1 PeerJ 2013 Vol: 1():. DOI: 10.7717/peerj.163

Microgeographic maladaptive performance and deme depression in response to roads and runoff

Despite theoretical understanding and empirical detection of local adaptation in natural environments, our knowledge of such divergence in fragmented habitats remains limited, especially in the context of microgeographic spatial scales and contemporary time scales. I used a combination of reciprocal transplant and common garden exposure experiments to evaluate potential microgeographic divergence in a pool-breeding amphibian occupying a landscape fragmented by roads. As indicated by reduced rates of survival and increased rates of malformation, I found evidence for maladaptation in road adjacent populations. This response is in direct counterpoint to recently described local adaption by a cohabiting species of amphibian. These results suggest that while divergence might commonly follow habitat modification, the direction of its outcome cannot be generalized even in identical habitats. Further, maladaptive responses can be associated with a more generalized depression effect that transcends the local environment. Alongside recent reports, these results suggest that maladaptive responses may be an emerging consequence of human-induced environmental change. Thus future studies should carefully consider the population unit as a key level for inference.

Mentions
Figures
Figure 1: Population locations and reciprocal transplant design.Locations of the 10 pools comprising transplants are shown. Asterisk on inset indicates approximate study location in northeastern Connecticut, USA. Like symbols indicate paired pools (red  =  roadside; blue  =  woodland). Interstate highway (I-84) and on/off-ramp infrastructure is indicated in yellow. Primary roads are heavily shaded, while secondary and unpaved roads are lightly shaded. Bar graph shows average specific conductance (µS; ±1 SE) in the woodland and roadside pools. From among a suite of abiotic variables, I found that pools differed only with respect to specific conductance, which indicates elevated ionic concentration, reflecting the increased presence of chloride ions in roadside pools that originate from road salt runoff. Figure 2: Embryonic Rana sylvatica survival across the G × E interaction.Survival (±1 SE) is shown here as the mean proportion of individuals surviving to feeding stage across all experimental units (N = 99). The woodland deme is represented by open circles while the roadside deme is represented by filled squares. The environment in which the animals were grown out is on the x-axis. Figure 3: Malformations and survival following road salt exposure.Rana sylvatica responses to experimental road salt exposure. Values (±1 SE) represent means of container level responses. (A) Proportion of embryos surviving to hatchling with axial malformations following chronic exposure to three concentrations of road salt: Low  =  175 µS; medium  =  1,000 µS; high  =  4,000 µS (n = 90). Open bars represent woodland deme; filled bars represent roadside deme. (B) Proportion of hatchling survival following acute exposure to road salt (n = 200). Open circles represent woodland deme; filled squares represent roadside deme. Concentrations of total roadsalt (g/L) are shown. From lowest to highest, these concentrations yielded specific conductance values of 1,300, 6,430, 9,370, 13,150, and 17,830 µS, respectively. Figure 4: Wood frog fecundity in relation to female size and deme.Female Rana sylvatica fecundity. Number of eggs laid per female (n = 56) is shown in relation to female body size (snout-vent length [SVL]). Open circles represent females from the woodland deme; filled squares represent females from the roadside deme.
Altmetric
References
  1. JA Bantle; RA Finch; DT Burton; G Linder Atlas of abnormalities: a guide for the performance of FETAX , (1991) .
    • . . . When all embryos hatched or died, I estimated survival, developmental stage, SVL, and prevalence of axial malformations (Bantle et al., 1991). . . .
  2. J Bernardo Maternal effects in animal ecology American Zoologist 36, 83-105 (1996) .
    • . . . For example, maternal effects, broadly defined as “a direct effect of a parent’s phenotype on the phenotype of its offspring” (Bernardo, 1996) may be associated with the roadside deme’s reduced ability to survive . . .
  3. KA Berven Factors affecting variation in reproductive traits within a population of wood frogs (Rana sylvatica) Copeia 1988, 605-615 (1988) .
    • . . . For example, if we assume that roadside adults face higher mortality risk—either through carryover effects from the aquatic environment, or through terrestrial road pressures such as roadkill—then the average age of roadside breeders should be lower, a condition associated with poorer offspring provisioning in wood frogs (Berven, 1988) . . .
  4. KA Berven Factors affecting population fluctuations in larval and adult stages of the wood frog (Rana Sylvatica) Ecology 71, 1599-1608 (1990) .
    • . . . Because adult wood frogs exhibit high site fidelity (Berven, 1990) and have shown phenotypic differentiation across microgeographic environmental gradients (Skelly, 2004), I assumed that the embryos I collected represent a lineage of wood frogs specific to the collection pool (i.e., ‘population’) . . .
  5. KA Berven; TA Grudzien Dispersal in the wood frog (Rana sylvatica)- implications for genetic population-structure Evolution 44, 2047-2056 (1990) .
    • . . . The wood frog is an explosive breeder, with the majority of oviposition for a given population occurring in a single night (Berven & Grudzien, 1990) . . .
  6. BM Bolker; ME Brooks; CJ Clark; SW Geange; JR Poulsen; MHH Stevens; JSS White Generalized linear mixed models: a practical guide for ecology and evolution Trends in Ecology & Evolution 24, 127-135 (2009) .
    • . . . I used a combination of MCMC randomization methods and log-likelihood approaches to conduct inference (Bolker et al., 2009) . . .
  7. SP Brady Road to evolution? Local adaptation to road adjacency in an amphibian (Ambystoma maculatum) Scientific Reports 2, 235 (2012) .
    • . . . In human-altered contexts, where novel selection pressures and reduced gene flow can yield adaptive responses (Brady, 2012) and/or genetic drift (Delaney, Riley & Fisher, 2010), and the rates of phenotypic change are high relative to undisturbed settings (Darimont et al., 2009; Hendry, Farrugia & Kinnison, 2008), the potential for population divergence should be elevated . . .
    • . . . My method for selecting populations for this study is previously reported in a reciprocal transplant study that indicated local adaptation to roads in the spotted salamander (Brady, 2012) . . .
    • . . . S1). (Details of the experimental enclosures are described elsewhere (Brady, 2012) and an image can be found in Fig . . .
    • . . . Specific methods are reported elsewhere (Brady, 2012) . . .
    • . . . Regardless of mechanism, the maladaptive performance of the roadside deme of wood frogs occurs in direct contrast to the local adaptation of the roadside deme of spotted salamanders breeding and dwelling in the very same pools (Brady, 2012) . . .
  8. M Christie; M Marine; R French; M Blouin Genetic adaptation to captivity can occur in a single generation Proceedings of the National Academy of Sciences of the United States of America 109, 238-242 (2012) .
    • . . . Similarly, hatchery-reared fish have become locally adapted to the hatchery environment at the cost of maladaptation to the wild (Christie et al., 2012) . . .
  9. BJ Crespi The evolution of maladaptation Heredity 84, 623-629 (2000) .
    • . . . However, gene flow and genetic drift can constrain local adaptation, rendering populations equivalently adapted between environments, or even maladapted to their local environment (Crespi, 2000; Falk et al., 2012; Stearns & Sage, 1980; Urban, 2006) . . .
    • . . . Maladaptation can be defined most generally as a deviation from an adaptive peak caused by genetic processes (Crespi, 2000) . . .
  10. CT Darimont; SM Carlson; MT Kinnison; PC Paquet; TE Reimchen; CC Wilmers Human predators outpace other agents of trait change in the wild Proceedings of the National Academy of Sciences of the United States of America 106, 952-954 (2009) .
    • . . . In human-altered contexts, where novel selection pressures and reduced gene flow can yield adaptive responses (Brady, 2012) and/or genetic drift (Delaney, Riley & Fisher, 2010), and the rates of phenotypic change are high relative to undisturbed settings (Darimont et al., 2009; Hendry, Farrugia & Kinnison, 2008), the potential for population divergence should be elevated . . .
    • . . . For example, phenotypes of species harvested for human consumption have undergone rapid and substantial change (Darimont et al., 2009) in directions believed to oppose adaptation to the increased environmental variation predicted to occur with climate change (Koons, 2009) . . .
  11. KS Delaney; SPD Riley; RN Fisher A rapid, strong, and convergent genetic response to urban habitat fragmentation in four divergent and widespread vertebrates PLoS ONE 5, (2010) .
    • . . . In human-altered contexts, where novel selection pressures and reduced gene flow can yield adaptive responses (Brady, 2012) and/or genetic drift (Delaney, Riley & Fisher, 2010), and the rates of phenotypic change are high relative to undisturbed settings (Darimont et al., 2009; Hendry, Farrugia & Kinnison, 2008), the potential for population divergence should be elevated . . .
  12. P Edelaar; AM Siepielski; J Clobert Matching habitat choice causes directed gene flow: a neglected dimension in evolution and ecology Evolution 62, 2462-2472 (2008) .
    • . . . For example, younger or otherwise misfit adults may breed more frequently in the suboptimal conditions found in roadside as compared to woodland pools, potentially through competitive exclusions or the process of ‘matching habitat choice’ (Edelaar, Siepielski & Clobert, 2008) . . .
  13. JJ Falk; CE Parent; D Agashe; DI Bolnick Drift and selection entwined: asymmetric reproductive isolation in an experimental niche shift Evolutionary Ecology Research 14, 403-423 (2012) .
    • . . . However, gene flow and genetic drift can constrain local adaptation, rendering populations equivalently adapted between environments, or even maladapted to their local environment (Crespi, 2000; Falk et al., 2012; Stearns & Sage, 1980; Urban, 2006) . . .
    • . . . While this generalized depression effect (hereafter ‘deme depression’) is not well described in the literature (but see Falk et al., 2012), there are conceptual analogues worth considering . . .
  14. G Garcia-Ramos; M Kirkpatrick Genetic models of adaptation and gene flow in peripheral populations Evolution 51, 21-28 (1997) .
    • . . . Where it has been described, maladaptation at the population level is typically the consequence of migration load (Hanski, Mononen & Ovaskainen, 2011), resulting when asymmetrical gene flow from non-adapted source populations yields genotypes of sub-optimal fitness in the recipient population (Garcia-Ramos & Kirkpatrick, 1997) . . .
  15. KL Gosner A simplified table for staging anuran embryos and larvae with notes on identification Herpetologiea 16, 183-190 (1960) .
    • . . . At the conclusion of the experiment, when all eggs had either hatched and reached feeding stage or died, I estimated survival, developmental stage (Gosner, 1960), and snout-vent length (SVL) at the container level . . .
    • . . . Upon approaching feeding stage (Gosner developmental stage 25 (Gosner, 1960)), larvae were fed a mixture of 3:1 ground rabbit chow and fish flakes at a daily ration of 10% average body mass . . .
  16. I Hanski; T Mononen; O Ovaskainen The American Naturalist 177, 29-43 (2011) .
    • . . . Where it has been described, maladaptation at the population level is typically the consequence of migration load (Hanski, Mononen & Ovaskainen, 2011), resulting when asymmetrical gene flow from non-adapted source populations yields genotypes of sub-optimal fitness in the recipient population (Garcia-Ramos & Kirkpatrick, 1997) . . .
  17. AP Hendry; TJ Farrugia; MT Kinnison Human influences on rates of phenotypic change in wild animal populations Molecular Ecology 17, 20-29 (2008) .
    • . . . In human-altered contexts, where novel selection pressures and reduced gene flow can yield adaptive responses (Brady, 2012) and/or genetic drift (Delaney, Riley & Fisher, 2010), and the rates of phenotypic change are high relative to undisturbed settings (Darimont et al., 2009; Hendry, Farrugia & Kinnison, 2008), the potential for population divergence should be elevated . . .
  18. J Hereford A quantitative survey of local adaptation and fitness trade-offs American Naturalist 173, 579-588 (2009) .
    • . . . For example, spatially heterogeneous selection pressures coupled with heritable phenotypic variation has often been found to result in local adaptation (Hereford, 2009; Kawecki & Ebert, 2004) . . .
    • . . . In a review of this literature, Hereford (2009) found the frequency of maladaptation to be 0.29, as compared to the frequency of local adaptation, which was 0.71 . . .
  19. WA Hopkins; SE DuRant; S Brandon Patrick; CL Rowe; BP Jackson Reproduction, embryonic development, and maternal transfer of contaminants in the amphibian Gastrophryne carolinensis Environmental Health Perspectives 114, 661-666 (2006) .
    • . . . This possibility is supported by outcomes from studies in which coal combustion waste contaminants such as strontium and selenium are maternally transferred, negatively affecting offspring survival and development in two species of toads (Hopkins et al., 2006; Metts et al., 2012). . . .
  20. N Karraker; J Gibbs Contrasting road effect signals in reproduction of long- versus short-lived amphibians Hydrobiologia 664, 213-218 (2011) .
    • . . . While roadside egg masses appeared somewhat more compact than woodland egg masses (Karraker & Gibbs, 2011), there were no conspicuous differences in the embryos themselves between sites, and selection of embryos from within each egg mass was haphazard . . .
  21. TJ Kawecki; D Ebert Conceptual issues in local adaptation Ecology Letters 7, 1225-1241 (2004) .
    • . . . For example, spatially heterogeneous selection pressures coupled with heritable phenotypic variation has often been found to result in local adaptation (Hereford, 2009; Kawecki & Ebert, 2004) . . .
    • . . . Evidenced by a specific form of gene-by-environment interaction (G × E), local adaptation results when the local population has higher fitness than a foreign population measured within the local environment (Kawecki & Ebert, 2004) . . .
  22. DN Koons Does harvest select for maladaptation in an increasingly variable world? Proceedings of the National Academy of Sciences of the United States of America 106, (2009) .
    • . . . For example, phenotypes of species harvested for human consumption have undergone rapid and substantial change (Darimont et al., 2009) in directions believed to oppose adaptation to the increased environmental variation predicted to occur with climate change (Koons, 2009) . . .
  23. AT Laugen; A Laurila; J Merilä Maternal and genetic contributions to geographical variation in Rana temporaria larval life-history traits Biological Journal of the Linnean Society 76, 61-70 (2002) .
    • . . . All models of performance variables were composed with and without embryo size as a covariate in order to estimate the potential influence of egg size mediated maternal effects (Laugen, Laurila & Merilä, 2002) . . .
  24. SA Levin The problem of pattern and scale in ecology Ecology 73, 1943-1967 (1992) .
    • . . . Identifying informative scales and levels of biological organization for investigation has long been regarded as a critical challenge in ecology, limiting our ability to describe variation in the natural world (Levin, 1992) . . .
  25. BS Metts; KA Buhlmann; DE Scott; TD Tuberville; WA Hopkins Interactive effects of maternal and environmental exposure to coal combustion wastes decrease survival of larval southern toads (Bufo terrestris) Environmental Pollution 164, 211-218 (2012) .
    • . . . This possibility is supported by outcomes from studies in which coal combustion waste contaminants such as strontium and selenium are maternally transferred, negatively affecting offspring survival and development in two species of toads (Hopkins et al., 2006; Metts et al., 2012). . . .
  26. J-S Moore; A Hendry Can gene flow have negative demographic consequences? Mixed evidence from stream threespine stickleback Philosophical Transactions of the Royal Society Biological Sciences 364, 1533-1542 (2009) .
    • . . . Though evident in wild populations (Moore & Hendry, 2009), such empirical examples of population level maladaptation are rare (Spitzer, 2006) . . .
  27. TE Mousseau; CW Fox Maternal effects as adaptations Maternal effects, developmental plasticity, and life history evolution , 244-260 (1998) .
    • . . . Yet, I found no evidence that egg size—a maternally mediated trait known to influence amphibian performance (reviewed by Mousseau & Fox, 1998)—varied across deme or had any effect on performance variables . . .
  28. D Newman; D Pilson Increased probability of extinction due to decreased genetic effective population size: experimental populations of Clarkia pulchella Evolution 51, 354-362 (1997) .
    • . . . As one such example, inbreeding depression can arise in the context of reduced effective population size (Newman & Pilson, 1997) that could result from intense selection (Robertson, 1961) in the roadside environment . . .
  29. M Pigliucci Evolution of phenotypic plasticity: where are we going now? Trends in Ecology & Evolution 20, 481-486 (2005) .
    • . . . Population level approaches—which are bringing to light profound variation that occurs across small spatial and temporal scales (Pigliucci, 2005)—offer a promising advancement on this challenge . . .
  30. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at Link , (2012) .
    • . . . All statistical analyses were conducted in R V. 2.15.0 (R Development Core Team, 2012) . . .
  31. WS Rasband ImageJ. Bethesda, Maryland: US National Institutes of Health. Available at Link , (1997–2012) .
    • . . . Specifically, I used ImageJ (Rasband, 1997–2012) to measure the two-dimensional surface area of 15 embryos per cluster represented by a best-fit ellipse (Brady, 2012) . . .
    • . . . Following oviposition, egg masses were photographed and fecundity was estimated from each photo using ImageJ (Rasband, 1997–2012) to manually count the total number of eggs oviposited per female. . . .
  32. JL Richardson Divergent landscape effects on population connectivity in two co-occurring amphibian species Molecular Ecology 21, 4437-4451 (2012) .
    • . . . However, the types of roads in this system do not appear to limit gene flow between wood frog populations in surrounding landscapes separated by similar or larger distances of space (Richardson, 2012) . . .
  33. A Robertson Inbreeding in artificial selection programmes Genetical Research 2, 189-194 (1961) .
    • . . . As one such example, inbreeding depression can arise in the context of reduced effective population size (Newman & Pilson, 1997) that could result from intense selection (Robertson, 1961) in the roadside environment . . .
  34. O Ronce; F Shaw; F Rousset; R Shaw Is inbreeding depression lower in maladapted populations? A quantitative genetics model Evolution 63, 1807-1819 (2009) .
    • . . . However, inbreeding depression is predicted to be minimal in the context of intense maladaptation (Ronce et al., 2009) . . .
  35. D Sanzo; SJ Hecnar Effects of road de-icing salt (NaCl) on larval wood frogs (Rana sylvatica) Environmental Pollution 140, 247-256 (2006) .
    • . . . These concentrations were selected to represent previously documented acute exposure assays of the wood frog (Sanzo & Hecnar, 2006) . . .
  36. GG Simpson Tempo and mode in evolution , (1944) .
    • . . . Likewise, phenotypic homogeneity would be expected if we view maladaptation through the lens of the adaptive phenotypic landscapes framework (Simpson, 1944), which predicts that maladaptation is a ubiquitous outcome of dynamic environments because the response to selection always lags one generation behind selection itself; therefore, the adaptive high point is constantly shifting before phenotypes can reach their adaptive peak (Crespi, 2000) . . .
  37. DK Skelly Microgeographic countergradient variation in the wood frog, Rana sylvatica Evolution 58, 160-165 (2004) .
    • . . . Variation observed at this level can reveal dramatic differences among local populations, even across microgeographic spatial scales (Skelly, 2004; Urban, 2010) . . .
    • . . . Given the suite of novel selection pressures associated with road fragmentation and runoff—along with previously documented contemporary local adaptation to temperature by the wood frog in this landscape (Skelly, 2004)—I expected that road adjacent populations of this species would be locally adapted to the roadside environment . . .
    • . . . Because adult wood frogs exhibit high site fidelity (Berven, 1990) and have shown phenotypic differentiation across microgeographic environmental gradients (Skelly, 2004), I assumed that the embryos I collected represent a lineage of wood frogs specific to the collection pool (i.e., ‘population’) . . .
  38. B Spitzer Local maladaptation in the soft scale insect Saissetia coffeae (Hemiptera: Coccidae) Evolution 60, 1859-1867 (2006) .
    • . . . Though evident in wild populations (Moore & Hendry, 2009), such empirical examples of population level maladaptation are rare (Spitzer, 2006) . . .
    • . . . As an empirical phenomenon, maladaptation is less common than local adaptation (Crespi, 2000; Spitzer, 2006) . . .
  39. SC Stearns; RD Sage Maladaptation in a marginal population of the mosquito fish, Gambusia-Affinis Evolution 34, 65-75 (1980) .
    • . . . However, gene flow and genetic drift can constrain local adaptation, rendering populations equivalently adapted between environments, or even maladapted to their local environment (Crespi, 2000; Falk et al., 2012; Stearns & Sage, 1980; Urban, 2006) . . .
  40. J Thompson Specific hypotheses on the geographic mosaic of coevolution The American Naturalist 153, S1-S14 (1999) .
    • . . . Rather, most examples of maladaptation in general describe reduced fitness across higher levels of organization, typically in host-parasite or host-pathogen systems, where coevolution can generate reciprocating dynamics of maladaptation and adaptation (Thompson, 1999). . . .
  41. SC Trombulak; CA Frissell Review of ecological effects of roads on terrestrial and aquatic communities Conservation Biology 14, 18-30 (2000) .
    • . . . This may be especially likely in harsh roadside environments, where mutagens such as heavy metals are known to accumulate from road runoff (Trombulak & Frissell, 2000). . . .
  42. MC Urban Maladaptation and mass effects in a metacommunity: consequences for species coexistence The American Naturalist 168, 28-40 (2006) .
    • . . . However, gene flow and genetic drift can constrain local adaptation, rendering populations equivalently adapted between environments, or even maladapted to their local environment (Crespi, 2000; Falk et al., 2012; Stearns & Sage, 1980; Urban, 2006) . . .
  43. MC Urban Microgeographic adaptations of spotted salamander morphological defenses in response to a predaceous salamander and beetle Oikos 119, 646-658 (2010) .
    • . . . Variation observed at this level can reveal dramatic differences among local populations, even across microgeographic spatial scales (Skelly, 2004; Urban, 2010) . . .
  44. GA Wellborn; DK Skelly; EE Werner Mechanisms creating community structure across a freshwater habitat gradient Annual Review of Ecology and Systematics 27, 337-363 (1996) .
    • . . . Finally, to investigate potential mechanisms by which road proximity influences wood frog performance, I characterized the environment at each pool by estimating a suite of abiotic variables associated with amphibian distribution and performance (Wellborn, Skelly & Werner, 1996) . . .
    • . . . In each pool, I measured seven environmental characteristics associated with amphibian distribution and performance (Wellborn, Skelly & Werner, 1996) . . .
Expand