Molecular Ecology (2015) 24, 525–544 doi: 10.1111/mec.13048 INVITED REVIEWS AND SYNTHESES Species are hypotheses: avoid connectivity assessments based on pillars of sand 1 1 1 ERIC PANTE,* NICOLAS PUILLANDRE,† AMELIA VIRICEL,* SOPHIE ARNAUD-HAOND,‡ DIDIER AURELLE,§ MAGALIE CASTELIN,¶ ANNE CHENUIL,§ CHRISTOPHE DESTOMBE,**†† DIDIER FORCIOLI,‡‡§§ MYRIAM VALERO,**†† FREDERIQUE VIARD**¶¶ and SARAH SAMADI† *Littoral, Environnement et Societes (LIENSs), UMR 7266 CNRS - Universite de La Rochelle, 2 rue Olympe de Gouges, 17042 La Rochelle, France, †ISYEB – UMR 7205 – CNRS, MNHN, UPMC (University Paris 06), EPHE – Museum national d’Histoire naturelle, Sorbonne Universites, CP26, 57 rue Cuvier, F-75231 Paris Cedex 05, France, ‡IFREMER, UMR 212 Ecosystemes marins Exploites, F-34203 Sete, France, §Aix Marseille Universite, CNRS, IRD, Avignon Universite, IMBE UMR 7263, 13397 Marseille, France, ¶Aquatic Animal Health Section, Fisheries and Oceans Canada, Pacific Biological Station, 3190 Hammond Bay Road, Nanaimo BC, Canada, V9T 6N7, **Sorbonne Universites, UPMC, University Paris 06, Station Biologique de Roscoff, F-29680 Roscoff, France, ††CNRS, Laboratory Evolutionary Biology and Ecology of Algae, Sorbonne Universites, Universite Pierre et Marie Curie (UPMC) Univ Paris 06, UMI 3614, UPMC, PUCCh, UACh, Station Biologique de Roscoff, F-29680 Roscoff, France, ‡‡Faculte des Sciences, Universite Nice-Sophia-Antipolis, Equipe Symbiose Marine UMR 7138, Parc Valrose, 06108 Nice Cedex 2, France, §§UMR 7138 Evolution Paris Seine, Universite Pierre et Marie Curie – CNRS, 7 Quai St Bernard, 75252 Paris Cedex 05, France, ¶¶Centre National de la Recherche Scientifique (CNRS), Laboratory Adaptation and Diversity in the Marine Environment, Team Diversity and Connectivity in Coastal Marine Landscapes, UMR 7144, Station Biologique de Roscoff, F-29680 Roscoff, France Abstract Connectivity among populations determines the dynamics and evolution of populations, and its assessment is essential in ecology in general and in conservation biology in particu- lar. The robust basis of any ecological study is the accurate delimitation of evolutionary units, such as populations, metapopulations and species. Yet a disconnect still persists between the work of taxonomists describing species as working hypotheses and the use of species delimitation by molecular ecologists interested in describing patterns of gene flow. This problem is particularly acute in the marine environment where the inventory of bio- diversity is relatively delayed, while for the past two decades, molecular studies have shown a high prevalence of cryptic species. In this study, we illustrate, based on marine case studies, how the failure to recognize boundaries of evolutionary-relevant unit leads to heavily biased estimates of connectivity. We review the conceptual framework within which species delimitation can be formalized as falsifiable hypotheses and show how con- nectivity studies can feed integrative taxonomic work and vice versa. Finally, we suggest strategies for spatial, temporal and phylogenetic sampling to reduce the probability of inadequately delimiting evolutionary units when engaging in connectivity studies. Keywords: connectivity, marine organisms, molecular systematics, taxonomy Received 30 September 2014; revision received 6 December 2014; accepted 13 December 2014 A Scopus search in November 2014 revealed 2137 Context and problems documents published since 1991, including 1776 Population connectivity has been the subject of an during the last 5 years. Applications of connectivity increasing number of studies for the last two decades. studies include the following: (i) identifying geneti- cally isolated populations of protected or endangered Correspondence: Sarah Samadi, Fax: +33 1 40 79 38 44; species that should be monitored as separate conser- E-mail: [email protected] vation units (e.g. Palsbøll et al. 2007), (ii) optimizing 1Authors with equal contributions. the size and location of protected areas to create © 2014 John Wiley & Sons Ltd 526 E. PANTE ET AL. well-connected reserve networks (e.g. Kritzer & Sale dant, commercially important or ecologically well-stud- 2004; Jones et al. 2007), (iii) identifying stocks for fisheries ied marine taxa, species were often described long ago, management (e.g. Fogarty & Botsford 2007; Waples et al. but are not necessarily adequately delineated (e.g. 2008; Reiss et al. 2009) and (iv) evaluating the potential Uthicke et al. 2010; Jaafar et al. 2012; Mantelatto et al. impacts of resource exploitation on population dynamics 2014; Thomas et al. 2014). For instance, Ciona intestinalis of local communities (e.g. Bors et al. 2012; Plouviez et al. (Linnaeus, 1767) is a model organism in evolutionary 2013). Besides these important applications to manage- developmental biology and was among the first ani- ment and conservation, connectivity studies can also help mals to have its genome fully sequenced (Dehal et al. addressing long-standing questions such as the status of 2002). However, this taxon was recently revealed as a marine populations as open or closed systems (e.g. Co- complex of four cryptic species (Suzuki et al. 2005; Cap- wen et al. 2000), the identification of factors that create uti et al. 2007; Nydam & Harrison 2007; Zhan et al. and maintain genetic differentiation (e.g. Bilton et al. 2010), including two widespread species that diverged 2002; Shank 2010) or understanding of how local adapta- ca. 3–4 Ma ago (Roux et al. 2013) and are not yet for- tion can occur in high gene flow species (e.g. Nielsen mally named. Knowledge gaps for marine organisms et al. 2009). are so important (e.g. Knowlton 2000; Webb et al. 2010) In this review, we underline the paramount impor- that, beyond the discovery of recently separated lin- tance of establishing the state of taxonomic treatment eages (i.e. the ‘emergent phylogenetic systems’ defined and available background knowledge on the biology of in Carstens et al. 2013), taxonomic confusions at higher organisms when engaging in connectivity assessments. rank may also impact the definition of species. For These aspects are sometimes overlooked despite the fact instance in 2009, Johnson et al. revealed that three fami- that they should condition the sampling design and the lies of deep-sea fishes artificially separated juveniles, inferences made from genetic data. One of the main males and females. This confusion was resolved by considerations we develop is that properly estimating examining the morphology of additional specimens and connectivity at the population level requires assessing using molecular data. This result greatly changes the the robustness of accepted taxonomic hypotheses. This understanding of connectivity in terms of life-history is not trivial as for many taxa, taxonomic knowledge is traits and ecology for these taxa. inadequate or even lacking. Indeed, a great portion of The main objective of this review was thus to stress the world’s biodiversity remains to be described, partic- the fact that, when conducting connectivity assessments, ularly in underexplored, difficult-to-access habitats such taxonomic status should be critically reassessed, and as tropical rainforests and the marine environment revised if necessary, in the light of newly acquired bio- (Appeltans et al. 2012). In addition, in these habitats, logical data. This point is particularly crucial in marine species descriptions are often based on few specimens, systems, which are the focus of this review, both and species distribution ranges as well as ecological because the marine biota accumulates the greatest num- requirements are poorly known. ber of deep evolutionary lineages, but also because it The development of DNA sequencing techniques has remains vastly undersampled (Costello et al. 2010; Not considerably accelerated the rate of discovery and the et al. 2012). We outline that the population genetics data documentation of species distributions. This is notably collected for estimating connectivity may in turn be true in the marine realm, especially for poorly studied used to refine taxonomic knowledge at the species rank. eukaryotic phyla for which few or no other characters Cryptic species are typically revealed by population are available as reference for taxonomic delimitations. genetic studies, phylogeographic analyses or barcoding For example, analyses of DNA characters revealed that (e.g. Uthicke et al. 2010; Jaafar et al. 2012; Mantelatto macro- and micro-algae include many cryptic lineages et al. 2014; Thomas et al. 2014). In fact, the patterns that are subject to convergent morphological evolution emerging from connectivity and species delimitation (Leliaert et al. 2014). However, turning DNA-based dis- studies belong to the same divergence continuum coveries into robustly and formally named taxonomic because population and species divergence result from entities remains challenging (Satler et al. 2013). Conse- common ecological and micro-evolutionary processes. quently, the literature includes a large number of unde- First, we illustrate the importance of accurate taxonomic scribed and un-named cryptic or pseudo-cryptic species evaluation in the analysis of connectivity by examining for which identification from one study to another is one specific question: the correlation between pelagic uneasy or impossible (e.g. Pante et al. 2015). As recently larval duration (PLD) and genetic structure estimates. underlined by Fontaine et al. (2012), the time lapse We then highlight the main issues in relation to
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages20 Page
-
File Size-