Search

International audience

International audience; The intense and unidirectional selection pressure caused by the massive deployment of resistant plants can result in rapid adaptation of pathogen populations1. Such rapid evolution was documented for the poplar rust fungus. In 1994, poplar rust populations suddenly overcame the RMlp7 qualitative resistance carried by several poplar cultivars planted widely in Western Europe2,3. This recent event of adaptation from standing genetic variation caused a selective sweep on the rust genome4. We study a 25-year temporal sampling of poplar rust populations to decipher the demographic and evolutionary history of this pathogen while overcoming its host genetic resistance. Using resurrection ecology, we examine phenotypic and genetic variations throughout this adaptive event. Our analyses reveal that a unique and homogeneous genetic group overcame RMlp7 resistance and replaced the ancestral genetic group within five years. We then use forward simulations to 1) understand the interplay between demography and genetic evolution underpinning the rapid evolution of poplar rust populations, and 2) disentangle the polymorphism signatures of selection from that of stochastic processes due to demographic changes. We show high stochasticity in evolutionary trajectories with the notable effect of evolutionary rescue scenarios on polymorphism signatures. Finally, we integrate our simulator in an Approximate Bayesian approach (ABC) to infer the demographic and selection parameters from temporal genetic data. Our statistical framework coupling modelling with temporal data is powerful to understand recent events of rapid adaptation.1 Saubin et al. DOI:10.24072/pcjournal.102 Louet et al. DOI:10.1111/mec.162943 Persoons et al. DOI:10.1111/mec.13980 4 Persoons et al. DOI:10.1093/gbe/evab279

International audience

The demography and genetic structure of a population are closely linked. The study of this interplay is crucial, especially for organisms with frequent demographic fluctuations such as pathogen species responsible for emerging diseases. Classical population genetics models have been used to explore this link for simplified demographic processes. In this thesis, I further investigate the interplay between demography and genetic evolution in pathogen species that display complex life cycles. For this purpose, I focus on two ecological systems that strongly deviate from demographic equilibrium, each of which allows for realistic modelling assumptions. The first ecological system focuses on a major selection event with the temporal dimension being of prime importance. The second system reports recurrent colonisation events in which the spatial aspect is decisive.In the first part, I study an event of resistance overcoming by a pathogen population. Firstly, a modelling approach is used to establish the conditions for observing resistance overcoming and to identify the determinants of resistance durability. The results highlight the antagonistic effect of the proportion of resistant hosts deployed in the agricultural landscape, which decreases the probability of overcoming but increases the speed of overcoming when it occurs. Secondly, this model is implemented to account for genetic evolution at neutral loci. The results identify three demographic scenarios associated with distinct genetic signatures during resistance overcoming: 1) small variations in population sizes and small changes in genetic structures, 2) a strong founder event on the resistant host that in turn creates a genetic structure on the susceptible host, and 3) an evolutionary rescue event that results in a strong founder event on the resistant host, preceded by a bottleneck on the susceptible host. Finally, this theoretical framework of demogenetic analysis is applied to empirical data to infer the parameters underlying the overcoming of resistant RMlp7 poplars by the pathogen Melampsora larici-populina. Two parameters are particularly well estimated and the inferred values are in agreement with our biological knowledge: a high proportion of resistant hosts in the landscape (more than 80%) and an initial frequency of virulent alleles in the pathogen population between 5 and 10%.In the second part, I study colonisation and its genetic consequences. These analyses focus on the recurrent invasion of the Durance River valley by Melampsora larici-populina. Firstly, a mechanistic-statistical model is coupled to epidemiological data to infer the parameters underlying the pathogen's expansion dynamics. This approach shows that colonisation results from frequent long-distance dispersal events, with an average dispersal distance of more than two kilometres. Secondly, the characterisation of several annual colonisation events highlights a similar genetic structure which amplitude however varies greatly between years. Two extremes are identified: in 2011, strong conservation of the initial genetic diversity along the colonised domain; in 2004, rapid erosion of genetic diversity. The inter-annual variations in these structures can be explained by variations in the frequency of long-distance dispersal events.This work highlights the importance of contextualised models that take into account demogenetic variations for a better understanding of biological systems. The information obtained is then crucial for developing relevant control strategies against pathogen populations responsible for emerging diseases., La démographie et la structure génétique d'une population sont étroitement liées. L’étude de ce lien est cruciale, ce d’autant plus pour les organismes présentant des fluctuations démographiques fréquentes tels que les agents pathogènes responsables de maladies émergentes. Les modèles classiques de génétique des populations ont permis d’explorer ce lien pour des processus démographiques simplifiés. Dans cette thèse, j'examine de manière plus approfondie l'interaction entre la démographie et l'évolution génétique de populations d'agents pathogènes aux cycles de vie complexes. Pour ce faire, deux situations écologiques hors équilibre démographique sont étudiées, chacune permettant de poser des hypothèses de modélisation réalistes. L’une présente un évènement majeur de sélection où l’aspect temporel est primordial. L’autre présente des évènements récurrents de colonisation où l’aspect spatial est déterminant.Dans une première partie, je m'intéresse à l'étude d'un contournement de résistance par une population d'agents pathogènes. En premier lieu, une approche par modélisation permet d’établir les conditions d’observation d’un contournement et d'identifier les déterminants de la durabilité d’une résistance. Les résultats mettent en évidence l'effet antagoniste de la proportion d'hôtes résistants déployés dans le paysage agricole, qui diminue la probabilité de contournement mais augmente la vitesse du contournement lorsqu'il a lieu. Ce modèle est ensuite complété par la prise en compte de l'évolution génétique aux loci neutres. Les résultats permettent d'identifier trois scénarios démographiques associés à des signatures génétiques distinctes lors du contournement : 1) de faibles variations de tailles de populations et peu d'évolution de la structure génétique, 2) un événement fondateur fort sur l’hôte résistant qui crée en retour une structure sur l’hôte sensible et 3) un événement de sauvetage évolutif qui se traduit par un fort événement de fondation sur l’hôte résistant, précédé par un goulot d'étranglement sur l’hôte sensible. Enfin, ce cadre théorique d'analyse démogénétique est appliqué à des données empiriques pour inférer les paramètres régissant le contournement des peupliers résistants RMlp7 par l'agent pathogène Melampsora larici-populina. Deux paramètres sont particulièrement bien estimés et en accord avec nos connaissances biologiques : une forte proportion d'hôtes résistants dans le paysage (plus de 80 %) et une fréquence initiale d'allèles virulents dans la population d'agents pathogènes entre 5 et 10 %.Dans une seconde partie, je m'intéresse à l'étude de la colonisation et ses conséquences génétiques. Ces analyses portent sur un système écologique particulier, l’invasion récurrente de la vallée de la Durance par Melampsora larici-populina. Premièrement, un modèle mécanico-statistique est couplé aux données de suivi épidémiologique afin d'inférer les paramètres régissant la dynamique d'expansion de l’agent pathogène. Cette approche montre que la colonisation résulte de fréquents évènements de dispersion à longue distance, pour une distance moyenne de dispersion de plus de 2 km. Deuxièmement, la caractérisation de plusieurs évènements annuels de colonisation montre une structure génétique similaire mais qui varie fortement en amplitude d'une année sur l'autre. Deux extrêmes sont identifiés : en 2011, une forte conservation de la diversité génétique initiale le long du domaine colonisé ; en 2004, une érosion rapide de la diversité génétique. Les variations inter-annuelles de ces structures peuvent être expliquées par des variations dans la fréquence des évènements de dispersion à longue distance.Ce travail permet de discuter l’importance de modèles contextualisés et prenant en compte les variations démogénétiques dans la compréhension des systèmes biologiques. Les informations obtenues sont déterminantes pour le renforcement des stratégies de lutte contre les agents pathogènes responsables de maladies émergentes.

Cropping genetically resistant plants allows to control pathogen populations while substantially reducing chemical inputs. However, resistances are often quickly defeated by pathogens. In this context, how can sustainable crop protection be achieved? This question has shaped the debate about the durability of genetic resistances in agriculture for decades, and, despite active research efforts, has not been satisfactorily answered yet. Here we demonstrate from a bibliography analysis that the research field of resistance durability evolved into two non-overlapping directions: (i) the subfield of 'epidemiology and evolution', which aims to forecast and explain pathogen population dynamics; (ii) the subfield of 'molecular interactions', which studies the molecular processes involved in the overcoming of resistance and in the dialogue between plants and pathogens. After reviewing briefly these two subfields and the gap between the corresponding research communities, we propose strategies to merge these approaches into one by using the concepts of population genetics. Ultimately, such new eco-evolutionary studies could be used to determine the best strategy for the deployment of genetically resistant cultivars by integrating, from gene to landscape, all relevant and contextual biological knowledge into sound theoretical models.

International audience

International audience

International audience; Melampsora larici-populina is a partially asexual basidiomycete fungus responsible for poplar rust, with its sexual stage on larch (Larix sp.) and its asexual stage on poplar (Populus sp.). Resistant 7 poplar cultivars were widely planted in northern France until a major resistance breakdown which occurred in 1994. This selective event lead to a rapid and complete shift in the pathogen population. Two independent mutations are responsible for the virulent recessive allele: a SNP and a deletion. Here we study the dynamics of virulence allele fixation both theoretically and empirically, by modeling virulence fixation in different evolutionary scenarios, and by tracing back the polymorphism evolution at the candidate locus before, during and after the resistance breakdown. The modeling part highlighted the interplay between the life cycle (host alternation or not) and the extent of resistance deployment on the probability and dynamics of fixation. Fluorescence ratio methods allowed us to successfully characterize genotypes of reference strains. This method will be applied to nearly 300 isolates collected from 1988 to 2016. By linking our model results and the observed polymorphism evolution in the studied populations, we will investigate whether the combined presence of the deletion and the mutation conferring virulence may have led to such a rapid and complete replacement of the pathogen population.