Search

White lupin (Lupinus albus L.) is a pulse whose 49 domestication started about 3000 - 4000 years ago in the Mediterranean region [1]. It is cultivated for its seeds that contain high levels of proteins and are used both for food and feed [2]. The wild forms of the species can only be found in the Balkan region and evidence of its earliest use as a green manure and grain crop come from that same region [3]. Early Greek farmers selected larger seeds and white flowers, and presumably soft seededness (water permeable seeds) was the earliest domestication trait. Greek and Roman literature suggests that seed indehiscence (i.e. resistance to pod shattering) had not yet been incorporated by the first century A.D. [4]. Wild collections and landraces of white lupin contain high levels of quinolizidine alkaloids that accumulate in the seed, resulting in a bitter taste and possible toxicity. Lysine-derived alkaloids are characteristic of the Genistoids [5–7], a monophyletic basal clade belonging to the Fabaceae family. Traditionally these bitter compounds are removed from white lupin seeds by soaking in water, a practice that is still carried out today across the Mediterranean and Nile regions [1]. However, this is uneconomic on a broad-scale, which motivated the identification of low alkaloid mutants in Germany in the 1930s, aided by advances in chemistry [4]. Modern cultivars of white lupin incorporate low alkaloid genes, hence the term ‘sweet’ lupins. Breeding efforts have rarely been intensive or sustained over long periods. As a result, white lupin yields remain low and highly variable, in comparison to similar pulses like soybean for which important breeding efforts have been made internationally. Although white lupin cultivation represents a promising crop for Europe, in a political context aiming towards plant protein independence, the lack of well characterised genetic resources has been hampering a fast deployment of white lupin as an alternative crop to soybean imports. The recent sequencing of white lupingenome [8,9] demonstrated a resurgence of interest for this 74 “old” crop. We believe that white lupin intragenomic diversity might reflect the early traces of its slow and sporadic domestication history. Here we report a pangenome for white lupin that reveals important aspects of the species diversity, single nucleotide polymorphisms (SNPs) and gene presence– absence variations (PAVs). We construct a species pangenome consisting of ‘core’ genes that are present in all individuals and ‘variable’ (soft-core or shell) genes that are absent in some individuals [10,11]. Building on this comprehensive dataset, we were able to identify a deletion in the QTL region associated with late flowering in Ethiopian white lupins. The deleted gene is a homolog of the FT (Flowering Time) gene, suggesting that this deletion is at the origin of the need for vernalization in these accessions. Our analyses provide new perspectives on white lupin intra species diversity and domestication history.

White lupin (Lupinus albus L.) is a legume that produces seeds recognized for their high protein content and good nutritional value (lowest glycemic index of all grains, high dietary fiber content, and zero gluten or starch)1–5. White lupin can form nitrogen-fixing nodules but has lost the ability to form mycorrhizal symbiosis with fungi6. Nevertheless, its root system is well adapted to poor soils: it produces cluster roots, constituted of dozens of determinate lateral roots that improve soil exploration and phosphate remobilization7. As phosphate is a limited resource that comes from rock reserves8, the production of cluster roots is a trait of interest to improve fertilizers efficiency. Using long reads sequencing technologies, we provide a high-quality genome sequence of a modern variety of white lupin (2n=50, 451 Mb), as well as de novo assemblies of a landrace and a wild relative. We describe how domestication impacted soil exploration capacity through the early establishment of lateral and cluster roots. We identify the APETALA2 transcription factor LaPUCHI-1, homolog of the Arabidopsis morphogenesis coordinator9, as a potential regulator of this trait. Our high-quality genome and companion genomic and transcriptomic resources enable the development of modern breeding strategies to increase and stabilize yield and to develop new varieties with reduced allergenic properties (caused by conglutins10), which would favor the deployment of this promising culture.

International audience

Plants display considerable developmental plasticity in response to changing environmental conditions. The adaptations of the root system to variations of growth conditions are an excellent model to study developmental plasticity. In an effort to identify mutants altered in root perception to salt stress, we screened an activation tagging collection (Weigel et al.2000) and identified the lasso1mutant that displays a strong agravitropic root growth on standard medium. Interestingly, the lasso1mutant phenotype is partially reverted by salt and osmotic stress (sorbitol) application. The aux1mutant, harboring a mutation in the auxin influx transporter AUX1 (Bennett et al.1996), displays a strikingly similar root agravitropic phenotype. We demonstrated that (i) lasso1is not allelic to aux1, (ii) theAUX1protein localization is not altered in lasso1,and (iii) neither salt nor osmotic stress alter aux1agravitropic root growth. We thus make the assumptionthat different molecular mechanisms are responsible for the phenotypic behavior of lasso1and aux1. Auxin transport is at the core of root gravitropic response, so we analyzed the perturbation of the auxin response marker DR5::GFP in the root tips of lasso1. Our results suggest a strong accumulation of auxin in the root columella and lateral root cap and we are now focusing on the auxin efflux transporters from the PIN family. Segregation studies have demonstrated that the lasso1 phenotype is independent of the activation tagging T-DNA and therefore the sequencing oflasso1genome is in progress to search for the origin of the mutation. This study will help us to identify new molecular mechanisms involved in the perception and integration by roots of their environment (gravity, salt, water).

Master; Le séquençage ciblé par capture est une application récente des NGS qui permet de séquencer des zones précises de génomes dans un grand nombre d’individus. Bien que l’analyse bioinformatique de ces données reste très similaires à celles réalisées classiquement en NGS, les procédés mis en oeuvre lors de la procédure de capture induisent des spécificités qu’il est essentiel de prendre en compte. De fait, il devient nécessaire d’adapter les outils bioinformatiques à l’analyse de données de séquençage par capture et d’en traiter les sorties afin d’établir un diagnostic sur l’efficacité de la capture. Suite à une demande croissante des projets utilisant la capture au sein de l’UMR AGAP, j’ai définit une procédure d’analyse générique implémentée dans un outil d’intégration permettant une utilisation facile en routine par des utilisateurs non bio-informaticien. De plus, le pipeline intègre un diagnostic graphique du nombre de séquences considérées à chaque étape de l’analyse (nettoyage, alignement et appel de variants) ainsi que de la profondeur de séquençage selon différentes spécificités de la capture ciblée (« sur cible », « hors cible », position sur la cible, taille de la cible). Ce pipeline est implémenté et prêt à l’utilisation sur le cluster de calcul du CIRAD, accompagné d’une documentation permettant la diffusion de cet outil.