Molecular Ecology (2015) doi: 10.1111/mec.13094 SPECIAL ISSUE: INVASION GENETICS: THE BAKER AND STEBBINS LEGACY Relationship between invasion success and colony breeding structure in a subterranean termite 1 E. PERDEREAU,* A.-G. BAGNERES,* E.L. VARGO,† G. BAUDOUIN,* Y. XU,‡ P. LABADIE,† S. DUPONT* and F. DEDEINE* *Institut de Recherche sur la Biologie de l’Insecte UMR 7261 CNRS - Universite Francßois-Rabelais,UFR Sciences, Parc Grandmont, Tours 37200, France, †Department of Entomology and W.M. Keck Center for Behavioral Biology, Box 7613, North Carolina State University, Raleigh NC 27695-7613, USA, ‡Department of Entomology, South China Agricultural University, Guangzhou 510642, China Abstract Factors promoting the establishment and colonization success of introduced popula- tions in new environments constitute an important issue in biological invasions. In this context, the respective role of pre-adaptation and evolutionary changes during the invasion process is a key question that requires particular attention. This study com- pared the colony breeding structure (i.e. number and relatedness among reproductives within colonies) in native and introduced populations of the subterranean pest termite, Reticulitermes flavipes. We generated and analysed a data set of both microsatellite and mtDNA loci on termite samples collected in three introduced populations, one in France and two in Chile, and in the putative source population of French and Chilean infestations that has recently been identified in New Orleans, LA. We also provided a synthesis combining our results with those of previous studies to obtain a global pic- ture of the variation in breeding structure in this species. Whereas most native US populations are mainly composed of colonies headed by monogamous pairs of primary reproductives, all introduced populations exhibit a particular colony breeding structure that is characterized by hundreds of inbreeding reproductives (neotenics) and by a propensity of colonies to fuse, a pattern shared uniquely with the population of New Orleans. These characteristics are comparable to those of many invasive ants and are discussed to play an important role during the invasion process. Our finding that the New Orleans population exhibits the same breeding structure as its related introduced populations suggests that this native population is pre-adapted to invade new ranges. Keywords: breeding structure, invasion genetics, microsatellites, neoteny, Reticulitermes flavipes, sociality Received 1 October 2014; revision received 9 January 2015; accepted 14 January 2015 although a general understanding of how introduced Introduction populations become established and invade their new Biological invasions involve the introduction of species environments is still lacking (Novak 2007; Bock et al. in regions where they were previously absent, followed this issue). In particular, the ecological and genetic fac- by population growth and range expansion (Sax et al. tors promoting the colonization success of introduced 2005). The negative ecological and economic impacts of populations in new environments constitute a crucial many invasions generated strong research interest, but yet unresolved issue (Baker & Stebbins 1965; Sakai et al. 2001; Lee 2002; Facon et al. 2006). Certain intro- duced populations possess phenotypic and life history Correspondence: Franck Dedeine, Fax: +33(0)247367356; E-mail: [email protected] traits that pre-adapt them to successful colonization of 1Present address: Department of Entomology, Texas A&M Uni- the invaded habitats (Kolar & Lodge 2001; Pysek & versity, 2143 TAMU, College Station, TX 77843-2143, USA Richardson 2007). In contrast, some populations are not © 2015 John Wiley & Sons Ltd 2 E. PERDEREAU ET AL. necessarily pre-adapted to the invaded habitats at the 2009) and some of the most damaging invasive ants: time of introduction, but evolve phenotypic traits that Anoplolepis gracilipes, Linepithema humile, Solenopsis in- facilitate their survival, reproduction and range expan- victa, Lasius neglectus, Pheidole megacephala and Wasmannia sion (Wares et al. 2005; Dlugosch & Parker 2008a; Cola- auropunctata (Morel et al. 1990; Vanloon et al. 1990; Ross utti & Barrett 2013; Vandepitte et al. 2014). Despite the et al. 1996; Espadaler & Rey 2001; Holway et al. 2002; fact that founding event(s) often reduce the genetic Tsutsui & Suarez 2003; Le Breton et al. 2004; Fournier diversity in introduced populations, there is neverthe- et al. 2009; Blight et al. 2012). Interestingly, the introduc- less growing evidence that introduced populations con- tion of social insects is often accompanied by important tain enough additive genetic variance to efficiently modifications of their breeding structure (i.e. the number respond to selection and adapt to the ecological factors and relatedness of reproductives within social units) and present in the invaded habitats (Dlugosch & Parker mode of dispersal allowing colonies to rapidly grow and 2008b; Rollins et al. 2013; Moran & Alexander 2014). attain high densities and occupy large spatial expanses. Natural selection is often considered as the principal Invasive social insects provide promising model systems evolutionary force acting on introduced populations for studying evolutionary changes that occur during (Moran & Alexander 2014), but other mechanisms such invasions as well as for identifying ecological factors that as genetic drift or admixture can also cause important shape the evolution of breeding structures. genetic and phenotypic changes (Suarez et al. 2008). This study aims to compare the colony breeding Although the study of biological invasions is clearly structure of native and introduced populations of an important from an applied perspective to mitigate invasive pest termite, Reticulitermes flavipes (Rhinoter- impacts and prevent future introductions, Baker & Steb- mitidae), in order to test the hypothesis that the evolu- bins (1965) were among the first authors to develop the tion of the breeding structure following an introduction idea that invasive species provide unique model sys- event is not limited to the social Hymenoptera but also tems for understanding micro-evolutionary mechanisms occurs in the Isoptera. R. flavipes is a subterranean ter- and for identifying the ecological factors that shape mite that lives in forest ecosystems and in urban areas complex phenotypic traits (Barrett this issue). where it can cause significant damage to human-built The evolution of introduced populations, more espe- wooden structures. Originating from the US, this spe- cially adaptive evolution, has recently been considered cies is widespread in the eastern part of the US from an important component leading to invasiveness (Cola- Texas to Nebraska in the west and from Florida to Mas- utti & Barrett 2013; Blows & McGuigan this issue), sachusetts in the east. This species has been accidently although empirical evidence of adaptations in invading introduced by humans to several areas both in the populations at a molecular level remains very limited Americas and in Europe. In the Americas, R. flavipes (Vandepitte et al. 2014; Hodgins et al. this issue). To has been found in California, the Bahamas, Canada, detect evolutionary change during invasions, compara- Chile and Uruguay (Ripa & Castro 2000; Austin et al. tive analyses of source versus introduced populations 2002, 2005; Su et al. 2006; Scaduto et al. 2012). In have been conducted. So far, most of these comparative Europe, R. flavipes is today widespread throughout studies included measures of neutral genetic variation, France and has been locally found in Austria (Vienna), phenotypic variation and, more rarely, quantitative Germany (Hamburg) and Italy (Varese) (Kollar 1837; genetic variation and phenotypic plasticity (Geng et al. Weidner 1937; Ghesini et al. 2010). Recently, a global 2007; Suarez et al. 2008). These studies principally used genetic analysis using R. flavipes samples collected in invasive plants as model systems (Bossdorf et al. 2005; the native and introduced ranges revealed that the most van Kleunen et al. this issue), although other taxa likely source of French populations is located in the among which fungi (Gladieux et al. this issue) or region of New Orleans, Louisiana (Perdereau et al. insects, and more particularly invasive social insects, 2013). This study also suggests that French introduc- have proved useful. tions occurred directly from Louisiana and that the first Social insects count among the most successful invad- introduction(s) started during the 18th century at the ers, often causing severe damage to local communities earliest (i.e. period of first trade between New Orleans and ecosystems (Moller 1996). The colonization success and France). In addition, French populations showed of social insects has been largely attributed to their com- lower genetic diversity than US populations, suggesting plex and flexible social organization (Moller 1996; Chap- that founding event(s) resulted in genetic bottlenecks man & Bourke 2001; Holway et al. 2002). There has been (Perdereau et al. 2013). Finally, this global analysis indi- considerable empirical work on various social Hymenop- cates that the population of New Orleans might also be tera including wasps (Vespula spp.) (Donovan et al. 1992; the source of Chilean populations (Perdereau et al. Kasper et al. 2008; Hanna et al. 2014), bumble bees (Bom- 2013), a hypothesis that recently has received support bus terrestris) (Buttermore 1997; Nagamitsu