (incompatible attachment) and at stage 3 (tubercle necrosis).

FIGURE 2

Description of the resistance mechanism inhibiting O. cumana development at stage 2 in the RIL population HA89xLR1. The photos show (A) compatible attachment and (B) incompatible attachment between O. cumana and sunflower roots. The arrows indicate the attachment area between the radicle of O. cumana race F and the sunflower roots. (C) Distribution of the rate of IA counted 2 weeks after infection by O. cumana race F. The data are calculated from at least 3 replicates for each sunflower line. Replicates with less than 5 attachments (IA + CA) were discarded. The two parental lines HA89 and LR1 showed 3.5 and 46.5% of IA, respectively.

FIGURE 3

Evaluation of the resistance level to O. cumana race F in the RILs population HA89xLR1 at stage 3. Photos illustrate the different developmental stages observed for either (A) necrotic tubercle or (B) healthy tubercle after 5 weeks post infection. (C) Distribution of the number of healthy tubercle in the RILs population HA89xLR1 5 weeks after infection. The two parental lines HA89 and LR1 show 3.4 and 2.4 healthy tubercle per plant, respectively (D) Distribution of the rate of necrotic tubercle in the RILs population HA89xLR1 5 weeks after infection. No necrotic tubercles were observed for the two parental lines. Value represents the number of RILs for each range of the number of healthy tubercle or rate of necrotic tubercle per sunflower plant. Five independent experiments were performed with one single plant per RIL in each experiment.

FIGURE 4

Characterization of the resistance level in field at the stage 4 on the RILs population HA89xLR1. Distribution of the number of O. cumana emerged shoots per sunflower plant in 2014 (A) and in 2015 (B). The data represent the mean of three replicates of 15 sunflower plants in field (Cordoba, Spain). (C) Relationship between the number of O. cumana emerged shoots in 2014 and 2015.

FIGURE 5

Relationship between the different sunflower resistance traits to O. cumana race F. PCA was performed with all phenotypic data from the 101 RILs. The two first axis of the PCA explain 68.5% of the variability.

FIGURE 6

Phenotypic diversity within the complete RIL population for O. cumana race F. Three groups were found according to the first dimension of the PCA. They correspond to resistant lines (A), susceptible lines (C) and partially resistant lines (B). The four control lines HA89, LR1, P-96 and 2603 have been added as additional samples to be shown on the figure but they have not been taken into account in the PCA analysis.

FIGURE 7

Quantitative trait loci (QTL) mapping of the different quantitative resistance traits to O. cumana race F and G. (A) The genetic map used to detect QTLs covered the 17LGs (951 genetic bins and 1795.8 cM lengths). QTLs are indicated at the left of each Linkage Group. (B) Summary of the different QTL found for the resistance to O. cumana race F and G. QTL name nomenclature is: trait-broomrape race-linkage group-QTL number on the linkage group. Additive effect, (-) be equivalent to a decrease of the value due to LR1 alleles. The traits used for QTL mapping were: IA, ratio of incompatible attachments (IA); Emergence, the number of emerged shoot broomrape per sunflower plant under field conditions in 2014 and 2015; Healthy Tubercle, the number of healthy tubercles; Necrotic Tubercle, ratio of necrotic tubercles; ACP-Dim1, the coordinates of the first axis from the PCA analysis. QTL detection was performed with MCQTL (Jourjon et al., 2005) with a threshold corresponding to a Type I error rate of 1%.