Direct and indirect fitness effects of plant metabolites and genetic constraints limit evolution of allelopathy in an invading plant

Abstract

Upon introduction to new continents, invading species encounter novel communities of consumers, pathogens, and competitors. Both phenotypic plasticity and rapid evolution can facilitate adaptation across these heterogenous communities, facilitating further invasion. However, the rate and extent of adaptive evolution on contemporary timescales can be constrained by phenotypic plasticity and limits imposed by genetic co-variation for traits under selection. We measured phenotypic plasticity and quantified genetic co-variation for growth, competition, and fitness among 23 naturally inbred seed families of Alliaria petiolata (garlic mustard) collected across its invasive range in eastern North America. After growing a self-pollinated generation in a uniform common garden to reduce maternal effects, we reared second-generation plants in a two-year greenhouse and field experiment with naive soil from an uninvaded habitat. We measured selection gradients and lifetime fitness when reared alone, with an intraspecific competitor, and under interspecific competition with naive Acer saccharum (sugar maple) saplings. Total glucosinolate production was strongly correlated with the production of chlorophyll a (Chl a) (R2 = 0.45) such that first principal component (PC1) accounted for 84% of variation in these two traits. Furthermore, PC1 exhibited high plasticity across growing environments (p < 0.001) with limited broad-sense heritability (H2 = 2.91; p = 0.08). In contrast, investment in glucosinolate production relative to Chl a (PC2) was significantly heritable (H2 =16.91, p < 0.001) with minimal plasticity across treatments. Causal analysis revealed that plastic variation for higher Chl a + glucosinolate production (PC1) had an indirect positive effect on A. petiolata fitness via a direct, negative effect on A. saccharum performance. In contrast, heritable variation for higher glucosinolate investment (PC2) had a direct, positive effect on A. saccharum performance and an indirect negative effect on A. petiolata fitness. Applying causal inference, we find that evolution of allelopathy in A. petiolata has been constrained by (i) a lack of genetic variation, (ii) selection against glucosinolate investment under interspecific competition, and (iii) phenotypic plasticity. These factors limit adaptive evolution but maintain fitness during population growth as plants switch from interspecific to intraspecific competition during invasion.