The Macroevolutionary Consequences of Plant Breeding Systems

On Earth, you can pretty much find flowering plants anywhere on land, from palm trees in the tropics to dandelions in the cracks of sidewalks. There are roughly 6,000 species of mammals — but an estimated 400,000 species of flowering plants. Except, these two groups appeared around the same time in evolutionary history. Huh? Why are flowering plants so diverse? How did they get that way, and get everywhere?

One reason may be the incredible diversity of plant sexual systems. That is to say, how flowering plants have sex. Many sexually reproducing organisms (like mammals) split up sexual function into separate male and female organisms. But almost all flowering plants are hermaphroditic, meaning their flowers contain both male and female parts.

But if flowers usually contain both male and female parts why don’t they just mate with themselves? In fact, many do. These plants are self-compatible (SC), meaning they can successfully fertilize their own ovules (eggs). Yet there are also many plants — some estimates suggest almost half of all flowering plants — that cannot mate with themselves at all.

Why not? How do flowering plants — stationary organisms with no behavior — control what pollen (sperm) fertilizes their ovules? Tons of ways, as it turns out. Some do this morphologically: the shape of the flower prevents self-fertilization by separating male parts from female parts (no touching!), or some split sexual functions into separate male and female flowers. Some do this temporally, meaning that male and female parts are not “active” at the same time.

But many more use invisible genetic mechanisms that allow a plant to recognize and reject its own pollen. These plants are self-incompatible (SI). Several such mechanisms have been uncovered, and some are even shared across many distantly-related groups of plants.

Sex is important: it governs the evolutionary pathways available to organisms and the distribution of all traits. It makes sense then that differences in sexual systems have profound evolutionary consequences. What kind of consequences? How do these differences influence the distribution and diversity of flowering plants?

To answer these questions, we first need to know something about how SI and SC are distributed across groups of flowering plants — we must first determine the observed pattern. From there, we may evaluate what processes are responsible for generating such a pattern. Currently we know very little about which species are SI and which are SC in many groups. Thus, one part of my dissertation is concerned with uncovering this distribution in two of the largest flowering plant families: the bean family (Fabaceae) and the orchid family (Orchidaceae), which account for 16% of all flowering plant species.

Breeding System Distribution in Fabaceae

Self-incompatibility (SI) is thought to be widespread among flowering plants. While many mechanisms can underlie the SI phenotype, independent gains and the persistence of many such mechanisms across vast timescales suggests a strong role of natural selection and profound evolutionary consequences. But how widespread is SI really? How many genetic and molecular mechanisms underlie the SI phenotype, and what are their evolutionary relationships? Answering such questions requires first uncovering the distribution of these traits across all flowering plants.

One particular flavor of SI, T2 S-RNase-based GSI, is found in several distantly-related core eudicot groups including the rose family (Rosaceae). Because Rosaceae is a close relative of the bean family (Fabaceae), and both groups are known to contain SI species, it has long been hypothesized that this particular SI system is also operating in Fabaceae.

But... is it? The bean family contains nearly 20,000 species. How many of these species are SI? What is the distribution of breeding systems across this group? Are the characteristic molecular markers of T2 S-RNase-based GSI found in any Fabaceae species?

Arroyo (1981) published the most definitive exploration to date of breeding system distribution in Fabaceae with a dataset containing approximately 350 species. This impressive collection accounts for slightly less than 2% of the family — not nearly enough data to evaluate important questions about the evolution of SI.

To address this gap in data, we revist Arroyo’s original dataset and add new data, yielding a database with 1285 species — 6.6% of those recognized in the family — across 190 genera spanning five of the six currently recognized subfamilies. We use this data to summarize what we currently know about the distribution of SI in Fabaceae, and to evaluate the long-standing hypothesis that Fabaceae shares the same SI system as the closely-related Rosaceae.

Breeding System Distribution in Orchidaceae

One of the largest families of organisms on Earth, the orchids are known for their stunning diversity of plant-pollinator relationships. While most species are thought to be self-compatible, self-incompatibility has been uncovered in several genera — most notably Dendrobium.

Several SI mechanisms have been uncovered in eudicots, some across distantly-related groups. But what about monocots? How many SI mechanisms are present in this group and what are their relationships?

To evaluate this question, more data is needed on the distribution of SI in monocot groups. We address this lack of of data by compiling a breeding system dataset from the orchid family — the second largest family of flowering plants, containing approximately 28,000 species. Our dataset contains 1051 species across 295 genera and all five subfamilies.

Stay tuned for more!