Received: 9 October 2018 | Revised: 29 January 2019 | Accepted: 4 February 2019 DOI: 10.1111/fwb.13295 ORIGINAL ARTICLE Phenotypic variability of rusty crayfish (Faxonius rusticus) at the leading edge of its riverine invasion Mathis L. Messager | Julian D. Olden School of Aquatic and Fishery Sciences, University of Washington, Seattle, Abstract Washington, U.S.A. 1. Species around the globe are undergoing phenotypic shifts at ecologically rele- Correspondence vant timescales as they invade new ecosystems and respond to changing environ- Mathis L. Messager, School of Aquatic and ments. Disentangling the contribution of environmental gradients from the Fishery Sciences, University of Washington, Seattle, WA, U.S.A. process of range expansion in shaping these changes, and identifying the specific Email: [email protected] traits undergoing selection, are both critical to anticipate the secondary spread Funding information and impact of invasive species. School of Aquatic and Fishery Sciences, 2. Here, we investigate phenotypic changes in rusty crayfish (Faxonius rusticus), a Grant/Award Number: John N. Cobb Scholarship in Fisheries; University of nuisance invasive species, through an extensive survey of their invasion gradient Washington, Grant/Award Number: in multiple tributaries of the John Day River (JDR, Oregon, U.S.A.), a major tribu- H. Mason Keeler Endowed Professorship; The Crustacean Society, Grant/Award tary of the Columbia River. Number: Fellowship in Graduate Studies; 3. Rusty crayfish in the JDR have developed better physiological condition (intrinsic Society for Freshwater Science, Grant/ Award Number: Simpson Fund growth and/or reproductive potential measured as RNA/DNA ratio) but less com- petitive morphology (lighter body and smaller claws) as they spread upstream and downstream from their location of initial introduction. In addition, rusty crayfish in invasion front populations are at a lower trophic level than conspecifics closer to core areas. 4. By accounting for variations in temperature, primary productivity, and prey (mac- roinvertebrates) biomass throughout the invasion extent of rusty crayfish, our findings suggest that low conspecific densities at the invasion edge and spatial sorting primarily drive these phenotypic changes. The trends observed here are thus likely to intensify over time as rusty crayfish continues to rapidly spread throughout the JDR and reach the mainstem Columbia River. 5. Our study shows that phenotypic shifts can manifest rapidly over environmental gradients experienced during the range expansion of aquatic invasive species. Patterns in both morphological and functional traits documented in the JDR dem- onstrate the importance of both environmental factors and dispersal processes in shaping these responses in riverine networks. KEYWORDS aquatic invasive species, Faxonius rusticus, invasive crayfish, John Day River, range margin Freshwater Biology. 2019;1–14. wileyonlinelibrary.com/journal/fwb © 2019 John Wiley & Sons Ltd. | 1 2 | MESSAGER AND OLDEN 1 | INTRODUCTION potential for directional selection (Burton, Phillips, & Travis, 2010), although dispersal ability may evolve independently of other life- Species around the globe are exposed to changing selection pres- history traits (Bonte & Dahirel, 2017). The traits involved in pro- sures as they invade new landscapes or shift their range to track moting dispersal and growth at the invasion front are so numerous environmental change (Moran & Alexander, 2014). When their geo- that natural selection and spatial sorting, when enhancing these graphic distribution shifts or expands, the individuals in the vanguard traits, can impact morphology, physiology, behaviour, immunol- of these populations often face novel environmental conditions, ogy, and life history, among others (Chuang & Peterson, 2016). predators, and competitors (Chuang & Peterson, 2016). Mounting Understanding the specific traits undergoing selection in invasive evidence suggests that these factors, in combination with low con- species and disentangling the influence of environmental condi- specific densities relative to those experienced by core populations, tions from contemporary evolution therefore requires empirical promote rapid changes in species phenotypes at range boundaries studies that are specific to the species and systems at hand. (Chuang & Peterson, 2016). Crayfish are among the most widely introduced freshwater an- Phenotypic changes at the leading edge of invasive popu- imals worldwide (Lodge et al., 2012). Following their introduction, lations’ range have been observed in many taxonomic groups, non- native crayfish can cause severe ecological impacts across including amphibians (e.g. cane toad Rhinella marina, Perkins, entire food webs to a greater extent than native crayfish because, Phillips, Baskett, & Hastings, 2013), insects (e.g. ground beetle like their native counterparts, they have polytrophic feeding hab- Merizodus soledadinus, Laparie, Renault, Lebouvier, & Delattre, its, but also often reach much greater densities and heightened 2013), fish (e.g. round goby Neogobius melanostomus, Brandner, levels of foraging activity (Hansen et al., 2013; Pintor & Sih, 2009; Cerwenka, Schliewen, & Geist, 2013), and decapods (e.g. signal Twardochleb, Olden, & Larson, 2013). In invaded ecosystems, crayfish Pacifastacus leniusculus, Hudina, Hock, Žganec, & Lucić, native crayfish species can be displaced within a few years, and 2012). These changes have been manifested in traits ranging from populations of macrophytes, insects, snails, and fish often decline body length and fecundity to boldness (Chuang & Peterson, 2016), (Bobeldyk & Lamberti, 2010; McCarthy, Hein, Olden, & Vander and have been associated with accelerated invasion rates (Phillips, Zanden, 2006; Olden, McCarthy, Maxted, Fetzer, & Vander Zanden, Brown, & Shine, 2010a; Weiss- Lehman, Hufbauer, & Melbourne, 2006; Rosenthal, Stevens, & Lodge, 2006). Changes in population 2017) and increased impacts to recipient ecosystems (Brandner structure, behaviour, morphology, and physiology have already et al., 2013; Iacarella, Dick, & Ricciardi, 2015). Enhanced insight been reported between core and edge populations in several on- into the processes leading to phenotypic changes in invasive spe- going river invasions by crayfish (Hudina, Zganec, & Hock, 2015; cies is thus essential for anticipating their future spread and im- Hudina et al., 2012; Pârvulescu, Pîrvu, Moroşan, & Zaharia, 2015; pact (Phillips, 2015), as well as predicting the outcome of species Rebrina, Skejo, Lucić, & Hudina, 2015). At the biogeographical level, range shifts in response to climate change (Caplat et al., 2013; differences in growth, survival, feeding habits, and behaviour are Travis et al., 2013). also common among crayfish congeners between their native and Phenotypic plasticity (Davidson, Jennions, & Nicotra, 2011), non- native range, further demonstrating the potential phenotypic natural selection (Brown, Kelehear, & Shine, 2013), and spatial changes wrought by the invasion process (Glon, Reisinger, & Pintor, sorting (Shine, Brown, & Phillips, 2011) are the three dominant pro- 2018; Pintor & Sih, 2009; Reisinger, Elgin, Towle, Chan, & Lodge, cesses responsible for observed trait variability at range edges, yet 2017; Sargent & Lodge, 2014). Even though the consequences of their respective contributions are seldom understood. Phenotypic these changes on invaded ecosystems often remain unexplored, in- plasticity, the ability for multiple phenotypes to arise from a sin- creased invasion rates alone could challenge our ability to respond gle genotype in response to changing environmental conditions, is to new and ongoing crayfish invasions. In addition, given that the particularly prevalent in invasive species (Davidson et al., 2011). It impact of an invasive species on the recipient ecosystem is not only is crucial in allowing populations to spread and adapt to changing a function of its range size but also of its abundance, per- capita environments faster than would otherwise be possible by evo- effect, and other factors (Thomsen, Olden, Wernberg, Griffin, & lution through natural selection alone (Chevin, Lande, & Mace, Silliman, 2011), changes in its somatic and reproductive growth 2010). Furthermore, abiotic and biotic forces at the invasion front rates or trophic niche could have severe consequences for native can lead to trait evolution by natural selection. Low intra- specific communities. Knowledge of the potential phenotypic shifts occur- density at the leading edge is most likely to shift selective pres- ring at the front of crayfish invasions could thus shed light on both sures towards higher growth and reproduction (Phillips, Brown, & the selection pressures exerted upon dispersing populations and on Shine, 2010b). Lastly, the range expansion process itself can lead the future impact of these invasions. to adaptive changes in traits through spatial sorting, whereby the In this study, we investigated changes in rusty crayfish (Faxonius fastest dispersing individuals at the expanding edge of the popula- rusticus, Girard 1852, previously Orconectes rusticus, Crandall & De tion systematically interbreed, resulting in selection for enhanced Grave, 2017) traits across their invasion gradient in the John Day dispersal ability in their offspring if dispersive traits are heritable River (JDR), the only known occurrence of this species west of the (Shine et al., 2011). This runaway process continues in subsequent North American