Title: THE HIDDEN SIDE OF INVASIONS: MASSIVE INTROGRESSION BY LOCAL GENES
Abstract: EvolutionVolume 62, Issue 8 p. 1908-1920 Free Access THE HIDDEN SIDE OF INVASIONS: MASSIVE INTROGRESSION BY LOCAL GENES Mathias Currat, Mathias Currat Computational and Molecular Population Genetics Laboratory, Institute of Zoology, University of Bern, Baltzerstrasse 6, 3012 Berne, Switzerland Laboratory of Anthropology, Genetics and Peopling History, Department of Anthropology & Ecology, University of Geneva, 12 rue Gustave-Revilliod, 1227 Carouge, Switzerland E-mail: [email protected] for more papers by this authorManuel Ruedi, Manuel Ruedi Natural History Museum of Geneva, Malagnou 1, 1208 Geneva, SwitzerlandSearch for more papers by this authorRémy J. Petit, Rémy J. Petit INRA, UMR 1202 Biodiversity, Genes & Communities, 69 Route d'Arcachon, F-33610 Cestas, FranceSearch for more papers by this authorLaurent Excoffier, Laurent Excoffier Computational and Molecular Population Genetics Laboratory, Institute of Zoology, University of Bern, Baltzerstrasse 6, 3012 Berne, SwitzerlandSearch for more papers by this author Mathias Currat, Mathias Currat Computational and Molecular Population Genetics Laboratory, Institute of Zoology, University of Bern, Baltzerstrasse 6, 3012 Berne, Switzerland Laboratory of Anthropology, Genetics and Peopling History, Department of Anthropology & Ecology, University of Geneva, 12 rue Gustave-Revilliod, 1227 Carouge, Switzerland E-mail: [email protected] for more papers by this authorManuel Ruedi, Manuel Ruedi Natural History Museum of Geneva, Malagnou 1, 1208 Geneva, SwitzerlandSearch for more papers by this authorRémy J. Petit, Rémy J. Petit INRA, UMR 1202 Biodiversity, Genes & Communities, 69 Route d'Arcachon, F-33610 Cestas, FranceSearch for more papers by this authorLaurent Excoffier, Laurent Excoffier Computational and Molecular Population Genetics Laboratory, Institute of Zoology, University of Bern, Baltzerstrasse 6, 3012 Berne, SwitzerlandSearch for more papers by this author First published: 23 July 2008 https://doi.org/10.1111/j.1558-5646.2008.00413.xCitations: 300AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Despite hundreds of reports involving both plants and animals, the mechanisms underlying introgression remain obscure, even if some form of selection is frequently invoked. Introgression has repeatedly been reported in species that have recently colonized a new habitat, suggesting that demographic processes should be given more attention for understanding the mechanisms of introgression. Here we show by spatially explicit simulations that massive introgression of neutral genes takes place during the invasion of an occupied territory if interbreeding is not severely prevented between the invading and the local species. We also demonstrate that introgression occurs almost exclusively from the local to the invading species, especially for populations located far away from the source of the invasion, and this irrespective of the relative densities of the two species. This pattern is strongest at markers experiencing reduced gene flow, in keeping with the observation that organelle genes are often preferentially introgressed across species boundaries. A survey of the literature shows that a majority of published empirical studies of introgression during range expansions, in animals and in plants, follow the predictions of our model. Our results imply that speciation genes can be identified by comparing genomes of interfertile native and invading species pairs. Molecular analyses have revealed that many related species share some elements in their genome, potentially due to the retention of neutral or balanced ancestral polymorphisms or to recent hybridization followed by introgression (Rieseberg et al. 1999; Machado et al. 2002; Charlesworth et al. 2005; Mallet 2005). Patterns of gene introgression are commonly thought to depend on the fitness of hybrids and their ability to backcross with parental species (Broyles 2002; Borge et al. 2005). Similarly, the observation that introgression is often asymmetrical between hybridizing species (Barton and Hewitt 1989; Orive and Barton 2002) seems to implicate selection as a cause (Lehman et al. 1991; Borge et al. 2005), with the more adapted alleles spreading from one species to the other (Barton 2001; Whitney et al. 2006). However, the spatial and dynamic contexts of introgression have rarely been considered, even though massive introgression has frequently been reported in species having recently colonized a new habitat (Melo-Ferreira et al. 2005; Rieseberg et al. 2007), thereby questioning the interpretations invoking selection. Hence, a model incorporating demographic details of species encounters might help decide if it is necessary to invoke selection (Ballard and Whitlock 2004; Rieseberg et al. 2007) or differences in relative species abundance (Cianchi et al. 2003) to explain actual patterns of introgression. An analysis of the introgression dynamics at neutral genes experiencing different effective rates of gene flow should also provide some hints to the actual mechanisms involved in asymmetric introgression. Most models of hybridizing species have considered static hybrid zones, or they have insisted on the importance of selection to prevent or enhance introgression, without specifying a null model that would be valid for neutral markers (e.g., Huxel 1999; Ferdy and Austerlitz 2002; Tsitrone et al. 2003; Chan and Levin 2005). Nevertheless, it has been suggested that introgression of neutral genetic markers should affect mostly the advancing taxon as compared to the already established one (e.g., Baker 1948; Moran 1981; Barton and Hewitt 1985; Buggs 2007). For instance, in the context of moving hybrid zones, Buggs (2007) argued that "a moving zone will leave in its wake a tail of clines of unlinked neutral markers," which can be interpreted as introgression of genes from the native species into the colonizing one, as suggested by Moran (1981). This type of explanation appears rather intuitive: as the wave of advance spreads forwards, neutral alleles or traits will flow in the opposite direction, into the invading population, and the frequency of introgressed alleles will steadily increase behind the advancing wave front, until introgression is complete. However, in these previous studies, it was unclear if introgression was truly asymmetric from the native toward the invading species, or if introgression was symmetric but could only be tracked in the part of the range in which the native species was displaced by the invading (i.e., where the two species have been in contact). Moreover, the recent literature on hybridization between exotic and native species during biological invasions has focused on introgression into the native species, rather than on the expected direction of introgression (Huxel 1999; Epifanio and Philipp 2000; Wolf et al. 2001; Hall et al. 2006). As a consequence, no formal analysis has been performed to specify if introgression is generally biased toward the invading species or if introgression is affected by the coexistence of the two species. In the present study, we try to provide answers to these questions by explicitly considering the spatial dynamics of the invasion process. Although several authors have studied population spatial expansions allowing for different dispersal kernels (see e.g., Mollison 1977; Shaw 1995; Bialozyt et al. 2006), and detailed models of genetic exchanges between local and incoming populations have been proposed at the intraspecific level (see e.g., Aoki et al. 1996; Barton 2000; Ackland et al. 2007), two-species models are still missing. There are clear analogies between the invasion of an occupied territory by a new population and the invasion of an advantageous allele in a population, because in both cases migrant individuals phenotypically distinct from local individuals are the vector of the invasion process. The spatial diffusion of an advantageous allele in a population has been investigated by Fisher (1937). Quite recently, the fate of neutral genes linked to a favorably selected mutant spreading in a continuously distributed population has been studied analytically (Barton 2000). This latter work has shown that only the genes located very close to a selected locus would hitchhike and get displaced away from their geographical origin with the wave (i.e., for r/s<<0.1, where r is the recombination rate and s is the selective advantage of the heterozygotes). Therefore, Barton's model predicts that most of the genetic background of a species should not be affected by the spatial expansion of an advantageous allele, but it is unclear to which extent this spatial model of hitchhiking in a single population could be extended to the case of an invading species interbreeding infrequently with a local species. Another intraspecific model that is also relevant for the study of interspecific introgression is the demic diffusion model (Ammerman and Cavalli-Sforza 1973). This model was introduced to explain the occurrence of allele frequency clines over a large portion of Europe by the progressive dilution of the gene pool of Neolithic farmers when they colonized the continent. Previous implementations of this model have shown that demic diffusion would lead to allele frequency clines for very small levels of admixture between invading and local populations (0.024%, as used in Rendine et al. 1986; Barbujani et al. 1995). Note however that allele frequency clines can also occur without any introgression after a spatial expansion (Currat and Excoffier 2005; Klopfstein et al. 2006; Travis et al. 2007). In connection to demic diffusion, Ackland et al. (2007) have recently extended Fisher's equations to model the wave of advance of a neutral trait associated to an advantageous (cultural) trait in humans. In line with previous theoretical work (Aoki et al. 1996), their simulations show that for low migration rates of the carriers of the beneficial technology, the diffusion of the beneficial trait can be quite distinct from that of the neutral trait. It thus suggests a possible disconnection between the diffusion of beneficial and neutral variation, in agreement with the work of Barton (2000) on genetic hitchhiking. In previous models, the importance of intraspecific gene flow (among local or among invading populations) on introgression levels has not been investigated. Organelle genes typically experience reduced gene flow compared to nuclear genes, as a direct consequence of their uniparental and clonal mode of inheritance, both in animals (Birky et al. 1989) and in plants (Petit et al. 1993). Interestingly, organelle genes have long been shown to readily move across species barriers (e.g., Ferris et al. 1983; Powell 1983; Rieseberg and Soltis 1991; Whittemore and Schaal 1991; Bernatchez et al. 1995), but the reasons for this increased permeability are unclear. In the specific context of human evolution, Currat and Excoffier (2004) have simulated the range expansion of modern humans into Europe already occupied by Neanderthals. They modeled interaction between these two taxa in the form of admixture and direct competition between populations, leading to the extinction of the Neanderthals. Although the goal of that study was to estimate the maximum level of Neanderthal genes into the modern human gene pool, these authors noted that the modern human population should present a large fraction of Neanderthal genes if admixture had been possible between the two species. Here, we propose to use similar spatially explicit simulations to further investigate the pattern of introgression at neutral loci resulting from the range expansion of a species into an already occupied territory. We model a range expansion process with and without competition with the local species, and we monitor introgression levels in invading and resident species as a function of their ability to interbreed. We then check our model predictions against published empirical studies of introgression for organelle and nuclear genes during range expansions in animals and plants. Materials and Methods SIMULATION OF SPATIAL EXPANSIONS WITH INTERBREEDING Simulations were performed using a modified version of the SPLATCHE program (Currat et al. 2004) in a virtual world with 10,000 demes made up of haploid individuals and arranged on a two-dimensional lattice. Note that these haploid simulations apply equally well to diploid loci, but in that case population sizes do not represent the number of individuals (N) but the number of gene copies at those loci (2N). At the onset of each simulation, the whole world is already occupied by a local species. A new invading species appears in the lower left corner of the square world, in a deme arbitrarily located at position [5,5] on the lattice (see Fig. 1). This newly founded population then sends migrants to empty neighboring demes, which are logistically filled to their carrying capacity, and which send further migrants to adjacent demes, thus progressively colonizing the whole world (see Fig. 1). During and after this colonization process, the invading species can locally interact with the resident species by exchanging migrants through interbreeding and by competing with it for local resources. In more details, each intersection point of the lattice can harbor both a local and an invading deme, which can exchange genes by interbreeding. Here we define an introgression event as the transfer of a gene from the gene pool of one species to that of the other species. For simplicity, this event is supposed to occur in a single generation, although in reality it can take place over a few generations by hybridization and repeated backcrossing. Therefore, these interbreeding events should be viewed as effective interbreeding events. As discussed elsewhere (Currat and Excoffier 2004), the rate of interbreeding is assumed to be density-dependent. At any location of the lattice, the probability of a successful introgression event is thus defined as A=γ (2NiNj)/(Ni+Nj)2, where Ni and Nj are the current deme densities of the two species. It results in the introgression of ANi genes from species i to species j, whereas ANj genes are transferred in the other direction each generation. In our model, the parameter γ is a general measure of the strength of barriers to gene flow between species. We do not explicitly model the nature of these barriers, but they could be either prezygotic (and γ could for instance be considered as a measure of disassortative mating), postzygotic (and γ would be a measure of the fitness of the hybrid individuals), or any combination of factors preventing the successful mating of members of both species. In any case, a γ value of 0 corresponds to a total absence of interbreeding between the two species, a γ value of 1 corresponds to random mating between the two species, and any value in between implies that mating is locally nonrandom between the two species. With this simple parameterization, we do not need to specify how mating is prevented or if it is controlled by one or several loci. Note however that the effect of genes involved in reproductive isolation on nearby neutral loci should decrease monotonically with recombination distance (Barton and Bengtsson 1986). In other words, one would expect that the γ parameter would increase with distance from loci controlling the fitness of hybrids or disassortative mating. Assuming that γ values are identical in both species, deme densities are then updated as N′i=Ni(1 −A) +ANj to reflect these episodes of interbreeding. Then, deme densities are logistically regulated (see Currat and Excoffier 2004) as N″i=N′i[1 +r(Ki−N′i−N′j)/Ki], where Ki is the local carrying capacity of the ith population and r is the deme intrinsic rate of growth. Finally, migration can occur between neighboring demes of the same species, and deme densities are again updated as N‴i=N″i(1 −m) +Ii, where m is the probability of emigration, and Ii is the number of migrants received from the neighboring demes (four neighboring demes for central demes, and less for demes located on the border of the lattice). Note that at carrying capacity, each deme sends on average Km emigrants per generation to its neighbors. These updates are repeated over all demes for 1500 generations. Figure 1Open in figure viewerPowerPoint Schematic representation of the simulated invasion process with or without competition between the local and the invading species. Light gray and black pixels represent demes in which only the local or only the invading species occurs, respectively. Dark gray pixels represent locations where both species coexist. The first frame shows the situation after 40 generations. The three other frames depict the invasion process over time, depending on the demographic parameters reported in Table 1. INVASION SCENARIOS We studied seven scenarios (C1–C7) of invasion with competition between local and invading species as well as five scenarios without competition (NC1–NC5) (see Table 1). These scenarios explore the effects of differences in local carrying capacity (K) and number of migrants exchanged between neighboring demes of the same species (Km) on introgression levels. In scenarios with competition, the invading species has a competitive edge over the local one. This advantage is implemented by assigning it a larger carrying capacity, which eventually drives the local species to extinction (see Shigesada and Kawasaki 1997; Currat and Excoffier 2004 for details). In that case, interbreeding will only occur at the edge of the expansion wave, where the two species still coexist (see Fig. 1). An absence of competition implies that the two species occupy different ecological niches, and both will coexist and potentially interbreed until the end of the simulation. Table 1. Parameters of the simulated invasion scenarios. K: Carrying capacity. Km: Number of emigrants sent to neighboring demes of the same species at carrying capacity. The intrinsic rate of growth (r) was set to 0.5 in all cases. The cohabitation time is the average number of generations during which the two species coexist at a given location. The colonization time is the number of generations taken by the invading species to colonize the square world. Scenario Local species Invading species Cohabitation time Colonization time K Km K Km Competition C1 50 1 500 10 9.5–10.0 1495–1500 C2 50 1 5000 100 7.5–8.0 1240–1270 C3 500 10 5000 100 17.0–19.5 1140–1330 C4 50 10 500 100 10.0 540–570 C5 50 10 5000 1000 8.5 500–510 C6 50 1 100 10 19.0–23.0 1095–1270 C7 500 10 1000 100 29.0–32.0 735–1010 No competition NC1 50 10 50 10 1135–1150 680–715 NC2 500 10 500 10 705–790 1425–1500 NC3 500 100 500 100 1215–1245 505–555 NC4 50 10 500 100 1215–1225 540–555 NC5 500 100 50 10 1135–1170 645–715 COALESCENT SIMULATIONS For each scenario and various levels of interbreeding γ, we performed 10,000 backward coalescent simulations to assess the final proportion of introgressed genes into a given species. For each coalescent simulation, we sample 40 genes in 25 equally spaced demes of the invading population for scenarios C1–C7 and in 25 demes (see Fig. 5) of the invading and of the local species for scenarios NC1–NC5. The ancestry of the samples is traced back in time on the simulated genealogies to infer the species' origin of these genes. The average fraction of genes originating from the other species is used as an estimator of the introgression level in each species. Note that for very small γ values, introgression levels can vary among the 25 demes and lead to clines from the introduction point toward the far end of the range (see Figs. 5 and 6). Figure 5Open in figure viewerPowerPoint Spatial variation in introgression levels. Proportion of introgressed genes (black pie) in the invading species for each of the 25 simulated samples and for the 12 scenarios C1-C7 and NC1-NC5. Simulations were done by assuming that γ= 0.015. The source of the invading species expansion is close to the bottom left pie-chart (see Fig. 1). Figure 6Open in figure viewerPowerPoint Variation in introgression level into the invading population along the expansion path in absence of competition. The average introgression level is reported as filled circles for different interbreeding levels in 20 equally spaced demes located along the expansion path going from the lower left to the upper right of the simulated world. The modeled scenario corresponds to scenario NC1 in Table 1 and Figure 3B. The dashed lines correspond to the limits of a 90% empirical confidence interval obtained from 10,000 coalescent simulations. The expansion starts from deme 1. Note that a gradient of introgression is only visible over the whole transect for very limited levels of interbreeding (γ= 0.01). LITERATURE SURVEY We surveyed the recent literature for cases of significant introgression after a range expansion of one species into the range of another one. We looked for examples based both on chloroplast and mitochondrial markers and on nuclear markers. However, examples based on organelle markers were more frequent, because they are particularly abundant in the introgression literature and the specific origin of the lineages can be more easily inferred, thus making it possible to deduce the direction of the introgression. Among over 200 papers reporting introgression, many were restricted to recent zones of contact or to taxa showing only F1 hybrids (hence without effective introgression). These cases were discarded because they do not correspond to the modeled process of range expansion with effective introgression measured over a large portion of the range of the species. A further difficulty was to objectively infer which species was resident and which one was invading. This information was sometimes unavailable, or the two species were thought to have both expanded, but it was unclear which one had arrived first. These ambiguous cases were also excluded. Results DYNAMICS OF THE INTROGRESSION To understand the introgression process, it is instructive to monitor the dynamics of the interbreeding events at the wave front, where the invading species is initially at low densities. The change in deme densities upon invasion and the number of introgression events between the local and the invading species are illustrated in Figure 2 with or without competition between the two species. In both cases, local genes introgress the invading species when it is still at very low density. Therefore these few introgressed genes are amplified by the logistic growth of the invading population. This is in contrast with the situation of invading genes entering the local species, which is already at carrying capacity (without competition) or declining (with competition). Therefore, a single gene introgressing the invading population may be found at multiple copies when the invading population reaches its carrying capacity, which introduces a net asymmetry in effective introgression between the invading and the local population. Note that this asymmetry only occurs if interbreeding events are frequent enough to occur on the wave front, when the invading population is still at low density, which explains why the final introgression levels are symmetric for low interbreeding rates and become asymmetric with increasing interbreeding levels (see Fig. 3B). Large overall introgression levels in the invading population are therefore due to the progressive dilution of the gene pool of the invading species by the few interbreeding events occurring at the wave front (Chikhi et al. 2002). Note however that this dilution is greatly accelerated when one takes into account the logistic growth occurring in newly founded demes (Currat and Excoffier 2004). Therefore, as colonization proceeds, the gene pool of invading populations is increasingly diluted by resident genes, such that the invasion process is carried out by progressively more introgressed individuals, leading to higher introgression levels away from the source of the invasion (see 4, 5, 6). Figure 2Open in figure viewerPowerPoint Illustration of the local demographic and introgression dynamics. We plot the evolution of population densities and introgression events over time at a given location of the lattice, assuming no migration from neighboring demes for simplicity. At generation zero, the local species is at carrying capacity, and the invading species appears. (A) Demographic parameters correspond to the scenario NC2 without competition and in (B) demographic parameters correspond to the scenario C1 with competition. In both cases, γ= 4%. Note that introgression first occurs from the local to the invading species. Note that in (B) the number of introgression events is smaller than (A) due to the smaller population size of the local species (see Material and Methods). Loc → Inv: Introgression events from the local to the invading species. Inv → Loc: Introgression events from the invading to the local species. Figure 3Open in figure viewerPowerPoint Proportion of genes introgressed into a given species as a function of the level of interbreeding (γ). (A) Cases with competition between a local and an invading species. (B) Cases without competition. loc. documents introgression level in the local species, whereas inv. documents the introgression level in the invading species. Reported rates are averages over 10,000 simulations. Case simulation parameters are those reported in Table 1. Interbreeding success can be considered as the fitness of genes transferred into a new background by interbreeding. Figure 4Open in figure viewerPowerPoint Variance in introgression levels across the invaded area. Proportion of introgressed genes in the invading species (black dots) and in the local species (gray triangles) for three samples located at various distances from the origin of the expansion. Sample 1 is located at the source of the expansion (pane A), sample 13 is in the middle of the invaded area (B) and sample 25 is most distant from the origin of the expansion (C). Dotted lines represent the limits of an empirical 90% confidence interval for introgression levels obtained from 10,000 simulations. The variance in introgression levels is particularly large close to the origin of the expansion and when the admixture rate (γ) between species is low. The introgression proportions have been obtained for scenario NC1 (Table 1). INVASION WITH COMPETITION When the resident and the invading species compete for local resources, the invading species progressively replaces the local species if it has a larger carrying capacity (Shigesada and Kawasaki 1997), but in our simulations it gets almost completely introgressed by local genes if more than 10% of effective interbreeding events are successful (Fig. 3A). Note that this 10% limit is rather arbitrary, as there are many scenarios for which complete introgression occurs for much lower interbreeding levels. It may be quite specific to our simulation sets, and the final level of introgression depends on where it is measured (e.g., see Fig. 5). Indeed, demes located further away from the source show higher levels of introgression, because more interbreeding events occured during the expansion at the wave front. Final introgression levels can actually reach up to 95% with as little as 3% interbreeding success (cases C1, C3, C6, and C7). Even when the invading species is 100 times more numerous than the local one, complete introgression of the invading species occurs when interbreeding rate (γ) exceeds 10% (cases C2 and C5). When the invading species is only two times more abundant than the local species, high levels of introgression also occur for moderate levels of interbreeding (cases C6 and C7). This is explained by the fact that the cohabitation period at the wave front is longer for reduced competitive advantage of the invader (see Table 1), allowing for more interbreeding events to occur during this period. Overall, the final level of introgression is positively correlated with the interbreeding rate and the size of the local population (compare cases C2 and C3), and it is negatively correlated with the size of the invading population (compare cases C1 and C2 or C4 and C5). Introgression is also favored when gene flow between adjacent demes is restricted (compare cases C1 and C4, C1 and C6, as well as C2 and C5). This makes sense, because genes introgressed from the local population at the wave front will compete with migrant genes from the invading populations if gene flow is high, and will be thus less amplified by the logistic regulation. A similar negative effect of migration on the spread of new variants has been documented in the case of "allele surfing" in expanding populations (Klopfstein et al. 2006). Allele surfing refers to the spread of a previously rare allele during a range expansion. An allele can surf on a wave of advance, reaching high frequencies and occupying a large area, despite having no selective advantage (Edmonds et al. 2004; Klopfstein et al. 2006; Hall
Publication Year: 2008
Publication Date: 2008-06-01
Language: en
Type: article
Indexed In: ['crossref', 'pubmed']
Access and Citation
Cited By Count: 774
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot