1. Which genetic differences contribute to reproductive isolation and how did they arise?
2. Can new species form in the face of gene flow?
3. What is the structure of the fitness landscape?
I am addressing these questions in the genus Antirrhinum (snapdragons), a group of short lived plants in Spain. In collaboration with the John Innes Centre and the University of Toulouse, we are investigating the evolutionary and ecological dynamics of hybrid zones between two subspecies that different in flower colour (yellow and magenta - pictured). Here, two major loci control flower colour, giving rise to six colour phenotypes across a narrow hybrid zone. This raises the question as to how the alleles that control the distinct phenotypes arose and spread despite strong barriers to invasion into parental populations. The clear link between genotype and phenotype in Antirrhinum and the well characterised hybrid zone provides a unique opportunity to directly examine the role of selection, epistasis and drift in the evolution and maintenance of distinct species in nature.
To address these questions we are using a combination of ecological and genetic data to construct a detailed pedigree across the hybrid zone to determine the fitness landscape. We are also using Whole Genome Sequencing to investigate the topology of divergence across the genome, particularly for areas of genome under different forms of selection.
Hybridization and the maintenance of distinct species
Many plants and animals capable of interbreeding with related taxa exhibit dramatic variation in hybrid frequency among populations. It is important to understand the basis of this variation as the evolutionary consequences (homogenization or further divergence) will, in part, depend on the frequency of hybridization. Surprisingly few studies have examined the ecological and demographic basis of this variation.
My PhD research in Australia (2002-2008) with Prof. Andrew Young (CSIRO) examined the mechanisms governing patterns of hybridization and the barriers to gene flow in the genus Eucalyptus. This work used a combination of ecological field work, population genetics, and manipulative experiments to examine the processes governing variation in hybridization between the uncommon E. aggregata, and the more abundant E. rubida and E. viminalis (Field et al 2009 Cons Gen) (pictured).
In my disertation work, I uncovered evidence that the relative abundance of species is an important parameter determining the frequency of hybrid formation, seed production and seedling performance (Field et al 2008 J Ecol). By examining the spatial genetic structure across multiple hybrid zones, I also found evidence of asymmetrical gene flow from the common towards the rare species.This likely reflects a combination of demography and differences in style lengths (E. rubida: 7 mm, E. aggregata: 4 mm) preventing pollen tubes of smaller-flowered species from fertilizing larger-flowered species (Field et al 2011 Heredity). In a following study, I used pedigree reconstruction in hybrid zones to show that large-scale landscape process, local demography and pre-mating phenological barriers can govern frequencies of hybrid formation at an individual level (Field et al 2011 Mol Ecol*) [*Featured in a perspective article by Berthold Heinz].
Mating system evolution
Why do some dioecious plant species exhibit male or female biased sex ratios?
Theory predicts that when the cost of producing males and females is equal, a 1:1 sex ratio should be maintained by negative frequency-dependent selection. However, there is evidence from surveys of plant populations of frequent departures from equality. As a PDF in the Barrett lab in collaboration with PDF Melinda Pickup, I investigated this problem with both theoretical modeling approaches and phylogenetic comparative analysis. Initial modelling results have demonstrated the importance of non-equilibrium conditions and life history as explanations of sex ratio variation (Barrett et al. 2010 Phil Trans Royal Soc). Our following synthesis of sex ratio variation in 242 angiosperm species (representing 125 genera and 62 families) indicates that gender-based differences in the cost of reproduction, sex determining mechanisms and non-equilibrium conditions each play important roles in affecting flowering sex ratios in dioecious plant species (Field et al. 2013 Evolution; Field et al. 2013 Annals Bot.).
Polyploidy is common throughout much of the plant kingdom and is also present in a range of animals. It is well recognized that both allo- and auto-polyploidy are of major evolutionary and ecologically importance in plants. As a consequence there has been increasing interest in investigating the ecological and genetic attributes of polyploids and developing our understanding of its role in plant evolution and importance in biological conservation. However, this has not translated to the availability of analytical tools for population genetic analysis for polyploids.
The development of population genetic software for polyploid organisms have been hampered by the complexities of polyploid inheritance and difficulties in interpreting molecular marker data. I am bridging this gap with development of new theoretical approaches and associated software for a range of population genetic analyses (Field et al. in review). In collaboration with Prof. Andrew Young and Dr. Linda Broadhurst at CSIRO Plant Industry (Australia), I am also testing these new approaches with genetic marker data in both hexaploid and tetraploid populations.
Autopolyploids pose a number of significant challenges for population genetic analyses due to polysomic inheritance and genotype uncertainty.
Mechanisms governing female biased sex ratios
Whereas the causes of male biased sex ratios have been well studied, the mechanisms behind female biased sex ratios remain poorly understood. For species with sex chromosomes, competition between female- and male-determining microgametophytes (pollen grains) has been proposed as one of the causes of female biased sex ratios (certation hypothesis). This is thought to be due to Y-chromosome degeneration resulting differential performance of male- and female- determining pollen grains and preferential fertilization by female-determining pollen grains under intense competition. In the Barrett lab, in collaboration with PDF Melinda Pickup, I am exploring the mechanisms governing female bias in Rumex hastatulus, an annual species native to the southern USA ranging in distribution from Texas to North Carolina. Using controlled crosses we have found significant female biased progeny arrays (sex ratio = 0.62) with the probability of producing female offspring increasing with pollination intensity (Field et al. 2012 IJPS). We are investigating this relation further by examining the importance of plant density and local sex ratio on progeny sex ratios in manipulated experimental arrays at the University of Toronto.
Evolution and maintenance of separate sexes
The evolutionary transition from hermaphroditism (combined sexes) to dioecy (co-occurrence of males and females) has occurred independently at least 100 times in the flowering plants. However, in many species, a low to moderate frequency of labile hermaphrodites can remain, with some populations consisting of males, females and hermaphrodites. This condition is typically referred to as subdioecy or trioecy. This tendency for hermaphrodites to not be entirely replaced by male phenotypes raises the question of whether this state represents a stable sexual system and what role gender plasticity plays in hindering the transition to full dioecy. In the Barrett lab, I used manipulative greenhouse experiments to investigate the maintenance of subdioecy using Sagittaria latifolia, an aquatic species common in N. America with some populations consisting of all three sex phenotypes (males, females, hermaphrodites- pictured).
A. majus. striatum (left), A. majus. pseudomajus (right) and their hybrids (middle)
Antirrhinum parentals and some of their hybrids in the hybrid zone in the Pyrenees.
Black gum - Eucalyptus aggregata (left) and Candlebark - E. rubida (right) of eastern Australia.
Purebred (A,R,V) and hybrid (AR, AV) seedling morphology
Beetles are a primary herbivore on these Eucalyptus (top) and can cause severe leaf damage (bottom)
The bird pollinated Eremophila glabra of central Australia which has diploid, tetraploid and hexaploid races.