Phenotyping for grapevine QTL identification. The case of resistance to Plasmopara viticola and Erysiphe necator . A review.

Grapevine species belonging to the genus Vitis display susceptibility, partial or total resistance to Plasmopara viticola Berl. & De Toni (the causal agent of downy mildew DM) and to Erysiphe necator Sch. (the causal agent of powdery mildew PM). Due to the impact of DM and PM diseases on Vitis vinifera L. cultivation, research has focused for a long time on the identification and introgression of wild grapevine resistance into cultivated varieties. Since 2000, genetic studies have provided significant progress for grapevine breeding. To date, thirty-one loci of resistance to P. viticola (called Rpv) and thirteen loci of resistance to E. necator (called Ren or Run) have been identified in American and Asian Vitis species but also in some Vitis vinifera varieties (Table 1; https://www.vivc.de). Some American Vitis species such as V. aestivalis, V. cinerea, V. rotundifolia, V. riparia, V. rupestris and a few Asian species such as V. amurensis or V. piasezkii have been studied. Despite their limited level of resistance compared to other Vitis species, it is only recently that an interest in Vitis vinifera varieties has appeared because new germplasms have been investigated.


Introduction
Grapevine species belonging to the genus Vitis display susceptibility, partial or total resistance to Plasmopara viticola Berl. & De Toni (the causal agent of downy mildew -DM) and to Erysiphe necator Sch. (the causal agent of powdery mildew -PM). Due to the impact of DM and PM diseases on Vitis vinifera L. cultivation, research has focused for a long time on the identification and introgression of wild grapevine resistance into cultivated varieties. Since 2000, genetic studies have provided significant progress for grapevine breeding. To date, thirty-one loci of resistance to P. viticola (called Rpv) and thirteen loci of resistance to E. necator (called Ren or Run) have been identified in American and Asian Vitis species but also in some Vitis vinifera varieties (Table  1; https://www.vivc.de). Some American Vitis species such as V. aestivalis, V. cinerea, V. rotundifolia, V. riparia, V. rupestris and a few Asian species such as V. amurensis or V. piasezkii have been studied. Despite their limited level of resistance compared to other Vitis species, it is only recently that an interest in Vitis vinifera varieties has appeared because new germplasms have been investigated.
Resistance loci have been located in different chromosomes and usually in genomic regions rich in NBS-LRR-like resistance genes, called R genes (Di Gaspero et al., 2007, Moroldo et al., 2008. Some loci co-localized in grapevine genome but showed different resistance haplotypes, resistance levels and/or specificities toward pathogen isolates. P. viticola and E. necator are both biotrophic pathogens that can infect all grapevine green tissue: leaves, shoots, inflorescences and young berries. The resistance loci identified, induce various levels of resistance. The resistance conferred by most of the loci is partial. The observation of the symptoms, shows that pathogen sporulation is present but in a lower amount than for a susceptible variety. Very few loci, such as those identified in V. rotundifolia or V. piasezkii, induce a total resistance to P. viticola or E. necator. No sporulation is observed but hypersensitive response (HR) is frequently produced.

QTL characterization
The characterization of quantitative-trait-loci (QTL) involves the association between individual allelic variants and phenotypes. Nowadays, the availability of the grapevine reference genome and the developments in the genetic field (e.g. genotyping by sequencing technologies) allow the discovery of thousands of molecular markers, the unification of grapevine linkage groups and the production of precise and dense genetic maps. For these reasons, the *Corresponding author: tyrone.possamai@crea.gov.it genotyping in QTL studies no longer represents an obstacle to loci discovery and location.
Phenotyping is currently the main bottleneck in resistance QTL mapping. This is due to the several biological components involved and their direct impact on the resistance expression. Moreover, the technical difficulties in managing and standardizing the evaluations limit also the process automation. Finally, the consequential costs and the time associated with the practice contribute to this bottleneck. Phenotyping for grapevine resistance may take place in the vineyard, the greenhouse, or the laboratory.

QTL
Phenotyping in the vineyard needs several years of observation, depending on favorable climate conditions for disease development. The number of observations per year varies between studies. Leaves and bunches are generally assessed for the presence of disease. Sometimes, the symptoms are also observed on rachis and stems.   The variability in the resistance phenotyping methods used reflects the continuous search by grapevine breeders, geneticists and pathologists for reliable data able to predict the performance of the different genotypes and loci.
This review is based on the phenotyping strategies implemented to map the Rpv and Run/Ren loci in the last twenty years. The detailed analysis of the methods used, shows that whatever the environment considered: the vineyard, the greenhouse or the laboratory, some elements can be considered as sources of variation. The plant material (experimental design, type and age), the inoculum (natural or controlled) and the assessment method (date, frequency, type of scoring) can influence the reliability of the resistance response.

Conclusions
This analysis reveals the strengths and weaknesses of phenotyping methods used for resistance loci identification and mapping. This analysis proposes also to standardize experimental protocols by providing some defined informations. Biological results emerging from grapevine resistance mapping studies will be discussed in the light of the above phenotyping considerations.