• Nenhum resultado encontrado

Article

published

in

Molecular

Ecology,

2019,

28,

4573-4591.

doi:

https://doi.org/10.1111/mec.15249

1 Departamento de Biologia da Faculdade de Ciências da Universidade do Porto, Rua Campo

Alegre, 4169-007 Porto, Portugal.

2 CIBIO/InBIO, Centro de Investigacao em Biodiversidade e Recursos Genéticos da

Universidade do Porto, Instituto de Ciências Agrárias de Vairão, Rua Padre Armando Quintas 7, 4485-661 Vairão, Portugal

3 Department of Zoology and Entomology, School of Biological and Environmental Sciences,

University of Fort Hare, Private Bag X1314, Alice 5700, South Africa

4 Department of Environmental Science, Policy and Management, University of California, 130

4.1 – Abstract

Evolutionary changes in reproductive mode may affect co‐evolving traits, such as dispersal, although this subject remains largely underexplored. The shift from aquatic oviparous or larviparous reproduction to terrestrial viviparous reproduction in some amphibians entails skipping the aquatic larval stage and, thus, greater independence from water. Accordingly, amphibians exhibiting terrestrial viviparous reproduction may potentially disperse across a wider variety of suboptimal habitats and increase population connectivity in fragmented landscapes compared to aquatic‐breeding species. We investigated this hypothesis in the fire salamander (Salamandra salamandra), which exhibits both aquatic‐ (larviparity) and terrestrial‐ breeding (viviparity) strategies. We genotyped 426 larviparous and 360 viviparous adult salamanders for 13 microsatellite loci and sequenced a mitochondrial marker for 133 larviparous and 119 viviparous individuals to compare population connectivity and landscape resistance to gene flow within a landscape genetics framework. Contrary to our predictions, viviparous populations exhibited greater differentiation and reduced genetic connectivity compared to larviparous populations. Landscape genetic analyses indicate viviparity may be partially responsible for this pattern, as water courses comprised a significant barrier only in viviparous salamanders, probably due to their fully terrestrial life cycle. Agricultural areas and, to a lesser extent, topography also decreased genetic connectivity in both larviparous and viviparous populations. This study is one of very few to explicitly demonstrate the evolution of a derived reproductive mode affects patterns of genetic connectivity. Our findings open avenues for future research to better understand the eco‐evolutionary implications underlying the emergence of terrestrial reproduction in amphibians.

Keywords: genetic structure, haplotypes, landscape genetics, larviparity, pueriparity,

Salamandra salamandra.

4.2 – Introduction

The evolution of derived phenotypic traits enables individuals to exploit novel resources and colonise new areas, often entailing profound eco-evolutionary implications to taxa (Losos 2010). One remarkable life-history adaptation is the transition from egg-laying (oviparous) reproduction to live-bearing (viviparous) reproduction, which occurred more than 150 times in vertebrates (mostly in reptiles) and involved major phenotypic, genetic, and ecological changes, especially in females (e.g. Pincheira-Donoso et al. 2013; Blackburn 2015; Wake 2015; Helmstetter et al. 2016; Halliwell et al. 2017; Gao et al. 2019). Strong environmental

pressures on offspring (e.g. stressful environmental conditions or predation) generally selected for longer periods of embryo retention (i.e. viviparity) to increase offspring survival rates, thus allowing viviparous taxa to thrive in harsher environments and disperse to areas previously inaccessible (e.g. Pincheira-Donoso et al. 2013; Helmstetter et al. 2016; Ma et al. 2018).

The phenotypic and ecological changes underlying the evolution of viviparity may also affect co‐evolving traits, such as dispersal. This is because the dispersal ecology of organisms is intimately linked with reproductive biology, as it comprises a key mechanism for finding mates and breeding sites (Bonte et al. 2012; Pittman et al. 2014; Cosgrove et al. 2018). Additionally, individuals must also adjust dispersal decisions and pathways in accordance with environmental conditions encountered during the dispersal process to increase reproductive success (Bonte et al. 2012). Hence, the changes in reproductive biology and behaviour entailed by the evolution of viviparity can alter the ways individuals interact with the surrounding environment (Shine 2015), which in turn may affect overall patterns of dispersal, gene flow, and population dynamics. However, the effects of reproductive mode on dispersal ecology remain largely underexplored.

For amphibians in particular, shifts to viviparous or pueriparous reproduction (hereafter we use “pueriparous” for amphibians; see Greven 2003) result in significant life‐history changes, making amphibians good systems in which to examine the effects of changes in reproductive mode on dispersal and genetic connectivity. Most amphibians exhibit a biphasic life cycle, in which an aquatic larval stage is followed by metamorphosis into terrestrial juveniles (Wells 2007). In aquatic‐breeding amphibians, dispersal behaviour and success are largely driven by the quality and availability of aquatic breeding sites for the deposition and development of offspring (Pittman et al. 2014). However, some amphibians shifted from ancestral oviparous or larviparous aquatic reproduction (delivery of eggs or larvae in water, respectively) to pueriparous terrestrial reproduction (parturition of juveniles), possibly in response to a lack of suitable water bodies in their environments for depositing offspring (Velo‐Antón et al. 2015; Liedtke et al. 2017). Their fully terrestrial lifestyle enables them to cope better with the challenges imposed by water‐limited environments and to potentially disperse successfully across a wider variety of unsuitable habitats in comparison to aquatic‐breeding amphibians (Liedtke et al. 2017; Lourenço et al. 2017).

Previous studies in aquatic‐ and terrestrial‐breeding amphibians (including pueriparous and direct‐developing species) have indeed suggested a lower dependency on water may reduce population genetic divergence in heterogeneous and fragmented landscapes (Measey et al. 2007; Mims et al. 2015; Sandberger‐Loua et al. 2018), although other studies on direct‐ developing amphibians have reported substantial levels of genetic differentiation among

populations (e.g. Peterman et al. 2014a; Paz et al. 2015). Comparative studies involving species showing intraspecific variation in reproductive modes are crucial (e.g. aquatic vs. terrestrial reproduction), as they can overcome the potentially confounding factors involved in comparisons of species with pronounced phenotypic and ecological differences (e.g. Garcia

et al. 2017; Hendrix et al. 2017).

Here, we examine patterns of gene flow among populations of the fire salamander (Salamandra salamandra, Linnaeus 1758), which, with its sister species, the North‐African fire salamander (S. algira, Bedriaga 1883), is one of the only known amphibians exhibiting both aquatic and terrestrial reproduction (Velo‐Antón et al. 2015; Dinis and Velo‐Antón 2017).

Salamandra salamandra exhibits two reproductive strategies: larviparity, in which females

deliver up to ca. 90 larvae in water bodies after a gestation period of approximately 90 days; and pueriparity, in which the larval aquatic stage is skipped and females deliver 1–35 fully metamorphosed terrestrial juveniles after the same gestation period (Buckley et al. 2007; Velo‐ Antón et al. 2015). The ancestral reproductive mode, larviparity, is present throughout most of its range, while pueriparity is currently restricted to a section of northern Spain in the subspecies S. s. bernardezi and S. s. fastuosa (Figure 4.1A; Velo‐Antón et al. 2015). Pueriparity probably arose in S. s. bernardezi in the Cantabrian Mountains during the Pleistocene, possibly in response to the lack of surface water in karstic limestone substrates (García‐París et al. 2003). This trait later introgressed eastwards with S. s. fastuosa during population expansions following cycles of warm and cold climates (García‐París et al. 2003). More recently (Holocene, <8 kya), pueriparity has also emerged independently in two insular populations of S. s. gallaica in north‐western Spain (Velo‐Antón et al. 2007; Velo‐Antón et al. 2012).

A previous study contrasting patterns of fine‐scale genetic structure and dispersal between larviparous and pueriparous fire salamanders on intact, natural landscapes (1-km transects) did not find significant differences in spatial genetic autocorrelation between reproductive modes (Lourenço et al. 2018a). However, whether such patterns hold at broader scales in heterogeneous and fragmented landscapes remains unknown. For instance, water bodies may be important dispersal corridors for larviparous populations, while water‐limited habitats may be less resistant to dispersal for pueriparous salamanders, as they can survive in harsher environments (Lourenço et al. 2017). The field of landscape genetics has comprised a useful analytical framework for inferring the potential role of phenotypic traits and environmental heterogeneity in maintaining genetic connectivity among different taxa (e.g. Richardson 2012; Manel and Holderegger 2013; Garcia et al. 2017). These approaches may help us to determine whether a shift to pueriparity altered the way fire salamanders disperse across the landscape.

Here, we use a mitochondrial marker and a set of nuclear microsatellite loci in a comparative landscape genetics framework to: (i) characterise and evaluate differences in patterns of genetic diversity and structure between pueriparous and larviparous populations; and (ii) identify the environmental variables governing genetic connectivity in populations exhibiting different reproductive strategies. Because pueriparous individuals are expected to survive and disperse across a greater range of habitats, we expect to observe reduced genetic structure and higher genetic connectivity among pueriparous populations compared to their larviparous counterparts across two fragmented and heterogeneous focal landscapes.

Documentos relacionados