“Sexually antagonistic variation and the evolution of dimorphic sexual systems”

Posted on

Colin Olito and Tim Connallon (May 2019)

Read the Article

Sexually antagonistic variation promotes the evolution of separate sexes from hermaphroditism beyond classic predictions

Separate sexes can evolve from outcrossing hermaphrodites too!

Understanding when and why separate sexes (dioecy) evolve from hermaphroditism is a fundamental question in evolutionary biology with a long and storied history. A key advance came in the late 1970’s, when Brian and Deborah Charlesworth developed a series of theoretical models, published here in The American Naturalist, describing the evolution of separate sexes from hermaphroditism via the invasion of nuclear unisexual sterility alleles, which abolish either the male or female sex-function in hermaphrodites (Charlesworth & Charlesworth 1978). This influential theory predicts that the evolution of separates sexes is most likely when female and male sex-functions genetically trade-off with one another, and rates of self-fertilization and inbreeding depression are high among hermaphrodites. Yet, while some studies of single species support this longstanding prediction, empirical evidence for an association between dioecy and selfing across taxa remains underwhelming. In their forthcoming article, Colin Olito and Tim Connallon reconsider this issue, extending the theory with a series of two-locus models involving both a ‘sterility locus’ (where a unisexual sterility mutation may occur), and another ‘sexually antagonistic locus’ (where alleles beneficial for one sex function are deleterious for the other). Their models show that genetic linkage of unisexual sterility alleles to a sexually antagonistic locus facilitates the initial step in the evolution of separate sexes and inverts the predicted relation between self-fertilization and dioecy relative to the predictions of the classical theory by essentially allowing unisexual sterility alleles to ‘hitchhike’ with SA alleles. Their findings suggest that dioecy may evolve from hermaphroditism under much broader conditions than previously thought and suggest a new role for sexually antagonistic genetic variation in the evolutionary origins of new sex-chromosome systems.


Charlesworth, B., and D. Charlesworth. 1978. A model for the evolution of dioecy and gynodioecy. The American Naturalist 112:975–997.

In the gynodioecious shrub Daphne jezoensis (Thymelaeaceae), unisexual females produce anthers containing very small amounts of inviable pollen. In most populations of D. jezoensis, female frequencies are around 50%, and fruit set is quite low in hermaphrodite individuals, suggesting that this species is at a later stage in the gynodioecy pathway towards full dioecy. The above photos, taken during field research in Napporo Forest Park, Sapporo, Japan, illustrate hermaphrodite and unisexual female individuals (left-hand photo) and cross-sectioned flowers (right-hand photo). (Credit: Lawrence D. Harder)


Multicellular Eukaryotes use a broad spectrum of sexual reproduction strategies, ranging from simultaneous hermaphroditism to complete dioecy (separate sexes). The evolutionary pathway from hermaphroditism to dioecy involves the spread of “sterility alleles” that eliminate female or male reproductive functions, producing unisexual individuals. Classical theory predicts that evolutionary transitions to dioecy are feasible when female and male sex functions genetically trade-off with one another (allocation to sex functions is “sexually antagonistic”), and rates of self-fertilization and inbreeding depression are high within the ancestral hermaphrodite population. We show that genetic linkage between sterility alleles and loci under sexually antagonistic selection significantly alters these classical predictions. We identify three specific consequences of linkage for the evolution of dimorphic sexual systems. First, linkage broadens conditions for the invasion of unisexual sterility alleles, facilitating transitions to sexual systems that are intermediate between hermaphroditism and dioecy (andro- and gynodioecy). Second, linkage elevates the equilibrium frequencies of unisexual individuals within andro- and gynodioecious populations, which promotes subsequent transitions to full dioecy. Third, linkage dampens the role of inbreeding during transitions to andro- and gynodioecy, making these transitions feasible in outbred populations. We discuss implications of these results for the evolution of dimorphic reproductive systems and sex chromosomes.