Drivers and consequences of in- and outbreeding in plants

Abstract

Outbreeding and inbreeding have important genetic consequences for plants in general, and more specifically during plant invasions. Some studies have found that admixture (between-population outbreeding) could promote the invasiveness of alien plants. However, most of these studies were restricted to the first generation (F1), and few have considered further generations.In chapter 2 of my thesis, I focused on the admixture effect on invasive plants in the second generation (F2). However, plant breeding system is not fixed, but is a highly labile trait. The switch of breeding system is the first step towards mating system evolution. In other words, the evolution from outcrossing to selfing on the population level results from the loss of self-incompatibility on the individual level. In chapter 3 and 4, I focused on the effect of inbreeding depression on mating system evolution, and the genetic mechanics of the loss of self-incompatibility in Arabidopsis lyrata. In chapter 2, I tested whether admixture-benefits can be maintained in the F2 generation of the invasive species Mimulus guttatus. We follow up on a previous study, in which we made crosses between plants of M. guttatus from native- (western North America) and invaded-range populations (New Zealand and Scotland), and showed that admixture increases F1 performance. Here, we performed further crosses to create non-admixed progeny, F1 progeny resulting from within- and between-range admixture, and subsequent F2 progeny both through outcrossing and through self-fertilization. Our main findings are that non-admixed progeny of M. guttatus were outperformed by admixed progeny (averaged across F1 and F2), particularly by progeny from between-range admixture. However, the benefit of admixture was stronger in F1 than in F2 progeny, especially when the F2 was produced by self-fertilization. Our findings indicate that increased performance of admixed F1 progeny is partly maintained in the F2 progeny. Admixture might thus boost the performance of an invasive plant across multiple generations.In chapter 3, I tested whether sibling competition (between self- and cross-progeny) can magnify inbreeding depression, and whether inbreeding depression could be different between below- and aboveground biomass. We set a greenhouse experiment to assess the performance and estimate inbreeding depression of plants from outcrossing and selfing populations in North American A. lyrata. We found that sibling competition did not significantly change the magnitude of inbreeding depression, which never exceeded δ=0.34 in biomass traits. Combined with previous findings that drought stress and inducing defense also did not magnify inbreeding depression, this suggests that the relatively low estimates of inbreeding depression for biomass are indeed realistic estimates of the true inbreeding depression in North American A. lyrata. Moreover, there is similar pattern for below- and aboveground biomass in inbreeding depression, indicating aboveground could be used as a sufficient proxy for belowground for estimation of inbreeding depression.In chapter 4, to explore the genetic mechanism of the loss of self-incompatibility, we made intra- and inter-population crosses using self-incompatible (SI) plants from six predominantly outcrossing populations and self-compatible (SC) plants from six predominantly selfing populations of North-American A. lyrata. We found that most progeny from crosses between SC parents from different selfing populations are also self- compatibility. The lack of restoration of self-incompatibility shows that the loss of self-incompatibility in the different selfing populations in A. lyrata cannot be explained by independent recessive loss-of-function mutations. Progeny from crosses between parents with different breeding systems resulted in approximately even frequencies of SC and SI plants. This result strongly rejects the hypothesis that the loss of self-incompatibility is determined by only one genetic factor, and indicates that it must be co–determined by two or more unlinked genetic factors. We propose the hypothesis that the loss of self-incompatibility in A. lyrata resulted from the interaction between an S-locus modifier and specific S-haplotypes. More specifically, our data fit a model in which a recessive allele at a modifier-locus affects the dominant haplotype S19 and the recessive haplotype S1.