How Did the Cliff Swallow Cross the Road? A Genomic Study of Wing Length Variation Under Traffic-Mediated Selection in a Highway Nesting Bird

Abstract

Human modification has become a near-ubiquitous force of environmental change globally, challenging wildlife with novel environmental pressures on form and function and, in some cases, necessitating rapid evolutionary response. Perhaps one of the largest elements of the contemporary human “footprint” is the network of approximately 40 million miles of roads and highways that now encircle the globe, around one-tenth of which inundate the United States alone. Ironically, this extensive infrastructure that connects our communities over large geographic distances fractures the natural environments they traverse, introducing novel barriers to migration and geographic connectivity among otherwise similar habitat types. Additionally, high volumes of traffic have introduced novel acoustic, chemical, and physical disruptions to organisms living adjacent to roads, resulting in elevated stress levels, genetic isolation, water and air pollution, and the threat of vehicular collision. However, despite the geographic range and multifaceted threats of roadway systems, the evolutionary implications of such extensive human habitat disturbance remain understudied. Brown and Brown (2013) provides among the most well-documented evidence for highways as a persistent source of evolutionary change in a wild population. Using phenotypic and demographic data collected across a 30-year period, they documented rapid shifts in wing morphology in highway commensal nesting populations of cliff swallows (Petrochelidon pyrrhonota) in southwestern Nebraska. They additionally observed that the number of individuals killed by vehicles dramatically decreased over the study period, with the overall population shifting towards a smaller wingspan during that period and the road-killed individuals having a significantly longer wingspan than the rest of the population (Brown and Brown 2013). In this study, we sought to identify candidate genes that may have been targeted by selection during the thirty year time period. Using whole genome sequencing (WGS) to search for loci displaying extreme genetic divergence between short-winged and long-winged individuals within the focal population, we found several distinct genetic peaks differentiating long- and short-winged groups, with the most robust signatures of divergence occurring on chromosomes 1, 3, 5, and the Z sex chromosome. Investigation into the genes lying within the most significant divergence peaks of the genome uncovered several loci with known function in determining vertebrate limb length, osteoblast formation, and joint morphology. Additionally, we found that the strongest peak genome-wide contained a suite of physically proximate genes associated with other interacting developmental and physiological functions know to be crucial for determining flight power, maneuverability, and endurance in birds, including metabolic control, lipogenesis, keel length, chest width, muscle growth, and muscle fiber composition. Taken together, these data suggest that intraspecific variation in wing length may be one subordinate component of a larger suite of traits that determine variation in flight performance, which may have collectively been targeted by selection in the species due to traffic-mediated selection over a 30-year period.