Geropogon Hybridus (L.) Sch.Bip
Total Page：16
File Type：pdf, Size：1020Kb
Mathematisch-Naturwissenschaftliche Fakultät Christina M. Müller | Benjamin Schulz | Daniel Lauterbach | Michael Ristow | Volker Wissemann | Birgit Gemeinholzer Geropogon hybridus (L.) Sch.Bip. (Asteraceae) exhibits micro-geographic genetic divergence at ecological range limits along a steep precipitation gradient Suggested citation referring to the original publication: Plant Systematics and Evolution 303 (2017) 91–104 DOI https://doi.org/10.1007/s00606-016-1354-y ISSN (print) 0378-2697 ISSN (online) 1615-6110 Postprint archived at the Institutional Repository of the Potsdam University in: Postprints der Universität Potsdam Mathematisch-Naturwissenschaftliche Reihe ; 832 ISSN 1866-8372 https://nbn-resolving.org/urn:nbn:de:kobv:517-opus4-427061 DOI https://doi.org/10.25932/publishup-42706 Plant Syst Evol (2017) 303:91–104 DOI 10.1007/s00606-016-1354-y ORIGINAL ARTICLE Geropogon hybridus (L.) Sch.Bip. (Asteraceae) exhibits micro-geographic genetic divergence at ecological range limits along a steep precipitation gradient 1 2 3 4 Christina M. Mu¨ller • Benjamin Schulz • Daniel Lauterbach • Michael Ristow • 1 1 Volker Wissemann • Birgit Gemeinholzer Received: 11 March 2016 / Accepted: 9 September 2016 / Published online: 31 October 2016 Ó The Author(s) 2016. This article is published with open access at Springerlink.com Abstract We analyzed the population genetic pattern of the study area, which indicates that reduced precipitation 12 fragmented Geropogon hybridus ecological range edge toward range edge leads to population genetic divergence. populations in Israel along a steep precipitation gradient. In However, this pattern diminished when the hypothesized the investigation area (45 9 20 km2), the annual mean gene flow barrier was taken into account. Applying the precipitation changes rapidly from 450 mm in the north spatial analysis method revealed 11 outlier loci that were (Mediterranean-influenced climate zone) to 300 mm in the correlated to annual precipitation and, moreover, were south (semiarid climate zone) without significant temper- indicative for putative precipitation-related adaptation ature changes. Our analysis (91 individuals, 12 popula- (BAYESCAN, MCHEZA). The results suggest that even tions, 123 polymorphic loci) revealed strongly structured on micro-geographic scales, environmental factors play populations (AMOVA UST = 0.35; P \ 0.001); however, prominent roles in population divergence, genetic drift, and differentiation did not change gradually toward range edge. directional selection. The pattern is typical for strong IBD was significant (Mantel test r = 0.81; P = 0.001) and environmental gradients, e.g., at species range edges and derived from sharply divided groups between the north- ecological limits, and if gene flow barriers and mosaic-like ernmost populations and the others further south, due to structures of fragmented habitats hamper dispersal. dispersal or environmental limitations. This was corrobo- rated by the PCA and STRUCTURE analyses. IBD and Keywords Environmental association studies Á IBE were significant despite the micro-geographic scale of Fragmented habitats Á Isolation by distance (IBD) Á Isolation by environment (IBE) Á Range edge populations Handling editor: Christoph Oberprieler. Electronic supplementary material The online version of this Introduction article (doi:10.1007/s00606-016-1354-y) contains supplementary material, which is available to authorized users. Species’ distribution ranges are commonly defined by & Birgit Gemeinholzer ecological limits, which are most often determined by [email protected] ecological gradients. As species approach their range lim- 1 its, their populations typically become smaller and more Systematic Botany, Justus-Liebig-University Giessen, fragmented (Bridle and Vines 2007). As a consequence, Heinrich-Buff-Ring 38, 35392 Giessen, Germany 2 range edge (margins or verge) populations often feature Institute of Landscape Ecology and Resource Management, decreased genetic diversity due to reduced gene flow as Interdisciplinary Research Centre (IFZ), Justus-Liebig- University Giessen, Heinrich-Buff-Ring 26-32, result of the more fragmented distribution or random 35393 Giessen, Germany genetic drift effects enhance genetic diversity due to neu- 3 Botanical Garden, University of Potsdam, Maulbeerallee 2a, tral, negative, or positive mutations which can appear and 14469 Potsdam, Germany increase in frequency over time (e.g., Ellstrand and Elam 4 Plant Ecology and Nature Conservation, University of 1993). If gene flow patterns align with geographic distance Potsdam, Maulbeerallee 3, 14469 Potsdam, Germany (isolation by distance, IBD; Wright 1943), this affects the 123 92 C. M. Mu¨ller et al. distribution of variation within and among range edge geographic distances to analyze IBD, while more recent populations (Eckert et al. 2008; Sexton et al. 2009). If studies also incorporate multiple environmental variables ecological gradients cause range limits, populations are with a main focus on environmental gradients (e.g., cli- generally more likely to maintain specialized genotypes matic-, elevation-, environmental- and habitat gradients) that are well adapted to particular ecological conditions with climatic variables being the most commonly used (e.g., Rehm et al. 2015). This is especially the case when (Gerber et al. 2004; Nahum et al. 2008; Nakazato et al. populations that are geographically closer together are 2008; Manel et al. 2012; Jones et al. 2013; Gray et al. 2014; genetically more similar than populations that are further Harter et al. 2015). Environmental association studies by apart (IBD) and where genetic and environmental differ- means of climatic variables generally comprise tempera- ences among populations (isolation by environment, IBE) ture and precipitation that over large geographic distances are positively correlated. Especially in smaller and more normally change simultaneously. Thus, it is often impos- fragmented range edge populations, positive, advantageous sible to detangle one factor from another to determine the mutations are more likely to become fixed, as directional exact driving forces for putative adaptation (Linhart and selection within and among range edge populations might Grant 1996). As a consequence, only a few studies could be stronger than in central populations, where stabilizing detect outlier loci in non-model plant species. In Cam- selection through higher rates of gene flow tend to oppose panula (Jones et al. 2013) and Cotinus (Lei et al. 2015), effects of local selection and therefore limits adaptation these outlier loci are clearly associated with precipitation, (Hoffmann and Blows 1994; Lenormand 2002; Bridle and while Manel et al. (2012) in alpine plant species, Gray et al. Vines 2007; Sexton et al. 2009). Under an IBD and IBE (2014)inAndropogon, and Hu¨bner et al. (2009)inHor- pattern, range edge populations eventually diversify and deum showed that precipitations in combination with undergo niche evolution during adaptation to novel envi- temperature were the best environmental predictors. ronments, or alternatively may depauperate where adapta- Here, we investigated the genetic diversity and structure tion is prevented by small population size (Sexton et al. of range edge populations of the annual Geropogon 2009). hybridus (L.) Sch.Bip. along a steep precipitation gradient For decades, evolutionary ecologists have investigated (450–300 mm) on a micro-geographic scale (45 km) local adaptation across different systems and scales, how- without any significant temperature change. Using ampli- ever, rarely on micro-geographic scales. This is due to the fied fragment length polymorphism (AFLP) analysis (Vos assumption that high rates of gene flow prevent adaptive et al. 1995), we asked the following questions: (1) Does divergence at fine spatial grains (Richardson et al. 2014). genetic diversity and differentiation change gradually However, several investigations on small geographic scales toward range edge? (2) Can we identify significant IBD or suggest that micro-geographic divergence is more wide- IBE pattern despite the micro-geographic scale of the study spread than commonly assumed (Kettlewell 1955; Anto- area? (3) Is there putative precipitation-related adaptation novics and Bradshaw 1970; Steiner and Berrang 1990; among the surveyed populations? Kavanagh et al. 2010; Willi and Hoffmann 2012; Krueger- Hadfield et al. 2013; Richardson and Urban 2013; Richardson et al. 2014). Materials and methods Environmental association studies allow the investiga- tion of the nature of local adaptation by identifying its Study species leading causes (Savolainen et al. 2013; Gray et al. 2014). This is achieved by linking genetic variation to environ- Geropogon hybridus is a diploid annual herbaceous mental variables (Manel et al. 2012); however, in non- Asteraceae species, 10–40–(80) cm high. The flowering model species with limited or absent genomic information, stem is erect, usually glabrous with narrowly linear, grass- the identification of adaptive genetic variation can only be like leaves. The pedicule is hollow or swollen below the achieved indirectly. By investigating anonymous loci in capitula. The linear, long involucral bracts often exceed the numerous individuals, it is possible to detect outlier loci of capitulum. The zygomorphic flowers (from March to May) ecological relevance that may be linked to adaptive genes have pink to violet corollas, with dark purple anthers. The (Haldane 1948; Endler