Causes of variation in population sex ratio
NSF-DEB Evolutionary Processes (EP):
Understanding the causes of geographic variation in sex ratio of a gynodioecious plant (2009–2014) Fabulous Co-PIs: Chris Caruso, Maia Bailey Fabulous students: Stephanie Hovatter (M.Sc), Julie Proell (M. Sc.), Hannah Madson (M.Sc.), Binaya Adhikari (Ph.D.) Variation in population sex ratio provides an excellent model for investigating the mechanisms of population differentiation because it is tightly linked to reproductive fitness and has clear consequences for patterns of genetic variation within populations. We investigated the evolutionary causes and consequences of variation in. population sex ratio in the sexually dimorphic wildflower, Lobelia siphilitica (Campanulaceae). L. siphilitica is gynodioecious, a relatively rare breeding system in which hermaphrodites and females co-exist within populations. L. siphilitica exhibits substantial variation in the frequency of female plants among populations (see figure to the right). Population sex ratios are also strongly geographically structured in this species, which we are still working to explain! |
Distribution and population sex ratios for 93 populations of L. siphilitica. Grey shading shows county-level presence data based on collection records. Pie charts show the proportion of plants in each population that are female (red) or hermaphrodite (black). All populations were censused in 2009 & 2011 (Madson, M.Sc. thesis, Kent State University, 2012)
|
Photo credits (left to right): Hannah Madson; Maia Bailey; Catherine Walsh
Key outcomes of this project:
- Population sex ratios of L. siphilitica vary from 0–100% female and are strongly geographically structured (see figure above). Females are much more abundant in the southern-central part of the species range compared to peripheral and northern locales. Based on genetic-marker analysis, this pattern more likely reflects selective processes than stochastic ones (Madson et al. 2022).
- However, we did a growth chamber experiment to test whether or not females perform better under warm conditions likely to be present in southern populations, and warmer temperatures did NOT increase female fitness (Bailey, Case & Caruso, 2017).
- We have strong evidence that variation in population sex ratio reflects variation in the cost of male-fertility restoration rather than variation in the relative seed fitness of females and hermaphrodites (Case & Caruso 2010; Caruso et al. 2012; Caruso & Case 2013). Ours was the first test of this important hypothesis in natural populations!
- Populations with more females have higher mitochondrial haplotype diversity than populations with fewer females. In particular, populations with >50% female plants carry rare, recombinant mitotypes that are likely to carry novel male sterility genes, suggesting that male sterility genes either arise more frequently or are less likely to be lost from those populations (Adhikari et al. 2019)! We also found that mitochondrial and plastid genomes are not in complete linkage disequilibrium, and female prevalence tracks mitochondrial diversity rather than plastid diversity!
- Females and hermaphrodites of L. siphilitica are differentially plastic in their flower production: females appear to respond plastically to neighborhood sex ratio, opening more flowers when more hermaphrodites are nearby! This novel mechanism does NOT depend on the amount of pollen received by female plants, and may mediate frequency-dependent selection (Rivkin et al. 2015)!
- Crosses between populations of L. siphilitica that experience different climates (i.e., annual mean temperatures and annual precipitation) don't work as well, a signature of outbreeding depression (Caruso et al. 2015). This means that restoration projects that reintroduce native wildflowers (like L. siphilitica) may be more successful if they use plant genotypes from similar climatic conditions.
- Pollen is dispensed from individual L. siphilitica flowers at an exceptionally low rate — only about 9% of available pollen is removed in the first floral visit by a simulated pollinator; the amount decreased with additional simulated visits (Eisen et al. 2017)! Relative to other floral traits, the amount of pollen dispensed per visit was not highly heritable, meaning that it may not have high potential to evolve.