The evolutionary origin of paired fins, the forerunners of our own arms and legs, was a major developmental innovation for vertebrates. The emergence of these new appendages marked the beginning of a new period of anatomical diversification, during which the basic skeletal components were sculpted into an impressive array of structures, ranging from skate fins and frog legs to bird wings and whale flippers.
One of the biggest surprises of modern developmental biology was the discovery that the evolution of novel forms did not require the evolution of new genes, but often involved recycling of the same ancient genetic tools. On occasion, the number of tools was increased, either by gene or entire genome duplications, but modulating the regulation of existing genes was evolution's preferred route to novel form. We are interested in the cellular and molecular mechanisms that underpin the evolution of two kinds of appendages, limbs and genitalia, and in how the evolutionary history of a developmental system relates to its susceptibility to congenital malformations.
Early Evolution of Fins and Limbs
One focus of my lab is the molecular basis for the origin of fins and for the modification of fins into limbs. Recent discoveries of early vertebrate fossils revealed that median fins (such as dorsal, caudal, and anal fins) evolved before paired fins. These findings led us to hypothesize that the genetic program for fin (and later, limb) development originated in an unexpected position, the dorsal and ventral midline, rather than in the body wall. To test this hypothesis, we compared development of two key groups of basal vertebrates, lampreys and sharks. Lampreys are jawless fish that diverged from our lineage after median fins evolved, but before the origin of paired fins. Therefore, lampreys can provide insights into primitive mechanisms of body wall patterning. Sharks are the most primitive living vertebrates with paired fins, and because some shark species retain primitive fin anatomy, they provide a window into how early fins may have formed.
We cloned and characterized shark orthologs of multiple genes involved in limb development and found that median fin development is regulated by the same gene networks that operate in paired fins and limbs. To determine whether these mechanisms evolved before the origin of paired fins, we compared the expression of orthologous genes in lampreys and sharks. This analysis revealed that lamprey and shark median fins have the same embryonic origin and utilize the same genetic program as paired fins and limbs. These results provide evidence that the mechanisms of fin and limb development were assembled in the median fins of early vertebrates before the origin of paired appendages. Therefore, paired fins originated by redeployment of an ancient genetic circuit to a new position. This indicates that the molecular instructions for appendage development can elicit a similar response in cells with different embryonic origins.
Development of External Genitalia
The genetic machinery for appendage development was recycled yet again during the evolution of external genitalia. Within vertebrates, true external genital organs are found only in tetrapods (vertebrates with fingers and toes at the ends of their limbs). The origin of external genitalia allowed vertebrates to transition from external fertilization to internal fertilization, an important step toward an obligate terrestrial lifestyle. This new type of appendage underwent another modification near the origin of mammals, when the open groove used for delivery of sperm was remodeled into a closed urethral tube. It is formation of this urethral tube that is affected in a condition called hypospadias, one of the most common birth defects in humans. Hypospadias affects approximately 1 in 250 live births and is characterized by a failure of urethral tube closure. Affected children have ectopic or oversized urethral openings and, in the most severe cases, males have ambiguous or feminized genitalia.
Unlike many organ systems, in which numerous candidate genes were known to be expressed in interesting patterns, external genital development was rather poorly understood when we began this project. Given how conservative evolution has been when it comes to inventing new ways of solving biological problems, we hypothesized that the same suite of mechanisms that controls development of fins and limbs might also operate in the genitalia. The simplicity of the hypothesis made it relatively easy to test. By using microsurgical manipulations and transcriptional profiling, we identified the urethral epithelium as an organizing center in the genital tubercle, the embryonic precursor of the penis and clitoris. One gene in particular, Sonic hedgehog, captured our attention because it is expressed in urethral epithelial cells and it controls pattern in numerous other structures, including the limb and spinal cord. We tested the function of Sonic hedgehog in the genital tubercle and found that deletion of the gene results in complete absence of external genitalia. After identifying an essential early role of Sonic hedgehog in genital tubercle formation, we dissected its function at each day of genital development by temporally controlled inactivation. These experiments revealed a continuous requirement for Sonic hedgehog over the course of genital development; it coordinates anorectal and genitourinary development at early stages and regulates patterning of the genital tubercle at later stages.
Gene-Environment Interactions in the Embryo
The incidence of hypospadias has approximately doubled, without explanation, since the late 1960s. It is widely suspected that environmental endocrine-disrupting chemicals (EDCs) contribute to the rising incidence of genitourinary malformations, although how these contaminants interact with the gene networks that regulate development is unknown.
Our search for genes with potential roles in urethral tube closure led us to a fibroblast growth factor receptor, FgfR2, that is expressed in two tissues involved in hypospadias, the urethra and the prepuce (or foreskin). FgfR2 and its ligand, Fgf10, are also required for the initiation of limb budding. Our studies showed that when either Fgf10 or FgfR2 is functionally inactivated in mouse embryos, the mice develop hypospadias. The genital phenotype of FgfR2-null mutants is strikingly similar to that which results from inactivation of the androgen receptor (AR), which causes genital feminization and hypospadias. We found a connection between androgen signaling and FgfR2 activity by chemically blocking AR function in mouse genital tubercles, which led to a dose-dependent down-regulation of FgfR2 expression. Additionally, we identified an AR-binding site in the FgfR2 promoter, suggesting that FgfR2 is a direct target of AR. These results provide the first direct link between systemically circulating endocrine signals and the local regulation of gene expression during external genital development.
It seems unlikely that mutations in developmental control genes can account for the large number of children who develop hypospadias. Indeed, screens for mutations in candidate genes in affected individuals have not been fruitful, which suggests that nongenetic factors may be at play. Our finding that a gene required for urethral tube closure can be transcriptionally silenced by exposure to an anti-androgenic chemical raised a new question: Could hypospadias result from exposure of an embryo to EDCs that transiently down-regulate the expression of genes required for urethral tube closure? We are now screening mouse models with EDCs that induce hypospadias and using microarrays to identify transcriptional responses to these chemicals. Our hope is that discovery of the genetic pathways that respond to EDCs in the maternal environment will provide a foundation for the development of preventive strategies.
Grants from the National Institutes of Health and the National Science Foundation provided partial support for these projects.
As of May 30, 2012