Genetic potential for changes in breeding systems: Predicted and observed trait changes during artificial selection for male and female allocation in a gynodioecious species

Abstract

Premise: Evolution of separate sexes from hermaphroditism often proceeds through gynodioecy, but genetic constraints on this process are poorly understood. Genetic (co-)variances and between-sex genetic correlations were used to predict evolutionary responses of multiple reproductive traits in a sexually dimorphic gynodioecious species, and predictions were compared with observed responses to artificial selection.

Methods: Schiedea (Caryophyllaceae) is an endemic Hawaiian lineage with hermaphroditic, gynodioecious, subdioecious, and dioecious species. We measured genetic parameters of Schiedea salicaria and used them to predict evolutionary responses of 18 traits in hermaphrodites and females in response to artificial selection for increased male (stamen) biomass in hermaphrodites or increased female (carpel, capsule) biomass in females. Observed responses over two generations were compared with predictions in replicate lines of treatments and controls.

Results: In only two generations, both stamen biomass in hermaphrodites and female biomass in females responded markedly to direct selection, supporting a key assumption of models for evolution of dioecy. Other biomass traits, pollen and ovule numbers, and inflorescence characters important in wind pollination evolved indirectly in response to selection on sex allocation. Responses generally followed predictions from multivariate selection models, with some responses unexpectedly large due to increased genetic correlations as selection proceeded.

Conclusions: Results illustrate the power of artificial selection and utility of multivariate selection models incorporating sex differences. They further indicate that pollen and ovule numbers and inflorescence architecture could evolve in response to selection on biomass allocation to male versus female function, producing complex changes in plant phenotype as separate sexes evolve.

Keywords: Caryophyllaceae; Schiedea; artificial selection; between-sex correlation; dioecy; genetic correlation; inflorescence; sex allocation.

Figure 1

Schiedea salicaria. (A) Photograph of inflorescence. (B) Diagram of inflorescence illustrating the inflorescence traits included in the selection study. (C) Hermaphroditic flower showing dehisced anthers of a flower in a male stage, with three stigmas that have not yet expanded and become receptive. (D) Female (pistillate) flower showing expanded, receptive stigmas and vestigial stamens.

Figure 2

Selection program using Schiedea salicaria to measure genetic variances and covariances and between sex genetic correlations for biomass and traits associated with reproduction in a sexually dimorphic species. Full‐sib families (n = 91) from partial diallel crosses were used to create three types of lines: greater male (stamen) biomass in hermaphrodites (High M), greater female (carpel+capsule) biomass in females (High F), or Control (C, no selection for biomass). Each type had two replicate lines. The High M lines were formed by selecting the top 10 full sibships (from the 91 sibships in the baseline generation) with the highest male biomass in hermaphrodites, with the exception that those sibships could not share a parental sibship (to reduce inbreeding). As a result, the 10 sibships selected for the High M lines were drawn from the top 12 sibships for male biomass. Similarly, for the High F lines, the 10 full sibships that had the highest allocation to female biomass were chosen from the top 13 sibships that did not share a parent sibship. Sibships for the Control lines were randomly chosen as long as they did not share a parental sibship. Replicate lines used the same sibships but not the same plants. Each selection line yielded 30 full sibships (180 full sibships for the six lines from Gen 1). At least 10 progeny (5 females and 5 hermaphrodites) were grown from each cross, yielding more than 1800 plants. The entire process was repeated to produce the second generation after selection (Gen 2), although no attempt was made to avoid selecting sibships with shared parents for crossing to form the second generation because of the smaller number of sibships available to choose from relative to the baseline generation.

Figure 3

Responses to selection over two generations for (A) average male biomass in hermaphrodites and (B) average female biomass in females. Means and standard errors across full‐sib family averages are shown for two control lines, two high female lines, and two high male lines.

Figure 4

Mean trait values for carpel mass, capsule mass, and sepal mass after two generations of selection. Selection lines included two control lines, two lines selected for high average female biomass in females (F1, F2), and two lines selected for high average male biomass in hermaphrodites (M1, M2). Values plotted are means and standard errors across 60 full sib by progeny sex combinations.

Figure 5

Mean trait values for seed mass and inflorescence traits after two generations of selection. Selection lines included two control lines, two lines selected for high average female biomass in females (F1, F2), and two lines selected for high average male biomass in hermaphrodites (M1, M2). Values plotted are means and standard errors across 60 full sib by progeny sex combinations.

Figure 6

Mean trait values for pollen number per flower and ovule number per flower after two generations of selection. Selection lines included two control lines, two lines selected for high average female biomass in females (F1, F2), and two lines selected for high average male biomass in hermaphrodites (M1, M2). (A, B) Values plotted are means and standard errors across 30 full‐sib combinations. (C, D) Mean and standard errors across 60 full sib by progeny sex combinations. High female lines (F1, F2) differed significantly from the control lines (C1, C2), except for lateral pollen number (panel B). High male lines (M1, M2) differed significantly from the control lines in all cases. High male lines differed significantly from the high female lines, except for lateral ovule number (panel D). Statistical details are in Table 4.

Figure 7

Comparison of responses to selection over two generations to predicted values from baseline estimates of genetic parameters. (A) Responses in hermaphrodites to selection on average male biomass in hermaphrodites. (B) Responses to selection in hermaphrodites on average female biomass in females. (C) Responses to selection in females on average male biomass in hermaphrodites. (D) Responses to selection in females on average male biomass in hermaphrodites. Means and standard errors across two lines are shown. Responses are relative to the mean for the two control lines. Both responses and predicted values are expressed in units of SD of the trait in the Baseline generation. Values used in plotting are available by trait in Appendix S1. Large blue circle: average male biomass. Large red circle: average female biomass. Small blue circle: pollen number. Small red circle: ovule number. Upward triangle: biomass traits (carpel mass, capsule mass, sepal mass, and seed mass for terminal and lateral flowers). Diamond: inflorescence architecture (total flower number, inflorescence condensation, pedicel length for terminal and lateral flowers).