Complex genetic determinism of male-fertility restoration in the gynodioecious snail Physa acuta

Elpida Skarlou 1,2 & Fanny Laugier 1,2, Kévin Béthune 1, Timothée Chenin 1, Jean-Marc Donnay 1, Céline Froissard 1, Patrice David 1 1 CEFE, CNRS, Univ Montpellier, EPHE, IRD, Montpellier, France 2 co-first authors Abstract Male fertility in plants is often controlled by the interaction between mitochondrial and nuclear genes. Some mitotypes confer cytoplasmic male sterility (CMS), making the individual male-sterile, unless the nuclear background contains alleles called restorers, that suppress the effects of CMS and restore the hermaphroditic phenotype. Restorers in cultivated crops are often alleles with strong and dominant effect, but in wild plants, data often suggest more complex systems. Here, we characterized the inheritance and specificity of restoration in a new CMS model, the freshwater snail Physa acuta. We explored two different populations (i) a naive population i.e., without contact with CMS in the past 80 generations, (ii) a non-naive population, where CMS is present and largely restored. Although we found male fertility of individuals with CMS mitogenomes to be heritable in both contexts, this heritability was of a different nature depending on population history. In naïve populations not coevolved with CMS the background variation may include alleles that happen to act as weak quantitative modifiers of the penetrance of CMS, while in populations coevolved with CMS, selection may have favored, when such variants were available, the emergence of strong alleles with a dominant effect. FL and PD designed the experiment. ES, FL, KB, TC, JMD, CF, PD conducted the experiments. ES, PD wrote the code. ES, FL and PD wrote the manuscript. Preprint: https://doi.org/10.1101/2024.08.08.607164 Files: Our objective was to characterize, within both the naive and the non-naive backgrounds, genetic variation for the potential to restore male fertility in CMS mitotypes. We did this in two ways (i) first, we tested whether we could select for high versus low restoration potential within the naive population (HFRLFR dataset) (ii) second, we produced inbred lines from both populations and characterized genetic variance in restoration potential among lines, specially aiming to identify lines fixed for either restorer or maintainer (i.e., non-restorer) alleles (hereafter, “inbred lines” dataset). Inbredlines dataset: inbredlines.csv, inbredlines_mothers.csv HFR and LFR dataset: "G0" file contains all data of babies produced by "G0" mothers of the experimental individuals (G1 file) analysis_inbredlines&HFRLFR_final2.Rmd: all analyses of both inbredlines and HFR&LFR dataset + figures