Thèse Architecture Génétique des Traits d'Anatomie Racinaire Impliqués dans l'Acquisition d'Eau chez la Vigne H/F - 33

Établissement : Université de Bordeaux
École doctorale : Sciences de la Vie et de la Santé
Laboratoire de recherche : Ecophysiologie et Génomique Fonctionnelle de la Vigne
Direction de la thèse : Marina DE MIGUEL VEGA ORCID 0000000312857549
Début de la thèse : 2026-10-01
Date limite de candidature : 2026-05-20T23:59:59

Un approvisionnement en eau suffisant est essentiel à la croissance et à la survie des plantes, mais il devient de plus en plus limité avec le changement climatique. Renforcer la capacité d'adaptation des cultures au déficit hydrique est donc crucial pour une agriculture durable. Les racines jouent un rôle central dans l'acquisition de l'eau, car elles sont les premiers organes à percevoir l'assèchement du sol. Pourtant, les programmes de sélection ciblent rarement les traits racinaires - en particulier chez les espèces pérennes - en raison des difficultés techniques du phénotypage.

La vigne, culture pérenne majeure à forte valeur économique et dotée d'une grande diversité d'espèces sauvages apparentées, constitue un modèle idéal pour étudier les bases génétiques de l'adaptation racinaire au déficit hydrique. L'acquisition de l'eau repose sur trois processus interdépendants : l'exploration du sol, l'absorption et le transport interne de l'eau. Les traits anatomiques racinaires conditionnent l'efficacité de ces processus en modulant l'équilibre entre coûts métaboliques et conductivité hydraulique.

Ce projet vise à déterminer si ces traits peuvent être intégrés dans les programmes d'amélioration de la vigne afin d'optimiser son acquisition d'eau en conditions de déficit hydrique. En combinant l'imagerie à haut débit pour le phénotypage anatomique, l'apprentissage profond pour l'analyse d'images, la modélisation structure-fonction pour estimer la conductivité hydraulique et la génomique, le doctorant identifiera les traits anatomiques favorisant l'acquisition d'eau, en déchiffrera le contrôle génétique et évaluera le potentiel réservoir d'allèles favorables dans des espèces sauvages de Vitis. Ces travaux révéleront les bases génétiques et la plasticité des racines face au déficit hydrique et fourniront des outils pour la sélection de cultures pérennes plus résistantes à la sécheresse.

The root system is central to whole-plant functioning, playing key roles in water and nutrient uptake, anchorage, and interaction with the soil microbiome. Consequently, understanding how roots adapt and respond to environmental constraints is crucial for improving plant resilience. Roots are the first organ to detect the onset of drought conditions and consequently they play a key role in maintaining plant hydration through both constitutive traits and through morphological, physiological and biochemical adjustments in response to the stress (i.e. plasticity). Considering that drought is a major limitation for plant growth and productivity, in both natural and agricultural ecosystems, and that it will be exacerbated in some regions as a consequence of climate change, identifying the root traits that underlie the ability of plants to cope with drought will provide essential knowledge to develop drought resilient crops.
Grapevine is one of the most important fruit species in the world. In France (the 2nd wine producer country in the world), viticulture represents 15% of the Agriculture Surface (CNIV). Since the end of the XIXth century, the grapevine species Vitis vinifera L. has been grafted in most vineyards in the world to cope with the introduction of the soil-borne aphid Phylloxera from North America. Rootstocks are accessions or hybrids of American Vitis species naturally resistant to Phylloxera. This has been the main trait, together with other agronomically relevant characteristics, such as rooting and grafting ability, targeted in traditional rootstock breeding programs. Most of the currently used rootstocks were selected more than 100 years ago and they have a narrow genetic background that may compromise their long-term adaptive and productive response in new environments. In the current context of global change, new threats may limit the sustainability and production of viticulture, such as intensified abiotic (drier and poorer soils) and biotic stresses (related with new and invasive pathogens). In particular, drought is one of the most important effects of climate change that limits yield and affects the composition of grapes. In addition, there is a social demand for environmentally-friendly agricultural management practices based on low addition of chemicals. In this sense, breeding and selecting rootstocks better adapted to the expected new environments represents a leverage for the sustainable production of vineyards without modifying the wine varieties and consequently the wine typicity. Therefore, there is a need to provide new genetic resources and next-generation breeding tools to improve the performance of rootstocks today and in the future.

The main objective of this proposal is to evaluate the potential of root anatomical traits to be incorporated in breeding programs aimed to improve grapevine water supply under drought conditions.
For this purpose, we will combine modeling, high-throughput phenotyping and genomics to:
1.Identify root anatomical traits for improved water uptake capacity in grapevine.
2.Identify genomic regions controlling the expression of key anatomical traits.
3.Assess the contribution of root anatomy to drought adaptation in a wild Vitis spp. diversity panel.

Two populations will be studied: a biparental F1 cross and a multispecies diversity panel.
Root anatomy will be analyzed using Laser Ablation Tomography (LAT) which allows rapid and quantitative imaging of internal structures. The system employs a UV laser to ablate a thin layer of the sample while a high-resolution camera images the exposed surface, producing a series of full-color images. The UV excitation indices differential autofluorescence emission in root tissues, facilitating the classification of distinct anatomical features. Deep-learning segmentation algorithms wil be used to analyze the generated images.
Root radial hydraulic conductivity will be estimated from LAT-derived traits through structure-function models.