Schematic illustrating how testosterone facilitates sex-specific gene expression. Testosterone (T), which circulates at higher levels in males (right) than in females (left), binds to the androgen receptor (AR), which translocates to the nucleus, recruits sex- and region-specific cofactors, and binds to an androgen response element (ARE), leading to altered transcription of an ARE-proximal color gene.
COLOR EVOLUTION AND HORMONE-PHENOTYPE INTERACTIONS
I am interested in using the evolution of colorful traits to study hormone-phenotype interactions. Lizards in the genus Sceloporus provide an excellent opportunity to understand how interactions between testosterone and down-stream processes evolve because, across the phylogeny, there has been multiple independent losses of testosterone-induced color development. Currently, I am testing the hypothesis that the melanin synthesis pathway can become decoupled from androgen regulation, which alters patterns of gene expression.
Using testosterone manipulation experiments and RNAseq, I am examining how patterns of gene expression differ between S. undulatus, which has vibrant blue ventral coloration (above), and S. virgatus, which lacks the ability to develop blue on their abdomens. In unmanipulated animals, adult male S. undulatus have stacked guanine platelets within iridophores that lay superficial to melanized melanophores. These iridophores are also present in adult female S. undulatus and both sexes of S. virgatus, suggesting that the lack of melanin strongly contributes to the loss of this phenotype.
I am interested in using the evolution of colorful traits to study hormone-phenotype interactions. Lizards in the genus Sceloporus provide an excellent opportunity to understand how interactions between testosterone and down-stream processes evolve because, across the phylogeny, there has been multiple independent losses of testosterone-induced color development. Currently, I am testing the hypothesis that the melanin synthesis pathway can become decoupled from androgen regulation, which alters patterns of gene expression.
Using testosterone manipulation experiments and RNAseq, I am examining how patterns of gene expression differ between S. undulatus, which has vibrant blue ventral coloration (above), and S. virgatus, which lacks the ability to develop blue on their abdomens. In unmanipulated animals, adult male S. undulatus have stacked guanine platelets within iridophores that lay superficial to melanized melanophores. These iridophores are also present in adult female S. undulatus and both sexes of S. virgatus, suggesting that the lack of melanin strongly contributes to the loss of this phenotype.
Interaction between testosterone and ventral skin tissue in S. undulatus and S. virgatus. Our working model is one where AR+ cells (shown as keratinocytes for ease of visualization, but melanophores are equally likely as the target cell) produce proopiomelanocortin (POMC) in response to testosterone, inducing melanin synthesis through α-melanocyte stimulating hormone (α-MSH) interacting with the melanocortin-1 receptor (not shown) on melanophores.
CONTRIBUTION OF THE ANDROGEN RECEPTOR
Working with Matt Milnes at Georgia College, I am quantifying the distribution and abundance of androgen receptor (AR) in the dermis of these two species to test if loss of melanin-related gene expression can be partially explained by evolution of where and how many AR+ nuclei exist among cell types.
Several relevant cell types can express AR and influence whether melanin can and will be synthesized in the dermis. For example, keratinocytes can be local sources of proopiomelanocortin, a prohormone typically produced in the pituitary gland that is cleaved into several smaller peptide hormones, including α-melanocyte stimulating hormone. Therefore, the loss of AR+ keratinocytes in S. virgatus represents one (of many) potential avenue by which testosterone-regulated melanin synthesis could be altered relative to the ancestral state, represented by the vibrant blue abdominal coloration in S. undulatus.
Working with Matt Milnes at Georgia College, I am quantifying the distribution and abundance of androgen receptor (AR) in the dermis of these two species to test if loss of melanin-related gene expression can be partially explained by evolution of where and how many AR+ nuclei exist among cell types.
Several relevant cell types can express AR and influence whether melanin can and will be synthesized in the dermis. For example, keratinocytes can be local sources of proopiomelanocortin, a prohormone typically produced in the pituitary gland that is cleaved into several smaller peptide hormones, including α-melanocyte stimulating hormone. Therefore, the loss of AR+ keratinocytes in S. virgatus represents one (of many) potential avenue by which testosterone-regulated melanin synthesis could be altered relative to the ancestral state, represented by the vibrant blue abdominal coloration in S. undulatus.
Reaction norms of sprint speed, abdominal hue, and throat hue in male prairie lizards.
COLOR AS A SIGNAL
Using the prairie lizard (Sceloporus consobrinus), which has thermally sensitive ventral coloration, Matt Gifford and I showed that individual sensitivity to thermal conditions covaried between ventral coloration and sprint speed at the individual level. Further, we showed that individuals who were "bluer" at one temperature were "bluer" at all temperatures. Together, these results suggested to us that the bluest individuals are the fastest, which may have important consequences for fitness.
Read about this work here!
Using the prairie lizard (Sceloporus consobrinus), which has thermally sensitive ventral coloration, Matt Gifford and I showed that individual sensitivity to thermal conditions covaried between ventral coloration and sprint speed at the individual level. Further, we showed that individuals who were "bluer" at one temperature were "bluer" at all temperatures. Together, these results suggested to us that the bluest individuals are the fastest, which may have important consequences for fitness.
Read about this work here!
