Periodic Reporting for period 2 - PHENOVIGOUR (Elucidating the phenotypic determinants of hybrid vigour)
Okres sprawozdawczy: 2022-10-01 do 2024-03-31
How organismal traits are inherited in offspring is a long-standing, open question. Hybrids between unrelated individuals often exhibit higher growth, disease resistance and fertility than their parents. This phenomenon, called hybrid vigour or ‘heterosis’, is widely observed and agronomically exploited in both plants and animals. However, its phenotypic mechanisms continue to puzzle biologists. A strong bottleneck to our understanding of heterosis, and trait inheritance more broadly, is that most studies have only focused on its genetic mechanisms. The PHENOVIGOUR project proposes a novel approach to explain the phenotypic mechanisms of trait inheritance. The main hypothesis is that trait deviation in a hybrid is the geometric result of the nonlinear effects of traits at lower levels of integration. To test this hypothesis, the research plan will determine how nonadditive inheritance varies between traits, test whether heterosis is explained by modelling trait relationships and build a predictive model of hybrid performance.
The innovative nature of PHENOVIGOUR is to examine the causes of trait inheritance taking a holistic approach based on phenotypic integration. By comparing multiple traits in contrasting species, we will determine whether the nonadditive inheritance of phenotypic traits increases with the degree of trait integration. Moreover, we will determine whether hybrids are constrained by the same trait relationships than parental lines. To address these questions, we will investigate inter- and intra-specific trait variation with the help of (1) large plant databases, and (2) through experimental approaches performed on three model species: Arabidopsis thaliana, sorghum, and maize. We will focus on allometric scaling relationships because they rely on mechanistic assumptions and they offer testable mathematical equations of trait covariation.
The innovative nature of PHENOVIGOUR is to examine the causes of trait inheritance taking a holistic approach based on phenotypic integration. By comparing multiple traits in contrasting species, we will determine whether the nonadditive inheritance of phenotypic traits increases with the degree of trait integration. Moreover, we will determine whether hybrids are constrained by the same trait relationships than parental lines. To address these questions, we will investigate inter- and intra-specific trait variation with the help of (1) large plant databases, and (2) through experimental approaches performed on three model species: Arabidopsis thaliana, sorghum, and maize. We will focus on allometric scaling relationships because they rely on mechanistic assumptions and they offer testable mathematical equations of trait covariation.
During the first half of the project, we first worked on the role that scaling relationships play in the integration of phenotypic traits across organizational scales. In an opinion paper (Vasseur et al., 2022), we advocate that modelling scaling relationships from genome to whole-plant traits is the key for unravelling phenotypic integration in both plants and animals. Moreover, we explored the consequences of allometric scaling in crop plants, and highlighted how modelling allometry may help crop breeding and predictive modelling (Westgeest et al., submitted). In addition, we reanalyzed previously published datasets to evaluate the magnitude of genetic variability and plasticity on allometric relationships. Interestingly, we found a much larger effect of genetic variability than plasticity (Vasseur et al., 2023), suggesting that local adaptation favors the selection of specific allometries that weakly respond to environmental variations. During this period, we also launched an ambitious experiment on sorghum and maize with more than 14,000 plants individually phenotyped for traits at the molecular, foliar, and whole-plant levels. Traits are currently measured and first analyses of this massive dataset will be performed in 2024.
We also examined the effect of plant-plant interactions, as it occurs in the field, on plant response to herbivores. Interestingly, we found that some genotypes, when interacting, have a repellent effect on herbivores, while other genotypes exhibit the opposite trend: they tend to attract herbivores when interacting with each other (Estaragues et al., 2023). This suggests that complex interactions can occur within plant communities and cultivated field, with unpredictable effect on pests and herbivores. Finally, as part of the databasing task of WP1, we gathered trait measurements on a large collection of A. thaliana genotypes that we made public through a data paper (Przybylska et al., 2023). We are currently developing a similar trait database for plant crops (Vaillant et al., in prep.).
We also examined the effect of plant-plant interactions, as it occurs in the field, on plant response to herbivores. Interestingly, we found that some genotypes, when interacting, have a repellent effect on herbivores, while other genotypes exhibit the opposite trend: they tend to attract herbivores when interacting with each other (Estaragues et al., 2023). This suggests that complex interactions can occur within plant communities and cultivated field, with unpredictable effect on pests and herbivores. Finally, as part of the databasing task of WP1, we gathered trait measurements on a large collection of A. thaliana genotypes that we made public through a data paper (Przybylska et al., 2023). We are currently developing a similar trait database for plant crops (Vaillant et al., in prep.).
The analysis of scaling relationships at different organizational scales initiated during the first months of the project revealed that the approach and the core hypothesis of the project (i.e. that trait inheritance can be predicted from trait relationships) could have valuable outcomes if applied at the cellular scale, to investigate the scaling relationships between genome size, RNA content and cell size. Indeed, such approach could allow building predictive models from the lowest level of organization (genome) to the highest (whole-plant performance). To explore this promising avenue, we started to collect molecular traits (ribosomal DNA amount, RNA content, cell size) on the maize and sorghum plants grown and phenotyped for leaf and whole-plant traits. In the coming month, we expect that the analysis of trait relationships from genome to whole-plant will provide key insights into the mechanisms of phenotypic integration and trait inheritance, potentially giving rise to new and powerful predictive approaches. This should led to major publication achievements before the end of the project.