Evol Ecol (2013) 27:165–184 DOI 10.1007/s10682-012-9577-z ORIGINAL PAPER

Does variation in host plant association and symbiont infection of pea aphid populations induce genetic and behaviour differentiation of its main parasitoid, Aphidius ervi? Emilie Bilodeau • Jean-Christophe Simon • Jean-Fre´de´ric Guay Julie Turgeon • Conrad Cloutier

•

Received: 27 October 2011 / Accepted: 3 May 2012 / Published online: 4 June 2012 Ó Springer Science+Business Media B.V. 2012

Abstract The host-associated differentiation (HAD) hypothesis states that higher trophic levels in parasitic associations should exhibit similar divergence in case of host sympatric speciation. We tested HAD on populations of Aphidius ervi the main parasitoid of the pea aphid Acyrthosiphon pisum, emerging from host populations specialized on either alfalfa or red clover. Host and parasitoid populations were assessed for genetic variation and structure, while considering geography, host plant and host aphid protective symbionts Regiella insecticola and Hamiltonella defensa as potential covariables. Cluster and hierarchical analyses were used to assess the contribution of these variables to population structure, based on genotyping pea aphids and associated A. ervi with microsatellites, and host aphid facultative symbionts with 16S rDNA markers. Pea aphid genotypes were clearly distributed in two groups closely corresponding with their plant origins, confirming strong plant associated differentiation of this aphid in North America. Overall parasitism by A. ervi averaged 21.5 % across samples, and many parasitized aphids producing a wasp hosted defensive bacteria, indicating partial or ineffective protective efficacy of these symbionts in the field. The A. ervi population genetic data failed to support differentiation Electronic supplementary material The online version of this article (doi:10.1007/s10682-012-9577-z) contains supplementary material, which is available to authorized users. E. Bilodeau  J.-F. Guay  J. Turgeon  C. Cloutier (&) De´partement de Biologie, Universite´ Laval, Que´bec, QC G1V 0A6, Canada e-mail: [email protected] E. Bilodeau e-mail: [email protected] J.-F. Guay e-mail: [email protected] J. Turgeon e-mail: [email protected] J.-C. Simon INRA, UMR 1349, Institut de Ge´ne´tique, Environnement et Protection des Plantes, Domaine de la Motte, BP 35327, 35653 Le Rheu cedex, France e-mail: [email protected]

123

166

Evol Ecol (2013) 27:165–184

according to the host plant association of their pea aphid host. Potential for parasitoid specialization was also explored in experiments where wasps from alfalfa and clover aphids were reciprocally transplanted on alternate hosts, the hypothesis being that wasp behaviour and parasitic stages should be most adapted to their host of origin. Results revealed higher probability of oviposition on the alfalfa aphids, but higher adult emergence success on red clover aphids, with no interaction as expected under HAD. We conclude that our study provides no support for the HAD in this system. We discuss factors that might impair A. ervi specialization on its divergent aphid hosts on alfalfa and clover. Keywords Coevolution  Tritrophic system  Parasitism  Endoparasitoid  Host specialization  Aphid bacterial symbiosis  Medicago sativa  Trifolium pratense

Introduction Coevolutionary processes are major mechanisms generating species diversity on Earth (Ehrlich and Raven 1964; Farrell 1988; Mitter et al. 1988; Schluter 2000). Evidence of bitrophic arm races between plants and insect herbivores are numerous, e.g., apple maggot fly, Rhagoletis pomonella (Feder et al. 1988), brown planthopper, Nilaparvata lugens (Claridge et al. 1985) or European corn borer, Ostrinia nubilalis (Bourguet et al. 2000) but those on multitrophic interactions remain sparse (Blair et al. 2005; e.g., Stireman et al. 2006; Forbes et al. 2009). Determining how higher trophic levels evolve in response to specialization of lower trophic levels could highlight our understanding of species interactions and diversification and improve biological control strategies (Roderick and Navajas 2003). In this paper, we examine tritrophic interactions between a parasitoid wasp (Aphidius ervi), and its phytophagous insect host (the pea aphid, Acyrthosiphon pisum) whose populations are specialized on different legume species, while explicitly considering potential mediation by aphid bacterial symbionts. The pea aphid is a valuable model of host plant related specialization and incipient speciation (Ferrari et al. 2006; Peccoud et al. 2009a, b; Via 2009). Probably introduced from Europe to North America (Mackauer 1971; Angalet and Fuester 1977), it is considered a mild pest of forage and food crops (Hales et al. 1997). Pea aphid feeds exclusively on Fabaceae (Ferrari and Godfray 2006), but host plant specific and genetically differentiated biotypes have been identified, such as widespread alfalfa and clover races known in the USA (Via 1991a), Europe (Sandstro¨m 1996; Simon et al. 2003; Ferrari et al. 2006), and South America (Peccoud et al. 2008). In eastern North America, studies have documented extensive host plant based sympatric reproductive isolation of the pea aphid, expressed as higher reproduction and survival, and preference for their respective host plant (Via 1991a, b, 1999; Caillaud and Via 2000; Via et al. 2000; Hawthorne and Via 2001). If pea aphid host plant specialization affects A. ervi fitness, it should exert divergent selection pressures on this parasitoid in North America. In Europe A. ervi has a greater diversity of hosts, where a genetic mosaic of distinct races exploiting different aphid species on unrelated host plants coexist (see Emelianov et al. 2011 and references therein). In contrast, in North America, A. ervi diversity is presumably more restricted (see e.g., Nguyen et al. 2008, where the listed host Macrosiphum euphorbiae was found to be highly resistant to A. ervi) as it was introduced around 50 years ago in a large scale biological control programme, apparently from collections mostly made in France on pea aphids from alfalfa (Mackauer and Finlayson 1967; Angalet and Fuester 1977; Hufbauer 2002). The

123

Evol Ecol (2013) 27:165–184

167

introduction of A. ervi from Europe may have started a large scale coevolutionary process in a tritrophic system that is still going on, but whose consequences remain to be determined. Interactions with A. pisum or closely related hosts are absolutely mandatory for A. ervi, requiring wasp behavioural and physiological adaptations to multiple host traits (detection, attack, host penetration to inject egg and venom) and particularly to host invasion and exploitation by immature forms (virulence, physiological manipulation, and parasitic efficiency expressed as host conversion efficiency into parasitoid growth). In particular, adaptation to survive potential host immunological defenses including defensive bacterial symbionts during early parasitism is critical (see below). Stireman et al. (2006) hypothesized that where sympatric speciation occurs in phytophagous insect hosts, their specialized parasitoids should experience similar genetic divergence pressure, a hypothesis referred to as cascading (or sequential, see Abrahamson and Blair 2007) host-associated differentiation (HAD). In the A. ervi–A. pisum system, there is clear potential for HAD given genetic variation and evolutionary potential both in aphid resistance and parasitoid virulence (Henter 1995; Henter and Via 1995; Dion et al. 2011). The possibility of divergent selective pressures on the parasitoid by the alfalfa versus the clover aphid races was first explored by Hufbauer and Via (1999), Hufbauer et al. (2001) and Hufbauer (2002). In elegant reciprocal transplant tests where (New York and France) aphids and parasitoids from alfalfa and clover were matched, no evidence was found that wasps were more adapted to their host aphid of origin (Hufbauer 2002, Fig. 2). Intriguingly, however, the New York wasps (unlike the French) experienced twice the resistance (immunity) of the alfalfa than the clover aphids, which remained unexplained possibly because the role of defensive symbionts in pea aphid resistance to parasitoids was unknown at that time. Indeed more recent work has shown that pea aphid resistance to Aphidius depends primarily on facultative symbionts, particularly H. defensa, which can stop or impair development of A. ervi (reviewed in Oliver et al. 2010; see also Guay et al. 2009). Another symbiont known as PAXS has been identified in pea aphids from Que´bec (Guay et al. 2009) and its association with H. defensa confers high resistance to A. ervi, even under heat stress. It should also be noted that the facultative symbiont Regiella insecticola, known in the pea aphid, has also been linked to parasitoid resistance in the green peach aphid Myzus persicae (Vorburger et al. 2010). Defensive aphid symbionts are common and vertically transmitted, although they can also be acquired horizontally (Moran and Dunbar 2006; Oliver et al. 2010). Their potential role in natural aphid populations and evolutionary consequences for A. ervi populations cannot be ignored. Aphids are well known for a long history of coevolution with their primary nutritional symbiont Buchnera aphidicola (Baumann et al. 1995). Functional traits of aphid defensive symbionts (e.g., H. defensa) also make them good candidates as key players in aphid evolution, and consequently in HAD of pea aphid endoparasitic wasps such as A. ervi as their most important parasites. As intracellular microorganisms, they exhibit high intimacy with both the aphid and its primary symbionts, whose fitness in turn they directly affect. Recent holobiotic concepts of evolution explicitly recognize aphid symbiont associations as essential components of the aphid phenotype (Zilber-Rosenberg and Rosenberg 2008), implying that species such as H. defensa, are key players of aphid evolution. High anti-parasitoid resistance conferred by H. defensa (and/or PAXS), is a matter of life and death in pea aphid populations, because of frequent parasitism by A. ervi, its most abundant and widespread parasitic wasp. Defensive bacteria could affect HAD in this system by being more or less prevalent, or their defenses being expressed at higher levels in some host plant races of the pea aphid than others, thus creating the potential to select for A. ervi specialization on pea aphid host plant races that are exempt from them. They can destroy A. ervi at infection time (oviposition) or

123

168

Evol Ecol (2013) 27:165–184

during early development of the initial stages (embryo, first stage larva). Thus the A. ervi wasp should reproduce more on pea aphids that are free from anti-parasitoid bacteria, thus favouring strategies developed either to avoid infected hosts, or to evolve anti-bacterial virulence that overcome symbiont dependent immunity. Potential for rapid evolutionary response in A. ervi was recently demonstrated in laboratory experiments (Dion et al. 2011). Here, we tested the HAD hypothesis in A. ervi by first studying genetic variation using microsatellites in alfalfa and clover field populations of pea aphids and associated A. ervi. Genetic evidence for HAD remains sparse (Stireman et al. 2006; Forbes et al. 2009), and mostly negative (e.g., Cronin and Abrahamson 2001; Daza-Bustamante et al. 2002; Medina 2005; Lozier et al. 2009). However, genetic evidence alone might overlook ongoing divergence (Via 2001). Indeed, in cases of recent divergence, genetic signature may be delayed if genetic drift operates slowly because of large population sizes and/or ongoing, albeit reduced, gene flow (Via 2009). Divergence would be better tested from complementary vantage points of genetics and ecology (Via 2001, 2009; Dorchin et al. 2009; Lozier et al. 2009; Feder and Forbes 2010). Thus we also tested HAD predictions on host parasite interactions using a reciprocal transplant design with hosts and parasitoids associated with alfalfa and red clover, and which we examined at two levels. First, wasp host selection was examined for cues of divergence, as behaviour is shaped by natural selection to quickly respond to changing environments (Caro 2007; Morris et al. 2009). Host selection in A. ervi is critically expressed when wasps detect aphids at close range (Michaud and Mackauer 1994). But attacks can fail as pea aphids have a range of antiparasitoid defense behaviours (Gerling et al. 1990; Mackauer et al. 1996; Schwo¨rer and Vo¨lkl 2001; Kunert et al. 2010; Le Ralec et al. 2010), and can avoid ovipositor contact by kicking, running away or dropping, resulting in a vanished opportunity. Second, we used immature survival and adult emergence and sex ratio as proxys to test for divergence in A. ervi’s ability to exploit the aphids. Given egg laying in a host, the ultimate barrier to parasitism is host immunity that kills or inhibits the egg or larva (Guay et al. 2009; Le Ralec et al. 2010; Oliver et al. 2010). Failure to hatch or to emerge is a cue to the inability of parasitoid progeny to exploit hosts selected by the wasp. We tested the HAD prediction that probability of A. ervi emergence (conditional on successful attack) should be higher in aphids similar to those that hosted the wasp. Sex of progeny is an additional indicator of adaptative host selection based on sex allocation theory (Charnov et al. 1981). For koinobionts such as A. ervi studies support the prediction that fertilized eggs (daughters) are preferentially allocated to higher quality, larger or more advanced host stages (Cloutier et al. 1991, 2000; Sequeira and Mackauer 1993; Morris and Fellowes 2002). If HAD occurs here, the probability of a female emergence may be higher in aphids from the host plant that hosted the parasitoid.

Materials and methods Field sampling and aphid rearing In summer 2007, we collected pea aphids in three pairs of field crops in different localities in southern Que´bec, Canada. Each pair included an alfalfa field and a red clover field. The first pair was located at the agricultural experiment station of Universite´ Laval in Saint-Augustinde-Desmaures (referred as Saint-Augustin in Tables) (46°430 55.8400 N, 71°300 54.1400 W), the second in the field station of McGill University in Montre´al (45°250 48.8200 N, 73°550 53.500 W),

123

Evol Ecol (2013) 27:165–184

169

and the third one in the Agriculture and Agri-Food Canada, Lennoxville field site (45°220 9.5500 N, 71°490 45.2600 W). The three sites were separated by 160–240 km. We collected a total of 389 pea aphids over all sites and field crops by sweep netting crop vegetation (only one aphid collected per sweep), and moving 10 m between catches to minimize risk of resampling aphid clones (Supplementary Table 1). Green and pink pea aphids were present, and each color morph was sampled to be represented nearly equally among collections. Collected aphids were summer parthenogens and were bred clonally (Blackman and Eastop 1984) under controlled conditions (18 °C, 16L: 8D photoperiod), on their host plant of origin, to obtain adults for molecular work. We thus obtained 302 pea aphid clonal lineages i.e., 48–53 per crop for each site (Supplementary Table 1). Many collected aphids were parasitized and formed mummies before emerging as adults, the great majority of which could safely be assigned to A. ervi based on morphology (Mackauer and Finlayson 1967). Among the emerging A. ervi, 157 were females that could be linked to a known field collected A. pisum host. Also, among the A. ervi-parasitized aphids that mummified, 70 produced clonal daughters before dying, which were reared to maturity for molecular characterization (Supplementary Table 1). Aphid symbionts detection For each pea aphid clone, we determined the occurrence of the obligatory symbiont B. aphidicola, and facultative symbionts R. insecticola and H. defensa, using diagnostic PCR based on bacterial 16S rDNA (Sandstro¨m et al. 2001; Tsuchida et al. 2002). Because of their possible contribution to A. ervi resistance, aphids were also characterized for presence of Serratia symbiotica and PAXS (Oliver et al. 2003, 2006; Guay et al. 2009). Aphid DNA samples where Buchnera could not be confirmed were considered unreliable and rejected for quantitative analyses. We related the symbiont associations in aphids (thereafter symbiotypes) with aphid population ecology (site, host plant association), and with genetic structure using cluster analysis. Aphid facultative symbiosis is inherited in asexually reproducing aphids, because of high symbiont heritability across aphid clonal generations (Russell and Moran 2005). Symbiotype structure based on presence/absence of four facultative symbionts in aphid samples was examined using FLOCK_AFLP 2.0, a non Bayesian clustering algorithm for mapping genetic similarity in natural populations (Duchesne and Turgeon 2009). The mixed aphid sample (n = 277) was subdivided into K C 2 groups based on symbiotype similarity, which were tested for possible relations with site, host plant association, and aphid genetic population structure. Aphid and parasitoid population structure in the field We used microsatellite markers to assess the degree of genetic variation among aphid and wasp samples. For the 302 pea aphid clones, we extracted DNA from reared adults using the salting-out method modified from Sunnucks et al. (1996) and Aljanabi and Martinez (1997). Microsatellite analyses of pea aphids were performed at seven loci isolated from A. pisum or from its close relative Acyrthosiphon loti (ApF08M, ApH08M, ApH10M, AlA09M, AlB07M, AlB08M, and AlB12M) (Simon et al. 2007). For A. ervi wasps, we extracted DNA from 157 females using the DNeasyTM Tissue Kit (Qiagen, Mississauga, Ontario, Canada) with the following modifications: 5 ll of Proteinase K was added 1 h before the end of lysis, and DNA was resuspended in 50 ll of Buffer AE. We assessed microsatellite variation at eight loci isolated from A. ervi (Ae4, Ae51, Ae74, Ae129,

123

170

Evol Ecol (2013) 27:165–184

Ae47A, Ae47B, Ae78A, Ae78B) (Hufbauer et al. 2001). For data analysis, missing aphid and wasp genotypes at some loci were inferred using the hot deck imputation technique (Fuller and Kim 2005), where missing genotypes were replaced by randomly selected alleles from the overall distributions, thus allowing use of all available information from field samples. We compared two approaches of clustering analysis to infer population structure of A. pisum and A. ervi. First, we used FLOCK_MSAT 2.0, which relies on maximum-likelihood iterative allocation to form K clusters (Duchesne and Turgeon 2009). For any value of K, Flock provides the likelihood difference that an individual belongs to each of the K clusters, and the difference in allocation log-likelihoods to the two most likely reference groups (LLOD). Flock results were used to examine how groups of individuals on the basis of sampling site, host plant (and symbiotype, see above) were allocated among the K clusters using v2 tests. This test, as well as the mean LLOD value, allow validating which K value(s) is (are) biologically relevant. Second, we used the STRUCTURE software (version 2.3, Pritchard et al. 2000), a well-known Bayesian clustering algorithm allowing estimation of the proportion of an individual’s genome having ancestry in the K clusters. Posterior probabilities were determined three times for each K value, using 500,000 burn-in and 1,000,000 iterations with the correlated allele frequency and alpha estimation options turned on. We also performed analyses of molecular variance (AMOVA) to determine relative contributions of crop plant and site in accounting for the observed genetic variation in pea aphids and the A. ervi females. We used Arlequin v. 3.11 based on Excoffier et al. (1992) to partition covariance among individuals, populations, plants and sites. Significance tests were done using F-tests based on sequential sums of squares from 10,000 permutations of the data, which give a precision of ±0.0001 to the p values. In addition, we performed a separate AMOVA on the 70 A. ervi that emerged from field collected pea aphids whose genotypes could be matched to that of their pea aphid host. Here, we used the genetic clusters of pea aphids identified by FLOCK_MSAT 2.0 (see ‘‘Results’’) as the grouping factor (results with STRUCTURE were identical). We calculated differentiation coefficients (FST) across samples using ARLEQUIN (Excoffier et al. 2005). FST is a relatively robust descriptor of genetic differentiation to evaluate the extent of realized gene exchange between clusters (Via 1999). We corrected p values for multiple comparisons using sequential Bonferroni adjustments (Rice 1989). Host selection behaviour and host exploitation experiment We examined A. ervi specialization in a reciprocal design using wasps from alfalfa and clover pea aphid host sympatric populations. Two experimental A. ervi lines were started from 94 A. ervi collected at the mummy stage in alfalfa and clover in Saint-Augustin-deDesmaures. We randomly selected 15 females and 10 males among emerging wasps from each crop to establish each wasp line as a laboratory colony (alfalfa or red clover). Each colony was maintained at about 100 individuals on susceptible pea aphids (infected with Regiella insecticola but free of other screened symbionts) collected in their field of origin, on greenhouse alfalfa and red clover potted plants. Experimental wasp individuals were isolated at the mummy stage and visually observed to mate before testing when aged 48–72 h post-emergence. Experimental host aphids were from 10 randomly selected clones from the SaintAugustin-de-Desmaures site, five from each host plant. Among the five clones from alfalfa two were pink and three were green morphs, and among the five clones from red clover three were pink and two were green. Aphid clones were genotyped and symbiotyped and

123

Evol Ecol (2013) 27:165–184

171

were bred in isolation on their respective host plant over the four-month experiment (18 °C; 16L: 8D). We used 3rd stage aphid hosts to control for effects of host size on wasp preference, and progeny survival, emergence and sex ratio (Sequeira and Mackauer 1992, 1993; Mackauer et al. 1996; Mackauer 1997; Cloutier et al. 2000; Colinet et al. 2005; Henry et al. 2005, 2009; Barrette et al. 2009). Host selection behaviour was assessed by the same observer (EB) in a 5-min test conducted at 20 °C in a 60 9 15 mm Petri dish positioned on a white surface under fluorescent room light. We introduced a 3rd stage aphid randomly selected from one of the 10 clonal lineages, with an A. ervi female randomly selected from one of the two colonies. We recorded all ovipositor contacts on the aphid, and delay before the first such contact using THE OBSERVER 5.0. (Noldus Information Technology, Wageningen, Netherlands). For each wasp individual we replicated the test six times using six different aphids of the same clone. Replication allowed controlling for potential short term change in behaviour of the young inexperienced wasps (e.g., Morris and Fellowes 2002; Langley et al. 2006; Takemoto et al. 2009). After testing a first wasp from one colony, a wasp from the second colony was similarly tested, so that wasps from both colonies were always tested on the same occasion on each clone. We thus tested 100 A. ervi wasps, i.e., 50 from each colony and conducted a total of 600 host selection tests. Following behavioural observations, experimental aphids of replicates 1–3 (n = 300) were dissected shortly after to determine rate of host acceptance based on presence of parasitoid eggs in the haemolymph. Host acceptance was the ratio of number of eggs/ number of wasp ovipositor contacts for each aphid. Probability of ovipositor contact and delay before it (in seconds) were used as measures of adaptive host selection (Almohamad et al. 2008; Henry et al. 2009). If HAD occurs in this system, probability of ovipositor contact should be higher, and delay before such contact shorter, on aphid clones from the same host plant as the aphid host of the wasp. Adaptative host exploitation by the A. ervi progeny was examined using replicate aphids 4–6 (n = 300) that had been contacted C1 times with the ovipositor. They were reared on cut foliage from their respective host plant (18 °C, 16L: 8D) to estimate adult emergence rate and sex ratio. From a statistical viewpoint, the experiment comprised five replicate A. ervi wasps from each colony (blocks); two A. ervi lines (associated with alfalfa and red clover aphids); 10 pea aphid clones, i.e., five from alfalfa and red clover; and six aphids from each clone that were presented in succession within 30–40 min. We examined factors that explain (1) probability of ovipositor contact; (2) delay before first contact; (3) probability of emergence of an adult A. ervi conditional to ovipositor contact; and (4) probability of a female parasitoid emergence. We used mixed logit models (GLIMMIX, SAS 9.2) for binary dependent variables (i.e., 1, 3, and 4) and mixed ANOVA (MIXED procedure, SAS 9.2) for delay before ovipositor contact. Pea aphid clone (n = 10), parasitoid colony (n = 2), and replicate order (n = 1–6) were fixed effects. Random effects were block (n = 5), and block 9 pea aphid clone 9 parasitoid colony (n = 100). We also tested for effects of covariables: host plant of the aphid (n = 2, clover or alfalfa), color morph of the pea aphid (n = 2, green or pink), and presence of facultative symbionts R. insecticola and H. defensa. We followed the method of Allison (1999) for probabilities that can be safely estimated at 0 % or 100 % because of lack of variation for some treatments. Also, to estimate probability of ovipositor contact and delay, we had to reduce the six replicates to three levels (i.e., first and second, third and fourth, and fifth and sixth) for model convergence. All models assumptions were verified and the significance level was set at a = 5 %.

123

172

Evol Ecol (2013) 27:165–184

Results Genetic structure of aphid and parasitoid populations For pea aphids, both clustering analyses clearly identified two groups of genotypes closely associated with their host plant (alfalfa and red clover). Using FLOCK, 302 pea aphids collected formed two genotype clusters (mean LLOD = 9.32) differentially distributed among the two host plants (v25 = 172.46; p \ 0.001). Using STRUCTURE, the highest posterior probability was also clearly for K = 2. Nearly all pea aphids collected on alfalfa had genotypes belonging to the first genetic cluster, while the second cluster comprised 80 % of aphids collected in red clover fields (Fig. 1), the other individuals having genotypes belonging to the (first) cluster dominated by the alfalfa aphids. AMOVA indicated 11.2 % of pea aphids genetic variation was explained by the among crop plants effect, compared to 1.2 % for the sites within plants effect, both sources being highly significant (p B 0.0001; Table 1). Pairwise comparisons of FST (Table 2) show that

Fig. 1 Estimated population structure of 302 A. pisum individuals collected in three sites in Que´bec, 2007. Each genotype is represented by a thin vertical line, with segments representing the individual’s estimated membership fractions in the two clusters (i.e., red for cluster 1 and green for cluster 2). Black lines separate individuals of different samples labelled according to site (bottom) and host plant (top). Clustering analysis was performed with STRUCTURE using genotypes based on 7 microsatellite loci Table 1 Analyses of molecular variance (AMOVA, Arlequin 3.11) of Acyrthosiphon pisum (N = 302) and Aphidius ervi females (N = 157) emerged from them Source of variation

D.F.

Sum of squares

Variance component

Variation (%)

p

A. pisum Among plants

1

98.997

0.30856

11.19

B0.0001

Among sites within plants

4

23.218

0.03370

1.22

B0.0001 B0.0001

Within populations

598

1,443.755

2.41431

87.58

Total

603

1,565.970

2.75657

100.00

A. ervi Among plants

1

2.057

-0.01144

-0.46

Among sites within plants

4

14.235

0.02212

0.88

0.91436 0.04980

Within populations

308

769.205

2.49742

99.57

0.12554

Total

313

785.497

2.50810

100.00

Aphid hosts were randomly collected from the three Que´bec study sites, with host plant (i.e., alfalfa or red clover) as the top level population grouping factor

123

Evol Ecol (2013) 27:165–184

173

Table 2 Genetic differentiation according to pairwise FST values between all pairs of six A. pisum and A. ervi samples from Que´bec, corresponding to individuals pooled by host plant and site Alfalfa

Red clover

Lennoxville

Saint-Augustin

Montre´al

Lennoxville

0.00245NS

0.01286*

0.11249*

0.17212*

0.00749*

0.10137*

0.15487*

0.07024*

0.12933*

0.19010*

0.09865*

0.02287*

0.00924*

Saint-Augustin

A. pisum Alfalfa Montre´al Lennoxville Saint-Augustin Red clover Montre´al Lennoxville

0.07974*

0.02575*

A. ervi Alfalfa Montre´al Lennoxville Saint-Augustin Red clover Montre´al

0.01282NS

0.01371* NS

0.00798

0.00707NS

0.00343NS

-0.00344NS

NS

NS

0.00128NS

NS

0.00470

0.01596NS

0.00509NS

0.00973NS

0.01161

NS

0.00663

-0.00214

Lennoxville

0.00281NS

The asterisk indicates significance based on Bonferroni corrected critical p-values for alpha = 0.05. NS stands for non-significant

comparisons between host plants were all significant, being often one order of magnitude higher that those between distant sites involving the same host plant. In contrast with pea aphid populations, we did not detect any genetic signature of differentiation among the 157 A. ervi collected in the six field crops. Both cluster analyses failed to indicate that A. ervi genotypes could be grouped into K C 2 clusters (FLOCK: mean LLOD B2.15; v25 = 7.05; p C 0.22; STRUCTURE: highest ln (Prob) for K = 1). Moreover AMOVA (Arlequin 3.11, Table 1) indicated no significant effect of host plant (p = 0.91436), despite a slightly significant effect of site within plants (p = 0.0498). For the 70 wasps whose genotype was associated with their aphid host genotype, 45 wasps were from aphids of the alfalfa cluster and 25 from the red clover cluster. This grouping factor, representing the aphid host genetic type, did not influence the total variance in the AMOVA (p = 0.59). There was only slight evidence for A. ervi genetic differentiation among localities for alfalfa aphids (Montreal vs. St-Augustin FST = 0.01371, p = 0.03, Table 2), but there was evidence of inbreeding within each site (FIS = 0.11–0.14; p \ 0.001). Symbionts in field collected aphids and co-occurrence with parasitism Facultative symbiont determination was realized for 277 field-collected aphids (Supplementary Table 1). Single infections of R. insecticola were present in a much higher proportion (&79) in pea aphids collected on red clover. In comparison, we found H. defensa (alone or with R. insecticola or PAXS) in a significantly higher proportion (&3.59) in aphids on alfalfa than on red clover (0.60 ± 0.19 SD, 0.16 ± 0.19 SD, v22 = 32.38; p \ 0.001). S. symbiotica was found on all sites on red clover, but at very low frequency (Supplementary Table 1).

123

174

Evol Ecol (2013) 27:165–184

Mummies and emerging wasps assigned to A. ervi Haliday dominated by far (94 %) among parasitoid species in field-collected aphids, the additional 6 % being accounted by Praon sp. (3 %), and Lysiphlebus sp. plus unidentified hyperparasites (3 %). The overall A. ervi parasitism rate based on numbers of collected aphids forming mummies was 21.5 % (ranging from a mean of 19.1 % in Montre´al to 31.9 % in Lennoxville on both crops), which are minimum rates because sampling was not specifically designed to precisely estimate field parasitism. Close to 50 % of the field collected aphids that were parasitized by A. ervi reached maturity and produced progeny (70 out of 157 individuals), which were PCR screened for symbionts. They exhibit a rate of infection with presumably defensive bacteria (H. defensa, PAXS) of 49 % (31 out of 63 individuals; 7 Buchnera negative individuals were excluded), showing that many pea aphids in the field are not systematically protected from A. ervi by their symbionts. When comparing this rate of infection to the rate taking into account all collected aphids (n = 277), which is 40 %, no significant difference was observed (v21 = 1.906; 0.25 \ p \ 0.10). However, the infection rate with defensive symbionts for reproducing parasitized aphids was significantly higher on alfalfa (73 %; n = 27) than on clover (15 %; n = 4) (v21 = 10.29; 0.005 \ p \ 0.001), which is consistent with global symbiont occurrence (see above), H. defensa being more present on alfalfa than clover. FLOCK_AFLP was used to search for population structure among symbiotypes (i.e., cooccurrence of C2 facultative bacteria) for aphids that were both genotyped and determined for presence of the four facultative bacteria (n = 277). Among the eight observed symbiotypes, four groups were formed by FLOCK, which were tested with contingency Chisquares for association with the two main aphid genotype clusters recognized by FLOCK (Table 3). A strong association of aphid genotype cluster #2 (80 % clover aphids, Fig. 1) with aphid symbiotype #3 was found (p B 0.0001), showing that most clover pea aphids carried R. insecticola (characteristic of symbiotype group #3) (Table 3). Similar analysis for anti-parasitoid bacteria (H. defensa and PAXS, present in symbiotype groups #1 and #2) showed that they were strongly associated with the alfalfa aphids (p B 0.0001), i.e., the other main aphid genotype cluster. Host selection behaviour and host exploitation experiment Genotypes of the experimental pea aphid clones were representative of the genetic clusters associated with each host plant. The five alfalfa clones were included in the first cluster, clearly associated with alfalfa, while four of the red clover clones belonged to the second cluster. Of the 600 5-min tests, 14 (2 %) failed to provide any valuable information because of escape of some A. ervi females while being tested. Ovipositor contact was observed in 51.7 % of all completed tests (303/586). The rate of successful oviposition based on dissection (replicate aphids 1–3), was 71.0 % for the first aphid (27 parasitoids laid C1 eggs in 38 ovipositor contacts), 82.2 % for the second (37/45), and 83.7 % for the third one (41/49), a trend consistent with increasing tendency to attack during early encounters with hosts. Probability of ovipositor contact was strongly influenced by aphid clone, indicating high inter-clone variation for being selected as hosts by A. ervi (F9 76 = 5.00, p \ 0.001; Table 4). Color was also a strong determinant with nearly five times more contacts on green (0.82 ± 0.07) than on pink clones (0.17 ± 0.06; F1 76 = 42.82, p \ 0.001; Fig. 2b). Aphids from alfalfa were contacted significantly more frequently (0.63 ± 0.11) than red clover aphids (0.36 ± 0.10; F1 76 = 5.05, p = 0.028; Fig. 2c). Most notably, there was no

123

Evol Ecol (2013) 27:165–184

175

Table 3 Symbiont communities in A. pisum (N = 277) on alfalfa and red clover in Que´bec populations (2007), related with host genetic structure as revealed by FLOCK Symbiotype cluster (n)

Association strength (contingency v2 p value)

Symbiont species S. symbiotica

H. defensa

R. insecticola

PAXS

Site

12 (41)

0

1

0

1

B0.0001

2a3 (53)

0

1

0

0

0.13618

Aphid cluster1

Host plant

B0.0001

0.0008

B0.0001

B0.0001

2b (17)

0

1

1

0

0.55401

0.0016

0.1254

3 (105)

0

0

1

0

0.09469

B0.0001

B0.0001

44 (61)

0

0

0

0

Mixed (277)

– B0.0001

– B0.0001

– B0.0001

Associations of symbiotype groups with site and aphid genotype or host plant were statistically tested for significance using contingency Chi-square. p values in bold are highly significant 1

Genotype groups (K = 2) formed by FLOCK_MSAT and confirmed by STRUCTURE (see Fig. 1)

2

Five individuals in cluster 1 also hosted R. insecticola

3

One individual in cluster 2a also hosted S. symbiotica

4

Two individuals in cluster 4 hosted S. symbiotica

significant aphid host plant 9 parasitoid colony (i.e., aphid host of parasitoid) interaction (F1 76 = 1.75, p = 0.190), as expected under the hypothesis of behaviour-related HAD. We found a significant interaction between parasitoid colony and the aphid replicate order (F2 464 = 5.53, p = 0.004; Table 4), suggesting that attack tendency of wasps from clover aphids increased faster (Fig. 2a). We found no significant effect of pea aphid clone (F9 76 = 0.42, p = 0.915) on delay before ovipositor contact among the 300 tests where a successful ovipositor contact had occurred, nor any significant effects of other independent variables on this variable (p [ 0.90). The probability of emergence was significantly influenced by pea aphid clone (F9 40 = 2.25, p = 0.039; Table 4), and significantly higher (1.79) for clones from red clover (0.71 ± 0.08) than clones from alfalfa (0.42 ± 0.08; F1 40 = 5.32, p = 0.026; Fig. 3a). Successful emergence was also significantly higher for the clover aphid parasitoids (0.69 ± 0.08) than the alfalfa colony (0.45 ± 0.07; F1 40 = 4.14, p = 0.049; Fig. 3b). However, there was no significant interaction between wasp colony and the aphid host plant (F1 40 = 0.40, p = 0.529), as would be expected under the hypothesis of host resistance or host quality-driven HAD. We found no effect of aphid clone (F8 24 = 0.51, p = 0.836), nor any other independent variables in statistical modeling sex of emerging parasitoids (p = 0.84). We found no significant relationship between offspring sex of the mated wasps and attributes of their hosts, with female sex ratio varying on average between 0.27 and 0.82. Nevertheless, female sex ratio was 10–20 % lower in tests of wasps with aphids of the other host plant, suggesting that wasps facing aphids from the other crop might evaluate them as slightly inferior quality hosts.

Discussion Our aphid population genetic data show that at least two host plant (crop) specialized pea aphid biotypes are present over a wide area in Que´bec, one associated with alfalfa and the

123

176

Evol Ecol (2013) 27:165–184

Table 4 Statistical modeling (GLIMMIX) of probability of ovipositor contact by A. ervi on A. pisum in a 5 min test (n = 586), and probability of adult emergence from an aphid that was contacted (n = 166 tests with C 1 attack), as function of aphid clone (n = 10), color morph (green, pink), host plant (red clover, alfalfa), bacterial symbiont (R. insecticola, H. defensa), parasitoid colony (alfalfa, red clover), and replicate order of host presentation (i.e., 1–6) Effect

Ovipositor contact df

F value

Progeny emergence Pr [ F

df

F value

Pr [ F

Block

4

1.81

0.136

4

0.01

0.999

Clone

9

5.00

\0.001

9

2.25

0.039 0.026

Contrasts Host plant

1

5.05

0.028

1

5.32

Color morph

1

42.82

\0.001

1

1.38

0.248

H. defensa

1

3.42

0.068

1

2.76

0.105

R. insecticola

1

0.81

0.372

1

0.39

0.535

Colony (A. ervi)

1

0.01

0.958

1

4.14

0.049

Clone 9 colony

9

0.60

0.792

5

1.07

0.390

Contrasts Host plant 9 colony

1

1.75

0.190

1

0.40

0.529

Color morph 9 colony

1

1.14

0.288

1

0.00

0.962

H. defensa 9 colony

1

0.13

0.718

1

1.04

0.313

R. insecticola 9 colony

1

0.62

0.434

1

0.40

0.530

2

7.95

\0.001

18

0.38

0.990

Error1 Replicate Replicate 9 clone

76

40

Contrasts Replicate 9 host plant

2

0.01

0.988

Replicate 9 color morph

2

0.23

0.795 0.610

Replicate 9 H. defensa

2

0.50

Replicate 9 R. insecticola

2

0.50

0.738

2

5.53

0.004

Replicate 9 colony Error2

464

Corrected total

585

59

Block represents observation series (n = 5), where six aphids of each clone were exposed to a randomly selected wasp from one of two parasitoid colonies

other associated with red clover, as reported in the northeastern US (Via 1991b). According to the HAD hypothesis (Stireman et al. 2006), such divergence of the pea aphid as the main host of A. ervi could be reflected in divergence of the parasitoid. But in fact, our A. ervi genetic data as well as reciprocal transplant data on host selection behaviour and host exploitation, fail to support the HAD hypothesis for this parasitoid, which is closely associated with pea aphids. Local confirmation for genetic divergence between aphid populations sympatrically associated with alfalfa and red clover was needed, given geographical distance from previously known populations. Cluster analysis consistently identified two groups of pea aphids associated with red clover and alfalfa. Host plant explained much more variation in genetic structure than did geographic origin. This adds to evidence of widespread

123

Evol Ecol (2013) 27:165–184

177

Fig. 2 Statistical modeling of the probability (mean ± SD) of ovipositor contact of Aphidius ervi on Acyrthosiphon pisum during 5-min host selection tests (see Table 4) as a function of a interaction between replicate order of the test aphid and parasitoid plant origin, b pea aphid color morph, and c pea aphid’s host plant

Fig. 3 Probability of adult emergence (mean ± SD) of Aphidius ervi progeny, from an Acyrthosiphon pisum that was contacted by the wasp’s ovipositor in a 5-min test, as a function of a pea aphid’s host plant, and b host aphid of the parasitoid (see Table 4 for model structure and F and p values of factors)

sympatric, host plant related genetic divergence in A. pisum (e.g., Via 1991a; Hawthorne and Via 2001, Simon et al. 2003; Ferrari et al. 2006). Interestingly, pea aphids of the alfalfa genetic cluster were also collected in red clover fields, but conversely, aphid genotypes typical of red clover fields were not found in alfalfa. Similar results were obtained in France by Frantz et al. (2006), and these observations agree with those of Via et al. (2000) who found that fitness of pea aphid migrants from alfalfa to red clover is higher than fitness of migrants from red clover to alfalfa. This could be due to infection by R. insecticola, which is commonly associated with clover types, and also known in some cases to expand aphid diet breadth (see Tsuchida et al. 2004). When reexamining our results, we indeed found that the majority (67 %) of those atypical ‘‘migrant’’ clones were infected with R. insecticola, alone or in association with other symbionts. The remaining clones either had no facultative symbionts (24 %) or hosted other types of symbionts (9 %). There was no parallel genetic differentiation among A. ervi female wasps sampled on parasitized aphids from any of the three pairs of red clover and alfalfa fields, in contrast to their pea aphid hosts. Moreover, the lack of genetic differentiation extended to wasps whose host genotype could be directly assigned to the alfalfa or red clover aphid genetic type. We conclude that there is no genetic evidence at all for A. ervi HAD in these A. ervi populations, despite clear evidence for genetically based host plant differentiation of their pea aphid hosts.

123

178

Evol Ecol (2013) 27:165–184

Failure to detect parasitoid genetic differentiation either represents a biological fact, or may be due to a lack of power with our markers. Here, a main concern is that there may have been insufficient time since this wasp’s introduction to North America for natural populations to differentiate and adapt to pea aphid biotypes (Hufbauer 2001). Also, sympatric speciation with no physical barriers to gene flow requires strong and continuous disruptive selection on variable populations. This type of divergence is also favoured by assortative mating on each host plant and weak gene flow (Rice and Hostert 1993; Via 2001; Abrahamson and Blair 2007). While selection on A. ervi could be effective given sufficient genetic variation (Henter 1995), the selection regime may be inconsistent given that A. ervi in North America also, can use other hosts than pea aphids (Hufbauer 2001). High dispersal may counteract any host driven divergence that might temporarily occur locally in A. ervi. Indeed, Aphidius wasps or their early stages in parasitized alate aphids, can disperse over large distances (Cameron et al. 1981; Langhof et al. 2005 and references therein). This is clearly evidenced for A. ervi from its rapid displacement of the then recently introduced A. smithi, as the principal parasitoid of pea aphids in the Northeastern US, soon after release in 1959 (Angalet and Fuester 1977). In our aphid samples A. smithi was not detected among mummies that formed and emerged, A. ervi and rare Praon sp. and Lysiphlebus sp. being the only primary parasitoid wasps found. The biological significance of apparent lack of genetic differentiation driven by pea aphid host plant specialization is also difficult to interpret on the basis of past studies on parasitoid/host parallel structure, given their idiosyncratic results. Despite evidence for some systems (e.g., Stireman et al. 2006; Forbes et al. 2009), failure to detect genetic structure was reported for highly specialist parasitoids where HAD was expected, such as for Aleiodes nolophanae parasitizing green cloverworm moth (Plathypena scabra) (Medina 2005). Our results are similar to other studies on Aphidius species where no evidence of HAD was found (Daza-Bustamante et al. 2002; Lozier et al. 2009). The neutral A. ervi markers used were reasonably variable (2–23 alleles per locus, Supplementary Table 2) and Hufbauer et al. (2004) were successful at detecting A. ervi differentiation between European and North American populations. The large, mobile populations of the parasitoid wasps might, however, decelerate genetic drift, as suggested by the generally low FST values. It is also not excluded that the spatial scale of this study was too small (distance between sites: 160–240 km) and/or that many more markers would be necessary to detect ongoing differentiation. Most studies on our host-parasitoid system preceded the discovery of the potential role of aphid bacterial symbionts in their relationships, which we clearly acknowledge in this study. Single infections by R. insecticola, which was common in our samples, were over seven times higher on aphids collected on red clover than on alfalfa aphids (see presence of R. insecticola in Table 3), which is consistent with several studies elsewhere (Simon et al. 2003; Ferrari et al. 2004; Frantz et al. 2009). There is evidence that R. insecticola increases pea aphid fitness on clover, in some pea aphid genotypes (Leonardo and Muiru 2003; Tsuchida et al. 2004; see Oliver et al. 2010). Thus our findings add further support to the hypothesis that this symbiont contributes to pea aphid fitness on clover, although aphid 9 symbiont genotype interactions are suspected (Janson et al. 2008; Oliver et al. 2010). We found the H. defensa symbiont alone or with other symbionts (e.g., PAXS) more frequently on alfalfa pea aphids (about 39) than clover aphids, which agrees with similar findings elsewhere (Simon et al. 2003; Frantz et al. 2009). Because H. defensa has been linked to Aphidius parasitoid resistance in other aphids (Vorburger et al. 2009) and especially in A. pisum (Oliver et al. 2003, 2005; Ferrari et al. 2004; Guay et al. 2009), its

123

Evol Ecol (2013) 27:165–184

179

occurrence here, in field parasitized aphids is particularly interesting. On average 45 % of all mummified aphids produced progeny that were infected with ‘defensive’ symbionts inherited from their parasitized mother before she died (details not shown). Therefore, H. defensa (alone or in multiple symbiosis) in pea aphid was ineffective or partly effective in preventing A. ervi field parasitism. Still, reproduction of a substantial fraction of the successfully parasitized aphids can preserve the genotypes of both the pea aphid host and its defensive symbionts, while the parasitized aphid dies to produce an A. ervi wasp. In natural populations, this may considerably expand the time scale of any arms race between the host and the parasitoid. Our work suggests that selection pressures exerted by A. ervi on its host (and beneficial symbionts) in this system can be lower than generally assumed (e.g., Henter 1995, p. 439). This is consistent with Henter and Via (1995) who could find no field evidence of pea aphid clonal selection over a season (*5–6 generations of asexual reproduction) in alfalfa pea aphid in New York, despite relatively high rates of parasitism. An obvious explanation is that the aphid individual does not represent the unit of selection, which is the aphid clone to which it belongs. In this respect, even genes of the parasitized aphid itself and its symbionts often survive in born progeny of parasitized aphids as we found, although the extent to which this might affect coevolutionary interactions remains to be determined. We found no further support for HAD in the parasitoid host selection behaviour and host exploitation experiment. Interaction between wasp colony and pea aphid biotype was expected under HAD, but it had no significant effects on probability of ovipositor contact, delay before contact, survival to adult emergence, and sex ratio. This pattern cannot be explained by lack of power, as direct effects of most factors were found on most response variables. The higher probability of ovipositor contact (about 29, Fig. 2a) on pea aphid clones from alfalfa than clover indicates that irrespective of their host origin, wasps generally preferred the alfalfa aphids. Our no choice test assumes that wasps naturally search for aphids within clonal colonies where choice is limited to aphid sizes or stages, which we controlled for. This might be the outcome of evolutionary history of A. ervi in eastern North America, which was introduced from collections made on alfalfa in France (Mackauer and Finlayson 1967; Angalet and Fuester 1977; Hufbauer 2002). However, the opposite pattern was observed for survival and adult emergence, which was lower on aphids from alfalfa than red clover, for both wasp colonies (Fig. 3a). This intriguingly suggests contrasting A. ervi traits of compatibility from the wasp behavioural versus the host exploitation perspectives (immature stages), with the clover aphids being the best to exploit, but the least preferred by the wasps. Pea aphid color strongly affected wasp behaviour, with ovipositor contacts being four times higher on green than pink aphids. Similarly, Libbrecht et al. (2007) found in a binary choice test that A. ervi preferentially selected green over pink pea aphids. Most field studies reported higher parasitism on green than pink pea aphids (Losey et al. 1997; Powell et al. 1998; Langley et al. 2006), except Henter and Via (1995) who found the opposite on alfalfa. In reference to close range behaviour, we conclude that irrespective of aphid or parasitoid host plant origin, green pea aphids are intrinsically more attractive to A. ervi wasps than pink morphs. We speculate that host selection on the basis of aphid color might also contribute to absence of detectable parasitoid genetic differentiation. Wasps from red clover aphids increased ovipositor contacts rate faster than females of the alfalfa population (Fig. 2a) suggesting learning, which has already been reported in A. ervi (Langley et al. 2006; Takemoto et al. 2009). More work is needed to explore how behaviour plasticity affects A. ervi’s potential for coevolution, and also considering the

123

180

Evol Ecol (2013) 27:165–184

cues from the plant–host complex which are used by aphid parasitoids to select their hosts (Battaglia et al. 1995; Powell et al. 1998). To summarize, we detected a strong genetic structure associated with plant origin in pea aphid populations from Que´bec and identified two host races, one associated with alfalfa and the other with red clover. We also found that aphids from alfalfa mostly hosted antiparasitoid defensive symbionts (H. defensa, PAXS), while those from clover aphids were primarily infected by the symbiont R. insecticola. However, contrarily to the HAD hypothesis, we found no obvious genetic and behavioural differentiation between A. ervi populations parasitizing either alfalfa or clover specialized pea aphids in Que´bec. Many factors may explain the absence of A. ervi population divergence in relation with its plant specialized pea aphid hosts, including high rates of dispersal that may prevent ecological specialization, short evolutionary history of A. ervi–A pisum interactions in North America, little difference in host quality between pea aphid races and variation in host availability across time and space. Acknowledgments This study was supported by a NSERC Discovery grant to Conrad Cloutier. We thank summer students (Sophie Laliberte´, Joseph Moisan-De Serres, Anne Bogeto, Aure´lie Guy and Claudia Zdenka-Ficher) for technical help with aphid and parasitoid collecting and rearing. We also thank Lucie Mieuzet for pea aphid genotyping and help with aphid symbiont detection, Je´roˆme Lemaıˆtre for help with data analysis, Pierre Duchesne from Universite´ Laval for help with cluster analysis, Gae´tan Daigle for help with AMOVA and Marie-Claude Gagnon for help with genetic analysis.

References Abrahamson AG, Blair CP (2007) Sequential radiation through host race formation: herbivore diversity leads to diversity in natural enemies. In: Tilmon KJ (ed) Specialization, speciation, and radiation: the evolutionary biology of herbivorous insects. University of California Press, Berkeley, pp 188–202 Aljanabi SM, Martinez I (1997) Universal and rapid salt-extraction of high quality genomic DNA for PCRbased techniques. Nucl Acids Res 25:4692–4693 Allison PD (1999) Logistic regression using the SAS system—theory and application. SAS Institute Inc, Cary Almohamad R, Verheggen FJ, Francis F et al (2008) Discrimination of parasitized aphids by a hoverfly predator: effects on larval performance, foraging, and oviposition behavior. Entomol Exp Appl 128:73–80 Angalet GW, Fuester R (1977) The Aphidius parasites of the pea aphid Acyrthosiphon pisum in the eastern half of the United States of America. Ann Entomol Soc Am 70:87–96 Barrette M, Wu GM, Brodeur J et al (2009) Testing competing measures of profitability for mobile resources. Oecologia 158:757–764 Battaglia D, Pennacchio F, Romano A et al (1995) The role of physical cues in the regulation of host recognition and acceptance behavior of Aphidius ervi Haliday (Hymenoptera, Braconidae). J Insect Behav 8:739–750 Baumann P, Baumann L, Lai CY et al (1995) Genetics, physiology and evolutionary relationships of the genus Buchnera: intracellular symbionts of aphids. Ann Rev Microbiol 49:55–94 Blackman RL, Eastop VF (1984) Aphids on the world’s crops. Interscience Publication, Toronto Blair CP, Abrahamson WG, Jackman JA et al (2005) Cryptic speciation and host-race formation in a purportedly generalist tumbling flower beetle. Evolution 59:304–316 Bourguet D, Bethenod MT, Trouve C et al (2000) Host-plant diversity of the European corn borer (Ostrinia nubilalis): what value for sustainable transgenic insecticidal Bt maize? Proc R Soc B 267:1177–1184 Caillaud MC, Via S (2000) Specialized feeding behavior influences both ecological specialization and assortative mating in sympatric host races of pea aphids. Am Nat 156:606–621 Cameron PJ, Walker GP, Allan DJ (1981) Establishment of the introduced parasite Aphidius eadyi (Hymenoptera: Aphidiidae) in the north island of New Zealand, and its initial effect on pea aphid. N Z J Zool 8:105–112

123

Evol Ecol (2013) 27:165–184

181

Caro T (2007) Behavior and conservation: a bridge too far? Trends Ecol Evol 22:394–400 Charnov EL, Losdenhartogh RL, Jones WT et al (1981) Sex ratio evolution in a variable environment. Nature 289:27–33 Claridge MF, Denhollander J, Morgan JC (1985) The status of weed-associated populations of the brown planthopper, Nilaparvata lugens (Sta˚l)—host race or biological species. Zool J Linn Soc 84:77–90 Cloutier C, Levesque CA, Eaves DM et al (1991) Maternal adjustment of sex-ratio in response to host size in the aphid parasitoid Ephedrus californicus. Can J Zool 69:1489–1495 Cloutier C, Duperron J, Tertuliano M et al (2000) Host instar, body size and fitness in the koinobiotic parasitoid Aphidius nigripes. Entomol Exp Appl 97:29–40 Colinet H, Salin C, Boivin G (2005) Host age and fitness-related traits in a koinobiont aphid parasitoid. Ecol Entomol 30:473–479 Cronin JT, Abrahamson AG (2001) Do parasitoids diversify in response to host-plant shifts by herbivorous insects? Ecol Entomol 26:347–355 Daza-Bustamante P, Fuentes-Contreras E, Rodriguez LC et al (2002) Behavioural differences between Aphidius ervi populations from two tritrophic systems are due to phenotypic plasticity. Entomol Exp Appl 104:321–328 Dion E, Ze´le´ F, Simon JC et al (2011) Rapid evolution of parasitoids when faced with the symbiontmediated resistance of their hosts. J Evol Biol 24:741–750 Dorchin N, Scott ER, Clarkin CE et al (2009) Behavioural, ecological and genetic evidence confirm the occurrence of host-associated differentiation in goldenrod gall-midges. J Evol Biol 22:729–739 Duchesne P, Turgeon J (2009) FLOCK: a method for quick mapping of admixture without source samples. Mol Ecol Resour 9:1333–1344 Ehrlich P, Raven P (1964) Butterflies and plants: a study in coevolution. Evolution 18:586–608 Emelianov I, Hernandes-Lopez A, Torrence M et al (2011) Fusion–fission experiments in Aphidius: evolutionary split without isolation in response to environmental bimodality. Heredity 106:798–807 Excoffier L, Smouse PE, Quattro JM (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131:479–491 Excoffier L, Laval G, Schneider S (2005) Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol Bioinform Online 1:47–50 Farrell B (1988) ‘‘Inordinate fondness’’ explained: why are there so many beetles? Science 281:555–559 Feder JL, Forbes A (2010) Sequential speciation and the diversity of parasitic insects. Ecol Entomol 35:67–76 Feder JL, Chilcote CA, Bush GL (1988) Genetic differentiation between sympatric host races of the apple maggot fly Rhagoletis pomonella. Nature 336:61–64 Ferrari J, Godfray HCJ (2006) The maintenance of intraspecific biodiversity: the interplay of selection on resource use and on natural enemy resistance in the pea aphid. Ecol Res 21:9–16 Ferrari J, Darby AC, Daniell TJ et al (2004) Linking the bacterial community in pea aphids with host-plant use and natural enemy resistance. Ecol Entomol 29:60–65 Ferrari J, Godfray HCJ, Faulconbridge AS et al (2006) Population differentiation and genetic variation in host choice among pea aphids from eight host plant genera. Evolution 60:1574–1584 Forbes AA, Powell THQ, Stelinski LL et al (2009) Sequential sympatric speciation across trophic levels. Science 323:776–779 Frantz A, Plantegenest M, Mieuzet L et al (2006) Ecological specialization correlates with genotypic differentiation in sympatric host-populations of the pea aphid. J Evol Biol 19:392–401 Frantz A, Calcagno V, Mieuzet L et al (2009) Complex trait differentiation between host-populations of the pea aphid Acyrthosiphon pisum (Harris): implications for the evolution of ecological specialisation. Biol J Linn Soc 97:718–727 Fuller WA, Kim JK (2005) Hot deck imputation for the response model. Survey Methodology (Statistics Canada, Catalogue No. 12-001) 31:139–149 Gerling D, Roitberg BD, Mackauer M (1990) Instar-specific defense of the pea aphid, Acyrthosiphon pisum—influence on oviposition success of the parasite Aphelinus asychis (Hymenoptera, Aphelinidae). J Insect Behav 3:501–514 Guay JF, Boudreault S, Michaud D et al (2009) Impact of environmental stress on aphid clonal resistance to parasitoids: role of Hamiltonella defensa bacterial symbiosis in association with a new facultative symbiont of the pea aphid. J Insect Physiol 55:919–926 Hales DF, Tomiuk J, Wohrmann K et al (1997) Evolutionary and genetic aspects of aphid biology: a review. Eur J Entomol 94:1–55 Hawthorne DJ, Via S (2001) Genetic linkage of ecological specialization and reproductive isolation in pea aphids. Nature 412:904–907

123

182

Evol Ecol (2013) 27:165–184

Henry LM, Gillespie DR, Roitberg BD (2005) Does mother really know best? Oviposition preference reduces reproductive performance in the generalist parasitoid Aphidius ervi. Entomol Exp Appl 116:167–174 Henry LM, Ma BO, Roitberg BD (2009) Size-mediated adaptive foraging: a host-selection strategy for insect parasitoids. Oecologia 161:433–445 Henter HJ (1995) The potential for coevolution in a host-parasitoid system. II. Genetic variation within a population of wasps in the ability to parasitize an aphid host. Evolution 49:439–445 Henter HJ, Via S (1995) The potential for coevolution in a host-parasitoid system. I. Genetic variation within an aphid population in susceptibility to a parasitic wasp. Evolution 49:427–438 Hufbauer RA (2001) Pea aphid-parasitoid interactions: have parasitoids adapted to differential resistance? Ecology 82:717–725 Hufbauer RA (2002) Evidence for nonadaptive evolution in parasitoid virulence following a biological control introduction. Ecol Appl 12:66–78 Hufbauer RA, Via S (1999) Evolution of an aphid-parasitoid interaction: variation in resistance to parasitism among aphid populations specialized on different plants. Evolution 53:1435–1445 Hufbauer RA, Bogdanowicz SM, Perez L et al (2001) Isolation and characterization of microsatellites in Aphidius ervi (Hymenoptera: Braconidae) and their applicability to related species. Mol Ecol Notes 1:197–199 Hufbauer RA, Bogdanowicz SM, Harrison RG (2004) The population genetics of a biological control introduction: mitochondrial DNA and microsatellite variation in native and introduced populations of Aphidius ervi, a parasitoid wasp. Mol Ecol 13:337–348 Janson E, Stireman JO, Singer MS et al (2008) Phytophagous insect-microbe mutualisms and adaptive evolutionary diversification. Evolution 62:997–1012 Kunert G, Belz E, Simon JC et al (2010) Differences in defensive behaviour between host-adapted races of the pea aphid. Ecol Entomol 35:147–154 Langhof M, Meyhofer R, Poehling HM et al (2005) Measuring the field dispersal of Aphidius colemani (Hymenoptera: Braconidae). Agric Ecosyst Environ 107:137–143 Langley SA, Tilmon KJ, Cardinale BJ et al (2006) Learning by the parasitoid wasp, Aphidius ervi (Hymenoptera: Braconidae), alters individual fixed preferences for pea aphid color morphs. Oecologia 150:172–179 Le Ralec A, Anselme C, Outreman Y et al (2010) Evolutionary ecology of the interactions between aphids and their parasitoids. C R Biol 333:554–565 Leonardo TE, Muiru GT (2003) Facultative symbionts are associated with host plant specialization in pea aphid populations. Proc R Soc B 270:S209–S212 Libbrecht R, Gwynn DM, Fellowes MDE (2007) Aphidius ervi preferentially attacks the green morph of the pea aphid, Acyrthosiphon pisum. J Insect Behav 20:25–32 Losey JE, Ives AR, Harmon J et al (1997) A polymorphism maintained by opposite patterns of parasitism and predation. Nature 388:269–272 Lozier JD, Roderick GK, Mills NJ (2009) Molecular markers reveal strong geographic, but not host associated, genetic differentiation in Aphidius transcaspicus, a parasitoid of the aphid genus Hyalopterus. Bull Entomol Res 99:83–96 Mackauer M (1971) Acyrthosiphon pisum (Harris), pea aphid (Homoptera: Aphididae). In: Biological control programmes against insects and weeds in Canada 1959–1968. Technical communication no 4. Commonwealth Institute in Biological Control, Trinidad, pp 3–10 Mackauer M (1997) Growth and development in parasitoid wasps. In: Dettner K, Bauer G, Vo¨lkl W (eds) Vertical food web interactions: evolutionary patterns and driving forces. Ecological studies, vol 130. Springer, Berlin, pp 191–203 Mackauer M, Finlayson T (1967) The hymenopterous parasites (Hymenoptera: Aphidiidae) of the pea aphid in eastern North America. Can Entomol 99:1051–1082 Mackauer M, Michaud JP, Vo¨lkl W (1996) Host choice by aphidiid parasitoids (Hymenoptera: Aphidiidae): host recognition, host quality, and host value. Can Entomol 128:959–980 Medina RF (2005) The role of host-plant species in the differentiation of sympatric populations of hymenopteran parasitoids. Dissertation, University of Maryland, College Park Michaud JP, Mackauer M (1994) The use of visual cues in host evaluation by aphidiid wasps. I. Comparison between three Aphidius parasitoids of the pea aphid. Entomol Exp Appl 70:273–283 Mitter C, Farrell B, Weigmann B (1988) The phylogenetic study of adaptive zones: has phytophagy promoted insect diversification? Am Nat 132:107–128 Moran NA, Dunbar HE (2006) Sexual acquisition of beneficial symbionts in aphids. Proc Natl Acad Sci USA 103:12803–12806

123

Evol Ecol (2013) 27:165–184

183

Morris RJ, Fellowes MDE (2002) Learning and natal host influence host preference, handling time and sex allocation behaviour in a pupal parasitoid. Behav Ecol Sociobiol 51:386–393 Morris DW, Kotler BP, Brown JS et al (2009) Behavioral indicators for conserving mammal diversity. Ann NY Acad Sci 1162:334–356 Nguyen TTA, Boudreault S, Michaud D et al (2008) Proteomes of the aphid Macrosiphum euphorbiae in its resistance and susceptibility responses to differently compatible parasitoids. Insect Biochem Mol Biol 38:730–739 Oliver KM, Russell JA, Moran NA et al (2003) Facultative bacterial symbionts in aphids confer resistance to parasitic wasps. Proc Natl Acad Sci USA 100:1803–1807 Oliver KM, Moran NA, Hunter MS (2005) Variation in resistance to parasitism in aphids is due to symbionts not host genotype. Proc Natl Acad Sci USA 102:12795–12800 Oliver KM, Moran NA, Hunter MS (2006) Costs and benefits of a superinfection of facultative symbionts in aphids. Proc R Soc B 273:1273–1280 Oliver KM, Degnan PH, Burke GR et al (2010) Facultative symbionts of aphids and the horizontal transfer of ecologically important traits. Annu Rev Entomol 55:247–266 Peccoud J, Figueroa CC, Silva AX et al (2008) Host range expansion of an introduced insect pest through multiple colonizations of specialized clones. Mol Ecol 17:4608–4618 Peccoud J, Ollivier A, Plantegenest M et al (2009a) A continuum of genetic divergence from sympatric host races to species in the pea aphid complex. Proc Natl Acad Sci USA 106:7495–7500 Peccoud J, Simon JC, McLaughlin HJ et al (2009b) Post-Pleistocene radiation of the pea aphid complex revealed by rapidly evolving endosymbionts. Proc Natl Acad Sci USA 106:16315–16320 Powell W, Pennacchio F, Poppy G et al (1998) Strategies involved in the location of hosts by the parasitoid Aphidius ervi Haliday (Hymenoptera: Braconidae: Aphidiinea). Biol Control 11:104–112 Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155:945–959 Rice WR (1989) Analyzing tables of statistical tests. Evolution 43:223–225 Rice WR, Hostert EE (1993) Laboratory experiments on speciation—what have we learned in 40 years? Evolution 47:1637–1653 Roderick GK, Navajas M (2003) Genes in new environments: genetics and evolution in biological control. Nat Rev Genet 4:889–899 Russell JA, Moran NA (2005) Horizontal transfer of bacterial symbionts: heritability and fitness effects in a novel aphid host. Appl Environ Microbiol 71:7987–7994 Sandstro¨m J (1996) Temporal changes in host adaptation in the pea aphid, Acyrthosiphon pisum. Ecol Entomol 21:56–62 Sandstro¨m JP, Russell JA, White JP et al (2001) Independent origins and horizontal transfer of bacterial symbionts of aphids. Mol Ecol 10:217–228 Schluter D (2000) The ecology of adaptative radiation. Oxford University Press, Oxford Schwo¨rer U, Vo¨lkl W (2001) Foraging behavior of Aphidius ervi (Haliday) (Hymenoptera: Braconidae: Aphidiinae) at different spatial scales: resource utilization and suboptimal weather conditions. Biol Control 21:111–119 Sequeira R, Mackauer M (1992) Nutritional ecology of an insect host-parasitoid association: the pea aphid—Aphidius ervi system. Ecology 73:183–189 Sequeira R, Mackauer M (1993) Seasonal variation in body size and offspring sex ratio in field populations of the parasitoid wasp, Aphidius ervi (Hymenoptera, Aphidiidae). Oikos 68:340–346 Simon JC, Carre´ S, Boutin M et al (2003) Host-based divergence in populations of the pea aphid: insights from nuclear markers and the prevalence of facultative symbionts. Proc R Soc B 270:1703–1712 Simon JC, Sakurai M, Bonhomme J et al (2007) Elimination of a specialised facultative symbiont does not affect the reproductive mode of its aphid host. Ecol Entomol 32:296–301 Stireman JO, Nason JD, Heard SB et al (2006) Cascading host-associated genetic differentiation in parasitoids of phytophagous insects. Proc R Soc B 273:523–530 Sunnucks P, England PR, Taylor AC et al (1996) Microsatellite and chromosome evolution of parthenogenetic Sitobion aphids in Australia. Genetics 144:747–756 Takemoto H, Powell W, Pickett J et al (2009) Learning is involved in the response of parasitic wasps Aphidius ervi (Haliday) (Hymenoptera: Braconidae) to volatiles from a broad bean plant, Vicia faba (Fabaceae), infested by aphids Acyrthosiphon pisum (Harris) (Homoptera: Aphididae). Appl Entomol Zool 44:23–28 Tsuchida T, Koga R, Shibao H et al (2002) Diversity and geographic distribution of secondary endosymbiotic bacteria in natural populations of the pea aphid, Acyrthosiphon pisum. Mol Ecol 11:2123–2135 Tsuchida T, Koga R, Fukatsu T (2004) Host plant specialization governed by facultative symbiont. Science 303:1989

123

184

Evol Ecol (2013) 27:165–184

Via S (1991a) Specialized host plant performance of pea aphid clones is not altered by experience. Ecology 72:1420–1427 Via S (1991b) The genetic structure of host plant adaptation in a spatial patchwork - demographic variability among reciprocally transplanted pea aphid clones. Evolution 45:827–852 Via S (1999) Reproductive isolation between sympatric races of pea aphids. I. Gene flow restriction and habitat choice. Evolution 53:1446–1457 Via S (2001) Sympatric speciation in animals: the ugly duckling grows up. Trends Ecol Evol 16:381–390 Via S (2009) Natural selection in action during speciation. Proc Natl Acad Sci USA 106:9939–9946 Via S, Bouck AC, Skillman S (2000) Reproductive isolation between divergent races of pea aphids on two hosts. II. Selection against migrants and hybrids in the parental environments. Evolution 54:1626–1637 Vorburger C, Sandrock C, Gouskov A et al (2009) Genotypic variation and the role of defensive endosymbionts in an all-parthenogenetic host-parasitoid interaction. Evolution 63:1439–1450 Vorburger C, Gehrer L, Rodriguez P (2010) A strain of the bacterial symbiont Regiella insecticola protects aphids against parasitoids. Biol Lett 6:109–111 Zilber-Rosenberg I, Rosenberg E (2008) Role of microorganisms in the evolution of animals and plants: the hologenome theory of evolution. FEMS Microbiol Rev 32:723–735

123