How does pollination mutualism affect the evolution of prior self-fertilization? A model

Abstract

The mode of pollination is often neglected regarding the evolution of selfing. Yet the distribution of mating systems seems to depend on the mode of pollination, and pollinators are likely to interfere with selfing evolution, since they can cause strong selective pressures on floral traits. Most selfing species reduce their investment in reproduction, and display smaller flowers, with less nectar and scents (referred to as selfing syndrome). We model the evolution of prior selfing when it affects both the demography of plants and pollinators and the investment of plants in pollination. Including the selfing syndrome in the model allows to predict several outcomes: plants can evolve either toward complete outcrossing, complete selfing, or to a stable mixed-mating system, even when inbreeding depression is high. We predict that the evolution to high prior selfing could lead to evolutionary suicides, highlighting the importance of merging demography and evolution in models. The consequence of the selfing syndrome on plant-pollinator interactions could be a widespread mechanism driving the evolution of selfing in animal-pollinated taxa. K E Y W O R D S :Adaptive Dynamics, demography, inbreeding depression, mixed-mating, pleiotropy, selfing syndrome.Flowering plants exhibit a great diversity in mating systems, ranging from obligate outcrossing to complete selfing through mixed-mating species (simultaneous selfing and outcrossing). The recurrent transitions from outcrossing to selfing, despite negative effects of selfing on long-term diversification (Igic and Busch 2013;Wright et al. 2013), have motivated numerous empirical and theoretical studies Busch and Delph 2012). Seminal studies have investigated how the genetic implications of self-pollination should impact the evolution of selfing. On the one hand, selfers benefit from a 50% transmission advantage of their genome compared to outcrossers, known as the "automatic transmission advantage" (Fisher 1941): selfers transmit two copies of their genes through their own seeds and one copy through pollen export, whereas outcrossers transmit only one copy through their own seeds, and one copy through pollen export. On the other hand, inbreeding depression is assumed to prevent the evolution of selfing because of a reduction of fitness in selfed offspring compared to outcrossed ones (Charlesworth 2006). More recently, theoretical studies have started to consider the impact of several ecological mechanisms of pollination on selfing evolution, such as pollen limitation (Holsinger 1991;Cheptou 2004;, pleiotropic effects of selfing on viability or fertility components (e.g., pollen discounting: Lloyd 1992; Harder and Wilson 1998;Johnston et al. 2009;Jordan and Otto 2012; or simultaneous increase in the number of selfed and outcrossed ovules: Johnston et al. 2009) and indirect effect of floral display on geitonogamous self-pollination (Devaux et al. 2014a). Because they predict that mixed-mating systems can be evolutionarily stable under some conditions, such models allow a be...

The double edged sword: The demographic consequences of the evolution of self-fertilization

“…Though it has also previously been shown that interspecific interactions can cause evolutionary suicide of selfing populations (Lepers et al. ) due to feedback mechanisms between demography and genetics, our work is the first to explicitly show that even without such complex interactions initially stable populations continuously evolving from outcrossing to very high self‐fertilization rates can go to extinction.…”

Too attractive to self: How pollinators can interfere with the evolution of selfing

“…Because the 162 mutant arises from the resident population, it is initially very rare (one individual) and arises in a 163 resident population that is initially at equilibrium with its animal pollinator. Consistent with 164 previously validated phenotypic models (Cheptou, 2004;Lepers, et al, 2014), we assume that 165 when outcrossing occurs between resident and mutant genotypes, 50% of the offspring exhibit 166 the resident phenotype, while 50% exhibit the mutant phenotype. 167…”

“…However, reproductive assurance alone 49 cannot explain why outcrossing persists in spite of low inbreeding depression, underscoring that 50 we have not yet fully explored the conditions that favor or constrain the evolution of selfing. 51 Explicit consideration of pollination as a mutualism involving plant and pollinator as 52 equally interdependent actors can highlight how pollinators might constrain the evolution of 53 selfing (Devaux et al, 2014;Lepers, et al, 2014;Spigler and Kalisz, 2017). Pollinators depend 54 critically on floral resources for their own metabolic demands and for provisioning their broods.…”

“…We stress that this single-locus model does not account for the feedback 70 between plant and pollinator populations that is inherent to the mutualism. As an alternative, 71 consumer-resource models (e.g., Holland and DeAngelis, 2010) allow us to ask how pollinators 72 can influence the evolution of selfing by explicitly connecting plant and pollinator dynamics 73 (e.g., Lepers, et al, 2014). 74…”

Eco-evolutionary dynamics further weakens mutualistic interaction and coexistence under population decline

“…Theoretical studies have investigated the ecological (seminal article by Goh in 1979 [14]) [15][16][17] and evolutionary dynamics [18][19][20][21][22] of plant-pollinated communities. In particular, evolution of plant selfing with changing pollinator communities has been studied in several papers [23][24][25]. Here we focus on the consequence of declining pollinator populations on the eco-evolutionary process within a plant-pollinator community with evolving plant attractiveness.…”