Figures
Abstract
The oriental fruit moth Grapholita ( = Cydia) molesta is a key fruit pest globally. Despite its economic importance, little is known about its population genetics in its putative native range that includes China. We used five polymorphic microsatellite loci and two mitochondrial gene sequences to characterize the population genetic diversity and genetic structure of G. molesta from nine sublocations in three regions of a major fruit growing area of China. Larval samples were collected throughout the season from peach, and in late season, after host switch by the moth to pome fruit, also from apple and pear. We found high numbers of microsatellite alleles and mitochondrial DNA haplotypes in all regions, together with a high number of private alleles and of haplotypes at all sublocations, providing strong evidence that the sampled area belongs to the origin of this species. Samples collected from peach at all sublocations were geographically structured, and a significant albeit weak pattern of isolation-by-distance was found among populations, likely reflecting the low flight capacity of this moth. Interestingly, populations sampled from apple and pear in the late season showed a structure differing from that of populations sampled from peach throughout the season, indicating a selective host switch of a certain part of the population only. The recently detected various olfactory genotypes in G. molesta may underly this selective host switch. These genetic data yield, for the first time, an understanding of population dynamics of G. molesta in its native range, and of a selective host switch from peach to pome fruit, which may have a broad applicability to other global fruit production areas for designing suitable pest management strategies.
Citation: Zheng Y, Peng X, Liu G, Pan H, Dorn S, Chen M (2013) High Genetic Diversity and Structured Populations of the Oriental Fruit Moth in Its Range of Origin. PLoS ONE 8(11): e78476. https://doi.org/10.1371/journal.pone.0078476
Editor: Nicholas J. Mills, University of California, Berkeley, United States of America
Received: June 4, 2013; Accepted: September 12, 2013; Published: November 4, 2013
Copyright: © 2013 Zheng et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by grants of the National Natural Science Foundation of China (No. 31071687 and No. 31272036), the Joint Project of Northwest A&F University and ETH Zurich, the Talent Recruitment Project of Northwest A&F University, the National Key Technology R&D Program (No. 2012BAK11B03)from the Ministry of Science and Technology, the Doctoral Fund (No. 20110204110001) from the Ministry of Education of China and the Scientific Research Foundation for Returned of Ministry of Education of China and International Exchange Project of Northwest A&F University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In insect field populations, genetic diversity and genetic structure can be affected by various factors, such as range of origin as opposed to invaded regions, geographical distance and host plant species [1]–[3]. Some studies have documented that genetic diversity and structure are associated with the dispersal ability of the insect concerned [4]–[6]. Generally, populations in their native range show a particularly high genetic diversity, whilst populations in invaded regions usually show reduced genetic diversity [7], [8].
An insect species of global importance is the oriental fruit moth, Grapholita ( = Cydia) molesta Busck (Lepidoptera: Tortricidae), a key pest of rosaceae fruit trees wordwide [9]. Its putative native range includes China, where it is thought to have co-evolved with its primary host peach [9], [10], but little is known yet about its population genetics in this area. This moth completes three to seven generations per year depending on latitude and elevation [11], [12]. Remarkably, it makes a host switch from peach to pome fruit in the late part of the season, particularly after peach harvest, in several areas of the world including China [11], [12], Italy [13], [14], Southern Switzerland [15] and USA [16]. The changing abiotic environment in the late season with decreasing daylight period experienced in the larval stage promotes flight performance in this species [17]. In addition, changing compositions of volatile blends emitted by its primary host peach and its secondary hosts apple and pear further contribute to this host switch, as volatiles from the pome fruit apple and pear become particularly attractive to female moths in the late season [18]–[21]. Despite the importance of the seasonal dynamics given by this host switch, consequences on patterns of population structure in the oriental fruit moth are yet unknown.
Dispersal of G. molesta is considered limited based on estimation of its flight range [22], [23], though a proportion of individuals within a population, in particular females, have the ability to disperse between orchards [17]. It is expected that G. molesta individuals have an ecological advantage when they remain and oviposit within the orchard where they emerge, unless environmental change triggers displacement of some individuals [4]. Whereas no detailed information is available on population structure in the native range, structured populations of G. molesta have been reported twice from invaded countries. In South Africa, G. molesta populations collected from closely situated orchards (<1 km) could be distinguished [4], and significant intra-regional structure among five populations from eight sampling sites were reported for this species in Italy [24]. A recent paper analyzed the invasion routes of G. molesta on different continents, and documented that populations of this species were geographically structured on a continental scale [10]. Three populations from China were used in that above work to illustrate Asian populations, however, they were not representative of a broad geographic sampling range or of a main fruit growing region in this country.
Population genetic studies can reveal the micro-evolution and ecological adaption strategies of an insect pest in agroecosystems [3], and thereby greatly advance our understanding of its genetic diversity and genetic structure. Such knowledge is important for the design and optimization of sustainable pest management strategies [4], [5], [24]. Our objective was to investigate the genetic diversity of G. molesta in its putative range of origin and its genetic structure in a key fruit growing area of China, as well as the consequence of seasonal dynamics with host switch from primary to secondary hosts. We analyzed samples collected from peach throughout the season and from pome fruit after host switch in the late season, using a combination of microsatellite data and mitochondrial gene sequences. We hypothesized that populations will show (a) a number of typical characteristics reflecting a native species, (b) a genetic structure between different regions and possibly also between sublocations within these regions, reflecting the limited flight capacity of this moth species, and (c) little genetic structure across the season irrespective of the host infested.
Materials and Methods
Ethics Statement
No specific permissions were required for the described field studies for this wide spread agriculture pest. We confirm that the locations were not privately owned or protected in any way. The field studies did not involve endangered or protected species.
Insect sampling
Oriental fruit moth larval samples were collected in 2011 from a major Chinese fruit growing area, where this species is found on peach until peach harvest, which is only in exceptional cases after July, and on the pome fruit apple and pear, where it is found nearly exclusively in the late season, i.e. in August. Samples from 9 sublocatons in 3 regions were used (Fig. 1), and all orchards sampled (>5 ha) consisted of trees of the same host plant species. To test for potential seasonal variations, samples were taken at different times from the same orchard wherever possible (Table 1). The distance between sampled fruit trees was at least 5 m, and only one larva per tree was used to minimize sibling collection. At least 50 2nd to 3rd instar larvae were collected per population (sublocation, host species and time; Table 1), and preserved in 10 ml Falcon tubes filled with ethanol.
The codes for sublocations are explained in Table 1. (A) map of China (dashed box indicates the sampling regions); (B) enlarged map of sampling regions (real line box indicates three sampling regions); sampling sublocations in (C) YanTai region; (D) WeiFang region; (E) TaiAn region.
DNA extraction
Genomic DNA was extracted from 8–10 mg of larval material using the DNeasy Tissue Kit (Qiagen, Hilden, Germany). Extraction was performed according to the bench protocol for animal tissues. DNA was eluted in TE buffer and stored at −20°C.
Microsatellite amplification and genotyping
Ten previously reported microsatellite markers [24] were used in all the samples. Data of five loci (Gm01, Gm03, Gm04, Gm05 and Gm07) had to be rejected either because the loci were monomorphic (Gm01, Gm04, and Gm05), or because of the presence of stutter peaks and/or imperfect repeats (Gm03 and Gm07) in genotyping results of many samples. Five loci (Gm02, Gm06, Gm08, Gm09 and Gm10) that were highly polymorphic were selected for further analysis. The forward primer of each pair was labeled at the 5′ end with FAM fluorescent dye [25]. Amplifications were carried out in a total volume of 25 µl, containing 1×PCR amplification buffer (Takara, Dalian, China), 2 µM MgCl2, 0.2 µM of each dNTP (Takara, Dalian, China), 0.5 µl of each forward primer (2 µM), 2 µl of each reverse primer (2 µM), 2 µl M13 primer (2 µM), 1.0 U Taq polymerase (Takara, Dalian, China), and 2 µl genomic DNA (10–30 ng/µl). Amplification comprised a 10 minutes initial denaturation at 95°C, followed by 35 cycles of 30 seconds at 95°C, 45 seconds annealing at a primer-specific temperature, and 45 seconds at 72°C, then 10 cycles of 30 seconds at 95°C, 45 seconds at 53°C and 45 seconds at 72°C, and a final step 10 minutes at 72°C. To examine the length and genotype of the amplified PCR products, an ABI3730XL automated DNA sequencer (Applied Biosystems, Foster City, CA, USA) and GENESCAN version 4.0 (Applied Biosystems, Foster City, CA, USA) were used.
Mitochondrial DNA amplification
The 740 bp mitochondrial cytochrome oxidase subunit I (COI) was amplified with the primer pairs C1-J-2183 (5′-CAACATTTATTTTGATTTTTTGG -3′) and TL2-N-3014 (5′-TCCAATGCACTAATCTGCCATATTA -3′) [26], while 700 bp cytochrome oxidase subunit II (COII) was amplified using the primer pairs TL2-J-3037 (5′-ATGGCAGATTATATGTAATGG-3′) and TK-N-3785 (5′-GTTTAAGAGACCAGTACTTG-3′) [26]. Amplifications were carried out in a total volume of 25 µl, containing 1×PCR amplification buffer (Takara, Dalian, China), 2 µM MgCl2, 0.2 µM of each dNTP (Takara, Dalian, China), 3.0 µl of each oligonucleotide primer (2 µM), 2 µl genomic DNA (10–30 ng/µl). The thermal profile used a 10 minutes initial denaturation at 95°C, followed by 35 cycles of 30 seconds at 95°C, 45 seconds annealing at a primer-specific temperature (48°C for COI and 57°C for COII), 45 seconds at 72°C, and a final step of 15 minutes at 72°C. After verification via gel electrophoresis, the PCR templates were purified and then sequenced in both directions using the same primer pairs on an ABI3730XL automated DNA sequencer (Applied Biosystems, Foster City, CA, USA).
Data analysis
We used MICRO-CHECKER version 2.2.3 [27] to check the data for null alleles, stutter-errors and large allele drop-out. Null allele frequencies at each locus were calculated using the R PACKAGE GENELAND 3.1.4 [28], [29]. GENEPOP version 4.0.1 [30] was used to calculate allele frequencies, number of alleles per locus, the observed (Ho) and expected heterozygosity (He), inbreeding coefficient (FIS), the genotypic linkage disequilibrium, deviations from Hardy-Weinberg equilibrium, and the Hardy-Weinberg equilibrium P values. ADZE (Allelic Diversity Analyzer) version 1.0 [31] was utilized for calculating the mean allelic richness per locus for all the sampling populations with a rarefaction method that adjusted the observed number of alleles to a common size for each population sample.
Population structure from microsatellite data was analyzed using STRUCTURE version 2.3.3 [32] based on Bayesian clustering approach. We used the admixture ancestry model and the correlated allele frequency model. The correct number of clusters (K) was determined by calculating the probability of the data given a certain prior value of K. Markov chain Monte Carlo (MCMC) was carried out using 20 replicate runs of 1 000 000 iterations following a burn-in period of 50 000 iterations for each K-value from 1 to 10. The most likely value of K was estimated according to Evanno et al. (2005) [33]. The Nei's standard genetic distances among populations were utilized to construct Neighbor-joining (NJ) tree using PHYLIP version 3.6a [34]. The tree was visualized using TREEVIEW version 2.0 software [35]. In addition, we determined private allele estimates using HP-RARE [36].
We used ARLEQUIN version 3.5.1.2 [37] to estimate the molecular variance (AMOVA) between the groups of populations, along with the pair fixation indices (FST), and significance levels were assessed (P-values) using 10,000 permutations of diploid multilocus genotypes between samples [38]. A two-tailed t test at the significant level 0.05 was used to test whether differentiation between pome fruit populations was less than differences between pome fruit and peach populations; pairwise FST values between pome fruit populations were compared to pairwise FST values difference between pome fruit and peach populations in the test.
For AMOVA, samples were grouped according to three models: (A) ‘comparison of variance among peach populations from different regions’. In this model, 13 populations from peach were divided into three groups by geographic regions: (i) TaiAn (6TNJ-Ph, 6TYT-Ph, 6TNJ-Ph, 7TNM-Ph, 7TYT-Ph, 7TCL-Ph and 8TNM-Ph); (ii) WeiFang (6WHS-Ph, 7WHS-Ph, 8WHS-Ph and 8WPY-Ph); (iii) YanTai (7YFS-Ph and 8YXS-Ph). (B) ‘comparison of variance among pome fruit populations from different regions’. In this model, genetic differentiation of four pome fruit populations from different regions were separated into two groups according to geographic regions: (i) TaiAn (8TCL-Ap); (ii) YanTai (8YXS-Ap, 8YFS-Pr and 8YLY-Pr). (C) ‘comparison of variance among populations from peach and pome fruits. In this model, all 17 populations were divided into two groups by the types of host plants: (i) peach (6TNJ-Ph, 6TYT-Ph, TNJ-Ph, 7TNM-Ph, 7TYT-Ph, 7TCL-Ph, 8TNM-Ph, 6WHS-Ph, 7WHS-Ph, 8WHS-Ph, 8WPY-Ph, 7YFS-Ph, and 8YXS-Ph); (ii) pome fruit (8TCL-Ap, 8YXS-Ap, 8YFS-Pr and 8YLY-Pr).
In order to test for isolation by distance (IBD), the matrices of genetic distance FST/(1- FST) of microsatellite data and the geographic distance (lnKm) were compared using the Mantel test with10,000 permutations [39]. This analysis was performed using the ZT-software [40].
The obtained mitochondrial DNA sequences were aligned using CLUSTAL_X version1.8 [41]. All population genetic parameters including number of haplotypes (Nh), nucleotide diversity (π) and haplotype diversity (Hd) were calculated using the program DnaSP version 5.0 [42].
To study the genetic relationships among haplotypes, we used the program NETWORK 4.6.1 [43] to construct the Median-joining networks of mtDNA haplotypes basing on statistical parsimony, and Program MEGA 5.0 [44] was used to reconstruct the NJ tree under the p-distances simple assumptions.
Results
Microsatellite marker and mitochondrial DNA
We genotyped 399 individuals at five microsatellite loci. Population statistics for the 399 analyzed individuals of 17 investigated populations are given in Table 2. The mean number of alleles per locus ranged from 7.0 to 16.4, reflecting relative higher numbers of alleles in comparison with the invasive populations of the pest [24]. No significant linkage disequilibrium (P<0.05) was found of all possible combinations after Bonferroni correction, which suggests that the five loci represented independent information across all the samples. The frequency of null alleles ranged from 0.13 to 0.28, which are similar values to previously reported ones from this tortricid pest [3], [24] and are typical for lepidopterans [5], [24], [45], [46].
In all the 399 individuals analyzed, we obtained 385 COI (740 bp) and 396 COII (700 bp) gene sequences, and observed 34 haplotypes for the COI gene (GenBank accession numbers are KF552033 to KF552066) and 38 haplotypes for the COII gene (GenBank accession numbers are KF551995 to KF552032).
Genetic variation and genetic diversity
Microsatellite analysis.
In the analyzed G. molesta populations sampled in China (Table 2), a total of 126 alleles were found across the five genotyped loci. The mean number of alleles per locus ranged from 6.4 to16.4. We found a high number of private alleles (between 1.4 and 8.8 per population) as well as high allele richness (between 6.2 and 14.9). The mean value of He was between 0.66 and 0.89, that of Ho between 0.27 and 0.63. Eight of the 17 populations revealed significant departures from Hardy-Weinberg equilibrium, together with high positive mean values of FIS ranging from 0.23 to 0.65, indicating the existence of heterozygote deficiencies (Table 2).
mtDNA gene analysis.
Five of the observed 34 G. molesta COI gene haplotypes were shared between populations, 29 were private. Sixteen of 38 COII gene haplotypes were shared between populations, 22 were private. The COII and COI alignment had 21 (3%) and 28 (4%) variable characters, respectively, and no indels were observed in either of the two gene regions. As is common for an insect mitochondrial gene, the base composition of each of the two genes was biased toward As and Ts (76.4% of COI and 72.7% of COII). The COI gene covered 28 variable sites, 18 of which were parsimony informative, whilst 13 of the 21 polymorphic sites in COII gene were parsimony informative.
The haplotype diversity (Hd) ranged from 0.63 to 0.95 for COII, and from 0.34 to 0.87 for COI (Table 3). The number of haplotypes was remarkably high. The average number of haplotypes (Nh) of COI over all populations was 5.06, ranging from 3 to 10, and for COII, the mean number of haplotypes was 8.12, ranging from 5 to 14. The average nucleotide diversity (π) of COI was 0.14%, ranging from 0.11% to 0.24%, for COII the mean nucleotide diversities (π) was 0.26%, ranging from 0.13% to 0.38% (Table 3).
Genetic structure
Microsatellite cluster analysis.
The most likely value of K was 4 [46] indicating that the 17 populations included in this study can be assigned to four clusters, which are hereafter referred to as cluster 1, cluster 2, cluster 3 and cluster 4. The proportions of each population that contributed to each of the four clusters are shown in Fig. 2. Cluster 1 mainly consisted of individuals from the host plant peach in the TaiAn region (6TNJ-Ph, 0.652, implying that 65.2% of individuals from 6TNJ-Ph contributed to this cluster; 6TYT-Ph, 0.489; 7TNJ-Ph, 0.451; 7TNM-Ph, 0.701; 7TYT-Ph, 0.527; 7TCL-Ph, 0.925 and 8TNM-Ph, 0.800). Cluster 2 comprised all four populations sampled from peach at two sublocations in the WeiFang region (6WHS-Ph, 0.629; 7WHS-Ph, 0.939; 8WHS-Ph, 0.632 and 8WPY-Ph, 0.832). Cluster 3 was mainly made up of two populations sampled from peach in the YanTai region (7YFS-Ph, 0.876 and 8YXS-Ph, 0.90). Cluster 4 primarily contained individuals from the populations sampled in the late season from apple and pear in several regions (8TCL-Ap, 0.623; 8YXS-Ap, 0.576; 8YFS-Pr, 0.870 and 8YLY-Pr, 0.460).
Proportion of membership coefficient for 17 G. molesta populations falling into the four clusters is depicted by four different colors, respectively. The codes for populations are explained in Table 1.
Neighbor-joining tree derived from microsatellite allele frequencies.
To assess whether the four clusters obtained by Bayesian clustering analysis using Structure were genetically distinct, we further analysed the relationship among populations using NJ tree [47] based on Nei's genetic distance (Fig. 3). The NJ tree resulted in four major clades which were consistant with the Bayesian clustering results of the microsatellite data. All seven populations (6TNJ-Ph, 6TYT-Ph, 7TNJ-Ph, 7TNM-Ph, 7TYT-Ph, 7TCL-Ph, and 8TNM-Ph) collected from peach in TaiAn made up one clade. Similarly, all four populations (6WHS-Ph, 7WHS-Ph, 8WHS-Ph, and 8WPY-Ph) sampled from peach in WeiFang were found in the same clade, and the two populations (7YFS-Ph and 8YXS-Ph) from peach in YanTai formed a clade. Populations (8TCL-Ap, 8YXS-Ap, 8YFS-Pr and 8YLY-Pr) sampled in the late season from the pome fruit apple and pear of various regions constitute a single clade, which revealed the same genetic structure as the afore mentioned Bayesian clustering analysis.
Seventeen G. molesta populations were assigned to four clades indicated by I, II, III, IV in correspondence to clusters 1–4 of the Bayesian clustering analysis. I, samples from peach in TaiAn; II, samples from peach in WeiFang; III, samples from peach in YanTai; IV, samples from pear and apple in TaiAn, WeiFang or YanTai. The codes for sublocations are explained in Table 1.
Pairwise FST analysis from microsatellite loci.
Pairwise FST analysis was calculated for each pair of populations over five microsatellite loci with values ranging from 0.021 to 0.255. The significance of pairwise FST was determined with a Markov chain analysis. The results showed significant differentiation between all pairs of populations with the exception of one pair (6TYT-Ph and 7TYT-Ph) (Table 4). The two-tailed t test results showed that the differentiation among pome fruit populations (mean pairwise FST = 0.022±0.0002) was less than the differentiation between pome fruit and peach populations (mean pairwise FST = 0.050±0.002) with a P value of 0.010 (t = 1.967, df = 325).
Isolation by distance.
The Mantel test provided an r value of 0.363 (P = 0.005) based on all 13 populations from peach, indicating that there was a significant isolation-by-distance effect present among different geographic populations from this host. In contrast, when the four populations from the pome fruit hosts were added, this correlation between genetic distances and geographic distances was no longer significant (r = 0.086, P = 0.253).
Haplotype network of the combined mtDNA.
A total of 85 haplotypes were obtained with the combination of COI and COII genes. The network of the haplotypes are shown in Fig. 4. The nine most common haplotypes were shared by the populations of all the three regions (YanTai, TaiAn and WeiFang), of which eight were obtained from both peach and pome fruits (PP5, PP15, PP16, PP17, PP18, PP20, PP23 and PP29), one from peach (PE7). In TaiAn, WeiFang and YanTai, 40, 37 and 35 haplotypes were observed, respectively. Populations from all the three regions showed a high number of private haplotypes (with 24, 21 and 24 in TaiAn, WeiFang and YanTai, respectively). TaiAn and WeiFang populations shared five haplotypes (PE1, PE6, PE27, PE28 and PP4). No other haplotypes were observed to be shared by two regions.
Each circle represents a haplotype, and the area of a circle or square is proportional to the number of observed individuals. Colors within the nodes refer to the G. molesta sampling regions: green, TaiAn; red, YanTai; blue, WeiFang; black, the inferred missing haplotypes. The two capital letters each, PE, PO and PP indicate haplotpyes from peach, pome fruit (apple and pear) and both peach and pome fruit, respectively.
NJ tree of the combined COI and COII haplotypes.
NJ tree of combined G. molesta COI and COII haplotypes based on p-distance is shown in Fig. 5. The analysis based on the combination of COI and COII gene divided the haplotypes into eight major clusters (W1, W2, Y1, Y2, TW1, TW2, TW3 and TY). The W1 and W2 cluster mainly contained the haplotypes identified in the WeiFang region. The Y1 and Y2 cluster mainly comprised haplotypes from the YanTai region. The TW1, TW2 and TW3 cluster was primarily a mixture of haplotypes from both the TaiAn and WeiFang region. The TY cluster was mainly made up of haplotypes from both the TaiAn and YanTai region.
Eight clusters are indicated by TY, TW1,TW2, TW3, W1, W2, Y1 and Y2.
Analysis of molecular variance (AMOVA)
AMOVA results for the three models are shown in Table 5. Significant, albeit low level genetic variances were found in all three models, with populations grouped according to (A, B) region and (C) type of host plants.
Discussion
We investigated the genetic diversity and genetic structure of 17 G. molesta populations from three regions of a main fruit growing area in China, using microsatellite and mitochondrial DNA markers. High genetic diversity was documented in these moth populations. All populations from peach were geographically structured. Whereas genetic structure of populations from peach was similar throughout the growing season within each region, populations collected from apple and pear in the late season surprisingly differed in genetic structure from populations collected from peach.
Genetic diversity
This study provided an extended sample analysis from the putative native region of G. molesta. The mean number of alleles per locus (NA) averaged 10.4 over all populations, which was about twice as high as reported for G. molesta populations that invaded Italy (average 5.8) [24], Australia, South Africa, South and North American and European countries (average 4.4) [3]. Moreover, we found high numbers of private alleles of microsatellite loci and high allelic richness at each sublocation sampled. All these characteristics of high genetic diversity found with microsatellites are indicators of a native species [48]–[50]. Furthermore, all populations displayed large numbers of mitochondrial haplotypes, with a total of 86 haplotypes obtained for COI (average 5.1 over populations), and of 138 haplotypes for COII (average 8.1 over populations). High numbers of private haplotypes existed in each of the three regions. Both these characteristics found with mitochondrial markers are again indicators of native species, especially for native species with low dispersal capacity [7], [50]. In fact, the criterion of low dispersal capacity applies also to G. molesta, which has the flight ability to move between non-contiguous orchards but not over long distances [17], [23], [51]. Previous molecular studies on the oriental fruit moth relied on populations sampled globally or in invaded regions [3], [24], whereas our data provide for the first time comprehensive empirical evidence for the assumption [9] that China lies indeed in the range of G. molesta's origin.
Geographical genetic structure
Populations from peach were geographically structured in the three sampled regions of China, as documented by the well defined clusters formed by populations from each region based on microsatellite markers. There was weak but significant correlation between geographical distance and genetic distance among these populations. Our microsatellite data provided stronger information on geographical population structure than mitochondrial data did in the NJ tree of the combination of COI and COII haplotypes. This difference may be due to the diploid nature of microsatellites as opposed the haploid nature of mitochondrial markers, as was concluded in a review on population genetics information obtained from these markers [7]. The sampling in the present study was carried out in one of the most important fruit growing areas of China with a long history of fruit tree planting. The geographically structured populations found are likely the result of the limited flight capacity of G. molesta [17], [23], [51], so that isolation sufficed to restrict spread of the species. Our findings from China are in line with results from South Africa on localized populations [52], from Italy on significant intra-regional structure [24], and from a global survey reporting geographical structure on a continental scale [3]. Similar results on genetic structure were found in a closely related moth species with low flight capacity, Cydia pomonella [5], [52]–[54]. In the current study, we found consistently similar genetic structures within a region (Fig. 2), and human-mediated displacement [55] might have contributed to this result. Actually, local Chinese governments promote centralized purchase of fruit, what might facilitate genetic exchange of G. molesta populations within regions, as is also the case for cooperatives in Italy [24].
Genetic variance among all populations sampled existed (Table 5). The only exception (6TTY-Ph and 7TYT-Ph) without significant genetic variance (P = 0.216) was in a heavily infested abandoned orchard, which seemed to have provided a particularly stable environment for the moth. Otherwise, environmental conditions in single orchards sufficiently changed between sampling times to allow for detection of significant genetic variance. Such environmental changes include the considerable changes in emitted volatile blends from host trees with progressing season, which elicit altered behavioral responses of the oriental fruit moth [18], [20].
Genetic structure across host plants
Populations sampled in the late season from the pome fruit apple and pear differed in genetic structure from populations sampled from the primary host peach, and they form a single clade exclusively comprising populations from pome fruit. Three factors might have contributed to this remarkable finding: the increased flight performance of the moth in the late season [56], the altered olfactory preference of gravid female to changing volatile blends emitted by peach and pome fruit with progressing season [18]–[20], and the recently detected existence of different olfactory genotypes in this moth species [21]. The first two factors are associated with the movement of the moths from the stone fruit peach to the pome fruit apple and pear. Decreasing length of daylight experienced by the growing larvae promote enhanced flight performance of the emerged G. molesta moths, and this effect is even stronger when ambient temperature is still high [56]. Furthermore, the volatiles emitted by apple and pear become attractive for the female moths towards the late season, facilitating a host switch during this period [18], [20]. Beyond these two factors, which increase flight and facilitate orientation towards pome fruit trees, the following third factor may contribute to explain the different genetic structure of the populations from pome fruit compared to populations from peach in the late season. In the oligophagous G. molesta, there are different olfactory genotypes, and this natural genetic diversity facilitates host switch [21] and likely contributes to the differential population structure documented in the current study for larvae from pome fruit compared to those from peach. Our results from the population structure analysis and from the NJ tree consistently suggest a selective host switch of a certain part of the population only. Certain genotypes from peach, possibly those particularly responsive to pome fruit odors, seem to be the major contributors to the host switch.
Implication for pest management
The oriental fruit moth is not only a key pest of rosaceae fruit trees in its native range in Asia, requiring major pest management input [11], [12], [57], but has invaded all other fruit growing continents, posing major challenges to sustainable fruit tree protection [10], [51], [58]. The current study is the first to show that population genetic structure of this pest may be changing with the host switch from the primary host peach to the secondary hosts pear and apple in the late part of the season, and this finding might provide useful information for the design of future season-long pest management programs. For example, late season pest management in pome fruit might rely on different pest management tools than those used in the early season on peach. This strategy holds the potential for two benefits: first, it can be adjusted optionally to the genotypes prevailing in the respective pest populations, and second, the alternation of the management approach in the second part of the season might contribute to limit univocal selection pressure and thus forced evolution of resistant pest genotypes [20], [59], [60].
Conclusions
Genetic diversity of G. molesta is high in its region of origin. Significant genetic structure exists among populations from peach in the different regions. The weak flight capacity of the moth underlies the pattern of genetic structure observed. Contrary to our original expectation, host switch of the moth results in different genetic structure of populations on the secondary host plants, to which the recently detected different olfactory genotypes of the moth likely contribute. Design and implementation of sustainable pest management strategies could therefore be adjusted to the population genetics and dispersal of this pest insect.
Acknowledgments
We thank Dr. Heather Kirk, Dr. Kathrin Tschudi-Rein, Dr. Karina L. Silva-Brandão and a further reviewer for helpful comments on the manuscript. We are grateful to Ming-Jing Qu, Yan-Lin Jiao, Rui-Hong Sun for their help with the collection of oriental fruit moth samples.
Author Contributions
Conceived and designed the experiments: YZ SD MC. Performed the experiments: YZ. Analyzed the data: YZ MC. Contributed reagents/materials/analysis tools: XP GL HP. Wrote the paper: YZ SD MC.
References
- 1. Miller NJ, Birley AJ, Overall ADJ, Tatchell GM (2003) Population genetic structure of the lettuce root aphid, Pemphigus bursarius (L.), in relation to geographic distance, gene flow and host plant usage. Heredity 91: 217–223.
- 2. Sword GA, Joern A, Senior LB (2005) Host plant-associated genetic differentiation in the snakeweed grasshopper, Hesperotettix viridis (Orthoptera: Acrididae). Mol Ecol 14: 2197–2205.
- 3. Kirk H, Dorn S, Mazzi D (2013) Molecular genetics and genomics generate new insights into invertebrate pest invasions. Evol Appl 6: 842–856.
- 4. Timm AE, Geertsema H, Warnich L (2008) Population genetic structure of Grapholita molesta (Lepidoptera: Tortricidae) in South Africa. Ann Entomol Soc Am 101: 197–203.
- 5. Chen MH, Dorn S (2010) Microsatellites reveal genetic differentiation among populations in an insect species with high genetic variability in dispersal, the codling moth, Cydia pomonella (L.) (Lepidoptera: Tortricidae). Bull of Entomol Res 100: 75–85.
- 6. Karsten M, van Vuuren BJ, Barnaud A, Terblanche JS (2013) Population Genetics of Ceratitis capitata in South Africa: implications for dispersal and pest management. PLoS ONE 8: e54281.
- 7. Puillandre N, Dupas S, Dangles O, Zeddam JL, Capdevielle-Dulac C, et al. (2008) Genetic bottleneck in invasive species: the potato tuber moth adds to the list. Biol Invasions 10: 319–333.
- 8. Yang XM, Sun JT, Xue XF, Li JB, Hong XY (2012) Invasion genetics of the western flower thrips in China: evidence for genetic bottleneck, hybridization and bridgehead effect. PLoS ONE 8: e34567.
- 9.
Rothschild GHL, Vickers RA (1991) Biology, ecology and control of the oriental fruit moth. In: Tortricid pests: their biology, natural enemies and control (eds van der Geest LPS, Evenhuis HH). Elsevier Amsterdam: 389–412.
- 10. Kirk H, Dorn S, Mazzi D (2013) Worldwide population genetic structure of the oriental fruit moth (Grapholita molesta), a globally invasive pest. BMC Ecol 13: 12.
- 11. Feng MX, Jiang RD, Wang P, Wang JQ (2002) Population dynamics of Grapholita molesta (Busck) in peach orchards. China Fruits 4: 30–31.
- 12. Li T, Lian M, Ma PS, Zhang YX, Zhou YN (2010) Damage and control of the oriental fruit moth, Grapholitha molesta (Busck) in peach orchard. J Shanxi Agric Sci 5: 47–50.
- 13. Pollini A, Bariselli M (1993) Cydia molesta: pest on the increase and defence of pome fruits. Inf Agrar 14: 19–21.
- 14. Dorn S, Hughes J, Molinari F, Cravedi P (2001) Cydia molesta and Cydia pomonella: comparison of adult behaviour. IOBC WPRS Bull 24: 133–138.
- 15.
Bovey A (1979) La Défense des Plantes Cultivées. Lausanne: Éditions Payot.863p.
- 16. Myers CT, Hull LA, Krawczyk G (2007) Effects of orchard host plants (apple and peach) on development of oriental fruit moth (Lepidoptera: Tortricidae). J Econ Entomol 100: 421–430.
- 17. Hughes J, Dorn S (2002) Sexual differences in the flight performance of the oriental fruit moth, Cydia molesta. Entomol Exp Appl 103: 171–182.
- 18. Piñero JC, Dorn S (2009) Response of female oriental fruit moth to volatiles from apple and peach trees at three phenological stages. Entomol Exp Appl 131: 67–74.
- 19. Lu PF, Huang LQ, Wang CZ (2012) Identification and field evaluation of pear fruit volatiles attractive to the oriental fruit moth, Cydia molesta. J Chem Ecol 38: 1003–1016.
- 20. Najar-Rodriguez A, Bellutti N, Dorn S (2013) Larval performance of the oriental fruit moth across fruits from primary and secondary hosts. Physiol Entomol 38: 63–70.
- 21. Najar-Rodriguez A, Schneeberger M, Bellutti N, Dorn S (2012) Variation in attraction to host plant odors in an invasive moth has a genetic basis and is genetically negatively correlated with fecundity. Behav Genet 42: 687–697.
- 22. Sziraki G (1979) Dispersion and movement activity of the oriental fruit moth Grapholita molesta (Busck) in large-scale orchards. Acta Phytopathologica Academiae Scientiarum Hungaricae 14: 209–228.
- 23. Sciarretta A, Trematerra P (2006) Geostatistical characterization of the spatial distribution of Grapholita molesta and Anarsia lineatella males in an agricultural landscape. J Appl Entomol 130: 73–83.
- 24. Torriani MVG, Mazzi D, Hein S, Dorn S (2010) Structured populations of the oriental fruit moth in an agricultural ecosystem. Mol Ecol 19: 2651–2660.
- 25. Schuelke M (2000) An economic method for the fluorescent labeling of PCR fragments. Nat Biotechnol 18: 233–234.
- 26. Simon C, Prati F, Beckenbach A, Crespi B, Liu H (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann Entomol Soc Am 87: 651–701.
- 27. Van Oosterhout C, Hutchinson WE, Wills DPM, Shipley P (2004) Micro-checker: software for identifying and correcting genotyping errors in microsatellite data. Mol Eco Notes 4: 535–538.
- 28. Guillot G, Estoup A, Mortier F, Cosson J (2005) A spatial statistical model for landscape genetics. Genetics 170: 1261–1280.
- 29. Guillot G, Frédéric M, Arnaud E (2005) GENELAND: a computer package for landscape genetics. Mol Ecol Notes 5: 712–715.
- 30. Rousset F (2008) GENEPOP'007: a complete re-implementation of the GENEPOP software for Windows and Linux. Mol Ecol Resour 8: 103–106.
- 31. Szpiech ZA, Jakobsson M, Rosenberg NA (2008) ADZE: a rarefaction approach for counting alleles private to combinations of populations. Bioinformatics 24: 2498–2504.
- 32. Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155: 945–959.
- 33. Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol 14: 2611–2620.
- 34. Felsenstein J (1989) Phylip-phylogeny inference package (version 3.2). Cladistics 5: 164–166.
- 35. Page RDM (1996) Treeview: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 12: 357–358.
- 36. Kalinowski ST (2005) HP-RARE 1.0: a computer program for performing rarefaction on measures of allelic richness. Mol Ecol Notes 5: 187–189.
- 37. Excoffier L, Lischer HL (2010) Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol 10: 564–567.
- 38. Schneider S, Roessli D, Excoffier L (2000) Arelquin: A software for population genetic data analysis. User manual ver 2: 2496–2497.
- 39. Mantel N (1967) The detection of disease clustering and a generalized regression approach. Canc Res 27: 209–220.
- 40. Bonnet E, Van der Peer Y (2002) ZT: a software tool for simple and partial Mantel tests. J Stat Software 7: 1–12.
- 41. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The clustal_X Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25: 4876–4882.
- 42. Librado P, Ozas JR (2009) DnaSP ver. 5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451–1452.
- 43. Bandelt H-J, Forster P, Röhl A (1999) Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol 16: 37–48.
- 44. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA 5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739.
- 45. Dakin EE, Avise JC (2004) Microsatellite null alleles in parentage analysis.Heredity. 93: 504–509.
- 46. Meglécz E, Petenian F, Danchin E, D'Acier AC, Rasplus JY, et al. (2004) High similarity between flanking regions of different microsatellites detected within each of two species of Lepidoptera: Parnassius apollo and Euphydryas aurinia. Mol Ecol 13: 1693–1700.
- 47. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425.
- 48. Grapputo A, Boman S, Lindström L, Lyytinen A, Mappes J (2005) The voyage of an invasive species across continents: genetic diversity of North American and European Colorado potato beetle populations. Mol Ecol 14: 4207–4219.
- 49. Lehrian S, Balint M, Haase P, Pauls SU (2010) Genetic population structure of an autumn-emerging caddisfly with inherently low dispersal capacity and insights into its phylogeography. J N Am Benthol Soc 29: 1100–1118.
- 50. Singh YT, Mazumdar-Leighton S, Saikia M, Pant P, Kashung S, et al. (2012) Genetic variation within native populations of endemic silkmoth Antheraea assamensis (Helfer) from Northeast India indicates need for in situ conservation. PLoS ONE 7: e49972.
- 51. Ellis NH, Hull LA (2012) Factors influencing adult male Grapholita molesta dispersal in commercial Malus and Prunus host crops. Entomol Exp Appl 146: 232–241.
- 52. Timm AE, Geertsema H, Warnich L (2006) Gene flow among Cydia pomonella (Lepidoptera: Tortricidae) geographic and host populations in South Africa. J Econ Entomol 99: 341–348.
- 53. Fuentes-Contreras E, Reyes M, Barros W, Sauphanor B (2007) Evaluation of azinphosmethyl resistance and activity of detoxifying enzymes in codling moth (Lepidoptera: Tortricidae) from central Chile. J Econ Entomol 100: 551–556.
- 54. Thaler R, Brandstätter A, Meraner A, Chabicovski M, Parson W, et al. (2008) Molecular phylogeny and population structure of the codling moth (Cydia pomonella) in Central Europe: II. AFLP analysis reflects human-aided local adaptation of a global pest species. Mol Phylogenet Evol 48: 838–849.
- 55. Mazzi D, Dorn S (2012) Movement of insect pests in agricultural landscapes. Ann App Biol 160: 97–113.
- 56. Hughes, Hern A, Dorn S (2004) Preimaginal environment influences adult flight in Cydia molesta (Lepidoptera: Tortricidae). Environ Entomol 33: 1155–1162.
- 57. Li B, Qin YC, He L, Wu XM (2008) The control oriental fruit moth with sex pheromone dispensers and mixtures of sugar-acetic acid-ethanol. Acta Phytophylacica Sinica 35: 285–286.
- 58. Il'ichev AL, Williams DG, Gut LJ (2007) Dual pheromone dispenser for combined control of codling moth Cydia pomonella L. and oriental fruit moth Grapholita molesta (Busck) (Lep., Tortricidae) in pears. Appl Entomol 131: 368–376.
- 59. Loxdale HD, Schöfl G, Wiesner KR, Nyabuga FN, Heckel DG, et al. (2011) Stay at home aphids: comparative spatial and seasonal metapopulation structure and dynamics of two specialist tansy aphid species studied using microsatellite markers. Biol J Linn Soc Lond 104: 838–865.
- 60. Domingues FA, Silva-Brandão KL, Abreu AG, Perera OP, Blanco CA, et al. (2012) Genetic structure and gene flow among Brazilian populations of Heliothis virescens (Lepidoptera: Noctuidae). J Econ Entomol 105: 2136–2146.