Parallel Habitat-Driven Differences in the Phylogeographical Structure of Two Independent Lineages of Mediterranean Saline Water Beetles
Total Page:16
File Type:pdf, Size:1020Kb
Molecular Ecology (2009) 18, 3885–3902 doi: 10.1111/j.1365-294X.2009.04319.x Parallel habitat-driven differences in the phylogeographical structure of two independent lineages of Mediterranean saline water beetles P. ABELLA´ N,*† A. MILLA´ N‡ and I. RIBERA*† *Departamento de Biodiversidad y Biologı´a Evolutiva, Museo Nacional de Ciencias Naturales (CSIC), Jose´ Gutie´rrez Abascal 2, 28006 Madrid, Spain, †Instituto de Biologı´a Evolutiva (UPF-CSIC), Passeig Maritim de la Barceloneta 37-49, 08003 Barcelona, Spain, ‡Departamento de Ecologı´a e Hidrologı´a, Universidad de Murcia, Campus de Espinardo, 30100, Murcia, Spain Abstract It has been hypothesized that species living in small lentic water bodies, because of the short-term geological persistence of their habitat, should show higher dispersal ability, with increased gene flow among populations and a less pronounced phylogeographical structure. Conversely, lotic species, living in more geologically stable habitats, should show reduced dispersal and an increased phylogeographical structure at the same geographical scales. In this work we tested the influence of habitat type in two groups of aquatic Coleoptera (Nebrioporus ceresyi and Ochthebius notabilis groups, families Dytiscidae and Hydraenidae respectively), each of them with closely related species typical of lotic and lentic saline Western Mediterranean water bodies. We used mitochondrial cox1 sequence data of 453 specimens of 77 populations through the range of nine species to compare a lotic vs. a lentic lineage in each of the two groups. Despite the differences in biology (predators vs. detritivorous) and evolutionary history, in both lotic lineages there was a higher proportion of nucleotide diversity among than within groups of populations, and a faster rate of accumulation of haplotype diversity (as measured by rarefaction curves) than in the lentic lineages. Similarly, lotic lineages had a higher absolute phylogenetic diversity, more remarkable considering their smaller absolute geographical ranges. By comparing closely related species, we were able to show the effect of contrasting habitat preferences in two different groups, in agreement with predictions derived from habitat stability. Keywords: aquatic Coleoptera, habitat constraints, habitat stability, nucleotide diversity, phylo- geographical structure Received 25 February 2009; revision received 1 July 2009; accepted 7 July 2009 these two types of water bodies. Most species inhabit- Introduction ing aquatic environments are specialized to one of The habitat determines the spatial matrix in which these habitat types, and these differences in habitat species persist, constraining many aspects of species’ preference occur at all phylogenetic levels, often ecology and affecting biogeography and population among sister species, or groups of closely related spe- structure (Southwood 1977, 1988; Korfiatis & Stamou cies (Illies 1978). Both habitat types differ in their 1999). In freshwater habitats, a particularly important long-term persistence: small lentic water bodies, typi- ecological difference is whether they are standing (len- cal of many macroinvertebrates, tend to fill with sedi- tic) or running (lotic), with a wide range of physical ment over a time period of decades or centuries, and biological characteristics associated to each of whereas rivers and streams persist over geologically defined periods, even if their exact location may Correspondence: Pedro Abella´n, Fax: +34 932309555; change or if they dry up seasonally (Hutchinson E-mail: [email protected] 1957). Ó 2009 Blackwell Publishing Ltd 3886 P. ABELLA´ N, A. MILLA´ N and I. RIBERA The contrasting habitat permanence of lotic and small saline water bodies in the Western Mediterranean, the lentic water bodies is predicted to have a cascade of Ochthebius notabilis and Nebrioporus ceresyi groups (fami- ecological and evolutionary implications ranging from lies Hydraenidae and Dytiscidae respectively, Ja¨ch 1992; the individual to the lineage (see Ribera 2008 for a Fery et al. 1996), each of them including closely related review). Critically, species living in geologically long- species exclusive of either lotic or lentic saline habitats. lived lotic habitats will be expected to have reduced Both species groups have independently colonized dispersal ability compared with closely related species inland saline or hyper-saline waters and they are often living in geologically more ephemeral lentic habitats. It found together, although only very rarely more than must be stressed that ‘dispersal’ is understood here as one species of each group is found in the same place. the ability of a species to establish a new population in More specifically, we ask whether (i) lotic species show a non-contiguous habitat patch (i.e. emigration plus more geographical structuring than closely related lentic inter-patch movement plus immigration, in the termi- species over comparable geographical scales, and (ii) in nology of Bowler & Benton 2005), not the ability of an lotic species, genetic variability is preferentially distrib- individual to move in space. Usual correlates of ‘dis- uted among populations, whereas in lentic species, it is persal’ only measure the ability to move, which is not more evenly distributed among and within populations. necessarily equivalent to dispersal in the sense used It is important to note that we do not claim that habitat here (Stenseth & Lidicker 1992). The higher mobility persistence is the only factor determining genetic struc- and colonization ability of populations of lentic species ture, nor even the most influential one: when compar- is expected to lead to more dynamic and larger geo- ing species with large differences in ecology, biology or graphical ranges, increased gene flow between popula- evolutionary background, the expectation is to find tions and reduced inter-population genetic variability, wide differences in population structure, even if all of leading to a greater evolutionary cohesion of species. them occupy the same physical space (as shown by e.g. Over comparable geographical scales, lotic species Short & Caterino 2009 with three species of three differ- would have a more pronounced phylogeographical ent families of water beetles). By comparing closely structure, with a higher inter-population diversity, related species, we try to homogenize confounding fac- stronger geographical association and an increased tors other than exclusively the habitat difference (the probability of peripatric or allopatric speciation than principle behind the sister group comparison, Harvey & lentic species (Marten et al. 2006; Ribera 2008). Pagel 1991; Barraclough et al. 1998). The generality of Previous results have established the relationship the results is tested through the study of two different between habitat type, size of the geographical range lineages of closely related species: if both show parallel and the shape of latitudinal gradients in a number of differences associated to the same habitat type, despite freshwater invertebrates (Ribera & Vogler 2000; Ribera large differences in evolutionary history, biology and et al. 2003; Hof et al. 2006, 2008; Ribera 2008). The rela- ecology, this will add strong evidence to support the tionship between habitat type and phylogeographical role of habitat type as a constraint of phylogeographical structure has received less attention. Using allozyme structure. data on more than 150 species of freshwater molluscs, Saline inland waters seem suitable systems for testing insects and crustaceans Marten et al. (2006) found that the effect of differences in habitat permanence on lentic invertebrates exhibit on average lower genetic phylogeographical patterns, as the spatial matrix of population differentiation than lotic species, in a consis- habitat patches represent clearly circumscribed localities tent way across the three studied groups. More in an otherwise diffusely structured landscape. The recently, Papadopoulou et al. (2008) explored with sim- phylogeographical information of organisms living in ulations how the migration rate might affect variation saline environments is scarce and limited to a few spe- in the coalescence of mitochondrial DNA (mtDNA) in cies. Prior molecular analyses have reported deep taxa occupying habitat types that differ in their stability. phylogeographical structure in both planktonic Rotifer Using one aquatic and one terrestrial genus of beetles, and brine shrimp species from Iberian salt lakes and they show that lineages in more stable habitats had salt-pans (Go´mez et al. 2000; Mun˜oz et al. 2008), as well greater levels of population sub-division and geographi- as in one of the species of the O. notabilis group, Ochthe- cal structure, a result confirmed for the terrestrial bius glaber, in SE Spain (Abella´n et al. 2007). Phylogeo- beetles in a subsequent, more detailed study (Papado- graphical analyses revealed a surprisingly high degree poulou et al. 2009). of geographical structure in these species, detectable Here we aim to test more thoroughly the effect of among populations separated by relatively short geo- contrasting habitat preferences on the genetic diversity graphical distances. Our paper will also contribute to and phylogeographical structure of populations. We use the knowledge of the origin of the saline inland water two phylogenetically distant lineages of beetles living in biota in the west Mediterranean, a largely neglected Ó 2009 Blackwell Publishing Ltd HABITAT STABILITY AND GENETIC STRUCTURE