Convergent changes in the trophic ecology of extremophile fish along replicated environmental gradientsFreshwater Biology


Michael Tobler, Kristin Scharnweber, Ryan Greenway, Courtney N. Passow, Lenin Arias-Rodriguez, Francisco J. García-De-León
Aquatic Science


Convergent changes in the trophic ecology of extremophile fish along replicated environmental gradients


LENIN ARIAS-RODRIGUEZ? AND FRANCISCO J. GARCIA-DE-LE ON? *Division of Biology, Kansas State University, Manhattan, KS, U.S.A. ?Evolutionary Biology Centre, Department of Ecology and Genetics, Limnology, Uppsala University, Uppsala, Sweden ?Division Academica de Ciencias Biologicas, Universidad Juarez Autonoma de Tabasco, Villahermosa, Tabasco, Mexico ?Centro de Investigaciones Biologicas del Noroeste (Laboratorio de Genetica para la Conservacion), La Paz, Baja California, Mexico

SUMMARY 1. Divergent selection along environmental gradients connecting locally restricted extreme habitats and adjacent benign habitats can shape convergent evolution of traits involved in coping with physiochemical stressors and can drive speciation. At the same time, the presence of such stressors alters aspects of the biotic environment, including resource availability and competitive regimes.

However, it remains unclear whether and how the ecology of populations occurring in both extreme and benign environments varies in a predictable fashion. 2. We investigated the trophic ecology of live-bearing fishes of the genus Poecilia that have independently colonised multiple springs containing toxic hydrogen sulphide in southern Mexico. Sulphide spring fish are adapted to the unique environmental conditions and are reproductively isolated from ancestral populations in adjacent non-sulphidic habitats. We used gut content analyses to test whether colonisation of extreme habitats was accompanied by shifts of trophic resource use and expansions of trophic niche width. Furthermore, we tested whether dietary shifts were reflected in trophic morphology by comparing intestinal tract lengths among populations using both wild-caught and common garden-raised individuals. 3. Gut content analyses revealed that fish inhabiting toxic springs expanded their trophic niche width and changed their dietary resource use from detritus and algae to sulphide bacteria and invertebrates. This dietary shift was paralleled by changes in intestinal tract morphology, whereby sulphide spring fish had shorter intestines than fish from adjacent non-sulphidic habitats. Analysis of common garden-raised fish indicated that morphological differences between sulphidic and non-sulphidic populations are at least in part due to genetic differentiation. Both patterns of trophic resource use and differentiation in trophic morphology were consistent across replicated pairs of sulphidic and non-sulphidic populations, although the magnitude of differentiation varied among river drainages. 4. Our results suggest that colonisation of and adaptation to sulphide springs in southern Mexico was paralleled by convergent changes in trophic ecology. This highlights the complexity of environmental gradients and the necessity of considering multiple sources of selection when studying the evolution of complex phenotypes.

Keywords: dietary niche, ecological diversification, extreme environment, hydrogen sulphide springs, Poecilia (Poeciliidae)

Correspondence: Michael Tobler, Kansas State University, Division of Biology, 116 Ackert Hall, Manhattan, KS 66506, U.S.A.

E-mail: 768 ? 2015 John Wiley & Sons Ltd

Freshwater Biology (2015) 60, 768?780 doi:10.1111/fwb.12530


Successful maintenance of homeostasis in organisms tolerating extreme environments requires costly adaptations absent in closely related taxa (Sibly & Calow, 1989;

Townsend, Begon & Harper, 2003). Studies of organisms inhabiting extreme environments have mostly focused on identifying mechanisms that allow them to survive exposure to particular physiochemical stressors, and strategies for coping involve modifications of biochemical and physiological pathways, morphology, behaviour, life history and even symbioses with other organisms (Waterman, 1999; Van Dover, 2000; Bergman et al., 2003;

Ip, Kuah & Chew, 2004). Divergent selection in extreme environments not only drives population differentiation in phenotypic traits associated with tolerating physiochemical stressors, but may also lead to the emergence of reproductive isolating barriers among populations inhabiting extreme and adjacent benign environments (Gross et al., 2004; Lexer & Fay, 2005; Gompert et al., 2006). Hence, like other sources of divergent selection, extreme physiochemical stressors can contribute to both phenotypic evolution and speciation.

Adjacent extreme and benign habitat patches usually exhibit complex environmental differences with suites of ecologically relevant factors covarying between them.

This is particularly true for aspects of the biotic environment, because the presence of physiochemical stressors can impose constraints for the persistence of populations and accordingly act as a filter affecting the composition of biological communities (Belyea & Lancaster, 1999).

Indeed, extreme habitats are usually less productive and characterised by reduced species diversity (McMullin,

Bergquist & Fisher, 2000; Tsurumi, 2003). Colonisation of extreme environments by populations tolerant of physiochemical stressors may therefore lead to pronounced differences in resource availability and quality, changes in competitive interactions within and between species and changes in exposure to predators and parasites. It follows that phenotypic differentiation between populations in extreme and benign habitats may include not only adaptive traits directly involved in coping with physiochemical stressors, but also traits affected by correlated differences in the biotic conditions, which can create an interactive network of selective forces (Grether et al., 2001; Tobler & Plath, 2011). Acknowledging the complexity of selective regimes in extreme environments is critical to elucidate potential cause and effect relationships between environmental variation and the evolution of complex phenotypes (Kaeuffer et al., 2012).