An integrative genetic study of the bunch compactness trait in grapevine

Resumen

Grapevine (Vitis vinifera L.) bunch compactness is an important trait affecting the quality and sanitary status of table and wine grapes. In spite of its agronomic and commercial relevance, little is known about the molecular and genetic mechanisms underpinning this trait. Some reasons are the great number of factors affecting the trait (it arises from the integration of numerous bunch and berry attributes), and the lack of a reliable and globally accepted method for its objective and quantitative evaluation. One of the main aims of this work was the dissection of bunch compactness at a multicultivar level to identify the main bunch and berry attributes affecting the trait. Thus, a large number of bunches of a diverse grapevine collection of wine and table grape varieties was evaluated for many traits during three consecutive seasons (2011, 2012 and 2013). Univariate statistical analyses indicated that most of the studied traits might have an influence on bunch compactness natural variation, confirming its multifactorial nature. Further multivariate analyses showed that the number of berries per bunch and the length of the rachis ramifications have a major influence on bunch compactness, whereas berry dimensions play a secondary role. Consequently, they arise as the most appropriate target traits to unravel the genetic determinism of bunch compactness. On the other hand, a series of quantitative and objective compactness indexes were designed from the combination of different bunch and berry metrics. They were tested in a set of highly diverse bunches, and compared to other selected published indexes. Some of the new indexes proved to be more adequate for the multivarietal study of this trait than those previously published, so they are proposed as objective estimators for the viticulture sector and the scientific community. Moreover, the use of novel approaches (2D-image based technologies and 3D-scanning) was also tested for the accurate estimation of bunch compactness. Since those novel systems allow the precise determination of some bunch attributes highly related to bunch compactness that cannot be assessed by hand, they provide a new framework for the fast and automatic quantification of bunch compactness. Transcriptomic comparative analyses between loose and compact grapevine clones obtained in our laboratory generated a series of candidate genes for bunch compactness or bunch compactness-related traits. Their genotyping in the grapevine varieties previously characterized allowed the identification of a set of novel genetic variants. They were further analyzed by association mapping to test their relationship with bunch compactness and the two most determining factors influencing this trait: the number of berries per bunch and the length of the first ramification of the bunch. This approach allowed the identification of a reduced number of SNPs recursively associated with bunch compactness or bunch compactness-related traits in genes not previously related to them, like a MYB transcription factor (associated with berry number) and a gene encoding for an uclacyanin protein (associated with ramification length and bunch compactness). Consequently, these genes/polymorphisms are proposed as suitable candidates for future works aimed to verify the association results obtained in this work. Lastly, the gene VvNAC26 [the grapevine closest homologue to Arabidopsis NAP (NAC-LIKE, ACTIVATED BY APETALA3/PISTILLATA)] was selected as a candidate gene to perform an association analysis with different bunch and berry traits. Agreeing with literature data that suggests a role for this gene in flower and berry development, we found that some VvNAC26 polymorphisms (and their combination in minihaplotypes) significantly associate with berry dimensions (berry length, width, weight and volume), suggesting the role of this gene in the final size of the berry in the cultivated grapevine. Moreover, the phylogenetic analysis of the VvNAC26 haplotypes indicated that the associated polymorphisms could have been involved in the early domestication and selection processes driving to the differentiation between table and wine grape varieties.